Newborn Screening for Spinal Muscular Atrophy in China Using DNA Mass Spectrometry

Background: Spinal muscular atrophy (SMA) is the most common neurodegenerative disorder and the leading genetic cause of infant mortality. Early detection of SMA through newborn screening (NBS) is essential to selecting pre-symptomatic treatment and ensuring optimal outcome, as well as, prompting the urgent need for effective screening methods. This study aimed to determine the feasibility of applying an Agena iPLEX SMA assay in NBS for SMA in China. Methods: We developed an Agena iPLEX SMA assay based on the matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, and evaluated the performance of this assay through assessment of 167 previously-genotyped samples. Then we conducted a pilot study to apply this assay for SMA NBS. The SMN1 and SMN2 copy number of screen-positive patients were determined by multiplex ligation-dependent probe amplification analysis. Results: The sensitivity and specificity of the Agena iPLEX SMA assay were both 100%. Three patients with homozygous SMN1 deletion were successfully identified and conformed by multiplex ligation-dependent probe amplification analysis. Two patients had two SMN2 copies, which was correlated with severe SMA type I phenotype; both of them exhibited neurogenic lesion and with decreased muscle power. Another patient with four SMN2 copies, whose genotype correlated with milder SMA type III or IV phenotype, had normal growth and development without clinical symptoms. Conclusions: The Agena iPLEX SMA assay is an effective and reliable approach for population-based SMA NBS. The first large-scale pilot study using this assay in the Mainland of China showed that large-scale implementation of population-based NBS for SMA is feasible.

6 which may be detectable with the P021 SMA MLPA probemix (Arkblad et al. 2006). The great majority of SMA carriers can be identified by the presence of only a single SMN1 exon 7 copy. The one copy frequency in the US is estimated to be 1:37 for Caucasians, 1:46 for Ashkenazi Jews, 1:56 for Asians, 1:91 for African-Americans and 1:125 for Hispanics. Additionally, about 3-8% of SMA carriers (27% of African Americans) have two SMN1 copies on one chromosome and 0 copies on the other (2+0) (Alias et al. 2014, Ben-Shachar et al. 2011, Hendrickson et al. 2009, Miskovic et al. 2011, Smith et al. 2007). Dosage analysis cannot determine the difference between '1+1' and '2+0' (silent carriers) arrangements. Both situations are simply detected as having two SMN1 copies leading to false negative results. A thorough molecular analysis should be performed in parents and blood relatives of SMA patients when initial results indicate two SMN1 copies. Recently, Luo et al. (2014) reported that a haplotype block specific for SMN1 duplications is present in a large percentage of Ashkenazi Jews and in other ethnic groups. Identifying this haplotype will help distinguish silent carriers. If interested in more information regarding the identification of this haplotype, please contact info@mlpa.com.
The SMN2 copy number is very variable with only 60-70% of individuals having two copies. Provided that at least one functional SMN1 copy is present, complete absence of the centromeric SMN2 gene seems to have no clinical consequences. However, determining the SMN2 copy number is important for SMA patients: the more SMN2 copies, the less severe the disease is expected to be. More information on spinal muscular atrophy can be found in http://www.ncbi.nlm.nih.gov/books/NBK1352/.
Gene structure: SMN1 and SMN2, each having 9 exons, are part of a 500 kb inverted duplication on chromosome 5q13. SMN1 LRG_676 is available at http://www.lrg-sequence.org/ and is identical to Genbank NG_008691.1. SMN2 LRG_677 is pending approval and is identical to GenBank NG_008728.1. Exon numbering: The exon numbering used in this P060-B2 SMA product description and in the P060-B2 SMA lot-specific Coffalyser.Net analysis sheet is the traditional exon numbering (exons 1, 2a, 2b, and 3-8). Please note that the SMN1 and SMN2 exon numbering is different in the SMN1 and SMN2 LRG sequence and in the NCBI NG_008691.1 and NG_008728.1 reference sequence. P060-B2 probemix content: This SALSA ® MLPA ® probemix P060 SMA contains 21 MLPA probes with amplification products between 154 and 342 nt (Table 2) including 2 probes each for SMN1 and SMN2 (Table 2) and 17 reference probes that detect sequences outside this region. The identity of the genes detected by the reference probes is available online (www.mlpa.com).
-The SMN1 Exon 7 probe 14919-L17081 (183 nt) is the most important probe as it can be used to determine SMN1 copy number, which is important for deducing SMA carrier status. This probe is specific for SMN1 and will give no significant signal on SMN2. The probe has its ligation site at the C-to-T transition in exon 7, which is the site that affects RNA splicing in SMN2.
-The SMN1 Exon 8 probe 14881-L17082 (218 nt) is able to distinguish between SMN1 and SMN2 at exon 8 (G-to-A transition). The signal of this probe indicates the copy number of SMN1 exon 8. In approximately 95% of the samples, the copy number detected by the SMN1 exon 7 probe and the SMN1 exon 8 probe is identical. This SMN1 exon 8 probe cannot be used to quantify the number of SMN1 copies, as an exon 8 mutation will still result in a functional protein. Only the SMN1 exon 7 probe should be used to determine the SMN1 copy number. In the majority of the remaining 5% of samples, gene conversion between SMN1 and SMN2 has resulted in a chimeric gene containing the SMN1 exon 7 sequence and the SMN2 exon 8 sequence. Such a hybrid gene results in a functionally identical protein to the SMN1 protein. -The SMN2 Exon 7 probe 14921-L17083 (282 nt) identifies the SMN2 copy number, which is important for SMA patients, but has no influence on SMA carrier status.
-The SMN2 Exon 8 probe 14878-L17084 (301 nt) confirms the results obtained with the SMN2 exon 7 probe in most individuals. In case the copy number detected by this exon 8 probe does not correspond to that found by the exon 7 probe, only the exon 7 probe should be used to determine SMN2 copy number.
The summary of these findings and what they mean for carrier/patient status can be found in Table 1.
This probemix contains nine quality control fragments generating amplification products between 64 and 105 nt: four DNA Quantity Fragments (Q-fragments), three DNA Denaturation Fragments (D-fragments), and one chromosome X and one chromosome Y-specific fragment ( Table 2). The Q-fragments are only visible when less than 100 ng sample DNA is used. Low signal of the 88 or 96 nt fragment indicates incomplete DNA denaturation. More information on how to interpret observations on these control fragments can be found in the MLPA General Protocol.

MLPA technique:
The principles of the MLPA technique (Schouten et al. 2002) are described in the MLPA General Protocol (www.mlpa.com).

MLPA technique validation:
Internal validation of the MLPA technique using 16 DNA samples from healthy individuals is required, in particular when using MLPA for the first time, or when changing the sample handling procedure, DNA extraction method or instruments used. This validation experiment should result in a standard deviation <0.10 for all reference probes over the experiment.
Required specimens: Purified DNA from peripheral blood, (un)cultured amniotic fluid obtained in week 16 of the pregnancy or later and free from blood contamination, (un)cultured chorionic villi free from maternal contamination, or fetal blood. Samples should be free from impurities know to affect MLPA reactions. For more information please refer to the section on DNA sample treatment found in the MLPA General Protocol.

Reference samples:
The choice of reference samples is important for the correct determination of the SMN1 and SMN2 copy numbers. MRC-Holland is not able to provide reference DNA samples. One reason is that for MLPA reactions, the reference DNA samples should be derived from the same tissue type, handled using the same procedure, and prepared using the same DNA extraction method as the patient samples. It is strongly advised to first make a selection of suitable reference samples with known copy numbers before SMA testing is started. In particular, suitable reference samples are essential when testing patients with African ancestry. One method of doing this is to test a number (e.g. 16) of healthy individuals who are from families without a history of SMA for two copies of both SMN1 and SMN2. More information regarding the selection and use of reference samples can be found in the MLPA General Protocol. Interpretation of results: The expected results for SMN1 and SMN2 specific MLPA probes are allele copy numbers of 2 (normal), 0 (homozygous deletion), 1 (heterozygous deletion), 3 and occasionally 4. Allele copy numbers of 3 or 4 are usually due to gene conversion.
The standard deviation of all probes in the reference samples should be <0.10 and the dosage quotient (DQ) of the reference probes in the patient samples should be between 0.80 and 1.20. When these criteria are fulfilled, the following cut-off values for the DQ of the probes can be used to interpret MLPA results:  SMA patient SMN1 is absent, as no copies of the determining SMN1 exon 7 sequence are found. Due to gene conversion, 1 or more copies of the characteristic SMN1 exon 8 sequence appear to have become incorporated in the SMN2 gene.

SMA patient
If the patient has SMA symptoms, but one copy of SMN1 exon 7 is present, the patient may belong to the group presenting compound heterozygosity. Sequencing might reveal a defect in the remaining SMN1 copy.

SMA carrier
One copy of SMN1 is absent, making the person a carrier. The absence of one copy of the SMN1 exon 8 sequence confirms this.

SMA carrier
One copy of SMN1 is absent, making the person a carrier. A: due to gene conversion, 1 (or more) copies of the characteristic SMN1 exon 8 have become incorporated in the SMN2 gene. B: an SMN2 exon 8 copy has replaced the characteristic SMN1 exon 8 copy.

Most likely not a SMA carrier
Most likely this person is not a carrier. However, there is a possibility that both SMN1 copies lie on one chromosome. If there is a reason to believe that the person is a carrier (i.e. child is SMA-patient), he/she may belong to the 3-8% of carriers where this is indeed the case.
-Analysis of parental samples may be necessary for correct interpretation of complex results.
-False positive results: Please note that abnormalities detected by a single probe (or multiple consecutive probes) still have a considerable chance of being a false positive result. Abnormalities detected by a single probe may be due to a SNP very close to the ligation site. Sequence analysis can establish whether mutations or polymorphisms are present in the probe target sequence.  (Varga et al. 2012). Analysis of an independently collected secondary DNA sample can exclude these kinds of contamination artefacts. -Copy number changes detected by reference probes are unlikely to have any relation to the condition tested for.
Notes SMA results: SMA carrier screening: -False negative results: The presence of two SMN1 exon 7 copies suggests that the person tested is not a carrier. However, this result can also be due to the presence of two SMN1 copies on one chromosome and 0 on the other, in which case the person tested is in fact a SMA carrier. Detection of some carriers is therefore compromised, as MLPA and other techniques are not able to identify carriers who have one chromosome lacking SMN1 with the other chromosome carrying two copies of SMN1. -False positive results: Please note that individual MLPA probes can be affected differently by changes in experimental procedures or impurities in samples. Highly unlikely results such as an unusual high frequency of SMN1 exon 7 loss (carrier) or SMN1 exon 7 gain, without loss or gain of the exon 8 probe in most of these samples, should be treated with caution.

Limitations of the procedure:
-MLPA cannot detect any changes that lie outside the target sequence of the probes and will not detect copy number neutral inversions or translocations. Even when MLPA did not detect any aberrations, the possibility remains that biological changes in that gene or chromosomal region do exist but remain undetected. -Sequence changes (e.g. SNPs, point mutations, small indels) in the target sequence detected by a probe can cause false positive results. Mutations/SNPs (even when >20 nt from the probe ligation site) can reduce the probe signal by preventing ligation of the probe oligonucleotides or by destabilising the binding of a probe oligonucleotide to the sample DNA.
Please report copy number changes detected by the reference probes, false positive results due to SNPs, and unusual results to MRC-Holland: info@mlpa.com.

Confirmation of results:
As a way to confirm positive results, it is recommended to repeat the MLPA reaction with an independent DNA sample (independent DNA extraction). However, an apparent deletion detected by a single probe can be due to e.g. a mutation/polymorphism that prevents ligation or destabilises the binding of probe oligonucleotides to the DNA sample. Sequence analysis can establish whether mutations or polymorphisms are present in the probe target sequence.  Reference probe 01042-L17093 8q24 331 Reference probe 01043-L17094 8q13 342 Reference probe 13399-L17297 6q12

P060 Product history
Version Modification

B2
The 88 and 96 nt DNA denaturation control fragments have been replaced (QDX2). B1 Completely redesigned product. SMN2 exon 7 and 8 probes are now included. A2 Two extra control fragments at 100 and 105 nt, specific for chromosome X and Y, have been added. A1 First release.

Implemented changes in the product description
Version B2-04 -07 October 2016 (03) -Intended use and required specimens sections updated.