Spinocerebellar Ataxia Type 12 (SCA 12): Clinical Features and Pathogenetic Mechanisms

Spinocerebellar Ataxia 12 (SCA12) is a rare disease that was first identified in a family in the United States. Patients suffered from classical spinocerebellar ataxia symptoms with an age of disease onset ranging from 8-55 years. A trinucleotide (CAG) repeat expansion was confirmed in all the affected individuals. The CAG expansion mapped to the 5’ untranslated region (UTR) of the PPP2R2B gene. This gene encodes a regulatory subunit, B of the heterotrimeric protein phosphatase 2A (PP2A). The function of this particular PP2A complex is not well understood, and the underlying molecular mechanism of SCA12 remains unclear. Additional pedigrees have been identified throughout the world but SCA12 remains a rare disease. In this chapter we will discuss the clinical manifestation of the disease and the known functions of the PP2A regulator B.


Molecular genetics and Incidence
SCA12 is defined as an autosomal dominant cerebellar ataxia (ADCA) of otherwise unknown cause concurrent with a CAG repeat expansion within chromosome 5q31-33 upstream of the PPP2R2B gene (Holmes et. al., 1999). The PPP2R2B gene product, termed B , is a neuron specific regulatory subunit of the heterotrimeric PP2A (Strack et. al., 1998). PP2A has been shown to play an essential role in many cellular functions (Janssens & Goris, 2001). The CAG repeat expansion associated with SCA12 was first identified through an unbiased repeat expansion detection study and found to occur within the noncoding region of the PPP2R2B gene (Holmes et. al., 1999). The nonpathalogical range of allele expansion is quite large (7-45 repeats) and is highly dependent on ethnic background (Fujigasaki et. al., 2001;Holmes et. al., 1999). The lower extreme of the range of pathological allele expansion has been established as 51 repeats. As is common to all ADCA disorders, inheritance of SCA12 follows an autosomal dominant pattern wherein a CAG repeat expansion of pathological length in just one allele is sufficient to induce the SCA12 disease state. Unlike other neurodegenerative diseases associated with a CAG repeat expansion, such as Huntingon disease, the number of CAG repeats associated with SCA12 does not correlate with the age of disease onset (Srivastava et. al., 2001). In addition, nondirectional vertical instability in the length of the expanded allele has been observed, however its clinical significance is unknown (Srivastava et. al., 2001). One individual has been identified with www.intechopen.com pathological repeat expansions in both alleles; however, due to the young age of this patient, it is unclear what effect homozygosity will have on the disease phenotype (Bahl et. al., 2005).

Clinical features
At present, SCA12 confirmed by genetic testing remains a very rare illness. However, as genetic testing, including whole genome sequencing, becomes common practice, the true incidence of SCA12 may prove to be much higher among previously categorized ADCA patients of unknown cause. Indeed, among a cohort of ADCA patients in India the incidence of SCA12 has proven to be much higher than in other geographical locales (Bahl et. al., 2005;Srivastava et. al., 2001). Given this observation, those who encounter ADCA patients should be aware of SCA12 and develop an index of suspicion informed by careful history taking, detailed neurological examination and deliberate laboratory testing.
As SCA12 has only been recognized as a distinct pathology for the last decade and, at present, only a very few patients have been described in the primary literature, an appreciation for the natural history of the disease is still evolving. By careful consideration of those cases that have been well characterized in the American, Indian, Italian and Chinese cohorts, a clinical picture of the SCA12 patient will be developed here. The descriptions provided here are intended to inform the clinician who encounters ADCA patients of unknown cause and to guide clinical decision-making.

Patient reported history of illness
Early in the course of the disease the prototypical SCA12 patient will present with postural and action tremor of the upper limbs. Age of onset of this tremor is highly variable with a range between 8 and 55 years, but seems to cluster primarily between the third and fifth decade of life (Brussino et. al., 2010;Fujigasaki et. al., 2001;Holmes et. al., 1999;Srivastava et. al., 2001). The first manifestations of the action tremor of the upper limbs have been described by patients as difficulty with activities requiring fine motor coordination, such as writing, as well as difficulties with activities requiring gross motor coordination such as attempting to hold and purposefully manipulate objects like a cup (Fujigasaki et. al., 2001;. Observers describe the tremor as slowly progressive in nature with an increase in amplitude and involvement of the head and neck have been observed over the course of a decade . The action tremor of the upper limbs as the harbinger of the disease is unique to SCA12 and differentiates SCA12 from other ADCA disorders (Schols et. al., 2004;Teive, 2009). This tremor is not, however, universal among SCA12 patients, and its absence does not rule out SCA12 (Srivastava et. al., 2001;Wang, J et. al., 2011). Presentation of the upper limb action tremor is very similar to that of essential tremor and has previously been misdiagnosed as such early in the SCA12 course . Differentiating the SCA12 associated upper limb action tremor from isolated essential tremor requires an appreciation of the complete constellation of SCA12 associated symptoms as well as a family history consistent with ADCA.

Neurological examination
The time elapsed since disease onset has been reported to directly correlate with the number of neurological abnormalities . The examination of an SCA12 patient should therefore be informed by the patient reported history. To fully characterize the constellation of symptoms associated with SAC12 early in the course of the disease, care should be taken to elicit mild neurological abnormalities that may otherwise be subclinical in nature. Characterizing the gross neurological deficits present late in the course of the disease can serve to chart disease progression.

Motor skills deficits
As indicated above, the action tremor associated with SCA12 is one of the earliest hallmarks of the disease. Action tremor features include postural and kinetic properties, as well as a low frequency (3 Hz) , and are similar to a tremor subset associated with cerebellar lesion termed "cerebellar postural tremor" (Hallett, 1991). As such, the postural features of the tremor can be elicited in the clinical setting by asking the patient to maintain their arms in an outstretched position and observing for limb tremor. The kinetic features of the tremor can be assessed by having the patient engage in a goal-directed movement of the upper limbs, such as finger-to-nose testing. Tremor should disappear www.intechopen.com completely while the upper limbs are at rest and not maintaining position against the force of gravity.
Loss of motor coordination due to cerebellar dysfunction associated with SCA12 manifests when the patient engages in a number of activities. During finger-to-nose testing, rather than smooth, rapid, accurate movements, the SCA12 patient will display slow, hesitant, inaccurate movements consistent with upper limb dysmetria. Further, the SCA12 patient has been reported to be unable to engage in rapid alternating movements (dysdiadochokinesia) such as alternating between turning the palms or the back of the hand face up . Motor deficits also disrupt speech and can result in dysarthria Srivastava et. al., 2001).
Parkinsonain features have also been described in SCA12 patients from the American Cohort. These manifest as paucity of spontaneous movements, mild bradykinesia, upper limb rigidity and postural anteroflexion .
A great deal of heterogeneity has been observed in the symptoms of SCA12 patients from different ethnic backgrounds. Unique to the Indian cohort, facial myokymia has also been described in a small number of SCA12 patients (Srivastava et. al., 2001). Although the proband of the Chinese cohort developed generalized ataxia during the third decade of life, action tremor has not been observed (Wang, J et. al., 2011).

Gait abnormalities
The ataxic gait of the SCA12 patient has been described as being very similar to that observed in other diseases with cerebellar dysfunction. The SCA12 patient maintains stability by adopting a broad based stance. Parkinsonian features have also manifest in the gait among individuals of the American Cohort . Initiation of movement is delayed.
Steps have been described as hesitant, small and slow. When turning, the SCA12 patient has been described as engaging in an "en bloc" approach. A mild ataxic phenotype can be exaggerated by having the patient maintain a tandem gait, wherein the patient walks in a straight line with the heel of the front foot touching the toes of the back foot at each step.

Cranial nerve assessment
With the exception of oculomotor nerve (CNIII) abnormalities, the cranial nerves are largely intact and function without deficit in the SCA12 patient. Horizontal nystagmus has been described and may represent an early manifestation of the disease (Fujigasaki et. al., 2001;Holmes et. al., 2003;Srivastava et. al., 2001). In addition slow saccades and broken pursuit have been described in SCA12 patients from the Indian cohort (Fujigasaki et. al., 2001;Srivastava et. al., 2001).

Assessment of reflexes
Diffuse hyperreflexia has been described for SCA12 patients from the American, Indian and Italian cohorts (Brussino et. al., 2010;Fujigasaki et. al., 2001;Srivastava et. al., 2001). A return of primitive reflexes in the otherwise mature SCA12 patient has also been described. These reflexes include an extensor plantar response (positive Babinski sign), grasp reflex, rooting reflex and glabellar blink reflex (Myerson sign).

Mental Status
Psychiatric disorders have been reported to occur concurrently with SCA12. Anxiety and depression have been reported in members of the American cohort, but not the Indian or Italian cohorts (Brussino et. al., 2010;Srivastava et. al., 2001). Whether these disorders result as a direct consequence of the SCA12 disease process or represent an individual response to the presence of the disease is unclear. Paranoid delusions have also been reported in one SCA12 patient . A decline in cognition has been described in SCA12 patients two to three decades after initial onset of the disease (Fujigasaki et. al., 2001;.

Neuroimaging studies
Computerized tomography (CT) and magnetic resonance imaging (MRI) studies of symptomatic SCA12 patients reveal that mild to moderate cerebellar and cortical atrophy is a near universal finding of the disease (Brussino et. al., 2010;Fujigasaki et. al., 2001;Srivastava et. al., 2001;Wang, J et. al., 2011). An example of these findings from imaging studies performed on members of the American cohort of SCA12 patients is shown (Figure 1). The cerebellar vermis appears to be more vulnerable to atropy than the cerebellar hemispheres . Atrophy of subcortical structures has not been described. Additional characterization by single-proton emission computed tomography (SPECT) revealed metabolic deficiencies in atrophic cortical areas; however, the value of this test is uncertain in the symptomatic patient (Fujigasaki et. al., 2001). Proton magnetic resonance spectroscopy has been used to demonstrate neurometabolic and microstructural changes in the SCA12 patient (Brussino et. al., 2010), and this technique represents a noninvasive method that may longitudinally describe the asymptomatic SCA12 patient.

Genetic testing
Genetic testing for the presence of CAG repeat expansion is available. The reader is directed to the GeneTests Laboratory Directory available online (http://www.ncbi.nlm. nih.gov/sites/GeneTests/lab) for a list of available testing centers. The small sample size of affected individuals currently identified has left the question of penetrance of the disease open. Therefore, a great deal of care should be exercised when interpreting the results of a genetic test from an asymptomatic patient.

Medical management
Currently, management of SCA12 is limited to providing symptomatic relief for the action tremor. Treatment of the SCA12 action tremor is very similar to that provided for essential tremor. A reduction in tremor amplitude has been achieved with beta-blockers and barbiturates . When appropriate, pharmacological relief for symptoms associated with the disease such as depression and anxiety should be offered to the SCA12 patient.

PP2A and B regulatory subunit
Protein phosphorylation is the most common posttranslational modification of proteins, and it plays a role in nearly every cellular function. The addition of phosphate is mediated through a large group (>500) of enzymes called kinases and requires ATP as a substrate. The reverse reaction is mediated by a smaller number of protein phosphatases in which, in most cases, specificity is provided through the formation of multimeric protein complexes. One of the most abundant protein phosphatase is PP2A, which is an essential, ubiquitously expressed phosphatase that targets phospho-serine and phospho-threonine. PP2A exists as a heterotrimer composed of one member of four diverse families of regulatory subunits (B), a scaffolding subunit (A) and a catalytic subunit (C) (Figure 2). Humans express 4 families of regulatory subunits termed B, B', B'', and B''', which determine both cellular localization and substrate specificity (Slupe et. al., 2011). The B family, also known B55, consists of 4 distinct genes () that encode proteins containing a highly conserved core WD40 domain, which has propeller like structure, with over 90% amino acid identity among the family members ( Figure 2B). The B regulatory subunit is encoded by the PPP2R2B gene, which has several splice-variants that are expressed exclusively in neuronal tissue.

Gene structure and expression
The exon arrangement of the PPP2R2B gene is highly conserved among mammals and spread over more than 500,000 base pairs (Dagda et. al., 2003;Schmidt et. al., 2002). Exon 1.1 and 1.2 are alternatively expressed first exons containing the ATG start site for the splice variants B1 and B2, respectively. These first exons, which contain the unique aminotermini, are spliced to common exons 2-9 that encode the WD40 domain found in all the B family of regulatory subunits (Figure 3) (Dagda et. al., 2003). At the mRNA level, B1 and B2 are expressed prominently in brain tissue, and B1 can also be found in the testis (Dagda et. al., 2003). At the protein level, western blot analysis indicates that the B1 is exclusively expressed in brain tissue and not in the testis, despite the high mRNA expression in that tissue. Closer analysis of specific brain regions has shown high levels of the B1 protein throughout the brain (Strack et. al., 1998).

Transcriptional regulation
The CAG trinucleotide repeat expansion associated with the SCA12 disease is situated just upstream of the transcriptional start site of the B1 specific exon 1.1. A recent study identified the apparent transcriptional regulators for basal expression of the B1 promoter and the effect of the CAG repeat on basal expression (Lin et. al., 2010). Luciferase assays using deletions of the B1 promoter and chromatin immunoprecipitation assays reveal that Fig. 3. Schematic representation of PPP2R2B gene structure, splice variant transcripts and proteins. The gene structure shows the CAG repeat expansion location, the B1 (exon 1.1; red) and B2 N-terminal coding sequences (exon 1.2; green). Transcripts and proteins indicate the B1 (red) and B2 (green) splice variant specific transcripts and encoded proteins. Modified from Dagda et. al. (2003). CREB1, SP1 and TRAP4 bind to and regulate the B1 promoter. Higher luciferase activity is seen in neuronal cell lines and correlates well with the known B1 neuronal expression. Interestingly, increasing the size of the CAG repeat in the B 1 promoter increased the promoter activity two-fold. The increased activity is specific to the expansion of the CAG and not a result of changing the spacing of promoter since no change is seen in an AT expansion control (Lin et. al., 2010). A normal length CAG repeat does appear to be important for basal promoter activity since decreasing the number of CAG repeats reduced the promoter activity (Chen et. al., 2009). Independent studies conducted in Japan and Taiwan found that patients suffering from Alzheimer's disease had an increased likelihood of having a reduced number of CAG trinucleotide repeats compared to healthy control subjects (Chen et. al., 2009;Kimura et. al., 2011). Overall, these studies have identified important aspects of the PPP2R2B transcriptional regulation and help to discriminate between the role of the CAG repeat in providing basal transcriptional activation and the pathological effects of increasing or decreasing the trinucleotide repeat number.
A recently identified Japanese autosomal dominant cerebellar ataxia raises more uncertainty about the role of PPP2R2B gene in SCA12. The disease locus for this new ataxia included the PPP2R2B gene but contained no CAG expansion (Sato et. al., 2010). Additionally, all exons and intron/exon borders were sequenced for the entire PPP2R2B gene, including the both first exons (1.1 and 1.2), and no mutations were discovered. Several neuronally expressed genes are within the identified locus and may contain the genetic insult resulting in the ataxia (Sato et al 2010). This does raise the possibility that some of the effects of the CAG expansion in the PPP2R2B gene may be mediated through dysregulation of other nearby genes and not just changes in Bgene expression.

PPP2R2B regulation and cancer
Another important form of regulation of B1 occurs in colorectal cancer (CRC) wherein developed cell lines show a decrease or complete absence of B1 expression (Tan et. al., 2010). Furthermore, gene array comparisons of matched patient-derived mucosa controls and CRC tumors indicate a significant decrease in B1 expression in 90% of the tumors. The loss of B1 expression is mediated through hypermethylation of a CpG island that occurs in the B1 promoter. Aberrant methylation of the PPP2R2B gene also appears to be important in breast cancer, as seen in recent reports (Dejeux et. al., 2010;Muggerud et. al., 2010). Finally, an intronic SNP of the PPP2R2B gene, with unknown functional consequence, is correlated with improved prognosis in a breast cancer cohort (Vazquez et. al., 2011). These studies clearly indicate that regulation of the PPP2R2B gene is important in multiple cancers and may provide additional insight into the function of the PPP2R2B gene.

Protein function
The B1 and B2 splice variants encode proteins that share a common WD40 repeat domain that mediates the recruitment of the A and C subunits of PP2A to make a functional trimeric protein phosphatase. The B1 and B2 proteins differ only in the first 21 and 24 amino acids, respectively, but this leads to a dramatic difference in the protein distribution within the cell.

B1 protein function
B1 has a cytoplasmic distribution and overexpression in cultured primary neurons does not change the morphology, survival or sensitivity to toxic treatments (Figure 4) (Dagda et. al., 2008). Overexpression of B1 in a neuroblastoma cell line does result in increased autophagy (Cheng et. al., 2009). In CRC the loss of B1 following methylation of the CpG island leads to aberrant phosphorylation of several proteins, including the oncogene c-myc. Reexpression of B1 in a colorectal cell line decreases xenograft growth (Tan et. al., 2010). This represents the first described pathway regulated specifically by a B1 containing PP2A trimer. Since some of the proteins regulated by B1 in CRC are also expressed in neuronal tissues, it may be of interest to examine whether the B1-mediated changes in phosphorylation also play a role in SCA12.

B2 protein function
The B2 N-terminus encodes a mitochondrial targeting sequence that results in recruitment of the trimeric PP2A enzyme to the outer mitochondrial membrane (OMM) (Dagda et. al., 2003). In primary hippocampal neurons, PP2A-mediated phosphatase activity at the OMM, through recruitment by B2, results in mitochondrial fragmentation and increased basal death and sensitivity to neurotoxic insults ( Figure 4) (Dagda et. al., 2005;Dagda et. al., 2008). Expression of B2 mutants, that either do not target to the OMM or cannot recruit the A and C subunits, prevents the mitochondrial fragmentation and increased neuronal death ( Figure  4) (Dagda et. al., 2008). Epitasis experiments indicate that the PP2A/B2-mediated mitochondrial fragmentation precedes and is obligatory to the increased neuronal cell death (Dagda et. al., 2008). An additional study, utilizing neuroblastoma cells, confirmed the increased sensitivity of cells expressing B2 but implicated an increase in autophagy as the culprit in the increased cell death (Cheng et. al., 2009).
Mitochondrial dysfunction is a hallmark of several neurodegenerative diseases, including Alzheimer disease. It can therefore be postulated that the CAG trinucleotide repeat expansion, which is known to increase B1 promoter activity, amplifies both B1 and B2 expression. The B2 upregulation may lead to increased mitochondrial fragmentation and increasing mitochondrial dysfunction in SCA12. Indeed, several other ataxias involve mitochondrial dysfunction. In patients suffering from SCA7, both liver and skeletal muscle biopsies show abnormal mitochondria (Han et. al., 2010). Heterozygous knockout mice for AFG3L2, a mitochondrial-targeted AAA-protease, develop abnormal mitochondria with decreased function and are a model of SCA28 (Maltecca et. al., 2009). Finally, in clinical trials pharmacological treatments with idebenone, an antioxidant thought to counteract mitochondrial dysfunction, have shown some promise in treatment of the genetic neurological disorder Friedreich ataxia (Marmolino, 2011). These examples highlight some of the ataxias associated with mitochondrial dysfunction and exemplify why mitochondrial dysfunction could be an important aspect of SCA12.

Animal models of SCA12
While characterization of the PPP2R2B gene products has suggested possible pathogenic mechanisms, animal models of SCA12 are urgently needed to test the predictions of the in www.intechopen.com Fig. 4. Mitochondrial targeting of PP2A/B2 is neurotoxic. Hippocampal neurons were transfected with the indicated GFP fusion proteins (om, outer mitochondrial; WT, wildtype) and scored for apoptotic nuclei. B2 mutants that block mitochondrial localization (R6A) or AC dimer recruitment (RR168EE) also block apoptosis induction. Modified from Dagda et. al. (2008). vitro studies discussed above. A recently developed fly model of SCA12 does display some neuropathies that may be homologous to the human disease (Wang, YC et. al., 2011). In this model, Drosophila overexpresses the human B2 or tws, the fly homolog of B, which results in a dramatic increase in neuronal apoptosis and, for the highest level of tws, a decrease in fly life span. Overexpression of tws results in mitochondrial fragmentation and dysfunction, observed as an increase in reactive oxygen species (ROS) production. Expression of superoxide dismutase 2 or antioxidants treatments reduces ROS production and attenuates the effects of tws overexpression. How the neuropathies and their reversal by pharmacological treatments seen in the fly SCA12 model relate to the human disease remains to be seen.

Conclusion
The CAG trinucleotide repeat expansion that occurs in the PPP2R2B gene is now well established as the cause of the autosomal dominant SCA12. This is a rare disease that shows a classical ataxia phenotype. The CAG repeat occurs in the promoter of a neuronally expressed protein, B1, and expansion of the CAG results in increased B1 promoter activity. Aberrant expression of B1 also correlates with several cancers. Expression of another neuronal splice variant of PPP2R2B, B2, increases neuronal death, but its role in SCA12 remains unknown. Despite the identified PPP2R2B gene functions, the underlying molecular basis of the SCA12 disease is not known. Animal models are needed to address the complexity of SCA12 and develop potential therapeutic treatments. The fly model of SCA12 does show mitochondrial dysfunction and recapitulates some neuron specific cell death (Wang, YC et. al., 2011); however, the development of a mammalian model system will likely be required to understand the molecular basis of SCA12 pathogenesis.

Acknowledgment
The authors would like to thank Drs. Nicole Worden and Melissa Bose for critically evaluating this manuscript.