Statistics from Altmetric.com
The clinical syndrome of frontotemporal dementia is associated with several neurodegenerative disorders: Pick's disease, corticobasal degeneration, motor neuron disease-associated dementia (MND-dementia), frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17), and frontal lobe degeneration (FLD). These disorders, although they do not match the frequency of Alzheimer's disease, are far from uncommon, and present clinicians and neuropathologists with formidable, if not insurmountable diagnostic difficulties. However, recent advances in cellular and molecular pathology, biochemistry, and molecular genetics have been instrumental in their nosological definition. The discovery of a mutation in the tau protein gene on chromosome 17 in 1998 has established that several phenotypically heterogeneous familial dementias with a confusing variety of names all belong to FTDP-17.1 2 Most but not all frontotemporal dementias are characterised by intracellular inclusions formed by abnormal cytoskeletal components, both in neurons and in glial cells. Pick's disease, corticobasal degeneration, and FTDP-17 belong to the larger group of tauopathies, as their hallmark lesions contain tau protein, distinguishing them from MND-dementia and FLD, two disorders without tau pathology. Of the three tauopathies, FTDP-17 can be defined by its genetic abnormality, whereas the differential diagnosis of Pick's disease and corticobasal degeneration remains difficult and controversial.
Clinically Pick's disease is characterised by frontal and anterior temporal lobe dysfunction and progressive dementia, whereas neuropathologically the underlying frontotemporal atrophy is complemented histologically by the presence of large, swollen, achromatic neurons, the Pick cells, and by tau positive intraneuronal inclusions, the Pick bodies. The clinical features of corticobasal degeneration include asymmetric extrapyramidal signs, parkinsonism, and the “alien limb” phenomenon (apparent purposeful movements which are not under voluntary control) followed by cognitive impairment. Histologically there is neuronal loss, astrocytosis, and tau positive neuronal and glial inclusions, but the most prominent feature is the presence of large, swollen neurons, morphologically indistinguishable from Pick cells. It is the occurrence of these cells in both disorders which causes most of the differential diagnostic problems.
In the human central CNS six tau isoforms are generated by alternative splicing and these are then postranslationally phosphorylated. At molecular level the electrophoretic profiles of aggregated tau proteins in these neurodegenerative disorders are disease specific. For example, although both Pick's disease and corticobasal degeneration are tauopathies, characterised by tau positive cellular inclusions, their tau profile is apparently different: the tau doublet is 64 and 69 kDa in corticobasal degeneration, but 55 and 64 kDa in Pick's disease.3 Here we report the use of tau antibodies which, exploiting the different phosphorylation patterns of these two disorders, has made a more accurate neuropathological diagnosis possible.
Sections from the frontal lobe and the temporal lobe (including the hippocampus) from 10 cases of corticobasal degeneration and 10 cases of Pick's disease were examined by immunohistochemistry, using phosphorylation independent (SMI51, TP007, TP70, 304, 189) and phosphorylation dependent (AT180, AT270, AT8, 12E8) anti-tau antibodies. All sections were immunostained according to a standardised protocol using the avidin-biotin complex (DAKO) with appropriate positive and negative controls. All the swollen neurons and neuronal and glial inclusions were positively immunostained with all the anti-tau antibodies in corticobasal degeneration. However, in Pick's disease all the antibodies but one immunostained the Pick bodies and the large swollen Pick cells: antibody 12E8 gave negative results (figure A and B).
Phosphorylation at the site of Ser262/Ser356 is thought to be one of the most prominent factors affecting the biological activity of tau.3 In corticobasal degeneration antibody 12E8 detected the phosphorylated epitopes Ser262 and/or Ser356, whereas in Pick's disease these sites remain unphosphorylated and consequently this antibody did not label any of the Pick bodies or Pick cells.4 These findings are important both from a practical and from a theoretical point of view. The immediate practical implication is a more accurate neuropathological diagnosis, enabling Pick's disease to be distinguished from corticobasal degeneration; this, in turn, is essential for improving clinical investigations. In addition, new insights into the molecular pathology of these disorders contribute to the nosological definition and better understanding of a complex group of neurodegenerative diseases.