KIR2DS2/KIR2DL2/HLA-C1 Haplotype Is Associated with Alzheimer’s Disease: Implication for the Role of Herpesvirus Infections

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder, where neuroinflammation and immune cells are key pathological factors. Recently, it was suggested a possible association between AD and human herpesvirus 6 (HHV-6) infection. Since we recently observed that multiple sclerosis patients with KIR2DL2 expression on natural killer (NK) cells are more susceptible to herpesvirus infection, we tested the possible implication of KIR/HLA genetic for HHV-6A infection. We identified, for the first time, a possible implication of a specific KIR/HLA subset in AD. The combination KIR2DS2/KIR2DL2/C1 correlated with a lower MMSED_i score, representative of a severe AD status and an increased susceptibility to HHV-6A infection. Therefore, the results seem to converge on the hypothesis that herpesvirus infection might play a role in AD. If this hypothesis finds experimental confirmation, a new therapeutic strategy, modulating KIR2DL2 expression on NK cells, for AD might be envisaged.

Keywords

Alzheimer’s disease herpesvirus HHV-6A KIR NK cell

INTRODUCTION

Alzheimer’s disease (AD) is the most common neurodegenerative disease worldwide, representing 60–70% of all dementia cases [1]. Neuroinflammation is a key factor in AD pathology, and immune cells play a functional role in modulating soluble molecules including neuroendocrines, neurotransmitters, and cytokines [2, 3]. Several studies have reported that the presence of proinflammatory cytokines and lymphocyte subset distribution can be disturbed in AD [2, 3]. Natural killer (NK) cells are key effectors of the immune system, due to their human leukocyte antigen (HLA)-restricted cytotoxic activity. NK inhibitory receptors prevent NK cell-mediated attack against normal cells whereas cells with compromised HLA class I expression (e.g., by tumor transformation or viral infection) are susceptible to NK cell-mediated killing. There are two main types of NK inhibitory receptors: 1) killer Ig-like receptors (KIRs), specific for determinants shared by groups of HLA-A, HLA-B, or HLA-C allotypes [4] and 2) CD94/NKG2A, recognizing HLA-E [5]. Inhibitory KIRs have long cytoplasmic tails (KIR-L) containing two immunoreceptor tyrosine-based inhibition motifs (ITIMs) that, upon phosphorylation, shut down NK cell activation. Each KIR displays two (KIR2DL) or three (KIR3DL) extracellular Ig-domains conferring specificity for HLA allotypes. The inhibitory KIR2DL1, KIR2DL2, and KIR2DL3 receptors recognize distinct HLA-C allotypes, KIR3DL1 ligands are HLA-B molecules with the Bw4 epitope, and KIR3DL2 binds HLA-A3 and HLA-A11.

Although the immunobiology of NK cells in neurodegenerative diseases such as multiple sclerosis (MS) is well studied, little is known regarding their role in AD [3]. NK cells have two main functions: killing of target cells and cytokine production. NK cells killing takes place by secretion of perforin and granzymes and mediators of apoptosis (TNF-related apoptosis-inducing ligand and FasL). NK cells may secrete both immunostimulatory and immunomodulatory cytokines such as interferon (IFN)-γ, TNF-α, granulocyte monocyte colony-stimulating factor, IL-5, IL-13, IL-10, and transforming growth factor (TGF)-β. The immune system in AD patients is severely affected and several studies reported that proinflammatory circulating cytokines and lymphocyte subset distribution can be disturbed in AD [3]. The implication of NK cells in AD immunopathogenesis was observed by Krishnaraj [6] who showed that cholinesterase inhibitors could suppress expansion and cytotoxic function of NK cells in AD patients, with an improvement of clinical status. Araga et al. [7] showed that the functional potential of NK cells in AD patients was significantly lower than in normal controls. Also, other researchers reported decreased NK cytotoxic function [3]. Thus, evaluating the role of NK cells in the immunopathogenesis of AD may open new horizons for designing novel therapeutic methods.

In our recent studies on MS, we observed that patients with KIR2DL2 expression are more susceptible to herpesvirus infection [8, 9]. In particular, MS patients positive for KIR2DL2 expression on NK cells failed to control herpes simplex virus 1 (HSV-1) infection and secreted high levels of Th17 cytokines, while MS patients negative for KIR2DL2 on NK cells had an efficient killing of virus-infected cells and released Th1 cytokines. Furthermore, we have evidence that the lack of activation of KIR2DL2 positive NK cell takes place in the presence of other human herpesviruses such as EBV and HHV-6 (Rizzo et al., manuscript in preparation). A lower NK cells intracerebral surveillance triggered by herpesvirus infection could have an important role in MS pathogenesis.

AD has been associated with a possible infective etiology [10], and almost three decades ago was suggested for the first time a possible involvement of herpes simplex virus type 1 (HSV-1) [11]. The ever-increasing number of these studies warrants re-evaluation of a possible association between HSV infection and AD. In humans, herpesvirus brain infection has been associated with AD-like pathology [12 –14]. In mice and in cell culture, amyloid-β (Aβ) deposition and tau abnormalities typical of AD are observed after infection with HSV-1 [15 –18], and a direct interaction between AβPP and HSV-1 has been reported [19]. Antivirals, including acyclovir, block in vitro HSV-1-induced Aβ and tau pathology [20]. HSV-1 reactivation and seropositivity correlate with incident AD [12, 13] and the use of antiherpetic drugs reduces the relative risk of senile dementia by a factor of 10 [14]. Apolipoprotein E (APOE), a risk factor for AD, increases HSV-1-induced brain damage [21]. Inside the brain, ApoE, characterized by three allelic forms: APOE2, APOE3, APOE4, helps to clear Aβ, a plaque component. ApoE ɛ2 appears to perform this function more effectively than ApoE ɛ4, with ApoE ɛ3 in the middle. Interestingly, ApoE ɛ4 is a risk factor for vascular causes of cognitive impairment and dementia [22, 23], can increase the risk of atherosclerosis and stroke [24], may induce the response to stress or injury [24], increases the risk for AD, and lowers the age of onset [25].

Furthermore, it has been suggested that there is a cumulative correlation between multiple herpesvirus infections and cognitive decline in patients with vascular disease [26]. Also, a connection between cytomegalovirus (CMV) and AD was made, showing that CMV IgG and IFN-γ levels in cerebrospinal fluid correlate with neuropathologic characteristics of AD [27] and with an increased rate of cognitive decline [28].

Recently it was suggested a possible association between AD and human herpesvirus 6 (HHV-6) infection [29]. The authors, constructing multiscale networks of late-onset AD, observed increased presence of HHV-6A in AD patients than in controls, suggesting viral activity as a general feature of AD. HHV-6 is classified in two distinct species, HHV-6A and HHV-6B [30], but the term HHV-6 remains in usage for both species. The two viral species are quite similar, sharing many similarities, but differ for cell receptors, specific tropism, and disease association. HHV-6 preferentially replicates in activated CD4+ T lymphocytes [31, 32], but can infect also NK cells, endothelial cells, astrocytes, oligodendrocytes, and microglial cells [31]. The in vivo host tissue range of HHV-6 includes brain, tonsils, salivary glands, kidneys, liver, lymph nodes, endothelial cells, and monocytes/macrophages [31 , 33–35].

Interestingly, similarly to MS, AD patients also show decreased NK cell activity [2]. These two observations (i.e., HHV-6A presence and NK cell decreased activation) might support the hypothesis that NK cells and herpesvirus infection might contribute to AD immune-pathogenesis.

Our aim was to test the possible implication of KIR/HLA genetic background in AD, and its association with HHV-6A infection.

MATERIALS AND METHODS

Study subjects

We recruited 98 patients (55–87 years old) with an AD clinical diagnosis in accordance with the criteria of the National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer’s Disease and Related Disorders Association (NINCDS–ADRDA) [36]. These criteria include a mild-to-moderate severity, neuroimaging analysis, and a Mini-Mental State Examination (MMSE_i) [37] score of 11.0 to 29.0 at diagnosis (Table 1). MMSE values were corrected by age and educational level as proposed by Magni et al. [38]. After correction, the MMSE_i scores ranged from 12.3 to 24.7. We recruited 1 severe, 42 moderate, and 55 mild patients. All patients were recruited at the Alzheimer Center of the University Hospital, Arcispedale S. Anna, during the 2004-2006 period. Disease severity was re-evaluated after one year of follow up. The study was approved by the Ethics Committee of the University of Ferrara, and the subjects and the caregivers consented to the study.

Table 1

Baseline characteristics

Baseline characteristics	AD (mean±SD)	Control (mean±SD)
Female: Male	65:33	66:34
Age (y)	75.9±5.5	74.3±4.2
Age at diagnosis (y)	73.3±5.5	–
MMSED_I score	19.4±3.6	–
ADLD_I score/6^a	5.4±0.88	–
IADLD_I score/19^a	13.1±2.3	–
MMSED_I score	19.3±3.2	–

MMSED_I, Mini-Mental State Examination at diagnosis, corrected by age and education; ADLDI, Activities of Daily Living at diagnosis; IADLDI, Instrumental Activities of Daily Living at diagnosis. Higher score indicates better performances in normal daily activities (ADL, IADL) or better cognitive functions.

Healthy members of societies for retired people were recruited for the cognitively-normal control group. Only ethnic Italians without known dementia in first degree relatives were included.

Genotyping KIR and HLA

Genomic DNA was isolated from whole blood using the QIAamp DNA Blood Mini kit (QIAGEN, Hilden, Germany). Genotyping of KIR and HLA alleles was performed by Olerup typing kit (West Chester, PA, USA) following manufacturer’s protocols.

Haplotype reconstruction

Maximum-likelihood KIR haplotype frequencies were estimated using an expectation–maximization (EM) algorithm implemented in Arlequin 3.1 [39]. KIR-inferred haplotypes were compared with those available on the web site of the official allele nomenclature (https://www.ebi.ac.uk/ipd/kir/nomenclature.html).

Genotyping APOE

APOE alleles were determined by genotyping at two single nucleotide polymorphisms (SNPs), rs7412 and rs429358. Genotyping was performed using predesigned Taqman Assays (C_904973_10 and C_3084793_20, Applied Biosystems) on the Applied Biosystems 7900 HT Fast Real Time PCR. Deviation from Hardy Weinberg Equilibrium (HWE) was calculated for the APOE locus, and both cases (p = 0.86) and controls (p = 0.36) were found to be in HWE.

NK92 cell nucleofection

NK92 (ATCC CRL-2407) cell line was grown in MEM-Alpha medium (Gibco BRL, Invitrogen Corporation, Carlsbad, CA, USA) supplemented with 20% of FCS (fetal calf serum), 0.1 mM 2-Mercaptoethanol, 100 U/mL of penicillin and streptomycin, and 150 U/mL of IL-2. Cell cultures were maintained at 37°C in humidified atmosphere of 5% CO₂.

KIR2DL2-GFP-tagged was inserted in the plasmid pCMV6-AC-GFP (OriGene Technologies, Rockville, MD USA). The recombinant plasmid was amplified in One Shot^® TOP 10F’ E. Coli (Invitrogren). Plasmid DNA was extracted with plasmid MAXI kit (QIAGEN, Hilden, Germany) and nucleofected in NK92 cells (Ingenio^®Electroporation Kit, Mirus Bio LLC, WI USA). Stable cell colonies were selected in the presence of 1000μg/mL of Geneticin^® (G418 Sulfate, Gibco^®). KIR2DL2 NK92 cells and NK92 control cells were co-cultured with infected cells in a 1:5 ratio and collected for subsequent analysis [8, 9]. Each experiment was performed in triplicate.

Images were acquired with Nikon Eclipse TE2000S fluorescence microscope equipped with a digital camera.

KIR2 e beta-actin mRNA analysis

Total RNA was extracted from PBMC by TRIzol (Invitrogen). RNA samples were digested with DNase (1 U/ug RNA) and stored at –85°C in ethanol until cDNA synthesis. Two micrograms of total RNA were reverse transcribed using SuperScripttrademark First-Strand Synthesis System (Invitrogen). cDNA was amplified by specific oligonucleotide primers for KIR2DL2 [40] and house-keeping control gene beta-actin [41, 42]. Amplification products were analyzed by ethidium bromide–stained agarose gel electrophoresis.

HHV-6 infection

The human T cell line J-Jhan was cultured in RPMI-1640 (Gibco BRL, Invitrogen Corporation, Carlsbad, CA, USA) with 10% FCS, supplemented with 100 U/ml each of penicillin and streptomycin and maintained at 37°C in humidified atmosphere of 5% CO₂.

J-Jhan cells were infected with an HHV-6A (U1102) cell free inoculum as previously described [42]. The day after infection, J-Jhan cells were harvested for DNA and RNA quantification and for APOE mRNA analysis or NK92 cells were added at a 1:5 ratio. Each experiment was performed in triplicate.

HHV-6 analysis

HHV-6 DNA viral load was analyzed by real time quantitative (qPCR) in duplicate, as described [43]. RNA cell extraction was performed using the RNeasy kit (Qiagen, Hilden, Germany). No DNA contaminated the RNA samples, as shown by control β-actin PCR without retrotranscription [41, 42]. Reverse transcription was performed by the RT2 First strand kit (Qiagen, Hilden, Germany). cDNA aliquots corresponding to 200 ng RNA were used for virus transcription analysis, performed by qPCR detecting the expression of U42 gene, as previously reported [44].

Sp1 phosphorylation status analysis

Sp1 phosphorylation was evaluated by western blot. Proteins were extracted by RIPA buffer and ready-to-use cocktails of protease and phosphatase inhibitors (Abcam, UK). Protein concentration was quantified by Bradford assay (Bio-Rad Laboratories). Similar amounts of protein were loaded in 12% TGX-Pre-cast gel (Biorad, Segrate, MI, Italy), with subsequent electroblotting. The membrane was incubated with 1:200 Sp1 (ProSci Incorporated; Poway, CA, USA) or anti-phospho Sp1 (pThr) (Sigma-Aldrich; Saint Louis, MO, USA) monoclonal antibodies. A horseradish peroxidase (HRP)-conjugated antimouse antibody (1:5000; Amersham Biosciences, NJ, USA) was added and developed with the ECL kit (Amersham Biosciences, NJ, USA). The images were acquired by GelDoc EZ System (BioRad; Milano; Italy).

PKA analysis

PKA was extracted from 10∧7 cells with PepTag^® Non-Radioactive Protein Kinase Assays (Promega, Milano, Italy). The inhibition of PKA activity was obtained with increasing concentrations (0.25, 0.5, 1.0, 2.0, 2.5μM) of KT 5720 (Sigma-Aldrich, St. Louis, MO, USA), a specific inhibitor of protein kinase A.

Carboxyfluorescein diacetate succinimidyl ester (CFSE) analysis

Infected cells were assessed for cell-mediated cytoxicity, using 7AAD/CFSE Cell-mediated cytoxicity assay kit (Cayman Chemicals, Ann Arbor, MI, USA), according to the manufacturer’s protocol. Labelled cells were analyzed with FACSCantoII flow cytometer (Becton Dickinson, Milano, Italy) and FlowJo software (FlowJo LLC, OR, USA).

APOE expression

J-Jhan cells, infected with HHV-6A (U1102) were harvested after 1, 3, 7, 14, and 21 days post infection. Total RNA was extracted (RNeasy kit, Qiagen, Hilden, Germany) and checked for the absence of contaminating DNA [41]. RNA reverse transcription was performed as described above. APOE expression analysis was performed on DNA aliquots corresponding to 200 ng RNA using Applied Biosystems Gene expression analysis (Hs00171168_m1).

Statistical analyses

To detect significant departure from Hardy–Weinberg equilibrium, we used the extended version of Fisher’s exact test implemented in Arlequin 3.01 [39]. Tests of association between genotypes and cognitive changes were performed with a logistic regression adjusted for age, sex, baseline MMSED_i, to test the possible association between disease status and KIR. Biological variables were evaluated by Student t test by Graph pad software. p-value <0.05 was considered to be statistically significant. p values were corrected (p_c) for multiple comparisons, using Bonferroni correction.

RESULTS

KIR/HLA combinations

Patient characteristics are shown in (Table 1). Frequencies and distributions of KIR alleles were in Hardy Weinberg equilibrium and are summarized in (Table 2). Patients with AD showed a significantly increased frequency of the activating receptor KIR2DS2 (62% versus 37.5%; pc: 2.2×10^–³) and a reduced frequency of the activating receptor KIR3DS1 (16.6% versus 32.7%; pc:2.9×10^–³). A higher number of AD patients had increased inhibitory receptor KIR2DL2 (62.2% versus 36.4%; pc: 2.2×10^–³). No significant differences were found in the distribution of KIR genotypes (AA or Bx) but there was an increase in HLA-C1/C2 genotype in AD patients (45.9% versus 22.7%; pc: 2×10^–³). When analyzing KIRs and their respective ligands, the combination KIR2DS2/KIR2DL2 present/C1 present, was significantly increased in AD patients (48% versus 28.2%; pc:0.037) while the combination KIR2DS2/KIR2DL2 absent/C1 present was significantly decreased in AD patients (6.1% versus 30.9%; pc:5.1×10^–⁵). After including all variables in logistic regression, the increased frequency of KIR2DL2 and increased frequency of the combination KIR2DS2/KIR2DL2 present/C1 present maintained a strict correlation with AD (p = 0.02; p = 0.01, respectively).

Table 2

KIR/HLA frequency

KIR and HLA	AD (N = 98)	%	Control (N = 110)	%	p	p _c
Activating KIR
KIR2DS1	49	50	45	40.9	0.21
KIR2DS2	61	62	41	37.2	2.8×10^–4	2.2×10^–3
KIR2DS3	35	35.7	31	28.2	0.29
KIR2DS4	85	87	91	82.7	0.45	3.6×10^–4
KIR2DS5	46	46.9	33	30	0.015
KIR3DS1	12	16.6	36	32.7	4.9×10^–4	2.9×10^–3
Inhibitory KIR
KIR2DL1	88	89.8	95	86.4	0.52
KIR2DL2	61	62.2	40	36.4	2.8×10^–4	2.2×10^–3
KIR2DL3	71	72.4	85	77.3	0.52
KIR2DL4	98	100	110	100	0
KIR2DL5	59	60.2	50	45.5	0.038
KIR3DL1	92	93.9	91	82.7	0.018
KIR3DL2	98	100	110	100	0
KIR3DL3	98	100	110	100	0
KIR genotype
AA	26	26.5	32	29	0.35
Bx	72	73.5	65	59.1
HLA genotype
C1/C1	25	25.5	22	20	0.41
C1/C2	45	45.9	25	22.7	4×10^–4	2×10^–3
C2/C2	28	28.5	47	42.7	0.042
HLA-Bw4	77	78.6	78	70.9	1
HLA-Bw6	91	92.9	92	83.6	0.054
KIR and their ligands
KIR2DS1/KIR2DL1 pres/HLA-C2 pres	39	39.8	31	28.2	0.08
KIR2DS1/KIR2DL1 pres/HLA-C2 abs	10	10.2	9	8.2	0.64
KIR2DS1/KIR2DL1 abs/HLA-C2 pres	2	2	1	0.9	0.60
KIR2DS2/KIR2DL2 pres/HLA-C1 pres	47	48	31	28.2	0.0041	0.037
KIR2DS2/KIR2DL2 pres/HLA-C1 abs	14	14.3	15	13.6	1
KIR2DS2/KIR2DL2 abs/HLA-C1 pres	6	6.1	34	30.9	5.7×10^–6	5.1×10^–5
KIR3DS1/KI3DL1 pres/ HLA-Bw4 pres	29	29.6	27	24.5	0.88
KIR3DS1/KI3DL1 pres/ HLA-Bw4 abs	9	9.2	8	7.3	0.80
KIR3DS1/KI3DL1 abs/ HLA-Bw4 pres	1	1	3	2.7	0.62

p, p value obtained by Fisher’s exact test. p_c, corrected p value for multiple comparisons (Bonferroni correction).

There was a significant lower MMSED_I score at diagnosis in patients with the KIR2DS2/KIR2DL2/C1 combination (p = 0.0002) (Fig. 1a). Disease severity at diagnosis, ADL_I, and IADL_I did not present any correlation with KIR/HLA combinations (Fig. 1b, c). Interestingly, the worsening of the disease status after one-year follow-up was more evident in patients with the KIR2DS2/KIR2DL2/C1 combination. In fact, 14 of 20 patients with a worsened disease status had KIR2DS2/KIR2DL2/C1 (p = 0.0045).

Fig.1

J-Jhan T cell lines were infected with HHV-6A at a multiplicity of infection of 100 genome equivalent/cell. Virus presence (DNA) and transcription (RNA) were evaluated respectively by qPCR and RT-qPCR performed on U94 and U42 virus genes respectively, at 1, 3 and 7, 14, and 21 d.p.i., as already detailed. mRNA ApoE ɛ4 expression was evaluated in J-Jhan cells at 1, 3 and 7, 14, and 21 d.p.i. RQ, relative quantity of ApoE ɛ4 mRNA.

KIR/HLA combinations and APOE alleles and expression

Frequencies and distributions of APOE alleles were in Hardy Weinberg equilibrium. The frequency of APOE ɛ4 allele was highest in AD patients (p_c: 0.00018) (Table 3). APOE ɛ4 allele was significantly more frequent in AD patients with the KIR2DS2/KIR2DL2/C1 combination (p = 3.9×10^–5) (Table 4).

Table 3

APOE allele frequency

	Allele frequencies
	ɛ2	ɛ3	ɛ4
AD patients (98)	7.2	52.4	40.4
Control subjects (110)	11.4	74.6	14.0

Table 4

KIR/HLA frequency in combination with APOE allele frequency

	AD (N = 98)			Control (N = 110)
	ɛ2	ɛ3	ɛ4	ɛ2	ɛ3	ɛ4
KIR2DS2/KIR2DL2 pres/HLA-C1 pres	6.4	42.6	46.8	9.7	64.5	16.9
Other combinations	7.8	62.7	35.3	10.1	68.4	11.4

We evaluated the possible effect of HHV-6A infection on APOE ɛ4 expression. The results, shown in (Fig. 1), evidence that APOE ɛ4 mRNA synthesis is enhanced by HHV-6A infection, in a time dependent correlation. In particular, there is a clear correlation between the enhancement in HHV-6A RNA production and the increase in APOE ɛ4 mRNA at day 3 p.i. with a peak at day 14 p.i. (Fig. 1).

KIR2DL2 receptor and HHV-6 infection

Since the KIR2DS2/KIR2DL2/C1 combination has been shown to increase the susceptibility to HSV-1 infections [9], we evaluated if this might also be the case for HHV-6A. NK92 cell line, negative for KIR expression, was nucleofected with KIR2DL2 or control plasmid (GFP) (Fig. 2d). Efficiency of transfection was 89% (Fig. 2e). NK92-GFP and NK92-KIR2DL2 cells were co-cultured with HHV-6A infected T cells, expressing HLA-C1 allele. NK92-GFP cells were able to control HHV-6A infection, reducing viral titer, while NK92-KIR2DL2 cells did not (Fig. 2f) (p < 0.001). The blocking of KIR2DL2 receptor with CD158b moAb, specific for KIR2DL2 receptor, reverted the effect of KIR2DL2 expression on NK92 cells, reducing HHV-6A titer to values similar to those obtained in NK92-GFP cells (Fig. 2f). To ensure that NK92-GFP cells killed HHV-6A infected cells and not only to prevent viral replication, cells were analyzed by flow cytometry. We observed no modification in HHV-6A infected cell percentage when co-cultured with NK92-GFP cells (Fig. 2g). On the contrary, NK92-KIR2DL2 cells reduced CFSE⁺ cell percentage, indicating a higher proliferation of the cells and suggesting the inability of NK92-KIR2DL2 cells to kill infected cells.

Fig.2

a) MMSED_I (Mini-Mental State Examination at diagnosis), b) ADLD_I (Activities of Daily Living at diagnosis), and c) IADLD_I (Instrumental Activities of Daily Living at diagnosis) distribution according with the KIR2DS2/KIR2DL2 present/C1 present combination. ^**p value <0.05, obtained by logistic regression. d) NK92 nucleofected cells with pCMV6-AC-GFP (RG224797, OriGene Technologies, Inc., Rockville, MD, USA) (NK92-GFP) or pCMV6-AC-KIR2DL2 (GFP- tagged- Human killer cell immunoglobulin-like receptor, two domains, long cytoplasmic tail, 2) (NK92-KIR2DL2) plasmid. GFP expression was evaluated after 12 days of selection. e) Beta actin and KIR2DL2 mRNA expression in NK92 (WT), NK92 nucleofected cells with pCMV6-AC-GFP (GFP) or pCMV6-AC-KIR2DL2 (KIR2DL2) plasmid. f) HHV-6A infection of J-Jhan cells and co-culture with NK92 nucleofected cells with pCMV6-AC-GFP (NK92-GFP) or pCMV6-AC-KIR2DL2 plasmid with (NK92-KIR2DL2 + CD158b) or without (NK92-KIR2DL2) the addition of anti-KIR2DL2 moAb (CD158b). Virus DNA presence was evaluated using an amount of 100 ng of DNA. Results are expressed as mean Log10 target molecules. g) Mean results obtained for CFSE Fluorescent cell staining of HHV-6A infection of J-Jhan cells co-cultured with NK92 nucleofected cells with pCMV6-AC-GFP (NK92-GFP) or pCMV6-AC-KIR2DL2 plasmid with (NK92-KIR2DL2 + CD158b) with (NK92-KIR2DL2 + CD158b) or without (NK92-KIR2DL2) the addition of anti-KIR2DL2 moAb (CD158b). ^*p value <0.05, obtained by Student t-test. Each experiment was performed in triplicate.

Involvement of PKA/Sp1 pathway

Since KIR2DL2 expression is controlled by specific transcription factors [45], we evaluated the modulation of Sp1 phosphorylation (pThr), highly represented in NK92 cell line [46, 47]. We observed an increase in Sp1 phosphorylation in NK92 cells co-cultured with HHV-6A infected cells (Fig. 3a). Since Sp1 DNA binding properties are stimulated by cAMP-dependent protein kinase (PKA) [48], we looked at possible differences in PKA activity. PKA activity increased in NK92 cells co-cultured with HHV-6A infected cells (Fig. 3b). Interestingly, Sp1 phosphorylation was decreased in the presence of KT 5720, a PKA inhibitor, even in the presence of HHV-6A infected cells (Fig. 3c). These data suggest that the PKA/Sp1 pathway might be involved in KIR2DL2 expression on NK cells, inhibiting their activation during HHV-6A infection.

Fig.3

a) Western blot analysis for the evaluation of Sp1 and pThrSp1 expression in NK92 cells in co-culture with HHV-6A infection of J-Jhan cells (HHV-6A). b) PKA activity in NK92 cells in co-culture with HHV-6A infection of J-Jhan cells (HHV-6A). c) Fold increase of pThrSp1 in NK92 cells in co-culture with HHV-6A infection of J-Jhan cells (HHV-6A) in the presence of increasing concentrations (0.25, 0.5, 1.0, 2.0, 2.5μM) of KT 5720, a specific, cell-permeable inhibitor of protein kinase A. ^*p value <0.05, obtained by Student t-test. Each experiment was performed in triplicate.

Conclusions

In this study, we describe a possible role for a specific KIR/HLA subset in AD. We observed an increased frequency of KIR2DS2/KIR2DL2/C1 in AD patients that correlates with a lower MMSED_i score at diagnosis and a worsening of the disease status in the one-year follow-up. Since we have previously shown that the combination KIR2DS2/KIR2DL2/C1 results in a higher susceptibility of MS patients to herpesvirus infections, this genetic background might have a role also in herpesvirus infection in AD. Interestingly, both activating KIR2DS2 and inhibiting KIR2DL2 receptors resulted increased in AD patients. KIR2DL2 and KIR2DS2 are in linkage disequilibrium, implying they are “preferentially associated” with one another. However, when KIR2DS2 and KIR2DL2 are co-expressed, NK cell inhibition via KIR2DL2 over-rides NK cell activation via KIR2DS2 [49]. We hypothesized that AD patients with KIR2DS2/KIR2DL2/C1 might be less prone to control HHV-6A infection and, in more general terms, herpesvirus infection. The in vitro experiments on NK92-KIR2DL2 cells showed an impaired control of HHV-6A infection in comparison with wild type NK92-GFP. Interestingly, NK92 cells challenged with HHV-6A infected cells increased the phosphorylation status of Sp1due to both PKA activity and the direct transactivating effect of HHV-6 on Sp1 binding sites [50]. Since KIR2DL2 promoter has a Sp1 binding site [45], the activation of Sp1 could affect the expression of KIR2DL2 on the surface of NK cells in the presence of herpesvirus infected cells [9]. Meanwhile, the activation of PKA could have a role in AD neurofibrillary pathology, directly participating in the tau pathological hyperphosphorylation [51]. We hypothesize that HHV-6A infection might induce PKA activation, tau hyperphosphorylation, and Sp1 phosphorylation, thereby increasing its binding with the KIR2DL2 promoter. Overexpression of KIR2DL2 will consequently inhibit NK cells [9], possibly with a direct role in AD. Previous studies have already described the possible involvement of NK cells in AD immune-pathogenesis [7], showing a lower NK cell functional potential in AD patients [2]. Solerte et al. suggested that the overactivity of NK cells AD patients could be related to the dysregulation of PKC (protein kinase C) expression [52], similarly to what we observed for PKA.

Taking into consideration the genetic background of AD patients, it was necessary to look at APOE allelic distribution. Inside the brain, ApoE, characterized by three allelic forms: APOE2, APOE3, APOE4, helps to clear Aβ, a plaque component. We confirmed that APOE ɛ4 allele has an increased frequency in AD patients (Tables 3 and 4), and, interestingly, this was more frequent in KIR2DS2/KIR2DL2/C1 AD patients. These results suggest that APOE ɛ4 allele and KIR2DS2/KIR2DL2/C1 could have a combined effect in a multi-gene form of AD. In particular, APOE is able to inhibit the activation of immune cells, and therefore it could enhance the NK inhibition associated to KIR2DS2/KIR2DL2/C1. We also observed that HHV-6A infection up-modulates ApoE ɛ4 expression. Moreover, ApoE increases PKA activity and, conversely, the secretion of ApoE occurs via a PKA-dependent pathway [54]. Therefor a loop could be established, with HHV-6 increasing ApoE ɛ4 expression and this would further decrease an already reduced KIR2DS2/KIR2DL2/C1 NK patrolling activity.

The involvement of NK cell in AD could include their defensive role toward HHV infections, leading to the deregulation of immune system and resulting in the AD progression [55]. Our results with NK92-KIR2DL2 cells, expressing only KIR2DL2 and not other KIR receptors that could confound the analysis, supports this possibility. Higher herpesvirus infection might lead to hyperphosphorylation and polymerization of tau proteins, induce Aβ peptide fibrilization and deposition [56], suggesting that herpesviruses might have an important role also in several features of AD pathology.

In conclusion, the possible role of herpesvirus infection in AD development/control is supported by: 1) the increase of KIR2DL2/C1 genotype in AD patients (Table 2), 2) a lower anti-herpetic activity of KIR2DL2 positive NK cells (Fig. 1f) [9], 3) a lower NK activity in AD [2], 4) an increased abundance of neurotropic herpesviruses in AD [29] that promote Aβ deposition [56], and 5) the creation of a molecular environment favorable for tau fibrilization (Fig. 2b).

To confirm this hypothesis, further studies are needed on ex vivo samples from AD patients. If this hypothesis finds experimental confirmation, a new therapeutic strategy for AD might be envisaged. In fact, modulation of KIR2DL2 expression on NK cells, as already proposed for other pathological conditions [57, 58], could restore their anti-herpetic surveillance.

Footnotes

ACKNOWLEDGMENTS

We thank Kristin Loomis and Linda Sartor for the support in paper writing and English revision. This research was supported by University of Ferrara FAR (RR, DDL) and FISM grant 2015/R/20 (RR).

Authors’ disclosures available online ().

References

Hort

, O’Brien

, Gainotti

, Pirttila

, Popescu

, Rektorova

, Sorbi

, Scheltens

, EFNS Scientist Panel on Dementia (2010) EFNS guidelines for the diagnosis and management of Alzheimer’s disease. Eur J Neurol 17, 1236–1248.

Richartz-Salzburger

, Batra

, Stransky

, Laske

, Kohler

, Bartels

, Buchkremer

, Schott

(2007) Altered lymphocyte distribution in Alzheimer’s disease. J Psychiatr Res 41, 174–178.

Speciale

, Calabrese

, Saresella

, Tinelli

, Mariani

, Sanvito

, Longhi

, Ferrante

(2007) Lymphocyte subset patterns and cytokine production in Alzheimer’s disease patients. Neurobiol Aging 28, 1163–1169.

Thielens

, Vivier

, Romagne

(2012) NK cell MHC class I specific receptors (KIR): From biology to clinical intervention. Curr Opin Immunol 24, 239–245.

Cheent

, Jamil

, Cassidy

, Liu

, Mbiribindi

, Mulder

, Claas

, Purbhoo

, Khakoo

(2013) Synergistic inhibition of natural killer cells by the nonsignaling molecule CD94. Proc Natl Acad Sci U S A 110, 16981–16986.

Krishnaraj

(1991) Immunomodulation by 9-amino-1,2,3,4-tetrahydroacridine (THA): 1. Down-regulation of natural cell-mediated cytotoxicity. Immunopharmacology 22, 69–76.

Araga

, Kagimoto

, Funamoto

, Takahashi

(1991) Reduced natural killer cell activity in patients with dementia of the Alzheimer type. Acta Neurol Scand 84, 259–263.

Rizzo

, Bortolotti

, Fainardi

, Gentili

, Bolzani

, Baldi

, Casetta

, Granieri

, Rotola

, Furlan

, Di Luca

(2016) KIR2DL2 inhibitory pathway enhances Th17 cytokine secretion by NK cells in response to herpesvirus infection in multiple sclerosis patients. J Neuroimmunol 294, 1–5.

Rizzo

, Gentili

, Casetta

, Caselli

, De Gennaro

, Granieri

, Cassai

, Di Luca

, Rotola

(2012) Altered natural killer cells’ response to herpes virus infection in multiple sclerosis involves KIR2DL2 expression. J Neuroimmunol 251, 55–64.

10.

Itzhaki

, Lathe

, Balin

, Ball

, Bearer

, Braak

, Bullido

, Carter

, Clerici

, Cosby

, Del Tredici

, Field

, Fulop

, Grassi

, Griffin

, Haas

, Hudson

, Kamer

, Kell

, Licastro

, Letenneur

, Lovheim

, Mancuso

, Miklossy

, Otth

, Palamara

, Perry

, Preston

, Pretorius

, Strandberg

, Tabet

, Taylor-Robinson

, Whittum-Hudson

(2016) Microbes and Alzheimer’s disease. J Alzheimers Dis 51, 979–984.

11.

Jamieson

, Maitland

, Wilcock

, Craske

, Itzhaki

(1991) Latent herpes simplex virus type 1 in normal and Alzheimer’s disease brains. J Med Virol 33, 224–227.

12.

Letenneur

, Peres

, Fleury

, Garrigue

, Barberger-Gateau

, Helmer

, Orgogozo

, Gauthier

, Dartigues

(2008) Seropositivity to herpes simplex virus antibodies and risk of Alzheimer’s disease: A population-based cohort study. PLoS One 3, e3637.

13.

Lovheim

, Gilthorpe

, Adolfsson

, Nilsson

, Elgh

(2015) Reactivated herpes simplex infection increases the risk of Alzheimer’s disease. Alzheimers Dement 11, 593–599.

14.

Tzeng

, Chung

, Lin

, Chiang

, Yeh

, Huang

, Lu

, Chang

, Kao

, Yeh

, Chiang

, Chou

, Tsao

, Wu

, Chien

(2018) Anti-herpetic medications and reduced risk of dementia in patients with herpes simplex virus infections-a nationwide, population-based cohort study in Taiwan. Neurotherapeutics 15, 417–429.

15.

Wozniak

, Itzhaki

, Shipley

, Dobson

(2007) Herpes simplex virus infection causes cellular beta-amyloid accumulation and secretase upregulation. Neurosci Lett 429, 95–100.

16.

De Chiara

, Marcocci

, Civitelli

, Argnani

, Piacentini

, Ripoli

, Manservigi

, Grassi

, Garaci

, Palamara

(2010) APP processing induced by herpes simplex virus type 1 (HSV-1) yields several APP fragments in human and rat neuronal cells. PLoS One 5, e13989.

17.

Martin

, Aguila

, Araya

, Vio

, Valdivia

, Zambrano

, Concha

, Otth

(2014) Inflammatory and neurodegeneration markers during asymptomatic HSV-1 reactivation. J Alzheimers Dis 39, 849–859.

18.

Piacentini

, Li Puma

, Ripoli

, Marcocci

, De Chiara

, Garaci

, Palamara

, Grassi

(2015) Herpes Simplex Virus type-1 infection induces synaptic dysfunction in cultured cortical neurons via GSK-3 activation and intraneuronal amyloid-beta protein accumulation. Sci Rep 5, 15444.

19.

Cheng

, Ferland

, Webster

, Bearer

(2011) Herpes simplex virus dances with amyloid precursor protein while exiting the cell. PLoS One 6, e17966.

20.

Wozniak

, Frost

, Preston

, Itzhaki

(2011) Antivirals reduce the formation of key Alzheimer’s disease molecules in cell cultures acutely infected with herpes simplex virus type 1. PLoS One 6, e25152.

21.

Harris

, Harris

(2018) Molecular mechanisms for herpes simplex virus type 1 pathogenesis in Alzheimer’s disease. Front Aging Neurosci 10, 48.

22.

Dwyer

, Skrobot

, Dwyer

, Munafo

, Kehoe

(2013) Using Alzgene-like approaches to investigate susceptibility genes for vascular cognitive impairment. J Alzheimers Dis 34, 145–154.

23.

Sun

, Tan

, Wang

, Tan

, Li

, Xu

, Zhu

, Jiang

, Yu

(2015) Genetics of vascular dementia: Systematic review and meta-analysis. J Alzheimers Dis 46, 611–629.

24.

Liu

, Liu

, Kanekiyo

, Xu

, Bu

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

25.

Michaelson

(2014) APOE epsilon4: The most prevalent yet understudied risk factor for Alzheimer’s disease. Alzheimers Dement 10, 861–868.

26.

Strandberg

, Pitkala

, Linnavuori

, Tilvis

(2003) Impact of viral and bacterial burden on cognitive impairment in elderly persons with cardiovascular diseases. Stroke 34, 2126–2131.

27.

Lurain

, Hanson

, Martinson

, Leurgans

, Landay

, Bennett

, Schneider

(2013) Virological and immunological characteristics of human cytomegalovirus infection associated with Alzheimer disease. J Infect Dis 208, 564–572.

28.

Aiello

, Haan

, Blythe

, Moore

, Gonzalez

, Jagust

(2006) The influence of latent viral infection on rate of cognitive decline over 4 years. J Am Geriatr Soc 54, 1046–1054.

29.

Readhead

, Haure-Mirande

, Funk

, Richards

, Shannon

, Haroutunian

, Sano

, Liang

, Beckmann

, Price

, Reiman

, Schadt

, Ehrlich

, Gandy

, Dudley

(2018) Multiscale analysis of independent Alzheimer’s cohorts finds disruption of molecular, genetic, and clinical networks by human herpesvirus. Neuron 99, 64–82e67.

30.

Ablashi

, Agut

, Alvarez-Lafuente

, Clark

, Dewhurst

, DiLuca

, Flamand

, Frenkel

, Gallo

, Gompels

, Hollsberg

, Jacobson

, Luppi

, Lusso

, Malnati

, Medveczky

, Mori

, Pellett

, Pritchett

, Yamanishi

, Yoshikawa

(2014) Classification of HHV-6A and HHV-6B as distinct viruses. Arch Virol 159, 863–870.

31.

De Bolle

, Naesens

, De Clercq

(2005) Update on human herpesvirus 6 biology, clinical features, and therapy. Clin Microbiol Rev 18, 217–245.

32.

Braun

, Dominguez

, Pellett

(1997) Human herpesvirus 6. Clin Microbiol Rev 10, 521–567.

33.

Donati

, Akhyani

, Fogdell-Hahn

, Cermelli

, Cassiani-Ingoni

, Vortmeyer

, Heiss

, Cogen

, Gaillard

, Sato

, Theodore

, Jacobson

(2003) Detection of human herpesvirus-6 in mesial temporal lobe epilepsy surgical brain resections. Neurology 61, 1405–1411.

34.

Roush

, Domiati-Saad

, Margraf

, Krisher

, Scheuermann

, Rogers

, Dawson

(2001) Prevalence and cellular reservoir of latent human herpesvirus 6 in tonsillar lymphoid tissue. Am J Clin Pathol 116, 648–654.

35.

Fox

, Briggs

, Ward

, Tedder

(1990) Human herpesvirus 6 in salivary glands. Lancet 336, 590–593.

36.

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.

37.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state". A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

38.

Magni

, Binetti

, Bianchetti

, Rozzini

, Trabucchi

(1996) Mini-Mental State Examination: A normative study in Italian elderly population. Eur J Neurol 3, 198–202.

39.

Excoffier

, Laval

, Schneider

(2007) Arlequin (version 3.0): An integrated software package for population genetics data analysis. Evol Bioinform Online 1, 47–50.

40.

Uhrberg

, Valiante

, Shum

, Shilling

, Lienert-Weidenbach

, Corliss

, Tyan

, Lanier

, Parham

(1997) Human diversity in killer cell inhibitory receptor genes. Immunity 7, 753–763.

41.

Caselli

, Bortolotti

, Marci

, Rotola

, Gentili

, Soffritti

, D’Accolti

, Lo Monte

, Sicolo

, Barao

, Di Luca

, Rizzo

(2017) HHV-6A infection of endometrial epithelial cells induces increased endometrial NK cell-mediated cytotoxicity. Front Microbiol 8, 2525.

42.

Rizzo

, Soffritti

, D’Accolti

, Bortolotti

, Di Luca

, Caselli

(2017) HHV-6A/6B Infection of NK cells modulates the expression of miRNAs and transcription factors potentially associated to impaired NK activity. Front Microbiol 8, 2143.

43.

Comar

, D’Agaro

, Horejsh

, Galvan

, Fiorentini

, Andolina

, Caruso

, Di Luca

, Campello

(2005) Long-lasting CD3+ T-cell deficiency after cord blood stem cell transplantation in a human herpesvirus 6-infected child. J Clin Microbiol 43, 2002–2003.

44.

Caselli

, Zatelli

, Rizzo

, Benedetti

, Martorelli

, Trasforini

, Cassai

, degli Uberti

, Di Luca

, Dolcetti

(2012) Virologic and immunologic evidence supporting an association between HHV-6 and Hashimoto’s thyroiditis. PLoS Pathog 8, e1002951.

45.

Liu

, Kuick

, Hanash

, Richardson

(2009) DNA methylation inhibition increases T cell KIR expression through effects on both promoter methylation and transcription factors. Clin Immunol 130, 213–224.

46.

, Vallejo

, Jiang

, Weyand

, Goronzy

(2005) Distinct transcriptional control mechanisms of killer immunoglobulin-like receptors in natural killer (NK) and in T cells. J Biol Chem 280, 24277–24285.

47.

Guo

, Eviatar-Ribak

, Miskimins

(2010) Sp1 phosphorylation is involved in myelin basic protein gene transcription. J Neurosci Res 88, 3233–3242.

48.

Rohlff

, Ahmad

, Borellini

, Lei

, Glazer

(1997) Modulation of transcription factor Sp1 by cAMP-dependent protein kinase. J Biol Chem 272, 21137–21141.

49.

David

, Djaoud

, Willem

, Legrand

, Rettman

, Gagne

, Cesbron

, Retiere

(2013) Large spectrum of HLA-C recognition by killer Ig-like receptor (KIR)2DL2 and KIR2DL3 and restricted C1 SPECIFICITY of KIR2DS2: Dominant impact of KIR2DL2/KIR2DS2 on KIR2D NK cell repertoire formation. J Immunol 191, 4778–4788.

50.

Wang

, Jones

, Norcross

, Bohnlein

, Razzaque

(1994) Identification and characterization of a human herpesvirus 6 gene segment capable of transactivating the human immunodeficiency virus type 1 long terminal repeat in an Sp1 binding site-dependent manner. J Virol 68, 1706–1713.

51.

Jicha

, Weaver

, Lane

, Vianna

, Kress

, Rockwood

, Davies

(1999) cAMP-dependent protein kinase phosphorylations on tau in Alzheimer’s disease. J Neurosci 19, 7486–7494.

52.

Solerte

, Fioravanti

, Pascale

, Ferrari

, Govoni

, Battaini

(1998) Increased natural killer cell cytotoxicity in Alzheimer’s disease may involve protein kinase C dysregulation. Neurobiol Aging 19, 191–199.

53.

Zhu

, Hui

(2003) Apolipoprotein E binding to low density lipoprotein receptor-related protein-1 inhibits cell migration via activation of cAMP-dependent protein kinase A. J Biol Chem 278, 36257–36263.

54.

Kockx

, Guo

, Huby

, Lesnik

, Kay

, Sabaretnam

, Jary

, Hill

, Gaus

, Chapman

, Stow

, Jessup

, Kritharides

(2007) Secretion of apolipoprotein E from macrophages occurs via a protein kinase A and calcium-dependent pathway along the microtubule network. Circ Res 101, 607–616.

55.

Holtzman

, Morris

, Goate

(2011) Alzheimer’s disease: The challenge of the second century. Sci Transl Med 3, 77sr71.

56.

Eimer

, Vijaya Kumar

, Navalpur Shanmugam

, Rodriguez

, Mitchell

, Washicosky

, Gyorgy

, Breakefield

, Tanzi

, Moir

(2018) Alzheimer’s disease-associated beta-amyloid is rapidly seeded by herpesviridae to protect against brain infection. Neuron 99, 56–63e53.

57.

Vey

, Karlin

, Sadot-Lebouvier

, Broussais

, Berton-Rigaud

, Rey

, Charbonnier

, Marie

, Andre

, Paturel

, Zerbib

, Bennouna

, Salles

, Goncalves

(2018) A phase 1 study of lirilumab (antibody against killer immunoglobulin-like receptor antibody KIR2D; IPH2102) in patients with solid tumors and hematologic malignancies. Oncotarget 9, 17675–17688.

58.

Yalniz

, Daver

, Rezvani

, Kornblau

, Ohanian

, Borthakur

, DiNardo

, Konopleva

, Burger

, Gasior

, Pierce

, Kantarjian

, Garcia-Manero

(2018) A pilot trial of lirilumab with or without azacitidine for patients with myelodysplastic syndrome. Clin Lymphoma Myeloma Leuk 18, 658–663.e2.