Article Text


A gene locus for branchio-otic syndrome maps to chromosome 14q21.3-q24.3
  1. R G Ruf1,
  2. J Berkman2,
  3. M T F Wolf1,
  4. P Nurnberg3,4,
  5. M Gattas2,
  6. E-M Ruf5,
  7. V Hyland6,
  8. J Kromberg2,
  9. I Glass7,
  10. J Macmillan2,
  11. E Otto1,
  12. G Nurnberg3,
  13. B Lucke,
  14. H C Hennies,
  15. F Hildebrandt1
  1. 1Departments of Pediatrics and Human Genetics, University of Michigan, Ann Arbor, USA
  2. 2Queensland Clinical Genetics Service, Royal Children’s Hospital, Brisbane, Queensland, Australia
  3. 3Gene Mapping Centre and Department of Molecular Genetics, Max-Delbrueck Centre for Molecular Medicine, Berlin-Buch, Germany
  4. 4Institute of Medical Genetics, Charité University Hospital, Humboldt University, Berlin, Germany
  5. 5University Children’s Hospital, Freiburg, Germany
  6. 6Molecular Genetics Laboratory, Queensland Health Pathology Service, Queensland, Australia
  7. 7Departments of Pediatrics and Medicine, University of Washington School of Medicine, Seattle, USA
  1. Correspondence to:
 Dr F Hildebrandt, University of Michigan Health System, 8220C MSRB III, 1150 West Medical Center Drive, Ann Arbor, MI 48109, USA; 

Statistics from

Branchio-oto-renal syndrome (BOR, OMIM 113650) is an autosomal dominant disorder characterised by the association of hearing loss (HL), structural ear anomalies, branchial arch defects, and renal anomalies.1 The prevalence approximates 1:40 000 in the general population, and has been reported in about 2% of deaf children.2 Age of onset for deafness varies from childhood to early adulthood.3 The clinical expression of BOR exhibits wide intra- and interfamilial variability. In addition, reduced penetrance for BOR has been assumed.4 The major feature of BOR, which occurs in 93% of patients, is HL, which can be conductive, sensorineural, or mixed. Besides the classical ear, kidney, and branchial arch anomalies, different developmental manifestations of BOR in other organ systems have been described. Among these, dysfunction of the lacrimal duct system is a common association.5–10 Thus, BOR represents a clinically and genetically heterogeneous disease complex that manifests predominantly during organogenesis. A gene locus for autosomal dominant BOR had been localised on chromosome 8q13.11,12 Subsequently, mutations in the human homologue of the Drosophila eyes absent gene (EYA1) have been shown to be causative for BOR (OMIM 601653).13 Branchio-otic syndrome (BOS) (OMIM 602588) was initially described as a disorder distinct from BOR, featuring the same clinical symptoms as BOR with the exception of renal anomalies.1 The large variety of clinical phenotypes and the description of mutations in the EYA1 gene for BOR and BOS patients13–15 show that BOR and BOS can represent allelic defects of the EYA1 gene. The identification of a second gene locus in a large BOS pedigree on chromosome 1q31 established the presence of genetic locus heterogeneity for BOS.4 No linkage to this locus has been published for BOR families and the gene defect is still to be identified. The issue of genotype/phenotype relationships regarding clinical features of BOR or BOS remains unsolved. We describe here a genome wide search for linkage in a large pedigree with BOS, in which linkage to the EYA1 locus on chromosome 8q13 had been excluded, resulting in a new locus (BOS3) on chromosome 14q.

Key points

  • Branchio-oto-renal syndrome (BOR) is an autosomal dominant developmental disorder characterised by the association of hearing loss, branchial arch defects, and renal anomalies. Branchio-otic syndrome (BOS) represents a related disorder presenting with the same clinical features without renal anomalies.

  • Recessive mutations in the human homologue of the Drosophila eyes absent gene (EYA1) have been shown to cause BOR and BOS. A locus (BOS2) for autosomal dominant BOS has been localised to chromosome 1q31.

  • We performed a genome wide search for linkage in a large pedigree with BOS with more than 40 affected subjects and mapped a new gene locus (BOS3) to chromosome 14q21.3-q24.3. The highest multipoint lod score was Zmax=4.81 (θ=) for marker D14S980.

  • Identification of the gene causing branchio-otic syndrome type 3 will offer new insights into the development and molecular mechanisms of hearing.


Blood samples and clinical data for a large multigeneration family with over 40 affected subjects with BOS were obtained after informed consent was given by patients and unaffected relatives. The ethnic origin of the family was Anglo-Saxon Australian. Clinical examinations and renal ultrasound were performed in 32 affected family members. Twenty blood samples were collected (14 from affected subjects, six from unaffected relatives or partners) and DNA was extracted for molecular analysis. All 32 affected subjects had deafness (100%). In 17 affected family members, precise audiometric data were available. Diagnosis was sensorineural HL in 14 of them (82%) and mixed HL in three of them (18%). The affected frequencies varied from low to high frequencies as well as the presence of HL in all frequencies. Severity varied from mild to severe HL being still progressive in six cases. Among subjects IV.14, IV.17, and V.2 differences in the HL between the right and left ear were found. In IV.14 sensorineural HL was mild in the right ear whereas it was moderate to severe in the left ear. In IV.17, in addition to moderate to severe sensorineural HL on both sides, moderate to severe conductive HL was present only on the left side. In V.2 high frequency HL in the right ear differed from low frequency in the left ear. Age of onset was very variable with an average of 9.5 years, ranging from 3 weeks to 22 years. Eight subjects (25%) had branchial arch defects, three with branchial cysts, and six with ear pits as external ear manifestation (table 1). In three affected subjects (9%) lacrimal duct stenosis was diagnosed as a common feature associated with BOR/BOS. No congenital renal anomalies were found, although two adult affected sibs had renal carcinomas, which most likely was coincidental. The absence of congenital renal anomalies suggests a diagnosis of BOS rather than BOR in this family. Genomic DNA was isolated, by standard methods,16 either directly from blood samples or after Epstein-Barr-virus transformation of peripheral blood lymphocytes.17 DNA was available for haplotype analysis in 14 affected and six unaffected subjects for the genome wide search for linkage. In the other subjects haplotypes were inferred if possible (fig 1). A total of 380 microsatellite markers from the Genethon final linkage map18 with an average spacing of 11 cM were used. For further fine mapping on chromosome 14q21.3-q24.3, six additional markers, with an average distance of 3.5 cM, were used. Order and sex averaged distances (in parentheses) between these markers from centromeric to q telomeric are as follows: D14S599 (2.9 cM), D14S306 (2.8 cM), D14S1013 (3.2 cM), D14S748 (4.5 cM), D14S587 (4.1 cM), D14S980 (2.9 cM), D14S274 (3.9 cM), D14S592 (8.0 cM), D14S588 (2.9 cM), D14S1002 (4.5 cM), D14S1025 (3.7 cM), D14S53 (5.1 cM), and D14S606.18 Semiautomated genotyping was performed with a MegaBACE-1000 analysis system. Data were analysed by Genetic Profiler Software, version 1.1. Two point lod score calculations were performed by the LINKAGE program package,19 with the help of the LINKRUN computer program (T F Wienker, unpublished data), using an autosomal dominant model with 100% penetrance and a gene frequency for BOS of 0.0001. The “lodmax - 1 support interval” was defined as the genetic map positions intersecting the lod score curve at Zmax=1.20 For haplotyping and computation of multipoint lod scores, the program SIMWALK21 was used, assuming equal allele frequencies. Because of the reduced penetrance described in BOR/BOS, the calculations were performed on basis of an “affecteds only” strategy.

Table 1

Clinical data of affected subjects from the BOS kindred

Figure 1

Haplotypes on chromosome 14q12-q23 of the BO family. Haplotypes are shown for the subjects where DNA was available (indicated by an arrow) or haplotypes could be inferred. Thirteen microsatellite markers are shown on the left from cen to qter (top to bottom). Filled upper right quadrant indicates diagnosis of hearing loss, filled lower right quadrant ear pits, filled upper left quadrant lacrimal duct stenosis, and filled lower left quadrant branchial cysts. Haplotypes are interpreted as differently coloured bars. Paternal haplotypes are drawn to the left, maternal ones to the right. Segments of haplotypes which could not unambiguously be assigned to the paternal or maternal haplotype are represented by a thin line. Inferred haplotypes are indicated in parentheses. The black haplotype cosegregates with the affected status. Note that marker D14S1013 is flanking the BOS3 locus on its centromeric borders, as defined through a recombination in II.2, and that marker D14S53 is flanking the BOS3 locus at its q terminal border as defined by a recombination in V.5. Flanking markers are underlined.


Before starting the genome wide search for linkage, the EYA1 gene locus was excluded by linkage and mutational analysis. By evaluating the results of the genome wide search, the locus for BOS on chromosome 1q31 was also excluded for this kindred (data not shown). From the total genome search for linkage, only for one locus was cosegregation of the haplotype pattern in all affected subjects found for markers D14S587, D14S592, and D14S588 on chromosome 14q21.3-q24.3, yielding a maximum two point lod score of Zmax=3.27 (θ=0) for marker D14S587 (table 1). Further fine mapping with an additional six markers confirmed the locus. Haplotype analysis showed clear evidence that the disease allele cosegregated with all affected subjects and was absent from unaffected subjects (fig 1). A recombination event in II.2 defined marker D14S1013 as proximally flanking, and a recombination in V.5 identified marker D14S53 as distally flanking the critical genetic region within a 37.7 cM interval on chromosome 14q21.3-q24.3. Multipoint analysis of the 11 markers resulted in a Zmax=4.81 at marker D14S980 at relative position 50.9 (fig 2). The 95% confidence interval at Zmax=120 extends over a 33.9 cM interval between the markers D10S1013 and D10S53 within the set of 11 microsatellites. Marker D14S980 also showed the highest two point lod score value Zmax=4.11 (θ=0) (table 2).

Table 2

Two point lod scores generated in the BOS kindred at various recombination fractions for markers at the BOS3 locus

Figure 2

Multipoint lod scores for the BOS3 locus on chromosome 14q21.3-q24.3 versus the 13 markers shown in fig 1. Relative position is given in cM according to the Genethon map.18 The two markers D14S1013 and D14S53 flanking the BOS3 region (fig 1) are underlined. cen, centromeric orientation; qter, q terminal orientation.


Here we have reported a third gene locus for BOS, BOS3, which maps to chromosome 14q21.3-q24.3. According to the UCSC Genome Browser, the interval between markers D14S1013 and D14S53 spans a physical distance of approximately 33 Mb, relative marker positions are 41 383 995 and 74 328 130, respectively. A recombination in the healthy subject III.8 could define marker D14S587 as proximally flanking. As reduced penetrance for BOS is known, this does not represent a secure border. Further fine mapping with more affected members of this pedigree and examination of other familial cases with BOS will help to refine this region. In contrast to the pedigree described by Kumar et al4 linked to chromosome 1q31, where HL was diagnosed in 50% of the affected subjects, deafness seems to be a major feature in this pedigree. As 25% of the patients show an association with branchial arch defects, a non-syndromic form of deafness is unlikely. The diagnosis of lacrimal duct stenosis, a common association of BOR and BOS, further confirms the diagnosis of BOS. The HL varied in form, severity, frequency, and the age of onset among the different family members and even between the ears of one patient, a characteristic feature of BOR and BOS. The low percentage of branchial arch defects compared to previously described families with BOS and BOR can be explained either by the known variable expressivity or by the genetic heterogeneity of BOS.

Genes encoding proteins involved in renal and otic morphogenesis and organogenesis are good candidates. EYA1 deficient mice have been shown to lack ears and kidneys and show abnormal apoptosis of organ primordia.22 Another member of the EYA gene family, EYA4, is responsible for late onset deafness.23EYA2 and EYA3 are excluded from the BOS3 locus on chromosome 14, as they are localised on chromosomes 20 and 1, respectively. Gene loci for non-syndromic forms of autosomal dominant24 and autosomal recessive deafness25 were mapped to the critical interval on chromosome 14q21.3-q24.3. Whether the autosomal dominant form is an allelic variant of BOS requires the identification of the causative gene. Identification of the gene causing BOS3 in this pedigree and other patients with BOS will lead to new insights into the pathophysiology and development of auditory function.


We thank all the patients, their family members, and their physicians for their participation in this study. FH was supported by a grant from the German Research Foundation (SFB 592).

Data access. URL for data in this article are as follows: Genethon map: Online Mendelian Inheritance in Man (OMIM): UCSC Genome Browser:


View Abstract

Request permissions

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.