Systemic inflammation as a central player in the initiation and development of Alzheimer’s disease

Abstract

Alzheimer’s disease (AD) is an age-related neurodegenerative disorder and the most common cause of dementia. While the amyloid cascade hypothesis has long dominated AD research, emerging evidence suggests that neuroinflammation may play a more central role in disease onset and progression. Increasingly, AD is recognized as a multifactorial disorder influenced by systemic inflammation and immune dysregulation, shifting focus toward peripheral immune mechanisms as potential contributors to neurodegeneration. This review explores the hypothesis that inflammaging, the age-related increase in pro-inflammatory mediators, combined with lifelong exposure to infections, injuries, metabolic changes, and chronic diseases, among others, may prime the immune system, amplifying neuroinflammation and influencing the progression and exacerbation of AD pathology. To this end, we examined how systemic immune disturbances, including chronic pain, post-operative cognitive dysfunction, viral and bacterial infections, gut microbiome dysregulation, and cardiovascular disease, may act as risk factors for AD. Overall, evidence suggests that modulating peripheral inflammation, accompanied by early diagnosis, could significantly reduce the risk of developing AD. Furthermore, we highlight key immune signaling pathways involved in both central and peripheral immune responses, such as the NLRP3 inflammasome and TREM2, which represent promising therapeutic targets for modulating inflammation while preserving protective immune functions. Strategies aimed at reducing systemic inflammation, identifying early biomarkers, and intervening before significant neurodegeneration occurs may provide novel approaches to delay or prevent AD onset. In conclusion, this review underscores the crucial role of systemic inflammation in AD pathogenesis and progression. By targeting peripheral immune dysfunction, we may advance our understanding of AD mechanisms and develop more effective therapeutic interventions to mitigate disease risk and progression.

Background

Alzheimer’s disease (AD) is a progressive neurodegenerative disease and the most common form of dementia, clinically characterized by progressive memory loss and impairments in cognitive and behavioral skills [1]. AD is classified into two main subtypes: sporadic or late-onset AD (LOAD) and early-onset AD (EOAD) [2]. The pathological features in both AD forms are generally represented by the buildup of amyloid-β (Aβ) plaques and neurofibrillary tangles (NFTs), leading to synaptic dysfunction and progressive neuronal loss. As a result, AD is associated with significant brain atrophy in both cortical and hippocampal regions, accompanied by enlargement of lateral ventricles [3].

In addition to these hallmark features, alterations in the immune system and the resulting neuroinflammation have been increasingly recognized as crucial factors in AD pathology [4]. However, despite these similarities, LOAD and EOAD differ in their onset, progression, and underlying mechanisms [5]. LOAD, the most common form, accounts for 90–95% of all AD cases and typically develops after age 65, with a strong association with the APOE ε4 allele [6, 7]. In contrast, EOAD represents 5–10% of AD cases, develops before age 65, and is often linked to autosomal dominant mutations in APP, PSEN1, or PSEN2 genes, leading to a more aggressive disease course, characterized by higher Aβ burden and immune system hyperactivation [5, 8].

Importantly, despite extensive research efforts, the cause of AD remains largely unknown. To date, the amyloid cascade hypothesis, which identifies Aβ accumulation as the primary contributor to AD pathology, has been the central causative explanation of AD. According to this model, the progressive buildup of Aβ in brain tissue leads to the eventual formation of Aβ plaques, which further spread into different brain regions, resulting in neurotransmitter loss, synaptic dysfunction, and cell death [9]. However, the amyloid cascade hypothesis has been challenged due to the lack of correlation between amyloid deposition and severity of cognitive deficits, as well as the relative inefficacy of the newly developed monoclonal antibodies that target Aβ [10, 11]. Moreover, it has been suggested that while Aβ is an important contributor to AD pathogenesis, there is limited evidence supporting its role in the disease’s etiology [12]. Nevertheless, Aβ remains a central feature of AD pathology. In particular, growing evidence suggests that Aβ oligomers, rather than Aβ plaques, are the more neurotoxic species and may be involved in the early stages of the disease [13, 14]. Indeed, Aβ oligomers have shown to disrupt synaptic function and activate glial cells, contributing to inflammation [15]. Thus, even though the amyloid cascade hypothesis may not fully explain all aspects of AD pathogenesis, the role of Aβ in in the disease progression remains critical.

In parallel, other markers of progressing AD pathology, including NFTs, synaptic loss, and microglial activation, closely correlate with the disease course, indicating that, besides Aβ aggregation, additional mechanisms contribute to the disease pathogenesis [16]. This has led to the emergence of alternative hypotheses, such as the neuroinflammation hypothesis, which proposes that innate immune responses may act both downstream and upstream of Aβ pathology [17]. Indeed, the neuroinflammation hypothesis proposes that immune activation may not only be a consequence of the pathological events present in the AD brain but could also contribute to the disease onset and progression [18, 19]. It is important to note that even though the involvement of neuroinflammation is well-supported, the possibility that peripheral inflammation may act as a trigger is still under investigation. Importantly, much of the evidence supporting this hypothesis has been shown in transgenic mouse models containing familial AD mutations, where Aβ overproduction represents the main hallmark and initiating factor. Therefore, determining whether immune activation precedes or follows Aβ accumulation remains challenging.

In this review, we investigate the current evidence regarding how changes in peripheral inflammation caused by different pro-inflammatory triggers, may influence the onset and progression of AD. To this end, we will specifically focus on LOAD, as it is more stronglylinked to age-related immune dysfunction and chronic low-grade inflammation. Unlike EOAD, which is often driven by genetic mutations, LOAD is heavily influenced by systemic inflammatory processes associated with aging, metabolic disorders, and cardiovascular diseases, among others, representing key risk factors that contribute to neuroinflammation and disease progression [20–22]. Additionally, the role of systemic inflammation in LOAD has been more extensively studied, providing a stronger foundation for understanding how peripheral immune responses interact with neurodegenerative mechanisms. By focusing on LOAD, we aim to highlight the critical role of inflammatory pathways in AD onset and progression, as well as potential therapeutic strategies targeting systemic inflammation.

In the following sections, we will explore the neuroinflammation hypothesis along with the proposed mechanisms and sources that contribute to its onset and its role as a triggering factor for AD.

Neuroinflammation and glial cells

Neuroinflammation is initiated by the main immune cells of the brain, namely astrocytes and microglia. Astrocytes are responsible for synapse and neurotransmitter maintenance, as well as playing an essential role in forming and maintaining the blood-brain-barrier (BBB) [23, 24]. On the other hand, microglia are the first line of defense in the CNS, and their main function consists of clearing and degrading harmful agents present in the brain via phagocytosis [25]. In the context of AD, microglia and astrocytes are crucial for the efficient removal of Aβ from the brain [26–28]. Indeed, one of the main mechanisms behind the increased accumulation of Aβ during AD is the reduced efficiency of its clearance from the brain, especially by microglia [29]. Moreover, microglia are critical for the maintenance of homeostasis and tissue repair. As mentioned, microglia also play an essential role in regulating neuroinflammation via the production of pro-inflammatory cytokines [30].

However, neuroinflammation is a double-edged sword: while microglia and astrocyte activation are initially protective and necessary for the activation of pro-survival mechanisms, as well as for uptake and degradation of Aβ, chronically activated microglia can lead to numerous adverse effects that contribute to AD pathology [31]. Firstly, an excessive release of pro-inflammatory cytokines by microglia has been linked to an increase in amyloid precursor protein (APP) synthesis, as well as in BACE-1 transcription, both resulting in an enhancement of Aβ production [32, 33]. Chronically activated microglia can also release reactive oxygen species (ROS) and free radicals, contributing to neuronal death [34]. Moreover, the release of pro-inflammatory cytokines can affect neurotransmitter levels and synapse maintenance, resulting in synaptic dysfunction [35]. Additionally, the presence of Aβ oligomers at early stages of the disease can also trigger microglial activation and pro-inflammatory responses [13, 15], which in turn promotes further Aβ production and aggregation. The previously mentioned molecular mechanisms gradually result in Aβ accumulation and a vicious cycle of chronic inflammation and cell death.

Altogether, maintaining optimal microglial function appears crucial for preventing Aβ accumulation and breaking the cycle of chronic inflammation in AD. Therefore, a key aspect of understanding the etiology of AD pathology may lie in uncovering the reasons and mechanisms behind the progressive chronic activation of microglia and their subsequent reduced ability to clear Aβ. In contrast to the amyloid cascade hypothesis, which suggests that microglial dysfunction is a consequence of Aβ deposits, the neuroinflammation hypothesis argues that impaired microglial function may precede and contribute to Aβ accumulation [36].

Peripheral inflammation and inflammaging

In this model, it is proposed that as we age, the immune system gradually transitions into a more pro-inflammatory state, also known as ‘’inflammaging’’, characterized by an increase in proinflammatory mediators and low-level systemic inflammation [19, 37]. It is suggested that this increase in peripheral inflammation might eventually lead to neuroinflammation and microglial dysfunction. In fact, it has been extensively shown that the systemic immune system can affect the CNS and that inflammatory mediators in the blood, such as cytokines, can enter the brain [38–40]. For instance, increased peripheral inflammation caused by different mediators, such as infections, changes in microbiota composition, or injuries, have been proposed as major contributors to neuroinflammation and microglial overactivation [38, 41]. This immune activation may act synergistically with early Aβ pathology, exacerbating microglial responses [42]. As a result, Aβ and other cellular debris will accumulate, which in turn drives further microglial activation. This excessive activation will result in the release of more pro-inflammatory factors, ultimately contributing to the production of Aβ plaques and inducing cell death [36].

Thus, while neuroinflammation driven by CNS immune cells has long been recognized as a key factor in AD pathogenesis [4, 18], there is a growing interest in the role of the peripheral immune system as a potential initiator of microglial dysfunction, neuroinflammation, and AD pathology. One proposed mechanism linking inflammaging, peripheral inflammation, and neuroinflammation in AD is the activation of the NOD-like receptor protein 3 (NLRP3) inflammasome. The NLRP3 inflammasome is a signaling complex present in immune cells which becomes activated after injury or cell stress [43]. Upon activation, it triggers the cleavage of pro-caspase-1 into caspase-1. This, in turn, leads to the release of pro-inflammatory cytokines like IL-1β and IL-18, promoting an inflammatory response [44, 45]. Inflammaging and peripheral inflammation, arising from various factors (discussed in the following sections), can result in dysregulated activity of the NLRP3 inflammasome in both the periphery and the brain [37]. This activation triggers inflammatory signaling cascades and creates an inflammatory environment in the brain, leading to chronic microglial activation. Persistent activation of the NLRP3 inflammasome in microglia causes microglial dysfunction, reducing their ability to effectively clear Aβ plaques, ultimately leading to their accumulation [37]. Moreover, it has been demonstrated that NLRP3 inflammasome activation is essential for the development of Aβ plaques and tau pathology in AD mouse models [46, 47], making the NLRP3 inflammasome an attractive therapeutic target against neuroinflammation and AD.

Hypothesis

Taken together, it seems that AD may be more influenced by systemic triggers than it was initially thought. We believe that inflammaging (natural aging-related increase in pro-inflammatory mediators) along with other changes in the peripheral immune system caused by different pro-inflammatory challenges during our lifespan, such as metabolic changes, disorders like chronic pain or cardiovascular disease, injuries or infections may lead to a primed immune system that contributes to the progression and exacerbation of AD pathology. However, this hypothesis should not be interpreted as independent from Aβ-centered mechanisms. Instead, we propose that systemic inflammation may act as a primary modulator of Aβ pathology, particularly through its influence on microglial overactivation, Aβ clearance and NFT aggregation. Thus, we hypothesize that modulation of peripheral inflammation, early diagnosis, and treatment of the conditions that contribute to systemic inflammation might decrease the risk of developing AD.

In the following section, we will delve into the changes in peripheral immune cells occurring during AD progression, as well as potential sources of systemic inflammation that may play a role in the priming of the immune system and the potential consequent development of AD.

Overview

As we discussed in previous sections, peripheral or systemic immunity may play an important role in the development and progression of AD. Direct evidence of elevated peripheral inflammation in AD patients has been demonstrated in multiple meta-analyses [48, 49], which show elevated levels of pro-inflammatory cytokines and other inflammatory markers compared to healthy controls. These include interleukin-12 (IL-12), IL-1β, IL-6, tumor necrosis factor-alpha (TNF-α), TNF receptors 1 and 2, as well as inflammatory molecules such as C-reactive protein and C-X-C motif chemokine-10 [48, 49]. In particular, the levels of TNF- α and IL-6 have been associated with cognitive decline and disease severity [50, 51].

T cells

CD4 and CD8 T cells

Distinct changes in peripheral immune cells have been related to the pathophysiology of AD. For instance, adaptive immune T cells have been largely studied in the context of AD. Overall results agree that both CD8 + and CD4 + effector T cells, mainly involved in coordinating immune responses against pathogens, are elevated in AD patients [52, 53]. In fact, a recent study performed in an AD mouse model reported a potential role of CD4 + effector T cells in contributing to AD development [54]. Similarly, research performed on post-mortem human AD brains, in vivo AD mouse models, and in vitro systems demonstrated that CD8 + T cells exacerbate AD pathology [55–57].

Regulatory T cells (Tregs)

On the other hand, regulatory T cells (Tregs) are a subset of CD4 + T cells essential for regulating immune responses, including the maintenance of immune tolerance and prevention of excessive immune reactions harmful to the host [58]. Tregs have been suggested to play a protective role in different neurodegenerative diseases, such as amyotrophic lateral sclerosis (ALS) and Parkinson’s diseases (PD). Regarding AD, previous studies have reported decreased Tregs in AD patients, which may contribute to an exacerbated immune response [53, 59]. In animal models, it was shown that depleting Tregs from the APP/PS1 mouse model of AD resulted in an acceleration of Aβ-driven cognitive impairments [60]. Conversely, amplification of Tregs in the same mouse model led to an improvement in cognitive functions, as well as an increase in plaque-associated microglia [60]. Another study, by Yeapuri et al., also demonstrated the therapeutic benefit of Tregs on cognitive functions and neuropathology in the APP/PS1 mouse model of AD [61]. These studies are supported by our own work, where we showed that immune modulation via systemic administration of a tumor necrosis factor receptor 2 (TNFR2) agonist rescued cognitive functions, decreased amyloid plaque deposition and increased plaque-associated microglia in the J20 mouse model of AD [62, 63]. Furthermore, treatment with the TNFR2 agonist promoted the expansion of Tregs, leading to an increased Treg infiltration in the brains of the mice [63, 64]. This increased Treg presence may have potentially shifted the microglial phenotype towards a more restorative, rather than inflammatory, function. Importantly, our study showed that this systemic treatment showed an equal or even superior protective effect than central administration, highlighting again the importance of the peripheral immune system in AD.

Besides specific neurodegenerative diseases, Tregs have also been proven protective and essential for recovery in various conditions and disorders, such as neuropathic pain, arthritis, ischemic stroke, and traumatic brain injury [65–68]. Importantly, a recent study has identified age-associated subsets of Tregs with senescent features [69]. Altogether, these findings emphasize the pivotal role of peripheral immune modulation, particularly through Tregs expansion, in neuroprotection and cognitive function improvement. Given the broad protective effects of Tregs across neurodegenerative and inflammatory conditions, therapeutic strategies should not only aim to increase Treg numbers but also focus on enhancing their functional quality to ensure effective immunosuppression and anti-inflammatory activity.

Theories

After reviewing these changes in peripheral immune cells observed in AD, one important aspect would be to elucidate the origin of the observed dysregulated immune system and chronic peripheral inflammation in AD. Even though the exact etiology remains unknown, there are several factors that contribute to the increase in peripheral inflammation. First, the gut microbiome has been extensively studied, and it has recently been proposed as one of the driving mechanisms in AD (reviewed in Sect. 2.3 of this review). Briefly, disturbances and changes in the composition of the gut microbiome can increase intestinal permeability, leading to the translocation of lipopolysaccharide (LPS) into the peripheral circulation [70]. In turn, LPS can have extensive effects on different subsets of immune cells. For instance, different studies have reported that translocated LPS can lead to the expansion and differentiation of CD4 T helper cells, CD8 T cells, as well as B cells [71–73]. Furthermore, changes in the gut microbiome can also influence the differentiation trajectory of naïve CD4 T cells, potentially causing an imbalance in Tregs [74]. All these abnormalities in immune cell clusters may, as previously mentioned, impact peripheral inflammation and AD development.

Another point of interest has been dedicated to epigenetic dysregulation in some cell clusters. Specifically, epigenetic changes in T cells and macrophages, including increased expression of microRNA-155 (miR-155) and altered DNA methylation, have been linked to heightened peripheral inflammation in AD patients [75, 76]. It has been proposed that one of the causes for this epigenetic dysregulation might also be linked to LPS translocation, which has been shown to increase miR-155 in immune cells, supporting a shift towards a pro-inflammatory phenotype and exacerbation of AD pathology [77]. Finally, triggering receptor expressed in myeloid cells 2 (TREM2) is a receptor that, when activated, plays a key role in the induction and regulation of the immune response [78]. In the brain, TREM2 is mainly expressed by microglia and regulates its functions, such as microglial activation and phagocytosis [79]. Mounting evidence has linked mutations in TREM2 to a higher risk of developing AD, mainly due to the decreased protective role of microglia in phagocytizing Aβ plaques [78, 80–82]. Furthermore, TREM2 has been suggested as a contributor to peripheral inflammation. It has been reported that in certain peripheral immune cells, such as dendritic cells, TREM2 is upregulated, which leads to the differentiation of Th2 and Th17 cells [83]. Moreover, elevated expression of TREM2 mRNA in the periphery was negatively correlated to both Mini-Mental State Examination and Montreal Cognitive Assessment scores in AD patients [84], highlighting that peripheral immune alterations may significantly influence cognitive deficits in AD patients.

Finally, an additional potential contributor to AD exacerbation is the changes in the BBB that occur due to peripheral inflammation. For instance, it has been reported that elevated levels of pro-inflammatory cytokines can lead to changes in tight junctions and damage to endothelial cells [85–87], causing the disruption of the BBB. This can, in turn, lead to the infiltration of peripheral immune cells, which may also directly contribute to the activation of microglia and astrocytes [88], further amplifying neuroinflammation.

Collectively, it has become increasingly evident that systemic inflammation and changes in systemic immune responses can have long-lasting effects on the brain. Inflammaging and priming of the immune system due to pro-inflammatory stimuli occurring during an individual’s life may be key in understanding the onset and development of AD. The described changes leading to peripheral inflammation can also be influenced by different disorders and diseases, which have also been linked to the exacerbation of AD pathology.

Disorders associated with systemic immunity and their link to AD

As indicated in the previous sections, it is increasingly evident that systemic inflammation may significantly influence both the development and advancement of AD. When the immune system responds to disease, infection, or injury, the changes in the immune system at the systemic level may contribute to AD. Systemic inflammation may not only serve as a driving force for AD, but is also implicated in other disorders, such as cardiovascular disorders, obesity, gut inflammation, chronic pain, and viral and bacterial infections. These disorders can collectively contribute to inflammaging [89]. In this section, we delve into the interplay between systemic immune disorders and their potential impact on the pathology and progression of AD, proposing that chronic immune activation might serve both as an initiator and accelerator of neuroinflammation in AD. By examining these conditions, we aim to reinforce the idea that AD should be considered a systemic immune disorder rather than a purely brain-specific pathology.

Systemic inflammatory insults

Several systemic events, such as chronic pain, viral and bacterial infections, surgeries, and trauma-related insults, can trigger neuroinflammation, which is a significant contributor to the development of AD (Fig. 1). These conditions not only contribute to transient immune activation but may also lead to long-term changes in immune function that predispose individuals to neurodegenerative diseases like AD.

Fig. 1.

Various systemic diseases and conditions —including oral infections, obesity, type 2 diabetes, cardiovascular diseases, gut inflammation, chronic pain, infections, and surgery— contribute to chronic inflammation, which may in turn exacerbate the pathology and progression of Alzheimer’s disease (AD). Prolonged immune activation due to these conditions can lead to persistent low-grade inflammation, which primes the immune system and increases the susceptibility to neuroinflammation in AD. These peripheral immune disturbances can compromise the blood-brain barrier (BBB), allowing inflammatory mediators to enter the brain and may contribute to microglial activation, amyloid-β plaque accumulation, neurofibrillary tangle formation, neuronal loss, and neuroinflammation. Created with BioRender

Chronic pain – Alzheimer’s disease

Chronic pain, defined as pain lasting more than 3 months, is most commonly experienced as back, neck, or joint pain in older adults. It is associated with both cognitive and emotional decline [90–92]. Chronic pain is commonly seen in the elderly, with a prevalence rate of 38.5%. Interestingly, this rate increases up to 45.8% in AD patients, suggesting a correlation between chronic pain and AD [93, 94].

Multiple studies link chronic pain to increased dementia risk in the future [95–97]. For instance, a 10-year cohort study found that fibromyalgia patients, a chronic pain syndrome, had a significantly higher dementia risk [97]. Other longitudinal studies showed that chronic pain in multiple sites (such as back, knee, head) elevates the risk of dementia and AD [98, 99] and accelerates cognitive decline [99]. Since a single chronic pain site increases the risk of getting dementia by 8% [98], appropriately managing chronic pain could potentially contribute to lowering the prevalence of AD.

Although the link between chronic pain and AD is evident, the underlying mechanism remains unclear. Evidence increasingly suggests common factors and pathologies, including aging, reduced gray matter, and both systemic and neuroinflammation [99–104]. For instance, systemic inflammation in chronic pain is evidenced by elevated pro-inflammatory cytokine blood levels [105–108], which might predispose individuals to its development and pathogenesis [109]. Moreover, systemic inflammation can disturb pain regulation by decreasing endogenous pain inhibition and increasing pain [110–112], as shown in a clinical study [112].

Neuroinflammation is another common factor between chronic pain and AD. It can be triggered by systemic inflammation [113], and plays a role in inducing and maintaining chronic pain via microglial activation [92, 114–116]. Neuroinflammation induces chronic pain by causing central sensitization through the release of pro-inflammatory cytokines such as TNF-α, IL-6, and IL-1β [115, 117, 118]. Activated microglia are reported in brain regions like the hippocampus, prefrontal cortex, and amygdala in chronic pain [119], and they have a pivotal role in the transition from acute to chronic pain [120]. The pro-inflammatory state of microglia and elevated release of pro-inflammatory cytokines seen in chronic pain might potentially contribute to rendering chronic pain patients more susceptible to developing dementia, including AD.

Anti-inflammatory therapies targeting microglial activation

Given the shared role of systemic inflammation in both AD and chronic pain, anti-inflammatory therapies may help in managing and mitigating the progression of both diseases. One approach to reduce inflammation is to inhibit pathological activation of microglia [121].

For example, the antibiotic minocycline reduces microglial activation and exhibits neuroprotective, anti-amyloidogenic, anti-inflammatory, and antioxidant properties [122–125], with studies suggesting that it may alleviate the symptoms of both AD and chronic pain [126–129]. In animal models, it reduced Aβ plaque deposition, hyperphosphorylation of tau, improved memory deficits, and reduced neuronal cell death [126, 127, 130, 131]. However, one clinical trial in AD reported no significant cognitive improvement after two years of treatment and noted side effects at higher doses [132].

Since microglial activation also contributes to central sensitization in chronic pain [133] and the transition from acute to chronic pain [120], reducing microglial activation may also be therapeutic for chronic pain. Supporting this, upregulation of TREM2 and its adaptor protein DNAX-activating protein of 12 (DAP12) in spinal cord microglia has been shown to induce neuropathic pain by enhancing proinflammatory cytokine release from microglia [134]. In this manner, pharmacological inhibition of TREM2 reduced fracture-associated chronic pain [135].

Animal studies suggest minocycline`s anti-nociceptive effects in chronic pain conditions such as diabetic and chemotherapy-induced peripheral neuropathy, and visceral pain [136–140]. Moreover, in a clinical study, patients with breast cancer receiving 200 mg of minocycline daily throughout the 12 weeks of chemotherapy reported less pain and less use of opioid pain medication than the placebo group [141]. Additionally, a current clinical trial is testing the effect of a 14-day treatment with minocycline on chronic low back pain [122].

Similarly, common pain medications, including paracetamol and nonsteroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen and aspirin, have been studied for their potential impact on AD [142]. Preclinical studies suggest that ibuprofen reduces microglial activation, alleviates AD pathology, and improves cognitive function in different AD mouse models [143–146]. Some clinical studies also indicate that long-term NSAID use may lower the risk of developing AD [147–150] and prevent cognitive decline, especially when initiated in midlife (before 65 years of age) [151]. However, NSAIDs may accelerate AD progression in individuals with advanced pathology [152], suggesting that their protective effects depend on both the timing and duration of treatment [153]. Notably, one study found that aspirin and paracetamol were linked to a higher dementia risk in people with chronic pain, whereas ibuprofen did not show a significant association [154]. Moreover, another study showed that regular paracetamol use, but not ibuprofen, significantly increased dementia risk in the elderly, regardless of genetic predisposition [155]. These findings highlight the importance of carefully selecting pain management strategies, as their effects on AD and cognitive health may be protective or harmful depending on the context.

In conclusion, minocycline’s and NSAIDs’ potential as a treatment for AD and chronic pain remains subject to ongoing research, with certain studies suggesting their effectiveness while others raise questions about their clinical application, highlighting the need for further investigation and optimization. Some of the above-mentioned studies suggest that targeting systemic inflammation through the reduction of microglial activation may be beneficial in addressing AD and chronic pain [121, 127, 129, 138, 139, 141].

However, as elaborated in prior sections, microglia are essential for the uptake and clearance of Aβ and for the removal of debris from dying or damaged cells, thus playing a critical role in maintaining brain homeostasis [156, 157]. For example, TREM2-deficient microglia or microglial depletion have been shown to exacerbate Aβ and tau pathology in mouse models of AD [158–160]. Indeed, microglia exhibit a dual role in AD, being either neuroprotective or neurotoxic depending on the disease stage and pathological insults. Consequently, interventions to reduce microglial activation, such as with minocycline, should be carefully timed to avoid exacerbating pathology.

Anti-inflammatory therapies targeting TNF-α

Modulating proinflammatory cytokines like TNF-α offers another strategy to reduce systemic inflammation in AD and chronic pain.

In addition to its involvement in AD as discussed in previous sections, TNF-α plays a role in both central and peripheral mechanisms underlying chronic pain [161–163]. The level of TNF-α is elevated in the periphery in chronic pain [117, 161, 164], highlighting its therapeutic potential. A study showed that both a TNF-α blocker and minocycline reduced neuropathic pain-like behavior and microglial activation after spinal cord injury [165].

However, as discussed previously, blocking TNF-α can have detrimental effects due to the opposing functions of the TNF receptor 1 (TNFR1) and TNF-α receptor 2 (TNFR2) [166]. Therefore, instead of inhibiting TNFα, selectively modulating TNFR1 or TNFR2 might be a more effective approach [63, 167, 168]. For instance, Fischer et al. found that i.p. administration of a TNFR2 agonist promoted long-term recovery in neuropathic pain in both sexes [68], while TNFR1 inhibition with an antagonist was therapeutic in only male mice [167].

Given the promising results of selective TNFR modulation in treating AD and chronic pain, further research should explore the use of a TNFR2 agonist or TNFR1 antagonist in an AD mouse model that recapitulates clinical chronic pain conditions.

Overall, chronic pain and AD may be linked through shared inflammatory mechanisms. While certain therapies show promise, further research should carefully consider the dual roles of microglia and inflammation, ensuring that interventions are precisely timed and targeted to avoid unintended detrimental effects.

Post-operative cognitive dysfunction – Alzheimer’s disease

Tissue injuries such as trauma and surgeries may also induce systemic inflammation [169–171].

While acute postoperative pain is expected, over 10% of surgical patients experience chronic pain 12 months after surgery, with procedures such as orthopedic and spine surgeries, knee arthroscopy, and total knee or hip arthroplasty having a particularly high incidence [172]. This prolonged pain may be linked to systemic inflammation, which plays a role in turning acute pain into chronic pain post-surgery [173]. Systemic inflammation is also believed to contribute to post-operative cognitive impairments following surgery. Post-operative cognitive impairments can manifest in both short-term disturbances, such as postoperative delirium, or long-term disturbances as post-operative cognitive dysfunction (POCD), on which we will focus in the next sections.

Interestingly, the link between surgery and cognitive decline extends beyond POCD. For example, people over the age of 55 who underwent coronary artery bypass graft surgery showed a significantly increased risk of developing dementia [174, 175], with elderly and individuals with existing cognitive impairment being particularly being vulnerable [175].

A preclinical study found that abdominal surgery in aged wild-type mice caused spatial memory deficits and increased hippocampal Aβ levels, while younger mice were unaffected following the same surgery [176]. Moreover, 9-month-old AD mice displayed increased cognitive impairment, supported by reductions in freezing time in contextual fear conditioning and increased Aβ accumulation in the hippocampus after the peripheral surgery compared to Sham AD mice at the same age. These results suggest that the combination of surgery with factors such as aging or carrying an AD gene mutation can lead to POCD [176].

Although mechanisms remain unclear, surgery-induced inflammation is believed to play a role in this process [176]. The initiation of an inflammatory response due to surgical trauma can result in neuroinflammation and synaptic impairment, and impaired neurogenesis [177, 178], which predisposes patients to POCD and dementia [175] as well as accelerates the disease progression of AD [179]. Furthermore, elevated proinflammatory cytokine levels after surgery, such as TNF-α and IL-1β, might also contribute to POCD [180].

A recent clinical study observed that individuals undergoing major surgical procedures showed an elevation of plasma pTau181 levels [181], which is a key AD biomarker [181–183]. Importantly, the study highlighted that cardiopulmonary bypass surgery exhibited a heightened risk for deviant biomarker levels and postoperative neurocognitive disorder when compared to noncardiac surgery [181].

Since neuroinflammation and systemic inflammation are shared hallmarks of both AD and POCD, mitigating inflammation may help prevent cognitive decline [184, 185]. Various factors can influence neuroinflammation, including gut microbiota, which will be discussed in the following sections.

Preclinical studies showed that microbiome-targeted therapies, such as supplementation with probiotics and prebiotics, decrease inflammation and improve cognition in POCD [186–188], as well as AD models [189–191]. For example, rats that were administered the prebiotic Bimuno for three weeks before abdominal surgery displayed improved learning and memory performance in the hippocampus-dependent novel object recognition test in early post-surgery compared to the control group [188]. Moreover, the surgery resulted in a significant increase in microglial activation in the hippocampus, as indicated by Iba-1 (Ionized calcium-binding adaptor molecule 1) expression. However, prebiotic administration mitigated the surgery-induced microglial overactivation [188].

In addition to microbiome-targeted approaches for addressing POCD, reducing microglial activation with the antibiotic minocycline also suppressed postoperative neuroinflammation and mitigated the hippocampal-dependent memory impairment caused by a non-cardiac surgery as described in an animal study [192].

Clinically, older patients receiving probiotics before a non-cardiac surgery had lower POCD incidence, likely due to reduced inflammation and stress responses [193]. However, a recent clinical trial found no significant benefit of minocycline on POCD outcomes at 1 week and 3 months following a knee surgery [194].

In conclusion, while surgery may pose a risk, particularly for the elderly and those with pre-existing cognitive deficits, a careful approach to managing excessive inflammation after surgery may offer therapeutic benefits. However, it is crucial to balance the timing and extent of anti-inflammatory interventions, as microglial activation also plays a protective role. Strategies that modulate inflammation, especially before and after surgery, without completely inhibiting beneficial immune responses, could be key in developing effective treatments for POCD.

Infections – Alzheimer’s disease

Emerging research suggests a novel perspective on the etiology of AD by linking it to infections. One possible explanation of how the viral and bacterial agents may elevate the risk of AD is by stimulating an innate immune response and initiating inflammation [195]. Upon the initiation of an inflammatory cascade following an infection, amyloid production and deposition are upsurged, which then induce further inflammatory responses, ultimately leading to the development of AD pathology [195].

Various bacterial species and several viruses have been associated with AD pathogenesis, potentially through their ability to induce neuroinflammation and Aβ production. In this section, we will explore the hypothesis that infections initiated by different bacterial and viral pathogens may serve as persistent immune stressors that prime microglia into a pro-inflammatory state, thereby contributing to the neuroinflammatory cycle in AD.

Viral Infections – Alzheimer’s disease

Multiple studies have linked viral infections to neurodegenerative disorders, including AD. A recent study revealed that exposure to 45 different viral agents increased the risk of developing neurodegenerative diseases, with viral encephalitis showing the most significant association, resulting in a 30-fold increased risk of developing AD [196]. Several viruses, including different types of human herpesvirus (HHV), human immunodeficiency virus (HIV), and hepatitis C virus (HCV), have been associated with AD development, possibly by promoting neuroinflammatory processes.

Herpes simplex virus type 1 (HSV1)

The herpesviruses are highly prevalent, infecting over 90% of adults at some point in their lives [197]. Herpes simplex virus type 1 (HSV1) is a common virus that might cause cold sores and persists in a latent state in the nervous system throughout life, with periodic reactivation and recurring infections in the event of immunity impairment [197, 198].

Cumulative HSV1 exposure diminishes cognitive performance regardless of other health status, as shown in a clinical study [199]. In addition to that, HSV1 exposure is linked to AD and dementia as a susceptibility factor in several studies. A 16-year study of over 30,000 individuals found that HSV infection resulted in an 2.56-fold hazard ratio for the development of dementia [200]. Several studies link HSV1 in the brain to AD, especially in apolipoprotein E4 (APOE4) carriers [201–204], with one study showing a 12-fold increase in AD risk when both HSV1 and APOE4 were present compared with individuals having just one or neither of these factors [204]. Moreover, a large cohort study revealed an increased episodic memory decline, which is an early AD-related symptom, in participants aged over 65 years old who were carriers of both HSV and the APOE4 allele, in contrast to APOE4 carriers without HSV1 infection. Furthermore, a mouse study demonstrated that HSV1 DNA levels were 13-fold lower in the brain of APOE knock-out mice compared to wild-type mice that were injected with the same dosage of HSV1 [205], underscoring the combined impact of HSV1 and APOE4 on AD risk [201–205].

Itzhaki and colleagues suggest that HSV1 can migrate into the brain from the periphery via retrograde axonal transport, olfactory nerve route, or bloodstream [206–208] when there is a decline in the immune system function, particularly with aging [209, 210]. Once in the brain, HSV1 can reactivate, and it can contribute to neuroinflammation [211, 212] and Aβ aggregation [213], increasing AD and cognitive decline risk in those with frequent peripheral infections [214].

HSV1 can directly stimulate Aβ production in vitro in neuronal and glial cells [213, 215], but also in by upregulating the Aβ-producing enzymes, such as β-secretase 1 and γ-secretase [213]. While an increase in APP amyloidogenic processing may provide initial protection against HSV1, recurrent HSV1 reactivation over a lifetime can lead to Aβ accumulation in the brain [216]. Furthermore, as the Aβ clearance by microglia is compromised by inflammaging [217], elderly people become more vulnerable to Aβ production following HSV1 infection, potentially exacerbating AD pathology [216]. Recent evidence also showed that HSV1 infection impaired microglial antiviral defense by downregulating TREM2 expression, which disrupted interferon signaling and phagocytosis [218]. TREM2 deficiency increased susceptibility to HSV1 in both cell culture and mice, highlighting its crucial role in microglial antiviral response [218].

Notably, HSV1 DNA was found in 90% of Aβ plaques in AD brains, with 72% of the viral DNA localized to plaques [209].

While the role of herpesviruses in AD etiology remains unclear, Aβ aggregation and neuroinflammation, which are the key hallmarks of AD pathology, might be the antimicrobial response to HSV1 exposure in the brain [202]. Some evidence supporting this hypothesis includes the antiviral properties exhibited by Aβ oligomers against HSV1 infection in neuronal cells [216, 219] and the 5xFAD mouse model [219].

HSV1 may also trigger tau hyperphosphorylation. Human neuroblastoma cells that were treated with HSV1 had elevated levels of AD-specific tau phosphorylation [220]. Moreover, hyperphosphorylated tau (p-tau) expression showed a transient increase in HSV1-infected primary murine hippocampal neuronal cultures on days 1 and 3 following injection of HSV-1 [221]. However, the levels of p-tau diminished by day 7, whereas Aβ accumulation continued to progressively increase over the course of 7 days, with barely detectable levels of Aβ on day 1 compared to the highest levels observed on day 7 after HSV1 infection, suggesting p-tau may be an acute danger response, while Aβ provides long-term defense against a continued stressor such as an infection [221].

Human immunodeficiency virus (HIV)

Besides HSV1, HIV, and HCV infection might be risk factors for AD, AD-type pathology, and dementia [222–225]. A recent population study reported a higher prevalence of AD and related dementias among HIV-positive individuals, particularly older women, compared to HIV-negative individuals [223]. HIV regulatory proteins such as trans-activator of transcription (Tat), envelope glycoprotein (Gp120), viral protein R (Vpr), and negative factor (Nef) can directly activate neuroinflammatory pathways [226] and induce oxidative stress through mitochondrial dysfunction, which contributes to neuronal damage [227]. Furthermore, the HIV-1 Gag protein has been shown to interact with APP, promoting its processing into amyloid-beta, thereby linking HIV infection to AD-like pathology [228]. Supporting this, Aβ has been observed in the brains of HIV patients [222, 229].

Moreover, inflammasomes, particularly NLRP3, have been identified as central mediators of inflammation during HIV-1 infection, with recent reviews highlighting its potential as a therapeutic target for HIV [230, 231].

Hepatitis C virus (HCV)

A study on bipolar disorder (BD) patients found that those with AD had a 2.3-fold higher prevalence of prior HCV infection compared to BD patients without AD [224]. Another population-based study identified an elevated risk of dementia among HCV-infected patients within an 11-year time frame compared to the non-HCV cohort and identified HCV infection as a risk factor for dementia [225]. Cognitive impairment in HCV patients may be associated with systemic inflammation caused by the virus [232].

Antiviral therapeutic strategies targeting AD pathology

These interactions support the significant role of viral infections in AD pathology and suggest that antiviral agents or vaccination might be considered viable preventive therapeutic strategies in AD. For instance, HSV patients treated with anti-herpetic medications had a significantly reduced risk of developing AD or dementia over 10 years compared to untreated individuals [200]. This aligns with another cohort study involving over 200,000 individuals, which revealed that herpes zoster increased dementia risk, while antiviral treatment (acyclovir, famciclovir, or valaciclovir) significantly decreased that risk [233].

This reduced risk of AD and dementia following an anti-herpetic treatment may be attributed to the potential of some anti-viral compounds to directly mitigate AD-related pathology. Supporting this hypothesis, different antiviral agents used for HSV1 treatment, such as acyclovir, penciclovir, and foscarnet, have been shown to significantly reduce Aβ and p-tau accumulation in HSV1-infected cells [234]. Given this growing body of research, some authors suggest that early-life vaccination against HSV-1 may be protective against AD [235].

In summary, the above-mentioned findings provide evidence that viral infections, particularly HSV1, could serve as risk factors for the development of AD and dementia by contributing to AD pathology. Nevertheless, recognizing and treating viral infections promptly may mitigate the risk of AD development.

Bacterial Infections – Alzheimer’s disease

Various bacterial species have been associated with AD pathogenesis, potentially through their ability to induce neuroinflammation and Aβ production.

Borrelia burgdorferi

Over 30 years ago, Borrelia burgdorferi, a spirochete causing Lyme disease, was detected in the neocortex of a postmortem AD brain [236], underscoring additional research to explore the potential link between Borrelia and possibly other bacteria to dementia and AD. Like herpetic viruses, spirochetes, including B. burgdorferi and Tropenoma pallidum, can enter a latent state with persistent infection. Importantly, chronic exposure to spirochetes can lead to chronic inflammation and amyloid deposition [198, 237]. To test the hypothesis, Miklossy et al. infected rat neuronal cultures with Borrelia, finding increased Aβ and p-tau levels, particularly after prolonged exposure (8 vs. 2 weeks) [238]. Moreover, they suggested that B. burgdorferi and T. pallidum may contribute to amyloid deposition, cortical atrophy, and dementia [239, 240].

C. pneumoniae

C. pneumoniae is an intracellular respiratory pathogen that might cause pneumonia, bronchitis, and various upper respiratory tract illnesses [241] and it is also associated with AD.

C. pneumoniae can induce amyloid deposits in mice. The analysis of non-transgenic mice brains that were treated with C. pneumoniae demonstrated the presence of amyloid deposits, which were detected 1 month after inoculation. As the infection advanced, both the size and quantity of these deposits increased, supporting its potential to drive or accelerate AD-like pathology [242]. Moreover, C. pneumoniae can progressively increase the protein levels of BACE1 and PSEN1, which can result in further neuroinflammation and increased Aβ production [243].

When transitioning to human studies, clinical evidence further supports the relevance of C. Pneumoniae in AD. A recent cohort study found a 1.6-fold increased risk of AD in patients with C. Pneumoniae within 8 years of tracking, which was mitigated by timely antibiotic treatment [244]. In a different study by Balin et al., C. pneumoniae was detected in 17 out of 19 AD brains in regions such as the hippocampus and temporal cortex by using PCR analysis. In contrast, no evidence of C. pneumoniae was found in control patients [245].

Similarly, 20 of 27 AD brains exhibited C. pneumoniae DNA, whereas only 3/27 control brains were positive. Moreover, immunohistochemical analysis revealed that cells infected with C. pneumoniae were found in close proximity to both Aβ and NFTs in the AD brain [246].

Although these studies point to a strong association between C. pneumoniae and AD, it does not prove that C. pneumoniae directly causes AD. The increased risk of AD and observed presence of C. pneumoniae in AD brains may reflect a secondary effect, where the bacterium triggers neuroinflammation, which subsequently accelerates or exacerbates AD pathology. Thus, more controlled trials are necessary to confirm this relationship.

Antibacterial therapeutic strategies targeting AD pathology

Considering the potential role of bacteria in AD risk and pathogenesis, antibacterial treatments have been tested as a therapeutic strategy in AD. Antibiotics are commonly used to treat bacterial infections, and some of them have also been investigated in the context of AD. As discussed earlier, minocycline may improve cognitive impairments and AD-related pathology [126, 127, 130].

Additionally, rifampicin and doxycycline are among the antibiotics studied as a treatment approach for AD. Rifampicin exhibits anti-inflammatory, anti-oxidative, and neuroprotective properties [247, 248]. It can prevent Aβ aggregation and protect neuronal cells from Aβ-induced cytotoxicity in vitro [249]. In AD mouse models, rifampicin reduced Aβ-oligomers, synapse loss, and microglial activation and improved spatial memory in a dose-dependent manner [250]. It also reduced tau pathology and improved memory in tauopathy models, likely by inhibiting protein oligomerization [250].

Additionally, rifampicin has shown a promising effect in a retrospective FDG-PET (Fluorodeoxyglucose-Positron Emission Tomography) study in a clinical study, suggesting preventive effects on the progression of AD with a daily dose of at least 450 mg for 1 year [251].

Similarly, doxycycline inhibits Aβ formation in vitro, highlighting its anti-amyloidogenic activity [252]. Furthermore, doxycycline treatment in APP/PS1 mice at 15 to 16 months of age rescued memory deficits. Surprisingly, the treatment did not affect the plaque load, suggesting that doxycycline restores memory without affecting insoluble Aβ pools. The memory recovery was linked to a reduction in glial activation and neuroinflammation, leading the authors to propose that doxycycline might operate by returning glial cells to a state that enables the regular processing of new memories or by interfering with the oligomeric Aβ species [253].

Combined treatment with rifampicin (300 mg) and doxycycline (200 mg) over three months significantly reduced cognitive decline in patients with mild to moderate AD in a clinical study. While the authors speculated that the effect of the antibiotics might be attributed to the reduced cerebral amyloid deposition, it is noteworthy that the study did not include measurements of Aβ plaques [254].

Yet, some studies found an association between antibiotic use and cognitive impairment, AD, and dementia risk. For example, a year of treatment with rifampicin and doxycycline led to a decline in cognition among AD patients, though the decline was not significant when both drugs were combined [255]. It is plausible that some properties of rifampicin and doxycycline might be beneficial in the short term, whereas other properties may have a negative impact that becomes more evident during long-term (12-month) treatment [255].

A large retrospective cohort study in South Korea (n > 300,000) found that antibiotic use—particularly cephalosporins—for three months or longer increased the risk of AD and dementia, with longer cumulative exposure linked to even higher risk [256]. This may be due to neuroinflammation and neurodegeneration induced by antibiotics through the gut microbiota-brain axis [256, 257]. As will be discussed in the coming sections, changes in gut microbiota can affect the pathogenesis of neurodegenerative diseases, including AD. Prolonged antibiotic exposure can lead to short and long-term changes in gut microbiota that might induce changes in the gut-brain axis, potentially affecting the incidence of dementia [256, 257]. Importantly, the authors suggest that long-term prescription of antibiotics may be attributed to the poor health status of the participants, which might be a factor in increasing their vulnerability to dementia [256]. While this study links cumulative antibiotic use to dementia risk, further research is needed to determine whether antibiotics are a direct cause or if underlying health conditions contribute to this association.

Similarly, in a cohort study, long-term antibiotic use (minimum 2 months) in midlife was associated with cognitive impairment assessed seven years later, particularly in women treated for respiratory or urinary tract infections [258].

Thus, the therapeutic use of antibiotics in AD remains controversial. Although eliminating chronic infections might help with disease prevention and AD pathology by reducing neuroinflammation, many antibiotics can cause changes in the composition of gut microbiota, which might worsen the disease course [257]. Interestingly, Aβ itself may have antimicrobial properties, suggesting its accumulation could also reflect a protective response to chronic infections [259].

As a complementary strategy to antibiotics, targeting host immune pathways may offer additional benefits. For instance, Brucella abortus, a bacterium that causes chronic host infections, was shown to upregulate TREM2 expression in M2 macrophages, suppressing mitochondrial ROS production and preventing pyroptosis, thus enhancing bacterial survival [259]. Downregulation of TREM2 reversed these effects [259]. Similarly, genetically deleting or blocking TREM2 reduced the survival of M. tuberculosis through enhanced ROS production, which highlights TREM2 as a potential therapeutic target in infection-driven immune evasion and inflammation [261]. However, TREM2`s role in pathogen removal appears to show conflicting results, as it enhanced bacterial clearance in S. typhimurium and P. aeruginosa infections, suggesting that its effect may vary by microbial context and immune environment [262–264].

In summary, the findings suggest that different bacteria may contribute to a sequence of processes, including neuroinflammation and amyloid accumulation, ultimately leading to AD-like pathology. This further supports the infection hypothesis of AD, which proposes that microbial factors play a role in disease onset and progression.

Oral infections - Alzheimer’s disease

In the preceding sections, we discussed the association between non-oral bacteria and AD. We now focus on the potential etiological link between oral infections—particularly periodontitis—and AD.

Periodontitis, a prevalent oral inflammatory disease that damages tooth-supporting structures and leads to tooth loss [265, 266], affects around 11% of the global population [268].

In addition to being a source of local inflammation, periodontitis can trigger systemic inflammation and act as a driver for a chronic immune response [268]. This occurs through the release of pro-inflammatory cytokines into the bloodstream, which can exert effects on a systemic level or enter the brain. Moreover, periodontitis can facilitate the transport of intact bacteria and viruses to the brain [269–271]. The systemic inflammation triggered by untreated periodontitis might trigger or exacerbate AD [270–273].

Different population-based retrospective studies showed that chronic periodontitis patients had an increased risk of developing AD and overall dementia [274, 275]. Notably, periodontitis has been observed to accelerate the cognitive decline in AD patients during a six-month follow-up period, likely due to the systemic inflammation it induced [273].

Dysbiosis of the oral microbiome has a crucial role in the development of periodontitis and is notably evident in AD patients. Pathogenic bacteria linked to periodontitis such as Porphyromonas gingivalis, Fusobacterium nucleatum, and Actinomyces naeslundii are significantly elevated in AD patients compared to healthy controls [276–278], making them potential risk factors for AD pathogenesis.

For example, P. gingivalis, the keystone pathogen in periodontitis, impairs spatial memory and elevates brain levels of TNF-α, IL-6, and IL-1β in wild-type mice and rats [279–283].

It also increases hippocampal and cortical microglia and astrocytes [281, 284, 285], which might result from the ability of P. gingivalis to induce the formation of ASC specks —protein aggregates that amplify inflammasome activation and neuroinflammation— through activation of the NLRP3 inflammasome [286]. NLRP3 activation in periodontal tissues enhanced the secretion of proinflammatory cytokines such as IL-1β and IL-18 [287]. However, P. gingivalis-induced alveolar bone loss was significantly reduced in NLRP3-deficient mice, suggesting that NLRP3 inflammasome mediates the inflammatory response driving periodontal disease [287].

Furthermore, TREM2 expression is upregulated in the alveolar bone of chronic periodontitis patients. TREM2 was found to promote osteoclast differentiation and bone resorption by enhancing intracellular ROS levels [288]. This effect was further exacerbated by Aβ oligomers, suggesting a potential link between amyloid pathology and periodontal bone loss via TREM2 signalling [288].

Moreover, oral administration of P. gingivalis increases Aβ and p-tau in the cortex and hippocampus and upregulates APP and BACE1 in both AD and wild-type mice [280, 281, 284, 289]. These data implicate that periodontitis might promote AD pathology in the brain.

Notably, after oral administration of P. gingivalis, the bacterium and its toxic proteases (gingipains) were detected in the brains of the wild-type rats and wild-type and ApoE^−/− mice, where they are localized in microglia, astrocytes, and neurons [280, 284, 290].

DNA of P. gingivalis was also detected in the CSF of clinical AD patients by Dominy et al. In this study, the researchers demonstrated that gingipains were present in 96% of post-mortem AD brains. The levels were significantly higher in the AD brains, including the hippocampus, where the gingipains colocalized with neurons, astrocytes, and tau and ubiquitin pathology compared to control brains, which may provide direct evidence of the pathogen`s involvement in AD [278]. The presence of LPS from P. gingivalis in post-mortem AD brains further supports this link [291].

Beyond P. gingivalis, F. nucleatum infection in 5xFAD mice exacerbated memory impairment, increased Aβ and p-tau, and tripled TNF-α and IL-1β levels compared to controls [266].

These findings suggest that treating oral pathogens could reduce AD risk and pathology. In this manner Dominy et al. tested gingipain inhibitors (e.g., atuzaginstat) in P. gingivalis-infected mice and found reduced brain bacterial load, neuroinflammation, Aβ1–42 levels, and neuronal loss—highlighting therapeutic potential despite clinical trial challenges [278].

In addition to bacterial protease inhibitors, targeting inflammatory pathways such as the NLRP3 inflammasome might offer therapeutic benefits. For instance, the NLRP3 inflammasome inhibitor MCC950 has been shown to reduce alveolar bone loss and osteoclast differentiation in mouse models of periodontitis [292]. Similarly, glyburide, a diabetes drug known to block NLRP3 activation, decreased inflammation and bone resorption in animal models [293, 294].

In summary, the relationship between periodontitis and AD underscores the importance of maintaining good oral health as a preventive measure against AD and neurodegenerative diseases [295]. By monitoring specific oral pathogens, such as P. gingivalis, we could potentially identify early indicators of AD before the clinical manifestations emerge.

Gut inflammation - Alzheimer’s disease

Gut microbiota, which is the system of microorganisms colonizing the gastrointestinal tract, can interact with the brain, a phenomenon which is referred to by the term microbiota-gut-brain axis [296]. It has been recognized that the modulations in the gut-brain axis can affect the pathogenesis of neurodegenerative diseases, including AD. Systemic inflammation and intestinal permeability can be induced by changes in gut microbiota composition, which is also called dysbiosis. The gut dysbiosis may lead to increased gut permeability, which facilitates increased translocation of LPS and bacteria into the systemic circulation, which might trigger a persistent state of systemic inflammation [297, 298]. Ultimately, systemic inflammation caused by the alterations in the gut microbiota may lead to or worsen AD pathology, including Aβ deposition and cognitive impairments [299]. It is suggested that this effect might be partially mediated through NLRP3 inflammasome signaling, which links gut dysbiosis to neuroinflammatory responses [300]. Notably, transplantation of gut microbiota from AD patients into mice activated the NLRP3 inflammasome in the intestines of mice, leading to increased peripheral inflammation and microglial activation in the brain, resulting in exacerbated cognitive impairment [301].

Aging can influence gut microbiota composition. For example, pro-inflammatory bacterial species in the gut microbiota have been found to increase with aging, whereas beneficial bacteria with anti-inflammatory properties, such as F.prausnitzii, decrease [302].

AD pathology itself influences gut microbiota composition. Additionally, lifestyle changes commonly observed in AD patients, such as altered diet, reduced physical activity and sleep disturbances, may also contribute to shifts in gut microbiota composition, further reinforcing the bidirectional relationship between AD and microbial dysbiosis [259]. The pro-inflammatory gut bacteria populations increase in AD patients, whereas the diversity of gut microbiota decreases [303]. The increased intestinal permeability induced by microbial dysbiosis elevates plasma LPS, serum IL-1, and TNF-α levels. Gut dysbiosis can occur before the onset of dementia, as shown in a study where there were no differences in microbial diversity between patients with AD and those with mild cognitive impairment (MCI) [303].

Interestingly, recent studies indicated sex-specific differences in the gut microbiome and its impact on AD pathology. Changes in gut bacterial composition, as well as reductions in amyloid plaques and neuroinflammation following interventions such as antibiotic treatment or sodium oligomannate (GV-971), were observed mainly in male mice, but not in females [304, 305]. Estrogen levels might modulate these microbiome changes and subsequent amyloid pathology, suggesting complex crosstalk between sex hormones, gut microbiota, and AD progression [304, 305]. This finding highlights the importance of considering sex differences when investigating interventions for AD.

Long-term antibiotic use might also result in gut dysbiosis and can alter systemic inflammatory components, impacting cognitive functions [255, 257, 258, 296]. For example, a recent study using the 3xtg-AD mouse model demonstrated that Klebsiella pneumoniae, a common hospital-acquired infection, was able to translocate from the gut to the bloodstream and further reach the brain by crossing the blood-brain barrier. This effect was especially under antibiotic-induced dysbiosis, which subsequently exacerbated AD pathogenesis, neuroinflammation, and cognitive impairment [306].

Furthermore, clinical studies revealed that microbiota dysbiosis caused by gastrointestinal tract (GT) diseases such as irritable bowel syndrome (IBS) heightens the risk of developing AD [307, 308]. Therefore, identifying individuals with GT disorders is crucial as interventions to restore healthy gut microbiota can reduce the risk of future AD development.

Notably, regional differences in AD prevalence have been reported. For instance, Japanese-American men in Hawaii showed higher AD rates than those in Japan, likely due to dietary differences—high fat intake being a risk factor, while fish consumption slightly reduced risk [309].

There are some microbiome-targeted therapies to restore a healthy gut microbiota, such as dietary or non-dietary interventions to reduce neuroinflammation [190, 310, 311].

Using probiotics is a promising strategy for re-establishing healthy gut microbiota and delaying AD progression [190, 312]. Prebiotics and probiotics can enhance the GI barrier function, decrease intestinal permeability, and reduce inflammation [310].

In a preclinical study, 3xTg-AD mice treated with SLAB51 (a probiotic mix of nine bacterial strains) for four months showed improved cognition, modified microbiota, and reduced Aβ plaques compared to controls [190].

Furthermore, in a study exploring the potential impact of gut microbiota on AD pathogenesis, a transgenic AD mouse displaying both Aβ and tau pathology (ADLP^APT) exhibited differences in the gut microbiota composition compared to wild-type mice [311]. Notably, when ADLP^APT was transplanted with the faecal microbiota from healthy wild-type mice almost every day for 4 months, there was a notable reduction in Aβ and tau pathology, as well as glial activity and cognitive impairment. Consequently, the authors suggested that microbiota-mediated intestinal and systemic inflammation contribute to AD pathology, and targeting systemic factors, including the re-establishing of a healthy gut microbiota, may serve as a good therapeutic target in AD [311].

The effect of probiotics on AD has also been investigated in a few clinical studies, although the number of studies conducted so far remains limited. For instance, AD patients supplemented with 200 ml/day probiotic milk containing Lactobacillus and Bifidobacterium for 12 weeks displayed a significant improvement in cognitive functions as assessed by the Mini-Mental State Examination, compared to AD patients who received a placebo [313]. Moreover, probiotic treatment before undergoing surgery can reduce POCD induced by surgery and reduce surgery-induced neuroinflammation by manipulating gut microbiota [188, 193].

In conclusion, the interplay between gut microbiota, systemic inflammation, and AD pathology indicates the importance of therapeutic interventions targeting gut health. Gut microbiota profiling could be a potential biomarker for future AD risk association. Addressing and managing gut inflammation as a lifestyle intervention, given its ease and affordability, may help alleviate AD symptoms and could play a role in delaying or preventing the onset of the disease when combined with other therapies.

Cardiovascular disease and risk factors - Alzheimer’s disease

There are several cardiovascular diseases and risk factors linked to AD risk, with systemic inflammation being the common underlying factor.

For example, obesity is considered a significant risk factor for AD. Mid-life obesity, particularly characterized by a body mass index (BMI) higher than 30, increases the risk of developing AD as supported by several studies [314–317]. Interestingly, this correlation appears to reverse as people age, leading to what is known as the obesity paradox, where older obese individuals exhibit a reduced risk of AD [318–320]. Weight loss, which often occurs before cognitive decline and is considered a non-cognitive sign of preclinical AD, might help explain the obesity paradox in AD [321, 322]. Weight loss has been linked to brain atrophy and appears to alter the relationship between late-life obesity and brain atrophy [322]. The obesity paradox may possibly result from reverse causation, where early weight loss related to preclinical AD masks the detrimental impact of obesity in older adults [321].

Similarly, metabolic syndrome (MetS), a cluster of conditions including obesity, insulin resistance, hypertension, and dyslipidemia, has been associated with an increased risk of cognitive decline and AD [323–327]. MetS contributes to neurodegeneration through chronic systemic inflammation, vascular dysfunction, and impaired insulin signaling, which can promote Aβ accumulation and tau pathology [326].

A recent large population-based study in the Netherlands examined whether different changes in MetS status differ in cognitive functioning, but found no significant differences between groups [328].

Obesity triggers low-grade inflammation in peripheral tissues and the circulation, marked by elevated levels of proinflammatory cytokine levels including TNF-α and IL-1β [329–332]. For instance, a high-fat diabetogenic diet can induce AD pathogenesis through immune system activation via toll-like receptor 4 (TLR4), a key element in the inflammatory response [333]. The glial activation in the brain could be directly influenced by the saturated fatty acids that cross the BBB, which could be one of the mechanisms of how obesity can stimulate AD pathogenesis [334]. Furthermore, this inflammatory state is partly driven by NLRP3 inflammasome, which senses obesity-associated danger signals, leading to caspase-1 activation and IL-1β and IL-18 secretion [335]. Calorie restriction and exercise-mediated weight loss in obese individuals with the 2 diabetes significantly reduced NLRP3 expression in adipose tissue, leading to decreased inflammation and improved insulin sensitivity [335]. Moreover, a recent study demonstrated that NLRP3 inflammasome inhibitors NT-0249 and NT-0796 reversed high-fat-diet-induced obesity and reduced systemic inflammation and astrogliosis in mouse models [336]. In addition to NLRP3, macrophages expressing TREM2 have been associated with the onset and progression of obesity [337]. In a recent study, TREM2 was shown to regulate astrocytic lipid metabolism and reduce neuroinflammation in an AD mouse model, supporting its potential as a therapeutic target [338].

The systemic inflammation observed in obesity can lead to insulin resistance and type 2 diabetes (T2D) [334]. Similar to obesity, diabetes is a cardiovascular risk factor for AD, with studies showing that T2D patients have an increased risk of developing AD [339–342]. Insulin dysfunction, a hallmark of T2D, can directly influence AD pathology by regulating amyloid-beta metabolism and tau phosphorylation [343–345]. Insulin and Aβ are both substrates for the insulin-degrading enzyme (IDE), highlighting a molecular connection between AD and T2D [346–349].

Moreover, T2D induces systemic inflammation, evidenced by elevated proinflammatory cytokines, such as TNF-α, which can further exacerbate insulin resistance [332, 350, 351].

In support of this, increased NLRP3 activation and IL-1β and IL-18 expression have been observed in T2D patients, suggesting an increased innate immune response that may contribute to systemic inflammation [352].

These effects of T2D can contribute to elevating the risk of developing AD or worsening the disease.

Therefore, targeting systemic inflammation could be a suitable therapeutic strategy both for T2D and AD [353], as certain antidiabetic medications have shown promise in mitigating AD pathology [354, 355].

Other cardiovascular diseases and risk factors, including coronary heart disease (CHD), stroke, hypertension, and high cholesterol, are also linked to AD, with inflammation being a significant contributor to the diseases [356]. A large meta-analysis study showed that individuals with CHD and heart failure had a 27% and 60% increased risk of developing dementia, respectively [357]. Interestingly, serum soluble TREM2 levels are increased in CHD patients and associated with several cardiovascular risk factors, suggesting that soluble TREM2 might show good diagnostic potential in identifying CHD [358].

Additionally, hypertension and high serum cholesterol levels in midlife significantly raise the subsequent AD risk in later life, particularly when both factors are present [359–361]. Another study demonstrated that high systolic blood pressure in mid-life was linked to the formation of plaques and neurofibrillary tangles, whereas another study found that midlife hypertension might impair the clearance of Aβ plaques from the brain, suggesting a molecular connection to AD pathology [362, 363].

A low-grade chronic systemic and CNS inflammation is a contributing factor to the physiopathology of hypertension [364, 365]. A recent review discusses that inflammation acts as a mediator between hypertension and neurodegenerative diseases, possibly through a reduction in cerebral blood flow or disruption of the BBB, which may allow inflammatory cells to enter the brain tissue [365].

Furthermore, arterial stiffness is considered an independent risk factor for both AD and cardiovascular diseases, and it is thought to be mechanistically linked to AD through inflammation [366].

Considering the link between cardiovascular risk factors and diseases with dementia and AD, intervention of these factors with pharmacology or lifestyle changes can be a preventative strategy against AD [89, 359, 367, 368]. Several studies showed that the use of anti-hypertensive drugs such as angiotensin-converting enzyme inhibitors, β-blockers, calcium channel blockers, and diuretics reduced the risk of AD and dementia [369–374]. Furthermore, following a healthy lifestyle including regular physical exercise and a Mediterranean diet during early and mid-adulthood can effectively lower inflammation and prevent dementia and AD [375–379].

Conclusions and future perspectives

AD is increasingly recognized as a multifactorial condition involving systemic processes. We propose that inflammaging combined with various lifelong exposures to peripheral immune system challenges such as metabolic changes, injuries, infections and some diseases, may prime the immune system in ways that amplify or aggravate AD pathology. In this context, systemic inflammation emerges as a central factor connecting AD with other diseases and conditions.

Cardiovascular factors, particularly obesity and T2D, are among the strongest peripheral contributors to AD risk, due to their promotion of chronic low-grade inflammation and metabolic dysfunction. Viral infections like HSV1, especially in APOE4 carriers, show a strong and well-documented link to AD risk, whereas bacterial pathogens such as P. gingivalis may promote direct CNS invasion and inflammatory cascades. Chronic pain is a moderate, sustained inflammatory stressor that primes the CNS over time, while POCD involves acute systemic immune activation, which might be particularly harmful in vulnerable older adults. Gut dysbiosis is emerging as a significant modulator of systemic and neuroinflammation that can impair the BBB, with growing evidence linking impaired gut-brain barrier function to AD development.

These comorbidities can precede or coexist with AD, significantly impacting the disease’s clinical course and progression. Therefore, we suggest that managing peripheral inflammation, along with the early identification and treatment of the conditions that promote systemic inflammation, could potentially reduce the risk of developing AD.

Even though chronic microglial overactivation can exacerbate AD and other diseases, microglia are essential for the uptake and clearance of Aβ. Therefore, when developing treatments targeting neuroinflammation and microglia, it is important to find the balance to keep the neuroprotective effect of microglia and even enhance it, while mitigating detrimental inflammation.

For example, increased TREM2 is protective in AD patients [380–382], whereas loss-of-function mutations in TREM2 increase the risk for late-onset AD by impairing inflammatory processes [383, 384].

Considering TREM2’s role in the phagocytic activity of microglia on amyloid plaques, reduced TREM2 might result in brain damage by hindering the clearance of Aβ and other toxic substances [384, 385].

Hence, enhancing TREM2 signalling may be an effective therapeutic strategy to support the microglial response to the Aβ plaques and preserve the neuroprotective actions of microglia [385]. Studies have shown that increased TREM2 activation through TREM2 agonistic antibodies or overexpression in mouse models improves cognition and reduces amyloid pathology by enhancing microglia-mediated phagocytosis [385–392]. The stage of the disease may be a moderating factor for TREM2-associated microglial activity, with early symptomatic stages potentially being optimal for therapeutic intervention [380].

In addition to TREM2, modulating the TNF receptor pathways can boost the neuroprotective effects of TNFR2, while still allowing TNFR1 to maintain its role in mediating neuroinflammation. This dual approach helps in modulating neuroinflammation without completely suppressing the necessary immune responses of microglia.

Another promising target for reducing detrimental microglial activation could be the NLRP3 inflammasome, as Aβ-induced NLRP3 inflammasome activation plays a key role in inflammation, AD pathology, and memory deficits [393–397]. NLRP3 inhibitors or Nlrp3-/- mice models have demonstrated neuroprotective effects and reduced neurodegeneration [393, 398–401].

All these approaches could help modulate detrimental inflammation while keeping the neuroprotective effects of microglia. Targeting downstream molecules in inflammatory mechanisms can be a good option instead of inhibiting final cytokines and molecules, as they might have beneficial effects [394]. Moreover, as the dynamic of microglia changes during disease progression, it is crucial to determine the optimal intervention time before or during the disease course, as inappropriate modifications might lead to even harmful effects.

Given the therapeutic relevance of TREM2 and NLRP3 in AD, future research should aim to validate these pathways as clinically useful biomarkers. A prospective longitudinal study could recruit individuals at risk for AD, such as APOE4 carriers, those with mild cognitive impairment, or individuals with comorbidities described in this review, and perform regular measurements of soluble TREM2 and NLRP3-related markers (such as IL-1β, ASC specks, caspase-1 activity) in CSF or plasma. These biomarkers could be tracked together with cognitive outcomes and neuroimaging to assess their relevance as predictive indicators and potential clinical trial endpoints [402–406].

A limitation of the evidence shown in this review is that most studies supporting the involvement of peripheral immune alterations in AD pathogenesis have been performed in transgenic mouse models carrying familial AD mutations. In these models, Aβ overproduction represents the primary and initiating pathological hallmark. Consequently, the use of these mouse models limits our ability to determine the sequence of events leading to the disease onset and progression. Specifically, it remains unclear whether inflammaging and neuroinflammation act as upstream triggers that lead to Aβ accumulation or whether they arise as a secondary response to Aβ aggregation. This distinction is critical for understanding the disease etiology and for developing effective therapeutic approaches. Future research employing alternative models, such as sporadic AD models or human-based studies, is essential to clarify whether immune dysregulation precedes or follows Aβ pathology in the context of AD.

Furthermore, future human studies must also address the challenges of proving causality between peripheral immune challenges and AD pathology. This requires designing longitudinal studies, establishing well-defined inclusion and exclusion criteria, accounting for variables such as age, sex, comorbidities, and history of systemic inflammation, such as surgeries and infections [42]. Baseline cognitive status, APOE genotype, peripheral inflammation markers, and lifestyle factors might further help control for heterogeneity and identify individuals at greater risk for AD and neurodegeneration following peripheral immune insults. Yet, proving causation remains difficult due to AD`s long latency period, its multifactorial etiology, and the confounding effects of comorbidities and reverse causation [407].

In conclusion, viewing AD as a systemic disease involving dynamic interactions within peripheral and central immune compartments offers a comprehensive perspective. This approach deepens our understanding of disease pathogenesis and risk factors and may facilitate the development of improved diagnostics and novel therapeutics. Early diagnosis and modulation of the diseases and factors that contribute to peripheral inflammation could mitigate the risk of developing AD. This insight is crucial for advancing AD research and care, highlighting the importance of addressing systemic inflammation and comorbidities in managing AD.

Abbreviations

Aβ

Amyloid-beta

AD

Alzheimer’s disease

ALS

Amyotrophic lateral sclerosis

APP

Amyloid precursor protein

APOE4

Apolipoprotein E4

BBB

Blood-brain barrier

BACE1

Beta-site amyloid precursor protein cleaving enzyme 1

BD

Bipolar disorder

CNS

Central nervous system

CHD

Coronary heart disease

CSF

Cerebrospinal fluid

FACS

Fluorescence-activated cell sorting (Flow Cytometry)

HCV

Hepatitis C virus

HIV

Human immunodeficiency virus

HSV1

Herpes simplex virus type 1

IDE

Insulin-degrading enzyme

IL-1β

Interleukin-1 beta

IL-6

Interleukin-6

IL-12

Interleukin-12

LPS

Lipopolysaccharide

MetS

Metabolic syndrome

MGB axis

Microbiota-gut-brain axis

MCI

Mild cognitive impairment

MRI

Magnetic resonance imaging

MWM

Morris Water Maze

NFTs

Neurofibrillary tangles

NLRP3

NOD-like receptor protein 3

NSAIDs

Nonsteroidal anti-inflammatory drugs

pTau

Phosphorylated tau

POCD

Post-operative cognitive dysfunction

PD

Parkinson’s disease

ROS

Reactive oxygen species

T2D

Type 2 diabetes

Tat

Trans-activator of transcription (HIV protein)

TREM2

Triggering receptor expressed on myeloid cells 2

TNF-α

Tumor necrosis factor-alpha

TNFR1

Tumor necrosis factor receptor 1

TNFR2

Tumor necrosis factor receptor 2

Tregs

Regulatory T cells

Author contributions

I.B. and N.O-C. wrote the manuscript. D.V.D, P.D.D. and U.E. reviewed it. All authors read and approved the final manuscript.

Funding

The work in this review was funded by a grant received from Alzheimer Nederland (WE.03-2021-05) granted to P.D.D and U.E.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.