On Thursday, 24 February 2011, Ruth Itzhaki, University of Manchester, U.K.; Elisa Porcellini, University of Bologna, Italy; Luc Letenneur, INSERM, Bordeaux, France, and Richard Smeyne, St. Jude Children’s Research Hospital, Memphis, Tennessee, shared some of their latest research and field audience questions. Paul Klapper, herpes virologist at Manchester Royal Infirmary, U.K., and Terrence Town, Cedars-Sinai Medical Center, Los Angeles, were on hand for discussion afterward.
Herpes simplex virus 1 (HSV1) is attracting attention from a growing number of research groups as a possible trigger for Alzheimer's disease. Recent work has tied the virus to AD biomarkers, and offered some epidemiological and genetic support for the long-proposed connection between HSV reactivation and AD risk.
To one group of investigators, the data are strong enough to warrant a treatment trial with antiviral drugs—which are readily available and inexpensive—and they applied for funding to the British government this month. What do you think? Share your comments.
- See responses to some questions raised during the Webinar.
- Listen to the Webinar
Ruth Itzhaki's Presentation
Elisa Porcellini's Presentation
Luc Letenneur's Presentation
Richard Smeyne's Presentation
By Esther Landhuis
Could the same bugs that give us cold sores set off Alzheimer’s disease in some people? Bizarre as it may sound at first blush, the notion that microbes such as herpes simplex virus may contribute to AD has buzzed around for years, perhaps drawing as much skepticism as intrigue. The Alzforum hosted a Live Discussion on this topic in 2004, and now, a spate of recent literature and budding developments on the human trials front gives us reason to take another look. The popular press on both sides of the Atlantic is starting to pay attention, too (see The Times; The Huffington Post).
Viruses are blamed primarily for fever blisters and other short-lived afflictions, so it’s no wonder they receive scant consideration as potential triggers for fatal neurodegenerative diseases that develop over decades. However, this possibility may not be so far-fetched—after all, many bugs lurk within the body for years in a resting state. “This process of dormancy followed by activation makes infectious agents prime candidates as factors in chronic disease,” Ruth Itzhaki and Matthew Wozniak of the University of Manchester, United Kingdom, suggest in a recent review (Wozniak and Itzhaki, 2010). Their commentary mentions prior long-fought battles to prove correspondingly “heretical” ideas—such as Helicobacter pylori as a cause of stomach ulcers, and human papillomaviruses as a cause of cervical cancer.
The 2004 Live Discussion The Pathogen Hypothesis featured heated discussion of Itzhaki’s work on herpes simplex virus type 1 (HSV1) and Brian Balin’s data showing induction of amyloid plaques in the brains of wild-type mice infected with Chlamydia pneumoniae (Little et al., 2004). More recently, Balin, Philadelphia College of Osteopathic Medicine, Pennsylvania, and colleagues have detected Chlamydia antigens alongside amyloid deposits and tau tangles in postmortem AD brain tissue (Hammond et al., 2010).
Itzhaki and Wozniak, and lately other groups as well, are making similar inroads with HSV1. The UK researchers have detected HSV1 DNA in elderly brains by PCR (Jamieson et al., 1991), and via antibodies to HSV proteins in the cerebrospinal fluid (Wozniak et al., 2005). They showed it localizes to amyloid plaques (Wozniak et al., 2009), and linked the virus to other AD biomarkers. In their cell culture system, HSV1 infection drives up AD-specific tau phosphorylation (Wozniak et al., 2009), increases levels of intracellular Aβ, BACE1, and the γ-secretase component nicastrin, and triggers plaque formation in the brains of wild-type mice (Wozniak et al., 2007). Other labs working on HSV1 have also reported links to tau and Aβ. Carola Otth and colleagues at University Austral de Chile, Valdivia, found that HSV1 induces tau cleavage by caspase-3 in primary neuron and astrocyte cultures (Lerchundi et al., 2010). Just last month, researchers led by Jesus Aldudo at CIBERNED in Madrid, Spain, reported that HSV1 increases intracellular Aβ production in autophagosomes and inhibits their breakdown within those compartments (Santana et al., 2011). And Elaine Bearer and colleagues at the University of New Mexico Health Sciences Center, Albuquerque, have evidence that intracellular HSV1 interacts with amyloid precursor protein (APP), and that this interplay enhances viral transport and disrupts APP transport and distribution. These data are in press at PLoS ONE. Taken together, these findings, some say, rise to a level of evidence where other laboratories should attempt to reproduce these data and test in their own experiments whether HSV1 could play a causal role in AD.
Recent epidemiological studies by independent groups appear to support the argument. Federico Licastro, first author Elisa Porcellini, and colleagues at the University of Bologna, Italy, analyzed data from a highly regarded genomewide association study of several thousand European AD patients and controls (see Lambert et al., 2009). They found a set of eight AD-linked gene variants that may increase the brain’s susceptibility to viral infections (Porcellini et al., 2010). The genes identified were nectin-2 (NC2), also known as poliovirus receptor-related protein 2 (PVRL2); apolipoprotein E (ApoE); glycoprotein carcinoembryonic antigen-related cell adhesion molecule-16 (CEACAM-16); B cell lymphoma-3 (BCL3); translocase of outer mitochondrial membrane 40 homolog (TOMM40); complement receptor-1 (CRl); ApoJ or clusterin (CLU); and C-type lectin domain A family-16 member (CLEC16A). These variants form a genetic signature that may determine individual brain susceptibility to pathogen infection, or susceptibility to pathogen damage, particularly by HSV and related viruses. This “may be one complex genetic trait influencing the risk of neurodegeneration leading to clinical AD in old age,” the authors write.
In another study, first author Luc Letenneur at INSERM in Bordeaux, France, and colleagues reported a connection between HSV reactivation and AD risk in 512 French seniors who were cognitively normal when they enrolled in the large, prospective PAQUID study that followed them for 14 years (Letenneur et al., 2008). Previous analyses had measured serum anti-HSV antibodies in small cohorts of AD patients and controls (e.g., Renvoize et al., 1987; Ounanian et al., 1990). The studies produced inconsistent results, possibly because they only examined immunoglobulin G (IgG) antibodies, which characterize past infections or reactivations. In the recent study, Letenneur and colleagues monitored both IgG and IgM anti-HSV antibodies in serum and found that only the latter, which reveals recent HSV reactivation, was associated with elevated AD risk.
How might this work? On the mechanistic front, Claudio Grassi of Catholic University Medical School, Rome, Italy, and colleagues last July reported that HSV1 infection disrupts calcium homeostasis, leading to APP phosphorylation and Aβ42 accumulation, in rat primary cortical neurons (Piacentini et al., 2010). A month earlier, Isamu Mori of Shubun University in Aichi, Japan, proposed that HSV1 may use viral accessory genes to promote its spread within the nervous system (Mori, 2010). Walter Lukiw and James Hill of Louisiana State University Health Sciences Center, New Orleans, have data for a different means by which the herpes virus may evade destruction. They found that HSV1 infection upregulates miRNA-146a, a brain-enriched microRNA associated with pro-inflammatory signaling in stressed neurons and AD (Hill et al., 2009). Last October, they followed with a report that they were able to dampen this miRNA-146a boost in human primary brain cells with aciclovir, an antiviral drug used to treat HSV infections (Lukiw et al., 2010; see also Itzhaki comment on Piacentini et al., 2010; Lukiw et al., 2010; and Porcellini et al., 2010).
Itzhaki and Wozniak themselves have unpublished data on antiviral drug effects on signature AD pathologies as well. They found that antivirals greatly reduce the amounts of Aβ and phospho-tau induced by HSV1 (manuscript in preparation). Moreover, Itzhaki and seven other investigators in three British cities on 9 February submitted a proposal for an antiviral drug trial in AD to the British Medical Research Council. The placebo-controlled study would test an aciclovir prodrug in people with mild to moderate AD who are seropositive for HSV1, measuring cognition and daily function. The team comprises PI Nitin Purandare and co-applicants Alistair Burns, Ruth Itzhaki, Graham Dunn, Julie Morris (all of the University of Manchester, U.K.), Paul Klapper (Manchester Royal Infirmary, Manchester), Clive Holmes (University of Southampton), and Naji Tabet (Institute of Post Graduate Medicine, Brighton & Sussex Medical School, U.K.). In a recent review (Holmes and Cotterell, 2009), Clive Holmes has argued for early intervention against infection as an AD prevention strategy. So do Nicolaas Verhoeff, Kunin-Lunenfeld Applied Research Unit in Toronto, Canada, and colleagues in an independent commentary (Honjo et al., 2009).
It is hard to estimate the proportion of AD patients who may benefit from antiviral treatment, Itzhaki told ARF, but there are some epidemiological data suggesting that AD risk among people with HSV1 DNA in the brain is higher in ApoE4 carriers than in non-carriers (see Itzhaki et al., 1997; Itzhaki et al., 2001). The ApoE link is controversial, however, as one other epidemiological study confirmed the association (Itabashi et al., 1997), whereas another merely found a trend (Beffert et al., 1998). Incidentally, Itzhaki and colleagues found that ApoE4 is a risk factor for HSV-induced cold sores (Itzhaki et al., 1997), and for infection by HSV2, the genital herpes virus (Jayasuriya et al., 2008). Another study confirmed the HSV2 finding (Koelle et al., 2010).
Recent animal work seems to support the idea that the ApoE isoform does influence HSV infection in the brain. Researchers led by Howard Federoff, who moved from the University of Rochester, New York, to Georgetown University in Washington, DC, analyzed HSV1-infected transgenic mice expressing the ApoE2, 3, or 4 allele exclusively. They reported that the ApoE4 mice expressed higher levels of the viral early genes and less of the latency gene, conditions that would promote more frequent reawakening of the virus (Miller and Federoff, 2008). Others have also infected ApoE-transgenic mice with HSV1, and found greater viral load in the brains of ApoE4 animals (Burgos et al., 2007; Bhattacharjee et al., 2008). And in a review published last year, Gerald Rimbach of Christian-Albrechts University, Kiel, Germany, analyzed available data from human trials and basic science research to propose possible mechanisms for how ApoE4 may drive up AD risk in people with active HSV infection (Kuhlmann et al., 2010).
So, esteemed scientists all across the field, here you have it. In your mind, does this new data since 2004 shift your prior stance? In preparation for our live hour, Alzforum asked around among AD researchers. As is to be expected for a controversial topic, some replied on the record, others off. David Holtzman of Washington University, St. Louis, wrote: “I think this topic is of interest. In order for there to be more interest, there need to be experiments which demonstrate (or not) in animals and in humans that there is or isn’t a cause/effect relationship. For example, in animal models, do particular types of herpes virus actually accelerate AD-type pathology and its associated neurodegeneration? If so, what type? In humans, is there evidence in living humans that those developing AD pathology and neurodegeneration (pre- and post-symptomatically) have differences (or not) in prior or current CNS infections with different herpes or other viruses?”
Others said more epidemiology is needed—in particular, to show that active HSV1 is seen more commonly, or that its effects are more damaging, before and during the onset of typical AD symptoms than in age-matched elderly who did not develop AD or its precursor, mild cognitive impairment (MCI). This could be done by demonstrating links between HSV1 titers and AD biomarkers (e.g., low cerebrospinal fluid Aβ42/40 ratio, high PIB signal in brain amyloid imaging). Some people reinforced the need for more human data showing a true relationship by pointing out that the high prevalence of HSV in human brain combined with the great stickiness of plaques makes it unsurprising that the two should sometimes be found together. Yet others pointed out that for a theory in this large and noisy field to gain traction, it is usually necessary that at least three independent labs show similar data.
How about it? Are these studies possible? Does existing evidence make a strong enough case for other labs to attempt independent confirmation? In what ways is the proof of causation by Aβ stronger than for HSV? What would be reasonable to do next?
Q: The virus is safe as long as it is latent, and indeed cytoprotective. What is waking it up and producing the IgM+ effects? Other pathogens?
Paul Klapper: I'm not sure I would agree with the concept of a virus being "safe." In latency, herpes simplex virus is believed to only exist as episomal circularized DNA. In this form, it is dormant but can be reactivated. It therefore always carries the potential for cell destruction. In its circularized form, it is essentially dormant, and thus to suggest that latency is “cytoprotective” seems inappropriate.
The production of IgM antibody in peripheral blood could indicate reactivation at any peripheral site. As Letteneur explained, it is not possible to routinely sample CSF, which might give a better indication of intrathecal reactivation. Itzhaki does, however, have data showing intrathecal synthesis of HSV antibody in Alzheimer’s patients. In a known HSV infection of the brain—herpes encephalitis—we know there is a profound and vigorous immune response to the virus. This response is prolonged in survivors of the disease. The finding of intrathecal synthesis of HSV antibody in the CNS indicates the virus has been there and has replicated, i.e., has at least produced HSV proteins. In latency, no proteins are produced.
Ruth Itzhaki: We have suggested that HSV1 is reactivated in the CNS, in the same way as it is known to be in the PNS, by events such as stress, immunosuppression, and peripheral infections. It seems unlikely that in the latent state, the virus is “safe,” as there is evidence that inflammation continues then. Anyway, it is difficult to distinguish between a very low level of persistent infection and true latency.
Q: Are there any fluoridated antivirals that could be used in PET scans to detect the virus in vivo?
Paul Klapper: The short answer is no. Richard Price in New York did produce work that we also followed in Manchester during the 1980s. We were trying to come up with a non-invasive method of diagnosing herpes encephalitis. In those days, we used radio-iodinated antivirals. We used iodo-vinyl deoxyuridine (a derivative of bromo-vinyl deoxyuridine, which has the same mode of action as aciclovir) labeled with Iodine -131. We were able to visualize intracranial accumulation of the drug (following intravenous administration) using a gamma camera. We had work in progress to produce fluoro-vinyl deoxyuridine with the intention of using this initially in NMR CT scanning (it is possible to detect accumulation of fluoro compounds using NMR spectroscopy) and then PET (at the time PET was not practicable because there was only one operational PET scanner in the whole of the U.K.). Work on this was suspended when we switched to using PCR on CSF as a means of detecting the virus in herpes encephalitis. Labeling does, however, remain a possible way of detecting and studying active virus replication within the brain. However, within the present context, the mechanism that Itzhaki proposes is not that the virus is continuously replicating in brain. It may reactivate infrequently, and there may be only limited replication during reactivation. Thus, to use PET with a targeting antiviral to demonstrate this reactivation would be problematic. It would essentially require continuous scanning to be able to detect reactivation at the time that it occurred. This would not be practicable.
Q: PET imaging of humans using positron-labeled HSV probes is a great idea.
Paul Klapper: True, we had the idea more than 20 years ago. For the reasons outlined above, it is not going to get us much further forward in studies of Alzheimer's.
Q: Chimerix has an antiviral that gets into the brain....
Paul Klapper: There are numerous antivirals that enter the brain and many anti-HSV compounds. None of them achieves free passage to the brain. Usually we can predict their penetrative ability using the octanol:water partition coefficient of the drug. This essentially measures how well the drug will traverse lipid membranes. In relation to the planned clinical trial, the decision was made to use aciclovir (administered as the pro-drug valaciclovir in order to allow oral dosing) because of the extensive experience and pharmacokinetic data available on the use of aciclovir in treatment of herpes encephalitis.
Q: Does aciclovir readily enter the brain? What level of dosing do you think would be required?
Paul Klapper: When administered intravenously, the drug does not readily enter the brain. It has to traverse the blood-brain barrier. This means that, in order to achieve adequate therapeutic concentrations within the CNS, the intravenous level has to be adequately high. The trials of treatment of herpes encephalitis during the 1980s showed that 10 mg/kg dosage every eight hours was required to achieve adequate intra-CNS concentration of the drug. Huge amounts are not required because the beauty of aciclovir is that the drug will selectively concentrate within virus-infected cells as a result of phosphorylation of the drug by the virus-specific enzyme thymidine kinase. Intravenous treatment is not practicable in elderly demented patients; hence, the pro-drug valaciclovir will be used. Some 55 percent of the orally delivered drug will appear in systemic circulation, meaning that adequate systemic levels can be achieved to ensure therapeutic concentrations of the drug within the CNS.
Q: We treated our pediatric patient with cerebral HSV2 with ganciclovir and suppressed the HSV2 replication; hence, it must cross the blood-brain barrier.
Paul Klapper: Ganciclovir has good activity against HSV and penetrates the blood-brain barrier with similar efficiency to aciclovir. Given a choice, however, I would always opt for aciclovir. Aciclovir has a much better safety profile than ganciclovir. Even a single dose of ganciclovir carries a risk of causing infertility in a patient and can commonly cause neutropenia. While it is appropriate for treatment of infections with a virus not effectively treated by aciclovir, for example, cytomegalovirus, first-line treatment of HSV will always be aciclovir. In neonates, as in the elderly, the blood-brain barrier is more permeable than in adults; hence, monitoring for excessive drug concentrations (which can cause reversible neurotoxicity) through inadequate renal function becomes important.
Ruth Itzhaki: The Chimerix compound (CMX001) is one of several antivirals that can enter the brain. I wrote to Chimerix about investigating it, but they never replied. Incidentally, another bonus of using antivirals to treat AD is that, presumably, they would restore Aβ to its normal very low level rather than eliminating it. (Whether Aβ is elicited by infection as an innate immune response of the cell or to aid virus replication is an issue we’d like to investigate, but the damage it causes is likely due to its subsequent overproduction.)
Q: For antivirals to be effective, there needs to be ongoing viral replication. Any thoughts about this with respect to AD patients?
Paul Klapper: Absolutely right. So late-stage Alzheimer’s is not going to be treatable with antivirals. What needs to happen is treatment of early-stage HSV reactivation that, as shown in Itzhaki's model, is a trigger for the development of plaques. As it is not possible to predict when reactivation will occur, long-term suppressive treatment is the only hope. This works in the peripheral nervous system where we use continuous aciclovir/valaciclovir treatment to suppress reactivation in patients who have repeated and frequent reactivation of oro-pharyngeal or genital herpes—hence, the rationale for a trial of suppressive treatment in mild to moderate Alzheimer’s. If we suppress reactivation, do we slow or arrest the process of disease development?
Q: On the question of selective vulnerability of the entorhinal cortex and hippocampus, is this a trigeminal-olfactory connection with HSV1 infection? We think Chlamydia goes through the olfactory system to damage the EC and hippocampus.
Paul Klapper: There seems to be some confusion about herpes simplex and its interaction with the peripheral nervous system. Neuroanatomists will tell you that there is no clear route for HSV to track from the trigeminal ganglion back to the limbic areas of the brain—the site of herpes encephalitis. However, does the virus actually need to reactivate from this site to reach the brain? Not necessarily. It is possible that during primary infection, herpes simplex virus is delivered to sensory nerves other than the trigeminal nerve. It is entirely credible that the virus enters olfactory nerves at this time and establishes latency within the CNS as well as the PNS. The triggers that cause the virus to reactivate in the PNS must not be the same as within the CNS; otherwise, there would be more evidence of CNS latency. However, as Ruth and others have shown, virus can be found in normal brain, supporting the proposition that virus is in the brain and obviating the more tortuous proposal that virus reactivated within the trigeminal nerve ganglion then undergoes anterograde transport through the spinal cord to reach the brain.
Q: I would suggest that the olfactory pathway is a key to the entry into the vulnerable areas of the brain having earliest involvement in AD.
Paul Klapper: The olfactory nerve route for delivery of virus to the CNS is, as outlined above, a credible route for delivery of virus to the CNS during primary HSV infection.
Q: We proposed the idea that HSV would travel into the trigeminal nucleus in the brain stem (Bearer, 2004).
Paul Klapper: There is an extremely large literature on HSV in nervous tissue because of studies over more than 50 years. Studies on peripheral nerve transit, and reactivation, have continued during this time with even more vigor. Many of these studies provide a very strong experimental basis for the proposal that HSV infection is a potential trigger for Alzheimer's.
Q: If 85 percent of elderly humans are infected with HSV, why do only a few (10 percent) get AD?
Paul Klapper: We ask a similar question of herpes encephalitis. Why, when 85 to 95 percent of the population carry one or both types of HSV, is herpes encephalitis so rare? This is one of the most complex of human viruses. It has very complex interactions with both the innate and adaptive immune system. We are unlikely to properly explain either herpes encephalitis or Alzheimer’s in the near future. The more we uncover, the more we realize how little we know. Herpes encephalitis does, however, exist and must be treated. The accumulated evidence amassed by Itzhaki and other researchers suggest HSV may be a trigger for Alzheimer’s. Given this, we should be conducting a trial of treatment even if we cannot completely understand the process.
Ruth Itzhaki: Our studies on human brains indicate that it is HSV1 in brain, plus ApoE4 genotype status, that confers a high risk of AD. Therefore, although HSV1 reaches the brains of ApoE2 and 3 carriers as well as ApoE4 carriers, i.e., all are probably equally susceptible to infection, we propose that, when the virus reactivates, the ApoE4 carriers suffer much greater damage, and after repeated reactivations, eventually develop AD.
Q: There is a lot of work on the swine alpha herpes virus, pseudorabies virus, that is used in animal models. How well this replicates human disease is a question.
Paul Klapper: There is a very wide variety of work on the use of 160+ herpes viruses in a huge range of animal models of PNS and CNS infection over the years. The reality is that none of the models can properly mimic human disease. We have tried in herpes encephalitis, and while we can model selected facets of the human disease, they are all incomplete. I suspect the same is true in trying to model HSV in Alzheimer’s. The best model is the human disease itself.
Q: Have you discussed HSV vaccines?
Paul Klapper: Vaccines for HSV have been in development since the 1930s. We still do not have one, and there is a very real difficulty in delivering the vaccine before we become infected. Most of us acquire HSV1 soon after we lose passively acquired maternal immunity. There is then a very short time in which to deliver protective vaccination, and it would have to be a very effective vaccine to ensure lifelong immunity. I do not believe vaccination is a practicable proposition for this virus.
Q: HSV does not infect mice the same way as in humans.
Paul Klapper: Agreed. There is no exact mimic of human disease.
Q: Are there mouse herpes viruses that can be used in these models? Does HSV infect mouse cells, such as N2a cells?
Paul Klapper: I’m not sure about these particular cells, but while HSV is exclusively a human virus, under experimental conditions in the laboratory, it can infect a wide variety of cell types (including mouse cells)—unlike some other viruses, such as varicella-zoster or cytomegalovirus, which even under experimental conditions can only infect a very select series of cells and cell types.
Ruth Itzhaki: It was suggested that only immunocyto-histochemical methods for Aβ, which are not specific and might detect APP rather than Aβ, have been used with HSV1-infected cells or mouse brains. Several studies detected Aβ using ELISA and/or Western blot, as well ICC with several different Aβ antibodies (see Wozniak et al., 2007; Piacentini et al., 2010; Santana et al., 2011). Our study (Wozniak et al. 2007) also used an APP antibody and showed that in infected SH-SY5Y cells, the APP level decreased while Aβ increased. Piacentini et al. showed a similar effect in rat cortical neurons. On the idea that neuronal damage might reactivate HSV1, why invoke an unknown damaging agent when we know that HSV1 is present in many elderly brains and that it does reactivate there—and that HSV1 damages cells? Also, infected ApoE-transfected mice would be a better model for sporadic AD than HSV1-infected transgenic mice overproducing Aβ.
Q: Why are there tens of thousands of lymphocytes in everyone's trigeminal ganglia, even from birth, through very old age? What is their function?
Paul Klapper: Immune defense.
Q: Do we have access to a group of elderly adults who, geographically, have not been exposed to HSV1?
Paul Klapper: HSV1 has a worldwide distribution. Following primary infection, a lifelong latent infection is established. Periodically, the virus reactivates, causing continued immune activation. Presence or absence of HSV-specific antibody therefore provides a good indicator of prior infection. The prevalence of antibody is linked to socioeconomic status: In poor, overcrowded populations, the virus spreads readily and the prevalence of antibody approaches 100 percent. In populations of high socioeconomic status, virus spreads less readily and antibody prevalence is lower. Your target population for sero-negative elderly adults is, thus, wealthy elderly adults (with the proviso that they are not nouveau riche!).
Ruth Itzhaki: If there are any, they’ve not yet been identified. However, as I said in my talk, we know nothing about risk factors for AD in the 40 percent of people who do not have the joint risk factors of brain HSV1 and ApoE4 carriage.
Q: Richard Smeyne's presentation reminded me of a recent study (Braak and del Tredici 2011) showing AD-like tau changes in the locus ceruleus at quite a young age. Is this compatible with your findings?
Paul Klapper: Smeyne mentioned Von Economo's encephalitis (also known as encephalitis lethargica) that occurred in epidemic form during the 1920s, and by 1940 had all but died out. The reports of the Matheson Commission make really interesting reading in relation to the neuropathology of this condition. The epidemic followed the pandemic of influenza A after World War I. While a specific definition of tau changes was not possible in 1920, the descriptions are consistent. Though influenza was not demonstrated in the brains of patients dying with this encephalitis, herpes simplex was found in some cases, and this was long before herpes encephalitis was first properly described in 1947. Smeyne’s results are interesting, but HSV rather than influenza provides a more understandable trigger for Alzheimer’s. Perhaps what Smeyne’s results show us is that influenza, and possibly other viruses, are actually the trigger for reactivation of HSV latent within the CNS.