RESEARCH ARTICLE

Outfoxed by RBFOX1-A Caution About Ascertainment Bias Benjamin Kamien,1,2* Anath C. Lionel,3,4 Nicole Bain,5 Stephen W. Scherer,3,4 and Matthew Hunter1,2,6 1

Hunter Genetics, Newcastle, New South Wales, Australia

2

The University of Newcastle, School of Medicine and Public Health, Newcastle, New South Wales, Australia The Centre for Applied Genomics and Program in Genetics and Genome Biology, The Hospital for Sick Children, Toronto, Canada

3 4

Department of Molecular Genetics and McLaughlin Centre, University of Toronto, Toronto, Canada

5

Hunter Area Pathology Service, Department of Molecular Medicine, Newcastle, Australia Genetics of Learning Disability (GOLD) Service, Newcastle, New South Wales, Australia

6

Manuscript Received: 6 June 2013; Manuscript Accepted: 6 January 2014

We report on two patients with intragenic deletions of RBFOX1 and one patient with an intragenic duplication of RBFOX1. These patients, by report, all had autism spectrum disorder and/or developmental delay and had strong family histories of these conditions. We initially hypothesized that RBFOX1 was another susceptibility locus for autism spectrum disorder or developmental delay. However, epidemiological evidence examining large numbers of individuals did not support this hypothesis and the data presented here suggests that RBFOX1 intragenic copy number variants are not pathogenic. This contradicts previous reports that examined smaller numbers of patients and controls. Ó 2014 Wiley Periodicals, Inc.

Key words: RBFOX1; A2BP1; FOX1; genetic susceptibility; autism spectrum disorder; DNA copy number variation; ascertainment bias

INTRODUCTION More than 200 autism susceptibility genes have been identified to date and copy number variation (CNV) susceptibility loci can be identified by chromosomal microarrays [Heil and Schaaf, 2013]. That is, these CNVs increase the chance of a child developing intellectual disability, autism spectrum disorders, epilepsy, schizophrenia, attention deficit hyperactivity disorder (ADHD), or congenital anomalies, but may not be sufficient to cause a phenotype by themselves [Palmer et al., 2011]. A unifying feature of several identified susceptibility loci is localization of the involved protein at neural synapses [Cottrell et al., 2011]. RBFOX1 at 16p13.3 (also known as A2BP1 or FOX1) encodes a RNA binding protein that regulates pre-mRNA splicing in developmental genes expressed in muscle, heart, and neuronal cells. There are previous case reports implicating chromosomal translocations through RBFOX1 as possibly pathogenic, associated with combinations of autism, intellectual disability, seizures, and mild

Ó 2014 Wiley Periodicals, Inc.

How to Cite this Article: Kamien B, Lionel AC, Bain N, Scherer SW, Hunter M. 2014. Outfoxed by RBFOX1-A caution about ascertainment bias. Am J Med Genet Part A 164A:1411–1418.

dysmorphism [Bhalla et al., 2004; Martin et al., 2007]. Intragenic deletions of RBFOX1 have also been reported as having similar clinical effects [Mikhail et al., 2011; Davis et al., 2012]. In a cohort of 335 families (parent–child trios) with ADHD, three probands had intragenic CNVs (one duplication and two deletions) involving RBFOX1. Two of three parents with the same CNV were also reported as having features of ADHD and none of 2,026 healthy controls had CNVs involving RBFOX1 [Elia et al., 2010]. Exonic RBFOX1 CNVs have also been implicated as causing a susceptibility to epilepsy. Lal et al. [2013], found five of 1,408 (0.36%) exon-disrupting RBFOX1 deletions in patients with idiopathic generalized epilepsy, whereas none were observed in 2,256

Conflict of interest: none. Grant sponsor: University of Toronto McLaughlin Centre; Grant sponsor: NeuroDevNet; Grant sponsor: Genome Canada Ontario Genomics Institute; Grant sponsor: Canadian Institutes for Health Research (CIHR); Grant sponsor: Canadian Institute for Advanced Research; Grant sponsor: Canada Foundation for Innovation; Grant sponsor: Government of Ontario, Autism Speaks; Grant sponsor: Hospital for Sick Children Foundation.  Correspondence to: Benjamin Kamien, Hunter Genetics, PO Box 84, Waratah 2298, Australia. E-mail: [email protected] Article first published online in Wiley Online Library (wileyonlinelibrary.com): 24 March 2014 DOI 10.1002/ajmg.a.36458

1411

1412 controls. Three of 1,408 (0.21%) patients had intronic CNVs and two of 2,256 (0.09%) controls had intronic CNVs. Recently Zhao [2013], examined a mixed pediatric cohort who underwent aCGH analysis for a variety of developmental or congenital abnormalities and found that 14/2,124 (0.66%) patients had RBFOX1 CNVs. Five CNVs involved exons (0.24%) and nine involved only introns (0.42%). Despite discussing ascertainment bias, the conclusion reached was that these CNVs probably cause a susceptibility to neurodevelopmental and neuropsychiatric disorders. We report here three patients with RBFOX1 CNVs and obtain further epidemiological data to examine whether CNVs in RBFOX1 are clinically significant.

METHODS Three patients were seen in our clinical genetics unit where a RBFOX1 CNV was detected using a 60K ISCA oligonucleotide comparative genomic hybridization (CGH) array (BlueGnome, Cambridge, UK), analyzed with BlueMulti v2.6 software. A minimum of four consecutive oligonucleotides was used for a CNV call to be made and all CNVs were validated using FISH analysis or 15K custom design oligonucleotide array [Agilent, Santa Clara, CA]. The DECIPHER database was then examined for information about other patients with RBFOX1 CNVs [Firth et al., 2009]. An audit of high-resolution chromosomal microarray data from a cohort of 2,446 individuals with autism was also undertaken [Pinto et al., 2010] and unpublished data from the Autism Genome Project. Lastly, we performed a search using the Database of Genomic Variants (DGV) (accessed 30.11.2013), examining the RBFOX1 region from chr16:6069132 to 7763340 (GRCh37) [Iafrate et al., 2004].

CLINICAL REPORT Patient 1 is a female born at 40 weeks gestation after an uncomplicated pregnancy with no known exposures in utero and a birth weight of 3.60 kg (AGA). Slow developmental progress, particularly speech delay was noted at age 2 years. Weschler Preschool and

AMERICAN JOURNAL OF MEDICAL GENETICS PART A Primary Scale of Intelligence; third edition (WPPSI-3) assessment at age five showed a verbal IQ of 54 (0.1 centile), nonverbal/visual skills of 44 (

Outfoxed by RBFOX1-a caution about ascertainment bias.

We report on two patients with intragenic deletions of RBFOX1 and one patient with an intragenic duplication of RBFOX1. These patients, by report, all...
400KB Sizes 2 Downloads 3 Views