Peter Hayward

Lithium reverses tau pathology in Drosophila

Inhibitors of glycogen synthase kinase (GSK-3 ) can reverse the symptoms caused by tau pathology—one of the signs of Alzheimer’s disease (AD) and frontotemporal dementia (FTD)—in Drosophila. The established use of lithium, a GSK-3 inhibitor, to treat psychiatric disorders opens the potential for fast-track therapy for these dementias. Small-scale trials in patients with mild to moderate AD funded by the UK Medical Research Council will begin in the next few months. Simon Lovestone (Institute of Psychiatry, Kings College London, UK) and colleagues investigated the effects of overexpression and phosphorylation of tau protein in Drosophila larvae and adult flies. Overexpression of human tau caused disruption of axonal transport, which was worsened by tau phosphorylation in the presence of GSK-3 (Mol Psychiatry 2004; published online March 2, DOI: 10.1038/sj.mp. 4001483). “The axonal transport defect resulted in functional loss in intact larvae and adult flies”, explains Lovestone. Larvae with excess phosphorylated tau were less able to crawl and some had total hind-section paralysis; adult flies did less well on a

Word recognition localised to left occipitotemporal cortex

Postsurgical language defi cits in a patient treated for epilepsy have led researchers to suggest that a specifi c site in the left occipitotemporal cortex allows recognition of all the letters in a word at once. Before the surgery the patient, who had had epilepsy for over 30 years, could read normally. But after removal of a small epileptogenic focus in a region thought to be involved in word recognition, the length of time taken to read individual words increased with word length, suggesting that the patient could no longer recognise words as whole entities, but instead had to tackle each word letter-by-letter—a disorder known as “pure alexia”. “We showed that a region, for which we provide precise anatomical and functional delineation, is both causally necessary and specifi c to an essential cultural competence, namely expert reading”, says researcher Laurent Cohen (Hopital Salpetriere, Paris, France). Patients with vascular lesions in this general area have shown similar defi cits in reading; however, because vascular lesions tend to cover a large cortical area, precise delineation of the visual word form area in these patients is not possible. In the patient studied by Cohen and colleagues, not only could the researchers record specifi c activity related to word recognition before treatment, but also the resection was small enough to allow pinpointing of the area involved in word recognition (Neuron 2006; 50: 191–204). In an accompanying Preview of the paper, Alex Martin (National Institute of Mental Health, Bethesda, MD, USA) describes some of the controversies surrounding the existence of visual object-recognition modules. “Of all the claims for a category specifi c brain region, perhaps the most controversial has been the visual word form area”, says Martin. “How could there be a piece of neural tissue dedicated to a recently invented cognitive skill like word recognition?” Cohen suggests that, because it takes years for children to develop expert word recognition, “there is no ready made ‘module‘ for visual word recognition, but a progressive specialisation process that capitalises on the plasticity for the human ventral inferotemporal cortex to build the visual word form area.” The researchers suggest that the specifi c localisation to the left occipitotemporal cortex may develop due to the lateralisation of language areas of the brain.

The neuroanatomy of fatherhood

Parenthood is as much a state of mind as it is simply looking after one’s off spring—at least that is the evidence from a study of marmoset fathers. Males of these small primates show substantial increases in dendritic spines and vasopressin receptors in the prefrontal cortex in response to fatherhood. Previous MRI studies have shown that human parents have activity related stimuli in response to their own children, so the Princeton researchers decided to investigate the changes further. Marmosets provided an ideal model because of the level of parental care provided by the males. “Rat fathers would be just as likely to eat their own babies”, explains Yevgenia A Kozorovitskiy (Psychology Department, Princeton University, NJ, USA), “but in marmosets the fathers are a great model of very involved fathers—a model of an ideal human father.” Directly after giving birth, marmoset females hand over their twins to the males, who will carry the off spring, in fact they off er nearly all parental care except for providing milk. The researchers found that, in male marmosets that had given parental care to their off spring the number of dendritic spines and vasopressin receptors in the prefrontal cortex were higher than in non-fathers (Nat Neurosci 2006; 9: 1094–95). “You might expect to see changes in areas responsible for maintaining hormonal balance, but to see it in such a high-level area involved in goal-directed behaviour was a real surprise”, Kozorovitskiy told The Lancet Neurology. “The fact that the prefrontal cortex is so highly responsive to endocrine and environmental infl uences is particularly important, since it mediates the highest levels of cognitive function”, says John Morrison (Mount Sinai School of Medicine, New York, USA). Given the similar endocrine responses to stress, reproduction, and childrearing of mammals, Morrison suggests that human beings would likely show similar adaptive changes in the prefrontal cortex in response to major parenthood and other fundamental transitions in life, such as puberty, menopause, and intense stress. Kozorovitskiy feels that the changes in the brain are likely brought about by interaction with young and hope to investigate what changes occur in nonparental care-givers. The researchers suspect that the changes in vasopressin receptors are in some way related to dopaminergic reward systems, and would also like to investigate the mechanisms behind these changes.

Neurotrophic factor linked to Rett's syndrome

Findings in a mouse model of Rett’s syndrome suggest that brain-derived neurotrophic factor (BDNF) is involved in the pathogenesis of this X-linked mental-retardation disorder. This new insight into the pathogenesis of the disorder may suggest possible targets for therapeutic intervention. Rett’s syndrome (also known as RTT disease), which affects about one in 10 000–15 000 girls, causes massive deterioration in health starting at about 6 months of age. Patients can survive to adulthood with proper care, but have severe mental retardation, seizures, and ataxia. 80% of human cases are associated with a mutation in MECP2 on the X chromosome. This gene encodes a binding protein that regulates transcription of other genes. Researchers at Massachusetts Institute of Technology developed Mecp2 knock-out mice to investigate the disorder, and early studies suggested that BDNF expression was altered in these mice. In the latest study, Qiang Chang (Whitehead Institute for Biomedical Research, Cambridge, MA, USA) and colleagues show that BDNF expression is low in Mecp2 knock-out mice (Neuron 2006; 49: 341–48). When the researchers completely knocked out the Bdnf gene in Mecp2 knock-out mice, the mice developed symptoms earlier. Furthermore, when BDNF was overexpressed, symptoms of the disorder in mice were ameliorated. “This is the first instance that the Rett disease progression has been altered by changing expression of another gene”, says reseacher Rudolf Jaenisch in a press release issued by the Rett Syndrome Research Foundation. “Our results suggest therapeutic opportunites through the manipulation of BDNF expression”, state the authors. “This paper is very interesting”, says Michael Johnston (Kennedy Krieger Institute at Johns Hopkins University School of Medicine, Baltimore, USA), “in that it links BDNF with the pathogenesis of Rett’s syndrome in these mice.” However, there are several differences between the disorder in mice and Rett’s syndrome in girls. The mice seem to have lower neuronal activity, but in girls, hyperventilation, seizures, abnormal electroencephalograms, and hyperkinetic hand movements point to increased neuronal activity. “Although Bdnf is one of many genes regulated by Mecp2, it is the first one shown to modulate disease progression. Future studies will be aimed at understanding its effect on the course of RTT disease”, the authors conclude.

FOXP2 sequence starts talking

Carrying an APOE4 allele may increase the risk of normal (nondemented) cognitive decline and also make the brain more vulnerable to Alzheimer’s disease, report Scottish researchers. “We know that mechanistically, APOE is involved in neuronal repair, and it’s been shown in head injury patients that if you’ve got the E4 allele, then neuronal repair seems to be less effective”, explains John Starr of the Royal Victoria Hospital in Edinburgh. “If you hypothesise that part of the agerelated changes in cognition have to do with insults to the brain—such as oxidative stress—then that would lead to neuronal damage. So neuronal repair may be one of the things that we need to maintain the brain in a good state.” Starr and co-workers retested 491 non-demented 80 year olds who had participated in the Scottish Mental Survey of 1932. Individuals with and without the APOE4 gene had similar test scores at age 11 years (99 vs 100, respectively). But at age 80 years, those with APOE4 scored significantly lower than those without APOE4 (97 vs 101, respectively; Nature 2002; 418: 932). “An IQ difference of about 4 points might not make a big difference for a particular individual”, says Starr. “But looking at the bigger picture, you might say that these people have more vulnerable brains for some reason, or had more insults and didn’t deal with them quite so well.” Since the researchers took pains to exclude from the study any older adults with even a hint of Alzheimer’s disease or dementia, at issue, he explains, is whether there is “a group of people with vulnerable brains who are older and may be at risk of progressing to Alzheimer’s but aren’t there yet”. Such individuals may have additional risk factors for cognitive decline, such as type-2 diabetes, hyperlipidemia, cerebrovascular disease, or other genetic variants, continues Starr. He also notes that his group recently received funding to study NMDA receptors and superoxide dismutase variants with a view to “getting a background of the kinds of genotypes that might put you at risk.” “You’d then have two therapeutic targets: trying to prevent people from getting to a vulnerable state in the first place, and—at least for older people at risk of sporadic Alzheimer’s disease—preventing a progression from vulnerability to the actual disease.” Marilynn Larkin Gene linked to normal cognitive decline