Insulin is a peptide hormone composed of 51 amino acids, generated from the precursor proinsulin through a series of processing (Rahman et al., 2021). Primarily produced by β cells of the pancreatic islets, insulin is the main anabolic hormone that regulates the energy metabolism throughout the body, i.e., promoting glucose uptake into the liver, fat, and muscle cells (White, 2003). In addition, acting through common receptors (insulin receptor [InsR] and IGF-1 receptor [IGF-1R]), insulin and insulin-like growth factors (IGFs) regulate proliferation and survival of various cell types throughout the body during development, in adulthood, and in aging processes (Nakae et al., 2001; Richardson et al., 2004). InsR and IGF-1R are tyrosine kinase receptors that can be present as homodimers (InsR/InsR, IGF-1R/IGF-1R) or heterodimer (InsR/IGF-1R), sharing the majority of downstream pathways, namely phosphoinositide 3-kinase (PI3K), serine threonine kinase Akt, glycogen synthase kinase 3β (GSK3β), and mammalian target of rapamycin (mTOR) (O’Neill et al., 2012). InsR and IGF-1R are highly expressed in the CNS, including the hippocampus and hypothalamus, and central insulin regulates peripheral energy metabolism (O’Neill et al., 2012; Tiedemann et al., 2017; Barrios et al., 2021; Scherer et al., 2021). Furthermore, it is now evident that insulin and IGF-1 play important roles in higher-order brain functions, such as memory and cognitive processing, and in neuroprotection in response to brain injury (Stewart and Rotwein, 1996; Gerozissis, 2003; Fernandez and Torres-Aleman, 2012), indicating pleiotropic roles for insulin/IGF signaling in the CNS.

Insulin crosses the BBB using a saturable transporter, which is regulated by insulin itself and altered by a number of factors including hyperglycemia and diabetes (Banks et al., 2012). In addition, the choroid plexus, a highly vascularized tissue responsible for the production of cerebrospinal fluid (CSF) at the interface of the CNS and periphery, releases insulin and IGFs (Salehi et al., 2009; Ziegler et al., 2012; Mazucanti et al., 2019; Dani et al., 2021). In the parenchyma, microglia generate IGF-1 during development (Wlodarczyk et al., 2017) as well as in adulthood (Myhre et al., 2019), while astrocytes can produce insulin, which is negatively regulated by Aβ and bacterial lipopolysaccharide (LPS) (Takano et al., 2018). InsR and IGF-1R are expressed on neurons and non-neuronal cells (Shaughness et al., 2020), and are involved in the regulation of synaptic plasticity (Dyer et al., 2016), astroglial energy metabolism, microglial inflammatory phenotypes (Haas et al., 2020), as well as the self-renewal and maintenance of neural stem cells (Ziegler et al., 2015), indicating crucial roles of insulin through multiple targets and mechanisms in the brain parenchyma (Gabbouj et al., 2019). Although the hypothalamus is the most well-studied target region regulating systemic energy metabolism (Mitchell and Begg, 2021), specific inactivation of InsR and IGF-1R in the hippocampus and amygdala led to increased anxiety and cognitive impairment in animal models (Soto et al., 2019), indicating the significance of extrahypothalamic and non-metabolic roles for insulin. Moreover, hippocampal microglia responded to insulin treatment in young rats but not in the aged group (Haas et al., 2020), suggesting that microglia develop insulin resistance during aging. Indeed, experimental evidence suggests the link between impaired insulin signaling in the CNS and cognitive impairment. Intracerebral streptozotocin (STZ) induces AD-like brain pathology in mice, which has been widely used as a model for sporadic AD (Lester-Coll et al., 2006). In mixed AD and diabetic mice using APPswe/PS1dE9 (APP/PS1) transgenic AD model or triple-transgenic model of AD (3xTg-AD) with either STZ treatment or InsR deficiency (db/db), genetic background and dysfunctional insulin signaling cooperatively exacerbated CNS inflammation and AD pathology (Hierro-Bujalance et al., 2020; Imamura et al., 2020; Sankar et al., 2020). In addition, intranasal insulin treatment can alleviate AD pathogenesis and cognitive impairment via reduced neuroinflammation and enhanced neural plasticity (Chen Y. et al., 2014; Guo et al., 2017), possibly via the actions through receptors on microglia and astrocytes (Spielman et al., 2015). Although what roles insulin/IGF signaling play in the CNS could be context-dependent, i.e., either beneficial (Carro et al., 2003; Sukhanov et al., 2007; Tien et al., 2017) or damaging (Labandeira-Garcia et al., 2017), fine-tuning of these signaling pathways is crucial to maintaining brain functions. Considering the anti-inflammatory effects of insulin and IGF-1 on microglia (Labandeira-Garcia et al., 2017; Shaughness et al., 2020) and observations that age-related changes of microglia toward pro-inflammatory phenotypes are implicated in brain aging and neurodegenerative disorders (Gemma et al., 2010; Chowen and Garcia-Segura, 2020), microglial insulin/IGF resistance may underlie chronic inflammation in the brain, which is associated with dementia (Lutshumba et al., 2021). Consequently, dysregulated insulin/IGF signaling and insulin resistance in the CNS have been linked to an increased risk of dementia, including AD, which is sometimes referred to as type 3 diabetes (Biessels et al., 2006; Whitmer et al., 2008; Zemva and Schubert, 2011; Banks et al., 2012; Smolina et al., 2015; Biessels and Despa, 2018; Kuo et al., 2018; Shinjyo et al., 2020).

Acute and chronic infections induce insulin resistance (Yki-Järvinen et al., 1989; Fernández-Real et al., 2006), whereas metabolic imbalance (e.g., higher body fat mass and diabetes) is associated with higher infection susceptibility (Rayfield et al., 1982; Fernández-Real et al., 2007), indicating bidirectional interactions between the immune system and energy metabolism in human body. Chronic low-grade inflammation and immune dysregulation likely mediate the mutual and possibly synergetic relationship. The strong interaction between the immune and metabolic pathways is apparently rooted in their common evolutionary origin, as represented by the fat body in Drosophila, which senses both infectious and metabolic stresses and perform the functions of the liver, adipose tissue, and immune system (Hotamisligil, 2017). Insulin plays a key role in such evolutionarily conserved immunometabolism, partly through the interactions with tumor necrosis factor (TNF) receptor (Uysal et al., 1997) and Toll-like receptor (TLR) pathways (DiAngelo et al., 2009; Hotamisligil, 2017), as well as through modulating metabolic pathways in immune cells (van Niekerk et al., 2020). Through metabolic regulation, insulin alleviates the harmful effects of hyperglycemia (Sun et al., 2014). For example, due to its glucose-lowering effect, insulin exert anti-inflammatory effects through modulating the release of inflammatory mediators. As glucose is pro-inflammatory, insulin deficiency activates inflammatory reactions in the body, leading to the release of inflammatory mediators including reactive oxygen species and pro-inflammatory cytokines from leukocytes (Mohanty et al., 2000; Esposito et al., 2002). In addition, insulin directly activates phagocytic and bactericidal activity of immune cells and diabetes-induced infection susceptibility is partly mediated by impaired immune responses due to the lack of insulin signaling, as shown in rodent studies (Yano et al., 2012).

Leptin is a 16 kDa polypeptide that regulates metabolic balance and fat storage. Mainly produced by the white adipose tissue (WAT), leptin acts via leptin receptor (LepR) in the brain and plays a pivotal role in the control of appetite and energy expenditure. There are six LepRs (LepRa to LepRf), with identical extracellular N-terminal domain and distinct intracellular C-terminal regions generated by alternative splicing of db (Lee et al., 1996). LepRb, the long isoform with high ligand affinity, is the major isoform expressed in the brain and activates intracellular pathways, including JAK/STAT, ERK/MAPK, and IRS/PI3K (Allison and Myers, 2014). The hypothalamic nuclei, the regulatory center of energy homeostasis, are highly enriched with LepRb (Elmquist et al., 1998; Balthasar et al., 2004; Leshan et al., 2009), and leptin exerts its effects through the action in the hypothalamus to regulate food intake and energy metabolism (Friedman, 2019). Leptin enters the hypothalamus through fenestrated capillary and acts on LepR expressed on neurons in arcuate nucleus (ARC), dorsomedial hypothalamus (DMH), and ventromedial hypothalamus (VMH), enabling a feedback mechanism to maintain energy balance, thereby preventing obesity and metabolic disorders (Pandit et al., 2017). In addition, leptin can cross BBB using a saturable transport system (Banks and Farrell, 2003), as well as the blood-CSF barrier (choroid plexus epithelia) (Merino et al., 2006; Dietrich et al., 2008). LepRb in extrahypothalamic brain regions also plays significant roles in the regulation of energy metabolism (Scott et al., 2009). For example, LepRb in the ventral tegmental area (VTA) regulates energy balance via mesolimbic dopaminergic system (Fulton et al., 2006; Hommel et al., 2006), suggesting that leptin targets multiple brain regions and cellular components. Consequently, dysregulation of leptin signaling results in obesity, diabetes, and associated comorbidities (Myers et al., 2008; Wasim et al., 2016; Fischer et al., 2020). Leptin resistance is the condition where diminished leptin sensitivity occurs, resulting in a defect in satiety detection despite high leptin levels, which has been linked to obesity (Izquierdo et al., 2019). Leptin deficient (ob/ob) and LepR deficient (db/db) mice, carrying mutations in leptin (ob) and LepR (db) genes, respectively, exhibit excessive eating, develop obesity and diabetes, and are widely used as animal models of T2DM (Chen et al., 1996; Ninomiya et al., 2002; Gautron and Elmquist, 2011). Growing evidence suggests the contribution of leptin resistance to neurodegeneration in AD (Bonda et al., 2014). Serum leptin levels showed negative correlation with cognitive decline in the elderly (Holden et al., 2009). In the brain autopsy, CSF leptin levels were significantly higher in AD compared to control and mild cognitive impairment cases, and CSF leptin concentration was correlated with pathological neurofibrillary tangle burden (Bonda et al., 2014), suggesting that leptin resistance develops during AD progression.

Adipokines are the adipose tissue-derived factors that affect whole body homeostasis in autocrine and paracrine manners, targeting a number of biological processes such as glucose metabolism, lipid metabolism, insulin sensitivity, as well as immune response (Fasshauer and Blüher, 2015). Adipokines include leptin, adiponectin, vaspin, fibroblast growth factor 21 (FGF21), and many more, each exerting specific biological effects, and mediate diverse actions throughout the body (Fasshauer and Blüher, 2015). Originally identified as an adipocyte-derived hormone that regulates neuroendocrine axis, leptin is one of the most studied adipokines linking the immune system and energy metabolism (Abella et al., 2017; Jiménez-Cortegana et al., 2021).

Indeed, leptin belongs to the family of long-chain helical cytokines and has similarity to IL-6, IL-12, and granulocyte colony-stimulating factor (G-CSF). LepR is a type I cytokine receptor (La Cava and Matarese, 2004). Although neurons are the most well-established cellular targets of leptin, LepR is also expressed by non-neuronal cells throughout the body. Importantly, most immune cells, including hematopoietic bone-marrow precursors, monocytes/macrophages, and lymphocytes, express LepR, suggesting that leptin directly modulate immune responses and inflammation (La Cava and Matarese, 2004; Procaccini et al., 2012, 2017). Adipose tissue and lymphoid organs are often anatomically associated, and the contiguity between adipocytes and lymphoid cells supports the functional interactions (Matarese et al., 2002). For example, leptin affects thymic function and growth/survival of bone-marrow CD34⁺ precursors and CD4⁺ T cells (Lord et al., 1998; Howard et al., 1999; Martín-Romero and Sánchez-Margalet, 2001; Papathanassoglou et al., 2006; Cohen et al., 2017), and modulates both innate and adaptive immunity through diverse mechanisms (Maurya et al., 2018), such as enhancement of neutrophil oxidative burst (Mancuso et al., 2002; Caldefie-Chezet et al., 2003), phagocytosis by monocytes/macrophages (Faggioni et al., 1999; Fantuzzi and Faggioni, 2000; Sánchez-Margalet et al., 2003), and cytotoxic activity of natural killer (NK) cells (Tian et al., 2002), as well as mobilization of macrophages, lymphocytes (Abella et al., 2017), and neutrophils (Souza-Almeida et al., 2018).

Leptin deficiency and resistance are associated with increased susceptibility to infectious diseases, including bacterial (e.g., Mycobacterium tuberculosis and Streptococcus pneumonia), viral [e.g., coronaviruses, influenza A virus, and human immunodeficiency virus (HIV)], and parasitic (e.g., Trypanosoma brucei, Trypanosoma cruzi, and Entamoeba histolytica) infections (Amole et al., 1985; Sánchez-Pozo et al., 2003; Wieland et al., 2005; Hsu et al., 2007; Nagajyothi et al., 2010; Tschöp et al., 2010; Vedantama and Viswanathan, 2012; Zhang et al., 2013; Radigan et al., 2014; Alti et al., 2018; Guglielmi et al., 2021), while infection with certain pathogens, such as Plasmodium spp. and Toxoplasma gondii, can cause dysregulated leptin secretion, independently of adiposity (Pulido-Mendez et al., 2002; Baltaci and Mogulkoc, 2012). The significant association between obesity-induced inflammation (meta-inflammation) and the severity of infectious disease has been highlighted by COVID-19 pandemic (Huizinga et al., 2020; Rebello et al., 2020). It is notable that Streptococcus pneumoniae is a major cause of meningitis, which potentially leads to persistent cognitive disability (Yau et al., 2018), and M. tuberculosis can disseminate into the brain and induce CNS tuberculosis, a cascade of inflammatory responses that can potentially cause brain damage (Leonard, 2017; Davis et al., 2019). HIV-positive individuals frequently suffer neurocognitive disorders (HIV-associated neurocognitive disorders, HAND), and influenza A virus (H1N1) have been associated with neurological manifestation in both young and adult patients (Cárdenas et al., 2014; Wilking et al., 2014), with some developing permanent sequelae (Cárdenas et al., 2014), suggesting that defective leptin signaling could affect brain functions through increased infection susceptibility. While the mechanistic link between leptin resistance and infection susceptibility is complex and multifactorial (Maurya et al., 2018; Rebello et al., 2020), leptin’s actions through immune cells in the periphery and CNS, and dysregulation thereof, likely play a significant role.

The brain is the most energy-demanding organ, consuming glucose at a disproportionately high rate compared to the rest of the body. Glucose metabolism in the brain is founded on intricate interactions between the cellular components, i.e., neurons and glia. Neurons are highly aerobic and heavily dependent on OXPHOS, while astrocytes and oligodendrocytes, the major neuroglial populations, are predominantly glycolytic. These cell-type specific metabolic profiles support the overall power system in the CNS: lactate generation via glycolysis in neuroglia fueling OXPHOS in neurons (Fünfschilling et al., 2012; Jha and Morrison, 2018; Rosko et al., 2019) (Figure 2). To protect such elaborate systems, the brain harbors resident innate immune cells, namely resident macrophage populations: microglia, perivascular macrophages, meningeal macrophages, and choroid plexus macrophages. The former three, including microglia, have embryonic origins and are maintained through local self-renewal while choroid plexus macrophages are replenished via constant supply of bone marrow-derived monocytes (Prinz et al., 2017). These brain-resident macrophage populations play significant roles in the brain physiology, as well as in defense against infection (Ransohoff and Brown, 2012; Heneka et al., 2015; Herz et al., 2017; Kierdorf and Prinz, 2017).

While aberrant activation of resident macrophages contributes to CNS pathology via inflammation (De Strooper and Karran, 2016; Herz et al., 2017), evidence is growing that brain-resident macrophages play homeostatic roles in the brain. Residing in the brain parenchyma, microglia play pivotal roles in the development and maintenance of neuronal networks. The most notable roles of microglia include complement-dependent synaptic pruning during CNS development (Schafer et al., 2012) and synaptic reorganization throughout life (Herz et al., 2017). In addition, microglia-derived neurotropic factors, such as brain-derived neurotrophic factor (BDNF), promote learning-dependent synaptic formation (Parkhurst et al., 2013) and protect neurons from brain injuries (Madinier et al., 2009). Furthermore, CNS macrophages are the major component of glymphatic system, which plays a crucial role in the clearance of waste products and toxic materials, including Aβ (Gordleeva et al., 2020). In the context of AD pathogenesis, microglial activation is a double-edged sword; it can facilitate the clearance of Aβ and tau, while potentially inducing neuroinflammation leading to neuronal damages associated with AD (Lue et al., 2010; Leyns and Holtzman, 2017; Shippy and Ulland, 2020).

Meanwhile, growing evidence suggests that astrocytes are part of the innate immunity in the brain (Sofroniew, 2020). Responding to the changing microenvironment and diverse stimuli (e.g., microbial infections, exposure to environmental toxins, tumor formation, and neurodegenerative diseases), astrocytes undergo morphological and functional changes and influence the outcome of a number of brain disorders (Drögemüller et al., 2008; Soung and Klein, 2018; Katsouri et al., 2020; Sofroniew, 2020). In addition, crosstalk between astrocytes and microglia in the hippocampus is responsible for sensitivity to insults (Lana et al., 2020), pointing to the importance of communication between astroglia and brain-resident macrophages in cognitive dysfunctions (Liddelow et al., 2017; Linnerbauer et al., 2020). Furthermore, astrocytes and microglia participate in Aβ clearance and protection of synaptic connectivity in AD models, suggesting highly context-dependent immunological roles of these glial cell populations against AD development (Davis et al., 2021).

Microglia and astrocytes, also called macroglia, are major resident non-neuronal cells in the brain. As described above, evidence suggests that the interplay between neurons and these glial population plays pivotal roles in the brain, including the hippocampus, in physiological and pathological conditions (Lana et al., 2021). While neuronal cells express InsR/IGF-1R at high levels and many studies have focused on the roles of neuronal InsR/IGF-1R (Frölich et al., 1998), microglia and astrocytes also express InsR and IGF-1R (Shaughness et al., 2020). Insulin and IGFs support brain functions through regulating astroglial glucose metabolism (Sonnewald et al., 1996; Fernandez et al., 2018). InsR signaling modulates astroglial glucose uptake and bioenergetics (Heni et al., 2011), and specific ablation of astroglial InsR led to altered mood and cognition in mice (González-García et al., 2021). In addition, insulin can modulate inflammatory responses of astrocytes (Spielman et al., 2015), suggesting the immunometabolic regulation of astrocytes by insulin. IGF-1R can also modulate astroglial metabolic and immunological signatures (Hernandez-Garzón et al., 2016), and astroglial IGF-1R signaling mediates synaptic plasticity of cortical inhibitory neurons (Noriega-Prieto et al., 2021). These data suggest that impairment in InsR/IGF-1R signaling pathways may disrupt the metabolic network between glia and neurons, as well as immunological roles of astrocytes, eventually leading CNS dysfunctions. Microglial InsR/IGF-1R also mediate the effects of insulin. Low-dose insulin exerted pro-inflammatory effects on microglial cells in vitro (Spielman et al., 2015), while it was anti-inflammatory at a higher concentration (Brabazon et al., 2018), suggesting the variable roles of microglial InsR/IGF-1R signaling. In addition, microglia are the major source of IGF-1 in the brain, and microglial IGF-1 was found to be increased in AD model mice compared to wild-type (Myhre et al., 2019). These data suggest that insulin and IGFs play crucial roles in immunometabolism via the phenotypic regulation of glial cells, potentially mediating the link between bioenergetics and immunity in the brain.

Microglia (Tang et al., 2007; André et al., 2017; Fujita and Yamashita, 2019), and astrocytes (Naranjo et al., 2020; Pratap and Holsinger, 2020) express LepR. Leptin can enhance microglial pro-inflammatory responses, including IL-6 production through a mechanism involving insulin receptor substrate-1 (IRS-1), PI3K, and Akt (Tang et al., 2007; André et al., 2017; Fujita and Yamashita, 2019), IL-1β release via a caspase 1-independent mechanism (Pinteaux et al., 2007), and lipopolysaccharide (LPS)-induced pro-inflammatory responses (Lafrance et al., 2010). On the other hand, in a spinal cord injury model, leptin reduced microglial inflammatory responses, while inducing neuroprotective phenotypes, Fernández-Martos et al. (2012). In myeloid cell-specific LepR deficient mice, hypothalamic microglia exhibited less ramified morphology and impaired phagocytic capacity (Gao et al., 2018), suggesting that leptin directly regulates homeostatic microglial phenotypes (Davis et al., 2017). Considering multifaceted physiological roles of microglia and potential harm (i.e., neuroinflammation) caused by their aberrant activation (Colonna and Butovsky, 2017; Li and Barres, 2018), it is conceivable that impaired leptin signaling in microglia could lead to significant homeostatic imbalance in the CNS. On the other hand, specific LepR depletion in astrocytes (GFAP-LepR–/–) resulted in impaired neurotransmission in the hippocampus, suggesting that leptin regulates hippocampal plasticity via astroglial LepR, possibly by regulating the glucose and glutamate up-take capacity (Naranjo et al., 2020).

In addition to resident macrophages, the CNS harbors multiple leukocyte populations, including dendritic cells (DCs) and T lymphocytes. As a professional antigen-presenting cells, DCs play an essential role in the regulation of adaptive immunity. Although initially believed to be absent in the healthy brain parenchyma, only appearing in response to aging, injury, and infections (Fischer and Reichmann, 2001), studies have shown the presence of DCs within the healthy steady-state brain (Bulloch et al., 2008; Prodinger et al., 2011). Upon activation, brain DCs can migrate out of the brain and induce T-cell homing into the CNS where antigen-specific immune responses may take place (Karman et al., 2004). Correlation between aging and the accumulation of DCs in the CNS (Kaunzner et al., 2012) suggests potential involvement of brain DCs in immunological changes and inflammation associated with aging. In response to immunological stimuli, such as viral and parasitic infections, T cells infiltrate into the brain, where these cells play beneficial roles by eliminating pathogens. In addition, studies have shown that long-lived memory T cells are established after CNS infections (Wakim et al., 2012; Landrith et al., 2017; Mockus et al., 2018), and these tissue-resident memory CD8 cells can provide frontline defense against re-infection (Netherby-Winslow et al., 2021). However, persistence of these cells can also cause neuronal damages (Ai and Klein, 2020; Ghazanfari et al., 2021).

Although it is unknown how these CNS resident immune cells respond to and regulated by leptin and insulin, InsR is expressed on the surface of activated T cells (Helderman et al., 1978). Acting through InsR, insulin enhances and maintains T cell functions after immunological challenges (Helderman, 1984) and potentially induces anti-inflammatory polarization in the periphery (Viardot et al., 2007). These data suggest that insulin may play an immunomodulatory role via actions on T cell populations in the brain.

Growing evidence supports the infectious etiology of dementia, and a number of causative agents have been proposed, including viruses (e.g., Herpes simplex virus type 1 (HSV-1), cytomegalovirus (CMV)] and bacteria (e.g., Chlamydophila pneumoniae, spirochetes, Helicobacter pylori, Porphyromonas gingivalis) (Harris and Harris, 2015; Sochocka et al., 2017; Nazareth et al., 2021). Suggested mechanisms include persistent inflammation caused by chronic infection, leading to a vicious cycle of neuroinflammation and neurodegeneration, as well as BBB disruption, which may further enhance the entry of infectious agents and proinflammatory mediators from the periphery (Sochocka et al., 2017) (Figure 3).

Some infectious agents can enter and persistently inhabit the brain, directly causing chronic activation of resident immune cells (Sochocka et al., 2017). For example, Chlamydia pneumoniae, obligate intracellular bacteria, are frequently found in microglia and astrocytes in the brain of AD patients (Balin et al., 1998; Nazareth et al., 2021). Microglia and astrocytes are also the target of some viruses, such as CMV (Cheeran et al., 2001) and HIV (Avalos et al., 2017; Valdebenito et al., 2021). In addition, T. gondii, a neurotropic protozoan parasite causing a lifelong CNS infection, has been implicated in various neurological disorders, including dementia (Nayeri Chegeni et al., 2019). T. gondii is an obligate intracellular parasite, potentially infecting all nucleated cell types. Chronic Toxoplasma infection is linked to microglial activation and persistent neuroinflammation (Li et al., 2019), and CNS energy metabolism is significantly skewed toward glycolysis upon infection (Hargrave et al., 2019), suggesting that T. gondii interferes with host immunometabolism in the brain. Indeed, it has been shown that Toxoplasma infection may lead to reduced IGF-1 signaling in the brain (El Saftawy et al., 2020). On the other hand, In vitro, high glucose and up-regulation of glycolysis in the host cell had a significant impact on the life-stage conversion of T. gondii (i.e., from fast-growing tachyzoite stage into dormant bradyzoite stage) (Weilhammer et al., 2012), indicating that host cell metabolism could modulate the virulence of intracellular pathogens. Intriguingly, T. gondii infection can enhance Aβ clearance by increasing phagocytic activity of recruited monocytes (Möhle et al., 2016), suggesting that the ultimate outcome may depend on the extent and the aggressiveness of the infection. Of note, SARS-CoV-2 is frequently found in astroglia and microglia, rather than neurons, in the autopsy brain, indicating potential involvement of glial activation in the neurological manifestation of COVID-19 (Solomon, 2021). These data suggest that infectious agents could produce immunometabolic imbalance in the CNS via affecting microglial and astroglial phenotypes (Figure 2), which may lead to chronic inflammation and an increased risk of neurodegenerative disorders (Stephenson et al., 2018).

West Nile virus (WNV) is a single-stranded RNA virus, genetically related to the Japanese encephalitis virus (JEV). Transmitted by mosquitoes, WNV potentially causes life-threatening encephalitis or meningitis, especially in the elderly (Alli et al., 2021). In addition, it can cause persistent cognitive impairments (Murray et al., 2014; Vittor et al., 2020). While CNS infiltration of lymphocytes, including CD8⁺ T cells, is essential in eliminating viruses, persistent inflammation can cause synaptic loss and neuronal death, impairing cognitive ability (Garber et al., 2019; Vittor et al., 2020). Of note, diabetes is a frequent comorbidity of severe WNV diseases and considered a risk factor for developing WNV encephalitis (Badawi et al., 2018). In db/db obesity model, WNV infection-induced leukocyte infiltration into the brain was significantly lower, suggesting that obesity compromises protection against viral infection in the brain (Kumar et al., 2014). In particular, infiltration of CD8⁺ T cells was significantly reduced in obese mice, which was associated with higher viral load and enhanced inflammatory responses in the brain (Kumar et al., 2014), suggesting a role of leptin signaling in protecting the brain via leukocyte recruitment (Rummel et al., 2010).

Porphyromonas gingivalis is a periodontal disease-causing Gram-negative bacteria found in the oral cavity. Evidence suggests a strong association between P. gingivalis infection and sporadic AD (Kanagasingam et al., 2020), and its presence has been identified in the brain of AD patients (Dominy et al., 2019). P. gingivalis infection can trigger inflammation both in the periphery and CNS in affected individuals, leading to cognitive decline (Dominy et al., 2019; Kanagasingam et al., 2020). In mice, oral P. gingivalis infection led to brain colonization and AD-like pathogenesis, including complement activation and Aβ_1–42 formation (Poole et al., 2015; Dominy et al., 2019), suggesting potential mechanistic links between periodontal disease and AD. A systematic review of pre-clinical studies also found that P. gingivalis infection induced inflammatory responses and tissue degeneration in the brain, which were associated with cognitive impairment (Costa et al., 2021).

Periodontitis is also associated with obesity (Suvan et al., 2011; Keller et al., 2015; Nascimento et al., 2015; Khan et al., 2018). High fat diet (HFD) significantly enhanced systemic inflammation induced by periodontal pathogens in rodents (Virto et al., 2018). P. gingivalis infection up-regulated the expression of genes associated with insulin/IGF-1 signaling and induced inflammatory responses in the brain of db/db mice (Virto et al., 2018), suggesting that metabolic imbalance due to dysregulated insulin and leptin signaling pathways may exacerbate the outcome of periodontitis and associated inflammation in the brain.

Despite the effectiveness of antiretroviral therapy in saving the lives of many from acquired immunodeficiency syndrome (AIDS), HIV-positive individuals frequently suffer neurocognitive disorders (HAND) (Antinori et al., 2007; Foley et al., 2011). While inflammation and brain atrophy due to persistent viral presence in the CNS may partly provide explanations (Pemberton and Brew, 2001; Woods et al., 2010; Ances et al., 2012; Brew and Barnes, 2019), exact mechanisms underlying HAND remain unclear.

Metabolic syndromes, including diabetes, is prevalent in HIV-infected individuals (Calza et al., 2011; Paik and Kotler, 2011), and insulin resistance was associated with lower cognitive scores in a HIV-1 cohort (Valcour et al., 2005, 2006), suggesting that HIV infection may increase the risk of cognitive impairment via metabolic dysregulation. In fact, in a murine model of HAND, intranasal insulin administration restored hippocampal dendritic integrity and cognitive functions, independently of HIV burden in the brain (Kim et al., 2019). In primary human neurons and microglia in vitro, insulin suppressed infection-induced inflammatory responses and HIV-1 growth in microglia, and prevented infection-induced neuronal death (Mamik et al., 2016). In addition, in a feline HIV model in vivo, intranasal insulin enhanced the preservation of cortical neurons and improved cognitive performance (Mamik et al., 2016), suggesting that impaired insulin signaling in the CNS may underlie cognitive impairment in HIV positive individuals.