Publication / Source: Neurology Central
Authors: Robert D. Moir and Rudolph E. Tanzi (Massachusetts General Hospital and Harvard Medical School, MA, USA)
An association between chronic infection and amyloidosis, including β-amyloid deposition in Alzheimer’s disease (AD) , has been recognized for almost a century . However, since the discovery in the mid 1980’s that the pathological hallmark of AD, β-amyloid plaques, are generated from neurotoxic amyloid-β peptide (Aβ) the suggestion of infection as an etiology factor in AD has most often been met with skepticism. Nevertheless, recent findings such as the report by Pisa et al.  of fungi in the brains of AD patients have prompted a renewed interest in the ‘infection hypothesis’ of Alzheimer’s disease. Thanks to the decades long efforts of a small subset of AD researchers, diverse epidemiological and experimental data now supports an association between infections in the central nervous system (CNS) and AD. Publication of data consistent with the infection hypothesis has also surged in recent years, with over 80 publications supporting increased CNS infection in AD, cited on PubMed in 2014. However, concerns continue to be voiced about the conclusions that can be drawn from these data.
At least three arguments typically arise in various forms opposing the conclusion that increased CNS infection in AD indicates pathogens are etiological factors for the disease.
First, infections are a consequence of AD and not the driving etiological factor. Infections may arise in AD brain because the failing immunity and compromised blood-brain barrier associated with the disease allows opportunistic pathogens to spread to the CNS. Second, no clear mechanism has been identified for how infection may drive amyloidosis and other pathologies. Infection associated with inflammation alone does not appear to be sufficient. Indeed, Aβ oligomerization appears to drive inflammation rather than the reverse . Third, no single pathogen has been identified that is present in an overwhelming number of AD cases. For any particular pathogen, uninfected AD patients can be identified with extensive β-amyloid deposition. This suggests AD pathology develops independent of the pathogen. These longstanding concerns have stymied the serious consideration of infection as an etiological factor for AD. However the new data from Pisa et al. once again raises the issue that infection needs to be more seriously considered as an etiological agent in AD.
Much of the historical skepticism for pathogens as etiological factors comes from the leading model of amyloidogenesis. Aβ is most often characterized as a functionless byproduct of the catabolism of the β-amyloid protein precursor (APP). Thus, the activities of Aβ peptides are considered intrinsically abnormal, particularly the propensity for self-association. The view that Aβ activities are anomalous was a reasonable hypothesis following the peptides discovery. Initially, disease conditions were thought to induce abnormal processing pathways that lead to Aβ generation . This was shown not to be the case in the 1990’s and the prevailing model shifted to an ‘inadvertent’ constitutive origin for Aβ . However, contrary to this model, almost all other disease-associated proteins have been found to also play normal physiological roles. The contention that Aβ is an incidental product of catabolism, and by extension, the peptides activities serve no physiological function, has persisted almost unchallenged for over two decades. This is in spite of evidence highlighting that Aβ is conserved across most vertebrate species up to at least 400 million years (humans share Aβ42 sequences with coelacanths, an ancient fish taxon) , as well as extensive data indicating broad activity spectra for the peptide. The remarkable conservation of the Aβ sequence and associated activities suggests that the peptide mediates an important normal physiological pathway.