P. gingivalis alters lung microbiota and aggravates disease severity of COPD rats by up-regulating Hsp90α/MLKL

ABSTRACT

Background

Epidemiological evidence has confirmed that periodontitis is an essential and independent risk factor of chronic obstructive pulmonary disease (COPD). Porphyromonas gingivalis, a major pathogen implicated in periodontitis, may make a vital contribution to COPD progression. However, the specific effects and molecular mechanism of the link between P. gingivalis and COPD are not clear.

Methods and Results

A COPD rat model was constructed by smoke exposure combined intratracheal instillation of E. coli-LPS, then P. gingivalis was introduced into the oral cavity of COPD rats. This research observed that lower lung function, more severe alveolar damage and inflammation occurred in COPD rats with P. gingivalis group. Meanwhile, P. gingivalis/gingipains could colonize the lung tissues and be enriched in bronchoalveolar lavage fluid (BALF) of COPD rats with P. gingivalis group, along with alterations in lung microbiota. Proteomic analysis suggested that Hsp90α/MLKL-meditated necroptosis pathway was up-regulated in P. gingivalis-induced COPD aggravation, the detection of Hsp90α and MLKL in serum and lung tissue verified that Hsp90α/MLKL was up-regulated.

Conclusion

These results indicate that P. gingivalis could emigrate into the lungs, alter lung microbiota and lead to aggravation of COPD, which Hsp90α/MLKL might participate in.

KEYWORDS:

• Porphyromonas gingivalis
• chronic obstructive pulmonary disease
• lung microbiota
• Hsp90α
• MLKL
• necroptosis

Introduction

As the main cause of tooth loss, periodontitis impairs people’s quality of life and is an essential and independent risk factor of chronic obstructive pulmonary disease (COPD) [1,2]. COPD is a common and chronic respiratory disease which is characterized by irreversible airflow limitation due to airway and/or alveolar abnormalities [3]. Sometimes COPD patients suffer from severe exacerbations which is associated with significantly lower survival outcomes [4,5]. Poor periodontal health, especially higher plaque scores and fewer teeth, is associated with the increased frequency of exacerbations in COPD patients [6]. Some researches focused on the involvement of periodontal pathogens in the development of COPD [7–9]. Porphyromonas gingivalis, a major pathogens in periodontitis development, may play a crucial role in COPD progress [10]. A prospective follow-up study found that COPD patients with high serum P. gingivalis IgG titers had significantly lower incidence and frequent of exacerbation than patients with normal IgG titers [11]. Our previous studies also found a high detection rate of P. gingivalis in respiratory secretions of patients with COPD exacerbation, which is homologous to P. gingivalis in dental plaque, supported the hypothesis that P. gingivalis may contribute to the pathology of COPD exacerbation [12]. However, the specific effects and molecular mechanism of the link between P. gingivalis and COPD are not clear.

In healthy individuals, the lung microbiota is relatively simple and low in abundance that an ecological balance between lung microbiota and host was maintained through mechanical and immune clearance [13]. However, there is a significant disruption of the lung microbiota, characterized by dysbiosis in microbial composition and quantity in COPD [14,15]. Multi-omics analysis found that the altered lung microbiome in COPD interact with host gene expression by their metabolites to promote inflammation [16]. The changes in airway microbes also shift COPD from neutrophilic to eosinophilic inflammation [17]. Pseudomonas aeruginosa and Streptococcus pneumoniae are common respiratory condition pathogens that may induce acute exacerbation of COPD [18,19]. P. gingivalis was known to enhance the adhesion of Streptococcus pneumoniae to alveoli by promoting the expression of Platelet-Activating Factor Receptor through gingipain [20]. P. gingivalis or its virulence factors can also cooperate with Streptococcus pneumoniae and Pseudomonas aeruginosa to promote alveolar epithelial cell apoptosis and aggravate respiratory inflammation [21,22]. However, no study has reported the impact of P. gingivalis on lung microbiota in COPD progress.

Inflammation is the main mechanism of COPD progression, which can lead to small airway damage, alveolus destruction, and lung function reduced [23]. As a cytosolic molecular chaperone, Heat shocked protein 90 α (Hsp90α) could regulate the phosphorylation, oligomerization and membrane transport of Mixed Lineage Kinase Domain Like Pseudokinase (MLKL) [24,25]. Activated MLKL can translocate to the cell membrane, leading to plasma membrane rupture and cell necroptosis [26]. With the rupture of the plasma membrane, cells swell, chromatin shrinks, and damaged cells release a significant quantity of damage associated molecular patterns (DAMPs), inflammatory factors and cellular contents which directly causing inflammation and indirectly spreading inflammatory responses [27–29]. Early study has found that Hsp90α in serum of patients with moderate-to-severe COPD are significantly higher than those of healthy smokers and non-smoking controls [30]. The expression of MLKL/p-MLKL protein is increased in respiratory epithelial cells and macrophages of patients with severe COPD [31]. Therefore, Hsp90α/MLKL may participate in the occurrence and development of COPD.

In the present study, by constructing COPD rat model and P. gingivalis inoculation in COPD rats’ oral cavity, we want to investigate the specific effects and molecular mechanism of the link between P. gingivalis and COPD. So as to clarify the role of P. gingivalis in COPD progression and provide a novel idea on the prevention and treatment of COPD.

Materials and methods

Bacterial cultures

P. gingivalis W83 (Provided by the Department of Periodontology, School of Stomatology, China Medical University) were cultured in fresh brain heart infusion (BHI) agar plate with 0.5% hemin, 0.1% menadione (vitamin K1),0.05% yeast extract and 5% sterile defibrinated sheep blood at 37°C in an anaerobic jar for 3–5 days. Monoclonal colonies in good growth condition (dark black and shiny) were selected to BHI liquid medium without sheep blood. After 16–18 h culturation, bacteria were collected by centrifugation at 3500 rpm for 5 min, and resuspended in sterile PBS. OD was measured using a UV spectrophotometer at 600 nm, and 1 × 10 [9] CFU/mL was used as the bacterial concentration at an absorbance of one.

Animal model

The Sprague-Dawley (SD) rats (180–200 g, 6–7 weeks-old) were purchased from Changsheng Biotechnology Co., Ltd. (Liaoning, China). After being fed adaptively in SPF laboratory animal room for 1 week, the rats were randomly divided into three groups (n = 6 per group): Control, COPD, COPD with P. gingivalis (COPD+P.g). The method of COPD rat modeling was minor modified according to some references [32,33]. Briefly, COPD group and COPD+P.g group rats were exposed to cigarette smoke in an automatic smoke exposure device (C-100, Shanghai Yuyan Instruments Co. Ltd. 50 × 40 × 29 cm) at a rate of 30 minutes of 12 cigarettes (Hong He, Yunnan, China; tar: 18 mg, nicotine content: 1.1 mg) twice a day for successive 4 weeks, and intratracheal instillation of 200 μg of Escherichia coli lipopolysaccharide (E. coli-LPS) (O55:B5, Sigma, America) under anesthetize condition on day 1 and day 14. The Control group was feed in fresh air and received intratracheal instillation of 200 μg of sterile PBS under anesthetize condition on day 1 and day 14. (The detail of intratracheal instillation is in the supplementary material) At the end of week 4, lung function tests were performed to estimate if the COPD model was successfully made on the basis of pulmonary functional impairment. At the begin of week 5, COPD+P.g group rats were anesthetized with sodium pentobarbital, a 0.2-mm wire was placed in the dentogingival area of both maxillary first molars, and received1 × 10⁹ CFU/mL(1 mL) of P. gingivalis W83 with 5% carboxymethylcellulose sodium (CMC-Na) in the gingival sulcus and oral cavity once every other day for 2 weeks. Other groups received normal breeding for 2 weeks (Figure 1a). At the end of week 6, lung function tests were carried out to demonstrate the effects of P. gingivalis on COPD rats. Then all rats were sacrificed for following experiments. All procedures were approved by the Animal Research Committee of China Medical University (Permit number: K2020031).

Figure 1. P. gingivalis exacerbates alveolus destruction, impairs lung function and promotes inflammation in COPD rats. (a) Flowchart of animal experiments including COPD modeling and P. gingivalis inoculation in COPD rats’ oral cavity. (b) Lung function test at the end of week 4(n = 6). (c) Lung function test at the end of week 6(n = 6). (d) Representative HE images of lung tissue. Scale bar:200 µm. (e) Quantification of MLI (mean linear intercept) in the lung tissue (n = 6). (f-g) the content of IL-8 and TNF-α in BALF and serum (n = 6). ns: no significant difference; *p < 0.05; **p < 0.01; ***p < 0.001. PEF: peak expiratory flow; PIF: peak inspiratory flow; TV: tidal volume.

Lung function test

At the end of week 4 and 6, lung function was measured using the EMKA pulmonary system (GYD-003). After calibration of the parameters, the rats were placed in a closed plethysmograph box, and the data were recorded when the rats were quiet and breathing smoothly. The detection time for each rat was 4 minutes, and the parameters included tidal volume (TV), peak inspiration flow (PIF) and peak expiration flow (PEF).

Histological analysis

For histological analysis, the right-lung tissue of rat was formalin-fixed and paraffin-embedded. The deparaffinized and rehydrated lung tissue slices (3 μm) of rat were stained with hematoxylin and eosin (H&E), we calculated the mean linear intercept (MLI) to reflect the mean alveolar diameter [34]. Briefly, we captured five histological fields (upper, middle, lower, left, and right) per rat at 200× magnification from lung section, Using Image J software to measure the total length of the diagonal line (L) and count the number of alveolar septa (Ns) across the diagonal line of the visual field, and MLI=L/Ns.

Immunofluorescence staining

Immunofluorescence staining was performed to detect the colonization of P. gingivalis/gingipain in lung tissue. First, antigen retrieval was performed by using citrate buffer for 2 minutes at 100°C on deparaffinized and rehydrated lung sections (3 μm) of rat. Then incubated by 3% hydrogen peroxide to remove endogenous peroxidase and blocked with 5% goat serum for 15 minutes. Finally, lung sections were incubated with monoclonal mouse anti-rgpA/kgp/hagA antibodies (61BG1.3, DSHB, USA) at a 1:100 dilution overnight at 4°C, followed by incubation with goat anti-mouse IgG Dylight 800 secondary antibody (1:100, Abbkine, USA) and DAPI in the dark. Images were captured by a fluorescence microscope (Nikon).

16S rRNA sequencing and data analysis

To clarify the effects of P. gingivalis on lung microbiota, we collected bronchoalveolar lavage fluid (BALF) by lavaging the left lung three times with 3 mL of cold sterile PBS and remove the cellular debris from lavage fluid. Four BALF samples were randomly picked out from each COPD and COPD+P.g group to investigate the lung microbiota difference between the two groups. The 16S rRNA sequencing of BALF was performed by BGI Genomics Co. Ltd (Shanghai, China). After DNA extraction and amplification of the V3-V4 region of bacterial 16S rRNA in BALF, the validated libraries qualified by the Agilent 2100 bioanalyzer were sequenced using the Illumina HiSeq2000 platform. The tags were clustered into OTUs (Operational Taxonomic Units) and compared with the RDP database to annotate the taxa. Alpha diversity and beta diversity analysis were conducted to compare the diversity difference of bacteria between COPD and COPD+P.g group. Linear discriminant analysis effect size (LEfSe) method was used for identifying differentially abundant taxa between two groups with LDA threshold >4.

Label-free LC-MS/MS proteomics analysis

To further explore the potential molecular mechanism in the effects of P. gingivalis on COPD rats, we collected serum samples from all rats and randomly picked out three serum samples from each COPD and COPD+P.g group for Label-free proteomics analysis, that was performed by PTM Bio (Hangzhou, Zhejiang, China). The raw mass spectral data was processed by MaxQuant software (v.1.5.2.8) based on the standard protocol [35]. Tandem mass spectra were searched against the protein sequence of rat in the Uniport database (54,764 entries), and the LFQ (label-free quantification) intensity of each protein was calculated based on the database search results [36]. These quantitative values were then log2-transformed, and statistical tests (three-sample, two-tail t-test) were applied to test for differential expression between COPD and COPD+P.g group. A fold change cutoff of >1.5 or <1/1.5 and a p-value cutoff of <0.05 were used to identify significant differentially expressed proteins and visualized via Volcano plot. KEGG pathway enrichment analysis was performed using the ClusterProfiler package in RStudio to identify enriched biological pathways associated with upregulated differential expressed proteins. Principal Component Analysis (PCA) was also conducted using the factoextra and factoMineR packages in RStudio to visualize the differences between COPD and COPD+P.g groups.

Immunohistochemistry staining

We employed Immunohistochemistry staining to evaluate the expression of Hsp90α and MLKL in the lung tissue of each group. The deparaffinized and rehydrated lung sections (3 μm) of rat were prepared. After antigen retrieval and blocking, the lung sections were incubated with anti-MLKL dilution (1:1,000, Proteintech Group, China) or anti-Hsp90α dilution (1:800, Proteintech Group, China) overnight at 4°C, followed by incubation with enzyme-labeled sheep anti-mouse/rabbit IgG-polymer (Gene Tech, China). Finally, DAB solution was added to visualize the anti-body complexes and sections were counterstained with hematoxylin dye. IHC images were obtained using an optical microscope (Nikon). The mean optical density of the positive staining area was evaluated by ImageJ software.

ELISA

BALF and serum samples were collected from each group as described above. According to the manufacturer’s protocols, using ELISA kit to detect the content of IL-8 (ABIN2535650, 4A Biotech, Germany), TNF-α (SEA133Ra, Cloud-Clone, China) in BALF and serum of rats. We also detect the content of Hsp90α (SED523Mi, Cloud-Clone, China) and MLKL (SER645Ra, Cloud-Clone, China) in serum of rats.

Statistical analysis

Statistical analysis of 16S rRNA sequencing and proteomics has been described above. The other results were presented as mean ± standard deviation (SD) and analyzed by statistical software GraphPad prism 9.0. The normality of the data was assessed using the S–W test. In case of normal distribution, One-way ANOVA analysis was employed to compare the differences among groups followed by the Tukey multiple comparison test. p-Values <0.05 were considered significant.

Results

P. gingivalis exacerbates alveolar destruction, impairs lung function, and promotes inflammation in COPD rats

To investigate the effect of P. gingivalis on COPD progression, we constructed a COPD rat model by smoke exposure combined with endotracheal instillation of E. coli-LPS (Figure 1a), and measured lung function at the end of week 4 to verify the PEF, PIF and TV were significantly lower in the COPD+P.g and COPD groups than that in the Control group (p<0.001) (Figure 1b). After P. gingivalis was inoculated into the oral cavity of the COPD rats for 2 weeks (Figure 1a), we observed an increase in the severity of COPD. Compared with these in the COPD group, the PEF, PIF and TV in COPD+P.g group were significantly lower (p<0.01, p<0.05, p<0.05) (Figure 1c). For the histopathological changes in the lungs, HE staining images revealed that the Control group rats had relatively intact alveolar structures with orderly cells. However, the alveoli of the COPD and COPD+P.g group were enlarged, and some of the alveolar walls had ruptured, leading to the formation of pulmonary bullae with inflammatory cell infiltration in the alveolar septa (Figure 1d). Quantitative analysis of the alveolar mean linear interval (MLI) showed that the mean alveolar diameter in the COPD+P.g group was significantly greater than that in the COPD and Control group (p<0.001) (Figure 1e). Moreover, increased levels of TNF-α and IL-8 were found in BALF and serum of the COPD+P.g group versus COPD group (Figure 1f,g) (p < 0.05, p < 0.01). These data indicated that P. gingivalis exacerbates alveolar destruction, impairs lung function and promotes inflammation in COPD rats.

P. gingivalis colonizes the lungs and alters the lung microbiota of COPD rats

To elucidate the impact of P. gingivalis on the lung microbiota in COPD rats, we first observed by immunofluorescence staining that P. gingivalis/gingipains could colonize the lungs, not only in the bronchus but also the alveolus (Figure 2a). Then we conducted 16S rRNA sequencing of the BALF from the COPD and COPD+P.g groups. A total of 21.5% of OUTs were distinct from those in the COPD+P.g group according to the Venn diagram of bacteria (Figure 2b). Principal coordinates analysis (PCoA) based on weighted UniFrac revealed that the differences of the beta diversity in lung microbiota composition between the COPD and COPD+P.g groups were significant (ANOSIM, R = 0.4479, p = 0.027) (Figure 2c). There was also a noticeable increase in the Shannon index of the alpha diversity of the lung microbiota in the COPD+P.g group compared with that in the COPD group (p = 0.029) (Figure 2d). At the phylum level, the abundances of Veillonella and Streptococcus were greater in the COPD+P.g group than in the COPD group (Figure 2f). Consistent with immunofluorescence staining results, P. gingivalis abundance increased in COPD+P.g group at the species level (Figure 2e). Notably, Porphyromonas was the remarkable differential microbiota constituent between the COPD and COPD+P.g groups, according to the LEfSe results (Figure 2g). These data indicated that P. gingivalis can colonize the lungs and change the lung microbiota of COPD+P.g rats.

Figure 2. P. gingivalis colonizes the lungs and alters lung microbiota of COPD rats. (a) Representative immunofluorescence images of P. gingivalis/gingipains in lung tissues. Scale bar:100 µm. (b) Venn diagram of shared and unique OTUs between COPD and COPD+P.G group. (c) PCoA plot of beta diversity of lung microbiota based on weighted unifrac. (ANOSIM, r = 0.4479 p = 0.027). (d)Boxplot of alpha diversity (Chao1, observed species, and Shannon index) of lung microbiota in BALF between COPD and COPD+P.G group (n = 4). (e-f) relative abundance of lung microbiota at the species (e) and phylum (f) levels between COPD and COPD+P.G group. (g) Cladograms of LEfSe showing differences in the bacterial taxa between COPD and COPD+P.G group (LDA score > 4).

Hsp90α and MLKL are up-regulated during P. gingivalis-induced COPD aggravation

To explore the underlying molecular mechanism involved in P. gingivalis-induced COPD aggravation. we identified differentially expressed proteins in the serum of the COPD group and COPD+P.g group by proteomics analysis with a threshold of FC > 1.5 or <0.667 and a P. value <0.05. There were 79 up-regulated proteins and 77 down-regulated proteins in the COPD+P.g group (Figure 3b). PCA revealed a distinct difference between the COPD and COPD+P.g group, which can be divided into two groups (Figure 3a). However, KEGG enrichment analysis of the up-regulated proteins suggested that necroptosis pathway participate in P. gingivalis-induced COPD aggravation (Figure 3c). Furthermore, Hsp90α and MLKL were involved in the necroptosis pathway (Figure 3d). Consistent with the proteomics results, the serum levels of Hsp90α and MLKL in COPD+P.g group were significantly greater than those in the COPD group (p < 0.001) (Figure 4a,b). Spearman’s correlation analysis revealed that the serum levels of Hsp90α and MLKL were correlated positively with the MLI (p < 0.01) (Figure 4c). Accordingly, the expression of Hsp90α and MLKL in lung tissue was measured by immunohistochemistry staining (Figure 4d,f), and the average optical density of positive cells of Hsp90α and MLKL in the lung tissue of the COPD+P.g group was significantly greater (p < 0.01, p < 0.05) (Figure 4e,g). However, the changes in Hsp90α in both the serum and lung tissue were not significantly different between the COPD group and the Control group (Figure 4a,e). On the other hand, heat-inactive P. gingivalis promotes the expression of Hsp90α/MLKL in A549 cell under COPD condition (see in supplementary materials). These data revealed that Hsp90α and MLKL are up-regulated during P. gingivalis-induced COPD aggravation.

Figure 3. Hsp90α and MLKL are up-regulated in P. gingivalis-induced COPD aggravation revealed by proteomics. (a) PCA results between COPD and COPD+P.G group(n = 3). (b) Volcano plot for the differentially expressed proteins (gene names). The X-axis shows the Log2 FC, and the Y-axis shows the -Log10 P. value. The red and blue circles represent upregulated proteins and down-regulated proteins in COPD+P.G group, respectively. (c) The KEGG pathways analysis of all the upregulated differentially expressed protein in COPD+P.G group. (d) Visualization of the relationship between necroptosis pathways and up-regulated proteins (gene names).

Figure 4. Hsp90α and MLKL are up-regulated in lung tissue and serum of COPD+P.G rats. (a,b) The content of Hsp90α and MLKL in serum (n = 6). (c) Spearman’s correlation analysis of Hsp90α, MLKL, IL-8, TNF-α, PIF, PEF and TV. Color from blue to red showing the negative to positive correlations. (d,f) Representative immunohistochemical staining of Hsp90α and MLKL in lung tissues. Scale bar:120 µm,30 µm. (e,g) Quantification of average absorbance of Hsp90α and MLKL positive cells in the lung tissue (n = 6). ns: no significant difference; *p < 0.05; **p < 0.01; ***p < 0.001.

Discussion

P. gingivalis, as one of major pathogens implicated in the development of periodontitis, can adhere to and colonize the periodontal tissues, expressing various of virulence factors to cause inflammatory response and destroy the periodontal tissues [10,37]. What was more, P. gingivalis was found to be closely related to COPD. They can be detected in BALF and lower respiratory tract secretions of COPD [8,12]. The level of P. gingivalis IgG titers in serum of COPD patients is positively correlated with the frequency of COPD exacerbation [11]. Importantly, the detection rate of P. gingivalis in subgingival plaque was negatively correlated with lung function [38]. Therefore, it is particularly important to investigate the impact of P. gingivalis on the progress of COPD.

Animal experiments found that periodontitis promoted the inflammation of COPD, but it is mainly considered the cause of periodontitis-related inflammatory factors [39]. Another study established an animal model of COPD by elastase induced emphysema, and found that Fusobacterium nucleatum aggravates COPD inflammatory response after it is injected into the trachea [40]. Considering the pathogenesis of COPD is complex, it is associated primarily with smoking, early-life respiratory infections and genetics [41], moreover the location of P. gingivalis is mainly in the periodontal pocket. We constructed the rat model of COPD by cigarette smoke exposure combined with endotracheal instillation of E. coli-LPS, and made P. gingivalis inoculated in COPD rats’ oral cavity by orthodontic wire ligation and local application of P. gingivalis suspension. After P. gingivalis inoculating in oral cavity of COPD rats for 2 weeks, we found the lung function of COPD+P.g group was significantly impaired comparing to COPD group. Accordingly, we measured the MLI of lung tissue to reflect the alveolar size, and the results showed that P. gingivalis worsened alveolar damage in COPD+P.g group. Importantly, the levels of TNF-α and IL-8 in BALF and serum of COPD+P.g group were significantly higher than those of COPD and Control group, which means P. gingivalis aggravated local and systemic inflammation of COPD rats. On the other hand, the inflammatory endotype of COPD will be observed in our subsequent studies.

Although inflammation is the main mechanism of COPD progression, alterations in the lung microbiome are correlated with decreased lung function and enhanced airway inflammation of COPD [17,42]. The ‘vicious circle hypothesis’ explains how the lung microbiome play a role in COPD pathogenesis [43]. Inhalation of toxic gases or acute changes in the lung microbiota disrupt innate defenses, allow dysbiosis of lung microbiota from diversity and abundance, and lead to adverse inflammatory response. The continuous inflammatory response further causes serious imbalance of microbial composition and quantity, and finally forms a vicious cycle of continuous destruction of lung tissue. In our study, we observed P. gingivalis colonized and were enriched in lung of COPD rats after P. gingivalis inoculating in oral cavity for 2 weeks, which indicated P. gingivalis could emigrate and flourish in lungs of COPD rats. From 16S rRNA gene sequencing of lung microbiome in the BALF, we found a significant discriminate of beta diversity between COPD and COPD+P.g according to PCoA, and an increased Shannon index of alpha diversity in COPD+P.g. Although a decreased Shannon index was observed with increasing COPD severity in some clinical researches [44,45]. A multicenter cohort study found great differences in microbial alpha and beta diversity variability of COPD exacerbation across study centers compared to baseline [46], and the lung microbiome diversity was increased in COPD of eosinophilic inflammation [13]. In addition, we observed that the relative abundance of Veillonella and Streptococcus was increased in the COPD+P.g group at phylum level. Multi-omics study found that lower lung function and more severe symptoms were positively associated with Streptococcus and Veillonella in moderate state of COPD [47]. More importantly, Porphyromonas was the remarkable differential microbiota between the COPD and COPD+P.g groups according to LEfSe, which means P. gingivalis is the biomarker in altered lung microbiome of COPD+P.g rats. Therefore, we conduct that P. gingivalis could alter lung microbiota and aggravate disease severity of COPD rats.

To further explore the potential molecular mechanism of how P. gingivalis aggravates COPD progression, serum proteomic analysis of rats was performed and the results suggested that Hsp90α/MLKL-involved necroptosis pathway played an important role in. As the one of isoforms of Hsp90 cytosolic molecular chaperone, Hsp90α is associated with COPD progression [30,48]. It has been found that Hsp90α is involved in necroptosis by regulating the phosphorylation, oligomerization and membrane transport of MLKL, the inhibition of Hsp90 decreases the expression of MLKL [24,25]. In our study, Hsp90α and MLKL were up-regulated in COPD+P.g group compared to COPD and Control group, which was positive correlated with MLI. At the same time, we also noticed that the changes of Hsp90α were not significant in COPD group versus Control group. Study has found that the persistent airway goblet cell metaplasia (a common pathological manifestation of respiratory diseases) is related to Hsp90 activity, the inhibition of Hsp90 can reverse goblet cell metaplasia in vivo and in vitro [49]. Therefore, Hsp90 may be a target for COPD therapy. An in vitro study has reported that P. gingivalis-LPS can promote the expression of Hsp90α in human gingival fibroblasts to cause inflammation [50]. Our previous research has revealed that P. gingivalis-LPS can induce necroptosis of oral epithelial cells and release a large number of DAMPs to further activate macrophages [51]. So, we next conduct that P. gingivalis-induced COPD aggravation is related to the up-regulated Hsp90α and MLKL. On the other hand, our in-vitro experiments in supplementary materials suggested that P. gingivalis had a direct and complex pathogenic mechanism for COPD, which was similar in peridontits [52]. However, whether it is the indirect effect of P. gingivalis on COPD progression needs further verification.

In conclusion, P. gingivalis could colonize and be enriched in the lungs of COPD rats, alter lung microbiota and lead to aggravation of COPD. Proteomic analysis of serum suggested that Hsp90α/MLKL-meditated necroptosis was up-regulated in P. gingivalis-induced COPD aggravation. The detections of Hsp90α and MLKL in serum and lung tissue verified that Hsp90α/MLKL was up-regulated in COPD+P.g group, which might provide a new idea for the molecular mechanism of the link between P. gingivalis and COPD progression.

Acknowledgments

We sincerely thank Yanqing Liu (School and Hospital of Stomatology, China Medical University), Xin Qiao (Department of Pulmonary and Critical Care Medicine, the First Hospital of China Medical University) and HaoShen Feng (Shengjing Hospital of China Medical University) for providing key technical support for this experiment.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/20002297.2024.2334588

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [81700970], [82170969] and [82101027], Doctoral Research Foundation of Liaoning Province [2021-BS-01].