Accuracy of Marker Analysis, Quantitative Real-Time Polymerase Chain Reaction, and Multiple Ligation-Dependent Probe Amplification to Determine SMN2 Copy Number in Patients with Spinal Muscular Atrophy

Abstract

Spinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disorder caused by absence of or mutations in the survival motor neuron1 gene (SMN1). All SMA patients have a highly homologous copy of SMN1, the SMN2 gene. Severe (type I) SMA patients present one or two SMN2 copies, whereas milder chronic forms (type II-III) usually have three or four SMN2 copies. SMN2 dosage is important to stratify patients for motor function tests and clinical trials. Our aim was to compare three methods, marker analysis, real-time quantitative polymerase chain reaction using the LightCycler instrument, and multiple ligation-dependent probe amplification (MLPA), to characterize their accuracy in quantifying SMN2 genes. We studied a group of 62 genetically confirmed SMA patients, 54 with homozygous absence of exons 7 and 8 of SMN1 and 8 with SMN2-SMN1 hybrid genes. A complete correlation using the three methods was observed in 32 patients (51.6%). In the remaining 30 patients, discordances between the three methods were found, including under or overestimation of SMN2 copies by marker analysis with respect to the quantitative methods (LightCycler and MLPA) because of lack of informativeness of markers, 3' deletions of SMN genes, and breakpoints in SMN2-SMN1 hybrid genes. The technical limitations and advantages and disadvantages of these methods are discussed. We conclude that the three methods complement each other in estimating the SMN2 copy number in most cases. However, MLPA offers additional information to characterize SMA cases with particular rearrangements such as partial deletions and hybrid genes.

Full text

ORIGINAL ARTICLES
Accuracy of Marker Analysis, Quantitative Real-Time
Polymerase Chain Reaction, and Multiple Ligation-Dependent
Probe Amplification to Determine SMN2 Copy Number
in Patients with Spinal Muscular Atrophy
Laura Alı´as,
1
Sara Bernal,
1
Maria J. Barcelo´,
1
Eva Also-Rallo,
1
Rebeca Martı´nez-Herna´ ndez,
1
Francisco J. Rodrı´guez-Alvarez,
2
Concepcio´ n Herna´ ndez-Chico,
2
Montserrat Baiget,
1
and Eduardo F. Tizzano
1
Spinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disorder caused by absence of or
mutations in the survival motor neuron1 gene (SMN1). All SMA patients have a highly homologous copy of
SMN1, the SMN2 gene. Severe (type I) SMA patients present one or two SMN2 copies, whereas milder chronic
forms (type II-III) usually have three or four SMN2 copies. SMN2 dosage is important to stratify patients for
motor function tests and clinical trials. Our aim was to compare three methods, marker analysis, real-time
quantitative polymerase chain reaction using the LightCycler instrument, and multiple ligation-dependent probe
amplification (MLPA), to characterize their accuracy in quantifying SMN2 genes. We studied a group of 62
genetically confirmed SMA patients, 54 with homozygous absence of exons 7 and 8 of SMN1 and 8 with SMN2-
SMN1 hybrid genes. A complete correlation using the three methods was observed in 32 patients (51.6%). In the
remaining 30 patients, discordances between the three methods were found, including under or overestimation
of SMN2 copies by marker analysis with respect to the quantitative methods (LightCycler and MLPA) because of
lack of informativeness of markers, 3¢deletions of SMN genes, and breakpoints in SMN2-SMN1 hybrid genes.
The technical limitations and advantages and disadvantages of these methods are discussed. We conclude that
the three methods complement each other in estimating the SMN2 copy number in most cases. However, MLPA
offers additional information to characterize SMA cases with particular rearrangements such as partial deletions
and hybrid genes.
Introduction
Spinal muscular atrophy (SMA) is an autosomal reces-
sive disease characterized by degeneration of a-motor
neurons of the spinal cord. Mutations in or absence of the
survival motor neuron 1 gene (SMN1) (NM_000344) are re-
sponsible for the disease (Lefebvre et al., 1995). A highly ho-
mologous copy of SMN1, the SMN2 gene (NM_017411), is
found in all patients. Approximately 90% of SMA patients show
a homozygous absence of exons 7 and 8 of SMN1, and another
5% of patients have SMN2-SMN1 hybrid genes (Wirth et al.,
1999; Cusco et al., 2001). The vast majority of the remaining 5% of
patients present one allele with an SMN1 deleted copy and the
other allele with a point mutation (Alias et al., 2009).
According to motor milestones and age of onset, SMA has
been classified into four clinical groups. The Werdnig-
Hoffmann syndrome (MIM #253300), or type-I SMA, is the
most severe form: the disease begins before 6 months and
patients rarely live beyond 2 years of age. In intermediate
forms, or type-II SMA (MIM # 253550), the disease appears
before 18 months of age; patients are able tosit but are unable to
walk unaided. Type-III SMA, also called Kugelberg-Welander
disease (MIM#253400), is a milder form and it manifests after
18 months of age; patients are able to stand and walk but often
become wheelchair-bound during youth or adulthood.
SMN2 is considered a phenotypic modifier of the disease.
Acute type-I SMA patients usually present one or two SMN2
copies, whereas most of chronic type II and III patients have
three or four copies (Feldkotter et al., 2002; Cusco et al., 2006;
Wirth et al., 2006). It is therefore very important to estimate
the SMN2 copy number so as to be able to stratify SMA pa-
tients for motor function tests (Tiziano et al., 2007) and clini-
cal trials (Sumner, 2006). Further, the analysis of subgroups
of patients with different SMA types but identical number of
SMN2 copies would be helpful to identify additional genes
that may modify the phenotype.
1
Servicio de Gene
´tica, Hospital de la Santa Creu i Sant Pau, CIBERER (U705), Barcelona, Spain.
2
Unidad de Gene
´tica, Hospital Ramo
´n y Cajal, CIBERER (U728), Madrid, Spain.
GENETIC TESTING AND MOLECULAR BIOMARKERS
Volume 00, Number 00, 2011
ªMary Ann Liebert, Inc.
Pp. 1–
DOI: 10.1089/gtmb.2010.0253
1

The number of SMN2 copies in SMA individuals with a
homozygous deletion of the SMN1 gene can be estimated by
screening of multicopy markers (C212 and C272) located in
the 5¢region of the SMN genes. However, it can now be more
reliably calculated since the appearance of real-time quanti-
tative polymerase chain reaction (qPCR) by LightCycler (LC)
methodology (Cusco et al., 2002; Feldkotter et al., 2002), Taq-
man technology (Anhuf et al., 2003), the DHPLC instrument
(Su et al., 2005), and the more recent multiple ligation-
dependent probe amplification (MLPA) (Tomaszewicz et al.,
2005; Arkblad et al., 2006).
We compared the accuracy of C212 and C272 marker
analysis, qPCR, and MLPA to estimate the SMN2 copy
number in a group of Spanish SMA patients.
Materials and Methods
Patients
Genomic DNA from peripheral blood of 62 Spanish SMA
patients (Type I =18, Type II =30, Type III =14) was extracted
by the salting-out method (Miller et al., 1988). The study in-
cludes 54 patients with absence of exons 7 and 8 of the SMN1
gene, and 8 more patients who were carriers of SMN2-SMN1
hybrid genes. All patients or parents signed their corre-
sponding informed consent.
Haplotype analysis
We amplified two microsatellite markers, D5S1556 (Ag1-
CA =C272) (DiDonato et al., 1994) and D5F149S1 (C212)
(Melki et al., 1994), located upstream of the 5¢region of the
SMN genes, with primers labeled with two different fluoro-
chromes: NED (yellow) and FAM (blue), respectively. The
PCR products were mixed with ROX-500 molecular weight
marker (red) and analyzed using the ABI Prism 3100 Avant
Capillary DNA Sequencer. The genotypes were determined
using the GeneScan software package (Perkin Elmer-Applied
Biosystems

, Foster City, CA) (Fig. 1A).
Quantitative real-time PCR
To determine the SMN2 gene copy number, we used a
quantitative real-time PCR assay (LightCycler instrument;
Roche Diagnostics, Basel, Switzerland) based on specific
amplification of exon 7 of this gene, as previously described
(Cusco et al., 2002; Feldkotter et al., 2002). Briefly, we used
primers designed to specifically anneal exon 7 of the SMN2
gene based on the nucleotide differences with the SMN1 gene
at nucleotide +6 from exon 7 and at +214 from intron 7 (Fig.
1B, C). The PCR was performed in a total volume of 10 mL
containing 7.5 ng of genomic DNA, 10 pmol of each primer,
1mL of Roche Molecular Biochemicals Fast-start DNA Kit
FIG. 1. Schematic representation of the SMA locus. (A) Organization of the SMA locus with the centromeric and telomeric
repeated elements. Markers C212 and C272 are located -13 kb and -468 bp, respectively, from the 5¢end of the SMN genes.
Information was according to Map Viewer NCBI database (www.ncbi.nlm.nih.gov). (B) Distribution of the MLPA probes
(upper part) and LC primers (lower part) along the SMN genes. The forward LC primer overlaps 24 nucleotides (16 intronic
and 8 exonic including the C >T transition that differentiates SMN2 exon 7 from SMN1 exon 7) with the specific exon 7 probe
of the SMN2 gene of MLPA. Common SMN1/SMN2 MLPA probes (black rectangle); specific SMN2 probes of the MLPA
(gray rectangle); UTR regions (dotted boxes); forward and reverse specific SMN2 primers (black arrows). (C) Partially
magnified scheme of exons 7 and 8 and intron boundaries with nucleotide differences between SMN1 and SMN2. Expected
SMN2 hybridization regions of LC primers and the specific exon 7 probe of MLPA. [Note that the SMN2LC reverse primer is
modified in one base to allow specific amplification (Feldkotter et al., 2002)]. SMA, spinal muscular atrophy; MLPA, multiplex
ligation-dependent probe amplification; LC, LightCycler.
2 ALI
´AS ET AL.

SYBR GreenI, and 4 mM MgCl
2
. The associated software tool
is based on the Second Derivative Maximum method, which
performs online measurements of the SYBR Green I molecule
signal in each PCR cycle. The results of the SMN2 copy
number were extrapolated from a regression line built by
dilutions of a sample with homozygous deletion of the SMN1
gene and with a known SMN2 copy number.
The PCR program contains a melting step to ensure that
the entire detected signal is exclusively obtained by amplifi-
cation of SMN2 exon 7. Further, as PCR products were un-
detected in individuals without SMN2 (data not shown),
cross-amplification of the SMN1 with specific SMN2 primers
was ruled out.
MLPA technique
The p021 SMA kit, manufactured by MRC-Holland
(www.mrc-holland.org), is designed to amplify 37 regions, 16
located at SMA locus on 5q13 and 21 distributed throughout
the genome. Eight of the SMA locus probes are complemen-
tary to SMN gene sequences, either in common regions (in
exons 1, 4, 6, and 8) or in specific fragments for each SMN gene
(in exons 7 and 8) (Fig. 1B, C). The MLPA

method is based on
a multiplex PCR strategy. Each amplified fragment was sep-
arated and quantified by capillary electrophoresis on an ABI
Prism 3130 Avant Capillary DNA Sequencer and data were
visualized with the GeneMapper software v4.0 package
(Perkin Elmer-Applied Biosystems). Internal normalization to
obtain the relative peak area of each sample was achieved by
dividing the peak area value of each SMN probe by the
combined peak areas of the control probes. The relative peak
area value was set according to one of two external controls:
(1) known samples with two SMN1 and two SMN2 copies and
(2) known samples with no SMN1 and two SMN2 copies.
Intervals to define the copy number of each sample were
defined following the concept of Huang et al. (2007) with
slight modifications in function of MLPA screening in our
Spanish population (further explanation in Supplementary
Fig. S1; Supplementary Data are available online at
www.liebertonline.com/gtmb).
Sequencing methods
Genomic DNA was amplified using R111 as forward pri-
mer in intron 6 and 541C1120 as reverse primer in exon 8
(Lefebvre et al., 1995). The 845-bp amplification product was
purified by ExoSAP-IT

from USB (product from Affymetrix,
Cleveland, OH) before direct forward and reverse sequencing
using BigDye v1.1 Terminator Reaction Kit on an ABI PRISM
3100 Avant DNA automatic sequencer unit following the
manufacturer’s protocol (Applied Biosystems). The sequences
obtained were compared with the reference sequence from
GenBank (NT_006713).
Results
After screening of 62 SMA patients, we detected complete
concordance between the three methods to estimate SMN2
copies in 32 patients (51.6%). Twenty-eight of these patients
had absence of exons 7 and 8 of SMN1, whereas four showed
SMN2-SMN1 hybrid genes. In the remaining 30 patients, three
main types of discrepancies were detected (Table 1):
1. Lower SMN2 copies by C212 and C272 marker analysis:
in 20 patients with homozygous deletion (for sample
details, see Table 2) and in 3 with SMN2-SMN1 hybrid
Table 1. Summary of the Comparative Results Between the Three Methods Used in This Work
Methods Patients
EX7/EX8 SMN1 Results Markers LC MLPA Type I Type II Type III Total Remarks
-/-Concordants C C C 10 11 7 28 25 patients with total concordance in
SMN2 copy number between the
three methods.
3 patients with a borderline result in
some of the MLPA probes.
Discordants U C C 4 11 5 20 15 patients with noninformative
markers.
5 patients with a noninformative
marker and a borderline value in
some of the MLPA probes.
O C C 4 2 – 6 Underestimation of the SMN2 by LC
and the 3¢MLPA probes with
respect to markers and 5¢MLPA
probes (exons 1, 4, and 6).
-/+Concordants C C C – 3 1 4 Concordance considering marker,
LC, and the specific exon 7 SMN2
probe of MLPA.
Discordants U C C – 2 1 3 Noninformative markers.
C U C – 1 – 1 Coincidence of SMN1 and SMN2
sequences that involved the
reverse LC primer.
-/-, SMA patients with homozygous deletion of exons 7 and 8 of the SMN1 gene; -/+, SMA patients with SMN2-SMN1 hybrid genes
(without exon 7 but with the presence of exon 8 of the SMN1 gene); U, underestimation; O, overestimation; C, concordance of results between
methods. Further information is given in Tables 2 and 3.
SMA, spinal muscular atrophy; MLPA, multiplex ligation-dependent probe amplification; LC, LightCycler.
COMPARISON OF THREE METHODS TO QUANTIFY SMN2 GENES 3

Table 2. SMN2 Copy Number Estimation by Three Different Methods Investigated in 54 Spinal Muscular Atrophy
Patients with Homozygous Deletion of Exons 7and 8of the SMN1 Gene
SMN2 copy number
MLPA
Allele number SMN 2 SMN1/SMN2 probes
No. SMA Type C212 C272 LC Exon 7 Exon 8 Exon 1 Exon 4 Exon 6 Exon 8
1 I 2222 22222
2 I 2222 22222
3 I 2222 22222
4 I 2222 22222
5 I 2222 22222
6 I 2222 22222
7 I 2222 22222
8 I 1122 22222
9 I 1122 22222
10I 1222 22222
11I 2122 22222
12 I 2 2 2 2 1–2 2222
13 I 2 2 2 2 1–2 2222
14I 2222 22222
15II 2222 22222
16II 3333 33333
17II 3333 33333
18II 3333 33333
19II 3333 33333
20II 3333 33333
21II 3333 33333
22II 3333 33333
23II 2233 33333
24II 2233 33333
25II 2333 33333
26II 3233 33333
27II 3233 33333
28 II 3 2 3 3 3 3–4 333
29 II 3 2 3 3 3 3–4 333
30 II 3 2 3 3 3 3–4 333
31 II 3 1 3 3 3 2–3 333
32 II 3 2 3 3 3 3 3 2–3 3
33II 2333 33333
34II 3333 33333
35 II 3 3 3 3 2–3 3333
36II 3333 33333
37III 3333 33333
38III 3333 33333
39III 3333 33333
40III 3333 33333
41III 3333 33333
42III 3333 33333
43III 2233 33333
44III 2233 33333
45III 1133 33333
46III 3233 33333
47III 3233 33333
48III 3333 33333
49
a
I 2211 12221
50
a
I 4422 24442
51
a
I 2322 23332
52
a
I 3222 23332
53
a
II 3 4 3 3 3 4 3–4 3–4 3
54
a
II 3222 23332
Discrepant results are depicted in gray boxes and borderline results in bold.
a
Samples assumed as 3¢partial deletions.
4 ALI
´AS ET AL.

genes (Table 3), the SMN2 copy number obtained from
haplotype analyses was underestimated with respect to
the values obtained by LC (exon 7 only) and MLPA
techniques (specific exon 7 and 8 probes). This was as-
sumed as lack of informativeness in one or both
markers. In 15 of these cases, the discordance was ob-
served in either C212 or C272 CA-microsatellites. In the
remaining eight patients, the lack of informativeness
was detected in both multicopy markers (an example is
illustrated in Fig. 2). In our patients with complete ab-
sence of SMN1 (n=48), from 129 expected SMN2 copies
(compiled from the results of quantification of the exon
7 by LC and MLPA; Table 2), C212 alleles were able to
distinguish 118 SMN2 copies (91.5%), whereas C272
was able to distinguish 110 (85.3%).
2. Higher SMN2 copies by marker analysis: in six other
patients the SMN2 copy number obtained by marker
analysis was higher than that obtained by quantitative
LC methods and specific SMN2 probes of MLPA. In
contrast, results obtained with the common MLPA
probes of the SMN1/SMN2 genes (exons 1, 4, and 6)
were in agreement with the marker studies. We con-
sidered that these cases represented partial deletions of
the 3¢end of the SMN genes. Given that the results
between the specific SMN2 exon 8 probes and the
common SMN1/SMN2 exon 8 probe of the MLPA were
in agreement (Table 2) and the absence of specific
MLPA probes to discriminate the 5¢region, it was not
possible to ascribe these partial deletions to either of the
SMN genes.
3. LC underestimation in hybrid genes. When we ana-
lyzed SMN2-SMN1 hybrids by MLPA, the specific
probes showed an extra copy of SMN1 exon 8 that
corresponded to a missing copy of SMN2 exon 8. The
results of the common SMN1/SMN2 MLPA probes
agreed with these findings. However, in one case (hy-
brid number 6 from Table 3), LC showed one less exon
7SMN2 copy than MLPA. LC primers are designed to
hybridize the specific sequences that differentiate SMN1
and SMN2. Given that the forward LC primer overlaps
with the SMN2 exon 7 probe of MLPA (Fig. 1B, C), we
speculated that a possible mismatch (a instead of g)
may have influenced the annealing of the reverse LC
primer in our patient. A genomic sequence analysis
from intron 6 to exon 8 of this patient showed that only
SMN2 sequences were detected up to the nucleotide
position +214 from the exon 7. From this position on-
ward, heterozygous peaks at the clue positions in
SMN1 and SMN2 were present (Fig. 3A, B). In partic-
ular, the nucleotide +215 from exon 7 (that corresponds
to the second nucleotide of the reverse LC primer)
predicted a mismatch in this sample.
Discussion
For several years we have been routinely analyzing SMA
patients and family members with C272-C212 markers to es-
tablish the at-risk haplotype and with LC to determine their
SMN2 copy number (Cusco et al., 2002; Feldkotter et al., 2002).
The emergence of the MLPA technique opened new ap-
proaches to quantify SMN genes as well as define possible
structural changes in patients (Tomaszewicz et al., 2005;
Arkblad et al., 2006; Huang et al., 2007). The initial results
obtained with markers and LC were further expanded with
the values obtained by MLPA analysis, either from the specific
SMN2 probes (exons 7 and 8) or with the common SMN
probes (exons 1, 4, 6, and 8).
In this first systematic comparison of three methods (C212
and C272 marker analysis, LC, and MLPA) to estimate the
SMN2 copy number in a large group of Spanish SMA patients,
we found a total concordance between the three approaches in
51.6% of our 62 patients. We observed three main types of
discrepant results in the remaining patients.
The first discrepancy was associated with multicopy
markers C212 and C272, underestimating SMN2 copy num-
ber. These markers are useful to determine the at-risk haplo-
type in carrier studies in SMA families and they were key to
discover de novo deletions in affected patients (Melki et al.,
1994). C272 is closer (468 bp) to the 5¢end of the SMN genes
than C212, which is located 13 kb upstream SMN (Fig. 1). In
our study, C212 alleles distinguished more SMN2 copies
(94.5%) than C272 (89.4%). This difference may be due to the
fact that SMN deletions often encompass the closer C272 mar-
ker. However, the alleles detected by C212 were concordant
Table 3. SMN2 Copy Number Estimation by the Three Different Methods in Eight Spinal Muscular Atrophy
Patients with SMN2-SMN1 Hybrid Genes
SMN2 copy number
MLPA
Allele number SMN1 SMN2 SMN1/SMN2 probes
SMA type C212 C272 LC Exon 7 Exon 8 Exon 7 Exon 8 Exon 1 Exon 4 Exon 6 Exon 8
1II 2220121 2222
2II 3330132 3333
3II 3330132 3333
4II 22302313–4 333
5II 1220121 2222
6II 2220132 3333
7 III 3 3 3 0 2 3 1 3 3 3 3
8 III 3 2 3 0 1 3 2 2 2 2 3
Discrepant results are depicted in gray boxes and borderline results in bold.
COMPARISON OF THREE METHODS TO QUANTIFY SMN2 GENES 5

with the MLPA results (see Table 2), indicating that these
alleles were associated with SMN2 and not with the deleted or
absent SMN1. We consider that the lack of informativeness of
these two markers was responsible for the discrepancy ob-
served in comparison with the other two methods. To avoid
pitfalls in the initial interpretation of the results, it is necessary
to study family members as illustrated in Figure 2. Thus, the
results of marker analysis in SMA patients to estimate SMN2
copies should always be cautiously considered and be com-
plemented with family results and a quantitative analysis.
The second discrepancy concerns an SMN2 overestimation
by C212 and C272 in comparison with the results obtained
through LC and MLPA in six cases. In these cases, marker
estimation agreed with results of the common MLPA probes
for exons 1, 4, and 6 but was discordant with respect to MLPA
probes of the 3¢end (common exon 8 probe and specific exon 7
and 8 probes) and to LC quantification. We considered these
rearrangements as partial deletions of the 3¢end. This mo-
lecular event has been described as a polymorphism (Arkblad
et al., 2006) in SMA carriers and the general population, albeit
not in patients. Interestingly, four of our six patients with this
rearrangement had severe type I and the other two had type
II. These SMA forms usually have fewer SMN2 copies, sup-
porting the idea that these six patients may have truncated
SMN2 genes.
Quantitative results in patients with hybrid genes may be
difficult to interpret. Our patient 6 (Table 3), who had an
SMN2-SMN1 hybrid gene structure, showed LC values that
were lower than values obtained using the other two ap-
proaches (markers and MLPA). A detailed sequence of
the involved hybrid region revealed SMN1/SMN2 genes in
the +215 nucleotide from exon 7. In this position, which
FIG. 2. Example of noninformative alleles in an SMA family. Markers C212 (upper electropherograms) and C272 (lower
electropherograms) analyses show markedly different peak areas in the two alleles of individual II-2. The LC and MLPA
results in this patient were compatible with three SMN2 copies. The 217 allele (C212) and the 183 allele (C272) have been
inherited from both progenitors and their peak areas are bigger than the other two alleles (221 and 191, respectively),
indicating the presence of two SMN2 copies in relation to the 217 and 183 alleles. In contrast, her brother II:1 shows more
homogeneous peak areas, compatible with a single dose of each allele. MW, molecular weight marker.
6 ALI
´AS ET AL.

corresponds to the second nucleotide of the reverse primer
used with LC, we would usually find only the SMN2 gene. A
lower efficiency of amplification may explain the underesti-
mation with LC. In contrast, MLPA results were accurate
because the specific MLPA exon 7 probe for the SMN2 gene
was able to hybridize and amplify the complementary se-
quence under analysis (Fig. 3B). As the breakpoints of hybrid
genes may be different in each patient (Cusco et al., 2001), the
nature of the hybrid structure may influence the final estimate
of SMN2 copies by quantitative methods.
Other possible technical limitations should also be consid-
ered. The results of MLPA may be borderline for the specific
exon 8 probe (see, for example, cases 12, 13, and 35 in Table 2).
Such cases can be resolved, however, by considering the re-
sults from the remaining probes. Another possible limitation
in MLPA that may generate uncertain results is the quality of
the DNA. This is chiefly influenced by the extraction method,
as previously described with the LC quantification (Cusco
et al., 2002; Feldkotter et al., 2002). We suggest that test sam-
ples should be normalized with respect to controls by the
same method of extraction. The same concept applies to the
intervals employed to categorize results, such as the common
exon 1 probe that produces several borderline results as in
patients 28–31. It could be good practice in each laboratory to
test a number of control and known SMA samples to better
redefine the intervals employed (see Supplementary Fig. S1).
Cumulative knowledge about genotype–phenotype corre-
lations in SMA confirms an inverse relationship between the
SMN2 copy number and disease severity (Feldkotter et al.,
2002; Cusco et al., 2006; Wirth et al., 2006) and shows that
quantification of this gene is imperative in affected patients.
Such information is also useful when stratifying patients for
entry into clinical trials. The results of our study provide
helpful information and assessment to be considered for im-
plementation of SMN2 dosage in molecular genetic labora-
tories.
In conclusion, marker analysis, LC, and MLPA complement
each other in estimating the SMN2 copy number. If concor-
dance between the three methods is not complete, results
should be carefully evaluated according to informativeness of
markers, type of mutation, SMN structure, and DNA quality of
the samples. Although quantitative LC is a reliable and acces-
sible method, MLPA further characterizes SMA cases with re-
arrangements such as partial deletions or hybrid genes.
Acknowledgments
This workwas supported by CIBERER (to L.A.and F.J.R.-A.),
GENAME Project (to S.B., R.M.-H., E.F.T., and C.H.-C.),
FIS05-2416 (to E.A.-R.), and grants FIS 05-2416 and 08-0729 (to
E.F.T.). The authors thank the consenting patients and parents
who made this study possible.
Disclosure Statement
No competing financial interests exist.
References
Alias L, Bernal S, Fuentes-Prior P, et al. (2009) Mutation update
of spinal muscular atrophy in Spain: molecular characterization
FIG. 3. Sequence analysis of patient 6 (Table 3) with a particular hybrid structure and discordant LC vs. MLPA results. Only
nucleotides differences between SMN genes are depicted (further information in Fig. 1). (A) Electropherogram of the hybrid
region around position +215 in intron 7. The sequence upstream is not depicted. In this position the first heterozygous peak
(black arrow) is detected corresponding to the nucleotide differences between SMN1 and SMN2 genes. The nucleotides where
the reverse LC-modified primer for SMN2 hybridizes are within a rectangle. (B) Schematic representation of two SMN2
complete genes (upper part) and the SMN2-SMN1 hybrid gene (lower part) of this patient. In the two complete SMN2 genes,
the reverse LC-modified primer is hybridized, having only one mismatch (see Fig. 1), whereas in the hybrid gene this primer
does not recognize the adenine (a) corresponding to the SMN1 gene (2 bp mismatch). The MLPA probes (black rectangles) are
completely hybridized. This event may explain the discordant result between LC (two SMN2 copies) vs. MLPA (three SMN2
copies) (further explanation in the text).
COMPARISON OF THREE METHODS TO QUANTIFY SMN2 GENES 7

of 745 unrelated patients and identification of four novel
mutations in the SMN1 gene. Hum Genet 125:29–39.
Anhuf D, Eggermann T, Rudnik-Schoneborn S, Zerres K (2003)
Determination of SMN1 and SMN2 copy number using Taq-
Man technology. Hum Mutat 22:74–78.
Arkblad EL, Darin N, Berg K, et al. (2006) Multiplex ligation-
dependent probe amplification improves diagnostics in spinal
muscular atrophy. Neuromuscul Disord 16:830–838.
Cusco I, Barcelo MJ, Baiget M, Tizzano EF (2002) Implementa-
tion of SMA carrier testing in genetic laboratories: comparison
of two methods for quantifying the SMN1 gene. Hum Mutat
20:452–459.
Cusco I, Barcelo MJ, del Rio E, et al. (2001) Characterisation of
SMN hybrid genes in Spanish SMA patients: de novo, homo-
zygous and compound heterozygous cases. Hum Genet
108:222–229.
Cusco I, Barcelo MJ, Rojas-Garcia R, et al. (2006) SMN2 copy
number predicts acute or chronic spinal muscular atrophy but
does not account for intrafamilial variability in siblings. J
Neurol 253:21–25.
DiDonato CJ, Morgan K, Carpten JD, et al. (1994) Association be-
tween Ag1-CA alleles and severity of autosomal recessive proxi-
mal spinal muscular atrophy. Am J Hum Genet 55:1218–1229.
Feldkotter M, Schwarzer V, Wirth R, et al. (2002) Quantitative an-
alyses of SMN1 and SMN2 based on real-time lightCycler PCR:
fast and highly reliable carrier testing and prediction of severity
of spinal muscular atrophy. Am J Hum Genet 70:358–368.
Huang CH, Chang YY, Chen CH, et al. (2007) Copy number
analysis of survival motor neuron genes by multiplex ligation-
dependent probe amplification. Genet Med 9:241–248.
Lefebvre S, Burglen L, Reboullet S, et al. (1995) Identification and
characterization of a spinal muscular atrophy-determining
gene. Cell 80;155–165.
Melki J, Lefebvre S, Burglen L, et al. (1994) De novo and inherited
deletions of the 5q13 region in spinal muscular atrophies.
Science 264:1474–1477.
Miller SA, Dykes DD, Polesky HF (1988) A simple salting out
procedure for extracting DNA from human nucleated cells.
Nucleic Acids Res 16:1215.
Su YN, Hung CC, Li H, et al. (2005) Quantitative analysis of
SMN1 and SMN2 genes based on DHPLC: a highly efficient
and reliable carrier-screening test. Hum Mutat 25:460–467.
Sumner CJ (2006) Therapeutics development for spinal muscular
atrophy. NeuroRx 3:235–245.
Tiziano FD, Bertini E, Messina S, et al. (2007) The Hammersmith
functional score correlates with the SMN2 copy number: a
multicentric study. Neuromuscul Disord 17:400–403.
Tomaszewicz K, Kang P, Wu BL (2005) Detection of homozy-
gous and heterozygous SMN deletions of spinal muscular
atrophy in a single assay with multiplex ligation-dependent
probe amplification. Beijing Da Xue Xue Bao 37:55–57.
Wirth B, Brichta L, Schrank B, et al. (2006) Mildly affected pa-
tients with spinal muscular atrophy are partially protected by
an increased SMN2 copy number. Hum Genet 119:422–428.
Wirth B, Herz M, Wetter A, et al. (1999) Quantitative analysis of
survival motor neuron copies: identification of subtle SMN1
mutations in patients with spinal muscular atrophy, genotype-
phenotype correlation, and implications for genetic counsel-
ing. Am J Hum Genet 64:1340–1356.
Address correspondence to:
Eduardo F. Tizzano, M.D., Ph.D.
Servicio de Gene
´tica
Hospital de la Santa Creu i Sant Pau
CIBERER (U705)
Sant Quintı
´89
08041 Barcelona
Spain
E-mail: etizzano@santpau.cat
8 ALI
´AS ET AL.