Complex mechanisms linking neurocognitive dysfunction to insulin resistance and other metabolic dysfunction


Scientific evidence has established several links between metabolic and neurocognitive dysfunction, and epidemiologic evidence has revealed an increased risk of Alzheimer’s disease and vascular dementia in patients with diabetes. In July 2015, the National Institute of Diabetes, Digestive, and Kidney Diseases gathered experts from multiple clinical and scientific disciplines, in a workshop entitled “The Intersection of Metabolic and Neurocognitive Dysfunction”, to clarify the state-of-the-science on the mechanisms linking metabolic dysfunction, and insulin resistance and diabetes in particular, to neurocognitive impairment and dementia. This perspective is intended to serve as a summary of the opinions expressed at this meeting, which focused on identifying gaps and opportunities to advance research in this emerging area with important public health relevance.


Emerging data have established links between systemic metabolic dysfunction, such as diabetes, and neurocognitive impairment, including dementia. The current epidemic of dementia is driven, at least in part, by the concurrent epidemics of obesity, insulin resistance, diabetes, and metabolic syndrome. Early research sought to elucidate the cause(s) for the apparent role of metabolic dysfunction in the increased prevalence of neurocognitive dysfunction and dementia, tentatively attributed to vascular contributions to cognitive impairment and dementia (VCID), and Alzheimer’s disease (AD).
More recent research has revealed that the relationships linking metabolism and brain dysfunction are bidirectional. On the one hand, diabetes increases risk of dementia by about two-fold, while on the other hand, the development of neurocognitive disorders has been linked to an increased risk of metabolic disease (Biessels et al., 2014). While the relationships linking metabolism and brain dysfunction in humans are more complex and more difficult to study in isolation, mouse models of isolated cerebral amyloidosis and cerebral amyloid angiopathy have been discovered to have peripheral insulin resistance (Ruiz et al., 2016).
Genetic, epigenetic, environmental, and other mechanisms (likely in combination) are being implicated as underpinning these relationships and, in some cases, with behavior acting as a disease modifier. Of particular interest, both peripheral insulin resistance (the central pathophysiologic feature of type 2 diabetes) and “brain insulin resistance” (a much-discussed but poorly characterized entity) have been suggested to play important roles in neurocognitive dysfunction and dementia. Insulin resistance and the associated metabolic dysregulation can be driven by high-fat (“Western”) diet, physical inactivity, and obesity, and have deleterious effects on neurocognition (Ishii & Iadecola, 2015a).
Because the pathophysiological processes that lead to metabolic and neurocognitive dysfunction precede the development of clinical syndromes by years, early intervention and preventive measures are possible and an area of intense study (Norton et al., 2014). Indeed, AD neuropathology is detectable more than 10 years before the onset of the first clinical symptoms (Bateman et al., 2012) and pre-diabetes also precedes diabetes by years in most cases. There is now a focus on biomarkers and intervention research at pre-disease states, as there may be a higher chance that diseases are modifiable at this stage and prevention may be attainable (Sperling et al., 2015).
Confounding the picture is evidence that diabetes worsens VCID and, in turn, that VCID increase the likelihood of developing AD pathology. Nonetheless, diabetes and dementia are complex diseases, with heterogeneous risk factors, underlying mechanisms, and clinical expressions, often with different contributing co-morbidities.
In this era of precision medicine, it is imperative that we identify the various metabolic-neurocognitive phenotypes in order to understand the mechanisms that drive these diseases so that we can develop targeted therapeutic strategies to successfully manage and, hopefully, prevent these complex, multifactorial diseases.
In July 2015, the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) at the National Institutes of Health (NIH) gathered a group of cross-disciplinary scientific experts in Bethesda, MD for purposes of:
  • (1) Reviewing the state-of-the-science;
  • (2) Identifying key knowledge gaps and critical unanswered questions; and
  • (3) Developing ideas for future directions to advance investigation at the intersection of metabolic and neurocognitive dysfunction.
The meeting focused on clarifying leading candidate mechanisms that may explain relationships between metabolic and neurocognitive dysfunction, with a special emphasis on the bidirectional relationships between diabetes and/or insulin resistance (either peripheral, central, or both) and cognitive impairment and AD and/or VCID. Although there has been increased attention to the neurodegenerative and/or vascular contributions to cognitive impairment and dementia (Snyder et al., 2015), it is important to consider the contributions of mechanisms that may not fall into either category as well as mechanisms that span multiple categories.
In this White Paper, we have attempted to survey and summarize the opinions among the experts in attendance with regard to the most likely mechanisms to investigate in order to improve our understanding of diseases and disorders at the interface of metabolism and cognition; in other words, to address the questions of “what do we know now?” and “what directions should be pursued in order to advance that knowledge?”. In so doing, we aim (a) to improve public health at the level of community medical care that considers patients within the context of their behaviors and environment; and (b) to inform public policy that would be required for a major effort toward prevention of these common co-morbid conditions of high societal burden.

Summary of current knowledge

Diabetes and neurocognitive dysfunction (including dementia)

The high prevalence of type 2 diabetes (~26%) and dementia (~11%) in individuals > 65 years old, and the increased relative risk of all-cause dementia (~2 to ~2.5-fold) conferred by type 2 diabetes make understanding this connection a major public health imperative. Even individuals with type 2 diabetes without dementia have subtle impairment in performance (a third of a standard deviation, compared to individuals without diabetes) in a range of cognitive functions, including learning and memory, processing speed, and executive function across the lifespan (Biessels et al., 2014). These decrements in cognitive function are accompanied by structural and functional brain changes, with one estimate attributing five years of brain aging (based on volumetric magnetic resonance imaging, MRI) to diabetes relative to chronological age and estimates in those without diabetes (Franke et al., 2013). The most common neuroimaging findings include modest white and gray matter atrophy (most pronounced in the temporal and frontal cortices), cerebral small vessel disease, and disrupted structural and functional connectivity (Biessels & Reagan, 2015). Of relevance to understanding pathophysiologic mechanisms linking diabetes and neurocognition is that some of these brain changes may be observed prior to the development of diabetes (Brundel et al., 2014Convit et al., 2003).
The neurocognitive profile of diabetes shares features with aspects of cerebrovascular disease and AD. However, it is still unclear if the relationship between diabetes and neurocognitive dysfunction and dementia is due to classic AD pathology, vascular processes such as infarcts or vessel pathology, other pathologies less closely related to AD or vascular disease (e.g., impaired brain insulin signaling), or some combination of these. Some have proposed that “all-cause” dementia be employed in order to avoid the implication that we can classify etiologies accurately based on current knowledge. Understanding the mechanism(s) that explain the neurocognitive complications of diabetes and other metabolic disease will be important if we are going to be successful in developing therapeutic targets and approaches to mitigate the effects of metabolic disease on the brain and cognitive function. This is especially the case if metabolic disease impacts the brain in ways that differ from its effects on other end organs impacted by diabetes and other metabolic disease.
The CNS pathology associated with metabolic disease has largely been characterized as a cognitive disorder with most of the research focused on brain regions and networks that support higher-order cognitive functions. Research on the central regulation of energy balance and peripheral metabolism, on the other hand, has almost exclusively focused on the hypothalamus and brainstem, brain regions and networks whose contributions to higher-order cognitive functions may not have been fully defined. The recent appreciation of the bidirectional relationship between diabetes and neurocognitive dysfunction/dementia presents an opportunity to combine efforts and resources to tackle both of these complex and related problems together. Cerebral amyloidosis can precede clinical cognitive decline for up to 30 years, and during this early phase may occur in limited regions of the brain. Therefore, it is plausible that AD pathology (amyloid-beta or Aβ, tauopathy) may be present in the hypothalamus prior to the onset of cognitive symptoms, and that this hypothalamic pathology could disrupt homeostatic functions such as energy balance and peripheral metabolism prior to the onset of detectable cognitive dysfunction (Ishii & Iadecola, 2015b).

Mechanisms linking diabetes to dementia

While diabetes is known to increase risk for dementia, the underlying mechanisms linking these conditions is less clear. Undoubtedly, given the well-established micro- and macro-vascular complications of diabetes, the least disputed mechanism of brain injury involves cerebrovascular disease. Indeed, it is now well-established that diabetes, and elevation in the most frequently used biomarker of diabetes (hemoglobin A1c), increases the risk of stroke, and that stroke in turn, increases the risk of cognitive impairment and dementia, including vascular dementia (Abbott et al., 1987Banerjee et al., 2012Gorelick et al., 2011Ivan et al., 2004Li et al., 2012Ramirez et al., 2015). Aside from brain infarction, the central pathologic feature of stroke (Arvanitakis et al., 2006a), other vascular and related processes in the brain plausibly play a pathophysiologic role in relating diabetes to dementia, including white matter disease, breakdown of the blood brain barrier, inflammation, and others (Hsu et al., 2012Shimizu et al., 2013). Also, co-morbid conditions to diabetes, such as hypertension, hypercholesterolemia, and obesity, and complications of diabetes such as myocardial infarction, can lead to cerebrovascular disease and dementia, either alone or in concert with one another (Yaffe, 2007). The published literature as a whole suggests about a two-fold increased risk of stroke in individuals with diabetes, pointing to other pathways also being involved in linking diabetes to dementia.
While AD pathology is often cited to be present in about two-thirds of patients with dementia, this estimate includes both patients with AD pathology alone, as well as those with both AD pathology along with other pathologies such as the pathology of VCID and/or other neurodegenerative dementias (e.g., synucleinopathy, TDP-43, etc). Overall, the most common form of dementia is “mixed dementia”, and that mixture is most commonly one of both vascular and neurodegenerative pathology (Schneider et al., 2007). Links between diabetes and AD pathology are also reported but remain poorly understood. Some data from in vivo biomarker studies such as those using neuroimaging and cerebrospinal fluid, suggest that persons with diabetes have more brain atrophy, reduced glucose metabolism, and other changes in markers of neurodegeneration, including alterations in cerebrospinal fluid and phosphorylated tau in particular, that are in keeping with AD (Baker et al., 2011Moran et al., 2015Roberts et al., 2014). Mice with knockout of the insulin receptor in the brain show increased tau phosphorylation (Schubert et al., 2004). Data from postmortem brain tissue from elderly individuals have demonstrated the presence of some biochemical features reminiscent of insulin resistance in the hippocampal formation of persons with AD (with and without diabetes) compared to cognitively normal controls, and this insulin resistance has been reported to be associated with pathology that involves both Aβ oligomers and fibrillar amyloid plaques (Talbot et al., 2012). Consistent with this is the report that a mouse model that accumulates only Aβ oligomers showed metabolic flexibility that remained relatively intact while another mouse model that accumulated both Aβ oligomers and fibrils developed features of insulin resistance (Ruiz et al., 2016). A more recent postmortem study has shown that “brain insulin resistance” (a term in need of a precise definition) is associated with tau pathology in AD as well as other neurodegenerative tauopathies such as corticobasal degeneration (Yarchoan et al., 2014). Finally, some genetic data now suggest a link between diabetes and AD, as in the case of the insulin degrading enzyme gene or, more recently the SorCS1 gene which is thought to regulate Aβ metabolism (Lane et al., 2010Lane et al., 2013). Thus, amyloidosis, tauopathy, typical AD pathology and/or other neurodegenerative mechanisms -- as well as “brain insulin resistance” (perhaps even apart from peripheral diabetes per se) -- appear to play roles in linking diabetes to dementia, apart from the cerebrovascular mechanisms.
Nonetheless, neuropathological studies have largely failed to confirm a consistent association between diabetes and AD pathology. Indeed, only a few studies suggest associations with increased AD pathology, including in subsets of persons such as those who are apolipoprotein E epsilon 4 carriers (Malek-Ahmadi et al., 2013Matsuzaki et al., 2010Peila et al., 2002), but most studies show either no relationship or a relationship with decreased pathology that is modulated by antidiabetic therapies (Ahtiluoto et al., 2010Alafuzoff et al., 2009Arvanitakis et al., 2006aBeeri et al., 2005Heitner & Dickson, 1997Janson et al., 2004Nelson et al., 2009Sonnen et al., 2009). There are reports that the clinical cognitive syndrome in diabetes is more dysexecutive rather than amnestic (Arvanitakis et al., 2004Arvanitakis et al., 2006bNandipati et al., 2012), arguing against a strong diabetes-AD connection. In the largest clinicopathological study to date of about 2,400 persons, diabetes was not associated with AD pathology when assessed using overall measures of AD and more specific measures of severity of tangle and amyloid pathology (Abner et al., 2016). Whether some individual features of AD pathology (e.g., Aβ oligomer and/or fibril accumulation) play roles remains unclear. In summary, while an AD mechanism is less established (or perhaps less clinically important) than the vascular mechanism, further elucidation of this pathway may bring insight into other opportunities for treatment and prevention of neurocognitive dysfunction and dementia. Along these lines, both observational studies and experimental clinical trials have examined the role of conventional and unconventional (e.g., intranasal) administration of anti-diabetes medications for treatment of clinical and pathologic AD (Beeri et al., 2008Craft et al., 2012Gold et al., 2010Watson et al., 2005).
Chronic hyperglycemia has been associated with reductions in endothelial glucose transporters such as GLUT1 that could lead to glucose deprivation and excitotoxicity, and could account for some of the neurocognitive complications of diabetes (Gjedde & Crone, 1981Matthaei et al., 1986). This is unlike diabetic retinopathy, which has been associated with upregulation of GLUT1 glucose transporter (Kumagai et al., 1996) and is another example of how diabetic complications in the brain may differ from other organ systems.
Yet other mechanisms underlying the association of diabetes with neurocognitive dysfunction and dementia need to be further explored. While not a widely accepted concept, some researchers suggest that some AD may even be considered a metabolic disease caused by insulin resistance in the brain, separate from diabetes (Rivera et al., 2005), raising yet again the need for an evidence-based definition for “brain insulin resistance”. Given the range of effects of insulin in the brain, including neurogenesis, neurite outgrowth, and modulation of catecholamine release/uptake, other possible links between insulin resistance and dementia need to be considered. For examples, mice lacking insulin signaling in brain have been shown to have altered dopamine turnover associated with behavioral changes (Kleinridders et al., 2015). Downstream effects from insulin resistance, whether peripheral or central, could involve increased glucose and advanced glycation end products (AGEs), and alterations in adipokines (e.g., leptin) and other factors (Cherbuin et al., 2012Holden et al., 2009Yaffe et al., 2011). Inflammation-related events in the periphery and brain may be another important factor explaining the connection between diabetes and neurocognitive dysfunction (see (De Felice & Lourenco, 2015Ferreira et al., 2014) for recent reviews of this topic). Effects of insulin and other factors, including inflammation in the brain, need further consideration as plausible underlying mechanisms linking insulin resistance, with and without diabetes/pre-diabetes, to dementia.
Even in the absence of vascular and other complications, diabetes can alter synaptic plasticity in the mouse hippocampus resulting in cognitive deficits (Stranahan et al., 2008), and mice with diabetes are less likely to recover from stroke due to impaired neuroplastic mechanisms (Sweetnam et al., 2012). Ongoing studies in humans support the hypothesis that cognitive dysfunction in type 2 diabetes is related to alterations in the mechanisms of cortical brain plasticity (Fried et al., unpublished observations). The mechanisms for altered synaptic plasticity in diabetes are unclear but may be linked to reductions in efficacy of N-methyl-D-aspartate receptor (NMDAR)-dependent mechanisms of plasticity. Any downregulation in post-synaptic NMDARs to moderate this risk would consequently reduce the efficiency of long-term potentiation and alter any NMDAR-dependent measures. Alterations in the efficacy of the mechanisms of plasticity have also been identified as early pathophysiologic steps in Alzheimer’s disease and these may thus represent another link between metabolic disorders and dementia.

Pathophysiology unique to diabetes and its impact on neurocognition

Studies of the relationships linking diabetes, insulin resistance, and neurocognitive dysfunction have largely focused on how peripheral metabolism impacts brain function. Unlike other complications of diabetes (e.g., neuropathy, retinopathy, nephropathy), neurocognitive complications of diabetes have not clearly been demonstrated to correlate with measures of peripheral glycemia (with the exception of very poor glycemic control, HbA1c > 10% and a potential role for glycemic variability) and there is only limited evidence for a modest association with other measures of peripheral glucose regulation (e.g., insulin concentration, insulin action, insulin resistance) (Geijselaers et al., 2015). It is possible that lower levels of hyperglycemia (i.e., HbA1c < 10%) may have detectable effects on neurocognitive function in adequately powered studies or that HbA1c levels may not reflect highly fluctuating blood glucose concentrations, which could have independent deleterious effects on glucose-sensitive tissues such as brain endothelium. It is also possible that there are central mechanisms that will better account for the neurocognitive dysfunction observed in diabetes and other metabolic disease. One promising line of research has central insulin and insulin-like growth factor 1 (IGF-1) in the spotlight. There is evidence that disrupted central insulin and IGF-1 signaling may lead to disrupted neurotransmitter (e.g., dopamine) and astroglial cell function, brain endothelial cell function involved in formation and regulation of blood-brain barrier (BBB), mitochondrial metabolism and oxidative stress, regulation of the phosphorylation of microtubule-associated tau protein and clearance of Aβ and/or amyloid fibrils, cholesterol synthesis in the brain (important for myelination and membrane function), glucose and lipid metabolism in select regions of the brain, and regulation of central energy balance, which could relate to both metabolic and neurocognitive dysfunction (Bingham et al., 2002Brüning et al., 2000Convit et al., 2003Kleinridders et al., 2015Liu et al., 2013Montagne et al., 2015Schubert et al., 2004Stouffer et al., 2015Suzuki et al., 2010Suzuki et al., 2013). Restoring insulin function in the brain via intranasal insulin is now being tested as a potential therapeutic option for neurocognitive dysfunction in both diabetes and dementia separately (Claxton et al., 2015Novak et al., 2014) and repurposing diabetes drugs for use in dementia has also been proposed (Yarchoan & Arnold, 2014). However, more research in this area is critically needed. For example, it has been suggested that the reduced insulin signaling observed in aging, and exacerbated in AD, results from a neuroprotective mechanism (Steculorum et al., 2014). This view derives from evidence that decreased insulin signaling promotes life-extension in C. elegans (Kenyon et al., 1993) and that genetically induced insulin resistance has positive effects on cognition in an animal model of AD (Killick et al., 2009). If so, then intranasal insulin treatment may produce short term benefits but deleterious effects in the long term. The apparent paradox between pathological insulin resistance and the benefits of acute disruption of insulin signaling is a complex topic requiring further research attention.
The physiological neuroendocrine mechanism(s) that may participate in the intersection between diet, obesity, diabetes, and dementia is less clear; however, there are a few intriguing mechanisms to consider. Insulin modulates synaptic plasticity of dopamine neurons to regulate reward function and food intake behavior (Stouffer et al., 2015), as well as behaviors of anxiety and depression (Kleinridders et al., 2015) and hyperinsulinemia may lead to ineffective regulation of dopamine and increased food intake (Liu et al., 2013). It is also possible that hormonal changes that occur in obesity and diabetes, such as disrupted glucocorticoid function, could impact synaptic plasticity and brain function (Bocarsly et al., 2015). Impaired neuroplasticity itself, via a glucocorticoid-related or other mechanism(s), could account for metabolic and neurocognitive dysfunction (Oberman & Pascual-Leone, 2013Pascual-Leone et al., 2011Stranahan et al., 2008Wosiski-Kuhn et al., 2014). There is also evidence that corticotropin-releasing factor regulates cerebral amyloid accumulation along an as yet undefined pathway that is glucocorticoid-independent (Macauley et al., 2015).
Microglial dysfunction has become a central point for investigation in AD, since about one-third of genes associated with AD are expressed at high levels or exclusively by microglia. Glucocorticoids appear to impact microglia activation and the release of pro-inflammatory cytokines, which may result in neuroinflammation and cognitive dysfunction (Dey et al., 2014Erion et al., 2014Hao et al., 2016). Somewhat unexpectedly, the gut microbiome has been recently revealed as a potential modulator of physiological function of microglia, and research in other neurodegenerative diseases and Parkinson Disease in particular, have suggested a gut-brain mechanism of disease (Erny et al., 2015).

Mechanisms linking diet, obesity, and neurocognition

There has been even less appreciation for how diet and obesity may adversely impact brain function in the absence of diabetes or dementia. Obesity has been associated with altered brain structure and function in animal models and in metabolically and neurologically healthy adults and children (Bocarsly et al., 2015Hsu & Kanoski, 2014Yau et al., 2014). Compared to adults with normal body mass index (BMI = 18.5 – 24.9), overweight and obese adults in midlife (ages 40–59 years) have a relative risk (RR) of 1.35 (overweight) and 2.04 (obese) for AD and 1.26 (overweight) and 1.64 (obese) of any cause dementia (Anstey et al., 2011). This association between overweight and obesity is not observed in late life (ages 60 years or older) at a time when weight loss is more common and a risk factor for dementia (Anstey et al., 2011). Consumption of a high fat diet (HFD) can also negatively impact the brain and cognitive function well before obesity onset (Hsu & Kanoski, 2014). Indeed, obesity can impact cognition independently from metabolic disease and diet can impact metabolic function and cognition independently of obesity. Recent data suggest that weight reduction may even improve neurocognitive function (Horie et al., 2016). It will therefore be important to look beyond disease states to examine the influence of diet and obesity on neurocognitive function more generally.
Although obesity is occasionally associated with global measures of brain atrophy (Brooks et al., 2013Enzinger et al., 2005Raji et al., 2010) and cognitive decline, many studies suggest that executive function and learning and memory are most affected in both adults and children (Fitzpatrick et al., 2013Francis & Stevenson, 2013Gunstad et al., 2007Sabia et al., 2009). Corresponding with this neuropsychological profile, structural changes are observed in the parietal and prefrontal cortex (important for executive function), as well as the entorhinal cortex and hippocampus (important for learning/memory) (Enzinger et al., 2005Fotuhi et al., 2012Hao et al., 2016Miller & Spencer, 2014Pannacciulli et al., 2006Raji et al., 2010). These changes may be a consequence rather than a cause of obesity, as supported by neuroimaging data in the minipig animal model (Val-Laillet et al., 2011). Cerebral blood flow, a marker of neuronal activity, is significantly lower in the dorsolateral and anterior prefrontal cortex in minipigs with diet induced obesity compared to their lean counterparts. Likewise, in humans, there is reduced cortical thickness in the prefrontal, temporal and parietal cortex in adults but not in children (Sharkey et al., 2015), suggesting that structural changes occur after chronic HFD, metabolic disease and/or obesity. Teasing apart these three factors is an important avenue for future work.
HFD and obesity are also associated with dopamine-dependent mesocorticolimbic-prefrontal alterations that may impact reward learning, motivation, and executive functions (Johnson & Kenny, 2010Sevgi et al., 2015Stice et al., 2008Sun et al., 2015Vainik et al., 2013). Overweight/obese compared to healthy weight individuals show reduced change in striatal D2R binding potential in response to glucose ingestion (consistent with reduced dopamine release) and several studies have reported a negative association between basal metabolic index and the blood oxygen level dependent (BOLD) response to milkshake consumption in the dorsal striatum. Although BOLD does not directly measure dopamine release, the effect is dependent upon the Taq1a A1 polymorphism, which affects D2 receptor density, thus linking the BOLD response to abnormal D2R signaling which can also be linked to heightened impulsivity (Babbs et al., 2013). Along this line, there are emerging data that suggest specific macronutrients or combinations of macronutrients have selective effects on dopamine adaptations. One can speculate on the development of “vicious cycles” that lead to obesity, driven by both genetic makeup, dietary composition, and other factors.
Similar to the association between diabetes and AD, “brain insulin resistance” may play a key role. Insulin action directly affects central glucose metabolism and is posited to influence learning and memory by modulating synaptic plasticity, density, neurotransmission and adult neurogenesis. As such, intranasal administration improves cognition in healthy individuals (Benedict et al., 2004), raising the possibility that the reverse is also true; that is, decreased insulin signaling associated with sub-clinical insulin resistance accounts for the cognitive dysfunction observed following a HFD or obesity in otherwise healthy individuals. Supporting this hypothesis, a functional type of “brain insulin resistance” has been associated with hypo-active responses in obese non-diabetic, cognitively intact, people within fronto-parietal circuits, which are critical for working memory (the ability to hold and manipulate new and old information “on-line”). Working memory is markedly impaired in overweight and obese individuals and insulin sensitivity mediates a negative association between BMI and response in the fronto-parietal circuit during a working memory task (Gonzales et al., 2010). Diet-induced obesity can also influence insulin resistance in the hippocampus and likely elsewhere in the brain where insulin receptors are expressed, such as the amygdala and midbrain (McNay et al., 2010). Thus, metabolic alterations may impact brain function well before the onset of metabolic disorders, and weight reduction may improve both metabolic state and cognition (Horie et al., 2016).
Typical peripheral insulin resistance is often correlated with a number of other factors associated with diet-induced obesity that may also cause neurocognitive dysfunction. Obesity and HFD lead to systemic and central inflammation with elevated circulating IL-12 and IL-6, both of which have been linked to impaired processing speed and executive function, even independently of metabolic risk factors (Gregor & Hotamisligil, 2011Thaler & Schwartz, 2010Trollor et al., 2012). Thus, several groups have highlighted inflammation as “the pathway to cognitive impairment” (Miller & Spencer, 2014Steculorum et al., 2014). Hypersecretion of glucocorticoids may also contribute. The hypothalamic-pituitary-adrenal axis, which is altered in obesity, regulates glucocorticoid secretion, and has in turn been associated with mood changes, memory impairment, and reduced hippocampal volume (MacQueen & Frodl, 2011Raber, 1998).
Emerging work also suggests that diet-induced BBB disruption may provide a mechanistic link involving diet, obesity and hippocampal-dependent cognitive functions (Hsu & Kanoski, 2014). Rats fed a HFD for 90 days show decreased expression of BBB tight junction proteins and the presence of sodium fluorescin, a fluorescent-tagged molecule normally excluded from the brain by the BBB and choroid plexus, was found in the hippocampus but not in the prefrontal cortex or striatum (Kanoski et al., 2010). More recently, Davidson and colleagues found sodium fluorescin in the hippocampus of rats that failed to show a reversal of diet-induced impairments in hippocampal-dependent tasks upon HFD discontinuation, while rats demonstrating recovery of function did not (Davidson et al., 2012). Some evidence indicates that BBB integrity may influence learning and memory by affecting the passage of nutrients and neuroendocrine signals such as insulin, leptin and ghrelin from the periphery to the brain.

Comments

Popular posts from this blog

High prevalence of diabetes mellitus and impaired glucose tolerance in liver cancer patients: A hospital based study of 4610 patients with benign tumors or specific cancers

Nephron filtration rate and proximal tubular fluid reabsorption in the Akita mouse model of type I diabetes mellitus