The actors and events which dictate the ratios of tau exons 2 and 10 are in the process of being defined. In contrast to the rapid increase in understanding at the molecular level, there is still or no clear sense of how tau deposition results in neurodegeneration and cognitive decline. Results from animal, cellular, and in vitro models give inconsistent results: some suggest that tau aggregated in tangles is the toxic species, while others indicate that aggregated tau is in fact an inert pool and a safe "warehouse" of otherwise toxic soluble tau species (reviewed by Brandt et al. 2005; Feinstein and Wilson 2005; King 2005). The latter theory gains support from both invertebrate and vertebrate animal models that overexpress tau, in which neurodegeneration occurs without obvious tangle formation.
Tau transgenic mice only recapture a subset of the FTDP symptoms, and the details of the pathology that they develop depend on the specifics of their transgene (Brandt et al. 2005). However, although the causal chain between the tau molecule and the neuronal death that heralds dementia is still unclear, it is certain that the tau splicing mutations are sufficient to cause neurodegeneration by altering the ratio of tau isoforms.
It is intriguing that disturbance in the relative tau isoform abundance should result in tangle formation and neuronal death. This correlates with the finding that mice which overexpress tau develop severe neuropathies or gliopathies regardless of transgene details (Brandt et al. 2005). This stands in stark contrast to tau null mice, which are viable though defective in neuronal maturation, muscular strength, and cognition (Dawson et al. 2001; Harada et al. 1994; Ikegami et al. 2000). These results suggest that tauopathies are variants of a dosage disease, like chromosomal trisomies.
The tau splicing variants affect specific functions of the protein: exon 2 modulates interactions with the membrane (Brandt et al. 1995) and may be involved in signal transduction mechanisms and growth cone dynamics
(Liu et al 1999). In this connection, it is interesting that the tau-homozygotes have morphologically altered small-caliber neurons, which have the highest proportion of microtubules close to the membrane (Harada et al 1994). Exon 10 increases affinity to microtubules (Lovestone and Reynolds 1997) and its presence may affect cytoskeletal fluidity. The two regulated domains also include potential sites for in vivo phosphorylation. Inclusion of these domains can alter tau affinity for microtubules, and influence its interactions with other cytoskeletal or membrane components, including regulatory kinases and phosphatases.
Skewed ratios of tau isoforms clearly influence the activity and effects of tau: besides the well-documented results of altered exon 10 inclusion, excess inclusion of exons 2 and 3 causes gliopathy and spinal cord degeneration (Higuchi et al. 2002) and tau is cleaved to its N-terminal fragment very early in neuronal apoptosis, in turn becoming an effector of the process (Canu et al. 1998; Fasulo et al. 2000). Moreover, tau transcription decreases greatly in neuronal cells vulnerable to apoptosis (Esclaire et al. 1998), whereas it increases in sufferers of Down syndrome, who often develop early-onset dementia (Mehta et al. 1999; Oyama et al. 1994).
One theory that would accommodate the behavior of all tau mutations is that tau polymerization (whether with itself or with other ligands) may require a precise stoichiometry of isoforms to correctly discharge one or more functions, which extend beyond just interaction with microtubules. Any disturbance in tau ratios may lead to the accumulation of tau unasso-ciated with microtubules and its eventual precipitation into NFTs. Incorrect levels or species of soluble tau may sequester other cytoskeletal or membrane components, thereby disturbing axonal transport or architecture. Also, since tau in NFTs is irreversibly withdrawn from circulation, increase of NFTs may decrease tau levels below a limit critical to neuronal viability. Thus, altering the tau isoform ratio may correspond to either loss of function or gain of toxic function. At the cellular level, toxicity is defined by the final outcome of axonal disarrangement and eventual neuronal death. If enough neurons die, the brain can no longer rewire and reroute local functions, eventually resulting in the clinical presentations of dementia. Short-lived mammals such as rodents appear largely immune to such diseases, which require chronic accumulation. Disease onset is accelerated in pedigrees bearing mutations that make them more susceptible to deposition of insoluble proteins. Formation of aggregates has been proposed as the underlying unifying cause of neurodegenerative diseases (Singleton et al. 2004).
Increasing knowledge of the specific functions of the tau isoforms, as well as identification of the factors that regulate their splicing, will give significant insights into the normal and abnormal operation of the cooperative networks that establish and maintain neuronal morphology. Results from such work may reveal the processes common to types of dementia in which NFTs are the sole or major pathological manifestation and in the long term give us a handle for ameliorating, preventing, or even reversing dementia - a specter that looms ever darker as the human lifespan lengthens.
Was this article helpful?