Mutational analysis of GJB2, GJB6 and 12S rRNA genes in Vietnamese non-syndromic deaf children

Nguyen Thuy Duong1, Nguyen Thi Xuan1, Nguyen Hai Ha1, Nguyen Dang Ton1, Huynh Thi Thu Hue1, Phi Thi Thu Trang1, Duong Thi Thu Ha1, Nguyen Van Phong1, Nguyen Tuyet Xuong2, Ian Holt3, Nong Van Hai1* 1Institute of Genome Research, Vietnam Academy of Science and Technology, 18 Hoang Quoc Viet, Cau Giay, Hanoi, Vietnam 2National Hospital of Pediatrics, 18/ 879 De La Thanh, Dong Da, Hanoi, Vietnam 3Wolfson Centre for Inherited Neuromuscular Disease, RJAH Orthopaedic Hospital, Oswestry, United Kingdom.


ABSTRACT :
Congenital deafness is one of the most common disorders worldwide with a prevalence of one in every 300-1000 newborns (1). In Vietnam, 15,000-20,000 babies are born with congenital hearing loss (HL) each year [unpublished data]. Gene alterations including autosomal dominant, autosomal recessive, X-linked or mitochondrial mode of inheritance contribute about 50% of children with HL (2), and 70% of which are non-syndromic HL since HL is the only symptom and no clinical signals are found. The other half is caused by non-genetic factors (3) including ototoxic drugs specifically aminoglycosides, which increase the phenotypic expression of the mutations. Alterations in GJB2, GJB6 and mitochondria DNA 12S rRNA genes are thought to be the major causes of HL (4,5,6).
The GJB2 and GJB6 genes encoding the gap junction proteins, connexin (Cx) 26 and Cx30 respectively, are located on chromosome 13q11-q12. Cx26 and Cx30 are colocalized in gap junction plaques in the cochlea, form intercellular communication channels facilitating the circulation of potassium ions and small molecules between the cytoplasm of cochlear non-sensory epithelial cells (7). Cx26 and Cx30 share 77% identity in amino acid sequence (8) and have an important role in the cochlear physiology of hearing as ablation of the Cx26 or Cx30 proteins in cochlea leads to deafness in humans (5). Mutations in the GJB2 and GJB6 genes have been reported to be linked predominantly to HL. Concerning GJB2, a total of about 100 mutated variations causing HL have been detected varying from mild to profound in hearing impaired individuals (6). The predominance of each mutation varies between different ethnicities or countries. Up to 85% of Caucasians carry a 35delG mutation (8), while the 235delC mutation was found to be the common pathogenic mutation among Asians including Chinese (9), Korean (10) and Japanese (11) populations with the carrier rates of 20,3%, 6,9% and 49,8%, respectively. A c.167delT mutation was present in 4,03% of Ashkenazi Jewish (12) and 2% of Argentianian (13) populations. Many amino acid substitutions in the GJB6 gene have been identified to be responsible for HL. A 309-kb large deletion in GJB6, called del (GJB6-D13S1830), is the second most common mutation after the GJB2 mutations. The frequencies of this deletion detected in Western populations such as Spain, France, the United Kingdom and Brazil were 25,5%, 5%, 22,2% and 6,3%, respectively (14,15). Conversely, no mutation in the GJB6 was detected in Austrian, Chinese, Iranian and Italian populations (16,17,18). Mutations in the mitochondrial genes including the mitochondrial (mt) DNA 12S rRNA play an important role in causing maternally inherited syndromic and aminoglycoside-induced HL (19). The most common mutation in the 12S rRNA gene is a homoplasmic m.1555A>G mutation that was detected in European children (19), Chinese (20) and Korean (21) populations at frequencies of 0,19%, 1,48% and 0,9%, respectively. The m.1555A>G mutation is located at A-site, the target for aminoglycoside antibiotics (22). Therefore, this mutation is believed to be related to aminoglycoside induced-HL (19). In our study, mutational analysis of the deafness-related genes, GJB2, GJB6 and 12S rRNA was investigated in 76 patients with congenital HL and 78 healthy individuals. All subjects were examined to determine the prevalence of these gene mutations in Vietnamese populations.

Study subjects
The study subjects were a total of 76 non-syndromic congenital deaf children who were enrolled at Nhan Chinh School for the Deaf and the Dumb, Hanoi, Vietnam. As a control group, 78 healthy Vietnamese blood donors were analyzed. The parents or guardians of the subjects signed written informed consent forms prior to their participation in the study. The subjects were all from the Kinh ethnic background, in the age range of 2 to 15 years old and clinical data were recorded about their mother's heath during pregnancy. A detailed history of disease and use of aminoglycoside antibiotics were obtained from all subjects in patient and control groups to exclude other causes of HL. The study was approved by the Scientific Committee of Institute of Genome Research under reference number 02/QD-NCHG. DNA sequencing Genomic DNA was isolated from peripheral blood samples using a DNeasy blood and tissue kit (Qiagen, Hilden, Germany). To determine mutations of the GJB2, GJB6 and mitochondrial 12S rRNA genes, polymerase chain reaction (The Veriti® Thermal Cycler, Applied Biosystems, USA) and DNA sequencing (ABI Prism@ 3100 and/or 3500 Genetic Analyzers, Thermo Scientific, USA) were performed as previously described (23). DNA sequence variations were identified by comparing subject DNA sequence to the GJB2, GJB6 and 12S rRNA reference sequences: Genbank Accession Numbers NM_004004.5, NM_001110219.2 and NC_001807.3, respectively. The coding region of the GJB2 was amplified by using primers: GJB2-F: 5'-CGTCTTTTCCAGAGG-CAAACCG-3' and GJB2-R: 5'-GCTAGCGACT-GAGCCTTGAC-3' attaining a-802 base pair (bp) fragment. The GJB6 was amplified by using primers: GJB6-F: 5'-CACCGTGTACATTTCCCAAG-3' and GJB6-R: 5'-CCAGAAGGCAATCCCAACCT-3' . The amplification product length was 933-bp. The 12S rRNA was amplified by using primers: 12S rRNA-F: 5'-CCCACAGTTTATGTAGCTTAC-3' and 12S rR-NA-R: 5'-CAGAACACTACGAGCCACA-3' . The amplification product length was 1100-bp. All obtained PCR fragments were purified with a GeneJET PCR purification kit (Thermo Scientific, USA). The PCR products were sequenced on both strands with the same primers used for the PCR.

Statistical analysis
Fisher's exact test was used for group association. The level of significance in all cases was set at p < 0.05. RESULTS Mutational analysis of the GJB2 and the GJB6 genes Sequencing of the GJB2 coding region identified four nucleotide changes in this gene (Table.1 and Fig.1). These were a 235delC pathogenic mutation (Fig.1A) and three known variants including p.V27I (c.79G>A), p.V37I (c.109G>A) and p.E114G (c.341A>G) identified (Fig.1B-D). The 235delC mutation was detected in three affected subjects, accounting for 3,95% of the patient group and none of the control group carried this mutation. In the present small study, the occurrence of the 235delC mutation was not significantly different between patient and control groups (P>0.1), but this mutation has been found at higher frequencies in several Asian populations such as Chinese (20,3%) (9), Korean (6,9%) (10) and Japanese (7%) (11) populations, indicating different genetic backgrounds among Asians. The non-pathogenic variants in the GJB2, considered polymorphisms (p.V27I, p.V37I and p.E114G) have been described previously (24). As shown in Table 1 the p.V27I, p.V37I and p.E114G variants were identified in all patient and healthy control groups with the carrier frequencies ranging from 1,28% to 17,1%. The frequencies of the p.V27I, p.V37I and p.E114G variants all found in French population were less than 2% (25). We additionally found both a two nucleotide change containing p.V27I and p.E114G and a variant with a three nucleotide change containing p.V27I, p.V37I and p.E114G which were carried by both patient and healthy control groups. The variant with a two nucleotide change has been reported to occur at a frequency of 0,78% in an Iranian population (26). The GJB6 gene has been shown to cause deafness mainly through the large deletion called del(D13S1830). The frequencies of the deletion detected in Spain, France, the United Kingdom and Brazil were 25,5%, 5%, 22,2% and 6,3%, respectively (14,15). In our study, none of the patient or control group subjects carried mutations or deletions in the GJB6 gene.

Mutational analysis of the mitochondrial 12S rRNA gene
In the current study, a total of seven nucleotide changes in the 12S rRNA gene were identified ( Table 2). Of these nucleotide changes, a m.1438A>G variant (Fig.  2G) was present at a frequency of 100% in our population. This frequency was comparable to that reported in Chinese, Korean and Tunisian populations (27,21,20). In contrast, this variant was found in less than 2% of Iranian subjects (28). The remainder were six known variants including m.709G>A, m.813A>G, m.980T>C, m.1048C>T, m.1107T>C nd m.1119T>C ( Fig. 2A-G and Table. 2). The m.709G>A and m.1119T>C variants were identified in both patient and control groups with the carrier frequencies of (21% versus 12,82%) and (3,95% versus 12,82%), respectively. Similarly, the m.709G>A variant ocurred in both patient and control groups at frequencies of 21,1% and 19,7% in Korean, of 2,2% and 1,23% in Chinese and of 14,9% and 15% in Iranian populations, respectively (21,20,28). The m.1119T>C variant was found in both patients and control subjects in Korean and Iranian populations at a rate ranging from 1,4 to 4,4% (21,28), whereas 2,2% of affected subjects and none of a control group carried this variant in a Chinese population (20). In addition, we observed a two nucleotide change containing m.709G>A and m.980T>C and also a variant with a three nucleotide change containing m.709G>A, m.813A>G and m.1119T>C which were identified in both patient and healthy control groups. In other studies, the m.980T>C variant detected in both patient and control groups in an Iranian population were 2,8% and 2%, respectively (28) and the m.813A>G variant was carried by 1% of hearing impaired subjects in a Tunisian population (27). We additionally found a variant with a two nucleotide change containing m.1048C>T and m.1107T>C which was carried by 1,32% of the patient group but not by the control group. Other studies indicated that the m.1048C>T variant found in the Tunisian population was present in 2% of control subjects (27) and also in 1,48% of Chinese (20) and 0,9% of Iranian (28) populations with HL. Therefore, different variants of this gene occur in different ethnic backgrounds.

DISCUSSION
The present study is the first report of the association between congenital deafness and mutations in the GJB2, GJB6 and 12S rRNA genes in the Vietnamese Kinh population. In this study, we observed that the pathogenic 235delC mutation in GJB2 occurred with a frequency of 3,95% in the affected subjects and mutation in GJB6 were absent in all subjects in patient and control groups. Previous reports showed that the frequencies of the 235delC mutation were higher in several populations, such as Chinese (20,3%) (9), Korean (10) and Japanese (7%) (11) populations (Table.3). Therefore, the 235delC mutation needs further analysis to determine whether it is a predominant mutation causing HL in the Vietnamese population.  jor causes of congenital deafness. In this study we observed three nucleotide changes in the GJB2 which were non-pathogenic polymorphisms. Among these, the p.V37I variant has been considered as both pathogenic and non-pathogenic and occurred with the high relative carrier frequency of 17,1% and 10,26% in the patient and control groups, respectively (Table. 1). The results were comparable to those reported in other investigations in Japanese (11) and Chinese (9) hearing impaired subjects. Whereas, the variant was detected at lower frequencies in a French population in less than 2% of patients with HL (25). A very common mutation in the GJB2 gene causing HL, a 35delG mutation, was found in most hearing impaired individuals. In a Caucasian population, the 35delG accounted for up to 85% of all the GJB2 mutations in deaf children (8). However, the patient group in our observation did not carry this mutation, which was similarly to previous studies in Korean and Japanese populations (10,11,24). Therefore, this observation suggested that there was a different geographic distribution of the GJB2 mutations causing HL.
Concerning GJB6, the present study was in agreement with previous reports in Austrian, Chinese, Iranian and Italian populations (16,17,18) showing that no mutation in the GJB6 was detected in the Vietnamese population, so indicating that the common cause of HL in our cohort was not due to alterations in the GJB6 gene. Mutations in the 12S rRNA gene has been indicated to be responsible for both non-syndromic and aminoglycoside-induced pediatric HL (29). In the current study, there were seven kinds of nucleotide changes in the 12S rRNA identified. Among the variants in the 12S rRNA gene found, the m.1438A>G variant was present in 100% of all subjects, indicating that this variant might be one of the most common polymorphisms in the Vietnamese population. In addition, the mutation m.1555A>G is thought to be the most common mitochondrial mutation associated with HL found in many populations such as European children (19), Chinese (20) and Korean (21) populations (Table.3). However, this mutations was not identified in our population as well as several populations including Mexican (30). In this study, no difference in seven nucleotide alterations in the 12S rRNA between patient and control groups was found, suggesting that these variants were not related to HL in Vietnamese population.
Our data is an initial investigation of genetic causes of non-syndromic deafness in the Vietnamese population. However, further investigation with a larger sample size is necessary to determine the significance of these gene mutations in Vietnamese deafness.