The oral risk factors link to periodontal pathogens accessing the brain (Riviere et al., 2002; Poole et al., 2013). These environmental factors may have contributed to dementia development in the elderly. Several studies have supported the links between life-style choices, specifically engaging with oral health practices and AD development (Kondo et al., 1994; Stein et al., 2007; Paganini-Hill et al., 2012; Luo et al., 2015; Ide et al., 2016). More recently, a Taiwanese study found a strong link between chronic periodontal disease (exposures of around 10-year period) and AD (Chen et al., 2017). This retrospective study (Chen et al., 2017) correlates with a previous prospective study in which circulating antibodies to two oral bacteria were linked to cognitive deficit 10 years later (Sparks Stein et al., 2012). SES (poor oral health, smoking, unhealthy diet) factors are very important when considering infections of the brain. According to literature, periodontal disease does contribute to systemic markers of inflammation (Loos, 2005). Research has shown that systemic inflammatory markers mediate the development of AD (Schmidt et al., 2002; Holmes et al., 2003; Engelhart et al., 2004; Kamer et al., 2009; Sparks Stein et al., 2012). This finding is the crucial link that explains the association of periodontal disease with both the etiology of cardiovascular pathologies (Libby et al., 2002), and insulin resistance (Craft, 2005). Rather surprisingly, Escherichia coli K12 possesses insulin that is similar to the mammalian functional protein (LeRoith et al., 1981). The fact that E. coli lipopolysaccharide (LPS) in AD brains co-localizes with Aβ hallmark protein (Zhan et al., 2016; Zhao et al., 2017) suggests extrinsic sources of insulin have the potential to contribute to excess insulin and/or its resistance development. It may be plausible to suggest that microbial endocrinology plays a role in the development of diabetes in the AD brain.

The oral cavity is important for a number of nutritional functions that are required for a healthy existence and, as such, links it to the GI tract. As noted previously, the GI tract microbiome can affect AD independently of periodontal disease. However, the GI tract microbial challenges also pose an interesting analogy through the dysbiotic microbiome concept to the intestinal mucosal barriers becoming permeable (Daulatzai, 2014).

The hypothalamic-pituitary-adrenal (HPA) axis responds to periodontal disease polymicrobial infections. Related inputs from the CNS may modify GI tract function, while inputs from the alimentary canal can modulate symptom processing via sensory pathways to the brain. Since AD onset progresses slowly to the clinical expression, Brandscheid et al. (2017) examined whether the asymptomatic neuropathology correlated with changing bacteria in the GI tract of a familial (5xFAD) mouse model. Interestingly the study found reduced protein breakdown and this was related to poor trypsin secretions. Lower levels of trypsin affected this niche as demonstrated by a changing fecal microbial composition in an age-dependent manner (Brandscheid et al., 2017). The consequences of a changing gut microbiome are that specific bacterial species taking up residency may acquire the potential to break down mucosal barriers.

The most important underlying metabolic disturbances that raise the risk for both AD and periodontal disease are insulin resistance and a state of chronic inflammation. Based on the variability in the inflammatory component of AD cases, three main subtypes of this disease are described, which are distinguishable by metabolic profiling. Two of these are described as the “inflammatory” type and “non-inflammatory” type (Bredesen, 2015). The “inflammatory” type often encompasses an ApoE4 allele carrier, which presents with markers of systemic and cerebral inflammation such as C-reactive protein (CRP) and IL-6 levels, insulin resistance and hyperhomocysteinemia. In the “non-inflammatory” type AD insulin resistance, hypovitaminosis D, hyperhomocysteinemia and reduced hormones dominate while indicators of systemic inflammation are absent. This does not however exclude the existence of intra-cerebral inflammation. Since the inflammatory and non-inflammatory subtypes are insulin resistant, and often FDG-PET scans show temporal parietal reductions in glucose utilization (Cunnane et al., 2011; Bredesen, 2015), their management by nutritional support is plausible.

Good nutrition plays an important role in health. The five leading global risks for mortality appear to be high blood pressure, tobacco usage, hyperglycemia, physical inactivity and obesity. Nutrition can address three of these five global risks successfully. For example, if individuals consume low amounts of plant-derived foods, they are at increased risk of developing T2DM, metabolic syndrome, some cancers, and AD (Salas-Salvadó et al., 2016). Although a relationship between nutrition and AD onset is proposed, the mechanisms, which may explain how poor nutrition leads to AD, are incomplete. Here we explore how poor nutrition may not only act as a risk factor for poor oral health, but may also influence the mechanisms that mediate the relationship between periodontal disease and AD development. Furthermore, we explore interventions that can support good nutrition.

The MEND™ protocol is diet based, and shows reversal of the functional loss in the prodromal AD cases (Bredesen et al., 2016). This study refers to a cohort (N = 10) of patients who were able to re-gain memory by avoiding risk factors as suggested in the MEND™ protocol (Bredesen et al., 2016). However, it is unclear if all the foods found within a Western diet directly disturb brain function, or if it is single nutrients or a combination of different, dietary elements that have an adverse impact on brain health. For example, there is no strong argument (Ravnskov et al., 2014; Harcombe et al., 2016), for avoiding saturated fats, however, when combined with high glycaemic carbohydrates, saturated fats do have negative health consequences (Varela-López et al., 2016). However, the prospective PURE cohort study in 18 countries from five continents found that total fat intake and individual types of fatty acids are associated with a decreased risk of total mortality, and saturated fat intake correlates inversely with risk for strokes, an established risk factor for AD (Dehghan et al., 2017).

Diet also has an impact on periodontal disease, an established risk factor for AD (Kamer et al., 2015; Chen et al., 2017; Harding et al., 2017). Therefore, promoting a healthy diet that supports good oral health is an important preventative measure for AD. A recent epidemiological study on a Japanese population showed a more advanced periodontal disease in people who had a lower fat intake (Hamasaki et al., 2017). The authors concluded that a diet low in fat and high in carbohydrates might be associated with a higher prevalence of periodontal disease. On the other hand, the consumption of vegetables, fruit, β-carotene, vitamin C, β-tocopherol, EPA and DHA helped patients to reduce their levels of periodontal disease (Dodington et al., 2015).

For the prevention of insulin resistance and inflammation, diets high in monounsaturated (e.g., olive oil, avocados, macadamias, chicken fat) and omega-3 polyunsaturated fatty acids of marine origin are recommended, e.g., a Mediterranean style diet (Schwingshackl and Hoffmann, 2014). Epidemiological and interventional studies have repeatedly shown superior outcomes in weight loss (Hession et al., 2009), weight maintenance (Larsen et al., 2010), insulin resistance, triglyceride lowering and HDL cholesterol raising (Shai et al., 2008; Bazzano et al., 2014), and lowering of CRP (Bazzano et al., 2014) with lower carbohydrate diets containing more protein and fat. The same applies for the adoption of a Mediterranean type diet, which by definition is not low in fat (Schwingshackl and Hoffmann, 2014). The PREDIMED intervention trial in individuals with elevated cardiovascular risk compared a Mediterranean eating pattern supplemented with olive oil or nuts with a lower fat diet. The results showed significantly lower numbers of cardiovascular events in those who ate the diets supplemented with nuts or olive oil (Estruch et al., 2013) and slower cognitive decline (Estruch et al., 2013; Martínez-Lapiscina et al., 2013). In addition, foods high in important micronutrients, such as omega-3 fatty acids, vitamins C, D, E, B12 and B6, zinc, iron, iodine and magnesium, potassium, magnesium and dietary fiber can support brain health (Cunnane, 2006).

Beside their negative impact on insulin resistance and inflammation, diets high in simple carbohydrates may impair the integrity of the BBB (Hsu and Kanoski, 2014). Compromised BBB function in individuals susceptible to periodontitis will allow microbes to enter the brain and this could increase the chances of developing AD. Adopting a healthier Mediterranean style diet (Schwingshackl and Hoffmann, 2014), or eating fish, vegetables, nuts, cocoa and a variety of berries (Poulose et al., 2014), which are rich in antioxidants and have anti-inflammatory properties, may help to reduce gingival inflammation. Therefore, a diet, which reduces inflammation and insulin resistance, may help to maintain oral and brain health and thus may reduce the risk of AD (Singh et al., 2014).

Vitamins are an important element of a well-balanced diet and vitamin deficits can have important consequences for both periodontal disease and cognitive performance. Epidemiological studies show that low serum levels of 25-hydroxyvitamin D (25OHD) are associated with increased risk of dental caries and periodontal disease (as well as other risk factors such as cardiovascular diseases, T2DM and depression), which, in turn, are risk factors for developing AD (Grant, 2009; Toffanello et al., 2014a,b). Therefore, monitoring vitamin D levels in populations such as the elderly, who are at risk of AD, is important. Some individuals may be at a greater risk to a vitamin D deficiency because of their dietary choices. For example, vegans could have lower intakes because they avoid vitamin D rich food such as fatty fish and free-range eggs, or individuals who live in climates where exposure to the sun (UV-B radiation) is scarce throughout the year. More importantly, mature vitamin D results from the pro-hormone synthesis initiated by UV-B radiation from the sun and the ability to do so declines during aging. Properly targeted nutrition, careful sun exposure and supplements can help to achieve health-supporting values of 25OHD ≥75 nmol/l.

Vitamin B also appears to be important for protecting against the onset and progression of AD. Inadequate levels of B vitamins (folate, B12, B6) relate to cognitive impairment and incidence of dementia in the elderly (Morris, 2012), possibly because lower levels of these micronutrients elevate plasma homocysteine (HCys). HCys is a neuro and vasotoxic metabolite that may lead to impaired S-adenosylmethionine-dependent methylation reactions, which is vital to CNS functioning (Selhub et al., 2010; Köbe et al., 2016). Even without clinically manifested deficiency of vitamin B12, levels in the lower normal range were associated with poorer memory performance and a reduced microstructural integrity of the hippocampus, a brain area, typically affected first in AD (Köbe et al., 2016).

Foods like meat, eggs and fish provide sufficient amounts of vitamin B12 to the body (Cunnane, 2006). Strict vegetarians on plant foods lacking in B12 must consider supplementing their diet with B12. However, interventional studies with single or few nutritional supplements have yielded mixed results (Eussen et al., 2006; Cunnane et al., 2009; Quinn et al., 2010; Yurko-Mauro et al., 2010) and one explanation for this discrepancy maybe that in the elderly, methylation reactions are less effective. Therefore, some researchers recommend offering methylated folate and cobalamin to seniors to raise vitamin B12 or lower homocysteine blood levels (Bredesen, 2014; Hara et al., 2016). In a recent prospective case-control study, Hara et al. (2016) found patients with AD or related diseases who took a supplement containing L-methylfolate, methylcobalamine and N-acetyl-cysteine, showed a noticeable benefit. The authors suggest that the methylated vitamins are fully reduced and in bioactive functional form to combat hyperhomocysteinaemia (Hara et al., 2016). Another nutritional formulation containing sulfur containing N-acetyl-cysteine and S-adenosyl-methionine, in addition to vitamins, also showed some promise in slowing the cognitive decline seen in prodromal AD cases (Remington et al., 2015). In summary, careful monitoring of vitamin D and B levels, and dietary interventions as required, may help to prevent the onset and progression of AD. Additional research is needed to identify other vitamins and nutrients, which can support oral health and preserve cognitive functioning.

Recently attention has focused on the gut-brain axis and the role of probiotics in maintaining both physical and mental health. Diets rich in fruit and vegetables and low in meat are associated with a greater prevalence of Prevotella species rather than Bacteroides species, both of which express genes for different digestive enzymes (Greiner et al., 2014). A vegetarian diet has been shown to reduce gut inflammation, increasing commensal microbes and decreasing pathobionts (Kim et al., 2013). This in turn reduces intestinal lipocalin-2, an adipokine, associated with inflammation and insulin resistance development (Zhang et al., 2008; Kim et al., 2013).

Supplementation with probiotics appeared to be useful in the maintenance of oral health (Krasse et al., 2006); with the daily consumption of a probiotic milk drink reducing plaque induced gingival inflammation (Slawik et al., 2011). Furthermore, Riccia et al. (2007) have shown the probiotic Lactobacillus brevis to have an anti-inflammatory effect on patients with chronic periodontitis. However, a recent systematic review concluded that current probiotics based treatments for periodontal disease only produce short-term benefits (Jayaram et al., 2016; Bakarcic et al., 2017). Therefore, further research regarding whether the long-term prophylactic use of probiotics is useful to support oral health or whether probiotics to treat active periodontal disease is desirable.

Probiotics may also reduce the onset of AD by stimulating the production of GABA, the inhibitory neurotransmitter in the CNS, via glutamate signaling (Bhattacharjee and Lukiw, 2013). This is significant as studies performed on post mortem brain tissues show reduced GABA concentrations in AD patients. In addition, a recent randomized, double blind, controlled, clinical trial showed that consumption of a mixture of probiotics over a 12-week period had a positive effect on cognitive function in AD patients (Pistollato et al., 2016). Probiotics decrease the levels of proinflammatory cytokines IL-1β, IL-5, IL-6, IL-8 and TNF-α, which are upregulated in the elderly, and increase numbers of activated lymphocytes, natural killer cells and phagocytosis (Rincon et al., 2014; Wang et al., 2015). All these factors indicate an improvement in the adaptive immune response, along with a reduction in inflammation. Overall, it would appear that probiotics have the potential to provide a beneficial effect on cognitive decline and oral health. The fact that probiotics show an anti-inflammatory effect, along with the established GI tract-brain axis, would provide a possible explanation to this positive effect and suggest a potential mechanism for a reduction in onset of AD (Pistollato et al., 2016).

Not only do patients with T2DM and other states of insulin resistance show a higher risk of periodontitis, but they also show a higher risk for AD (Ahmed et al., 2017; Górska et al., 2017). In addition, high blood glucose emerges as a risk factor for both diseases (Crane et al., 2013; Miranda et al., 2017). Crane et al. (2013) reported a study of a large cohort of seniors without dementia over a period of 6.8 years. After adjusting for potential confounding variables, participants with higher blood glucose readings in the preceding years had a significantly increased risk of dementia (Crane et al., 2013). The negative impact of high glucose levels has implications for non-diabetic elderly individuals who have higher glucose and HbA1c levels on memory performance, hippocampal volume and microstructure (Kerti et al., 2013). Therefore, strategies aimed at lowering glucose levels, not only in diabetics but also individuals with blood glucose within in the normal range, may beneficially influence cognition and oral health in the older population. The best way to achieve optimal blood glucose levels is to reduce the intake of highly processed carbohydrate rich foods, high glycaemic carbohydrates and to increase fiber, fat and protein (Volek et al., 2009).

Lack of brain insulin or the intact functioning of the insulin signaling cascade leads to problems with Aβ clearing, learning and memory, satiety and regulation of feeding behaviors. Additionally, disturbed glucose utilization is seen as part of the normal aging process (Yao et al., 2011). Fortunately, the elderly, including those with mild cognitive impairment and mild forms of AD, are able to take up ketone bodies, like β-hydroxybutyrate (BHB) and acetoacetate, through insulin independent monocarboxylate transporters (Cunnane et al., 2016). Ketone bodies serve as an alternative and efficient fuel for neurons and other brain cells, as well as protecting neurons from oxidative damage. Moreover, the ketone body BHB is able to raise the levels of brain derived neurotrophic factor (BDNF), which is important for growth of new neurons in the hippocampus (Cunnane et al., 2016; Kullmann et al., 2016).

Ketone bodies are metabolites of fatty acid oxidation. The liver is the main peripheral organ to produce ketone bodies, but the astrocytes in the brain are also able to build ketones from medium chain fatty acids (Nonaka et al., 2016; Thevenet et al., 2016), which are saturated fatty acids with chain lengths of 8–12 carbon atoms (C8–C12). Research on healthy individuals has shown that small amounts of medium chain triglycerides (MCTs, 20–30 g C10/d) are able to close an energy gap typically found in aging individuals (Courchesne-Loyer et al., 2013) and delay the onset of AD symptoms (Reger et al., 2004). The achieved degree of ketonemia (ca. 0, 5 mmol/l) was estimated to contribute to up to 9% of brain energy metabolism. For higher ketone body levels it would be necessary to eat more MCTs, adopt a strict diet (very low-carbohydrate ketogenic diet) or to fast.

As MCTs, such as those found in virgin coconut oil, are able to raise ketone body levels mildly, even in the presence of an ordinary diet, they offer an easy way to achieve supply of alternative fuel for neurons with compromised glucose utilization. In a case study, Newport et al. (2015) reported improvements in a patient with early onset AD using coconut oil, MCTs and ketone supplements. As even the healthy aging brain shifts from glucose-driven bioenergetics towards a compensatory ketogenic pathway (Yao et al., 2011), it seems sensible to use ketogenic foods (high quality, natural fatty foods and fats) to reduce the risk of cognitive decline as a result of energy deficiencies and insulin resistance. Although coconut oil raises hepatic ketone levels only mildly, it appears beneficial and potentially helpful for astrocytes to synthesize ketone bodies and then deliver them to neighboring neurons. A study with astrocytes in culture demonstrated that lauric acid (C12), the major medium chain saturated fatty acid constituent of coconut oil, is able to cross the BBB and activate ketogenesis in astrocytes (Nonaka et al., 2016). Alternatively, oil pulling (whereby oil is held in the mouth) with coconut oil was able to reduce the S. mutans populations in saliva, and this could be a means to aid the maintenance of oral health (Kaushik et al., 2016). A preliminary report by Peedikayil et al. (2015) described the effects of oil pulling with coconut oil, which contains a high proportion of the anti-inflammatory and antimicrobial compound lauric acid. It led to a decrease in plaque formation and improved gingival indices after 30 days of its application in African adolescents.

Smoking tobacco increases the severity with which periodontal disease progresses. For example, the use of nicotine is shown to select for periodontal pathogens, including P. gingivalis, Treponema denticola and Tannerella forsythia, which will encourage progression of this oral disease (Zambon et al., 1996). The periodontal indices change dramatically, as smoking not only leads to harboring higher proportions of oral bacteria that colonize rapidly, but also encourages colonization of species of the more virulent, red complex pathogens such as T. forsythia and P. gingivalis strains than not smoking (Kamma et al., 1999). The highly virulent forms of T. forsythia and P. gingivalis in greater numbers are capable of maintaining severe inflammation, and their invasive property continues to exacerbate pathology of the periodontium (Kamma et al., 1999; Bergström, 2004). The inflammation arising from the microbes which result from poor hygiene along with secondary mediators from bone regulatory factors (Takahashi, 2005; Olsen et al., 2016) and the innate immune response in smokers ultimately succumb to a higher rate of alveolar bone erosion (Grossi et al., 1995). Therefore, inflammatory burden due to smoking (Kamma et al., 1999) can undermine the developing chronicity of periodontitis alongside the impact on health of the distant organs affected by periodontal pathogens.

Drinking alcohol can lead to a moderate increase in the risk of periodontal disease (Tezal et al., 2004; Genco and Borgnakke, 2013). High alcohol consumption affects the innate immune system in a negative manner; with chronic alcohol consumption being predictive of higher levels of inflammatory markers (McDade et al., 2006; Barr et al., 2016). Conversely, drinking alcohol in moderation is thought to reduce periodontal disease, by a positive effect on the innate immune system, and so, reduce the risk of vascular diseases, and AD. In addition, moderate amounts of wine (Evers et al., 2009) can have beneficial anti-inflammatory effects and so may reduce the risk of developing chronic diseases, such as periodontitis and AD.

Recent research by Sprecher et al. (2017) has highlighted the link between poor sleep and AD development. The study found higher levels of pre-clinical AD biomarkers in cerebrospinal fluid samples in people reporting several measures of poor sleep such as higher levels of daytime somnolence. Therefore, attending to good sleep hygiene is an important preventative measure against AD development (Harding et al., 2017). Specifically, adequate sleep in advancing age is important for retaining memory, and for the effective functioning of the glymphatic system mediated cleansing of the brain (Iliff et al., 2013). The HPA axis has been shown to play a role in the maintenance of sleep/wake patterns. Infections from dysbiotic microbiomes including the GI tract microbiome will modulate the HPA through changes in the neurochemicals and inflammatory mediators (Galland, 2014). For example, proinflammatory cytokines are potent inducers of non-rapid eye movement (nREM) sleep (Galland, 2014). Deterioration of the circadian system that controls sleep/wake cycles in the elderly may contribute to the excess accumulation of abnormal proteins in AD brains and impairments to the glymphatic cleansing system (Yesavage et al., 2003). Active periodontal disease and reduced glymphatic system functioning, due to restricted sleep periods, can also impair the brain’s ability to clear microbes, such as P. gingivalis (Harding et al., 2017). Furthermore, there is some evidence to suggest P. gingivalis, the keystone pathogen of chronic periodontitis, accesses the brains of experimental mice (Poole et al., 2015; Singhrao et al., 2017), and can disturb the microglial cell day/night activity (Takayama et al., 2016). This implies that P. gingivalis infection through chronic periodontitis can further contribute to poor quality sleep.

Humans are diurnal mammals and hence many of their metabolic activities have evolved for activity and storage during the day, and rest, repair and use of stored nutrients during the night. This is supported by hormonal rhythms, like the melatonin rhythm, which is mainly regulated by light and heavily disturbed by blue wavelength emitting devices like smartphones, tablets, TV screens and e-readers (Heo et al., 2017). The daily rhythms make us more insulin resistant in the evening and during the night (Carrasco-Benso et al., 2016; Morris et al., 2016), therefore large dinners, especially those with a heavy carbohydrate load should be avoided in the evening when the body is meant to fast and to use stored fats. In addition, a heavy dinner can disturb sleep and impair sleep quality. Therefore, it is recommended that people fast for at least 3 h before going to bed and to reduce the carbohydrate content, especially of the evening meal (Bredesen, 2014).

Other dietary interventions found to help promote sleep include, avoiding caffeine-containing foods and drinks, if not well tolerated, before bedtime, and alcohol, which can increase sleepiness but disturbs regenerative deep sleep phases (Ebrahim et al., 2013; Clark and Landolt, 2017). Alternatively, eating cherries or kiwifruits can promote sleep via their naturally high levels of melatonin (Pigeon et al., 2010; Lin et al., 2011). Drinking milk before bed may also increase drowsiness due to its high level of tryptophan and B vitamins (Peuhkuri et al., 2012). Finally, exercise taken during the day can have a positive impact on sleep, especially when undertaken outside in natural daylight (Reid et al., 2010).