Impairment of vascular pathways of cerebral β-amyloid (Aβ) elimination contributes to Alzheimer disease (AD). Vascular damage is commonly associated with diabetes. Here we show in human tissues and AD-model rats that bloodborne islet amyloid polypeptide (amylin) secreted from the pancreas perturbs cerebral Aβ clearance. Blood amylin concentrations are higher in AD than in cognitively unaffected persons. Amyloid-forming amylin accumulates in circulating monocytes and co-deposits with Aβ within the brain microvasculature, possibly involving inflammation. In rats, pancreatic expression of amyloid-forming human amylin indeed induces cerebrovascular inflammation and amylin-Aβ co-deposits. LRP1-mediated Aβ transport across the blood-brain barrier and Aβ clearance through interstitial fluid drainage along vascular walls are impaired, as indicated by Aβ deposition in perivascular spaces. At the molecular level, cerebrovascular amylin deposits alter immune and hypoxia-related brain gene expression. These converging data from humans and laboratory animals suggest that altering bloodborne amylin could potentially reduce cerebrovascular amylin deposits and Aβ pathology.

No abstract

Hypersecretion of amylin, a β-cell hormone that regulates satiation, is common in individuals with prediabetes and is associated with pancreatic amyloid deposition and type-2 diabetes. Evidence has emerged that increased circulating levels of amyloid-forming human amylin may potentially impair brain function. Because mouse amylin is non-amyloidogenic, we generated transgenic mice with conditional pancreatic expression of amyloid-forming human amylin to study how in vivo knockdown of human amylin expression influences brain function during the development of type-2 diabetes. Males and females were fed a high-fat diet starting at 3 months of age to induce amylin hypersecretion and glucose dysregulation. Males developed hyperglycemia at 5 months of age, whereas females showed glucose dysregulation 3-4 months later. At 5 months of age, human amylin-expressing male mice were randomly assigned to either amylin downregulation group (by peritoneal tamoxifen injection) or control group (maintained amylin expression) (n = 10/group). Two months later, we assessed brain function with the novel object recognition test and performed comparative non-targeted metabolomics and global RNA-seq analyses of hippocampal tissue. Mice with downregulated human amylin show enhanced recognition memory index (p < 0.001) and lower blood glucose levels (p < 0.001) compared to those that continued to express human amylin. This was associated with increased hippocampal levels of glycolysis metabolites, including lactic acid (p < 0.01), glucose-6-phosphate (p = 0.06), and fructose (p = 0.07). Hippocampal gene-expression patterns between the two mouse groups revealed extensive compensatory changes in gene expression related to glucose metabolism. In conclusion, amylin downregulation in diabetic mice improves systemic glucose homeostasis and memory. Molecular processes associated with improved memory involve increased hippocampal glycolytic fluxes and compensatory gene expression.

Emergent large vessel occlusions result in severe ischemic stroke without appropriate treatment with thrombolysis and/or mechanical thrombectomy. Type-2 diabetes mellitus (T2DM) is a major risk factor in stroke, with 25% of ischemic attacks occurring in individuals with T2DM. T2DM diagnosis is also associated with poorer functional outcomes, prolonged hospitalizations, and increased risk of recurrent stroke. Amylin, a peptide co-secreted with insulin in pancreatic β-cells, is hypersecreted in T2DM and readily forms neurotoxic oligomers which deposit in brain parenchyma. Due to amylin’s role in T2DM and T2DM’s relationship to stroke, we anticipated an increased level of amylin would be deposited on red blood cells (RBCs) of stroke patients when compared to non-stroke patients. Additionally, we anticipated an increased level of amylin immunoreactivity (AIR) in clot lysates when compared to RBC lysates and plasma. Blood samples and thrombi ( n =47) were collected from patients undergoing mechanical thrombectomies for stroke while blood samples ( n =21) were collected from patients with non-stroke neurological conditions. Samples were lysed and assayed for total protein concentration and intensity of AIR. Amylin uptake coefficients (AUCs) demonstrating the proportionality of amylin deposited on RBCs compared to total circulating amylin were calculated. After normalizing to total protein concentration, analysis revealed a significantly increased level of AIR in stroke clots when compared to stroke and non-stroke plasma and RBC lysates (p<0.001 for each). Additionally, a significant increase (p<0.0073) in AUC was found in stroke versus non-stroke. In summary, amylin accumulates in thrombi and deposits on RBCs of stroke patients. Further research into amylin’s potential role in thrombus formation is justified. Future studies are also needed to determine if stroke severity is associated with amylin level in thrombi and if T2DM exacerbates amylin-stroke pathology.

BackgroundAmylin, a 37‐residue amyloidogenic peptide, is co‐secreted with insulin by pancreatic beta‐cells (Verma et al. 2020). Vascular deposition of amylin has been noted in patients with diabetes and neurocognitive disorders (Ly et al., 2017) Emergent large vessel occlusions (ELVOs) result in severe ischemic strokes without appropriate treatment. Growth factors, including erythropoietin (EPO), are involved in post‐stroke recovery as a potential neuroprotectant target, inhibiting apoptosis and decreasing inflammation (Jerndal et al., 2010). Here, amylin accumulation in plasma, red blood cells (RBCs), and ELVOs (where available) and plasma EPO levels were compared between stroke and control patients. Due to amylin’s amyloidogenic propensity, we hypothesized that there would be an association between amylin uptake and stroke incidence.MethodA prospective registry (Blood and Clot Thrombectomy Registry and Collaboration; BACTRAC) was developed to analyze blood and thrombus directly in patients presenting with an ELVO. Total protein concentration and amylin concentrations of clot lysates, plasma, and RBC lysates were obtained via bicinchoninic acid assay (BCA) and competitive enzyme‐linked immunosorbent assay (ELISA), respectively. Plasma EPO concentrations were obtained via ELISA. An amylin uptake coefficient (the level of amylin in RBC lysates divided by the sum of amylin present in RBC lysates and plasma) was calculated to determine the degree to which circulating amylin is depositing on RBCs.ResultDue to a significant difference between control and stroke RBCs (p = 0.0012, Figure 1) in total protein concentration, all downstream analyses were normalized to respective protein concentrations. A significant difference in amylin uptake by RBCs between stroke and control patients is shown in Figure 2 (p = 0.0073). An uptake coefficient greater than 0.50 indicates uptake of amylin by RBCs. Finally, a significant difference in plasma EPO concentration was observed between stroke and control (p = 0.0017; Figure 3).ConclusionConsistent with our hypotheses, the present data indicates that amylin uptake by RBCs is significantly increased in stroke patients than in control patients. However, both groups show evidence of accumulation, as indicated by uptake coefficients greater than 0.50. Further, levels of erythropoietin are significantly increased in stroke patients. Further investigation into whether amylin may be thrombogenic is warranted.