09-Assessing the role of Porphyromonas gingivalis in periodontitis to determine a causative relat - Enfermagem (2024)

Prévia do material em texto

Full Terms & Conditions of access and use can be found athttps://www.tandfonline.com/action/journalInformation?journalCode=zjom20Journal of Oral MicrobiologyISSN: (Print) 2000-2297 (Online) Journal homepage: www.tandfonline.com/journals/zjom20Assessing the role of Porphyromonas gingivalis inperiodontitis to determine a causative relationshipwith Alzheimer’s diseaseSim K. Singhrao & Ingar OlsenTo cite this article: Sim K. Singhrao & Ingar Olsen (2019) Assessing the role of Porphyromonasgingivalis in periodontitis to determine a causative relationship with Alzheimer’s disease,Journal of Oral Microbiology, 11:1, 1563405, DOI: 10.1080/20002297.2018.1563405To link to this article: https://doi.org/10.1080/20002297.2018.1563405© 2019 The Author(s). Published by InformaUK Limited, trading as Taylor & FrancisGroup.Published online: 29 Jan 2019.Submit your article to this journal Article views: 9505View related articles View Crossmark dataCiting articles: 28 View citing articles https://www.tandfonline.com/action/journalInformation?journalCode=zjom20https://www.tandfonline.com/journals/zjom20?src=pdfhttps://www.tandfonline.com/action/showCitFormats?doi=10.1080/20002297.2018.1563405https://doi.org/10.1080/20002297.2018.1563405https://www.tandfonline.com/action/authorSubmission?journalCode=zjom20&show=instructions&src=pdfhttps://www.tandfonline.com/action/authorSubmission?journalCode=zjom20&show=instructions&src=pdfhttps://www.tandfonline.com/doi/mlt/10.1080/20002297.2018.1563405?src=pdfhttps://www.tandfonline.com/doi/mlt/10.1080/20002297.2018.1563405?src=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1080/20002297.2018.1563405&domain=pdf&date_stamp=29 Jan 2019http://crossmark.crossref.org/dialog/?doi=10.1080/20002297.2018.1563405&domain=pdf&date_stamp=29 Jan 2019https://www.tandfonline.com/doi/citedby/10.1080/20002297.2018.1563405?src=pdfhttps://www.tandfonline.com/doi/citedby/10.1080/20002297.2018.1563405?src=pdfREVIEW ARTICLEAssessing the role of Porphyromonas gingivalis in periodontitis to determine acausative relationship with Alzheimer’s diseaseSim K. Singhraoa and Ingar OlsenbaDementia and Neurodegenerative Diseases Research Group, Faculty of Clinical and Biomedical Sciences, School of Dentistry, Universityof Central Lancashire, Preston, UK; bDepartment of Oral Biology, Faculty of Dentistry, University of Oslo, Oslo, NorwayABSTRACTChronic periodontitis of 10 years’ duration is reported to become a twofold risk factor for thedevelopment of Alzheimer’s disease (AD). Periodontitis is modifiable, and this fits with thecurrent action plan for preventing AD. However, until periodontitis, becomes acknowledgedas a firm risk factor for AD, this risk will continue. Here, we put forward our own argumentbased on the current literature for in vivo infection-mediated periodontal disease modelssupporting the antimicrobial protection hypothesis of AD and interventional studies support-ing the causal links. Oral infections with Porphyromonas gingivalis, or introduction of itslipopolysaccharide (LPS), in various mouse models has demonstrated the development ofkey neuropathological hallmark lesions defining AD. These are extracellular amyloid-betaplaques, phosphorylated tau, neurofibrillary tangles, widespread acute and chronic inflam-mation, blood–brain barrier defects together with the clinical phenotype showing impairedlearning and spatial memory. Live P. gingivalis and its LPS (commercial or from ‘microbullets’)are powerful peripheral and intracerebral inflammatory signalling initiators, and this hasdirect implications on memory and lesion development. Maintaining a healthy oral micro-biome and managing periodontal disease with regular surveillance and good oral hygienethroughout life is likely to reduce the unnecessary burden of AD in some individuals.ARTICLE HISTORYReceived 20 August 2018Revised 5 December 2018Accepted 13 December 2018KEYWORDSAlzheimer’s disease; chronicperiodontitis; cause;infection; P. gingivalis;lipopolysaccharide riskfactor; interventionIntroductionPeriodontitis is a highly prevalent oral disease in humansaffecting nearly 50% of the population worldwide [1].The polymicrobial infectious aetiology of chronic period-ontitis is due to the host’s sub-gingival pathobiome,which initiates hard/soft tissue destruction that worsenswith advancing age. Furthermore, the dysbiotic oral bio-film consortia can affect the functioning of the brain,potentially causing depressive illnesses, as implicated inthe development of dementia [2]. Risk factors for period-ontal disease are smoking [3], alcohol consumption [4],and poor oral hygiene [5]. Paganini-Hill et al. [6] high-lighted that behavioural factors involving oral hygienewere significant in dementia onset, stating that, dentateindividuals who did not brush their teeth daily, had a22–65% greater risk of developing dementia comparedwith individuals who brushed their teeth three timesdaily [6]. These statistics suggest a proportion of indivi-duals are particularly susceptible to Porphyromonas gin-givalis infections, because not everyone suffers fromperiodontitis and not all who develop Alzheimer’sdementia suffer from periodontitis [7]. Periodontal dis-ease is modifiable by both professional intervention andpersonal behavioural changes associated with oralhygiene [8,9], and this offers an avenue for reducingunnecessary mental health suffering for some individualsin their old age. In addition, therapeutic elements ofprevention and treatment of dementia with a view toanti-P. gingivalis therapy are being sought. For example,Cortexyme Inc®, a USA based company is seeking theirlead compound, COR388; to treat dementia in Phase 1clinical trials. Such preventive measures are vital forwhen standard periodontal therapy becomes a challengefor both the patient (the ‘vulnerable category’ of patientsaccording to the mental health act), and the treatingdentist [www.cortexyme.com. https://www.cdc.gov/chronicdisease/resources/publications/aag/alzheimers.htm]Alzheimer’s disease (AD), the most common formof dementia, is the leading cause of cognitive andbehavioural impairment worldwide [10]. As theelderly population, keeps increasing so does the inci-dence of AD manifesting in two different forms:familial and sporadic. The latter form is most fre-quent, constituting about 95% of the cases but itscause remains open to debate. Both forms have iden-tical neuropathological hallmarks, which are accumu-lations of hyper-phosphorylated tau composed ofneurofibrillary tangles, and extracellular amyloid-beta (Aβ) deposits called ‘amyloid plaques’. Tau pro-tein is prone to hyper-phosphorylation at serine andthreonine residues due to the activity of multipleCONTACT Sim K. Singhrao SKSinghrao@uclan.ac.uk Dementia and Neurodegenerative Diseases Research Group, Faculty of Clinical andBiomedical Sciences, School of Dentistry, University of Central Lancashire, Preston, UKJOURNAL OF ORAL MICROBIOLOGY2019, VOL. 11, 1563405https://doi.org/10.1080/20002297.2018.1563405© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permitsunrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.http://www.cortexyme.comhttps://www.cdc.gov/chronicdisease/resources/publications/aag/alzheimers.htmhttps://www.cdc.gov/chronicdisease/resources/publications/aag/alzheimers.htmhttps://www.cdc.gov/chronicdisease/resources/publications/aag/alzheimers.htmhttp://www.tandfonline.comhttp://crossmark.crossref.org/dialog/?doi=10.1080/20002297.2018.1563405&domain=pdfkinase enzymes orchestrating various signalling path-ways for normal and key infection-related cellularfunctions [11–13]. Managing AD is a financial andmedical challengeworldwide and prevention viamodifiable factors is one of the key ways to avoidand/or decelerate progression of this disease [8,9].Since the acceptance of any hypothesis explainingthe cause of AD must involve the two hallmark pro-teins (Aβ and phosphorylated-tau tangles), and inter-ventional studies in humans having tested beneficialoutcomes, we asked the question, how does sub-gin-gival dysbiosis under the influence of P. gingivalis, thekeystone pathogen of periodontal disease, relate tothe cause of AD?The main risk factors for the sporadic form of AD,which correlates with chronic periodontitis, areadvancing age and loss of up to nine teeth [14].Another study in which patients with chronic period-ontitis and/or gingivitis, were monitored for 10 yearsdemonstrated an increased risk for dementia (1.13%)compared with the control (0.92%) group [15]. Whileperiodontitis correlates with other inflammatorypathologies such as cardiovascular diseases [16–18]and diabetes [19], there is overlap of AD withischemic stroke disease [20] and insulin resistanceor as some scientists like to call it, type 3 diabetes[21]. Periodontal disease is modifiable, and this pro-vides the rationale for assessing causative associationswith AD. The focus here is on summarising keyepidemiological studies that have established thetimeline and size of risk, as well as research on invivo infection with P. gingivalis and its LPS support-ing AD clinicopathological causal links and interven-tional trials showing clinical benefits.10-year exposure to chronic periodontitisdoubles the risk for ADAn epidemiological study by Kondo et al. [22] pro-vided the initial concept, which proposed that personswith premature tooth loss were a significant risk factorfor AD. Gatz et al. [23], investigated the risk of devel-oping AD due to tooth loss in identical twins. For thisunivariate human model, tooth loss around 35 years’age provided an odds ratio of 1.74 (95% CI 1.35, 2.24)for developing AD. Stein et al. [14] linked periodonti-tis to cognitive deficit thereby proposing periodontaldisease as a risk factor for developing AD later in life.They concluded that missing up to nine teeth carriedthe highest risk for developing late-onset AD with anodds ratio of 2.2 (95% CI 1.1, 4.5). A retrospectivestudy conducted by Chen et al. [24] found a stronglink between chronic periodontal disease (exposures ofaround 10-years) and AD and this correlates with aprospective laboratory-based study in which circulat-ing antibodies to two oral bacteria (Fusobacteriumnucleatum and Prevotella intermedia) were linked toa cognitive deficit 10 years later [25]. In summary,Tzeng et al. [15], Chen et al. [24] and Sparks Stein etal. [25] suggested that gingivitis, and chronic period-ontal disease of 10 years duration can promote AD.Furthermore, Stein et al. [14] showed tooth loss due toperiodontal disease can double the risk for AD onset.Concept of systemic inflammationcontributing to memory lossDunn et al. [26] noted (from medical records), theoccurrence of repeated systemic infections in elderlysubjects prior to their clinical diagnosis of dementia.This led to the assumption that inflammation plays anegative role on the health of the brain. With focus onperipheral inflammation, Holmes et al. [27] suggestedthat circulating systemic inflammatory markers (cyto-kines) were negatively influencing memory in sporadicAD cases. Further studies examined the links withspecific periodontal pathogens to the functional cog-nitive loss seen in clinical AD cases. To this end, Nobleet al. [28] found that P. gingivalis infection was asso-ciated with impaired spatial/episodic memory in ADwith an odds ratio of 2.00 (95% CI 1.19 to 3.36) afteradjusting for confounders. Subsequent studiesfocussed on detecting acute phase inflammatory med-iators in the plasma of blood taken from confirmedAD cases in relation to periodontal pathogens/period-ontitis and confirmed systemic inflammatory markercontribution from oral bacteria [29–31].Ide et al. [31] set out to test the hypothesis thatcirculating inflammatory cytokines due to periodontaldisease bacteria were associated with greater rates ofcognitive decline in clinical AD cases. The studyrecruited 59 participants with mild to moderate AD inwhich cognition and circulating inflammatory markerswere tested. The majority of participants (52) were fol-lowed-up at 6 months when they all underwent repeatassessment of their initial biomarkers. The study revealedthat the presence of periodontitis at baseline was asso-ciated with a six-fold increase in the rate of cognitivedecline in participants over the 6-month follow-up per-iod. Periodontitis at baseline was also associated with arelative increase in the pro-inflammatory state over thesix-month follow-up. The authors concluded that peri-odontitis is associated with an increase in cognitivedecline in AD. In this study, the scientific hypothesislinking cognitive decline with the body’s inflammatoryresponses was supported. The weaknesses of the studywere the small number of participants and absence ofcontrol participants with intact cognition.Host's microbiome dysbiosis is an importantenvironmental factorA symbiotic oral microbiome plays a role in healthyliving and successful ageing [32,33]. However, a2 S. K. SINGHRAO ET AL.pathobiome prevails where pathogenic bacteria havechanged their original commensal status, as with P.gingivalis-associated periodontal disease being a typicalexample. A pathobiome in any part of a host’s bodyrepresents an environmental risk factor. Although aber-rant infections such as those linking Lyme disease(Borrelia burgdorferi transmitted by a bite from infectedticks, and syphilis (also known as an atrophic form ofgeneral paresis caused by Treponema pallidum), even-tually lead to AD [34,35], they represent a separate riskfactor to the so-called environmental risk that isrestricted to pathobiomes and lifestyle choices. This dis-tinction is important in view of the apolipoprotein geneallele 4 (APOE ε4) as a susceptibility gene that interactswith environmental risk factors such as a sub-gingivalpathobiome that results in periodontal disease and/orcombined with smoking, poor choices of diet and asedentary lifestyle. This will likely enhance its biologicalfunction in favour of AD. Hence, the current discussion,where possible, is limited to P. gingivalis infectionsbecause its interactome shows overlaps with AD suscept-ibility genes making this bacterium an excellent candi-date for confirming the environmental risk factor status[2]. With the demonstration of P. gingivalis LPS exclu-sively in AD brains, Poole et al. [36] have provided therationale for in vivo proof of concept studies.The (simplified) amyloid cascadeThe insoluble Aβ deposits (amyloid plaques) in theAD brain [37] are the consequence of amyloidprecursor protein (APP) proteolysis along the Nterminus (start of the protein) to the cytoplasmictail at the C terminus (end of the amino acid chainterminated by a free carboxyl group). The enzymesgenerating Aβ are known as beta-secretase 1 orBACE 1, which couples with γ-secretase in thefamilial form of AD [38–40]. BACE 1 in this con-text therefore, recognizes the cleavage site of themutated (mt)APP gene in the familial form of ADand results in enhanced Aβ production [41]. Thisgenetic trait is the basis for generating transgenicmouse models for evaluating human AD. However,APP in the sporadic form of AD is not mutated[42], and the results of infections can vary accord-ing to the genetic make-up of the host animal. Thisfact has to be considered by researchers whenselecting animal models to test their hypothesisand by readers when comparing experimental out-comes. Whilst, Aβ40 is the most prominent species(80–90 %) found in AD brains, the amyloidogenicAβ42, overall represents the lesser component(5–10 %) [37,43]. Other species ofAβ fibrils(Aβ39, 38, 34, 33) also occur in the AD brain buttheir presence is generally neglected [44] for rea-sons poorly understood.AD-transgenic mice support experimentalperiodontitis as a nominal riskCurrently, there is only one report that employed theAPP-transgenic model (APP-Tg) carrying the Swedishand Indiana mutations [45] infected with P. gingivalisto assess the role of periodontitis in the developmentof AD hallmark pathology (Table 1). After the leadauthor’s communication with the senior author ofreference [45], the experimental regime was clarifiedas the one in their published article was somewhatmisleading. It is therefore time to correctly state thatIshida et al. [45], induced experimental periodontitisvia an oral, mono-infection with live P. gingivalisATCC 33277T. This is a seronegative strain commonlyused by global scientists in laboratory investigations.The mice in this study were infected every other daywith 1 × 109 CFU over the first 10 days (five infec-tions/10 days), and sacrificed five weeks after the firstinfection. They noted enhanced deposition of Aβ40,and Aβ42 amyloid plaques in the hippocampus andhigher levels of IL-1β and TNF-α in the infected APP-Tg group compared to the control/sham-infectedgroup. Despite showing statistically significant differ-ences in the Aβ40 and Aβ42 amyloid plaques (deter-mined by image analysis) and protein (determined byELISA assay), Ishida et al. [45] concluded that fivemono-P. gingivalis infections over the first 10 days ofthe experiment were likely to have exacerbated thedisease process rather than having contributed to theoverall Aβ hallmark pathology. Behavioural testingdemonstrated that the cognitive function was signifi-cantly impaired in periodontitis-induced APP-Tg micecompared to the sham-infected group. A mechanisticexplanation for the greater cognitive deficit in theinfected APP-Tg group was an increased intracerebralinflammation following experimental periodontitis.In vitro findings of Mueller-Steiner et al. [44] withAβ42 peptide incubation with cathepsin B (a lysosomalenzyme that can mimic the enzymatic activity of BACE1 in wild type APP), suggests that Aβ42 can be degradedto Aβ40 and Aβ38. Taken together, the P. gingivalis-infected APP-Tg group of the Ishida et al. [45] studyresulting in greater total Aβ load detected with their82e1 antibody may have been the product of Aβ42truncation. Therefore, Aβ oligomer distinction wouldhave been a better way to evaluate plaque data in theinfected group. Another weakness of this model wasthat the method of entry of P. gingivalis into the mousebrain remained unclarified. However, greater amountsof LPS in the infected mouse brains were recordedimplying that this P. gingivalis virulence factor hadreached the brain and was responsible for stimulatingintracerebral antigen presenting cells (glia) to upregu-late cytokine expression and liberation. However, func-tional tests provide useful causal links with the ADphenotype even in the familial form of AD.JOURNAL OF ORAL MICROBIOLOGY 3Mouse models with wild type APP support P.gingivalis as a risk factor for ADIn contrast to the familial form of AD, the sporadicform results mainly from the enzymatic processing ofthe wild type (wt)APP. Cathepsin B coupled with thecommon γ-secretase activity releases Aβ [42,46,47].Both BACE 1 (mtAPP) and cathepsin B (wtAPP)cleave at their specific amino acid motifs [46,48–51].The insoluble Aβ amyloid plaques are unequivocallyextracellular, but it is not clear if their precursorprotein processing is extracellular or intracellular orboth. Hook et al. [46] hypothesized intracellular pro-cessing of wtAPP via the regulated secretory vesiclepathway mediated by cathepsin B and its exit fromthe neuron (Figure 1). Unlike the mtAPP, thatimplies extracellular processing by BACE 1 resultingin enhanced Aβ, cathepsin B processing of wtAPPappears to be intracellular yielding less Aβ.Neuropathology diagnosis relies on a specific thresh-old of Aβ amyloid plaques in both forms of AD; thissuggests that the sporadic form is a result of defectiveclearance whilst the familial form results fromenhanced deposition of this insoluble proteinaccounting for time differences in their early (famil-ial) and late (sporadic) onsets. In this respect, theTakayama et al. [52] study links P. gingivalis tosleep disturbances and subversion of microglial cellfunction, and paves the way for Aβ build-up.P. gingivalis is an intracellular bacterium thatenters lysosomal bodies of cells following an endocy-tic entry and the trans-Golgi network via the endo-plasmic reticulum to avoid immune surveillance andextend its viability [53]. Rather surprisingly, P. gingi-valis and its LPS can manipulate endosomal/lysoso-mal enzymes [54] to degrade wtAPP, intracellularly(Figure 3), whilst IL-1β cytokine [55] provides theinflammophilic sustenance (Figure 3). Since IL-1β ispart of the inflammasome assembly, P. gingivalis canmodify related activity in several ways [56]. Amongthem are ATP/P2X7-signalling molecules, which areassociated not only with periodontitis but also withthe development of several systemic diseases [55,56].Both TLR-2 and NLRP3 can recognize a functionalbacterial amyloid known as curli, within Aβ amyloidplaques [57] and AD brain Aβ deposits transduce theactivation of the NLRP3 inflammasome in microglialcells in vitro and in vivo [58–61].Wild type mouse model of experimentalperiodontitis supports neuroinflammation andAD phenotype according to advancing ageDing et al. [62] investigated periodontal disease effectson the brain in C57BL/6J wild-type mice in four(young) and 52-week (middle-aged) groups (Table 1).Induction of experimental periodontitis was by an oralTable1.SummaryofpathologicalhallmarkproteinappearanceandfunctionaltestingfollowingintroductionofP.gingivalisand/orLPSinexperimentalmice.Invivomodel,no.ofinfectionsanddurationTypeofinfectionPathologyoutcomeBehaviouraltesting.AD-likephenotypeoutcomeAllsupportingreferencesApolipoproteinEknockout(ApoE−/−)mice.109CFU,withfourinfectionsover12weeks'[ref65]andeightinfectionsover24weeks'[refs64,66],durationPeriodontitis-inducedoral,(periodontal)mono-infectionwithP.gingivalisFDC381ATCC53977[ref63].Inflammation(complementactivationandoxidativestress)DefectivehippocampalBBBNotdone63–66Amyloidprecursorprotein-transgenic(APP-Tg)micecarryingtheSwedishandIndianamutations.109CFU,fiveinfectionsoverfirst10daysofexperiment.Entiredurationofexperimentwas5weeksPeriodontitis-inducedoral,mono-infectionwithP.gingivalisATCC33277TGreaterdepositionofAβ40andAβ42amyloidplaquesinhippocampusandlevelsofIL-1βandTNF-αininfectedAPP-Tgmicecomparedtocontrol(sham-infected)APP-TggroupCognitivefunctionwassignificantlyimpairedinperiodontitis-inducedAPP-Tgmice45C57BL/6Nwild-type(2monthsold),middle-agedmice(12monthsold)andage-matchedcathepsinBsufficient/knockout(CatB−/−)miceinoculatedwithPgLPS(1mg/kg)daily,for5weeksSystemicexposuretopurifiedP.gingivalisLPS(PgLPS)fromATCC33277TNeuroinflammation(activatedastrocytesandmicroglia);intracellularAβinmiddle-agedWTmiceonlyInducedlearningandmemorydeficitsinmiddle-agedWTmiceonly55FemaleC57BL/6Jwild-typemiceC57BL/6J,at4and52weeks’agegroups109CFU,withrepeatinfectionevery48hover6weeksPeriodontitis-inducedoral,mono-infectionwithP.gingivalisATCC33277TInflammationasevidencedbyinflammatorymediator(cytokine)releaseImpairedlearningandmemoryatmiddleaged(52weeks’group)62C57BL/6wildtypemice.6weeks'ageInoculumsizewas109CFU,3infections/week(66infectionsinall)over22weeks'durationPeriodontitis-inducedoral,mono-infectionwithP.gingivalis(strainW83)Inflammation,extracellularAβ42amyloidplaquesandser396residueoftauproteinphosphorylationandneurofibrillarytangleformationinhippocampusN/A72C57BL/6,8weekoldmalemiceP.gingivalis-LPSandvariousinhibitorsoftheTLRsignallingpathwayintarperitonealsingleinjection(s)5mg/kgInducedglialcellactivationandinducedinflammationviasynthesisofinflammatorycytokinesLearningandmemoryimpairmentinitiatedviaTLR4signallingpathway744 S. K. SINGHRAO ET AL.dose of live P. gingivalis (ATCC 33277T) mono-infec-tion using 1 × 109 CFU, which was repeated every 48 hover six weeks in each group. Evidence of intracerebralinflammation was obtained by inflammatory cytokinegene expression (using molecular biology methodolo-gies), and protein release (using ELISA assay).Cytokine levels (IL-1β, IL-6 and TNF-α) by bothmethods were significantly higher in the P. gingivalismono-infected older age group. Ding et al. [62] alsoperformed behavioural tests and demonstrated statis-tically significant outcomes for impaired spatial learn-ing and memory in the older age (middle-aged,52 week) group infected with P. gingivalis comparedto the younger (4-week old) and middle-aged unin-fected mice. Increased intracerebral inflammationaccounted for the impaired spatial learning and mem-ory following experimental periodontitis. The lack ofdemonstration of P. gingivalis or its LPS entry into themouse brain alongside omission of observationstowards Aβ protein in the neuronal cell body is aweakness of the study. This may imply that anysequestered Aβ in the regulated secretory vesicle path-way generated by cathepsin B processing was belowdetection limits, provided the bacterium and or its LPSentered the brain. However, functional tests provideuseful causal links with AD phenotype according toadvancing age, but imply that this is an inflammation-mediated event secondary to infection.Apolipoprotein E knockout mouse model of acuteand chronic periodontitis for AD neuropathologyThe apolipoprotein E knockout (ApoE−/−) mouse har-bours wtAPP in which periodontitis is induced follow-ing an oral P. gingivalis mono-infection [63] (Table 1).The mice received repeat infections (6 in 12 weeks’duration and 12 in 24 weeks’ duration) with 1 × 109CFU. The two time points corresponded to acute andchronic periodontitis in mice. This model was the firstto evaluate migration, to the brain, of the same P.Figure 1. The schematic suggests that the insoluble Aβ plaques are the result of intracellular processing of the wild type amyloidprecursor protein via the regulated secretory vesicle pathway mediated by cathepsin B (Taken from ref. 46 with permission).Figure 2. The schematic representation suggests the role of peripheral inflammation that in this example is contributed by a period-ontitis pocket infected by P. gingivalis. NF-κB signalling gives rise to cytokines and IL-1β and TNF-α appear to weaken the blood-brainbarrier (BBB) [65,66]. Bacteraemia allows P. gingivalis and its virulence factors to access the systemic circulation and enter the brain. P.gingivalis and its virulence factors enter cells and pass along the endosome/lysosomes where they encounter cathepsin B [46,55].JOURNAL OF ORAL MICROBIOLOGY 5gingivalis strain used to infect the host orally at the 24-weeks’ infection timeline, when the oral disease (peri-odontitis) had become chronic [63,64]. Acute phaseinflammation in the form of the activated complementsystem was detected at 24 weeks’ post infection withevidence of neuronal vulnerability to necrotic celldeath [64]. At 12 weeks’ post infection, oxidative stressand damage to the microvasculature in the hippocam-pus were noted [65], but subsequent blood–brain bar-rier (BBB) permeability in the cerebral (hippocampus)cortex became apparent only after 24 weeks [66].These findings support the chronic nature of period-ontitis that subsequently converts to risk for AD ashighlighted by former prospective and retrospectivestudies [24,25]. Poole et al. [64] and Rokad et al. [65]endorsed the innate immune responses in the form ofcomplement activation and reactive oxygen species(ROS)-mediated damage respectively, as described forhuman AD [67]. Furthermore, BBB defects [66] linkedaging and AD protective barrier breakdowns [68–71].The weaknesses of the ApoE−/− P. gingivalis-mono-infection model are the inability to monitor insolubleAβ due to the total apolipoprotein gene knockoutbecause apolipoprotein E is one of the three essentialproteins required for amyloid fibril formation.However, the positive outcomes of this study, in rela-tion to the similarity between the sporadic form of ADpathology and the ApoE−/− mouse model are many.Firstly to proof of concept testing that P. gingivalisfollowing bacteraemia can translocate from its oralniche to the brain [64]. Secondly, the peripheralinflammation had a negative impact on the BBB integ-rity [66]. Thirdly, P. gingivalis entry related directly tothe innate immune responses impacting on intracer-ebral inflammation in the form of ROS, and comple-ment activation [64,65]. These outcomes aresignificant findings that outweigh the inability to assessAβ pathology in this periodontitis-AD model [63].Wild type mouse model of periodontitisdemonstrates the cardinal AD lesionsThe infection periodontal model of Ilievski et al. [72] isC57BL/6 wild-type mice of 6 weeks’ age (Table 1). Thisis the youngest cohort of animals tested in any of theinfection models thus far. Induction of experimentalperiodontitis was by an oral dose of live P. gingivalis(W83), serotype 1 encapsulated, bacterial mono-infec-tion using 1 × 109 CFU, repeated every other day(Monday-Friday) of every week for 22 weeks.Evidence of intracerebral entry of the same P. gingivalis(W83) used to infect the host orally at the 22-weeks’infection timeline demonstrated its intracellular pre-sence using fluorescent in-situ hybridisation or FISHas used by Singhrao et al. [66]. FISH in the Ilievski et al.[72] mice brain sections clearly demonstrates the sitesFigure 3. Illustrates the direct pathways by which P. gingivalis activity gives rise to the Alzheimer phenotype. Zhang et al. [74]suggest this involves toll like receptor 4 (TLR-4) signalling) and Wu et al. [55] suggest microglial cytokine IL-1β for neuronalfunction that bears its receptor on the membrane for intracellular processing of the amyloid precursor protein (APP) andamyloid beta release as hypothesized by Hook et al. [46]. Ilievski et al. [72] demonstrated the build-up, in the mice brains, ofamyloid plaques outside the neurons and phosphorylation of tau on ser396 residue, an activity likely to be the result of theglycogen synthase-3β (GSK-3β) enzyme. GSK-3β can be activated via the NF-κB signalling cascade or the increasing amyloidplaque burden leading to neurofibrillary tangles (NFTs) forming. Inflammation followingP. gingivalis entry into brain from anoral niche is supported by Poole et al. [64] and subsequently by Ilievski et al. [72]. The encapsulated P. gingivalis W83 strain cancitrullinate structural proteins in glia and in neurons as described in reference [73].6 S. K. SINGHRAO ET AL.of infection in the various brain cells and in some casesthe nucleus. This clearly suggests the 22-week timelineof their infection regime was somewhat excessive.However, as a proof of concept study to determinehallmark lesion formation, this is acceptable. Ilievski etal. [72] detected inflammation by glial cell activationsupported by an elevated cytokine milieu within thehippocampus. However, most interestingly, theydemonstrated the AD defining hallmark lesions. Theseare amyloid plaques and de novo phosphorylated ser396residue in tau protein that was bound to neurofibrillarytangles within the hippocampus of the infected micegroup only. Although, this study lacked functional test-ing in the presence of hallmark lesions in the hippo-campus, it unequivocally demonstrated the causativerelationship of P. gingivalis with emerging AD pathol-ogy, which in itself is a significant milestone.P. gingivalis can citrullinate proteinsAs the Ilievski et al. [72] study on infections favouredP. gingivalis W83, which is a serotype 1 and capsu-lated strain, it is important to mention that such astrain has the capacity to produce citrullinated epi-topes, which could be detrimental to the health of thehost [73]. Currently, there are no reports of P. gingi-valis derived peptidyl arginine deiminase (PPAD)activity in the brain possibly because these studiesare still novel and because antibodies to PPAD arenot widely available. In the future P. gingivalis W83infections must examine this aspect of P. gingivalisvirulence to see if the disease follows an autoimmunecourse or not.P. gingivalis LPS and its effect on the brainLPS is the major surface membrane component ofvirulent Gram-negative bacteria such as P. gingivalis(Figure 3). P. gingivalis LPS is an important contributorto inflammation and neurodegeneration in AD becausepattern recognition receptors (PRRs), like Toll-likereceptors (TLRs), expressed by glia (as antigen present-ing cells) can recognize pathogen-associated molecularpatterns (PAMPs) in microorganisms to trigger anti-bacterial responses [45,55,62,64,74]. P. gingivalis LPScan stimulate CD14, TLR-2 or −4 and send signals tothe nucleus by the MyD88 pathway, which initiates acascade of events that involve an increased expressionof proinflammatory cytokines [45,55,62,72,74]. This isthe background to the identification of the mechanismof memory loss by the Zhang et al. [74] investigationdiscussed below. Another noteworthy feature of P. gin-givalis is that its LPS exists in at least two differentforms, O-LPS and A-LPS. The latter shows heterogene-ity occurring as two isoforms, LPS1435/1449 and LPS1690[75]. These isoforms can produce opposing effects onTLR-2 and −4 activation. The capacity to change its LPSto LPS1435/1449 or LPS1690 helps P. gingivalis in adjustingto the local inflammatory milieu, enabling it to survivein primary and distant sites, especially in lysosomalcompartments of different tissue cell types [54].P. gingivalis-LPS model links with intracellularAβ in cathepsin B sufficient miceWu et al. [55] (Table 1) examined the effect of sys-temic exposure to purified P. gingivalis-LPS (PgLPS)in wild-type (C57BL/6N), young (2 months old) andmiddle-aged (12 months old) mice and age-matchedcathepsin B sufficient and knockout mice for Aβ. Theexperimental procedure involved intraperitonealinjection of PgLPS from ATCC 33277T (1 mg/kg)daily dose, for 5 weeks. The timespan for introduc-tion of the endotoxin was determined from theirprevious evaluation of systemic inflammation toinduce deficits in the hippocampal long-term poten-tiation in middle-aged rats through microglia-mediated neuroinflammation [76]. Inflammation(activated astrocytes and microglia) and intracellularAβ in middle-aged cathepsin B sufficient mice werereported together with learning and memory deficits.An explanation for the neuronal Aβ and the memorydeficit included P. gingivalis using the endocytic/lyso-somal pathway to enter cells and the consequence ofglial cell activation and cytokine release. P. gingivalisLPS can mediate inflammation via cytokine (IL-1β)release, and then generate intracellular Aβ by activa-tion of cathepsin B in an age-dependent manner [55].This study reported AD like behaviour following P.gingivalis ATCC 33277T LPS introduction in mice,and above all demonstrated intracellular release ofAβ according to the hypothesis of Hook et al. [46]via the regulated secretory vesicle pathway mediatedby cathepsin B (Figure 3), which consolidated thecausative relationships of P. gingivalis with AD [55].This supports the Ilievski et al. [72] study linkingextracellular Aβ amyloid plaques following P. gingi-valis infection in mouse models expressing wtAPP. Inaddition, the Hook et al. [46] hypothesis suggestedthe slower accumulation of extracellular amyloid pla-ques in wtAPP protein. This distinction is very clearfrom the Ilievski et al. [72] images, which show fewerplaques compared with the images from the infectedAPP-Tg group [45].P. gingivalis LPS administration once in wildtype mice supports TLR-4 signalling leadingto AD phenotypeA study by Zhang et al. [74] aimed at dissecting out themechanism that leads to loss of memory following P.gingivalis LPS intraperitoneal administration. The LPSused here is the same as that administered in otherstudies [55] confirming experimental consistency withJOURNAL OF ORAL MICROBIOLOGY 7previous phenotype related data outcome. The Zhang etal. [74] study used 8-week-old C57BL/6 mice. A singleintraperitoneal injection of 5 mg/Kg P. gingivalis-LPSwas administered with/without the TLR-4 inhibitorTAK-242. Seven days following the LPS and/or TLR-4-inhibitor challenge, cognitive tests were performed,which included the Open Field, Morris Water Maze(MWM) and Passive Avoidance, Following sacrifice,the brain tissues were examined for inflammatory mar-kers by molecular biology, protein biochemistry andimmunohistochemistry. Compared with the controlgroup, Zhang et al. [74] found the test group receivingP. gingivalis LPS had impaired spatial learning andmem-ory during the MWM test and a weak desire for facingfear in their ‘Passive Avoidance Test’. They also observedglial (microglia and astrocytes) cell activation in thecortex and the hippocampus regions of these mice brainsalongside of upregulated cytokines (TNF-α, IL-1β, IL-6and IL-8). The TLR-4 inhibitor (TAK-242) group micebrains demonstrated suppressed TLR-4/NF-κB signal-ling pathways, and the same group of animals did notshow cognitive impairments. This confirmed the role ofTLR-4 signalling in poor memory development.LPS links with tau protein phosphorylation inAD transgenic miceBacterial products appear to play a detrimental role inthe onset and development of tau pathology.Currently, there is only one report linking tau proteinphosphorylation due to P. gingivalis infection in mice.In another study, the AD transgenic mouse model(3xTg-AD and rTg4510) harbouring mutated taugenes [11,13] has been administered with an intra-peritoneal injection of purified commercial LPS fromEscherichia coli (12 doses over 6 weeks) in 4-months-old 3xTg-AD mice; the authors firmly implicated therole of microglial cytokines (IL-1β) in tau phosphor-ylation [11]. These researchers also found increasedtau phosphorylation at Ser202/Thr205 and Thr231/Ser235 residues in hippocampal neurons comparedwith sham-infected transgenic control mice in kinasespecific activity [11]. What is intriguing is that Wu etal. [55] have demonstrated a role for IL-1β in theircathepsin B sufficient mice following introduction ofP. gingivalis LPS but without reporting tauphosphor-ylation. This leaves gaps in our knowledge as to thediffering mechanisms of E. coli LPS and P. gingivalisLPS [75] in initiating innate signalling cascades.Lee et al. [13] injected commercial, purified LPSfrom Salmonella abortus-equi (an aberrant infectionequivalent), demonstrating that phosphorylated tauincreases on select serine residues (Ser199/202 andSer396) in the mutated (rTg4510) group of micecompared with controls. Differential tau phosphory-lation at specific and plausible serine/threonine sitesis acceptable as it reflects the activity of specifickinases that phosphorylate their respective aminoacid residues in the tau protein [11]. This suggeststhat vulnerability in the tau gene, together with infec-tions by Gram-negative bacteria, are able to influenceneurofibrillary tangle formation. However, AD doesnot harbour mutations in the tau gene per se, and tothis end, Kitazawa et al. [11] observed that LPSaffected the phosphorylation of tau in the wild typelittermates over sham-infected non-transgenic mice,which they attributed to inflammation. The resultsconfirm a causative role of bacterial LPS in the devel-opment of the pathology in both the familial andsporadic forms of AD with inflammation playing apivotal role in both APP processing and tau proteinphosphorylation at select residues.Expression of AD phenotype in infected miceCognition describes a person’s mental ability to processinformation, reasoning, and learning of new skills,remembering them, and relating to them. Episodicmem-ory, accounts for neuronal plasticity, whereby neuronscan modify the patterns of connectivity of functionalneurons through enhanced development of their den-drites and axon elongation [77]. This ability appears to belost in AD and is the basis of the cognitive testing regimedesigned to assess new patients for dementia. Synapticplasticity refers to the ability of synapses to modify toadapt to challenges posed by ageing and disease pro-cesses. These changes include size, morphology, densityand even complete loss of synapses within defined para-meters of disease [78]. Spatial memory loss correlateswith synaptic loss [79] and earlier reports assigned Aβoligomer toxicity to synaptic loss in AD [80]. P. gingivalisinfection models described here are also displayingimpaired spatial and learning and memory deficits inthe younger and older age groups [55,62,74], and in theAD transgenic and wild type infectedmouse models [45]where inflammation is a common feature. Therefore,inflammatory mediators such as IL-1β, which also hasdetrimental effect on synapses, is a likely challenge frominfections that may correlate with plasticity and synapseloss resulting in cognitive dysfunctional displays byinfected mice [45,55,62,74].Interventional studies support periodontitisas a risk factor for ADThe earliest interventional study was performed byRolim et al. [81], which included 29 participants withclinically mild AD. A dentist performed a completeevaluation involving: clinical questionnaire; researchdiagnostic criteria for temporomandibular disorders;McGill pain questionnaire; oral health impact profile;decayed, missing and filled teeth index and completeperiodontal examination before and after the inter-vention. The study found a reduction of orofacial8 S. K. SINGHRAO ET AL.pain, and improvement in the mandibular functionand periodontal indices in patients with AD. Theseimprovements were maintained until the last evalua-tion after 6 months and were followed by a reductionin the functional impairment due to cognitive com-promise [81]. The limitation of this study is that itlacked a bigger cohort and appropriate (non-AD)controls.Concluding remarksP. gingivalis infections and its LPS appear to be clo-sely associated with the development of the sporadicform of AD. Data presented here are showing a con-sistent causative role of P. gingivalis infections and itsLPS in mice for the development of cardinal hallmarklesions and cognitive impairment via systemic andintracerebral inflammation. In addition, inflamma-tion is playing a pivotal role in APP processing togenerate Aβ, neurofibrillary tangle formation anddeteriorating memory. The time of onset (early vslate) can be explained by modes of Aβ deposition.For example, Aβ deposition in the familial AD brainmay arise directly by extracellular release frommtAPP processing by BACE 1. However, wtAPPappears to undergo intracellular processing andrelease of Aβ mediated by the regulated secretoryvesicle pathway inflicted by cathepsin B processingresulting in lesser yields. Neuropathology diagnosisrelies on a specific threshold of Aβ plaques in bothforms of AD; this may imply that the sporadic formhas a defective clearance while the familial form hasincreased deposition of this insoluble protein. Thissuggests that the 10-year lag phase of chronic period-ontitis to become a risk factor for the sporadic formof AD is plausible. Why the long lag phase? This maybe due to initial weakening of the protective BBBthrough aging allowing easier access of bacteria intothe brain. Alternatively, unlike the oral cavity, whichembraces a range of diverse bacterial phylotypes anddevelops chronic infection only after a few weeks, thehealthy brain may be slow to respond to nominallyvirulent (seronegative) P. gingivalis stains in youngerhuman hosts because of their immune system. Incontrast, seropositive 1 strains such as W83 appearto reach the brain with speed but the importantfeature is the bacterial load. This goes back to poororal hygiene habits seen in patients.Tau phosphorylation causal links with P. gingivalisinfection completes the intrigue that P. gingivalis caninitiate and produce both of the defining lesions ofAD. Ideally, functional tests on all related periodon-titis infection for AD would be desirable but notalways possible, as the studies described here haveshown. In all instances, live P. gingivalis and its LPSare powerful peripheral and intracerebral inflamma-tory signalling initiators, and this has direct and earlyimplications on memory. The data presented here aresignificant, contributing to our growing knowledge ofthe causal associations between the sub-gingivalpathobiome under the influence of P. gingivalis anddevelopment of AD. Maintaining an oral microbiomesymbiosis and preventing periodontal disease withregular surveillance and good oral hygiene through-out life is likely to reduce the incidence of unwantedsuffering from AD.Disclosure statementNo potential conflict of interest was reported by theauthors.FundingSKS has received a PreViser award from the Oral andDental Research Trust, [2018], and also acknowledges thecontinued financial support from the School of Dentistry,University of Central Lancashire, UK.References[1] Eke PI, Dye BA, Wei L, et al. Update on prevalence ofperiodontitis in adults in the USA: NHANES 2009 to2012. J Periodontol. 2015;86(5):611–622.[2] Carter CJ, France J, Crean S, et al. The Porphyromonasgingivalis/host interactome shows enrichment inGWASdb genes related to Alzheimer’s disease, dia-betes and cardiovascular diseases. Front AgingNeurosci. 2017;9:408.[3] Eke PI, Wei L, Thornton-Evans OG, et al. Risk indi-cators for periodontitis in US adults: NHANES 2009to 2012. J Periodontol. 2016;87(10):1174–1185.[4] Wang J, Lv J, Wang W, et al. Alcohol consumptionand risk of periodontitis: a meta-analysis. J ClinPeriodontol. 2016;43(7):572–583.[5] Lertpimonchai A, Rattanasiri S, Arj-Ong VallibhakaraS, et al. The association between oral hygiene andperiodontitis: a systematic review and meta-analysis.Int Dent J. 2017;67(6):332–343.[6] Paganini-Hill A, White SC, Atchison KA. Dentition,dental health habits, and dementia: the Leisure Worldcohort study. J Am Geriatr Soc. 2012;60:1556–1563.[7] Farhad SZ, Amini S, Khalilian A, et al. The effect ofchronic periodontitis on serum levels of tumor necro-sis factor-alpha in Alzheimer disease. Dent Res J(Isfahan). 2014;11(5):549–552.[8] Harding A, Robinson S, Crean S, et al. Can bettermanagement of periodontal disease delay the onsetand progression of Alzheimer’s disease? J AlzheimersDis. 2017;58:337–348.[9] Harding A, Gonder U, Robinson SJ, et al. Exploringthe association between Alzheimer’s disease, oralhealth, microbial endocrinology and nutrition. FrontAging Neurosci. 2017;9:398.[10] Hill JM, Clement C, Poque AI, et al. Pathogenicmicrobes, the microbiome, and Alzheimer’s disease(AD). Front Aging Neurosci. 2014;6:127.[11] Kitazawa M, Oddo S, Yamasaki TR, et al.Lipopolysaccharide-induced inflammation exacerbatestau pathology by a cyclin-dependent kinase 5-JOURNAL OF ORAL MICROBIOLOGY 9mediated pathway in a transgenic model ofAlzheimer’s disease. J Neurosci. 2005;25:8843–8853.[12] Hanger DP, Byers HL, Wray S, et al. Novel phosphor-ylation sites in tau from Alzheimer brain support arole for casein kinase 1 in disease pathogenesis. J BiolChem. 2007;282(32):23645–23654.[13] Lee DC, Rizer J, Selenica ML, et al. LPS-inducedinflammation exacerbates phospho-tau pathology inrTg4510 mice. J Neuroinflam. 2010;7:56.[14] Stein PS, Desrosiers M, Donegan SJ, et al. Tooth loss,dementia and neuropathology in the Nun study. J AmDent Assoc. 2007;138:1314–1322.[15] Tzeng NS, Chung CH, Yeh CB, et al. Are chronicperiodontitis and gingivitis associated with dementia?A nationwide, retrospective, matched-cohort study inTaiwan. Neuroepidemiology. 2016;47:82–93.[16] Blaizot A, Vergnes JN, Nuwwareh S, et al.Periodontal diseases and cardiovascular events:meta-analysis of observational studies. Int Dent J.2009;59(4):197–209.[17] Virtanen E, Nurmi T, Söder PÖ, et al. Apical period-ontitis associates with cardiovascular diseases: a cross-sectional study from Sweden. BMC Oral Health.2017;17(1):107.[18] Macedo Paizan ML, Vilela-Martin JF. Is there anassociation between periodontitis and hypertension?Curr Cardiol Rev. 2014;10(4):355–361.[19] Graziani F, Gennai S, Solini A, et al. Systematic reviewand meta-analysis of epidemiologic observational evi-dence on the effect of periodontitis on diabetes. Anupdate of the EFP-AAP review. J Clin Periodontol.2018;45(2):167–187.[20] Leira Y, Seoane J, Blanco M, et al. Association betweenperiodontitis and ischemic stroke: a systematic reviewand meta analysis. Eur J Epidemiol. 2017;32(1):43–53.[21] de la Monte SM. Type 3 diabetes is sporadic Alzheimer’sdisease: mini-review. Eur Neuropsychopharmacol.2014;24(12):1954–1960.[22] Kondo K, Niino M, Shido KA. Case-control study ofAlzheimer’s disease in Japan—significance of life-styles. Dementia. 1994;5(6):314–326.[23] Gatz M, Mortimer JA, Fratiglioni L, et al. Potentiallymodifiable risk factors for dementia in identical twins.Alzheimers Dement. 2006;2:110–117.[24] Chen C-K, Wu Y-T, Chang Y-C. Association betweenchronic periodontitis and the risk of Alzheimer’s dis-ease: a retrospective, population-based, matched-cohort study. Alzheimers Res Ther. 2017;9:56.[25] Sparks Stein P, Steffen MJ, Smith C, et al. Serum antibo-dies to periodontal pathogens are a risk factor forAlzheimer’s disease.AlzheimersDement. 2012;8:196–203.[26] Dunn N, Mullee M, Perry H, et al. Associationbetween dementia and infectious disease: evidencefrom a case control study. Alzheimer Dis AssocDisord. 2005;19(2):91–94.[27] Holmes C, El-Okl M, Williams AL, et al. Systemicinfection, interleukin 1β and cognitive decline inAlzheimer’s disease. J Neurol Neurosurg Psychiatry.2003;74:788–789.[28] Noble JM, Borrell LN, Papapanou PN, et al.Periodontitis is associated with cognitive impairmentamong older adults: analysis of NHANES-III. J NeurolNeurosurg Psychiatry. 2009;80:1206–1211.[29] Watts A, Crimmins EM, Gatz M. Inflammation as apotential mediator for the association between period-ontal disease and Alzheimer’s disease. NeuropsychiatrDis Treat. 2008;4:865–876.[30] Kamer AR, Craig RG, Pirraglia E, et al. TNFα andantibodies to periodontal bacteria discriminatebetween Alzheimer’s disease patients and normal sub-jects. J Neuroimmunol. 2009;216:92–97.[31] Ide M, Harris M, Stevens A, et al. Periodontitis andcognitive decline in Alzheimer’s disease. PLoS One.2016;11(3):e0151081.[32] Aas JA, Paster BJ, Stokes LN, et al. Defining thenormal bacterial flora of the oral cavity. J ClinMicrobiol. 2005;43(11):5721–5732.[33] Rogers GB, Bruce KD. Exploring the parallel develop-ment of microbial systems in neonates with cysticfibrosis. MBio. 2012;3(6):e00408–e00412.[34] Miklossy J. Chronic inflammation and amyloidogen-esis in Alzheimer’s disease – role of spirochetes. JAlzheimers Dis. 2008;13:381–391.[35] Miklossy J. Historic evidence to support a causal rela-tionship between spirochetal infections and Alzheimer’sdisease. Front Aging Neurosci. 2015;7:46.[36] Poole S, Singhrao SK, Kesavalu L, et al. Determiningthe presence of periodontopathic virulence factors inshort-term postmortem Alzheimer’s disease brain tis-sue. J Alzheimers Dis. 2013;36:665–677.[37] Murphy MP, LeVine H. Alzheimer’s disease and the β-amyloid peptide. J Alzheimers Dis. 2010;19:311–323.[38] Price DL, Sisodia SS. Mutant genes in familialAlzheimer’s disease and transgenic models. AnnuRev Neurosci. 1998;21:479–505.[39] Selkoe DJ. Alzheimer’s disease: genes, proteins, andtherapy. Physiol Rev. 2001;81:741–766.[40] Selkoe DJ. Deciphering the genesis and fate of amyloidbeta-protein yields novel therapies for Alzheimer dis-ease. J Clin Invest. 2002;110:1375–1381.[41] Rossor MN, Newman S, Frackowiak RS, et al.Alzheimer’s disease families with amyloid precursorprotein mutations. Ann NY Acad Sci. 1993;695:198–202.[42] Hook V, Kindy M, Reinheckel T, et al. Genetic cathe-psin B deficiency reduces beta-amyloid in transgenicmice expressing human wild-type amyloid precursorprotein. Biochem Biophys Res Commun.2009;386:284–288.[43] Mori H, Takio K, Ogawara M, et al. Mass spectro-metry of purified amyloid beta protein in Alzheimer’sdisease. J Biol Chem. 1992;267:17082–17086.[44] Mueller-Steiner S, Zhou Y, Arai H, et al.Antiamyloidogenic and neuroprotective functions ofcathepsin B: implications for Alzheimer’s disease.Neuron. 2006;51:703–714.[45] Ishida N, Ishihara Y, Ishida K, et al. Periodontitisinduced by bacterial infection exacerbates features ofAlzheimer’s disease in transgenic mice. NPJ AgingMech Dis. 2017;3:15.[46] Hook V, Schechter I, Demuth HU, et al. Alternativepathways for production of beta-amyloid peptides ofAlzheimer’s disease. Biol Chem. 2008;389:993–1006.[47] Hook V, Hook G, Kindy M. Pharmacogenetic featuresof cathepsin B inhibitors that improve memory deficitand reduce beta-amyloid related to Alzheimer’s dis-ease. Biol Chem. 2010;3391:861–872.[48] Hook G, Hook V, Kindy M. The cysteine proteaseinhibitor, E64d, reduces brain amyloid-β and improvesmemory deficits in Alzheimer’s disease animal modelsby inhibiting cathepsin B, but not BACE1, β-secretaseactivity. J Alzheimers Dis. 2011;26:387–408.[49] Hook V, Funkelstein L, Wegrzyn J, et al. Cysteine cathe-psins in the secretory vesicle produce active peptides:10 S. K. SINGHRAO ET AL.cathepsin L generates peptide neurotransmitters andcathepsin B produces beta-amyloid of Alzheimer’s dis-ease. Biochim Biophys Acta. 2012;1824:89–104.[50] Cataldo AM, Barnett JL, Pieroni C, et al. Increasedneuronal endocytosis and protease delivery to earlyendosomes in sporadic Alzheimer’s disease: neuro-pathologic evidence for a mechanism of increasedbeta-amyloidogenesis. J Neurosci. 1997;17:6142–6151.[51] Cataldo A, Rebeck GW, Ghetri B, et al. Endocytic dis-turbances distinguish among subtypes of Alzheimer’s dis-ease and related disorders. Ann Neurol. 2001;50:661–665.[52] Takayama F, Hayashi Y, Wu Z, et al. Diurnal dynamicbehaviour of microglia in response to infected bacteriathrough the UDP-P2Y6 receptor system. Sci Rep.2016;6:30006.[53] DornBR, Dunn WA, Progulske-Fox A. Porphyromonasgingivalis traffics to autophagosomes in human coronaryartery endothelial cells. Infect Immun. 2001;69:5698–5708.[54] Furuta N, Tsuda K, Omori H, et al. Porphyromonasgingivalis outer membrane vesicles enter humanepithelial cells via an endocytic pathway and aresorted to lysosomal compartments. Infect Immun.2009;77:4187–4196.[55] Wu Z, Ni J, Liu Y, et al. Cathepsin B plays a criticalrole in inducing Alzheimer’s disease-like phenotypesfollowing chronic systemic exposure to lipopolysac-charide from Porphyromonas gingivalis in mice.Brain Behav Immun. 2017;65:350–361.[56] Olsen I, Yilmaz Ö. Modulation of inflammasome activityby Porphyromonas gingivalis in periodontitis and asso-ciated systemic diseases. J Oral Microbiol. 2016;8:30385.[57] Rapsinski GJ, Wynosky-Dolfi MA, Oppong GO, et al.Toll-like receptor 2 and NLRP3 cooperate to recog-nize a functional bacterial amyloid, curli. InfectImmun. 2015;83(2):693–701.[58] Heneka MT, Kummer MP, Stutz A, et al. NLRP3 isactivated inAlzheimer’s disease and contributes to pathol-ogy in APP/PS1 mice. Nature. 2013;493(7434):674–678.[59] Halle A, Hornung V, Petzold GC, et al. The NALP3inflammasome is involved in the innate immuneresponse to amyloid-beta. Nat Immunol. 2008;9(8):857–865.[60] Sheedy FJ, Grebe A, Rayner KJ, et al. CD36 coordi-nates NLRP3 inflammasome activation by facilitatingintracellular nucleation of soluble ligands into parti-culate ligands in sterile inflammation. Nat Immunol.2013;14(8):812–820.[61] Gold M, Khoury JE. β-amyloid, microglia, and theinflammasome in Alzheimer’s disease. SeminImmunopathol. 2015;37:607–611.[62] Ding Y, Ren J, Yu H, et al. Porphyromonas gingivalis, aperiodontitis causing bacterium, induces memoryimpairment and age-dependent neuroinflammationin mice. Immun Ageing. 2018;15:6.[63] Velsko IM, Chukkapalli SS, Rivera MF, et al. Activeinvasion of oral and aortic tissues by Porphyromonasgingivalis in mice causally links periodontitis andatherosclerosis. PLoS One. 2014;9:e97811.[64] Poole S, Singhrao SK, Chukkapalli S, et al. Activeinvasion of Porphyromonas gingivalis and infection-induced complement activation in ApoE−/- micebrains. J Alzheimers Dis. 2015;43:67–80.[65] Rokad F, Moseley R, Hardy RS, et al. Cerebral oxidativestress and microvasculature defects in TNF-α expres-sing transgenic and Porphyromonas gingivalis-infectedApoE−/- mice. J Alzheimers Dis. 2017;60:359–369.[66] Singhrao SK, Chukkapalli S, Poole S, et al. ChronicPorphyromonas gingivalis infection accelerates theoccurrence of age-related granules in ApoE−/- mice. JOral Microbiol. 2017;9:1270602.[67] AkiyamaH, Barger S, Barnum S, et al. Inflammation andAlzheimer’s disease. Neurobiol Aging. 2000;21:383–421.[68] Tran L, Greenwood-Van Meerveld B. Age-associatedremodeling of the intestinal epithelial barrier. JGerontol A Biol Sci Med Sci. 2013;68:1045–1056.[69] Marques F, Sousa JC, Sousa N, et al. Blood-brain-barriers in aging and in Alzheimer’s disease. MolNeurodegener. 2013;8:38.[70] Montagne A, Barnes SR, Sweeney MD, et al. Blood-brain barrier breakdown in the aging human hippo-campus. Neuron. 2015;85:296–302.[71] Halliday MR, Rege SV, Ma Q, et al. Accelerated peri-cyte degeneration and blood-brain barrier breakdownin apolipoprotein E4 carriers with Alzheimer’s disease.J Cereb Blood Flow Metab. 2016;36:216–227.[72] Ilievski V, Zuchowska PK, Green SJ, et al. Chronic oralapplication of a periodontal pathogen results in braininflammation, neurodegeneration and amyloid beta pro-duction in wild type mice. PLoS One. 2018;13(10):e0204941.[73] Olsen I, Singhrao SK, Potempa J. Citrullination as aplausible link to periodontitis, rheumatoid arthritis,atherosclerosis and Alzheimer’s disease. J OralMicrobiol. 2018;10(1):1487742.[74] Zhang J, Yu C, Zhang X, et al. Porphyromonas gingi-valis lipopolysaccharide induces cognitive dysfunc-tion, mediated by neuronal inflammation viaactivation of the TLR4 signalling pathway in C57BL/6 mice. J Neuroinflammation. 2018;15(1):37.[75] Olsen I, Singhrao SK. Importance of heterogeneity inPorhyromonas gingivalis lipopolysaccharide lipid A intissue specific inflammatory signalling. J OralMicrobiol. 2018;10:1440128.[76] Liu Y, Wu Z, Nakanishi Y, et al. Infection ofmicroglia with Porphyromonas gingivalis promotescell migration and an inflammatory responsethrough the gingipain-mediated activation of pro-tease-activated receptor-2 in mice. Sci Rep. 2017;7(1):11759.[77] Bäckman L, Small BJ, Fratiglioni L. Stability of thepreclinical episodic memory deficit in Alzheimer’sdisease. Brain. 2001;124(1):96–102.[78] Huttenlocher PR, Dabholkar AS. Regional differencesin synaptogenesis in human cerebral cortex. J CompNeurol. 1997;387(2):167–178.[79] Terry RD, Masliah E, Salmon DP, et al. Physical basisof cognitive alterations in Alzheimer’s disease: synapseloss is the major correlate of cognitive impairment.Ann Neurol. 1991;30:572–580.[80] Walsh DM, Selkoe DJ. Deciphering the molecularbasis of memory failure in Alzheimer’s disease.Neuron. 2004;44:181–193.[81] Rolim TDS, Fabri GM, Nitrini R, et al. Evaluation ofpatients with Alzheimer’s disease before and after dentaltreatment. Arq Neuropsiquiatr. 2014;72(12):919–924.JOURNAL OF ORAL MICROBIOLOGY 11AbstractIntroduction10-year exposure to chronic periodontitis doubles the risk for ADConcept of systemic inflammation contributing to memory lossHost's microbiome dysbiosis is an important environmental factorThe (simplified) amyloid cascadeAD-transgenic mice support experimental periodontitis as a nominal riskMouse models with wild type APP support P. gingivalis as a risk factor for ADWild type mouse model of experimental periodontitis supports neuroinflammation and AD phenotype according to advancing ageApolipoprotein E knockout mouse model of acute and chronic periodontitis for AD neuropathologyWild type mouse model of periodontitis demonstrates the cardinal AD lesionsP. gingivalis can citrullinate proteinsP. gingivalis LPS and its effect on the brainP. gingivalis-LPS model links with intracellular Aβ in cathepsin B sufficient miceP. gingivalis LPS administration once in wild type mice supports TLR-4 signalling leading to AD phenotypeLPS links with tau protein phosphorylation in AD transgenic miceExpression of AD phenotype in infected miceInterventional studies support periodontitis as a risk factor for ADConcluding remarksDisclosure statementFundingReferences