| Type 2 diabetes(T2D)and Alzheimer’s disease(AD)are two common age-related diseases.As the global population ageing,their incidence and prevalence are increasing.To date,nearly 463 million people suffer from T2D and more than 50 million people are living with dementia worldwide(AD patients accounting for 50-60%).A large amount of epidemiological evidence has suggested a bidirectional association between T2D and AD.The risk of AD for T2D patients is 1.53-fold compared with individuals without T2D.Meanwhile,individuals with AD also exhibit greater impairments in glucose and insulin metabolism than those with normal cognitive function.The molecular mechanism underlying the relationship between T2D and AD remains unclear,although several shared pathophysiological features have been proposed for the two conditions,including insulin resistance,protein misfolding,inflammation,oxidative stress,and vascular dysfunction.Therefore,exploring the mechanism underlying the relationship between T2D and AD may provide new ideas for the prevention and therapy of the two diseases.β-amyloid(Aβ),mainly consisting of Aβ40 and Aβ42,is one of the characteristic bio markers for AD.In the brain of AD patients,Aβ is generated in large quantities and aggregates to form senile plaques,and the detection of the cerebrospinal fluid Aβ levels and angiography of Aβ deposition in the brain are usually used as important diagnostic methods for AD.Notably,more than 40%of Aβ in the brain can be transported into peripheral blood through several pathways.Meanwhile,considering the invasiveness of lumbar puncture and the expensive cost of Aβ imaging,plasma Aβ has attracted much attention as a potential biomarker for AD.A meta-analysis based on cohort studies has shown that AD patients had significantly higher plasma Aβ40 and Aβ42 levels than cognitively normal individuals.Furthermore,animal experiments demonstrated that increased plasma Aβ levels could induce peripheral insulin resistance.Researchers further found that active or passive immunization against plasma Aβ could improve insulin resistance in experimental animals.To date,several cross-sectional studies have examined the differences in plasma Aβ40 and Aβ42 concentrations between individuals with T2D and those without T2D but yielded controversial results,and no prospective study has been reported.Peripheral insulin resistance is one of the main pathogenesis of T2D.Early studies demonstrated that Aβ could induce neuronal insulin resistance through several mechanisms.Meanwhile,in vitro study found that Aβ25-35 could directly attenuate the inhibitory effect of insulin on hepatocyte gluconeogenesis,however,it is unclear whether Aβ40 and Aβ42 have the same effect.Except for inhibiting gluconeogenesis,promoting the uptake and utilization of glucose by cells is another important pathway for insulin to regulate blood glucose,and skeletal muscle cell is one of the main target cells.However,no study has explored the effects of Aβ40 and Aβ42 on the insulin sensitivity of skeletal muscle cells.Therefore,this study intends to conduct a large sample case-control study to explore the association of plasma Aβ40 and Aβ42 concentration with risk of impaired glucose regulation(IGR)and T2D;to further explore the association between plasma Aβ40 and Aβ42 concentration and risk of incident IGR and T2D with a prospective nested case-control design;to explore the effects of Aβ40 and Aβ42 on the insulin sensitivity of human liver cancer cell line HepG2 and mouse myoblast cell line C2C12 through in vitro studies,which could provide the mechanism evidence for epidemiological studies.The main content of this study is divided into the following three parts.Part 1.Association of plasma β-amyloid 40 and 42 concentration with risk of impaired glucose regulation and type 2 diabetesObjective:To explore the association of Aβ40 and Aβ42 concentration with risk of IGR and T2D.Methods:We conducted a case-control study including 571 newly-diagnosed IGR cases,1063 newly-diagnosed T2D cases,and 1063 controls matched by age(±3 years),sex,and apolipoprotein E ε4(APOE ε4)carrying status.All participants were recruited from population screening for diabetes at Tongji Medical College Hospital,Wuhan,China,between 2012 and 2015.Plasma Aβ40 and Aβ42 concentrations were simultaneously measured with electrochemiluminescence immunoassay(ECLIA).Multivariable conditional logistic regression was used to evaluate the association of plasma Aβ40 and Aβ42 concentrations with the likelihood of IGR and T2D.Adjustments were made for potential confounding factors,including age,sex,body mass index(BMI),APOE ε4 carrying status,current smoking status,current drinking status,physical activity status,family history of diabetes and history of hypertension.Results:The medians(interquartile ranges[IQRs])of plasma Aβ40 concentrations were 134.09(118.70-153.61)ng/L,134.45(117.99-154.58)ng/L,and 126.99(114.36-144.85)ng/L for IGR,T2D,and controls,respectively;the medians(IQRs)of plasma Aβ42 concentrations were 13.25(10.78-16.59)ng/L,13.25(11.04-16.14)ng/L,and 12.21(10.00-14.94)ng/L for IGR,T2D,and controls,respectively.Plasma Aβ40 and Aβ42 concentrations were significantly higher in IGR and T2D cases compared with the controls(P<0.001).The multivariable-adjusted odds ratios(ORs)and 95%confidence intervals(CIs)of IGR and T2D for the highest quartile of plasma Aβ40 concentrations compared with the lowest were 2.06(95%CI:1.38-3.06)and 1.96(95%CI:1.45-2.65),respectively.Each 30 ng/L increment of plasma Aβ40 was associated with 19%(95%CI:4%-37%)higher odds of IGR and 28%(95%CI:14%-43%)higher odds of T2D.The multivariable-adjusted ORs of IGR and T2D for the highest quartile of plasma Aβ40 concentrations compared with the lowest were 1.80(95%CI:1.25-2.60)and 2.01(95%CI:1.50-2.69),respectively.Each 5 ng/L increment of plasma Aβ42 was associated with 39%(95%CI:19&-62%igher odds of IGR and 37%(95%CI:21%-55%)higher odds of T2D.The positive associations of plasma Aβ40 and Aβ42 with IGR&T2D were consistent among almost all subgroups.The associations of plasma Aβ40 with IGR&T2D seemed to be stronger in subjects without physical activity(P for interaction=0.008).The association of plasma Aβ42 with IGR&T2D seemed to be stronger in subjects with age≤50(P for interaction<0.001),males(P for interaction=0.016),or subjects without physical activity(P for interaction=0.004).Subjects in the highest tertile of both plasma Aβ40 and Aβ42 concentrations had 3.38-fold greater odds of IGR and 2.96-fold greater odds of T2D compared with those in the lowest tertile of both plasma Aβ40 and Aβ42 concentrations.Conclusions:Our findings suggested significant and positive associations of plasma Aβ40 and Aβ42 concentration with risk of IGR and T2D.However,the associations of plasma Aβ40 and Aβ42 with T2D need further validation in prospective studies.Part 2.Prospective association of plasma β-amyloid 40 and 42 concentration with risk of incident impaired glucose regulation and type 2 diabetesObjective:To explore the prospective association of Aβ40 and Aβ42 concentration with risk of incident IGR and T2D.Methods:We conducted a nested case-control study based on the Tongji-Ezhou cohort,including 100 incident IGR cases,121 incident T2D cases,and 442 controls matched by age(±3 years)and sex.Baseline plasma Aβ40 and Aβ42 concentrations were simultaneously measured with ECLIA.Multivariable conditional logistic regression was used to evaluate the association of plasma Aβ40 and Aβ42 concentrations with the likelihood of incident IGR and T2D.Adjustments were made for potential confounding factors,including age,sex,BMI,current smoking status,current drinking status,physical activity status,family history of diabetes and history of hypertension.Results:The medians(IQRs)of plasma Aβ40 concentrations were 143.98(123.73-175.24)ng/L,142.91(122.18-177.23)ng/L,and 128.82(112.33-152.32)ng/L for IGR,T2D,and controls,respectively;the medians(IQRs)of plasma Aβ42 concentrations were 13.96(11.03-17.50)ng/L,13.92(11.29-17.86)ng/L,and 12.21(10.31-15.10)ng/L for IGR,T2D,and controls,respectively.Plasma Aβ40 and Aβ42 concentrations were significantly higher in IGR and T2D cases compared with the controls(P<0.001).After adjustment for several confounding factors,ORs of IGR and T2D for subjects in the highest quartile of plasma Aβ40 concentrations were 3.33(95%CI:1.56-7.13)and 3.79(95%CI:1.81-7.94),respectively,comparing to subjects in the lowest quartile of plasma Aβ40 concentrations.Each 30 ng/L increment of plasma Aβ40 was associated with 54%(95%CI:23%-92%)higher odds of IGR and 47%(95%CI:15%-88%)higher odds of T2D.ORs of IGR and T2D for subjects in the highest quartile of plasma Aβ42 concentrations were 2.85(95%CI:1.27-6.39)and 2.88(95%CI:1.44-5.75),respectively,comparing to subjects in the lowest quartile of plasma Aβ42 concentrations.Each 5 ng/L increment of plasma Aβ42 was associated with 64%(95%CI:20%-124%)higher odds of IGR and 53%(95%CI:19%-96%igher odds of T2D.The positive associations of plasma Aβ40 and Aβ42 with IGR&T2D were consistent among all subgroups except for the subgroups stratified by age and current smoking status.Conclusions:Our findings suggested significant and positive associations of plasma Aβ40 and Aβ42 concentration with risk of incident IGR and T2D.However,further studies are warranted to elucidate the mechanisms underlying plasma Aβ40 and Aβ42 increasing risk of T2DPart 3.Effects of β-amyloid 40 and 42 on the insulin sensitivity of HepG2 cells and C2C12 cellsObjective:To explore the effects of Aβ40 and Aβ42 on the insulin sensitivity of HepG2 cells and C2C12 cells.Methods:HepG2 cells and C2C12 cells were seeded in petri dishes after recovery.We observed the state and density of the cells daily,and conducted passage culture when the confluence of cells reached 80%.The cells were seeded in 96-well or 12-well plates required for the experiment and HepG2 cells could be used for the experiment when they were completely adherent.C2Cl2 cells were differentiated into myotubes before they were used for experiments.CCK-8 kit was used to detect cell viability,and the intervention time was determined to be 48 h.The experimental groups were divided as follows:(1)blank control:equal volume medium;(2)insulin group:100 nM insulin;(3)Aβ40 low-dose group(Aβ40-L):2 μM Aβ40+100 nM insulin;(4)Aβ40 medium-dose group(Aβ40-M):10 μM Aβ40+100 nM insulin;(5)Aβ40 high-dose group(Aβ40-H):20 μM Aβ40+100 nM insulin;(6)Aβ40-1 group(the reverse amino acid sequence control peptide for Aβ40):20 μM Aβ40-1+100 nM insulin;(7)Aβ42 low-dose group(Aβ42-L):2 μM Aβ42+100 nM insulin;(8)Aβ42 medium-dose group(Aβ42-M):10 μM Aβ42+100 nM insulin;(9)Aβ42 high-dose group(Aβ42-H):20 μM Aβ42+100 nM insulin;(10)Aβ42-1 group(the reverse amino acid sequence control peptide for Aβ42):20 μM Aβ42-1+100 nM insulin.12-well plates were treated with Aβ according to the above group,and glucose production experiment was used to evaluate the effects of Aβ on the insulin sensitivity of HepG2 cells;glucose uptake experiment was used to evaluate the effects of Aβ on the insulin sensitivity of C2C12 cells.Results:(1)Compared to the blank control,the glucose production of HepG2 cells in the insulin group was reduced by 27%(P<0.05).Aβ40 and Aβ42 attenuated the effect of insulin in a dose-dependent manner,and glucose production in the Aβ40-H was not significantly different from the blank control.Compared to the insulin group(73%±6%),the glucose production in the Aβ40-M group(84%±1%)and Aβ40-H group(95%±11%)was significantly increased(P<0.05).Consistently,the glucose production in the Aβ42-M group(84%±1%)and Aβ42-H group(88%±4%)was also significantly higher than the insulin group(P<0.05).In contrast,the glucose production in the Aβ40-1 and Aβ42-1 groups was not significantly different from the insulin group.(2)Compared to the blank control,the glucose uptake of C2C12 cells in the insulin group was increased by 153%(P<0.05).Aβ40 and Aβ42 attenuated the insulin effect in a dose-dependent manner,but compared to the blank control,glucose uptake in Aβ40-H and Aβ42-H groups was still increased by 100%and 60%(P<0.05),respectively.Compared to the insulin group(253%±25%),the glucose uptake in the Aβ40-H group(202%±15%)was significantly reduced(P<0.05),while the glucose uptake in the Aβ42-M group(199%±9%)and Aβ42-H group(160%±6%)was also significantly reduced(P<0.05).However,the glucose uptake in the Aβ40-1 and Aβ42-1 groups was not significantly different from that in the insulin group.Conclusions:Our study found that Aβ40 and Aβ42 could attenuate the insulin sensitivity of HepG2 cells and C2Cl2 cells,and the strength of effects became stronger with increasing concentrations.Aβ40-1 and Aβ42-1 had no significant effect on the insulin sensitivity of HepG2 cells and C2C12 cells,which indicated that the amino acid sequence might be an important basis for Aβ40 and Aβ42 inducing insulin resistance.However,the molecular mechanisms underlying Aβ40 and Aβ42 inducing insulin resistance in hepatocytes and skeletal muscle cells need to be further explored. |
