Development of a single-tube molecular screen for five common spinocerebellar ataxiasBay Jia Wei, Li YuyaoBackground and Purpose of Research1.1. Spinocerebellar ataxias (SCAs)Spinocerebellar ataxias (SCAs) are a group of autosomal dominant neurodegenerative disorders, characterised by progressive cerebellar ataxia in conjunction with oculomotor dysfunction, dysarthria, pyramidal signs, extrapyramidal signs, pigmentary retinopathy, peripheral neuropathy, cognitive impairment among other symptoms1. Individual subtypes are differentiated by the specific genes in which the mutation occurs. The prevalence of individual subtypes may vary in different regions, however worldwide the most common subtypes are SCA 1, 2, 3, 6 and 72 (Annex 1), all of which are caused by abnormal expansion of polyglutamine-encoding CAG repeat3, leading to the formation of intranuclear aggregates and inclusions in the brains of patients. As the SCAs are progressive, severity and age of onset are determined by repeat size, with larger CAG repeat expansions indicative of earlier age of onset and more debilitating symptoms. The range of CAG repeats varies across each SCA subtype (Annex 2). 1.2. The need for a single-tube molecular screenDue to the overlap in clinical phenotypes observed between those with distinct genotypes in the different SCA subtypes9, blanket genetic testing, consisting of individual assays as it is currently impossible to amplify multiple loci simultaneously10, is needed in order to accurately determine the subtype of SCA in an affected person. Such tests involve the identification of the number of CAG repeats on specific causative genes. Yet, such a method is time-consuming and expensive, with a single SCA diagnostic test costing approximately US$30011. It is also unviable in preimplantation genetic diagnosis (PGD), a major clinical application of the diagnosis of the age-onset diseases, where single cell PCR (SCPCR) is routine and genetic testing has to be performed on just a single or a few biopsied embryonic cells12. To overcome the above limitations, there is a clear need for a multiplex SCA PCR assay able to amplify multiple loci simultaneously, much like that practiced by Lawrence et al13. Such an assay can ideally detect amplicon lengths and number of CAG repeats simultaneously and accurately across multiple SCA subtypes14, thus allowing for a rapid single-tube molecular screen. With the above criteria in mind, in this study, we aim to optimise this novel approach to screen for the five common SCA subtypes SCA 1, 2, 3, 6 and 7 by multiplex amplification.1.3. Our strategy utilising TP-PCR Conventional molecular tests are usually performed by amplifying the region containing the CAG repeat, followed by size determination of the resultant PCR products through electrophoresis15. However, even with the development of Long Range PCR16, result analysis using may still be complicated by PCR failure to amplify an expanded allele in heterozygotes15, causing heterozygous samples to appear homozygous. Southern blotting is a method of choice to eliminate the possibility of identification oversight, but is labor-intensive, time-consuming14 and not cost-effective especially in the handling of a few samples. As a result, Warner et al has described the triplet-repeat primed PCR (TP-PCR) strategy for any disorder with a triplet repeat expansion17. Recommended by the European Genetics Quality Network18 to confirm for circumstances of homoallelism for the molecular testing of SCAs when samples show apparent homozygosity, it is more sensitive yet less labour-intensive and has been proven to amplify CAG repeat expansions reliably. Such a method uses a fluorescently labelled locus-specific forward primer together with a pair of primers which have a common 5′ sequence (tail), which is a random, artificial DNA sequence that is not complementary to human DNA. On the other hand, a similar approach has been developed by Jama et al in their investigation on Huntington Disease15, yet it does away with the artificial tail sequence and only requires 2 primers. Therefore, in order to develop a robust and effective single-tube TP-PCR multiplex assay capable of rapidly diagnosing the five common SCA subtypes, SCA 1, 2, 3, 6 and 7, our study aims to investigate and compare both the 3-primer and 2-primer approaches of TP-PCR described by Warner et al and Jama at el respectively by evaluating their specificity, sensitivity and amplification efficiencies.HypothesisGiven the same principles behind both methods, we expect to see both methods perform similarly in terms of specificity, sensitivity and amplification efficiencies. Method and Materials3.1 TP-PCR Model and Primer DesignBoth TP-PCR methods consist of a locus-specific forward primer complementary to a sequence upstream of the repeat region, in addition to a reverse primer which, in the 3′ end, is complementary to and able to anneal anywhere with 5 consecutive CAG repeats within the repeat region. This generates stutters representing PCR products differing by 1 CAG repeat in amplicon size and enables amplification over the entire CAG repeat. For the 3-primer TP-PCR, the 3′ end of the reverse primer (5′ TPR) consists of 15 bp of CTG, complementary to five CAG repeats, with a few bp junction specific to the targeted SCA gene locus followed by a 21 bp tail section at the 5′ end (Figure 1). The tail section of TP primer has the same genomic sequence to an additional tail primer, which can anneal to amplified PCR products between cycles. Figure 1. Depiction of the 3-primer TP-PCR primers designOn the other hand, the 2-primer TP-PCR utilizes a chimeric reverse primer made up of 15 bp of CTG complementary to five CAG repeats followed by a 6-8 bp sequence specific to each SCA gene locus (Figure 2). Figure 2. Depiction of the 2-primer TP-PCR primers designThe forward and reverse primers were designed to have a melting temperature of approximately 60oC, 40-60% GC ratio, 18-22 bp primer length and staggered expected amplicon sizes to facilitate convenient fluorescence tagging and single-tube PCR amplification. Polymorphisms, primer dimers and reported variants within targeted exons were avoided to maintain specificity of the PCR assay.3.2 SamplesGenomic DNA obtained from National Institute of General Medical Sciences Human Genetic Cell Repository at the Coriell Institute for Medical Research (Camden, New Jersey) were used for assay optimization. SCA negative samples used were GM05164, GM06075 and GM07426. For screening purposes, a total of 5 SCA samples (Table 3) were used in this study: NA06926, NA13536, NA13537, GM06907, and GM06151. Table 3. The samples used in this study3.3 Simplex SCA Gene Amplification and Agarose Gel ElectrophoresisWe tested the performance and compared the efficiencies of various designed primers in simplex through SCA gene amplification conducted using the Applied Biosystems GeneAmp PCR System 9700 in a 25.00 µL PCR mixture. The reaction mixture contained Q solution (Qiagen), 10ng genomic DNA, 1 U HotStarTaq® DNA polymerase (Qiagen, Hilden, Germany), 1X supplied PCR buffer (containing 1.5mM MgCl2, Qiagen), 0.2 mM deoxyribonucleotide triphosphates (Roche, Penzberg, Germany), and respective primers. Amplification was performed with a 15-minute initial denaturation at 95oC, followed by 30 cycles of 98oC for 45 seconds, 60oC for 1 minute, and 72oC for 2 minutes, and a final extension at 72oC for 20 minutes. After which, 10 µL of amplified PCR products were mixed with 2 µL of formamide-loading dye and electrophoresed through a pre-prepared agarose gel at 100V and 100mA for 45 min. The resultant bands were viewed under the UVP GelDoc-It Imaging System and compared against a 1kB DNA ladder. PCR assays generating thicker and more defined bands were preferred and the corresponding primers were then chosen. Gene amplification was then carried out again using the chosen primers, this time fluorescently labelled, and PCR products run through capillary electrophoresis (CE) to confirm the peak pattern of the simplexes. 3.4 Optimization of Single-tube SCA Multiplex and CE AnalysisFluorescence-tagged forward primers were synthesised and labeled using standard dye sets: NEDTM for SCA1F, FAMTM for SCA3F and SCA7F, HEXTM for SCA2F and SCA6F. Fluorescent labels were selected to ensure that amplicons with overlapping size ranges were given different labels (Figure 4). For example, the primers for SCA2 and SCA6 were labelled with HEX because their sizes do not overlap.Figure 4. Fluorescence labels for different SCA subtypes. Bolded boxes indicate the amplicon lengths of normal to full penetrance of various SCA subtypes. Shaded boxes indicate amplicon lengths with reduced penetrance.Different parameters including annealing temperature, Q solution concentration, forward primer concentrations, reverse primer concentrations for 2-primer TP-PCR and tail primer concentrations for 3-primer TP-PCR were systematically tested to optimize the performance of the single-tube multiplex. The final optimized reagent concentrations in the 25.00 µL pentaplex PCR assay are as follows: 1.5X Q solution, 1X PCR buffer (containing 1.5mM MgCl2, Qiagen), 0.2 mM deoxyribonucleotide triphosphates (Roche, Penzberg, Germany), 2 U HotStarTaq® DNA polymerase (Qiagen, Hilden, Germany), 1X supplied PCR buffer (containing 1.5mM MgCl2, Qiagen), 0.2 mM deoxyribonucleotide triphosphates (Roche, Penzberg, Germany) and respective concentrations of primers as shown in Table 5.Table 5. Final optimised primer concentrations in SCA multiplexesPCR amplification experimental conditions were as previously described for simplex amplification. A total of 4µL PCR product was added to a mixture of 0.5µL of GS500-ROX internal molecular weight standard (Applied Biosystems Inc.) and 9µL of Hi-Di Formamide (Applied Biosystems Inc.). The mixture was denatured at 95oC for 5 minute and rapidly cooled to 4oC. The PCR fragments were resolved by electrophoresis in a Applied BioSystem (AB) 3130 Genetic Analyzer (GA) System using performance optimized polymer (POP-7), with a 36-cm array (Applied Biosystems Inc.). Samples were electrokinetically injected at 1kV for 15 seconds and electropherosed for 40 minutes. GeneScan analysis was performed using GeneMapper version 4.0 software (Applied Biosystems Inc.).Results and Discussion4.1. SimplexOur simplex electropherograms demonstrated that our primers were working well and displayed the expected TP-PCR stuttering peaks with 3 bp difference in size. For SCA6, it can be concluded as a heterozygous sample with two alleles because there are 2 prominent peaks indicating the strongest annealing regions at the 3’end of the CAG repeat region with the locus-specific junction. The repeat size can then be determined by counting from 5 (as the reverse primer anneals to 5 CAG repeats at the 5′ end of the CAG repeat region) until the prominent peak is reached. In the case of SCA6, the repeat sizes are 8 and 13. While it was expected that the 2-primer method would generate taller last peaks as a result of annealing more strongly at the last position due to the greater number of nucleotides on the reverse primer that are complementary to the gene sequence than that of the 3-primer method, our simplex electropherograms (Figure 6) showed otherwise. In SCA1, for example, there is no prominent peak indicating that the reverse primer annealed more strongly at the last position. This made it difficult to call the repeat size using the 2-primer method in the case of SCA1. When comparing the two TP-PCR methods, different peak patterns were generated although there is close concordance of the number of CAG repeats, except for SCA3, where the 2-primer method showed very prominent peaks at two positions, leading to a wipe-out of the smaller peaks in the expected stuttering pattern. This could possibly be attributed to the high G-C content of the 8 nucleotides immediately following the CAG repeat region that the chimeric reverse primer could have preferentially annealed to over the CAG repeat region (Annex 3).Figure 6. Electropherograms of 3-primer and 2-primer simplexes for the five SCA subtypes*6 CAG repeats which were affected by SNPs and thus not complementary to the reverse primers were accounted for4.2. Optimisation of pentaplexAfter running 3-primer and 2-primer TP-PCR on previously characterised samples, we obtained the results of the optimized assay (Figure 7). Using the 3-primer method, the peaks could be easily seen and counted, and peaks for each SCA subtype were roughly of the same size. Expanded alleles in all 5 SCA samples were accurately sized within ± 2 CAG repeats, giving 100% specificity, sensitivity and accuracy. Yet, we encountered a number of problems using the 2-primer method. For assay interpretation, prominent peak(s) are used to call CAG allele sizes. Yet, this feature is noticeably absent in the peak patterns of the 2-primer pentaplex of SCA1 positive samples (NA06926, NA13536, and NA13537), as well as SCA3 positive sample (GM06151). In these cases, the peak corresponding to the correct repeat size is not prominent unlike with the same samples using the 3-primer pentaplex. The absence of such an important feature is a detriment to the sensitivity of the 2-primer assay. Furthermore, as can be seen in Figure 7, the peaks in the electropherograms of the 2-primer assays were jagged as compared to the singular peaks of the 3-primer assays. Such jagged peaks can be easily confused to be 2 peaks rather than just 1 peak, giving rise to the possibility of false positives. Additionally, a substantial number of peaks in the electropherogram of the 2-primer assay were simply too small to be determined accurately. This problem is particularly severe if it occurs at the tail end of electropherograms, such as in the SCA3-positive sample, where stuttering causes peak heights to decrease so much that peaks belonging to larger amplicons cannot be seen, much less counted. In such cases it is simply impossible to determine the number of peaks and therefore CAG repeats accurately. Thus, this assay cannot be used to determine the repeat sizes of very large expansions. Figure 7. Electropherograms of 3-primer and 2-primer multiplexes for SCA 1, 2, 3, 6 and 7*6 CAG repeats which were affected by SNPs and thus not complementary to the reverse primers were accounted forConclusion and Recommendations for Future WorkThrough the optimisation and subsequent comparison of both 3-primer and 2-primer assays, we have concluded that the 3-primer assay can accurately size CAG repeat regions, as well as reliably detect apparent homozygous samples and eliminate possibilities of failure in the identification of expanded alleles for heterozygotes. Thus, it can be used as a robust and effective single-tube molecular screen for five common SCAs. Yet, we acknowledge the relatively limited sample pool that was used for screening. Given more time in the future, we can expand our sample pool, taking note to source for SCA6 and SCA7-positive samples in order to ensure the reliability of our assay such that it can be used as an accurate and reliable diagnostic tool in the future. Furthermore, although our assay was unaffected by this phenomenon, the discovery of interruptions within the CAG repeat region in certain SCA subtypes has led to the possibility of TP-PCR failure due to the inability of the reverse primer to anneal, potentially leading to false-negatives. In anticipation of this, further research in bidirectional TP-PCR targeting both the 5′ and 3′ ends of the CAG repeat can be done. Additionally, as we successfully multiplexed 5 common SCA subtypes, it is reasonable to project future research in extending our PCR assay to include additional SCAs in addition to other related disorders without compromising on assay specificity and sensitivity.