Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Hum Genet. Author manuscript; available in PMC 2011 January 1.
Published in final edited form as:
PMCID: PMC2847447

Deletion of an enhancer near DLX5 and DLX6 in a family with hearing loss, craniofacial defects, and an inv(7)(q21.3q35)


Precisely regulated temporal and spatial patterns of gene expression are essential for proper human development. Cis-acting regulatory elements, some located at large distances from their corresponding genes, play a critical role in transcriptional control of key developmental genes and disruption of these regulatory elements can lead to disease. We report a three generation family with five affected members, all of whom have hearing loss, craniofacial defects, and a paracentric inversion of the long arm of chromosome 7, inv(7)(q21.3q35). High resolution mapping of the inversion showed that the 7q21.3 breakpoint is located 65 and 80 kb centromeric of DLX6 and DLX5, respectively. Further analysis revealed a 5115 bp deletion at the 7q21.3 breakpoint. While the breakpoint does not disrupt either DLX5 or DLX6, the syndrome present in the family is similar to that observed in Dlx5 knockout mice and includes a subset of the features observed in individuals with DLX5 and DLX6 deletions, implicating dysregulation of DLX5 and DLX6 in the family’s phenotype. Bioinformatic analysis indicates that the 5115 bp deletion at the 7q21.3 breakpoint could contain regulatory elements necessary for DLX5 and DLX6 expression. Using a transgenic mouse reporter assay, we show that the deleted sequence can drive expression in the ear and developing bones of E12.5 embryos. Consequently, the observed familial syndrome is likely caused by dysregulation of DLX5 and/or DLX6 in specific tissues due to deletion of an enhancer and possibly separation from other regulatory elements by the chromosomal inversion.

Keywords: DLX5, DLX6, enhancer, inversion, 7q21.3


The profoundly complex processes underlying human development are predicated upon precisely regulated temporal and spatial patterns and proper levels of gene expression. Important developmental genes are subject to strict regulation and often reside in a complex and extended genomic regulatory landscape (Kleinjan and van Heyningen 2005). Most of these genes, for example SHH and SOX9, have compact transcriptional units, but are surrounded by large non-coding regions, sometimes hundreds of kilobases long, that must remain intact for appropriate expression. These extended domains likely contain multiple regulatory elements necessary for expression of their corresponding genes in specific tissues. Most developmental genes with associated large regulatory domains have been identified through chromosomal rearrangements. For example, translocations up to 265 kb 5’ of SHH cause the same phenotype, holoprosencephaly type 3, as do mutations within the SHH coding sequence (Belloni et al. 1996; Roessler et al. 1996; Roessler et al. 1997).

Recent whole genome sequencing of many species has now made identification of these long range regulatory elements possible. Because these regulatory elements have important biological function, their sequence is expected to be evolutionarily conserved. Non-genic regions with a high degree of conservation between distant species, such as human and zebrafish, are attractive candidate sequences for regulatory elements. Large scale investigation of these non-genic evolutionarily conserved regions using transgenic mouse reporter assays has revealed that many of these sequences can act as enhancers (Pennacchio et al. 2006). Additionally, specific investigations of the developmental regulatory regions identified by chromosomal rearrangements have uncovered multiple enhancers driving the expression of SHH, PAX6, and SOX9 (Bagheri-Fam et al. 2006; Jeong et al. 2006; Kleinjan et al. 2006). In one study, the breakpoint of a translocation causing preaxial polydactyly was mapped to a region approximately 1 Mb 5’ of SHH and a SHH enhancer driving expression in the limbs was subsequently identified in this region (Lettice et al. 2002; Lettice et al. 2003). Point mutations within this enhancer have been found in several families with preaxial polydactyly (Lettice et al. 2003; Gurnett et al. 2007), unequivocally demonstrating that disruption of the action of long range enhancers can cause developmental malformations.

The homeobox transcription factors DLX5 and DLX6, which are important for craniofacial, inner ear, and limb development, also seem to be regulated by cis-acting long range enhancers. Chromosomal rearrangements up to 450 kb away from DLX5 and DLX6 cause phenotypes similar to complete deletion of these genes (Scherer et al. 1994, Chromosome 7 Annotation Project, In this study, we have mapped and sequenced the breakpoints of an inversion, inv(7)(q21.3q35), in a family with hearing loss and craniofacial defects. We have shown that the 7q35 breakpoint disrupts CNTNAP2 and the 7q21.3 breakpoint is located 65 and 80 kb from DLX6 and DLX5, respectively. Investigation of these genes implicates dysregulation of DLX5 and/or DLX6 in the pathogenesis of the family’s phenotype. Additionally, deletion of a 5115 bp sequence at the 7q21.3 breakpoint seems to contribute to the phenotype as it may contain an enhancer essential for expression of DLX5 and DLX6 in the ear and in developing bones.


Patient Ascertainment

The DGAP115 family was ascertained as part of the Developmental Genome Anatomy Project (DGAP). The human study protocol for this project has been reviewed and approved by the Partners Health Care System Human Research Committee.

Fluorescence in situ hybridization (FISH) analysis

Peripheral blood samples were collected from the five affected family members and the unaffected grandmother of the proband. Lymphoblastoid cell lines were generated from each blood sample at the Massachusetts General Hospital Cell Transformation Core using standard protocols. Based on the reported karyotypes, BAC clones spanning the breakpoint regions were selected for FISH mapping using the University of California Santa Cruz (UCSC) Genome Browser ( BACs from the RP11 library were obtained from Children’s Hospital Oakland Research Institute (Oakland, CA) and BACs from the CTD library were acquired from Invitrogen (Carlsbad, CA). Metaphase chromosome spreads were prepared from the lymphoblastoid cells using standard cytogenetic protocols and FISH performed as previously described (Ney et al. 1993). BACs were directly labeled with either Spectrum Orange or Spectrum Green conjugated dUTP using a nick translation kit (Vysis, Downers Grove, IL) and differentially labeled pairs were hybridized overnight to metaphase chromosome preparations. After washing, chromosomes were counterstained with DAPI and analyzed with a Zeiss Axioskop (Thornwood, NY) microscope and Applied Imaging CytoVision software (Santa Clara, CA). At least 10 metaphases were scored per BAC probe.

Southern Blot Analysis

The 27 kb of overlapping sequence from BACs CTD-2366L13 and RP11-737n20 was analyzed using RestrictionMapper ( to select restriction enzymes digesting in staggered patterns across the sequence and using RepeatMasker ( to identify regions in which unique probes could be generated. DNA was isolated from lymphoblastoid cell lines using the PureGene system (Gentra, Minneapolis, MN). 7.5 ug of DGAP115 and control genomic DNA were digested overnight. Digested DNA was electrophoresed in a 0.7% agarose gel and transferred overnight to Hybond-N+ membrane (Amersham Biosciences, Piscataway, NJ) via capillary action. DNA was cross-linked to the membrane with ultraviolet light. Probes of 300–600 bp were created by PCR, radioactively labeled with α32P-dCTP by random priming using the Megaprime DNA Labeling Kit (Amersham Biosciences), and purified using Microspin G-25 columns (Amersham Biosciences). Following prehybridization for at least an hour in UltraHybe (Ambion, Austin, TX) at 42°C, blots were hybridized overnight at 42°C with labeled probes that had previously been denatured at 95°C for 10 minutes. Blots were washed several times, exposed to Kodak Biomax XAR film in cassettes with intensifying screens at −80°C, and developed with a Kodak X-OMAT 2000A processor. Bands detected in patient but not in control DNA indicate localization of the breakpoint within these restriction fragments.

Breakpoint Cloning

The 7q21.3 breakpoint was cloned using the suppression PCR strategy described in (Siebert et al. 1995). The 7q21.3 specific primer sequences were 5’-ATTCTGGAAACGCAGGTGGCTCTGT-3’ and 5’-CGCCCTTACCCCTCTTAGAC-3’ (nested). DGAP115 and control reactions were electrophoresed in a 1% agarose gel and the band detected only in the DGAP115 reaction was extracted using the QIAquick Gel Exaction Kit (Qiagen, Valencia, CA). The extracted band was cloned into pCRII-TOPO using the Invitrogen TOPO-TA cloning kit and sequenced at the Dana-Farber/Harvard Cancer Center (DFHCC) DNA Resource Core. Based on the sequence of the 7q21.3 breakpoint, PCR primers were designed to amplify the 7q35 breakpoint. The 7q35 PCR product was TOPO-TA cloned and sequenced at the DFHCC DNA Resource Core. Breakpoints were confirmed in all affected family members by PCR and sequencing.

Array Comparative Genomic Hybridization (aCGH)

aCGH was performed by the DFHCC Cytogenetics Core facility using Spectral Genomics 2600 BAC arrays (Houston, TX) with ~1 Mb resolution.

Mouse Whole Mount In Situ Hybridization

For mouse whole mount in situ hybridization experiments, a 640 bp fragment from the 3’UTR of Cntnap2 was amplified by PCR from mouse embryonic cDNA and was TOPO-TA cloned into pCRII-TOPO. The vector was linearized with either BamHI or EcoRV and both antisense and sense riboprobes with digoxigenin conjugated UTP were generated by in vitro transcription using Sp6 and T7 RNA polymerases (MAXIscript kit, Ambion). Riboprobes were purified with P-30 Tris RNase-free Micro Bio-Spin Columns (Bio-Rad, Hercules, CA). E9.5, E10.5, and E11.5 ICR embryos (Taconic, Germantown, NY) were collected in cold PBS and fixed overnight in 4% paraformaldehyde at 4°C. Embryos were dehydrated in a series of methanol/PBST washes at room temperature and stored in 100% methanol at −20°C until ready for processing. Embryos were rehydrated, bleached in 6% hydrogen peroxide in PBST, and treated with proteinase K for 10 minutes (E9.5), 15 minutes (E10.5), or 20 minutes (E11.5). After postfixing in 4% paraformaldehyde/0.2% glutaraldehyde in PBST, embryos were equilibrated in 50% hybridization solution (50% formamide, 5XSSC at pH 4.5, 1% SDS, 50 ug/ml yeast tRNA, and 50 ug/ml heparin) in PBST and then prehybridized in 100% hybridization solution at 65°C for one hour. Riboprobe was added to the embryos and they were incubated overnight at 65°C. Embryos were washed, blocked with 10% heat inactivated sheep serum and 1% blocking reagent (Roche, Indianapolis, IN) in MABT, and then incubated with anti-digoxigenin-AP antibody (Roche) at a concentration of 1:5000 in 1% heat inactivated sheep serum and 1% blocking reagent overnight at 4°C. Embryos were then washed thoroughly, soaked in NTMT, and developed in BM Purple AP Substrate (Roche) until the reaction was judged complete (about 2 hours). Finally, the embryos were washed, postfixed in 4% paraformaldehyde/0.1% glutaraldehyde for 20 minutes, and washed twice more.

Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR)

RNA was isolated from lymphoblastoid cell lines using Trizol (Invitrogen). 1 ug of RNA was reverse transcribed using the SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen). The resulting cDNA was amplified with the following primers: 5’-CTCGCTGGGATTGACACAA-3’ and 5’-GGTCATAGATTTCAAGGCACCA-3’ for DLX5 and 5’-CGAACTGGCAGCTTCCTTAG-3’ and 5’-AATGCAGGAGTCCAAAATGC-3’ for DLX6.

Analysis of Deleted Sequence

The 5115 bp deleted sequence from 7q21.3 was submitted to GenomeVista (; Bray et al. 2003; Couronne et al. 2003) to assess conservation of the sequence between human, mouse, rat, and chicken genomes. The 5115 bp deleted sequence was also compared to the mouse and rat genomes using both BLASTN and cross-species megaBLAST to identify highly similar regions between the species (; Altschul et al. 1990). Additionally, the sequence was submitted to the Genomatix MatInspector program (; Quandt et al. 1995) to search for known transcription factor binding sites.

Transgenic Embryo Assay

The entire 5115 bp sequence plus the conserved regions at each end of the sequence (chr7:96,401,902-96,407,985 on the March 2006 assembly of the human genome) and a minimal beta globin promoter (5’-AGCTTCCCGGGCTGGGCATAAAAGTCAGGGCAGAGCCATCTATTGCTTACATTTGCTTCTGCA-3’) were cloned between the XhoI and NcoI sites of MCS_pNASSβ, which contains a LacZ cassette (gift of Amy Donner, Brigham and Women’s Hospital). 20 ug of the construct was digested overnight with NarI and SphI. Digested DNA was electrophoresed in a 1% agarose gel, the 10 kb band was extracted using the QIAquick Gel Extraction kit (Qiagen), and the extracted band was eluted in 10 mM Tris/0.25 mM EDTA pH 8.0. Following elution, this linearized construct was diluted, injected into the pronucleus of FVB fertilized eggs, and transferred into pseudopregnant mothers by the Brigham and Women’s Hospital Transgenic Core facility.

Embryos were collected at E12.5 and their yolk sacs were saved for genotyping. The embryos were washed in PBS and fixed for one hour in 1% formaldehyde, 0.2% glutaraldehyde, 2 mM MgCl2, 5 mM EGTA, and 0.02% NP-40 in PBS. Then they were washed three times in PBS/0.02% NP-40 and stained at 37°C with 5 mM K3Fe(CN)6, 5mM K4Fe(CN)6, 2 mM MgCl2, 0.01% NaDeoxycholate, 0.02% NP-40, and 1mg/ml X-gal in PBS until the reaction was complete. Finally, embryos were washed three times in PBS and postfixed in 4% paraformaldehyde overnight at 4°C. Images of embryos were taken with a dissecting microscope using a digital camera. Ears were dissected and whole mounted images were taken using a digital camera mounted onto a Nikon E800 microscope.

DNA was isolated from the yolk sacs using the DNAeasy Blood and Tissue Kit (Qiagen). PCR was performed on the resulting DNA using a primer in the LacZ gene and a primer in the 5115 bp deleted sequence to identify the transgenic embryos.


We have ascertained a family (designated the DGAP115 family) consisting of five affected members in which hearing loss and craniofacial anomalies segregate with a paracentric inversion of the long arm of chromosome 7. The family consists of five affected members spanning three generations, the proband (designated DGAP115), the proband’s mother (DGAP116), and the proband’s three children (DGAP117, DGAP118, and DGAP119). All five affected family members have hearing loss, micrognathia, and abnormal pinnae (small and posteriorly rotated with overfolded superior helices) (Figure 1). Some of these individuals also have additional anomalies, including cleft palate, mild developmental delay, early puberty, kyphosis, and abnormal dentition, as detailed in Table 1. X-rays taken for unrelated purposes have also shown femoral head necrosis in one family member. An MRI of the brain and auditory structures of DGAP115 revealed abnormalities of the inner ear. The right inner ear showed incomplete segmentation of the cochlea with an underdeveloped basal turn and a poorly formed modiolus, while the left ear showed hypoplasia of the cochlea with absent segmentation, a poorly formed modiolus, and cochlear aperture stenosis. Vestibular structures and the vestibulocochlear nerve appeared normal on both sides. Structural abnormalities of the middle ear are also likely; however, a follow-up temporal bone CT scan to evaluate the morphology of the ossicles was not performed. The reported karyotype for all affected individuals, based on GTG-banded analysis of metaphase chromosomes, was 46,XX,inv(7)(q22.1q35) or 46,XY,inv(7)(q22.1q35). The proband’s maternal grandmother is phenotypically normal and does not have the inv(7). Her maternal grandfather also had a normal phenotype but was not available for further evaluation.

Figure 1
Side views of the DGAP115 family members (DGAP115-119) and a family pedigree with affected individuals denoted. Note the micrognathia and abnormal pinnae with overfolded superior helices present in all family members (color figure online)
Table 1
Phenotype of affected family members

Breakpoint Mapping

To investigate the molecular basis of this family’s clinical features, we further mapped the breakpoints of the inversion and examined potential candidate genes in the breakpoint regions. We began by performing FISH on metaphase chromosomes from DGAP115 lymphoblastoid cells. Based on the reported karyotype, inv(7)(q22.1q35), BACs localized to 7q21-q31 and 7q34-q36 on the UCSC website were selected and sequential FISH experiments performed to narrow the breakpoint regions. BAC clones CTD-2350n7 and RP11-879e11 from 7q21.3 hybridized to 7q21.3 on the normal chromosome 7 and to both 7q21.3 and 7q35 on the inv(7) indicating that the breakpoint occurs within the 100 kb region where these BACs overlap (Figure 2A and Figure 3). The ends of a BAC centromeric to the 7q21.3 breakpoint, CTD-2366L13, and a BAC telomeric to this breakpoint, RP11-737n20, slightly overlap but neither shows a split hybridization signal, suggesting that the breakpoint lies within the 27 kb of overlapping sequence (Figure 3). BACs RP11-5p2 and RP11-643a21 from 7q35 hybridized to 7q35 on the normal chromosome 7 and to both 7q21.3 and 7q35 on the inv(7) narrowing the 7q35 breakpoint to 140 kb (Figure 2B). Based on our FISH studies, which provide a higher resolution analysis than a traditional GTG-banded karyotype, the breakpoints of the familial inversion were revised to inv(7)(q21.3q35).

Figure 2
Mapping of the DGAP115 family inversion breakpoints using FISH and Southern blots. A) FISH on metaphase chromosomes showing BAC RP11-879e11 (green), which maps to 7q21.3 and spans the 7q21.3 breakpoint. B) FISH on metaphase chromosomes showing BAC RP11-643a21 ...
Figure 3
Strategy used to clone and sequence the DGAP115 family inversion breakpoints. FISH mapping with overlapping BACs narrowed the 7q21.3 breakpoint region to 27 kb. BACs with split hybridization signals are colored red. Southern blots with probes spanning ...

To delineate the location of the 7q21.3 breakpoint further, Southern blots were performed on control and DGAP115 DNA using probes spanning the 27 kb FISH-defined breakpoint region. A band was detected in the AflII, KpnI, and StuI DGAP115 digests that was not present in the control digests indicating that the breakpoint occurs within the 3255 bp where these restriction fragments overlap (Figure 2C and Figure 3). No aberrant bands were detected in the EcoRI digests, further narrowing the breakpoint to 331 bp (Figure 3). The 7q21.3 breakpoint was then amplified using suppression PCR, cloned, and sequenced. Using this sequence as a guide for primer design, the 7q35 breakpoint was amplified directly by PCR, cloned, and sequenced. Sequencing of both breakpoints revealed a 5115 bp deletion (chr7:96,402,577-96,407,691 on the March 2006 assembly of the human genome) of 7q21.3 sequence, a 2 bp deletion of 7q35 sequence, and a 15 bp duplication of 7q35 sequence on the inv(7) (Figure 3). Sequencing also confirmed that all affected family members have the same exact breakpoints. To ensure that there were no deletions or duplications elsewhere in the genome, DNA from DGAP115 was examined by aCGH. No clinically significant copy number gains or losses were detected.

Identification and Investigation of Genes near the Breakpoints

The 7q35 breakpoint disrupts CNTNAP2 between exons nine and ten (Figure 4). To determine if CNTNAP2 is a strong candidate gene for the family’s phenotype, its expression pattern during early development was investigated using whole mount in situ hybridization on mouse embryos. At E9.5, E10.5, and E11.5, Cntnap2 expression was seen primarily in the brain (Figure 5). Cntnap2 expression was absent from most tissues relevant to the family’s phenotype, such as the branchial arches and the otic placode/vesicle, suggesting that malformation of these tissues is unrelated to disruption of this gene. Additionally, Cntnap2 knockout mice have been reported to have “no apparent phenotype” (Poliak et al. 2003) and specific examination of these null mice revealed no craniofacial or ear abnormalities. Recent reports have implicated CNTNAP2 as the cause of various neurologic disorders, including autism, epilepsy, and schizophrenia (Strauss et al. 2006; Alarcon et al. 2008; Arking et al. 2008; Bakkaloglu et al. 2008; Friedman et al. 2008); however, disruption of CNTNAP2 does not appear to be the primary cause of the DGAP115 family’s clinical features.

Figure 4
Map of the 7q35 breakpoint from the UCSC genome browser. The breakpoint is marked by the red line. The 7q35 breakpoint disrupts CNTNAP2 between exons nine and ten (color figure online)
Figure 5
Whole mount in situ hybridization on mouse embryos using a Cntnap2 riboprobe at E9.5 (A,B), E10.5 (C,D), and E11.5 (E,F). Cntnap2 expression is seen primarily in the brain at all three time points (color figure online)

There are no known genes, mRNAs, or ESTs at the 7q21.3 inversion breakpoint. The closest genes to the breakpoint are DLX6 (65.5 kb telomeric), DLX5 (80 kb telomeric), and SHFM1 (225 kb centromeric) (Figure 6). Dlx5 knockout mice have malformations of the middle and inner ear, abnormal pinnae, and micrognathia (Acampora et al. 1999; Depew et al. 1999; Merlo et al. 2002b), consistent with the phenotype seen in the DGAP115 family. Additionally, individuals with deletions of both DLX5 and DLX6 or with chromosomal rearrangements centromeric to DLX5 and DLX6 have severe phenotypes which generally include hearing loss, abnormal pinnae, micrognathia, cleft palate, and developmental delay (Chromosome 7 Annotation Project, To assess whether the inversion affects expression of DLX5 or DLX6, reverse transcriptase (RT) -PCR was performed for the DGAP115 family and control lymphoblastoid cell lines using primer sets for both genes. Because both DLX5 and DLX6 are expressed at very low levels in lymphoblastoid cell lines, they were not consistently amplified even when the PCR was performed for an increased number of cycles. Previous reports have also noted difficulty in assessing DLX5 and DLX6 expression in lymphoblastoid cell lines (Horike et al. 2005; Schule et al. 2007; Miyano et al. 2008) and profiles for both DLX5 and DLX6 in the UniGene database ( show no expression in blood.

Figure 6
Map of the 7q21.3 breakpoint region from the UCSC genome browser. The 7q21.3 breakpoint is 65 kb and 80 kb from DLX6 and DLX5, respectively. 5115 bp of 7q21.3 sequence is deleted at the breakpoint. This deleted sequence contains several regions that are ...

Contribution of 5115 bp Deleted Sequence to Family’s Phenotype

To investigate further a role for the 7q21.3 breakpoint in the family’s phenotype, the 5115 bp deleted sequence was analyzed. Comparing the deleted sequence with the syntenic regions of the mouse genome and the rat genome using GenomeVista ( revealed three regions (1-3MC) with over 75% conservation in both mouse and rat and an additional five regions (1-5C) with over 70% conservation in either mouse or rat when using 100 bp sliding windows (Figure 6). Comparing the deleted sequence to the mouse and rat genomes via BLAST ( revealed two regions with greater than 85% identity. These regions correspond with regions 1MC and 2MC identified by GenomeVista. BLAST also detected a 50 bp region in both mouse and rat with high basepair identity, 91% for rat and 83% for mouse, which corresponds to GenomeVista region 1C. MatInspector (, which searches for known transcription factor binding sites, identified four DLX1 and seven MEF2 consensus binding sites within the 5115 bp deleted sequence. Both DLX1 and MEF2C are known to regulate DLX5 and DLX6 expression (Anderson et al. 1997; Zerucha et al. 2000; Verzi et al. 2007).

Because the 5115 bp deleted sequence contains several conserved regions and binding sites for transcription factors that regulate DLX5 and DLX6, this sequence was tested for in vivo enhancer activity. The 5115 bp deleted region and a minimal beta globin promoter were cloned into the MCS-pNASSβ vector upstream of the LacZ cassette. This construct was then used to generate transgenic embryos to determine if the deleted region could drive expression of the LacZ reporter in developing tissues relevant to the family’s phenotype. Eight of the 72 F0 embryos harvested at E12.5 had the transgenic construct integrated into the genome. Two of these eight embryos (25%) showed LacZ expression. Interestingly, the LacZ expression appears in a subset of the tissues where DLX5 and DLX6 are expressed at E12.5 (Acampora et al. 1999; Robledo et al. 2002), including the developing bones of the vertebrae and ribs and the inner ear (Figure 7). Further dissection to localize the inner ear expression more specifically revealed intense staining of the endolymphatic sac and the tip of the cochlear duct and minor staining in the posterior semicircular canal (Figures 7D and 7E). The presence of LacZ expression in specific tissues suggests that the 5115 bp deleted region contains an enhancer which potentially regulates expression of genes in the vicinity of the 7q21.3 breakpoint, such as DLX5 and DLX6.

Figure 7
X-gal staining of E12.5 transgenic embryo shows LacZ expression in developing bones and the ear. A) View of transgenic embryo showing surface staining. B) Cleared view of transgenic embryo showing staining of internal structures. C) Dissection of the ...


Congenital anomalies in individuals with chromosomal rearrangements are often caused by disruption of genes at the rearrangement breakpoints. However, chromosomal rearrangements have also been useful for detecting genes that are subject to the action of long range regulatory elements. Several genes have been identified for which rearrangement breakpoints at a distance of up to 1 Mb away from the gene can cause the same phenotype as a direct disruption of that gene. These genes include PAX6, SOX9, POU3F4, SHH, SIX, MAF, FOXC1, FOXC2, FOXL2, and TWIST (Kleinjan and van Heyningen 2005). Expression of these genes is precisely regulated during development by long range regulatory elements and disruption of this regulation can cause severe phenotypes. Mapping the breakpoints of a chromosomal rearrangement in the family reported herein revealed that CNTNAP2 is directly disrupted at 7q35, but is unlikely to be responsible for the observed heritable phenotype. The 7q21.3 breakpoint does not disrupt any genes, but is located less than 100 kb from DLX5 and DLX6 which are compelling candidate genes for the family’s disorder.

Exclusion of CNTNAP2 as a Candidate Gene for the Family’s Phenotype

The 7q35 breakpoint disrupts CNTNAP2 between exons nine and ten effectively separating the first nine exons from the last 15 exons on the inv(7). CNTNAP2 is a member of the neurexin superfamily of transmembrane proteins that mediate cell-cell interactions in the nervous system. CNTNAP2 co-localizes with the voltage gated potassium channels, KCNA1, KCNA2, and KCNAB2, at the juxtaparanodal region of myelinated axons (Poliak et al. 1999). It is necessary for proper positioning of these channels in the juxtaparanodal region and possibly for formation and maintenance of distinct domains along the axon (Poliak et al. 2003). Expression of Cntnap2 in developing mice is seen predominantly in the brain. To assess the potential role of this gene in neurologic disease, Cntnap2 null mice were generated, but reported to have no noticeable phenotype (Poliak et al. 2003). No craniofacial, skeletal, or ear malformations were observed in the null mice.

Recent reports of individuals with other mutations of CNTNAP2 also support exclusion of CNTNAP2 as a candidate gene for hearing loss, craniofacial, and skeletal abnormalities. Chromosomal rearrangements disrupting CNTNAP2 have been reported in individuals with various neurologic conditions, including an insertion in a family with Gilles de la Tourette syndrome and obsessive compulsive disorder (Verkerk et al. 2003), an inversion in a child with autism (Bakkaloglu et al. 2008), and deletions in one individual with autism (Rossi et al. 2008), one individual with epilepsy (Friedman et al. 2008), and two individuals with both epilepsy and schizophrenia (Friedman et al. 2008). Additionally, a recessive mutation in CNTNAP2 causes cortical dysplasia-focal epilepsy syndrome (Strauss et al. 2006) and other CNTNAP2 variants increase risk for autism (Alarcon et al. 2008; Arking et al. 2008; Bakkaloglu et al. 2008). None of the individuals with these other mutations have been reported to have hearing loss, craniofacial defects, or skeletal abnormalities. Furthermore, a completely healthy individual with a translocation disrupting CNTNAP2 has also been reported (Belloso et al. 2007). Besides mild developmental delay, no neurological conditions have been noted for any of the DGAP115 family members and the family has been described as “very laid back.” Based on the phenotype of the knockout mouse and the other patients with CNTNAP2 mutations, disruption of CNTNAP2 does not appear to contribute to the syndrome present in the DGAP115 family.

DLX5 and DLX6 as Candidate Genes for the Family’s Phenotype

DLX5 and DLX6 are transcription factors containing a homeobox domain related to the Drosophila Distal-less homeobox domain and are important developmental genes involved in patterning of craniofacial structures, the inner ear, the limbs, and the brain (Merlo et al. 2000). They also play a major role in chondrocyte and osteoblast differentiation. In E9.5 mouse embryos, Dlx5 and Dlx6 are expressed in the branchial arches and otic placode (Acampora et al. 1999). By E11.5, expression is seen throughout the derivatives of the first branchial arch, in the vestibular portion of the inner ear, in the diencephalon and telencephalon of the brain, and in the apical ectodermal ridge (Acampora et al. 1999). Dlx5 and Dlx6 are also expressed in all skeletal elements from initial cartilage formation until ossification (Simeone et al. 1994). Dlx5 knockout mice are perinatal lethal and can be easily distinguished from their wild-type and heterozygous littermates due to their decreased nasal capsular breadth, micrognathia, and abnormal pinnae (Depew et al. 1999). Dlx5 knockouts are also characterized by cleft palate, ectopic bones in the middle ear, a shortened cochlear duct, absent semicircular canals, a reduced endolymphatic duct, misshapen teeth, and cranial bone defects (Acampora et al. 1999; Depew et al. 1999; Merlo et al. 2002b). Dlx5 and Dlx6 double knockout mice have more severe craniofacial, ear, and bone defects, including ectrodactyly (Merlo et al. 2002a; Robledo et al. 2002). Although neither DLX5 nor DLX6 is disrupted by the breakpoint, the DGAP115 family members and the Dlx5 knockout mice have many similar features.

Furthermore, multiple individuals with deletions of DLX5 and DLX6 have been reported. These individuals have severe phenotypes, including ectrodactyly, sensorineural hearing loss, conductive hearing loss due to fixation of ossicles, abnormal pinnae with overfolded helices, cleft palate, micrognathia, developmental delay, and abnormal teeth (Scherer et al. 1994; Haberlandt et al. 2001; Fukushima et al. 2003; Wieland et al. 2004). Affected members of the DGAP115 family also have sensorineural hearing loss, conductive hearing loss, abnormal pinnae with overfolded helices, cleft palate, micrognathia, developmental delay, and abnormal teeth. However, the clinical features of the DGAP115 family are less severe than those seen in the DLX5 and DLX6 deletion patients, as they have a milder hearing loss and they do not have ectrodactyly. Interestingly, several individuals with translocations centromeric to DLX5 and DLX6 have also been reported with similar phenotypes to the deletion patients (Chromosome 7 Annotation Project, suggesting the existence of long range cis-acting regulatory elements centromeric to DLX5 and DLX6 that are essential for proper DLX5 and DLX6 expression. It is possible that the family’s inversion leads to loss of function of an enhancer element which regulates patterning of the ear and craniofacial regions, but is not necessary for other functions of DLX5 and DLX6, such as development of the fingers and toes. Given that the DGAP115 family has clinical features similar to, but milder than, those caused by loss of the DLX5 and DLX6 genes, we propose that the inv(7) disturbs the function of an enhancer which regulates DLX5 and/or DLX6 expression in specific tissues during development.

Identification of an Enhancer near DLX5 and DLX6 Driving Expression in Developing Bones and the Ear

Since disruption of the function of an enhancer could be the pathogenic mechanism causing the phenotype of the DGAP115 family, we decided to test for the presence of long range enhancers in the vicinity of the 7q21.3 breakpoint using an in vivo reporter assay. Dysregulation of DLX5 and DLX6 could result from deletion of an enhancer within the 5115 bp deleted sequence at the breakpoint or from separation of the genes from enhancers on the opposite side of the breakpoint. Bioinformatic analysis of the 5115 bp deleted sequence for conserved regions and relevant transcription factor binding sites indicated that it could potentially harbor regulatory elements. Testing the deleted sequence in transgenic embryos showed that it could drive expression of a LacZ reporter gene in developing skeletal structures and the inner ear, which are areas of DLX5 and DLX6 expression and tissues affected in the DGAP115 family members. While these results support the idea that the deleted region contains an enhancer regulating DLX5 and/or DLX6 expression, we also expected to see LacZ expression in the developing mandible and middle ear tissues. Embryos were assayed at E12.5 because DLX5 and DLX6 are at their widest range of expression at this time. While no expression was seen in the mandible at this time point, the reporter may have been expressed in this area at an earlier time point as the mandible begins developing around E9.5. A MEF2C-dependent enhancer in the promoter region of DLX6 drives reporter gene expression in the mandible and other craniofacial regions at E9.5 (Verzi et al. 2007); however, in embryos assayed at E11.5 the same sequence drives expression in the midbrain, but not in any craniofacial areas (Vista Enhancer Browser,, demonstrating the precise temporal specificity of enhancers. Likewise, the middle ear primordium is present at E12.5, but development of the middle ear bones is not detectable until at least E14.5; therefore, the reporter gene may be active in the middle ear later in development. Alternatively, there could be additional enhancers located in the 7q21.3 region centromeric to the breakpoint regulating DLX5 and DLX6 expression in the mandible, middle ear, and other craniofacial areas. While two embryos showed similar X-gal staining patterns, the other six transgenic embryos did not show any X-gal staining. There are several possible explanations for this; for example, the construct may have inserted into regions of heterochromatin in the non-expressing embryos or the enhancer in the deleted sequence must act in concert with other elements to drive transcription effectively. The latter option seems likely as additional conserved areas were also present on both sides of the deleted sequence at 7q21.3. Furthermore, the deleted region is quite large, so it is possible that an enhancer element located at a distance from the minimal promoter will not be able to act on the promoter efficiently when it is outside of its normal genomic context, particularly if elements that promote chromatin looping are not present. The results of the reporter assay indicate that the 5115 bp deleted sequence contains an enhancer that could regulate DLX5 and DLX6 expression in the vertebrae, the ribs, the inner ear, and possibly other tissues. Deletion of this sequence from the inv(7) could potentially cause some, if not all, of the clinical features seen in the DGAP115 family.

While the tissues where this potential enhancer is active correlate with the tissues where DLX5 and DLX6 are expressed, further work will need to be done to establish conclusively that this enhancer plays a role in the regulation of DLX5 and/or DLX6. For example, transgenic studies using large BAC-based constructs with DLX5 and DLX6 in their normal genomic context and the enhancer region either present or absent could be used to show that the enhancer actually directs the expression of the DLX5 and/or DLX6 genes. Additional studies would also be needed to determine the exact sequence of the enhancer and the transcription factors that bind the enhancer. The 5115 bp deleted region could be split into smaller segments and those smaller constructs could be used in either transgenic experiments or in vitro reporter assays to localize more precisely the enhancer element. Once the enhancer is precisely identified, electrophoretic mobility shift assays (EMSA) could be performed to discover which transcription factors bind to the enhancer, which may provide further insight into the molecular pathways involved in otic, craniofacial, and skeletal development.

In summary, high resolution mapping and sequencing of the breakpoints of a familial inversion associated with hearing loss and craniofacial defects revealed a 5115 bp deletion at one of the breakpoints. This breakpoint occurs less than 100 kb from the DLX5 and DLX6 genes, which are regulated in a strict temporal and spatial pattern during development presumably due to the action of long range cis-acting regulatory elements. Disrupting the action of these regulatory elements may cause severe abnormalities, including hearing loss, craniofacial defects, and limb defects, like those observed from deletion of these genes. Similarly, dysregulation of DLX5 and/or DLX6 due to deletion of an enhancer at the 7q21.3 breakpoint and possibly separation from other regulatory elements is the likely cause of the syndrome observed in the DGAP115 family.


We would like to express our sincere gratitude to the DGAP115 family for making this study possible. We would also like to thank Elior Peles for access to Cntnap2 null mice, Orit Hermesh, Amiel Dror, and Karen B. Avraham for analysis of the Cntnap2 null mice, Richard Robertson for interpretation of MRI images, the Massachusetts General Hospital Cell Transformation Core for assistance with EBV transformation, the Brigham and Women’s Hospital Transgenic core for assistance with generation of transgenic mice, and the Dana-Farber/Harvard Cancer Center (P30CA06516) Cytogenetic Core and DNA Resource Core for assistance with aCGH and DNA sequencing. This work was supported by grants from the National Institutes of Health, P01 GM061354 (to C.C.M.) and F31 DC007540 (to K.K.B).


  • Acampora D, Merlo GR, Paleari L, Zerega B, Postiglione MP, Mantero S, Bober E, Barbieri O, Simeone A, Levi G. Craniofacial, vestibular and bone defects in mice lacking the Distal-less-related gene Dlx5. Development. 1999;126:3795–3809. [PubMed]
  • Alarcon M, Abrahams BS, Stone JL, Duvall JA, Perederiy JV, Bomar JM, Sebat J, Wigler M, Martin CL, Ledbetter DH, Nelson SF, Cantor RM, Geschwind DH. Linkage, association, and gene-expression analyses identify CNTNAP2 as an autism-susceptibility gene. Am J Hum Genet. 2008;82:150–159. [PubMed]
  • Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215:403–410. [PubMed]
  • Anderson SA, Qiu M, Bulfone A, Eisenstat DD, Meneses J, Pedersen R, Rubenstein JL. Mutations of the homeobox genes Dlx-1 and Dlx-2 disrupt the striatal subventricular zone and differentiation of late born striatal neurons. Neuron. 1997;19:27–37. [PubMed]
  • Arking DE, Cutler DJ, Brune CW, Teslovich TM, West K, Ikeda M, Rea A, Guy M, Lin S, Cook EH, Chakravarti A. A common genetic variant in the neurexin superfamily member CNTNAP2 increases familial risk of autism. Am J Hum Genet. 2008;82:160–164. [PubMed]
  • Bagheri-Fam S, Barrionuevo F, Dohrmann U, Gunther T, Schule R, Kemler R, Mallo M, Kanzler B, Scherer G. Long-range upstream and downstream enhancers control distinct subsets of the complex spatiotemporal Sox9 expression pattern. Dev Biol. 2006;291:382–397. [PubMed]
  • Bakkaloglu B, O'Roak BJ, Louvi A, Gupta AR, Abelson JF, Morgan TM, Chawarska K, Klin A, Ercan-Sencicek AG, Stillman AA, Tanriover G, Abrahams BS, Duvall JA, Robbins EM, Geschwind DH, Biederer T, Gunel M, Lifton RP, State MW. Molecular cytogenetic analysis and resequencing of contactin associated protein-like 2 in autism spectrum disorders. Am J Hum Genet. 2008;82:165–173. [PubMed]
  • Belloni E, Muenke M, Roessler E, Traverso G, Siegel-Bartelt J, Frumkin A, Mitchell HF, Donis-Keller H, Helms C, Hing AV, Heng HH, Koop B, Martindale D, Rommens JM, Tsui LC, Scherer SW. Identification of Sonic hedgehog as a candidate gene responsible for holoprosencephaly. Nat Genet. 1996;14:353–356. [PubMed]
  • Belloso JM, Bache I, Guitart M, Caballin MR, Halgren C, Kirchhoff M, Ropers HH, Tommerup N, Tumer Z. Disruption of the CNTNAP2 gene in a t(7;15) translocation family without symptoms of Gilles de la Tourette syndrome. Eur J Hum Genet. 2007;15:711–713. [PubMed]
  • Bray N, Dubchak I, Pachter L. AVID: A global alignment program. Genome Res. 2003;13:97–102. [PubMed]
  • Couronne O, Poliakov A, Bray N, Ishkhanov T, Ryaboy D, Rubin E, Pachter L, Dubchak I. Strategies and tools for whole-genome alignments. Genome Res. 2003;13:73–80. [PubMed]
  • Depew MJ, Liu JK, Long JE, Presley R, Meneses JJ, Pedersen RA, Rubenstein JL. Dlx5 regulates regional development of the branchial arches and sensory capsules. Development. 1999;126:3831–3846. [PubMed]
  • Friedman JI, Vrijenhoek T, Markx S, Janssen IM, van der Vliet WA, Faas BH, Knoers NV, Cahn W, Kahn RS, Edelmann L, Davis KL, Silverman JM, Brunner HG, van Kessel AG, Wijmenga C, Ophoff RA, Veltman JA. CNTNAP2 gene dosage variation is associated with schizophrenia and epilepsy. Mol Psychiatry. 2008;13:261–266. [PubMed]
  • Fukushima K, Nagai K, Tsukada H, Sugata A, Sugata K, Kasai N, Kibayashi N, Maeda Y, Gunduz M, Nishizaki K. Deletion mapping of split hand/split foot malformation with hearing impairment: a case report. Int J Pediatr Otorhinolaryngol. 2003;67:1127–1132. [PubMed]
  • Gurnett CA, Bowcock AM, Dietz FR, Morcuende JA, Murray JC, Dobbs MB. Two novel point mutations in the long-range SHH enhancer in three families with triphalangeal thumb and preaxial polydactyly. Am J Med Genet A. 2007;143:27–32. [PubMed]
  • Haberlandt E, Loffler J, Hirst-Stadlmann A, Stockl B, Judmaier W, Fischer H, Heinz-Erian P, Muller T, Utermann G, Smith RJ, Janecke AR. Split hand/split foot malformation associated with sensorineural deafness, inner and middle ear malformation, hypodontia, congenital vertical talus, and deletion of eight microsatellite markers in 7q21.1-q21.3. J Med Genet. 2001;38:405–409. [PMC free article] [PubMed]
  • Horike S, Cai S, Miyano M, Cheng JF, Kohwi-Shigematsu T. Loss of silent-chromatin looping and impaired imprinting of DLX5 in Rett syndrome. Nat Genet. 2005;37:31–40. [PubMed]
  • Jeong Y, El-Jaick K, Roessler E, Muenke M, Epstein DJ. A functional screen for sonic hedgehog regulatory elements across a 1 Mb interval identifies long-range ventral forebrain enhancers. Development. 2006;133:761–772. [PubMed]
  • Kleinjan DA, Seawright A, Mella S, Carr CB, Tyas DA, Simpson TI, Mason JO, Price DJ, van Heyningen V. Long-range downstream enhancers are essential for Pax6 expression. Dev Biol. 2006;299:563–581. [PMC free article] [PubMed]
  • Kleinjan DA, van Heyningen V. Long-range control of gene expression: emerging mechanisms and disruption in disease. Am J Hum Genet. 2005;76:8–32. [PubMed]
  • Lettice LA, Heaney SJ, Purdie LA, Li L, de Beer P, Oostra BA, Goode D, Elgar G, Hill RE, de Graaff E. A long-range Shh enhancer regulates expression in the developing limb and fin and is associated with preaxial polydactyly. Hum Mol Genet. 2003;12:1725–1735. [PubMed]
  • Lettice LA, Horikoshi T, Heaney SJ, van Baren MJ, van der Linde HC, Breedveld GJ, Joosse M, Akarsu N, Oostra BA, Endo N, Shibata M, Suzuki M, Takahashi E, Shinka T, Nakahori Y, Ayusawa D, Nakabayashi K, Scherer SW, Heutink P, Hill RE, Noji S. Disruption of a long-range cis-acting regulator for Shh causes preaxial polydactyly. Proc Natl Acad Sci U S A. 2002;99:7548–7553. [PubMed]
  • Merlo GR, Paleari L, Mantero S, Genova F, Beverdam A, Palmisano GL, Barbieri O, Levi G. Mouse model of split hand/foot malformation type I. Genesis. 2002a;33:97–101. [PubMed]
  • Merlo GR, Paleari L, Mantero S, Zerega B, Adamska M, Rinkwitz S, Bober E, Levi G. The Dlx5 homeobox gene is essential for vestibular morphogenesis in the mouse embryo through a BMP4-mediated pathway. Dev Biol. 2002b;248:157–169. [PubMed]
  • Merlo GR, Zerega B, Paleari L, Trombino S, Mantero S, Levi G. Multiple functions of Dlx genes. Int J Dev Biol. 2000;44:619–626. [PubMed]
  • Miyano M, Horike S, Cai S, Oshimura M, Kohwi-Shigematsu T. DLX5 expression is monoallelic and Dlx5 is up-regulated in the Mecp2-null frontal cortex. J Cell Mol Med. 2008;12:1188–1191. [PubMed]
  • Ney PA, Andrews NC, Jane SM, Safer B, Purucker ME, Weremowicz S, Morton CC, Goff SC, Orkin SH, Nienhuis AW. Purification of the human NF-E2 complex: cDNA cloning of the hematopoietic cell-specific subunit and evidence for an associated partner. Mol Cell Biol. 1993;13:5604–5612. [PMC free article] [PubMed]
  • Pennacchio LA, Ahituv N, Moses AM, Prabhakar S, Nobrega MA, Shoukry M, Minovitsky S, Dubchak I, Holt A, Lewis KD, Plajzer-Frick I, Akiyama J, De Val S, Afzal V, Black BL, Couronne O, Eisen MB, Visel A, Rubin EM. In vivo enhancer analysis of human conserved non-coding sequences. Nature. 2006;444:499–502. [PubMed]
  • Poliak S, Gollan L, Martinez R, Custer A, Einheber S, Salzer JL, Trimmer JS, Shrager P, Peles E. Caspr2, a new member of the neurexin superfamily, is localized at the juxtaparanodes of myelinated axons and associates with K+ channels. Neuron. 1999;24:1037–1047. [PubMed]
  • Poliak S, Salomon D, Elhanany H, Sabanay H, Kiernan B, Pevny L, Stewart CL, Xu X, Chiu SY, Shrager P, Furley AJ, Peles E. Juxtaparanodal clustering of Shaker-like K+ channels in myelinated axons depends on Caspr2 and TAG-1. J Cell Biol. 2003;162:1149–1160. [PMC free article] [PubMed]
  • Quandt K, Frech K, Karas H, Wingender E, Werner T. MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 1995;23:4878–4884. [PMC free article] [PubMed]
  • Robledo RF, Rajan L, Li X, Lufkin T. The Dlx5 and Dlx6 homeobox genes are essential for craniofacial, axial, and appendicular skeletal development. Genes Dev. 2002;16:1089–1101. [PubMed]
  • Roessler E, Belloni E, Gaudenz K, Jay P, Berta P, Scherer SW, Tsui LC, Muenke M. Mutations in the human Sonic Hedgehog gene cause holoprosencephaly. Nat Genet. 1996;14:357–360. [PubMed]
  • Roessler E, Ward DE, Gaudenz K, Belloni E, Scherer SW, Donnai D, Siegel-Bartelt J, Tsui LC, Muenke M. Cytogenetic rearrangements involving the loss of the Sonic Hedgehog gene at 7q36 cause holoprosencephaly. Hum Genet. 1997;100:172–181. [PubMed]
  • Rossi E, Verri AP, Patricelli MG, Destefani V, Ricca I, Vetro A, Ciccone R, Giorda R, Toniolo D, Maraschio P, Zuffardi O. A 12Mb deletion at 7q33-q35 associated with autism spectrum disorders and primary amenorrhea. Eur J Med Genet. 2008;51:631–638. [PubMed]
  • Scherer SW, Poorkaj P, Massa H, Soder S, Allen T, Nunes M, Geshuri D, Wong E, Belloni E, Little S, et al. Physical mapping of the split hand/split foot locus on chromosome 7 and implication in syndromic ectrodactyly. Hum Mol Genet. 1994;3:1345–1354. [PubMed]
  • Schule B, Li HH, Fisch-Kohl C, Purmann C, Francke U. DLX5 and DLX6 expression is biallelic and not modulated by MeCP2 deficiency. Am J Hum Genet. 2007;81:492–506. [PubMed]
  • Siebert PD, Chenchik A, Kellogg DE, Lukyanov KA, Lukyanov SA. An improved PCR method for walking in uncloned genomic DNA. Nucleic Acids Res. 1995;23:1087–1088. [PMC free article] [PubMed]
  • Simeone A, Acampora D, Pannese M, D'Esposito M, Stornaiuolo A, Gulisano M, Mallamaci A, Kastury K, Druck T, Huebner K, et al. Cloning and characterization of two members of the vertebrate Dlx gene family. Proc Natl Acad Sci U S A. 1994;91:2250–2254. [PubMed]
  • Strauss KA, Puffenberger EG, Huentelman MJ, Gottlieb S, Dobrin SE, Parod JM, Stephan DA, Morton DH. Recessive symptomatic focal epilepsy and mutant contactin-associated protein-like 2. N Engl J Med. 2006;354:1370–1377. [PubMed]
  • Verkerk AJ, Mathews CA, Joosse M, Eussen BH, Heutink P, Oostra BA. CNTNAP2 is disrupted in a family with Gilles de la Tourette syndrome and obsessive compulsive disorder. Genomics. 2003;82:1–9. [PubMed]
  • Verzi MP, Agarwal P, Brown C, McCulley DJ, Schwarz JJ, Black BL. The transcription factor MEF2C is required for craniofacial development. Dev Cell. 2007;12:645–652. [PMC free article] [PubMed]
  • Wieland I, Muschke P, Jakubiczka S, Volleth M, Freigang B, Wieacker PF. Refinement of the deletion in 7q21.3 associated with split hand/foot malformation type 1 and Mondini dysplasia. J Med Genet. 2004;41:e54. [PMC free article] [PubMed]
  • Zerucha T, Stuhmer T, Hatch G, Park BK, Long Q, Yu G, Gambarotta A, Schultz JR, Rubenstein JL, Ekker M. A highly conserved enhancer in the Dlx5/Dlx6 intergenic region is the site of cross-regulatory interactions between Dlx genes in the embryonic forebrain. J Neurosci. 2000;20:709–721. [PubMed]