Locked nucleic acid (LNA) based PCR approach for the diagnosis and screening of spinal muscular atrophy

Patients and samples

Peripheral blood samples (2 mL) were collected from 31 patients diagnosed with SMA, 37 validated healthy controls, and 47 carrier parents.

Blood samples were stored at -80^oC, until use. All patient and controls samples were collected from individuals who enrolled to the Pediatric Neurology Clinic, Marmara University Pendik Training and Research Hospital, Turkey between September and December 2021. All SMA patients were children under 18 years of age who were clinically suspected of having SMA and confirmed by MLPA. These children had no comorbid conditions, and none of the participants had any history or clinical signs of other muscular atrophy disorders. The majority of carriers were parents of the patient children, all of whom had MLPA results. Carriers who were not parents of affected children were healthy volunteers identified through population screening as potential SMA carriers, with carrier status subsequently confirmed by MLPA. Healthy controls consisted of children under 18 years of age who were confirmed as non-carriers through newborn screening and healthy volunteers identified through population screening as non-carriers. For all individuals included in the study, SMA diagnosis was either confirmed or excluded through both clinical evaluation and MLPA analysis.

Child number 24 was found to be heterozygous by MLPA; however, due to the presence of an SMA phenotype, DNA sequencing was also performed, revealing a frameshift mutation in exon 7. As the primary goal of this study was to develop a specific LNA-based assay aligned with the gold-standard MLPA for SMA diagnosis, the initial evaluation was conducted in a limited sample group. Supplementary Table I summarises the SMN1 and SMN2 status of the samples according to MLPA.

DNA purification

Genomic DNA was extracted from peripheral blood samples using the Roche High Pure PCR Template Preparation Kit (Germany) according to the manufacturer’s recommendations.

LNA PCR: All PCR reactions were performed separately in a final volume of 20 µL, using 20 ng of template DNA, 10 pmol of SMN1 primers, 15 pmol of Actin and SMN2 primers. All reactions were run in triplicate, in LightCycler 480 II instrument with the LightCycler 480 SYBR Green I Master mix (Roche, Germany). Ct values were calculated using the LightCycler 480 Software (release 1.5.1.62) using built in absolute quantification/2^nd derivative max analysis module. Each primer was 3’ LNA modified and sequences are given in the Supplementary Table II. The primers were designed using the Primer3 software and validated in silico with IDT OligoAnalyser, NCBI BLAST, and UCSC in silico PCR tools. The primers were synthesised by Integrated DNA Technologies (IDT, USA).

Thermocycling conditions were as follows: polymerase activation at 95°C for 10 min, followed by 40 amplification cycles of 95°C for 15 sec and 61°C for 30 sec. Melting curves were obtained from 55-95°C with continuous ramps of 0.1°C. Ct values above 35 were considered negative. All PCR reactions were independently repeated by three researchers, and no significant variation was observed in the obtained ΔCt values among them (Supplementary Figs. 2-7). All healthy controls had a ΔCt value<0 and all carriers had ΔCt value >0. The detection limit was determined to be 0.4 ng/µL.

LNA PCR optimization

We collected samples from 115 individuals, genomic DNA was purified from blood samples. DNA concentrations were between 15-100 ng/µL and all concentrations were set to 5ng/µL for PCR experiments. We ran LNA-PCR for samples with different genotypes. Supplementary Figure 8 represents the amplification curves and melting peaks for SMN1 and SMN2 genes.

Sample “sss” has 2 copies of SMN1 gene but was homozygous for the deletion of SMN2. Supplementary Figure 9 represents the PCR amplification curve for SMN2 gene, green lines show the amplification curve of sample “sss” for SMN2. All samples have amplifications about 26^th cycle but sample “sss” who does not carry SMN2 gene has no amplification.

We also analysed SMA patients who are homozygous for SMN1 deletion, as seen in the Supplementary Figure 10 patients had no amplification with SMN1 primers whereas control samples carrying 2 copies of SMN1 gene had a Ct value before cycle 30.

After analysing all the samples, we understood that samples with deletions had no amplification thus primers were specific, but we further analysed PCR products by Sanger sequencing. We analysed SMN1 LNA PCR products of healthy controls and carriers because there were no PCR products in the SMN1 LNA PCRs of patients. We analysed SMN2 LNA PCR products of the PCR reactions of healthy controls, carriers, and patients with Sanger sequencing. SMN1 LNA PCR amplified only SMN1 gene whereas SMN2 LNA PCR amplified only SMN2 gene.

Supplementary Figure 11 shows chromatogram of SMN1 LNA PCR product of sample “GG” who had 2 copies of SMN1 and 2 copies of SMN2 with F primer, giving the sequence of ‘ACTTCCTTTATTTTCCTTACAGGGTTTCAGA CAAAATCAAAAAGAAGGA’. Supplementary Figure 12 shows the chromatogram of SMN1 LNA PCR product of same sample with R primer giving the sequence ‘ATAAAGCTATCTATATATATAGCT ATCTATGTCTATATAGCTATTTT’. By combining the two sequences we found that sequence was identic with NCBI SMN1 reference sequence.

We also analysed the SMN2 LNA-PCR products by Sanger sequencing and confirmed that the SMN2 primers specifically amplified only the SMN2 gene. Since all Sanger sequencing results demonstrated that the LNA primers did not amplify any non-specific products, the specificity of the LNA assay was validated. Representative chromatograms from different samples are provided upon request.

SMN1 quantification for the detection of patients and carriers

After confirming the specificity of the LNA-PCR, we calculated the ΔCt values for all samples and analysed gene copy numbers accordingly. The Ct values for SMN1 and Actin, as well as the corresponding ΔCt values, are presented in Table I. Figure 1 illustrates the graph of copy number and ΔCt values. All patients and carriers were clearly distinguished using the SMN1 LNA-PCR assay. All individuals with ΔCt values greater than zero were classified as carriers, whereas those with ΔCt values less than zero were classified as healthy controls. Ct values for the SMN1 gene in patient DNA samples were all above 35, indicating no detectable amplification. In carrier samples, Ct values for SMN1 ranged from 27.74 to 31.92 (mean±SD: 29.68±0.66). All carriers showed ΔCt values greater than zero. In healthy control samples Ct values for SMN1 ranged from 26.91 to 29.01(mean±SD: 28.06±0.39). Notably, sample 24C, an SMA patient with both heterozygous deletion and point mutation in the SMN1 gene, showed a Ct value of 30.9, consistent with the presence of one SMN1 copy carrying a point mutation. The results demonstrated complete concordance with MLPA results, with an area under the ROC curve (AUC) of 1.0, indicating perfect discrimination. No false-positive or false-negative results were observed; only patient 24C yielded a carrier-like ΔCt value, which was also unclassified by MLPA ( Fig. 2A).

Close

Fig. 1.

SMN1 copy number-ΔCt Graph of samples. (A) Graph of controls, carriers, and patients. (B) Graph of controls and carriers. Patients are excluded from the graph for more precise data presentation. SMN1 copy number is shown on the Y-axis, while the ΔCt value is represented on the X-axis.

Export to PPT

Close

Fig. 2.

Diagnostic performance of ΔCt values. (A) ROC for the discrimination between carriers and controls using SMN1 ΔCt values. (B) ROC for the determination of SMN2 copy numbers using SMN2 ΔCt values.

Export to PPT

SMN2 quantification in patient samples

The second objective of the SMA LNA-PCR method was to determine the SMN2 copy number in affected children using the ΔCt method. In a well-optimised PCR experiment, the difference in Ct values between a reference gene with a known copy number and the target gene with an unknown copy number should allow for the estimation of gene dosage and copy number. The Ct values obtained from PCR reactions conducted under optimised conditions are presented in the Table II. In our dataset, the ΔCt values for patients carrying 2 copies of the SMN2 gene ranged from -0.74 to 0.07, for those carrying 3 copies ranged from -1.43 to -0.96, and for patients carrying 4 copies were completely discriminated, with ΔCt values ranging from -1.78 to -3.34. As the sample size was limited and the ΔCt ranges for each copy number group did not overlap, the distinction between groups was unambiguous (Fig. 3). Consistently, Kruskal-Wallis analysis confirmed a statistically significant difference among the groups (P<0.0001). Consequently, ROC analysis yielded an area under the curve (AUC) of 1.0. (Fig. 2B). Given our limited sample size, performing Youden’s index or other cut-off optimisation calculations was not appropriate in this study. Instead, our focus here was on designing assays specific for the SMN1 and SMN2 genes, which can be applied in future studies involving larger cohorts.

Close

Fig. 3.

SMN2 copy number-. ΔCt Graph of SMA Patients. SMN2 copy number is shown on the Y-axis, while the ΔCt value is represented on the X-axis.

Export to PPT