LPS is a unique component of the outer membrane of Gram-negative bacteria, composed of lipid A, a short core oligosaccharide and O-antigen. It is also called endotoxin due to its ability to induce a robust inflammatory response. LPS is the most representative virulence factor among periodontal pathogens. Clinical studies have shown that LPS levels were positively correlated with periodontal clinical parameters and inflammatory factors before and after periodontal treatment (Lee et al., 2008; Shaddox et al., 2013).

After the periodontal microbial homeostasis is disrupted, P. gingivalis proliferates in large numbers, and excessive proliferation and death cause the release of LPS (Zheng et al., 2021). P. gingivalis LPS is recognized by the Toll-like receptor 4 (TLR4) of macrophages and activates the NF-κB and MAPK signaling pathways, thereby inducing the release of inflammatory cytokines (Nativel et al., 2017). These factors can promote the expression of matrix metalloproteinases and osteoclast factors and then destroy soft tissue and bone.

P. gingivalis LPS can induce macrophage foam cell formation. It promotes the binding of macrophages to low-density lipoproteins (LDL) and induces macrophages to modify native LDL (Qi et al., 2003). P. gingivalis LPS promotes monocyte chemotaxis and adhesion to vascular endothelial cells through Akt and NF-κB signaling pathways (Wang et al., 2020). P. gingivalis LPS can also promote the high expression of angiotensin II and IL-6 in vascular endothelial cells and accelerates endothelial dysfunction (Viafara-Garcia et al., 2019). In animal models, P. gingivalis LPS increased the secretion of TNF-α from macrophages and up-regulated the expression of endothelial cell adhesion molecules, which aggravated the exacerbated effect of ligature-induced PD on AS (Suh et al., 2019). However, the role of other periodontal pathogenic LPS in AS, such as Treponema denticola (T. denticola) and Tannerella forsythia (T. forsythia), is still unknown.

From a broader perspective, PD related low-grade endotoxemia (LGE) causes phenotypic and transcriptional changes in myeloid cell populations that enhance their response to pathogens, a process known as trained immunity (Netea et al., 2020). LPS from periodontal pockets continue to enter the peripheral blood at low levels, activating neutrophil hyperresponsiveness (Vitkov et al., 2021) (The role of neutrophils in this process will be discussed in detail later). Therefore, periodontal pathogenic bacteria-derived LPS represented by P. gingivalis LPS and its induced LGE may be the link between PD and AS.

CpG DNA is a type of DNA sequence with immune activation function containing unmethylated CpG motif, including artificially synthesized oligodeoxynucleotides containing CpG (CpG ODN) and genomic DNA of bacteria, viruses, and invertebrates (Zhou and Deng, 2021). CpG DNA triggers immunostimulatory activity through TLR9 (Ohto et al., 2015). TLR9 is highly expressed in the gingival tissue of PD patients (Narayan et al., 2018), suggesting that CpG DNA could be actively involved in the progression of PD.

It was found in vitro that macrophages recognize CpG DNA from periodontal pathogens through TLR9 and then highly express IL-1β and TNF-α to induce osteoclastogenesis (Zou et al., 2002). In addition, TLR9-related autophagy may also be involved in the progression of PD (Wei et al., 2020). However, there are also studies showing that CpG ODN sometimes may positively affect PD. CpG ODNs can promote the proliferation and differentiation of MC3T3 cells in the early stage and up-regulate the expression levels of bone differentiation genes SP7 and OCN (Yu et al., 2020).

AS can associate with PD through CpG DNA. In ApoE^−/- mice infected by P. gingivalis, alveolar bone resorption and aortic plaque increased significantly (Xuan et al., 2017). The genomic DNA of P. gingivalis can be detected in the oral epithelium and the aorta (Velsko et al., 2014). In polymicrobial infection-induced periodontal disease, ApoE^−/- mice had enlarged aortic plaques, accumulation of macrophages around the arteries, increased serum cholesterol and triglycerides, while genomic DNA of P. gingivalis, T. denticola, and T. forsythia can be detected in the aorta and liver (Rivera et al., 2013). Intravenous injection of P. gingivalis in ApoE^−/- mice can also aggravate AS, and the ribosomal DNA of P. gingivalis can be detected in the aorta, liver and heart (Li et al., 2002). However, P. gingivalis DNA in the periodontal pocket may not be transferred to the heart valve area, causing the aortic valve and mitral valve to degenerate (Radwan-Oczko et al., 2014). Therefore, periodontal pathogens represented by P. gingivalis might colonize the arterial wall through blood circulation, and the CpG DNA released after bacterial cell lysis may regulate the development of AS by activating the corresponding TLR9 pathway.

However, the role of CpG DNA/ODN on AS is still controversial. The genetic deletion of the TLR9 gene exacerbated AS lesions in ApoE^−/- mice, and the use of CpG ODN 1668 can reduce this effect (Koulis et al., 2014). In other words, CpG ODN and TLR9 may protect the aorta in specific circumstances. Further studies demonstrated that TLR9 plays a negative role in vascular injury (Hirata et al., 2013), and systemic stimulation of TLR9 with high-dose CpG ODN will aggravate the development of AS (Krogmann et al., 2016). There are differences between CpG DNA released in PD and synthetic CpG ODN. The difference in concentration and sequence may lead to changes in the activation of the downstream inflammatory pathway of TLR9 in AS lesions. Therefore, it is necessary to screen PD-related CpG DNA to clarify its role in AS in future studies.

Peptidoglycan (PGN) is a common component of bacterial cell walls. Transcriptional pathways for peptidoglycan synthesis are significantly up-regulated in tongue and subgingival plaque in patients with periodontitis (Yost et al., 2015; Belstrom et al., 2021). PGN can be recognized by TLR2 on the cell membrane and the endogenous NOD1, NOD2, and NLRP3. The expression of NOD1 and NOD2 can be detected in human periodontal ligament cells (hPDLC). Under the stimulation of PGN, the production of IL-6 and IL-8 in hPDLC increased, and the NF-κB and MAPK signaling pathways were activated (Jeon et al., 2012). Injecting PGN into the gums of mice can induce osteoclastogenesis, and TLR2, NOD1, and NOD2 are activated (Kishimoto et al., 2012). N-acetylglucosamine can be recognized by the glycolytic enzyme hexokinase in the cytoplasm and subsequently activate the NLRP3 inflammasome, which may be involved in the progression of PD (Wolf et al., 2016). In summary, periodontal pathogens may activate excessive innate immunity through PGN, an exogenous pathogen-associated molecular pattern, and cause PD. The destruction of periodontal tissue damages the barrier function of the oral mucosa. The decisive invasion and migration capabilities of periodontal pathogens make it enter the circulation and spread the PGN in the cell wall to the cardiovascular system.

Although there are few studies on the relation between periodontal PGN and AS, as a ubiquitous substance in bacteria, PGN produced by lysis or ectopic colonization of periodontal pathogens around blood vessels can cause chronic inflammation. The long-term chronic inflammation of the vascular microenvironment is obviously beneficial to the formation of AS. Early studies found that PGN induced the production of pro-inflammatory cytokines through TLR2 and increased the vulnerability of AS plaques (Nijhuis et al., 2004). While the vascular endothelial dysfunction appeared in rats modeled by surgery and a high-cholesterol diet, the concentration of serum PGN was significantly increased (Tsunooka et al., 2005). It has been confirmed that TLR2 is expressed in macrophages in AS lesions. PGN activated monocytes to overexpress intercellular adhesion molecule-1 (ICAM-1) through the TLR2 and NF-κB pathways, promoting monocyte adhesion and chemotaxis to vascular diseases (Nijhuis et al., 2007; Xie et al., 2016). Using PGN to stimulate human coronary artery endothelial cells that were knocked out TLR2 through CRISPR-Cas9 technology, the expression levels of ICAM-1, IL-6, and IL-8 were significantly down-regulated (Wang et al., 2018). PGN can also be recognized by PGN recognition protein-1 (PGLYRP-1) in the innate immune system. The level of circulating PGLYRP-1 is associated with AS, coronary artery calcification, thickening of the abdominal aorta, and acute coronary syndrome (Brownell et al., 2016; Han et al., 2021). PGLYRP1 may promote the formation of AS plaques by regulating the overexpression of adhesion molecules in endothelial cells (Jin et al., 2021). PGN also triggers the up-regulation of vascular cell adhesion molecule-1 (VCAM-1) through the NOD1-RIP2-NF-κB axis, promotes the recruitment of myeloid cells, and leads to endothelial dysfunction (Gonzalez-Ramos et al., 2019).

Neutrophil extracellular traps (NETs) are an extracellular fibrous network structure produced by neutrophils after being stimulated, mainly composed of chromatin and cellular proteins (Brinkmann et al., 2004). The process of neutrophils forming NETs is called NETosis, including suicidal NETosis and survival vital NETosis, which is considered a cell death program different from apoptosis and necrosis (Yipp and Kubes, 2013). NETs recognize, trap, and restrict the spread of bacteria and other pathogens and highly express antimicrobial peptides and other antibacterial ingredients to delete pathogens ultimately (Papayannopoulos, 2018).

Numerous clinical studies have confirmed that NETs are closely related to the progression of PD. NETs expression in the inflamed gingival tissue was higher than that in the healthy control group (White P. C. et al., 2016). The expression of NETs in the gingival tissue of patients with periodontitis was higher than that of patients with gingivitis, indicating that the level of NETs is related to the severity of periodontal inflammation (Magan-Fernandez et al., 2019). The gingival biopsy samples of patients with PD and the purulent exudate in the periodontal pockets (Vitkov et al., 2009) found high expression of NETs, showing a fibrous network structure. There are many bacteria in the NETs and their mechanical entanglement, and they are closely arranged on the surface of the epithelium. A case-control study (Kaneko et al., 2018) found that the NETs level was positively correlated with the average probing depth and clinical attachment loss in patients with PD. NETs were detected in supragingival plaque biofilms, and NETs-related protein Myeloperoxidase (MPO) was found in saliva and biofilms, which confirmed that oral bacteria isolated from plaque biofilms could stimulate the formation of NETs (Hirschfeld et al., 2015). Neutrophils in PD are recruited by fibrin through myeloid integrin α_mβ₂-binding motif and activated to generate NETs (Silva et al., 2021). Studies have found that P. gingivalis (Jayaprakash et al., 2015) and Fusobacterium nucleatum (F. nucleatum) (White et al., 2014) can stimulate neutrophils to produce reactive oxygen species (ROS). Streptococcus sanguis also increased the level of NETs marker citH3 and up-regulated the level of MPO (Oveisi et al., 2019). In summary, specific periodontal pathogens can stimulate neutrophils to produce ROS and release NETs.

Multivariate logistic regression analysis showed that the peripheral blood NET level was significantly positively associated with moderate to severe PD (Kaneko et al., 2018). Degradation of NETs in plasma is increased after periodontal therapy in PD patients (Moonen et al., 2020). Periodontal pathogens and excessive NETs produced during PD may participate in the progression of AS after entering the circulation. After entering the bloodstream, P. gingivalis binds to erythrocytes to avoid ROS destruction, thereby further activating the Rho GTPase signaling pathway, up-regulating CD11b/CD18, and promoting the activation of neutrophils (Borgeson et al., 2011; Damgaard et al., 2017). Periodontal pathogen DNA can be detected in carotid plaque. T. forsythia is significantly related to intraplaque hemorrhage and neutrophil activation, reflected in the increased release of MPO, cell-free DNA, and NETs (Range et al., 2014). The systemic inflammatory state caused by PD promotes the adhesion of neutrophils and endothelial cells by increasing oxidative stress parameters (superoxide and mitochondrial membrane potential) (Martinez-Herrera et al., 2018). However, studies are still limited to the relationship between periodontal NETs and AS. The differences between the histone composition and DNA sequence of periodontal NETs and other systemic inflammatory NETs, as well as the mechanism of periodontal NETs in AS plaque formation, deserve more efforts to reveal both in vivo and in vitro.

High mobility group Box 1 (HMGB1) is a non-histone chromosome binding protein widely distributed in the nucleus of various cells. It plays an important role in stabilizing the structure of nucleosomes, regulating transcription factors, and DNA replication repair (Vijayakumar et al., 2019). HMGB1 can be released by necrotic or ruptured cells and activated immune cells (Andersson and Tracey, 2011). High levels of HMGB1 can be detected in the gingival crevicular fluid of patients with moderate to severe chronic PD (Luo et al., 2011; Paknejad et al., 2016), and it was positively correlated with periodontal clinical parameters (including plaque index, bleeding index, probing depth, and clinical attachment level) (Yu et al., 2019). Human gingival epithelial cells can increase the secretion of HMGB1 under the stimulation of TNF-α (Morimoto et al., 2008). Similarly, IL-1β promoted the secretion of HMGB1 in fibroblasts (Ito et al., 2012). Under the constant stimulation of periodontal infection, tissue cells continuously secrete HMGB1, which causes macrophages to be further activated and secrete cytokines, and amplifying the destruction of inflammation. This process can be inhibited by HMGB1 antibody (Yoshihara-Hirata et al., 2018). Excessive HMGB1 secreted by periodontal cells in PD or released by apoptosis can enter the circulation through osmosis and transmit the damage signal to the artery.

In AS, HMGB1 plays an important role. The increase of HMGB1 level directly leads to the mass production of cytokines, including IFN-γ, TNF-α, IL-1β, and IL-6, which promotes the formation of AS and reduces the stability of plaque (Su et al., 2015). In the AS model of rabbits, administration of HMGB1 and TNF-α can significantly aggravate the inflammation of advanced plaques (Kim J.-S. et al., 2016). In the ApoE^−/- mouse AS model induced by the Western diet, anti-HMGB1 antibodies were more than six times higher than regular diet ApoE^−/- mice and ApoE^+/+ mice, indicating that HMGB1 autoimmunity is involved in the progression of AS (Pan et al., 2016). It was speculated that HMGB1 produced by PD might be involved in systemic diseases by acting on monocytes, macrophages and vascular endothelial cells (Morimoto-Yamashita et al., 2012). Bioinformatics analysis shows that HMGB1 is a potential molecular mechanism of the association between PD and AS (Ning et al., 2021). There have been reports that P. gingivalis elevated HMGB1 levels after myocardial infarction in mice (Srisuwantha et al., 2017). In addition, the circular RNA PPP1CC of P. gingivali can regulate the apoptosis of vascular smooth muscle cells through the HMGB1/TLR9/AIM2 axis (Liu J. et al., 2021). The role of HMGB1 produced by PD in the AS model remains to be discovered.

S100 protein is a group of calcium-binding proteins. Its family has more than 20 members and participates in the metabolism of the cytoskeleton under physiological conditions (Vogl et al., 2014). When cells are damaged or phagocytes are activated, macrophages secrete S100A8, S100A9, and S100A12 (Foell et al., 2007). These proteins combine with PRRs as “alarmins”, activating immune cells and endothelial cells to promote inflammation (Goyette and Geczy, 2011). The levels of S100A8, S100A9, and S100A12 in saliva and gingival crevicular fluid of patients with PD were significantly increased (Kojima et al., 2000; Shin et al., 2019; Lira-Junior et al., 2020; Jimenez et al., 2021). Microbial infection can cause abnormal expression of S100 protein in gingival tissues and destroy the epithelial barrier function (Nishii et al., 2013). Immunohistochemistry and RNA sequencing showed that S100A8 and S100A9 were highly expressed in the ligature-induced PD (Maekawa et al., 2019). After mouse osteocyte-like cells (MLO-Y4-A2) were treated with S100A9, the expression of IL-6 and RANKL increased, and the p38/ERK/STAT3 signaling pathway was activated, indicating that alarmins were involved in periodontal bone destruction (Takagi et al., 2020). It is worth noting that the levels of S100A12 and C-reactive protein in the gingival crevicular fluid and serum of patients with PD are elevated and positively correlated with periodontal parameters (Pradeep et al., 2014), suggesting that alarmins may be the link between PD and AS.

The expression of S100A9 and SMemb protein (a phenotypic marker of smooth muscle cell proliferation) increased in aneurysm specimens from patients infected with P. gingivalis (Hokamura et al., 2010). This phenotype has been verified in mice. Studies have also found that P. gingivalis infection can up-regulate the expression of S100A9 in human aortic smooth muscle cells (hAOSMC), which makes hAOSMC change from a contractile to proliferative phenotype (Inaba et al., 2009). These may lead to a potential mechanism for PD to promote aortic intimal hyperplasia. Therefore, PD can accelerate AS through alarmins.