A toxic effector of T7SS facilitates bacterial competition and virulence through membrane damage in Streptococcus suis

TL;DR

This paper identifies a toxin from Streptococcus suis that helps the bacteria compete with others and cause disease by damaging cell membranes.

Contribution

The study reveals LXG-T2 as a novel T7SS effector that causes membrane disruption and enhances virulence in Streptococcus suis.

Findings

LXG-T2 shows strong bactericidal activity against E. coli and gives S. suis a competitive advantage.

LXG-T2 compromises bacterial membrane integrity, increasing permeability and depolarization.

LXG-T2 promotes S. suis survival in a murine infection model and exhibits cytotoxic effects on host cells.

Abstract

Streptococcus suis is a zoonotic pathogen that poses a significant threat to both the swine industry and human health. This bacterium utilizes a type VII secretion system (T7SS) to translocate effector proteins that mediate bacterial competition and contribute to virulence. However, the functions of T7SS effectors in S. suis remain poorly understood. In this study, we identified and characterized LXG-T2, a T7SS-secreted toxin from S. suis virulent strain WUSS351. Bioinformatics analysis revealed that LXG-T2 harbors a C-terminal glycine zipper motif, a structural feature commonly associated with membrane-disrupting toxins. Functional assays demonstrated that LXG-T2 exhibits strong bactericidal activity against E. coli and provides S. suis with a competitive advantage. Furthermore, the LXG-T2 has the capacity to compromise the integrity of bacterial membranes, as evidenced by the observed increase in membrane permeability and depolarization in target cells. Moreover, LXG-T2 exhibited cytotoxic effects on host cells and promoted S. suis survival in a murine infection model. Collectively, our findings establish LXG-T2 as a T7SS effector mediated membrane disruption to enhance both bacterial competition and virulence. This work not only reveals a novel mechanism by which S. suis manipulates microbial communities, but also highlights the significance of T7SS effectors as key mediators of pathogenesis in Gram-positive bacteria.

The online version contains supplementary material available at 10.1186/s13567-025-01706-6.

Figures

Full text

Introduction

Many bacterial species utilize specialized secretion systems to deliver effector proteins into host cells or release them into the extracellular environment [1]. The type VII protein secretion system (T7SS) was first discovered in Mycobacterium tuberculosis and has since been identified in a variety of Gram-positive bacteria [2–4]. As a specialized protein secretion pathway, the T7SS facilitates bacterial survival, proliferation, and dissemination in the host [5]. Moreover, it plays a key role in mediating interbacterial competition and contributes significantly to the pathogenesis of bacterial pathogens [6]. The core components of the T7SS apparatus include EsxA, EssA, EssB, EsaB, EssC, and EsaA, forming the essential secretion machinery within the bacterial envelope, as characterized in Staphylococcus aureus [5, 6].

LXG proteins, initially identified by Aravind et al., belong to a broad family of polymorphic toxins and are predicted to function as T7SS substrates [7]. These proteins typically possess an N-terminal LXG domain that mediates secretion via the T7SS, and a highly variable C-terminal region responsible for the toxic activity [8, 9]. Recent studies have shown that several Firmicutes, including S. aureus, Bacillus subtilis, and Streptococcus intermedius, secrete LXG effectors with diverse C-terminal toxic domains that contribute to bacterial competition and virulence in a T7SS-dependent manner. For example, LXG effectors function as a phospholipase A (TslA) and a membrane-depolarizing toxin (TspA) in S. aureus [10, 11], as a NADase (TelB) and a lipid II phosphatase (TelC) in S. intermedius [12], and as DNases in B. subtilis, which harbors six chromosomal LXG toxin–immunity operons, five of which encode DNase toxins that mediate competitive interactions [13].

Effectors secreted by various secretion systems typically target essential cellular components such as nucleic acids, membranes, and peptidoglycans [12, 14, 15]. Kim et al. identified a conserved sequence motif, known as the glycine zipper (GxxxGxxxG), which is frequently found in membrane-associated proteins [16]. These motifs facilitate the formation of amphipathic helices that insert into lipid bilayers and disrupt membrane integrity [17]. Effectors such as CdzC and CdzD from Caulobacter crescentus (type I secretion system) and Tse4 from Pseudomonas aeruginosa (type VI secretion system, T6SS) employ glycine zipper motifs to mediate antibacterial activity [18, 19]. However, the role of glycine zipper motifs in the T7SS-mediated toxicity remains poorly understood.

Streptococcus suis is one of the major pathogens threatening the swine industry that primarily colonizes the tonsils and nasal cavities of pigs [20, 21]. It can cause a range of severe diseases, including meningitis, arthritis, endocarditis, septicemia, and sudden death [22, 23]. Importantly*, S. suis* is recognized as a zoonotic pathogen that can lead to meningitis, sepsis, and endocarditis in humans [24].

In our previous research, we identified S. suis virulent strain WUSS351, which encodes three LXG effectors: LXG-T1, LXG-T2, and LXG-T3 [25]. In this study, we focused on LXG-T2, which contains a glycine zipper motif at its C-terminus. We demonstrate that LXG-T2 mediates bacterial competition and contributes to pathogenesis by compromising membrane integrity. Our findings enhance the understanding of T7SS-mediated competitive mechanisms in Gram-positive bacteria and elucidate the pathogenic roles of S. suis.

Materials and methods

Bioinformatics analysis

Pfam domains within the LXG-T2 protein were identified using the SMART database [26]. Multiple sequence alignment was performed using ClustalW, implemented within the MEGA X software package (version 10.1.8). Subsequently, a phylogenetic tree was constructed via Maximum-Likelihood analysis. Transmembrane helices in the LXG-T2 effector were predicted using TMHMM [27]. Regional sequence logos (probability-weighted Kullback-Leibler logo type) were generated using the web-based Seq2Logo 2.0 server [28]. The structural model of LXG-T2 was generated using AlphaFold2. Homolog identification was performed using the Protein Homology/analogY Recognition Engine (PHYRE2) version 2.2.

Bacterial strains and culture conditions

All strains and plasmids used in this study are listed in Additional file 1. S. suis was cultured in Todd–Hewitt broth (THB, Hopebio, Qingdao, China) or on 6% (v/v) sheep blood agar at 37 °C in the presence of 5% CO_2_. Spectinomycin (100 µg/mL) or chloramphenicol (5 µg/mL) was added when necessary. Gene deletion mutants of the S. suis strain WUSS351 were constructed following a previously published protocol [29], and all deletions were confirmed by PCR and sequencing. Complementation strains were generated in situ by introducing a nonsense mutation to enable differentiation from the wild type (WT) strain, as previously described [30]. The E. coli strains used included Top10 for toxicity assays. E. coli was grown in Luria–Bertani (LB, Becton Dickinson, USA) medium with shaking at 37 °C and supplemented with ampicillin (100 µg/mL), kanamycin (50 µg/mL), chloramphenicol (20 µg/mL), or spectinomycin (50 µg/mL) as required. Primers are listed in Additional file 2.

Bacterial toxicity assay

To evaluate the toxicity of LXG-T2 and identify the corresponding immunity protein, E. coli TOP10 cells were transformed with pBADHisA plasmids. Primers are listed in Additional file 2. Overnight cultures of E. coli carrying pBADHisA (empty vector), pBADHisA-LXG-T2, and pBADHisA-LXG-T2 & Imm2 were diluted 1:100 in fresh LB medium containing 100 µg/mL ampicillin and grown for 2 h. L-arabinose was added to a final concentration of 0.2% to induce protein expression. Bacterial growth was monitored by measuring optical density at 600 nm (OD_600_) every 2 h, and tenfold serial dilutions were spotted onto LB agar plates containing ampicillin (100 µg/mL). Plates were incubated overnight at 37 °C and visualized using the Gel Doc XR1 system (Bio-Rad).

Bacterial competition assay

For competition assays on agar plates, attacker and target strains were adjusted to an OD_600_ of 0.6 in THB. The colony-forming unit (CFU) of the attacker and target strains were mixed at a 10:1 ratio. The supernatants of both strains were discarded after centrifugation at 5000 × g, and the cell pellets were resuspended in 100 μL phosphate-buffered saline (PBS). Then, 8 μL of the mixture was spotted onto LB agar plates and incubated at 37 °C under 5% CO_2_ for 24 h. After incubation, bacteria were recovered by adding 1 mL of PBS to the agar surface and gently pipetting to resuspend them. The CFUs were determined by plating serially diluted suspensions on the appropriate selective agar plates. All experiments were performed with three independent biological replicates.

Renilla luciferase (Rluc) assay

The tet promoter was cloned into the pKSM410 plasmid containing the Rluc fluorescent gene. S. suis WT and ΔessC1 were transformed with the pKSM410 plasmid, encoding the Rluc fragment fused to proteins of interest for testing effector secretion. Primers are listed in Additional file 2. THB cultures (1 mL) was inoculated with S. suis colonies from a THB agar plate and grown overnight. The cultures were then subcultured in fresh THB containing spectinomycin (100 µg/mL) and incubated at 37 °C with shaking for 2 h before the induction of 200 ng/mL Anhydrotetracycline (ATc). After an additional 2 h incubation, cultures were harvested at OD_600_ = 1. Cells were centrifuged at 5000 × g for 10 min at 4 °C, and 100 µL of the supernatant was transferred to a 96-well black opaque plate. To these wells, equal volumes of luciferase working solution were added, and luminescence was measured after a 2-min incubation using the M200Pro (Tecan, Switzerland). The relative luminescence was calculated as the ratio of luminescence in the WT strain supernatant to that of the ΔessC1 supernatant at the same time point.

Western blot

Both the empty vector pBADHisA and the recombinant plasmid pBADHisA-LXG-T2, which carries an N-terminal His-tag fusion, were transformed into E. coli TOP10 cells. Overnight cultures of the transformed strains were inoculated into 100 mL of LB medium containing 100 µg/mL ampicillin and incubated at 37 °C with shaking until the optical density at OD_600_ reached 0.6. Protein expression was induced by adding L-arabinose to a final concentration of 0.2%, and the cultures were incubated for an additional 3 h under the same conditions. Cells were harvested by centrifugation at 5000 × g for 5 min, washed twice with PBS and resuspended in loading buffer. Samples were boiled for 10 min and analyzed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), followed by western blotting. Proteins were transferred onto methanol-activated polyvinylidene fluoride (PVDF) membranes using a Trans-Blot Turbo Transfer System (Bio-Rad). The membranes were blocked with 5% (w/v) skim milk in PBST (PBS containing 0.05% Tween-20) for 2 h at room temperature, then incubated overnight at 4 °C with an anti-His primary antibody (Abmart, China) diluted 1:2000 in PBST. After three washes with PBST, membranes were incubated with a horseradish peroxidase (HRP)-conjugated anti-rabbit IgG secondary antibody (Abmart, China) diluted 1:5000 for 1 h at room temperature. Following additional washes, the blots were developed using an enhanced chemiluminescence (ECL) detection kit (Vazyme, China) according to the manufacturer’s instructions.

Scanning electron microscope (SEM) analysis

E. coli TOP10 harboring pBADHisA vector were grown overnight and then transferred to fresh medium supplemented with ampicillin and incubated for 2 h at 37 °C. Expression was induced with 0.2% L-arabinose. Cells were collected by centrifugation at 5000 × g to an OD_600_ of 0.6 with PBS and fixed with 2.5% glutaraldehyde at 4 °C for 24 h to preserve their structural. Samples were then dehydrated in graded ethanol solutions (30%, 50%, 70%, 90%, and 100%) for 10–15 min each. After drying and sputter coating with gold, the samples were imaged using a SEM (Hitachi, Japan).

Membrane permeability assay

To analyze the integrity of the cell membrane using Molecular Probes’ LIVE/DEAD BacLight Bacterial Viability Kits (Thermo Fisher, USA), we utilized E. coli TOP10 strains harboring the pBADHisA vector, which expresses LXG-CT2 and coexpresses the cognate immunity protein Imm2. E. coli were grown overnight in LB medium supplemented with ampicillin before being diluted 1:100 in fresh medium and incubated at 37 °C. Once the E. coli cultures reached the logarithmic phase, they were diluted to a concentration of 1 × 10^8^ cells per mL in sterile 0.85% NaCl solution. To this suspension, 1.5 μL of SYTO 9 stain and 1.5 μL of propidium iodide (PI) were added, and the mixture was incubated at room temperature in the dark for 15 min. A permeabilized control sample, consisting of bacteria harboring an empty vector and treated with 70% isopropyl alcohol, was also prepared and stained with both dyes. Subsequently, 5 μL of the stained sample was sandwiched between a microscope slide and an 18 mm square coverslip for imaging. Finally, bacterial were observed using a Nikon A1 confocal microscope (Nikon, Japan), with excitation and emission wavelengths set at 488 nm and 500/635 nm, respectively.

Membrane depolarization assay

Membrane depolarization was assessed using 3,3′-dipropylthiadicarbocyanine iodide [DiSC3(5)] (MCE, Cat No. HY-D0085). Strains were grown under the same conditions as those used for the Membrane Permeability Assay. Log-phase bacterial cultures were washed and resuspended in PBS to an OD_600_ of 0.6. A total of 100 μL of each sample was aspirated into a 96-well plate, and DiSC3(5) was added to a final concentration of 10 μM, followed by incubation for 30 min. The membrane depolarization-sensitive dye DiSC3(5) to measure fluorescence intensity using a multi-mode microplate reader (M200Pro, Tecan, Switzerland). Bacteria harboring empty vector served as the negative control, while bacteria treated with polymyxin B were used as a positive control.

Lactate dehydrogenase (LDH) cytotoxicity assay

HEK293T cells (abcam, Cat# ab259776) were cultured in Dulbecco’s modified eagle medium (DMEM; Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco) and 1% penicillin–streptomycin at 37 °C in a humidified incubator with 5% CO_2_. Cells were seeded into 24-well plates at 500 μL per well and preincubated for 24 h. Upon reaching approximately 80% confluency, cells were transfected using Lipofectamine^TM^ 3000 (Thermo Fisher, Cat# L3000001) according to the manufacturer’s protocol. pcDNA3.1-LXG-T2 plasmid and empty vector control (pcDNA3.1) plasmids were transfected at a dose of 500 ng per well. Primers are listed in Additional file 2. After transfection, cells were further incubated at 37 °C with 5% CO_2_ for 36 h prior to LDH release induction. At the end of the incubation period, the following treatments were applied: high control and high control blank wells received 10 μL of lysis solution to induce maximum LDH release, while low control wells received 10 μL of complete culture medium to measure spontaneous LDH release. The treated plates were incubated at 37 °C with 5% CO_2_ for 30 min. After incubation, 50 μL of supernatant from each well was transferred to a corresponding well in a 96-well plate. Then, 50 μL of LDH working solution (Cytotoxicity LDH Assay Kit, MCE) was added to each well. The plate was gently shaken to ensure proper mixing and incubated at room temperature in the dark for 30 min. To stop the reaction, 50 μL of stop solution was added to each well. Absorbance was measured at 490 nm using a microplate reader. The following control wells were included for accurate interpretation of LDH release: high control, cells plus medium with lysis solution, representing maximum LDH release; high control blank, medium with lysis solution but without cells, used to subtract the background from high control; low control, cells plus medium without lysis solution, representing spontaneous LDH release; background blank, medium alone (no cells, no lysis solution), used to subtract background absorbance from low control and experimental samples. Cytotoxicity was calculated using the following formula: Cytotoxicity (%) = [(X−Z)/(Y−Z)] × 100%. Where: X = OD_490_ of sample well − OD_490_ of background blank, Y = OD_490_ of high control − OD_490_ of high control blank, Z = OD_490_ of low control − OD_490_ of background blank.

Animal infection experiments

Mouse infection experiments were conducted at the Laboratory Animal Center of Nanjing Agricultural University, under permit number NJAU.No20251107245. Six-week-old female specific pathogen-free (SPF) BALB/c mice were purchased from the Comparative Medicine Center of Yangzhou University (Yangzhou, China). S. suis strains (WT, ΔLXG-T2, and C-LXG-T2) were cultured until reaching the exponential phase (OD_600_ = 0.6), washed then in PBS. Mice were randomly divided into three groups, and each group contained five mice and was intraperitoneally injected with WT, either ΔLXG-T2 or C-LXG-T2 at a dose of 3 × 10^8^ CFU/mouse. At 12 h postinfection, mice were euthanized, and samples from the spleen, liver, kidney, brain, and blood were collected. Organs were weighed, and homogenized in PBS. The number of viable bacteria in the organs and blood was determined by plating serial dilutions on THB.

Statistical analyses

Statistical tests, number of events quantified, standard deviation of the mean, and statistical significance was reported in figure legends. Statistical analysis was conducted using GraphPad Prism 9 software, and statistical significance was assessed using one-way analysis of variance (ANOVA) followed by Dunnett’s multiple comparisons test. Data were presented as mean ± SD. Ns indicates no significance, * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001, and **** indicates p < 0.0001.

Results

S. suis strain WUSS351 encodes a T7SS effector with a glycine zipper motif

Our previous investigations of the virulent strain WUSS351 of S. suis revealed the presence of a functional T7SS, a conserved protein export machinery involved in bacterial pathogenesis and competition (Figure 1A) [25]. Whole-genome sequencing identified a T7SS gene cluster encoding three distinct LXG effector genes inWUSS351 [25]. Remarkably, bioinformatics analyses and transmembrane topology modeling identified a putative glycine zipper motif within the C-terminal region of E8M06_09950*,* which was subsequently renamed LXG-T2 (Figure 1B). Given the well-characterized “self-protection” mechanism in bacterial secretion systems, wherein cognate immunity proteins neutralize effector toxicity via specific molecular interactions [8], we hypothesized that the gene downstream of LXG-T2, hereafter named Imm2**,** encodes its corresponding immunity protein (Figure 1A).Figure 1The effector LXG-T2 from S. suis WUSS351 T7SS contains a conserved Gly-zipper motif found across various pathogenic bacteria. A Schematic diagram of the T7SS locus in strain WUSS351. Core component genes (esxA, esaA, essA, esaB, essB, and essC1) are highlighted in yellow, LXG-T2 in blue, and its cognate immunity gene Imm2 in green. B Domain organization of LXG-T2 showing an N-terminal LXG domain and a conserved C-terminal Gly-zipper motif (GxxxGxxxGxxxG). C Phylogenetic analysis of LXG-T2 and its homologs in S. suis, Listeria, and Streptococcus species.

Domain architecture analysis using SMART (v9.0) and the NCBI Conserved Domain Database [26] revealed two organizations of LXG-T2: (i) an N-terminal LXG domain (residues 13–198), characteristic of T7SS substrates; (ii) a C-terminal glycine zipper motif embedded within the membrane-spanning region (Figure 1B). Furthermore, multiple alignment of LXG-T2 with homologs from S. suis, Streptococcus agalactiae (GBS), Streptococcus equi subsp*. zooepidemicus*, confirmed the conservation of a repeating GxxxG motif, characteristic of glycine zippers (Figure 1B). Notably, homologous toxin domains of LXG-T2 were also identified in non-Streptococcus species, including members of the genus Listeria (Figure 1C). The presence of this structural motif, commonly associated with membrane-interacting proteins, suggests a potential role in membrane disruption [17, 31], warranting further functional analysis.

LXG-T2 exhibits bactericidal activity against E. coli and confers competitive advantage

To assess the functionality of the LXG-T2/Imm2 toxin-immunity pair, we performed heterologous expression assays in E. coli. The genes encoding LXG-T2 and its putative downstream immunity partner Imm2 were cloned into the arabinose-inducible vector pBADHisA. Growth curve analysis of E. coli TOP10 induced with 0.2% L-arabinose showed significant growth inhibition in cells expressing LXG-T2 alone, whereas coexpression with Imm2 restored growth to levels comparable to empty vector controls (Figure 2A). Consistent with these results, CFU assays revealed a substantial decrease in viability in LXG-T2-expressing strains, which was fully rescued by coexpression of Imm2 (Figure 2B). In addition, we have independently verified the expression of pBADHisA-LXG-T2 by western blotting, which confirmed a protein band of approximately 55 kDa, consistent with the predicted molecular weight of LXG-T2 (Additional file 3). These findings confirm the intrinsic bactericidal activity of LXG-T2 and identify Imm2 as its cognate immunity protein.Figure 2LXG-T2 is toxic to E. coli and functions as a T7SS effector mediating intraspecific competition. A Growth curves of E. coli TOP10 strains carrying pBADHisA (empty), pBADHisA-LXG-T2, or coexpressing LXG-T2 and its cognate immunity protein Imm2, after induction with 0.2% L-arabinose. B Viability of bacterial cells harboring the empty vector, LXG-T2 alone, or LXG-T2 coexpressed Imm2, following induction. C Rluc-Luciferase assay demonstrating LXG secretion. A fusion gene encoding Rluc fused to LXG was constructed in plasmid pKSM410. This plasmid was then expressed in both WT and ΔessC1 strains. Relative luminescence in culture supernatants was measured. Rluc-TrxA served as a negative (cytoplasmic) control, and Rluc-EsxA as a positive (secreted) control. D Growth curves of WT and the mutant strains ΔT7SS, ΔLXG-T2, ΔLXG-T2&Imm2, and the complemented strain C-LXG-T2 in THB medium. E Bacterial competition assays using WUSS351, ΔLXG-T2, and the complemented strain C-LXG-T2 as attacker strains, with the ΔLXG-T2&Imm2 mutant as the target strain. Points and bars show mean ± SD (n = 3 biological replicates). Statistical significance was assessed by one-way ANOVA followed by Dunnett’s multiple comparisons test. (*** indicates p < 0.001 and **** indicates p < 0.0001).

Given previous reports implicating the N-terminal LXG domain in effector secretion through the T7SS apparatus [8], we next tested whether functional LXG is associated with the T7SS. We constructed a secretion reporter by fusing the N-terminal LXG domain (residues 1–200) to Rluc and expressed it under the control of a tetracycline-inducible promoter in the pKSM shuttle vector. Culture supernatants were analyzed for Rluc activity using coelenterazine substrate. It offers quantitative analysis and eliminates the need to concentrate supernatant proteins through precipitation. The constructed pKSM shuttle vector was introduced into WT strain WUSS351 and the ΔessC1 mutant, which lacks the ATPase component essential for T7SS function. We then compared the luminescence intensity in the supernatant of the WT strain to that of the ΔessC1 mutant, using the expression of EsxA fused with Rluc as a positive control (secreted) and the expression of TrxA fused with Rluc as a negative control (cytoplasmic) [32]. As expected, robust luminescence from LXG-Rluc was observed in the WT background, comparable to the EsxA-Rluc positive control (Figure 2C). The data confirm that LXG-T2 secretion is T7SS-dependent, as secretion was significantly reduced in the ΔessC1 strain, which lacks the essC1 gene that encodes the core component.

To determine the ecological role of LXG-T2 in bacterial competition, we performed bacterial competition assays. The WT strain WUSS351, ΔLXG-T2 deletion mutant, and complemented strain (C-LXG-T2) served as attacker strains, while strain lacking LXG-T2 and Imm2 (ΔLXG-T2ΔImm2) was used as target strain. The mutant strains were generated by replacing the target DNA sequence with distinct antibiotic resistance cassettes, enabling their selection and quantification on appropriate antibiotic-containing media. Growth curve analysis confirmed that deletion or complementation of LXG-T2 did not affect growth under standard conditions (Figure 2D). Notably, the WT strain significantly reduced the viability of the target strains compared with the ΔLXG-T2 mutant, indicating a loss of competitive advantage in the ΔLXG-T2 mutant. The complemented strain restored the competitive phenotype to WT levels (Figure 2E). Collectively, these findings establish LXG-T2 as a key effector that promotes S. suis fitness during intraspecies competition.

LXG-T2 employs a glycine zipper motif to perforate bacterial membranes

Bioinformatics analysis predicted that LXG-T2 contains transmembrane helices and a C-terminal domain belonging to the Gly-zipper_Omp superfamily. Proteins of this superfamily harbor a conserved GxxxGxxxG motif within their transmembrane regions, which mediates hydrophobic helix–helix interactions and is implicated in membrane insertion and pore formation (Figure 3A) [17, 33]. Furthermore, structural modeling using Phyre2.2 revealed a notable similarity between LXG-T2 and the outer membrane lipoprotein SlyB [34], which contains a glycine zipper domain that facilitates the formation of a transmembrane α-helical hairpin structure. This structure presents distinct binding sites for both phospholipids and lipopolysaccharides [35]. Notably, both LXG-T2 and SlyB exhibit the glycine zipper motif (GxxxGxxxG), with the glycine residues highlighted in orange (Figure 3B). In addition, limited structural similarity was also predicted with the pore-forming domain of colicin (Additional file 4). These structural signatures suggest that LXG-T2 may integrate into and disrupt target cell membranes (Figure 3C). To investigate this hypothesis, we conducted SEM of E. coli TOP10 strains harboring either the empty pBADHisA vector, pBADHisA-LXG-T2, or pBADHisA-LXG-T2 & Imm2. Following induction with 0.2% L-arabinose for 2 h, cells were processed to preserve native membrane architecture for SEM observation. Bacteria expressing LXG-T2 displayed membrane perforations (red arrows) and collapsed morphology, whereas coexpression with Imm2 preserved normal cell morphology comparable to the vector control (Figure 3D). These findings indicate that LXG-T2 disrupts the membrane via a glycine zipper-mediated pore-forming mechanism and that Imm2 specifically neutralizes this toxic activity.Figure 3Structural prediction and membrane-disrupting activity of LXG-T2. A Prediction of multiple transmembrane domains in the C-terminal region of LXG-T2. B AlphaFold2 structural models of LXG-T2 and SlyB, shown in ribbon representation and the Gly-zipper motif is colored in orange. C Schematic model illustrating T7SS-mediated secretion of LXG-T2 and its proposed mechanism of membrane disruption via the glycine zipper. D SEM images of E. coli TOP10 cells carrying pBADHisA (empty), LXG-T2, or LXG-T2 coexpressed with Imm2, following induction with 0.2% L-arabinose. Red arrows indicate sites of membrane perforation and shrinkage.

LXG-T2 increases membrane permeability and causes membrane depolarization

To further characterize the membrane-disruptive function of LXG-T2, we assessed its impact on membrane permeability and polarization using fluorescence-based assays. E. coli TOP10 strains were induced with 0.2% L-arabinose to express LXG-T2 (pBADHisA-LXG-T2), LXG-T2&Imm2 (pBADHisA-LXG-T2&Imm2), and an empty vector (pBADHisA). The E. coli cells were stained with SYTO9 and PI and visualized using confocal microscopy. SYTO9 stains all cells green by penetrating intact membranes, whereas PI penetrates only compromised membranes and emits red fluorescence upon DNA binding, displacing SYTO9. Cells expressing LXG-T2 exhibited strong red fluorescence, indicative of membrane disruption, similar to the positive control. In contrast, cells coexpressing LXG-T2&Imm2 or harboring the empty vector exhibited predominantly green fluorescence, reflecting intact membranes (Figure 4A). These results confirm that LXG-T2 compromises membrane integrity and that Imm2 effectively counteracts this activity.Figure 4LXG-T2 increases membrane permeability and induces membrane depolarization. A Confocal laser scanning microscopy images of E. coli cells stained with SYTO 9 (green) and PI (red) after the indicated treatments. As a positive control, cells harboring the empty vector were treated with 70% isopropanol. (Scale bar: 5 μm). B Quantification of membrane depolarization using DiSC3(5) fluorescence during LXG-T2 expression. The fluorescence signal was recorded using a multimode microplate reader. Statistical significance was assessed by one-way ANOVA followed by Dunnett’s multiple comparisons test. Data are presented as mean ± SD. (**** indicates p < 0.0001).

To determine whether LXG-T2 causes membrane depolarization, we employed the potential-sensitive dye DiSC3(5) [36]. In polarized membranes, DiSC3(5) accumulates and self-quenches, while depolarization results in dye release and fluorescence increase. After a 2-h induction, the bacteria were stained with DiSC3(5), and the fluorescence intensity was quantified using a microplate reader. Bacteria expressing the empty vector were treated with Polymyxin B, an antimicrobial peptide that induces large pore formation and is used as a positive control for depolarization [37]. As expected, polymyxin B treatment resulted in high fluorescence signals. In contrast, cells harboring the empty vector displayed low DiSC3(5) fluorescence that was not affected by supplementation with L-arabinose. LXG-T2 expression triggered a significant increase in DiSC3(5) fluorescence, indicative of rapid membrane depolarization, whereas coexpression with Imm2 attenuated this effect to the level of the negative control (Figure 4B). Together, these data indicate that LXG-T2 compromises membrane integrity and disrupts membrane potential, likely through glycine zipper-mediated pore formation. The immunity protein Imm2 neutralizes the toxin effect, underscoring its essential protective role.

LXG-T2 shows cytotoxic effect on host cells and contributes to the virulence of S. suis

To evaluate the cytotoxic potential of LXG-T2 toward host cells, human embryonic kidney cells HEK293T were transfected with a pcDNA3.1 vector encoding LXG-T2. The concentration of LDH in the culture supernatant of cells expressing the LXG-T2 effector was significantly higher than that of the control group (Figure 5A). This finding suggests that the LXG-T2 effector may compromise the integrity of the cell membrane. Therefore, we hypothesize that LXG-T2 may serve as a crucial virulence factor of S. suis, directly contributing to its pathogenic processes in the host. To assess the contribution of LXG-T2 to S. suis pathogenicity, we employed a murine systemic infection model using 6-week-old female BALB/c mice. We evaluated the effect of LXG-T2 on the organ load of S. suis during infection. Mice were intravenously inoculated with 3 × 10^8^ CFU of either WT strain, ΔLXG-T2 mutant, and complemented strain C-LXG-T2. At 12 h postinfection, animals were euthanized, and blood along with major organs were harvested for bacterial burden quantification via serial dilution plating on THB agar. The ΔLXG-T2 mutant strain exhibited significantly reduced bacterial burdens in the blood, brain, liver, kidneys, and spleen compared with the WT strain (Figures 5B–F). Complementation of the mutant restored bacterial burdens to near WT levels, confirming that LXG-T2 is required for efficient systemic dissemination and tissue colonization (Figures 5B–F). These results identify LXG-T2 as a key virulence factor that enables S. suis to overcome host defenses and establish multi-organ infections.Figure 5LXG-T2 contributes to the virulence of S. suis. A LDH release from HEK293T cells into the culture medium following LXG-T2 expression. B–F Bacterial burdens in the blood, brain, liver, kidney, and spleen at 12 h postinfection. Each dot represents an individual mouse. Statistical significance was assessed using one-way ANOVA followed by Dunnett’s multiple comparisons test. Data are presented as mean ± SD. (ns indicates no significance, * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001, and **** indicates p < 0.0001).

Discussion

Glycine zipper-mediated membrane permeabilization represents a conserved mechanism exploited by a variety of bacterial toxins to translocate across target membranes. For instance, the LXG substrate TelE from Streptococcus gallolyticus subsp. gallolyticus (SGG) requires a C-terminal glycine zipper motif for its membrane-disruptive function [31]. Similarly, the T6SS substrate Tde1 facilitates the passage of larger molecules such as PI (668 Da), suggesting that its C-terminal glycine zipper motif forms a pore large enough to compromise membrane integrity [38]. In addition, the T6SS substrate Tse4, which contains a high content of glycine zipper residues from Pseudomonas aeruginosa, disrupts membrane potential by forming a cation-selective pore. However, this pore does not permit the passage of small molecules such as o-nitrophenyl-β-D-galactopyranoside (ONPG, 300 Da), indicating that it does not induce general membrane permeabilization [19]. Glycine zipper motifs are also prevalent in certain transmembrane and amyloid-forming proteins, such as the Aβ peptide implicated in Alzheimer’s disease and the human prion protein PrP, where they mediate oligomerization and aggregation within lipid bilayers [17, 39, 40]. Furthermore, bacteriocins often harbor glycine zippers that facilitate cytotoxicity via membrane insertion [17]. Collectively, these observations support the notion that the glycine zipper region facilitates both aggregation at the membrane surface and pore formation within lipid bilayers. Consistent with these findings, our results indicate that LXG-T2 functions as a membrane-depolarizing toxin. This effector features a C-terminal glycine zipper motif (GxxxGxxxG) responsible for its bactericidal activity. When induced with 0.2% L-arabinose, the bacteria lost their characteristic rod shape, suggesting a complete loss of cell envelope integrity. Moreover, the bacteria became permeable to PI, a strong indicator that inner membrane integrity had been compromised.

Secretion systems are key tools used by bacterial pathogens to eliminate competitors and establish colonization within host tissues. In S. aureus, the T7SS has been shown to contribute to virulence and persistence in diverse infection models, including nasal colonization, skin abscesses, pneumonia, and bacteremia [4, 41, 42]. Several membrane-permeabilizing toxins associated with the T7SS have been identified. These toxins specifically target the cell membrane structures of competing bacteria, compromising membrane integrity and inducing cell death. As a result, the toxins confer a competitive advantage to the producing bacteria within the microbial population. Palmer et al. demonstrated that the C-terminal domain of TspA dissipates membrane potential and promotes T7SS-dependent bacterial proliferation in a zebrafish infection model [11]. In addition, the N-terminal domain of the S. aureus toxin TslA acts as a phospholipase A, disrupting membranes via detergent-like activity, although it was found to be dispensable for virulence in a mouse abscess model [10]. In line with these findings, we demonstrated that LXG-T2 exhibits cytotoxicity toward host cells and may contribute to S. suis pathogenicity.

To date, research on the impact of effectors secreted by the T7SS on host cells remains limited. Notably, the GBS strain CJB111 produces EsxA1, a pore-forming protein that induces cell death in brain endothelial cells [43]. The damage inflicted on host cells during bacterial infections can exacerbate disease progression and lead to multi-organ failure [44]. In our study, the expression of LXG-T2 in HEK293T cells resulted in an increased release of LDH, thereby validating its cytotoxic effects. This membrane-disrupting activity is likely mediated through glycine zipper-facilitated insertion and pore formation, leading to destabilization of the plasma membrane and ultimately cell lysis. This mechanism may also explain its ability to compromise the integrity of the cell membrane. Furthermore, phylogenetic analyses suggest that LXG-T2 homologs are widely distributed among pathogenic bacteria, underscoring the potential evolutionary conservation of this mechanism in bacterial virulence.

LXG toxin gene clusters typically follow a conserved genomic organization in which the immunity gene is located immediately downstream of the effector gene [45]. For pore-forming toxins that utilize glycine zippers, such as CdzC/D in Caulobacter crescentus and Tse4 in Pseudomonas aeruginosa, the associated immunity proteins are usually multipass transmembrane proteins anchored to the inner membrane [18, 19]. Similarly, the inner membrane protein TsaI neutralizes the membrane-depolarizing activity of the S. aureus T7SS toxin TspA [11]. Consistent with these findings, the immunity protein Imm2, encoded directly downstream of LXG-T2, is predicted to contain a transmembrane domain and was shown to protect E. coli from LXG-T2-mediated toxicity. This suggests a specific and direct neutralization mechanism, likely involving membrane-associated protein interactions. Notably, specific LXG toxin-immunity pairs are likely essential for maintaining the integrity of antibacterial mechanisms during microbial competition. These systems likely contribute to the ecological fitness and niche adaptation of S. suis within host environments, such as the tonsillar crypts, where microbial competition is intense.

In summary, we characterized LXG-T2 as a membrane-targeting effector secreted by the T7SS of S. suis. Its glycine zipper-dependent pore-forming activity underlies both its antibacterial and cytotoxic functions. These results expand the functional repertoire of LXG effectors and provide important insights into how T7SS contributes to bacterial competition, tissue colonization, and host damage.

Supplementary Information

Additional file 1. Bacterial strains and plasmids used in this study. Additional file 2. Primers used in this study. Additional file **3. Detection of LXG-T2 expression by western blot using an anti-His antibody.**Additional file 4. Identification of LXG-T2 structural homologs using PHYRE2. Analysis with PHYRE2 v2.2 revealed significant structural similarity to the outer-membrane lipoprotein SlyB and the pore-forming domain of colicin.