Journal of Bone and Mineral Research

Genetic Association and Gene Expression Profiles of TGFB1 and the Contribution of TGFB1 to Otosclerosis Susceptibility



Otosclerosis (OTSC) is a common form of acquired hearing loss resulting from disturbed bone remodeling in the otic capsule of the middle ear. Transforming growth factor-beta1 (TGFB1) produced by osteoblasts is the most abundant growth factor in human bone. Previous studies have shown the contribution of single-nucleotide polymorphisms (SNPs) in TGFB1 toward the risk of developing OTSC in some ethnic populations. The present study was aimed at investigating the genetic association and expression profiles of TGFB1 in OTSC patients. Two SNPs (c.–800G > A and c.–509C > T) in the promoter region and three SNPs (c.29T > C, c.74G > C, and c.788C > T) in the coding region were genotyped in 170 cases and 170 controls. The genetic association analysis revealed the significant association between c.–509C > T (p = 0.0067; odds ratio [OR] = 1.562; 95% confidence interval [CI], 1.140–2.139) and OTSC. The increased minor allele “T” frequency in cases (0.42) compared to controls (0.31) indicates its possible role in the etiology of the disease. The minor allele frequencies for the SNPs c.–800G > A, c.29T > C, and c.74G > C were similar among the cases (0.04, 0.47, and 0.08, respectively) and controls (0.05, 0.42, 0.07, respectively). We found that c.788C > T was monomorphic in this population. Interestingly, a four-locus haplotype (G-T-T-G) from these SNPs was found to be significantly associated with OTSC (p = 0.0077). We identified a de novo heterozygous mutation c.–832G > A in the promoter region of TGFB1 in 1 patient. In a secondary analysis, we investigated the possibility of abnormal TGFB1 expression and irregular bone growth in OTSC by expression analysis of TGFB1 mRNA in disease tissue compared to control. We found relatively increased expression of TGFB1 mRNA in the stapes tissues of cases compared to controls (p = 0.0057). In conclusion, this study identified a risk variant c.–509C > T and a risk haplotype G-T-T-G in the TGFB1 gene that contribute toward the susceptibility to OTSC. © 2013 American Society for Bone and Mineral Research.


Otosclerosis (OTSC) is a condition producing hearing loss resulting from impaired stapedial movement that occurs as a result of abnormal bone growth in the otic capsule. Initially, otosclerotic outgrowth is located near the pole of the anterior or posterior crura and leads to conductive hearing loss. Later on, the otosclerotic lesion involves other parts of the otic capsule and causes sensorineural hearing loss.[1] The typical age of onset of OTSC is in the third decade; however, the range extends from the late teenage years to the sixth decade of life.[2] In the majority of patients, both ears are involved in hearing impairment.[3] The prevalence of clinical OTSC varies from race to race and has been found to be more frequent among whites; it is rare in Asians, Blacks, and Native Americans.[3] Various hypotheses regarding the etiology of OTSC have been proposed, including genetic and environmental factors. An environmental factor such as measles virus infection has been proposed as the causative factor for OTSC. However, the theory of virus etiology remains unsettled and controversial.[4]

The genetic predisposition to OTSC has been recognized after the identification of 10 monogenic loci (OTSC1–OTSC5, OTSC7, OTSC8, and OTSC10).[5-12] The causative genes for the condition within these regions have not yet been found. Over the past 10 years, population-based case-control studies have shown the genetic association of COL1A1 (MIM #120150), transforming growth factor-beta1 (TGFB1; MIM #190180), BMP2 (MIM #1122610), BMP4 (MIM #112261), and RELN (MIM #600514) genes in disease development.[13-16] Though studies have shown the association of the RELN gene single-nucleotide polymorphism (SNP) rs3914132 with OTSC,[14, 17, 18] our study[3] failed to detect a significant association between the RELN SNP and OTSC in the Indian population.

The association between TGFB1 SNPs and OTSC is a most interesting genetic observation because TGFB1's role is obvious in the early development of the otic capsule.[16] The active form of TGFB1 is a 25-kDa disulfide-linked dimmer of two identical chains containing 112 amino acids, synthesized in latent form as a protein of 390 amino acids. TGFB1 is a multifactorial cytokine produced by osteoblasts and stored in the bone matrix; it also acts as a regulator molecule for both skeletal development and homeostasis. TGFB1 released during bone resorption and activated in an acidic microenvironment stimulates the proliferation and differentiation of osteoblasts and induces bone formation.[19] In the early stage of ear development, TGFB1 acts as an epithelial-derived regulatory molecule that stimulates otic capsule chondrogenesis and later on inhibits this process to permit perilymphatic space formation and capsular modeling.[16] TGFB1 mRNA expression was observed in the epithelial-derived sensory anlagen of the developing mouse and human inner ear.[20] In OTSC, abnormal bone growth at the stapedial footplate reflects either increased activity of osteoblasts or reduced bone resorption by osteoclasts. Previous studies (in the literature) have suggested that TGFB1 released from bone matrix decreases bone resorption by reducing osteoclast activity, which is a result of reduced receptor activator of NF-κB ligand (RANKL) production by osteoblasts.[21] It has been found that TGFB1 strongly inhibits the mRNA expression level of RANKL in osteoblast cells.[22] Although RANKL is essential for osteoclast formation, factors such as TGFB1 may powerfully modify these osteoclastogenic stimuli. It has been found that a low concentration of TGFB1 stimulates osteoclast precursors for osteoclast differentiation and vice versa.[23] The exact molecular mechanism that links the bone resorption to bone formation is not yet known.

Recently, microarray-based gene expression profiling in human stapes tissue samples from OTSC cases and controls has shown the differential expression of two genes (PF4: upregulated, and IBSP: downregulated) playing roles in TGFB1 signaling.[24] PF4 acts as an inhibitory protein that inhibits the binding of TGFB1 to type I receptors but not to type II receptors.[25] TGFB1 activation element (TAE) in the rat BSP gene promoter regulates the expression of the IBSP homolog in rat. Differential expression of PF4 and IBSP genes indicates the possible role of TGFB1 signaling in the etiopathogenesis of OTSC.

Several genetic association studies have shown the association of TGFB1 SNPs with diseases like osteoporosis, asthma, cancers, osteoarthritis, hypertension, and nonsyndromic cleft lip.[16] Mutations in the TGFB1 gene were found to be associated with a rare autosomal-dominant disorder, Camurati-Engleman disease, which arises from increased TGFB1 signaling. Some common polymorphisms in the TGFB1 gene with possible functional significance have been reported. SNPs c.–800G > A and c.–509C > T, situated within the CREB and YY1 consensus binding site, regulates transcriptional activity.[26] The c.–509T allele has been associated with increased TGFB1 at the plasma level. The coding region polymorphisms c.29T > C and c.74G > C located in the signal peptide sequence of the TGFB1 precursor encoding nonsynonymous amino acid substitutions have been associated with elevated TGFB1 serum levels.[26] The c.788C > T polymorphism located within the cleaved part of TGFB1 was found to be significantly associated with OTSC.[16, 27] The activity of the c.788T allele was detected to be higher than the wild-type allele.[16]

Based on previous reports and the functional role of TGFB1 in bone remodeling, the present study was performed to scrutinize the genetic association of TGFB1 polymorphisms with OTSC and to investigate the expression of TGFB1 mRNA in stapes tissue of OTSC cases and controls.

Patients and Methods

Study subjects

A total of 170 unrelated nonsyndromic OTSC cases were recruited from Ear, Nose, and Throat (ENT) units of Capital Hospital, Bhubaneswar, India, and SCB Medical College, Cuttack, Odisha, India. The diagnosis of OTSC was based on family history, otoscopy, pure-tone audiometry, and impedance testing. Pure-tone audiometry was performed in a double-walled soundproof room using standard procedures. The frequencies tested for air conduction were 125, 250, 500, 1000, 2000, 4000, and 8000 Hz and those for bone conduction were 250, 500, 1000, 2000, and 4000 Hz. We also recruited 170 control individuals matching their sex and ethnicity. These individuals were selected at random from the population who had not reported any history of hearing impairment in their family. This study was approved by the Institutional Human Ethics Committee.


Venous blood samples were collected from all the cases and controls after obtaining their written consent. Total genomic DNA was isolated using the rapid nonenzymatic method,[28] and the amount and purity were determined by spectrophotometer. The genotype analysis carried for the SNPs and the methods followed are described in Table 1.

Table 1. TGFB1 SNPs, Their Location, Primer Sequences, Product Sizes, PCR Conditions, and Methods Used for Genotyping of Polymorphic TGFB1 Sites
SNPPositionPrimer sequence (5′–3′)Size (bp)Annealing temperature (°C)Genotyping
  1. TGFB1 = transforming growth factor-beta1; SNP = single nucleotide polymorphism; SSCP = single-strand conformation polymorphism; HRM = high-resolution melting curve.

Single-strand conformation polymorphism (SSCP) analysis was performed by mixing the PCR products with SSCP loading buffer. The mixture was denatured and the snap-cooled products were electrophoresed on a 10% native polyacrylamide gel. The separated strands were visualized by silver staining.

High-resolution melting (HRM) curve analysis was carried out on a Light Cycler480 (Roche Diagnostics, Penzberg, Bavaria, Germany) using a Type-it HRM PCR kit (Qiagen). Some of the sequenced samples were used as positive controls. PCR was performed under the following conditions: an initial denaturation step at 95°C for 10 minutes, then 45 cycles of 95°C for 10 seconds, 60°C for 10 seconds, and 72°C for 15 seconds. After the amplification phase, a melting curve analysis was performed at 95°C for 1 minute, 40°C for 1 minute, and 65°C for 1 second, followed by slow heating at 0.02°C/sec to 95°C with 25 acquisitions/1°C. The HRM curve analysis was performed using the Light Cycler 480 Gene Scanning software v1.2 (Roche Diagnostic, Penzberg, Bavaria, Germany). Melting curves were normalized, temperature was adjusted, and, finally, a difference plot was generated. The genotypes of the subset were defined according to known genotypes of controls. To address the possibility of genotyping error, 25% of the randomly selected subjects were sequenced, which ensured the absence of any genotyping error.

For sequencing, amplified products were cleaned by QIAquick gel extraction kit (Qiagen), according to the manufacturer's instructions. The same primers used for DNA amplification were used for sequencing in both directions (BigDye Terminator v3.1 Cycle Sequencing Kit; Applied Biosystems, Foster City, CA, USA).

TGFB1 expression in the stapedial tissue samples

Tissue sample collection and RNA isolation

Stapes tissue specimens (n = 6) were collected from six OTSC patients undergoing surgery for stapedectomy. Normal stapes tissue samples (n = 4) were obtained from four individual cadavers. None of the control specimens had shown the evidence of OTSC foci. All the samples were preserved in RNAlater (Qiagen) immediately upon surgical removal. Specimens were frozen in liquid nitrogen in RNase-free microcentrifuge tube and homogenized using a tissue-tearer (Biospec Products Inc., Bartlesville, OK, USA) in QIAzol lysis reagent (Qiagen, Maryland, USA). Total RNA was isolated using a RNeasy Lipid Tissue mini kit (Qiagen) following the manufacturer's instructions. All samples were DNase digested to remove any possible copurified DNA with the RNase-free DNase set (Qiagen). RNA quality and quantity were assessed using a Nanophotometer (Implen, GmbH, Munchen, Germany). Total RNA samples were stored in the DNase-free and RNase-free tubes at −80°C until further use.

cDNA synthesis and RT-PCR analysis

cDNA synthesis was performed with oligo (dT) 18 primer using 100 ng of total RNA in a total volume of 40 µL with the first-strand cDNA synthesis kit (Fermentas Inc., Glen Burnie, MD, USA) following the manufacturer's instructions. The target sequence of TGFB1 mRNA was amplified using previously[29] described gene specific primers (5′-AAGGACCTCGGCTGGAAGTG-3′, and 5′-CCCGGGTTATGCTGGT-TGTA-3′) covering the exon-exon boundaries. Amplification was performed with an initial denaturation at 95°C for 5 minutes, followed by 40 cycles of 95°C for 30 seconds, 60°C for 30 seconds, 72°C for 30 seconds, and extended for 72°C for 5 minutes (Eppendorf's Mastercycler, Hamburg, Germany). PCR products were electrophoresed on a 2% agarose gel and visualized upon staining with ethidium bromide. The reverse-transcription PCR analysis revealed the expression of TGFB1 mRNA in stapes tissues of OTSC cases and controls.

Real-time PCR analysis

To quantify the total mRNA expression of TGFB1, fluorescence-based real-time quantitative PCR was performed using QuantiTect SYBR Green RT-PCR Kit (Qiagen) with 4 µL of cDNA and 4 pmol of each primer in a total reaction volume of 20 µL. The amplification was initiated with an initial denaturation step at 95°C for 15 minutes, followed by 40 to 50 cycles of 95°C for 15 seconds, 60°C for 30 seconds, and 72°C for 30 seconds (DNA Engine Opticon2, Bio-Rad, USA). The specificity of the PCR products was verified by melting curve analysis at the end of the PCR. Each sample was assayed twice in duplicate. The comparative ΔCT method (ΔΔCT method) was used to quantity the TGFB1 mRNA level relative to the average expression of the two housekeeping genes (β-actin and 18S rRNA). The data of the two independent analyses for each sample and parameters were averaged and relative expression levels were presented as the relative fold change analyzed using unpaired two-tailed Student's t test.

Statistical analysis

To identify the association of TGFB1 variants with OTSC, genotype and allele frequency were calculated by the allele counting method. A Hardy-Weinberg equilibrium (HWE) calculator was employed to calculate HWE.[30] Chi square and Fisher's exact tests were used for comparison of genotype and allele frequencies between cases and controls. Bonferroni correction for multiple testing of four SNPs (0.05/4 = 0.0125) was used and p <0.0125 was considered statistically significant. To assess the effect of combinations of these SNPs with phenotype, haplotype analysis was performed using SNPalyzeV8.0.2 pro software (Dynacom, Chiba, Japan). For TGFB1 gene expression analysis, unpaired two-tailed Student's t test was carried out using GraphPad (La Jolla, CA, USA) Prism 5.01 software. In silico analysis of the novel TGFB1 promoter variant c.–832G > A was performed using the TFSEARCH (version 1.3) computer program to predict the gain/loss of putative transcription factor binding sites (TFBSs). For this variant, 50 bp of surrounding genomic DNA sequence (25 bp upstream and 25 bp downstream of both the major and minor alleles) were analyzed using the optimized matrix similarity thresholds to predict gain and/or loss of putative TFBSs.


To explore the genetic association of TGFB1 variants with OTSC, we genotyped 170 OTSC cases and 170 controls as described in Table 1. The genotype frequencies of all the SNPs were within HWE expectations in both cases and controls. A cutoff p value of 0.001 was used. The calculated genotype and allele frequencies of TGFB1 variants are displayed in Table 2.

Table 2. Allele and Genotype Frequency Distribution With OR and 95% CI for TGFB1 Polymorphisms in Nonsyndromic Otosclerosis Cases and Controls
SNPGenotype alleleCase (n = 170)aControl (n = 170)aOR (95% CI)pb
  • OR = odds ratio; CI = confidence interval; TGFB1 = transforming growth factor-beta1; SNP = single nucleotide polymorphism; NA = not applicable.
  • aValues are the calculated genotype and allele frequencies of TGFB1.
  • bp < 0.0125 was considered significant; significant values are in bold.
c.–800G>A (rs1800468)GG157 (0.92)155 (0.91)1.169 (0.538–2.538)0.8440
GA13 (0.08)15 (0.09)0.855 (0.394–1.858)0.8440
AA0 (0.00)0 (0.00)NANA
G327 (0.96)325 (0.95)1.161 (0.543–2.479)0.8474
A13 (0.04)15 (0.05)
c.–509C>T (rs1800469)CC54 (0.31)72 (0.43)0.633 (0.406–0.987)0.0561
CT90 (0.53)89 (0.52)1.024 (0.668–1.568)1.0000
TT26 (0.15)9 (0.05)3.431 (1.557–7.559)0.0021
C198 (0.58)233 (0.69)1.562 (1.140–2.139)0.0067
T142 (0.42)107 (0.31)
c.29T>C (rs1800473)TT36 (0.21)53 (0.31)0.593 (0.363–0.968)0.0480
TC108 (0.64)90 (0.53)1.548 (1.003–2.390)0.0614
CC26 (0.15)27 (0.16)0.956 (0.532–1.719)1.0000
T180 (0.53)196 (0.58)0.826 (0.610–1.119)0.2473
C160 (0.47)144 (0.42)
c.74G>C (rs1800471)GG143 (0.84)150 (0.88)0.706 (0.379–1.315)0.3459
GC27 (0.16)17 (0.10)1.699 (0.888–3.250)0.1453
CC0 (0.00)3 (0.02)0.140 (0.007–2.740)0.2478
G313 (0.92)317 (0.93)0.841 (0.472–1.499)0.6598
C27 (0.08)23 (0.07)

Initially, we performed an examination of the coding part of the gene (exon1 and exon5) by direct sequencing and HRM analysis. We found two known variants, c.29T > C and c.74G > C, in exon1 of the TGFB1 gene. The variant c.74G > C was rare, but c.29T > C was more common among the cases as well as in controls. When allele frequencies were compared between the two groups, no significant association was observed for either c.29T > C (p = 0.2473; odds ratio [OR] = 0.826; 95% confidence interval [CI], 0.610–1.119) or c.74G > C (p = 0.6598; OR = 0.841; 95% CI, 0.472–1.499) with OTSC. Comparison of genotype frequencies for c.29T > C assuming a dominant/recessive model showed weak association with OTSC (p = 0.0480; OR = 0.593; 95% CI, 0.363–0.968), but this association did not survive the multiple correction (Bonferroni cutoff p value = 0.05/4 = 0.0125). The SNP c.788C > T in exon5, which has been reported to be polymorphic in other populations, was found to be nonpolymorphic in this population (n = 340).

Screening of the TGFB1 promoter region has identified two known polymorphisms c.–509C > T and c.–800G > A. The genotype assignment of the TGFB1 promoter SNP c.–509C > T was determined by HRM curves, using the sequenced samples as control genotype. For this SNP, 90 of 170 (0.53) cases and 89 of 170 (0.52) controls were heterozygous, whereas 26 of 170 (0.15) cases and 9 of 170 (0.05) controls were homozygous for the risk allele “T” at this position. A highly significant difference in the allele count was obtained for the c.–509C > T variant in the patients and the controls. We confirmed the association between the c.–509C > T variant with OTSC by using chi square and Fisher's exact tests (p = 0.0067; OR = 1.562; 95% CI, 1.140–2.139). The distribution was statistically significant after multiple testing. Association testing with genotype frequency of the c.–509C > T variant of TGFB1 showed consistent association with OTSC, assuming a dominant/recessive model (p = 0.0021; OR = 3.431; 95% CI, 1.557–7.559).

For the variant c.–800G > A, we found that 13 of 170 (0.08) cases and 15 of 170 (0.09) controls were heterozygous and none of the subjects were homozygous “AA” for this SNP. No significant difference was observed when allele and genotype frequencies were compared between cases and controls for the c.–800G > A SNP in TGFB1 gene. SSCP mapping of the c.–800G > A SNP of TGFB1 has identified a novel heterozygote c.–832G > A mutation in an OTSC case. The identified mutation was validated by direct sequencing in both forward and reverse directions. This mutation is novel because it was not found in any of the other DNA samples of 169 cases and 170 controls. Screening of the 1000 Genomes database has not revealed any c.–832G > A mutation in the entire database. We suggest that this mutation is de novo because it was not detected in DNA isolated from the blood samples of patients' parents. In silico analysis for the mutation c.–832G > A predicted the altered binding of two transcription factors v-Myb and MZF1 in the mutated promoter sequence, supporting the alleged effect of this mutation in the etiology of OTSC.

Furthermore, to assess the role of TGFB1 gene with OTSC, the genotypes of all the SNPs for every sample were entered in to the SNPalyze software (Dynacom) for haplotype analysis. The haplotype G-T-T-G was found to be more frequent in cases (0.0874) compared to controls (0.0375). Testing the haplotype G-T-T-G frequencies in cases and controls gave a significant p value of 0.0077 (Table 3). The haplotype G-T-T-G is also the only haplotype that contains the associated c.–509T alleles of the causative SNP, suggesting the role of this variant in the etiology of OTSC.

Table 3. Frequency of Common Haplotypes in Otosclerosis Cases and Controls
HaplotypeaOverall (n = 340)Case (n = 170)Control (n = 170)χ2Normal p valuebPermuted p valueb
  • aHaplotypes with frequency <1% in both groups are not shown.
  • bValues of p are based on 1000 permutations; significant values are in bold.

We next studied TGFB1 mRNA expression in stapes tissue of cases and controls. Expression analysis by reverse-transcription PCR (RT-PCR) revealed the TGFB1 mRNA expression in stapes tissues of cases and controls. Further, comprehensive expression analysis by real-time quantitative PCR revealed that the expression of TGFB1 mRNA was significantly higher (p = 0.0057) in human stapes tissue of OTSC cases (Fig. 1). The reverse-transcription PCR results of two housekeeping genes (β-actin and 18S rRNA) showed no remarkable differences among the tissues from both groups.

Figure 1.

Real-time quantitative PCR analysis of the mRNA level of TGFB1 in four control stapes tissues (black circles) and six otosclerosis stapes tissues (black squares). RNA input was normalized to the average expression of the two housekeeping genes β-actin and 18S rRNA. The data from the two independent analyses for each sample and parameter were averaged and are presented as relative fold change. TGFB1 expression was significantly higher in cases (p = 0.0057). Results are expressed as mean ± SEM.

Further, to see the effect of the c.–509T allele on TGFB1 expression, DNA samples from two patients (with increased TGFB1 expression) were sequenced in both directions. One of the patients was found to be homozygous “TT” and another one was heterozygous “CT” for c.–509C > T.


The identification of candidate genes through genetic association studies is the method of choice for investigating the genetic basis of complex traits, and replication of these findings is essential for establishing the credibility of such results. In this replication study, a comprehensive investigation of TGFB1 polymorphisms was rigorously done to evaluate the positive findings from previous reports.[16, 27] TGFB1 plays an important role in cell growth and differentiation in a wide variety of cell types. TGFB1 secreted by osteoblasts is the most abundant cytokine in bone matrix, with a potent effect on bone metabolism.[22] Abnormal bone growth in OTSC is either a result of increased osteoblast activity or a result of reduced osteoclast activity. It has been reported that TGFB1 increases the mRNA expression of osteoclastogenesis-inhibitory factors in osteoblast cells.[31]

More than 100 polymorphisms have been reported in the TGFB1 gene, some of which are correlated with different races.[32] Several common polymorphisms in the TGFB1 gene with possible functional significance in the regulation of transcriptional activity have been reported. An association of TGFB1 polymorphism c.788C > T has been investigated with OTSC.[16, 27] The functional analysis has shown a higher activity of the causative allele c.788T than the wild-type allele.[16] In the present study, we found that the c.788C > T SNP in the TGFB1 gene was nonpolymorphic in this population (n = 340). Previous studies also suggested that the TGFB1 SNP c.788C > T is monomorphic in South African and Indian populations.[33-35] The absence of an OTSC-causing variant c.788C > T in Indian and South African populations indicates that the incidence of OTSC varies from race to race due to racial differences in allele frequency of the disease-causing variants. Of the two polymorphisms c.29T > C and c.74G > C in exon1, only c.29T > C was weakly associated with OTSC (p = 0.0480). However, this association did not survive the correction for multiple testing. These two coding polymorphisms encode nonsynonymous amino acid substitutions that have been associated with elevated TGFB1 serum levels.[26] Previously, the SNP c.29T > C has shown a trend toward association with OTSC in the French population.[16] However, no evidence of association between c.29T > C and OTSC have been reported in Belgian-Dutch and Tunisian populations.[16, 27]

Further, we hypothesized that the other variants in TGFB1 could be associated with OTSC in this population. We therefore were interested in extending our examination of the TGFB1 gene by screening the promoter variants (c.–509C > T and c.–800G > A) to identify the genetic counterparts for OTSC in this population. The TGFB1 variants c.–509C > T and c.–800G > A have possible functional significance. The c.–509C > T variant influences TGFB1 transcriptional activity in an allele-specific manner and alters DNA protein complex formation.[36] The c.–509C > T polymorphism located within an YY1 consensus binding site and the “T” allele was found to enhance YY1 binding and TGFB1 promoter activity.[37] In this study, we found a strong positive association between the c.–509T allele (p = 0.0067; OR = 1.562; 95% CI, 1.140–2.139) and the “TT” genotype (p = 0.0021; OR = 3.431; 95% CI, 1.557–7.559) of the c.–509C > T variant with OTSC. The relatively high OR in this association study (TT versus CC + CT genotype) suggests that this association may signify a mechanism leading to the development of OTSC. The minor allele “T” frequency was found to be higher in cases (0.42) than controls (0.31), which points to the disease-causing role of this allele “T” in the development of OTSC. The results of association testing survived multiple correction (Bonferroni correction cutoff p value = 0.0125), which suggests that association of TGFB1 variant c.–509C > T with OTSC is real and that TGFB1 might have an important role in the etiopathogenesis of OTSC. The risk allele “T” frequency for the c.–509C > T SNP in the control group (0.31) was found to be similar to the previously examined Belgian-Dutch population (0.32).[16]

It has been reported that the c.–509C > T variant is a genetic determinant of bone mass and the “T” allele is a risk factor for genetic susceptibility to a bone disorder like osteoporosis.[38] The risk allele “T” of c.–509C > T was found to be associated with increased TGFB1 plasma levels and it has been suggested that there may be a correlation between bone diseases and the risk allele “T” at c.–509C > T of TGFB1.[39] The 5′ flanking sequence of the human TGFB1 gene contains two negatively regulatory regions (–1362 to –1132 and –731 to –453) and an enhancer-like element located between –1132 and –731. Both the negative regulatory region and enhancer-like region play an important role in the cell-specific expression of human TGFB1.[40] TGFB1 promoter region variants influence gene expression and potentially contribute to the pathogenesis of TGFB1-related disease. It has been found that the transcriptional activity of the c.–509T allele of the TGFB1 gene is slightly greater than that of the wild-type “C” allele.[41]

Haplotype analysis of the four TGFB1 polymorphisms showed a moderate difference in the distribution of the G-T-T-G haplotype between cases (0.0874) and controls (0.0375). In comparison to the most common haplotype G-C-T-G, the G-T-T-G haplotype was associated with increased risk for OTSC (0.0077). This outcome appears logical if one considers the results of the genotype analysis for the individual c.–509C > T polymorphisms in the TGFB1 gene; the haplotype G-T-T-G is the only haplotype that contains the associated c.–509T variant of the causative SNP, suggesting the role of this variant in the etiology of OTSC. Haplotype analysis represents a much more powerful approach than analyzing only individual polymorphisms because the haplotype also provides information on recombination, which is vital for identification of disease-causing mutation by linkage analysis.[42]

The c.–800G > A variant situated within a consensus CREB half-site and an “A” allele has reduced affinity for the CREB family of transcription factors, which alter the transcription of other members of the TGFB gene family.[39] Genotype analysis for this variant has not revealed any significant association with OTSC. However, SSCP mapping of the promoter polymorphism c.–800G > A detected a novel heterozygous mutation c.–832G > A in an OTSC case. Interestingly, in silico analysis for this variation predicted the altered binding of two transcription factors v-Myb and MZF1 in the mutated promoter sequence. Further, functional analysis is warranted to assess the consequence of the 1-bp substitution (c.–832G > A) on TGFB1 expression because truncation or mutation in the promoter region often leads to impaired protein synthesis.[43, 44]

Many of the complex diseases such as OTSC that manifest their effect exclusively in certain tissues need expression measurements in tissue-type relevant to the disease for the confirmation of statistical observations in genetic association studies. To support our association findings, we demonstrated expression of TGFB1 mRNA in stapes tissues from OTSC cases and controls. In this study, mRNA expression analysis by reverse-transcription PCR and real-time quantitative PCR demonstrated that the TGFB1 gene expression was higher in the stapes tissues of OTSC cases than in controls, suggesting the role of TGFB1 in the abnormal bone turnover in OTSC. Genotyping of TGFB1 polymorphism c.–509C > T in 2 patients (with increased TGFB1 expression) showed the alleged effect of the “T” allele on TGFB1 expression. However, the exact mechanism that controls TGFB1 expression is unclear. Further, functional variants identified in this study, in the regulatory regions, warrant more investigation with a larger number of samples to elucidate their pathophysiological role in OTSC development.


In conclusion, our study suggests that the TGFB1 polymorphism c.–509C > T provides a disease-causing effect for OTSC. Profiling of mRNA in disease and control stapes tissue validated the evidence that TGFB1 might be an important candidate gene for OTSC. The information obtained from this study would be useful to identify the molecular mechanism by which TGFB1 plays a role in OTSC development.


All authors state that they have no conflicts of interest.


This work was financially supported by the Department of Biotechnology, New Delhi, Government of India (Grant Sanction #BT/PR8137/GBD/27/14/2006 dated 25.04.2007). We thank all the subjects who participated in the present study.