Neurodegeneration: tauopathies, AD and FTD

Aging, in conjunction with genetic and epigenetic factors can cause severe problems in synaptic transmission, eventually resulting in loss of synapses and even of neurons. Neurodegenerative diseases represent major medical, social and economical challenges. Cognitive, behavioral and sensori-motor functions can be differentially affected to different gradations in different types of neurons in different CNS-regions resulting in diverse cognitive, behavioral and/or motor deficits.

Dementias become most prevalent with aging, characterized by progressive dysfunction of different circuits of synapses and neurons, defining distinct symptoms and clinical entities, i.e. Alzheimer's disease (AD), fronto-temporal dementia (FTD), vascular dementia (VD). Despite considerable progress, the pathogenesis remains poorly understood and the options for the patients for early diagnosis and effective treatment are minimal, with a very grim prognosis.

Numerous pathological and functional studies point to the direct involvement of the neuronal cytoskeleton in many neurodegenerative diseases. Normal axonal transport is an absolute prerequisite for the maintenance of synaptic structural integrity and synaptic transmission (Geschwind, 2003; Ehlers, 2004; Mattson, 2004; Guzik and Goldstein, 2004; Baas and Qiang, 2005; Stokin et al, 2005; Lazarov et al, 2005).

Protein tau is a microtubule associated protein (MAP) that is thought to control microtubular integrity in the axons, needed for the bi-directional transport outlined above. Hyper-phosphorylation and aggregation of protein tau is the most obvious pathological hallmark of many neuro-degenerative diseases, including AD. A diverse set of mutations in the tau gene located on chromosome 17, cause a diverse set of tauopathies collectively known as fronto-temporal dementia with parkinsonism (FTDP-17) (Brion, 1998; Spillantini and Goedert, 1998; Buée et al, 2000; Delacourte and Buée, 2000; Heutink, 2000; Lee et al, 2001; Ingram and Spillantini, 2002; Geschwind, 2003; Mandelkow et al, 2003; Rao and Nixon, 2003).

Genetic research into rare genetic variants of neurodegenerative diseases have yielded important genes that constitute essential starting-points for fundamental research into the pathogenesis, i.e. mutations in tau in FTDP-17 and in amyloid precursor protein (APP) and in presenilin-1 (PS1) in AD.

1.Alzheimers Dementia (AD)

Among the dementias, more than 75% of all cases are due to AD. Particularly problematic is sporadic AD as its incidence increases exponentially with age. AD strikes more than 1 in 3 people over the age of 80, being the most important dementia in this, the fastest expanding age-group. Moreover, the success of modern medicine continues to increase our life expectancy and thereby creates additional, and dramatic social and financial burdens on society, on the families of the elderly, and on all coming generations. The estimated cost of an AD patient is about 15000 EUR per year and in the EEC we will have to care for more than 10 million AD patients by 2030.

Clinically is AD characterized by progressive deterioration of all mental functions, with memory processes primarily affected, together with dramatic affective and behavioral changes. AD patients experience a gradually worsening memory, decreasing attention, alterations in mood accompanied by varying periods of aggression, frustration, agitation, anxiety, stereotypic behaviour. As the disease progresses, usually over a period of 7 to 10 years, symptoms intensify and progressively, AD patients become vegetative and totally dependent for all bodily functions. They continue to weaken and ultimately die from secondary complications.

Definitive diagnosis of AD is reached only post-mortem, by microscopic analysis of the brain following silver impregnation (essentially as described by A. Alzheimer in 1906). Diagnostically AD is defined by the combination of pathological lesions in specific areas of the brain, mainly in hippocampus and cortex, i.e. amyloid plaques and neurofibrillary tangles (NFT) (fig. 1). The extracellular amyloid plaques begin diffusely and evolve into senile or neuritic plaques. They are mainly built of amyloid peptides of 40 and 42 amino acids long (Aβ40/42). NFT on the other hand form inside neurons as accumulations of paired helical filaments (PHF) that are semi-crystalline aggregates of hyper-phosphorylated protein tau.

Fig. 1: Amyloid plaques and neurofibrillary tangles (NFT).

Genetics of AD

The majority of AD cases (>99%) are sporadic with age of onset >65 and without any known genetic or epigenetic cause or origin. Aging is the major risk factor for AD, with an exponential increase with age and although many genes have been proposed as associated, many are not confirmed and most even contested. The only exception is the proven genetic contribution of ApoE4, since the vast increase in frequency of the ε4 allele of the ApoE gene (chromosome 19q13.2) in AD populations was confirmed by hundreds of studies. Thus, ApoE4 has gained status as “major genetic risk factor” whereby carriers of 1 or 2 ε4 alleles have a 5 and 10 times higher risk to become demented after 65 as opposed to carriers of the common ApoE3 alleles. The genetic and epigenetic data argue for a multifactorial and polygenic cause of sporadic AD. Less than 1% of all AD cases are familial, dominantly inherited with an early age of onset (25-55 years). These EOFAD cases are caused by mutations in Amyloid Precursor Protein (APP) or Presenilin (PS1/PS2) genes.

Most importantly, all AD patients have, independent of their genetic make-up the same brain lesions on post-mortem pathological examination, i.e. amyloid plaques and neurofibrillary tangles, which is in fact the definition of AD. Therefore, the study of familial AD with mutants of APP and PS yields information that is relevant not only for the amyloid pathology as "trigger" in EOFAD, but by extension also for all the sporadic AD cases. In addition, however, it provides the strongest argument for a causal link and hierarchy of amyloid over tau-pathology, which constitutes the basis of the "amyloid cascade hypothesis" (Selkoe, 2005). Nevertheless, at least as important and often forgotten, is the fact that EOFAD patients develop also the typical tauopathy concomitant with their early amyloid pathology, indicating that the tauopathy is not only an inherent part of the diagnostic definition of AD, but also of the pathology per se. In addition, it offers a strong argument for the hypothesis that tau is, if not the trigger, at least an important "executer" of the neurodegeneration.

Molecular aspects of AD

The undisputed breakthrough in “molecular” AD research was the biochemical identification of amyloid peptides (Aß) as the major constituents of senile and vascular amyloid deposits in AD brain (Glenner and Wong, 1984). This led to the identification of APP, their precursor from which they are proteolytically excised. The follow-up genetic studies of EOFAD identified mutations in APP, PS1 and PS2 as their cause, and in turn providing tools needed for fundamental research in molecular, cellular and transgenic analysis and model-building.

AD is not only the most frequent dementia but also the major frequent tau-opathy, sharing the important pathological feature of intracellular neurofibrillary tangles (NFT) with many other tau-opathies, that can differ widely in their clinical manifestations. This can be explained by the different types of neurons in different brain-regions that are actually struck by the tauopathy, despite the fact that the common denominator is the formation of NFT as final aggregates of hyper-phosphorylated protein tau. In AD the NFT are rather unordered assemblies of paired helical filaments (PHF), which are semi-crystalline aggregates of hyper-phosphorylated protein tau. In the various types of FTD, the filaments can also be straight or ribbon-like. Together with the amyloid plaques, diseased neurons filled with PHF in their processes and with NFT in their cell-bodies, constitute the diagnostic definition in post-mortem AD brain, appearing as argyrophilic neuronal cell-bodies, dystrophic neurites and fine neuropil threads. Their regional distribution and density correlates with cognitive decline and with indices of neuro-degeneration in hippocampus and cortex, the brain regions that are functionally and clinically directly implicated in AD.

Since the amyloid- and tau-pathology are inseparable in AD, we need to define their mechanistic relation. This will not only help to understand the ethiology of AD, but will also be essential for our fundamental understanding of the mechanisms and specific processes of normal axonal transport, of the functions of protein tau and of APP. The latter is in great need to be clarified in terms of its contribution to normal axonal and synaptic functions, the precise functional role of its complex proteolytic processing and of its cellular trafficking.

This package of fundamental knowledge that needs to be gathered at the fundamental level, will undoubtedly also have profound clinical implications in terms of novel diagnostic and therapeutic options and candidates in the tau-opathies in general, not only for AD but also for FTD.

2. Fronto-temporal Dementia (FTD)

Frontotemporal dementia (FTD) is clinically characterized by a gradual onset of changes in personality and social behavior, by language dysfunction and associated with Parkinson symptoms or motor neuron disease. FTD strikes between the age of 35 and 75 and while presenting as clinically heterogeneous, can be mistaken for AD or a psychiatric disorder. Depending on the study, the frequency of FTD is 5-15% of all types of dementia.

FTD presents either sporadically or in about half of the cases with a family history of dementia, but usually not well defined. Linkage of various cases to chromosome 17 was first established in 1994 in a pedigree with a disorder designated "desinhibition-dementia-parkinsonism-amyotrophy complex". Since then, about 25 different mutations have been identified in the tau gene in as many families with FTD. The different mutations can explain the diversity in symptoms despite a common mechanism of intra-cellular pathogenic inclusions of aggregated, hyper-phosphorylated protein tau. Other tauopathies with widespread but variable tau pathology include Pick's disease, progressive supranuclear palsy, corticobasal degeneration, … but no mutations are identified.

The human tau gene contains 15 exons, but not all are expressed in the tau mRNA in adult CNS. Exons 2, 3, and 10 are alternatively spliced to yield a total of six isoforms characterized by 0, 1 or 2 amino-terminal inserts, that are thought to exert or modulate the spacing and/or interaction of microtubules with membranes. Exon10 encodes the 2nd of the four microtubule-binding domains and its splicing in or out gives rise to either the tau-4R or tau-3R isoforms. Many of the FTD mutations are situated within exon 10 and are thought to decrease the binding of protein tau to microtubules. In addition, and interestingly several non-coding mutations are located in the introns and near the splice junction of exon 10, which evidently cannot cause expression of a mutant tau protein, but are proposed to increase the tau-4R/3R isoform ratio (reviewed by Goedert et al, 1998; Heutink, 2000; Ingram and Spillantini, 2002; Geschwind, 2003; Avila et al, 2004; Iqbal et al, 2005).

The mutations in the tau gene in FTDP-17 conclusively demonstrates that mutant tau is sufficient to produce neurodegeneration and dementia, in the absence of amyloid or other defects. Significantly, an aberrant increase in wild-type tau-4R isoform, alone or due to distortion of the tau-3R/4R isoform ratio as in the FTD-families with intronic mutations, provides the strongest impetus to test the hypothesis that tau-4R itself can be neurotoxic, as proposed in AD (Brion, 1998; Spillantini and Goedert, 1998; Heutink, 2000; Lee et al, 2001; Ingram and Spillantini, 2002; Geschwind, 2003; Mandelkow et al, 2003; Rao and Nixon, 2003; Spittaels et al, 1999; 2000; Terwel et al, 2002; 2005). Any experimental approach should therefore attempt to understand the common mechanism of mis-sense and of intronic mutations, i.e. mutant and wild-type tau-4R, as cause of FTD.

Protein tau aggregation and filament formation is the pathological hallmark of FTDP-17, similar to all tauopathies including AD as the most frequent. Classically, hyper-phosphorylation and mutations are proposed to lead to aggregation of tau and to disruption of microtubular transport systems, synaptic dysfunction and neurodegeneration. This simplified statement dramatically underlines the fact that molecular details of the mechanism at every step and each aspect, are lacking.