The pulmonary commensal Lactobacillus regulated lung γδ T cells to enhance resistance against bacterial infections

Abstract

Background

Pulmonary commensals play a crucial role in regulating host immune homeostasis and combating infections. Nevertheless, the deep mechanisms remain unclear.

Results

Long-term antibiotics pre-exposure enhanced the susceptibility to bacterial pneumonia, while intranasal reconstitution of the pulmonary microbiota mitigated these adverse effects, restoring host resilience to infections. We isolated two pulmonary commensals, Lactobacillus plantarum and Lactobacillus murinus, demonstrating that they induced IL-17A-mediated antibacterial immunity and promoted resistance to lung infections. Moreover, antibiotics-treatment reduced the frequency of pulmonary IL-17A secreting Vγ4⁺ γδ T cells and made the mice more susceptible to pneumonia, which was reversed by transferring pulmonary Lactobacillus commensals. In addition, our data indicated that L. plantarum and L. murinus-derived metabolites, particularly extracellular polysaccharides, can activate lung Vγ4⁺ γδ T cells to secrete IL-17A in defense against bacterial lung infections.

Conclusions

In this study, we report for the first time that pulmonary commensal Lactobacillus, specifically L. plantarum and L. murinus, activate Vγ4⁺ γδ T cells to secrete IL-17A, thereby mitigating susceptibility to Staphylococcus aureus and Pseudomonas aeruginosa infections. Additionally, we identified the metabolite of L. plantarum and L. murinus, extracellular polysaccharides, as the key immunomodulatory molecules. This research highlights the importance of pulmonary commensals in the regulation of anti-infection immunity and provides a theoretical foundation for clinical studies on the role of lung microbiota in combating infections.

Download video file ^{(55.1MB, mp4)}

Video Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s40168-025-02205-8.

Background

Antibiotics have revolutionized medicine by significantly extending human life expectancy and effectively combating pathogenic infections. However, their broad-spectrum antibacterial properties mean that, in addition to eradicating harmful bacteria, antibiotics also disrupt beneficial host microbiota [1]. In recent decades, the widespread use of antibiotics has contributed to the rise of drug-resistant pathogenic bacteria, such as methicillin-resistant Staphylococcus aureus (S. aureus), Streptococcus pneumoniae (S. pneumoniae), and Pseudomonas aeruginosa (P. aeruginosa) [2, 3]. Clinical evidence suggests that antibiotics often fail to resolve infections, and paradoxically, may increase the host’s susceptibility to secondary infections by disrupting microbiota-dependent immune defenses, further heightening vulnerability to subsequent pathogenic invasions [4].

The human body harbors diverse microbial communities that colonize mucosal surfaces such as the intestines, skin, and lungs, playing a vital role in maintaining immune homeostasis and protecting against pathogenic infections [5]. In the past three decades, significant progress has been made in research related to the gut microbiome, while research on the lung microbiome has been relatively scarce. The lungs, which are the largest mucosal tissue in the human body, were once thought to be sterile. However, with the advancement of high-throughput molecular sequencing technologies, the existence, composition, and relative abundance of the lung microbiota have gradually been revealed [6, 7]. Studies utilizing germ-free (GF) mice have demonstrated the critical role of microbiota in bolstering resistance to bacterial lung infections [8]. For instance, GF mice exhibit heightened susceptibility to pneumonia caused by P. aeruginosa, S. pneumoniae, and Klebsiella pneumoniae (K. pneumoniae) compared to specific-pathogen-free (SPF) mice [9, 10]. Similarly, antibiotic-treated mice display trends akin to GF mice, with increased vulnerability to S. pneumoniae and K. pneumoniae infections. Notably, microbiota-disrupted mice show impaired cytokine production in the lungs following S. pneumoniae pneumonia, a susceptibility linked to decreased levels of GM-CSF and IL-17A [10]. Replenishment of microbiota from healthy mice via nasal or oral routes partially restored resistance to infection, highlighting the contributory roles of both local (lung) and distal (gut) microbiota in enhancing pulmonary immune defenses. While both the lung and gut microbiomes influence lung homeostasis and infection resistance, the lung microbiome exhibits a more direct connection to pulmonary function. Although its role in regulating immune equilibrium and infection resistance is increasingly recognized, the precise mechanisms through which the lung microbiota interact with the mucosa to enhance host defenses against S. aureus or P. aeruginosa infections remain unclear.

γδ T cells, a highly heterogeneous subset of T lymphocytes, are predominantly located at mucosal barrier sites such as the skin, gut, and lungs. In these non-lymphoid tissues, γδ T cells participate in local physiological activities through the secretion of basic cytokines and initiate inflammatory responses when pathogens invade [11]. Research has demonstrated that following S. pneumoniae infection in the lungs, subsets of γδ T cells, including Vγ1⁺, Vγ4⁺, and Vγ6⁺ γδ T cells respond rapidly, with Vγ4⁺ γδ T cells playing a major role in fighting infection. Studies in TCR Vγ4-deficient mice revealed significantly higher bacterial burdens and reduced survival rates compared to wild-type mice after S. pneumoniae infection, which correlated with decreased neutrophil recruitment in the lungs [12]. Interleukin-17 (IL-17) is a pivotal molecule in inflammation, host defense against mucosal infections, and the maintenance of mucosal barrier integrity [13]. During inflammation, IL-17 exerts diverse effects by inducing epithelial cells, endothelial cells, and fibroblasts to produce pro-inflammatory cytokines, chemokines, antimicrobial peptides, and matrix metalloproteinases [14]. Notably, IL-17 stimulates the expression of IL-8 and granulocyte colony-stimulating factor (G-CSF), which promote neutrophil production and facilitate their migration from the bone marrow to infection sites, thereby amplifying the immune response [15]. γδ17 T cells, a subset of γδ T cells, are a major source of early IL-17 production in mucosal tissues, including the lungs, gut, and reproductive tract [16]. These cells have been shown to provide protection against bacterial and fungal infections as well as viral hepatitis [17–19]. Additionally, γδ17 T cells play a protective role in infections caused by herpes simplex virus, simian immunodeficiency virus, and human immunodeficiency virus (HIV) [20]. The main subsets of γδ T cells involved in lung anti-infection are Vγ4⁺ and Vγ6⁺ γδ T cells. Vγ4⁺ γδ T cells recruit neutrophils by secreting chemokine ligand 2 (CXCL2; MIP2) and tumor necrosis factor (TNF), while dendritic cells capturing pathogens produce IL-23 to promote the expansion of Vγ4⁺ and Vγ6⁺ γδ17 T cells, leading to granuloma formation. The resident Vγ4⁺ γδ17 T cells have memory, enabling them to rapidly produce IL-17 and clear pathogens upon re-infection [21].

Evidence increasingly suggests a bidirectional regulation between the microbiota and γδ T cells, particularly in mucosal tissues where these cells interact with commensal and pathogenic bacteria [22]. Microbiota influence the differentiation and activation of γδ T cells, with subsets like Vγ4⁺ γδ17 T cells relying on specific microbial communities. Certain commensal bacteria promote γδ T cell expansion via the VAV1 signaling pathway and activation of IL-1R1⁺ γδ17 T cells through MyD88-dependent IL-1 and IL-23 secretion. These cells serve as key IL-17 producers, recruiting neutrophils to infection sites and aiding pathogen clearance [23, 24]. However, the specific commensal bacteria involved in this process remain unclear.

Research on microbiota has transitioned into the “post-sequencing” era. While previous studies have predominantly focused on the gut microbiome, the lung microbiome remains comparatively underexplored. In this study, we employ systems biology approaches to comprehensively predict and identify microbiota and their metabolites that interact with lung γδ T cells. Using S. aureus and P. aeruginosa as representative pathogens to establish lung infection models, we aim to investigate the molecular mechanisms underlying immune regulation. Elucidating how commensal bacteria and lung γδ T cells interact is vital for maintaining mucosal homeostasis and enhancing defense against pulmonary pathogens, offering a theoretical foundation for the development of probiotic-based therapies targeting bacterial lung infections.

Methods

Experimental animals and ethics statement

Specific-pathogen-free (SPF) C57BL/6 mice, age 6–8 weeks, were purchased from Beijing Vital River Laboratory Animal Technology Co. (Beijing, China). γδ T cells knockout mice (TCR δKO, JAX Stock number: 003288) and IL-17A knockout mice (IL-17AKO, NCBI Stock number: 16171) on C57BL/6N background were purchased from cyagen. All animals were allowed to acclimate to their new environment for 1 week and housed under controlled conditions with a 12-h light/dark cycle and provided unlimited access to food and water. Animals were euthanized by cervical dislocation following administration of ketamine/xylazine cocktail at defined time-points post-infection or when humane endpoints (e.g. hypothermia, weight loss) had been reached, whichever occurred earlier.

Bacterial strains

Lactobacillus plantarum (L. plantarum), Lactobacillus murinus (L. murinus), Lactobacillus vaginalis (L. vaginalis), Bacillus zhangzhouensis (B. zhangzhouensis), and Bacillus pumilus (B. pumilus) were isolated from the lungs of SPF C57BL/6 mice. They are identified on the basis of physiological characteristics and 16S rDNA sequence analysis.

P. aeruginosa strains PAO1 was purchased from BeNa Culture Collection (Beijing, China).

S. aureus strains USA300 was purchased from BeNa Culture Collection (Beijing). S. aureus USA300 was isolated from an adolescent patient with severe sepsis syndrome at Texas Children's Hospital, Houston, TX, USA.

Mouse antibiotic treatments and bacterial reconstitution

Ampicillin (1 mg/mL), chloromycetin (1 mg/mL), neomycin (1 mg/mL), and vancomycin (0.5 mg/mL) were diluted in autoclaved water and delivered to mice ad libitum as their drinking water for up to 21 days [25]. Water was refreshed twice weekly. Antibiotic treatment was discontinued 3 days prior to infection. Collect intestinal contents and bronchoalveolar lavage fluid from No ABX mice respectively. Intestinal contents (200 µg in 50 µL PBS) or vehicle control (50 µL PBS) were administered to antibiotic-exposed mice via a single oral gavage using fine polyethylene tubing for three consecutive days [26]. Alternatively, BAL fluid (50 µL) or vehicle (50 µL PBS) was delivered intranasally for three consecutive days. Commensal reconstitution was halted 24 h before infection. Recombinant IL-17A (rIL-17A; R&D Systems) was administered intranasally to ampicillin- and vancomycin-treated mice 24 h prior to infection.

Mice were injected with a ketamine/xylazine cocktail and anesthetized. Then, 40 µL bacterial slurry (5 × 10⁶CFUs of S. aureus USA300 or 1 × 10⁶CFUs of PAO1) were inoculated intranasally via left nostril, and observed continuously for 48–72 h. To account for sucrose present in the antibiotic cocktail, control animals were given sucrose-supplemented drinking water and subsequently challenged with S. aureus or P. aeruginosa under the same conditions. To evaluate bacterial burdens, the lower concentration of bacterial slurry (5 × 10⁵CFUs of S. aureus USA300 or 1 × 10⁵CFUs of PAO1) that caused morbidity but not death was inoculated intranasally via left nostril. Lungs were homogenized in sterile PBS, and serial dilutions of lung homogenates or BAL fluid were plated on TSA and incubated overnight at 37 °C to determine CFU counts. Organs were harvested at specified time points post-infection, fixed in 10% formalin for 24 h, and embedded in paraffin wax. Tissue sections were stained with periodic acid-Schiff (PAS) and/or hematoxylin and eosin (H&E) for histological analysis.

Isolation, identification, and quantification of lung bacterial microbiota

To isolate lung bacteria, lung tissues from SPF mice were harvested under sterile conditions and placed in a 3.5 cm dish. One milliliter of sterile PBS was added, and the tissue was minced thoroughly before being filtered through a 100 µm cell strainer. The resulting homogenate was plated onto MRS agar and incubated at 37 °C under anaerobic conditions, or onto TSA and incubated at 37 °C under aerobic conditions. After 1–2 days of culture, individual bacterial colonies were isolated and further characterized. Bacterial species were identified through 16S rDNA sequencing, targeting the V3–V4 regions, as previously described [27].

16S ribosomal DNA gene sequencing

16S rDNA sequencing was performed as previously described [28]. Bacterial DNA from lung and fecal samples was extracted following the methods outlined above. The V3–V4 regions of the 16S rRNA gene were amplified by PCR using the primers 515 F (5ʹ-GTGCCAGCMGCCGCGGTAA-3ʹ) and 806R (5ʹ-GGACTACHVGGGTWTCTAAT-3ʹ). Each PCR reaction consisted of 15 µL Phusion High-Fidelity PCR Master Mix (New England Biolabs), 0.2 µM of each primer, and 10 ng of genomic DNA template. Amplification was carried out with an initial denaturation at 98 °C for 1 min, followed by 30 cycles of 98 °C for 10 s, 50 °C for 30 s, and 72 °C for 30 s, and a final extension at 72 °C for 5 min. The resulting PCR products were purified using magnetic beads, and equal volumes of the purified products were pooled based on their concentrations. The pooled PCR products were analyzed, and target bands were excised and recovered. The constructed library was quantified using Qubit and Q-PCR. After quality validation, paired-end (PE250) sequencing was performed on a NovaSeq 6000 platform.

Transcriptomics analysis

Total lung RNA (3 mice per group) was extracted using TRIzol reagent (Invitrogen, CA, USA) following the manufacturer’s protocol. RNA purity and concentration were assessed with a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA), and RNA integrity was evaluated using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). RNA-seq libraries were prepared with the VAHTS Universal V6 RNA-seq Library Prep Kit, as per the manufacturer’s instructions, and sequenced on an Illumina NovaSeq 6000 platform to generate 150 bp paired-end reads. Raw sequencing reads in FASTQ format were processed with fastp [29] to remove low-quality reads, yielding clean reads for further analysis. Clean reads were aligned to the reference genome using HISAT2 [30]. Gene expression levels were quantified as fragments per kilobase of transcript per million mapped reads (FPKM) [31], and read counts for each gene were obtained with HTSeq-count [32]. Principal component analysis (PCA) was performed using R (v3.2.0) to assess sample biological replicates. Differential expression analysis was conducted with DESeq2, applying thresholds of Q value < 0.05 and fold change > 2 or < 0.5 to identify differentially expressed genes (DEGs). Hierarchical clustering of DEGs was performed in R (v3.2.0) to visualize expression patterns across groups and samples. A radar plot of the top 30 DEGs was generated using the R package ggradar to highlight upregulated and downregulated genes. Enrichment analyses of DEGs were conducted using Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), Reactome, and WikiPathways databases based on hypergeometric distribution. Significant enriched terms were visualized with bar plots, chord diagrams, and bubble plots generated in R (v3.2.0). Additionally, gene set enrichment analysis (GSEA) was performed using the GSEA software. This analysis ranked genes by differential expression levels between sample groups to identify whether predefined gene sets were significantly enriched at the top or bottom of the ranking.

Cytokine determination

Lung and spleen tissues of the mice (3–5 mice per group) were homogenized and lysed in RIPA buffer (50 mM Tris, pH 7.4; 150 mM NaCl; 1% Triton X-100; 1% sodium deoxycholate; 0.1% SDS). Cytokine levels, including IL-17A, IL-17F (BioLegend), IFN-γ, CXCL2, CXCL1, IL-1β, and IL-23 (Thermo Fisher Scientific), were measured using ELISA kits according to the manufacturer’s protocols.

Flow cytometric analysis

We pooled and cut the fresh lungs from mice and incubated (37 °C, 30 min), then cut tissues with shaking (150 rpm) in digestion buffer (RPMI 1640 with 10% FBS, 15 mM HEPES, 1% penicillin/streptomycin (wt/vol) and 300 U mL⁻¹ collagenase VIII) and pressed through a 100-μm nylon strainer to obtain single-cell suspension. We then incubated (37 °C, 5 h, 5% CO₂) the cells (1 × 10⁷) in culture medium containing RPMI 1640 with 10% FCS, 1 × nonessential amino acids, 10 mM HEPES, 2 mM l-glutamine (all from Invitrogen), and 1% penicillin/streptomycin with 1:1000 Golgi Stop (554,724, BD Biosciences), 10 ng/mL phorbol 12-myristate 13-acteate (PMA), and 500 ng/mL calcium ionophore A23187 (both from Sigma-Aldrich). We washed and incubated (4 °C, 30 min) with anti-mouse CD45 antibody (45–0454-82), anti-mouse CD4 antibody (GK1.5), anti-mouse CD11b, anti-mouse Ly6G (Gr-1) antibody (1A8), anti-mouse CD117 antibody (2B8), anti-mouse NKp46 antibody (29A1.4), anti-mouse NK1.1 antibody (2G12), anti-mouse TCRγ/δ antibody (29-2L17), anti-mouse TCR vγ2/Cr4 antibody (2.11), and anti-mouse TCR vγ4 antibody (UC3-10A6) (all diluted 1:100, Biolegend). For intracellular staining, we washed and fixed (4 °C, 60 min) the surface-stained cells in 1 × Cytofix/Cytoperm buffer (BD Biosciences) and permeabilized them (4 °C, overnight) using 1 × permeabilization buffer (BD Biosciences) according to manufacturer instructions. We stained the cells intracellularly with anti-mouse IL-17A antibody (5164) (diluted 1:50, Biolegend) and then washed (2 ×) and resuspended them in flow cytometry buffer. We analyzed the data with flow cytometry (ACEA NovoCyte).

Depletion of Vγ2 and Vγ4 γδ T cells

Vγ2 (clone 2.11) and Vγ4 (clone UC3-10A6) γδ T cells were depleted using antibodies from Bio X Cell (Lebanon, NH, USA). Antibodies (dosage 200 μg per 100 μL PBS) and isotype control were i.v. injected in tail vein three days before S. aureus or P. aeruginosa pulmonary infection.

Extraction and purification of EPSs

The extraction and purification of polysaccharides from lactic acid bacteria were performed as previously described [33]. Briefly, 1 mL of an overnight bacterial culture was inoculated into 50 mL of DeMan, Rogosa, and Sharpe (MRS) broth. After 24 h of incubation at 37 °C, bacterial cells were separated from the extracellular polysaccharides (EPSs) preparation by centrifugation at 5000 rpm for 15 min at 4 °C. Ethanol (80%) was added to the supernatant, and the mixture was incubated at 4 °C for 12 h to precipitate the EPSs. The precipitated supernatants were collected by centrifugation at 5000 rpm for 15 min, dialyzed using a membrane with a molecular weight cutoff of 8000–12,000 Da, and subsequently lyophilized to obtain the crude polysaccharide. The crude EPS was dissolved in distilled water at a concentration of 10 mg/mL and loaded onto a DEAE Sepharose Fast Flow column (GE, USA; dimensions: 2.6 × 40 cm). Distilled water was used for initial elution at a flow rate of 2 mL/min, with fractions collected every 6 mL. The sample was then eluted with a 50 mM Tris–HCl buffer (pH 7.60) containing a linear NaCl gradient of 0–1 M. The final elution was performed with distilled water at a flow rate of 1 mL/min, with fractions collected every 6 mL. The eluate containing EPS was pooled, dialyzed against distilled water at 4 °C for 2 days, and lyophilized to obtain purified EPS.

Statistical analysis

GraphPad software program was used for data processing and analysis. P values < 0.05 were considered statistically significant. Comparisons between two normally distributed groups were performed by simple two-tailed unpaired Student’s t test, and other analyses for unequal variance were determined by Mann–Whitney U test. For multiple groups comparisons, we used one- or two-way analysis of variance (ANOVA) with Tukey’s post hoc analysis for equally distributed groups. The data for survival test were analyzed by Wilcoxon log-rank survival test. Values are presented as means ± SEM. P values are annotated as follows: *p < 0.05, **p < 0.01, ***p < 0.001.

Results

Long-term antibiotics pre-exposure enhanced the susceptibility to bacterial pneumonia

S. aureus and P. aeruginosa are major pathogens responsible for pneumonia in humans and animals, posing significant threats to public health and the livestock industry. Antibiotics have been the primary treatment for infections caused by these pathogens [34]. However, the broad-spectrum activity of antibiotics not only eliminates pathogenic bacteria but also disrupts beneficial host microbiota, resulting in microbiota imbalance and increased susceptibility to subsequent infections. Clinical studies have highlighted the critical role of commensal microbiota in maintaining homeostasis and providing defense against infections [35, 36]. To investigate whether antibiotic exposure affects susceptibility to respiratory pathogens, mice were treated with a combination of ampicillin, vancomycin, neomycin, and chloromycetin—four commonly used antibiotics—for 21 days, followed by intranasal infection with S. aureus or P. aeruginosa (Fig. 1A). This antibiotic regimen significantly reduced the bacterial loads of both aerobic and anaerobic microbiota in the intestinal contents and bronchial lavage fluid (BALF) (Fig. 1B, Fig. S1). Mice exposed to antibiotics (ABX) exhibited increased susceptibility to S. aureus- and P. aeruginosa-induced lung infections compared to those not exposed to antibiotics (No ABX) (Fig. 1C). Furthermore, antibiotic-treated mice demonstrated higher lung and BALF bacterial burdens of both S. aureus and P. aeruginosa at 12 h post-inoculation compared to non-antibiotic-treated controls (Fig. 1D).

Fig. 1.

The commensals promote resistance to S. aureus and P. aeruginosa in an IL-17A-dependent manner. A Study design. Day 21, Ampicillin (1 mg/mL), chloromycetin (1 mg/mL), neomycin (1 mg/mL), and vancomycin (0.5 mg/mL) were diluted in autoclaved water and delivered to mice ad libitum as their drinking water for up to 21 days. Day 23, 2 days after stopping antibiotic-treatment. B Feces and lungs bacteria enumerated in mice exposed to mixed antibiotics (ABX) or ABX-free (No ABX), quantified using real-time polymerase chain reaction (RT-PCR). C 40 µL bacterial slurry (5 × 10⁶CFUs of S. aureus USA300 or 1 × 10⁶CFUs of P. aeruginosa PAO1) were inoculated intranasally via left nostril, and observed continuously for 48 h. Representative survival rates from two independent experiments are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (**p < 0.01; ***p < 0.001). D The lower concentration of bacterial slurry (40 µL, 5 × 10⁵CFUs of S. aureus USA300 or 1 × 10.⁵CFUs of P. aeruginosa PAO1) were inoculated intranasally via left nostril, S. aureus and P. aeruginosa burdens in lung and BALF at 12 h post-intranasal inoculation were counted (n = 5). E The level of IL-17A in the lungs of ABX and No ABX mice before or post S. aureus or P. aeruginosa infection (n = 4–6). F Clustering of genes expressed differentially in the lungs of ABX and No ABX mice. Expression results are presented as log2 RPKM values ("reads per kilobase of transcript per million reads"; mean and maximum value for each gene) (n = 3). G Recombinant IL-17A (10 µg/mouse) inoculated intranasally to ABX mice for 3 days in a row. H Representative survival rates from two independent experiments of No ABX, ABX, and ABX treated with rIL-17A mice are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (**p < 0.01; ***p < 0.001). I Bacterial burdens of No ABX, ABX and ABX treated with rIL-17A mice at 12 h post infected with S. aureus or P. aeruginosa (the lower concentration of bacterial slurry). J Representative survival rates from two independent experiments of ABX (WT), No ABX (WT) and No ABX (IL-17AKO) mice are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (**p < 0.01; ***p < 0.001). K Bacterial burdens in lungs of ABX (WT), No ABX (WT), and No ABX (IL-17AKO) mice infected with S. aureus or P. aeruginosa (The lower concentration of bacterial slurry). Data are presented as means ± SEM. Significant differences were calculated with One-way ANOVA followed by Tukey’s multiple comparisons test. ns, not significant, **p < 0.01, ***p < 0.001

The microbiota enhances pulmonary immunity in an IL-17A-dependent manner

IL-17 plays a pivotal role in inflammation, host defense against mucosal infections, and maintaining mucosal integrity and dynamic balance [13, 14]. During the inflammatory process, IL-17 exerts multifaceted effects by inducing epithelial cells, endothelial cells, and fibroblasts to produce pro-inflammatory cytokines, chemokines, antimicrobial peptides, and matrix metalloproteinases. Recent studies have demonstrated that the microbiota promotes GM-CSF production during respiratory infections through IL-17A signaling [10]. Therefore, we hypothesized that the increased susceptibility of mice to respiratory pathogens following long-term antibiotic treatment may be linked to reduced IL-17 levels. To investigate, we analyzed IL-17A secretion in the lungs of ABX and No ABX mice before and after infection with S. aureus or P. aeruginosa. As shown in Fig. 1E, prior to infection, IL-17A levels were significantly higher in the lungs of No ABX mice compared to ABX mice. After infection, IL-17A levels in No ABX mice were further upregulated, whereas levels in ABX mice remained unchanged and significantly lower than those in No ABX mice. Gene expression analysis (Fig. 1F) confirmed that il17a expression was markedly higher in No ABX mice compared to ABX mice. To further test the role of IL-17A, ABX mice were treated with recombinant IL-17A (rIL-17A) prior to infection with S. aureus or P. aeruginosa (Fig. 1G). As shown in Fig. 1H, rIL-17A treatment significantly improved survival rates in ABX mice, making them comparable to those of No ABX mice. Additionally, bacterial burdens post infections in the lungs of rIL-17A-treated ABX mice were substantially lower than in untreated ABX mice (Fig. 1I). Furthermore, we used IL-17AKO mice to confirm this result. As shown in Fig. 1J, IL-17AKO mice exhibited lower survival rates following infection compared to No ABX wild-type mice. Moreover, bacterial burdens in the lungs of IL-17AKO mice were similar to those of the ABX wild-type mice but significantly higher than in No ABX wild-type mice (Fig. 1K). These studies demonstrate that long-term broad-spectrum antibiotic exposure in mice results in host microbiota dysbiosis and increased susceptibility to S. aureus and P. aeruginosa, with the microbiota resisting infection through an IL-17A-dependent mechanism.

The commensals of Lactobacillus plantarum and Lactobacillus murinus from lung enhance respiratory defenses against bacterial infection

Epidemiological studies have shown that patients receiving frequent antibiotic treatments not only exhibit altered intestinal commensal bacteria but also face an increased risk of developing pneumonia [37]. Reconstitution therapy targeting intestinal commensal bacteria has been proposed to mitigate the adverse effects of antibiotics on the host. However, whether lung commensal bacteria similarly educate the mucosal immune system, as intestinal commensal bacteria do, remains a topic of debate [38, 39]. Moreover, the role of lung commensal bacteria in enhancing resistance to pathogens has yet to be fully explored. To address this, we restored disrupted commensal microbiota in ABX mice by transferring intestinal contents orally or BALF contents intranasally from specific-pathogen-free (SPF) mice (Fig. 2A). Both fecal microbiota transplantation (FMT) and respiratory microbiota transfer can enhance the bacterial abundance in the feces and lungs of ABX mice (Fig S2). Reconstitution of intestinal or pulmonary commensal bacteria partially restored resistance to pneumonia in ABX mice. Notably, ABX mice receiving pulmonary commensal bacteria exhibited significantly higher survival rates compared to those receiving intestinal commensal bacteria following S. aureus or P. aeruginosa infection (Fig. 2B). Furthermore, ABX mice exposed to pulmonary commensals conferred greater resistance to both pathogens than ABX mice or ABX mice exposed to intestinal commensals (Fig. 2C).

Fig. 2.

Lactobacillus plantarum and Lactobacillus murinus from Lung enhance respiratory defenses against bacterial infection via IL-17A. A Study design. Day 26, intestinal contents or BAL fluid contents from No ABX mice transferred to ABX for 3 days in a row. B 40 µL bacterial slurry (5 × 10⁶CFUs of S. aureus USA300 or 1 × 10⁶CFUs of P. aeruginosa PAO1) were inoculated intranasally via left nostril, and observed continuously for 48 h. Representative survival rates from two independent experiments are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (*p < 0.05; **p < 0.01). C The lower bacterial slurry (40 µL, 5 × 10⁵CFUs of S. aureus USA300 or 1 × 10.⁵CFUs of P. aeruginosa PAO1) were inoculated intranasally via left nostril, S. aureus and P. aeruginosa burdens in lung at 12 h post-intranasal inoculation were counted (n = 7–8). D Representative survival rates of No ABX, antibiotics (Amp and Van)-exposed mice or antibiotics (Chl and Neo)-exposed mice infected with S. aureus or P. aeruginosa from two independent experiments are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (*p < 0.05; **p < 0.01). E S. aureus and P. aeruginosa burdens at 12 h post-intranasal inoculation in lung and BALF (n = 7–8). F Relative abundance of commensal bacteria obtained from 16S rDNA sequencing of the intestinal contents and lung of the mice exposed to antibiotics (Amp + Van) or no No ABX. Each bar represents the pooled intestinal contents or lungs from mice from 3 different litters (n = 3). G Representative survival rates of No ABX mice, Amp + Van treated mice (ABX), ABX mice exposed to L. plantarum, L. murinus, L. vaginalis, B. pumilus, or B. zhangzhouensis infected with S. aureus or P. aeruginosa from two independent experiments are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (***p < 0.001). H S. aureus and P. aeruginosa burdens at 12 h post-intranasal inoculation in lungs (n = 8). I The levels of IL-17A, CXCL1, CXCL2, IL-1β, and IL-23 in the lungs of No ABX mice, Amp + Van treated mice, ABX mice exposed to L. plantarum and ABX mice exposed to L. murinus before or post S. aureus or P. aeruginosa infection (n = 3–5). J The percentage of neutrophils in the lung of No ABX mice, Amp + Van treated mice, ABX mice exposed to L. plantarum and ABX mice exposed to L. murinus mice post S. aureus or P. aeruginosa infection (n = 4–6). K Representative survival rates of No ABX (WT) mice, Amp + Van treated mice, ABX mice (IL-17AKO) exposed to L. plantarum and ABX mice (IL-17AKO) exposed to L. murinus infected with S. aureus or P. aeruginosa from two independent experiments are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (***p < 0.001). L Bacterial burdens in lungs of No ABX (WT) mice, Amp + Van treated mice, ABX mice (IL-17AKO) exposed to L. plantarum and ABX mice (IL-17AKO) exposed to L. murinus infected with S. aureus or P. aeruginosa (The lower bacterial slurry). Data are presented as means ± SEM. Significant differences were calculated with one-way ANOVA followed by Tukey’s multiple comparisons test. ns not significant, *p < 0.05, **p < 0.01, ***p < 0.001

To identify potential commensal microbes involved in enhancing pulmonary defenses, we treated C57BL/6 mice with combinations of antibiotics targeting specific bacterial groups. A combination of chloramphenicol (Chl) and neomycin (Neo) is primarily used to eliminate Gram-negative bacteria, while the combination of ampicillin (Amp) and vancomycin (Van) is primarily used to eliminate Gram-positive bacteria [40]. Mice treated with Amp and Van exhibited significantly lower survival rates compared to those treated with Chl and Neo, not only following S. aureus infection but also after P. aeruginosa infection (Fig. 2D). Moreover, Amp and Van-treated mice showed increased lung bacterial burdens for both S. aureus and P. aeruginosa at 12 h post-inoculation, compared to non-antibiotic-treated controls or Chl and Neo-treated mice (Fig. 2E). These findings suggest that Gram-positive commensal bacteria may play a critical role in resistance to bacterial invasion. To further investigate, we sequenced 16S rDNA amplicons from lung tissues of Amp and Van-treated (ABX) and untreated (No ABX) mice to characterize bacterial taxa potentially involved in infection resistance. ABX mice exhibited substantial changes in bacterial composition in both the intestinal contents and lungs compared to No ABX mice (Fig. 2F). Notably, Lactobacillus species, especially Lactobacillus plantarum (L. plantarum), were significantly reduced in both lung and intestinal commensal microbiota in ABX mice. Based on these observations, we hypothesized that Lactobacillus species in the lungs are crucial for resistance to bacterial infections. To test this hypothesis, we isolated and identified potential lung microbiota, including Lactobacillus and Bacillus species, from No ABX SPF C57BL/6 mice. Five bacterial species, L. plantarum, L. murinus, Lactobacillus vaginalis (L. vaginalis), Bacillus zhangzhouensis (B. zhangzhouensis), and Bacillus pumilus (B. pumilus) were isolated and intranasally transferred into ABX mice. ABX mice exposed to L. plantarum or L. murinus demonstrated significantly increased resistance to both S. aureus and P. aeruginosa infections, as evidenced by higher survival rates compared to the mice treated with ampicillin and vancomycin (Fig. 2G). In contrast, ABX mice receiving L. vaginalis, B. pumilus, or B. zhangzhouensis showed no significant survival advantage post-infection (Fig. 2G). Additionally, bacterial burdens in the lungs of ABX mice exposed to L. plantarum or L. murinus were significantly lower after infection (Fig. 2H). These results collectively demonstrate that lung commensal bacteria, particularly L. plantarum and L. murinus, mitigates susceptibility to S. aureus and P. aeruginosa.

L. plantarum and L. murinus enhance respiratory defenses against bacterial infection via an IL-17A-dependent manner

It is widely recognized that direct competition between commensal microbiota and invading bacterial pathogens or resident pathobionts serves as a crucial host defense mechanism against mucosal bacterial infections [41, 42]. It was demonstrated for the intestine [43, 44] and for the skin [45], but data from the lung are currently not available. To investigate whether lung commensals such as L. plantarum and L. murinus directly inhibit the growth of bacterial lung pathogens, we conducted in vitro experiments. Co-culture studies were not feasible due to incompatible media requirements for optimal bacterial growth. Instead, we employed an experimental setup in which S. aureus, P. aeruginosa, or S. pneumoniae were cultured in the presence of either fresh MRS medium or MRS medium conditioned with L. plantarum, L. murinus, or B. pumilus, both diluted at a ratio of 1:10 with S. aureus or P. aeruginosa TSB media [46]. As shown in Fig. S3A, L. plantarum, L. murinus, as well as B. pumilus did not significantly inhibit the optical density (OD) or colony-forming units (CFUs) of P. aeruginosa compared to P. aeruginosa grown alone. After prolonged incubation, L. plantarum and L. murinus exhibited limited inhibitory effects on the growth of S. aureus (Fig. S3B). Interestingly, both L. plantarum and L. murinus significantly inhibited the growth of S. pneumoniae in vitro in TSB culture media (Fig. S3C), consistent with previously reported findings [46]. This result suggested that the effective lung commensal bacteria, L. plantarum and L. murinus, mediated protection against P. aeruginosa and S. aureus are likely mediated by mechanisms other than direct inhibition.

We carried out RNA sequencing analysis of lung RNA isolated from No ABX, ABX mice, ABX mice exposed to with L. plantarum or L. murinus intranasally at 12 h post infection. As shown in Fig. S4, antibiotic-treated mice exhibited significantly reduced intrapulmonary il17a production in response S. aureus or P. aeruginosa infection compared to non-antibiotic-treated mice or those with commensal exposure. Next, we measured key signaling molecules involved in the IL-17A pathway during the innate response to respiratory infection, including CXCL1, CXCL2, IL-1β, and IL-23. In ABX mice, intrapulmonary production of these molecules was markedly reduced both before and after infection with S. aureus or P. aeruginosa compared to No ABX mice (F.2I). However, exposure to L. plantarum or L. murinus restored their production to levels comparable to No ABX mice. Interestingly, the levels of these molecules in ABX mice did not differ significantly between pre- and post-infection states. Additionally, IL-17F and IFN-γ levels showed distinct trends among groups (Fig. S5A). Cytokine levels in the spleen showed no significant differences across groups, suggesting that pulmonary mucosal immunity, rather than systemic immunity, plays a critical role in resistance to S. aureus or P. aeruginosa pulmonary infections (Fig. S5B). IL-17A is known to be essential for inflammatory responses, particularly in promoting the development, maturation, and recruitment of neutrophils [47, 48]. Consistent with this, post-infection neutrophil counts in the lungs were significantly higher in No ABX mice and those exposed to L. plantarum or L. murinus compared to ABX mice (Fig. 2J, Fig. S6A). To further elucidate the lung commensals regulate IL-17A to resist bacterial infection, we transferred L. plantarum or L. murinus into IL-17AKO mice. As shown in Fig. 2K, ABX (IL-17AKO) mice exposed to L. plantarum or L. murinus exhibited survival rates comparable to ABX mice but significantly lower than No ABX wild-type (WT) mice post-infection. Correspondingly, bacterial burdens in the lungs of No ABX WT mice were significantly lower than in other groups (Fig. 2L). These data indicate that pulmonary commensal bacteria, especially L. plantarum and L. murinus, mediate protection against S. aureus and P. aeruginosa lung infections through activation of IL-17A-related pathways.

The commensals of L. plantarum and L. murinus regulate Vγ4⁺γδ T cells against bacterial infections

We sought to determine whether disruption of commensal colonization altered the repertoire of IL-17A–producing cells. We found significantly decreased numbers of IL-17A⁺γδ T cells, but not CD4 T cells, ILC3 cells, or NKT cells, in the lungs of ABX mice as compared to No ABX mice (Fig. 3A, Fig. S6B, C). To determine whether reversing commensal disruption could restore immune function, we exposed ABX mice to L. plantarum or L. murinus. Both treatments successfully restored the numbers of IL-17A⁺ γδ T cells in the lungs of ABX mice (Fig. 3A). Furthermore, we treated γδ T cell-deficient (TCR δKO) mice with antibiotics and exposed them to L. plantarum and L. murinus. By assessing survival rates, bacterial colonization and histopathological analysis, we aimed to elucidate the correlation between L. plantarum, L. murinus, and γδ T cells. Compared to No ABX (WT) mice, antibiotic-treated TCR δKO mice exposed to L. plantarum or L. murinus offered no effective protection against S. aureus or P. aeruginosa infection (Fig. 3B). ABX (WT) mice, L. plantarum treated ABX (TCR δKO) mice. and L. murinus treated ABX (TCR δKO) mice exhibited significantly greater bacterial proliferation in the lungs compared to No ABX (WT) mice (Fig. 3C). Histopathological analysis revealed that ABX (TCR δKO) mice treated with L. plantarum or L. murinus displayed severe pulmonary damage and edema post-infection, similar to WT mice treated with antibiotics, while No ABX (WT) controls exhibited considerably milder lung injury (Fig. 3D). Additionally, we detected the changes of IL-17A in lungs of these groups prior and post S. aureus or P. aeruginosa infection. As shown in Fig. S7A, no statistically significant difference was observed in the intrapulmonary IL-17A production between ABX (WT) mice and ABX (TCR δKO) mice exposed to L. plantarum or L. murinus. We found that IL-17A in these group were shown no significantly difference between prior and post infections. These findings demonstrate that long-term antibiotic exposure disrupts commensal bacteria, leading to a reduction in IL-17A–producing γδ T cells in the lungs and increased susceptibility to S. aureus and P. aeruginosa induced pneumonia. Importantly, lung commensal, L. plantarum and L. murinus, can partially rescue these immune defects and restore pulmonary defenses.

Fig. 3.

L. plantarum and L. murinus regulated Vγ4⁺ γδ T cells secreting IL-17A against infections. A Representative flow cytometry plots of distinct subsets of IL-17A⁺ cells (γδ T cells, CD4 T cells, NKT cells and ILC cells) in the lungs of ABX mice, No ABX mice, ABX mice exposed to L. plantarum, and ABX mice exposed to L. murinus. B Survival rates of No ABX (WT), ABX (WT), ABX (TCR δKO) + L. plantarum and ABX (TCR δKO) + L. murinus mice infected with S. aureus or P. aeruginosa. C S. aureus burdens and P. aeruginosa burdens in lung at 12 h post-intranasal inoculation in the lower concentration of bacterial slurry. D Histological analysis of lung at 12 h post-intranasal inoculation. E Absolute percentage and number of IL-17A⁺ Vγ2⁺ γδ T cells and IL-17A⁺ Vγ4.⁺ γδ T cells in ABX mice, No ABX mice, ABX mice exposed to L. plantarum, and ABX mice exposed to L. murinus. F Mice were administered the Vγ2 or Vγ4 neutralizing antibody (αVγ2, αVγ4) concomitant with S. aureus or P. aeruginosa infection. Survival rates of No ABX, ABX, No ABX (αVγ2), No ABX (αVγ4) mice post challenge. G S. aureus burdens and P. aeruginosa burdens in lung at 12 h post-intranasal inoculation in the lower concentration of bacterial slurry. Data are presented as means ± SEM. Significant differences were calculated with one-way ANOVA followed by Tukey’s multiple comparisons test. ns not significant, *p < 0.05, **p < 0.01, ***p < 0.001

Vγ2⁺, Vγ4⁺ γδ T cells have all been reported to exist in the lung, and their relative abundance seems to be context dependent during different types of infection or inflammatory conditions [49, 50]. As depicted in Fig. 3E, a significant reduction was observed in both the percentage and number of IL-17A-producing Vγ4⁺ γδ T cells in ABX-treated mice. Importantly, this decrease was reversible upon exposure to either L. plantarum or L. murinus. In stark contrast, IL-17A-producing Vγ2⁺ γδ T cells showed no discernible change following antibiotic treatment. These results imply that L. plantarum and L. murinus predominantly modulate IL-17A secretion by Vγ4⁺ γδ T cells, with less evident effects on Vγ2⁺ γδ T cells. Neutralization of Vγ4 γδ T cells significantly increased mortality and bacterial burden in No ABX mice during infection compared to antibiotic-treated controls. In contrast, neutralization of Vγ2 γδ T cells had no significant effect on survival rates or bacterial burdens in No ABX mice compared to controls (Fig. 3F, G). Moreover, IL-17A production in the lungs was markedly reduced following Vγ4 γδ T cell neutralization, regardless of antibiotic pre-exposure or microbiota transfer (Fig. S7B). This finding underscores the critical role of IL-17A-producing Vγ4⁺ γδ T cells in microbiota-mediated regulation of lung immunity. These results suggest that antibiotic treatment decreases the frequency of pulmonary IL-17A-secreting Vγ4⁺ γδ T cells, thereby increasing susceptibility to pneumonia. This susceptibility can be mitigated by exposure to pulmonary commensals.

Extracellular polysaccharides from Lactobacillus commensals activate Vγ4 ⁺ γδ T cells to secrete IL-17A

Dennis demonstrated that IL-17-producing γδ T cells can be activated by IL-23 and IL-1 in both infectious and noninfectious settings, both in vitro and in vivo [24]. To test the role of these cytokines in microbiota-mediated protection, ABX mice were administered IL-1β and IL-23 neutralizing antibodies, exposure to L. plantarum or L. murinus, and subsequently infected with S. aureus or P. aeruginosa (Fig. 4A). As shown in Fig. 4B, neutralization of IL-1β and IL-23 abolished the protective effects of L. plantarum or L. murinus exposure. Metabolites produced by symbiotic bacteria, such as lactate and propionate, are known to activate dendritic cells and modulate immune responses against infections. Previous studies from our lab demonstrated that EPS derived from Lactobacillus species stimulate γδ T cells to secrete IL-17A [33, 51]. To further explore how L. plantarum and L. murinus activate Vγ4⁺ γδ T cells, we isolated γδ T cells from mouse lungs and co-cultured them in vitro with L-lactate, purified EPS from L. plantarum, and purified EPS from L. murinus (Fig. 4C). The results, presented in Fig. 4D, indicated that EPSs activated γδ T cells to secrete IL-17A in a dose-dependent manner. Importantly, this activation required the presence of both IL-23 and IL-1β. To investigate further, EPS was administered intranasally to ABX-treated mice, and changes in IL-17A secretion by Vγ2⁺ and Vγ4⁺ γδ T cells were assessed. The results, shown in Fig. 4E, Fig. S6B, EPS-treated groups (EPS-L. p and EPS-L. m), exhibited significantly increased numbers of IL-17A-producing Vγ4⁺ γδ T cells in the lungs compared to the L-lactate group, while no significant differences were observed in Vγ2⁺ γδ T cells. In followed study, EPSs and L-lactate were intranasally administered to ABX mice, which were then challenged with S. aureus or P. aeruginosa (Fig. 4F). As shown Fig. 4G, ABX mice that exposed to EPS-L. p or EPS-L. m period likewise showed increased resistance to infections (both S. aureus infection and P. aeruginosa infection) compared to the mice received L-lactate. The ABX mice exposed to EPS-L. p or EPS-L. m shown the lower bacterial burdens after infection (Fig. 4H). We performed dissection on mice and found that in the ABX group, mice infected with S. aureus or P. aeruginosa had severe lung damage. In contrast, EPSs-treated mice exhibited markedly reduced lung injury, while L-lactate treatment provided no significant improvement (Fig. 4I). Additionally, post-infections neutrophil counts in the lungs were significantly higher in ABX mice received EPS-L. p or EPS-L. m compared to ABX mice and the ABX mice exposed to L-lactate (Fig. 4J). These findings demonstrate that EPSs, the metabolite of lung commensal bacteria L. plantarum and L. murinus, activated Vγ4⁺ γδ T cells to secrete IL-17A, thereby mitigating susceptibility to S. aureus and P. aeruginosa

Fig. 4.

Extracellular polysaccharides from Lactobacillus commensals activated Vγ4⁺ γδ T cells to secrete IL-17A recruiting neutrophils to resist S. aureus and P. aeruginosa infection. A Study design. IL-1β-neutralizing antibody (10 µg/mouse) and IL-23-neutralizing antibody (10 µg/mouse) inoculated intranasally to ABX mice for 3 days in a row. B 40 µL bacterial slurry (5 × 10⁶CFUs of S. aureus USA300 or 1 × 10⁶CFUs of P. aeruginosa PAO1) were inoculated intranasally via left nostril, and observed continuously for 48 h. Representative survival rates from two independent experiments are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (***p < 0.001). C Lung γδ cells were isolated using a TCR γ/δ isolation kit for mice according to the manufacturer’s instructions. D A total of 100,00 γδ T cells were co-cultured in vitro with L-lactate, purified EPS from L. plantarum, and purified EPS from L. murinus for 12 h. IL-17A levels in these groups were detected using IL-17A-ELISA Kit. E EPS-L. p, EPS-L. m or Lactic acid (200 µg/mouse) were administrated intranasally to ABX mice. Absolute number of IL-17A⁺Vγ2⁺ γδ T cells and IL-17A⁺Vγ4⁺ γδ T cells in ABX mice, No ABX mice, EPS-L. p received mice, EPS-L. m received mice, or lactic acid received mice. F EPS-L. p, EPS-L. m or lactic acid (200 µg/mouse) inoculated intranasally to ABX mice for 3 days in a row. G 40 µL bacterial slurry (5 × 10⁶CFUs of S. aureus USA300 or 1 × 10.⁶CFUs of P. aeruginosa PAO1) were inoculated intranasally via left nostril, and observed continuously for 48 h. Representative survival rates from two independent experiments are shown (n = 9–10). The data for survival test were analyzed by Wilcoxon log-rank survival test (**p < 0.01, ***p < 0.001). H S. aureus and P. aeruginosa burdens at 12 h post-intranasal inoculation in lung (n = 7–8). I Representative macroscopic views of lungs. J The percentage of neutrophils in the lungs of No ABX mice, ABX mice, ABX mice exposed to EPS-L. p, ABX mice exposed to EPS-L. m and ABX mice exposed to Lactic acid post S. aureus or P. aeruginosa infection (n = 4–6). Data are presented as means ± SEM. Significant differences were calculated with one-way ANOVA followed by Tukey’s multiple comparisons test. ns not significant, *p < 0.05, **p < 0.01, ***p < 0.001.

Discussion

In recent years, the incidence of pneumonia caused by bacteria such as methicillin-resistant Staphylococcus aureus (MRSA) and P. aeruginosa has been steadily rising, with high morbidity and mortality rates [34, 52]. While antimicrobial therapies remain the primary treatment, their effectiveness is limited, and in some cases, antibiotics may increase the risk of hospital-acquired pneumonia by disrupting microbial balance [53, 54]. Here, we demonstrated that long-term antibiotic treatment, particularly with ampicillin and vancomycin, enhanced susceptibility to S. aureus and P. aeruginosa, leading to pulmonary inflammation and death. Antibiotics significantly impair γδ T cell function in the lungs during bacterial infection, with protection against antibiotic-associated mortality partially dependent on IL-17A. Furthermore, we discovered that lung commensal bacteria play a unique role in combating S. aureus and P. aeruginosa lung infections. Exposure of ABX mice to lung commensal bacteria L. plantarum and L. murinus stimulated Vγ4⁺ γδ T cells to secrete IL-17A. Further in-depth research revealed that EPSs, metabolites of lung commensal bacteria, played a crucial role in this process.

This study focused on the early response to lung commensals in the context of acute lung infection. A recent study demonstrated that long-term antibiotic exposure increases mice’s susceptibility to S. pneumoniae and K. pneumoniae [10]. However, whether long-term antibiotic use enhances susceptibility to S. aureus and P. aeruginosa, and the underlying mechanisms, remain unclear. To investigate, we established an antibiotic-induced dysbiosis model by treating mice with a four-antibiotic regimen for 21 days. Upon lung infection with S. aureus and P. aeruginosa, ABX mice showed significantly lower survival rates compared to non-antibiotic-treated controls, confirming that prolonged antibiotic use increases susceptibility to these pathogens. Given the anatomical, developmental, and immune differences between GF and SPF mice, studying microbiota-mediated protection against pneumonia in GF models has proven challenging [55]. Broad-spectrum antibiotic-induced dysbiosis in SPF mice provides a practical alternative for investigating the protective roles of symbiotic microbiota.

Mucosal surfaces, such as the gut, skin, and lungs, host diverse microbiota essential for immune homeostasis and defense against infections [39, 56]. While lungs were once thought to be sterile, high-throughput sequencing has revealed a dynamic microbial ecosystem in healthy lungs, closely tied to local immunity [6, 57, 58]. This ecosystem is dynamic, similar to the gut, and closely linked to the local microbial environment, with corresponding immune characteristics. Antibiotic treatment has been shown to increase the risk of hospital-acquired pneumonia, likely due to the disruption of microbiota-dependent pulmonary immune responses caused by clinically used antibiotics [54]. Microbes primarily enter the lungs via microaspiration, a pathway evidenced by the greater similarity of the lower respiratory tract microbiota to oral flora than to nasal flora [7]. Recently, a study on human oral microbiota indicates that episodic aspiration with oral commensals is rapidly cleared from the lower airways but induces a prolonged immune response that secondarily decreases susceptibility to S. pneumoniae [59]. Investigations of murine pulmonary microbiota identified L. murinus as the predominant commensal under steady-state conditions, where it exerts essential protective effects against pneumococcal colonization [46]. Undoubtedly, the impact of the pulmonary microbiota on respiratory diseases is garnering growing attention. In this study, following microbiota reconstruction, mice with restored lung microbiota showed significantly higher survival rates and reduced bacterial loads after infection compared to those receiving gut microbiota reconstruction. This highlights the critical role of pulmonary commensal bacteria in combating bacterial pneumonia.

Currently, FMT is widely recognized as an effective treatment for Clostridioides difficile infections [60, 61]. Its potential for treating other gut and extra-gut diseases associated with microbiota dysbiosis has also become a research hotspot. However, findings from various studies are inconsistent, and significant controversies remain. Additionally, there is no consensus on the standardization of FMT procedures, safety assessments, or the mechanisms underlying its therapeutic effects, necessitating further in-depth research [62, 63]. Identifying specific symbiotic bacteria with anti-infective advantages is crucial for advancing microbiota-based therapies. Herein, we found that mice treated with ampicillin and vancomycin exhibited significantly lower survival rates after infection with pathogenic bacteria, suggesting that commensals sensitive to these antibiotics play an important role in resisting S. aureus and P. aeruginosa. Previous studies have reported that prolonged use of vancomycin alone increases the risk of fungal and bacterial co-infections [25]. Microbiota diversity analysis of antibiotic-treated and non-treated groups revealed similar trends in the changes of lung and fecal microbiota, potentially indicating a direct relationship with the gut-lung axis. Antibiotic-treated mice showed a significant reduction in Gram-positive bacteria, particularly Lactobacillus, in both lungs and feces. The therapeutic potential of Lactobacillus species in animal models of respiratory infections has been extensively documented. Oral supplementation with various Lactobacillus strains, such as L. casei [64], L. rhamnosus [65], L. gasseri [66, 67], L. plantarum [68, 69], has been demonstrated to mitigate influenza virus-induced pathology and mortality in mice. Likewise, L. rhamnosus [65] administered orally improved the control of respiratory syncytial virus (RSV) infection. Furthermore, a protective role for L. casei [70, 71] and L. rhamnossus [72] has also been evidenced in respiratory infections involving S. pneumoniae or P. aeruginosa [73]. In this study, we revealed that ABX-treated mice, when exposed to L. plantarum and L. murinus, exhibited enhanced survival following infection with S. aureus and P. aeruginosa, highlighting the importance of lung commensals L. plantarum and L. murinus in defending against bacterial pulmonary infections.

It is generally believed that symbiotic bacteria resist infection by exogenous pathogens through three main mechanisms: (a) Intestinal symbiotic microbiota deplete nutrients and energy sources, thereby depriving pathogens of ecological niches. (b) They secrete antimicrobial substances, such as antimicrobial peptides, to inhibit the growth and reproduction of pathogenic bacteria. (c) They modulate the immune microenvironment to combat pathogenic infections [74, 75]. We first co-cultured lung-derived L. plantarum, L. murinus, and S. aureus, P. aeruginosa in vitro. The results showed that lung-derived microbiota could not directly inhibit or had only limited ability to inhibit S. aureus and P. aeruginosa. However, L. murinus effectively inhibited the growth of S. pneumoniae, consistent with findings reported by Yildiz [46]. IL-17 is a crucial pro-inflammatory cytokine that plays a key role in immune defense, inflammatory responses, and maintaining mucosal barrier integrity [48]. Previous studies have shown that gut microbiota regulate host immune responses through metabolites (e.g., short-chain fatty acids) and microbe-associated molecular patterns (MAMPs), promoting IL-17 production. Certain bacteria (e.g., Mycobacterium species) can induce Th17 cell differentiation, increasing IL-17 secretion [76, 77]. Human oral commensals (MOC; Prevotella melaninogenica, Veillonella parvula, and Streptococcus mitis) induces a prolonged Th17 response [59]. Additionally, studies on lung infections caused by S. aureus and P. aeruginosa have demonstrated that IL-17 plays a critical role [78, 79]. Here, we demonstrated that the L. plantarum and L. murinus resist bacterial lung infections in an IL-17A-dependent manner. IL-17 stimulates epithelial cells to secrete chemokines (e.g., CXCL1, CXCL5), recruiting neutrophils to the infection site and enhancing bacterial phagocytosis [13, 15]. This aligns with our observation that ABX mice exposure to L. plantarum or L. murinus showed increased pulmonary neutrophil counts and upregulated associated chemokines.

In the gut, there are many cellular sources of IL-17A that are important for enhancing innate intestinal defense, including Th17 cells [80]. In the airways, potential sources of this cytokine include γδ T cells, invariant natural killer T cells, and alveolar macrophages themselves [81]. However, the source of this cytokine and the molecular mechanisms by which the microbiota regulate IL-17A production in our model remain to be determined. Herein, we observed that lung commensal L. plantarum and L. murinus predominantly activated γδ T cells to secrete IL-17A. Across various models, γδ T cells demonstrate diverse mechanisms in their anti-infectious and pro-tumor functions. Moreover, the phenotypes and subsets of the specific γδ T cell involved also differ considerably [82, 83]. A study on corneal commensal bacteria revealed that these bacteria protect the body from corneal infections by inducing γδ T cells in the cornea to secrete IL-17 [9]. Using a genetic model for primary sclerosing cholangitis, researchers showed how increased gut leakage and the translocation of L. gasseri from the gut to the liver activated and expanded hepatic Vγ6⁺ γδ17 T cells, resulting in a mature CD44^highCD62⁻ phenotype [84]. Expanding on analyses of microbiota-affected γδ T cell subsets, this study uniquely revealed that lung commensals modulate pulmonary Vγ4⁺ γδ T cells. This modulation promotes IL-17A secretion, which confers robust resistance against S. aureus or P. aeruginosa pneumonia. Human γδ T cells exhibit compartmentalized distribution, with Vδ1⁺ cells enriched in mucosal epithelia and Vγ9Vδ2 cells dominating peripheral blood [85, 86]. Conversely, in mice, γδ T cell subsets originating in the fetal thymus acquire distinct effector cytokine profiles (e.g., IFN-γ and IL-17), allowing for specialized responses against infections and tumors [87, 88]. Direct extrapolation of mouse study findings to human clinical pulmonary immune mechanisms is challenging, given the taxonomic disparities between human and mouse γδ T cells and the divergent compositions of their respiratory microbiomes.

The mechanism by which the gut microbiota metabolite acetate resists viral infection involves enhancing type I interferon production through NLRP3 [89]. Symbiotic bacteria can promote anti-tumor immunity via the metabolite trimethylamine N-oxide (TMAO) [90]. Gut symbiotic lactic acid bacteria stimulate dendritic cells (DCs) to extend dendrites into the intestinal lumen through the secretion of l-lactic acid, enabling pathogen recognition, activation of the intestinal lamina propria immune response, and the exertion of anti-infection effects [91]. The immune-priming potential of bacterial polysaccharides, acting as TLR agonists, has been investigated as a promising immunotherapeutic strategy for infections. For example, E. coli-derived LPS priming has been shown to enhance protection against pulmonary infections by K. pneumoniae and influenza A virus [92, 93]. Additionally, a recent study suggests that microbial exopolysaccharide produced by Lactobacillus delbrueckii can induce CD8 T cell immune responses and improve the tumor microenvironment [94]. In this study, we proved that metabolites of the lung commensals L. plantarum and L. murinus, especially EPSs, are capable of activating Vγ4⁺ γδ T cells to secrete IL-17A, which in turn mitigated susceptibility to S. aureus and P. aeruginosa. The combined use of antibiotics and EPSs could potentially enhance therapeutic efficacy, especially against the backdrop of increasing antibiotic resistance and severe side effects, although it will certainly bring some challenges.

All lung microbiota data in this study were based on mouse models rather than human clinical models. Considering the differences between mouse and human lung microbiota [95, 96] as well as their immune microenvironments (including γδ T cell subsets [97, 98]), future clinical research on lung microbiota should be conducted with great caution. Furthermore, probiotic strains with anti-infective immunomodulatory properties may exert antagonistic effects against S. aureus and P. aeruginosa infections in individuals. However, these interventions carry potential risks, including the exacerbation of chronic inflammation and inducing lung cancer [99]. In summary, our data reveal that lung commensals, L. plantarum and L. murinus mitigate susceptibility to S. aureus and P. aeruginosa infections through a molecular mechanism involving IL-17A secretion by γδ T cells. These studies broaden understanding of host-microbiota interactions, providing guidance for treating antibiotic-induced immunosuppression and lung infections, while also offering new insights into the development of immunomodulatory probiotics for infection resistance.

Additional files. Fig S1. Broad-spectrum antibiotic pre-exposure results in a reduction in the bacterial load of the intestinal and bronchial lavage fluid (BALF) microbiota. A B21, 21 days before antibiotics treatment. S1, 1 day after stopping antibiotic-treatment. B Gram’s staining of faeces and BALF from non-antibiotic-treated mice (No ABX) and antibiotic-treated mice (ABX). C Bacterial burdens in the faeces and BALF (D) of No ABX (n = 6) and ABX (n = 10) mice at day 21 post drinking antibiotic water for up to 21 days. Fig S2. The number of bacteria in feces and lungs. Faeces and lungs bacteria enumerated in ABX mice, No ABX mice, ABX mice transferred intestinal contents orally and ABX mice transferred BALF contents intranasally, quantified using real-time polymerase chain reaction (RT-PCR). Fig S3. Lung commensal bacteria, L. plantarum and L. murinus mediated protection against P. aeruginosa and S. aureus did not depend on direct inhibition. S. aureus, P. aeruginosa was grown in presence of fresh MRS medium or L. plantarum, L. murinus conditioned MRS medium, both at a dilution of 1:10 with S. aureus or P. aeruginosa TSB media. B. pumilus was grown in presence of fresh TSA medium and S. pneurmoniae was grown in presence of fresh TSB medium as controls. A P. aeruginosa cultures were grown in presence of fresh or B. pumilus or L. plantarum or L. murinus conditioned media. Culture growth was followed hourly by optical density at 600 nm (OD600) for 12 h. We quantified the viable counts of P. aeruginosa at 3 and 12 hours of co-culture, respectively. B S. aureus cultures were grown in presence of fresh or B. pumilus or L. plantarum or L. murinus conditioned media. Culture growth was followed hourly by optical density at 600 nm (OD600) for 12 h. We quantified the viable counts of S. aureus at 3 and 12 hours of co-culture, respectively. Pooled data from two independent experiments are given as mean ± SD. C S. pneurmoniae cultures were grown in presence of fresh or B. pumilus or L. plantarum or L. murinus conditioned media. Culture growth was followed hourly by optical density at 600 nm (OD600) for 12 h. We quantified the viable counts of S. pneurmoniae at 3 and 12 hours of co-culture, respectively. Pooled data from two independent experiments are given as mean ± SD. Data are presented as means ± SEM. Significant differences were calculated with One-way ANOVA followed by Tukey’s multiple comparisons test. ns, not significant, *p < 0.05, **p < 0.01, *** p < 0.001. Fig S4. Lung commensal bacteria, L. plantarum and L. murinus mediated protection against P. aeruginosa and S. aureus may be relate to IL-17 signal pathway. A PCA analysis and KEGG pathway analysis for the lungs of No ABX, ABX, ABX exposed to L. plantarum and ABX exposed to L. murinus mice post S. aureus infection (n=3). B PCA analysis and KEGG pathway analysis for the lungs of No ABX, ABX, ABX exposed to L. plantarum and ABX exposed to L. murinus mice post P. aeruginosa infection (n=3). Fig S5. Lung commensal bacteria promote IL-17A dependent mucosal defenses. A The levels of IL-17F and IFN-γ in the lung of ABX, No ABX, ABX mice exposed to L. plantarum and ABX exposed to L. murinus mice prior or post S. aureus or P. aeruginosa infection (n=3-5). B The levels of IL-17A, IL-17F, IFN-γ, CXCL1, CXCL2, IL-1β and IL-23 in the spleens of ABX, No ABX, ABX exposed to L. plantarum and ABX exposed to L. murinus mice prior or post S. aureus or P. aeruginosa infection. Fig S6. Lung commensals regulated Vγ4⁺ γδ T cells secreting IL-17A against infections. A Representative gating strategy for neutrophils (CD45⁺ CD11b+ Gr-1⁺). B Representative gating strategy for Th17 cells (CD45⁺ CD4⁺ IL-17A⁺), γδ 17 T cells (CD45⁺ TCR γ/δ⁺ IL-17A⁺), Vγ2⁺ γδ 17 T cells (CD45⁺ TCR γ/δ⁺ Vγ2⁺ IL-17A⁺) and Vγ4⁺ γδ 17 T cells (CD45⁺ TCR γ/δ⁺ vγ4⁺ IL-17A⁺). C Representative gating strategy for IL-17A-NKT cells (CD45⁺ NK1.1⁺ IL-17A⁺) and IL-17A-ILC3 cells (CD45⁺ NKp64⁺ CD117⁺ IL-17A+). Fig S7. Long-term antibiotic exposure alters the repertoire of IL-17A–producing Vγ4⁺ γδ T cells in the lung of mice. A The levels of IL-17A in lungs of No ABX (WT), ABX (TCR δKO) +L. plantarum, ABX (TCR δKO) +L. murinus and ABX (TCR δKO) mice before or at 12 h post infection (n=5-6). B The levels of IL-17A in lungs of No ABX, Vγ4 neutralization (No ABX), Vγ2 neutralization (No ABX), ABX, Vγ2 neutralization (ABX+L. plantarum), Vγ2 neutralization (ABX+L. murinus), Vγ4 neutralization (ABX+L. plantarum), Vγ4 neutralization (ABX+L. murinus) mice before or at 12 h post S. aureus infection (n=6).

Authors’ contributions

Haochi Zhang and Xiao Wang conceived the project. Haochi Zhang, Xuemei Bao, Yubing Fu, Xiao Wang designed the studies. Haochi Zhang, Xuemei Bao Chunhe Li, Yanchen Liang, Na Pan, Bin Ma, Ting Wang, Jian Chen performed most of the experiments and analyzed the data. Haochi Zhang, Xuemei Bao, Xiao Wang all other authors discussed the data and wrote the manuscript. Haochi Zhang and Xuemei Bao prepared the figures. Haochi Zhang and Xiao Wang revised the manuscript, and offered financial assistance. Chunhe Li, Yanchen Liang, Na Pan, Bin Ma, Yubing Fu and Lipeng Zhang provided procedural advice, and revised the manuscript. All authors approved the final manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (grant no. 32260184), Distinguished Youth Fund Project of Inner Mongolia Autonomous Region Natural Science Foundation (grant no. 2023JQ08), the Technology Major Project of Hohhot (grant no. 2023-JieBangGuaShuai), the China Postdoctoral Science Foundation (grant no.2023MD734180) and the Clinical Need Oriented Basic Research Project of Inner Mongolia Academy of Medical Sciences (grant no. 2024GLLH0347).

Data availability

16S Ribosomal DNA Gene Sequencing data were deposited at the NCBI Sequence Read Archive (BioProject ID: PRJNA1244771). Gene expression array data were deposited at the NCBI Sequence Read Archive (BioProject ID: PRJNA1245133). All other data relevant to the study are included in the article or uploaded as online supplemental information.

Declarations

Ethics approval and consent to participate

No human subjects. All animal experiments adhered to the ethical guidelines established by the Inner Mongolia University Ethics Committee (SYXK2020-0006).

Consent for publication

Not applicable.

Competing interest

Not applicable.