Index | Round
Tables | Selected
Abstracts | News
Scientists Smitten by Mad Tau Disease
by Luc Buee
22 July 1998. A new epidemic of "mad tau disease" swept Amsterdam at the workshop on Hereditary Fronto-Temporal Dementia and Pick's disease. Fronto-temporal dementia and parkinsonism linked to chromosome 17 (FTDP-17) has been linked to mutations on the tau gene, and, in the words of Peter Davies, most of speakers themselves exhibited symptoms of the disorder: euphoria, disinhibition and impulsivity. (One wonders about other FTD symptoms such as hyperactivity, hyperorality and hypersexuality....)
In addition to the preannounced program, three speakers were added: Peter Heutink, Gerry Schellenberg and Maria Grazia Spillantini. In the first part of the workshop, clinical and neuropathological characteristics of fronto-temporal dementia were described. The subgroup of FTDP-17 was then introduced and description of the mutations found on tau were reported. Finally, biological effects on tau proteins were also presented.
Cummings briefly described clinical chracteristics of FTD versus Alzheimer's disease (abstract 1232). AD increases with age whereas FTD is a typical presenile dementia. In FTD, behavior is particularly affected and there is a severe disinhibition. In contrast to AD, visuospatial ability and calculation are preserved. Some variants may be observed: a left temporal lobe variant characterized by aphasia and right temporal lobe variant showing irritability and impulsivity. Some sub-syndromes may be also distinguished in progressive non fluent aphasia and semantic aphasia. Hodges also emphasized the role of semantic dementia in FTD (abstract 1233). Van Swieten (abstract 1235) described the different subgroups of FTD and introduced the history behind FTDP-17. He indicated that Pick's disease (characterized by the presence of Pick bodies) is likely to be non familial. In other FTD, hereditary accounts for probably between 20 and 38%. Neuropathology and PET studies in sporadic FTD were described by Forster (abstract 1237). Neuropathologically, brains of FTD patients display cortical gliosis, severe neuronal loss and superficial laminar spongiosis. There is a relative sparing of hippocampus. By imaging, atrophy and hypermetabolism are observed in anterior temporal lobe.
Michael Hutton kicked off the genetics part of the session (abstract 1234). Tau mutations were always found in FTDP-17 patients with tau pathology. The presence of tau-immunoreactive intraneuronal and glial inclusions was very common. Among the families studied, three mismutations were found in coding regions G272V, N279K and R406W. Three other mutations affecting the splicing region after exon 10 were also reported at +13, +14 and +16. G272V and P301L affect the microtubule-binding domain of tau protein. On Tuesday, Michel Goedert reported the effects of both mutations in an in vitro system of microtubule assembly (abstract 926; See Audio Recording). Mutated tau isoforms do not bind microtubule and induce microtubule disassembly.
Regarding the mutations in the splicing region, Hutton et al. indicated that they affect the ratio of tau mRNA with exon 10 (E10+)/tau mRNA w/o exon10 (E10-). For these mutations, they demonstrated by RT-PCR that the ratio E10+/E10- was between 2 and 6. For patients with the P301L mutation and control cases, this ratio was between 0.57 and 1.25, suggesting that the mutations in the splicing site induce an increase in E10+ tau isoforms. By exon-trapping, they confirmed these observations and demonstrated that the most striking mutation is at +14 and induces almost a six-fold increase in E10+ tau isoforms. Mutations in the splicing site disturb a stem loop that may stabilize this region of the pre mRNA and decrease access of U1snRNP to this RNA region. Without this stem loop, access of U1snRNP may be facilitated and increase the formation of E10+ tau isoforms. Hutton et al. tested this hypothesis by a rescue method in which they added two new nucleotides to the mutated sequence in order to form a new stem loop. This new structure was tested in their exon-trapping experiment and they showed that there is a decrease in E10+ tau isoforms compared to non-modified mutated sequence. Finally, sequence analysis of this splicing in different animals indicates that if there is no stem loop structure, there is an increase in E10+ tau isoforms.
Heutink focused on FTD in The Netherlands (no abstract available). 38% of FTD cases have a family history. Three families were analyzed by Hutton's group (HFTD I, II and III). Only families I and II showed mutations (P301L and G272V), and the affected individuals are clinically characterized by a disinhibition. Neuropathologically, they exhibited a tau pathology. In family III, no mutation was found on the tau gene, there was no tau pathology and clinically, the characteristics were loss of initiative, rather than disinhibition.
Schellenberg (no abstract available) briefly summarized his group's data (P301L, Seattle families D and F, and Oregon EL; and V337M, Seattle A) and described a new mutation N279K in PPND. He noted that Virginia Lee had described aggregation of E10+ tau isoforms in PPND (abstract 617). Schellenberg also emphasized that the change in nucleotide for the P301L mutation may also create an exon-splicing enhancer sequence.
He reported that no coding mutation was found on the tau gene in PSP patients. He suggested that PSP may exhibit a polymorphism that increases susceptibility.
Wilhelmsen (abstract 1236) reported that FTDP-17 families are highly heterogeneous. In Sweden, the most common mutation in FTDP-17 families is the G272V one. Sequencing of tau gene in exons 9-13 was performed in FTD, tauopathies, PSP and CBD: no other mutation was found.
Finally, Spillantini made a summary of the biochemical analysis of phosphorylated tau isoforms in FTDP-17. In Dutch HFTD II (G272V mutation), tau-immunoreactive neuronal glial structures are found. The neuronal structures are very close to Pick bodies. In Dutch HFTD I (P301L mutation), both neurons and glia are tau-immunoreactive. At the electron microscopy level, tau aggregates into twisted ribbons and to a much lesser extent into PHF. Biochemically, phosphorylated aggregated E10+ tau isoforms and also the E10- isoform with the 29 aa insert (exon 2) are found. In MSTD (intronic mutation described by Spillantini in PNAS, abstract 1294), phosphorylated E10+ tau isoforms aggregate into twisted ribbons filaments and appear by immunoblotting as a major tau doublet and a minor 72 kDa tau variant. Tau-immunoreactivity is found in neurons and glia. In Seattle family A (V337M mutation), neurons are tau-immunoreactive. All six tau isoforms are phosphorylated (with a PHF-like electrophoretic profile) and aggregate into PHF and to a lesser extent into straight filaments. Finally, soluble tau proteins are always similar to those found in control cases, except in MSTD where an increase in E10+ tau isoforms was observed.