Keywords

16.1 Spinocerebellar Ataxia Type 3 (SCA3): Overview

Spinocerebellar ataxia type 3 (SCA3)/Machado-Joseph disease (MJD) is the most frequent of the autosomal dominant ataxias, displaying a worldwide prevalence of 1.5 cases per 100,000 inhabitants [1]. In the Portuguese Azores, to where the first descriptions of SCA3 trace (revised in [2]), an homogeneous cohort of patients has been described [3,4,5], whose importance for the understanding of several aspects of this disease has been reinforced by Raposo and collaborators [6]. A recent longitudinal epidemiological study of SCA3 in the Azores Islands reported an overall prevalence of 1 in each 2544 for the total archipelago, and of 1 in each 158 for the island of Flores [7].

SCA3 involves predominantly the cerebellar, pyramidal, extrapyramidal, motor neuron and oculomotor systems. Although the average age for the appearance of first symptoms is around 40 years, extremes from 4 to 70 years have been reported (revised in [8]). As a clinical entity, SCA3 is recognized as highly pleomorphic; pleomorphism is reflected on the marked variation of age at onset, the heterogeneity of the clinical features displayed by the different patients, as well as on the existence of particular/atypical clinical presentations (see, for example, [9]). Clinical heterogeneity is further evidenced by inter-patient differences in the rate of disease progression and magnitude of the associated disability.

SCA3 causative gene displays almost complete penetrance (~98%, in accordance with [10]), following an age-dependent pattern. The probability of being a mutation carrier and consequently a posteriori risk diminishes with age of asymptomatic at risk individuals, reaching approximately zero in 70 years old subjects [11].

ATXN3, SCA3 causative gene, was described as containing a polymorphic expanded and unstable CAG tract at exon 10 [12], consensually ranging from 14 to 42 repeats in normal chromosomes and from 52 to 91 repeats in chromosomes harboring its mutated form [13, 14]. ATXN3 encodes for the ubiquously expressed ataxin-3 , a cysteine protease whose main native role is that of a deubiquitinating enzyme (DUB) in the ubiquitin -proteasome pathway. Expansion of the polyglutamine (polyQ) tract above the pathological threshold initiates a cascade of pathogenic events that have been extensively studied (revised in [15, 16]). Aggregation of mutant ataxin-3 , proteolytic cleavage , transcription dysregulation, mitochondrial dysfunction , axonal transport impairment, dysregulation of intracellular Ca2+ homeostasis and impairment of protein degradation are considered the main mechanisms implicated in SCA3 pathogenesis [16].

Despite the progresses made in the understanding of the underlying molecular mechanisms, disease-modifying therapies for SCA3 are still lacking, and only symptomatic approaches are currently available. So far interventional studies undertaken have failed to demonstrate an impact of compounds tested on disease progression. Therefore, despite corresponding to the SCA for which the highest number of interventional studies has been conducted, the large majority has been clearly power-limited; the lack of randomization, the absence of a placebo group and the short duration of the trials are some of the limitations pinpointed (revised in [17]). Based on their own previous experience with an interventional study, Saute and collaborators [18] have provided insights on potential pitfalls in SCA3 clinical trials , drawing attention to the importance of improving their efficacy [18].

Empowering clinical trials emerges as a major goal of current SCA3 research, given the previously described scenario, in which anticipated trials should be lengthy and require large numbers of patients. Part of such empowerment is related with the increase in sensitivity of the instruments used to measure disease progression and detect subtle therapeutic benefits; it is therefore expected that molecular biomarkers (MBs) will have the potential to provide a crucial contribution to the quality of SCA3 interventional studies.

16.2 Molecular Biomarkers of SCA3

A biomarker is understood as “a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacological responses to a therapeutic intervention” [19]. Biomarkers are frequently subdivided into “trait biomarkers ” if they provide indications as to the presence/absence of disease and “state biomarkers ” if they can inform about the severity of a certain pathology [20]. Biomarkers are expected to be useful in diagnostic, staging and prognosis of disease, as well as to aid in the prediction and monitoring of intervention [19].

As it would be expectable, clinical biomarkers correspond to the most investigated markers so far in SCA3. Several standardized clinical tests, including rating scales, have been developed to measure different aspects of the SCA3 phenotype. Such measures have been validated in studies of the natural history of the disease, targeting different cohorts of patients. Advantages pinpointed for the widespread use of clinical markers are the relatively low requirement in time and their reduced cost, as well as the fact that they can be obtained without the need for any sophisticated equipment [20]. Regardless of the progresses made with the development of more objective clinical scales, it is assumed that clinical measures are to a certain extent subjective (the complexity of the SCA3 phenotype further aggravates this limitation), insensitive to subtle changes in small periods of time, as well as potentially subjected to observational bias [20]. Also, current clinical measures are limited as to their usefulness in the preataxic stage of the disease, a phase that should be extremely important on what concerns the development of therapeutics .

Neuroimaging information holds the promise of sensitivity and informativity; neuroimaging indicators, such as specific volumetric alterations, are already being used as primary endpoint in clinical trials of neurodegenerative diseases similar to SCA3 (www.clinicaltrials.gov—clinical trial NCT02336633). Notwithstanding, it has to be acknowledged that although neuroimaging alterations in SCA3 are well documented in cross-sectional studies, longitudinal data is still scarce, a limitation that will be necessary to overcome, if such type of biomarker is to be considered as a primary endpoint in future interventional trials of SCA3.

Because in slow progressing disorders, such as SCA3, the capture of small effects remains a challenge, the development of MBs is urgently needed. MBs should be crucial to improve sensitivity, when used in complement to clinical endpoints; they are also crucial because they can be applied to the presymptomatic stage, thus resolving one of the limitations of clinical markers. Furthermore, when ameliorating drugs will be available MBs being able to precociously detect pathogenic alterations will be of use to optimize therapeutics efficiency, since such compounds are expected to be more efficient if administrated to mutation carriers before disease.

16.2.1 Molecular Trait Biomarkers of SCA3

The CAG repeat at the ATXN3 locus constitutes the primary trait biomarker of SCA3. In fact, the identification of the ATXN3 gene opened prospects for the direct detection of the mutation, providing the grounds for molecular diagnosis and allowing presymptomatic (predictive) testing. The identification of carriers of the mutation before onset offers the possibility of understanding the full process of progression of the disease (Fig. 16.1), which in turn will be of major importance at a stage where therapeutic offers will be available, as previously referred. In SCA3 patients, a negative correlation between the number of CAG repeats in the expanded allele and age at onset has been widely reported, the number of CAGs accounting for nearly 50–75% of the variation in the age of appearance of first symptoms (revised in [8]); in the Azorean SCA3 cohort, for example, the number of CAG repeats in the expanded allele explains 68% of the onset variance [21]. In addition to the size of the CAG tract, age at onset, specific symptoms , rate of progression as well as degree of severity are notably modifiable by other factors. Such factors could theoretically be environmental or genetic; probably both contribute to the specific final phenotype. Although research on genetic modifiers of SCA3 is still at an incipient stage, several genetic factors have already been investigated as modifiers of SCA3: the number of CAG repeats at several expansion loci [21,22,23]; allelic variants at the interleukine 6 (IL6; [24]), apolipoprotein E (APOE) [25, 26] and glucosylceramidase beta (GBA) [27] genes; variation in the 3’UTR at the ATXN3 gene [28] and size of the normal SCA3 allele [29]. Moreover, for some of these genes/variants a possible cumulative effect has also been tested; in fact, Azorean SCA3 patients carrying the APOE*e2 and the IL6*C allele presented an onset which was anticipated by an average of 10 years [24]. More recently, exome sequencing has been proposed to study genetic modifiers of SCA3 (Manuela Lima, personal communication). The identification of genetic modifiers would allow: (1) targeting particular pathways/mechanisms for development of therapeutic interventions; (2) stratifying patients based on their genotype and incorporating this knowledge into the design of clinical trials . This type of stratification will empower clinical trials by controlling for the effect of the genetic background of participants [30]; and (3) improving the prediction of age at onset, aiming to provide better genetic counseling.

Fig. 16.1
figure 1

Stages, type of biomarkers and interventional studies which can be performed during the natural course of SCA3

Full size image

16.2.2 Molecular State Biomarkers of SCA3

16.2.2.1 Transcriptional biomarkers

As previously referred, abnormal conformation of mutated ataxin-3 promotes a gain of a toxic function compromising several cellular mechanisms, namely transcription. Transcriptional regulation seems to be affected by mutated ataxin-3 via two processes: (1) recruitment of transcription factors into polyQ-rich inclusions [31,32,33]; and (2) abnormal interactions with transcription factors and co-activators [34,35,36,37,38]. Transcriptional dysregulation was initially studied in cellular and animal models of SCA3 [35, 39, 40]. In such models, transcriptional alterations of genes involved in inflammatory processes, cell signaling and encoding cell-surface associated proteins has been described [35, 39, 40]. Based on this previous evidence Raposo and colleagues [6] hypothesized that in SCA3 patients the analysis of disease-specific transcriptional changes in blood, a peripheral tissue, had the potential to allow the identification of novel biomarkers . A cross-sectional study with SCA3 patients and controls, using the Illumina Human V4-HT12 array, confirmed the presence of differences in expression between the two groups [6]. Twenty six genes, found to be up-regulated in patients, were selected for a first step of validation by quantitative real-time PCR (technical validation). From these 21 genes, fourteen were subsequently selected for validation by qPCR in a new set of SCA3 patients and controls. In this second validation step, the expression levels of FCGR3B (Fc fragment of IgG, low-affinity IIIb, receptor (CD16b)), CSF2RA (Colony-stimulating factor 2 receptor, alpha, low-affinity (granulocyte-macrophage), CLC (Charcot-Leyden crystal protein), FPR2 (Formyl peptide receptor 2), SLA (Src-like-adaptor), GPR97 (G protein-coupled receptor 97), P 2RY13 (Purinergic receptor P 2Y, G-protein coupled, 13), TNFSF14 (Tumor necrosis factor (ligand) superfamily, member 14), SELPLG (Selectin P ligand) and YIPF6 (Yip1 domain family, member 6) were found to be 1.11–2.60-fold higher in patients when compared to controls. Noteworthy, FCGR3B, P 2RY13 and SELPLG genes were significantly up-regulated. Raposo and colleagues [6] further shown that, particularly for FCGR3B and CLC, patients with shorter disease duration tended to have higher expression levels when compared with patients with longer disease duration [6].

In a second approach Raposo and colleagues [41] listed molecules whose levels had been previously described in the literature as altered in SCA3 patients. The goal was to verify if expression levels of the selected genes HSPB1 (heat shock 27 kDa protein 1), DNAJB1 (DnaJ (Hsp40) homolog, subfamily B, member 1), DNAJB12 (DnaJ (Hsp40) homolog, subfamily B, member 12), DNAJB14 (DnaJ (Hsp40) homolog, subfamily B, member 14), BAX (BCL2-associated X protein), BCL2 (B-cell CLL/lymphoma 2), SOD2 (superoxide dismutase 2, mitochondrial), IL1B (interleukin 1, beta) and IL6 (interleukin 6) correlated with disease. The authors concluded that HSPB1 and BCL2 were significantly down-regulated in patients compared to controls. Given the previously highlighted importance of the preclinical stage of SCA3 Raposo and colleagues [41] expanded this analysis to include samples from preataxic SCA3 subjects. mRNA levels adjusted for age at blood collection were obtained for a set of premanifest SCA3 subjects, patients and controls. BCL2 levels were distinct in SCA3 subjects as compared to controls, although not being able to differentiate between premanifest carriers and patients. Moreover, lower levels of IL6 mRNA were also found in preataxic carriers.

In the quest for SCA3 biomarkers , longitudinal studies should provide the best quality data; such studies, however, represent an important effort for researchers and patients and therefore are lacking for SCA3. Due to the availability of a homogeneous subset of SCA3 patients with several blood collection points, Raposo and colleagues [41] investigated the behavior of HSPB1 and BCL2 mRNA levels in a longitudinal setup. Blood samples from SCA3 patients were collected at the baseline of the study and at a second moment. BCL2 and HSPB1 mRNA adjusted levels were found to be significantly different between the baseline and the second observational moment; during disease progression, the mRNA levels of BCL2 and HSPB1 increased.

16.2.2.2 Biomarkers of mitochondrial (mtDNA) depletion and damage

Amongst the several cellular pathways shown to be altered in the presence of mutated ataxin-3 are those associated with mitochondrial integrity and function. Because different compounds which improve energy metabolism defects or reduce oxidative stress (such as creatine and coenzyme Q10, amongst others) have been proposed as having the potential to ameliorate polyQ diseases, the investigation of the potential of mitochondrial alterations as biomarkers is pertinent. Metabolic alterations linked with mitochondrial function in SCA3 have been widely reported [42,43,44]. Furthermore, data on mtDNA depletion and increased damage (higher frequency of mutations, namely deletions) has also been produced [44,45,46,47,48]. Some of these previous reports indicated that alterations in the leukocytes mtDNA content had the potential to be used as biomarkers of SCA3; results from the several studies, however, were not consensual. Aiming to confirm previous findings of increased mtDNA depletion and damage, analysing a larger and independent set of patients, Raposo and colleagues [49] measured the levels of mitochondrial encoded NADH dehydrogenase 1 (MT-ND1), NADH dehydrogenase 4 (MT-ND4) and ribonuclease P RNA component H1 (RPPH1) genes, which enabled the determination of the mtDNA content and the evaluation of the common deletion (m.8470_13446del4977 deletion, which is considered a marker of age). In this study, along with blood samples collected from SCA3 patients and community controls, samples from preataxic carriers were also analysed. Although differences in mtDNA content were not evidenced in carriers of the mutation, as compared to controls, the common deletion was significantly more frequent in patients and preataxic subjects than in controls, after adjusting for age at collection. In preataxic subjects, moreover, a significant correlation between the number of CAG repeats in expanded allele and the frequency of common deletion was obtained; the frequency of common deletion was lower in preataxic subjects carrying a higher number of CAGs [49].

Alterations in several molecules related with other SCA3-associated mechanisms, quantified either at the mRNA or protein level have been reported and are synthetized in Table 16.1.

Table 16.1 Candidate biomarkers linked to several SCA3-associated mechanisms, either identified using cross-sectional and longitudinal analysis
Full size table

16.3 Conclusions

The quest for MBs of SCA3 has started; although studies are presently in an exploratory phase, longitudinal studies have begun to aid in the validation of results from cross-sectional analysis. Candidate transcriptional biomarkers evaluated so far will need to be further tested to warrant their reliability; this means that their behavior should be independent from inter-individual variation and comorbidities, for example. Similar to what is expected in similar neurodegenerative diseases, it is likely that for SCA3 not only a single biomarker, but a set of different molecules, will have to be identified. This set of MBs will be use, in complementarity with clinical and imaging markers, to fully track SCA3, from its presymptomatic phase to the disease stage.