Abstract
Background:
Alzheimer’s disease (AD), the most common form of dementia worldwide, is associated with impairment in the mechanisms of the clearing of amyloid-β within a scenario of neuroinflammation. The etiopathogenesis of the AD is unclear, but a role for viral infection is suspected to play a role in initiating the disease. We recently described a positive correlation between high titers of HSV-1-specific antibodies (Ab) and the volumes of brain regions typically affected in disease.
Objective:
The exploration of a possible role for Herpesviridae in AD was extended by analyzing HHV-6-specific humoral immunity in individuals with AD or a diagnosis of amnestic mild cognitive impairment (aMCI), a condition that is often prodromic of the development of AD.
Methods:
59 AD, 60 aMCI, and 61 age-matched healthy controls were enrolled in the study. Serum HHV-6 IgG antibody titers and avidity index were tested by ELISA. Two randomly selected subgroups of AD and aMCI in whom HHV-6 serum antibodies were detected underwent brain magnetic resonance imaging (MRI) by 1.5 T scanner.
Results:
HHV-6 seroprevalence, antibody titers, and avidity were similar in the three groups. No correlation was found between Ab titers or avidity and brain volumes, either overall or in the regions typically affected by disease.
Conclusions:
The lack of any relation between humoral immune response against HHV-6 and AD and aMCI seems to rule out a role for this virus in the pathogenesis of AD.
INTRODUCTION
Alzheimer’s disease (AD) is the most common form of dementia worldwide and is clinically characterized by progressive memory loss and impairment of other cognitive functions as well as the inability to perform daily living activities. Mild cognitive impairment (MCI) is an intermediate stage between the expected cognitive decline of normal aging and the pathologic decline of dementia. Although MCI can present with a variety of symptoms, when memory loss is the predominant symptom, it is termed amnestic MCI (aMCI) and is frequently seen as a prodromal stage of AD, with an annual conversion rate from aMCI to AD of 15% [1]. Although the abnormal features and lesions of AD brain are well studied and characterized, the etiopathogenesis of the disease is still unclear. Recent data lend support to a possible role of human herpes simplex virus type 1 (HSV-1) infection in AD, at least as a factor that, among others–genetics, environmental, immune dysfunction–could predispose to the disease [2, 3], a concept proposed for the first time decades ago [4].
We have focused on studying the relation between HSV-1 antibodies and cortical grey matter (GM)volumes of AD and aMCI patients. We found that high antibody titers are significantly more frequent in patients compared to age-matched healthy controls (HC), and correlate positively with the volumes of brain regions typically affected in AD as well as in HSV-1-associated acute encephalitis [5, 6].
Human herpes virus type 6 (HHV-6) has been suggested to be involved in the pathogenesis of AD as well. This double stranded DNA virus can infect the brain [7, 8], and, interestingly, HHV-6- as well asHSV-1-associated encephalitis, predominantly interest the fronto-temporal region of the brain: the same brain area that is most commonly affected in AD [8]. In particular, two different studies found a higher frequency of HHV-6 DNA in brain of AD patients than in those of age-matched non-demented controls [9, 10], although whether its presence is a cause or just an effect of the disease is still uncertain.
We investigated the possible relationships between HHV-6-specific antibodies (IgG) and grey matter volumes, as measured by magnetic resonance imaging (MRI), in AD and aMCI individuals.
MATERIAL AND METHODS
Patients and controls
One-hundred-eighty individuals were recruited: 59 AD patients, 60 aMCI individuals, and 61 gender- and age HC. All subjects were recruited at the Don C. Gnocchi Foundation – ONLUS, Milan, Italy. Patients were diagnosed with probable AD according to the NINCDS-ADRDA criteria [11] or with aMCI according to Petersen criteria [1]. AD patients were in mild stage of the disease as determined by both Clinical Dementia Rating (CDR) [12] scale (CDR range: 0.5–1.5) and Mini-Mental State Examination (MMSE) [13] score (MMSE mean ± standard deviation: 20.3 ± 3.0). aMCI individuals were required to meet the Grundman operational criteria [14]: memory complaint, confirmed by an informant; abnormal memory function, documented by previous extensive neuropsychological evaluation; normal general cognitive function, as determined by both CDR scale (CDR with at least a 0.5 in the memory domain and MMSE score ≥24); no impairment in functional activities of daily living as determined by clinical interview with the patient and informant; no significant cerebral vascular disease (Hachinski score ≤4; [15]); no major psychiatric illnesses with particular attention to exclude subjects with history of depression (Hamilton Depression Rating Scale score ≤12; [16]). All aMCI individuals performed the Montreal Cognitive Assessment (MoCA) test [17]. HCs were selected according to the SENIEUR protocol for immune-gerontological studies of European Community’s Control Action Program on Aging and were unrelated healthy spouses of AD and MCI patients [18]. The cognitive status of HC was assessed by administration of MMSE (score for inclusion as normal control subjects >28). The study conformed to the ethical principles of the Declaration of Helsinki; all subjects or their caregivers gave informed and written consent according to a protocol approved by the local ethics committee of the Don C. Gnocchi Foundation – ONLUS, Milan, Italy. For each subject, whole blood and serum samples were collected.
Immunological analyses
HHV-6 serum IgG titers were measured using commercial enzyme immunoassays (HHV-6 IgG ELISA kit, Abnova, Taiwan). HHV-6 antibody (Ab) titers were expressed as positive index (PI), calculated by dividing optical density (OD) measurement generated from the assay by OD of a cut-off calibrator. PI <0.90 is considered seronegativity, PI >1.10 is considered seropositivity. Values between 0.90–1.10 PI are considered equivocal and have to be discarded by analysis.
HHV-6 specific IgG avidity was measured with a protein-denaturing agent: the protocol used is the protocol for ELISA analysis, with the addition of 6 M urea to the washing solution at the washing step after plasma reaction. The avidity index (indicated as %) was calculated as follow: anti-HHV-6 Ab titer measured with washing including urea/anti-HHV-6 Ab titer measured with washing without urea. HSV-1 and cytomegalovirus (CMV) serum IgG titers and avidity were also analyzed in the same individuals to confirm our previous results. Commercial kit (BEIA HSV-1 IgG and BEIA CMV Quant, Technogenetics, Milan, Italy) were used following previously published protocols [5, 6]. For the HSV-1 titers, the unit of measure is antibody index (AI): AI >1.1 is considered positive; AI <0.9 is considered negative. For the CMV titers, the unity of measure is arbitrary unit/ml (AU/ml): AU/ml >10 is considered positive; AU/ml <8 is considered negative. Values in the equivocal range were discarded. For HSV-1 and CMV avidity, the same protocol as the one used for HHV-6 avidity was followed.
MRI analyses
Two randomly selected subgroups of HHV-6 seropositive AD (n = 44) and aMCI (n = 23) patients underwent MRI imaging acquisition (1.5 T scanner, Siemens Magnetom Avanto, Erlange, Germany) including conventional sequences (FLAIR, T2) and a high-resolution 3D-T1 image (TR/TE = 1900/3.37 ms, matrix size = 192×256, isometric voxel = 1×1×1; number of slices = 176) to investigate brain structural features. Using Voxel Brain Morphometry (VBM), part of SPM8 (http://www.fil.ion.ucl.ac.uk/spm/). MRI high-resolution 3D-T1 images of AD and aMCI patients were then included in a VBM analysis. The VBM analysis was conducted using VBM8 (http://dbm.neuro.uni-jena.de/vbm/) according to the Unified Method [19]. After GM segmentation all images underwent spatial smoothing using a Gaussian kernel (FWHM 8 mm). We directly compared AD and aMCI patients with a template of age-matched HC, to investigate the pattern of brain atrophy (ANOVA). The relationship between HHV-6 and brain features has been investigated in both subgroups of AD and aMCI patients, using separate design matrices. In more detail, we investigated the relationship between cortical volume and HHV-6 specific Ab (n = 44 AD; n = 23 aMCI) and between cortical volume and HHV-6 IgG avidity index (n = 34 AD; n = 21 aMCI). Normalized total intracranial volume and MMSE score have been included in these second-level analyses as covariates of no interest to remove the confounding effects of different brain sizes and different cognitive levels. Results have been considered as statistically significant if surviving p < 0.05FWE-corrected threshold.
ApoE genotyping
Genomic DNA was isolated from whole blood from each subject by phenol-chloroform extraction. Customer-design Taqman probes for the 112 and 158 codons were used to determine the genotype of apolipoprotein E gene (APOE). Primers and probes for the 112 codon are: 112 Forward primer: 5′-GGG CGC GGA CAT GGA G-3′; 112 Reverse primer: 3′-TCC TCG GTG CTC TGG CC-5′; 112 Arg Probe: 5′-CGT GCG CGG CCG-3′-FAM; 112 Cys Probe: 5′-ACG TGT GCG GCC GCC TG-3′-VIC. Primers and probes for the 158 codon are: 158 Forward primer: 5′-CCG CGA TGC CGA TG-3′; 158 Reverse primer: 3′-GCT CGG CGC CCT CG-5′; 158 Arg probe: 5′-CCT GCA GAA GCG CCT GGC A-3′-FAM; 158 Cys probe: 5′-CCT GCA GAA GGG CCT GGG AGT-3′-VIC.
Statistical analysis
The parametric data are expressed as mean ± standard deviation, whereas the non parametric data as median and interquartile range (IQR). AD, aMCI, and HC were compared on democgraphic data using Chi-square test and One-way ANOVA with Bonferroni post hoc test for categorical and continuous variables, respectively. Differences in immunological data among groups were tested using Kruskal-Wallis test and, when appropriate, the Mann Whitney U test, and the correlations using Spearman’s correlation coefficient. P-values corresponding to ≤0.05 are described as statistically significant in the text. For the MRI analysis, results were considered as significant if surviving p < 0.05 after FWE correction for multiple comparisons. The statistical analyses were accomplished using commercial software (MedCalc Statistical Software version 14.10.2, Ostend, Belgium).
RESULTS
Demographical and clinical characteristics of the subjects
Gender, educational level, and age were comparable in the three groups examined; global cognitive levels (MMSE) were, as per definition, significantly reduced in AD and aMCI compared to HC (p < 0.0001) and lower in AD than aMCI (p < 0.0001). As expected, AD patients had a higher frequency of APOE ɛ4 variant compared to HC (p = 0.002), whereas no differences were found between aMCI and HC, as well as between AD and aMCI. Demographic and clinical characteristic of the individuals enrolled in the study are summarized in Table 1.
Viral seroprevalence, titers, and avidity
No differences were observed between the three groups when HHV-6, HSV-1, and CMVseroprevalence was evaluated (Table 2); seropositivity for at least two of the three analyzed viruses was present in all the individuals analyzed (Table 3). HHV-6 Ab titers were comparable in AD patients (2.5, 1.8–3.3 PI), aMCI (2.9, 2.2–3.6) and HC (2.8, 1.7–3.8) (Fig. 1). HHV-6 Ab avidity was comparable between the three groups as well (AD = 100, 94–100% ; aMCI = 100, 85–100% ; HC = 100, 96–100%). No correlation was found between the titers and the avidity of Ab for HHV-6, CMV or HSV-1.
Results confirmed that higher HSV-1 titers are seen in AD and aMCI individuals compared to HC (AD: 9.1, 7.1–10.3 AI; aMCI: 8.4, 6.1–9.6 AI; (HC 6.0, 4.5–8.2 AI); (AD versus HC: p = 0.0001; aMCI versus HC: p = 0.002) (Fig. 2A), whereas no differences were found in CMV antibody titers (AD: 114.8, 88.1–136.1 AU/ml; aMCI: 94.9, 68.7–135.9 AU/ml; HC: 97.8, 63.1–146.4; Fig. 2B). No differences were found regarding the avidity of these two viruses in any of the analyzed groups (Table 2). Finally, no correlations were found between serological data and either MoCA or MMSE scores.
MRI results and correlations
Direct comparison between groups (AD, aMCI, HC) showed the presence of the typical brain atrophy associated with AD and aMCI pathologies. As expected, AD subjects presented with statistically significant decreased GM volumes in medial temporal lobe areas (left >right) when compared with both aMCI subjects and HC.
Correlation analyses, modeled separately for AD and aMCI subjects, did not reveal correlations surviving statistical threshold between cortical volume and neither HHV-6 specific Ab nor HHV-6 IgG avidity index.
DISCUSSION
Herpesviridae are a complex family of DNA viruses that have been suggested to play a role in the development of AD, findings are nevertheless contradictory and not univocal. Recent analyses focused on HSV-1 [20]. Our data, in particular, have shown a positive correlation between anti-HSV-1 Ab and grey matter brain volume in AD [5, 6]. Because neurodegenerative diseases, such as AD, are associated with alterations of the blood-brain barrier [21, 22] that facilitate the transit of peripheral blood cells into the cerebrospinal fluid, enhancing immune cell entry, our data suggest that HSV-1-specific antibodies might play a protective role at early stage of the disease, reducing the viral activity in brain regions where the blood-brain barrier is damaged.
To verify whether a relationship between Ab titers and brain volume parameters could be observed for herpesviruses other than HSV-1, we analyzed Ab titers and avidity of another herpesvirus, HHV-6, correlating the results with those obtained in MRI. HHV-6 was chosen because of its prevalence and because this virus was recently suggested to be associated with the pathogenesis of another neurodegenerative disease: multiple sclerosis [23].
The possible role of HHV-6 in AD is highly controversial: thus, although a high frequency of HHV-6 genome was described to be present in AD brains [9], HHV-6 DNA is also observed frequently in the brains of non demented elderly individuals [7, 24]. This virus, therefore, seems to be endowed with the ability to invade the central nervous system without provoking any sign of infection and to remain clinically asymptomatic infection in brain tissue in a latent form also in healthy subjects.
Results herein indicate that HHV-6-specific antibody titers do not differ when AD and aMCI individuals are compared to HC; notably, no correlations were detected as well between HHV-6 Ab titers and brain volume as detected by MRI in either AD or MCI individuals. These findings seem to indicate that a role for HHV-6 in the pathogenesis of AD is highly unlikely, although it cannot be ruled out that a latent HHV-6 brain infection might activate (or be activated by) other viruses that could be involved in thedisease.
To our knowledge, these are the first data that analyze possible relationships between HHV-6 antibody titers and brain volume parameters in AD and aMCI. An older study analyzed the intrathecal HHV-6 antibody synthesis in a limited number of AD patients (18), finding it only in 4 patients [10]. Two more recent analyses focused on HHV-6 humoral response alone in AD in relation with particular genetic pattern, with no significant difference being observed in AD patients [25, 26]. Importantly, no data were available on HHV-6 antibodies in aMCI, a condition that is often prodromic to the development of full blown AD.
Antibody responses against another herpesvirus, CMV, were analyzed as well in our patients and controls. CMV establishes a chronic and usually asymptomatic infection in the host that can be reactivated in times of stress or in the presence of immunosuppression due to chronic illness or medication [27, 28]; importantly, the risk of reactivation increases during old age [29]. Several studies have examined the possible relation between CMV infection and dementia, and also in this case results are controversial [30–36]. The results of the present study are in line with our previous work [5], in which no relation were observed between CMV humoral response and brain damage either in AD or in aMCI individuals.
Finally, to verify results of our previous analyses, HSV-1 specific Ab were evaluated as well in these individuals. Results confirmed that HSV-1 antibody levels were significantly increased both in AD and aMCI individuals compared to HC, confirming the observation that HSV-1 infection does occur in AD and in aMCI, and that a robust immune response is elicited by such infection not only in symptomatic AD, but also in the early phase of the cognitive impairment decline, before the diagnosis of the disease.
Correlating the viral and MRI data with cerebrospinal fluid levels of biomarkers for AD diagnosis would be very interesting, but unfortunately, for ethical reasons the collection of this type of samples was not possible.
In conclusion, these findings, although needing to be confirmed by independent studies and in patients with other neurological conditions, reinforce the hypothesis that HSV-1 infection, but not infection with other herpesviruses, play a role in the pathogenesis of AD.
