BackgroundWe recently reported that Parkinsonian and dementia phenotypes emerge between 7-12 months of age in tau-/- mice on a Bl6/129sv mixed background. These observations were partially replicated by another group using pure Bl6 background tau-/- mice, but notably they did not observe a cognitive phenotype. A third group using Bl6 background tau-/- mice found cognitive impairment at 20-months of age.ResultsTo reconcile the observations, here we considered the genetic, dietary and environmental variables in both studies, and performed an extended set of behavioral studies on 12-month old tau+/+, tau+/-, and tau-/- mice comparing Bl6/129sv to Bl6 backgrounds. We found that tau-/- in both backgrounds exhibited reduced tyrosine hydroxylase-positive nigral neuron and impaired motor function in all assays used, which was ameliorated by oral treatment with L-DOPA, and not confounded by changes in body weight. Tau-/- in the C57BL6/SV129 background exhibited deficits in the Y-maze cognition task, but the mice on the Bl6 background did not.ConclusionsThese results validate our previous report on the neurodegenerative phenotypes of aged tau-/- mice, and show that genetic background may impact the extent of cognitive impairment in these mice. Therefore excessive lowering of tau should be avoided in therapeutic strategies for AD.

The homeostasis of insulin-like growth factor-1 (IGF-1) is essential for metabolism, development and survival. Insufficient IGF-1 is associated with poor recovery from wounds whereas excessive IGF-1 contributes to growth of tumours. We have shown that cyclic glycine-proline (cGP), a metabolite of IGF-1, can normalise IGF-1 function by showing its efficacy in improving the recovery from ischemic brain injury in rats and inhibiting the growth of lymphomic tumours in mice. Further investigation in cell culture suggested that cGP promoted the activity of IGF-1 when it was insufficient, but inhibited the activity of IGF-1 when it was excessive. Mathematical modelling revealed that the efficacy of cGP was a modulated IGF-1 effect via changing the binding of IGF-1 to its binding proteins, which dynamically regulates the balance between bioavailable and non-bioavailable IGF-1. Our data reveal a novel mechanism of auto-regulation of IGF-1, which has physiological and pathophysiological consequences and potential pharmacological utility.

Lithium is a first-line therapy for bipolar affective disorder. However, various adverse effects, including a Parkinson-like hand tremor, often limit its use. The understanding of the neurobiological basis of these side effects is still very limited. Nigral iron elevation is also a feature of Parkinsonian degeneration that may be related to soluble tau reduction. We found that magnetic resonance imaging T 2 relaxation time changes in subjects commenced on lithium therapy were consistent with iron elevation. In mice, lithium treatment lowers brain tau levels and increases nigral and cortical iron elevation that is closely associated with neurodegeneration, cognitive loss and parkinsonian features. In neuronal cultures lithium attenuates iron efflux by lowering tau protein that traffics amyloid precursor protein to facilitate iron efflux. Thus, tauand amyloid protein precursor-knockout mice were protected against lithium-induced iron elevation and neurotoxicity. These findings challenge the appropriateness of lithium as a potential treatment for disorders where brain iron is elevated (for example, Alzheimer's disease), and may explain lithium-associated motor symptoms in susceptible patients.