Consensus paper: pathological role of the cerebellum in autism

Abstract

There has been significant advancement in various aspects of scientific knowledge concerning the role of cerebellum in the etiopathogenesis of autism. In the current consensus paper, we will observe the diversity of opinions regarding the involvement of this important site in the pathology of autism. Recent emergent findings in literature related to cerebellar involvement in autism are discussed, including: cerebellar pathology, cerebellar imaging and symptom expression in autism, cerebellar genetics, cerebellar immune function, oxidative stress and mitochondrial dysfunction, GABAergic and glutamatergic systems, cholinergic, dopaminergic, serotonergic, and oxytocin-related changes in autism, motor control and cognitive deficits, cerebellar coordination of movements and cognition, gene-environment interactions, therapeutics in autism, and relevant animal models of autism. Points of consensus include presence of abnormal cerebellar anatomy, abnormal neurotransmitter systems, oxidative stress, cerebellar motor and cognitive deficits, and neuroinflammation in subjects with autism. Undefined areas or areas requiring further investigation include lack of treatment options for core symptoms of autism, vermal hypoplasia, and other vermal abnormalities as a consistent feature of autism, mechanisms underlying cerebellar contributions to cognition, and unknown mechanisms underlying neuroinflammation.

Figure 1

Timing properties of the olivocerebellar network. a Organization of the olivocerebellar system. Inferior olive (IO) neurons are electrotonically coupled pacemakers that exhibit subthreshold oscillations in membrane potential and project directly to the Purkinje cells as climbing fibers. Purkinje cells are GABAergic and project to deep cerebellar nuclei, which in turn project to motor, autonomic, and limbic cerebral structures. A separate output pathway returns to the IO at the site of gap junctions and releases GABA to regulate the degree of electrotonic coupling and oscillation. b Intracellular fill showing the complex dendritic arbor of a macaque monkey IO neuron. C. An IO neuron from the macaque monkey shows subthreshold oscillations in membrane potential that entrain spiking and may provide a timing signal, as has been suggested in rodents. (After Welsh et al [47]).

Figure 2

Potential mechanisms depicting the role of oxidative stress and mitochondrial dysfunction in the development and pathophysiology of autism.

Figure 3

Reduced FMRP leads to a reduction of many GABAA receptor subunits, potentially contributing to altered GABAergic transmission and balance of GABA/glutamate in the brain. This effect may possibly explain the likelihood of seizure and cognitive deficits in subjects with autism and others with neuropsychiatric disorders. When activated by mGluR5, FMRP acts to inhibit protein synthesis. Without FMRP, protein synthesis is increased, resulting in internalization of AMPA receptors, leading to long term depression. In addition, increased protein synthesis may also be responsible for altered morphology of dendrites, epileptiform activity, and impaired synaptic pruning in autism. Reprinted from [289] with permission from Elsevier.

Figure 4

Left: Photomicrograph taken from 3[H]- sensitive film through the human inferior olivary complex (IO) from a normal control adult case. Right: In this pilot study, a statistically significant decreased density of 3[H]AFDX labeled cholinergic muscarinic Type 2 (M2) Receptors is demonstrated in the Medial Accessory Olive (MAO). Binding parameters: 5nM 3[H]-AFDX rinsed in 10mM Tris-HCl cold buffer; 10µM atropine used as a displacer ; Exposure time: 22 weeks. Abbrev.: DAO, dorsal accessory olive; fmol/mg, femtomoles per milligram protein; IO, inferior olive; MAO, medial accessory olive; POd, dorsal principal olive; POl, lateral principal olive; POv, ventral principal olive.

Figure 5

A representation of a posterior view of flattened human cerebellar cortex, vermis and deep nuclei shown in the coronal plane. Areas for which there is consistent evidence of involvement in higher cognitive functions, including executive control, memory processes, and language, are highlighted in dark gray and black. This includes posterior lobules VI-VII/Crus I–II which are separated from anterior lobules I–V by the primary fissure. Lighter shaded lobules (I-V and VIII) are dedicated to skeletomotor and oculomotor control, although it should be noted that motor control pathways have been documented throughout the cerebellum including posterior lobules, lobule IX and the flocculus-nodulus. Posterior vermian lobules VI-VII and their connections with caudal fastigial nuclei modulate conjugate eye movements. More inferior vermian lobules serve as termination sites of spinocerebellar pathways involved in proprioception and, in conjunction with the flocculus-nodulus, they organize vestibular control. The vermis also has been implicated in affect regulation [192]. Motor pathways synapse in caudal fastigial nuclei, interpositus nuclei, and dorsal segments of dentate nuclei. Cognitive systems involve ventrolateral cells within dentate nuclei. VIIAf: folium of vermis; VIIAt: tuber of vermis.

Figure 6

Schematic illustration of how spatiotemporal spike activity patterns in the neocortex are selectively detected by the cerebellum resulting in precisely timed cerebellar output responses. a Schematic drawing of several pools (P1 – PN) of neurons (drawn as circles, N1 –N7) in the neocortex connected through excitatory projections and propagating synchronous activity from one pool to the next. Gray filled circles represent neurons projecting to the pontine nuclei, the relay nuclei from where mossy fiber projections to the cerebellum originate. b Firing patterns of excitatory neurons in the pontine nuclei (Mf1 – Mfn) with axons projecting as mossy fibers to the cerebellum. The spatiotemporal activity patterns in the neocortex drive either unaltered or transformed spatiotemporal activity patterns in the pontine nuclei neurons. c Cerebellar granule cells receiving mossy fiber input generate action potentials that travel along the slow conducting parallelfibers (conductance velocity Vpf ≈ 0.5 m/s). If the combination of time delay and spatial separation in the mossy fiber inputs match the conductance velocity of the parallel fibers (e.g. Vpf = d1/Δt1) sequential mossy fiber activity will be synchronized by the delays introduced by parallel fibers and result in synchronized inputs to Purkinje cells (PC), triggering precisely timed Purkinje cell output to deep cerebellar nuclei (DCN) and eventually to the neocortex via the cerebellar-thalamo-cortical pathways.

Figure 7

Photomicrographs of cerebellar sections from Lurcher (Lc/+), wildtype (+/+), and Lc/+↔+/+ chimeras taken at both low (top row) and high (bottom row) magnification demonstrating the loss of Purkinje cells in Lc/+ and chimeric mice compared to a control sample. Purkinje cells were stained immunocytochemically with the anti-Calbindin antibody (Chemicon) and appear dark brown in a single monolayer above the blue-stained cerebellar granule cells (cresyl violet counterstain). a Section from a +/+ control cerebellum with normal Purkinje cell number and cerebellar size. b Section from the cerebellum of a high-percentage +/+ chimera with a relatively minor loss of Purkinje cells. c Section from the cerebellum of a low-percentage +/+ chimera showing a relatively major loss of Purkinje cells. d Section from a Lc/+ cerebellum demonstrating the complete loss of Purkinje cells and extreme decrease in cerebellar size. Scale bar = 500 µm a–d, 120 µm a'–d'. Copyright © 2006 by the American Psychological Association. Reproduced with permission from [290]. The use of APA information does not imply endorsement by APA.

Figure 8

In a series of experiments using chimeric mice with varying numbers of Purkinje cells we tested the relationship between Purkinje cell number and behaviors relevant to the autism phenotype. Scatterplots depicting significant relationships between Purkinje cell number and (a) ambulatory events in an open field, (b) breakpoint in a progressive ratio task, (c) learning errors during the third reversal of a conditional visual discrimination, and (d) percent correct on a delayed matching-to-position task (24 sec delay). The best fit regression line is shown in each graph. Mice used in these analyses did not have motor deficits. Collectively these results indicate that low percentage chimeric mice are hyperactive (a), show higher levels of repetitive behavior (b), and have larger deficits in executive function (c) than high percentage chimeric mice. Reduced numbers of Purkinje cells were also surprisingly related to enhanced short term memory (d), which may be analogous to the savant skills shown by some patients with autism. a and b are reprinted from [287] with permission from John Wiley & Sons, Inc. c is reprinted from [284] with permission from Elsevier. d is reprinted from [285] with permission from John Wiley and Sons, Inc.

Figure 9

(Top) Neuronal circuitry underlying cerebellar modulation of PFC dopamine neurotransmission. Cerebellar modulation of dopamine release in the PFC may occur via polysynaptic inputs from cerebellar nuclei to dopamine-containing cells in the ventral tegmental area or via a monosynaptic input to thalamic projections making close appositions with dopamine terminals synapsing onto PFC pyramidal cells. Glutamatergic pathways are shown as red lines and the dopaminergic pathway as a green line. Dashed lines represent PFC feedback to cerebellum via the pontine nuclei. Nuclei abbreviations are shown in the ovals. To develop an animal model of autism, we used electrochemical methods to determine how and by what neural circuits cerebellar activity modulates PFC dopamine release and electrophysiological techniques to determine the impact of this modulation on PFC cellular activity in Lurcher mice with a complete loss of Purkinje cells and wildtype controls. (Bottom) Electrical stimulation of the cerebellar cortex evokes a long lasting increase in dopamine release in the PFC of wildtype mice; an effect that is absent in Lurcher mice with 100% loss in Purkinje cells. Dopamine release in the PFC evoked by 25 pulses (black bar) of stimulation (200 µA) at 50 Hz of the (a) cerebellar cortex and (b) contralateral ventromedial portion of the dentate nucleus in urethane anesthetized wildtype mice (blue lines) and Lurcher mice (red lines) bearing a 100% deficit in Purkinje cell numbers. Serotonin and norepinephrine reuptake blockade fails to alter cerebellar cortex evoked dopamine release in the PFC. Compared to the pre-drug control response (c, green line), selective blockade of dopamine reuptake in wildtype mice with nomifensine (20 mg/kg i.p.) significantly enhanced PFC dopamine release evoked by 100 pulses (black bar) of stimulation (200 µA) at 50 Hz (d, red line), whereas blockade of serotonin and norepinephrine reuptake with a combined fluoxetine and desipramine injection (20 mg/kg i.p. each), respectively, failed to alter the evoked response (d, blue line). In (a to d) the blue, red, and green center lines and outer black lines are the mean+SEM, respectively (n=6 for wildtype and Lurcher mice). Reprinted from [180] with permission from John Wiley & Sons, Inc