CLINICAL REPORT

Identification of SPRED1 Deletions using RT-PCR,Multiplex Ligation-dependent Probe Amplificationand Quantitative PCREmily Spencer,1 Julia Davis,1 Fady Mikhail,1 Chuanhua Fu,1 Raymon Vijzelaar,2 Elaine H Zackai,3

Holly Feret,3 M Stephen Meyn,4 Andrea Shugar,4 Gary Bellus,5 Kristina Kocsis,5 Sirpa Kivirikko,6

Minna P€oyh€onen,7,8 and Ludwine Messiaen1*1University of Alabama at Birmingham, Dept of Genetics, Birmingham, Alabama2MRC-Holland, Amsterdam, The Netherlands3Children’s Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, Division Human Genetics, Philadelphia,

Pennsylvania4Hospital for Sick Children, Division of Clinical and Metabolic Genetics, Toronto, Canada5The Children’s Hospital, University of Colorado, Department of Clinical Genetics and Metabolism, Denver-Aurora, Colorado6Helsinki University Central Hospital, Department of Clinical Genetics, Helsinki, Finland7University of Helsinki, Department of Medical Genetics, Helsinki, Finland8University of Helsinki, Department of Clinical Genetics, HUSLAB, Helsinki, Finland

Received 25 July 2010; Accepted 22 December 2010

Legius syndrome, is a recently identified autosomal dominant

disorder caused by loss of function mutations in the SPRED1

gene, with individuals mainly presenting with multiple caf�e-au-

lait macules (CALM), freckling and macrocephaly. So far, only

SPRED1 point mutations have been identified as the cause of this

syndrome. To determine if copy number changes (CNCs) are a

cause of Legius syndrome, we have used a Multiplex Ligation-

dependent Probe Amplification (MLPA) assay covering all

SPRED1 exons in a cohort of 510 NF1-negative patients present-

ing with multiple CALMs with or without freckling, but no other

NF1 diagnostic signs. Four different deletions were identified by

MLPA and confirmed by quantitative PCR, reverse transcriptase

PCR and/or array CGH: a deletion of exon 1 and the SPRED1

promoter region in a proband and two first-degree relatives; a

deletion of the entire SPRED1 gene in a sporadic patient; a

deletion of exon 2-6 in a proband and her father; and an�6.6 Mb

deletion on chromosome 15 that spans SPRED1 in a sporadic

patient. Deletions account for�10% of the 40 detected SPRED1

mutations in this cohort of 510 individuals. These results indi-

cate the need for dosage analysis to complement sequencing-

based SPRED1 mutation analyses. � 2011 Wiley-Liss, Inc.

Key words: Legius syndrome; MLPA; copy number change;

SPRED1; Neurofibromatosis Type 1; Ras pathway

INTRODUCTION

Recently, heterozygous SPRED1 mutations have been identified

in individuals/families with a pigmentary phenotype of multiple

caf�e-au-lait macules (CALM), axillary freckling and macrocephaly

[Brems et al., 2007; Denayer et al., 2010; Messiaen et al., 2009;

Muram-Zborovski et al., 2010; Pasmant et al., 2009; Spurlock et al.,

2009]. This disorder is currently referred to as Legius syndrome

[OMIM 611431]. About half of patients with an identified mutation

Additional supporting information may be found in the online version of

this article.

COMPETING INTERESTS: Raymon Vijzelaar is a scientist employed by

MRC-Holland, provider of commercially available MLPA assays.

*Correspondence to:

Ludwine Messiaen, University of Alabama at Birmingham, Dept of

Genetics, 1530 3rd Ave S, Kaul 330, Birmingham Alabama 35294-0024.

E-mail: lmessiaen@uab.edu

Published online 5 May 2011 in Wiley Online Library

(wileyonlinelibrary.com).

DOI 10.1002/ajmg.a.33894

How to Cite this Article:Spencer E, Davis J, Mikhail F, Fu C, Vijzelaar

R, Zackai E, Feret H, Meyn S, Shugar A,

Bellus G, Kocsis K, Kivirikko S, P€oyh€onen M,

Messiaen L. 2011. Identification of SPRED1

deletions using RT-PCR, multiplex ligation-

dependent probe amplification and

quantitative PCR.

Am J Med Genet Part A 155:1352–1359.

� 2011 Wiley-Liss, Inc. 1352

in SPRED1 fulfill the NF1 diagnostic criteria based on presence of

>5 CALMs with/without freckling and/or family history of consis-

tent pigmentary changes; however, none of these individuals de-

veloped other diagnostic criteria (Lisch nodules, neurofibromas,

osseous lesions, or optic pathway gliomas). In individuals with

CALMs with or without freckling, and no other specific distinguish-

ing features, a diagnosis of either NF1 or Legius syndrome cannot be

made solely based on clinical grounds. Although the diagnosis may

become apparent in most cases as more symptoms appear with

time, the diagnosis will remain uncertain for individuals who do not

develop other signs. Molecular genetic testing can resolve the

diagnosis in most such cases, and has important implications for

prognosis and counseling. We have developed an approach to

identify SPRED1 copy number changes (CNCs) and report four

different multi-exon deletions in four unrelated families. These

findings indicate the need for dosage analysis to complement

SPRED1 sequence analysis for genetic diagnosis.

SPRED1 (sprouty-related, EVH1 domain containing 1) consists

of seven exons, is located on chromosome 15q14, and encodes a

3,809 base transcript. The protein is 444 amino acids long. Spred1, a

member of the Sprouty protein family (Spred1,2,3 and Sprouty

1,2,3,4), acts as a negative regulator of the Ras/Raf interaction, and

modulates Mitogen Activated Protein Kinase (MAPK) signaling

[Miyoshi et al., 2004]. Spred proteins form a family containing an

N-terminal Enabled/VASP homology 1 domain (EVH1), a central

c-Kit binding domain (KBD) and a C-terminal cysteine rich

Sprouty related domain (SPR).

RNA-based analysis of SPRED1 starting from a single Reverse

Transcriptase (RT)-PCR fragment spanning the entire coding

region and used as the starting template for sequence analysis,

detects intragenic deletions as well as missense, splice, frameshift

and nonsense alterations, if nonsense-mediated RNA-decay is

prevented in the starting template [Andreutti-Zaugg et al., 1997;

Messiaen et al., 2000]. Previous RT-PCR analysis starting from

puromycin-treated short-term lymphocyte cultures of 450 NF1-

negative individuals with a broad range of NF1-related signs/

features did not identify any SPRED1 copy number changes (CNCs)

[Messiaen et al., 2009] nor have other groups described copy

number changes in SPRED1 testing [Brems et al., 2007; Denayer

et al., 2010; Muram-Zborovski et al., 2010; Pasmant et al., 2009;

Spurlock et al., 2009]. However, deletions extending beyond the 50

and/or 30 borders of the RT-PCR primers are refractory to detec-

tion, since the deleted mutant allele will not be captured by the RT-

PCR. Hence, additional diagnostic approaches for detecting CNCs

that extend beyond the gene borders were developed and are

reported herein.

MATERIALS AND METHODS

PatientsBetween October 2008 and June 2010, 510 patients with >5

CALMs, with or without freckling and no other NF1-related signs,

in whom no NF1 mutation was previously identified after compre-

hensive analysis in cultured blood lymphocytes, were referred to the

Medical Genomics Laboratory for clinical SPRED1 testing. Com-

prehensive SPRED1 analysis started with direct sequencing of the

entire coding region using an RNA-based approach starting from

puromycin treated short-term lymphocyte cultures, as previously

described [Messiaen et al., 2009]. These studies were complemented

with additional dosage analysis approaches (MLPA, qPCR, aCGH

as described below) to detect deletions extending beyond the

borders of the RT-PCR product. Mutations are described according

to the recommendations of the Human Genome Variation Society

(www.hgvs.org/mutnomen/) using sequence NM_152594.2.

Phenotypic information was recorded using a standardized

phenotypic dataform (http://www.genetics.uab.edu/medgenom-

ics/documents/SPRED1Reqform.pdf) completed by the referring

physicians. Additional clinical information was obtained after

informed consent. The study was approved by the UAB human

subjects IRB.

Multiplex Ligation-Dependent Probe Amplification(MLPA)MLPA analysis to detect copy number changes was performed using

the SALSA MLPA probe mix for SPRED1 (P295, MRC-Holland,

Amsterdam, the Netherlands), consisting of 23 target probes

(Supplemental eTable I online) in addition to 15 reference probes

hybridizing to other human genes located on other chromosomes

in the genome and 9 assay quality control probes. Samples were

prepared and analyzed according to manufacturers’ directions.

PCR products were separated on an automated ABI PRISM�

3130 xl Genetic Analyzer. GeneMarker� v1.85 software

(SoftGenetics, State College, Pennsylvania, USA) was used for data

analysis. Relative amounts of probe-amplified products were com-

pared with reference sample to determine the copy number of the

target sequences. Values under a threshold of 0.7 and over a

threshold of 1.3 for multiple adjacent probes, indicate the presence

of a deletion or duplication respectively [Schouten et al., 2002; Van

Opstal et al., 2009].

Quantitative Real-time PCR (qPCR)qPCR analysis was performed on all coding exons with one primer

pair for each exon, except for the large 651 bp exon 7 which was

analyzed by three primer pairs spread across the exon. Additionally,

exon 1 of FAM98B was analyzed in the proband of Family 2 in order

to delineate the size of the deletion. Primers were designed with

Beacon Designer (Premier Biosoft, Palo Alto, CA) and evaluated for

mis-priming and SNP presence using NCBI BLAST (http://

blast.ncbi.nlm.nih.gov/Blast.cgi) and SNPCheck (http://ngrl.ma-

n.ac.uk/SNPCheck/SNPCheck.html). Primer sequences and effi-

ciencies are given in Table I. A melting curve was generated for every

PCR amplicon to check the specificity of the PCR reaction (absence

of primer–dimers or other nonspecific amplification products).

Additionally, fragments of TBP and B2M were amplified as an

internal control in each reaction. SPRED1 qPCR was performed

using the LightCycler 480 (Roche Diagnostics, Rotkreuz,

Switzerland) and the DNA-binding dye SYBR Green (Roche Diag-

nostics AG, Mannheim, Germany) with cycling conditions: 100 at

95�C, 40 cycles at 95�C for 1500 and 58�C for 6000. Analysis of

FAM98B was performed using the CFX384 Real-Time PCR Detec-

tion System (Bio-Rad Laboratories, Hercules, CA) and the DNA-

binding dye SYBR Green (Bio-Rad Laboratories, Hercules, CA)

SPENCER ET AL. 1353

with cycling conditions: 30 at 95�C, 40 cycles at 95�C for 1000 and

58�C for 3000. Each experiment was performed in triplicate with

10 ng of DNA per reaction in a total volume of 10 ml. Gene copy

number was calculated using the comparative (D-Ct) Ct method

and error propagation rules [Hoebeeck et al., 2005]. The relative

dosage ratio is calculated from the real-time PCR efficiencies and

the crossing point deviation of an unknown sample versus a

control/calibrator. Geometric averaging of multiple internal con-

trol genes allows accurate determination of the relative quantity of

experimental samples versus control.

Array Comparative Genomic Hybridization(Array CGH)The sizes of the deletions were further delineated using a custom

designed whole-genome coverage 4x44k oligo-array (Agilent Tech-

nologies, Santa Clara, CA), which is based on the ISCA

(International Standard Cytogenomic Array) Consortium design.

DNA labeling, slide hybridization, washing, and scanning were

performed following the manufacturer’s protocol. The arrays were

scanned using the GenePix 4000B scanner (Molecular Devices,

Sunnyvale, CA). The scanned arrays were analyzed using the

‘Feature Extraction’ and ‘DNA Analytics’ software (Agilent

Technologies).

Definition of the Breakpoint of the SPRED1Exon 2-6 DeletionRT-PCR was performed as previously reported [Messiaen et al.,

2009]. To define break points of the deletion involving the exons

2-6, a series of forward primers (Supplemental eTable II online)

were designed, located every 2-3 kb and covering the 31.6 kb-region

between MLPA probe 13048-L14321, not deleted in the proband

and MLPA probe 13051-L14234, deleted in the proband. The

reverse primer was located after the first MLPA probe in exon 7,

not deleted. As the minimal size difference between the deleted and

the normal allele is 50 kb, amplification of the shorter allele carrying

the deletion will be favored. Using primers SP1delF8 and SP1delR1,

a 2.4 kb fragment was obtained in the patient, but not in the normal

controls. Further analysis of the amplicon by sequence analysis

using nested primers (Supplemental eTable II online) was used

thereafter to define the deletion breakpoint at the nucleotide level.

RESULTS

Using MLPA and RT-PCR on 510 NF1-negative patients, we found

four different heterozygous deletions: a deletion of the promoter

region and exon 1 [Chr15:g.(36169075_36253662)_(36365367_

36372927)del (NCBI Build 36.1)], a deletion of the entire SPRED1

gene and flanking genes [Chr15:g.(35579596_35632785)_

TABLE I. qPCR Primers, Fragment Lengths, and Efficiency for qPCR Analysis of SPRED1 and Control Genes (ATP5B, TBP, and B2M)

Primer name Primer Length (nt) Sequence Product Length (nt) Average PCR EfficiencySPRED1Ex1_Fwd 23 TGTTGCTCCTCCATCTCCAGATC 81 2.00SPRED1Ex1_Rev 23 CGCTTACTCGTTGTCAGAAGTCGSPRED1Ex2_Fwd 24 TGACTCAAGTGGTGGATGGTTACC 91 1.97SPRED1Ex2_Rev 24 ACAGCCATTCTCTTCCTGATGAGGSPRED1Ex3_Fwd 23 TGGCTTTTGTCAGGTGGTTTTGG 82 1.92SPRED1Ex3_Rev 24 CCAGTGGTGAAATGTTGGAGTGACSPRED1Ex4_Fwd 22 TACCTTAATTGCCAGGCAGTCC 149 1.92SPRED1Ex4_Rev 24 ATTCGGGGCATCCTAAGAAAATAGSPRED1Ex5_Fwd 24 AGCAAGAGACAGTTGTTACCAGTG 94 1.98SPRED1Ex5_Rev 24 TGCATGTAGACTCTTCTGGCATTCSPRED1Ex6_Fwd 24 AGGCTTGGACATTCAGAGCAGAAG 78 1.97SPRED1Ex6_Rev 24 TGGGACTTTAGGCTTCCACATTCCSPRED1Ex7.1_Fwd 24 GTCGCTATGCAGACTACAGACATC 134 1.95SPRED1Ex7.1_Rev 24 AGTCTCATCCCCACAAGAGTACAGSPRED1Ex7.2_Fwd 24 GGTATTTAAGACGCAGCCTTCCTC 88 1.94SPRED1Ex7.2_Rev 21 GCAGTATACGCAGCGAGAACGSPRED1Ex7.3_Fwd 24 ACACTAGCGACGACAAGTTCTGCT 109 1.98SPRED1Ex7.3_Rev 24 CCACAGCGATGGCACATTCTCAAAFAM98Ex1 Fwd 18 CCGGACCGGAACCTTGGC 130 1.94FAM98EX1 Rex 20 CAGCACGTCTCCCTCCATCGGAPDH_Fwd 18 GGGAAACTGTGGCGTGAT 111 1.98GAPDH_Rev 19 GCTGAACGGGAAGCTCACTTBP_Fwd 19 CAGCCGTTCAGCAGTCAAC 74 1.97TBP_Rev 20 TGAGTGGAAGAGCTGTGGTGB2M_Fwd 24 AGCAAGGACTGGTCTTTCTATCTC 82 2.00B2M_Rev 20 GGTTCACACGGCAGGCATAC

1354 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

(36511058_36573378)del (NCBI Build 36.1)], a deletion of exons 2

to 6, and a large �6.6 Mb deletion on chromosome 15 that spans

SPRED1 [arr 15q13.3q14(30,965,692-37,586,958)x1] (NCBI Build

36.1). All four SPRED1 deletions presented with normalized values

of�0.5 and were found in a cohort of 510 unrelated NF1-negative

probands with a phenotype compatible with Legius syndrome.

In addition, a SPRED1 point mutation or 1–4 base insertion/

duplication/deletion was identified in 36/510 individuals

(Supplemental eTable III online), which allows estimating the

frequency of SPRED1 deletions to be �10% of all mutations

detected.

The proband of Family 1, a 4-year-old girl, was initially referred

to a clinical geneticist at 14 months old for evaluation of multiple

CALMs. At the age of 4 years, she has 6 CALMs over 1 cm in

diameter, several smaller CALMs, posteriorly rotated ears, and

ptosis of the left eye (Fig. 1a). She has no freckling, neurofibromas,

Lisch nodules or other abnormalities (including T2-

hyperintensities) on brain MRI. Speech and motor development

are normal. Her 7-year-old brother has 10 CALMs over 1 cm, flat

feet, no freckling, no neurofibromas, no Lisch nodules, a normal

brain MRI (without T2-hyperintensities), and normal develop-

ment with mild learning difficulties. The 41-year old father has 6

CALMs larger than 1.5 cm, a few xanthelasmas under the eyelids as

clinically deduced by the ophthalmologist and clinical geneticist, no

freckling, no neurofibromas, and a normal brain MRI (without T2-

hyperintensities). He has no Lisch nodules but a few patchy areas of

iris pigmentation were observed in addition to a few large deep iris

crypts (Fig. 1c), similar to findings reported by Muram-Zborovski

FIG. 1. (a) Proband of Family 1 face and profile showing ptosis of left eye and posteriorly rotated ears. (b) Brother of proband in Family 1 with multiple

CALM on back. (c) Father of proband, eye, arrow indicates area of iris hyperpigmentation, arrowhead indicates a large deep iris crypt.

SPENCER ET AL. 1355

et al. [2010]. A SPRED1 deletion of the first exon and part of the

promoter region was found in the proband, her brother and father

using MLPA. Confirmatory testing by qPCR and array CGH

indicated the size of the deletion to be 111–203.8 kb, without

involvement of the closest neighboring gene, TMCO5A (Figs 2a,

3a, and Supplemental eFigure 1a online).

In the second family, a SPRED1 total gene deletion was identified

in a 9-year-old boy with no family history and >6 typical well-

defined CALM and left axillary freckling, no Lisch nodules, no

macrocephaly, a pectus excavatum, normal development, and no

other NF1-related signs. Parents were not available for further

testing. Confirmatory testing by qPCR and array CGH indicated

the deletion to encompass 878.3–993.8 kb, including the TMCO5A

and FAM98B genes (Figs 2b, 3a and b, and Supplemental eFigure 1b

online).

The proband of the third family carries an intragenic deletion of

SPRED1 exons 2-6 identified through RT-PCR, which was con-

firmed by MLPA and cloning of the breakpoint (Supplemental

eFigures 2 Figures 2a-c online). The deletion is 71 kb in size: c.33-

20604_684þ401delinsGAAA. In the immediate vicinity of the

breakpoints there is a 6-bp sequence TTTAAA, which is known

to be able to induce a curvature in the DNA molecule [Singh et al.,

1997], and predispose it to recombination. Across the breakpoint,

there is a 4 bp insertion, GAAA, indicating breakage might have

been mediated through Non-Homologous End Joining (NHEJ)

[Vissers et al., 2009]. The proband, a 7.5-year-old girl with >6

typical well-defined CALMs and bilateral axillary and inguinal

freckling, was referred for a possible Noonan syndrome phenotype.

She is being pharmacologically treated for ADHD. Her 4.5-year-old

sister has no signs of Legius syndrome and does not carry the

deletion. Her 40-year-old father carries the same deletion and has

>6 CALMs and approximately 20 lipomas on his legs, abdomen,

and the back of his neck. He is developmentally normal but reports

many signs of ADHD that have persisted into adulthood; however,

no formal diagnosis has been made. His history is also notable for

high serum cholesterol and hypertension in addition to insulin-

dependent diabetes.

The 2-year-old female proband in Family 4 presented with 8-10

CALMs of typical appearance with well defined smooth borders

(including one large one on her mid-back), along with speech delay,

mild developmental delay, an isolated cleft palate, and no family

history of NF1. She has no obvious dysmorphic features aside from

the cleft palate. At her most recent follow-up at 28 months she

weighed 10.8 kg (5–10th centile), was 81.5 cm tall (<5th centile),

and had a head circumference of 45.3 cm (<5th centile). Both of her

parents also have short stature (father is 160 cm, mother is

152.4 cm). MLPA identified a total SPRED1 gene deletion

(Supplemental eFigure 1c online) in the proband, which was

confirmed by array CGH to be a large� 6.6 Mb deletion that spans

29 annotated reference genes in addition to SPRED1 [arr

15q13.3q14(30,965,692-37,586,958)x1] (Figs2c and 3b). Further

analysis of both parents indicated the deletion had occurred de novo.

DISCUSSION

This is the first report to present SPRED1 copy number alterations

and the associated phenotype in seven individuals carrying 4

different deletions. All deletions had different breakpoints. None

of these deletions were flanked by segmental duplications, and

therefore are the result of another molecular mechanism distinct

from non-allelic homologous recombination (NAHR). The 4

SPRED1 deletions were found in a cohort of 510 unrelated NF1-

negative probands with a phenotype compatible with Legius syn-

drome. In addition, other intragenic SPRED1 mutations were

identified in 36/510 patients: 22/36 were pathogenic Loss-Of-

Function mutations and 14/36 were missense mutations: two

previously described likely benign and 12 novel not previously

described [Brems et al., 2007; Denayer et al., 2010; Messiaen et al.,

2009; Muram-Zborovski et al., 2010; Pasmant et al., 2009; Spurlock

et al., 2009] and additional family and functional studies to assess

their pathogenicity, are needed as described [Messiaen et al., 2009].

These data demonstrate the frequency of SPRED1 deletions to be

�10% of detectable mutations in this gene.

In this study, all results obtained with MLPA were confirmed

using other methods (RT-PCR, qPCR and array CGH). SPRED1-

deletion positive individuals in this study presented with multiple

CAL-macules with or without freckling (7/7), pectus excavatum (1/

7), multiple lipomas (1/2 adults), mild learning disabilities (2/7)

and ADHD (2/7), which are consistent clinical manifestations

documented in previous reports [Brems et al., 2007; Messiaen

et al., 2009]. All previously reported patients with Legius syndrome

had point mutations that were identified through direct sequencing

of the entire coding region [Brems et al., 2007; Denayer et al., 2010;

Messiaen et al., 2009; Muram-Zborovski et al., 2010; Pasmant et al.,

2009; Spurlock et al., 2009]. The current study identifies the first

seven individuals from four unrelated families carrying multi-exon

to whole-gene SPRED1 deletions. It is encouraging that the

phenotype in these individuals also primarily consists of CALM,

intertriginous freckling and learning disorders and does not

differ compared to patients with point mutations, even though

co-deletion of two other genes (TMCO5A and FAM98B) was

present in the proband of Family 2.

However the proband of Family 4 presented with isolated cleft

palate in addition to the SPRED1 phenotype. Published accounts of

large deletions in this region (without the inclusion of SPRED1 in

the deleted region, or with unspecific deletion coordinates

provided) detail cleft lip/palate and developmental delay, but do

not specifically mention the presence of CALMs [Calounova et al.,

2008; Chen et al., 2008; Erdogan et al., 2007; Natrajan et al., 2003;

Schwartz et al., 1985]. Evaluation of SPRED1 in patients with large

deletions of band q14 on chromosome 15 and cleft palate may help

explain pigmentary findings in these patients. Developmental delay

has been reported in patients with large 15q14 deletions, both

including [Erdogan et al., 2007] and excluding SPRED1 [Chen et al.,

2008], as well as in patients with deletions and point mutations

limited to SPRED1. This suggests that while SPRED1 may play a role

in the etiology of the incidence of developmental delay previously

described in 15q14 deletions, it is unlikely to be the sole cause.

This report indicates the need for a diagnostic technique to detect

copy number changes to complement SPRED1 sequence analysis.

Various techniques can detect copy number changes, but MLPA has

been widely used as a fast and inexpensive screening method in

numerous disorders [Sellner and Taylor, 2004]. There are two

common silent polymorphisms in the SPRED1 coding region

1356 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

FIG. 2. Oligo-array CGH plots aligned with a genomic map of the same region generated using the UCSC genome browser (NCBI36/hg18), and showing

the minimally deleted regions (dashed line), genomic ruler, ISCA 4x44k array probes, cytogenetic band, and annotated RefSeq genes. Repetitive DNA

sequence blocks are not evident at the deletion breakpoints. a: Note the deletion of 3 probes in SPRED1 exon 1 and promoter region in proband of

Family 1; b: Note the deletion of all probes in the SPRED1 region in addition to deletion of TMCO5A in the proband of Family 2. FAM98B was also shown

to be deleted in this individual by qPCR (see Fig. 3b); c: Note the �6.6 Mb deletion at 15q13.3q14 in the proband of Family 4.

SPENCER ET AL. 1357

rs7182445 (c.291G>A; p.Lys97Lys) and rs3751526 (c.1044T>C;

p.Val348Val) that, if present, would rule out a deletion in sequenced

cDNA. However, when using an exon by exon DNA-sequencing

approach, heterozygosity at coding or non-coding SNPs only

indicates the lack of a whole gene deletion, whereas further testing

using MLPA or other copy number analysis method is mandatory

to detect single- or multi-exon deletions.

ACKNOWLEDGMENTS

We would like to thank the patients and their families for their

support and participation. We would also like to thank the Wilder-

man family for their donation to further research for children with

caf�e-au-lait spots-only.

REFERENCES

Andreutti-Zaugg C, Scott RJ, Iggo R. 1997. Inhibition of nonsense-medi-ated messenger RNA decay in clinical samples facilitates detection ofhuman MSH2 mutations with an in vivo fusion protein assay andconventional techniques. Cancer Res 57:3288–3293.

Brems H, Chmara M, Sahbatou M, Denayer E, Taniguchi K, Kato R, SomersR, Messiaen L, De Schepper S, Fryns J-P, Cools J, Marynen P, Thomas G,Yoshimura A, Legius E. 2007. Germline loss-of-function mutations inSPRED1 cause a neurofibromatosis 1-like phenotype. Nat Genet 39:1120–1126.

Calounova G, Hedvicakova P, Silhanova E, Kreckova G, Sedlacek Z. 2008.Molecular and clinical characterization of two patients with Prader-Willisyndrome and atypical deletions of proximal chromosome 15q. Am JMed Genet 146A:1955–1962.

Chen CP, Lin SP, Tsai FJ, Chern SR, Lee CC, Wang W. 2008. A 5.6-Mbdeletion in 15q14 in a boy with speech and language disorder, cleft palate,epilepsy, a ventricular septal defect, mental retardation and developmen-tal delay. Eur J Med Genet 51:368–372.

Denayer E, Chmara M, Brems H, Kievit AM, van Bever Y, Van denOuweland AM, Van Minkelen R, de Goede-Bolder A, Oostenbrink R,Lakeman P, Beert E, Ishizaki T, Mori T, Keymolen K, Van den Ende J,Mangold E, Peltonen S, Brice G, Rankin J, Van Spaendonck-Zwarts KY,Yoshimura A, Legius E. 2010. Legius syndrome in fourteen families. HumMutat 32:E1985–1998.

Erdogan F, Ullmann R, Chen W, Schubert M, Adolph S, Hultschig C,Kalscheuer V, Ropers HH, Spaich C, Tzschach A. 2007. Characterizationof a 5.3 Mb deletion in 15q14 by comparative genomic hybridizationusing a whole genome ‘‘tiling path’’ BAC array in a girl with heart defect,cleft palate, and developmental delay. Am J Med Genet A 143:172–178.

Hoebeeck J, van der Luijt R, Poppe B, De Smet E, Yigit N, Claes K, Zewald R,de Jong GJ, De Paepe A, Speleman F, Vandesompele J. 2005. Rapiddetection of VHL exon deletions using real-time quantitative PCR. LabInvest 85:24–33.

Messiaen L, Yao S, Brems H, Callens T, Sathienkijkanchai A, Denayer E,Spencer E, Arn P, Babovic-Vuksanovic D, Bay C, Bobele G, Cohen BH,Escobar L, Eunpu D, Grebe T, Greenstein R, Hachen R, Irons M, KronnD, Lemire E, Leppig K, Lim C, McDonald M, Narayanan V, Pearn A,Pedersen R, Powell B, Shapiro LR, Skidmore D, Tegay D, Thiese H, ZackaiEH, Vijzelaar R, Taniguchi K, Ayada T, Okamoto F, Yoshimura A, ParretA, Korf B, Legius E. 2009. Clinical and Mutational Spectrum of Neuro-fibromatosis Type 1-like Syndrome. Jama 302:2111–2118.

Messiaen LM, Callens T, Mortier G, Beysen D, Vandenbroucke I, Van RoyN, Speleman F, Paepe AD. 2000. Exhaustive mutation analysis of the NF1gene allows identification of 95% of mutations and reveals a highfrequency of unusual splicing defects. Hum Mutat 15:541–555.

Miyoshi K, Wakioka T, Nishinakamura H, Kamio M, Yang L, Inoue M,Hasegawa M, Yonemitsu Y, Komiya S, Yoshimura A. 2004. The Sprouty-related protein, Spred, inhibits cell motility, metastasis, and Rho-medi-ated actin reorganization. Oncogene 23:5567–5576.

Muram-Zborovski TM, Stevenson DA, Viskochil DH, Dries DC, WilsonAR, Mao R. 2010. SPRED1 Mutations in a Neurofibromatosis Clinic. JChild Neurol 25:1203–1209.

Natrajan R, Louhelainen J, Williams S, Laye J, Knowles MA. 2003. High-resolution deletion mapping of 15q13.2-q21.1 in transitional cell carci-noma of the bladder. Cancer Res 63:7657–7662.

Pasmant E, Sabbagh A, Hanna N, Masliah-Planchon J, Jolly E, Goussard P,Ballerini P, Cartault F, Barbarot S, Landman-Parker J, Soufir N, Parfait B,Vidaud M, Wolkenstein P, Vidaud D. 2009. SPRED1 germline mutationscaused a neurofibromatosis type 1 overlapping phenotype. J Med Genet46:425–430.

Schouten JP, McElgunn CJ, Waaijer R, Zwijnenburg D, Diepvens F, Pals G.2002. Relative quantification of 40 nucleic acid sequences by multiplexligation-dependent probe amplification. Nucleic Acids Res 30:e57.

Schwartz S, Max SR, Panny SR, Cohen MM. 1985. Deletions of proximal15q and non-classical Prader-Willi syndrome phenotypes. Am J MedGenet 20:255–263.

Sellner LN, Taylor GR. 2004. MLPA and MAPH: new techniques fordetection of gene deletions. Hum Mutat 23:413–419.

Singh GB, Kramer JA, Krawetz SA. 1997. Mathematical model to predictregions of chromatin attachment to the nuclear matrix. Nucleic Acids Res25:1419–1425.

Spurlock G, Bennett E, Chuzhanova N, Thomas N, Jim H, Side L, Davies S,Haan E, Kerr B, Huson SM, Upadhyaya M. 2009. SPRED1 mutations

FIG. 3. (a) qPCR shows deletion of exon 1 in the proband of Family 1

(pt 1) and total gene deletion in the proband of Family 2 (pt 2)

when compared to a normal control (Calib). (b) qPCR shows

deletion of exon 1 of FAM98B in the proband of Family 2 when

compared to normal control (Calib) and the proband of Family 4

(Pos Con). [Color figure can be viewed in the online issue, which is

available at wileyonlinelibrary.com]

1358 AMERICAN JOURNAL OF MEDICAL GENETICS PART A

(Legius syndrome): another clinically useful genotype for dissecting theNF1 phenotype. J Med Genet 46:431–437.

Van Opstal D, Boter M, de Jong D, van den Berg C, Bruggenwirth HT,Wildschut HI, de Klein A, Galjaard RJ. 2009. Rapid aneuploidy detectionwith multiplex ligation-dependent probe amplification: a prospectivestudy of 4000 amniotic fluid samples. Eur J Hum Genet 17:112–121.

Vissers LE, Bhatt SS, Janssen IM, Xia Z, Lalani SR, Pfundt R, Derwinska K,de Vries BB, Gilissen C, Hoischen A, Nesteruk M, Wisniowiecka-Kowalnik B, Smyk M, Brunner HG, Cheung SW, van Kessel AG, VeltmanJA, Stankiewicz P. 2009. Rare pathogenic microdeletions and tandemduplications are microhomology-mediated and stimulated by localgenomic architecture. Hum Mol Genet 18:3579–3593.

SPENCER ET AL. 1359