Mutation of SALL2 Causes Recessive Ocular Coloboma in Humans and Mice

Human Molecular Genetics, 2014, Vol. 23, No.
10 2511–2526
doi:10.1093/hmg/ddt643
Advance Access published on January 9, 2014
Mutation of SALL2 causes recessive ocular

coloboma in humans and mice
Daniel Kelberman1,2, Lily Islam1,2,3,{, Jörn Lakowski2, Chiara Bacchelli4, Estelle Chanudet4,
Francesco Lescai4, Aara Patel2, Elia Stupka4,{, Anja Buck6, Stephan Wolf6, Philip L. Beales4,
Thomas S. Jacques5,7, Maria Bitner-Glindzicz3, Alki Liasis1,8, Ordan J. Lehmann9,
Jürgen Kohlhase10, Ken K. Nischal1,2,8,11,∗ and Jane C. Sowden1,2,∗
Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

1
Ulverscroft Vision Research Group, 2Developmental Biology Unit, Birth Defects Research Centre, 3Clinical and
Molecular Genetics Unit, 4Centre for Translational Genomics – GOSgene and 5Neural Development Unit, UCL Institute
of Child Health, 30 Guilford Street, London WC1N 1EH, UK, 6Institute for Human Genetics, University of Goettingen,
Germany, 7Department of Histopathology and 8Clinical and Academic Department of Ophthalmology, Great Ormond
Street Hospital for Children NHS Foundation Trust, London WC1N 3JH, UK, 9Department of Ophthalmology
and Medical Genetics, University of Alberta, Edmonton, Canada T6G 2H7, 10Center for Human Genetics Freiburg,
Heinrich-von-Stephan-Str. 5, 79100 Freiburg, Germany and 11UPMC Childrens Hospital of Pittsburgh and Eye Center,
4401 Penn Avenue, Pittsburgh, PA 15224, USA
Received October 3, 2013; Revised November 29, 2013; Accepted December 17, 2013
Ocular coloboma is a congenital defect resulting from failure of normal closure of the optic fissure during
embryonic eye development. This birth defect causes childhood blindness worldwide, yet the genetic etiology
is poorly understood. Here, we identified a novel homozygous mutation in the SALL2 gene in members of a
consanguineous family affected with non-syndromic ocular coloboma variably affecting the iris and retina.
This mutation, c.85G>T, introduces a premature termination codon (p.Glu29 ∗ ) predicted to truncate the
SALL2 protein so that it lacks three clusters of zinc-finger motifs that are essential for DNA-binding activity.
This discovery identifies SALL2 as the third member of the Drosophila homeotic Spalt-like family of develop-
mental transcription factor genes implicated in human disease. SALL2 is expressed in the developing human
retina at the time of, and subsequent to, optic fissure closure. Analysis of Sall2-deficient mouse embryos
revealed delayed apposition of the optic fissure margins and the persistence of an anterior retinal coloboma
phenotype after birth. Sall2-deficient embryos displayed correct posterior closure toward the optic nerve
head, and upon contact of the fissure margins, dissolution of the basal lamina occurred and PAX2, known
to be critical for this process, was expressed normally. Anterior closure was disrupted with the fissure margins
failing to meet, or in some cases misaligning leading to a retinal lesion. These observations demonstrate,
for the first time, a role for SALL2 in eye morphogenesis and that loss of function of the gene causes ocular
coloboma in humans and mice.
∗
To whom correspondence should be addressed at: Developmental Biology Unit, Birth Defects Research Centre, UCL Institute of Child Health, 30 Guilford
Street, London WC1N 1EH, UK. Tel: +44 2079052121; Fax: +44 2079052953; Email: j.sowden@ucl.ac.uk (J.C.S.); Division of Pediatric Ophthalmol-
ogy, UPMC Children’s Hospital of Pittsburgh and Eye Center, 4401 Penn Avenue, Floor 3, Pittsburgh, PA 15224, USA. Tel: +1 4126928940;
Fax: +1 4126827220; Email: nischalkk@upmc.edu (K.K.N.)
†
Present address: West Midlands Regional Genetics Service, Birmingham Women’s Hospital NHS Trust, Mindelsohn Way, Edgbaston, Birmingham B15
2TG, UK.
‡
Present address: Center for Translational Genomics and Bioinformatics, San Raffaele Scientific Institute, Milan, Italy.
# The Author 2014. Published by Oxford University Press.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/
licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is
properly cited.
2512 Human Molecular Genetics, 2014, Vol. 23, No. 10
INTRODUCTION 600725) (18), RAX (OMIM 601881) (19), GDF3 (OMIM

606522) and GDF6 (OMIM 601147) (13,20) and recessive muta-
Coloboma is an ocular birth defect resulting from abnormal tions in STRA6 (OMIM 610745) (21) and SMOC1 (OMIM
development of the eye during embryogenesis. It represents an im- 608488) (22). Additionally, rare cases of non-syndromic coloboma
portant cause of congenital blindness and visual impairment, esti- have also been identified in patients with recessive mutations of
mated to account for 3–11% of blindness in children worldwide VSX2 (formerly CHX10; OMIM 142993) (23) and ABCB6
(1). The prevalence varies by population ranging from 4 to 19 (OMIM 605452) (24) more frequently associated with micro-
per 100 000 live births across Europe (2–6) with higher rates phthalmia, dominant mutations of PAX6 (OMIM 607108) general-
reported in populations with high degrees of consanguinity (6,7). ly associated with a range of ocular defects including aniridia, and
Coloboma is defined as a congenital defect in any ocular MAF (OMIM 610210) (cataract and anterior segment dysgenesis)
tissue(s), typically presenting as absent tissue or a gap, at a site con- (23–26). The most commonly identified genetic causes of isolated
sistent with aberrant closure of the optic fissure (1). The embryonic coloboma, without microphthalmia, are CHD7 (OMIM 608892)
optic fissure is a transient ventral opening that arises during inva- mutations associated with CHARGE syndrome (OMIM 214800)

gination of the optic vesicle in the formation of the bilayered and PAX2 mutations, which cause renal-coloboma syndrome
optic cup. It permits the migration of periocular mesenchymal (OMIM 167409) (12,27,28). However, in the majority of cases,
cells (mostly of neural crest origin) into the developing eye to the genetic contribution to ocular coloboma phenotypes remains
form the hyaloid artery (1). During subsequent growth of the to be determined (9,13,14,16,20). The findings to date indicate sig-
optic cup, the edges of the fissure align and fuse completing forma- nificant genetic heterogeneity and suggest perturbation at multiple
tion of the eye globe so that the ventral and dorsal retina become stages of eye development can result in failure of optic fissure
morphologically indistinguishable. In human development, the closure. A comprehensive understanding of pathogenic genetic
fissure narrows and begins to close during the 5th week, starting changes is required to resolve the molecular etiology of coloboma.
centrally and progressing anteriorly and posteriorly, and is normal- Here, we report the identification of a novel homozygous
ly completely fused by the end of the 7th week (1,8). Failure of mutation in the SALL2 gene using a combined strategy of
fusion can lead to coloboma of one or multiple regions of the infer- homozygosity mapping and exome sequencing in three siblings
ior portion of the eye affecting any part of the globe traversed by the from a consanguineous family with an iris and retinochoroidal
fissure, from the iris to the optic nerve including the ciliary body, coloboma phenotype. This mutation introduces a premature
retina and choroid (9). Coloboma is also frequently associated stop codon predicted to result in loss of function of the mutant
with small (microphthalmic) or absent (anophthalmic) eyes (2,5) protein. RNA in situ hybridization of human embryonic eyes
as part of an interrelated spectrum of developmental eye anomal- revealed expression of SALL2 transcripts within the developing
ies, and can affect either one or both eyes. retina at the time of, and subsequent to, closure of the optic
The molecular pathways controlling the events involved in fissure (between 5 and 7 weeks of development). Furthermore,
closure of the optic fissure are largely unknown. Current knowl- we show that mice with homozygous loss of the orthologous
edge of the mechanisms of tissue fusion throughout development Sall2 gene exhibit an abnormal closure process and a variable
are limited and are of considerable interest as failure of tissue ocular coloboma phenotype, implicating a conserved role for
fusion also accounts for other common human birth defects such the gene in eye morphogenesis, and specifically optic fissure
as cleft palate and neural tube defects (10,11). Clinical, closure, in humans and mice.
epidemiological and experimental evidence from animal models
indicate a strong genetic basis. Recurrence risks for siblings of
patients affected with optic fissure closure defects have been sug- RESULTS
gested to be between 8 and 13% (3). However, a recent UK study
Patient phenotypes
estimated that 40% of coloboma cases were likely to have a
genetic cause based on positive family history or suspected familial Three siblings born of first-cousin parents of Kuwaiti origin with a
syndrome (12). Genes associated with coloboma phenotypes have variable non-syndromic coloboma phenotype (see Table 1) were
been identified in at least 20 syndromes, where the coloboma arises referred to a pediatric ophthalmologist (K.K.N.) for evaluation
as an occasional feature of a complex multisystem birth defect of poor vision. The eldest affected child (patient IV:3, Fig. 1A)
(9,13–17); these include dominant mutations in SHH (OMIM was 13 years old on first examination and presented with visual
Table 1. Summary of ocular phenotype of affected individuals
Patient IV:1 IV:2 IV:3

Clinical feature Left eye Right eye Left eye Right eye Left eye Right eye
Globe size Normal Normal Small corneal diameter Normal Normal Normal
Iris Coloboma Coloboma Normal Normal Coloboma Coloboma
Retina Coloboma Coloboma Coloboma involving macula Normal Coloboma Coloboma
Optic disc Coloboma Coloboma Coloboma Normal Coloboma Coloboma
Lens Normal Dislocated Cataract Normal Mild cataract Normal
Electroretinogram Normal Normal Normal Normal Normal Normal
Visual-evoked potential Poor Moderate No response Good Poor Good
Nystagmus Yes Yes Yes No Yes Yes
Squint Convergent No Convergent No Divergent No
Visual acuity (LogMAR) 1.0 0.54 Hand movements only 0.16 1.0 0.56
Human Molecular Genetics, 2014, Vol. 23, No. 10 2513
Figure 1. Pedigree structure, clinical images and homozygosity mapping. (A) Simplified pedigree of the family with electropherograms of part of SALL2 exon 2
showing the c.85G.T mutation (arrowhead) homozygous in all three affected children and heterozygous in their parents. (B) Fundus photographs from all three
affected children showing varying degrees of retinochoroidal coloboma affecting the optic disc in all cases except the right eye of patient IV:2 which has a normal
appearance. (C) Homozygosity mapping output for each of the three affected siblings from Illumina Beadstudio v3.2 for chromosome 14 containing SALL2. Each
dot represents an individual SNP marker on the array plotted as a function of the frequency of the minor (B) allele. Regions of extended homozygosity (.100
SNPs) are indicated by pink shading.
acuity of 0.56 LogMAR in the right eye and 1.0 in the left. She with manifest latent nystagmus and poor fixation in the left eye.
exhibited bilateral typical inferior iris coloboma and marked reti- On electrodiagnostic testing, pattern visual-evoked potentials
nochoroidal coloboma (Fig. 1B). In addition, she had a mild lens (VEPs) were evident to a range of test checks with both eyes
opacity, divergent squint (exotropia of 35–30 prism diopters), open but reduced in amplitude indicating macular pathway
dysfunction. The responses from the left eye were degraded com- 5.15 Gb of mappable sequence data providing an average
pared with the right suggestive of poor vision. Her brother (IV:1) depth of 58 reads, yielding a minimum of 5-fold coverage at
examined initially at 11 years had a visual acuity of 0.54 and 1.0 90% of the captured exome. Average depth of coverage in all
LogMAR in the right and left eyes, respectively. He exhibited bi- candidate regions of shared homozygosity was sufficient for
lateral typical inferior iris coloboma as well as bilateral retinochor- comprehensive variant detection (29- to 114-fold, Supplemen-
oidal and optic disc colobomata. In addition, the crystalline lens tary Material, Table S1) of all exons and for all known annotated
was displaced (subluxed) in the right eye. He had a convergent genes within all candidate regions regardless of their size. From
squint (esotropia of 45 prism diopters) and hypertropia in the left the aligned reads across the entire captured exome, we identified
eye with rapid manifest latent nystagmus in both eyes. Although a total of 20 969 SNP variants of which 8522 were homozygous,
electrodiagnostic testing revealed essentially normal flash plus an additional 991 indels. Variants were filtered by excluding
responses from each eye, monocular pattern VEP testing revealed all those present in either dbSNP (build 137) or our local database
macular pathway dysfunction affecting the left eye. The youngest containing data from 172 exomes with a minor allele frequency
of the three siblings (IV:2) aged 9 years at first examination exhib- .0.1. This identified 26 novel homozygous SNPs and 21 novel

ited reduced visual acuities (right eye, 0.16 LogMAR, left eye, indel variants across the entire exome. Of these a single-coding
hand movements only) and a large retinochoroidal coloboma in- variant mapped to within one of the regions of homozygosity
volving the optic disc in the left eye, whereas the right fundus shared by the affected individuals. This variant is situated
appeared completely normal (Fig. 1B). In addition, there was a within the largest homozygous region on chromosome 14 at pos-
hypertropia and variable esotropia of 20–45 prism diopters, to- ition g.21993777 (Supplementary Material, Fig. S1) and intro-
gether with manifest latent horizontal, small amplitude pendular duces a premature stop codon within exon 2 (of 2) of the
nystagmus in the left eye. Normal flash electroretinograms were SALL2 gene (c.85G.T, NM_005407.1) (OMIM 602219). The
obtained from both eyes with pattern VEP evident to a range of c.85G.T mutation was confirmed by Sanger sequencing and
test checks in the right eye. There was no consistent pattern VEP shown to be homozygous in all three affected siblings and het-
from the left eye and, in addition, the flash VEP was also degraded erozygous in both parents (Fig. 1A). Analysis of DNA from
in the left eye compared with the right. Together, this is indicative the unaffected siblings (Fig. 1A) would have provided additional
of very rudimentary vision in the left eye, most likely as a result of genetic evidence supporting pathogenicity of the c.85G.T
the optic nerve coloboma. He had evidence of a small corneal variant, but was not available for analysis. This variant is also
diameter in the left eye but was not diagnosed as clinically micro- absent from the National Heart Lung and Blood Institute
phthalmic; refraction was mildly hyperopic, whereas a clinically Exome Variant Server database containing sequence data from
microphthalmic eye would be expected to exhibit marked hyper- 6500 exomes (ESP6500, August 2013; http://eversusgs.wa
opia. Axial length was not formally measured. All three affected shington.edu/EVS/). The co-segregation of the homozygous
children had a full pediatric review and there was no indication variant with the coloboma phenotype in the three affected chil-
for any neurological, developmental or renal abnormalities. dren, the unaffected status of the heterozygous carrier parents,
Renal function was assessed to be normal by urea and electrolyte the known parental consanguinity and the absence of other
function tests; renal ultrasound was not performed, and therefore pathogenic variants within the shared regions of homozygosity,
we cannot exclude the possibility of a structural kidney abnormal- together with the predicted severity of the mutation led us to con-
ity. Similarly, neuroimaging was not indicated and therefore not clude that we had identified the pathogenic mutation in this
performed on any member of the family. The parents were found family. This mutation is predicted to produce a severely trun-
to be normal on ophthalmological examination. Unaffected sib- cated protein of only 29 residues (p.Glu29∗ ), compared with
lings were reported normal after ophthalmological examination the 1007 amino acid SALL2 protein, resulting in the loss of trans-
and were unavailable for inclusion in the study. lation of 97% of the protein coding sequence including three
clusters of zinc-finger motifs shown to be essential for DNA-
binding activity (29). Alternatively, this mutation may evoke
Homozygosity mapping and exome sequencing nonsense-mediated decay mechanisms, which exist to eliminate
Assuming an autosomal recessive model of inheritance and in view aberrant mRNA species containing premature truncation
of the known consanguinity present in the family, we applied an codons; however, these mechanisms are not usually activated
homozygosity mapping strategy to identify the causative genetic if the mutation exists in the final exon (30,31). In either of
mutation underlying the coloboma phenotype. Analysis of geno- these predicted events, the effect of this mutation is complete
type data from all three affected individuals to identify regions of loss of any active protein function rendering an essentially
extended (.100) consecutive homozygous SNPs revealed two null allele. An alternative hypothesis is that a downstream
major loci .2 Mb in size common to all three affected individuals: ATG codon could rescue functional domains of the protein
5.3 Mb on chromosome 8p11.1 delineated by SNPs rs16891730– resulting in a hypomorphic mutation rather than complete
rs13279327 (chr8g.43106718–48442384; assembly Hg19 NCBI loss of function. However, analysis of mutant mRNA or
build 37.1) and 15.5 Mb on chromosome 14q11.2 (rs2792135– protein from patient cells to confirm the consequence of
rs3138056) (chr14g.20409258–35868514; Fig. 1C), in addition the identified SALL2 mutation was not feasible due to lack
to a number of smaller intervals (see Supplementary Material, of availability of suitable patient tissue.
Table S1). Combined, all of these regions contain 320 genes
annotated in the RefSeq database.
Given the large number of potential candidate genes, a whole- Mutation analysis of SALL2
exome sequencing strategy was adopted to identify the causative To determine the wider potential contribution of SALL2 muta-
mutation. Sequencing of patient IV:1 generated a total of tions to the incidence of coloboma, the coding region of a
further 178 patients with coloboma and associated anophthal- Sall2 null mutant mice showed no overt phenotypic abnormal-
mia/microphthalmia phenotypes were Sanger sequenced ities. Histological analysis of the eyes revealed a colobomatous
[comprising isolated coloboma (n ¼ 49), colobomatous micro- phenotype. Table 2 summarizes the analysis of hematoxylin and
phthalmia (n ¼ 10), syndromic coloboma (n ¼ 52), micro- eosin stained sections of the eye from E13.5 and E14.5 Sall2 2/2
phthalmia/anophthalmia (n ¼ 67)]. This analysis identified null and Sall2 +/2 heterozygous control embryos, which
three separate previously unreported synonymous coding var- revealed a penetrant bilateral phenotype in the homozygous
iants [c.33C.T (p.L11L); c.381C.T (p.V128V); c.315C.T Sall2 2/2 embryos. At E13.5, 12 out of 12 (100%) eyes from
(p.S105S)] in three individuals, each heterozygous for their Sall2 2/2 embryos (from two separate litters) showed a clear
respective variant, and therefore unlikely to represent pathogen- gap in the ventral neural retina and retinal pigmented epithelium
ic variation. No other novel potentially pathogenic coding vari- (RPE) at the anterior aspect of the eye (Fig. 3A). Sequential sec-
ation was identified in this cohort (Supplementary Material, tions towards the posterior pole showed the unfused optic fissure
Table S2). margins meeting and abutting each other (Fig. 3B and B′ ). At the
mid-lenticular level 6 out of 12 (50%) eyes examined showed

evidence of incomplete fusion where both sides of the fissure
Expression analysis of SALL2 margins were asymmetric with visible indentations at the inner
SALL2 has previously been reported to be widely expressed in and outer aspects of the neural retina where the margins meet
adult human brain as well as the heart, kidney and pancreas and a visible gap in the RPE (Fig. 3B and B′ ). In all Sall2 2/2
(32). In the mouse, expression of Sall2 has been observed in eyes examined at E13.5 (100%, n ¼ 12) from the mid-lenticular
the developing brain at embryonic day (E) 7.5, becoming region posteriorly to the developing optic disc, the optic fissure
restricted to distinct structures within the forebrain, midbrain appeared fully fused with an indistinguishable appearance of
and hindbrain by E10.5, as well as the developing kidney and the dorsal and ventral retina (Fig. 3C), which was morphologic-
neural tube at E11.5 (33,34). Expression has also been observed ally similar to that of control eyes. In contrast to the open fissure
in the developing lens placode and optic cup at E11.5 consistent observed in the Sall2 2/2 null eyes, sequential coronal sections
with the timing of optic fissure closure in the mouse (E11 – E13) of Sall2 +/2 control eyes showed complete fusion along the
(34). To analyse expression of SALL2 during human eye devel- entire length of the fissure (Fig. 3G –I; n ¼ 5). No defects were
opment, we performed RNA in situ hybridization using an anti- detected in other ocular components.
sense riboprobe for SALL2 on frozen cryosections prepared from Similar results were obtained for the histological analysis of
human embryonic and fetal eyes. SALL2 transcripts were eyes at E14.5 with 100% of Sall2 2/2 eyes examined (n ¼ 9)
detected throughout the retina and developing lens vesicle, in showing an open optic fissure in the anterior aspect of the
addition to the periocular mesenchyme, at 5 weeks of develop- retina, 33% showed evidence of incomplete fusion at the mid-
ment (Carnegie stage 15) (Fig. 2A and B), the time at which lenticular point and all eyes showed normal morphology in pos-
optic fissure closure commences. Expression was maintained terior sections compared with controls (Fig. 3J– L). By contrast
in the developing retina and following completion of fissure in all Sall2 +/2 eyes at E14.5 the retina appeared fully continuous
closure, at 8 weeks of development, becomes restricted to the with no morphological characteristics of the open optic fissure
inner neuroblastic layer (Fig. 2D and E). Concurrent PAX6 (Fig. 3M – O; n ¼ 10).
expression was detected across the neural retina from 5 to 8 During normal optic fissure closure, the neuroepithelial basal
weeks of development (Fig. 2C and F). We also confirmed lamina at the tips of the fissure margins dissolves in a contact-
SALL2 expression in the developing cornea, lens and retina at dependent manner as the tips fuse to form the continuous
different developmental stages by reverse transcription PCR neural retina and RPE (35). Failure of basement membrane dis-
analysis of RNA extracted from microdissected human embry- integration at the optic fissure margins has been reported as a
onic and fetal ocular tissue together with expression of PAX6 cause of coloboma in a recent study of Fatty Liver Shionogi
in these tissues at all time-points examined (Fig. 2G). These mice who present a phenotype similar to human ocular coloboma
data support a role for SALL2 during human eye development without microphthalmia, and also in two other coloboma models
prior to, and maintained during and after, optic fissure closure. Pax2 2/2 and Vax2 2/2 mice (36– 38). In order to assess whether
dissolution of the basal lamina is occurring at the point where the
optic fissure margins meet in Sall2 2/2 embryonic eyes, we used
Histological analysis of Sall2 null mice an anti-Laminin antibody to identify the basal lamina. Figure 3D
To further strengthen the evidence supporting the pathogenicity shows the basal lamina clearly surrounding the tips of the
of SALL2 mutation as a novel cause of coloboma, we investi- unfused fissure margins at the anterior aspect of the retina in
gated loss of Sall2 function in a mouse model. Sall2-deficient E13.5 Sall2 2/2 eyes. No laminin staining was observed
mice were previously reported to have no apparent abnormal between the converged tips of the optic fissure at the point of
phenotype when bred on a C57BL/6 genetic background, and contact in the mid-lenticular region (Fig. 3E and E′ ), or within
low penetrance strain-specific incidence (11 – 17%) of neural the ventral retina in posterior sections where laminin staining
tube defects and significant perinatal lethality when bred onto was continuous around the lens and outside of the neural retina
different mixed genetic backgrounds (33,34). Neural tube (Fig. 3F). This indicates dissolution of the basal lamina is occur-
closure involves fusion of the neural plate, and bears similarity ring correctly in the absence of Sall2 when the fissure margins
to the process of optic fissure closure. As previous reports did make contact, consistent with the complete fusion of the optic
not analyse the eyes of Sall2 2/2 mice in detail, we bred homo- fissure observed from the midline to the posterior pole of the eye.
zygous Sall2 2/2 mice (on a C57BL/6 background, shown to Given the observation that the optic fissure appeared to fuse
have 0% penetrance of neural tube defects) for further analysis. normally at the back of the eye, we next performed analysis of

Figure 2. Expression analysis of SALL2 during human eye development. (A and B) RNA in situ hybridization on sagittal sections showing expression of SALL2 tran-
scripts throughout the retina (R) and developing lens vesicle (lv) at 5 weeks of development in comparison to PAX6 expression (C). Expression of SALL2 is maintained
at 8 weeks (D) but becomes restricted to the inner neuroblastic layer (INBL) as shown magnified in (E) as compared with PAX6 (F) which is expressed uniformly
throughout the developing neural retina. (G) RT –PCR analysis using intron-flanking primers for SALL2 in total RNA extracted from microdissected cornea,
retina and lens tissue from varying stages of development. PAX6 expression is shown for comparison and GAPDH was used as a positive control. (H) SALL2
sense probe shown as a control. GCL, ganglion cell layer;, INBL, inner neuroblastic layer; ONBL, outer neuroblastic layer; RPE, retinal pigmented epithelium.
Scale bar represents 50 mm in (B, C, E, F and H), 100 mm in (A), 200 mm in (D).
Table 2. Summary of histological analysis of the extent of optic fissure closure in Sall22/2 null and Sall2+/2 mouse embryos
Stage Genotype Anterior Midline Posterior Total eyes

Open Closed Open Closed Open Closed
E13.5 Sall22/2 12 (100%) 0 (0%) 6 (50%) 6 (50%) 0 (0%) 12 (100%) 12

Sall2+/2 0 (0%) 5 (100%) 0 (0%) 5 (100%) 0 (0%) 5 (100%) 5
E14.5 Sall22/2 9 (100%) 0 (0%) 3 (33%) 6 (67%) 0 (0%) 9 (100%) 9
Sall2+/2 0 (0%) 10 (100%) 0 (0%) 10 (100%) 0 (0%) 10 (100%) 10
eyes from postnatal day (P) 20 Sall2 2/2 mice (from three separ- two distinct classes of retinal abnormality in homozygous
ate litters, n ¼ 18 eyes) to determine if the anterior coloboma- mutant animals. Six of 18 eyes (33%) displayed a retinal colo-
tous defect persists, or whether the delayed embryonic fusion boma phenotype similar to that seen in the embryo with a gap
process is ameliorated at later stages. This analysis identified in the peripheral ventral neural retina behind the ciliary body
Figure 3. Analysis of optic fissure closure in embryonic eyes from Sall22/2 mice. (A– C) Coronal sections of the eye from an E13.5 homozygous Sall22/2 embryo
showing a gap in the neural retina at the anterior aspect of the eye (arrowheads in A). At the mid-lenticular region (B) the retinal margins meet and touch each other with
visible indentations at the opposing tissue margins (arrowheads in B′ which shows a magnification of the boxed region in B). Posteriorly (C), the retina appears fused
and morphologically continuous and indistinguishable from the control (I). (D –F) Immunohistochemistry of sections from E13.5 Sall22/2 eyes using anti-laminin
(red) to identify the basement membrane. Laminin staining is observed surrounding the unfused fissure margins anteriorly (arrowheads in D). At the point of contact
between the tips of retina in the mid-lenticular region laminin staining is not detectable (E, E′ ) suggesting dissolution of the basal lamina. Posteriorly, no laminin
staining is observed, suggesting the fusion of the optic fissure in this region (F). Cell nuclei are stained with DAPI (shown in blue). (G– I) Comparative sections
from control Sall2 +/2 at E13.5 showing fusion of the optic fissure at all levels of the eye. (J– L) Coronal sections of the eye from E14.5 Sall2 2/2 embryo
showing similar failure of optic fissure fusion in the anterior (arrowhead in J) and mid-lenticular regions (K, K′ , arrowheads in K′ indicate indentations in the
inner and outer aspects of the retina consistent with incomplete fusion), with a normal appearance posteriorly (L). (M–O) Representative sections from a control
embryo showing the continuous morphological appearance of the fused retina at E14.5. In all images, the ventral aspect is shown at the bottom of the image. L,
lens; RPE, retinal pigmented epithelium; R, neural retina. Scale bars represent 50 mm in (B′ , E′ , H′ , K′ and N′ ), 100 mm in (A, B, D, E, G, H, J, K, M and N) and
200 mm in (C, F, I, L and O).

Figure 4. Histological analysis of eyes dissected from P20 Sall22/2 mice. Top right. Schematic representation of a mouse eye showing the plane of sectioning and
approximate anterior location of the observed retinal abnormality (dotted line). Co, cornea; I, iris; L, lens; R, neural retina; ON, optic nerve. (A–E) Coronal sections of
the eye from a P20 homozygous Sall2 2/2 mouse showing a gap in the neural retina visible anteriorly (A and B). In serial sections towards the posterior pole, the tips of
the retinal margins meet with residual pigment visible surrounding the tips of the margins (arrowheads in B and D which show magnifications of boxed region in A and
C, respectively). (E) More posteriorly, beyond the mid-lenticular region, the retina appears fused and indistinguishable from normal. Note, the lens was removed
during processing. (F) Representative coronal section from a homozygous Sall2 2/2 mouse showing localized elevated lesion of the retina as compared with (G)
equivalent coronal section of eye from P20 heterozygous Sall2 +/2 mouse showing no abnormality. (H–J) Consecutive anterior-to-posterior sections showing mag-
nification of boxed region in (F) showing progressive swelling and disorganization of ganglion cell layer (H), inner plexiform layer (I) and inner and outer nuclear
layers (M). In more posterior sections, the retina shows normal lamination (K). GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer;
OPL, outer plexiform layer; ONL, outer nuclear layer; IS/OS, photoreceptor inner and outer segments. Scale bars represent 200 mm in (A, C, F, G and E), 50 mm
in (B and D) and 100 mm in (H–K).
(Fig. 4A and B; Supplementary Material, Fig. S2). Sequential tips of the neural retina (arrowhead, Fig. 4B and D). In this case,
sections towards the posterior pole showed the unfused although retinal lamination appeared histologically normal at
margins meeting and abutting each other (Fig. 4C and D). the juxtaposed margins, the layers were mis-aligned and
During normal development, the presumptive retinal pigmented visible indentations were apparent at the inner and outer
epithelial (RPE) layer has been reported to invert into the aspects of the neural retina where the margins meet, consistent
merging fissure margins prior to fusion and eventual RPE separ- with incomplete fusion (Fig. 4D, arrowheads). Other regions
ation from the neural retina (35). In the Sall2 2/2 retina a residual where the retinal margins failed to make contact showed evi-
layer of pigment was observed at P20 between the two opposing dence of disorganization of the retinal lamina (Supplementary
Material, Fig. S2, arrowheads). Seven other Sall2 2/2 mutant assays did not provide evidence in support of specific binding of
eyes showed a single focal anteriorly localized lesion within either PAX2 or PAX6 (or in combination) to the SALL2 promoter
the neural retina (Fig. 4F, H and I), appearing in some cases as (Fig. 5C). To assess the opposite model of Pax2/Pax6 expression
if the outer layer on the one side had traversed the inner layer being altered by loss of Sall2 during fusion, we analyzed the distri-
on the other side (see Fig. 4I). Outside of the lesion, the histology bution of PAX6 and PAX2 proteins in Sall2 2/2 embryonic eyes by
was indistinguishable from that of wild-type and heterozygous immunohistochemistry. By E13.5 the downregulation of PAX2
eyes (Fig. 4K). Within the elevated lesion, which protruded within the ventral optic cup has occurred normally in mutant
into the vitreal space, the ganglion cell layer, and the inner and embryos; PAX2 was detected only in a wedge of cells surrounding
outer nuclear layers were abnormally thickened (Fig. 4I and J), the optic nerve head as previously described (36,41). PAX6 was
whereas the plexiform layers appeared disorganized with an detected within the retina and lens. These expression patterns
increased inner plexiform layer and reduced outer plexiform were unaltered in Sall2 2/2 embryos compared with controls
layer. In all eyes examined, the retina of homozygous Sall2-null (Fig. 5D–K). Pax2 mutant eyes show persistence of the basal
mice was indistinguishable from wild type from the mid- lamina when the converging tips make contact, as judged by the

lenticular region to the posterior pole where the optic nerve abnormal presence of laminin across the retina (36). Here, the
exits the eye (Fig. 4E and G) consistent with the conclusion normal regulation and distribution of PAX2 observed in the
that the retinal abnormalities in the Sall2 null mice are restricted Sall22/2 eye is consistent with the observed dissolution of the
to the anterior aspect of the ventral retina. Comparative analysis basement lamina in the posterior hemisphere and the normal for-
of both eyes from the same animal showed cases of bilateral colo- mation of the optic nerve. These data together suggest that
boma (Supplementary Material, Fig. S2), bilateral retinal lesion, SALL2 is not directly regulated by PAX6/PAX2 and may be
as well as unilateral coloboma with contralateral retinal lesion. either genetically downstream of, but not directly regulated by
The remainder of the homozygous Sall2 2/2 eyes examined his- PAX2/PAX6 binding to the proximal promoter, or alternatively
tologically at P20 (n ¼ 5), as well as heterozygous Sall2 +/2 SALL2 may function in an independent pathway required for
mice of equivalent age (n ¼ 4) showed no visible morphological correct alignment of the fissure margins.
abnormality. There was no evidence for a small eye/micro-
phthalmic phenotype in any of the homozygous Sall2 2/2
mutant eyes at either the embryonic or adult stages investigated.
Together, these data show that loss of Sall2 causes coloboma in DISCUSSION
mice as well as humans, and suggests that Sall2 is required for
correct growth of the optic fissure margins towards each other SALL family genes and human disease
and their alignment, rather than the process of dissolving the Our study identifies SALL2 as the third member of the spalt-like
basement membrane. family of transcription factor genes to cause a human birth
defect. SALL/Spalt family genes are widely studied in model
organisms and play a crucial role during the development of a
Analysis of PAX2 and PAX6 in relation to SALL2 expression number of systems (42,43). The SALL2 gene contains a zinc-
Finally in order to identify coloboma genes that might be finger domain homologous to the Drosophila region-specific
regulating SALL2 in the eye during optic fissure closure, we homeotic gene spalt (sal) (42). There are four human spalt-like
performed an in silico analysis of putative transcription factor- genes each encoding an N-terminal C2HC type zinc finger, in
binding sites in 1.6 kb of the SALL2 proximal promoter addition to three (SALL2, SALL4) or four (SALL1, SALL3)
(chr14:g.22005337–22007000). Matinspector (www.genomatix. clusters of C2H2 class zinc-finger domains required for inter-
de) analysis identified a number of potential ocular transcription action with DNA and other proteins (42,44). Phylogenetic ana-
factor-binding sites in this region (Fig. 5A), including adjacent pre- lysis identifies conserved orthologues of SALL1, SALL3 and
dicted binding sites for PAX6 (chr14:g.22005800–22005818) and SALL4 genes throughout most metazoa, whereas SALL2
PAX2 (chr14:g.22005773–22005801), two transcription factors appears to have evolved following the divergence of the mam-
implicated in human coloboma pathology, located 430– malian lineage as orthologous genes have not been identified
460 bp upstream of the start of transcription of SALL2 (Fig. 5A). in non-mammalian organisms (45). Previously, we reported
Comparative genomic analysis of the human and mouse Sall2 pro- SALL1 mutations cause the developmental disorder Townes-
moters showed limited overall sequence conservation; however, Brocks syndrome (OMIM 107480) and SALL4 mutations
Matinspector also identified a number of putative PAX2- and cause Okihiro (Duane-Radial Ray; OMIM 607323) syndrome
PAX6-binding sites in the murine Sall2 proximal promoter (46,47). These syndromes exhibit dominant inheritance,
region (Supplementary Material, Fig. S3). Alteration of PAX2 whereas we identify homozygous loss of SALL2 as the cause
function is a common cause of coloboma and studies in mouse of a recessive eye condition. Similar to the mutation we have
and zebrafish models have highlighted its function in morpho- identified in SALL2 (p.Glu29∗ ), premature termination muta-
genetic events required for optic fissure closure (28,36,39,40). tions lacking all paralogous zinc-finger domains have been
To test whether PAX2 and/or PAX6 are capable of regulating reported for both SALL1 and SALL4 resulting in loss of function
expression of SALL2 in vitro, we cloned a 1.0 kb region of the of the mutant proteins (46– 48). A recessive hypomorphic muta-
proximal promoter upstream of a luciferase reporter gene. Co- tion in SALL1 has also been reported in association with a variant
transfection of PAX2 or PAX6 expression constructs separately Townes-Brocks phenotype (49) implying that dosage of these
or in combination did not elicit a significant activation of the genes is critical during embryogenesis. In the case of SALL2,
SALL2 promoter when compared with the promoterless reporter we find that a single gene copy is sufficient for normal ocular
vector (Fig. 5B). Furthermore, electrophoretic mobility shifts development, as demonstrated by the lack of an observable
Figure 5. Analysis of SALL2 in relation to PAX2 and PAX6. (A) Schematic diagram of 1.6 kb of the human SALL2 proximal promoter region showing selected tran-
scription factor-binding sites relevant to ocular development. (B) Luciferase reporter assay in HEK293 cells showing that neither PAX2 nor PAX6 activate transcrip-
tion of a luciferase reporter construct (LUC) containing 1023 bp of the SALL2 promoter encompassing putative binding sites for both proteins. No significant
difference in activation was observed between the SALL2 promoter construct (dark gray bars) compared with the promoterless reporter (pGL4.10, light gray bars)
when co-transfected with PAX2 or PAX6 expression constructs either singly or in combination. Data are presented as mean + SD of three repeat experiments
each performed in triplicate. (C) EMSA showing that no specific protein:DNA complex is formed between in vitro translated PAX2 or PAX6, either singly or in com-
bination, and a DNA probe containing both predicted binding sites in the SALL2 promoter as shown in (A). Right hand panel shows in vitro translation of PAX2 and
PAX6 proteins incorporating [35S]-L-methionine showing equivalent amounts of protein are synthesized in each reaction; bands of the expected sizes of 44 and 46 kDa
are observed for PAX2 and PAX6, respectively (D– G) Immunohistochemistry of sagittal sections of the eye from E13.5 mouse embryos stained with anti-PAX2 (red).
Specific expression of PAX2 is observed in the developing optic nerve head (arrowheads in E and G). No difference in expression pattern is observed between Sall2 2/2
(D and E) and control Sall2 +/2 (F and G) eyes. (H and I) Immunohistochemical analysis of PAX6 (red) in sagittal eye sections from E13.5 Sall2 2/2 (H and I) and
Sall2 +/2 control (J and K) embryos. PAX6 expression is detected in the retina (arrows) with strong expression seen in the anterior periphery of the retina (arrowheads),
as well as the corneal epithelium and lens. No difference in PAX6 protein expression was observed between Sall2 2/2 and control eyes. Nuclei were counterstained
with DAPI (blue). C, cornea; L, lens; R, retina; ON, optic nerve. Scale bars represent 50 mm in (E, G, I and K) and 100 mm in (D, F, H and J).
phenotype in the unaffected parents and heterozygous Sall2 a number of genes have been implicated in different stages of
mutant mice. this process. For instance, midline Sonic hedgehog signaling is
Functional redundancy between Sall factors is likely to com- required for (i) and loss of SHH signaling causes holoprosence-
pensate for the heterozygous loss of Sall2 in the eye (33,34,50), phaly, which in milder forms includes coloboma and micro-
particularly Sall1 which is expressed in the developing eye and phthalmia (18). Both RA and bone morphogenetic protein
can be associated with retinochoroidal coloboma in some (BMP) signaling play key roles in (ii) and disruption of these
cases of Townes-Brocks syndrome (48,51). Similarly, redun- pathways can cause coloboma associated with microphthalmia
dancy is likely to restrict the phenotype to the eye, despite the and anophthalmia [RBP4 (OMIM 180250), STRA6, ALDH1A3,
wider expression pattern of the Sall2 gene; for example, homo- BMP4 (OMIM 112262), GDF6, GDF3 and SMOC1 mutations
zygous deletion of Sall1 results in severe kidney dysgenesis, disrupt RA and BMP signaling, respectively] (13,20– 22,55–
whereas Sall2-deficient mice have normal kidneys despite 59), whereas the transcription factors RAX, PAX6 (OMIM
both genes being expressed during kidney development (33,52). 607108), VSX2, OTX2 (OMIM 600037), SIX3 (OMIM
603714) and SOX2 (OMIM 184429) are involved in eye field

specification and neural progenitor proliferation (i – ii) and loss
Variable penetrance of SALL2/Sall2-related phenotype of function causes anophthalmia/microphthalmia, including
The variable penetrance of the coloboma phenotype observed in rare reports of coloboma (16,17,23,25,60,61). Mutations in
the adult Sall2 2/2 mice accords both with the variable pene- C12orf57 have recently been identified in patients with colobo-
trance of neural tube defects previously observed on different matous microphthalmia associated with developmental delay,
genetic backgrounds (34) and the variable expressivity and seizures and corpus callosum abnormalities suggesting a role
incomplete penetrance in other examples of both murine and at similarly early levels of development (62). Analysis of
human coloboma (37,38). However, at embryonic stages the PAX2, a cause of optic disc and retinal coloboma (OMIM
observed coloboma phenotype was fully penetrant and bilateral 120330) (28), has identified roles in both patterning the optic
in all embryos investigated, indicating some degree of amelior- stalk/neural retina boundary together with PAX6 (i), and for
ation at later stages of development when a third of postnatal fissure closure and correct development of the optic disc (v
animals exhibited no apparent abnormality. The phenotype is and vi) (63). In addition, a genetic signaling cascade for colo-
also variable within the pedigree we describe, as demonstrated boma centered around PAX2 has been elucidated involving
by individual IV:2 who had a normal fundus appearance in one SHH and BMP4 signaling as upstream regulators (39,64). Al-
eye, indicating that the optic fissure closure defect can be over- though coloboma is a common association with microphthalmia,
come in the absence of SALL2 by the effects of other environ- this does not necessarily mean that coloboma is a non-specific
mental, stochastic or genetic factors. Implicated environmental outcome of microphthalmia. A recent study of 141 patients
risk factors include vitamin A deficiency during gestation con- with microphthalmia, anophthalmia and coloboma phenotypes
sistent with the role of retinoic acid (RA) signaling in early eye revealed a similar number of microphthalmic cases without
patterning (53). Variability in the initiation site or timing of coloboma (24%) to those with optic fissure closure defects and
optic fissure closure, both between and within species, may normal globe size (23%) (65). None of the patients in the
explain the different presentation and location of the defect. In present study were clinically diagnosed as microphthalmic. Fur-
mice initiation of closure of the optic fissure has been reported thermore, analysis of homozygous mutant Sall2 2/2 eyes at
to occur at a posterior level close to the developing optic disc embryonic and adult stages failed to show any evidence of a
rather than from a central midpoint in a proportion of animals small eye or microphthalmia phenotype. These observations
(35). Whereas in human development fusion initiates in the combined with the localized nature of the retinal coloboma
center of the fissure and progresses both anteriorly towards the phenotype identified suggest that SALL2 mutation plays a specif-
rim of the optic cup and posteriorly towards the optic stalk ic role in disruption of the closure process at steps (iii) and (iv),
(54). Genetic variation influencing the expression levels of, as but not (v).
yet unidentified, targets of SALL2 or their downstream effectors
is also likely to contribute to phenotypic variability in both
species. Genetic pathways and mechanisms implicated in optic
fissure closure
Where does SALL2 fit within the molecular pathways important
Human genes required for optic fissure fusion
for fissure closure? Our analysis suggested that SALL2 plays a
Correct optic fissure closure requires six crucial steps: (i) early specific role in optic fissure closure required independently of
morphogenesis of the optic vesicle, (ii) growth and patterning PAX2. Prompted by the presence of putative PAX2-binding
across the dorso-ventral axis of the bilayered optic cup to form sites in the SALL2 promoter we investigated the idea that
an incomplete globe, (iii) alignment and contact between the SALL2 is genetically downstream of, and potentially directly
neuroepithelia on either side of the fissure, (iv) adhesion of the regulated by, PAX2. No evidence was found supporting PAX2
fissure margins, (v) dissolution of the basement membrane and regulation of the SALL2 promoter. Our findings also refuted
(vi) fusion and separation of the neural retina and RPE to form the alternative model of coloboma caused by disruption of
a complete globe. Comprehensive identification of the genes PAX2 function by showing normal localization of PAX2 in the
involved in these steps, their function and interactions may cells encircling the optic disc and normal dissolution of the
offer a basis for future therapeutic interventions, but there is basal lamina, which requires PAX2, in the Sall2-deficient
much to be learned in this respect. From previous human eyes. PAX2 deficiency primarily causes a posterior optic disc
genetic studies and complementary analysis in animal models, defect in mice (63), whereas in the Sall2 2/2 mice normal
closure of the posterior globe was observed. This finding is also MATERIALS AND METHODS
consistent with a previous observation that Pax2 expression is
Ethical approval
unaltered in the developing kidney of Sall2-deficient mice
(33). Interestingly, a JNK1/2.BMP4/SHH.PAX2 signaling Informed written consent was obtained from the parents prior to
pathway is implicated in both optic fissure and neural tube collection of blood samples, DNA extraction and analysis. The
closure defects (39). Jnk12/2 Jnk22/+ mice (lacking activity study was approved by the UCL Institute of Child Health/
of members of the Jun N-terminal Kinase, JNK, group of Great Ormond Street Hospital for Children Joint Research
mitogen activated kinases) exhibit coloboma associated with Ethics Committee. All murine experiments were carried out
loss of Shh and Bmp4 signaling, and reduced levels of Pax2, using protocols reviewed and approved under license by the
while addition of Bmp4 restored Shh and Pax2 expression United Kingdom Home Office under the Animal (scientific
(39). The JNK1/2.BMP4/SHH pathway may also be important procedures) Act 1986.
for regulating SALL2 expression, independently of PAX2. In
Sall22/2 /Sall4+/2 mouse embryos that develop exencephaly,

like JNK1/JNK2/PAX2-deficient embryos, increased levels of Homozygosity mapping and exome sequencing
apoptotic cells and failure of elevation and fusion of the neural DNA samples from all three affected individuals were geno-
folds were reported (34). Similarities in the molecular pathways typed using the Human610-Quad BeadChip (Illumina)
and mechanisms of these different tissue fusion processes are scanned using an iScan system and data analyzed using Beadstu-
likely to emerge from future analysis. Other recent studies dio v3.2 (Illumina) to identify regions of consecutive homozy-
have provided new evidence for the association between dysre- gous SNPs (.100) common to all affected individuals.
gulation of cell death and coloboma in zebrafish (40,66), though Whole-exome sequencing was performed to identify candidate
it is not yet clear how similar these fusion processes are to those genetic variants. Three micrograms of genomic DNA from
in mammals. patient IV:1 was subject to in-solution enrichment of exonic
What is the role of SALL2 in the eye? The molecular function sequences using the SureSelect Human All Exon Kit v.1
of the SALL2 protein is largely unknown, although a role as a (Agilent) according to the manufacturer’s instructions. Adaptors
tumor suppressor has been suggested (67). Previous in vitro were ligated for sequencing using a 76 bp paired-end run on an
studies demonstrated that SALL2 exhibits growth arrest and Illumina Genome Analyser IIx platform. Reads were aligned
pro-apoptotic properties and interacts directly with the p75 neu- to the human reference genome (Hg19 NCBI build 37.1) using
rotrophin receptor, which is implicated in the control of pro- Novoalign software (version 2.07.04; Novocraft) (http://www.
grammed cell death during retinal development (68– 70). novocraft.com/main/index.php). Variant calling (SNPs and
Perturbation of cell death pathways, and/or promotion of cell indels) was performed using SAMtools (version 0.1.7) together
proliferation (68,71), could plausibly explain both the open with Dindel (version1.01) (http://www.sanger.ac.uk/resources/
fissure, and the appearance of an abnormal retinal mass in the software/dindel/) using default values and a minimum SNP
Sall2 null mice. Future studies to investigate the expression and indel quality of 30 and minimum SNP and gap mapping
pattern of other coloboma-associated genes, and the identifica- quality of 50.
tion of the downstream targets and effectors of SALL2 function
during embryonic development in Sall2 2/2 mice will help to
further elucidate essential genetic pathways and their role in In situ hybridization
coloboma formation. Human embryonic and fetal eyes were fixed in 4% (w/v)
phosphate-buffered formaldehyde solution (PFA) and equili-
brated in 30% sucrose solution for cryoprotection prior to freez-
ing in an optimal cutting temperature (OCT) compound (RA
SUMMARY
Lamb). Sections were cut on a Leica CM1900 UV cryostat to
In summary, using a combined homozygosity mapping and 14 mm thickness and collected on Superfrost-Plus glass slides
exome sequencing approach, we have identified a novel homo- (VWR). In situ hybridization was performed for 18 h in hybrid-
zygous mutation in SALL2 in three siblings variably affected ization buffer containing digoxigenin-incorporated riboprobes.
with retinochoroidal coloboma. This mutation (c.85G.T) is Riboprobes were generated from an 840 bp fragment of the 3′ un-
predicted to result in a truncated protein of only 29 amino translated region of human SALL2 (NM_005407.1) amplified
acids lacking all important functional domains of SALL2. using primers 5′ -GACCTTGGTAGAGGAGCTGA-3′ and
We have shown that SALL2 is expressed in the developing 5′ -AGGAATGCCACATACTGGTT-3′ and cloned into
human retina during the period of optic fissure closure and pGEMT-Easy (Promega). The template for the PAX6 probe
that Sall2-deficient mice exhibit an ocular coloboma phenotype has been described elsewhere (72). Slides were incubated with
providing a new model to investigate developmental closure anti-digoxigenin conjugated with alkaline phosphatase
and fusion mechanisms. In Sall2-deficient embryos the (Roche) diluted 1:1000 in 2% sheep serum. Expression patterns
fissure margins fail to meet in the anterior portion of the were visualized with a Nitro-Blue Tetrazolium Chloride/
globe, whereas in the posterior region closure is completed 5-Bromo-4-Chloro-3′ -Indolyphosphate p-Toluidine Salt
as normal dissolution of the basal lamina and fusion occurs. (NBT/BCIP) system (Roche). Sections were mounted in
This is the first report of a critical role for SALL2 in eye mor- DPX-mounting medium (RA Lamb) and viewed on a Zeiss
phogenesis in both humans and mice, and the gene should Axioplan 2 and images were captured with a Jenoptik C14
therefore be considered for mutation screening in patients digital camera (OpenLab, Improvision). In situ data are from
with coloboma. the analysis of five human eyes, tested in at least
three-independent experiments for each antisense probe; sense Plasmid constructs

probes were used as controls and gave no hybridization signal.
Full-length PAX2 (NM_003987) and PAX6 (IMAGE Consor-
tium Clone ID 3880468) (73) cDNA clones in mammalian
expression vectors were purchased from Cambridge Bioscience
Immunohistochemistry and Source Bioscience UK, respectively. PCR primers were
designed to amplify a 1023 bp region of the SALL2 proximal
OCT compound was removed from cryosections by washing for promoter (Chr14:g.22005060– 22006082) as follows: SALL2-
15 min in phosphate-buffered saline (PBS) at 378C. Sections 1023F 5′ -GTGCTCAGCCTAATTTTTGTG-3′ and SALL2-
were then blocked with 10% (v/v) goat serum, 1% (w/v) 1023R 5′ -GGGTAGAGAGTTGGGAGAGG-3′ . PCR products
bovine serum albumin in PBS containing 0.1% (v/v) Triton were generated from human genomic DNA, purified and
X-100 for 1 h at room temperature preceding primary antibody cloned directly upstream of the luciferase gene in pGL4.10
incubation. Primary antibodies were incubated in blocking solu- (Promega).
tion for 2 h at room temperature at the following concentrations:

PAX6 (Covance) 1:300, PAX2 (Lifespan Biosciences) 1:100,
and Laminin (Abcam) 1:200; Triton X-100 was excluded from Electrophoretic mobility shift assay
the blocking buffer for incubations with anti-Laminin. After
several washes with PBS, sections were incubated with Goat PAX2 and PAX6 proteins were synthesized using the TnT T7
anti-rabbit AlexaFluor594 (Life Technologies), 1:800 dilution, and SP6 Coupled Reticulocyte Lysate Systems (Promega).
in blocking solution at room temperature for one hour, followed EMSAs were performed using following radiolabeled DNA
by a 1:3000 dilution of the nuclear dye 4′ ,6-diamidino-2- probe encompassing both the PAX2- and PAX6-binding sites
phenylindole, dihydrochloride (DAPI; Life Technologies) for in the proximal promoter of SALL2, double stranded by anneal-
10 min at room temperature. Slides were mounted following ing to the complimentary oligonucleotide: 5′ -GATCACCT
several washes in PBS with Citifluor AF-1 mountant (Electron GGATTCCAGGCCTGTGTCTGATACATCACAGCTGTGC
Microscopy Science). AACTT-3′ . Ten microliters of in vitro translated protein were
added to 50 pmol of probe in a total 40 ml binding reaction con-
taining 7.5 mM Tris, 37.5 mM KCl, 3.25% glycerol, 30 mg
bovine serum albumin, 75 mM dithiothreitol, 750 mM ethylene-
Reverse transcription –PCR diaminetetraacetic acid. Binding reactions were incubated at
RNA was extracted from microdissected human fetal cornea, room temperature for 20 min prior to electrophoresis through
retina and lens using Trizol reagent (Invitrogen) according to an 8% non-denaturing polyacrylamide gel.
the manufacturer’s protocol. First-strand cDNA synthesis was
performed with 50 ng total RNA via M-MLV Reverse Transcript- Cell culture and transient transfection
ase (Promega) and random hexamer oligonucleotide primers.
Second strand synthesis was performed with gene-specific Human embryonic kidney 293 cells were maintained in Dulbec-
primers from exon 1 to exon 2 (flanking intron 1) for SALL2, co’s modified Eagle medium supplemented with 10% fetal
5′ -AACCCCAACAGTTAATCTCG-3′ and 5′ -TATGCTCTGT bovine serum at 378C in humidified air containing 5% CO2.
GTCCATGACC-3′ , and exon 9 to exon 10 for PAX6, 5′ -CACC Transient transfection assays were performed using Lipofecta-
AGTGTCTACCAACCAA-3′ and 5′ -TTGCATAGGCAGGTT mine 2000 reagent (Life Technologies) following the manufac-
ATTTG-3′ . turer’s instructions. Briefly, 2.5 × 104 cells/well were seeded
into a 96-well tissue culture plate. Cells were transfected with
20 ng of luciferase reporter and 30 ng of each expression con-
struct either individually or in combination as indicated in
Histological analysis of Sall2 mutant mice Figure 5. pRLSV40 (Promega) was cotransfected in all experi-
Sall2 knockout mice were maintained on a C57BL/6 background ments to control for transfection efficiency, and the total
and genotyped by PCR of genomic DNA as described previously amount of transfected DNA was normalized to 120 ng/well by
(34). Experimental litters were generated by overnight mating, the addition of pBluescript. Cells were harvested 24 h following
with the day of finding a copulation plug designated as E0.5. transfection and assayed for luciferase activity using the Dual-
Litters were dissected from the uterus at Days 13.5 and 14.5 in Luciferase Reporter Assay System (Promega), using a BMG
PBS. Whole heads were dissected from embryonic mice and Fluostar Optima microplate reader. Luciferase activity for
individual eyes dissected from P20 mice following labeling of each expression vector was determined by normalization to
the dorsal edge with Tissue Marking Dye (Cancer Diagnostics) the level of baseline luciferase activity of the promoterless
to orientate the eye, fixed in 4% PFA, dehydrated through a pGL4.10 reporter. Data are presented as mean + SD from
graded alcohol series and chloroform and embedded in paraffin. three-independent experiments each performed in triplicate.
Parasagittal sections through the head cut the optic cup en face Two-way Student’s t-test was used to determine significance
and were performed to allow visualization of the optic fissure between means.
as a gap in the ventral retina. Microtome sections were cut
at 8 mm mounted on Superfrost-Plus slides and rehydrated
through ethanol series prior to staining with hematoxylin and Mutation analysis
eosin, followed by subsequent dehydration in ethanol, clearing PCR primer pairs were designed to amplify the coding sequence
in Histo-clear (National diagnostics) and mounting. of SALL2 (NM_005407.1) using Primer3 (http://frodo.wi.mit.
edu/). Primer sequences were as follows: 1, forward 5′ -TCT MRC UK (G0700089)/Wellcome Trust (GR082557) Human De-
GCTTCACAGTGATTTGC-3′ , reverse 5′ -TGCATCTCAACT velopmental Biology Resource (http://hdbr.org). Funding to pay
CCTTCAAA-3′ ; 1A, forward 5′ -TTTCTCACTCCAGCTTCT the Open Access publication charges for this article was provided
CC-3′ , reverse 5′ -CCTACGCAGAGAATCATGC-3′ ; 2.1, by UCL.
forward 5′ -GGTTACTGGCCTCCTTGTTA-3′ , reverse 5′ -GT
CGATTCTGGAGGTAATGG-3′ ; 2.2, forward 5′ -CAGAGAG
GAGAGGAGAGGAGT-3′ , reverse 5′ -GTGGTAAAGGTGG
AAGAAGG-3′ ; 2.3, forward 5′ -AAGACACTGGCATCTTC
CTC-3′ , reverse 5′ -AAACGGTTTCCACAGACATT-3′ ; 2.4, REFERENCES
forward 5′ -CCAAAGTATTTGGCAGTGAC-3′ , reverse 5′ -TA 1. Onwochei, B.C., Simon, J.W., Bateman, J.B., Couture, K.C. and Mir, E.
CTTTCTGCCACTCCACTG-3′ ; 2.5, forward 5′ -TTTGTGC (2000) Ocular colobomata. Surv. Ophthalmol., 45, 175–194.
TCATGAAAGCAGT-3′ , reverse 5′ -AGATTACCCCTGGTG 2. Dolk, H., Busby, A., Armstrong, B.G. and Walls, P.H. (1998) Geographical
GAGA-3′ ; 2.6, forward 5′ -CGAGTGCTTAGCTGTCCTC-3′ , variation in anophthalmia and microphthalmia in England, 1988–94. BMJ,

317, 905– 909.
reverse 5′ -TCTTCTGAATCACCTCTCACTG-3′ ; 2.7, forward 3. Morrison, D., FitzPatrick, D., Hanson, I., Williamson, K., van Heyningen,
5′ -AGGATGAGGAAGAAGAGGAA-3′ , reverse 5′ -GTCTTC V., Fleck, B., Jones, I., Chalmers, J. and Campbell, H. (2002) National study
TGATGCTCCTCCAG-3′ ; 2.8, forward 5′ -GACCTTGGTAGA of microphthalmia, anophthalmia, and coloboma (MAC) in Scotland:
GGAGCTGA-3′ , reverse 5′ -ATCAGGGCTCATAACTCTG investigation of genetic aetiology. J. Med. Genet., 39, 16– 22.
4. Bermejo, E. and Martinez-Frias, M.L. (1998) Congenital eye
G-3′ . PCR products treated with Shrimp Alkaline Phosphatase malformations: clinical-epidemiological analysis of 1,124,654 consecutive
(GE Healthcare) and Exonuclease I (New England Biolabs) births in Spain. Am. J. Med. Genet., 75, 497– 504.
were directly sequenced using BigDye Terminator v1.1 sequen- 5. Stoll, C., Alembik, Y., Dott, B. and Roth, M.P. (1997) Congenital eye
cing chemistry (Applied Biosystems) using a 3730XL DNA malformations in 212,479 consecutive births. Ann. Genet., 40, 122– 128.
6. Shah, S.P., Taylor, A.E., Sowden, J.C., Ragge, N.K., Russell-Eggitt, I., Rahi,
Analysis System (Applied Biosystems). Sequences were J.S. and Gilbert, C.E. (2011) Anophthalmos, microphthalmos, and typical
analyzed using Sequencher v4.8 software (Gene Codes Corp.) coloboma in the United Kingdom: a prospective study of incidence and risk.
and compared with the reference sequence for SALL2 Invest Ophthalmol. Vis. Sci., 52, 558– 564.
(NM_005407.1). 7. Hornby, S.J., Dandona, L., Jones, R.B., Stewart, H. and Gilbert, C.E. (2003)
The familial contribution to non-syndromic ocular coloboma in south India.
Br. J. Ophthalmol., 87, 336– 340.
8. O’Rahilly, R. (1975) The prenatal development of the human eye. Exp. Eye
SUPPLEMENTARY MATERIAL Res., 21, 93– 112.
9. Gregory-Evans, C.Y., Williams, M.J., Halford, S. and Gregory-Evans, K.
Supplementary Material is available at HMG online. (2004) Ocular coloboma: a reassessment in the age of molecular
neuroscience. J. Med. Genet., 41, 881– 891.
10. Garne, E., Dolk, H., Loane, M. and Boyd, P.A. (2010) EUROCAT website
ACKNOWLEDGEMENTS data on prenatal detection rates of congenital anomalies. J. Med. Screen., 17,
97– 98.
We thank the family and all patients for participating in this 11. Yu, H., Smallwood, P.M., Wang, Y., Vidaltamayo, R., Reed, R. and Nathans,
study. We gratefully acknowledge the assistance of Wolfgang J. (2010) Frizzled 1 and frizzled 2 genes function in palate, ventricular
Engel, Institut für Humangenetik der Universität Göttingen septum and neural tube closure: general implications for tissue fusion
processes. Development, 137, 3707–3717.
for the provision of the Sall2 2/2 line. We thank the additional 12. Shah, S.P., Taylor, A.E., Sowden, J.C., Ragge, N., Russell-Eggitt, I., Rahi,
members of the GOSgene Committee (G.E. Moore, R. Kleta, J.S. and Gilbert, C.E. (2012) Anophthalmos, microphthalmos, and
B.G. Gaspar, M. Hubank, R.H. Scott), as well as members of Coloboma in the United kingdom: clinical features, results of investigations,
the GOSH Ophthalmology and Clinical Genetics Departments and early management. Ophthalmology, 119, 362–368.
(I. Russell-Eggitt, A. Kumar, A. Barnicoat, M. Lees, A. Male) 13. Ye, M., Berry-Wynne, K.M., Asai-Coakwell, M., Sundaresan, P., Footz, T.,
French, C.R., Abitbol, M., Fleisch, V.C., Corbett, N. and Allison, W.T., et al.
who contributed to the coloboma cohort. (2010) Mutation of the bone morphogenetic protein GDF3 causes ocular and
skeletal anomalies. Hum. Mol. Genet., 19, 287–298.
Conflict of Interest statement. None declared. 14. Ragge, N.K., Brown, A.G., Poloschek, C.M., Lorenz, B., Henderson, R.A.,
Clarke, M.P., Russell-Eggitt, I., Fielder, A., Gerrelli, D. and
Martinez-Barbera, J.P., et al. (2005) Heterozygous mutations of OTX2 cause
severe ocular malformations. Am. J. Hum. Genet., 76, 1008–1022.
FUNDING 15. Fantes, J., Ragge, N.K., Lynch, S.A., McGill, N.I., Collin, J.R.,
This research was supported by the Ulverscroft Foundation, the Howard-Peebles, P.N., Hayward, C., Vivian, A.J., Williamson, K., van
Heyningen, V. and Fitzpatrick, D.R. (2003) Mutations in SOX2 cause
Child Health Research Appeal Trust, Action Medical Research, anophthalmia. Nat. Genet., 33, 461–463.
the University of London Central Research Fund, the Rosetrees 16. Kelberman, D., Rizzoti, K., Avilion, A., Bitner-Glindzicz, M., Cianfarani,
Trust and the National Institute for Health Research (NIHR) S., Collins, J., Chong, W.K., Kirk, J.M., Achermann, J.C. and Ross, R., et al.
Biomedical Research Centre (BRC) at Great Ormond Street (2006) Mutations within Sox2/SOX2 are associated with abnormalities in
the hypothalamo-pituitary-gonadal axis in mice and humans. J. Clin. Invest,
Hospital for Children NHS Foundation Trust (GOSH) and Univer- 116, 2442–2455.
sity College London (UCL). J.C.S. is supported by the Great 17. Wallis, D.E., Roessler, E., Hehr, U., Nanni, L., Wiltshire, T., Richieri-Costa,
Ormond Street Hospital Children’s Charity. P.L.B. is supported A., Gillessen-Kaesbach, G., Zackai, E.H., Rommens, J. and Muenke, M.
by the Wellcome Trust. J.L. is supported by the Medical Research (1999) Mutations in the homeodomain of the human SIX3 gene cause
Council UK. L.I. was supported by an Action Medical Research holoprosencephaly. Nat. Genet., 22, 196–198.
18. Schimmenti, L.A., de la Cruz, J., Lewis, R.A., Karkera, J.D., Manligas, G.S.,
Training Fellowship. GOSgene at the UCL Institute of Child Roessler, E. and Muenke, M. (2003) Novel mutation in sonic hedgehog in
Health is supported by the NIHR BRC at GOSH and UCL. The non-syndromic colobomatous microphthalmia. Am. J. Med. Genet. A, 116A,
human embryonic and fetal material were provided by the Joint 215–221.
19. London, N.J., Kessler, P., Williams, B., Pauer, G.J., Hagstrom, S.A. and 39. Weston, C.R., Wong, A., Hall, J.P., Goad, M.E., Flavell, R.A. and Davis, R.J.
Traboulsi, E.I. (2009) Sequence alterations in RX in patients with (2003) JNK initiates a cytokine cascade that causes Pax2 expression and
microphthalmia, anophthalmia, and coloboma. Mol. Vis., 15, 162– 167. closure of the optic fissure. Genes Dev., 17, 1271– 1280.
20. Asai-Coakwell, M., French, C.R., Berry, K.M., Ye, M., Koss, R., Somerville, 40. Viringipurampeer, I.A., Ferreira, T., DeMaria, S., Yoon, J.J., Shan, X.,
M., Mueller, R., van Heyningen, V., Waskiewicz, A.J. and Lehmann, O.J. Moosajee, M., Gregory-Evans, K., Ngai, J. and Gregory-Evans, C.Y. (2012)
(2007) GDF6, a novel locus for a spectrum of ocular developmental Pax2 regulates a fadd-dependent molecular switch that drives tissue fusion
anomalies. Am. J. Hum. Genet., 80, 306– 315. during eye development. Hum. Mol. Genet., 21, 2357– 2369.
21. Casey, J., Kawaguchi, R., Morrissey, M., Sun, H., McGettigan, P., Nielsen, 41. Nornes, H.O., Dressler, G.R., Knapik, E.W., Deutsch, U. and Gruss, P.
J.E., Conroy, J., Regan, R., Kenny, E. and Cormican, P., et al. (2011) First (1990) Spatially and temporally restricted expression of Pax2 during murine
implication of STRA6 mutations in isolated anophthalmia, microphthalmia, neurogenesis. Development, 109, 797– 809.
and coloboma: a new dimension to the STRA6 phenotype. Hum. Mutat., 32, 42. de Celis, J.F. and Barrio, R. (2009) Regulation and function of Spalt proteins
1417– 1426. during animal development. Int. J. Dev. Biol., 53, 1385–1398.
22. Rainger, J., van, B.E., Ramsay, J.K., McKie, L., Al-Gazali, L., Pallotta, R., 43. Sanchez, J., Talamillo, A., Gonzalez, M., Sanchez-Pulido, L., Jimenez, S.,
Saponari, A., Branney, P., Fisher, M. and Morrison, H., et al. (2011) Loss of Pirone, L., Sutherland, J.D. and Barrio, R. (2011) Drosophila Sal and Salr are
the BMP antagonist, SMOC-1, causes Ophthalmo-acromelic (Waardenburg transcriptional repressors. Biochem. J., 438, 437 –445.

Anophthalmia) syndrome in humans and mice. PLoS Genet., 7, e1002114. 44. Sweetman, D. and Munsterberg, A. (2006) The vertebrate spalt genes in
23. Bar-Yosef, U., Abuelaish, I., Harel, T., Hendler, N., Ofir, R. and Birk, O.S. development and disease. Dev. Biol., 293, 285–293.
(2004) CHX10 mutations cause non-syndromic microphthalmia/ 45. Camp, E., Hope, R., Kortschak, R.D., Cox, T.C. and Lardelli, M. (2003)
anophthalmia in Arab and Jewish kindreds. Hum. Genet., 115, 302–309. Expression of three spalt (sal) gene homologues in zebrafish embryos. Dev.
24. Wang, L., He, F., Bu, J., Liu, X., Du, W., Dong, J., Cooney, J.D., Dubey, S.K., Genes Evol., 213, 35–43.
Shi, Y. and Gong, B., et al. (2012) ABCB6 mutations cause ocular coloboma. 46. Kohlhase, J., Heinrich, M., Schubert, L., Liebers, M., Kispert, A., Laccone,
Am. J. Hum. Genet., 90, 40–48. F., Turnpenny, P., Winter, R.M. and Reardon, W. (2002) Okihiro syndrome
25. Azuma, N., Yamaguchi, Y., Handa, H., Tadokoro, K., Asaka, A., Kawase, E. is caused by SALL4 mutations. Hum. Mol. Genet., 11, 2979– 2987.
and Yamada, M. (2003) Mutations of the PAX6 gene detected in patients 47. Kohlhase, J., Wischermann, A., Reichenbach, H., Froster, U. and Engel, W.
with a variety of optic-nerve malformations. Am. J. Hum. Genet., 72, (1998) Mutations in the SALL1 putative transcription factor gene cause
1565– 1570. Townes-Brocks syndrome. Nat. Genet., 18, 81–83.
26. Jamieson, R.V., Perveen, R., Kerr, B., Carette, M., Yardley, J., Heon, E., 48. Kohlhase, J., Taschner, P.E., Burfeind, P., Pasche, B., Newman, B., Blanck,
Wirth, M.G., van Heyningen, V., Donnai, D., Munier, F. and Black, G.C. C., Breuning, M.H., ten Kate, L.P., Maaswinkel-Mooy, P., Mitulla, B. et al.
(2002) Domain disruption and mutation of the bZIP transcription factor, (1999) Molecular analysis of SALL1 mutations in Townes-Brocks
MAF, associated with cataract, ocular anterior segment dysgenesis and syndrome. Am. J. Hum. Genet., 64, 435– 445.
coloboma. Hum. Mol. Genet., 11, 33–42.
49. Vodopiutz, J., Zoller, H., Fenwick, A.L., Arnhold, R., Schmid, M., Prayer,
27. Lalani, S.R., Safiullah, A.M., Fernbach, S.D., Harutyunyan, K.G., Thaller,
D., Muller, T., Repa, A., Pollak, A. and Aufricht, C., et al. (2013)
C., Peterson, L.E., McPherson, J.D., Gibbs, R.A., White, L.D. and Hefner,
Homozygous SALL1 mutation causes a novel multiple congenital
M., et al. (2006) Spectrum of CHD7 mutations in 110 individuals with
anomaly-mental retardation syndrome. J. Pediatr., 162, 612–617.
CHARGE syndrome and genotype-phenotype correlation. Am. J. Hum.
50. Baba, Y., Iida, A. and Watanabe, S. (2011) Sall3 plays essential roles in
Genet., 78, 303–314.
horizontal cell maturation through regulation of neurofilament expression
28. Bower, M., Salomon, R., Allanson, J., Antignac, C., Benedicenti, F., Benetti,
levels. Biochimie, 93, 1037–1046.
E., Binenbaum, G., Jensen, U.B., Cochat, P. and DeCramer, S., et al. (2012)
51. Botzenhart, E.M., Green, A., Ilyina, H., Konig, R., Lowry, R.B., Lo, I.F.,
Update of PAX2 mutations in renal coloboma syndrome and establishment
Shohat, M., Burke, L., McGaughran, J. and Chafai, R., et al. (2005) SALL1
of a locus-specific database. Hum. Mutat., 33, 457–466.
mutation analysis in Townes-Brocks syndrome: twelve novel mutations and
29. Gu, H., Li, D., Sung, C.K., Yim, H., Troke, P. and Benjamin, T. (2011)
DNA-binding and regulatory properties of the transcription factor and expansion of the phenotype. Hum. Mutat., 26, 282.
putative tumor suppressor p150(Sal2). Biochim. Biophys. Acta, 1809, 52. Nishinakamura, R., Matsumoto, Y., Nakao, K., Nakamura, K., Sato, A.,
276–283. Copeland, N.G., Gilbert, D.J., Jenkins, N.A., Scully, S. and Lacey, D.L., et al.
30. Maquat, L.E. (2005) Nonsense-mediated mRNA decay in mammals. J. Cell (2001) Murine homolog of SALL1 is essential for ureteric bud invasion in
Sci., 118, 1773– 1776. kidney development. Development, 128, 3105– 3115.
31. Schoenberg, D.R. and Maquat, L.E. (2012) Regulation of cytoplasmic 53. Hornby, S.J., Ward, S.J., Gilbert, C.E., Dandona, L., Foster, A. and Jones,
mRNA decay. Nat. Rev. Genet., 13, 246–259. R.B. (2002) Environmental risk factors in congenital malformations of the
32. Kohlhase, J., Schuh, R., Dowe, G., Kuhnlein, R.P., Jackle, H., Schroeder, B., eye. Ann. Trop. Paediatr., 22, 67–77.
Schulz-Schaeffer, W., Kretzschmar, H.A., Kohler, A. and Muller, U., et al. 54. Chang, L., Blain, D., Bertuzzi, S. and Brooks, B.P. (2006) Uveal coloboma:
(1996) Isolation, characterization, and organ-specific expression of two clinical and basic science update. Curr. Opin. Ophthalmol., 17, 447–470.
novel human zinc finger genes related to the Drosophila gene spalt. 55. Cukras, C., Gaasterland, T., Lee, P., Gudiseva, H.V., Chavali, V.R.,
Genomics, 38, 291– 298. Pullakhandam, R., Maranhao, B., Edsall, L., Soares, S. and Reddy, G.B.,
33. Sato, A., Matsumoto, Y., Koide, U., Kataoka, Y., Yoshida, N., Yokota, T., et al. (2012) Exome analysis identified a novel mutation in the rbp4 gene in a
Asashima, M. and Nishinakamura, R. (2003) Zinc finger protein sall2 is not consanguineous pedigree with retinal dystrophy and developmental
essential for embryonic and kidney development. Mol. Cell Biol., 23, 62– 69. abnormalities. PLoS ONE, 7, e50205.
34. Bohm, J., Buck, A., Borozdin, W., Mannan, A.U., Matysiak-Scholze, U., 56. Aldahmesh, M.A., Khan, A.O., Hijazi, H. and Alkuraya, F.S. (2013)
Adham, I., Schulz-Schaeffer, W., Floss, T., Wurst, W., Kohlhase, J. and Mutations in ALDH1A3 cause microphthalmia. Clin. Genet., 84, 128 –131.
Barrionuevo, F. (2008) Sall1, sall2, and sall4 are required for neural tube 57. Fares-Taie, L., Gerber, S., Chassaing, N., Clayton-Smith, J., Hanein, S.,
closure in mice. Am. J. Pathol., 173, 1455– 1463. Silva, E., Serey, M., Serre, V., Gerard, X. and Baumann, C., et al. (2013)
35. Hero, I. (1990) Optic fissure closure in the normal cinnamon mouse. An ALDH1A3 mutations cause recessive anophthalmia and microphthalmia.
ultrastructural study. Invest. Ophthalmol. Vis. Sci., 31, 197–216. Am. J. Hum. Genet., 92, 265 –270.
36. Torres, M., Gomez-Pardo, E. and Gruss, P. (1996) Pax2 contributes to inner 58. Roos, L., Fang, M., Dali, C., Jensen, H., Christoffersen, N., Wu, B., Zhang, J.,
ear patterning and optic nerve trajectory. Development, 122, 3381–3391. Xu, R., Harris, P. and Xu, X., et al. (2013) A homozygous mutation in a
37. Tsuji, N., Kita, K., Ozaki, K., Narama, I. and Matsuura, T. (2012) consanguineous family consolidates the role of ALDH1A3 in autosomal
Organogenesis of mild ocular coloboma in FLS mice: failure of basement recessive microphthalmia. Clin. Genet, doi: 10.1111/cge.12277.
membrane disintegration at optic fissure margins. Exp. Eye Res., 94, 59. Bakrania, P., Efthymiou, M., Klein, J.C., Salt, A., Bunyan, D.J., Wyatt, A.,
174–178. Ponting, C.P., Martin, A., Williams, S. and Lindley, V., et al. (2008)
38. Barbieri, A.M., Broccoli, V., Bovolenta, P., Alfano, G., Marchitiello, A., Mutations in BMP4 cause eye, brain, and digit developmental anomalies:
Mocchetti, C., Crippa, L., Bulfone, A., Marigo, V., Ballabio, A. and Banfi, S. overlap between the BMP4 and hedgehog signaling pathways. Am. J. Hum.
(2002) Vax2 inactivation in mouse determines alteration of the eye Genet., 82, 304 –319.
dorsal-ventral axis, misrouting of the optic fibres and eye coloboma. 60. Voronina, V.A., Kozhemyakina, E.A., O’Kernick, C.M., Kahn, N.D.,
Development, 129, 805– 813. Wenger, S.L., Linberg, J.V., Schneider, A.S. and Mathers, P.H. (2004)
Mutations in the human RAX homeobox gene in a patient with anophthalmia 67. Ma, Y., Li, D., Chai, L., Luciani, A.M., Ford, D., Morgan, J. and Maizel, A.L.
and sclerocornea. Hum. Mol. Genet., 13, 315– 322. (2001) Cloning and characterization of two promoters for the human HSAL2
61. Wyatt, A., Bakrania, P., Bunyan, D.J., Osborne, R.J., Crolla, J.A., Salt, A., gene and their transcriptional repression by the Wilms tumor suppressor
Ayuso, C., Newbury-Ecob, R., Abou-Rayyah, Y. and Collin, J.R., et al. (2008) gene product. J. Biol. Chem., 276, 48223–48230.
Novel heterozygous OTX2 mutations and whole gene deletions in 68. Pincheira, R., Baerwald, M., Dunbar, J.D. and Donner, D.B. (2009) Sall2 is a
anophthalmia, microphthalmia and coloboma. Hum. Mutat., 29, E278–E283. novel p75NTR-interacting protein that links NGF signalling to cell cycle
62. Zahrani, F., Aldahmesh, M.A., Alshammari, M.J., Al-Hazzaa, S.A. and progression and neurite outgrowth. EMBO J., 28, 261– 273.
Alkuraya, F.S. (2013) Mutations in c12orf57 cause a syndromic form of 69. Pincheira, R. and Donner, D.B. (2008) The Sall2 transcription factor is a
colobomatous microphthalmia. Am. J. Hum. Genet., 92, 387–391. novel p75NTR binding protein that promotes the development and function
63. Otteson, D.C., Shelden, E., Jones, J.M., Kameoka, J. and Hitchcock, P.F. of neurons. Ann. N. Y. Acad. Sci., 1144, 53–55.
(1998) Pax2 expression and retinal morphogenesis in the normal and Krd 70. Harada, C., Harada, T., Nakamura, K., Sakai, Y., Tanaka, K. and Parada,
mouse. Dev. Biol., 193, 209 –224.
L.F. (2006) Effect of p75NTR on the regulation of naturally occurring
64. Gregory-Evans, C.Y., Wallace, V.A. and Gregory-Evans, K. (2013) Gene
cell death and retinal ganglion cell number in the mouse eye. Dev. Biol.,
networks: dissecting pathways in retinal development and disease. Prog.
290, 57– 65.
Retin. Eye Res., 33, 40–66.
71. Li, D., Tian, Y., Ma, Y. and Benjamin, T. (2004) p150(Sal2) is a

65. Skalicky, S.E., White, A.J., Grigg, J.R., Martin, F., Smith, J., Jones, M.,
Donaldson, C., Smith, J.E., Flaherty, M. and Jamieson, R.V. (2013) p53-independent regulator of p21(WAF1/CIP). Mol. Cell Biol., 24,
Microphthalmia, Anophthalmia, and Coloboma and Associated Ocular and 3885– 3893.
Systemic Features: Understanding the Spectrum. JAMA Ophthalmol., 131, 72. Walther, C. and Gruss, P. (1991) Pax-6, a murine paired box gene, is
1517– 1524. expressed in the developing CNS. Development, 113, 1435–1449.
66. Lee, J., Lee, B.K. and Gross, J.M. (2013) Bcl6a function is required during 73. Lennon, G., Auffray, C., Polymeropoulos, M. and Soares, M.B. (1996) The
optic cup formation to prevent p53-dependent apoptosis and colobomata. I.M.A.G.E. Consortium: an integrated molecular analysis of genomes and
Hum. Mol. Genet., 22, 3568– 3582. their expression. Genomics, 33, 151–152.

Mutation of SALL2 Causes Recessive Ocular Coloboma in Humans and Mice

Uploaded by

Copyright:

Available Formats

Mutation of SALL2 Causes Recessive Ocular Coloboma in Humans and Mice

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Mutation of SALL2 Causes Recessive Ocular Coloboma in Humans and Mice

Uploaded by

Copyright:

Available Formats

Human Molecular Genetics, 2014, Vol. 23, No.

Mutation of SALL2 causes recessive ocular

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

# The Author 2014. Published by Oxford University Press.

INTRODUCTION 600725) (18), RAX (OMIM 601881) (19), GDF3 (OMIM

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Table 1. Summary of ocular phenotype of affected individuals

Patient IV:1 IV:2 IV:3

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Stage Genotype Anterior Midline Posterior Total eyes

E13.5 Sall22/2 12 (100%) 0 (0%) 6 (50%) 6 (50%) 0 (0%) 12 (100%) 12

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

three-independent experiments for each antisense probe; sense Plasmid constructs

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

Downloaded from https://academic.oup.com/hmg/article/23/10/2511/612937 by guest on 19 January 2021

You might also like