Serum Adiponectin in Alzheimer’s Disease (AD): Association with AD Biomarkers and Cognitive Outcome

Abstract

Background:

The association between dementia and serum adiponectin has been evaluated in many studies; however, conclusions remain mixed.

Objective:

We investigated the cross-sectional associations of adiponectin with cognitive function and Alzheimer’s disease (AD) biomarkers and whether serum adiponectin levels can predict cognitive outcomes.

Methods:

This study included 496 participants from the Alzheimer’s Disease Neuroimaging Initiative 1 (ADNI1) with available serum adiponectin levels at baseline and ≥65 years of age. Subjects were stratified based on sex and apolipoprotein ɛ4 (APOE4) carrier status to determine associations between adiponectin and cognitive function. The linear mixed model was used to analyze associations between adiponectin level and cognitive outcome in amnestic mild cognitive impairment (aMCI) patients.

Results:

Serum adiponectin levels were higher in aMCI and AD than in CN subjects among APOE4 non-carrier males (adiponectin in CN, aMCI, and AD: 0.54±0.24, 0.74±0.25, and 0.85±0.25, respectively, p < 0.001). In this group, serum adiponectin levels were associated with age (p = 0.001), ADAS13 (p < 0.001), memory function (p < 0.001), executive function (p < 0.001), total tau (p < 0.001), and phosphorylated tau (p < 0.001) measures in cerebrospinal fluid (CSF). Higher adiponectin level was not associated with cognitive outcome in aMCI patients in the linear mixed model analysis over 5.3±2.6 years of mean follow-up.

Conclusion:

Serum adiponectin level was associated with cognitive function and CSF AD biomarkers among APOE4 non-carrier males. However, serum adiponectin level was not associated with longitudinal cognitive function outcome in aMCI.

Keywords

Adiponectin Alzheimer’s disease apolipoprotein E4 mild cognitive impairment tau proteins

INTRODUCTION

Dementia is one of the major causes of disability and dependency globally among the elderly. Approximately 50 million people worldwide are suffering from dementia, with nearly 10 million new cases occurring every year. Currently, there is no treatment to cure dementia or to alter its progressive course. Metabolic syndrome, a modifiable factor that includes type 2 diabetes mellitus (T2DM) and obesity, has been cited as a risk factor for dementia [1, 2]. T2DM was associated with an increased risk of Alzheimer’s disease (AD) and vascular dementia [1]. People with high midlife body mass index (BMI) and central obesity are three times more likely to develop dementia in later life [2].

Adiponectin is a multi-functional adipocyte-derived hormone that regulates lipid and glucose metabolism [3], and paradoxically decreases in obesity. The physiological functions of adiponectin in obesity, diabetes, inflammation, atherosclerosis, and cardiovascular disease have been elucidated in many studies [4]. Hypoadiponectinemia is associated with insulin resistance, obesity, metabolic syndrome including T2DM, and cardiovascular disease [5]. In addition, adiponectin exerts anti-inflammatory effects by decreasing the production of pro-inflammatory cytokines [3]. Based on the relation between metabolic syndrome and adiponectin, the association between adiponectin and dementia has been investigated in several studies. However, the results lack consistency. In the longitudinal population-based Framingham Heart Study, an association with adiponectin and increased risk of all cause dementia and AD in females was observed [6]. High plasma adiponectin was also associated with neuroimaging and cognitive outcomes among females in the Mayo Clinic Study of Aging [7]. However, in elderly subjects, correlations were not observed between serum adiponectin levels, mild cognitive impairment (MCI) and cognitive normal (CN) subjects [8]. In another study, serum adiponectin levels were significantly lower in MCI and dementia subjects compared with controls and did not predict cognitive decline [9].

In the present study, the effects of adiponectin on the diagnosis and progression of dementia in the elderly were investigated. The cross-sectional association between serum adiponectin levels and AD biomarkers were examined based on sex and apolipoprotein E ɛ4 (APOE4) carrier status stratification. Furthermore, longitudinal analysis on cognitive function change in MCI subjects was performed to assess the prognostic effects of serum adiponectin baseline levels.

MATERIALS AND METHODS

Alzheimer’s disease neuroimaging initiative study design

Data used in the preparation of this article were obtained from the Alzheimer’s disease Neuroimaging Initiative (ADNI) database (http://adni.loni.usc.edu). The ADNI was launched in 2003 as a public-private partnership, led by Principal Investigator Michael W. Weiner, MD. The primary goal of ADNI has been to test whether serial magnetic resonance imaging (MRI), positron emission tomography (PET), other biological markers, and clinical and neuropsychological assessment can be combined to measure the progression of MCI and early AD. ADNI participants were recruited from more than 50 sites across the USA and Canada. Regional ethical committees of all institutions approved of the study, and all participants provided written informed consent. For up-to-date information, see http://www.adni-info.org.

Study subjects

All individuals included in this study were participants from the ADNI 1 with available baseline proteomics data. Serum adiponectin levels were analyzed in the elderly ≥65 years of age. Three patients were excluded from final analysis due to uncertain diagnostic information. Demographic information, educational level, APOE4 carrier status, clinical information, neuroimaging, and cerebrospinal fluid (CSF) biomarker data were downloaded from the ADNI data repository. BMI was calculated as the weight in kilograms divided by the square of the height in meters. By definition, individuals in the MCI group scored ≥24 on the Mini-Mental State Examination (MMSE) and exhibited subjective memory loss (> 1 standard deviation (SD) below the normal mean of the delayed recall of the Wechsler Memory Scale Logical Memory II), received a clinical dementia rating-sum of boxes (CDR-SB) score of 0.5, and had preserved activities of daily living and the absence of dementia. Those who were suspected as having vascular, traumatic, or inflammatory causes of MCI, or any significant neurologic disease other than AD, were excluded from the study cohort. The diagnosis of probable AD was based on the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer’s Disease and Related Disorders Association (NINCDS/ADRDA) criteria [10].

Laboratory analyses of adiponectin

Plasma samples were obtained in the morning following an overnight fast during the baseline visit. Plasma samples in 500μL EDTA were frozen within 120 min. Plasma samples were analyzed in a 190-analyte multiplex immunoassay panel (Human Discovery Map; Myriad RBM). The panel was developed on the Luminex xMAP platform by Rules-Based Medicine (RBM). The Luminex xMAP technology uses a flow-based laser apparatus and fluorescent polystyrene microspheres to detect biomarker concentrations. The adiponectin assay has a least detectable dose of 0.028μg/mL and a lower assay limit of 0.016μg/mL. The least detectable dose is considered the lowest reliable level for the assay, and anything lower is associated with greater error. All adiponectin levels were greater than the least detectable dose. The detailed data processing methods are described in Biomarkers Consortium Plasma Proteomics Project RBM multiplex data (http://adni.loni.ucla.edu/wp-content/uploads/2010/11/BC_Plasma_Proteomics_Data_Primer.pdf).

Cognitive assessment

The cognitive function was assessed at baseline, and annual follow-up. We used the modified Alzheimer’s Disease Assessment Scale cognition subscale (ADAS-Cog 13) [11] and composite scores of memory and executive functioning (ADNI-MEM and ADNI-EF) as indices of cognitive performance. ADAS-Cog 13 includes all items from ADAS-Cog (11 items assessing memory, language, praxis, and orientation) plus delayed recall and cancellation tasks. ADNI-MEM is a memory composite score calculated from the items in several memory tasks, including the Rey Auditory Verbal Learning Test, ADAS-Cog, Logical memory, and the MMSE [12]. ADNI-EF included Category Fluency-animals, Category Fluency-vegetables, Trails A and B, Digit Span Backward, Wechsler Adult Intelligence Scale-Revised Digit Symbol Substitution, and five Clock Drawing Test items (circle, symbol, numbers, hands, time) [13].

Neuroimaging analysis

MRI

Hippocampal atrophy on MRI was used as a biomarker of neurodegeneration, and white matter hyperintensity volume as a biomarker for microvascular disease burden, as described in the National Institute on Aging-Alzheimer’s Association Research Framework [14]. The hippocampal volume was measured on the 1.5 Tesla (1.5 T) MRI using a T1-weighted sagittal volumetric magnetization rapid gradient echo sequence (1.25 mm×1.25 mm×1.20 mm). The volume of white matter hyperintensity was measured based on run-time PD-, T1-, and T2-weighted structural MRI of the brain with labeled training examples. Detailed data processing methods are described online (adni.loni.usc.edu/methods/mri-analysis/).

F-18 Fluorodeoxyglucose (FDG) PET

Brain metabolism was used as a biomarker of neurodegeneration [14]. Five regions of interests (ROIs) (Left Angular Gyrus, Right Angular Gyrus, Bilateral Posterior Cingular Gyrus, Left Inferior Temporal Gyrus, and Right Inferior Temporal Gyrus) were used as the metaROI. The mean metabolism of each metaROI was normalized by dividing it by pons/vermis reference region mean. Detailed data processing methods are as described (http://adni.loni.usc.edu/methods/pet-analysis-method/pet-analysis/).

CSF biomarker analysis

The CSF values for amyloid-β₄₂ (Aβ₄₂), total tau (tTau) protein, and phosphorylated tau (pTau) protein at baseline were generated using a single lot number of the novel, fully automated, electrochemiluminescent Elecsys® immunoassays (Roche Diagnostics, Basel, Switzerland) downloaded from a single data set.

Statistical analyses

Serum adiponectin levels were log-transformed before analysis. The chi-square test was used to compare categorical variables. Continuous variables are expressed as the mean±SD. To compare continuous variables based on diagnosis (CN, MCI, and AD), one-way analysis of variance (ANOVA) with Tukey’s post-hoc test were used for normally distributed variables and the Kruskal-Wallis test with Mann-Whitney U test for skewed variables. The linear association of adiponectin level and AD biomarkers was assessed using Pearson’s correlation analysis. The Bonferroni correction was used to reduce multiple comparison problems. A linear mixed effects model including a random slope and intercept for each participants was used to investigate longitudinal changes in cognitive function over time based on the following independent variables: adiponectin level, time, age, and educational level. Patients with ≥24 months of follow up and ≥3 cognitive function evaluations were included in the linear mixed effects model analysis. A p-value < 0.05 was considered statistically significant. SPSS for Windows (version 25.0; SPSS Inc., Chicago, IL, USA) was used for statistical analyses.

RESULTS

Patient characteristics

The final analysis included 496 ADNI 1 participants: 311 (62.7%) males and 185 (37.3%) females, with a mean age of 76.5±5.7 ears (range 65.1–89.6 years). Baseline characteristics of each diagnosis (CN, MCI, and AD) are shown in Table 1. Age and educational level were not different between diagnosis groups. BMI in MCI (25.9±3.9, p = 0.044) and AD (25.3±3.4, p = 0.012) patients were lower than in CN subjects (27.2±4.0). Several ADNI participants had missing data for specific biomarkers. The specific number available for each biomarker is presented in Table 1.

Table 1

Demographic, Cognitive, and Biomarker Data of Study Subjects

	n	CN	n	MCI	n	AD	p
Age (y)	56	75.5±5.3	345	76.6±5.7	95	77.1±6.0	0.288
Sex – Female n (%)	56	27 (48.2)	345	119 (34.5)	95	39 (41.0)	0.101
Education (y)	56	15.7±2.7	345	15.6±3.0	95	14.9±3.2	0.096
APOE 4 – carrier n (%)	56	4 (7.1)	345	186 (53.9)	95	66 (69.4)	< 0.001
Adiponectin	56	0.69±0.29	345	0.76±0.24	95	0.81±0.24^a	0.021
BMI (kg/m²)	55	27.2±4.0	344	25.9±3.9^a	95	25.3±3.4^a	0.016
Hypertension n (%)	56	27 (48.2)	345	172 (49.8)	95	52 (54.7)	0.652
Dyslipidemia n (%)	56	25 (44.6)	345	170 (49.2)	95	50 (52.6)	0.636
CAD n (%)	56	7 (12.5)	345	50 (14.4)	95	14 (14.7)	0.917
DM n (%)	56	4 (7.1)	345	30 (8.6)	95	5 (5.2)	0.534
Modified Hachinski scores	56	0.6±0.7	345	0.6±0.7	95	0.6±0.6	0.983
		(range 0–3)		(range 0–4)		(range 0–3)
Neuropsychological test
MMSE	56	28.9±1.1	345	27.0±1.7^a^,c	95	23.6±1.9^a^,b	< 0.001
ADAS13	56	9.61±4.14	342	18.78±6.16^a^,c	93	28.19±6.82^a^,b	< 0.001
CDR-SB	56	0.01±0.09	345	1.60±0.88	95	4.24±1.56	< 0.001
		(range 0–0.5)		(range 0–5)^a^,c		(range 1–9)^a^,b
ADNI-MEM	56	0.875±0.463	345	–0.096±0.572^a,c	95	–0.831±0.532^a,b	< 0.001
ADNI-EF	56	0.638±0.664	345	–0.079±0.828^a,c	95	–0.955±0.841^a,b	< 0.001
Structural Magnetic Resonance Imaging
Hippocampal volume (cm³)	51	7.30±0.81	270	6.26±1.01^a,c	68	5.55±0.95^a^,b	< 0.001
White matter hyperintensity volume (cm³)	56	0.420±0.715	345	0.995±2.944	95	1.138±2.180	0.246
Cerebrospinal fluid
Amyloid-β₄₂ (pg/ml)	55	1445.03±264.93	160	808.14±412.92^a^,c	82	653.36±319.71^a^,b	< 0.001
Total-Tau (pg/ml)	55	225.34±74.21	160	320.10±148.79^a	82	350.95±128.83^a	< 0.001
Phosphorylated-Tau (pg/ml)	55	19.86±6.57	160	31.73±16.37^a	82	35.15±14.50^a	< 0.001
F-18 Fluorodeoxyglucose Positron Emission Tomography
MetaROI SUVr	25	1.292±0.116	178	1.185±0.117^a^,c	55	1.097±0.114^a^,b	< 0.001

n, number; CN, cognitive normal; MCI, mild cognitive impairment; AD, Alzheimer’s disease with dementia; APOE4, Apolipoprotein E ɛ4; BMI, body mass index; CAD, coronary artery disease; DM, diabetes mellitus; MMSE, Mini-Mental State Examination; ADAS 13, Alzheimer’s disease assessment scale-cognitive 13; CDR-SB, clinical dementia rating-sum of boxes; ADNI-MEM, a memory composite score calculated from the items in several memory tasks, including the Rey Auditory Verbal Learning Test, ADAS-Cog, Logical memory, and the MMSE; ADNI-EF, a executive composite score calculated from Category Fluency-animals, Category Fluency-vegetables, Trails A and B, Digit span backwards, Wechsler Adult Intelligence Scale-Revised Digit Symbol Substitution, and 5 Clock Drawing items (circle, symbol, numbers, hands, time); MetaROI, five regions of interests consist of left angular, right angular, bilateral posterior cingular, left inferior temporal, and right inferior temporal gyri; SUVr, standardized uptake value ratio normalized by pons/vermis reference region mean. Data are mean±standard deviation or number (%). Log-transformed adiponectin levels were used for statistical analysis. p of the One-way ANOVA with post-hoc Tukey’s test; ^ap < 0.05 compared to CN; ^bp < 0.05 compared to MCI; ^cp < 0.05 compared to AD.

Serum adiponectin level stratified based on sex and APOE4

The mean serum adiponectin level was 6.94±4.27μg/mL (range 1.1–32.0μg/mL). Adiponectin levels in females were higher than in males (0.85±0.23 versus 0.72±0.25, p < 0.001). Adiponectin levels were different in each diagnosis group; AD subjects had higher levels than CN subjects (0.81±0.24 versus 0.69±0.29, p = 0.021). When statistical analysis was performed on stratification based on sex and APOE4 carrier status, group difference in adiponectin level was observed in APOE4 non-carrier males (p < 0.001; Fig. 1 and Table 2). A higher adiponectin level was observed in MCI (0.74±0.25, p = 0.001) and AD subjects (0.82±0.25, p < 0.001) than in CN subjects (0.54±0.24) among APOE4 non-carrier males. In the same group, statistical difference in adiponectin levels was not observed between AD and MCI subjects. Different serum adiponectin levels were not observed among male APOE4 carriers in the diagnostic groups. Group differences were not observed in females.

Fig. 1

Serum adiponectin level stratified based on sex and apolipoprotein E ɛ4 carrier status (left: male; right: female). Adiponectin levels were log-transformed for the statistical analysis. A higher level of adiponectin was observed in MCI (p = 0.001) and AD (p < 0.001) than in CN (serum adiponectin in CN, MCI, and AD: 0.54±0.24, 0.74±0.25, and 0.85±0.25, respectively, p < 0.001) among APOE4 non-carrier males. Serum adiponectin levels failed to show difference according to diagnosis in APOE4 carrier males. In female participants, no association was found between serum adiponectin levels and cognitive diagnosis in both APOE4 carrier and non-carrier groups. APOE4, Apolipoprotein E ɛ4; CN, cognitive normal; MCI, mild cognitive impairment; AD, Alzheimer’s disease with dementia.

Table 2

Adiponectin levels stratified by sex and Apolipoprotein E ɛ4 carrier status

Sex	APOE4	N	CN	n	MCI	n	AD	p
Male	Non-carrier	26	0.54±0.24	105	0.74±0.25^*	17	0.85±0.25^*	< 0.001
	Carrier	3	0.59±0.36	121	0.72±0.22	39	0.70±0.25	0.571
Female	Non-carrier	54	0.86±0.25	26	0.81±0.25	12	0.95±0.20	0.130
	Carrier	1	0.56	65	0.85±0.22	27	0.88±0.19	0.540^**

APOE4, Apolipoprotein E ɛ4; n, number; CN, cognitive normal; MCI, mild cognitive impairment; AD, Alzheimer’s disease with dementia. Adiponectin levels were log-transformed for the statistical analysis. Data are shown in mean±SD. p of the Kruskal Wallis test with Mann Whitney U-test; ^*p < 0.05 compared to CN; ^**p value of the Student’s t test between MCI and AD.

Correlation between serum adiponectin and AD biomarkers

Pearson correlation analysis was performed between serum adiponectin levels and AD biomarkers in APOE4 non-carrier males. Serum adiponectin levels were correlated with age (r = 0.266, p = 0.001, n = 148) and cognitive function: ADAS13 (r = 0.373, p < 0.001, n = 148), ADNI-MEM (r = –0.289, p <0.001, n = 148), and ADNI-EF (r = –0.365, p < 0.001, n = 148). Serum adiponectin levels were positively correlated with tTau titer in CSF (r = 0.392, p < 0.001, n = 91) and with pTau titer in CSF (r = 0.409, p < 0.001, n = 091). Serum adiponectin levels were negatively correlated with Aβ₄₂ titer in CSF (r = –0.302, p = 0.004, n = 91) but only with borderline significance. Serum adiponectin levels were negatively correlated with BMI (r = –0.224, p =0.006, n = 147) and hippocampal volume (r = –0.263, p = 0.006, n = 109) but without statistical significance. Serum adiponectin levels were not significantly associated with white matter hyperintensity volume or brain metabolism.

Cognitive outcomes based on adiponectin level

To investigate the prognostic value of adiponectin, patients were divided into three groups based on serum adiponectin levels: low (adiponectin level in first tertile, adiponectin < 0.67, n = 32); intermediate (adiponectin in second tertile, 0.67≤adiponectin < 0.89, n = 26); high (adiponectin in third tertile, 0.89≤adiponectin, n = 22). In APOE4 non-carrier males diagnosed with MCI, linear mixed model analysis was performed to assess the difference in longitudinal cognitive function change based on serum adiponectin levels (Fig. 2). Cognitive functions decreased longitudinally in all groups during the median follow up of 63.4±32.2 months. The predicted mean values of ADNI-MEM in adiponectin groups were 0.040, –0.166, and –0.247 in low, intermediate, and high, respectively, at 31.3 months from baseline, 77.6 years of age, and educational level of 15.7 years. The predicted mean values of ADNI-EF in adiponectin groups were 0.043, –0.212, and –0.488 in low, intermediate, and high, respectively, at 31.3 months from baseline, 77.6 years of age, and educational level of 15.7 years. However, statistical difference was not observed in the reduction of longitudinal cognitive decline among adiponectin groups.

Fig. 2

Longitudinal changes in cognitive function across adiponectin levels. Apolipoprotein E ɛ4 non-carrier males with mild cognitive impairment were divided into three groups based on serum adiponectin levels: low (adiponectin level in first tertile, adiponectin < 0.67, n = 32); intermediate (adiponectin in second tertile, 0.67≤adiponectin < 0.89, n = 26); and high (adiponectin in third tertile, 0.89≤adiponectin, n = 22). Longitudinal trajectories of measured ADNI-MEM scores, predicted ADNI-MEM scores (A), measured ADNI-EF scores and predicted ADNI-EF scores (B) are shown. There were no statistical differences in the reduction of longitudinal cognitive decline among adiponectin groups. ADNI-MEM, a memory composite score calculated from the items in several memory tasks, including the Rey Auditory Verbal Learning Test; ADAS-Cog, Logical memory, and the MMSE, ADNI-EF, a executive composite score calculated from Category Fluency-animals, Category Fluency-vegetables, Trails A and B, Digit span backwards, Wechsler Adult Intelligence Scale-Revised Digit Symbol Substitution, and 5 Clock Drawing items (circle, symbol, numbers, hands, time).

DISCUSSION

In the present study, plasma adiponectin levels in elderly subjects differed across cognitive diagnosis groups. Adiponectin levels were higher in subjects with AD compared with cognitively healthy elderly subjects which was confirmed in APOE4 non-carrier males. In this group, high serum adiponectin levels correlated with older age, poor cognitive function, and high tTau and pTau levels in CSF. However, adiponectin levels measured in older subjects were not associated with future cognitive decline when considering age and educational level.

The present study was performed with APOE4 carrier status and sex stratification in the elderly ≥65 years of age. The ɛ4 allele of the apolipoprotein E gene is the main genetic risk factor for AD. APOE4 carriers have enhanced AD pathology, accelerated age-dependent cognitive decline and worse memory performance than non-carriers [15]. The difference in serum adiponectin levels based on sex has been previously reported [16] and serum adiponectin levels negatively correlate with testosterone levels [17]. On the other hand, serum adiponectin levels are positively correlated with age [16].

We demonstrated that serum adiponectin levels were significantly higher in females than in males, in agreement with previous studies. The values increased evenly in female patients regardless of diagnosis. For males, serum adiponectin levels in dementia patients were similar to the average value in females. Accumulated evidence has shown females are more susceptible to AD than males (mainly in the oldest-aged patients) [18]. This finding is not fully explained by longevity in females [19]. Women show faster cognitive decline after diagnosis of MCI [20], and those with AD have more severe neuropathology burden [21] and experience faster progression of hippocampal atrophy [22]. These findings can be explained by several factors: socioeconomic factors, such as education and occupation [23], differences in brain structure [24], brain function [25], sex hormones such as estrogen and testosterone [26], and inflammatory susceptibility [27]. The result of this study can be interpreted cautiously as evidence showing a woman’s unbalanced vulnerability to dementia.

Serum adiponectin levels were significantly higher in AD than in CN subjects and BMI was lower in AD than in CN subjects. Consistent with our findings, results in previous studies showed higher adiponectin levels in dementia subjects than in CN subjects [28, 29] and meta-analysis with 727 subjects (254 AD and 473 controls) showed higher adiponectin levels in AD patients compared with controls [30]. Adiponectin secreted by adipose tissue is paradoxically reduced in obesity, especially in visceral fat accumulation [31]. The relationship between cognitive function and BMI varies with age. People with high BMI or central obesity in midlife are twice as likely to develop dementia in later life [2]. Conversely, in older MCI patients, high BMI was associated with slower cognitive decline and weight loss was associated with faster progression [32]. Obesity is strongly associated with insulin resistance [33]. Like BMI, midlife insulin resistance is an independent risk factor for brain amyloid accumulation in the elderly; however, that association was not found in late-life insulin resistance [34]. In our study result, BMI and serum adiponectin level were negatively correlated in the APOE4 non-carrier male group but only with borderline significance. Considering the evidence to date, high adiponectin levels in elderly dementia patients can be interpreted as reflecting weight loss associated with dementia. Serum adiponectin levels did not predict the longitudinal changes in cognitive function in MCI patients. We presume that research on adiponectin in middle-age or measuring the longitudinal variation of adiponectin levels in MCI patients will provide meaningful results.

In the present study, associations were observed between serum adiponectin levels with cognitive functions and tTau and pTau proteins in CSF; however, high adiponectin levels in MCI patients did not predict rapid cognitive deterioration. The possible explanation supporting this finding is that peripheral adiponectin secretion is stimulated for protection against neurodegeneration. The pathologic hallmarks of AD are extracellular amyloid-β plaques and intracellular hyperphosphorylated tau aggregation, observed as neurofibrillary tangles [35]. Low CSF Aβ₄₂ is currently a widely accepted valid indicator of the abnormal pathologic condition associated with cerebral amyloid accumulation [36]. The amount of tTau protein in the CSF reflects the intensity of neuronal damage and neurodegeneration at a specific point [37], and elevated pTau protein in CSF is considered an indicator of abnormal pathologic state associated with AD-specific cerebral tau pathology [14]. In postmortem AD brains, adiponectin was co-localized with pTau in neurofibrillary tangles [28]. Evidence indicates serum adiponectin has neuroprotective and neurogenesis properties. Adiponectin is neuroprotective against cytotoxicity caused by Aβ in vitro [38]. Reduced adult hippocampal neurogenesis was found in adiponectin-deficient mice and intracerebroventricular treatment of adiponectin promoted adult hippocampal neurogenesis [39]. In another study, chronic adiponectin deficiency in aged adiponectin-knockout mice led to AD-like pathology and cognitive deficits, reinforcing the relevance of adiponectin in this disease [40].

The present study had several limitations. First, adiponectin is released from adipose tissue into circulation as complex multimetric isoforms comprised of full-length trimers, hexamers, high-molecular-weight multimers, and a globular fraction called globular adiponectin. The adiponectin level was analyzed in the present study; however, detailed information on the different proportions of adiponectin isomers was not available. Serum adiponectin level was measured only once at baseline, and this may not have reflected the possibility of fluctuation. Second, whether peripheral adiponectin can cross the blood-brain barrier and affect brain processes remains debatable. The adiponectin concentration in CSF is extremely low compared with peripheral tissue, and adiponectin in the central nervous system is mostly present in the trimeric form which may be important for neural function [41]. Although CSF concentration was much lower, CSF and serum adiponectin concentration were positively correlated [29], especially in males [41]. Further research is required to clarify whether adiponectin is synthesized intrathecally or flows into the intrathecal space from plasma passing through blood-brain barrier. The strength of this study is that longitudinal data of participants were analyzed from a large-prospective study cohort recruited for the dementia study and the diagnosis of MCI was limited to aMCI which is considered a prodrome of AD dementia. However, 30% of aMCI patients who progress to dementia have a primary brain pathology that is not AD [42]. The present study was based on clinical diagnosis of probable AD. Probable AD is a clinical diagnosis, and the field is moving further toward the biological definition of AD, requiring the presence of Aβ and tau pathology regardless of the presence of clinical symptoms [14]. Thus, the third limitation in the present study is the diagnosis of probable AD applied in the present study is synonymous to a multidomain amnestic dementia, which is likely enriched in subjects with AD pathologic findings; however, other pathologic characteristics could also be the primary cause. Lastly, the data used in the analysis were derived mostly from white subjects, and because the subgroup analysis was performed by stratifying study subjects based on sex and APOE4 genotype, a relatively small number of subjects were analyzed in this study. Further studies with a larger study cohort and pathologic correlation are needed to define the role of adiponectin in development and progression of dementia.

CONCLUSION

In the present cross-sectional study, AD patients were shown to have elevated serum adiponectin levels. Serum adiponectin levels were associated with cognitive function and CSF AD biomarkers among APOE4 non-carrier males. However, in longitudinal analysis, serum adiponectin levels were not associated with cognitive function outcome in subjects with MCI. The results support using serum adiponectin levels as a disease status biomarker, not a risk factor. To identify the role of adiponectin in onset and development of AD, further studies with a larger cohort in which sex and genetic information of middle-aged subjects are considered are needed.

Footnotes

ACKNOWLEDGMENTS

Data collection and sharing for this project was funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health (). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California.

Authors’ disclosures available online ().

References

Muller

, Tang

, Schupf

, Manly

, Mayeux

, Luchsinger

(2007) Metabolic syndrome and dementia risk in a multiethnic elderly cohort. Dement Geriatr Cogn Disord 24, 185–192.

Whitmer

, Gustafson

, Barrett-Connor

, Haan

, Gunderson

, Yaffe

(2008) Central obesity and increased risk of dementia more than three decades later. Neurology 71, 1057–1064.

Brochu-Gaudreau

, Rehfeldt

, Blouin

, Bordignon

, Murphy

, Palin

(2010) Adiponectin action from head to toe. Endocrine 37, 11–32.

Wang

, Scherer

(2016) Adiponectin, the past two decades. J Mol Cell Biol 8, 93–100.

Chandran

, Phillips

, Ciaraldi

, Henry

(2003) Adiponectin: More than just another fat cell hormone? Diabetes Care 26, 2442–2450.

van Himbergen

, Beiser

, Ai

, Seshadri

, Otokozawa

, Au

, Thongtang

, Wolf

, Schaefer

(2012) Biomarkers for insulin resistance and inflammation and the risk for all-cause dementia and Alzheimer disease: Results from the Framingham Heart Study. Arch Neurol 69, 594–600.

Wennberg

, Gustafson

, Hagen

, Roberts

, Knopman

, Jack

, Petersen

, Mielke

(2016) Serum adiponectin levels, neuroimaging, and cognition in the Mayo Clinic Study of Aging. J Alzheimers Dis 53, 573–581.

Roberts

, Geda

, Knopman

, Boeve

, Christianson

, Pankratz

, Kullo

, Tangalos

, Ivnik

, Petersen

(2009) Association of C-reactive protein with mild cognitive impairment. Alzheimers Dement 5, 398–405.

Teixeira

, Diniz

, Campos

, Miranda

, Rocha

, Talib

, Gattaz

, Forlenza

(2013) Decreased levels of circulating adiponectin in mild cognitive impairment and Alzheimer’s disease. Neuromolecular Med 15, 115–121.

10.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939–944.

11.

Mohs

, Knopman

, Petersen

, Ferris

, Ernesto

, Grundman

, Sano

, Bieliauskas

, Geldmacher

, Clark

, Thal

(1997) Development of cognitive instruments for use in clinical trials of antidementia drugs: Additions to the Alzheimer’s Disease Assessment Scale that broaden its scope. The Alzheimer’s Disease Cooperative Study. Alzheimer Dis Assoc Disord 11 Suppl 2, S13–21.

12.

Crane

, Carle

, Gibbons

, Insel

, Mackin

, Gross

, Jones

, Mukherjee

, Curtis

, Harvey

, Weiner

, Mungas

(2012) Development and assessment of a composite score for memory in the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Brain Imaging Behav 6, 502–516.

13.

Gibbons

, Carle

, Mackin

, Harvey

, Mukherjee

, Insel

, Curtis

, Mungas

, Crane

(2012) A composite score for executive functioning, validated in Alzheimer’s Disease Neuroimaging Initiative (ADNI) participants with baseline mild cognitive impairment. Brain Imaging Behav 6, 517–527.

14.

Jack

Jr. , Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

(2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

15.

Liu

, Liu

, Kanekiyo

, Xu

, Bu

(2013) Apolipoprotein E and Alzheimer disease: Risk, mechanisms and therapy. Nat Rev Neurol 9, 106–118.

16.

Obata

, Yamada

, Takahi

, Baden

, Saisho

, Tamba

, Yamamoto

, Umeda

, Furubayashi

, Matsuzawa

(2013) Relationship between serum adiponectin levels and age in healthy subjects and patients with type 2 diabetes. Clin Endocrinol (Oxf) 79, 204–210.

17.

Böttner

, Kratzsch

, Müller

, Kapellen

, Blüher

, Keller

, Blüher

, Kiess

(2004) Gender differences of adiponectin levels develop during the progression of puberty and are related to serum androgen levels. J Clin Endocrinol Metab 89, 4053–4061.

18.

Snyder

, Asthana

, Bain

, Brinton

, Craft

, Dubal

, Espeland

, Gatz

, Mielke

, Raber

, Rapp

, Yaffe

, Carrillo

(2016) Sex biology contributions to vulnerability to Alzheimer’s disease: A think tank convened by the Women’s Alzheimer’s Research Initiative. Alzheimers Dement 12, 1186–1196.

19.

Chêne

, Beiser

, Au

, Preis

, Wolf

, Dufouil

, Seshadri

(2015) Gender and incidence of dementia in the Framingham Heart Study from mid-adult life. Alzheimers Dement 11, 310–320.

20.

Lin

, Choudhury

, Rathakrishnan

, Marks

, Petrella

, Doraiswamy

(2015) Marked gender differences in progression of mild cognitive impairment over 8 years. Alzheimers Dement (N Y) 1, 103–110.

21.

Barnes

, Wilson

, Bienias

, Schneider

, Evans

, Bennett

(2005) Sex differences in the clinical manifestations of Alzheimer disease pathology. Arch Gen Psychiatry 62, 685–691.

22.

Ardekani

, Convit

, Bachman

(2016) Analysis of the MIRIAD data shows sex differences in hippocampal atrophy progression. J Alzheimers Dis 50, 847–857.

23.

Mielke

, Vemuri

, Rocca

(2014) Clinical epidemiology of Alzheimer’s disease: Assessing sex and gender differences. Clin Epidemiol 6, 37–48.

24.

Ruigrok

, Salimi-Khorshidi

, Lai

, Baron-Cohen

, Lombardo

, Tait

, Suckling

(2014) A meta-analysis of sex differences in human brain structure. Neurosci Biobehav Rev 39, 34–50.

25.

, Singh

(2014) Sex differences in cognitive impairment and Alzheimer’s disease. Front Neuroendocrinol 35, 385–403.

26.

Vest

, Pike

(2013) Gender, sex steroid hormones, and Alzheimer’s disease. Horm Behav 63, 301–307.

27.

Mangold

, Wronowski

, Du

, Masser

, Hadad

, Bixler

, Brucklacher

, Ford

, Sonntag

, Freeman

(2017) Sexually divergent induction of microglial-associated neuroinflammation with hippocampal aging. J Neuroinflammation 14, 141.

28.

Waragai

, Adame

, Trinh

, Sekiyama

, Takamatsu

, Une

, Masliah

, Hashimoto

(2016) Possible involvement of adiponectin, the anti-diabetes molecule, in the pathogenesis of Alzheimer’s disease. J Alzheimers Dis 52, 1453–1459.

29.

Une

, Takei

, Tomita

, Asamura

, Ohrui

, Furukawa

, Arai

(2011) Adiponectin in plasma and cerebrospinal fluid in MCI and Alzheimer’s disease. Eur J Neurol 18, 1006–1009.

30.

, Zhang

, Wang

, Jiang

, Tan

, Yu

, Tan

(2016) Peripheral blood adipokines and insulin levels in patients with Alzheimer’s disease: A replication study and meta-analysis. Curr Alzheimer Res 13, 223–233.

31.

Arita

, Kihara

, Ouchi

, Takahashi

, Maeda

, Miyagawa

, Hotta

, Shimomura

, Nakamura

, Miyaoka

, Kuriyama

, Nishida

, Yamashita

, Okubo

, Matsubara

, Muraguchi

, Ohmoto

, Funahashi

, Matsuzawa

(1999) Paradoxical decrease of an adipose-specific protein, adiponectin, in obesity. Biochem Biophys Res Commun 257, 79–83.

32.

Besser

, Gill

, Monsell

, Brenowitz

, Meranus

, Kukull

, Gustafson

(2014) Body mass index, weight change, and clinical progression in mild cognitive impairment and Alzheimer disease. Alzheimer Dis Assoc Disord 28, 36–43.

33.

Ferrannini

, Natali

, Bell

, Cavallo-Perin

, Lalic

, Mingrone

(1997) Insulin resistance and hypersecretion in obesity. European Group for the Study of Insulin Resistance (EGIR). J Clin Invest 100, 1166–1173.

34.

Ekblad

, Johansson

, Helin

, Viitanen

, Laine

, Puukka

, Jula

, Rinne

(2018) Midlife insulin resistance, APOE genotype, and late-life brain amyloid accumulation. Neurology 90, e1150–e1157.

35.

Braak

, Braak

(1991) Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol 82, 239–259.

36.

Blennow

, Mattsson

, Schöll

, Hansson

, Zetterberg

(2015) Amyloid biomarkers in Alzheimer’s disease. Trends Pharmacol Sci 36, 297–309.

37.

Blennow

, Hampel

(2003) CSF markers for incipient Alzheimer’s disease. Lancet Neurol 2, 605–613.

38.

Song

, Choi

, Whitcomb

, Kim

(2017) Adiponectin controls the apoptosis and the expression of tight junction proteins in brain endothelial cells through AdipoR1 under beta amyloid toxicity. Cell Death Dis 8, e3102.

39.

Zhang

, Wang

, Lu

(2016) Adiponectin exerts neurotrophic effects on dendritic arborization, spinogenesis, and neurogenesis of the dentate gyrus of male mice. Endocrinology 157, 2853–2869.

40.

, Cheng

, Jian

, Kwan

, Ho

, Cheng

, Yeung

, Zhou

, Hoo

, Chung

, Xu

, Lam

, Chan

(2016) Chronic adiponectin deficiency leads to Alzheimer’s disease-like cognitive impairments and pathologies through AMPK inactivation and cerebral insulin resistance in aged mice. Mol Neurodegener 11, 71.

41.

Kusminski

, McTernan

, Schraw

, Kos

, O’Hare

, Ahima

, Kumar

, Scherer

(2007) Adiponectin complexes in human cerebrospinal fluid: Distinct complex distribution from serum. Diabetologia 50, 634–642.

42.

Jicha

, Parisi

, Dickson

, Johnson

, Cha

, Ivnik

, Tangalos

, Boeve

, Knopman

, Braak

, Petersen

(2006) Neuropathologic outcome of mild cognitive impairment following progression to clinical dementia. Arch Neurol 63, 674–681.