Porphyromonas gingivalis infection may contribute to systemic and intracerebral amyloid-beta: implications for Alzheimer’s disease onset

KEYWORDS:

• Inflammation
• microbiota
• periodontitis
• systemic
• amyloid
• Aβ_3-42
• cathepsin B

1. Introduction

Generalized (‘chronic’) periodontitis, a common inflammatory disease affecting the supporting tissues of teeth, has been associated with several systemic diseases, e.g. cardiovascular diseases, diabetes, adverse pregnancy outcomes, rheumatoid arthritis, respiratory diseases, and Alzheimer’s disease (AD) [1–7]. Bacteria of the periodontal pocket can spread through the bloodstream, which is the common but not the only way of systemic bacterial dissemination in periodontitis[8]. Dental treatment, tooth brushing, flossing, chewing, and the use of tooth-picks in a patient with periodontitis will release bacteremia[9]. This can occur several times during the day and has been estimated to last for up to 3 hours[10]. Tooth-related bacteremia contains a wide spectrum of bacteria [11] among which the Gram-negative anaerobic rod Porphyromonas gingivalis seems to have a key role in the adult form of generalized periodontitis [12,13].

A plethora of studies firmly place P.gingivalis but not its companion species (for example, Tannerella forsythia and Treponema denticola in the red complex [13]) as a risk factor for AD. This is because P.gingivalis is adept at modifying the peripheral and intracerebral immune responses [14–16]. Furthermore, this bacterium has a range of enzymes including cathepsin B [17] and gingipains [18] that are, respectively, shown to interact with the amyloid precursor protein (APP) and neuronal tau resulting in the formation of amyloid-beta (Aβ) and neurofibrillary tangles (NFTs) [19,20], which are the cardinal hallmarks of AD. Prospective, retrospective population-based, and nested control studies have shown that the risk of developing the sporadic form of AD doubles when periodontal disease persists for about 10 years [21–23]. This is evident from the fact that a large section of individuals who go on to developing clinical AD also suffers from periodontitis.

Brain inflammation, characterized by increased activation of microglia and astrocytes, increases during aging, and is a key feature of AD[24]. This has been explained in terms of the hallmark lesions of AD, which are Aβ_40/42 extracellular deposits in the form of plaques and hyperphosphorylated tau protein associating with intraneuronal lesions called NFTs. Accumulation of Aβ plaques results from the proteolytic cleavage of the APP by β- and γ-secretase enzymes [25,26]. These secretases are different in AD driven by bacterial infections compared to the classically described site-specific secretases in the mutated APP of AD [27,28]. Similarly, toxic proteases from P.gingivalis called gingipains have been identified in the brain of AD patients, and the levels correlated with tau and ubiquitin pathology[15].

Aβ is classically believed to be produced by neurons within the AD brain irrespective of the trigger that causes its release. However, this view is changing, as some researchers believe the peripheral/systemic Aβ pool is also a contribution from platelets, skeletal muscle cells, skin fibroblasts, and monocyte/macrophages [29–31] and this has implications for AD pathogenesis over time. Production of inflammagens such as gingipains and lipopolysaccharide (LPS) secreted by P.gingivalis also occurs in the periodontal pocket where inflammatory macrophages are reported to bear Aβ[32]. Gil-Montoya et al. [33] have reported increased plasma Aβ_1-42 levels in individuals who have severe periodontal disease. Thus, Leira et al. [34] found when experimental periodontitis was induced in Sprague-Dawley rats, a strong positive correlation between alveolar bone loss and Aβ_1-40 serum levels at 7 days (r = 0.695, P = 0.012) and with serum Aβ_1-42 concentrations at 21 days (r = 0.968, P = 0.002). Taken together, Aβ also being generated peripherally in platelets, skin fibroblasts, and skeletal muscles [29,30] may enter the circulating blood[31]. The present editorial aims to discuss whether P.gingivalis can contribute to systemic and intracerebral pools of Aβ.

2. P.gingivalis induces systemic Aβ production in infected mice

Nie et al. [32] recently reported that chronic, systemic P.gingivalis infection increased the inflammatory responses and proteins associated with Aβ-production in the liver of mice. The liver was chosen for the peripheral Aβ source in macrophages because of the general abundance of these cells[32]. Nie et al. [32] observed that P. gingivalis infection in mouse liver macrophages caused a rapid production of interleukin 1-beta (IL-1β) and thereafter an intracellular accumulation of Aβ through activation of Toll-like receptor 2/nuclear factor kappaB (TLR2/NF-κB) signaling. NF-κB-dependent cathepsin B appeared crucial for cleaving pro-IL-1β and processing APP to induce the accumulation of pathogenic Aβ_3-42, which was significantly increased in liver macrophages of the P. gingivalis-infected mice. This original study demonstrated peripheral pools of Aβ due to periodontitis in macrophages within the periodontal tissue and in mice hepatic macrophages following P.gingivalis infection. In a follow-up study, Zeng et al. [17] induced systemic P.gingivalis infection in mice by intraperitoneal injections containing 1 x 10⁸ CFU/mouse every 3 days for 3 weeks. This significantly increased the expression of the advanced glycation end products (RAGE) receptor in the cluster of differentiation 31 (CD31)-positive endothelial cells. This implied that P.gingivalis systemic infection up-regulated RAGE expression in cerebral endothelial cells and facilitated Aβ entry into the mouse brain. Cathepsin B was suggested to be a contribution from the bacterium and the host with a critical role in regulating the NF-ĸB/RAGE expression and in the processing of APP. This study further supported the Nie et al. [32] concept for the potential in systemic spread of peripheral Aβ to the brain from P.gingivalis infection. In a proof of concept study, Bu et al. [31] had demonstrated the plausibility of peripheral Aβ entry to the brain being facilitated by the RAGE receptor within cerebral endothelial cells[17]. An alternative mode of peripheral Aβ entry into the brain is via macrophages of the lymphatic system[35].

Another focus of Nie and colleagues [32] was Aβ_1-42, which is classically considered as the toxic form of Aβ. They observed that Aβ_3-42 (Figure 1) not only occurred earlier but was also two-fold higher than Aβ_1-42 in the AD brain[32]. In AD, cathepsin B stimulated intracellular production of Aβ in the brain, including the Aβ_3-42. Interestingly, Aβ_3-42 following P.gingivalis-infection in mice generated IL-1β, which is a proinflammatory cytokine[32]. IL-1β participated in increasing the in vivo levels of Aβ_3-42 in the hepatic macrophages of P.gingivalis-infected mice and in vitro P.gingivalis-infected macrophages. Furthermore, Aβ_3-42 was induced by P.gingivalis infection, which had caused significant death of macrophages and reduced their phagocytic capacity compared to that of Aβ_1-42, suggesting Aβ_3-42 is very toxic. Aβ_3-42 was also detected exclusively in the AD brain, and this corroborates with the significantly more toxic form than Aβ_1-42[32]. This study agreed with that of Leira et al. [34] who reported that LPS from P.gingivalis increased Aβ protofibrils in the serum of rats. After experimental periodontitis had been induced in male Sprague-Dawley rats, it caused an acute elevation of Aβ_1-40 in serum that lasted during the whole experiment. Aβ_1-42 peptide levels, however, peaked at the end of the study.

Figure 1. Summarizes the Nie et al.[32]. vision as interpreted by Olsen and Singhrao for the contribution to AD of peripheral pools of Aβ, specifically Aβ_3-42. It is generated by P.gingivalis (Pg) oral infection that eventually reaches the liver and the brain. The proposed signaling pathway (TLR2,4/NF-ĸB) is also indicated where it is likely to act liberating interleukin-1β (IL-1β) cytokine that facilitates the amyloid precursor protein cleavage of Aβ via secretase enzymes, one of which is cathepsin B. The low-density lipoprotein receptor-related protein 1 (LRP1) is the receptor for Aβ transport from the brain to the peripheral blood. The Aβ from the systemic circulation can enter the brain using the advanced glycation end products (RAGE) receptor. Nie et al. [32] have shown Aβ within the gingival tissues of periodontitis patients and in the liver of middle-aged mice after chronic systemic P. gingivalis infection, thereby contributing to the peripheral pools of Aβ. Some researchers believe the peripheral Aβ also comes from platelets, skeletal muscle cells, skin fibroblasts, and monocyte/macrophages. The implications of the peripheral Aβ is that it can also enter the brain and contribute to AD pathology as shown by Bu et al. [31]

3. P. gingivalis also generates Aβ in the periodontium and within the brain

Systemically produced Aβ probably occurs in addition to locally generated Aβ in the periodontium and in the brain induced by P.gingivalis. As mentioned, Leira et al. [34] found a strong positive correlation between alveolar bone loss and Aβ_1-40 serum levels at 7 days (r = 0.695, P = 0.012) and with serum Aβ_1-42 concentrations at 21 days (r = 0.968, P = 0.002). Intracerebral production of Aβ generated by P.gingivalis has been seen in the brain of experimental wild-type animals and with AD transgenes [19,30–32]. Ilievski et al. [19] found that chronic oral application of P.gingivalis to wild-type mice resulted in the deposition of extracellular Aβ_1-42 together with neurodegeneration and intracerebral inflammation, as demonstrated previously by Poole et al. [36]. Similarly, Wu et al. [37] found that chronic exposure to LPS from P.gingivalis for five consecutive weeks caused learning and memory deficits together with intracellular accumulation of Aβ in neurons of middle-aged wild-type mice. Taken together, these reports suggest that P.gingivalis can induce both a local periodontal and a systemic Aβ production, thereby contributing to a pool of Aβ that can enter the brain facilitated by the endothelial RAGE receptor.

4. P. gingivalis interferes with components of the peripheral immune system aimed to defend the brain

Unexpectedly, recent research has shown that even components of the peripheral immune system, such as macrophages can participate in defending the brain from insults occurring outside the brain[38]. However, P.gingivalis has the ability to abolish the anaphylatoxin complement component 5a (C5a) in macrophages thereby undermining TLR2/4 immunity and degrade some of the complement receptor 1 (CR1) molecules that help clear amyloid via the spleen [39]. Whether this affects other macrophages in a similar way is not known. Further immune evasion strategies of P.gingivalis in relation to AD are discussed elsewhere[40].

5. Concluding remarks

We have communicated that monocytes/macrophages from the periodontium and the liver may provide an additional circulating pool of unique Aβ_3-42 fragments in patients with periodontitis. Entry of P. gingivalis and/or its gingipains and LPS into the brain due to a defective blood-brain barrier can lead to intracerebral deposition of Aβ plaques. These findings support the notion that the adult form of generalized periodontitis via P. gingivalis, contributes to both an oral and hepatic cellular source of cells that add to the systemic pool of Aβ. This peptide can also be a contribution of other cell sources of peripheral organs like skin smooth cells and platelets which have the potential to transport Aβ to the brain and over time may play a role in AD pathogenesis. Deposits of Aβ in the brain can start 10–20 years before the cognitive decline and the diagnosis of AD. This agrees with the timeline of at least 10 years required for periodontitis to initiate AD and emphasizes the need for meticulous dental hygiene as a feasible prophylaxis for AD.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.