Search

ELOVL5 (Elongase of Very-Long Fatty Acid 5) gene encodes for an enzyme that elongates long chain fatty acids, with a marked preference for polyunsaturated molecules. In particular, it plays an essential role in the elongation of omega-3 and omega-6 fatty acids, precursors for long-chain polyunsaturated fatty acids (PUFAs). Mutations of ELOVL5 cause the spino-cerebellar ataxia type 38 (SCA38), a rare autosomal neurological disease characterized by gait abnormality, dysarthria, dysphagia, hyposmia and peripheral neuropathy, conditions well represented by a mouse model with a targeted deletion of this gene (Elovl5–/– mice). However, the expression pattern of this enzyme in neuronal and glial cells of the central nervous system (CNS) is still uninvestigated. This work is aimed at filling this gap of knowledge by taking advantage of an Elovl5-reporter mouse line and immunofluorescence analyses on adult mouse CNS sections and glial cell primary cultures. Notably, Elovl5 appears expressed in a region- and cell type-specific manner. Abundant Elovl5-positive cells were found in the cerebellum, brainstem, and primary and accessory olfactory regions, where mitral cells show the most prominent expression. Hippocampal pyramidal cells of CA2/CA3 where also moderately labeled, while in the rest of the telencephalon Elovl5 expression was high in regions related to motor control. Analysis of primary glial cell cultures revealed Elovl5 expression in oligodendroglial cells at various maturation steps and in microglia, while astrocytes showed a heterogeneous in vivo expression of Elovl5. The elucidation of Elovl5 CNS distribution provides relevant information to understand the physiological functions of this enzyme and its PUFA products, whose unbalance is known to be involved in many pathological conditions.

The role of the cerebellum in Alzheimer’s disease (AD) has been neglected for a long time. Recent studies carried out using transgenic mouse models have demonstrated that amyloid-β (Aβ) is deposited in the cerebellum and affects synaptic transmission and plasticity, sometimes before plaque formation. A wide variability of motor phenotype has been observed in the different murine models of AD, without a consistent correlation with the extent of cerebellar histopathological changes or with cognitive deficits. The loss of noradrenergic drive may contribute to the impairment of cerebellar synaptic function and motor learning observed in these mice. Furthermore, cerebellar neurons, particularly granule cells, have been used as in vitro model of Aβ-induced neuronal damage. An unexpected conclusion is that the cerebellum, for a long time thought to be somehow protected from AD pathology, is actually considered as a region vulnerable to Aβ toxic damage, even at the early stage of the disease, with consequences on motor performance.

Purkinje cell (PC) dysfunction or degeneration is the most frequent finding in animal models with ataxic symptoms. Mutations affecting intrinsic membrane properties can lead to ataxia by altering the firing rate of PCs or their firing pattern. However, the relationship between specific firing alterations and motor symptoms is not yet clear, and in some cases PC dysfunction precedes the onset of ataxic signs. Moreover, a great variety of ionic and synaptic mechanisms can affect PC signaling, resulting in different features of motor dysfunction. Mutations affecting Na+ channels (NaV1.1, NaV1.6, NaVβ4, Fgf14 or Rer1) reduce the firing rate of PCs, mainly via an impairment of the Na+ resurgent current. Mutations that reduce Kv3 currents limit the firing rate frequency range. Mutations of Kv1 channels act mainly on inhibitory interneurons, generating excessive GABAergic signaling onto PCs, resulting in episodic ataxia. Kv4.3 mutations are responsible for a complex syndrome with several neurologic dysfunctions including ataxia. Mutations of either Cav or BK channels have similar consequences, consisting in a disruption of the firing pattern of PCs, with loss of precision, leading to ataxia. Another category of pathogenic mechanisms of ataxia regards alterations of synaptic signals arriving at the PC. At the parallel fiber (PF)-PC synapse, mutations of glutamate delta-2 (GluD2) or its ligand Crbl1 are responsible for the loss of synaptic contacts, abolishment of long-term depression (LTD) and motor deficits. At the same synapse, a correct function of metabotropic glutamate receptor 1 (mGlu1) receptors is necessary to avoid ataxia. Failure of climbing fiber (CF) maturation and establishment of PC mono-innervation occurs in a great number of mutant mice, including mGlu1 and its transduction pathway, GluD2, semaphorins and their receptors. All these models have in common the alteration of PC output signals, due to a variety of mechanisms affecting incoming synaptic signals or the way they are processed by the repertoire of ionic channels responsible for intrinsic membrane properties. Although the PC is a final common pathway of ataxia, the link between specific firing alterations and neurologic symptoms has not yet been systematically studied and the alterations of the cerebellar contribution to motor signals are still unknown.

Spino-Cerebellar-Ataxia type 38 (SCA38) is caused by missense mutations in the very long chain fatty acid elongase 5 gene, ELOVL5. The main clinical findings in this disease are ataxia, hyposmia and cerebellar atrophy. Mice in which Elovl5 has been knocked out represent a model of the loss of function hypothesis of SCA38. In agreement with this hypothesis, Elovl5 knock out mice reproduced the main symptoms of patients, motor deficits at the beam balance test and hyposmia. The cerebellar cortex of Elovl5 knock out mice showed a reduction of thickness of the molecular layer, already detectable at 6 months of age, confirmed at 12 and 18 months. The total perimeter length of the Purkinje cell (PC) layer was also reduced in Elovl5 knock out mice. Since Elovl5 transcripts are expressed by PCs, whose dendrites are a major component of the molecular layer, we hypothesized that an alteration of their dendrites might be responsible for the reduced thickness of this layer. Reconstruction of the dendritic tree of biocytin-filled PCs, followed by Sholl analysis, showed that the distribution of distal dendrites was significantly reduced in Elovl5 knock out mice. Dendritic spine density was conserved. These results suggest that Elovl5 knock out mice recapitulate SCA38 symptoms and that their cerebellar atrophy is due, at least in part, to a reduced extension of PC dendritic arborization.

The parallel fiber-Purkinje cell synapse represents the point of maximal signal divergence in the cerebellar cortex with an estimated number of about 60 billion synaptic contacts in the rat and 100,000 billions in humans. At the same time, the Purkinje cell dendritic tree is a site of remarkable convergence of more than 100,000 parallel fiber synapses. Parallel fibers activity generates fast postsynaptic currents via AMPA receptors, and slower signals, mediated by mGlu1 receptors, resulting in Purkinje cell depolarization accompanied by sharp calcium elevation within dendritic regions. Long-term depression and long-term potentiation have been widely described for the parallel fiber-Purkinje cell synapse and have been proposed as mechanisms for motor learning. The mechanisms of induction for LTP and LTD involve different signaling mechanisms within the presynaptic terminal and/or at the postsynaptic site, promoting enduring modification in the neurotransmitter release and change in responsiveness to the neurotransmitter. The parallel fiber-Purkinje cell synapse is finely modulated by several neurotransmitters, including serotonin, noradrenaline, and acetylcholine. The ability of these neuromodulators to gate LTP and LTD at the parallel fiber-Purkinje cell synapse could, at least in part, explain their effect on cerebellar-dependent learning and memory paradigms. Overall, these findings have important implications for understanding the cerebellar involvement in a series of pathological conditions, ranging from ataxia to autism. For example, parallel fiber-Purkinje cell synapse dysfunctions have been identified in several murine models of spinocerebellar ataxia (SCA) types 1, 3, 5 and 27. In some cases, the defect is specific for the AMPA receptor signaling (SCA27), while in others the mGlu1 pathway is affected (SCA1, 3, 5). Interestingly, the parallel fiber-Purkinje cell synapse has been shown to be hyper-functional in a mutant mouse model of autism spectrum disorder, with a selective deletion of Pten in Purkinje cells. However, the full range of methodological approaches, that allowed the discovery of the physiological principles of PF-PC synapse function, has not yet been completely exploited to investigate the pathophysiological mechanisms of diseases involving the cerebellum. We, therefore, propose to extend the spectrum of experimental investigations to tackle this problem.