The Cerebellum: Brain for an Implicit Self (6 page)

Comparison of cerebella in various vertebrate species reveals the evolutionary origin of the lobal and lobular structures of the cerebellum. The flocculonodular lobe is phylogenetically the oldest (thus called the archicerebellum) and closely associated with the vestibular organ (thus called the vestibulocerebellum). The vermis is also old (paleocerebellum), and it is closely associated with the spinal cord (spinocerebellum). The cerebellar hemispheres are “new,” being expanded in mammals and primates in association with the development of the cerebral neocortex (neocerebellum or cerebrocerebellum). Particularly notable are the large paraflocculus in the porpoise and whale, the wide ansoparamedian lobule in the monkey, and the width of the entire hemisphere in the human.

In addition to the right-left transversal lobular structure, the cerebellum is also divided into a number of longitudinal zones by various landmarks (
Voogd, 1964
;
Groenewegen and Voogd, 1977
;
Groenewegen et al., 1979
). First, the connections from the cerebellar cortex to the cerebellar nuclei by Purkinje cell axons are organized in three parts. The vermis is connected to the medial (fastigial) nucleus, the intermediate part of the hemispheres to the interpositus nucleus (in humans, emboliform and globose nuclei), and the lateral part of the hemispheres to the lateral nucleus (dentate nucleus in primates) (Color Plate III). In addition, a part of the vermis and the flocculonodular lobe are connected directly to the vestibular and other nuclei in the medulla oblongata. Second, in the afferent projection from the inferior olive (IO) to the cerebellar cortex, each small area of the IO projects to an anteroposteriorly extended longitudinal zone of the cerebellar cortex (Color Plate III). In this projection, seven major zones (A, B, C
₁
, C
₂
, C
₃
, D
₁
, and D
₂
) are distinguished. The A and B zones are in the vermis, whereas the C
₁
, C
₂
, and C
₃
zones cover the intermediate part of the hemispheres. The D
₁
and D
₂
zones are in the lateral part of the hemispheres. Third, a peptide, zebrin, distributes unevenly in the cerebellar cortex and marks the seven zones (
Leclerc et al., 1992
). A number of marker molecules are now available to label the longitudinal bands. Additional zones such as A
₂
-, X-, Y-, and D
₀
- zones have been defined (see
Apps and Hawkes, 2005
); they are not shown in Color Plates II and III. An interesting finding used intraventricular injection of adenovirus-carried markers, which labeled neuronal progenitor cells in a birth date-specific manner. Such injection in embryonic mice revealed that a cohort of Purkinje cells generated by mitosis on the same day formed a specific set of longitudinal bands, whereas Purkinje cells born one day earlier or later formed different sets of such bands (
Hashimoto and Mikoshiba, 2004
).

The horizontal lobules and longitudinal bands form a latticed map that divides the cerebellar surface into nearly a hundred small areas (Color Plate II)). This lattice provides a guide map for exploration of the cerebellum. Earlier, Bolk (
1906
) noticed in the giraffe that the lobule simplex (lobule VI) was extraordinarily large. In view of the giraffe’s long neck, he pointed out the possibility that this cerebellar area was for the precise control of the long neck by powerful shoulder muscles (Glickstein and Voogd, 1955). Bolk’s idea of a single somatotopical map in the cerebellum does not hold for the entire cerebellum, but it pointed to the presence of functional localization in the cerebellar cortex (
Manni and Petrosini, 2004
). The elephant with a long trunk that is used like a human hand has an extraordinarily large cerebellum (18.6% of the total brain, as contrasted to 10% to 11% in humans), but no regional expansion has been described yet as specifically related to this animal’s nose (
Shoshani et al., 2006
).

Another interesting and long-standing question concerns the prominent difference in the maps of the cerebellum of whales versus primates. In human and nonhuman primates, the cerebellum expands laterally, particularly in crus I and crus II (part of lobule HVII). The large cerebellum in whales (20% to 25% of the total brain), however, is due to an expansion of the paraflocculus, which occupies about three-fourths of the cerebellar surface (
Oelschläger, 2008
;
Oelschläger and Oelschläger, 2009
). It has been suggested that the whale’s large paraflocculus is a result of its adaptation to aquatic life, in which echolocation and acoustic communication are essential for survival and a meaningful social life (
Oelschläger, 2000
). In this regard, it is interesting that in the rat, the auditory cortex projects to the paraflocculus via the pontine nucleus (
Azizi et al., 1985
). In bats, which possess supersonic echolocation, the paraflocculus neurons respond to acoustic stimuli, specifically to their first harmonics (
Horikawa and Suga, 1986
). Hence, it is
probable that the paraflocculus is involved in echolocation and acoustic communication. The question of how the cerebellum contributes to such sonar systems will be discussed later in connection with its role in “sensory cancellation” (i.e., cancelling sensory perturbations evoked by self-initiated movements;
Chapters 15
and
18
). A large expansion of the ventral paraflocculus has been recognized in the endocast of a Cretaceous multituberculate, a mouse-sized mammal that appeared 130 million years ago and became extinct 90 million years later (Kielan-Jaworowska, 1986). It is interesting to speculate that this mammalian species also developed a sonar system for communication in the dark with the aid of the paraflocculus.

Another unique development is observed in the cerebellar valvula of mormyrid fish. This is basically a rostral protrusion of the cerebellum in the midbrain ventricle and much enlarged and folded over the brainstem and the telencephalon (Shi et al., 2008). The valvula is characterized by the prominent pattern of ridges on its dorsal surface. The valvula has been called a gigantocerebellum and intensively studied by neuroanatomists (
Nieuwenhuys and Nicholson, 1967
). It shares the general and basic organizational features with all other cerebella and cerebellar subdivisions but has in addition a number of unique features. For example, the efferent cells (corresponding to cerebellar nuclear neurons in
Chapter 6
, “
Pre- and Post-Cerebellar Cortex Neurons
”) are located close to Purkinje cells, and there are numerous deep stellate cells supplying specific inhibitory projections to efferent cells (
Meek et al., 2008
). This does not occur in usual cerebella (see
Chapter 5
, “
Inhibitory Interneurons and Glial Cells in the Cerebellar Cortex
”). Because the valvula receives much of its input from the electrosensory system, its role in electrosensation is probable (Finger et al., 1981).

The latticed areas have common homogenous microscopic structures, as will be examined later. In functional terms, they play diverse roles in connection with other divisions of the brain and spinal cord. It appears that numerous small “computers” of uniform structure and function are utilized individually for diverse purposes. The still remaining large blank in the mosaic map of the cerebellum implies that many more concrete roles of the cerebellum are yet to be identified.

In addition to the comparative anatomy mentioned previously, valuable strategy used widely in neuroscience research is to place a lesion in a brain and test for subsequent dysfunctions. Cutting, ablating, coagulating, and injecting certain toxic amino acids into brain tissues have been used to make lesions. Clinical cases with discrete cerebellar lesions also provide similarly useful data. Dow and Moruzzi (
1958
)
compiled such data from lesion studies collected up to the middle of the twentieth century. Transient blockade of functions by applying various pharmacological inhibitors or antagonists is a further development, and most recently, genetic manipulation has become a powerful tool to form a lesion in a specific element of neuronal circuits. Elaborate test paradigms for detecting lesion-induced behavioral disorders have also been developed recently.

It is interesting that in phrenology, as initiated by Franz Joseph Gall (1758–1828), the occipital part of the cranium that overlies the cerebellum was assigned as the center of love. This assertion was groundless, but it apparently stimulated pioneers of the nineteenth century. Luigi Rolando (1773–1831) removed the cerebellum in various animals and found subsequent motor disturbances. Jean Pierre Flourens (1794–1867) ablated the cock cerebellum, but seeing the cock still seeming to be attracted to a hen, he discarded the phrenological hypothesis (
Glickstein et al., 2009
). Flourens (
1822
) also observed that animals with a lesioned cerebellum still moved spontaneously, but only clumsily. He concluded that the cerebellum was responsible for movement coordination, in contrast to the cerebrum, which initiated movements via the spinal cord. This was amazing insight for a nineteenth century scientist!

In the early twentieth century, clinical neurology revealed that cerebellar dysfunction in humans was characterized by the loss of smooth, precise movements. Babinsky (1857–1932) defined dysmetria as a characteristic symptom of cerebellar dysfunction. A simple clinical test for dysmetria is a patient’s failure, with the eyes closed, to quickly and accurately touch the nose with an index finger (
Chapter 15
, “
Internal Models for Voluntary Motor Control
”). During World War I, Gordon Holmes (1876–1965) examined soldiers with a discrete gunshot wound in the cerebellum. He found that they exhibited a significantly slower onset of arm retraction on the damaged side. The cerebellum thus appeared to be required for a movement that was too quick to be influenced by sensory feedback. Intention tremor is another symptom frequently observed in cerebellar patients. It is characterized by coarse trembling of a forelimb, which is accentuated by the execution of purposeful voluntary movements such as reaching by the hand. It may expand to involve the head, eyes, and the upper half of the body. Intention tremor has been reproduced in monkeys by cooling the dentate and interpositus nuclei (
Flament and Hore, 1988
). It thus appears that the cerebellum normally prevents intrinsic potential oscillations to ensure smooth movements.

Early on, such subtle control of movements suggested the need for a form of motor learning in the cerebellum. In fact, there were classic observations that suggested the learning capabilities of the cerebellum. Flourens (
1842
) removed the
superior half of the cerebellum in a young cock and observed 15 days later that equilibrium was totally reestablished. However, when he removed the entire cerebellum from a hen, it did not recover its equilibrium even four months after the operation. Luciani (
1891
) reported a notable observation in 1891. He made a partial lesion in the cerebellum of a dog. After the animal’s full recovery, he placed a second lesion adjacent to the original one. A severe motor disturbance ensued as if to suggest that the first and second lesions were made at the same time. This was interpreted to mean that the second lesioned area was involved in recovery from the first lesion. Moruzzi (1910–1986) and his group postulated that cerebellar circuits either have a learning capability in and of themselves, or they are required for recalling a memory stored somewhere else in the CNS (Batini et al., 1976). They apparently considered that the cerebral cortex was a possible memory site, because of the observation that a later cerebral lesion cancelled motor recovery after cerebellar ablation.

The capability of cerebellar circuits to recover from and compensate for a functional deficit, as demonstrated in animals, argues against a perplexing question we sometimes face: why does a human who lacks a cerebellum, as is sometimes reported, exhibit no obvious dysfunction? Glickstein (
1994
) examined a number of such individuals and reported that the cerebellum was still present, albeit severely atrophied. He also emphasized that individuals with such an atrophied cerebellum definitely exhibited certain abnormal motor behavior. It seems likely that the viable portion of the cerebellum may compensate for deficits produced by its damaged portions. Possibly, the cerebral cortex provides additional compensation.

Experimental lesion studies and clinical observations have certainly highlighted the involvement of the cerebellum in the control of movement. As a result, the popular idea that the cerebellum is solely a motor center has tended to prevail. However, Moruzzi (
1940
) recognized several decades ago that the cerebellum contributed to both cardiovascular and respiratory regulation. The involvement of the cerebellum in mental activities has been suggested, albeit only occasionally and mainly on the basis of clinical observations that focused until recently on disturbances in expression using spoken words and gestures (
Chapter 17
, “
Cognitive Functions
”).

In 1837, Jan E. Purkinje (1787–1869) observed cerebellar tissues under a microscope and found oval objects. These were the first individual neurons observed in the literature, and they are now called “
Purkinje cells
.” In the early twentieth century, using the amazing silver staining method, Cajal observed and drew intricate
neuronal network structures that he found throughout the brain (
Ramón y Cajal, 1911
;
Sotelo, 2003
). In the cerebellum, he depicted the characteristic morphology of Purkinje cells, Golgi cells, basket cells, stellate cells, and granule cells (Color Plate IV). He also identified mossy fiber and climbing fiber terminals. His drawings even indicated the possible directions of neuronal signals conducted and transmitted from one cell to another. These elements are arranged in three layers (molecular layer, Purkinje cell layer, and granular layer). The structure is homogenous throughout the cerebellar cortex except for some regional differences. The tradition of Cajal’s fine microscopic anatomy was continued by a number of distinguished anatomists, and in the 1960s it was further advanced by the widespread contributions of electron microscopists.