Expression of Calcium-binding Proteins in Purkinje Cells of Spinocerebellar Ataxia-1 (SCA-1) Transgenic Mice.

Calcium ([Ca.sup.2+]) is an ubiquitous and important intracellular messenger for the actions of a number of hormones, growth factors and neurotransmitters. Intracellular concentrations of free [Ca.sup.2+] are closely regulated and are nominally 100nM. However, in excitable tissues, depolarization leads to the opening of [Ca.sup.2+]-channels with the result that free [Ca.sup.2+] levels rise 10 to 100 fold. Such events are short-lived under normal physiological conditions since resting levels are restored by [Ca.sup.2+]-dependent ATPases. Cytosolic regulation of [Ca.sup.2+] levels is also achieved by special [Ca.sup.2+] binding proteins (CBPs) such as calbindin-D 28k (CAB) and parvalbumin (PV). In the central nervous system PV exists in a subpopulation of [Gamma]-aminobutyric acid (GABA) ergic neurons, which are considered electrically and metabolically more active than companion neurons. In cerebellum, PV is highly localized in Purkinje cells and their dendritic arborizations, and in the interneurons, in the molecular layer. In Purkinje cells PV may act as an intracellular [Ca.sup.2+]-buffer. Calbindin, a 28kDa protein is also present widely in dendrites, soma and axons of Purkinje cells. However, the other cerebellar neurons lack CaB. In addition to [Ca.sup.2+] buffering, PV has been shown to form protein-nucleotide complexes under in vivo conditions, and it has also been proposed that PV plays an important role in translational regulation in cultured cells transfected with PV cDNA, PV expression influences cell division, shape and motility. However, it is not known if PV has similar function in Purkinje cells of cerebellum.

Disturbances in [Ca.sup.2+] regulation can have potentially lethal effects. Sustained increase in intracellular [Ca.sup.2+] levels such as those observed after excitatory amino acid treatment or due to reduced transport activity have been correlated with increased protein and lipid degradation events which ultimately lead to cell death. A growing body of literature suggests that neurons reach in specific CBPs may be relatively resistant to degeneration in a variety of acute and chronic disorders.

Alterations in the immunoreactivity (IR) to CaB and PV have been observed in spinocerebellar ataxias. Our recent studies have demonstrated that there is a significant reduction in PV IR in surviving Purkinje cells of patients with SCA-1 (Vig et al. 1996; Neurology 47:249-253). Furthermore, the recent report that SCA-6 is caused by mutations in the human a 1A [Ca.sup.2+]-channel also support the hypothesis that Purkinje cell degeneration may be associated with perturbed [Ca.sup.2+] homeostasis. SCA-1 is a progressive neurological disorder characterized by the degeneration of cerebellar Purkinje cells and the selective loss of neurons within the brainstem and spinocerebellar tract. The SCA-1 gene on the short arm of chromosome 6 has a trinucleotide (CAG) repeat that is expanded and unstable in affected individuals. SCA-1, like other CAG repeat expansion disorders, is believed to be caused by a dominant gain of function by the mutant protein containing the expanded polyglutamine tract. However, the precise role of the expanded polyglutamine tract in the pathogenesis of SCA-1 and other CAG repeat expansion disorders is unknown. Perutz and co-workers have shown by molecular modeling that polyglutamine stretches form B-strands. These investigators have speculated that pathologic effects might arise if extended repeats cause proteins to acquire high affinity for each other and/or for other nonspecific proteins.

Dr. Orr and his group reported the establishment of transgenic mice that express human SCA-1 cDNA transgenes with either a normal or an expanded CAG tract within the cerebellar Purkinje cells. While transgenic animals expressing the unexpanded SCA-1 allele had morphologically and functionally normal Purkinje cells, transgenic animals expressing the expanded SCA-1 allele developed ataxia and Purkinje cell degeneration. We used these transgenic mice as an animal model for human SCA-l, and to determine if the alterations in the levels of CBPs, PV and CaB precede the onset of ataxia in SCA-1 transgenic mice.

In the present study we determined PV and CaB expression (by immunohistochemistry and immunoblot analysis) in Purkinje cells of transgenic mice (TM) expressing the human SCA-1 gene with expanded (line B05) and normal (line A02) CAG tract, as well as in age-matched nontransgenic mice (nTM). Heterozygotes in the B05 line develop progressive ataxia beginning around 12 weeks of age. A02 animals are phenotypically indistinguishable from wildtype (non-transgenic) animals. In the cerebella of 8, 9, and 12 wks old TM-B05 there was a progressive decrease in PV immunoreactivity (IR) in Purkinje cells as compared to the nTM and TM-A02. PV immunostaining in interneurons was well preserved in all the groups. A progressive decrease was also observed in CaB-IR in Purkinje cells of 8, 9 and 12 wks old TM-B05. Cerebellar Purkinje cells of 6 wks old TM-B05 which exhibit no ataxia and even lack demonstrable Purkinje cell loss also revealed reduction in PV IR, This change was matched by a significant decrease in the amount of cerebellar PV in 6 wks old TM-B05 as determined by western blot analysis. CaB immunohistochemistry did not detect any marked changes in CaB IR within Purkinje cells at 4 wks. However, at 6 wks, immunostaining and immunoblot analysis revealed a significant decrease in CaB in TM-B05 compared to controls. The other neuronal markers like synaptophysin, GAP-43 and Purkinje cell specific protein kinase C-[Gamma] were not altered in 6 wks old TM-B05 cerebellum. These observations suggest that the changes in PV and CaB are highly specific and are not due to the changes in the architecture of TM-B05 cerebellum.

These data are in agreement with our earlier findings where we demonstrated a significant decrease in PV IR in surviving Purkinje cells of SCA-1 patients. In the areas which are not involved in SCA-1 like hippocampus, temporal and cerebral cortex, PV-positive neurons were seen, similar to those in controls. Thus, the biochemical alterations associated with Purkinje cell degeneration in human SCA-1 can also be seen in the transgenic mouse Purkinje cells, suggesting a similar pathogenic pathway.

CB and PV have been widely studied in a number of neurodegenerative disease. Previous studies have also indicated that brain areas that are particularly affected in Huntington's disease (corpus striatum), and Alzheimer's disease (nucleus basalis) exhibited a specific reduction of CaB gene expression. Motor neurons that normally lack PV may be preferentially lost in amyotrophic lateral sclerosis (ALS). A recent study points out that expression of CaB in motor neuron hybrid cells after retro viral infection with CaB CDNA prevents ALS IgG-mediated cytotoxicity. Our studies demonstrate a significant decrease in IR to PV and CaB when there is no clinical ataxia and demonstrable Purkinje cell loss in TM-B05. These changes may reflect an important step in the cascade of events triggered by the SCA-1 mutation that finally results in the loss of these cells. This hypothesis is to some extent supported by the occurrence of ataxia and altered dendrite [Ca.sup.2+] signaling in cerebellar Purkinje cell of knock out mice lacking CaB gene. We speculate that double mutants lacking both CaB and PV may show more severe ataxia and Purkinje cell dysfunction and pathology as seen in TM-B05 cerebellum. In Purkinje cells, intracellular [Ca.sup.2+] signaling con tributes to many cellular functions including regulating synaptic transmission by parallel and climbing fibers, control of excitability, and induction of long-term changes in efficacy of excitatory and inhibitory synapses. Thus, alterations in [Ca.sup.2+] homeostasis due to decreased CBPs would impair some of these vital functions, resulting in Purkinje cell degeneration.

The mechanism by which CAG repeat expansions in SCA-1 cause neurodegeneration in humans or in transgenic mice is not understood. The widely accepted hypothesis is that the Ataxin-1 protein (SCA-1 gene product) may self-interact or interact with other proteins and modify their function. Ataxin-1 has been shown to form complexes with glycolytic enzyme 3phosphate dehydrogenase (GAPDH). An impairment in energy metabolism can result in neuronal degeneration. GAPDH is widely expressed, whereas neuronal loss in SCA-1 is highly selective; therefore, the alternate hypothesis may be interaction of Ataxin-1 with other cell specific proteins. At this time, we do not know if the alterations in the levels of CaB and PV are due to their interactions with ataxin-1 or with other related proteins which are directly affected by ataxin-1. However, our preliminary observations suggest that the CaB mRNA levels are also reduced in 6 wks old TM-B05 suggesting that the alterations in CaB expression maybe at the transcriptional level. How these changes are related to SCA-1 mutation is currently being explored by using cultured Purkinje cells.

The results of this study have been recently published in Neurology, 1998;50:106-113. This research was supported, in part, from a grant from National Ataxia Foundation.

Parminder. J.S. Vig, Ph,D., Assistant Professor, Department of Neurology University of Mississippi Medical Center Jackson, Mississippi