We have recently shown that Okihiro syndrome results from mutation in the putative zinc finger transcription factor gene SALL4 on chromosome 20q13.13-13.2. There is considerable overlap of clinical features of Okihiro syndrome with other conditions, most notably Holt-Oram syndrome, a condition in part resulting from mutation of the TBX5 locus, as well as acro-renal-ocular syndrome. We analysed further families/patients with the clinical diagnosis of Holt-Oram syndrome and acro-renal-ocular syndrome for SALL4 mutations. We identified a novel SALL4 mutation in one family where the father was originally thought to have thalidomide embryopathy and had a daughter with a similar phenotype. We also found two novel mutations in two German families originally diagnosed as Holt-Oram syndrome and a further mutation in one out of two families carrying the diagnosis acro-renal-ocular syndrome. Our results show that some cases of “thalidomide embryopathy” might be the result of SALL4 mutations, resulting in an increased risk for similarly affected offspring. Furthermore we confirm the overlap of acro-renal-ocular syndrome with Okihiro syndrome at the molecular level and expand the phenotype of SALL4 mutations.
- Okihiro syndrome
- Holt-Oram syndrome
- acro-renal-ocular syndrome
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
The SALL gene family, derived from the Drosophila gene spalt,1 represents a related group of probable zinc finger transcription factors. Four such genes have been identified to date in man, of which SALL22 and SALL33 remain to be associated with human disease. However, the importance of this gene family to human malformation syndromes is indisputable, mutations at SALL1 on chromosome 16q12.1 being associated with Townes-Brocks syndrome and related phenotypes4,5 and, more recently, mutations at SALL4 being shown to be causative in patients with Okihiro syndrome.6,7 Kohlhase et al7 reported frameshift and nonsense mutations in five of eight families segregating the Okihiro syndrome phenotype. A further report6 identified two frameshift mutations and one nonsense mutation in three affected kindreds, including the family reported by Okihiro et al.8
Okihiro syndrome (OMIM 607323) refers to an autosomal dominant condition characterised by radial malformations associated with Duane congenital eye movement disorder. The Duane eye anomaly is characterised by limitation of abduction and narrowing of the palpebral fissure with retraction of the globe on adduction. Okihiro syndrome has also been described by other authors and is known by other names in published reports. Temtamy et al9 used the term DR syndrome, D representing Duane anomaly and deafness, R identifying the radial and renal manifestations of the condition.9 The first report is likely to have been that of Crisp.10
Familial occurrence of radial sided hand malformations, noteworthy for their variability, in association with Duane eye anomaly have been the central clinical features of several reports.10–13 The range of associated features observed in these patients is extensive, embracing anal stenosis, pigmentary disturbance, hearing deficit, renal malformations, external ear malformations, facial asymmetry, and cardiac lesions, particularly atrial septal defect. In reporting mutations of the SALL4 gene as the basis of Okihiro syndrome in their patients, Kohlhase et al7 similarly drew attention to associated clinical features of choanal atresia, triphalangeal thumb, renal agenesis, unilateral deafness, thumb reduplication, external ear malformations, and ventricular septal defect in their cohort.7 Not only do these clinical observations confirm that there is an extraordinary variability in phenotype in these patients with allelic conditions, but also prompt consideration as to what overlapping clinical entities, previously considered on a clinical basis to represent distinct conditions, might also be caused by mutation at the SALL4 locus.
In reporting SALL4 mutations in families with Okihiro syndrome, Kohlhase et al7 identified a number of possible candidate clinical conditions which might be tested for allelic mutations. Foremost among those conditions thus identified was Holt-Oram syndrome, an autosomal dominant disorder characterised by radial forelimb malformations, congenital heart abnormalities, and associated features.11 Although a major locus for Holt-Oram syndrome has been identified at TBX5, only approximately 30% of clinically identified cases of Holt-Oram syndrome have been found to have mutations at that locus.12 Attention was also drawn to a previous cytogenetic report of a de novo pericentric inversion of chromosome 20q13.2, where SALL4 is located, associated with a clinical presentation of bilateral absence of the thumbs and an atrial septal defect.13 This phenotype was taken to represent Holt-Oram syndrome and the suggestion was made that there were likely to be other cases of apparent Holt-Oram syndrome but without TBX5 mutation, in whom SALL4 mutations might be expected to be observed.7 Similarly, other candidate clinical conditions which might represent mutation at the SALL4 locus were suggested, comprising acro-renal-ocular syndrome14 and patients previously thought to represent thalidomide embryopathy but in whose offspring limb and cardiac malformations had also been observed.15
The purpose of this report is to communicate our initial experience of SALL4 mutation analysis across a much wider range of patients representing several different clinical diagnoses which overlap with Okihiro syndrome. In doing so we identified three new mutations and one recurrent mutation also reported by Al-Baradie et al6 at this locus and confirm our hypothesis that mutation of SALL4 results in a range of clinical conditions, hitherto considered to represent distinct entities.
MATERIALS AND METHODS
Venous blood was collected from patients and unaffected relatives after obtaining their informed consent.
Genomic DNA was prepared from peripheral lymphocytes by routine procedures. Mutation analysis of SALL4 exons 1–4 (complete coding region) was performed as described previously.7 GenBank accession numbers were SALL4 cDNA, NM_020436; genomic contig NT_011362.
In family 1, the index patient (II.1, fig 1A) showed bilateral absent thumbs, absent radii, and shortened humeri, the left arm being more severely shortened.14 She has a bilateral Duane anomaly, fine nystagmus, and slightly dysplastic optic discs. A mild bilateral conductive hearing loss was noted. Her affected mother (I.1, fig 1A) showed absent thumbs and shortened forearms. She has a hypoplastic dystopic left kidney and a moderate conductive and significant sensorineural hearing loss.
In family 2, the index patient (II.1, fig 1B) was born with bilateral asymmetrical radial ray defects consisting of absence of the thumb and first fingers, radial aplasia, and ulnar hypoplasia. Her father is similarly affected and also suffers from a left sided sensorineural and right sided mixed hearing loss. Both have structurally normal hearts with normal echo and normal ECG. Renal ultrasound and eye examination showed no abnormality. Thalidomide embryopathy was diagnosed in the father in childhood based on the nature of the abnormalities in the presence of a history of possible exposure during pregnancy. Occurrence of abnormalities in a subsequent generation was attributed to mutagenicity of thalidomide.15
In family 3, the affected boy (II.1, fig 1C) was born to unaffected non-consanguineous parents. A left thumb and radial hypoplasia together with a right sided club hand and aplasia of the thumb and radius was noted. Further abnormalities were a persistent foramen ovale and a crossed dystopia of the kidneys. He showed a simian crease on the left hand, deep creases of the feet, broadened and slightly elongated first toes, elongated fourth toes, mild epicanthic folds, and a broad flat nasal bridge. Psychomotor development was normal at the age of 2 years. Growth was on the 90th centile, weight on the 3rd, and head circumference on the 50th centile. An initial diagnosis of Holt-Oram syndrome was suggested, but after SALL4 mutation detection and re-examination, a bilateral Duane anomaly was noted and the diagnosis revised to Okihiro syndrome.
In family 4, the affected girl (II.1, fig 1D) showed bilateral thumb aplasia and radial club hand (fig 2G, H). Both arms were shortened. On the left side, all carpal bones were present and the radius was hypoplastic. On the right, a subtotal aplasia of the radius was noted together with a malformed olecranon. Further abnormalities were a VSD, hypertelorism with slight epicanthic folds, slit-like openings of the external auditory meatus (fig 2C), flat feet with large sandal gaps, and broad, short first toes. She had no kidney malformation or Duane anomaly. The initial diagnosis was Holt-Oram syndrome. Her father (I.2, fig 1D) showed less severe limb malformations, with bilateral thumb aplasia and hypoplastic radii (fig 2E, F). On re-examination, a right sided Duane anomaly was noted. He also had slit-like openings of the external auditory meatus (fig 2D), flat feet with large sandal gaps, and broad, short first toes. Facial abnormalities were not seen. Further anomalies include a bifid uvula, horseshoe kidney, and limited movements of the joints of fingers 2–5 bilaterally. Following mutation analysis and detection of the Duane anomaly in the father, a revised diagnosis of Okihiro syndrome was made in this family. The phenotypic features of all participating patients are summarised in table 1.
SALL4 mutation analysis was carried out in two families presenting with overlapping features of Okihiro syndrome and acro-renal-ocular syndrome.14 In one of these families (family 1 in this report and in the report of Becker et al14) we found the heterozygous nonsense mutation c.2593C>T (R865X) in exon 3 of the SALL4 gene. The mutation was detected in both affected family members (fig 1A).
We also analysed three families with the clinical diagnosis of Holt-Oram syndrome. Two of these families were negative for a TBX5 mutation and one had not been tested. Family 2 had been described previously as a possible example of thalidomide mutagenicity.15 Because of the clinical overlap with Holt-Oram syndrome, TBX5 mutation analysis was performed, but no mutation was detected. We found the heterozygous frameshift mutation c.326delC in exon 2 of SALL4 in both affected persons (father and daughter), resulting in a premature stop codon 31 bp 3′ of the mutation (fig 1B). In family 3, both parents are unaffected. Their only son was diagnosed with Holt-Oram syndrome, but was negative for a TBX5 mutation. We detected the heterozygous nonsense mutation c.523A>T (K175X) in exon 2 of SALL4 (fig 1C) in the patient, but not the parents. Family 4 had also been diagnosed with Holt-Oram syndrome. Here, the father and his daughter are affected. We identified the heterozygous nonsense mutation c.1849C>T (R617X) in exon 2 of SALL4. The mutation segregates with the abnormal phenotype in both affected subjects (fig 1D).
In this report we show that SALL4 mutations cause a range of phenotypes. Three out of four families reported here show that kidney anomalies are part of the phenotypic spectrum of SALL4 mutations, thereby confirming the hypothesis that Okihiro syndrome and acro-renal-ocular syndrome are allelic to some extent. However, we did not find a mutation in family 2 reported by Becker et al14 leaving open the possibility that another locus might be involved in this family.
Our results further show that in some cases reported as possibly resulting from thalidomide mutagenicity, SALL4 mutations could be the true cause of the phenotype. The mutation in family 2 reported here (case 2 in the report of McBride15) was found in both affected family members. The heritability of the abnormal phenotype in this pedigree has been previously represented as evidence for thalidomide as a mutagen. However, in view of the extended phenotype of limb malformations we now observe with SALL4 mutations, it appears much more likely that the phenotype in generation I of this family was misdiagnosed as representing thalidomide embryopathy when the phenotype was in fact the result of a mutation of SALL4.
The families described here also expand the phenotype of SALL4 mutations. We observed facial characteristics (hypertelorism, epicanthic folds) in two of our patients and not previously described slit-like openings of the external auditory meatus, which might help to discriminate Okihiro syndrome from Holt-Oram syndrome if no Duane anomaly is present. We also describe characteristic flat feet and large sandal gaps in both affected members of one family.
Our results suggest that all patients with Duane anomaly and radial ray malformations should be investigated for SALL4 mutations. Furthermore, as shown by our cohort, patients with typical radial ray malformations without Duane anomaly might carry a SALL4 mutation as well. These patients obviously might be diagnosed with Holt-Oram syndrome instead and require TBX5 mutation analysis. However, if TBX5 mutation analysis fails to show a mutation, SALL4 analysis should be considered. Our results further suggest that in such cases all affected family members should be clinically examined for Duane anomaly and other features associated with Okihiro syndrome because only one of the affected family members might have a Duane anomaly.
Together with this report, in total 11 different SALL4 mutations have now been described.6,7 These are one small insertion (c.940-941insC), five deletions (c.326delC, c.842delG, c.1053delG, c.1904delT, c.2425delG), and five nonsense mutations (c.523A>T, c.1849C>T, c.1954C>T, c.2288T>A, c.2593C>T).
Compared to the mutational spectrum of SALL1 mutations,5 most of which reside in exon 2, 5′ of the region encoding the first double zinc finger domain, the SALL4 mutations detected seem to be distributed over the gene.
It is not possible to correlate the severity of the phenotype with mutation position. Severe limb malformations with shortening of the upper limbs have been described for the mutations c.523A>T, c.842delG, c.1849C>T, and c.1904delT. Milder malformations have been seen with the mutations c.940-941insC, c.1053delG, and c.2405delG, and the mildest (thenar hypoplasia only) with c.1954C>T. The observation of a strong limb phenotype in family 1 of this report and a mild phenotype in pedigree DA in the report of Al-Baradie et al6 with the same causative SALL4 mutation c.2593C>T suggests that the phenotypic variability in these two families is the result of differences in the genetic background. With respect to the pathogenesis of the anomalies caused by SALL4 mutations, the nature of these mutations, all causing premature termination of translation, would suggest haploinsufficiency Also, the similar phenotype of patients with cytogenetic deletions of chromosome 20q probably encompassing the SALL4 locus would be consistent with this.16,17 However, the fact that the SALL1 knock-out mouse shows only kidney defects but no TBS-like phenotype18 raises serious doubts about the assumption that SALL1 mutations which are exclusively truncating5 cause Townes-Brocks syndrome via haploinsufficiency. An alternative explanation which is being discussed now is that truncating SALL1 mutations could lead to the TBS phenotype by a dominant negative action,19 with truncated proteins interfering with nuclear transport of the wild type proteins. This is thought to result from dimerisation of wild type and mutant proteins mediated by the evolutionarily highly conserved glutamine rich domain within the amino-terminal part of all known SAL-like proteins (fig 3).1–3,20–27 With respect to SALL4, the mutations c.326delC and c.523A>T show that the Glu rich domain (encoded by nucleotides 661–714 of the SALL4 cDNA) cannot be crucial for a possible dominant negative action since the domain would be absent from truncated proteins resulting from these mutations.
We would like to thank the families for their participation. JK would also like to thank W Engel for review and support. WR is supported by The Children’s Medical and Research Foundation at Our Lady’s Hospital for Sick Children. This work was funded by the Wilhelm-Sander-Stiftung (grant 98.075.2 to JK).