The human microbiome comprises a wide variety of genes composing the bacteria, viruses, funghi, and archea (Lloyd-Price et al., 2016; Dominguez-Bello et al., 2019). The microbiome plays a vital role in our development, physiology, immunity and nutrition (Dethlefsen et al., 2007). It is estimated that only the host’s cells and bacteria ratio is closer to 1:1, and not 1:10 as was described previously (Dethlefsen et al., 2007; Sender et al., 2016). Otherwise, the term microbiota is the collection of microbes such as bacteria, viruses, funghi, and archea, that cover all the body surfaces. Recently, it has been proposed that many chronic non-communicable diseases originate from an unbalanced microbiota (Dethlefsen et al., 2007; Hajishengallis et al., 2012; Lamont et al., 2018). Thus, the composition of the microbiota could predispose or favor the onset or progression of certain diseases or health conditions, such as insulin resistance, diabetes mellitus, dyslipidemia, periodontitis, and Alzheimer’s Disease (AD) (Turnbaugh et al., 2006; Cani et al., 2007; Aemaimanan et al., 2013; Komazaki et al., 2017; Jin et al., 2018; Díaz-Zúñiga et al., 2020). Thus, the imbalance of the microbiota known as dysbiosis, is defined as a compositional and functional alteration in the behavior of both one or a group of microorganisms due to quantitative or qualitative changes (Hajishengallis et al., 2012; Levy et al., 2017). Dysbiosis can be characterized by a loss or increase of certain microorganism, or by a loss of overall microbial diversity (Petersen and Round, 2014).

Several studies have shown that periodontitis affects cognitive status, and this effect was proposed to be a consequence of a chronic low-grade inflammatory state (Poole et al., 2015; Ishida et al., 2017; Wu et al., 2017; Ilievski et al., 2018; Zhang et al., 2018; Dominy et al., 2019; Díaz-Zúñiga et al., 2020; Kantarci et al., 2020). Periodontitis is an infectious bone-resorptive chronic disease caused by the dysbiosis of the periodontal microbiota (Hajishengallis et al., 2012; Hajishengallis, 2015). Among the bacteria associated with dysbiosis, the Gram-negative bacteria Porphyromonas gingivalis, Aggregatibacter actinomycetemcomitans, Tannerella forsythia, and Prevotella intermedia have been associated with its onset and progression (Hajishengallis et al., 2012; Hajishengallis and Lamont, 2012; Hajishengallis, 2014; Hajishengallis, 2015). These bacteria produce several virulence factors that give them the ability to induce host tissues’ invasion, adhesion, phagocytosis evasion, migration, and a consequent pro-inflammatory response in different tissues or organs (Perry et al., 1996; Laine and Winkelhoff, 1998; Foschi et al., 2006; Kim et al., 2007; Vernal et al., 2008; Darveau et al., 2012; Shimotahira et al., 2013; Vernal et al., 2014a; Vernal et al., 2014b; Díaz-Zúñiga et al., 2015a; Díaz-Zúñiga et al., 2015b; Doke et al., 2017). Periodontal dysbiotic bacteria have also been detected in other distal tissues, such as decidua basalis of the placenta, intima layer of atherosclerotic plaques, in the hippocampus of people who died due to AD, and in the feces of people with ulcerative colitis (Riviere et al., 2002; Poole et al., 2013; Szulc et al., 2015; Vanterpool et al., 2016; Udagawa et al., 2018; Dominy et al., 2019; Fischer et al., 2019). Although it has recently been recognized that the oral cavity and the intestine are connected by the microbiota and that a bidirectional relationship may exist, it is still unclear whether oral dysbiosis can affect both gut microbiota and the brain. In this way, the present review aims to determine if the dysbiosis of the oral microbiota trigger changes in the gut microbiota, creating a higher predisposition for the development of neuroinflammation or neurodegenerative diseases.

Patients affected by periodontitis can have an area of ​​ulceration of 20 cm² within their periodontium (Hujoel et al., 2001). Interestingly, periodontitis has been linked to other pathological conditions by causing transient bacteremia (Schenkein et al., 2000; Komine-Aizawa et al., 2019). Although it is known that the presence of bacteria in the bloodstream causes sepsis, there is no evidence of sepsis due to periodontal bacteria in the bloodstream. Leukocytes cannot recognize, fix, and engulf bacteria in high-velocity liquids such as the bloodstream (Minasyan, 2014). However, bacteria can be attracted (by electrical charges) to the erythrocytes’ surface and can be killed by contact through the release of oxygen from the oxyhemoglobin (Minasyan, 2017a). If bacteria survive this oxidative attack, they are consecutively filtered in the liver and spleen (Minasyan, 2017a). In some cases, bacteria can overload the liver and the spleen, and induce hepato- or splenomegaly (Minasyan, 2017a). Also, it has been proposed that bacteria enter the erythrocytes by creating membrane pores, and once inside these cells, they can be killed by oxidation or either be resistant (Minasyan, 2014; Minasyan, 2016). Only bacteria capable of resisting this outbreak would cause host death by sepsis (Minasyan, 2016). In this sense, once in the bloodstream, oral bacteria are rapidly killed by erythrocytes or filtered at the liver level so that transient bacteremia could induce a low-grade inflammatory liver response, and thus influence other tissues or organs by cytokine secretion (Minasyan, 2017b).

It is noteworthy to mention that recent studies have shown that the amount of bacteria present in the saliva is about 106/ml, which means that a person regularly swallows around 1012–1013 bacteria per day (von Troil-Lindén et al., 1995; Boutaga et al., 2007; Saygun et al., 2011; Arimatsu et al., 2014). After inoculating 10⁹ CFU of P. gingivalis by oral gavage, at 3 h, this bacterium was detected in the ileum and at 16 h in the colon, inducing gut dysbiosis shortly after being ingested (Arimatsu et al., 2014). Thus, the presence of periodontitis and the consequent swallowing of high loads of anaerobic bacteria could generate imbalances in the gut microbiota (Arimatsu et al., 2014). In this context, a particular cell type in the intestine, the enterochromaffin cells, permanently senses the invading pathogenic bacteria (Bellono et al., 2017). The dysbiosis-associated bacteria can be recognized by the enterochromaffin cells, which release cytokines and neurotransmitters to the afferent fibers of the vagus nerve and induce an intestinal sympathetic response (Figure 1) (Raybould et al., 2004).

Generally, the enterochromaffin cells are the most abundant endocrine cells in the intestine and are distributed widely from the stomach to the rectum (Gunawardene et al., 2011). The primary role of enterochromaffin cells is to synthesize, store, and secrete serotonin (Bellono et al., 2017). Also, enterochromaffin cells produce the corticotropin-releasing hormone, cholecystokinin, and somatostatin in response to virulence factors of pathogenic bacteria (Gershon and Tack, 2007; Hansen and Witte, 2008). In general terms, enterochromaffin cells have an essential role as a regulator of the secretion and motility of the intestine, and are considered a chemosensory to modulate neural pathways (Bellono et al., 2017). It has been demonstrated that these cells express on their surface the pituitary adenylate-cyclase-activating peptide, α-adrenergic, β-adrenergic, cholinergic, corticotropin-releasing hormone, and γ-aminobutyric acid (GABA) receptors, and can respond to neurotransmitters secreted by the gut microbiota (Gershon and Tack, 2007; Hansen and Witte, 2008). These cells also produce catecholamines, dynorphins, norepinephrine, and cytokines to the intestinal lumen in order to restore homeostasis, favoring the growth of commensal bacteria and inducing the death of pathogenic bacteria (Alonso et al., 2008). Any variability in the quality or quantity of the gut microbiota can be rapidly sensed by the enterochromaffin cells through Toll-like receptors (TLR) that recognize certain virulence factors (Wheatcroft et al., 2005; Bogunovic et al., 2007; Hansen and Witte, 2008). After stimulation, enterochromaffin cells secrete serotonin that the afferent nerves recognize to establish synapsis with these cells (Rhee et al., 2009). The vagus nerve fibers possess serotonin receptors and are located in the proximity of these cells (Raybould et al., 2004). In response, the vagus nerve increases intestinal motility and permeability, stimulates mucous secretion, induces diarrhea, and triggers intestinal inflammation (Mayer, 2000). Consequently, the intestinal sympathetic response allows macrophages and mast cells to migrate due to the increased intestinal permeability and bacteria invasion (Rhee et al., 2009). Finally, considering their dual role as endocrine and excitatory, the enterochromaffin cells have also been named neuropodal cells (Kaelberer et al., 2018).

Several studies using experimental periodontitis induced by P. gingivalis oral gavage demonstrated intestinal inflammatory events characterized by changes in the gut microbiota composition, intestinal barrier permeability, and the modulation of the intestinal immune response (Figure 2) (Arimatsu et al., 2014; Nakajima et al., 2015; Sato et al., 2017; Sato et al., 2018).

There is a close link between the enteric nervous system and the gut microbiota (Veiga-Fernandes and Pachnis, 2017). Indeed, the development of the enteric nervous system is modulated by the gut microbiota (Veiga-Fernandes and Pachnis, 2017). TLR2 and TLR4 are expressed on enteric nervous cells and, when activated, increase the expression of glial-derived neurotrophic factor (GDNF), released against bacterial challenge (Anitha et al., 2012; Heiss and Olofsson, 2019). During growth, development and maturation, the enteric neurons and glial cells constantly sense intestinal bacteria (Heiss and Olofsson, 2019). Also, the smooth muscle cells express TLR2, TLR3, TLR4, and TLR9, and together with enteric neurons, are involved in the response against bacteria (Anitha et al., 2012; Brun et al., 2013; Kobayashi et al., 2019). Enteric neurons and intestinal macrophages have established critical crosstalk that communicates the immune and nervous systems (Figure 3) (Müller et al., 2014). Intestinal macrophages are distributed in the lamina propria, submucosal layer, and muscular plexus of the intestine (Gabanyi et al., 2016). The muscle macrophage function depends on both exogenous signals such as virulence factors of bacteria, viruses, or fungi, and endogenous molecules, such as damage-associated molecular patterns (DAMPs) or stress molecules (Müller et al., 2014; Gabanyi et al., 2016). In the healthy intestine, muscle macrophages secrete bone morphogenic protein 2 —a growth factor for enteric neurons—, and maintain enteric nervous system homeostasis by engulfing senescent enteric neurons (Müller et al., 2014). The static position of muscle macrophages, primarily alongside neuronal cell bodies and nerve fibers, provides an interface for optimal crosstalk (Gabanyi et al., 2016). In addition, muscle macrophages express β2-adrenergic receptors on its surface, which allows them to have a neuroprotective role, similar to that described by microglia in the central nervous system (Figures 1, 3) (Huuskonen et al., 2004; Matteoli et al., 2014; Erny et al., 2015; Gabanyi et al., 2016; Ifuku et al., 2016). It is known that the β2-adrenergic signal blocked the immune response, while the α-adrenergic signal can stimulate it (Guereschi et al., 2013). The counterpart is due to the catecholamine release by enterochromaffin cells or neurons of the intestine. Also, the mesenteric lymph nodes are innervated by sympathetic fibers, which release epinephrine or nor-epinephrine to the immune cells that express β2-adrenergic receptors (del Rey and Besedovsky, 2008). These data suggest that the high production of norepinephrine in intestinal tissues—by the neurons from the celiac and superior-mesenteric ganglia—is related to the constant activation of muscle macrophages towards a modulatory phenotype of inflammation (M2) produced by macrophages residing in the lamina propria (Gabanyi et al., 2016). Thus, the macrophages of the lamina propria respond to bacteria or virulence factors entering the connective tissue, while the muscle macrophages respond to neuronal signals (Gabanyi et al., 2016). Besides, neutrophils and CD8⁺ T lymphocytes express more β2-adrenergic receptors than CD4⁺ T cells do not, so inflammation would be lacking of a Tregs response, which require more intense signals (del Rey and Besedovsky, 2008). Additional to macrophages, the mast cells in the gut submucosal and myenteric plexuses can respond to neuron-derived factors such as substance P, immune signals such as immunoglobulin (Ig)-E, and regulate both neuronal and immune cell activity through various mediators, including histamine, serotonin, and TNF-α release (Suzuki et al., 1995; van Diest et al., 2012).

The human gastrointestinal tract is colonized by 100 trillion microorganisms (Furness et al., 1999; Hill and Artis, 2010) and contains an effective neuroimmune barrier that constantly monitors and responds to potentially dangerous gut microbiota changes (Furness et al., 1999; Hill and Artis, 2010; Branzk et al., 2014). The intestinal barrier is in a dynamic balance, integrating inputs from the epithelium, immune cells, neurons of the enteric nervous system, enteric glia, and microbiota (Figures 1, 2, 3) (Rhee et al., 2009; Cryan and Dinan, 2012; Furness et al., 2014; Müller et al., 2014; Hergott et al., 2016; Strandwitz, 2018; Yu et al., 2020). This barrier is not impermeable, and some commensal species such as B. fragilis induce an immune response in the intestinal epithelium and thus control the bacterial gut balance (Zafar and Saier, 2021). The colonization of pathogenic bacteria induces the differentiation and local infiltration of Th17 subsets, which produce IL-17 and protect against infection by enteric pathogens (Ivanov et al., 2009). Commensal bacteria associated with the epithelium also activate the production of IL-22 and IgA in CD4⁺ T and plasma cells, respectively (Figure 2) (Hill and Artis, 2010; Gommerman et al., 2014; Ramirez et al., 2020). In the connective tissue, the bacteria are also recognized by infiltrating dendritic cells that migrate to the mesenteric lymph node and trigger the differentiation toward Th17 subsets (Hill and Artis, 2010). Besides, the activation of TLR2 promotes the expression of S100β1, inducible nitric oxide synthase, and GDNF in the enteric glia, which can promote IL-22 release (Brun et al., 2013; Benarroch, 2019). Several metabolites of intestinal bacteria are relevant not only for intestinal protection but also for the microbiota interactions with the immune and nervous systems (Dass et al., 2007; Jenkins et al., 2016; Szentirmai et al., 2019). The short-chain fatty acids (acetate, propionate, and butyrate) derived from the anaerobic fermentation of indigestible carbohydrates play an important role in counteracting inflammation and maintaining intestinal homeostasis (Figure 2) (Dass et al., 2007; Smith et al., 2013). Lactobacilli spp metabolizes dietary tryptophan in humans and generates indole ligands of aryl hydrocarbon receptors, expressed in Th22 cells (Trifari et al., 2009; Rothhammer et al., 2016). Also, several microbiota members produce neurotransmitter molecules and neuropeptides, such as serotonin, dopamine, GABA, and brain-derived neurotrophic factor (Strandwitz, 2018). Thus, bacterial products can stimulate epithelial cells to release molecules of signaling that will regulate the function of the enteric nervous system, establishing a communication between the gut microbiota and the brain.

Gut bacteria can interact with the brain by communicating with the mesenteric and vagus nerve afferent fibers (Rhee et al., 2009). Interestingly, some bacteria produce and secrete different neurotransmitters, inducing a nerve impulse in the neurons underlying the intestinal epithelium (Strandwitz, 2018; Cui et al., 2020). Commensal bacteria interact with enterocytes through the α2-receptor and can either increase or decrease the ability to eliminate pathogenic bacteria in the intestinal lumen (Rhee et al., 2009). Gut bacteria interact directly with afferent nerves in the presence of intestinal permeability caused by inflammation or stress (Turnbaugh et al., 2006; Bravo et al., 2012; Foster and McVey Neufeld, 2013; Sarkar et al., 2016). Besides, the enterochromaffin cells in the intestine can translocate bacteria from the lumen to the afferent neurons (Rhee et al., 2009).

Interestingly, it was proposed that the recomposition of the microbiota, in particular, the Bifidobacterium and Lactobacillus genera, would allow the regulation of serotonin levels in the brain and the re-establishment of the intestinal homeostasis by parasympathetic stimulation (Mayer, 2000; Leonard, 2006; Turnbaugh et al., 2007; Rhee et al., 2009). Thus, the existence of direct and bidirectional communication between the gut and the brain is recognized (Clarke et al., 2013; Sarkar et al., 2016; Müller et al., 2020). Indeed, the microbiota would play a neurodevelopmental role (Clarke et al., 2013). Germ-free animals have an exaggerated stress response compared to usually colonized animals, and stress is reversed by reconstituting the microbiota (Sudo et al., 2004). The incorporation of Bifidobacterium or Lactobacillus can benefit stressful or depressive alterations in both health conditions or illness (Desbonnet et al., 2008). In fact, probiotics can increase the bioavailability of tryptophan, a precursor of serotonin (Desbonnet et al., 2008).

One of the theories that explain the communication between the gut and the brain is the existence of parallel outputs (Rhee et al., 2009). The sympathetic and parasympathetic nerve fibers, the hypothalamic-pituitary-adrenal axis, and the endogenous pathways that regulate pain are the key regulators of gastrointestinal function (Rhee et al., 2009). These three parallel outputs can alter the microbiota by modifying the environment or by microbiota-host signaling. First, the sympathetic and parasympathetic systems regulate intestinal motility, secretion of acid, HCO3-, mucus, and the immune response (Mayer, 2000). As there is a stimulation of the sympathetic response, it can increase intestinal motility, affecting the processing of nutrients by the gut microbiota. On the contrary, a parasympathetic response decreases motility, associated with bacterial overgrowth in the small intestine (van Felius et al., 2003).

Second, enterochromaffin cells secrete catecholamines, serotonin, norepinephrine, dynorphin, and cytokines into the intestinal lumen in response to changes in both the concentration of nutrients or the bacteria’s balance up the gut microbiota (Rhee et al., 2009). Although it is not yet well understood how bacteria interact with the nervous system, it has been proposed that the concentration of short-chain fatty acids is directly related to the presence of a symbiotic microbiota (Müller et al., 2020). Although bacteria constitute a crucial node in the gut-brain axis’s bidirectional relationship, it has also been called the microbiota-gut-brain axis (Sarkar et al., 2016). In general terms, gut dysbiosis produced by oral bacteria derived from oral diseases could be linked to neuroinflammatory events in the brain through this interaction.

The gut microbiota and immune system are involved in gut-brain communication (Sharon et al., 2016; Shi et al., 2017). The pathogens-associated molecular patterns (PAMPs), such as LPS, activate enteric glial cells and neural afferents fibers (Pochard et al., 2018). The vagal and dorsal root ganglion afferent fibers innervating the intestine express TLRs and cytokine receptors, communicating local intestinal signals to the central nervous system (Furness et al., 2014). Studies on germ-free intestines of mice treated with antibiotics indicate that the products of gut microbiota are potent regulators of immune response in the central nervous system (CNS) and are essential for intestinal barrier maintenance (Hernández-Chirlaque et al., 2016). In this sense, the presence of short-chain fatty acids in the bloodstream promotes microglial maturation and differentiation, and the uptake increase of tryptophan metabolites activates aryl hydrocarbon receptors which modulate the astrocyte activation during inflammation (Erny et al., 2015; Rothhammer et al., 2016). Also, the small chain fatty acid reduces the expression of pro-inflammatory factors, including C-C chemokine ligand (CCL)-2, IL-6, TNF-α, and nitric oxide synthase in both astrocytes and microglia (Huuskonen et al., 2004; Cho et al., 2019).

Peripheral signals from immune cells and microbial products can also be transmitted to the CNS via sensory vagal neurons of the nodose ganglion and by nociceptive and visceroceptive neurons of the dorsal root ganglion (Ordovas-Montanes et al., 2015; Yu et al., 2020). The gastrointestinal tract also contains intrinsic primary afferent neurons that initiate local reflexes, contribute input signals to the CNS, and participate in host-microbe interactions (Lai et al., 2017). In addition to the receptors of voltage-dependent excitatory cells, these afferent nervous fibers express TLRs and receptors for cytokines, which include IL-1β, IL-6, IL-17, TNF-α, and prostaglandins, as well as other receptors for molecules released by immune cells (Ramirez et al., 2020). One of the most important effects of TLR priming in nerve fibers is the decrease in the excitatory threshold (Ramirez et al., 2020). Thus, the nerve fibers of the intestine can respond to both bacterial virulence factors or cytokines released by immune cells after pathogenic bacteria priming, making the gut-brain axis a most complex communication (Ramirez et al., 2020). After crosstalk between sensory neurons and immune cells, neurons modulate tissue inflammation through the release of substance P, the peptide related to the calcitonin gene, the vasoactive intestinal peptide, and other interacting molecules (Lai et al., 2017). The substance P induces immune cells to secrete pro-inflammatory cytokines and contributes to tissue repair (O’Connor et al., 2004). The peptide related to the calcitonin gene is present in the terminals that innervate Peyer’s patches and regulates maturation, proliferation, migration, antigen presentation, and cytokine production by lymphocytes (Lai et al., 2017). The abdominal vagal afferent fibers express receptors for IL-1β and are activated by the systemic administration of IL-1β (Ek et al., 1998; Holzer et al., 2004). Finally, the vagal afferent fibers transmit peripheral signals to the solitary tract nucleus, which relays this information to the brainstem and anterior areas of the brain that contain neurons that regulate immune and inflammatory responses (Cutsforth-Gregory and Benarroch, 2017).

The oral cavity is one of the main entry routes for microorganisms, mainly bacteria (Li et al., 2000; Loesche and Lopatin, 1998). Some studies have detected P. gingivalis in the bloodstream after tooth brushing, flossing, or chewing food, in subjects with or without periodontitis, causing transient bacteremia (Li et al., 2000; Loos, 20015; Forner et al., 2006). This event allows oral bacteria to migrate and establish themselves in other tissues or organs, such as the intimal layer of the coronary arteries, the liver, or the placenta (Marcelino et al., 2010; Vanterpool et al., 2016; Komazaki et al., 2017; Udagawa et al., 2018; Fischer et al., 2019). Indeed, a recent study showed that the most prevalent bacterium colonizing the intima layer of both coronary and femoral arteries was P. gingivalis (Mougeot et al., 2017). Once in the bloodstream, bacteria induce an acute liver inflammatory phase response characterized by increased pro-inflammatory cytokines that could enter or, even without entering, influence brain function (Wu and Nakanishi, 2014; Wu et al., 2017; Ding et al., 2018; Ilievski et al., 2018; Zhang et al., 2018). Although the blood-brain barrier (BBB) generally prevents substances from entering the brain, molecules such as cytokines can enter through capillaries from the circumventricular organs by using specific cytokine transporters, increasing BBB’s permeability through transporters of brain endothelial cells (Pan and Kastin, 1999; Banks et al., 2002; Perry, 2004). In patients with AD, a BBB breakdown was associated with cognitive decline and inflammation (Bowman et al., 2018). Thus, the first potential link between oral and brain might be due to the breakdown of the BBB induced by the low-systemic inflammatory mediators in the bloodstream (Figure 4) (Hawkins and Davis, 2005).

No scientific study reports an association between the oral cavity, the gut, and the brain. Approaches to this possible link may lie in the existing evidence between the oral and gut microbiota with the brain through the antecedents described above (Figure 6). Also, there is rising evidence that suggests oral dysbiotic microbiota can lead a gut dysbiosis, and together, trigger neuroinflammation. When evaluating the experimental periodontitis induced by ligature, some studies detected increased levels of serum α amyloid (SAA), IL-6, vascular endothelial growth factor (VEGF), receptor of activator NF-κB ligand (RANKL), CCL5, CXCL10, and B-cell activating factor (BAFF) at the serum level (Arimatsu et al., 2014; Matsuda et al., 2015; Komazaki et al., 2017; Jia et al., 2018; Palioto et al., 2019; Hamamoto et al., 2020). Curiously, two studies demonstrated that gut microbiota dysbiosis or alteration of the intestinal epithelium’s integrity was not induced, in comparison with others that did show the presence of these phenomena (Matsuda et al., 2015; Jia et al., 2018; Palioto et al., 2019; Huang et al., 2020). An explanation for this difference may be in the observation time, while one study installed the ligature for 5 days (Palioto et al., 2019), another group did it for 10 days (Matsuda et al., 2015), and two independent research groups did it for 28 days (Jia et al., 2018; Huang et al., 2020). The evidence indicates that ligature creates measurable bone resorption and pro-inflammatory mediators from day ten onwards, together with gut microbiota dysbiosis (Garlet, 2010; Matsuda et al., 2015; Palioto et al., 2019).

Additionally, the two independent studies evaluate the association between experimental periodontitis with AD or Parkinson’s Disease-like pathologies, using different models to induce bone resorption and enough inflammatory response to trigger neurodegeneration (Feng et al., 2020; Kantarci et al., 2020; Xue et al., 2020). By characterizing gut microbiota dysbiosis induced by experimental periodontitis at day 28 post-ligation, it is possible to detect alterations in the Firmicutes/Bacteroidetes ratio, which is essential to determine intestinal inflammation (Ley et al., 2005; Koren et al., 2011; Nakajima et al., 2015). Thus, experimental periodontitis can trigger gut dysbiosis after 28 days, and the neuroinflammatory events can be related to both periodontitis and gut dysbiosis effects. However, and as we have described throughout this article, it is unclear whether periodontitis induces neuroinflammation due to oral dysbiosis, gut dysbiosis, or both.

Additionally, when evaluating the different studies that induced periodontitis through oral gavage or ligation, intending to demonstrate the presence of gut microbiota dysbiosis, it was observed that only six studies confirmed the presence of periodontitis (Arimatsu et al., 2014; Matsuda et al., 2015; Jia et al., 2018; Sato et al., 2018; Palioto et al., 2019; Huang et al., 2020). All of them demonstrated the alteration of the gut microbiota, intestinal integrity loss, or pro-inflammatory immune response. Specifically, experimental periodontitis trigger gut microbiota dysbiosis by Clostridium, Firmicutes, Bacteroidetes or Lactobacillus spp quantitative changes (Arimatsu et al., 2014; Nakajima et al., 2015; Jia et al., 2018; Sato et al., 2018; Ohtsu et al., 2019; Hamamoto et al., 2020; Huang et al., 2020; Kobayashi et al., 2020). Besides, some of them detected inflammatory mediators such as IL-1β, IL-6, IL-17, TNF-α, among others, in serum, adipose tissue, and small or long intestine, or liver (Arimatsu et al., 2014; Matsuda et al., 2015; Komazaki et al., 2017; Jia et al., 2018; Palioto et al., 2019; Hamamoto et al., 2020). When evaluating the Th17/Treg lymphocytes or M1/​​M2 macrophages ratio, only two studies analyzed the presence of these cells in the mesenteric or cervical lymph nodes and the intestine (Sato et al., 2018; Kobayashi et al., 2020). Finally, three studies evaluated the secondary effect of the application of oral bacteria, determining the presence of insulin resistance or increases in serum triglyceride levels (Arimatsu et al., 2014; Komazaki et al., 2017; Ohtsu et al., 2019). P. gingivalis in the intestine was associated with a higher M1/​M2 ratio, triggering receptor expressed on myeloid cells 1 (TREM1) and NF-κB expression (Kobayashi et al., 2020). The increased production of IL-1β and IL-17 produces a decrease in zo-1 and dopaminergic neurons degeneration (Ding et al., 2017; Feng et al., 2020; Kobayashi et al., 2020). Interestingly, IL-17 can cross the BBB and induce apoptotic death of dopaminergic neurons that possess the IL-17RA (Kebir et al., 2007; Arimatsu et al., 2014; Nakajima et al., 2015). IL-17A is a cytokine that allows microglia, astrocytes, and neurons to communicate constantly, and the IL-17A/IL-17RA pathway increases the production of pro-inflammatory mediators, chemokines, and antimicrobial peptides production (Chen et al., 2020). Considering that during periodontitis, the Th17 lymphocytes producing IL-17 plays a central role as the main activator of osteoclasts and in the breakdown of the BBB, the levels of IL-17 in trigeminal ganglia or hippocampus should be studied in the future (Kebir et al., 2007; Vernal et al., 2014b; Monasterio et al., 2018a; Monasterio et al., 2018b).

Given that all the models used until today use different forms and concentrations to inoculate the bacteria or induce periodontitis, it is challenging to compare them. Although all demonstrate a cognitive deterioration to a greater or lesser extent, it is debatable whether these effects are a consequence of periodontitis, bacteria, pro-inflammatory cytokines, BBB breakdown, brain bacteria colonization, or gut microbiota dysbiosis. Thus, further studies are necessary to understand the so-called Oral-Brain axis. In the present review, the collected evidence indicates that the Oral-Brain axis may be due to the bacterial-trigeminal nerve interaction through the TLR4/CD14-MyD88-NF-κB pathway, which can be driven in a bi-directional manner (Figure 4B) (Wadachi and Hargreaves, 2005; Diogenes et al., 2011; Go et al., 2016).

Also, it remains to be demonstrated if the neuroinflammation induced by experimental periodontitis is not due to gut microbiota dysbiosis. This, because both ligation and oral gavage models produce gut microbiota dysbiosis through the Oral-Gut axis (Arimatsu et al., 2014; Matsuda et al., 2015; Jia et al., 2018; Sato et al., 2018; Palioto et al., 2019; Huang et al., 2020). Although palatal inoculation would avoid the dysbiotic effect on the gut microbiota, its effects at the periodontal and brain level would be through the Oral-Brain axis (Díaz-Zúñiga et al., 2020). However, it was not demonstrated if palatal inoculation of a single bacterium induced gut microbiota dysbiosis. Future studies using experimental periodontitis models and trying to elucidate the hypothesis mentioned above should consider the measurement of bacteria or mediators at the intestinal level to confirm or rule out the gut microbiota dysbiosis.

On the other hand, it is interesting that two studies that induced periodontitis by ligation or oral gavage observed two different results (Kantarci et al., 2020; Feng et al., 2020). One of them detected microglial activation in the hippocampus and AD-like pathology, and the other detected microglial activation in the substantia nigra pars compacta and a PD-like pathology (Kantarci et al., 2020; Feng et al., 2020). Together, these data allow us to speculate that periodontitis may be related to proteinopathies, including AD and PD and maybe others (Kwon et al., 2020). If so, periodontitis must be reconsidered as a public health problem, being not only the leading cause of tooth loosening but also a low-grade inflammatory disease that could predispose to relevant diseases affecting the brain.

The present review demonstrates that experimental periodontitis in different models can induce gut microbiota dysbiosis, affect the intestinal barrier permeability, induce an intestinal immune response, and trigger neuroinflammatory and neurodegenerative diseases. Secondly, there is evidence suggesting that oral bacteria can be recognized by trigeminal fibers and induce a response at the trigeminal ganglion level. Also, oral bacteria can trigger a sympathetic response capable of stimulating more osteoclastogenic functions.

In this way, the Oral-Gut-Brain axis could be defined based on anatomical communications, where the mouth and the intestine are in constant cross-talking. The oral-brain axis is mainly established from the trigeminal nerve and the gut-brain axis from the vagus nerve. In this context, the immune system is transversal, and an increase in inflammatory mediators at the oral or intestinal level will immediately affect brain homeostasis and vice-versa. Through the qualitative analysis of the selected papers, we observed that experimental periodontitis is capable of producing both neurodegenerative pathologies and intestinal dysbiosis, and periodontitis is likely to induce both conditions simultaneously. The severity of the neurodegenerative disease could depend, at least in part, on the effects of periodontitis in the gut microbiota, which could strengthen the immune response and create an injurious inflammatory and dysbiotic cycle. Thus, dementias would have their onset in dysbiotic phenomena that affect the oral cavity or the intestine.