Characterization of the murine Dfna5 promoter and regulatory regions
Introduction
Mutations in DFNA5 cause a non-syndromic, progressive, autosomal dominant, sensorineural type of hearing loss. Patients from the four families segregating DFNA5 mutations all have a highly similar phenotype. At the genomic DNA level, four different mutations were detected in DFNA5 in different families. All these mutations led to skipping of exon 8 at the mRNA level (Van Laer et al., 1998, Yu et al., 2003, Bischoff et al., 2004, Cheng, 2007). Moreover, a mutation in exon 5 that did not affect exon 8 splicing did not cause hearing loss in an Iranian family (Van Laer et al., 2007). The discovery of four mutations all leading to exon 8 skipping, led to the formulation of the hypothesis that DFNA5-related hearing loss was associated with a very specific gain of function event, in which only skipping of exon 8 causes disease while other mutations in the gene do not (Van Laer et al., 2004). This hypothesis was supported by experimental evidence demonstrating that transfections with mutant DFNA5 lead to cell death, both in yeast (Gregan et al., 2003) and in mammalian cells (Van Laer et al., 2004).
DFNA5 is ubiquitously expressed. The highest expression levels were detected in placenta, while lower expression was observed in brain, heart and kidney (Thompson and Weigel, 1998, Van Laer et al., 1998). Using RT-PCR, DFNA5 expression, albeit at low levels, was found in every tissue investigated so far (nine in total, unpublished results).
The physiological function of DFNA5 remains largely unknown. For a long time, DFNA5 was considered an orphan gene. Recently, DFNB59, a paralogue of DFNA5, has been described. Mutations in DFNB59, encoding Pejvakin, cause auditory neuropathy and non-syndromic hearing impairment in mice and men (Delmaghani et al., 2006, Hashemzadeh Chaleshtori et al., 2007, Schwander, 2007). As DFNB59 mutations may cause hearing loss through diverse pathogenic mechanisms, unfortunately the identification of DFNB59 did not shed light on the mechanism by which DFNA5 may cause hearing loss.
Although not much is known regarding the physiological function of DFNA5, we can certainly state that it is not only related to hearing loss. A clear link with cancer exists. In a recent paper DFNA5 was even catalogued as tumour suppressor gene (Kim et al., 2008b). DFNA5 has also been designated ICERE-1 (inversely correlated with estrogen receptor expression) due to its lower expression in estrogen receptor (ER)-positive breast cancers compared to ER-negative tumours (Thompson and Weigel, 1998). In melanoma cells, etoposide resistance was associated with decreased expression of DFNA5 (Lage et al., 2001). Etoposide is used as an anti-neoplastic agent and causes DNA strand breakage via inhibition of DNA topoisomerase II (Glisson and Ross, 1987). These results may indicate that DFNA5 plays a role in melanoma and breast cancer progression and in the resistance to chemotherapy. P53 is a well-known tumour suppressor gene. Upon DNA damage or other cellular stress situations, various pathways will lead to the activation of p53, inducing either cell cycle arrest to allow repair and survival or apoptosis, depending on the damage. Expression of DFNA5 was strongly induced both by endogenous and exogenous p53 in human cancer cell lines and a functional p53-binding site was identified in intron 1 of DFNA5. These observations suggested that DFNA5 may be involved in the p53-mediated cellular response to genotoxic stress (Masuda et al., 2006). Further evidence supporting a role for DFNA5 in cancer came from methylation studies. DFNA5 was silenced in breast cancer cell lines by methylation of the 5′ flanking region of the gene (Kim et al., 2008b). Previously, methylation of the DFNA5 promoter region was associated with gastric (Akino et al., 2007) and colorectal cancer (Kim et al., 2008a).
So far, the promoter region and the regulatory motifs for DFNA5 expression have not been characterized. However, knowledge of the regulatory regions of DFNA5 is particularly interesting as the above-mentioned studies implicate that pathophysiological downregulation of DFNA5 plays a role in breast, gastric and colorectal cancer. This implies that therapies based on the activation of DFNA5 expression might provide an effective treatment for these types of cancers. Here, we describe the identification of the Dfna5 transcription initiation site (TIS), the characterization of the regions responsible for basal transcriptional activity, and the presence of enhancing and silencing elements in the Dfna5 5′-flanking region. Furthermore, we provide evidence for the presence of a Dfna5 antisense transcript.
Section snippets
In silico analysis
Promoter regions were predicted in silico using different computer modelling programs, listed in Supplementary Table 1. CpG island prediction was performed using both the CpG island searcher (www.uscnorris.com/cpgislands2/cpg.aspx) (Takai and Jones, 2003) and the EMBOSS CpG prediction tool (www.ebi.ac.uk/cpgplot/). Transcription factor (TF) binding sites were scored using three programs, MatInspector (www.genomatix.de), NSITE (http://softberry.com/) and ProScan (//www-bimas.cit.nih.gov/molbio/proscan/
In silico investigation of the Dfna5 upstream region
In order to localize the promoter region, the 5′ flanking region of Dfna5 was first investigated in silico. The Eldorado, Proscan, TSSG, NNPP and Promoter 2,0 modelling programs respectively predicted 2, 3, 3, 21 and 3 potential promoter regions in an 11 kb fragment flanking Dfna5 (Refseq Genbank accession no. NT_039353.7, nt2523000 to nt2534000). Promoser and McPromoter did not predict any promoter elements in the investigated region. A detailed view of the predicted promoter regions contained
Discussion
Here we report the localization of the TIS of cochlear dfna5 mRNA, the identification of the core promoter and additional regulatory regions required for the expression of dfna5. Furthermore we report the identification of an antisense Dfna5 transcript.
The computer programs that we have used to identify the putative core promoter in silico resulted in diverging predictions. As recent studies that compared eight different promoter prediction programs indicated that true prediction rates vary
Acknowledgments
The authors thank Dr F. Kalinec from the House Ear Institute, LA, USA for providing the OC-k3 cell line. This work is supported by FWO grant 1.5.048.05N and by the European Commission FP6 Integrated Project: Eurohear (LSHG-CT-20054-512063).
References (39)
- et al.
Nucleofection of primary neurons
Methods Enzymol.
(2006) - et al.
DNA topoisomerase II: a primer on the enzyme and its unique role as a multidrug target in cancer chemotherapy
Pharmacol. Ther.
(1987) - et al.
A yeast model for the study of human DFNA5, a gene mutated in nonsyndromic hearing impairment
Biochim. Biophys. Acta
(2003) - et al.
The adenovirus oncoprotein E1a stimulates binding of transcription factor ETF to transcriptionally activate the p53 gene
J. Biol. Chem.
(1999) - et al.
Regulatory roles of AP-2 transcription factors in vertebrate development, apoptosis and cell-cycle control
Gene
(2000) - et al.
Establishment and characterization of conditionally immortalized organ of corti cell lines
Cell Biol. Int.
(1999) - et al.
DFNA5 (ICERE-1) contributes to acquired etoposide resistance in melanoma cells
FEBS Lett.
(2001) - et al.
Pattern of degeneration of the spiral ganglion cell and its processes in the C57BL/6J mouse
Hear. Res.
(2000) - et al.
EnsemPro: an ensemble approach to predicting transcription start sites in human genomic DNA sequences
Genomics
(2008) - et al.
A 3-nucleotide deletion in the polypyrimidine tract of intron 7 of the DFNA5 gene causes nonsyndromic hearing impairment in a Chinese family
Genomics
(2003)
An ENU-induced mutation in AP-2alpha leads to middle ear and ocular defects in Doarad mice
Mamm. Genome
Identification of DFNA5 as a target of epigenetic inactivation in gastric cancer
Cancer Sci.
A novel mutation identified in the DFNA5 gene in a Dutch family: a clinical and genetic evaluation
Audiol. Neurootol.
A novel DFNA5 mutation, IVS8 + 4 A > G, in the splice donor site of intron 8 causes late-onset non-syndromic hearing loss in a Chinese family
Clin. Genet.
Mutations in the gene encoding pejvakin, a newly identified protein of the afferent auditory pathway, cause DFNB59 auditory neuropathy
Nat. Genet.
CpG island hypermethylation and tumor suppressor genes: a booming present, a brighter future
Oncogene
Compilation of vertebrate-encoded transcription factors
Nucleic Acids Res.
Clustering of DNA sequences in human promoters
Genome Res.
Novel mutations in the pejvakin gene are associated with autosomal recessive non-syndromic hearing loss in Iranian families
Clin. Genet.
Cited by (9)
Identification of duck GSDME: Tissue distribution, proteolysis and cellular location
2022, CytokineCitation Excerpt :Human GSDME is often expressed in the placenta, heart, brain and kidney [11–13]. Mouse GSDME is expressed in the large intestine and brain [11,14,15]. The characteristics and function of GSDME in waterfowl have not yet been elucidated.
Transcriptional and Epigenetic Regulation of Gasdermins
2022, Journal of Molecular BiologyCitation Excerpt :Importantly, specific expression pattern may contribute to explain specific phenotypes associated with GSDM activation or GSDM mutations. Indeed, GSDME is expressed in cochlear hair cells.15 Specific gain of function mutations in GSDME, lead to exon 8 skipping, shortening of its inhibitory domain and result in a constitutive truncated GSDME mutant, causing cochlear cell death.16
The Versatile Gasdermin Family: Their Function and Roles in Diseases
2021, Frontiers in ImmunologyGasdermins: Pore-forming activities and beyond
2021, Acta Biochimica et Biophysica SinicaDFNA5, a gene involved in hearing loss and cancer: A review
2012, Annals of Otology, Rhinology and LaryngologyGasdermin superfamily: A novel gene family functioning in epithelial cells
2012, Endothelium and Epithelium: Composition, Functions and Pathology