Antioxidant Status and APOE Genotype As Susceptibility Factors for Neurodegeneration in Alzheimer's Disease and Vascular Dementia

Abstract

Different factors interact to develop neurodegeneration in patients with dementia and other neurodegenerative disorders. Oxidative stress and the ε4 allele of apolipoprotein E (ApoE) are associated with significant alteration in lipid metabolism, in turn connected to a variety of neurodegenerative diseases and aging. Thus, a better understanding of the pathogenetic pathways associated with lipid dyshomeostasis may elucidate the causes of neurodegenerative processes. To address this issue, we evaluated the effects of antioxidant status and APOE genotype on neurodegeneration in patients with dementia of the Alzheimer type (AD), with vascular dementia (VaD), and in elderly healthy controls. Eighty-two AD, 42 VaD patients, and 26 healthy controls were recruited and underwent medial temporal lobe atrophy (MTA) assessment, white matter hyperintensities rating (WMH), serum total antioxidant status assaying (TAS), and APOE genotyping. A logistic regression algorithm applied to our data revealed that a 0.01 mmol/L decrease of TAS concentration increased the probability of MTA by 24% (p=0.038) and that carriers of the APOE ε4 allele showed higher WMH scores (p=0.018), confirming that small variations in antioxidant systems homeostasis are associated with relevant modifications of disease risk. Furthermore, in individuals with analogous TAS values, the presence of the ε4 allele increased the predicted probability of having MTA. These outcomes further sustain the interaction of oxidative stress and APOE genotype to neurodegeneration.

Introduction

Atrophy of discrete and distributed areas of the brain is a well-known effect of the progressive loss of neurons and synapses leading to the clinical phenotype of dementia. It is a classic marker of Alzheimer's disease (AD),^1,2 but it is also seen in patients affected by dementia of vascular origin (VaD), the second most common cause of acquired dementia in western countries,³ as well as in some healthy elderly individuals.⁴ Although the regions typically affected by atrophy are distinguished between the two forms of dementia, especially in the earliest phases, it is not yet possible to exclude that the biological mechanisms underlying them are different as well.⁵ Several authors have recently described how atrophy of the medial temporal lobe structures (medial temporal lobe atrophy, MTA), including the hippocampal formation, the parahippocampal gyrus, and the entorhinal cortex, effectively contributes to cognitive impairment.⁶ On the other hand, a “vascular hypothesis” has become the object of intense investigation and debate, as researchers share the belief that phenomena originating in the white matter play important roles in the pathogenesis of both AD and VaD.^7–9

A number of conditions seem to contribute to white matter damage, such as micro- or macro-angiopathy, residual gliosis, diabetes, atherosclerosis, elevated blood pressure, increased homocysteine levels, and general disruptions of oxidative homeostasis.¹⁰ In AD, a major pathogenic role is probably played by impaired energy metabolism and oxidative stress.⁹ Studies carried out so far on the interplay among diet, cognition, and dementia delivered mixed results. Observational studies have suggested an association among these elements,¹¹ and clinical trials have demonstrated only modest but consistent effect of antioxidants on cognition.¹² Furthermore, different studies revealed that functional polymorphisms in anti-oxidant genes might be a risk factor for AD development.^13,14 Nevertheless, the question of whether the anti-oxidant status has a relationship with the rate of susceptibility to dementia and specifically with typical signs of neurodegenerative or neurovascular damage have been still poorly investigated.¹⁵

Another important pathogenetic pathway leading neurodegenerative processes is the APOE genotype. Apolipoprotein E (ApoE) has a key role in the regulation of the metabolism of low-density lipoproteins (LDL), cholesterol, and triglycerides.¹⁶ This protein acts as a ligand for the LDL receptor and is also involved in the maintenance and repair of neuronal cell membranes in the central and peripheral nervous systems.¹⁷ The ε4 allele of the APOE gene is associated with a variety of complex and age-related disorders.¹⁸ Moreover, the APOE ε4 alllele has been shown to play an important role in antioxidant status.^19,20 Furthermore, both oxidative stress and APOE ε4 have been shown to have significant associations with lipid dyshomeostasis.^21,22 Lipid metabolism plays a key role in the brain, because this organ has the highest concentration of lipids, except for adipose tissue, and several studies have demonstrated that an alteration in lipid homeostasis is involved and associated with a variety of neurodegenerative diseases and aging.²²

To verify the interactive effects on susceptibility to neurodegeneration of both oxidative stress and ApoE, we have analyzed total antioxidant status (TAS) and APOE genotype in 150 individuals—124 of whom affected by diverse grade and type of dementia—and also evaluated their associations with the outcomes of brain imaging measures.

Materials and Methods

The study was approved by the Ethical Committee of the Fatebenefratelli Hospital and was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. All subjects signed an informed consent prior to their inclusion in the study.

Patients

Between 2005 and 2007, 180 individuals with a diagnosis of suspected or ascertained dementia entered a screening phase at the Fatebenefratelli Hospital in Rome. Of them, 124 were eventually enrolled in the study: 82 (66 women) diagnosed as AD and 42 (21 women) as VaD, according to National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA)²³ and National Institute of Neurological Disorders and Stroke and Association Internationale pour la Recherché et l'Enseignement en Neurosciences (NINDS-AIREN)²⁴ criteria, respectively. A sample of 26 healthy controls (14 women), mostly spouses of patients and a few individuals complaining about subjective memory difficulties without clinical confirmation of MCI, was recruited in parallel to assess normative values of MTA, deep-seated ischemic lesions, and TAS. The exclusion criteria included: (1) Evidence of concomitant dementia of different types, such as frontotemporal, reversible dementias (including pseudo-depressive dementia), fluctuations in cognitive performance, and/or features of mixed dementias; (2) evidence of concomitant extra-pyramidal symptoms; (3) clinical and indirect evidence of depression as revealed by Geriatric Depression Scale scores higher than 13; (4) other psychiatric diseases, epilepsy, drug addiction, alcohol dependence, and use of drugs strongly interfering with the cytochrome c oxidase system; and (5) current or previous uncontrolled or complicated systemic diseases (including diabetes mellitus) or traumatic brain injuries.

All 150 people recruited underwent physical and neurological examination including the Mini-Mental State Examination (MMSE), laboratory and psychometric tests, and brain magnetic resonance imaging (MRI). Medical histories showed that no patients had ever suffered from major illnesses in the past. They were seen again after 12 months by a multidisciplinary team of neurologists, neuropsychologists, and radiologists to check for condition and to confirm the initial diagnosis. Brain MRI was also used to rule out territorial strokes or other major neurological conditions, such as hemorrhage or tumor. MTA scores were not used for final diagnoses. This subject sample partially overlaps previous studies.^25,26

MRI data evaluation

The MRI examination was performed on a 1.5-Tesla MR scanner operating with a quadrature head coil (Achieva, Philips Medical Systems, Best, The Netherlands) and included the acquisition of standard clinical sequences aimed at assessing brain parenchyma integrity and vascular damage (dual T2W spin-echo sequence acquired on axial plane), as well as putative cortical atrophy (T1W spin-echo). The plane of acquisition encompassed the bi-commisural line; all images had an in-plane resolution of 256×256, were contiguous, and were 3 mm thick. All MR images were entered on the digital server of the department and were independently examined by two experienced neuroradiologists, who were blinded to clinical diagnosis (inter-rater reliability=0.98). As mentioned above, this procedure was repeated after 12 months. Atrophy and white matter lesions were both graded according to standardized and widely accepted visual rating scales on plain MRI.^1,27,28 The degree of MTA was assessed with a ranking procedure (5-point rating scale)²⁷ and validated by linear measurements of the medial temporal lobe structures, including the hippocampal formation and surrounding spaces occupied by cerebrospinal fluid.

The visual rating scale of WMH included both anatomic distribution and lesion severity. On the basis of the anatomic distribution, a distinction was made between areas of periventricular hyperintensities (PVH, caps and rims), deep-seated hyperintensities (DWMH, which included frontal, parieto-occipital, and temporal white matter regions), and basal ganglia (BGH). Large vessel cortical infarcts in the territories of anterior, posterior, and middle cerebral artery, as well as infratentorial regions, were also evaluated. The presence of mass lesions and lobar hemorrhage were additional exclusion criteria from the study.^29,30 Global brain atrophy (ventricular and sulcal atrophy) was rated as present (1) or absent (0).

Biological analyses

Serum from fasting peripheral blood samples was collected in the morning and rapidly stored at −80°C. Total reactive anti-oxidant potential (TRAP) was assayed by the TAS kit (Randox Laboratories, Crumlin, UK), based on the method of Rice-Evans and Miller.^31,32 The assay is based on the quenching by the antioxidants present in serum of the radical cation of 2′2-azinobis-(3-ethyl-benzothiazoline-6-sulfonic acid) (ABTS), produced by the reaction between ABTS with the free radical ferryl myoglobin. In turn, ferryl myoglobin is produced by the reaction between metmyoglobin and hydrogen peroxide (H₂O₂). The relative ability of antioxidants to scavenge the ABTS^•+ was compared to that of a known amount of 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox), a synthetic antioxidant. The reference range in serum is 1.30–1.77 mmol/L.³²

APOE genotyping was performed according to standardized procedures based on restriction isotyping, which uses oligonucleotides to amplify APOE gene sequences containing amino acid positions 112 and 158. The amplification products were digested with HhaI and subjected to polyacrylamide gels electrophoresis.³³

Statistical analysis

A chi-squared or Fisher exact test, when necessary, was used to compare frequencies between groups. The Shapiro–Wilk test was used to evaluate if data were normally distributed. The Student t-test was used to compare continuous variables normally distributed. Mann–Whitney tests were used in the case of lack of normal distribution and to compare neuropsychological variables among AD and VaD groups. A logistic regression analysis on MTA (binomial variable, absence or presence) was used to evaluate the influence of biological, demographic, and clinical variables, according to the forward method (0.08 included, 0.10 excluded). This regression was only made on AD and VaD subjects in which MTA was the dependent variable and TAS, APOE, and age were the independent variables. PVH, DWMH, and BGH scores were collapsed into a single factor defining the white matter hyperintensity (WMH) through principal component analysis. This mathematical algorithm, through an orthogonal transformation, converts a set of observations of possibly correlated variables into a set of values of linearly un-correlated variables called principal components, whose number is less than or equal to the number of original variables.³⁴ Linear regression analysis with forward method was used to evaluate the effects of TAS, APOE, and demographic variables on WMH (quantitative variable). A p value<0.05 was considered as statistically significant. All analyses were performed on SPSS 17.0 software (SPSS Inc., Chicago, IL).

Results

The demographic and clinical characteristics of the study populations are reported in Table 1. There were no significant age differences between the AD, VaD patients, and controls. Average MMSE scores were 18.8±5.4 in AD and 22.6±4.4 in VaD patients (p=0.004). At least one copy of the APOE ε4 allele was found in 37.2% of the AD and 30.8% of the VaD patients, without statistical differences between the two groups, and in 8% of the control (p=0.022). Table 2 reports neuroradiological evaluations of patients with AD and VaD and healthy elderly controls. MTA was observed in 51.2% of the AD patients, 40.5% of the VaD patients, without differences between the two groups (p=0.257), and in 8% of controls group (p=0.001). Control subjects showing MTA were non-APOE ε4 carriers. WMH in VaD patients was 2.93 standard deviations (SD) higher than in controls. The AD and VaD patients did not differ in global atrophy (p=0.464) or APOE ε4 frequency (p=0.493), but significant differences among the groups are present in TAS levels, which were lower in the AD group (p=0.049). No significant differences in TAS levels were observed between carriers and non-carriers of the APOE ε4 allele, considering both the whole population under study and the different disease stages. The effects of clinical, biological, and demographic variables upon MTA were explored with a logistic regression analysis. TAS (p=0.038), APOE (p=0.079), and age (p=0.032), but not diagnosis, were the only variables that entered the final model, showing a significant effect on MTA. Specifically, TAS level and MTA appeared statistically linked by an inverse relationship: The logistic regression analysis showed that a 0.01 mmol/L decrease in TAS concentration determined a 24% increase in probability of having MTA. The APOE ε4 allele also seemed to contribute to MTA, although close to significance. In fact, Fig. 1 shows that for similar TAS values, the presence of the ε4 allele increased the predicted probability of having MTA. A linear regression model was built to explore the correlations of TAS levels and APOE genotype on the neurovascular burden as expressed by the WMH index. The APOE ε4 allele seemed to correlate to cerebral white matter damage because carriers of the APOE ε4 allele had higher WMH score values (p=0.018).

Table 1.

Demographic and Clinical Characteristics of the Subjects Included in the Present Study

	AD (n=82)	VaD (n=42)	HC (n=26)
Female, n (%)	66 (80.5)	21 (50)	14 (53.8)
Age (years), mean (SD)	75.4 (1.4)	76.6 (8.3)	71.2 (8.9)
MMSE, mean (SD) Disease severity	18.8 (5.4)	22.6 (4.4)	28.3 (1.3)
Moderate, n (%)	37 (45)	28 (67)	—
Severe, n (%)	41 (50)	14 (33)	—
Mild, n (%)	4 (5)	—	—
APOE ε4 carriers, n (%)	29 (37.2)	12 (30.8)	2 (8)
TAS (mmol/L), mean (SD)	1.3 (0.2)	1.4 (0.2)	1.4 (0.1)

AD, Alzheimer disease; VaD, vascular dementia; HC, healthy controls; n, total count; (%), ’n’ in percentage; SD, standard deviation; TAS, total antioxidant status (expressed in mmol/L); MMSE, Mini-Mental State Examination; APOE, apolipoprotein E.

Table 2.

Neuroradiological Variables of the Study Population

	AD (n=82)	VaD (n=42)	HC (n=26)
MTA (present), n (%)	42 (51.2)	17 (40.5)	2 (8)
PVH, mean (SD)	1.6 (2.2)	4.5 (3.4)	0.7 (1.1)
DWMH, mean (SD)	2.5 (5)	9.8 (9.2)	2.3 (3.3)
BGH, mean (SD)	1.1 (2.4)	3.9 (6.1)	0.5 (1.3)

MTA, Medial temporal atrophy; PVH, periventricular hyperintensities; DWMH, deep white matter hyperintensities; BGH, basal ganglia hyperintensities; n, total count; (%), ’n’ in percentage; SD, standard deviation.

FIG. 1.

Predicted probability of medial temporal lobe atrophy (MTA) derived from multivariate logistic regression analysis upon serum concentration of total antioxidant status (TAS) and APOE ε4 genotype, considering age as a confounding factor. (Triangles) APOE ε4 allele carriers (dotted line is corresponding average); (circles) represent APOE ε4 allele non-carriers (continuous line is corresponding average). The model revealed that for similar TAS values, the APOE ε4 allele raised the probability of having MTA.

Discussion

Neurodegeneration is a pathologic process referred to as loss of structure and function in neurons, eventually leading to neuronal death. It is common in several neurodegenerative diseases, and many pathogenetic pathways appear to be shared among them. Improved knowledge, even on factors conferring susceptibility to such disarranged processes, can offer hope for therapeutic advances to researchers and novel opportunities to patients.³⁵

Oxidative stress is strongly linked to neuronal death and networking dysfunctions within the brain. Specifically, oxidative stress–related damage of polyunsaturated fatty acids in biological membranes may partially explain this pathogenic association.³⁶ ApoE is strongly involved in the maintenance and repair of neuronal cell membranes both in central and peripheral nervous systems, and the APOE ε4 allele causes functional changes on the protein that impact its physiological properties.¹⁷ From the above, some interactive mechanism between oxidative stress susceptibility and APOE genotype toward neurodegeneration may exist.

Studies on activity of antioxidant enzymes in blood of AD patients revealed mixed and non-conclusive results: So far Cu/Zn superoxide dismutase 1 activity, for example, has been described to be not changed across numerous studies.^37,38 It was the same for several investigations on glutathione peroxidase or reductase and catalase activity in AD.^37,39,40 In contrast, data on the analysis of total antioxidant defense were more univocal through the literature. The TAS assay usually works as an indirect measure of the resistance of serum to be oxidized by an oxidant (see Materials and Methods). It examines the total antioxidant status of serum, measuring the suppression of the signal inducted from an external radical reagent, and this suppression is performed by all the antioxidants in the bloodstream, compounding either enzymatic or non-enzymatic antioxidant species present in general circulation. Uric acid and ascorbic acid (along with albumin) are the major soluble contributors among the non-enzymatic antioxidants to serum TAS, whereas vitamin A, tocopherols/tocotrienols (vitamin E), and α- and β-carotene are lipid soluble and key mediators for limiting lipid oxidation. Decreased TAS in serum of AD has been demonstrated in numerous studies, suggesting an overall decrease in AD,^41,42 even though antioxidant enzymes solely were found unchanged. However, the notion that antioxidant capacity is significantly elevated in AD and directly related to disease severity has been highlighted also in a recent post-mortem case–control study.⁴³ Further study will be necessary to elucidate this issue.

In line with the results of living AD patients, we found decreased levels of TAS in AD independent of APOE genotype, and to deepen such an aspect we also correlated TAS decrements to outcomes of an empirical and affordable evaluation of temporal atrophy, as revealed by visual inspection of the digital brain images. Specifically, our results revealed that MTA was present in a higher percentage of AD patients with respect to VaD and healthy controls, and that variations of MTA score are better explained on the basis of lower TAS levels. As a result, small decreases of TAS can account for a 24% increase of the probability of having MTA, as disclosed by the regression model developed ad hoc.

These outcomes highlight that small variations in antioxidant systems are associated with relevant modifications of disease risk, as reported also by other authors.^44–46 The above-mentioned TAS evaluation can be assumed to be the expression of the overall contribution of both non-enzymatic and endogenous enzymatic antioxidants systems in serum. For example, glutathione, catalase, including also dietary assumption of antioxidant nutrients (often evaluated as patients survey),⁴⁷ and metal proteins, specifically the ceruloplasmin–transferrin system,^26,48 are considered to be the main antioxidant system in serum. For similar TAS values, we observed that the ε4 allele of APOE increased the predicted probability of having MTA, confirming our hypothesis that increased susceptibility to oxidative stress and dysfunction in APOE partake in susceptibility to neurodegeneration. More specifically, these results suggest that two independent factors interact: A systemic antioxidant derangement, in concert with defective repair mechanisms of cell membrane upon cholesterol transport (as expressed by the APOE ε4 allele), worsens the susceptibility to neurodegeneration, further supporting a generally agreed upon scenario of an oxidative stress that hampers neuronal functionality in brain areas devoted to memory circuits, threatening their viability or survival.³⁰

Another facet of our study is that the APOE ε4 allele was associated with WMH scores. The degree to which the association of ε4 with dementia is mediated by vascular lesions in comparison with AD lesions is controversial.⁴⁹ Indeed, some studies hypothesized that the APOE genotype has a complex role in the vascular damage of AD patients.^50,51 The outcome of the present study seems to confirm the correlation of the APOE ε4 allele with the severity of cerebral amyloid angiopathy (CAA), commonly encountered in the walls of small and medium-sized vessels of the cerebral cortex and leptomeninges of most AD brains.⁵ However, the possibility that vascular lesions play a role in mediating the effects of the ε4 allele has been raised by studies showing associations between APOE ε4 and increased risk of VaD. Although studies on patients with clinically diagnosed VaD have been equivocal with regard to associations with APOE ε4,^52,53 an increased risk for ischemic and hemorrhagic stroke, as well as cardiovascular disease are widely recognized.⁵⁴

Besides this single-variable analysis, our main outcome is that TAS and APOE genotype seem to interact in damaging the brain. Although these results point to new insights in the pathogenetic basis of neurodegeneration, further studies are needed to validate our data and to explain these potential pathogenic connections. The confirmation of our hypothesis may furnish useful information for developing preventive and therapeutic strategies against neurodegeneration.

Acknowledgments

The authors are grateful to patients and their caregivers for irreplaceable support. The present work received funding from: ERA-Net NEURON [JTC 2008 nEUROsyn Grant]; Italian Health Department Grant “Profilo Biologico e Genetico della Disfunzione dei Metalli nella Malattia di Alzheimer e nel ‘Mild Cognitive Impairment” (RF 2006 conv.58).

Author Disclosure Statement

No competing financial interests exist.