A protective role for herpes simplex virus type-1-specific humoral immunity in Alzheimer’s Disease

KEYWORDS:

• Alzheimer’s disease
• antibody
• avidity
• human herpes simplex virus type-1
• humoral immunity
• magnetic resonance imaging
• mild cognitive impairment

Alzheimer’s disease (AD) is the most common form of senile dementia and, due to the worldwide increase of life expectancy, has rapidly become a major medical issue in industrialized countries. The economical toll of AD on public health is enormous, and is estimated to hover actually around US $ 818 billion, representing 1.09 % of global gross domestic product (GDP); this toll is expected to grow further, reaching US$ 3 trillion within the next 15 years [1].

AD is a progressive neurodegenerative disease characterized by the accumulation of abnormally phosphorylated tau (τ) protein, neuronal loss, and formation of senile plaques and neurofibrillary tangles; this results in memory disturbance, attention deficit, and a progressive degree of intellectual impairment. Senile plaques are extracellular and are predominantly constituted of amyloid-β (Aβ), a peptide produced by the proteolysis of amyloid precursor protein (APP), whereas neurofibrillary tangles are intracellular and are formed by τ protein. Aβ peptide accumulation, the major pathological hallmark of AD, has traditionally been considered to be the main trigger of pathology, as such accumulation triggers neuroinflammation and neurodegeneration. Notably, though, an intriguing alternate view of the role of Aβ has recently emerged. Thus, because this peptide is endowed with antimicrobial functions [2], its accumulation could be the result of an overproduction finalized at protecting the brain against infections with viruses or bacteria [3].

An impressive research effort is undergoing worldwide in the attempt to find therapies that could modulate the progression of AD or, in the best possible scenario, prevent it; success is nevertheless hampered by the consideration that the etiology of AD is still unknown. A role for infectious agents has long been suspected in AD [4]. Infectious agents would interact with inflammatory, environmental, and genetics factors, working as triggers that would initiate the processes leading to Aβ formation, abnormal τ phosphorylation, and neuronal loss. Human herpes simplex virus type-1 (HSV-1), in particular, a neurotropic virus with a lifelong latency in neurons, has been indicated as a possible culprit in the etiology of AD.

HSV-1 commonly infects humans, with up to 90% of the population harboring it by the sixth decade of life [5]. Thus, (1) HSV-1 DNA can be detected in a high proportion of both AD and elderly brains [6]; and (2) Intrathecal HSV-specific immunoglobulin G (IgG) can be observed in the majority of both AD patients and elderly controls [7]. The presence of a possible connection between HSV-1 and AD was first suggested more than 30 years ago [8]. This suggestion was subsequently strengthened by a number of in vitro data indicating that HSV-1 infection of neuron cell lines results in the same molecular alterations that are observed in AD, including APP processing [9] and intracellular distribution [10], and τ protein phosphorylation [11]. Notably, the possible connection between HSV-1 and AD has also been reinforced over the years by a number of epidemiological results (for a very recent state-of-the art document, see: Itzhaki et al. 2016 [12]).

In the past 2 years, we decided to investigate the possible role of HSV-1 in AD by merging together immunologic, virologic, and imaging techniques. Initial results showed the presence of a significant positive correlation between HSV-1 specific antibody (Ab) titers and cortical volumes, as measured by magnetic resonance imaging (MRI), in the right orbitofrontal cortex, the left temporal pole, and the right inferior frontal gyrus/temporal pole in patients with mild AD. Notably this correlation was lost in amnestic mild cognitive impairment (aMCI), a clinically intermediate situation between normal cognitive decline due to aging and AD [13].

What do these results mean? We believe that HSV-1-specific Ab could be protective in the early stages of AD [14]. Thus, the alterations of the blood–brain barrier (BBB) seen in AD would facilitate the penetration of peripheral lymphocytes and of HSV-1 Ab into the central nervous system (CNS). The higher titers of intra-CNS HSV-1-specific Ab obtained in this way would slow down and eventually block viral reactivation, limiting the damages in those brain regions where the BBB permeability increases. Of course an alternate way to explain these findings is that a (genetically determined?) weaker HSV-1-specific humoral response would result in the loss of control of viral reactivation, leading, over time, to the brain damage and the neurodegeneration that characterize AD.

Is the protective effect due to HSV-1 specific Ab alone, or could Ab directed toward other viruses play a role in this phenomenon? To verify this possibility, Abs for two other endemic viruses were analyzed in the same populations of AD and aMCI individuals. Results did not show the presence of any relation between cytomegalovirus (CMV)- and human herpesvirus 6 (HHV-6)-specific antibodies and any MRI or neurological parameter [13,15,16]. Because HHV-6 infection of the CNS, in particular, prevalently interests the frontotemporal region, the same brain area most commonly affected in AD [17]; these results support the possibility that HSV-1 is specifically involved in the MRI-detected brain impairment that characterizes AD.

A possible working model is that AD is initiated once the equilibrium between HSV-1 replication and its control by the immune response is lost, possibly because primary HSV-1 infection happens later in life, when the efficacy of the immune system is declining, or because HSV-1 reactivates once to many. Results from two independent longitudinal studies [18,19] seem to confirm this idea. In these cohorts, the presence of HSV-specific IgM was shown to associate with a significantly increased risk of developing AD. Because IgM Abs are present in serum for a limited time during primary infection, their presence indicates that primary HSV-1 infection happening later in life is not optimally contained by the immune response, possibly facilitating the development of AD.

Parameters other that Ab titers or their subclasses can be used to determine the efficacy of humoral responses; Ab avidity, in particular, the relative strength with which Abs bind antigen, is an important index of such efficacy. Recent results [20] showed that HSV-1 IgG avidity index are significantly increased in aMCI compared to AD, with the lowest indexes seen in healthy elderly individuals. These data suggest that HSV-1 humoral responses are more potent in aMCI, but do not furnish any arguments either supporting or denying a role for the virus in the development of AD. aMCI is nevertheless an excellent model to verify this hypothesis, as aMCI usually converts to AD over time. Conversion is unpredictable as it can happen very quickly in some individuals, whereas in other cases the neurological scenario does not evolve over prolonged periods of time. The obvious question, thus, is whether aMCI conversion to AD can be predicted by the efficacy of HSV-1-specific humoral responses. To try and find an answer to such question, we performed retrospective analyses of immune parameters in a cohort of aMCI individuals who did or did not progress to AD within a 24-months period. Results showed that HSV-1-specific Ab avidity was significantly higher at baseline in MCI-non-converters, in whom Ab titers were also increased. Importantly, a positive correlation was detected as well in these individuals between HSV-1 antibody titers and avidity, and MRI-evaluated cortical volumes. Again, the observed association was strictly HSV-1 specific, as no relationships were found between the avidity of HHV-6-specific Ab and conversion to AD [21]. Notably, the expression of the viral replication-specific sensors toll-like receptors (TLR) 8 and TLR9 on immune cells, and the production of interferon lambda, an antiviral cytokine, was significantly augmented as well in those aMCI individuals in whom AD conversion was not observed [22], indicating – not surprisingly – a role for innate immunity as well in preventing the progression of aMCI to AD.

These results seem to support the idea that the quality of HSV-1-specific host humoral response is crucial for the development of disease. Thus, progression to AD is prevented, or at least slowed down, in those aMCI individuals in whom strong HSV-1 humoral responses are observed. On the other hand, individuals in whom humoral immune responses are weaker, possibly because of genetic reasons or as a consequence of the aging-associated weaning of immune system, will more likely develop AD.

The smoking gun is still missing, but more and more bricks are added to the working hypothesis that HSV-1 infection/reactivation plays a role of primary importance in AD pathogenesis. These results raise the possibility that antiviral drugs could have a beneficial effect against the development of AD, in particular in the early phases of disease. Notably, recent data obtained in vitro or in the animal model using natural products known to be endowed with mild anti-herpetic activity, including brown algal polysaccharides [23], turmeric curcumin [24], and Brazilian nuts [25], indicate that these compounds might limit the accumulation of Aβ and τ protein, characteristic features of AD brains. Finally, these data also suggest that measurement of HSV-1 Ab avidity in serum – a simple noninvasive test performed on easy to collect clinical samples – could help fine-tuning disease prognosis.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with financial interest in or financial conflict with the subject matter or material discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Additional information

Funding

This paper was not funded.