Cerebellum

Overview

Phylogenetically, the initial development of the cerebellum (in fish) took place in relation to the vestibular labyrinth. With the development of quadrupedal locomotion, the anterior lobes of the cerebellum (in particular) became richly connected to the spinal cord. Assumption of the erect posture and achievement of a whole new range of physical and cognitive skills have been accompanied by the appearance of massive linkages between the anterior and posterior lobes of the cerebellum and the cerebral cortex. In general, cerebellar connections with the vestibular system, spinal cord, and cerebral cortex are arranged such that each cerebellar hemisphere is primarily concerned with the coordination of movements ipsilaterally.

The gross anatomy of the cerebellum is described briefly in Chapter 3 , where it may be reviewed at this time.

Functional Neuroanatomy

Phylogenetic and functional aspects can be combined (to an approximation) by dividing the cerebellum into functional regions ( Figs. 27.1 and 27.2 ). The flocculonodular lobe is also known as the vestibulocerebellum because it receives input and directs its output to the vestibular nuclei; it controls eye movements through the vestibular system. (Other contributions from the brainstem and through the pons from the parietal and occipital cortex provide further input and contribute to eye movement control.) Portions of the vermis receive input from the reticular formation, frontal eye fields, and superior colliculus and then project to the deep cerebellar nuclei (the fastigial nuclei ) in the white matter ( Fig. 27.2 ). The fastigial nuclei project to the gaze centres of the brainstem and vestibular nuclei to control saccadic eye movements ( Chapter 23 ).

The spinocerebellum includes the vermis and paravermal (next to the vermis) cerebellar cortex, and receives input from the spinocerebellar tracts (and through the pons cortical input and input from the vestibular nuclei and reticular formation). The vermis projects to the fastigial nuclei and, through projections to the reticular formation and vestibular nuclei where the reticulospinal and vestibulospinal tracts arise, influences postural reflexes of the head and trunk. The paravermal area receives cortical input via the pons and spinal cord input via the spinocerebellar tracts, and projects to the globose and emboliform nuclei ; the two nuclei together are called the interposed nucleus ( Fig. 27.2 ). The interposed nucleus projects to the parvocellular part of the red nucleus and ventrolateral thalamus. The parvocellular nucleus projects to the Inferior olivary nucleus through the rubro-olivary tract and with the cerebral cortex, involved in motor learning and correcting motor activity of the limbs.

The remaining lateral strip is the largest and projects to the dentate nucleus ( Fig. 27.2 ). This area is called the pontocerebellum because of its numerous inputs from the contralateral nuclei pontis. It is also called the neocerebellum because the pontine nuclei convey information from large areas of the cerebral neocortex (phylogenetically the most recent). It is uniquely large in the human brain and plays a significant role in the planning, initiation, control, and correction of voluntary movements as well as cognitive, speech, behaviour, affect, and executive functions.

Microscopic Anatomy

The structure of the cerebellar cortex appears uniform throughout. It is made up of three principal layers: the innermost granular layer (directly adjacent to the white matter), the Purkinje cell layer composed primarily of Purkinje cells, and the outermost molecular layer ( Fig. 27.3 ).

The granular layer contains billions of granule cells , whose somas are only 6 to 8 μm in diameter. Their short dendrites receive so-called mossy fibres from all sources except the inferior olivary nucleus. Before reaching the cerebellar cortex, the mossy fibres, which are excitatory in nature, give off collateral branches to the deep cerebellar nuclei.

The axons of the granule cells extend into the molecular layer where they divide in a T-shaped manner to form parallel fibres (they run parallel to the transverse fissures of the cerebellum) that are parallel to other parallel fibres, but perpendicular to the axes of the Purkinje cell dendritic tree. They make excitatory (glutamatergic) contacts with the distal dendrites of Purkinje cells. The granular layer also contains Golgi cells (see later) whose dendrites are stimulated by the parallel fibres of the granule cells. The Purkinje cell layer consists of very large Purkinje cells . The fan-shaped dendritic trees of the Purkinje cells are the largest dendritic trees in the entire nervous system and at right angles to the parallel fibres ( Fig. 27.4 ).

The dendritic tree of Purkinje cells receives synapses from a huge number of parallel fibre axons (as many as 100,000, but only a small number are active at any one time) of granule cells, each one making successive synapses (one or two per Purkinje cell) on the dendritic spines of about 400 (possibly more) Purkinje cells. Not surprisingly, stimulation of only a small number of granule cells by mossy fibres has a small facilitatory effect upon many Purkinje cells. Many thousands of parallel fibres must act simultaneously to bring the membrane potential of the Purkinje cell to threshold.

Each Purkinje neuron also receives a single climbing fibre from the contralateral inferior olivary nucleus. In stark contrast to the one-per-cell synapses of parallel fibres, the olivocerebellar fibre divides at the Purkinje dendritic branch points and makes numerous synaptic contacts (thousands) with those dendritic spines ( Fig. 27.4 ). A single action potential applied to one climbing fibre is enough to elicit a short burst of action potentials from its Purkinje cell, known as a complex spike (see later). The complex spike is so powerful that, for some time after it ceases firing, the synaptic effectiveness of these parallel fibres undergoes a long-term depression (LTD) . In this sense, the Purkinje cells remember that they have been excited by olivocerebellar fibres.

The axons of the Purkinje cells are the only axons to emerge from the cerebellar cortex. Remarkably, they are entirely inhibitory (the neurotransmitter is GABA) in their effects. Their principal targets are their corresponding deep cerebellar nuclei. They also give off collateral branches, mainly to Golgi cells .

The molecular layer is almost entirely taken up with Purkinje dendrites, parallel fibres, supporting neuroglial cells, and blood vessels. However, two sets of inhibitory interneurons are also found there, lying in the same plane as the Purkinje cell dendritic trees, but perpendicular to the granule cell axons or parallel fibres ( Fig. 27.3 ). Near the cortical surface are small stellate cells ; close to the Purkinje cell layer are larger basket cells . Both sets are contacted by parallel fibres, and they both inhibit Purkinje cells. The stellate cells synapse upon dendritic shafts, whereas the basket cells form a ‘basket’ of synaptic contacts around the soma, as well as forming axoaxonic synapses upon the initial segment of the axon. A single basket cell synapses upon some 250 Purkinje cells. As one group, or row, of Purkinje cells becomes active, the rows on either side will be inhibited by these interneurons.

The final cell type in the cortex is the Golgi cell , whose dendrites are contacted by parallel fibres and whose axons divide extensively before synapsing upon the short dendrites of granule cells. The synaptic ensemble that includes a mossy fibre terminal, granule cell dendrites, and Golgi cell boutons is known as a glomerulus ( Fig. 27.5 ). Golgi cells function to limit the output of the mossy fibre inputs onto the granule cells.

Spatial Effects of Mossy Fibre Activity ( Fig. 27.6 )

As already noted, cerebellar afferents (except for olivocerebellar ones) form mossy fibre terminals after giving off excitatory collaterals to one of the deep cerebellar nuclei. The afferents excite groups of granule cells, which in turn facilitate many hundreds of Purkinje cells arranged in rows beneath the parallel fibres of the granule cells. Along the row of excitation, known as a microzone (smallest efferent unit of the cerebellar cortex), the Purkinje cells begin to fire and to inhibit neurons in one of the deep nuclei. At the same time, weakly facilitated Purkinje cells along the edges of the microzone are inhibited by stellate and basket cells. As a result, a row of Purkinje cells is sharply focused, while those rows on either side will be inhibited. The excitation is terminated by Golgi cell inhibition via the granule cells that stimulated them (self-limiting). Powerful excitation will last longer because highly active Purkinje cells inhibit underlying Golgi cells, thereby allowing the granule cells to continue to fire.