Molecular aspects of tau proteins in Alzheimer's disease and frontotemporal dementia

Due to aggregation in the brain of dysfunctional tau proteins, both Alzheimer's disease (AD) and frontotemporal dementia (FTD) are referred to as tauopathies, an entity also including a number of other neurodegenerative disorders. In tauopathies, the normal microtubule-supportive function of tau is compromised, e.g. by abnormal phosphorylation or an altered ratio between tau isoforms with three (3Rtau) and four (4R tau) microtubule-binding repeat regions. Increased knowledge of the disease mechanisms in tauopathies has been achieved mainly by discoveries in molecular genetics. To date, approximately 20 mutations in the tau gene have been identified, all causing rare familial forms of FTD. The tau mutations are believed to exert the pathogenic effect either by increasing expression of 4R tau or by making the tau molecule more prone to aggregate. Moreover, being highly polymorphic, the tau gene is inherited as the two extended haplotypes A and B.

Previously, the tau A haplotype was found to confer susceptibility for several neurodegenerative disorders and was, in interaction with the apolipoprotein E (ApoE) epsilon4 allele, also identified as a risk factor for AD. AD patients display elevated tau levels in cerebrospinal fluid (CSF) and CSF-tau can thus be used as a diagnostic tool in the dementia investigation. However, the method's usefulness is restricted due to the invasive nature of the lumbar puncture and suboptimal disease specificity. There is, hence, a large need for easily accessible and dementia-specific biomarkers and we therefore aimed at exploring tau in peripheral tissues. In this study, fibroblasts from AD subjects with the Swedish mutation in the amyloid precursor protein were investigated. By ELISA, measuring all forms of tau, levels between 65-246 ng/ml were detected in fibroblast lysates but with no significant differences between AD and controls. By western blot, both AD and control fibroblasts were found to contain four of the traditional six tau isoforms as well as four additional bands corresponding to big tau, high molecular weight tau isoforms with an unknown function.

We also analyzed tau in plasma from patients with FTD, AD and vascular dementia. By ELISA, tau immunoreactivity was found in plasma from 21% of the subjects but levels were not associated to any of the dementias investigated. After further characterization, the immunoreactive species were estimated to have the approximate size of 160 kDa and were, like tau, stable for heat and perchloric acid. The influence of tau polymorphisms in FTD was investigated by microsatellite- and restriction fragment length polymorphism-analyses. For the tau alleles per se, no association to FTD or to any of its subgroups could be seen. However, a significant interaction between the tau A haplotype and the ApoE epsilon4 allele was identified as a risk factor for FTD (p= 0.006, OR 9.6).

Moreover, immunohistochemistry and western blot were performed on brain tissue from AD, FTD and control brains by adopting 3R tau- and 4R tau-specific antibodies. 3R tau was found to be the major neuronal tau species in the AD brains and other features specific for AD included subregional differences in hippocampal 3R tau and 4R tau staining, marked phosphorylation of 3R tau and 4R tau-reactive astrocytes. FTD brains, as well as control brains, displayed a much weaker staining with a slight 4R tau dominance. To conclude, the levels of tau in human fibroblasts and plasma were not altered in dementia patients. However, several of the traditional as well as big tau isoforms, were found in fibroblasts whereas either tau polymers or an unidentified tau-like protein were present in plasma from a minority of both demented and control subjects. Furthermore, as was recently shown for AD, the tau A haplotype and the ApoE epsilon4 allele were found to interactively increase the risk also for FTD, although the pathogenic explanation for this interaction remains to be established.

Finally, the observed dominance of neuronal 3R tau along with increased 3R tau phosphorylation and possible 4R tau scavenging by astrocytes may represent rescuing phenomena in the degenerating AD brain. The search for new biochemical markers in AD, FTD and other dementias should be strongly pursued as the use of new efficient drugs will depend on reliable diagnostic methods. Moreover, the pathogenic pathways associated with tau polymorphisms and alternative splicing of tau should be further investigated. An increased understanding of the underlying molecular mechanisms is fundamental for the development of new therapies against these devastating brain disorders.

List of scientific papers

I. Ingelson M, Vanmechelen E, Lannfelt L (1996). "Microtubule-associated protein tau in human fibroblasts with the Swedish Alzheimer mutation. " Neurosci Lett 220(1): 9-12
https://pubmed.ncbi.nlm.nih.gov/8977136

II. Ingelson M, Blomberg M, Benedikz E, Wahlund LO, Karlsson E, Vanmechelen E, Lannfelt L (1999). "Tau immunoreactivity detected in human plasma, but no obvious increase in dementia. " Dement Geriatr Cogn Disord 10(6): 442-5
https://pubmed.ncbi.nlm.nih.gov/10559557

III. Ingelson M, Fabre SF, Lilius L, Andersen C, Viitanen M, Almkvist O, Wahlund LO, Lannfelt L (2001). "Increased risk for frontotemporal dementia through interaction between tau polymorphisms and apolipoprotein E epsilon4. " Neuroreport 12(5): 905-9
https://pubmed.ncbi.nlm.nih.gov/11303757

IV. Ingelson M, Froelich Fabre S, Volkmann I, Sundstrom E, Naslund J, Lannfelt L, Bogdanovic N (2001). "Tau with three microtubule-binding repeats (3R tau) are the major tau isoforms in Alzheimer neuropathology." Mol Brain Res (Submitted)