The inferior olivary nucleus is a brainstem structure that projects to the Purkinje cells of the cerebellum. This olivocerebellar projection is known to be important for motor learning and the timing of movements. In a new study published in Developmental Biology, Zhu and colleagues turn their attention to its development — more specifically, how olivary axons are guided to the cerebellum. Zhu et al. present compelling evidence that the cerebellum itself produces long-range, diffusible guidance cues.

The olivary neurons originate in rhombomeres 7 and 8 of the hindbrain, and their axons grow across the ventral midline (floor plate) and project to the cerebellum on the contralateral side. It is thought that the growing axons are attracted to the floor plate by guidance molecules such as netrin-1, and their final topographic organization seems to be determined by local signals within the cerebellum that act over a short range. However, the origins of the cues that guide the axons once they have crossed the midline have remained largely obscure.

Zhu et al. examined the effects of cerebellar tissue on olivary axon growth and guidance by transplanting a cerebellar plate alongside the hindbrain at the same rostrocaudal level as the inferior olivary nucleus. They found that neurons in the caudolateral part of the nucleus, which normally project to the rostral cerebellum, projected to the transplant and could make the correct topographic connections, even when the rostrocaudal polarity of the grafted tissue was reversed. However, the rostromedial olivary neurons, which normally project to the caudal cerebellum, could only make the right connections with the ectopic cerebellum if its polarity was reversed, so that their target was rostral to their cell bodies. This led the authors to suggest that inhibitory signals in the caudal hindbrain might prevent olivary axons from projecting caudally. This idea was supported by experiments in which a cerebellum was grafted adjacent to the cervical spinal cord. In this case, none of the rostromedial neurons, and only a few of the caudolateral neurons, projected to the transplant.

So, the cerebellum clearly releases chemoattractants for olivary axons, but can these molecules diffuse over sufficiently long distances to be candidates for the endogenous cues? To find out, the authors sandwiched a slice of spinal cord tissue, which is normally non-permissive for olivary axon growth, between a caudal hindbrain explant and an explant of cerebellar tissue. They found that the olivary axons could cross the spinal cord 'bridge' to reach the cerebellum, indicating that the cerebellar chemoattractants act over a long range.

Like the axons of many other neuronal subtypes, the olivary axons change their sensitivity to guidance cues as they cross the midline. Zhu et al. found that exposure to floor plate signals made the axons more responsive to cerebellar guidance cues, possibly through upregulation of receptors for cerebellum-derived chemoattractants.

These findings show that the cerebellum not only produces locally acting signals that ensure that the olivary projections make the right topographic connections, but it also secretes long-range cues that guide olivary axons to the cerebellar primordium. The next step will be to identify these signals — a quest that will undoubtedly keep researchers occupied for some time to come.