Dementias, such as AD, are the most common types of NDs, with a higher prevalence in females than males (Mielke, 2018; Podcasy and Epperson, 2016; Y; Wang et al., 2020). The hallmark pathologies of AD, identified in both sexes, negatively affect cognition, predominantly long and short-term/working memory, and locomotor behavior (Clancy, 2013; Da Costa et al., 2013; Ju et al., 2013; Levenson, et al., 2014; Raskin et al., 2015; Podcasy and Epperson, 2016). The formation of aggregates of misfolded protein oligomers such as extracellular β-amyloid plaques or intracellular neurofibrillary tangles (NFTs), made of tau microtubule-associated protein (Mohandas et al., 2009; Kou et al., 2011; Hampel et al., 2021; Tracy et al., 2022) are late pathological hallmarks that cause damage to mitochondria and oxidative stress in neurons. Of protein aggregates in neurons and neuronal loss, observed as inclusion bodies that form lesions throughout the tissue (Takatori et a., 2019; Morley et al., 2018; Raskin et al., 2015; Sarantseva et al., 2009; Shadfar et al., 2015). The inclusion bodies spread over time to surrounding neurons through cell-to-cell prion-like propagation (for a detailed review on the cell-to-cell prione propagation, please refer to Hofmann et al., 2013), causing high oxidative stress and widespread inflammation of the brain, ultimately facilitating atrophy of the tissue as the disease progresses (Ashraf et al., 2014).

The Amyloid Precursor Protein (APP) is one member of a single-pass transmembrane protein family characterized by large extracellular domains. When APP is proteolytically cleaved by β/γ secretases, the product is amyloidogenic ABeta/β-amyloid peptide (Aβ) (Chen et al., 2017). The role of APP and its products is not very clear. The preeminent role of APP in the development of AD depends on the toxicity of the Aβ peptide since the loss of APP function does not seem deleterious. In AD patients, accumulation and aggregation of Aβ peptides forms Aβ fibrils that are acutely toxic to neurons. Aβ fibrils formation toxicity might explain also other pathological aspects of AD including neurofibrillary tangles, inflammation, and oxidative damage. For a detailed review on the pathogenesis of AD we refer the reader to a review by (O'Brien and Wong, 2011).

APP and β/γ secretases are not exclusive to the CNS. Recent studies reported regular APP expression in the Enteric Nervous System (ENS) of mice, suggesting an ENS involvement in AD (Van Ginneken et al., 2011; Semar et al., 2013; Chalazonitis and Rao 2018). The transgenic mice that over express a mutant form of human APP in the ENS are associated with early-onset familial AD, exhibiting an accumulation of Aβ in the enteric neurons leading to a decrease in enteric neuron abundance, dysmotility, and increased vulnerability to inflammation (Chalazonitis and Rao 2018). Preliminary data confirms that changes in location or amount of APP in the ENS correspond to disease expression in transgenic mice carrying APP mutation. APP aggregation in ENS appeared before any disease sign was detectable in the brain. Similarly, Αβ deposits have been observed in the Gastrointestinal (GI) tract of patients that overexpress APP (Semar et al., 2013; Sun et al., 2020). Although the study of the role of Aβ in the gut in the context of AD is still in its infancy, there have been increasing evidence that support a model where Aβ triggers NDs from the gut in the brain (Braak et al., 2003; Chalazonitis and Rao, 2018; Kowalski and Mulak., 2019).

The nature of this distal pathological seeding was hypothesized by Dr. Heiko Braak. His theory was based on observations that the pathogenesis of NDs, in the context of PD, was mediated by α-Synuclein inclusion bodies that develop first in the gut before translocate to the brain via the vagal and motor nerves by an unknown mechanism of gut-brain communication (Braak et al., 2003) (see next section). This hypothesis was then applied to β-amyloid pathogenesis by (Sun et al., 2020). In this study, 2-month-old ICR mice were injected with HiLyte Fluor 555-labelled Aβ42 (the APP isoform most prone to oligomerization) into the stomach and colon and exhibited defects in responsive/exploratory behavioral, short-term and long-term memory cognitive deficits 1-year post-injection. Despite evidence that the labelled Aβ42 monomers did not diffuse through the tissue, Aβ42 remained localized to cholinergic neurons at the injection sites 3 hrs and 3 days post-injection. Further, these mice displayed clear depositions of Aβ plaques throughout the brain, and vagus nerve (DMV), with plaques visible in the hippocampus, cortex, amygdala, and blood vessel walls. Thus, intra-GI administration of Aβ had a direct effect on the neuronal system, with β-amyloid deposits from the gut translocating to the CNS from the ENS via the vagal nerve by an unknown mechanism of gut-brain communication. Although this work hypothesized that a retrograde transport route had caused the presence of Αβ aggregates in the brain of the mice, there remains no direct evidence demonstrating that enteric Αβ seeds can retrogradely invade the CNS to induce AD symptoms. However, it is posited that the enteric Aβ seeds invade the brain and cause dementia by retrograde axonal transport through the vagal nerves and haematogenous routes, as seen in prion pathology (Watts et al., 2014). The enteric Αβ seeds did not seem to affect GI major functions when the animals were analysed at 1-year post-injection and only alteration of spontaneous contractions and neuronal couplings of the jejunum, which did not result in constipation, were observed. A difference in contraction frequency of the GI was detected also in another APP transgenic mouse (Semar et al., 2013). Other interesting metabolic defects such as weight gains were observed in various transgenic models. Weight changes (Wagner et al., 2019) have been associated with dementia in humans and are linked to disease progression (Ikeda et al., 2002). In a similar study (Galloway et al., 2007), identified neuronal tissue death alongside translocation of β-amyloid from the gut of mice.

While these murine studies demonstrated that Aβ42 aggregates can form in the GI tract and then enter the brain leading to cerebral amyloidosis and AD-like dementia, other studies showed the existence of a correlation between the amount Aβ42 protein and the extent of oxidative stress and inflammation in the brain of mice and humans (Tamagno et al., 2021). The brain’s vulnerability to oxidative stress is considered a crucial detrimental factor in AD. Aβ induces oxidative stress, and on the other hand oxidative stress can increase Aβ deposition. Notably, while the increase of soluble Aβ42 is correlated to elevated oxidative stress, inflammation, and tissue atrophy of the brain, neither the rate of dementia nor the extent of neurological damage is correlated with the Aβ amyloid. Studies on transgenic mice carrying AD-linked mutations in the gene for APP demonstrated the existence of soluble Aβ oligomers long before the deposition of β-amyloid, further supporting the hypothesis that in particular conditions, an over-production of soluble Aβ aggregates occurs in the human brain even in the absence of plaques. One of the main mechanisms that appears to break this balance is oxidative stress and neuroinflammation (Chang et al., 2014). There is growing consensus that oxidative stress represents a common mechanism that mediates the accumulation and toxicity of Aβ (Saeedi and Rashidy-Pour, 2021). Therefore, oxidative stress can be considered one of the factors responsible for the accumulation of Aβ.

Preliminary evidence from transgenic APP mice and AD human patients indicates that build-up of Aβ in enteric neurons causes inflammation of the gut before any sign of disease is detected in the brain. Enteric inflammation promotes “leaky” guts that release inflammatory mediators/bacteria-derived products into the blood circulation, leading to systemic/neuro-inflammation by weakening the BBB (Chalazonitis and Rao, 2018; Kowalski and Mulak, 2019). For a review of the role of the BBB in neurodegeneration, we refer the readers to (Hussain et al., 2021). Other studies support the theory that oxidative stress and inflammation in the gut may instead trigger β-amyloid aggregation in the brain by means of chronic, low-grade systemic inflammation (Braak et al., 2003). This was proposed from observations that a long period of gradual accumulation of oxidative damage precedes and results in the appearance of clinical and pathological AD symptoms, including Aβ deposition, NFT formation, metabolic dysfunction, and cognitive decline. This suggests that AD begins many years before its symptoms appear, and that antioxidant treatment can be an important therapeutic target to treat the disease (Saeedi and Rashidy-Pour, 2021). Like many neurological disorders, AD is associated with a variety of GI symptoms, raising the possibility that the ENS could also be affected. A clinical report showed that Aβ plaques were found in the submucosa of two AD patients (Joachim et al., 1989). However, with a paucity of evidence and some discordance in the findings, there remains a clear need for further studies of enteric neuronal pathology in AD.

The stimuli that trigger β-amyloid inclusion formation in the gut remain to be elucidated. The gut microbiota is a significant source of amyloids, changes in the microbiota may potentiate disease in the gut. Disruption of the microbiome occurs as a result of pathogenic infection, antibiotic treatment, aging, and local inflammation (Hakansson and Molin, 2011; Rinninella et al., 2019). Certain bacterial species in the gut, such as E. coli, can even produce bacterial amyloids that help the bacteria to form biofilm and survive mechanical and immune stressors (Kowalski and Mulak, 2019). Although bacterial amyloids differ from CNS amyloids in their primary structure, they share similarities in their tertiary structure (Friedland, 2015; Zhao et al., 2015). Shifts in commensal populations that favour species releasing amyloids would increase the concentration of bacterial-derived amyloid proteins in the gut triggering a prion-like cascade of β-amyloid oligomerization/aggregation in the gut (Hetz and Saxena, 2017) or promote cleavage of APP into its Aβ-amyloid peptides (Sarantseva et al., 2009; Bekris et al., 2010; Simon, 2018; Li et al., 2019). A pioneering study of PD carried out by (Chen et al., 2016) demonstrated that rats exposed to amyloid-producing Escherichia coli displayed increased neuronal α--Synuclein deposition in both the gut and brain and enhanced microgliosis and astrogliosis compared to rats exposed to bacteria without the ability to produce amyloids. Moreover, in the brain of animals exposed to amyloid-producing bacteria they measured an increased expression of inflammatory cytokines and reactive oxidative and nitrosative stressors which lead to neuronal and glial cell death (Chen et al., 2016; Li et al., 2016). Thus, bacterial amyloids can act as molecular mimics of prion proteins, eliciting cross-seeding, in which one amyloidogenic protein (bacteria amyloid protein) promotes another (e.g., host proteins) to also adopt a pathogenic β-sheet structure (Lundmark et al., 2005; Zhou et al., 2012).

Therefore, a combination of gut inflammation and dysbiosis is directly associated with gut barrier dysfunction and increased intestinal permeability (“leaky gut”) may contribute to the process of neurodegeneration (Marizzoni et al., 2017; Sochocka et al., 2019). Lipopolysaccharide (LPS) from commensals or pathogenic bacteria leaked from the gut can reach and accumulate in the brain and activate microglia, brain resident immune cells, and elicit inflammatory responses in the brain. LPS has been detected in the hippocampus and neocortex brain lysates from AD patients and has been found to colocalize with Aβ40/42 β-amyloid plaques (Zhao et al., 2017). Interestingly, the plasma concentration of LPS in AD patients is significantly higher than in healthy people (Zhang et al., 2009), and repeated systemic exposure to LPS in mice induced microglial priming and prolonged cytokine production.

Perturbation in microglia activation has been identified as a feature of NDs progression. For full reviews on the topic, we refer the readers to (Hickman et al., 2018) and (Muzio et al., 2021).

Tauopathies, are a group of NDs that include a form of AD, which are characterized by abnormal hyperphosphorylation of microtubule-associated protein tau that leads to the formation of NFTs. Similar to β-amyloid pathogenesis of NDs in the brain, tau monomers oligomerize and form aggregates known as NFTs and neuritic plaques, which form inclusion bodies that cause inflammation due to intra/extra-cellular damage in and around neurons (Anwal, 2021; Bălaşa et al., 2020; Takatori et al., 2019). NFTs have been found to accumulate in the gut and to cause gastrointestinal dysfunction such as alteration of gut motility before any clinical symptom is found in the brain (Kametani and Hasegawa, 2018). Moreover, both β-amyloid and also tau tangles/plaques promote local intestinal inflammation and weakening of the gut epithelial barrier (Kowalski and Mulak, 2019). The weakening of the gut epithelial barrier can cause metaflammation (low-grade chronic systemic inflammation) that has been linked to NDs (Komleva et al., 2021; Rydbom et al., 2021). For a review on metaflammation we refer the readers to (Itoh et al., 2022).

Although the mechanisms of DMGB axis pathogenesis in the context of ADs are poorly defined, concurrent evidence shows that the diseases may manifest/originate in the gut before appearing in the brain.

PD has been identified as one of the most rapidly growing NDs in the world (Dorsey et al., 2018) and is more prevalent in males than females (Moisan et al., 2016; Podcasy and Epperson, 2016; Feigin et al., 2019). This disease is primarily defined by progressive loss of motor-function (initiation and coordination) of the brain, along with bradykinesia. These behavioral disturbances are due to atrophy of the substantia nigra (Faustini et al., 2017; Dorsey et al., 2018), an essential medullary brain region that produces and projects dopamine to the limbic system, and higher cortical areas (Mayfeild, 2010). These projections are known as nigrostriatal pathways, activating regions of the brain that control motor-behavior (Zeighami et al., 2015; Faustini et al., 2017; Sonne, et al., 2021). Despite a similar pattern of atrophy in PD brains compared to healthy aging brains, the pathologies of the disease are what exacerbate and accelerate this phenotype (Zeighami et al., 2015). Hallmark pathologies of PD in the brain include intracellular misfolded protein aggregates of α-Synuclein oligomers (synucleinopathies) that form plaques called Lewy Body inclusions in the substantia nigra (Huang et al., 2015; Wang et al., 2016; Shahmoradian et al., 2019). These intracellular inclusions cause aberrant neurotransmission because, α-Synuclein is a SNAP-associated protein in neurons of the CNS, PNS and ENS that tethers and primes pre-synaptic vesicles for exocytosis (Killinger et al., 2019). Considering the similar effects of tau and β-amyloid to α-Synuclein proteins in neurotransmitter trafficking along the axon and release, it is not surprising that PD and AD are comorbid. Especially, since unresolved misfolded proteins with prion-like effects are characteristic of both diseases (Xie et al., 2014).

Recent works demonstrated that, α-Synuclein can be detected in the ENS years before clinical onset of PD, similar to tau and β-amyloid (Hawkes et al., 2010; Shannon et al., 2012; Devos et al., 2013; Driver-Dunkley et al., 2014; Hilton et al., 2014; Iranzo et al., 2014; Stokholm et al., 2016). This α-Synuclein accumulation causes constipation and inflammation of the gut that often leads to Inflammatory Bowel Disease (IBD) (Barbut et al., 2019; Brudek, 2019; Chen et al., 2019). As mentioned in the previous section, studies in PD rodent models have shown that α-Synuclein seeded in the enteric neurons of the colon, duodenum, and stomach (Shannon et al., 2012; Sánchez-Ferro et al., 2015; Kim et al., 2019; Challis et al., 2020) migrate via the vagal nerve to the brain (Braak et al., 2003; Holmqvist et al., 2014; Uemura et al., 2018). This was demonstrated in experiments where the α-Synuclein-rich lysate, prepared from severely affected PD patient substantia nigra brain tissue, was injected into the intestines of wild type rats leading to the detectable accumulation of α-Synuclein over time in both the DMV and the brain (Holmqvist et al., 2014). This study also showed that α-Synuclein was selectively up-taken by the DMV compared to BSA. Notably, α-Synuclein accumulated at the Dorsal Motor nucleus of the vagus nerve (DMV), a collection of cholinergic neurons. Furthermore, DMV migration of α-Synuclein was prevented by a vagotomy of the base-afferent vagal nerve, as demonstrated by (Kim et al., 2019) (Figure 2). Therefore, cholinergic neurons must facilitate long-distance translocation of α-Synuclein (monomer, oligomer, and fibril) via a microtubule-associated transport mechanism (Holmqvist et al., 2014). By this notion, the question that surfaces is whether the α-Synuclein inclusions in the DMV are the reason behind the GI issues observed in PD patients, or if the symptoms are the reason for the aggregation of α-Synuclein, exacerbating symptoms and pathology in the DMV.

Several studies have confirmed the migration of α-Synuclein from the ENS (Challis et al., 2020), by tracking fluorescently tagged-α-Synuclein, inoculated into the intestines of mice. The tagged-α-Synuclein translocated from the gut to the brain, and created human Parkinson-like symptoms, that affected sensorimotor behaviors including diminished ability of mice to climb down a pole or hang from a metal wire for an extended period of time. In addition, abnormalities in GI function as well as inflammation were reported in mice that received an intestinal injection of tagged-α-Synuclein. This recapitulated symptoms experienced by pre-clinical human PD patients (i.e. constipation and bowel inflammation) (Dinan and Cryan, 2017; Challis et al., 2020). Therefore, α-Synuclein pathogenesis can be seeded in the gut and ultimately, via the gut-brain axis, move to the brain. However, we still need to understand what triggers the accumulation of α-Synuclein in the intestine and leads to pathogenesis.

Since IBD and therefore intestinal inflammation occurs in PD patients before the brain pathologies begin, diet and commensals are suspected to play a role in the formation of α-Synuclein Lewy bodies (Rinninella et al., 2019; Mertsalmi et al., 2020). Mitochondrial abnormalities are observed in the gut of IBD (Jackson and Theiss, 2020), and PD (Brudek, 2019) patients. Protein α-Synuclein aggregates have been found to cause mitochondrial damage by interacting with complex-I of the electron transport chain, and slowing down mitochondrial metabolism (Kamp et al., 2010; Faustini et al., 2017). This type of mitochondrial stress increases the redox state of intestinal epithelial cells and creates inflammation (Friedman and Nunnari, 2014; Console et al., 2020). Therefore, α-Synuclein in PD might act just like β-amyloids and tau in AD, as a facilitator and amplifier of PD by seeding pathology in the gut, ultimately affecting the brain (Braak et al., 2003; Kitani-Morii et al., 2021). As for AD, the pathology in the gut could be influenced by oxidative stress and tissue inflammation of the gut that enhance protein aggregate formation and/or gut permeability that is translated to the brain, causing neuronal death and impairment of memory and locomotion. Thus, protein α-Synuclein aggregates in the gut can transfer to the brain or can generate local inflammation that distally impacts the BBB and neuronal activity as reported of Aβ amyloids.

Additionally, several genome-wide association studies revealed that mutations in leucine-rich repeat kinase 2 (LRRK2) that are one of the greatest genetic contributors to PD are also linked to increased incidence of Crohn’s Disease (CD), a form of IBD (Hui et al., 2018). Although to date, the pathology of CD does not include neuropathology, CD is characterized by an increase in leakage of the epithelium, inflammation, and dysbiosis in the GI, all features that seem to contribute to neuroinflammation and neuropathology in PD. Therefore, further studies looking at the role of LRKK2 in CD could help to define the mechanisms of the disease of PD and the role of gut-brain communication in the PD pathogenesis.

Overall, NDs such as AD and PD do not arise exclusively from defects in the brain, but signals coming from peripheral organ environments and inter-organ interaction can wire messages to the brain and effect the brain. Therefore, more studies are required to dissect the mechanisms of action of the DMGB in health and NDs. Fortunately, the use of multidisciplinary approaches and more amenable model organisms will fast-track our knowledge in this field.

The mouse mammalian model system in particular presents 90% synteny to the human genome (Monaco et al., 2015) and has been used in over 95% of studies of human disease (Vandamme, 2014). Mouse models of NDs have facilitated the study of neurodevelopmental and neurological diseases, and advanced our understanding of the molecular pathogenesis of NDs in humans (Dawson et al., 2018; Zhao and Bhattacharyya, 2018). In particular, the models that produce similar ND pathologies within the most conserved region of the mouse brain, the neocortex, have been extensively used (table 1). For example, transgenic or knockout (KO) mouse models for tau, β-amyloid, and α-Synuclein (table 1) have been used to investigate the physiological roles of the familiar forms of PD genes and to model dopaminergic degeneration caused by the dysfunction of these genes. Although these models exhibit various pathological and behavioral phenotypes, some aspects of PD pathogenesis such as the degeneration of distinct dopaminergic neurons or NFTs have not been reproduced (Harada et al., 1994; Cabin et al., 2002; Hasegawa et al., 2009; Chen et al., 2022).

In context of the intestinal epithelium, mice have also been extensively used in biomedical disease studies of intestinal dysplasia and inflammation such as IBD, or infection models to better understand the pathogenesis of gut-borne pathogens (Gkouskou et al., 2014; Vandamme, 2014; Stanford, et al., 2020). The fundamental gross/cellular anatomy of the mouse gut is also very similar and has been considered one of the best comparable animal models to the human gut (Figure 3) (Trinder et al., 2017; Stanford et al., 2020).

With regards to studies defining the effect of microbiota on animal physiology, the mouse has one of the most similar gut-commensal populations to the human when comparing the presence and abundance of major phylogenetic groups, such as Firmicutes, and Bacteroidetes (Figure 3) (Xiao et al., 2015; Trinder et al., 2017). However, studying the highly-intertwined DMGB axis is more challenging in a complex mammalian system where the cost and infrastructural hurdles of constructing germ-free mouse models is high. Moreover, it is very difficult to control multiple factors while performing experiments where environmental changes may greatly influence the results (Trinder et al., 2017). These challenges together with the high cost and time required to generate these models add further risk to the use of such models given the overarching difficulty in knowing if the phenotypes observed accurately recapitulate disease states in humans (Vandamme, 2014). These issues have prompted researchers to incorporate alternative model organisms for these studies.

The effects of dietary lipids on animal physiology are heavily regulated at the cellular level by individual metabolism. Cellular organelles, such as mitochondria and peroxisomes, play central roles in dietary lipid catabolism/anabolism and productions of lipid mediators vital for signaling and metabolism. These organelles influence the impact that dietary lipids have on animal physiology. Very long-chain fatty acids (VLCFAs) derived from vegetable oils, nuts, and seeds (Tenreiro et al., 2013) are normally catabolized by peroxisomes and can be lipotoxic to any tissue if not metabolized. Accumulation of VLCFAs is linked to neurodegeneration and has been found in AD brains along with reduced peroxisome activity (Kou et al., 2011). A study carried out in D. melanogaster (Di Cara et al., 2018), found that peroxisomal dysfunction specifically in the intestinal epithelial cells damages intestinal epithelial integrity resulting in dysplasia, and promotes the systemic accumulation of non-esterified fatty acids. These fatty acids trigger intestinal inflammation and decrease animal survival, confirming that abnormal lipid metabolism affects inflammation. Likewise, the oxidation of long-chain fatty acids by mitochondria is protective, as mitochondria defects lead to an accumulation of long chain fatty acids in the gut epithelium which induces high levels of oxidative stress (Swanson et al., 2012; Custers et al., 2022) and inflammation that affects ENS neurotransmission (Chen et al., 2019).

The brain has the highest cholesterol content in the body as the main component of axonal myelin sheets, essential to maintain the physiological functions of the brain such as cognition and movement (Rhea and Banks, 2021). An excess of cholesterol can be damaging and therefore needs to be oxidated into the oxysterol 24S-OH-Chol by the neuronal enzyme CYP46A1. 24S-OH-Chol is then released into the circulation by neurons and eliminated. As oxysterols can diffuse through plasma membranes, they can then cross the BBB. AD patients show a reduction in circulating levels of 24S-OH-Chol, because of decreased expression of the CYP46 enzyme in neuronal cells. Oxysterols are also present in plasma mainly in the form of 27-OH-Chol as a product of the mitochondrial CYP27A1 enzyme activity in the liver. Recent works suggested that 27-OH-Chol accumulation triggers AD-related pathological changes. Rabbits fed a cholesterol-rich diet showed increased serum cholesterol, increased 27-OH-Chol in the hippocampus, and pathological hallmarks similar to those found in AD, such as microglia activation, brain atrophy, and cognitive deficits (Brooks et al., 2017). Therefore, it is suggested that dietary cholesterol could induce neurodegeneration if there is excessive production of 27-OH-Chol by the mitochondria in the liver.

Important lipids for most physiological functions are not just dietary derived but are also products of gut microbiota metabolism. In mice and humans, it has been shown that the gut microbiota can affect lipid metabolism and lipid levels in the blood and different tissues (Chatelier et al., 2013; Cotillard et al., 2013; Lam et al., 2015; Just et al., 2018). However, dietary lipids can also alter the gut microbiota composition either by acting as substrates for bacterial metabolic processes or by diminishing bacterial growth due to their toxicity towards certain bacterial species (e.g. Bacteroides, Clostridium, and Roseburia) (Agans et al., 2018; Schoeler and Caesar, 2019). As a result, dietary lipids can shape their own metabolism by altering the balance of enteric bacteria that are able to metabolize them. SCFAs in particular are the metabolic end-products of gut-commensal fermentation of fibrous food, that have neuroprotective roles (Onyango et al., 2015; Dinan and Cryan, 2017; Baxter et al., 2019; Custers et al., 2022). For example, in a study conducted by (Kong et al., 2018), 16S rRNA gene sequencing and gas chromatography-mass spectrometry analyses in a Drosophila model of AD, showed decrease of Lactobacillus and Acetobacter species that correlate to a dramatic decrease of SCFA acetate, which is the most abundant SCFA. Concurrently, in Drosophila models of PD, the administration of sodium-butyrate reduces degeneration of dopaminergic neurons and improves the locomotor defects in a pan-neuronal transgenic fly model expressing mutant-human-α-Synuclein (St Laurent et al., 2013). SCFAs have also been linked to the maintenance of gut and immune homeostasis in mammalian systems, demonstrating a neuro-immunoendocrine regulatory role in the brain (Silva et al., 2020; Chen et al., 2022). For example, dietary butyrate has an anti-inflammatory effect in the gut (Baxter et al., 2019) and in the brain by influencing BBB permeability (Chen et al., 2022).

One line of communication between the gut microbiota and the central brain occurs via the DMV as shown by studies in mice demonstrating the anxiolytic effect of intestinal Bifidobacterium longum on animals that were subject to vagotomy (Bercik et al., 2011). As discussed above, it has also been proposed that commensal bacteria can secrete prion-like peptides that could trigger other proteins in the gut to aggregate (Chen et al., 2016; Chen et al., 2019). These misfolded aggregated proteins then spread through the intestine and the ENS to the brain. Protein aggregation of NDs peptides reach the brain via the DMV and a prion-like mechanism (Braak et al., 2003). This hypothesis was confirmed from preliminary data in Rattus norvegicus fed with bacteria that produce the prion-like protein, curli, which resulted in α-Synuclein deposition accumulation in the gut and brain (Chen et al., 2016).

In its abundance, the human microbiome is composed of around 1.3 more cells than cells of the human body (>200 taxa) (Sender et al., 2016; Yang et al., 2020), acting like a major organ system that contributes to physiological health. In fact, dysbiosis of gut commensal populations leads to metabolic and inflammatory diseases (Dinan and Cryan, 2017). Neurological diseases, such as NDs are co-morbid with gastrointestinal alterations (Vandvik et al., 2004; Sampson and Sarkis, 2015). It has indeed been reported that in both mouse and Drosophila models of human PD, the neurodegenerative pathologies in the brain have been observed in concurrence with commensal dysbiosis in the gut (Erny et al., 2015; Challis et al., 2020; Peterson, 2020; Kitani-Morii et al., 2021; Rydbom et al., 2021).

At a phylum level, the microbiome is primarily defined by two dominant bacterial phylotypes, Bacteroidetes and Firmicutes with Proteobacteria, Actinobacteria, Fusobacteria and Verrucomicrobia present in relatively low abundance (Qin et al., 2010; Lankelma et al., 2015). In humans, mice, and Drosophila, the phyla Firmicutes dominate the gut, where the Firmicutes is composed of many Lactobacillus species (Newell and Douglas, 2014; Trinder et al., 2017; Douglas, 2018). These Lactobacillus species have been found to influence host metabolism and behavioral pathologies of NDs in various gut-commensal studies in mammals and Drosophila (Newell and Douglas, 2014). For example, L. plantarum in mammals influences gut-lipid metabolism, increasing butyrate which has anti-inflammatory effects (Storelli et al., 2011). In Drosophila, Lactobacillus plantarum also alters gut metabolism such as the insulin signaling pathway implicated in the pathogenesis of NDs (Storelli et al., 2011; Newell and Douglas, 2014; Spinelli et al., 2019; Komleva et al., 2021). Moreover, L. plantarum also ameliorates age-dependent memory impairment in the mouse (Castelli et al., 2021) and Drosophila models of NDs (Schretter et al., 2018; Schretter, 2020). Studies in Drosophila have identified Lactobacillus brevis as the culprit behind the locomotion deficit of PD pathogenesis in a Drosophila PD model (Schretter et al., 2018). Therefore, in humans, interaction of Lactobacillus, or other undiscovered or unrecognized species that influence the gut-brain axis may hold the key to unlocking the aetiologies of NDs. However, sequencing and understanding the human microbiome is much more complex than a mouse or a fly.

In humans, the gut microbiota has recently been sequenced in patients with PD in a study carried out by (Scheperjans et al., 2015). They performed metagenomic analyses of 72 PD patients and 72 matched controls. The results indicated a major reduction in the levels of Prevotellaceae in the patients and a positive association between the levels of Enterobacteriaceae and the severity of postural instability and gait difficulty. However, the authors did not claim either the temporal or causal relationship between the gut microbiota and the features of the disease. Another analysis of microbiota composition in PD pointed to a reduction in butyrate-producing bacteria in feces and Faecalibacterium in the mucosa of PD (Keshavarzian et al., 2015).

Despite all the encouraging studies pointing to the microbiota as a potential therapeutic target for PD, to date, there is no conclusive evidence as of yet showing that microbiota transplants can ameliorate PD pathology. More research is needed to determine the relative role of the microbiome in PD.

A detailed analysis of the microbiota in AD patients is lacking, preliminary data published in preprint format has implicated the microbiota in the accumulation of amyloid plaques in a mouse model of AD (Harach et al., 2015). In this study, the authors generated a transgenic AD mouse model under germ-free conditions and found a dramatic reduction of cerebral Aβ amyloid pathology when compared to AD animals, which had normal intestinal microbiota. Colonization of germ-free AD mice with microbiota harvested from conventionally raised AD mice dramatically increased cerebral Aβ pathology. This was similarly observed in another study showing that antibiotic treatment could limit Aβ pathology and neuroinflammation (Minter et al., 2016). These data offer hope for the future generation of a novel microbiota-based approach to ameliorate symptoms of AD.

Gut-brain communication also occurs via the immune system. Signaling from gut epithelial cell-secreted peptides, bacterial-derived molecules, and cytokines can activate gut resident immune cells to secrete gradients of cytokines that pass the BBB and communicate with cells in the brain. The activation of microglia, the brain’s resident immune cells, plays a central role in neuroinflammatory processes in aging and NDs (Jyothi et al., 2015). The activation of microglia appears to be influenced by the gut microbiome (Erny et al., 2015). This data suggested that neuroinflammation can be regulated by targeting the gut microbiome using probiotic approaches.

In conclusion, the microbiota-gut-brain axis is becoming increasingly recognized in the context of NDs. However, further research is needed to define how the microbes communicate with the host and to understand the interactions between the diet, the microbiota and the intestinal epithelium.

The GI tract is a very complex organ with essential functions such as absorbing and processing of ingested food to fulfill the energy demands of development, reproduction, and survival (Woods et al., 1998). The GI tract is also a major source of neuronal and endocrine signals, which modulate food intake and the activity of other organs, such as the pancreas and the brain (Kabouridis and Pachnis, 2015; Abot et al., 2018). Additionally, the GI tract forms the largest and most important immune epithelial barrier against external dangers posed by ingested pathogens. Under ideal circumstances, the GI tract maintains a mutualistic and symbiotic relationship between a diverse and dynamic community of microorganisms and the host (Peterson and Artis, 2014; Abot et al., 2018). Here we will review the mechanisms by which the gut and gut-microbiota interactions might influence the brain activities in health and in the context of NDs while we refer the readers to a review by (Richards et al., 2021) for a detailed description of the mechanisms of gut-brain communications. The GI tract is also host to the ENS, a highly dense neuronal network that acts independently of the CNS (Fleming et al., 2020). The ENS presents a similar organization to the CNS brain regarding structure, function and pluri-chemical transmission. ENS neurons modulate a range of physiological activity, such as gut motility, intestinal permeability, intestinal immunity, enteric reflex, and enteroendocrine signaling (Carabotti et al., 2015), therefore playing a role in local inflammation (Fleming et al., 2020). The ENS communicates with the CNS via three primary neuronal types, intrinsic primary afferent neurons, motor neurons, and interneurons (Fleming et al., 2020). Communication between the neurons of the ENS and the PNS/CNS are carried out by compounds like acetylcholine, and reactive nitrogen species like Nitric oxide (Fleming et al., 2020). The primary method of communication between the ENS and CNS is facilitated by the bidirectional, 10th cranial nerve, the DMV in mammals and Drosophila (Breit et al., 2018).

The DMV is the main contributor of the parasympathetic nervous system. Exiting the medulla oblongata, the DMV delivers impulses from the brain to innervate peristalsis/digestion during a parasympathetic relaxed state. In return, the intestinal epithelium delivers signals to the brain communicating the state of the gut through the ENS via the DMV. The efferent vagal nerves regulate gastrointestinal secretory and motor function, and also the activity in the endocrine system of the gut, while the vagal afferents allow gut-brain information flow from the gut to the CNS. Vagal afferent signaling has been implicated modulating mood and affect, including distinct forms of anxiety and fear and are involved in the activation/regulation of the Hypothalamic Pituitary Adreno axis, which coordinates the adaptive responses to stressors of any kind (Tsigos and Chrousos, 2002; Howland, 2014). The DMV also has an immunomodulatory role as the cholinergic anti-inflammatory pathway (Goverse et al., 2016; Breit et al., 2018), making the DMV a major component of the neuroendocrine-immune axis. The appearance of pathogens in the gut activates resident innate immune cells to release cytokines. These in turn activate sensory fibers that ascend the DMV. Increasing efferent signals in the DMV suppresses peripheral cytokine release through macrophage nicotinic receptors and the cholinergic anti-inflammatory pathway. Thus, experimental activation of the cholinergic anti-inflammatory pathway by direct electrical stimulation of the efferent DMV was found to inhibit the synthesis of TNF-α in the liver, spleen, and heart, and lowers serum concentrations of TNF-α (Borovikova et al., 2000; Bernik et al., 2002). This regulation of the local and systemic inflammatory signals is essential to reduce metaflammation linked to neuroinflammation.

Gut communication to the brain does not solely rely on neurotransmission through the DMV but it can also be humoral through the secretion of metabolites processed from the diet by the microbiota or the gut epithelium. Metabolites can have neuroprotective trophic action (Liu, 2018) promoting neuron survival and plasticity in both the ENS and the CNS (Figure 1) (Liu et al., 2019). In a study conducted by (Vidal-Martínez et al., 2016) and colleagues, transgenic mice overexpressing mutant human α-synuclein developed PD-like enteric neuropathology. The animals accumulated α-synuclein aggregates in the intestinal epithelium and ENS nerves and exhibited gut motility problems that result in constipation. Treatments with FTY720, a sphingosine analog rescued the defects in the α-synuclein transgenic mice. FTY720 mediated the accumulation of the brain-derived neurotrophic factor that activates TrkB receptors in the gut reducing ENS synucleinopathy and improving gut motility (Vidal-Martinez et al., 2016).

The ENS and the intestinal epithelial cells also release peptides that can have local or distal effects. The battery of neuropeptides released in the intestines can affect the immune system or the CNS to control inflammatory responses, hormonal secretion, and metabolic activities. Gut peptides are mainly secreted from enteroendocrine cells in the GI tract. Neuropeptides and peptide hormones are released mainly by cell of the CNS and ENS executing in response to stress, metabolic, and commensals signaling. The Neuropeptide Y (NPY) is involved in controlling inflammation, pain, emotion, mood, cognition, stress responses, and metabolism. In the gut, NPY is expressed mainly by the ENS, where it regulates enteric inflammation. This is demonstrated by the observation that NPY-containing nerve fibers of the ENS are in close contact with immune cells in the mouse ileum lamina propria (Shibata et al., 2008). Specifically, NPY promotes colonic inflammation. A growing body of literature, both in humans and rodent models, suggests that brain NPY levels are altered in some neurodegenerative and neuroimmune diseases. NPY stimulates neuronal survival and neuroproliferation, attenuates neuroinflammation, and counteracts depressive symptoms and weight loss present in NDs. Ultimately, the extent to which these enteric peptides expression/repression in the gut is linked to NDs pathogenesis has not been established yet and the topic requires further investigation.

The intestinal immune cells constantly face a large number of antigens that are obtained from either food or the intestinal microbiota. Intestinal CD4⁺ T helper cells are highly involved in mucosal immunity. Enteric glial cells are essential for the integrity of the bowel. The presence of different bacterial species determines the pro- or anti-inflammatory state of CD4⁺ T-cells (Ivanov et al., 2009), which highlights the importance of the intestinal microbiota in the preservation of mucosal homeostasis. Failure to inhibit pro-inflammatory immune responses increases intestinal inflammation and may contribute to the development of immune-mediated inflammatory diseases. A loss of enteric glial cells also leads to severe inflammation of the intestines. The increased inflammation of the gut affects the permeability of the intestinal barrier activating resident immune cells and leading to the accumulation of inflammatory mediators and microbial-derived metabolites such as LPS in the circulatory system triggering systemic inflammation and metabolic dysfunction (Lancaster et al., 2018). Humoral pathways of gut inflammation spreading to the brain can damage the BBB. Multiple reports have shown that the BBB is damaged in PD patients (Pisani et al., 2012) and LPS (Gray and Woulfe, 2015) induced PD animal models exhibited disrupted BBB (Varatharaj and Galea, 2017). If the BBB is damaged, proinflammatory cytokines and immune cells such as T-cells (Engelhardt, 2017) and mast cells (Jones et al., 2019) from peripheral inflammation are able to enter the brain. These series of events are now considered the current model mechanism of disease of NDs that could be applied to familiar and sporadic forms of NDs and support a scenario that points at peripheral inflammation and alteration of the DMGB as main triggers of pathogenesis.

Metabolic alteration and inflammation of the intestine are two features observed in the most common NDs. How metabolic alteration occurs and what pathway is principally involved in the intestinal epithelium when the diseases seed, is not understood. Damage or functional alteration of essential organelles such as mitochondria and peroxisomes, have been found to accompany different pathologies of disease. Mitochondria and peroxisomes are highly metabolic cellular organelles, ubiquitously conserved across eukaryotes, from yeast to humans (Osellame et al., 2012; Smith and Aitchison, 2013; Baron et al., 2016). Both organelles participate in a broad range of conserved cellular-metabolic processes, most notably β-oxidation of medium-chain, long chain and very long-chain fatty acids, the anabolism and catabolism of complex signaling lipids such as phospholipids, and synthesis and turnover of reactive oxygen and reactive nitrogen species. These metabolites are essential for supporting cellular energetic processes, limiting redox stress, mitigating inflammation, and maintaining cell structure and signaling (Osellame et al., 2012; López-Armada et al., 2013; Smith and Aitchison, 2013; Di Cara, 2020). As mentioned previously, α-Synuclein inclusion bodies have been identified in the gut and enteric neurons of PD patients prior to the onset of disease, as well as in cases of inflammatory diseases that compromise the gut epithelial barrier (Hawkes et al., 2010; Challis et al., 2020; Derkinderen et al., 2021). α-Synuclein inclusions damage cellular mitochondrial functions, therefore, linking the diseases to metabolic and signaling dysfunction of mitochondria. However other studies reported that mitochondrial damage occurs before α-Synuclein inclusions appear.

Peroxisome dysfunction has not been directly linked to the onset of NDs. Cumulative evidence has reported that peroxisome dysfunction in the intestinal epithelium causes intestinal inflammation and dysplasia as reported in Drosophila gut-specific peroxisomal knock-down models by (Di Cara et al., 2018). In the same study, peroxisomes were also found to modulate the gut microbiota which, as discussed herein, changes in gut microbiota are tightly connected to inflammatory diseases including NDs. Clinical reports described that peroxisome numbers and metabolic activities are lower within neurons of the post-mortem brains of patients affected by NDs such as AD (Kou et al., 2011; Cipolla and Lodhi, 2017). In particular, post-mortem brains of AD patients have high VLCFAs (substrates of peroxisome β-oxidation) and in a parallel a decrease in ether-phospholipids (products of peroxisomal metabolism) (Kou et al., 2011).

Despite these correlations, more studies are needed to establish whether alteration of mitochondria and peroxisomes in the intestinal cells could be the culprit of the metabolic dysfunction that triggers the cascade to NDs.