Note: we welcome your input. Please send comments to: Kristin_Loomis@HHV-6Foundation.org
PROPOSED NEW PRIORITIES FOR ALZHEIMER’S RESEARCH
From the HHV-6A/B & HHV-7 research community
This summary was compiled by the HHV-6 Foundation with input from experts on HHV-6A & HHV-7, many of whom serve on our Scientific Advisory Board.
Exploration of HHV-6A and HHV-7 in the development and progression of Alzheimer’s Disease
Background: Two recent papers published in Neuron have strongly implicated herpesviruses in the pathogenesis of Alzheimer’s disease. Moir and Tanzi showed that amyloid plaques develop rapidly in response to infection by HHV-6A, HHV-6B and HSV1 (Eimer 2018).
In addition, a comprehensive NIH funded “multiomic” evaluation of the Alzheimer’s brain tissue using next generation sequencing data strongly implicated two roseoloviruses: HHV-6A and HHV-7 (Readhead 2018). Among the findings:
- Viral DNA sequence analysis showed increased abundance of HHV-6A vs. controls.
- Analysis of transcripts showed an abundance of HHV-6A and HHV-7 RNA vs. controls.
- A strong correlation between HHV-6A and HHV-7 RNA was found with the clinical dementia rating, amyloid plaque density and the Braak pathology score.
- Causal inference testing of viral quantitative trait loci found that DNA variants that predispose to AD also predispose to key viral species, most notably HHV-6A.
- Host genes modified by HHV-6A overlap with AD risk and biomarker-associated genes.
- Neuronal loss analysis pointed to HHV-6A as the key driver of cell death.
- There was relative specificity for HHV-6A and HHV-7 for Alzheimer’s compared to other neurodegenerative diseases.
- HHV-6A suppression of miR-155 offers a mechanism linking viral activity with Alzheimer’s pathology and supports the hypothesis that viral activity contributes to Alzheimer’s disease.
Contrary to popular belief, HHV-6A is not ubiquitous. Only HHV-6B, which is spread via the saliva, is ubiquitous. HHV-6A is not found in the saliva, we do not know how it is transmitted and the prevalence is unknown since no virus-specific serological assay is available to measure seroprevalence. HHV-6A infects the brain directly via the olfactory route. Unlike HHV-6B, it efficiently infects and replicates in olfactory ensheathing cells (Harberts 2011).
HHV-6A, but not HHV-6B, successfully replicates in human progenitor derived astrocytes (Donati 2005). HHV-6B infection in the astrocytic cell line U251 leads to abortive infection while infection with HHV-6A leads to replication (Yoshikawa 2002). HHV-6A but not HHV-6B supports productive infection in oligodendrocyte progenitor cells (Ahlqvist 2005, Dietrich 2004).
HHV-6A is an immunosuppressive virus with a potent impact on production of chemokines and cytokines, increasing the production of inflammatory cytokines. It productively infects CD4+, CD8+ and natural killer cells and gamma/delta T cells (Dagna 2013).
Sixteen years ago, a group led by Ruth Itzhaki, PhD, found HHV-6 DNA in the brain tissue of 70% of AD patients but only 40% of controls. Like the Readhead study, the prevalence of HSV-1 did not differ significantly between AD patients and controls (Lin 2002).
A study in Japan suggests that acute infections of HSV-1, HSV-2 and VZV in older adults may play a role in senile dementia. Acute HSV1/2 and VZV infections in an elderly population resulted in a three-fold increase in the risk of senile dementia. Importantly, antiviral therapy reduced the risk of dementia by 90% (Itzhaki 2018). Most of those dementias were atypical dementias and not classical Alzheimer’s (Tzeng 2018). Individual herpesviruses may affect different parts of the brain, with differences in associated dementias. Certainly, these findings deserve strong attention in the priorities for the study of Alzheimer’s disease and other dementias.
The HHV-6/7 research community has worked together to come up with specific focus areas that we believe are critical to exploring the role of roseoloviruses in Alzheimer’s disease. If the association is confirmed, then there are a number of promising treatment options that could be considered for clinical trials. They can be divided into four steps:
- Confirm the virological association between HHV6A/7 and Alzheimer’s
- Explore the potential mechanisms that might account for this association
- Utilize animal models to inform clinical trials
- Interventional clinical trials
STEP 1: Confirm the virological association between HHV6A/7 and Alzheimer’s
|
Focus #1: Confirm the relative abundance of HHV-6A/HHV-7 DNA and RNA in the brain tissue of AD brains vs. controls, using multiple techniques. |
|
Confirm the increased HHV-6A/B/7 viral load in AD brain vs controls using standard qPCR and ddPCR techniques. |
Readhead et al. studied over 1000 post mortem brains from patients with AD, other progressive neurological diseases and healthy aging. Investigators found an increased abundance of HHV-6A and HHV-7 sequences in AD patients vs. controls. Four brain regions including the superior temporal gyrus, anterior prefrontal cortex, inferior frontal gyrus and parahippocampal gyrus were examined. A previous study found HHV-6 DNA in 70% of brain samples, and 41% were positive for HHV-6A (Lin 2002). Additional brain regions including the olfactory bulb should be evaluated. |
|
Confirm the finding that there is increased HHV-6A RNA in AD brains, using traditional techniques. |
Readhead et al. found differential abundance of HHV-6A RNA in U3/U4, DR1 and DR6 regions as well as U3/U4 and HHV-7 DR1 and HSV-1. Enriched RNAseq for qRT-PCR positive samples could be explored. |
|
Establish a threshold in copies/ug that identifies clinically significant HHV-6A/7 infection. |
Only quantitative and variant-specific studies should be utilized. Qualitative PCR or prevalence studies are not useful, since some patients may have very low-level virus that is not clinically significant. |
|
Focus #2: Explore the presence of HHV-6A/HHV-7 DNA/RNA in other compartments of AD patients such as PBMCs, plasma, CSF, and determine if they might be useful as biomarkers. |
|
Determine the viral load of HHV-6A/HHV-6B and HHV-7 DNA in plasma, CSF and PBMC samples of AD vs. controls. |
One previous study showed 23% of PBL samples in AD vs. 4% of controls were HHV-6 positive (Carbone 2013). A/B specific qPCR and multiplex ddPCR should be utilized to study this question in well characterized patient and control samples. |
|
Determine whether AD patients with high viral loads of HHV-6A or HHV-7 (or HSV-1, EBV) progress faster than those with low viral loads. |
One study of 150 elderly subjects showed that those who progressed to AD had a significantly higher rate of PBLs positive for HHV-6 or EBV (Carbone 2013). |
|
Determine the RNA load of HHV-6A/HHV-6B and HHV-7 in AD blood samples vs. controls. |
Appropriate assays need to be developed and standardized using well characterized clinical samples. |
|
Establish thresholds to define significant infection in AD patients.
|
Since there may be clinically insignificant low levels of virus, qualitative measures are inadequate, and quantitative measures such as ddPCR or qPCR with a ratio of genomes per number of cells are necessary to evaluate infection. |
|
Determine whether patients in various stages of AD have HHV-6A or HHV-7 in other compartments such as the thymus, olfactory bulb and bone marrow. |
High levels of HHV-6 are found in the olfactory bulb (Harberts 2011). Biopsy of the olfactory bulb is a possible diagnostic tool. HHV-6A may be present in the bone marrow progenitor cells and HHV-6 infection of the thymus may play a role in thymopoiesis and T cell reconstitution (Phan 2018). |
|
Focus #3: Validate HHV-6A/B specific serology assays to study the prevalence of HHV-6A in AD; determine whether HHV-6A/7 serology could useful as biomarkers. |
|
Validate HHV-6A/B specific serology assays to study HHV-6A vs. HHV-6B in AD patients vs. controls. |
Although we know the prevalence of HHV-6B from saliva studies, we know very little about HHV-6A prevalence because it is not found in the saliva and no virus specific serological assay has been available. Recently, several options for such an assay have emerged, but they need validation. Options for an HHV-6A/B specific serology assay might include: (1) a fluorescent bead-based multiplex assay based on IE-1A (Bassig 2018) (2) a LIPS assay using IE-1A proteins (3) immunoblotting assays |
|
Determine if there is a difference in humoral response of HHV-6A/B/7 in AD pts vs. controls, and whether they might represent an effective biomarker for AD progression. |
One study found that AD patients have reduced HHV-6 humoral response compared to controls while CMV, EBV, and HSV-1 IgG levels in AD patients were no different from controls (Westman 2017). |
STEP 2: Explore the potential mechanisms that might account for this association
|
Focus #4: Establish animal models to study roseoloviruses in Alzheimer’s disease. |
|
Determine whether murine roseolovirus (MRV) enhances disease progression in AD mouse models.
|
A murine homolog has been characterized for HHV-6/7 (Patel 2017), murine roseolovirus (MRV). This is the same virus as murine thymic virus (MTV/MTLV). HSV-1 does not enhance disease progression in AD mouse models. It is unknown if MRV would impact AD progression. |
|
Develop a CD46 transgenic mouse crossed with an AD mouse model to study the impact of HHV-6A infection on AD progression. |
HHV-6A persists in the brains of CD46 transgenic mice infected with HHV-6A, resulting in neuroinflammation in a preliminary study (Reynaud 2014). |
|
Determine whether elderly or middle-aged rhesus or green monkeys have accelerated progression to AD when infected with HHV-6A. |
Aging rhesus macaques and African green monkeys manifest AD-like pathology (Paspalas 2018, Cramer 2018). Both monkey species can be infected with HHV-6 and could be used as an animal model. |
|
Determine the importance of viral infection via the olfactory route in AD, and whether olfactory infection by HHV-6A reduces systemic immune response predisposing to AD. |
Infection via the olfactory route has been suspected in AD because the pathology in AD begins in the olfactory bulb and entorhinal cortex. Macaques infected with HHV-6A intranasally do not develop a strong HHV-6 IgG immune response (Leibovitch 2013). Humans with AD exhibit weak HHV-6 humoral response (Westman 2017). HHV-6A but not HHV-6B infects olfactory ensheathing cells and HHV-6 load in olfactory bulb is very high (Harberts 2011). |
|
Focus #5: Characterize the impact of HHV-6A/HHV-7 on host gene expression in AD. |
|
Characterize the effect of in vitro HHV-6A lytic and latent infection on gene expression in neural cells. |
HHV-6A can infect in vitro cells of neural origin establishing latent and low productive infection, with potential unknown effects on genes expressed during AD. |
|
Characterize the nature of HHV-6A-associated inhibition of miR-155 in AD patients. |
Caselli 2017 found that HHV-6A inhibits miR-155. miR-155 is reduced in AD. |
|
Characterize the impact of HHV-6 on MCP-1/CCL2 in AD patients. |
MCP-1/CCL2 correlates with cognitive decline in patients with AD and MCI. CMP-1 is found elevated in the CSF of patients with HHV-6 encephalopathy (Kawamura 2014). HHV-6A stimulates the production of CCL2, CCL5 and CXCL10 in primary brain glial cultures (Reynaud 2014). |
|
Characterize the interaction of HHV-6A and TLR9. |
Blocking TLR9 signaling inhibits HHV-6 chemokine expression. |
|
Other pathogenesis. |
Other studies to understand how HHV-6A/7 are influencing AD pathogenesis might include GWAS, genome sequencing, eQTL. |
|
Focus #6: Characterize the possible HHV-6A-associated immune suppression and NK cell dysfunction in AD patients vs. controls. |
|
Characterize HHV-6A/B/7 infection in CD4, CD8 and NK cells in AD patients vs. controls and determine whether AD patients have reduced cellular immune response. |
HHV6A has been shown to productively infect CD4+ and CD8+ T cells, natural killer cells and gamma/delta T cells inducing de novo expression of CD4. In addition, HHV-6A can interfere with immunologic function via cytokine modulation including blockade of IL-12 production by antigen presenting cells, modulating cell-surface molecules essential for T-cell activation, and expression of viral chemokines and chemokine receptors. Some of these effects are related to downregulation of the viral receptor CD46, a key molecule linking innate and adaptive immune responses (Dagna 2013). |
|
Focus #7: Pathology. Characterize the types of cells that are infected by HHV-6A/HHV-7 infection in the brains of Alzheimer’s patients vs. controls. |
|
Determine which cell types are infected in regions of the brain with high viral loads of HHV-6A and HHV-7. |
Technologies such as single cell sequencing or flow-FISH could be utilized. The HHV-6 Foundation maintains monoclonal antibodies and hybridomas for HHV-6A, HHV-6B and HHV-7 for several proteins. |
|
Analyze existing monoclonal antibodies and develop improved mAbs for HHV-6A/B/7 immunohistochemistry. |
Existing mAbs need to be characterized. New hybridomas need to be developed. |
|
Focus #8: Explore the impact of chromosomal integration by roseoloviruses in Alzheimer’s. |
|
HHV-6A/B is known to integrate into the chromosome as a routine form of latency. One report suggests integration is possible for HHV-7 as well. The possible impact of integration on AD progression is unknown. |
Readhead et al. found increased abundance of HHV-6A and HHV-7 “unique” regions of each virus (the full viral sequence excluding the flanking terminal repeats). It remains unclear why higher levels of the unique region are present. To assess if the DR was lost prior/during integration, next generation sequencing of enriched and unenriched samples from AD patients should be conducted to determine the impact on the virus genome. |
|
Focus #9: Conduct epidemiology studies that might support the finding of HHV-6A/7 involvement in AD pathogenesis. |
|
Determine whether there is an increase in the rate of AD among patients with conditions tentatively associated with HHV-6A infection. |
HHV-6A has been found in 82% of Hashimoto’s thyroiditis patients (Caselli 2012). HHV-6A has been found in the uterus of 43% of women with unexplained infertility (Marci 2016, Coulam 2018). |
|
Determine whether there is an increased risk of AD among individuals with inherited ciHHV-6A. |
Since inherited ciHHV-6A can activate in response to immunosuppression (Endo 2014), HDACi (Arbuckle 2010) and high dose steroids, this population may have a higher rate of AD. |
|
Focus #10: Since HHV-6 activates in the ICU, and HHV-6B reactivation correlates with delirium in transplant patients, determine whether delirium in elderly ICU patients may be triggered by HHV-6A/B/7 reactivation. |
|
Determine whether HHV-6A/B/7 plays a role in ICU delirium of the elderly. |
HHV-6B reactivation during transplantation is associated with delirium and cognitive dysfunction (Zerr 2011). HHV-6A reactivates at high levels in ICU patients (Razonable 2002, Roa 2015). |
STEP 3: Use animal models to inform possible clinical trials
|
Focus #11: Develop models for testing various antiviral compounds for efficacy against HHV-6A/B/7 in vitro and in animal models. |
|
Develop methods for testing antivirals against HHV-6A/B/7. Study FDA approved and pipeline compounds to determine which have the best efficacy for HHV-6A and HHV-7. |
Use HHV-6A GFP and mCHerry virus and other methods. Utilize newly identified HHV-6A/B/7 animal models (CD46 transgenic mice for HHV6A, murine roseolovirus, porcine CMV- now determined to be a roseolovirus). Foscarnet shows the best selectivity index for HHV-6A but the studies are old and could be repeated with improved methodology to compare it to valganciclovir and pipeline drugs IV brincidofovir, cyclopropavir and others. |
|
Study potential therapies in animal models.
|
Potential animal models include: · CD46 transgenic mice (HHV-6A) · AD model mice infected with murine roseolovirus · Rhesus macaques · African green monkeys |
STEP 4: Interventional clinical trials
|
Focus #12: Interventional clinical trials |
|
High dose valacyclovir |
Although the standard dose is considered ineffective for active HHV-6/7 infection, a high dose was successful in reducing the rate of reactivation of HHV-6 in HSCT patients (Hill 2015, Wang 1996) Dose: 500mg 2x daily in Hill study, 2 grams, 4x/day in Wang study. |
|
Pipeline antiviral drugs |
Options include: (1) IV brincidofovir which is effective for HHV-6A/B/7 and HSV1 and penetrates the brain and (2) cyclopropavir which is effective for HHV-6A/B/7. |
|
HHV-6 specific cytotoxic T cell therapy |
Baylor developed HHV-6 specific cytotoxic T cells that are effective against HHV-6 in transplant patients (Gerdemann 2012, Naik 2016). Viracyte licensed the IP and is developing Viralym M which is active against 5 viruses including HHV-6, currently in Phase 2 study. |
|
IVIG therapy |
CMV specific IVIG (Cytogam) is useful against persistent CMV infection. |
|
Antiviral therapy – existing drugs used for HHV-6A/7 |
IV foscarnet has the highest selectivity index for HHV-6A/7 but can cause renal toxicity. Valcganciclovir is not as effective for HHV6A/7 as it is for CMV and HHV-6B; and requires regular monitoring for possible bone marrow suppression. Artesunate, an antimalarial, has potent antiviral effects against HHV-6A & HHV-7. |