Spinocerebellar ataxia type 1 (
SCA1) is a dominantly inherited disorder characterized by progressive loss of coordination, motor impairment and the degeneration of cerebellar Purkinje cells, spinocerebellar tracts and brainstem nuclei. Many dominantly inherited
neurodegenerative diseases share the mutational basis of
SCA1: the expansion of a translated CAG repeat coding for
glutamine. Mice lacking
ataxin-1 display learning deficits and altered hippocampal synaptic plasticity but none of the abnormalities seen in human
SCA1; mice expressing
ataxin-1 with an expanded CAG tract (82
glutamine residues), however, develop Purkinje cell pathology and
ataxia. These results suggest that mutant
ataxin-1 gains a novel function that leads to neuronal degeneration. This novel function might involve aberrant interaction(s) with cell-specific
protein(s), which in turn might explain the selective neuronal pathology. Mutant
ataxin-1 interacts preferentially with a
leucine-rich acidic
nuclear protein that is abundantly expressed in cerebellar Purkinje cells and other brain regions affected in
SCA1. Immunolocalization studies in affected neurons of patients and
SCA1 transgenic mice showed that mutant
ataxin-1 localizes to a single,
ubiquitin-positive nuclear inclusion (NI) that alters the distribution of the
proteasome and certain chaperones. Further analysis of NIs in transfected HeLa cells established that the
proteasome and chaperone
proteins co-localize with
ataxin-1 aggregates. Moreover, overexpression of the chaperone HDJ-2/HSDJ in HeLa cells decreased
ataxin-1 aggregation, suggesting that
protein misfolding might underlie NI formation. To assess the importance of the nuclear localization of
ataxin-1 and its role in
SCA1 pathogenesis, two lines of transgenic mice were generated. In the first line, the
nuclear localization signal was mutated so that full-length mutant
ataxin-1 would remain in the cytoplasm; mice from this line did not develop any
ataxia or pathology. This suggests that mutant
ataxin-1 is pathogenic only in the nucleus. To assess the role of the aggregates, transgenic mice were generated with mutant
ataxin-1 without the self-association domain (SAD) essential for aggregate formation. These mice developed
ataxia and Purkinje cell abnormalities similar to those seen in
SCA1 transgenic mice carrying full-length mutant
ataxin-1, but lacked NIs. The nuclear milieu is thus a critical factor in
SCA1 pathogenesis, but large NIs are not needed to initiate pathogenesis. They might instead be downstream of the primary pathogenic steps. Given the accumulated evidence, we propose the following model for
SCA1 pathogenesis: expansion of the
polyglutamine tract alters the conformation of
ataxin-1, causing it to misfold. This in turn leads to aberrant
protein interactions. Cell specificity is determined by the cell-specific
proteins interacting with
ataxin-1. Submicroscopic
protein aggregation might occur because of
protein misfolding, and those aggregates become detectable as NIs as the disease advances.
Proteasome redistribution to the NI might contribute to
disease progression by disturbing proteolysis and subsequent vital cellular functions.