IDENTIFICATION OF A TRUNCATING MUTATION IN THE HOMEODOMAIN (HD) OF PAX3 IN THREE GENERATIONS OF A FILIPINO FAMILY WITH WAARDENBURG SYNDROME TYPE 1

Part of a scientist's duty is to inform people - in layman's terms -
scientific breakthroughs that are relevant to society and to each one.
Thus, while preparing for an oral presentation this week for an international
scientific and engineering conference (PAASE), I thought it's good to post
the abstract of my undergraduate thesis.

Enjoy reading!

**********

IDENTIFICATION OF A TRUNCATING MUTATION IN THE HOMEODOMAIN (HD)
OF PAX3 IN THREE GENERATIONS OF A FILIPINO FAMILY
WITH WAARDENBURG SYNDROME TYPE 1

Andrew Agunod, Jr. and Cynthia Palmes-Saloma, Ph.D.
Laboratory of Molecular and Cell Biology, National Institute of Molecular Biology and Biotechnology
University of the Philippines Diliman, Quezon City

ABSTRACT

Waardenburg Syndrome is a rare autosomal dominant genetic disease characterized by defects of neural crest (NC) origin arising from mutations in genes associated with NC cell migration, differentiation, and survival during embryogenesis. We present three generations of a Filipino family with eight (8) Waardenburg Syndrome type 1 (WS1) affected members showing congenital unilateral or bilateral hearing loss, eye pigmentation defects, patchy hypopigmentation of the skin, white forelock, and dystopia canthorum. Mutations in the PAX3 gene which encodes a trancsription factor have been implicated in WS1. In this study, we screened for mutations in the DNA-binding paired domain (PD) and homoeodomain (HD) of Pax3 in all affected members. The paired domain which is frequently involved in WS1 cases and is encoded by exons 2 and 3 contained no mutations. However, in all affected members, there was a C→T transition in nucleic acid 1033 (C1033T) resulting to a nonsense mutation in codon 233 (R223X) affecting the homeodomain region, thus indicating that a mutation in one of the alleles of PAX3 is sufficient to cause the repertoire of neural crest abnormalities in WS1-affected individuals.

INTRODUCTION

Waardenburg Syndrome (WS) is an autosomal dominant neural crest abnormality associated with most common cases of congenital deafness. It has an incidence rate of 1 in 40,000 individuals and is known to exhibit wide genetic and clinical heterogeneity [1]. One of the four types of the disease is Waardenburg Syndrome 1 (WS1) which is characterized by dystopia canthorum (lateral displacement of the inner canthi of the eyes), congenital unilateral or bilateral sensorineural deafness, heterochromia irides (defect in eye pigmentation), white forelock, and patchy hypopigmentation of the skin.

The only causative gene associated with WS1 is the PAX-3 gene, coding for a transcription factor expressed by developing cells during embryogenesis. The neural crest cells that originate from the dorsal region of the neural tube actively express Pax3 which is required for their survival and differentiation into various tissues including the neurons in the peripheral nervous system, ganglia in the gut, pigment cells in the skin and eyes, and facial cartilage. PAX-3 is a member of the paired type gene family conserved among different species. It contains a paired box domain, an octapeptide, a homeodomain, and a Ser-Thre-Pro-rich COOH terminus. The paired box domain and homeodomain are conserved regions that bind to DNA and are essential for proper functioning of the Pax3 transcription factor which directly activates, in synergy with Sox10, the MITF gene that encodes another transcription factor critically involved in melanocyte differentiation through its action on tyrosinase gene expression [2]. Mutations in the Splotch locus of chromosome 1 in mice generate an animal model for WS1.

Sequence analysis of more than 90% of individuals with WS1 showed mutations in the PAX3 gene varying from simple single point mutations to large deletions. Mutations are numerous in the conserved regions of the paired box and homeobox regions found in exons 2-4 and 5-6 respectively. Knowledge on how these mutations caused the phenotypic characteristics seen in WS1 as well as the epistatic relationship between MITF and PAX3 have provided new insights on the role of genetic factors in the emergence of a disease phenotype that could help in the prognosis and management of the disorder.

In this study, we show that WS1 in three generations of a Filipino family is due to a C → Ttransition mutation in nucleic acid 1033 (C1033T) in the homeodomain region of one of the alleles of PAX3 resulting to a stop codon TGA in amino acid 233 of the transcription factor.

METHODOLOGY

Identification of a WS1 family. A three generation family with WS1-affected members was diagnosed by physicians studying pigmentary defects and was referred to us for further genetic workup. Affected individuals exhibited dystopia canthourm, a defining feature for WS1. Blood samples were collected from eight (8) available members of the WS1 family. A written consent was signed by the family members granting the researchers permission to use pertinent data and photographs for research publication purposes.

Genotyping and sequence analysis. Total genomic DNA was extracted from whole blood using Wang’s method. Primers were designed to amplify exons 2, 3 and 4, coding for the paired box, and exons 5 and 6 for the homeodomain of Pax3. Primers were designed to anneal at the intron-exon boundaries. Large scale PCR amplification was performed and the resulting products were gel-purified using Wizard® Minicolumn (Promega). Direct DNA sequencing using ABI-PRISM™ 377 DNA sequencer was then performed for all the samples. Sequences obtained were analyzed using Chromas™, Sequencher™ version 4.0.5 (Gene Codes Corporation, Ann Arbor, Michigan, USA) and Multalin™ (Corpet, 1988; http://prodes.toulouse.inra.fr/multalin/multalin.html). The PAX3 reference sequence was obtained from the Ensembl (www.ensembl.org) database.

RESULTS and DISCUSSION

Waardenburg Syndrome Type 1 is a genetic disease characterized by defects in neural crest cells during embryogenesis and has an occurrence of 1 in 40,000 individuals worldwide [3]. The pedigree of the three-generation family in this study exhibits an autosomal dominant mode of inheritance, confirming early investigations. There are 8 WS1-affected members in the family. The proband (II-5) has an affected father (II-1) and 3 of her siblings (II-1, II-10, and II-12), including herself, are also affected. Two (2) out of her 4 children (III-4 and III-6) inherited the disease while all the children of her normal siblings did not show any signs of WS1 (Figure 1 A). This pattern of inheritance where 50% of the children of an affected parent inherits the disease is characteristic of an autosomal dominant disorder.

Waardenburg syndrome is known to exhibit wide clinical heterogeneity due to differences in disease penetrance [1]. Affected members shows variable degrees of skin hypopigmentation, hearing loss, and eye and hair pigmentation abnormalities (Figure 1B). All were diagnosed with dystopia canthorum which is the defining feature for WS1. The full blown WS1 phenotype was observed in the sister and daughter (middle photo, Fig 1B) of the proband, who are both deaf-mute due to profound bilateral deafness, and have blue eyes and distinctive white forelocks. These symptoms were completely absent in normal members. The phenomenon of WS1 disease penetrance remains highly unclear. Moreover, while it was observed that some individuals with the same PAX3 mutation have varied phenotypes, there are those with a single base mutation and whole gene deletion that have similar features [4]. Nevertheless, genetic background and nonrandom environmental factors are pointed out to influence disease etiology [5].

The paired domain region is a critical binding domain of PAX3. Since the paired domain is upstream of the homeodomain, an abrogation at exon 2, 3 or 4 can lead to the abrogation of the homeodomain as well. Though mutations are numerous in both areas, mutations in the paired domain seemed more critical and more frequent in studies of WS1 families. PCR amplification of exons 2, 3 and 4 from the proband yielded fragments of 330, 312 and 382 bps, respectively. Products from exons 2 and 3 were gel-purified and sequenced. DNA sequences were visualized and edited using Chromas™ and sequence alignment was performed using Multalin™. The two exons subjected to direct sequencing revealed clean chromatograms with no indication of any sequence variations from the control. This result indicates that the paired domain of the proband is unaffected and suggests the existence of mutation(s) in some other part of the gene.

Primers FP and RP designed to amplify exon 5 showed products 240 bp in size. Products of the same size and intensity were amplified for all samples (Fig. 2A), eliminating the possibility of a large-scale deletion in the gene. Direct sequencing of Pax3 exon 5 DNA from the proband revealed a double peak for C and T as shown in Fig. 2B. The wild type C is located at nucleic acid 1033 and is part of codon 233 coding for arginine. Substitution by a mutant T (C1033T) in the sequence results to a stop codon (R223X) truncating the protein. PCR products were also sequenced indirectly by cloning in TOPO® vector. Sequencing revealed two alleles: a wild-type exon 5 with a C at nucleic acid 1033, and a mutant allele with a T at the same location (Fig. 2C). The same non-sense mutation was first described by Baldwin et al. [4] and is documented in the Human Gene Mutation Database (HGMD CM984205 www.hgmd.cf.ac.uk) as one of the causative mutations of WS1. This truncating mutation is detrimental to the proper functioning of Pax3 as it produces a protein without a DNA-binding homeodomain. Fortin et al [6] showed that the paired domain and homeodomain regions of Pax3, though individually possessing different sequence preference and high affinity as modular units, function interdependently and cooperatively in binding DNA in vivo. This implies that a mutation in either domain can lead to an impaired or dysfunctional transcription factor.

Direct sequencing of samples from other members of the family showed the presence of the same transition mutation C1003T in all WS1 individuals. The father, sister, son2, and daughter are heterozygous at the same region in the chromatogram while the unaffected husband, son1, and son3 demonstrated normal peaks (Fig. 3). This showed that the heterozygous allele C1033T was inherited by the proband from the father and was passed on to the third generation of her family.

One of the known functions of Pax3 is to transactivate the expression of MITF which is known as the master gene for melanogenesis (OMIM 156845). An impaired Pax3 affects the development of neural crest cell-derived melanocytes due to deficient levels of expression of the MITF gene that codes for a transcription factor which directs the expression of genes indispensable for melanocyte development such as the tyrosinase gene, TRP1, and TRP2 [2]. Melanocytes are responsible for pigmentation in the skin, eyes, and hair. They also form part of the cochlea of the inner ear. This explains the disease phenotype of WS1 patients where patients suffer from pigmentary anomalies in the skin, hair, and eyes and partial or bilateral hearing loss. Dystopia canthorum is present in almost all cases of WS1 and is considered as the defining feature for the genetic disorder. The phenotype is a direct consequence of mutations in Pax3 which is involved in the proper formation of facial structures.

In mice, a homozygous mutation in PAX3 is embryonic lethal due to an impaired development of the central nervous system. In humans, heterozygous PAX3 mutations lead to WS3, a more severe type with limb abnormalities. In another study, Pax3 was revealed to act in synergy with Sox10 in transactivating MITF expression [7].This provided an explanation for the molecular basis of WS4, another WS variant with additional feature similar to Hirschsprung Disease, implicating SOX10.

Primers designed to amplify the paired and homeodomains have been demonstrated to successfully amplify the desired regions from both wildtype and affected individuals. These primers can be excellent tools for screening for mutations in the Pax3 gene of WS1 patients.

CONCLUSION

Waardenbrug Syndrome Type 1 (WS1), a rare autosomal dominant disease, is known to be caused by mutations in the PAX3 gene. Pax3, a transcription factor, important for the development of neural crest cells, contains two DNA binding domains, namely the paired box and the homeodomain. Mutations in both regions have been known to cause WS1. The WS1 phenotype was observed in a three-generation family exhibiting defects in neural crest cell development such as eye pigmentation abnormality, skin hypopigmentation, whit forelock, dystopia canthorum, and sensorineural defects. Sequence analysis successfully established that exons 2 and 3 of the preoband was similar to the wildtype control. Direct and indirect sequencing of the homeodomain region of PAX3 revealed a heterozygous transition mutation in nucleic acid 1033 (C1033T) that resulted to a nonsense mutation R223X. This point mutation resulted to a truncated protein responsible for the characteristic WS1 phenotype in the three-generation family.

REFERENCES
(1) Waardenburg PJ. 1951. A new syndrome combining developmental anomalies of the eyelids, eyebrows and nose with pigmentary defects of the iris and head hair and with congenital deafness. Am Jorn of Hum Genetics 3:195-253.
(2) Bondurand N, Pingault v et al. 2000. Interaction among Sox10, Pax3, and MITF, three genes altered in Waardenburg syndrome. Hum Nolec genet. 9:1907-1917.
(3) Read AP and Newton VE. 1997. Waardenburg syndrome. J. Med. Genet. 34: 656-665. Retrieved March 21, 2006 from http://jmg.bmjjournals.com/cgi/content/abstract/34/8/656
(4) Baldwin C, Hoth C, Macina R and Milunsky A.1995. Mutations in PAX3 that cause Waardenburg syndrome type 1: ten new mutations and review of the literature. Am J of med genet 58:115-122.
(5) Morell R, Friedman, TB, Asher JH, Robbins LG. 1997. The incidence of deafness is non-randomly distributed among families segregating for Waardenburg Synsrome type 1 (WS1). J. Med. Genet 34:447-452.
(6) Fortin A, Underhill D and Gruss P. 1997. Reciprocal effectof Waardenburg syndrome mutations on DNA binding by the Pax 3 paired domain and homeodomain. Hum Molec Genet 6:1781-1790.
(7) Potterf SB, Furumura M, Dunn KJ, Arnheiter H, Pavan WJ. 2000. Transcription factor hierarchy in Waardenburg syndrome: regulation of MITF expression by SOX10 and PAX3. Hum. Genet. 107:1-6. Abstract retrieved March 20, 2006 from Pubmed database at http://www.pubmedcentral.nih.gov