USP4 depletion-driven RAB7A ubiquitylation impairs autophagosome-lysosome fusion and aggravates periodontitis

ABSTRACT

Periodontitis, a prevalent and chronic inflammatory disease, is intricately linked with macroautophagy/autophagy, which has a dual role in maintaining periodontal homeostasis. Despite its importance, the precise interplay between autophagy and periodontitis pathogenesis remains to be fully elucidated. In this study, our investigation revealed that the ubiquitination of RAB7A, mediated by reduced levels of the deubiquitinating enzyme USP4 (ubiquitin specific peptidase 4), disrupts normal lysosomal trafficking and autophagosome-lysosome fusion, thereby contributing significantly to periodontitis progression. Specifically, through genomic and histological analysis of clinical gingival samples, we observed a decreased RAB7A expression and impaired autophagic activity in periodontitis. This was further substantiated through experimental periodontitis mice, where RAB7A inactivation was shown to directly affect autophagy efficiency and drive periodontitis progression. Next, we explored the function of active RAB7A to promote lysosomal trafficking dynamics and autophagosome-lysosome fusion, which was inhibited by RAB7A ubiquitination in macrophages stimulated by Porphyromonas gingivalis (P. g.), one of the keystone pathogens of periodontitis. Last, by proteomics analysis, we revealed that the ubiquitination of RAB7A was mediated by USP4 and validated that upregulation of USP4 could attenuate periodontitis in vivo. In conclusion, these findings highlight the interaction between USP4 and RAB7A as a promising target for therapeutic intervention in managing periodontal diseases.

Abbreviation: 3-MA: 3-methyladenine; Baf A1:bafilomycin A₁; BECN1: beclin 1, autophagy related; CEJ-ABC: cementoenamel junctionto alveolar bone crest; IL1B/IL-1β: interleukin 1 beta; KD:knockdown; LPS: lipopolysaccharide; MOI: multiplicity of infection;OE: overexpression; P.g.: Porphyromonasgingivalis; RILP: Rabinteracting lysosomal protein; ScRNA-seq: single-cell RNA sequencing; SQSTM1/p62: sequestosome 1; S.s.: Streptococcus sanguinis; USP4:ubiquitin specific peptidase 4

Introduction

Periodontitis, a prevalent oral inflammfatory disease ranking sixth globally in incidence, is a prevailing cause of tooth loss, leading to compromised masticatory function, esthetic impairment, and diminished quality of life [1]. The etiology of periodontitis is increasingly understood to involve complex interactions between oral pathogens and the host immune responses. These interactions encompass strategies pathogens use to evade the immune system and changes in the composition of oral microbiota [2–4]. Notably, Porphyromonas gingivalis (P. g.) is recognized as a key pathogen in the development of periodontitis due to its ability to interfere with the host’s innate immune defenses [5]. The virulence factors enable P. g. to enter the bloodstream or cerebrospinal fluid, linking it to systemic conditions like coronary heart disease, rheumatoid arthritis, and Alzheimer disease [6,7].

Macroautophagy/autophagy is a vital self-degradative cellular process that involves a series of steps: initiation, nucleation, formation and elongation of phagophore, fusion of the autophagosome with lysosomes. Autophagy is crucial for maintaining the balance of periodontal tissues by inhibiting inflammatory cytokines like IL1B/IL-1β, one of the earliest cytokines to be elevated during periodontal infection, which can accelerate alveolar bone loss [8–10]. Studies indicate that both classical and nonclassical forms of autophagy, such as LC3-associated phagocytosis/LAP, can suppress the production and release of IL1B. In the absence of autophagy, macrophages stimulated with lipopolysaccharide (LPS) produce elevated levels of IL1B in a manner dependent on NLRP3 activation [11–13]. Increasing evidence from in-vitro studies indicates that autophagy is activated in different types of periodontal tissue cells when exposed to P. g., including gingival epithelial cells, gingival fibroblasts, and macrophages [14–16]. Nevertheless, recent studies have revealed a contradictory finding: P. g. can disrupt autophagic flux by enhancing the efflux of lysosomes, which prevents the fusion of autophagosomes with lysosomes and induces the production of IL1B in oral epithelial cells [17]. Likewise, deficiencies in autophagy have been documented in human dendritic cells and rat cardiomyocytes, where mechanisms involving the activation of the AKT-MTOR axis or cleavage of VAMP8 allow P. g. to avoid degradation through autophagy, thereby exacerbating inflammation [18,19]. Notably, impaired autophagy has been linked to worsened inflammatory tissue damage in various organs, such as the liver, lungs, and kidneys [20–22]. Based on these findings, we propose that the state of autophagy (whether it is enhanced or compromised) might fluctuate during different phases of periodontitis.

As known, the autophagosome-lysosome fusion step is essential for the autophagic degradation function, and the fusion step is regulated by several multiple regulators of membrane dynamics, including soluble N-ethylmaleimide-sensitive factor attachment protein receptor proteins/SNAREs, tethering protein, V-ATPase, and RAB7A [23]. Among them, the importance of RAB7A in the autophagosome-lysosome fusion process has been emphasized in several studies, and it is fundamental for lysosomal biogenesis, positioning, and functions [24–26]. Similar to other GTPases, RAB7A cycles between the inactive form (GDP-bound) and active form (GTP-bound), which depends on GTP binding [25]. In its active state, RAB7A recruits RILP (Rab interacting lysosomal protein), facilitating the clustering of autophagosomes near the cell nucleus and their subsequent fusion with lysosomes [24,27,28]. Recent research has shown that deficiencies or disruptions in RAB7A function impair the functionality of pulmonary artery endothelial cells and peripheral sensory neurons. These impairments are associated with the development of pulmonary hypertension and Charcot-Marie-Tooth Type 2B disease [29,30]. These findings suggest that deficiencies or inactivation of RAB7A may also contribute to the development of periodontitis.

Therefore, it is crucial to thoroughly investigate the factors influencing RAB7A activity to pinpoint precise therapeutic targets. Ubiquitination modifications of RAB7A are known to control its function, including the coordination of membrane trafficking dynamics and its interaction with downstream effector proteins such as RILP [31–34]. Studies have revealed that PRKN/parkin-mediated ubiquitination of RAB7A plays a crucial role in stabilizing the protein, thereby enhancing its association with RILP for effective vesicular trafficking [32]. Similarly, E3 ubiquitin ligase TRAF6 mediated-RAB7A ubiquitination promoted it bounded to STX17 (syntaxin 17; a soluble N-ethylmaleimide-sensitive factor attachment protein receptor protein that is essential for mature autophagosome), and thus promoting the fusion of autophagosomes with lysosomes [33]. Additionally, studies have indicated that the deubiquitinating enzyme USP32 enables non-ubiquitinated RAB7A to effectively mediate the minus-end transport of late endosomes (nuclear transport). Interestingly, reversible ubiquitination acts as a molecular switch that allows RAB7A to alternate between its various functional roles [34].

In this study, we utilized bioinformatics and histological analysis of clinical gingival specimens to systematically compare autophagy levels and RAB7A expression between periodontitis patients and healthy controls. Subsequently, employing experimental periodontitis mouse models, we investigated how RAB7A inactivation and impaired autophagy contribute to periodontitis pathogenesis. Additionally, in-vitro experiments were conducted to analyze RAB7A activity and its role in regulating lysosomal trafficking, autophagosome-lysosome fusion, and IL1B release. In this process, we identified the deubiquitinating enzyme USP4 (ubiquitin specific peptidase 4) as a pivotal player in modulating RAB7A ubiquitination and its function. Finally, the expression of USP4 was validated and its correlation with IL1B in periodontitis was explored in vivo, collectively demonstrating that reduced USP4-mediated ubiquitination of RAB7A disrupts normal lysosomal trafficking dynamics, impairs autophagosome-lysosome fusion, and significantly contributes to the progression of periodontitis.

Results

Decreased RAB7A expression and impaired autophagy are found in periodontitis patients

Aberrantly low RAB7A expression has been reported in patients with inflammatory diseases including osteoarthritis, nonalcoholic fatty liver disease, and sepsis [22,35,36]. In this study, clinical gingival samples were collected from severe periodontitis patients (stage III C) and healthy donors to analyze RAB7A expression and autophagic activity (Figure S1A). Immunoblotting of gingival sample lysates revealed a decreased RAB7A protein level in individuals with severe periodontitis (Figure 1A). RNA-seq analysis consistently demonstrated reduced RAB7A expression in the severe periodontitis group (Figure S1,B). To evaluate potential autophagy impairment in periodontitis, we assessed the expression of LC3 and SQSTM1/p62 (sequestosome 1) as indicators of autophagic activity. During autophagosome maturation, unlipidated LC3 (LC3-I) converts to phosphatidylethanolamine (PE)-conjugated LC3 (LC3-II). While SQSTM1, acting as a cargo receptor, interacts with autophagic substrates and delivers them to phagophores for ultimate degradation, thus serving as an index of autophagy. Accumulation of both LC3-II and SQSTM1 indicates a blockade in the autophagosome-lysosome fusion and/or degradation steps [37]. Our findings revealed a significant increase in LC3-II and SQSTM1 expression levels in gingival lysates from individuals with periodontitis compared to those from healthy individuals (Figure 1A).

Figure 1.

Impaired autophagy and decreased RAB7A expression are found in periodontitis patients. (A) Western blot analysis of autophagy regulator RAB7A and autophagy markers (LC3, SQSTM1) in healthy and periodontitis gingiva (n = 3 for each group). Anti-RAB7A (cell signaling technology 95,746), anti-LC3 (cell signaling technology 12,741), anti-SQSTM1 (cell signaling technology 88,588). (B) immunofluorescence analysis of autophagy regulator RAB7A, LC3, SQSTM1, LAMP1 in lamina propria of healthy and periodontitis gingiva (n = 3 for each group). Scale bars: 50 μm; zoom: 20 μm. Anti-RAB7A (cell signaling technology 95,746). anti-LC3 (cell signaling Technology,12741), anti-SQSTM1 (cell signaling Technology,88588). Data were presented as mean ± SEM. p-values were calculated through two-tailed Student’s t-tests.

To assess the RAB7A level and potential autophagy impairment in the lamina propria and epithelium of gingival tissue from individuals with periodontitis and healthy controls, we performed immunofluorescence analysis using RAB7A, SQSTM1, and LC3 antibodies on gingival tissue samples from individuals with periodontitis and healthy controls. These antibodies were consistent with those used in the earlier immunoblotting analysis (Figure 1A). We observed a notable decrease in RAB7A expression in the lamina propria of gingival tissue obtained from individuals with severe periodontitis (Figure 1B). Additionally, compared to the healthy control group, we observed significantly increased expression of SQSTM1 and LC3 in the severe periodontitis group. However, the expression of LAMP1 is upregulated in periodontitis, suggesting that the inhibition of autophagic flux is not due to a decrease in lysosome numbers (Figure 1B).

To determine the cell types in which RAB7A expression was downregulated in periodontitis, we reanalyzed a single-cell RNA sequencing (scRNA-seq) dataset from a previously published study, which included 13 healthy and 8 periodontitis gingival samples [38]. Utilizing the UMAP method, we successfully clustered the major cell populations (Figure S1C,D). Our analysis revealed a notable decrease in the relative expression of RAB7A within the lamina propria, encompassing endothelial cells, fibroblasts, and immune cells. Collectively, these results indicated impaired autophagic activity and downregulated RAB7A expression in the gingival samples obtained from individuals with severe periodontitis.

RAB7A agonist ML098 promotes autophagy and inhibits alveolar bone loss in periodontitis mice

Studies have indicated that RAB7A-mediated autophagosome maturation can mitigate neuroinflammation associated with endoplasmic reticulum stress [39]. This suggests that downregulation of RAB7A, resulting in impaired autophagy, could potentially contribute to the development of periodontitis. To assess the impact of RAB7A downregulation on periodontitis pathogenesis, we induced a periodontitis mouse model using ligation along with local application of P. g. suspension [40]. Mice were randomly allocated into three groups: non-model control, periodontitis, and periodontitis treated with the RAB7A agonist ML098 via intraperitoneal injection. Micro-CT analysis was employed to assess the loss of alveolar bone, and immunostaining was performed to analyze the expression of RAB7A, SQSTM1, and LC3 (Figure 2A). Analysis of 2D and 3D micro-CT images showed a marked increase in cementoenamel junction to alveolar bone crest (CEJ-ABC) distance in the periodontitis group compared to both the control and ML098-treated groups. Importantly, there was no significant difference in CEJ-ABC between the control and ML098 groups, suggesting that the RAB7A agonist ML098 mitigated alveolar bone loss in periodontitis mice (Figure 2B,C).

Figure 2.

RAB7A agonist ML098 promotes autophagy and inhibits alveolar bone loss in periodontitis mice. (A) mouse experimental periodontitis scheme. The 6-8-week-old mice were randomly divided into three groups. The control group was not treated. The periodontitis group was subjected to ligature and P. g. on the maxillary second molars with or without RAB7A agonist ML098 (1 mg/kg) intraperitoneal injection (every 2 days). Mice were harvested 10 days later for periodontal tissue analysis. (B) Representative micro-ct analysis of alveolar bone in normal control mice, periodontitis mice, and periodontitis mice with ML098 group. Scale bars: 500 μm. (C) micro-ct analysis of the CEJ-ABC distance at the palatal side of the second molar (indicated by red lines) in three groups of mice (n = 5 for each group). CEJ-ABC, the cementoenamel junction to the alveolar bone crest. (D) Representative Immunofluorescence staining of autophagy regulator RAB7A and autophagy maker LC3 and SQSTM1 in three groups of mice. Scale bars: 50 μm; zoom: 10 μm. Anti-RAB7A (cell signaling Technology,95746), anti-LC3 (cell signaling Technology,12741), anti-SQSTM1 (CST 88,588). (E, F, G) expression of RAB7A, LC3, and SQSTM1 in the gingiva of three groups of mice (n = 5 for each group). Data were presented as mean ± SEM. p-values were calculated through one-way ANOVA test.

Subsequently, we assessed autophagic activity in mice from the three experimental groups using immunostaining analysis. Our findings revealed that compared with the control group, the levels of LC3 and SQSTM1 protein were both increased, while RAB7A protein levels were significantly decreased in the gingiva of mice with periodontitis, consistent with observations in clinical periodontitis patients (Figure 2D–G). Upon treatment with the RAB7A agonist ML098, RAB7A levels were significantly enhanced, whereas the expression of LC3 was reduced in the gingiva of mice with periodontitis, indicating that RAB7A agonist ML098 promoted the autophagic activity in periodontitis mice (Figure 2D–G).

Furthermore, to investigate the role of impaired autophagy in periodontitis pathogenesis, we categorized mice into three groups: non-model control, periodontitis, and periodontitis treated with the V-ATPase inhibitor bafilomycin A₁ (Baf A1) via intraperitoneal injection (Figure S2A). Micro-CT imaging revealed that mice in the Baf A1 group exhibited exacerbated alveolar bone loss (increased CEJ-ABC and decreased bone mineral density) and elevated levels of LC3 and SQSTM1 compared to both the control and periodontitis groups, suggesting that impaired autophagy exacerbates alveolar bone loss in periodontitis mice (Figure S2B-E).

In summary, our investigation of clinical periodontitis gingival samples and experimental periodontitis mice models reveals that decreased RAB7A expression impairs autophagy, thereby playing a role in the development of periodontitis.

Autophagy impairment is mediated by blocked autophagosome-lysosome fusion in macrophages

Macrophages are well-recognized as resident myeloid cells within gingival tissues, playing a pivotal role in orchestrating the immune response during periodontitis [41]. We also observed decreased expression of RAB7A in macrophages within human gingival tissue (Figure S1C, D). Thus, we conducted further analysis to observe gene ontology (GO) enrichment and differential expression of autophagy-related genes in macrophages using the scRNA-seq data (Figure S3A, B). We conducted a thorough analysis of ScRNA-seq data to investigate genes related to autophagy. This includes examining the MTOR complex, autophagy initiation, nucleation, the formation and elongation of phagophores, and the fusion of autophagosomes with lysosomes in macrophages. The Volcano plot revealed that genes involved in autophagosome-lysosome fusion, including RAB7A and VAMP8, were significantly reduced, while the gene associated with autophagy initiation, ULK1, was markedly increased in the periodontitis group. Other genes highlighted in the plot are related to the MTOR complex and autophagy induction. (Figure S3A). Besides, the GO analysis showed a notable enrichment of genes related to “regulation of autophagy,” “macroautophagy,” “phagosome,” and “positive regulation of autophagy” in macrophages from periodontitis-affected gingiva (Figure S3B). These indicate that in macrophages from periodontitis-affected gingiva, autophagy impairment primarily occurs at the fusion stage rather than at the initiation stage.

To determine whether the autophagy process was blocked at the fusion step, we examined the autophagy flux in THP-1-derived macrophages stimulated by P. g., a key pathogen in periodontal disease. Intriguingly, we noted a decrease in SQSTM1 protein levels at 10 and 30 min of stimulation, followed by a gradual increase over time. Conversely, LC3-II levels rose continuously with prolonged P. g. stimulation, suggesting that autophagosome formation was not impaired (Figure 3A). When we extended the stimulation to 8 h, we observed similar autophagy impairments with an increasing multiplicity of infection (MOI). Our findings also indicate a significant reduction in the expression of active RAB7A following P. g. stimulation at 8 h or an MOI of 100, suggesting that the decreased activity of RAB7A in P. g.-stimulated THP-1 cells may be associated with impaired autophagic flux. (Figure 3B; Figure S3C-E). To assess whether the initiation phase of autophagy was affected, we evaluated the levels of autophagy initiation-related proteins, MTOR and BECN1. Non-phosphorylated MTOR, which is crucial for initiating autophagy, activates the ULK1-RB1CC1/FIP200-ATG13-ATG101 complex and the PIK3C3/VPS34-BECN1-ATG14 complex sequentially, thereby promoting autophagic activity [42]. Compared to the control group, the ratio of phosphorylated MTOR to non-phosphorylated MTOR decreased, while BECN1 protein levels continuously increased in THP-1 cells stimulated with P. g. This suggests that autophagosome membrane initiation and elongation were not disrupted in macrophages following P. g. stimulation (Figure 3A,B).

Figure 3.

Autophagy impairment is mediated by blocking autophagosome-lysosome fusion in macrophages. (A-B) time-dependent (left panel) and dose-dependent (right panel) assay of P. g.stimulation were performed on THP-1-derived macrophages. Western blot analysis was performed for the expression RAB7A and active RAB7A, autophagy markers (SQSTM1 and LC3), autophagy initiation-related proteins (BECN1, MTOR, and p-mtor). One representative blot is shown in two independent experiments. (C) GFP-RFP-LC3 THP-1 cells were stimulated with P. g. (MOI = 100), and representative time-lapse confocal images were captured to detect autophagic flux. Red dots represent autolysosomes and yellow dots autophagosomes. Scale bars: 10 μm. Zoom, 5 μm. (D) GFP-RFP-LC3 THP-1 cells were treated with P. g. (MOI = 100), Rapa (10 nM), and Baf A1 (50 nM) for 8 h to detect autophagy flux. Representative images of fluorescent LC3 puncta were shown. Red dots represent autolysosomes and yellow dots autophagosomes. Scale bars: 10 μm. (E-F) the number of GFP⁺ RFP⁺ LC3 (autophagosome) and GFP⁻ RFP⁺ LC3 (autolysosome) was calculated from multicell images. Data were presented as mean ± SEM. p-values were calculated through one-way ANOVA test. Rapa, rapamycin. Baf A1, bafilomycin A₁.

To further investigate whether the fusion step in the autophagic pathway was affected, THP-1 cells were transfected with GFP-RFP-LC3 lentivirus to track autophagy flux (Figure 3C–F). In this system, GFP-LC3 fluorescence (green) diminishes in the acidic environment of lysosomes, leaving only RFP-LC3 fluorescence (red) visible. Consequently, red puncta indicate autolysosomes, while yellow puncta (showing both GFP and RFP fluorescence) indicate autophagosomes [43]. Time-lapse imaging showed that upon P. g. infection, more autolysosomes were observed at the initial 30-min stimulation, whereas autophagosomes were more prevalent when the infection lasted beyond 2 h (Figure 3C, Video S1). When the P. g. stimulation was fixed at 8 h, the number of autophagosomes increased, whereas the number of autolysosomes decreased in the stimulated group compared to the control group (Figure 3D–F), consistent with the immunoblotting results (Figure 3A,B). To further demonstrate that P. g. stimulation inhibits autolysosome biogenesis, we employed rapamycin to enhance autophagosome formation. We then analyzed the levels of LC3-II protein and the number of autophagosomes following P. g. stimulation in rapamycin-treated THP-1 cells. The results indicated that, compared to the rapamycin-alone group, the combination of P. g. and rapamycin treatment significantly promoted the formation of autophagosomes and increased LC3-II protein expression (Figure 3D–F, Figure S3J). This suggests that P. g. stimulation inhibits autophagic flux by affecting the fusion step between autophagosomes and lysosomes. Additionally, in the P. g.-stimulated groups, we added or omitted 3-MA to inhibit autophagosome formation and observed the impact of P. g. stimulation on autophagic flux. The results showed that the presence of 3-MA decreased the levels of LC3-II and increased the accumulation of SQSTM1, further suggesting that P. g. stimulation promoted inhibition of autophagic flux at the autophagosome-lysosome fusion stage (Figure S3F). To demonstrate that inhibition of autophagic flux is a result of effective intracellular infection by P. g., we utilized LysoTracker to label lysosomes, GFP-LC3 to label autophagosomes, and Alexa Fluor™ 647 NHS ester to label P. g.. This allowed us to observe P. g. entering autophagosomes and subsequently fusing with lysosomes (Figure S3G).

To further assess whether the inhibitory effect of P. g. on autolysosome biogenesis is specific, we employed inactivated P. g. and another periodontal pathogen, Streptococcus sanguinis (S. s.), to evaluate their effects on autophagy. The findings indicated that both live P. g. and heat-inactivated P. g. stimulation elevate the levels of LC3-II and SQSTM1 proteins. In comparison to live P. g., heat-inactivated P. g. group showed lower levels of LC3-II and SQSTM1 proteins, indicating that inactivated P. g. has a weaker inhibitory effect on autophagy compared to live P. g. (Figure S3H). Conversely, applying S.s. to THP-1 cells over time resulted in a decrease in SQSTM1 levels and an increase in LC3-II levels. This suggests that live P. g. and heat-inactivated P. g. inhibited autophagy, whereas S.s. appeared to promote it (Figure S3I). We also conducted immunoblotting and ELISA analysis to evaluate how autophagy impairment affects IL1B maturation and release, using Baf A1 and the autophagy activator rapamycin. Our findings showed that Baf A1 markedly increased IL1B maturation and release in the P. g. group, whereas rapamycin reduced these processes (Figure S3K-L). Overall, these results indicate that prolonged stimulation with P. g. inhibits autophagy at the autophagosome-lysosome fusion step in THP-1 cells, leading to enhanced secretion and maturation of IL1B.

Autophagosome-lysosome fusion is blocked by decreased active form RAB7A

Previous research has shown that autophagosomes and lysosomes are transported to the perinuclear region, where they fuse to form autolysosomes for degrading their contents [44]. This perinuclear trafficking of lysosomes is regulated by the RAB7A GTPase, which alternates between its active GTP-bound state and inactive GDP-bound state [45]. To investigate whether the autophagosome-lysosome fusion was promoted by the active form of RAB7A, we tested the levels of active form GTP-RAB7A using a specific anti-GTP-RAB7A antibody to affinity isolate the GTP-RAB7A protein. We observed a significant increase in the ratio of active GTP-RAB7A to total RAB7A protein in the P. g. stimulation group (Figure 4A,B). Additionally, we transfected cells with GST-RILP plasmids and used GST antibodies to affinity isolate RILP-binding proteins to measure GTP-RAB7A (RILP-bound RAB7A) (Figure 4C). However, there was no significant difference in total RAB7A protein levels between the two groups, indicating that P. g. stimulation led to RAB7A inactivation rather than altering its expression in THP-1 cells. We also observed a similar phenomenon in THP-1 cells overexpressing RAB7A, where P. g. stimulation resulted in a decrease in GTP-RAB7A levels (Figure S4A). Active GTP-RAB7A regulates lysosomal trafficking to the perinuclear region by interacting with RILP [27]. Subsequently, we observed a significant reduction in the overlap of RAB7A and RILP in P. g.-stimulated cells, as shown in immunostaining images, suggesting that the RAB7A-RILP interaction was inhibited (Figure 4D,E). Consistent with earlier findings on RAB7A’s role in autophagosome-lysosome fusion, we observed changes in RAB7A distribution during P. g. stimulation. Initially, RAB7A colocalized with lysosomes near the cell membrane. Between 1 and 3 h post-stimulation, the signal became diffuse in the cytoplasm, and by approximately 6 h, RAB7A was predominantly aggregated around the cell periphery area (Figure S4B, Video S2). Additionally, an increase in the ratio of membrane-bound RAB7A to cytosolic RAB7A was observed during P. g. stimulation, indicating peripheral aggregation of RAB7A (Figure S4C, D).

Figure 4.

Autophagosome-lysosome fusion is blocked by decreased GTP-RAB7A. (A) Western blot analysis of total RAB7A and GTP-RAB7A (active form, immunoprecipitated by anti-GTP-RAB7A antibody) in control vs. P. g-stimulated THP-1 cells (MOI = 100, 8 h). One representative blot is shown in three independent experiments. Anti-RAB7A (cell signaling Technology,95746), anti-GTP-RAB7A (NewEast 26,923). (B) quantity analysis of the ratio of GTP-RAB7A expression to total RAB7A. (C) Western blot analysis of co-ip from THP-1 cells transfected and treated as indicated. RAB7A pulled down by GST-RILP represents GTP-RAB7A. (D) confocal microscopy analysis for colocalizations of RAB7A with RILP in control vs. P. g-stimulated THP-1 cells (MOI = 100, 8 h). RAB7A (red) and RILP (green) were stained using anti-RAB7A and anti-rilp antibodies. Scale bars: 10 μm. (E) colocalization plots report Mander’s overlap quantified from multicell images (black dots). (F) Western blot analysis of SQSTM1, LC3, total RAB7A and GTP-RAB7A (immunoprecipitated by anti-GTP-RAB7A antibody) in four groups as indicated. CID-1067700 (100 μM), ML098 (100 nM), baf A1 (50 nM). (G) GFP-RFP-LC3 THP-1 cells were treated with P. g. (MOI = 100), ML098 (100 nm), and CID-1067700 (100 μM) for 8 h and the colocalization of GFP-LC3 and RFP-LC3 was captured to detect the autophagy. Red dots represent autolysosomes and yellow dots autophagosomes. Scale bars: 10 μm. (H-I) the number of GFP⁺ RFP⁺ LC3 (autophagosome) and to GFP⁻ RFP⁺ LC3 (autolysosome) was calculated from multicell images. One representative blot is shown in two independent experiments. Data were presented as mean ± SEM. p-values were calculated through two-tailed Student’s t-tests (B, E) and one-way ANOVA test (H, I).

We then applied the RAB7A activator ML098 and the RAB7A inhibitor CID-1067700 to modulate GTP-RAB7A levels for 8 h in a dose-dependent manner. We observed that the level of SQSTM1 decreased with increasing concentrations of ML098, whereas SQSTM1 levels were elevated in the CID-1067700 group. However, LC3-II formation remained unaffected in both groups, confirming that RAB7A regulated the fusion of autophagosomes with lysosomes rather than the initiation of the autophagic pathway and autophagosome formation (Figure 4F, Figure S4E-F). Additionally, we assessed autophagy flux in four groups: the control group, P. g. stimulation group, P. g. combined with ML098 group, and CID-1067700 group. Compared with the control group, the number of autophagosomes per cell was significantly increased in both the P. g. stimulation and CID-1067700 group, while ML098 decreased the number of autophagosomes per cell, indicating that autophagy was promoted by ML098 and inhibited by CID-1067700 (Figure 4G–I).

We also performed immunoblotting and ELISA analysis to assess how ML098 and CID-1067700 affect IL1B maturation and release (Figure S4G-J). ML098 significantly reduced IL1B maturation and release in P. g.-stimulated cells, while CID-1067700 increased both IL1B maturation and release. Overall, these results indicate that in P. g.-stimulated THP-1 cells, the decrease in the active GTP-bound form of RAB7A leads to impaired autophagy and altered lysosomal trafficking, which in turn enhances the secretion and maturation of IL1B.

USP4 promotes autophagosome-lysosome fusion by deubiquitination of RAB7A

To investigate the mechanism behind the reduction of active RAB7A in P. g.-stimulated THP-1 cells, we isolated lysosomal membrane proteins from both control and P. g.-stimulated THP-1 cells by precipitating VAMP8, a lysosomal membrane protein, for proteomic analysis. Interestingly, the deubiquitinating enzyme USP4 was found in the lysosomal interacting proteins (Figure 5A). Recent studies suggest that RAB7A ubiquitination plays a key role in regulating late endosomal transport and recycling [34]. Through co-immunoprecipitation (co-IP) and rigid protein-protein docking, we observed an interaction between USP4 and RAB7A in THP-1 cells (Figure S5A; Figure 5B,C). We then evaluated USP4 expression and the ubiquitination of RAB7A in response to P. g. stimulation in both normal and USP4 knockdown (KD) THP-1 cells. Compared to the control group, USP4 expression was reduced, and RAB7A ubiquitination was increased in both the P. g. stimulation and USP4 KD groups (Figure 5B, Figure S5B). We also performed in vitro assays to examine the polyubiquitination of RAB7A in response to P. g. stimulation. The results showed a decrease in K63-specific polyubiquitination of RAB7A, which was evidenced by an increased presence of RAB7A protein bands at 23 kD on the gel (Figure S5C). Additionally, co-treatment with the proteasome inhibitor MG132 and the protein synthesis inhibitor cycloheximide indicated that RAB7A degradation does not proceed via the proteasomal pathway (Figure S5D-E).

Figure 5.

USP4 promotes autophagosome-lysosome fusion by deubiquitination of RAB7A. (A) the whole lysates were extracted from THP-1 cells with or without P. g. stimulation. Lysosomal interacting proteins were immunoprecipitated and were subject to mass spectrometry. The data revealed multiple USP4 peptides in lysosome-bound protein samples, indicating the interaction between RAB7A and USP4. (B, C) Western blot analysis of Co-ip from THP-1 cells transfected and treated as indicated. (B) THP-1 cells were transfected with siRNA to knock down the USP gene. Ubiquitin-conjugated proteins were pulled down using anti-ub antibody, followed by Western blot analysis using anti-RAB7A antibody. (C) THP-1 cells were transfected with USP4 overexpression lentivirus and used GTP-RAB7A and USP4 antibodies to pull down GTP-RAB7A-binding and USP4-binding proteins to measure RAB7A (GTP-RAB7A and USP4 bounding RAB7A. (D) Representative confocal images of RAB7A with lysosome marker LAMP1. Scale bars: 10 μm. (E) LAMP1 fluorescent intensity distribution is expressed as distance along a straight line from the center of the nucleus to the peripheral area. Fluorescence intensities were quantified along a straight line extending from the center of the cell’s nucleus (designated as fractional distance = 0) to the plasma membrane (designated as fractional distance = 1.0). (F) GFP-RFP-LC3 THP-1 cells were transfected with USP4 siRNA and USP4 overexpressed lentivirus to detect autophagy flux. Representative images were captured to display the colocalization of GFP-LC3 and RFP-LC3 after being treated with P. g. (MOI = 100, 8 h). Red dots represent autolysosomes and yellow dots autophagosomes. Scale bars: 10 μm. (G-H) the number of GFP⁺ RFP⁺ LC3 (autophagosome) and to GFP⁻ RFP⁺ LC3 (autolysosome) was calculated from multicell images. Data were presented as mean ± SEM. p-values were calculated through one-way ANOVA test.

To determine if RAB7A ubiquitination affects its role in lysosomal trafficking, we measured LAMP1 fluorescence intensity along a line from the nucleus to the cell periphery. We observed an increased peripheral LAMP1 localization in cells upon P. g. stimulation. In contrast, LAMP1 peripheral localization was reduced in USP4 overexpressing cells, indicating that downregulation of USP4 enhances RAB7A-mediated lysosomal trafficking (Figure 5D,E, Figure S5F-G). We measured autophagy flux in four groups: control, USP4 knockdown (KD), P. g. stimulation, and USP4 overexpression (OE) combined with P. g. stimulation. We observed a significant increase in the number of autophagosomes in the USP4 KD group compared to the control cells. Conversely, the number of autolysosomes per cell increased significantly in the USP4 OE group during P. g. stimulation, suggesting that USP4 overexpression enhances autophagic flux (Figure 5F,H). Additionally, we observed a significant increase in IL1B release in USP4 KD cells and a decrease in USP4 OE cells (Figure S5H). In summary, these findings showed that reduced USP4 inhibits autophagosome-lysosome fusion and IL1B release. This effect appears to be mediated by the RAB7A ubiquitination and lysosomal peripheral trafficking.

Decreased USP4 is correlated with enhanced IL1B expression in periodontitis

To confirm the association of USP4 with periodontitis pathogenesis, we collected clinical gingival tissues from healthy donors and severe periodontitis patients for immunostaining and RNA-Seq analysis. Immunostaining revealed that USP4 protein levels were significantly lower in the periodontitis group compared to the healthy group, while IL1B levels were higher (Figure 6A). The correlation heatmap was used to illustrate significant statistical correlation values (R) between autophagy-related genes and inflammation-related genes in human gingival tissues. A correlation heatmap showed a significant negative correlation between USP4 expression and IL1B expression in human gingival tissues (Figure 6B). Similar trends were observed in periodontitis mice, where USP4 expression was downregulated in the periodontitis group compared to the control group (Figure 6C). Finally, we achieved local overexpression of USP4 in mouse gingiva using adeno-associated virus/AAV, thereby promoting localized USP4 overexpression. And inhibited RAB7A activity through intraperitoneal injection of CID-1067700. By employing these approaches to respectively enhance mouse USP4 expression and suppress RAB7A activity, we investigated their impact on periodontitis development in mice. The results demonstrated that USP4 overexpression attenuated alveolar bone loss in periodontitis mice, while CID-1067000 exacerbated alveolar bone resorption (Figure 6D). These findings, along with observations on RAB7A deubiquitination, support a model where USP4-mediated deubiquitination of RAB7A promotes autophagosome-lysosome fusion (Figure 7A). In contrast, P. g. infection downregulates USP4, leading to RAB7A ubiquitination, which impairs autophagy and exacerbates periodontitis (Figure 7B).

Figure 6.

Decreased USP4 and enhanced IL1B expression are found in periodontitis. (A) immunofluorescence analysis of USP4, IL1B in lamina propria of healthy and periodontitis gingiva (n = 3 for each group). Scale bars: 50 μm; zoom: 20 μm. (B) Spearman correlation for the autophagy related genes (MAP1LC3A, MAP1LC3B2, SQSTM1, RAB7A), inflammation relation genes (IL1B), and USP4 across clinical healthy and periodontitis gingiva samples. The data was obtained from rna-seq analysis of the total RNA extracted from clinical human gingival tissue lysates (n = 5 in the healthy group, n = 4 in the periodontitis group). (C) the immunofluorescence analysis of USP4 in control group mice and periodontitis mice (n = 5 for each group). Scale bars: 50 μm. Zoom, 10 μm. (D) Representative micro-ct analysis of alveolar bone in normal control mice and periodontitis mice treated with ML098 (1 mg/kg),CID-1067700 (10 mg/kg), and USP4 overexpression adenovirus. Scale bars: 500 μm. Data were presented as mean ± SEM. p-values were calculated through two-tailed Student’s t-tests (A, C) and one-way ANOVA test (D).

Figure 7.

Model of RAB7A ubiquitination-mediated impaired autophagosome-lysosome fusion in periodontitis pathogenesis. Research Paperhe periodontitis condition (right panel): in the absence of USP4, lack of RAB7A deubiquitination (GDP-RAB7A form) results in impaired autophagosome-lysosome fusion. The impaired autophagy leads to increased IL1B maturation and release, which contributes to periodontitis pathogenesis.

Discussion

RAB7A has recently garnered attention due to its crucial role in regulating autophagy. This study aimed to investigate the impact of RAB7A inactivation on autophagic dysfunction and periodontal disease and to identify potential therapeutic targets. The rationale of this study stemmed from our observations that decreased RAB7A expression alongside increased SQSTM1 and LC3 proteins in gingival tissues derived from patients with severe periodontitis, indicating impaired autophagic flux. In line with these findings, experimental validation in periodontitis mouse models demonstrated that RAB7A inactivation and autophagy impairment were indeed associated with exacerbated alveolar bone loss. Moreover, We examined RAB7A’s role in lysosomal trafficking and autophagosome-lysosome fusion, finding that these processes were impaired due to RAB7A ubiquitination. Additionally, we discovered that USP4 insufficiency leads to increased RAB7A ubiquitination and confirmed reduced USP4 expression in vivo. In summary, our findings collectively suggest that reduced USP4 levels mediate RAB7A ubiquitination, disrupting lysosomal trafficking and autophagosome-lysosome fusion, which contributes to the development and progression of periodontitis.

As evidenced by numerous studies, autophagy plays a dual and complex role in the preservation of periodontal homeostasis [46–48]. Our research underscores its protective effect against periodontitis. We found that Baf A1 administration worsened alveolar bone loss in periodontitis mouse models (Figure S2B-E), consistent with previous studies where autophagy inhibition with 3-methyladenine (3-MA) also led to increased bone loss in periodontitis mice [48]. Moreover, enhanced autophagic activity has been shown to reduce inflammatory cell infiltration and alveolar bone loss in periodontal ligament cells, highlighting autophagy’s protective role in periodontitis progression [47]. However, our research revealed a seeming paradox: both increased SQSTM1 and LC3 expression was detected in clinical samples from severe periodontitis patients, experimental periodontitis mice, and P. g.-stimulated cellular models, which contrasts with the traditional view of either fully enhanced or fully suppressed autophagy in periodontitis [47,48]. This discrepancy may be due to autophagic role varying with disease severity. Supporting this, previous research showed that BECN1 and LC3 levels increased in periodontitis mice with ligation for 3 and 7 days but decreased in those with 10 and 14 days of ligation [47]. This suggests while autophagic activity may initially rise, it declines as periodontitis progresses, indicating a reduction in autophagic efficiency over time. These results highlight the complex relationship between autophagy and periodontitis, emphasizing the need for understanding the dynamic changes in autophagic processes at different stages of the disease.

In our study, the findings suggested that the fusion step of autophagosomes and lysosomes rather than the formation of autophagosomes was blocked in periodontitis (Figures 1B, 3A–F, Figure S3A). Given RAB7A’s role in regulating autophagosome-lysosome fusion, it is important to consider its impact on autophagy and periodontitis. Recent research has shown that RAB7A deficiency disrupts endothelial function in pulmonary hypertension and impairs sensory neuron function in Charcot-Marie-Tooth Type 2B disease [29,30]. Our results support these findings, showing that reduced RAB7A expression contributes to the development of periodontitis (Figures 1B, 2B–E, Figure S1B,D). Additionally, we found that RAB7A-mediated lysosomal trafficking was abnormal, aligning with reduced autophagosome-lysosome fusion in P. g.-stimulated THP-1 cells. This provides further evidence that RAB7A dysfunction contributes to increased inflammation in periodontitis. However, we observed that while RAB7A expression was reduced in periodontitis samples, the total RAB7 protein levels were not significantly decreased in P. g.-stimulated THP-1 cells. This discrepancy may be due to two main factors: (a) The 8 h P. g. stimulation (MOI = 100) of THP-1 cells does not fully replicate the chronic and complex nature of clinical periodontitis, which involves prolonged exposure to a variety of pathogens. Therefore, the brief exposure to P. g. may not be sufficient to affect the transcription and translation of RAB7A in THP-1 cells. Additionally, similar findings have been reported where RAB7A expression is downregulated in bacterial chondronecrosis with osteomyelitis [49]. In some studies involving virus-infected cells over 10 days, total RAB7A expression remains unchanged, though RAB7A activity is reduced [50]. This suggests that prolonged infections, rather than short-term ones, are more likely to lead to a decrease in RAB7A expression in vivo. (b) Different cells: The clinical gingival samples include a variety of cell types such as epithelial cells, endothelial cells, fibroblasts, and immune cells, leading to a combined effect on RAB7A expression. While macrophages account for only a part of gingival tissues. Consequently, changes in RAB7A expression in clinical samples may differ from those observed in P. g.-stimulated macrophages. In summary, the initial phase of our study provided verification that impaired RAB7A function disrupts autophagic processes, leading to autophagy impairment and ultimately contributing to the pathogenesis of periodontitis.

We think that the ability of P. g. to inhibit autophagy is linked to its virulence factors. Our study demonstrated both live P. g. and heat-inactivated P. g. stimulation resulted in increased levels of SQSTM1 and LC3-II in THP-1 cells (Figure S3H). However, compared to live P. g., inactivated P. g. exhibits a slightly weaker inhibitory effect on autophagy, which is due to the loss of functionality in some virulence factors. In heat-inactivated P. g., virulence factors such as fimbriae, gingipains, outer membrane proteins, and the type IX secretion system lose their functionality. Research has reported that P. g. gingipain can cleave VAMP8, thereby inhibiting autophagy in neonatal rat cardiomyocytes [18]. The loss of function of the virulence factor gingipain in heat-inactivated P. g. reduces its ability to inhibit autophagy. However, heat-inactivated P. g. retains some virulence factors, including LPS and the capsule, which are heat-resistant and continue to exhibit toxicity. Other studies have also found that LPS can inhibit autophagic flux in THP-1 cells in a time- and dose-dependent manner [51]. In summary, we propose that P. g. inhibits autophagic flux in THP-1 cells through the combined effects of several virulence factors.

Following the above-discussed findings, the second phase of our study delves into the exploration of the affecter influencing RAB7A activity and function. Active RAB7A is known to recruit RILP, which in turn attracts dynein-dynactin motor complexes that facilitate lysosome transport toward the perinuclear region (minus-end transport) of the cell [52–54]. This fundamental role was supported by our observation of reduced co-localization between RAB7A and RILP in P. g.-stimulated cells (Figure 4A–E). Additionally, the transport of the autophagosome from the cell periphery to the perinuclear region is essential for its fusion with the lysosome [28,55,56]. In line with this, we observed an initial aggregation of RAB7A and lysosomes at the perinuclear region within the first two h of P. g. stimulation (less than 2 h). However, after more than 4 h of stimulation, RAB7A and lysosomes accumulated in the peripheral areas, coinciding with a block in autophagosome-lysosome fusion. This pattern indicates that autophagic activity was initially enhanced but later impaired as RAB7A transitioned from the perinuclear region to the cell periphery (Figure S4B-D, Video S2). The findings are consistent with earlier studies that RAB7A regulates the aggregation and fusion of late endocytic structures with lysosomes, maintaining the perinuclear lysosome compartment [57,58]. Additionally, RAB7A governs the minus-end transport of autophagosomes via the RAB7A – RILP – dynein complex, highlighting its role in regulating autophagosome-lysosome fusion through interaction with RILP [59]. Collectively, these observations elucidate how disruptions in RAB7A dynamics lead to impaired autophagic flux and contribute to the progression of periodontitis.

Recent evidence indicates that the function of RAB7A is regulated by ubiquitination, which impacts its interaction with RILP [31,32,34]. Ubiquitination of RAB7A has been shown to be mediated by PRKN/parkin (an E3 ubiquitin ligase) and USP32, both of which target lysine K38 on RAB7A [32,34]. PRKN-mediated ubiquitination of RAB7A appears to stabilize RAB7A and promote its interaction with RILP [32]. Another study goes exactly counter to the idea and they suggest RILP prefers ubiquitination-deficient RAB7A, and the ubiquitination of RAB7A inhibits late endosomal transport toward the perinuclear area, resulting in dispersion and swelling of the late compartment [34]. Our findings align more closely with the latter hypothesis involving USP32, where we observed that RAB7A ubiquitination impairs normal lysosomal trafficking and autophagic activity. This might be due to the physiological cycling of RAB7A between ubiquitination and deubiquitination. Both excessive and insufficient ubiquitination levels can impair RAB7A’s function. Non-ubiquitinated RAB7A facilitates efficient perinuclear transport of late endosomes and lysosomes, whereas ubiquitinated RAB7A may alter its functions, such as recycling from late endosomes [34].

The role of RAB7A ubiquitination in periodontitis progression has not been previously explored. Our study is pioneering in identifying USP4 as a key enzyme responsible for deubiquitinating RAB7A. We found that deleting USP4 impairs RAB7A’s function in mediating lysosomal trafficking to the perinuclear region and disrupts autophagy flux. These findings enhance our understanding of how RAB7A ubiquitination contributes to the development of periodontitis.

Our research offers new insights into the complex role of RAB7A in periodontal health and disease. However, it is important to acknowledge the limitations of this work. One major limitation is that the full spectrum of RAB7A ubiquitination remains unexplored. Future studies should investigate whether other deubiquitinating enzymes or ubiquitin ligases influence RAB7A and identify the specific ubiquitination sites involved. Additionally, the mechanism by which P. g. reduces USP4 expression or why USP4 is downregulated in periodontitis remains unclear. Although USP4 downregulation has also been observed in virus-infected cells, its interaction with pathogens is not yet fully understood [60,61]. Studies suggest that USP4 expression may be regulated by transcriptional factors, including microRNAs and DNA methylation [62–65]. Notably, increased DNA methylation was observed in the gingiva of mice infected with P. g. in a periodontitis model [66]. This suggests that USP4 downregulation in the P. g. infection model might be mediated by transcriptional regulation.

In conclusion, our findings demonstrated that reduced USP4-mediated ubiquitination of RAB7A disrupts normal lysosomal trafficking dynamics, thereby impairing autophagosome-lysosome fusion and contributing significantly to the progression of periodontitis pathogenesis. These findings highlight the RAB7A-USP4 interaction as a promising target for potential therapeutic interventions in periodontal disease.

Materials and methods

Mice

C57BL/6 mice were obtained from the Experimental Animal Center (Shanghai SLAC Laboratory Animal Co., Ltd., China). The mice were bred in single-sex cages at 23 ± 3°C and 40–70% humidity under specific pathogen-free conditions with a 12-h light-dark cycle.

Mice (6–8-week-old, 18–22 g) were subjected to an experimental periodontitis model. Mice received a 5–0 silk ligature pre-moistened with P. g. suspension (10⁹ colony-forming units, 24 h) around their maxillary second molars. The ligatures were re-applied with the bacterial suspension every two days [40]. For drug treatment groups, animals were intraperitoneally injected with ML098 (1 mg/kg; MedChemExpress, HY-19800), CID-1067700 (10 mg/kg; MedChemExpress, HY-13452), or Baf A1 (0.1 mg/kg; MedChemExpress, HY-100558) every 2 days to maintain therapeutic concentrations. Mice were ear-tagged to enable identification. Following a 10-day experimental period, mice were euthanized and their maxillary jaws were carefully harvested for further analysis.

Micro-ct analysis

Each maxilla sample was rinsed in sterile saline solution and fixed in a 4% neutral buffered formalin solution for 24 h. Fully fixed maxilla samples were used for micro-CT scans [67] (SCANCO Medical AG, Switzerland) to determine the distances between the cementoenamel junction and the alveolar bone crest (CEJ-ABC), bone volume:tissue volume ratio (BV:TV), and bone mineral density/BMD at the buccal of maxillary second molar by using image analysis software [68].

Histological analysis

Clinical gingiva tissues were collected from severe periodontitis patients (Stage IIIC) and healthy donors. Mice gingiva tissues were collected from the maxillary second molar of mice. All tissues were fixed in 4% paraformaldehyde for 24 h. The mice specimens needed to be decalcified in 10% EDTA solution for 4 weeks. As for immunohistochemical detection, the sections were incubated with primary antibodies overnight at 4°C and followed by secondary antibodies at 37°C for 30 min. Freshly prepared DAB solution was added and nuclei were counterstained with hematoxylin. The slices were scanned by Olympus Digital Slide Scanner (Olympus VS200, Japan), and five images were randomly intercepted from each sample and analyzed by Image-Pro Plus 6 Software for the average positive-stained areas. As for immunofluorescence staining, the sections were incubated with CoraLite488/594 conjugated secondary antibodies (Proteintech, SA00013) at 37°C for 2 h and nuclei were stained with DAPI. The slices were scanned by Zeiss Laser Scanning Confocal Microscope (Zeiss LSM980, Germany), and five images were randomly intercepted from each sample and were analyzed by Fiji Software for the integrated fluorescence intensity per area.

RNA sequencing (rna-seq) analysis

Clinical human gingiva tissues were collected from severe periodontitis patients (stage IIIC) and healthy individuals. Total RNA was extracted using Trizol reagent (ThermoFisher, 15596018CN) following the manufacturer’s procedure. The total RNA quantity and purity were analyzed by Bioanalyzer 2100 and RNA 6000 Nano LabChip Kit (Agilent, 5067–1511), and high-quality RNA samples with RIN number > 7.0 were used to construct the sequencing library. We performed the 2 × 150bp paired-end sequencing (PE150) on an Illumina NovaseqTM 6000 (LC-Bio Technology CO., Ltd., Hangzhou, China) following the vendor’s recommended protocol. Bioinformatic analysis was performed using the OmicStudio Tools at https://www.omicstudio.cn/tool. The data has been deposited into the GEO database (GSE273165).

Single-cell rna-seq analysis

We analyzed single-cell RNA sequencing datasets from a biorepository previously published (NCBI GEO database under accession GSE164241) [38]. The dataset included 13 samples from individuals with periodontitis and 8 from health controls. Employing the standard Seurat workflow, we conducted quality control, normalization, and dimensionality deduction. Cluster was identified based on marker genes outlined in the supplementary file. For the four major cell types, pseudobulk differential expression analysis was conducted to identify differentially expressed genes/DEGs between two groups. Normalized data were then extracted for t-test. Macrophages were specifically chosen for further investigation, and pseudobulk differential expression analysis was applied again. Volcano plots and box plots were generated using the ggplot2 R package, while GO enrichment analysis utilized the clusterProfiler R package.

Cell lines and transfection

The human THP-1 cell line was purchased from Procell Life Science&Technology Co., Ltd. (CL-0233) and cultured with their specialized media (Procell, CM-0233). GFP-LC3 THP-1, RFP-GFP-LC3 THP-1, and Flag-RAB7A cell lines were obtained from Genomeditech (GM-C33998). The cultures were maintained at 37°C in a humidified incubator with 5% CO₂. Phorbol 12-myristate 13-acetate (PMA, 100 ng/ml; MULTI SCIENCES, CS0001) was used to promote their differentiation into macrophages. After 24 h of incubation and 24 h of break, differentiated THP-1-derived macrophages were obtained and used for subsequent experiments.

USP4 siRNA (5′- AACATGTCCGAGTTTGTCTGT- 3′) and its negative control (5′-UUCUCCGAACGUGUCACGUTT- 3′) were designed and constructed by HanBio Technology (Shanghai, China). For siRNA transfections, each 1 × 10⁶ cells (6-well plate) were stimulated by a mixture of 4 μg siRNA, 125 μL high glucose DMEM medium (ThermoFisher 11,965,092), and 4 μL Lipo8000 transfection reagent (Beyotime, C0533). After 2 days of culture, the transfection efficiency was checked by western blots.

USP4 overexpression lentivirus and its negative control (lentivirus vector GV492) were designed and constructed by GENECHEM (Shanghai, China). THP-1 cells were transduced with the lentiviral particles at a MOI of 100:1 for 2 days. The transduction medium was then replaced with a fresh medium containing puromycin (1 μg/mL) for 2 days. The transfection efficiency was checked by western blots.

P. g. strain and cell stimulation

P. g. (ATCC 33,277) was cultured in trypticase soy broth (TSB; ATCC, MD112) supplemented with 5 μg/mL hemin (Solarbio, H8132) and 1 μg/mL vitamin K1 (Solarbio, V8151), and cultured in an anaerobic condition (80% N₂, 10% H₂, and 10% CO₂) at 37°C. The fresh bacterial suspension was centrifuged and resuspended using RPMI 1640 medium (ThermoFisher, C11875500BT) and used within 8 h. The optical density/OD at 630 nm was used to measure the concentration of P. g. suspension (1 OD = 1 × 10⁹ colony-forming units/mL). The fresh live P. g. was added to the medium for co-culturation with THP-1-derived macrophages and the MOI was calculated by the ratio of P. g. to THP-1 cells.

Immunofluorescence staining and time-lapse images

For fluorescence confocal microscopy of fixed samples, cells were seeded into 12-well plates containing glass coverslips. After treatment, the cells were fixed with 4% paraformaldehyde for 20 min and permeabilized with Triton X-100 (Beyotime, P0096) solution (0.01% in PBS [Biosharp, BL302A]) for 20 min. After blocking in 5% BSA (Solarbio, SW3015) for 1 h, the cells were incubated with primary antibodies overnight at 4°C. Next, excess primary antibodies were removed and the cells were incubated with fluorescein-conjugated secondary antibodies for 1 h at room temperature in the dark. Finally, the cells were mounted with a DAPI-containing mounting medium (Invitrogen, S36939). Images were taken using a Zeiss Laser Scanning Confocal Microscope with 63× oil immersion objectives (Zeiss LSM980, Germany). The colocalization was evaluated by Mander’s overlap coefficient. Fractional distances were calculated as follows: fluorescence intensities (above automated background threshold) were measured along the longest straight line drawn from the center of a cell’s nucleus to the plasma membrane using the line profile tool in the LAS-AF Software [34].

For fluorescence confocal microscopy of live samples, cells were seeded into 35 mm glass bottom dishes and transfected as indicated. Samples were incubated with LysoTracker Red (1:10,000; Beyotime, C1046) at 37°C 1 h before imaging. The samples were cultured at 37°C with 5% CO₂, and time-lapse images were taken every 5 min using 20× objectives. The fluorescent intensity was analyzed by ImageJ Software.

Protein extraction

For protein extraction from lysates, a mixture of RIPA lysis and extraction buffer (ThermoFisher 89,900) containing phenyl methane sulfonyl fluoride (Biosharp, BL507A) was used to homogenize cells for 30 min at 4°C. The samples were then centrifuged at 15,000 rcf for 10 min and the supernatant was retained.

For protein extraction from the supernatant, methanol and chloroform were mixed with cell culture supernatant in a 2:1:1 ratio for extraction and then centrifuged at 15,000× g for 5 min. The precipitates were washed with methanol, centrifuged at 15,000× g for 5 min, and then dried at 37°C until methanol was completely volatilized. The precipitates were finally dissolved in RIPA lysis buffer with PMSF.

For protein extraction from the membrane and cytoplasm, the progress followed the instructions of the Membrane and Cytoplasmic Protein Extraction Kit (Biosharp, BL671B). Thirty μg per lane cell lysate proteins were analyzed by western blotting using the ChemiDoc MP imaging system (Bio-Rad, USA). Each data presented above is repeated for more than three independent experiments.

Immunoprecipitation

THP-1 cell lysates were extracted and incubated overnight at 4°C with protein A+G magnetic beads (Beyotime, P2108) lacking antibody coating to preclear nonspecifically bound proteins. Subsequently, after removing the beads without antibody binding from the cell lysates, antibody-coated magnetic beads were added and incubated with gentle shaking at room temperature for 2 h. Antibodies were coupled to protein A+G beads overnight at 4°C. Finally, the proteins were separated by magnetic frame and washed thrice with TBST buffer (Biosharp, BL602A), followed by adding 5× SDS sample buffer and heated for 10 min at 100°C. The primary antibodies include ubiquitin (Proteintech 10,201–2-AP), RAB7A (Cell Signaling Technology 95,746), GTP-RAB7A (NewEast 26,923), USP4 (Proteintech 66,822–1-Ig).

Western blots analysis

Equal amounts of protein from THP-1 cells were separated using sulfate – polyacrylamide gel electrophoresis (SDS-PAGE) gels. The separated proteins were transferred onto polyvinylidene difluoride. Membranes were then incubated overnight with primary antibodies and then incubated with a secondary antibody for 1.5 h, and the catalysis was promoted using SuperSignal West Pico PLUS Chemiluminescent Substrate (ThermoFisher 34,580).

ELISA analysis

For cytokine analysis, the IL1B in the supernatant was measured respectively using IL1B ELISA kits (Proteintech, KE00021) following the manufacturer’s instructions. The absorbance at 450 nm and 600 nm was measured and the protein levels were calculated according to the standard curve.

LC-MS/MS proteomics

Lysosomal protein was immunoprecipitated with anti-VAMP8 antibody (Proteintech 15,546–1-AP) from the control group THP-1 cells and P. g. (MOI = 100, 8 h) stimulated THP-1 cells. Protein samples were extracted from the lysates as mentioned above. The protein samples were from Bio-Tech Pack Technology Company Ltd for LC-MS/MS proteomics analysis (Beijing, China).

Antibodies

For western blot, the primary antibodies include anti-RAB7A (Cell Signaling Technology 95,746), anti-LC3A/B (Cell Signaling Technology 12,741), anti-SQSTM1 (Cell Signaling Technology 88,588), anti-USP4 (Proteintech 66,822–1-Ig), anti-BECN1/Beclin-1(Cell Signaling Technology, 4122), anti-MTOR (Cell Signaling Technology, 2972), anti-phospho-MTOR (Ser2448) (Cell Signaling Technology, 5536), anti-NLRP3 (Cell Signaling Technology 15,101), anti-cleaved-IL1B (Cell Signaling Technology 83,186), anti-ubiquitin (Proteintech 10,201–2-AP), anti-ATP1A1/ATPase Na+/K+ transporting subunit alpha 1 (Santa Cruz Biotechnology, sc -21,712), anti-ACTB/β-actin (Proteintech 60,008–1-Ig). The secondary antibody used in this study includes Goat Anti-Mouse IgG (Biosharp, BL001A) and Goat Anti-Rabbit IgG (Biosharp, BL003A).

For immunofluorescence, the primary antibodies used in this study include anti-RAB7A (Cell Signaling Technology 95,746), anti-RILP (Proteintech 13,574–1-AP), anti-LAMP1 (Cell Signaling Technology, 9091), anti-USP4 (Proteintech 66,822–1-Ig), anti-LC3A/B (Cell Signaling Technology 12,741), anti-SQSTM1 (Cell Signaling Technology 88,588).

For histological analysis, the primary antibodies used in this progress include, anti-LC3A/B (Cell Signaling Technology 12,741), anti-SQSTM1 (Cell Signaling Technology 88,588), anti-IL1B (Bioss, bs-0812 R), anti-cleaved-IL1B (Cell Signaling Technology 83,186).

For immunoprecipitation, the primary antibodies used in this progress include anti-RAB7A (Cell Signaling Technology 95,746), anti-GTP RAB7A (NewEast 26,923), anti-ubiquitin (Proteintech 10,201–2-AP).

Statistical analyses

A two-tailed unpaired student’s t-test was used to compare the differences between two groups after the determination of data distribution. For multiple group comparisons, one-way ANOVA test was performed. In cases of unequal variances or non-normal distribution, appropriate non-parametric tests and Kruskal-Wallis test were used. All data are presented as mean ± SEM. All statistical analysis and figures were performed using GraphPad Prism 9 Software. The degree of significance was indicated as: *p < 0.05; ** p < 0.01; *** p < 0.001.

Human subjects or animals

Ethical approval was obtained from Research Ethics Committee of the Affiliated Stomatology Hospital of Zhejiang University School of Medicine (Ethics Approval No. 2019-74 R). Animal related experiments were conducted according to a protocol approved by the Laboratory Animal Welfare and Ethics Committee of Zhejiang University, with the approval of the protocol under animal research project (ZJU20210178).

Funding Statement

This work was supported by National Natural Science Foundation of China [82170953, 82470973, 81991500, 81991502], the Science Fund for Distinguished Young Scholars of Zhejiang Province [LR23H140001], Key R&D Program of Zhejiang Province [2022C03088], and the Fundamental Research Funds for the Central Universities [226-2024-00202].

Disclosure statement

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

Data availability statement

All data are available in the manuscript and the supplementary materials are available upon request from the authors.

Supplemental material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15548627.2024.2429371