Immunosenescence is a phenomenon of irreversible loss of the ability to divide, as a result of which damage to the immune defense is observed, which contributes to the progression of susceptibility to disease in the elderly. Immunosenescence occurs when the ability to respond to new antigens is reduced (Pawelec, 2017). In addition, a variety of factors affect the progression of immunosenescence—genetics, environment, lifestyle, and nutrition—leading to infections and progression of disease pathology (Figure 1) (Costantini et al., 2018).

Franceschi et al. (2000) first used the term “inflamm-aging,” proposing a hypothesis based on a series of observations showing that aging of many organisms, including humans, is accompanied by an increase in the level of inflammatory markers in the blood, cells, and tissues. This is chronic, sterile, not associated with the presence of an infectious agent, and primarily due to endogenous signals, or subclinical (asymptomatic), mild, or basal inflammation, which is associated with aging. Already today, more and more studies indicate that “inflammatory” aging or inflammation associated with age is a risk factor for many chronic non-communicable diseases, such as cardiovascular (coronary heart disease and arterial hypertension), metabolic [diabetes mellitus 2 type (T2D)], musculoskeletal (osteoarthritis, osteoporosis, and sarcopenia), neurological (depression, dementia, and AD), and hematologic (malignant neoplasms and anemia) diseases, which leads to adverse effects on human health (Giunta et al., 2008). Many researchers agree that the number of nosologies is not limited to this list (Deleidi et al., 2015; Pawelec, 2017). The approaches for the treatment of diseases in which inflammation predominates in the pathogenesis may include, in addition to limiting caloric intake and increasing physical activity, the use of drugs. The drug action is aimed at limiting it by interfering with the processes of intracellular and extracellular signaling at different stages, and not only involves the impact on the main clinical manifestations or targets (blood pressure, cholesterol, blood sugar, etc.) (Zotkin et al., 2020).

AD is a fatal neurodegenerative disorder that is pathologically defined by extensive neuronal loss and the accumulation of intracellular neurofibrillary tangles (NFTs) and extracellular amyloid plaques in the brain. Studies since the discovery of amyloid beta (Aβ) and tau protein have provided detailed information on molecular pathological events, but little is known about the causes of AD, as well as about the possible effective treatment of this pathology (Silva et al., 2019; Tiwari et al., 2019).

It is well known that the risk of late-onset AD is partially due to genetics. In 2019, the results of a large meta-analysis of associations across the genome of clinically diagnosed late-onset AD were published. Currently, 25 risk loci are affected, five of which were identified in the latest study. Another confirmation of the contribution of the immune system to the late onset of AD confirms that the neurological and immune-mediated disease haplotype HLA-DR15 (human leukocyte antigen) is a risk factor for AD. In addition to tau-binding proteins and amyloid precursor protein (APP) metabolism, pathway analysis includes immunity and lipid metabolism, which are also associated with late onset of AD (Kunkle et al., 2019).

Newly discovered evidence proposes that inflammation is an important feature of AD, diabetes mellitus, and other pathologies; and it is believed that this process plays an important role in the pathogenesis of these syndromes. Therefore, understanding the interactions between the nervous system and the immune system may be key to preventing or delaying the onset of most diseases of the CNS. Neuroinflammation is an important part of the brain’s defense mechanisms against a variety of pathological agents, such as infections and injuries, and includes both the formation of soluble factors and the activation of specialized cells that are mobilized to respond and restore the brain’s normal physiology. Neuroinflammation is characterized by neuronal death in certain areas of the CNS (Figure 2) (Salmina et al., 2015; Calsolaro and Edison, 2016).

In AD, central events seem to combine the inflammasome, NF-κB pathway, and the microglial activation by a variety of factors, including Aβ and proinflammatory cytokines (Harms et al., 2015; Rea et al., 2018). Aβ in the brain microenvironment causes the glial activation that leads to microgliosis and astrocytosis around pathological proteins. Thus, glial cells are chronically activated in the brain before the onset of AD, which is associated with the development of chronic inflammation and contributes to the pathogenesis of AD. In the AD brain, microgliosis and astrocytosis because of the presence of senile plaques and NFTs can be detected immunohistochemically, and these glial cells exhibit pathologically specific morphology. Although the degree of gliosis correlates with cortical thickness and neurodegeneration, the role of various glial cells in neurodegenerative processes remains unknown (Saito and Saido, 2018).

Microglia, the main immune cells of the brain’s innate immunity, perform movements similar to macrophages to remove pathogens and protect neurons from various factors. At the same time, microglial cells secrete reactive oxygen species (ROS) and nitric oxide, which are neurotoxic. They also release proinflammatory cytokines and chemokines in response to danger signals. Dysregulation of microglial activity is associated with the pathogenesis of AD during aging (Salmina et al., 2015; Clayton et al., 2017). Triggering receptor expressed on myeloid cells-2 (TREM2) plays an important role in supporting microglial cell survival. Previously, it was shown that TREM2 promotes microglial clustering around fibrillar Aβ plaques in AD mouse model and postmortem human brain sections (Ulrich et al., 2017). Furthermore, TREM2 is a Aβ receptor that mediates microglial function and removal of Aβ (Zhao et al., 2018). An increase in soluble TREM2 fragments in cerebrospinal fluid indicates coincidence with markers of neuronal damage and onset of clinical dementia in AD (Ulrich et al., 2017).

It should be noted that recent human positron emission tomography (Aβ-PET) data indicate that Aβ deposition begins years before memory impairment and cognitive decline (Hatashita and Wakebe, 2019). Given the fact that Aβ acts as strong damage-associated molecular patterns (DAMPs), it seems that the interval between early accumulation of Aβ and later signs of disease progression, such as tau pathology and brain atrophy, is influenced by innate immune responses (Table 1). One of the canonical pathways of this Aβ-induced innate immune response is the activation of the NOD-like receptor (NLR) family, a pyrin domain-containing 3 (NLRP3) inflammation, which has been the subject of intense research (Heneka et al., 2018) (Figure 3).

Studies have shown that Aβ oligomers can trigger the expression of the NLRP3 inflammasome and thus promote inflammation and intensify association between T2D and AD (Rea et al., 2018). Inflammasome is involved in the progression of metabolic syndrome due to impaired adipose tissue sensitivity. It has been conclusively demonstrated that obesity triggers NLRP3 activation and that secreted IL-1β that impairs insulin signaling, which contributes to IR in mice (Mori et al., 2011; Rea et al., 2018). Another study found that obesity was associated with NLRP3 activation in adipose tissue (Mori et al., 2011; Vandanmagsar et al., 2011; Rea et al., 2018).

The role of NLRP3 inflammasomes in the pathogenesis of obesity has been supported by data showing that Nlrp3^–/– and Asc^–/– knockout mice are protected from obesity and IR induced by a high-fat diet. In addition, NLRP3 activation by inflammasomes/caspase-1 appears to be a key regulator of adipocyte differentiation and directs adipocytes to an insulin-resistant phenotype (Stienstra et al., 2010).

Consequently, CR and weight loss in obese diabetic subjects decrease the expression of the Nlrp3 and IL-1 genes in adipocytes, improving insulin sensitivity (Vandanmagsar et al., 2011). However, some studies have failed to find an association between NLRP3 inflammasome formation and obesity or IR (Nishimoto et al., 2016). Understanding the molecular mechanisms of chronic inflammation remains a major medical problem (Nishimoto et al., 2016; Rheinheimer et al., 2017). Thus, further research is required to understand the relationship between NLRP3 inflammasome formation and IR.

Moreover, the inflammasome pathway is activated by a variety of intracellular processes and associated with increased age and age-related diseases. Both the inflammatory pathway and the senescent cell-related SASP activate the inflammasome through the NF-κB and IL-α cascade, causing the inflammatory response and cytokine production that delays resolution and healing (Chien et al., 2011; Mori et al., 2011; Rea et al., 2018).

Therefore, caspase-1 or inflammasome inhibitors have been proposed as novel treatments for pathologies associated with aging and metabolism deterioration (Stienstra et al., 2010; Kanbay et al., 2019).

Until now, the production of insulin in the brain remains a controversial issue. There is a lot of conflicting evidence about the production of insulin in brain structures and cell types. The initial hypothesis was that insulin is able to cross the blood–brain barrier (BBB) through a saturable transport system. However, this mechanism is limited and ineffective. Later, data appeared on the possible production of insulin in the brain. Thus, the expression of mRNA and insulin protein was found in the hippocampus, olfactory bulb, striatum, hypothalamus, and entorhinal and prefrontal cortices (Mehran et al., 2012). Insulin secretion has also been reported in cultured astrocytes (Takano et al., 2018). In a recent study, the authors describe the presence of not only insulin mRNA but also the protein itself in the epithelial layer of the choroid plexus of mice and humans, along with proteins associated with the processing and secretion of insulin (Mazucanti et al., 2019).

Since the stability of the cognitive reserve is largely determined by neurogenesis, the study of metabolic trophic factors that influence this process is important. Insulin at moderate concentrations is known to play a neurotrophic role. Insulin has a pivotal role in the brain development, functioning of neurogenic niches, and aging. Activation of the insulin/IGF-I (insulin-like growth factor) signaling pathway regulates the exit of neuroblasts from the quiescence state. This signaling cascade, insulin and IGF-I, has been shown to promote neurogenesis by modulating proliferation, differentiation, and survival of neural stem cells (NSCs) (rev. in Spinelli et al., 2019). In healthy metabolic conditions, acute increases in insulin levels have a valuable effect on cognition. Nevertheless, chronically elevated insulin significantly reduces the level of its mediated effects (Neumann et al., 2008). Moreover, chronic hyperactivation of the insulin/IGF-I pathway can cause premature depletion of the stem cell pool (Spinelli et al., 2019). In contrast, high insulin levels may be associated with increased Aβ deposition in the brain, as insulin and Aβ compete for the same enzyme that provides their clearance, namely, the insulin-degrading enzyme (Hölscher, 2019). Thus, insulin can have a trophic or deleterious effect on neurogenesis (Spinelli et al., 2019). This conclusion can be confirmed by studies demonstrating impaired learning in animals with a model of type 2 diabetes mellitus (T2DM), as well as the observed cognitive deficit in clinical studies among patients with this pathology (Zilliox et al., 2016).

Recently, the term “type 3 diabetes mellitus” has often been used to denote AD, since the pathological events accompanying this pathology are pathogenetically associated with central IR (Kandimalla et al., 2017). Aβ suppresses insulin expression in astrocytes (Pitt et al., 2017; Spinelli et al., 2019). These data show bi-directional changes between impaired brain insulin signaling and Aβ deposition in AD. According to these results, local IR and changes in central glucose metabolism may be considered as early markers for the diagnosis of AD (Hölscher, 2019).

It was shown that central glucose hypometabolism can be detected decades before the clinical onset of AD (Mosconi et al., 2009; Sperling et al., 2011; Neth and Craft, 2017). However, in the last few years, it is becoming increasingly obvious as a condition of reactive or compensatory glucose hypermetabolism in neurologic diseases as an initial reaction to trauma and developing pathological processes (Ashraf et al., 2015; Neth and Craft, 2017). According to Neth and Craft (2017), a glucose hypermetabolism could be a temporary solution to the injury problem with a permanent reduction in glucose utilization. If additional data confirm the occurrence of an initial increase and a final decrease in glucose metabolism in the brain, then this shift can be visualized at an early stage (before the onset of clinical symptoms) and work to prevent the pathology. This hypermetabolic glucose shift at the early stages is likely complemented by increased use of other fuels as well. As the disease progresses, a bioenergy shift may occur due to a decrease in glucose dependence and an increase in the use of alternative energy sources (Neth and Craft, 2017).

As it has already been mentioned, the most preferred energy substrate for the brain, except the prolonged fasting, is glucose. For normal functioning of the brain, a constant supply of glucose is necessary. In small concentrations, glycogen was also found in the brain, which is stored in astrocytes. Glycogen provides lactate as an energy source for neurons through monocarboxylate transporters to support neural functions such as hippocampal-regulated memory formation and learning (Rich et al., 2019). Under conditions of reduced glucose intake, ketone bodies, mainly formed as a result of fatty acid (FA) oxidation, are an alternative main source of energy. In addition, FAs and their metabolites are capable of influencing many brain functions. This also allows them to be considered as potential targets for pharmacological and/or dietary interventions in certain brain pathologies (Romano et al., 2017).

Nevertheless, glucose is of the greatest importance as a brain energy substrate, and therefore, disturbances in glucose metabolism have significant consequences on the functioning of the brain. Decreased hippocampal volumes have been described in elderly with impaired glucose tolerance. It accompanied with a lower cognitive test performance. Similarly, another study confirmed that patients with higher fasting glucose and glycosylated hemoglobin experienced decreased memory and learning ability (Grabenhenrich, 2014). Glucose hypometabolism was most obviously recorded in the frontal, parietal, and temporal cortices (Tondo et al., 2020). This suggests that IR affects similar areas of the brain as in AD, supporting the notion that central IR may contribute to neurodegeneration. Moreover, diabetes has been repeatedly shown to be a strong predictor of cognitive dysfunction in the elderly (Kong et al., 2018).

Cognitive decline is not limited to impairments in learning and memory; there are other impairments as well. Brain IR has been confirmed to be associated with decreased task processing speed, cognitive flexibility, and motor skills. Current evidence confirmed that IR should be considered as an pivotal risk factor for the progression of cognitive dysfunction (Moheet et al., 2015).

Thus, clinical work and experimental studies in animals propose that IR has destructive effects on cognitive functions, in particular on learning and memory. Therefore, it becomes the principal aim to study the metabolic pathways and their association with the progression of AD and other neurodegenerative disorders (Spinelli et al., 2019).

Changes in brain insulin signaling, and in particular in the hippocampus, can alter molecular pathways involved in synaptic plasticity and neurogenesis in adults, thereby leading to a decrease in cognitive reserve, an increased risk of neurodegeneration, and a shortened life span (Epel, 2020). Long-term persistent excess of nutrients is the cause of stress acceleration of aging. However, an excess of nutrients causes hyperactivation of insulin signaling and leads to desensitization of IR-dependent molecular cascades. Because of this influence, the brain stops responding to insulin and eliminates both the metabolic and cognitive effects of this hormone (Spinelli et al., 2019).

IR makes it difficult for cells to maintain energy homeostasis. The brain in AD neurodegeneration is accompanied by changes similar to those observed in peripheral tissues in diabetes mellitus, including metabolic stress and neuroinflammation (Talbot et al., 2012). Thus, it can be assumed that such mechanisms explain IR in T2DM and impair central insulin transduction in patients with AD. Significant similarities between neuropathogenic mechanisms are induced by Aβ oligomers and cause the loss of neurons and synapses, as well as mechanisms associated with peripheral IR in diabetes (Craft, 2012).

Many explanations have been proposed for the impaired insulin transduction in the AD brain. One of them is decreased extracellular insulin assessed in cerebrospinal fluid, decreased total or cell surface of insulin receptor expression, and decreased affinity of insulin receptors for insulin (Talbot, 2014). The deficiency of extracellular insulin in the AD brain remains unclear and identified the opposite results obtained in cerebrospinal fluid (Molina et al., 2002). Similarly, deficits in total insulin receptors in AD brain tissues were not found in studies using age-matched controls, and cell fractionation did not reveal deficiencies in insulin receptor levels on the cell surface. Although the binding of insulin to the insulin receptor may be reduced in the brain tissue in AD, insulin still manages to activate the catalytic domain of the insulin receptor at a level of 71–74% of normal levels even in the hippocampal formation in AD. As noted above, a significantly greater decrease in insulin sensitivity is observed already after binding to the insulin receptor in the brain in AD. In the hippocampus, which is responsible for memory and learning, insulin activated only 10% of the normal level of insulin receptor substrate (IRS) (Talbot et al., 2012).

Thus, the most likely cause of decreased insulin signaling in the brain in AD is IR due to dysfunctional IRS-1. This appears to reflect Aβ-induced secretion of proinflammatory cytokines by glial cells. Among the early changes in AD, there is an increased solubility of Aβ, the monomers of which combine to form oligomers, which can later assemble into fibrils and form amyloid plaques (Sengupta et al., 2016). Also at the AD onset, Aβ oligomers activate microglia, which leads to the secretion of proinflammatory cytokines such as IL-1, IL-6, and tumor necrosis factor α (TNF-α) (Hemonnot et al., 2019). Such activation of microglia may play a key role in the AD pathogenesis, given the recent discovery that disabling a gene in AD model encoding a microglial receptor (i.e., NOD-like receptor 3) and perceiving inflammatory pathogens, including Aβ, prevent the development of AD and cognitive abnormalities that usually occur in this AD animal model (Heneka et al., 2013). Through neuronal receptors, microglial IL-1, IL-6, and TNF-α activate serine kinases IRS-1, known as IkBa kinase (IKK), C-Jun N-terminal kinase (JNK), and extracellular signal-regulated kinase (Erk2) (Talbot and Wang, 2014). Thus, Aβ oligomers injected into neuronal cultures or into the cerebral ventricles markedly increase the phosphorylation of IRS-1 serine (IRS-1 pS) at several sites, namely, S312, S616, and/or S636 (S307, S612, and S632 in rodents’ Erk2) (Talbot and Wang, 2014).

Elevated neural IRS-1 pSer is significant in the cortex and hippocampal formation in AD and, apparently, is the main cause of IRS-1 dysfunction in AD (Tanokashira et al., 2019). The most common cause of IR is inhibition of downstream signaling due to serine phosphorylation of the IRS1 (Boucher et al., 2014). Similar changes occur in the AD brain. Insulin activation of IRS-1 is consistently decreased in tissues with significantly increased levels of IRS-1 pS616 and IRS-1 pS636. These molecules may act as potential markers of central IR (Talbot et al., 2012; Talbot and Wang, 2014). As expected, the levels of these candidate biomarkers correlate significantly with the Aβ deposition and are associated with cognitive decline (Kong et al., 2018).

This may explain crossroad peripheral IR due to obesity and/or diabetes mellitus and brain IR in AD (Liu et al., 2011). Obesity and T2DM are, in fact, risk factors for AD and are associated with increased vascular proinflammatory cytokines (Pugazhenthi et al., 2017). With the development of AD, the impaired integrity of the BBB promotes the penetration of cytokines. They, in turn, activate IRS-1 serine kinases in the same way as cytokines obtained from microglia (Ferreira et al., 2018).

The relationship between IR and AD may be enhanced due to a common etiology leading to an increased risk of AD (Spinelli et al., 2019). IR is a potentially modifiable risk factor for AD; in this regard, early diagnosis of IR remains highly relevant. Previous research has relied on measurements of systemic IR based on blood glucose and insulin values, such as the Homeostatic Model of Insulin Resistance (HOMA-IR) assessment (da Silva et al., 2019). Peripheral and central IR overlap to some extent, and this may explain why associations between HOMA-IR and glucose hypometabolism in the brain have been observed. Currently, a variety of proteins have been isolated that are consistently significantly related to IR and AD pathology. They are discoidin, CUB, and LCCL domain-containing protein 2 (DCBLD2); Ephrin-B2 (ENFB2); ciliary neurotrophic factor receptor subunit alpha (CNTFR); neuronal growth regulator 1 (NEGR1); leucine-rich repeat-containing protein 4B (LRRC4B); and SLIT and NTRK-like protein 4 (SLITRK4). However, a search for markers specific to central IR is also required (Westwood et al., 2017).

For these reasons, in recent years, research has focused on assessing glucose metabolism in the brain and analyzing extracellular brain vesicles extracted from blood as biomarkers of IR and early phase of cognitive decline. In a study in patients with IR, but without loss of cognitive functions, IR was associated with hypometabolism in the hippocampus and higher levels of blood pressure biomarkers in the cerebrospinal fluid (Westwood et al., 2017). Cerebral glucose metabolism is closely related to neuronal activity, and a decrease in the cerebral metabolic rate for glucose (CMRglc) is one of the main hallmarks of AD. In vivo imaging using 2-[¹⁸F]fluoro-2-deoxy-D-glucose PET (FDG-PET) demonstrates a consistent and progressive decrease in CMRglc in patients with AD. A decrease in CMRglc in preclinical stages of AD, with mild cognitive impairment (MCI) was also indicated, as well as in carriers of the allele of apolipoprotein E-epsilon-4, a strong genetic risk factor of AD (Mosconi et al., 2008).

In a recent study, Mullins et al. (2017) demonstrated that pSer312-IRS-1 (which stimulates uncoupling of IRS-1 and leads to its degradation) and p-panTyr-IRS-1 (which promotes insulin-stimulated responses) are biomarkers of AD. Based on these data, a methodology for the isolation of exosomes from plasma was developed, followed by immunoprecipitation against the cell surface adhesion protein L1-CAM to enrich neural origins. It was demonstrated that pathogenic and signal peptides in plasma exosomes expressing L1-CAM/NCAM effectively distinguish between AD and control group and can predict the diagnosis of the disease. In addition, it was confirmed that the peripheral IR is separated from the brain IR that occurs in AD (Mullins et al., 2017; Kapogiannis et al., 2019; Spinelli et al., 2019). These data indicate the presence of IR of the brain in patients with AD and, to a lesser extent, in people with diabetes (Mullins et al., 2017; Spinelli et al., 2019).

Inflammation is a feature of diabetes mellitus and AD, and it is believed that this process plays an important role in the pathogenesis of these two pathologies. Inflammation is an important part of the body’s defense mechanisms against many pathological agents, such as infections and trauma, and includes both the formation of soluble factors and the activation of specific cells that are mobilized to respond and maintain the normal physiology and homeostasis (Chatterjee and Mudher, 2018).

It is believed that similar inflammatory processes occur in the CNS and periphery. The presence of inflammatory markers in AD in the brain tissue, including increased levels of cytokines/chemokines, which accompany with gliosis, was confirmed in many studies (Cao, 2015). In addition, an increase of inflammatory mediators in the blood, such as TNF-α, IL-6, and IL-1β, was observed in patients suffering from AD (Ng et al., 2018). Moreover, increased production of proinflammatory cytokines in adipose tissue is a key feature of the pathogenesis of metabolic disorders. A recent study has shown that an elevated level of TNF-α expressed in adipose tissue of obese individuals is the reason of peripheral IR. Therefore, in both the central and peripheral tissues, uncontrolled or chronic inflammation accompanies IR (Saltiel and Olefsky, 2017). Interestingly, inflammation also underlies hypothalamic dysfunction in obesity (Samodien et al., 2019). New evidence indicates that it is the inflammation and stress of the endoplasmic reticulum that are critical pathogenetic events in the central and peripheral IR in metabolic disorders (Arruda et al., 2011). In an obese and diabetic animal model, the neuroinflammation, especially through activation of TNF-α and the IkBa kinase (IKK)-b/nuclear factor-kB pathway, is a principal mechanism underlying the disease pathogenesis (Milanski et al., 2009). Consequently, the pathology of the hippocampus in AD and the pathology of the hypothalamus in obesity have common pathogenetic pathways associated with inflammation (De Felice and Ferreira, 2014).

Our own data have shown the protective phenotype of Nlrp3 knockout mice in the development of brain IR (Chernykh et al., 2018). Using an experimental approach to modeling AD, we investigated new molecular mechanisms of insulin signaling dysregulation in the amygdala in association with neuroinflammation and emotional disorders. It has been established that experimental AD is accompanied by impaired expression and functional activity of molecules–components of insulin-mediated signaling pathways and the development of IR together with up-regulation of neuroinflammation in the brain amygdala. This cascade of pathological reactions is reflected in emotional behavior disorder. NLRP3-dependent mechanisms have been demonstrated in the basolateral amygdala in normal conditions and during the development of neurodegeneration. It has been experimentally proven that preventing the development of local IR by blocking NLRP3 inflammasomes should be considered an approach to correcting BIR and emotional disorders in AD. The studied molecular mechanisms linking the development of local IR with neuroinflammation (with the participation of pIRS, GLUT4, IRAP, and NLRP3) and impaired cognitive and emotional spheres open up new possibilities for the prevention and correction of neurodegeneration in AD.

Thus, suppression of neuroinflammation by preventing the expression of NLRP3 inflammasomes in Nlrp3-knockout mice has a protective role in the development of AD, accompanied by IR, due to modulation of the expression of pIRS-Ser and downstream of insulin signaling cascade (Gorina et al., 2019).

In peripheral IR, impaired TNF-α signaling results in JNK activation (Chen et al., 2015). Activation of the TNFa/JNK pathological pathway is associated with the main inflammatory and stress-signaling mechanisms, including tension of the endoplasmic reticulum and activation of stress kinases IKK (IkBa kinase) and PKR (double-stranded RNA-dependent protein kinase) (Nakamura et al., 2010). In T2DM, high levels of TNF-α initiate serine phosphorylation of IRS-1 by stress kinases, blocking insulin signaling (Nakamura et al., 2010). TNF-α levels increase in microvessels of the brain and cerebrospinal fluid in AD (Ruan et al., 2009). Initial information that impaired insulin signaling in neurons in AD is associated with proinflammatory signaling was based on the fact that oligomers of Aβ induce inhibition of IRS-1 through TNF-a/JNK activation (Bomfim et al., 2012). These ideas were confirmed, and it was shown that common mechanisms underlie damaged peripheral insulin transduction in diabetes mellitus and central local IR in the AD brain. Namely, it was shown that IKK and PKR were increased in AD in the brain, and they mediate the inhibition of IRS-1 in hippocampal neurons induced by oligomers of Aβ (Lourenco et al., 2013). IKK mediates neuronal inhibition of IRS-1 by Aβ oligomers (Bomfim et al., 2012).

Since the important role of inflammation in the pathogenesis of neurodegeneration and IR in various pathologies, including diabetes mellitus and AD, has been convincingly shown, approaches based on an anti-inflammatory strategy can be used to treat symptoms and to interrupt the vicious circle of metaflammation. As visceral fat is strongly linked to metabolic disorders, strategies for correcting IR in metabolic-cognitive states are very promising approaches as well (Kullmann et al., 2020).

There is epidemiological evidence that some anti-inflammatory approaches, in particular the use of non-steroidal anti-inflammatory drugs (NSAIDs), reduce the risk of AD. However, this category of drugs does not affect cognitive function (Rivers-Auty et al., 2020). In clinical trials, no evidence of the effectiveness of NSAIDs was found. The ineffectiveness of anti-inflammatory strategies may be due to inappropriate non-steroidal anti-inflammatory drugs or due to epidemiological results caused by confounding factors. However, there is evidence that, for example, the use of diclofenac is associated with a decrease in morbidity, as well as with a slower decline in cognitive function. However, this requires further research into the potential therapeutic effects of diclofenac in AD. Some antidiabetic drugs, which are aimed at lowering blood sugar levels, also have anti-inflammatory effects. This action is also associated with the hypolipidemic effect and direct modulation of immune responses. Despite promising results from clinical trials of anti-inflammatory drugs, salicylates, no clear guidelines have been established regarding the recommendation of these compounds for the prevention or treatment of T2DM. The use of other NSAIDs to combat metaflammation also requires clinical trials. However, most of the known methods of treating T2DM exhibit anti-inflammatory properties to varying degrees, which arise because of triggering various pathways and their effects, depending on many factors. Therefore, further clinical studies are needed to test new drugs and identify specific molecular pathways that could be therapeutically targeting metainflammation (Kuryłowicz and Koźniewski, 2020; Rivers-Auty et al., 2020).

There is evidence of a decrease in systemic inflammation with the applying of dietary protocols in clinical trials, associated primarily with a reduction in calorie intake (Lopez-Garcia et al., 2004; Kuryłowicz and Koźniewski, 2020). High-fat diets and high calories cause metainflammation, so the idea that dietary intervention can help reduce inflammatory the response in IR is very promising. Although the studies were not uniform in design, all reported that weight loss, improved glycemic control, and hepatic steatosis were associated with varying degrees with serum C-reactive protein (CRP) reduction (Kuryłowicz and Koźniewski, 2020).

Since it is a well-known fact that excess of nutrients in the course of obesity and IR impairs metabolism leading to the endoplasmic reticulum stress, possible CR may have a protective role (Park et al., 2012; Kuryłowicz and Koźniewski, 2020; Ma et al., 2020).

Calorie restriction is one of the most promising approaches for reducing the negative effects of metabolic disorders, age-related diseases, and pathologies associated with metaflammation (Park et al., 2012; Ma et al., 2020). In animal studies, it has been shown that reducing calorie intake increases life span and helps to lower blood glucose and insulin levels (Redman and Ravussin, 2011; Kim et al., 2020). There is currently evidence that CR for 2 years by 15% in healthy, non-obese people leads to a decrease in systemic oxidative damage (Redman et al., 2018). The main effects of CR in mammals include weight loss, improved insulin signaling by increasing hormone receptor sensitivity, normal lipid profiles, and increased adiponectin levels (Balasubramanian et al., 2017).

Thus, research data demonstrate the beneficial effects of CR in conditions such as diabetes, inflammation, obesity, and cardiovascular disease (Redman et al., 2018). However, the mechanisms underlying such changes remain unclear. An increase in adipose tissue is associated with the development of age-related metabolic changes, including the development of IR. In contrast, a decrease in adipose tissue during prolonged CR led to an improvement in age-related IR (Escrivá et al., 2007; Fabbiano et al., 2016; Corrales et al., 2019). CR slows down and restores age-related immunosenescence by regulating energy metabolism and oxidative stress and decreasing the production of proinflammatory cytokines and neuroendocrine homeostasis (Costantini et al., 2018).