PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Neurobiol Aging. Author manuscript; available in PMC 2017 August 1.
Published in final edited form as:
PMCID: PMC4913037
NIHMSID: NIHMS784996

Impaired fasting glucose is associated with increased regional cerebral amyloid

Abstract

The Alzheimer’s disease (AD) risk gene apolipoprotein E epsilon 4 (APOE ε4) is associated with increased cerebral amyloid. Although impaired glucose metabolism is linked to AD risk, the relationship between impaired glycemia and cerebral amyloid is unclear. To investigate the independent effects of APOE ε4 and impaired glycemia on cerebral amyloid, we stratified nondemented subjects (n=73) into 4 groups: normal glucose, APOE ε4 non-carrier (control (CNT); n=31), normal glucose, APOE ε4 carrier (E4 only; n=14) impaired glycemia, APOE ε4 non-carrier (IG only; n=18) and impaired glycemia, APOE ε4 carrier (IG+E4; n=10). Cerebral amyloid differed both globally (p=0.023) and regionally; precuneus (p=0.007), posterior cingulate (PCC;p=0.020), superior parietal cortex (SPC;p=0.029), anterior cingulate (ACC;p=0.027), and frontal cortex (p=0.018). Post-hoc analyses revealed that E4 only subjects had increased cerebral amyloid vs. CNT globally and regionally in the precuneus, PCC, SPC, ACC, and frontal cortex. In IG only subjects, increased cerebral amyloid compared to CNT was restricted to precuneus, PCC, and SPC. IG+E4 subjects exhibited higher cerebral amyloid only in the precuneus relative to CNT. These results indicate that impaired glycemia and APOE ε4 genotype are independent risk factors for regional cerebral amyloid deposition. However, APOE ε4 and impaired glycemia did not have an additive effect on cerebral amyloid.

Keywords: impaired fasting glucose, insulin, Alzheimer’s disease, dementia, metabolism, amyloid, apolipoprotein E

Introduction

Type 2 Diabetes is a known risk factor for Alzheimer’s Disease (AD)(Morris, et al., 2014a) but the relationship between impaired glucose metabolism and cerebral amyloid is not well understood. In cognitively normal subjects, peripheral hyperglycemia is associated with decreased cerebral glucose metabolism (FDG-PET) in several brain regions, including the precuneus, posterior cingulate, and parietal regions (Burns, et al., 2013,Ishibashi, et al., 2015a,Ishibashi, et al., 2015b,Kawasaki, et al., 2008). These areas comprise a set of connections called the Default Mode Network (DMN). They exhibit some of the highest metabolic rates in the brain, (Cavanna and Trimble, 2006,Gusnard, et al., 2001), hypometabolism in AD (Bailly, et al., 2015,Mosconi, 2005), and are among the first to accumulate amyloid (Hedden, et al., 2009).

The most widely recognized risk gene for sporadic AD, Apolipoprotein E epsilon 4 (APOE ε4) is consistently associated with increased cerebral amyloid levels in nondemented subjects.(Jansen, et al., 2015) However, because of the relationship between peripheral and cerebral metabolism, specifically in DMN regions of interest, it is possible that impaired glycemia is an additional risk factor for accumulation of cerebral amyloid. The effect of glucose on amyloid levels is of particular clinical relevance in the elderly, as cerebral amyloid is a risk factor for AD and three quarters of U.S. elderly individuals exhibit prediabetes or diabetes (Cowie, et al., 2009). Thus, we stratified our sample into 4 groups to compare the independent and combined effect of each risk factor on cerebral amyloid. This is the first study to investigate the relationship of glucose metabolism, APOE ε4 genotype, and cerebral amyloid in nondemented elderly. We hypothesized that impaired glycemia may be an independent risk factor for elevated cerebral amyloid.

Methods

Approvals and recruitment

This study was approved by the University of Kansas Medical Center’s IRB. All participants provided informed consent according to institutional guidelines and this project was performed in accordance with the Declaration of Helsinki. Participants (n=73) were age 65 and older, sedentary (Mayer, et al., 2008) and free of cognitive impairment (Clinical Dementia Rating of 0). Subjects could not participate if they exhibited insulin-dependent diabetes, uncontrolled hypertension, or recent history of major neuropsychiatric, musculoskeletal, or cardiorespiratory impairment (within 2 years). Cases were further reviewed at a consensus diagnosis conference to ensure normal cognition.

Metabolic measures

Blood was drawn after an overnight fast. Plasma glucose was quantified (YSI-2300, Yellow Springs Instruments) and subjects classified as normal (NG; FG<100mg/dL) or impaired glycemia (IG; FG≥100mg/dL) based upon the American Diabetes Association cut-point for impaired fasting glucose. Twelve subjects who met the cut-point for impaired glycemia had a prior diabetes diagnosis and all diabetic subjects were on diabetic medication. We also quantified additional metabolic biomarkers (Insulin (Genway), amylin (Millipore) and non-esterified fatty acids (NEFA, Wako Diagnositcs)) in plasma using enzyme linked immunosorbent assay. Body mass was assessed using a digital scale accurate to 0.1kg (Seca Platform Scale, model 707). HOMA-IR was calculated from the product of glucose and insulin divided by 405.

Subject Groups

Individuals were grouped into a control group (CNT; n=31) if they had normal glycemia and did not carry the APOE ε4 gene. E4 only (n=14) subjects had normal glycema and carried at least one copy of APOE ε4. The IG group exhibited impaired glycemia but did not carry APOE ε4, and the combined IG+E4 group were both carriers of APOE ε4 and exhibited impaired glycemia.

Florbetapir PET

We used Florbetapir PET imaging to measure fibrillar beta-amyloid (Aβ) burden. Participants underwent two 5-min scans, approximately 50min after intravenous injection of 10mCi (370 Mbq) of Florbetapir F18 (18F-AV-45). Images were acquired on a GE Discovery ST PET/CT scanner and reconstructed using an iterative reconstruction algorithm, with a 3mm full-with, half maximum Gaussian filter and were corrected for radiation attenuation and summed. We used SPM 12 (http://www.fil.ion.ucl.ac.uk/) to process images, manually recentering images and normalizing to Montreal Neurological Institute space using a AV-45 specific PET template. We smoothed the resulting image using a 6mm full-width, half maximum Gaussian filter. Standard uptake value ratios (SUVRs) for 6 a priori defined ROIs were calculated relative to whole cerebellum. The ROI masks were created from the Wake Forest Pick Atlas,(Maldjian, et al., 2003) and included the anterior cingulate (ACC), posterior cingulate (PCC), precuneus, medial and superior frontal cortex, lateral temporal, occipital, and superior parietal cortex (SPC).

Statistical analyses

Data are expressed as means ± SD for continuous variables or number and percent for categorical data. Normality was assessed using Shapiro-Wilk tests of normality and non-normally distributed variables were log-transformed prior to univariate analyses. Differences between groups were assessed by one-way analysis of variance controlling for age and sex. For significant results, post-hoc analysis was performed using the least significant difference test. Categorical data were analyzed by chi-squared analyses. Relationships between continuous variables were analyzed using linear regression. Statistical analyses were performed using SPSS version 22. Results were considered significant at p<0.05.

Results

There were no differences in age, sex, or education or genotype between groups. Overall, significant differences in cerebral amyloid between groups were observed both globally (p=0.023) and regionally, in the precuneus (p=0.007), PCC (p=0.020), SPC (p=0.029), ACC (p=0.027), frontal cortex (p=0.018). Post-hoc analyses revealed the most widespread increases in E4 only subjects. This group had higher amyloid vs. CNT globally (p=0.002) and regionally; precuneus (p=0.023), PCC (p=0.015), SPC (p=0.016), ACC (p=0.015), and frontal cortex (p=0.007). Increased amyloid in the IG only group compared to CNT was restricted to DMN regions; precuneus (p=0.013), PCC (p=0.022), and SPC (p=0.030). The IG+E4 group had the most isolated increase in cerebral amyloid; only differing from CNT in the precuneus (p=0.021). No other inter-group differences aside from those vs. the CNT group were significant for any groups. Post-hoc findings between groups are summarized in Table 2.

Table 2
Post-hoc comparisons of cerebral amyloid between groups

In pooled analyses of all subjects, linear regression did not reveal any relationships between metabolic biomarkers and cerebral amyloid in any region. However, analyses of APOE ε4 carriers and noncarriers separately revealed DMN-specific relationships. In APOE ε4 noncarriers, fasting glucose was positively related to increased cerebral amyloid in the precuneus (β=0.308, p=0.031), PCC (β=0.345, p=0.019), and SPC (β=0.320, p=0.029). In APOE ε4 carriers, no relationships between metabolic biomarkers and cerebral amyloid were observed in any region.

Metabolic differences between the stratified groups were observed for body mass index (p=0.007) and body weight (p=0.029). Post-hoc analyses revealed that the IG+E4 group had higher BMI compared to the CNT (p=0.003) and E4 only (p=0.010) groups, and higher body weight compared to the CNT (p=0.018) and E4 only (p=0.008) groups. The IG only group had higher BMI than the CNT group (p=0.017) and higher body weight compared to the E4 only subjects (p=0.032). Other measures, including fasting insulin, amylin, non-esterified fatty acids, and HOMA-IR were not significantly different between groups.

The number of individuals with a history of diabetes did not differ between the groups with impaired glycemia (IG only and IG+E4). Because all 12 diabetic subjects in this study were taking diabetic medication, (11 metformin, 1 sitagliptin) we re-analyzed the data excluding diabetics. The IG only group still showed increased amyloid vs. CNT in the precuneus and SPC, but not the PCC, and the effect of increased cerebral amyloid in the precuenus of the E4+IG subjects vs. control was also no longer significant. Whether this is due to loss of power or is an effect of diabetic medication warrants further study.

Discussion

This is the first study to examine the relationship between impaired glycemia, APOE ε4 genotype, and cerebral amyloid in cognitively-healthy elderly. Individuals with only the risk factor of impaired glycemia (IG only) exhibited higher cerebral amyloid than CNT subjects in three highly metabolic brain regions. Even mild increases of fasting glucose in cognitively-normal subjects are associated with decreased regional brain glucose metabolism (FDG-PET) in these regions (Ishibashi, et al., 2015a,Ishibashi, et al., 2015b,Kawasaki, et al., 2008) known to show glucose hypometabolism in AD (Minoshima, et al., 1997,Mosconi, 2005,Sakamoto, et al., 2002). Interestingly, APOE ε4 carriers with impaired glycemia (IG+E4) exhibited more isolated increases in cerebral amyloid, while APOE ε4 carriers with normal glucose levels exhibited the most widespread increases in cerebral amyloid.

Prior studies of peripheral metabolism and cerebral amyloid are mixed. In late middle aged adults with normal glucose, greater insulin resistance (HOMA-IR) is associated with greater amyloid burden in two broad ROI’s (frontal and temporal)(Willette, et al., 2015). A much smaller study of the relationship between diabetes diagnosis and cerebral amyloid in 4 ROI’s (anterior cingulate, frontal cortex, parietal cortex, and precuneus) showed no relationship (Moran, et al., 2015). An additional study found that insulin resistance was not predictive of cerebral amyloid in elderly subjects, and found no relationship between glucose and postmortem AD pathology (Thambisetty, et al., 2013). However, subjects were grouped in tertiles (not an established cut-point) based upon either fasting or 120 minute post-oral glucose tolerance test values. Impaired fasting glucose and impaired glucose tolerance are different states of insulin resistance that do not necessarily define the same subjects or state of metabolic impairment (Nathan, et al., 2007). Studies using glucose tolerance testing may differ compared to studies examining fasting measures; thus, inconsistent findings may due to differing methodology for defining impaired glucose or insulin resistance.

Our study used a clinically accepted, readily available cut-point for to analyze this relationship and yields interesting insight on the potentially separate effects of impaired glycemia and APOE ε4 on cerebral amyloid pathology. Elevated insulin often precedes high glucose by compensating for mild insulin resistance (Prentki and Nolan, 2006) and hyperglycemia develops with the onset of beta-cell dysfunction and loss of compensation. Here, fasting insulin was not significantly different across groups. This is noteworthly because both glucose (Macauley, et al., 2015) and insulin (Gasparini, et al., 2001,Kulstad, et al., 2006,Reger, et al., 2008) affect amyloid production and trafficking, but may involve different pathways or having different magnitudes of effect on Aβ (reviewed in (Sato and Morishita, 2015).

Impaired glycemia is associated with disease progression in mild cognitive impairment (Morris, et al., 2014b,Velayudhan, et al., 2010). This study extends these findings by showing that prior to cognitive impairment, impaired glycemia is associated with increased cerebral amyloid in DMN regions of subjects not at genetic (APOE-mediated) AD risk. It is postulated that because neuronal depolarization, repolarization and protein trafficking all involve high energy demand, bioenergetic failure may play a key role in AD (Pathak, et al., 2013,Swerdlow, et al., 2010). Peripheral hyperglycemia is associated with hypometabolism in the brain,(Ishibashi, et al., 2015a,Ishibashi, et al., 2015b,Kawasaki, et al., 2008) and animal studies suggest that deficits in cellular energy metabolism affect the processing and transport of Aβ in vitro and in vivo.(Chao, et al., 2016,Kong, et al., 2015,Macauley, et al., 2015) and excess Aβ may in turn exacerbate hypometabolism (Tarczyluk, et al., 2015).

An additional consideration of this study includes diabetes medication use. Laboratory studies have shown that the first-line diabetes medication metformin both increases (Chen, et al., 2009,Picone, et al., 2015) and decreases (Hettich, et al., 2014) BACE1 expression, which could affect cerebral amyloid levels. It is worth noting that autopsy studies have failed to show that diabetes diagnosis is associated with increased amyloid pathology, (Alafuzoff, et al., 2009,Arvanitakis, et al., 2006) although most studies have assessed global markers of neuropathology, and our findings of increased amyloid pathology in the groups with impaired glycemia were regional in nature. In fact, the E4 only group was the only group to show significantly higher global cerebral amyloid. Additional well-powered studies are needed to evaluate the potential effect of diabetes and diabetic medication on regional cerebral amyloid in humans.

An important limitation of this study is the small sample size. While Aβ loads seemed similar between IG only and IG+E4 groups, particularly in DMN regions, this effect was significant in IG only subjects in the precuneus, PCC, and SPC, but only in the precuneus in the IG+E4 group, possibly due to lack of power. In addition, our measures of glucose metabolism are limited to peripheral, rather than cerebral measures. However, this adds to the current literature by indicating that peripheral hyperglycemia, which is very easy to measure with noninvasive measures, is related to cerebral amyloid deposition in subjects not at genetic AD risk in highly metabolic brain regions. In conclusion, this study shows regional differences in cerebral amyloid deposition due to both genotype and glycemic status in cognitively-healthy elderly. This is important as it suggests that regional amyloid deposition tracks with impairments in glucose metabolism prior to cognitive impairment in elderly individuals not at genetic AD risk, and further substantiates the building evidence that impaired glucose metabolism is a significant risk factor for the development of cognitive impairment in particular cohorts.

Table 1
Demographic, metabolic, physiologic, and neuroimaging outcomes

Highlights

  • Hyperglycemia is associated with increased default mode network amyloid in E4 negatives
  • Fasting glucose correlates with cerebral amyloid in the precuneus, PCC and SPC
  • Hyperglycemia and APOE E4 genotype together do not exacerbate cerebral amyloid

Acknowledgments

This research was supported by P30 AG035982, F32 AG044953, K99AG050490, KL2 TR000119, R01DK088940, Merit Review Award #1I01BX002567-01 from VA BLRD, and R01AG043962. Space, nursing, and assay support were provided by UL1 TR000001 and NICHD HD02528.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • Alafuzoff I, et al. Beta-amyloid deposition in brains of subjects with diabetes. Neuropathology and applied neurobiology. 2009;35(1):60–68. [PubMed]
  • Arvanitakis Z, et al. Diabetes is related to cerebral infarction but not to AD pathology in older persons. Neurology. 2006;67(11):1960–1965. [PubMed]
  • Bailly M, et al. 18F-FDG and 18F-florbetapir PET in clinical practice: regional analysis in mild cognitive impairment and Alzheimer disease. Clinical nuclear medicine. 2015;40(2):e111–e116. [PubMed]
  • Burns CM, et al. Higher serum glucose levels are associated with cerebral hypometabolism in Alzheimer regions. Neurology. 2013;80(17):1557–1564. [PMC free article] [PubMed]
  • Cavanna AE, Trimble MR. The precuneus: a review of its functional anatomy and behavioural correlates. Brain. 2006;129(Pt 3):564–583. [PubMed]
  • Chao AC, et al. Hyperglycemia Increases the Production of Amyloid Beta-Peptide Leading to Decreased Endothelial Tight Junction. CNS neuroscience & therapeutics. 2016;22(4):291–297. [PubMed]
  • Chen Y, et al. Antidiabetic drug metformin (GlucophageR) increases biogenesis of Alzheimer's amyloid peptides via up-regulating BACE1 transcription. Proc Natl Acad Sci U S A. 2009;106(10):3907–3912. [PubMed]
  • Cowie CC, et al. Full accounting of diabetes and pre-diabetes in the U.S. population in 1988–1994 and 2005–2006. Diabetes care. 2009;32(2):287–294. [PMC free article] [PubMed]
  • Gasparini L, et al. Stimulation of beta-amyloid precursor protein trafficking by insulin reduces intraneuronal beta-amyloid and requires mitogen-activated protein kinase signaling. The Journal of neuroscience. e. 2001;21(8):2561–2570. [PubMed]
  • Gusnard DA, et al. Searching for a baseline: functional imaging and the resting human brain. Nat Rev Neurosci. 2001;2(10):685–694. [PubMed]
  • Hedden T, et al. Disruption of functional connectivity in clinically normal older adults harboring amyloid burden. The Journal of neuroscience. 2009;29(40):12686–12694. 10.1523/JNEUROSCI.3189-09.2009. [PMC free article] [PubMed]
  • Hettich MM, et al. The anti-diabetic drug metformin reduces BACE1 protein level by interfering with the MID1 complex. PloS one. 2014;9(7):e102420. [PMC free article] [PubMed]
  • Ishibashi K, et al. Reduced uptake of (18)F-FDG and (15)O-H2O in Alzheimer's disease-related regions after glucose loading. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism. 2015a;35(8):1380–1385. [PMC free article] [PubMed]
  • Ishibashi K, et al. Relationship between Alzheimer disease-like pattern of 18F-FDG and fasting plasma glucose levels in cognitively normal volunteers. Journal of nuclear medicine. 2015b;56(2):229–233. [PubMed]
  • Jansen WJ, et al. Prevalence of cerebral amyloid pathology in persons without dementia: a meta-analysis. JAMA : the journal of the American Medical Association. 2015;313(19):1924–1938. [PMC free article] [PubMed]
  • Kawasaki K, et al. Influence of mild hyperglycemia on cerebral FDG distribution patterns calculated by statistical parametric mapping. Annals of nuclear medicine. 2008;22(3):191–200. [PubMed]
  • Kong JJ, et al. Nicorandil inhibits oxidative stress and amyloid-beta precursor protein processing in SH-SY5Y cells overexpressing APPsw. International journal of clinical and experimental medicine. 2015;8(2):1966–1975. [PMC free article] [PubMed]
  • Kulstad JJ, et al. Differential modulation of plasma {beta}-amyloid by insulin in patients with Alzheimer disease. Neurology. 2006;66(10):1506–1510. [PubMed]
  • Macauley SL, et al. Hyperglycemia modulates extracellular amyloid-beta concentrations and neuronal activity in vivo. The Journal of clinical investigation. 2015;125(6):2463–2467. [PMC free article] [PubMed]
  • Maldjian JA, et al. An automated method for neuroanatomic and cytoarchitectonic atlas-based interrogation of fMRI data sets. Neuroimage. 2003;19(3):1233–1239. [PubMed]
  • Mayer CJ, et al. Developing a Telephone Assessment of Physical Activity (TAPA) questionnaire for older adults. Prev Chronic Dis. 2008;5(1):A24. [PMC free article] [PubMed]
  • Minoshima S, et al. Metabolic reduction in the posterior cingulate cortex in very early Alzheimer's disease. Ann Neurol. 1997;42(1):85–94. [PubMed]
  • Moran C, et al. Type 2 diabetes mellitus and biomarkers of neurodegeneration. Neurology. 2015 [PubMed]
  • Morris JK, et al. Is Alzheimer's disease a systemic disease? Biochimica et biophysica acta. 2014a
  • Morris JK, et al. Impaired glycemia increases disease progression in mild cognitive impairment. Neurobiology of aging. 2014b;35(3):585–589. [PMC free article] [PubMed]
  • Mosconi L. Brain glucose metabolism in the early and specific diagnosis of Alzheimer's disease. FDG-PET studies in MCI and AD. European journal of nuclear medicine and molecular imaging. 2005;32(4):486–510. [PubMed]
  • Nathan DM, et al. Impaired fasting glucose and impaired glucose tolerance: implications for care. Diabetes care. 2007;30(3):753–759. [PubMed]
  • Pathak D, et al. Energy failure: does it contribute to neurodegeneration? Ann Neurol. 2013;74(4):506–516. [PMC free article] [PubMed]
  • Picone P, et al. Metformin increases APP expression and processing via oxidative stress, mitochondrial dysfunction and NF-kappaB activation: Use of insulin to attenuate metformin's effect. Biochimica et biophysica acta. 2015;1853(5):1046–1059. [PubMed]
  • Prentki M, Nolan CJ. Islet beta cell failure in type 2 diabetes. The Journal of clinical investigation. 2006;116(7):1802–1812. [PMC free article] [PubMed]
  • Reger MA, et al. Intranasal insulin improves cognition and modulates beta-amyloid in early AD. Neurology. 2008;70(6):440–448. [PubMed]
  • Sakamoto S, et al. Differences in cerebral metabolic impairment between early and late onset types of Alzheimer's disease. Journal of the neurological sciences. 2002;200(1–2):27–32. [PubMed]
  • Sato N, Morishita R. The roles of lipid and glucose metabolism in modulation of beta-amyloid, tau, and neurodegeneration in the pathogenesis of Alzheimer disease. Frontiers in aging neuroscience. 2015;7:199. [PMC free article] [PubMed]
  • Swerdlow RH, et al. The Alzheimer's disease mitochondrial cascade hypothesis. J Alzheimers Dis. 2010;20(Suppl 2):S265–S279. [PMC free article] [PubMed]
  • Tarczyluk MA, et al. Amyloid beta 1–42 induces hypometabolism in human stem cell-derived neuron and astrocyte networks. Journal of cerebral blood flow and metabolism. 2015;35(8):1348–1357. [PMC free article] [PubMed]
  • Thambisetty M, et al. Glucose intolerance, insulin resistance, and pathological features of Alzheimer disease in the Baltimore Longitudinal Study of Aging. JAMA neurology. 2013;70(9):1167–1172. [PMC free article] [PubMed]
  • Velayudhan L, et al. Risk of developing dementia in people with diabetes and mild cognitive impairment. The British journal of psychiatry : the journal of mental science. 2010;196(1):36–40. [PubMed]
  • Willette AA, et al. Insulin resistance predicts brain amyloid deposition in late middle-aged adults. Alzheimers Dement. 2015;11(5):504 e1–510 e1. [PMC free article] [PubMed]