ABSTRACT Introduction Porphyromonas gingivalis, a Gram-negative anaerobe, is a key contributor to periodontal disease. Emerging evidence suggests a role for the P. gingivalis CRISPR-Cas system in disease progression, although the specific roles of its components remain unclear. Objectives Here we investigate the role of cas7, a Class 1 type I-B CRISPR-Cas system component, in P. gingivalis physiology and host interaction. Methods We compared P. gingivalis wild-type and ∆cas7 strains for growth, biofilm formation, oxidative stress resistance, and hemagglutination. Host interactions were assessed using THP-1 macrophage-like cells to evaluate intracellular survival and cytokine response. Dual RNA-seq enabled host and microbe transcriptomic profiling during cellular infection, and Galleria mellonella was used to assess virulence. Results The ∆cas7 mutant showed similar planktonic growth and biofilm formation compared to wild-type but was more sensitive to oxidative stress and had reduced hemagglutination. Although intracellular survival was unaffected, ∆cas7 altered the host cytokine production profile. Transcriptomic analysis revealed differential gene expression linked to oxidative stress and disease progression. In vivo, ∆cas7 infection led to a trend of increased larval mortality. Conclusion These findings reveal a previously unrecognized role for cas7 in modulating P. gingivalis virulence, offering new insights into CRISPR-Cas system functions in bacterial pathogenesis. Keywords: Porphyromonas gingivalis, CRISPR, periodontal disease, virulence Introduction Periodontal disease (PD) is a chronic progressive oral disease of the periodontium that leads to destruction of the support tissues of the teeth. This dysregulated immune response by the host against dysbiotic polymicrobial communities is elicited by the bacteria that reside in the subgingival plaque [[34]1]. Periodontitis is the sixth most prevalent disease worldwide and is the leading cause of tooth loss in adults [[35]2]. In healthy individuals, bacterial species colonising the periodontal tissues are held in check by a homeostatic immune response. However, if this ecosystem is disrupted, the microbial community becomes dysbiotic, which may result in the development of PD [[36]3,[37]4]. Porphyromonas gingivalis is a Gram-negative anaerobic bacterium closely associated with periodontitis [[38]5]. The success of P. gingivalis as a periodontal pathogen is thought to be directly related to its capacity to induce an imbalance between the host immune system and subgingival microbiome and survive in the harsh environment of the periodontal pocket under chronic inflammatory conditions [[39]4]. P. gingivalis is well adapted to this space and possesses an array of virulence traits, including resistance to oxidative stress, capsule, fimbriae, lipopolysaccharide (LPS), gingipains and others [[40]4]. Interestingly, this organism is known to release outer membrane vesicles, which deliver virulence-associated molecules to the local environment [[41]6,[42]7]. These features have been linked to periodontal tissue colonisation, biofilm formation, invasion and persistence in epithelial cells and immune system evasion. However, despite our current knowledge, the precise mechanisms of how P. gingivalis induces the development and progression of PD remain poorly understood [[43]8]. Resistance to oxidative stress is critical to the ability of P. gingivalis to survive in the periodontal pocket [[44]9]. There, the organism encounters numerous oxidative stress challenges, including the existing microbiome and as part of the immune response, inflammatory cells such as neutrophils and macrophages that are able to produce reactive oxygen species (ROS) [[45]10]. ROS are potent antimicrobial agents able to induce damage to bacterial nucleic acids, lipids and proteins [[46]10]. Known mechanisms used by P. gingivalis to resist oxidative stress include antioxidant enzymes, ruberythrin, hemin layer and DNA repair [[47]9,[48]11,[49]12]. The production of a hemin layer also provides protection against ROS, dependent on iron uptake through hemagglutination and is also directly correlated with the oxidative stress resistance of P. gingivalis [[50]9,[51]10,[52]12,[53]13]. Several clinical studies have aimed to clarify the complex host‒pathogen interactions that mediate PD. A meta-transcriptomic study of the microbial community from patients who were monitored clinically revealed that clustered regularly interspaced short palindromic repeats (CRISPR) class 1 type I-B-associated genes from P. gingivalis were among the most highly upregulated genes only in sites that clinically progressed to disease [[54]14]. Recently, a positive correlation between the expression of CRISPR-cas genes in Fusobacterium nucleatum and PD progression in patients was also reported [[55]15]. The CRISPR-cas system was initially described as a defence mechanism bacteria use against bacteriophages and exogenous nucleic acids [[56]16–18]. These systems use an array of CRISPR small RNA (crRNA) consisting of repetitive sequences with unique spacers to recognise their targets and Cas proteins to mediate targeted nucleic acid degradation [[57]19,[58]20]. Beyond the role of the CRISPR-cas system against bacteriophages and foreign nucleic acids, there is emerging evidence of its influence on the biological processes of bacteria, such as resistance against oxidative stress, virulence and evasion of the immune system [[59]16–18,[60]21,[61]22]. The type I-B CRISPR-cas system of P. gingivalis ATCC 33277 contains 119 spacers and repeats, followed by cas genes with distinct roles: cas1, cas2 and cas4 mediate the acquisition of new spacers; cas6 processes the expression of the CRISPR array transcript into mature crRNAs; and cas3, cas5, cas7 and cas8 form, together with the crRNA, an effector complex that recognises and interferes by degrading foreign nucleic acids [[62]23–25]. Previous work by our group identified that the deletion of the cas3 gene is correlated with an increase in the virulence of P. gingivalis in both in vitro and in vivo models [[63]18]. More recently, work from our group also found that a novel adhesin related to virulence is modulated by the CRISPR-cas system [64][26]. Besides cas3, the importance of other cas genes from the CRISPR system in virulence is poorly understood. However, the impact of other cas genes on the P. gingivalis fitness in the murine abscess model and epithelial cell colonisation environments has been described [[65]27]. Among these genes, cas7, a key agent in the nucleic acid degradation process of the CRISPR-cas system, revealed significant importance in the dynamics of in vivo and in vitro models. Although recent findings have identified the roles of the CRISPR-cas system in bacterial virulence, there is limited mechanistic understanding of the contribution of elements of this system in this context. In the present study, we investigated the impact of the deletion of the cas7 gene of P. gingivalis on the host‒pathogen interaction. Using a multifaceted approach that includes classical oxidative stress assays, cell infection, dual-transcriptomic analysis and virulence modelling, we demonstrate the influence that cas7 has on P. gingivalis virulence. Materials and methods Bacterial growth and culture P. gingivalis ATCC 33277 wild type (WT) frozen stocks were streaked onto blood trypticase soy agar plates supplemented with 5 µg/mL hemin and 1 µg/mL menadione (BAPHK) and incubated anaerobically at 37 °C for 5 days. Plate-grown P. gingivalis were collected and used to seed trypticase soy broths supplemented with 5 µg/mL hemin and 1 µg/mL menadione (TSBHK) and were incubated overnight anaerobically, centrifuged and used in the experiments as described below. For the growth and culture of the Δcas7 mutant (construction described below), BAPHK plates and TSBHK broths were supplemented with 10 µg/mL erythromycin. Growth curves Overnight cultures of P. gingivalis WT and ∆cas7 in TSBHK broth were adjusted to OD = 0.25–0.3 and incubated at 37 °C anaerobically. The growth of the cultures was monitored by measuring the OD[600] nm every 1 h and colony-forming units (CFU) in BAPHK of the cultures after 5, 10, 24 and 30 h. Construction of the ∆cas7 mutant The construction of a cas7 knockout strain (∆cas7) of P. gingivalis was done by replacing the entire gene with an erythromycin resistance cassette using a similar approach as we had done previously [[66]18,[67]28]. Briefly, a plasmid was constructed by cloning an erythromycin resistance cassette (ermF gene from pVA2198) flanked by the 1-kb region upstream and downstream of PGN_RS09280 (cas7, previously annotated as PGN_1962) into the multiple cloning site of pUC19 (pUC19_KOPGN_1962M) using the NEBuilder HiFi DNA assembly kit. The erythromycin cassette and the flanking regions were PCR amplified from the plasmid using a Pfu polymerase (Fermentas, Massachusetts, USA) following the manufacturer's protocol. The PCR product was used to electroporate P. gingivalis 33277 electrocompetent cells. After 5 days of growth on BAPHK + erm plates, colonies were selected, and PCR and sequencing confirmed the PGN_RS09280 gene knockout. The knockout strain stocks were stored frozen in broth media containing glycerol and dimethyl sulfoxide (DMSO) at −80 °C. A more detailed description of the mutant construction can be found in Supplementary material 1. Biofilm formation and biomass quantification Biofilms of P. gingivalis strains were cultivated in a chemically defined medium with tryptone and α-ketoglutarate (CDM) [[68]28], and biomass was quantified as described previously [[69]29]. In brief, plate-grown colonies of P. gingivalis were transferred to TSBHK broth and incubated anaerobically at 37 °C overnight. Following centrifugation of the overnight cultures, the pellets were then washed and adjusted to an OD[600] of 0.3 (approximately 1 × 10^7​​​​ CFU/mL) in CDM, where 200 μl was placed into the wells of a 96-well flat-bottom plate and incubated anaerobically at 37 °C for 48 h. For biomass analysis, the supernatant fluids of the mature biofilms were gently removed, the plates were dried at room temperature and stained with safranin (0.1%) at room temperature for 15 min. After washing with distilled water, safranin was solubilised with 100 µl 99% ethanol, and the plates were read at 492 nm using a spectrophotometer (Agilent, California, USA). Oxidative stress assays The oxidative stress assays were performed based on Boutrin et al. [[70]30]. In brief, overnight cultures of P. gingivalis WT and ∆cas7 in TSBHK broth were adjusted to OD = 0.25–0.3. The stress agent (0.25 mM H[2]O[2]) was added to the cultures, which were then incubated at 37 °C anaerobically. The growth was monitored by OD[600]nm and colony-forming units CFU in BAPHK of the cultures after 5, 10, 24 and 30 h. Untreated cultures served as controls. The assay was performed in triplicate. Hemagglutination assays The hemagglutination assay was performed based on Guo et al. [[71]31] with modifications. Overnight cultures of TSBHK broth were adjusted to 10^9 CFU/mL by centrifugation, and 50 µl serial 2-fold dilutions were placed in U-shaped 96-well plates. The control wells received 50 µl of phosphate-buffered saline (PBS). Next, 50 µl of 1% sheep red blood cells (Hemostat, California, USA) were added to all the wells, and the plates were incubated at 4 °C aerobically for 24 h. The agglutination of erythrocytes was observed, digital micrographs were taken, and the reciprocal of the maximum dilution multiple with positive agglutination of erythrocytes was recorded. The assay was performed in triplicate. P. gingivalis co-culture with THP-1 cells assays The human monocyte/macrophage cell line THP-1 (ATCC, Virginia, USA) was cultured at 37 °C in a 5% CO[2] incubator in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum, l-glutamine (2 mM), penicillin/streptomycin (100 U/100 μg/mL), HEPES (10 mM), sodium pyruvate (1 mM), glucose (4.5 mg/mL), sodium bicarbonate (1.5 mg/mL) and 2-mercaptoethanol (0.05 mM). For P. gingivalis infection assays, THP-1 cells were adjusted to 5 × 10^5 cells/mL and were treated with 100 ng/mL phorbol 12-myristate-13-acetate (PMA; Sigma-Aldrich, Missouri, USA) to induce differentiation into a macrophage-like state [[72]18]. These cells were then added to each well of 24-well cell culture plates and allowed to differentiate into a macrophage-like state for 48 h. At this point, the cell culture medium was replaced with antibiotic-free culture medium, and the cells were used in challenge studies [[73]32,[74]33] P. gingivalis WT and Δcas7 mutants harvested from overnight TSBHK broth culture were washed with RPMI 1640 media, adjusted to an OD[600] of 1.0 (approximately 1 × 10^9 CFU/mL), and added to PMA-activated THP-1 cells at a multiplicity of infection (MOI) 100, in complete antibiotic-free RPMI-1640 medium. After 2 h of coculture, the wells were washed with PBS to remove non-attached bacteria and incubated with complete RPMI-1640 media supplemented with gentamicin (300 μg/mL)/metronidazole (200 μg/mL) and incubated for 4, 6 and 24 h [[75]34]. P. gingivalis intracellular survival, cytokine and chemokine production assessments and dual-transcriptomics assays were performed temporally as described below. P. gingivalis intracellular survival assays To assess P. gingivalis intracellular survival, we first performed the antibiotic protection assay described above. After antibiotic treatment, the cells were washed to remove antibiotics, and THP-1 cells were lysed with sterile water for 15 min, then an equal volume of 2 × PBS was added to each well. Serial 10-fold dilutions of each lysate were made in PBS, triplicate samples were plated on BAPHK +/− erm plates, and the resulting colonies were counted (CFU/mL) after 5 days of anaerobic growth. THP-1 cell viability assay To assess the effect that P. gingivalis may have on the viability of THP-1 cells, we employed a CyQUANT™ XTT Cell Viability Assay (Invitrogen, Oregon, USA). We first performed the antibiotic protection assay described above in 96-well plates. After 24 h of antibiotic treatment, the cells' viability was measured following the manufacturer's instructions. Cytokine and chemokine production To understand the contribution of cas7 in inflammatory response, we performed an antibiotic protection assay as described above and collected cell supernatant fluids from the wells and stored them at −20 °C until use. The frozen samples were thawed, and the levels of human TNF-α, IL-1β, IL-6, IL-8, IL-10, RANTES, IFN- [MATH: γ :MATH] , CCL2/JE/MCP-1, CCL3/MIP-1-alpha, IL-17, GM-CSF and M-CSF were determined using a 12-plex kit (Biotechne, Minnesota, USA) following the manufacturer's instructions. Data were acquired on a Magpix Luminex system running xPONENT 3.1 software (Luminex, Texas, USA) and analysed using a five-parameter logistic spline-curve fitting method and Milliplex Analyst V5.1 software (Merck KGAa, Darmstadt, Germany). Dual transcriptomics In separate experiments, total RNA was extracted from THP-1 cells cocultured with either P. gingivalis WT, ∆cas7 for 2 h or 6 h using a mirVana RNA isolation kit (Life Technologies, New York, USA). Briefly, after removing the media from the wells, cells were incubated with 1 mL of RNAlater (Thermo Fisher, Pittsburgh, USA) for 5 min. Followed by the addition of mirVana lysis buffer, cells were harvested from the bottom of the wells using a cell scraper and transferred to tubes for cell lysis by bead-beating for 20 s three times at maximum speed with diethyl pyrocarbonate (DEPC)-treated zirconia‒silica beads (BioSpec Products, Oklahoma, USA). The extraction of RNA from the lysates was conducted as indicated by the manufacturer, following DNAse treatment with Invitrogen DNAse (RNAse free). The RNA samples were sequenced by the SeqCenter (Pittsburgh, USA), and data were analysed as described below. Library preparation was performed using Illumina's Stranded Total RNA Prep Ligation with Ribo-Zero Plus kit (Illumina, California, USA) and 10 bp unique dual indices (UDI). Sequencing was performed using an Illumina NovaSeq X Plus plataform, generating 150 bp paired-end reads. Demultiplexing, quality control and adapter trimming were performed with bclconvert 1. (v4.1.5). Bioinformatics pipeline The bioinformatics pipeline was executed using the HiperGator cluster at the University of Florida. Quality control of the raw sequences was performed using FASTQC [[76]35], which revealed consistent mean Phred scores of 36 for the forward and 35 for the reverse reads, indicating high overall quality. Adapter sequences were trimmed with Cutadapt [[77]36], and reads shorter than 100 nucleotides were discarded. Additionally, bases with Phred scores below 30 at the 5ʹ and 3ʹ ends were removed. The resulting trimmed reads were mapped with the Burrows–Wheeler Aligner (BWA) [[78]37] to the Homo sapiens GRCh38 primary reference genome (BioProject: PRJNA31257) first and unmapped reads were isolated and later mapped to the P. gingivalis ATCC 33277 reference genome ([79]https://www.ncbi.nlm.nih.gov/). Subsequent handling, sorting and file conversion of the alignment data were carried out using the Samtool suite [[80]38]. Transcript quantification across the reference genomes was completed with HTSeq-counts [[81]39]. The count files from all samples were consolidated into a matrix as input for statistical and clustering analyses in the R statistical package [[82]39]. Clustering was performed in three steps: t-SNE analysis using the Rtsne package [[83]40], principal component analysis (PCA) with the base ‘prcomp’ function and principal coordinate analysis (PCoA) using the ‘pco’ function from the labdsv package [[84]41]. Statistical analyses were conducted with the edgeR package [[85]42], and visualisations were created using base plotting functions and the ggplot2 package [[86]43]. Data availability Gene expression data have been deposited in the NCBI Gene Expression Omnibus (GEO) database ([87]https://www.ncbi.nlm.nih.gov/geo/) under GEO Series accession number [88]GSE300980. Galleria mellonella virulence assays The wax worms in the last instar larval stage were purchased from Carolina Biological Supply Company (North Carolina, USA). Upon arrival, healthy larvae with no signs of melanisation were selected, randomly separated by weight (0.2–0.3 g), placed in petri dishes with wood chips, and incubated at 4 °C overnight (n = 5 larvae/dish; n = 30 larvae/group). Bacterial inocula were prepared by adjusting P. gingivalis WT, Δcas3 (as in a previous study, Solbiati et al., 2020), or Δcas7 to approximately 10^9​​​​​ CFU/mL in TSBHK, as indicated above [[89]18]. Groups were injected with 5 μl of P. gingivalis WT, ∆cas3, or ∆cas7, approximately 10^6 CFU in the last left proleg using a Hamilton syringe. The control groups received TSBHK media alone or TSBHK medium + heat-killed bacteria (heat-killed by boiling at 100 °C for 10 min). After injection, Galleria were incubated at 37 °C in the dark, and survival was monitored for 5 days. Larvae were scored as dead when they displayed melanisation and no movement in response to touch [[90]18,[91]26]. Statistical analysis All experiments were performed at least twice on separate occasions. Statistical analyses were performed using GraphPad Prism (version 10 Software, Massachusetts, USA). Normality was determined with the Shapiro‒Wilk test. Statistical differences were assessed by an unpaired t-test or by one-way analysis of variance followed by Tukey's multiple comparison test (ANOVA, for multiple-comparison corrections). Kaplan–Meier survival curves were plotted for G. mellonella survival results, and estimations of survival differences were compared using a log-rank test. A p-value of ≤ 0.05 was considered statistically significant. Results Deletion of cas7 has no significant impact on in vitro growth and biofilm formation of P. gingivalis Previous studies indicate that cas7 plays a critical role in the fitness of P. gingivalis ATCC 33277 in experimental models of gingival keratinocyte invasion and murine abscess formation [[92]27]. Based on these findings, a P. gingivalis ATCC 33277 cas7 mutant was constructed to investigate the impact of cas7 on the virulence of P. gingivalis. Initial experiments were conducted to compare the planktonic growth of P. gingivalis WT and ∆cas7 in TSBHK media broth, at 37 °C under anaerobic conditions by absorbance measurement of the cultures and CFU/ml assays. Although a slight delay was noted for ∆cas7, no statistically significant differences were detected between the in vitro planktonic growth of the WT and ∆cas7 strains ([93]Figure 1A and [94]B). We next performed biofilm assays and found a trend of lower biomass by the ∆cas7 mutant; however, no statistical difference in biofilm formation between WT and ∆cas7 mutant was observed ([95]Figure 1C). Figure 1. [96]Figure 1. [97]Open in a new tab In vitro growth and biofilm formation of P. gingivalis WT and ∆cas7 strains. Growth curves of WT and ∆cas7 in TSBHK broth media at 37 °C under anaerobic conditions and measured by the absorbance (OD[600] (A)) and CFU/mL (B). Biomass of biofilms formed in CDM broth media at 37 °C under anaerobiosis for 48 h (C). Data are expressed as the mean ± SEM (n = 6 for growth curve experiments and n = 18 for biofilm of biological replicates). Deletion of cas7 impacts oxidative stress resistance and hemagglutination of P. gingivalis Resistance to oxidative stress is critical to the ability of P. gingivalis to survive in the periodontal pocket, as the organism likely encounters numerous oxidative stress challenges from the oral environment and locally from immune cells [[98]44]. The CRISPR-cas system has been reported to influence the bacterial response to oxidative stress [[99]20]. Therefore, we sought to understand if the deletion of cas7 affected the growth of P. gingivalis under H[2]O[2] oxidative stress. Strikingly, we found that Δcas7 was more susceptible to H[2]O[2] at 5 and 10 h time-points when compared to WT, supporting that the mutant is less resistant to oxidative stress ([100]Figure 2A and [101]B). Figure 2. [102]Figure 2. [103]Open in a new tab Oxidative stress resistance and hemagglutination assays of P. gingivalis WT and ∆cas7. WT and ∆cas7 were grown in TSBHK broth media or TSBHK broth media supplemented with 250 µM of H[2]O[2] at 37 °C under anaerobic conditions and measured by absorbance (A) and CFU/mL (B). Serial dilutions of P. gingivalis were incubated with 1% sheep red cells for 24 h at 4 °C aerobically (C). All the experiments were performed in triplicates. Data are presented as the means ± standard error of the mean (SEM), with significant differences between the WT and ∆cas7 groups at each time point analysed by t-test. (*) represents p < 0.05. Hemagglutination is appreciated to be part of the strategy used by P. gingivalis to acquire hemin from red blood cells [[104]45]. This hemin is then used to form a surface layer crucial for resisting oxidative stress. Therefore, we tested the capacity of WT and Δcas7 strains to hemagglutinate sheep red blood cells. We detected a more persistent capacity of WT to agglutinate red blood cells in lower concentrations (1:256, 1:512) when compared to Δcas7 ([105]Figure 2C). cas7 deletion did not impact the intracellular survival of P. gingivalis, but it affected proinflammatory cytokine and chemokines production in the early phase immune response of THP-1 cells To begin understanding the contribution of cas7 to P. gingivalis interaction with host cells, we added WT and the Δcas7 to PMA-activated THP-1 cells, collected the cell culture supernatant fluids for cytokine profiling by multiplex immunoassays, and assessed intracellular survival of P. gingivalis over 24 h using a classic antibiotic protection assay [[106]18]. We also assessed the viability of THP-1 cells. We did not detect any significant differences in THP-1 cell viability after 24 h of coculture with WT or Δcas7 (Supplemental Figure 1). Although we observed a trend towards reduced intracellular survival in the THP-1 cells of ∆cas7 compared to WT, this did not reach statistical significance ([107]Figure 3A). Figure 3. [108]Figure 3. [109]Open in a new tab P. gingivalis WT and ∆cas7 cocultured with THP-1 cells. Intracellular surviving P. gingivalis WT (blue) and ∆cas7 (red) in antibiotic-treated THP-1 cells measured by CFU/mL after 4, 6, 10 and 24 h (A). Temporal assessment of cytokine and chemokine profiles of THP-1 cells cocultured with P. gingivalis WT and Δcas7 (B). After 4, 6 and 24 h of incubation with P. gingivalis WT and ∆cas7 strains at MOI = 100, the cell culture supernatants fluids were collected and analysed using Luminex. The levels of TNF-α, IFN-γ, IL-17, CCL3-MIP-α and GM-CSF were measured by multiplex immunoassay. Control groups received media only (grey). All the experiments were performed with six biological replicates. Data are presented as the means ± standard error of the mean (SEM), with significant differences in the One-way ANOVA. (*) represents p < 0.05; (**) p < 0.001; (***) p < 0.0005; (***) p < 0.00001. At 4 h, we found that TNF-α and IL-17 levels were significantly higher in the cell culture supernatant fluids of Δcas7-challenged THP-1 cells than those co-cultured with WT, and both groups were higher than the media alone (p < 0.05). At 4 h, only the cells infected with Δcas7 significantly increased IFN-γ, CCL3/MIP-3 and GM-CSF compared to the control. However, at 6 h and 24 h, the levels of TNF-α, IFN-γ, IL-17, CCL3/MIP-3 and GM-CSF released by the THP-1 cells were similar between the WT and Δcas7 groups ([110]Figure 3B). No statistical difference between WT and ∆cas7 groups for the other cytokines and chemokines was detected, however, both strains elicited a response that was significantly different from the control (p < 0.05, Supplemental Figure 2). Deletion of cas7 resulted in significant changes in the expression of genes by P. gingivalis during intracellular invasion To begin to explore the response of P. gingivalis and host cells during infection and the impact of cas7 on this interaction, we investigated gene expression using a dual transcriptomic approach to understand whether P. gingivalis ∆cas7 responds differently than WT to the intracellular environment of THP-1 cells and if these changes impact host response. Principal component analysis was performed to compare similarities in gene expression profiles among groups through clustering. At 2 h, we found a modest yet significant separation between WT and ∆cas7, but some overlap is noted ([111]Figure 4A). However, at 6 h, we found greater divergence of clustering between the WT strain and the mutant was detected than what was observed at 2 h ([112]Figure 4A). At 2 h, ∆cas7 P. gingivalis presented a total of 94 differentially expressed genes, with 50 being upregulated and 44 downregulated compared to WT P. gingivalis ([113]Figure 4B). At 6 h, Δcas7 P. gingivalis presented a total of 112 differentially expressed genes, with 70 being up-regulated and 42 downregulated compared to WT P. gingivalis ([114]Figure 4C). Among these genes with p values < 0.05 and log[2] fold-change higher than 1, we found genes related to oxidative stress resistance, heme metabolism and virulence of P. gingivalis, ranked in logFold-change ([115]Table 1). Antioxidant proteins and DNA-repair genes presented significant changes in their expression: tpx (thioredoxin peroxidase activity), PGN_RS1450 (ruberythrin, rbr) and at 2 h, uvrA (DNA repair activity). Genes related to heme metabolism pathways, including flavodoxin (Fla), hemG, hmuy and hemT, had significantly decreased expression. Interestingly, Fe binding and transport-related genes 4Fe-4S and Ton-B receptors were overexpressed for ∆cas7 at both time points. In the context of hemagglutination, we found that the hemA gene was also overexpressed by the mutant compared to the WT in a later phase. Notably, both time points presented reduced expression of electron transfer activity genes and overexpression of the virulence-related genes tapC and T9SS components. At 6 h, a previously described adhesin regulated by the CRISPR-cas system (PGN_RS07375) was upregulated [[116]26]. The isoprenoid biosynthesis-related gene dxs was underexpressed by the mutant at both time points. Interestingly, several transposase-related genes showed increased expression. A gene related to the development of subcutaneous infection in mice, PGN_RS01435 (ompH) [[117]46], was significantly overexpressed by Δcas7 compared to WT in THP-1 cells. The CRISPR gene cas6 was also slightly overexpressed (see the complete transcriptomics in the Supplemental Table). Figure 4. [118]Figure 4. [119]Open in a new tab Comparison of gene expression between P. gingivalis ∆cas7 and WT during intracellular infection. Principal component analysis (PCA) of the transcriptomes of intracellular P. gingivalis WT and Δcas7 mutant strains 2 and 6 h following THP-1 cell invasion (A). The transcriptomes from the different biological replicates for each condition are circled. Volcano plot displaying differentially expressed genes of P. gingivalis Δcas7, 2 h (B) and 6 h (C) of the antibiotic protection assay compared to P. gingivalis WT. Cut-offs of p-value ≤ 0.05 and log[2]-fold-change (FC) ≥ 1.0. The blue dots represent downregulated genes, and the red dots represent upregulated genes. Table 1. Differently expressed genes of P. gingivalis ∆cas7 during intracellular infection. P. gingivalis ∆cas7 upregulated genes in intracellular invasion of THP-1 cells (2 h) __________________________________________________________________ Porphyrin (Heme) metabolism Gene Annotation log[2]FC p-value PGN_RS04275 4Fe-4S 1.68 6.89E-16 PGN_RS03040 4Fe-4S V2 1.26 3.25E-24 Oxidative stress resistance PGN_RS01450 rbr 1.67 1.07E-21 Virulence PGN_RS06315 T9SS type A sorting domain-containing protein 1.32 9.21E-17 PGN_RS00715 tapC 1.12 1.44E-14 PGN_RS01435 ompH 1.05 1.04E-09 Transposases PGN_RS05430 3.91 5.60E-06 PGN_RS04020 2.56 0.015 PGN_RS04650 2.52 0.025 P. gingivalis ∆cas7 downregulated genes in intracellular invasion of THP-1 cells (2 h) __________________________________________________________________ Porphyrin (Heme) metabolism Gene Annotation log[2]FC p-value PGN_RS08500 fla −2.90 9.09E-33 PGN_RS06540 hemT −1.35 1.21E-9 Isoprenoid biosynthesis PGN_RS09885 dxs −1.68 1.23E-13 P. gingivalis ∆cas7 upregulated genes in intracellular invasion of THP-1 cells (6 h) __________________________________________________________________ Porphyrin (Heme) metabolism Gene Annotation log[2]FC p-value PGN_RS04275 4Fe-4S 1.9 8.03E-30 PGN_RS08230 hemA 1.73 8.85E-18 Oxidative stress resistance PGN_RS01450 rbr 1.57 1.9E-14 PGN_RS01845 tpx 1.18 5.9E-13 Virulence PGN_RS07375 head GIN domain-containing protein 1.25 2.81E-10 Transposases PGN_RS02170 4.85 0.006 PGN_RS07710 1.77 0.012 PGN_RS02125 4.59 0.02 P. gingivalis ∆cas7 downregulated genes in intracellular invasion of THP-1 cells (6 h) __________________________________________________________________ Porphyrin (Heme) metabolism Gene Annotation log[2]FC p-value PGN_RS08500 fla −4.26 1.86E-16 PGN_RS02650 hmuY −1.41 4.42E-07 PGN_RS00955 hemG −1.03 0.004 Isoprenoid biosynthesis PGN_RS09885 dxs −1.94 4.69E-20 [120]Open in a new tab The ∆cas7 mutant induced changes in the expression of genes in the THP-1 cells associated with periodontal disease Next, from our dual transcriptomics experiments, we explored the gene expression changes between THP-1 cells cocultured with P. gingivalis WT and Δcas7. At both points analysed, the PCA revealed different gene expression profiles of THP-1 co-cultures with WT and ∆cas7 ([121]Figure 5A). At 2 h, THP-1 cas7 cells presented a total of 280 differentially expressed genes compared with those in THP-1 cells cultured with WT P. gingivalis, with 100 upregulated genes and 180 downregulated genes ([122]Figure 5B) while at 6 h, THP-1 ∆cas7 presented a total of 295 differentially expressed genes, with 124 upregulated genes and 171 downregulated genes ([123]Figure 5C; see the complete transcriptomics in supplemental data 3). At 2 h, the inflammatory-related genes PEL1 (pellino E3 ubiquitin protease ligase 1) and ACOD1 (aconitate decarboxylase 1) were significantly downregulated. Importantly, genes positively associated with periodontitis, IFI44 and OASL, were also found to have reduced expression [[124]47,[125]48], and the gene MT-ND3, whose mutation is associated with more aggressive periodontitis, was found to be overexpressed by the mutant [[126]49]. At 6 h, vascular endothelial growth factor (VEGFA) [[127]50] MX1, MX2 [[128]51], OASL [[129]48], OAS2 [[130]52], ISG15 [[131]53], THBS1 [[132]51] and IRF7 [[133]54], genes associated with periodontitis and inflammatory activation, displayed reduced expression ([134]Figure 5B and [135]C). On the other hand, the antimicrobial gene CCL25 (C–C-motif chemokine ligand 25), which is strongly related to mucosal immunity, was increased in cells cocultured with ∆cas7 P. gingivalis at 2 h. Moreover, at 2 h, the increase in the expression of the immunomodulatory gene IFNB1 and the arachidonic acid pathway key genes ALOXE3 and PLA2GF were detected ([136]Figure 5A and [137]B). Interestingly, the overexpression of the positively correlated peroxide biosynthesis-related gene MT-CO2, which encodes the second subunit of cytochrome oxidase, important for cell metabolism, was also detected ([138]Table 2). Figure 5. [139]Figure 5. [140]Open in a new tab Comparison of gene expression between THP-1 cells infected with P. gingivalis ∆cas7 and WT. Principal component analysis (PCA) of the transcriptomes of THP-1 cells cocultured with P. gingivalis WT and Δcas7 mutant strains for 2 and 6 h (A). The transcriptomes of the different biological replicates for each condition are circled. Volcano plot of THP-1 cells at 2 h (B) and 6 h (C) of intracellular infection by P. gingivalis ∆cas7 compared to WT. Cut-offs of p-value = 0.05 and log[2]-fold-change (FC) = 1.0. The blue dots represent downregulated genes, and the red dots represent upregulated genes. Table 2. Differentially expressed genes of THP-1 cells during intracellular infection with P. gingivalis Δcas7. Gene Annotation log[2]FC p-value Macrophage-like cells invaded with ∆cas7 upregulated genes (2 h) H1-2 H1.2 linker histone 1.47 2.77E-13 H2BC14 H2B clustered histone 14 1.38 2.05E-10 EFCAB10 EF-hand calcium binding domain 10 1.21 1.80E-16 MT-ND3 Mitochondrially encoded NADH: ubiquinone oxidoreductase core subunit 3 1.10 9.63E-11 EPHA5 EPH receptor A5 1.03 3.16E-10 MT-CO2 Mitochondrially encoded cytochrome c oxidase II 1.02 2.00E-20 Macrophage-like cells invaded with ∆cas7 downregulated genes (2 h) EDN3 Endothelin 3 −2.81 2.58E-09 IFI44 Interferon-induced protein 44 −1.15 4.84E-10 PELI1 Pellino E3 ubiquitin protein ligase 1 −1.01 4.48E-36 ACOD1 Aconitate decarboxylase 1 −1.01 2.04E-14 OASL 2′-5′-Oligoadenylate synthetase like −1.00 6.17E-10 Macrophage-like cells invaded with ∆cas7 upregulated genes (6 h) RPL39 Ribosomal protein L39 1.39 5.13E-13 Macrophage-like cells invaded with ∆cas7 downregulated genes (6 h) MX1 MX dynamin like GTPase 1 −1.48 1.80E-34 SIGLEC1 Sialic acid binding Ig like lectin 1 −1.44 1.97E-16 GLIS1 GLIS family zinc finger 1 −1.30 6.54E-12 ISG15 ISG15 ubiquitin like modifier −1.21 8.25E-17 OASL 2′-5′-oligoadenylate synthetase like −1.21 3.91E-15 ISG15 ISG15 ubiquitin like modifier −1.21 8.25E-17 HELZ2 Helicase with zinc finger 2 −1.07 3.15E-20 IRF7 Interferon regulatory factor 7 −1.07 9.29E-11 VEGFA Vascular endothelial growth factor A −1.06 1.62E-11 MX2 MX dynamin like GTPase 2 −1.05 1.60E-19 THBS1 Thrombospondin 1 −1.03 4.5E-12 OAS2 2′-5′-oligoadenylate synthetase 2 −1.03 5.31E-20 USP18 Ubiquitin specific peptidase 18 −1.02 3.57E-09 FAM174C Family with sequence similarity 174 member C −1.00 3.92E-12 [141]Open in a new tab GO term enrichment analysis at 2 h revealed the activation of ‘structural chromatin’ and the suppression of the ‘G-protein receptor activity’ and ‘transmembrane signalling receptor activity’ pathways ([142]Figure 6A). However, at 6 h, a greater number of GO terms were enriched. We can highlight by p-values ≥ 0.05 the suppression of the GO terms of ‘structural chromatin’, ‘receptor ligand activity’, ‘signalling receptor activity’ and ‘signalling receptor activity regulator’ and the activation of the terms 'olfactory receptor activity’, ‘neurotransmission receptor activity’, ‘amino acid sodium symporter activity’, ‘amino acid sodium monoatomic symporter activity’ and ‘organic acid symporter activity’ ([143]Figure 6B). To better understand pathway changes in THP-1 cells in the response to ∆cas7, we performed pathway enriched analysis using GO terms. No enriched pathways were detected from the GO terms at 2 h. However, at 6 h, pathways activated included ‘basal cell carcinoma’, ‘nicotine addiction’, ‘phenylamine metabolism’, ‘tryptophan metabolism’, ‘olfactory receptor activation’ and ‘pathways suppressed included systemic lupus erythematosus’, ‘human papillomavirus’ and ‘Cushing syndrome’ ([144]Figure 6C). Figure 6. [145]Figure 6. [146]Open in a new tab GO terms and KEGG pathway enrichment analysis. GO term enrichment of differentially expressed THP-1 cells, 2 h (A) and 6 h (B) of antibiotic protection assay with P. gingivalis ∆cas7 compared to WT. Pathway enrichment of differentially expressed genes of THP-1 cells after 6 h of antibiotic protection assay with P. gingivalis ∆cas7 compared to WT (C). The y‐axis denotes GO-terms or pathway names; x‐axis indicates the gene ratio. The bubble size indicates the number of genes in each GO term or pathway. The colour indicates the corrected p‐value. Deletion of the cas7 gene shifts the virulence of P. gingivalis in an invertebrate model To assess virulence, we used the G. mellonella model [[147]26]. Based on the differences found between the WT and ∆cas7 transcriptomic data of the present study and the cas3 data previously reported [[148]18], there appear to be nuanced differences in how the host responds to ∆cas7 compared to ∆cas3. Thus, we included the ∆cas3 mutant in our experimental groups as a virulence-positive control. Compared to our control (medium only), WT, ∆cas3 and ∆cas7 groups all showed a significant decrease in survival ([149]Figure 7). The cas7 mutant did not induce a statistically significant change in survival rate compared to WT; however, the trend of an increase in virulence was observed ([150]Figure 7). As anticipated, the cas3 mutant was more virulent than WT. Interestingly, despite cas7 playing a key role in the assembled cascade, the virulence of the cas7 mutant was significantly reduced compared with the ∆cas3 mutant. The heat-killed groups did not induce a significant change in survival compared to the control. Figure 7. [151]Figure 7. [152]Open in a new tab Survival curves in Galleria mellonella. Kaplan‒Meier survival curves were determined. G. mellonella larvae were injected with the P. gingivalis ATCC 33277 WT, Δcas3, or Δcas7 strains. Additional groups of larvae were inoculated with: TSBHK medium, TSBHK medium plus the heat-killed wild type or mutant. strains^. Statistically significant differences using the log-rank test (*p < 0.05). Discussion Our present study revealed that deleting the cas7 gene significantly impacted oxidative stress resistance and induced hemagglutination of P. gingivalis, affected its host‒pathogen interaction using a macrophage model, and had a slight impact on virulence in the G. mellonella model. Beyond its role in protecting the bacterial genome from phages and acquiring exogenous DNA, recent data reveal that CRISPR-cas systems also participate in physiology and virulence [[153]16–18,[154]22]. To this point, approximately 18% of 330 CRISPR-bearing species possess self-targeting spacers [[155]55]. The CRISPR-cas system has been shown to impact biofilm formation and acquisition of resistance genes of pathogens such as Streptococcus mutans [[156]56,[157]57] and Enterococcus faecalis [[158]58,[159]59]. Moreover, a significant association between the presence of CRISPR-cas systems and antibiotic resistance has been described in E. faecalis and Klebsiella pneumoniae [[160]60,[161]61]. CRISPR-cas genes have also been reported to be involved in the regulation of interspecies competitive interaction and intraspecies diversification among periodontal pathogens, such as P. gingivalis, Treponema denticola and Tannerella forsythia [[162]62,[163]63]. Interestingly, Yost et al. [[164]14] found a positive correlation between the progression of PD and the expression of cas genes from both CRISPR system type I and type III of P. gingivalis. Recently, Krieger et al. [[165]15] described an increase in the expression of several cas genes and iron uptake genes by Fusobacterium species colonising the oral cavity [[166]15]. It has been reported that the CRISPR-cas system is involved in oral bacterial environmental stress resistance, such as changes in pH, heat and reactive oxygen species [[167]20,[168]64]. Focusing on cas7, we identified a significant increase in P. gingivalis sensitivity to H[2]O[2.] The importance of oxidative stress resistance ability by P. gingivalis is supported by the clinical findings of Yost et al., who also detected an increase in the expression of the oxidative stress response pathway in periodontal sites that transitioned to more severe periodontitis. This pathway includes expressing antioxidant enzymes, DNA repair and the hemin layer [[169]9]. Understanding the dynamic interaction between bacteria and the host during infection is difficult. Many studies have focused on the expression of gene pathways of one member or the other, but less frequently on both. To address this dynamic relationship between host and bacteria, we employed a dual-transcriptomics approach to monitor both members of this infection interaction. Our dual-transcriptomics data support that the ∆cas7 mutant has a complex dynamic in expression of oxidative stress genes[.] During the intracellular infection, RNAseq analysis of P. gingivalis revealed that deletion of cas7 led to increased expression of antioxidant protein genes tpx,rbr and a decrease of hemin uptake-related genes flavodoxin, hemG, hmuY and hemT. As iron-binding proteins and hemagglutinin enzymes were significantly overexpressed and hemG and hmuY shared reduced expression, these data support a change in the dynamics of uptake of iron, which may be affecting the formation of the hemin layer that plays a crucial role in oxidative stress resistance [[170]45,[171]65]. Our hypothesis is that the effect on the hemagglutination and oxidative stress resistance properties reflects an imbalance of the function of these genes. Although we do not know why there are differences in gene expression patterns, we speculate that based on the functional data that the flavodoxin and hemG, hmuY and hemT may be more relevant. Further studies are needed to mechanistically understand the phenotype and the precise mechanisms of the increase oxidative stress sensitivity of P. gingivalis ∆cas7 while inside the host cell. Solbiati et al. [[172]18] also described this down-expression of hemin-uptake genes for the P. gingivalis cas3 mutant during intracellular infection. We found an attenuation of 1−2 dilutions of the hemagglutination when the cas7 gene is mutated. Previously, the decrease of hemagglutination in 2 dilutions between P. gingivalis strains had been associated with decreased ferric uptake and oral bone loss in mice [[173]66]. DNA-repair-related genes (urvA) also had an impact on their expression. Notably, we detected an increase in the expression of the T9SS (Type IX Secretion System) gene by the mutant, but no increase in expression in gingipains encoding genes rgpA, rgpB or KgP. In P. gingivalis and some other periodontopathogens, the T9SS is critical for translocating proteins, such as gingipains, across the outer membrane [[174]67]. From our data, we speculate that there may be an impact in the transport of these virulence factors of P. gingivalis to the surface of the organism, which could impact the differences in the host response we observed against P. gingivalis ∆cas7 compared to WT. However, we found similar in vitro growth and intracellular survival, which may suggest if this finding is specific for intracellular milieu. For future studies, quantifying the levels of gingipains in supernatants can lead to the clarity of these findings. Previously, Solbiati et al. [[175]18] investigated the impact of cas3 on the virulence of P. gingivalis. Similar to our current findings with ∆cas7, it was found that deletion of the cas3 gene is related to an increase in TNF-alpha release by the infected macrophage-like cells, without directly affecting P. gingivalis intracellular survival. This was anticipated as both cas3 and cas7 participate in the same cascade process of crRNA degradation [[176]20,[177]24]. The present study found that the cas7 mutant induced a higher expression of IL-17 and IFN-gamma and other proinflammatory cytokines. IL-17 is secreted by various innate and adaptive immune cells. It can activate a series of inflammatory cascade reactions mediating the occurrence and development of periodontitis and related systemic chronic inflammatory diseases [[178]68,[179]69]. Besides recruiting macrophages, IFN-gamma is related to producing reactive oxygen species against periodontal pathogens [[180]70,[181]71]. It has been demonstrated that IL-17 and IFN-gamma can be related to in vitro and in vivo osteoclastogenesis and periodontal destruction [[182]72]. However, compared to transcriptomic data for cas3 reported by Solbiati et al. [[183]18], the deletion of cas7 showed a more modest change in gene expression when compared to WT. Fewer genes were expressed differently, and no significant changes in Toll-like receptors or other factors were observed. Another contrast was that cas3 transcriptomics data showed increased expression of the isoprenoid biosynthesis dxs gene, while the cas7 mutant significantly downregulates it. Interestingly, this pathway has been described as being involved in oxidative stress resistance [[184]73] and as a potential new target for developing new antibiotics against gram-negative pathogens such as K. pneumoniae and E. coli [[185]74–76]. With the finding that there are subtle differences in the host response to cas7 mutation compared with cas3, we explored the possibility of differences in the virulence of ∆cas7 compared with ∆cas3. Another study from our group described a hypothetical protein (PGN_RS07375) regulated by the P. gingivalis cas3 gene that plays an important role in virulence as an adhesin using the wax worm model [[186]26]. The P. gingivalis ∆cas3 mutant presented an increase in expression of PGN_RS07375, which was also observed in the present study by the ∆cas7 mutant. That suggests that deletion of the cas7 gene follows a similar phenotype to that of the cas3 gene regarding regulating this specific adhesin. Using the G. mellonella wax worm virulence model, we found a trend of increase in virulence of Δcas7 compared to WT, although intermediate of the previous level reported for Δcas3. We speculate that the CRISPR-cas system may have a partial compensatory mechanism for the deletion of cas7 that may not be as efficient for cas3 deletion. Miller et al. [[187]27] described the importance of cas genes on intracellular invasion using the gingival keratinocyte model. However, we did not detect statistically significant differences in intracellular survival between WT and Δcas7 strains using macrophage-like THP-1 cells, as it was also observed for the cas3 mutant previously [[188]18]. We do not know why these differences exist, but it could suggest that the impact of the CRISPR-cas system in host‒pathogen interaction may be cell-specific. Interestingly, from our dual transcriptomics experiments, among the enriched pathways, we found activation of olfactory receptor activity. It has been suggested that these receptors have an extra-nasal function in modulating macrophage activities, such as transcription of inflammasome components, IL-1β secretion and tumour progression [[189]77]. The suppressed pathways include the inflammatory-mediated diseases systemic erythematous lupus and human papillomavirus, which have been reported to significantly impact the host–pathogen interaction dynamics of P. gingivalis [[190]78–80]. In an oral environmental scenario, heme appears critical to the virulence of P. gingivalis. Heme is an important agent in impacting oxidative stress resistance of this organism[[191]26,[192]65,[193]66]. Although our studies have uncovered many new contributions of cas7 to P. gingivalis virulence, we acknowledge limitations such as the rather focused nature of our experimental design using macrophages. Thus, exploring the capacity of P. gingivalis to invade different cells and additional complexity, including organotypic and animal modelling, may reveal more about the mechanistic role of P. gingivalis CRISPR-cas system in PD development. Indeed, the clinical study of Yost et al. [[194]14] supports that the increase expression of CRISPR genes is positively correlated with the progression of clinical markers of PD progression. We recognise that the present study is descriptive in nature. Despite our new information connecting cas7 to oxidative stress sensitivity and gene regulation during macrophage uptake, there are limitations. Some examples include lack of complementation experiments to verify whether the observed phenotypic changes are specifically due to cas7 deletion, we did not confirm expression of selected genes identified in transcriptomics an alternative technique such as RT-PCR; however, our functional assays do suggest what genes patterns may be responsible for the phenotype of oxidative stress sensitivity. It would be interesting to explore the gene expression of the P. gingivalis strains in vitro during oxidative stress and hemagglutination assays, but we will consider this in future studies. Mechanistically, how the CRISPR-cas system influences P. gingivalis expression profiling is not clear. We speculate that CRISPR-cas systems process guide RNAs that match endogenous sequences, to modulate transcription and/or translation, as previous suggested in other microorganisms [[195]81–83]. Currently, we are working on finding these matches on the P. gingivalis genome. In summary, the cas7 gene of P. gingivalis plays an important role in bacterial biological processes and their interaction with the host. We observed that cas7 deletion impacts bacterial resistance to oxidative stress and hemagglutination. Additionally, the cas7 mutant induced a higher release of proinflammatory cytokines from THP-1 in early stages of intracellular infection and a different expression of genes in both host and pathogen using an antibiotic protection assay. When taken together with our prior studies, disruption of the CRISPR cascade complex is important in P. gingivalis virulence, but that there are nuanced differences found for mutation of different elements of the cascade complex support that there is potential for additional mechanistic studies to investigate this region in the context of P. gingivalis virulence and host‒pathogen interaction. Acknowledgements