Cellular "tug-of-war" breaks brain symmetry

The brains of vertebrates are asymmetrical, both structurally and functionally. This asymmetry is believed to increase the efficiency of information processing - one hemisphere  is specialized to perform certain functions, so the opposite is left free to perform others. In the human brain, for example, the left hemisphere is specialized for speech. This has been known since the 1860s, when the French physician Paul Broca noted that the aphasia (or inability to speak) which is a common symptom of stroke is associated with damage to a discrete region of the left frontal lobe.

Very little is known about how such asymmetries develop. But now researchers from UCL report a molecular mechanism by which a small population of premature neurons in the developing zebrafish brain is actively pulled from one side of the brain to the other. This cellular "tug-of-war" breaks the anatomical symmetry of the embryonic nervous system, so that cells initially located on both sides end up in a left-sided structure. The findings, which are published in the journal Neuron, shed some light on the evolution of handedness. 

Steve Wilson and his colleagues of UCL's Department of Anatomy and Developmental Biology screened lines of zebrafish for mutants exhibiting symmetry in a part of the brain called the epithalamus. One component of the epithalamus is the light-sensitive the parapineal nucleus, which is normally found only on the left side of the brain. The precursor cells which form this nucleus are initially located on both sides of the embryo, but those on the right migrate across the midline and condense to form the left-sided nucleus.

The genetic screen identified one strain of mutants in which the parapineal nucleus develops symmetrically. This line has a mutation in the gene encoding fibroblast growth factor 8 (Fgf-8), a molecule which plays multiple roles in the developing nervous system, including patterning the anterior-posterior axis. Because Fgf-8 is essential for inducing formation of the cerebellum, zebrafish carrying mutations in the gene lack this structure, and are therefore called acerebellar (or ace) mutants.

Analysis of the dynamics of Fgf-8 activity showed that the gene is initially expressed on both sides of the embryo, in a small cluster of cells just in front of the position at which the parapineal nucleus will form. Later, just before the parapineal cells on the right initiate their migration across the midline, the activity of the gene increases subtly on the left side. One of the four known Fgf-8 receptors was also found to be expressed at increased levels in parapineal cells. These results therefore suggest that the cells are responsive to an Fgf-8 signal during their leftward migration.      

To test this prediction, the researchers blocked Fgf signalling in normal embryos, just before the onset of parapineal cell migration. This was found to disrupt the migration of the cells; the parapineal nuclei had an appearance like those in the ace mutants - they were not clustered together, but instead consisted of a few cells scattered on both sides of the midline. Furthermore, blocking Fgf signalling during migration arrested the movements of the cells en route, suggesting that continuous Fgf-8 activity is required.

These findings were further confirmed by the addition of exogenous Fgf-8 to ace mutants. Plastic microbeads soaked in an Fgf-8 solution implanted  on the midline or on either side of it at an early stage of development restored the migration of parapineal cells, so that the nucleus developed normally. Surprisingly, it was found that the nucleus developed on the left side in approximately three quarters of the embryos, regardless of which side the Fgf-8-soaked beads were placed. This suggests that, once parapineal cells begin their migration in response to Fgf-8, an additional signal is required to specify the direction of their movements.

Earlier work had shown that the leftward migration of parapineal cells is dependent on left-sided expression of another molecule called Nodal. Asymmetries still develop in zebrafish with disrupted signalling, but the side of the body on which they develop is randomized - approximately 50% of the mutants are "left-brained" and the rest are "right-brained". Thus Nodal is involved in lateralization - the side on which an asymmetric structure develops - and not symmetry, but is likely to be the additional signal required for the leftward migration of  the parapineal nucleus. In the present study, the resear