Transgenic mice expressing P301L mutant human tau develop neurofibrillary tangles and motor problems, and most die before 10 months of age (Terwel et al., 2005). Curiously, though, in their youth (eight to 10 weeks of age), these same mice have increased long-term potentiation in the dentate gyrus and outperform wild-type littermates in a test of object recognition memory (ARF related news story on Boekhoorn et al., 2006). “It was a surprise that mutant tau could actually be beneficial,” Van Leuven said.

To probe the underlying mechanism, his team looked at what tau was doing to dendritic spines. First author Anna Kremer and colleagues crossed tau P301L mice—as well as the tau 4R strain expressing wild-type human tau at similar levels—with yellow fluorescent protein (YFP) mice to visualize axons, dendrites, and dendritic spines in vivo in bigenic progeny. Using confocal microscopy, the researchers imaged four brain areas affected by tauopathy and AD—the stratum, radiatum, and stratum oriens in the hippocampus, and cortical layer III above the striatum and above the hippocampus. They did so in young (one- to two-month-old) and adult (seven- to eight-month-old) mice. They measured the spine density, spine length, and spine maturation index, the last being the ratio of mushroom spines—which have mature, functional synapses—to all other spine types.

Analyzing nearly 58,000 spines in total for 20 mice, Kremer and colleagues found that, relative to YFP-only controls, young tau P301L mice sprouted normal numbers of spines, but an unusually high proportion of them were mature. In adult P301L transgenics, maturation index and spine length returned back to control levels, yet spine density remained high. The team saw similar effects in adult tau 4R mice, suggesting that wild-type and mutant tau may affect spine formation and morphology in comparable ways.

The researchers analyzed tau P301L and wild-type mouse brain for biochemical markers that could explain how tau proteins drive spine changes. They found both human P301L tau and endogenous mouse tau in synaptosomes, as judged by Western blot. However, immunohistochemistry using a variety of methods failed to detect appreciable amounts of mouse tau in dendritic spines; only human transgenic tau was there. This suggests that “tau’s actions in physiological conditions are quite different from its actions in pathological conditions,” Van Leuven said.

Jochen Herms of Ludwig-Maximilians University in Munich, Germany, pointed out, as did the authors, that it is possible that spine changes could result not only from dendritic pathology, but also from axonal signals. Axonal pathology has been shown to cause secondary spine loss in other tau-overexpressing transgenic mice, Herms noted in an e-mail to ARF. Tau is predominantly an axonal protein (see Hoover et al., 2010; Zempel et al., 2010), only moving into soma or dendrites when abnormally phosphorylated (see ARF related news story).

The mechanism behind the accelerated spine maturation in young tau P301L mice remains unknown. Dezhi Liao of the University of Minnesota, Minneapolis, speculated the enhancement could be due to compensation effects. Liao and others have shown that spine density and maturation goes up when AMPA receptor activity is blocked in cultured neurons, for instance (Liao et al., 1999; O’Brien et al., 1998).—Esther Landhuis.

Reference:
Kremer A, Maurin H, Demedts D, Devijver H, Borghgraef P, Van Leuven F. Early Improved and Late Defective Cognition Is Reflected by Dendritic Spines in Tau.P301L Mice. J Neurosci. 7 Dec 2011;31(49):18036-47. Abstract