Expression patterns of Sema3A in developing amniote limbs: With reference to the diversification of peripheral nerve innervation.

Paired limbs were acquired in the ancestor of tetrapods and their morphology has been highly diversified in amniotes in relation to the adaptive radiation to the terrestrial environment. These morphological changes may have been induced by modification of the developmental program of the skeletal or muscular system. To complete limb modification, it is also important to change the neuronal framework, because the functions of the limbs rely on neural circuits that involve coordinated movement. Previous studies have shown that class 3 semaphorins (Sema3 semaphorins), which act as repulsive axonal guidance cues, play a crucial role in the formation of the peripheral nerves in mice. Here, we studied the expression pattern of Sema3A orthologues in embryos of developing amniotes, including mouse, chick, soft-shelled turtle, and ocelot gecko. Sema3A transcripts were expressed in restricted mesenchymal parts of the developing limb primordium in all animals studied, and developing spinal nerves appeared to extend through Sema3A-negative regions. These results suggest that a Sema3A-dependent guidance system plays a key role in neuronal circuit formation in amniote limbs. We also found that Sema3A partially overlapped with the distribution of cartilage precursor cells. Based on these results, we propose a model in which axon guidance and skeletogenesis are linked by Sema3A; such mechanisms may underlie functional neuron rearrangement during limb diversification.

Evolutionary divergence of trigeminal nerve somatotopy in amniotes.

The trigeminal circuit relays somatosensory input from the face into the central nervous system. In central nuclei, the spatial arrangement of neurons reproduces the physical distribution of peripheral receptors, thus generating a somatotopic facial map during development. In mice, the ophthalmic, maxillary, and mandibular trigeminal nerve branches maintain a somatotopic segregation and generate spatially organized patterns of connectivity within hindbrain target nuclei. To investigate conservation of somatotopic organization, we compared trigeminal nerve organization in turtle, chick, and mouse embryos. We found that, in the turtle, mandibular and maxillary ganglion neuron rostrocaudal segregation and trigeminal tract somatotopy are similar to mouse. In contrast, chick mandibular ganglion neurons are located rostrally to maxillary neurons, with some intermingling, supporting previous observations (Noden [1980], J Comp Neurol 190:429-444). This organization results in an inversion of the relative positions and less precise axonal sorting of the maxillary and mandibular branches within the trigeminal tract, as compared to mouse and turtle. Moreover, using the turtle and chick orthologs of Drg11 in combination with Hoxa2 expression and axonal tracings from the periphery, we mapped the chick PrV nucleus position to rhombomere 1, confirming previous studies (Marin and Puelles [1995], Eur J Neurosci 7:1714-1738) and in contrast to mouse PrV, which mainly maps to rhombomere 2-3 (Oury et al. [2006], Science 313:1408-1413). Thus, somatotopy of trigeminal ganglion and nerve organization is only partially conserved through amniote evolution, possibly in relation to the modification of facial somatosensory structures and morphologies.