SEARCH

SEARCH BY CITATION

Keywords:

  • DHPLC;
  • mutation screening;
  • SLC26A4;
  • Pendred syndrome;
  • DFNB4

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Mutations in SLC26A4 cause Pendred syndrome, an autosomal-recessive disorder characterized by sensorineural deafness and goiter, and DFNB4, a type of autosomal recessive nonsyndromic deafness in which, by definition, affected persons do not have thyromegaly. The clinical diagnosis of these two conditions is difficult, making mutation screening of SLC26A4 a valuable test. Although screening can be accomplished in a variety of ways, all techniques are not equally accurate, timely or cost effective. We found single-strand conformational polymorphism analysis (SSCP) to be 63% effective in detecting mutations a panel of different SLC26A4 allele variants when compared to data from direct sequencing. Because direct sequencing can be time consuming and expensive, especially for a gene with 21 exons, we studied DHPLC as an alternative screening method. We found DHPLC as accurate and reliable as direct sequencing but to be more rapid and cost effective. In addition, we report 11 novel disease-causing allele variants of SLC26A4. © 2003 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Mutations in SLC26A4 (formerly known as PDS) cause Pendred syndrome (PS; MIM no. 274600), an autosomal-recessive disorder characterized by sensorineural deafness and goiter [Everett et al., 1997]. The deafness is congenital and associated with temporal bone abnormalities that range from isolated enlargement of the vestibular aqueduct (dilated vestibular aqueduct, DVA) to Mondini dysplasia, a more complex malformation that also includes cochlear hypoplasia, an anomaly in which the normal cochlear spiral of 2½ turns is replaced by a smaller coil of 1½ turns. Both DVA and Mondini dysplasia are easily recognized by either computed tomography or magnetic resonance imaging [Phelps et al., 1998] (Fig. 1).

thumbnail image

Figure 1. Temporal bone findings (a, right ear; b, left ear) seen with dilation of the vestibular aqueduct (arrow) and dysplasia of the cochlea (chevron). Their occurrence together is known as Mondini dysplasia. For comparison, computed tomography of a normal temporal bone is also shown (right).

Download figure to PowerPoint

The thyromegaly in PS is the result of multi-nodular goitrous changes in the thyroid gland that usually develop in the 2nd decade of life [Reardon et al., 1999]. Most affected persons remain euthyroid, although serum thyroglobulin levels may be elevated and the perchlorate challenge may be abnormal. In this test, individuals are given radiolabeled iodide and its localization to the thyroid is measured. Potassium perchlorate, a competitive inhibitor of iodide transport into the thyroid, is then administered. Normally, the amount of iodide in the thyroid remains stable, reflecting rapid oxidation of iodide to iodine as it is incorporated into thyroglobulin. However, in persons with PS, iodide transport across the thyrocyte is delayed and when perchlorate is administered, the sodium-iodide symporter is blocked and iodide leaks back into the bloodstream. This back leakage is quantifiable as a change in thyroid radioactivity, with a positive result reflecting a drop in radioactivity of greater than 10% [Morgans and Trotter, 1958] (Fig. 2).

thumbnail image

Figure 2. If an SLC26A4 mutation is present, transport into the colloid space is delayed and iodide builds up in the cytoplasm of the thyrocyte. Potassium perchlorate blocks the sodium-iodide symporter and radioactive iodide leaks back into the bloodstream. The fall in radioactivity over the thyroid gland is measured, with a drop >10% consistent with the diagnosis of PS.

Download figure to PowerPoint

In addition to PS, mutations in SLC26A4 cause DFNB4 (MIM no. 600791), a type of autosomal recessive nonsyndromic deafness in which, by definition, affected persons do not have thyromegaly [Li et al., 1998]. No other physical abnormalities co-segregate with the deafness, although abnormal inner ear development, and in particular DVA, can be documented by temporal bone imaging. Together, DFNB4 and PS are estimated to account for 1–8% of congenital deafness.

Functional studies suggest that some of the phenotypic differences between PS and DFNB4 may be due to the degree of residual function of the encoded protein, pendrin. Mutations that abolish all transport function appear to be associated with the PS phenotype, while retained minimal transport ability seems to prevent thyroid dysfunction but not the sensorineural deafness or temporal bone anomalies [Scott et al., 2000]. However, this correlation is not absolute and intrafamilial variability can make the distinction between PS and DFNB4 ambiguous [Phelps et al., 1998; Campbell et al., 2001]. The perchlorate challenge is not a reliable metric to resolve phenotypic differences, as it is not consistently positive in PS [Reardon et al., 1999]. For example, in one family with two affected siblings, one child demonstrated the classic features of PS—severe-to-profound deafness, goiter, and a positive perchlorate discharge test—but the other child had only mild sensorineural deafness and no goiter [Johnsen et al., 1989]. In another study in which six individuals had confirmed PS, only three had a positive perchlorate washout of >10% [Yong et al., 2001].

Since thyroid enlargement is an unreliable clinical indicator of PS and the perchlorate challenge can be ambiguous, several investigators have recommended genetic testing of SLC26A4 to establish the clinical diagnosis of PS or DFNB4 [Reardon et al., 1999; Campbell et al., 2001]. In a large clinical study, Campbell et al. [2001] demonstrated disease-causing allele variants of SLC26A4 in approximately 80% of multiplex families segregating DVA or Mondini dysplasia and in 30% of simplex families. These data suggest that mutations in SLC26A4 are a major genetic cause of these temporal bone abnormalities.

To date, 65 mutations have been reported in a total of 128 families [Pendred syndrome homepage, 2003]. Most of these mutations have been found in single families, however, 15 mutations have been reported in more than one family and four (L236P, IVS8 + 1G > A, E384G, and T416P) account for approximately 60% of the total PS genetic load [Campbell et al., 2001]. The large number of disease-causing allele variants means that mutation screening of SLC26A4 must include an analysis of all 20 protein-encoding exons (2–21) in addition to the splice donor site of exon 1.

A number of mutation detection strategies can be used to screen these exons for nucleotide changes including single-strand conformational polymorphism analysis (SSCP), heteroduplex analysis, cleavase fragment length polymorphism analysis, hybridization analysis, denaturing high-performance liquid chromatography (DHPLC), and direct sequencing [Kristensen et al., 2001]. Of these methods, SSCP and heteroduplex analysis were among the first to be used to detect genetic polymorphisms and remain very popular because of their simplicity. However, the “gold standard” for establishing the identity of unknown nucleotide sequences is direct sequencing.

Because of its size, high throughput mutation screening of SLC26A4 is challenging. The application of some techniques is unnecessarily laborious (SSCP, heteroduplex analysis), expensive (direct sequencing) or insensitive (SSCP, heteroduplex analysis) for mutation screening. DHPLC represents an attractive consideration as an inexpensive option with a sensitivity that approaches that of direct sequencing [Liu et al., 1998; O'Donovan et al., 1998; Ellis et al., 2000].

Mutation detection by DHPLC is based on the rapid separation and visualization of homo- and hetero-duplex DNA molecules using an ion-pair reverse-phase liquid chromatography system [for an excellent review see Xiao and Oefner, 2001]. It can be carried out under complete or partial denaturing conditions, varying column temperature to modulate the degree of denaturation. Accuracy is improved by using an algorithm to predict the optimum temperature at which a simple sequence variant in a given melting domain is detected, and the DHPLC column can then be maintained at a temperature that favors partial strand denaturation in the presence of base pair mismatching. In this study, we report our experience using DHPLC for mutation screening of SLC26A4 in patients with the clinical diagnosis of DVA or Mondini dysplasia.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Subject Recruitment

Persons with DVA or Mondini dysplasia were ascertained clinically based on the presence of hearing loss and temporal bone computed tomography or magnetic resonance imaging. In addition to these studies, their evaluation included a complete history and physical examination. To be classified with DVA, enlargement of the vestibular aqueduct had to be >1.5 mm at a point midway between the endolymphatic sac and the vestibule; to be classified with Mondini dysplasia, the cochlea also had to be abnormal, with incomplete partition and a scala communis. Persons with dominant types of hearing loss were excluded. All procedures were approved by the IRB at the University of Iowa.

SLC26A4 Mutation Panel

A panel of 55 individuals with the clinical diagnosis of PS/DFNB4 segregating 41 sequence-verified allele variants of SLC26A4 was used to compare the sensitivity and specificity of SSCP and DHPLC as mutation screening methodologies.

SLC26A4 Mutation Screening

In consenting persons meeting inclusion criteria DNA was extracted from whole blood and SLC26A4 mutation screening was completed by DHPLC, SSCP, and direct sequencing. Polymerase chain reaction (PCR) amplification and SSCP were performed as previously described, resolving PCR amplification products on MDE gels (BioWhittaker Molecular Applications, Rockland, ME) and using silver staining with visual confirmation to verify amplification and identify band shifts [Campbell et al., 2001].

DHPLC primers differed from SSCP primers and were chosen to optimize mutation detection based on amplicon length and melting profile (Table I). PCR amplification of each exon was performed using 20 η g of genomic DNA in a 50 μl reaction containing 5 μl 10× buffer (160 mM (NH4)2SO4, 670 mM Tris-HCl pH8.8, 0.1% Tween-20), 1.65 μl (50mM MgCl2), 2.0 μl dNTP cocktail (2.5 mM each dATP, dCTP, dTTP, and dGTP), 10 pmol each forward and reverse primer, and 0.50U DNA polymerase (Bioline, Inc., Springfield, NJ). Usual amplification conditions were 94°C for 1 min, followed by three rounds of three cycles each of 94°C for 10 sec, 56°C (55°C, round 2; 54°C, round 3) for 20 sec, and 72°C for 30 sec, and 30 cycles each of 94°C for 10 sec, 54°C for 20 sec, and 72°C for 30 sec, ending with an extension cycle of 72°C for 10 min (for some primer pairs adjustments were made to optimize the reaction). All samples were pooled post-PCR with two parts unknown sample with one part wild-type sample, allowing heteroduplexes to form by denaturing at 95°C for 5 min in a thermal cycler and cooling at a rate of 1°C/min to room temperature.

Table I. SLC26A4 Primers Used for DHPLC Screening
ExonScreening pairPrimer sequenceProduct sizeSequencing primer
11FCCTGACCTCGCAACCCTTGA4031R
 1RCCACCTGATCCGTGACCACTT  
22FTCTTCCCCTCCGATCGTCCT2922F
 2RCCTCCCCAAGGCGTGGAC  
33FTGCAAATTGGTTGTGACTGAG2423F
 3RAACTCCTGCTGGAGACCAGA  
44FGGAACCATTGTAAGTTGAGGAC2434F
 4RGCCAAAACACTTTAAACATGAGC  
55FCCTATGCAGACACATTGAACATTTG3735R
 5RAATTTTGGGTTCCAGGAAAT  
66FAGCTTGATGTAATATTTCCAGAGAG2886F
 6RGGAATGAACAGTGACCCATC  
7, 87–8FGCGTGTAGCAGCAGGAAGTA4837–8R
 7–8RGGAGTATCAGTGAAATGAAGCTTG  
99FGGAAAAAGGATGGTGGTCAA2549F
 9RTTTTCCTGTTTCCAGCCCTA  
1010FGCAGAGTAGGCATGGGAGTT28810F
 10RCCTTCCTCTGTTGCCATTCC  
11, 1211–12FGACACAAGGGAGAAGGACGA48511–12R
 11–12RTCCTCTGGAGTTCCCAAAGCAC  
1313FCACATGATGGTACCTGATACA26413R
 13RAACGAAAGAAAGTGGCTTCA  
1414FCGATTCCACACAAACACCAG38714F
 14RTTCATGACACTCCCTGTGGA  
1515FCCTTGCTAAGTAGCCAGAAATG25415F
 15RTTGGACCCCAGTAAATACTTGT  
1616FCCTTTGAGAAATAGCCTTTCCAG24116F
 16RGCTCTCATCAGGGAAAGGAA  
1717FCCAAGGAACAGTGTGTAGGTC37417F
 17RATTGCCAAAGCTCCAAATGT  
1818FTCCTGAGCAAGTAACTGAATGC19018F
 18RGAAAGGGCTTACGGGAAAGT  
1919FGGCAATAGAATGAGACTCTGT31219F
 19RCTAGACTTGTGTAATGTTTGCC  
2020FCAGTGGAGCATCAGGTGGG24620R
 20RGTTCCCTGACAGTTCTTAATCAG  
2121FCTGGGCAACAGTGAGTGAGA29621F
 21RGCATTGAGGAAGTTTTGTCTTG  

DHPLC analysis of each amplicon was performed at three different temperatures. The amplicons were analyzed using the Wavemaker software package (Wavemaker™ 4.1.34 [Transgenomic™]) to estimate the optimal temperature, run time and acetonitrile gradient. The best predicted temperature was then bracketed by ±2°C to optimize sensitivity and maximize the likelihood that novel mutations would be detected.

Sequencing was completed on an Applied Biosystems (ABI, Foster City, CA) model 3700 automated sequencer. Sequence data were compared to published sequence for SLC26A4 using the Sequencer 4.1 software program package (Gene Codes, Ann Arbor, MI).

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

SLC26A4 Mutation Panel

All 41 deafness-causing allele variants of SLC26A4 presented to the DHPLC could be identified by their elution profile for a detection rate of 100%. Nineteen mutations were detected at all three partial denaturing temperatures, 10 mutations were detected at two of three temperatures, and 12 mutations were detected at only one temperature (Table II). Of the four common mutations, L236P and E384G mutants showed discrete elution profiles at all three temperatures, but the T416P and IVS8 + 1G [RIGHTWARDS ARROW] A mutations were detected at only two and one temperatures, respectively. Mutations were tested multiple times from different samples to confirm test-retest reliability (data not shown). Elution profiles of the four most common SLC26A4 mutants were distinct and could be differentiated from each other easily (Figure 3).

Table II. SLC26A4 Mutation Detection Comparing DHPLC, Sequencing and SSCP
ExonNucleotide changeAmino acid changeDHPLCSEQSSCP
T1T2T3
  • a

    Novel mutations

  • +, Detectable; −, not detectable; +/− occasionally detectable; T1, T2, T3, DHPLC temperatures.

2IVS2-5 +++++
 70C [RIGHTWARDS ARROW] GaR24G+++++
 85G [RIGHTWARDS ARROW] CE29Q+++++
3233A [RIGHTWARDS ARROW] GaY78C+++++
 279delTX96+++
4311C [RIGHTWARDS ARROW] TaA104V+++++
 314A [RIGHTWARDS ARROW] GY105C++++
 317C [RIGHTWARDS ARROW] AA106D+++
 412G [RIGHTWARDS ARROW] TV138F+++
5416G [RIGHTWARDS ARROW] CG139A++++
6626G [RIGHTWARDS ARROW] TG209V++++
 707T [RIGHTWARDS ARROW] CL236P+++++/−
 753-56delCTCTX286+++++
7811G [RIGHTWARDS ARROW] CD271H++++
 783-784insTX286+++++
8IVS8 + 1G [RIGHTWARDS ARROW] A +++/−
 919-2 A [RIGHTWARDS ARROW] G +++
 970A [RIGHTWARDS ARROW] TaN324T+++
91003T [RIGHTWARDS ARROW] CF335L+++++
 1149delCFS383++++
101151A > GE384G++++
 1197delTFS400++++
 1226G [RIGHTWARDS ARROW] AR409H++
 1229C [RIGHTWARDS ARROW] TT410M++
 1246A [RIGHTWARDS ARROW] CT416P++++
 1262A [RIGHTWARDS ARROW] GaQ421R+++++
111264-1G [RIGHTWARDS ARROW] Ca +++
 1284-1286del TGCA429del+++++
 1334T [RIGHTWARDS ARROW] GL445W++++
131440T [RIGHTWARDS ARROW] AV480D++++
 1536-38delAGX524+++++/−
141588T [RIGHTWARDS ARROW] CY530H++++
151667A [RIGHTWARDS ARROW] GY556C+++
 1694G [RIGHTWARDS ARROW] AC565Y++
161790T [RIGHTWARDS ARROW] CL597S+++
171898delAFS634++++
 1958T [RIGHTWARDS ARROW] CV653A+++
 2015G [RIGHTWARDS ARROW] AG672E++
182048T [RIGHTWARDS ARROW] CaF683S++++
192127delTX719+++
 2168A [RIGHTWARDS ARROW] GH723R+++
thumbnail image

Figure 3. DHPLC chromatograms for the most common deafness-causing SLC26A4 allele variants, L236P (a), IVS8 + 1G > C (b), E384G (c), and T416P (d).

Download figure to PowerPoint

SSCP detected 26 (63%) of these same allele variants. Of the missed mutations, five (V138F, G209V, FS400, G672E, H723R) have been reported in more than one family. Three of the most common mutations (L236P, IVS8 + 1G > A, T416P) were detected by SSCP, although the detection for two (L236P, IVS8 + 1G > A) was not consistent (Table II).

Subject Recruitment—Comparative Study of DHPLC, SSCP and Sequencing

A total of 25 unrelated persons with DVA or Mondini dysplasia meeting the study criteria were screened for SLC26A4 allele variants. Sequencing results differed from wild type 18 times, confirming 14 different SLC26A4 allele variants in 11 persons. DHPLC results were 100% concordant, showing abnormal elution profiles each time sequencing data were not wild type. However, abnormal band shifts were clearly visible on SSCP in only 7 of 11 persons carrying SLC26A4 mutations (64%), and of the 14 allele variants, only 6 were detected (43%). Two of the eight mutations missed by SSCP are benign polymorphisms, one of which (2090-60insAAAC) was outside SSCP primer boundaries. Of the disease-causing mutations, L236P and E384G have been reported in several families [Coyle et al., 1998; Van Hauwe et al., 1998; Campbell et al., 2001], G497S has been reported once previously [Li et al., 1998], and M1T, D711X, D724G, and G740S are novel.

Of persons carrying SLC26A4 allele variants, five were compound heterozygotes for disease-causing mutations (9, 19, 20, 23, 25); in the remaining six persons (10, 14, 17, 18, 21, 22) only a single disease-causing mutation was identified (Table III). The inability to identify a second SLC26A4 disease-causing mutation in persons with PS/DFNB4 is not uncommon [Coyle et al., 1998; Campbell et al., 2001]. Whether the ‘missed’ mutation represents a disease-causing intronic or promoter mutation is not known.

Table III. Comparison of SSCP, DHPLC and Sequencing in Mutation Screening of SLC26A4 in 25 Unrelated Persons
No.SSCPDHPLCSequencing
  • a

    Benign polymorphisms.

  • b

    Novel mutations.

  • wt, Wild type.

1wt-all exonswt-all exonswt-all exons
2wt-all exonswt-all exonswt-all exons
3wt-all exonswt-all exonswt-all exons
4wt-all exonswt-all exonswt-all exons
5wt-all exonswt-all exonswt-all exons
6wt-all exonswt-all exonswt-all exons
7wt-all exonswt-all exonswt-all exons
8wt-all exonswt-all exonswt-all exons
9wt-all exonsMut-ex6; mut-ex10L236P; E384G
10wt-all exonsMut-ex6L236P
11wt-all exonswt-all exonswt-all exons
12wt-all exonswt-all exonswt-all exons
13wt-all exonswt-all exonswt-all exons
14wt-all exonsMut-ex5; mut-ex19416-16G [RIGHTWARDS ARROW] Ta; 2090-60 Ins AAACa
15wt-all exonswt-all exonswt-all exons
16wt-all exonswt-all exonswt-all exons
17Mut-ex2Mut-ex2M1Tb
18Mut-ex9Mut-ex9F335L
19Mut-ex3; mut-ex11Mut-ex3; mut-ex11Y78C; A429del
20Mut-ex10Mut-ex10; mut-ex13T416P; G497S
21Mut-ex10Mut-ex10T416P
22Mut-ex2Mut-ex2E29Q
23Mut-ex10Mut-ex10; mut-ex19T416P; D724Gb
24wt-all exonswt-all exonswt-all exons
25wt-all exonsMut-ex192090-60 Ins AAACa; D711Xb; G740Sb

Novel SLC26A4 Mutations

Eleven novel mutations were identified (Tables II, III), bringing to 76 the number of SLC26A4 disease-causing mutations now known [Pendred syndrome homepage, 2003]. All novel mutations were confirmed by bi-directional sequencing of the relevant exons using a minimum of two different PCR amplifications; none was present in a screen of 100 nonaffected controls.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

Because the diagnosis of PS and DFNB4 is difficult, mutation screening of SLC26A4 is a valuable clinical test. Data suggest that mutations in this gene account for 1–8% of congenital deafness [Fraser et al., 1960; Fraser, 1965], making it the second leading cause of congenital deafness after GJB2. Although four mutations account for 60% of SLC26A4 disease-causing allele variants, mutations have been reported in 18 of its 21 exons [Campbell et al., 2001]. This distribution mandates mutation screening of the entire gene.

Screening can be accomplished using a variety of techniques, although all are not equally accurate, timely or cost effective [Kristensen et al., 2001]. One of the more popular methods is SSCP. Mutations are detected by migration differences in single stranded DNA, with confirmatory sequencing as required. Although inexpensive, we found SSCP to be only 63% effective in detecting mutations in a panel of different SLC26A4 allele variants. At the other extreme is direct sequencing. While this approach represents the gold standard against which other mutation screening strategies should be compared, it is time consuming and expensive, especially for a gene with 21 exons. DHPLC offers a potentially attractive alternative to SSCP and direct sequencing as it is reported to be accurate, reliable, timely, and cost effective (Table IV) [Liu et al., 1998; O'Donovan et al., 1998; Ellis et al., 2000].

Table IV. Post-PCR Cost-Based Comparison
DescriptionSSCPDHPLCSEQ
  • a

    Based on ∼10,000 injections.

  • b

    Each sample analyzed at three temperatures ($1.30 × 3).

Size limits (bp)∼200∼1,000∼800
Mutation identificationNoNoYes
Turnaround time (hr)/sample150.324
Post PCR time (min)/sample415
Sensitivity (%)∼63∼100100
Cost/sample ($)0.13.90a,b11.0

We found DHPLC sensitivity for detecting SLC26A4 allele variants to be 100%. This result is similar to that reported by Taliani et al. [2001], who used DHPLC to screen PROC, a gene that encodes protein C. More than 200 different mutations in this gene are associated with an increased susceptibility to venous thromboembolism [Taliani et al., 2001]. Other authors have reported the sensitivity of DHPLC to range from 95% to 100% [Liu et al., 1998; O'Donovan et al., 1998; Ellis et al., 2000].

To ensure this level of accuracy and reliability, and to optimize cost effectiveness and turnaround time, a number of parameters must be considered. Amplicon length should be chosen after reviewing the predicted melt profile. If the GC content across the amplicon is consistent, the melt profile will be flat and >1,000 bp products can be used; however, it is more common to see a 300–400 base pair stretch of 50% GC-rich DNA bounded by regions of much higher or lower GC content. This variability affects the melt profile, making detection of some allele variants difficult.

Wavemaker software will predict the optimal acetonitrile gradient and partial denaturing temperature but additional testing should be performed 2°C above and below the predicted temperature. Although we use a standard acetonitrile gradient, we extend clean up and equilibration durations to remove impurities and increase column life. We also maintain the column by hot washing and analyzing DHPLC standards after every 200 injections. The ability of the column to detect these standards correlates directly with its ability to detect sequence variants in SLC26A4 or other genes of interest.

Other investigators have been able to correlate DHPLC chromatogram profile with mutation type [Taliani et al., 2001] but we found these correlations difficult to make. In our experience, wave profile for a given SLC26A4 allele variant differs from column to column and even in the same column based on column life and buffer constitution. However, we could distinguish all heterozygote mutations from wild type samples. DHPLC does not distinguish homozygote allele variants due to limitations inherent in heteroduplex analysis. We recommend pooling unknown samples with sequence-verified wild type DNA at a ratio of 2:1 as DHPLC has been shown consistently to detect single nucleotide polymorphisms at 5% frequency (1/20 alleles) [Wolford et al., 2000].

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

By coupling additional automated instrumentation with DHPLC, high throughput mutation screening of SLC26A4 is possible in persons with a PS/DNFB4 phenotype. Our data indicate that DHPLC-based screening is accurate, reliable, robust and cost effective.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES

We thank the families who participated in this study. This study was supported in part by RO1-DC02842 (RJHS) and The National Lottery (UK) (RCT).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. REFERENCES