Monday, November 29, 2010

New insights into Rett syndrome

A pair of papers from the lab of Fred Gage has provided new insights into the molecular and cellular processes affected in Rett syndrome. This syndrome is associated with arrested development and autistic features. It affects mainly girls, who typically show normal development until around age two, followed by a sudden and dramatic deterioration of function, regression of language skills and the emergence of autistic symptoms. It is caused mainly by mutations in the gene encoding MeCP2, which resides on the X chromosome. Complete removal of the function of this gene is effectively lethal, explaining why Rett syndrome is not observed in boys – males who inherit that mutation are not viable. Females, who have a back-up copy of the X chromosome survive but subsequently show the symptoms of the disease.

The function of the MeCP2 protein seems very far removed from the kinds of symptoms observed when it is deleted. The job of MeCP2 is to bind to DNA that carries a specific chemical tag – a methyl group – which marks DNA for repression. When MeCP2 binds, it recruits a host of other proteins which shut down that section of DNA and prevent any genes within it from being expressed. How a defect in a process that is so fundamental could result in such specific symptoms has been a mystery.

A major barrier in understanding these processes has been the inability to assay the effects of the mutation in this gene in neurons of people who carry it. After all, unlike some other cell types, one cannot easily simply extract neurons from patients. (They tend to be using them). New stem cell technologies developed over the last few years offer a way around this problem. It is possible to extract fibroblasts from patients with a simple skin biopsy. By transfecting these cells with genes that are normally expressed in embryonic stem cells it is possible to “de-differentiate” them – to turn them back into a stem cell. (The difference between a skin cell and a stem cell lies in the genes that are being expressed – transfecting the cells with the master regulatory genes that determine embryonic stem cell identity forces the expression profile back to that state). These “induced pluripotent stem cells” (iPS cells) can then be encouraged to differentiate into any of the cell-types of the body, including neurons. In this way, a virtual biopsy of a patient’s neurons can be obtained.

Gage and colleagues did exactly that, generating neurons in a dish from patients with Rett syndrome. I make that technique sound simple, but of course it isn’t, and these experiments represent a technical tour de force. They were then able to characterise various parameters of these neurons to assay more directly the molecular and cellular effects of MeCP2 mutation. These experiments revealed a not unexpected defect in the formation of synapses between Rett mutation neurons. Neurons from Rett mutation-carriers developed normally and showed normal electrophysiological properties but made fewer synapses with each other and showed a concomitant decrease in network activity. I say not unexpected because it had previously been shown that mouse neurons carrying a MeCP2 mutation show similar effects. This fits with highly convergent findings from autism genetics showing that many other implicated genes function in synapse formation.

What is important about the iPS cells, compared to the information that can be learned from studying mouse cells with MeCP2 knocked out, is that they give a picture of the effects, first, of the specific mutation in this gene in each patient, and second, of the genetic background of each patient, which may modify the effects of the MeCP2 mutation. This gives a far more direct view of the specific effects of each patient’s complete genotype on the development and function of their neurons.

While defects in synapse formation suggest a fundamental role for MeCP2 in neural development, which might imply an irreversible defect, in fact several lines of evidence suggest that the requirement for the function of MeCP2 may be ongoing, in processes of activity-dependent wiring, where neurons within networks strengthen connections based on their patterns of activity. This fits with the apparently normal early development, prior to age two, of girls with Rett syndrome, and also with evidence from mouse models that restoring MeCP2 function in adults can largely reverse the symptoms. These discoveries therefore hold out the promise that intervention in Rett syndrome patients, even in older children, may be effective.

Gage and colleagues tested a couple potential therapies on the neuronal networks derived from Rett syndrome patients and were able to show some degree of rescue of the defects. One of these, the protein insulin-like growth factor-1 (IGF-1), was previously shown to be effective in partially rescuing the defects in MeCP2 mutant mice, most likely by stimulating greater synapse production and compensating for the loss of MeCP2 activity. Clinical trials are now planned to test the efficacy of this approach in patients. Having the cells derived from patients should also greatly facilitate screening for new drugs that can correct the neuronal network defects.

Another paper from the same group, also analysing these cells, revealed a far less expected effect – one that suggests (far more speculatively) the possible involvement of a totally different pathogenic mechanism. One of the functions of the system that methylates DNA is to defend the genome against invaders. Our genome is riddled with parasitic elements – pieces of DNA that can replicate themselves and “jump” around the genome. Fully 45% of our “human” genome is made up of these so-called transposable elements. Most of the copies of these elements are inactive but a subset can generate new copies that will integrate at random into the genome. What has this got to do with Rett syndrome?

Well, MeCP2 is apparently one of the proteins whose job it is to shut down these transposable elements. Gage and colleagues could show that one particular class of these elements, called L1 elements, was far more active in cells derived from Rett syndrome patients. The L1 elements expressed higher levels of the proteins they encode and they generated additional copies of themselves, which were scattered around the genome. Interestingly, this effect seems to be restricted to neurons, presumably because the function of MeCP2 is especially required in that cell-type.

Though highly speculative, this raises the idea that high rates of somatic mutation (somatic meaning it happens in the body, not in the germline and thus will not be inherited), caused by L1 elements jumping around and landing in the middle of genes, may contribute to the severity and also the variability of the phenotype caused by MeCP2 mutations. The alternative is that the L1 transposition has no pathogenic effect but is simply a consequence of the Rett syndrome mutations. Future experiments will be required to tell which of these possibilities is correct.

Marchetto MC, Carromeu C, Acab A, Yu D, Yeo GW, Mu Y, Chen G, Gage FH, & Muotri AR (2010). A model for neural development and treatment of rett syndrome using human induced pluripotent stem cells. Cell, 143 (4), 527-39 PMID: 21074045

Muotri AR, Marchetto MC, Coufal NG, Oefner R, Yeo G, Nakashima K, & Gage FH (2010). L1 retrotransposition in neurons is modulated by MeCP2. Nature, 468 (7322), 443-6 PMID: 21085180

Monday, November 22, 2010

A synaesthetic mouse?

An amazing study just published in Cell starts out with fruit flies insensitive to pain and ends up with what looks very like a synaesthetic mouse. Penninger and colleagues were interested in the mechanisms of pain sensation and have been using the fruit fly as a model to investigate the underlying biological processes. Like any good geneticist faced with profound ignorance of how a process works, they began by screening for mutant flies that are insensitive to pain. Making use of a very powerful genetic resource developed in Vienna (a bank of fly lines expressing RNA interference constructs for every gene in the genome) they screened through all these genes to see which ones were required in neurons for flies to respond to pain. (In particular, pain caused by excessive heat).

Why should anyone care how a fly feels pain? Well, like practically everything else you can think of, the basic physiology and molecular biology of pain sensation is very highly conserved from flies to mammals. It starts with specialized proteins called TRP channels, which are ion channels that span the cell membrane and allow ions to pass across it in response to various stimuli. Some of these TRP channels respond specifically to painful stimuli, some even more specifically to painful heat, and these molecules are highly conserved. The hope was that by screening for other genes they would identify additional conserved elements of the pathway.

This was exactly what they found. Among hundreds of new mutants that were insensitive to pain, they focused in this report on one, a gene called straightjacket. This gene codes for a protein called alpha2delta3, or CACNA2D3, which is a member of a conserved family of proteins that make up part of a calcium channel. These proteins are involved in modulating neurotransmission and also in some aspects of development, including the formation of synapses. Interestingly, mutations in other members of this gene family are associated with bipolar disorder, schizophrenia, Timothy syndrome (the symptoms of which include autism), epilepsy and migraine.

This particular gene is conserved in mammals and the authors show that mutation of the gene in mice also leads to insensitivity to pain induced by heat, but not to painful mechanical stimuli – a remarkably specific functional conservation. In addition, they show suggestive evidence that variants in the gene in humans are also associated with a higher pain tolerance. These latter data will have to be replicated but tantalizingly suggest that variation in this gene in humans may contribute to differences in pain sensitivity.

Mutation of this gene seems to cause pain insensitivity not by blocking pain responses in the sensory neurons or by blocking transmission of this signal to the brain, but by blocking transmission from the first relay station of the brain, the thalamus, to the cortex, where it must pass to be consciously perceived. The authors could show that the sensory neurons still respond to painful stimuli and that a spinal pain reflex was intact. They also used functional magnetic resonance imaging in mice to show that the thalamus was active as normal in response to painful stimuli. However, a network of areas in the cortex (the “pain matrix”) was completely unresponsive. Somehow, deletion of CACNA2D3 alters connectivity within the thalamus or from thalamus to cortex in a way that precludes transmission of the signal to the pain matrix areas.

This is where the story really gets interesting. While they did not observe responses of the pain matrix areas in response to painful stimuli, they did observe something very unexpected – responses of the visual and auditory areas of the cortex! What’s more, they observed similar responses to tactile stimuli administered to the whiskers. Whatever is going on clearly affects more than just the pain circuitry.

The authors suggest that this kind of sensory cross-activation may represent a model for synaesthesia, which is characterised by very similar effects. While this condition is highly familial, no genes have yet been isolated for it. Could CACNA2D3 be a viable candidate? It certainly seems possible, though one point suggests that whatever is happening, while similar to developmental synasthesia, may be somewhat distinct.

Synaesthesia usually involves an extra percept in response to some stimulus, without any decrement in the response to the stimulus itself. So, people who see colours when they hear music hear the music normally – the colour is just part of that experience. This is rather different from a situation where one sense is deficient and is taken over by another. That situation can arise due to injury, for example, and can even be surgically induced in animal models (used to study brain plasticity). One recent report (see below) described a patient who had a lesion in the thalamus in the somatosensory nucleus. This region was subsequently invaded by fibres carrying auditory information so that the patient was able to feel sounds. (The auditory fibres were activated by sound, which cross-activated the somatosensory area, which communicated this activity to the somatosensory cortex, where it was perceived as a touch on the surface of the body).

Could such an effect explain what was happening in these mice? Perhaps for the pain circuits, though one would typically expect that they would be invaded by other senses, rather than the other way around. But for the tactile stimuli, the message was apparently still getting through to the somatosensory cortex, it was just also activating visual and auditory areas. That starts to look like a pretty good model for synaesthesia. Whether it really is would most convincingly be demonstrated by finding a mutation in this gene in someone with synaesthesia. A good place to start might be testing the carriers of the variants in this gene in humans which affected pain sensitivity for any signs of synaesthesia.

Even if it does not correspond exactly to what we call developmental synaesthesia, one can predict that something pretty strange would result from mutation of this gene in humans. Given that every base of the genome is probably mutant in someone on the planet it seems certain that such mutations will eventually crop up.

It is not yet clear what cellular mechanism can explain the cross-activation observed in the mutant mice. One can imagine any number of scenarios, including structural rewiring between thalamic nuclei (which are specialized to transmit different types of sensory information) or from thalamus to cortex. Alternatively, changes in neurotransmission might explain the effects, for example by damping down cross-inhibitory processes that normally sharpen responses to one sense at a time. One way to dissociate these would be to see whether blocking the function of the protein just in adults is sufficient to induce the effect or if it has to be blocked during development. This might be achieved using drugs – a close relative of CACNA2D3 is blocked by gabapentin, a drug used in humans as an antiepileptic and also to block neuropathic pain (like that which can arise due to shingles, for example). Whether this or a similar drug could affect the A2D3 subunit is not, I think, known, but no doubt someone is now looking for a drug that can.

Neely GG, Hess A, Costigan M, Keene AC, Goulas S, Langeslag M, Griffin RS, Belfer I, Dai F, Smith SB, Diatchenko L, Gupta V, Xia CP, Amann S, Kreitz S, Heindl-Erdmann C, Wolz S, Ly CV, Arora S, Sarangi R, Dan D, Novatchkova M, Rosenzweig M, Gibson DG, Truong D, Schramek D, Zoranovic T, Cronin SJ, Angjeli B, Brune K, Dietzl G, Maixner W, Meixner A, Thomas W, Pospisilik JA, Alenius M, Kress M, Subramaniam S, Garrity PA, Bellen HJ, Woolf CJ, & Penninger JM (2010). A Genome-wide Drosophila Screen for Heat Nociception Identifies α2δ3 as an Evolutionarily Conserved Pain Gene. Cell, 143 (4), 628-38 PMID: 21074052

Beauchamp MS, & Ro T (2008). Neural substrates of sound-touch synesthesia after a thalamic lesion. The Journal of neuroscience : the official journal of the Society for Neuroscience, 28 (50), 13696-702 PMID: 19074042

Wednesday, November 3, 2010

Announcing the Wiring the Brain conference 2011

I am pleased to announce the Wiring the Brain conference, which will be held over the 12th-15th April 2011, in Ireland. This is an international scientific conference which aims to explore how the brain is wired and what happens when that wiring is faulty.

It will bring together world-leaders in developmental neurobiology, psychiatric genetics, molecular and cellular neuroscience, systems and computational neuroscience, cognitive science and psychology. A major goal is to break down traditional boundaries between these disciplines to enable links to be made between differing levels of observation and explanation.

We will explore, for example, how mutations in genes controlling the formation of synaptic connections between neurons can alter local circuitry, changing the interactions between brain regions, thus altering the functions of large-scale neuronal networks, leading to specific cognitive dysfunction, which may ultimately manifest as the symptoms of schizophrenia or autism. Though the subjects dealt with will be much broader than that, this example illustrates the kind of explanatory framework we hope to develop, level by level, from molecules to mind.

A list of confirmed speakers is provided below. We are excited to have an outstanding programme of leading researchers across many different fields. The full programme is available at Registration and abstract submission are now open. You can follow updates on the meeting and pre-meeting discussion topics on the Wiring the Brain Facebook group.

The conference is being held in association with Neuroscience Ireland and with BioMed Central and we are delighted to have them both involved. We have also received generous support from Science Foundation Ireland and from other sponsors (listed on the conference website).

The venue is the beautiful Ritz Carlton hotel in Powerscourt, Co. Wicklow, a convenient drive from Dublin airport and one of the most scenic areas of the country.

We hope to see some of you there!

The Organising Committee

Kevin Mitchell, Trinity College Dublin
Aiden Corvin, Trinity College Dublin 

Isabella Graef, Stanford University

Edward Hubbard, Vanderbilt University

Franck Polleux, The Scripps Research Institute

Keynote lectures

Gyorgy Buzsaki, Rutgers University
- brain oscillations and cognitive functions

Carla Shatz, Stanford University,
- activity-dependent mechanisms of neural development

Chris Walsh, Harvard Medical School
- genetics of cortical development and cortical malformations

Plenary speakers

Rosa Cossart, INSERM U901, Université de la Méditerranée, Marseilles
- neuronal network development and function

Ricardo Dolmetsch, Stanford University
- neuronal signaling pathways; molecular mechanisms in autism

Dan Geschwind, University of California, Los Angeles
- genetics and pathogenic mechanisms of autism; brain systems biology

Michael Gill, Trinity College Dublin
- genetics and pathogenic mechanisms of psychiatric disorders

Anirvan Ghosh, University of California, San Diego
- molecular mechanisms of neuronal connectivity

Melissa Hines, University of Cambridge
- sexual differentiation of the nervous system

Josh Huang, Cold Spring Harbor Laboratories
- molecular mechanisms of synaptogenesis

Heidi Johansen-Berg, University of Oxford
- diffusion-weighted tractography in the human brain

Mark Johnson, Birkbeck College, University of London
- cognitive development, neuroconstructivism

Maria Karayiorgou, Columbia University
- genetics and pathogenic mechanisms of schizophrenia

Isabelle Mansuy, University of Zurich
- epigenetic mechanisms of synaptic plasticity and dysfunction

Andreas Meyer-Lindenberg, University of Heidelberg
- functional and structural neuroimaging in psychiatric disorders

Bita Moghaddam, University of Pittsburgh
- network development and mechanisms of psychiatric dysfunction

Tomas Paus, University of Nottingham
- maturation of cortical connectivity in adolescence

Linda Richards, Queensland Brain Institute
- axon guidance, cortical connectivity

Akira Sawa, Johns Hopkins University
- molecular and cellular functions of psychiatric risk genes

Bradley Schlaggar, Washington University, St. Louis
- functional connectivity networks

Klaas Stephan, University of Zurich
- computational modeling of brain connectivity

Pierre Vanderhaeghen, University of Brussels
- molecular mechanisms of cortical development