UNIVERSITY OF CINCINNATI
Date:______
I, ______, hereby submit this work as part of the requirements for the degree of: in:
It is entitled:
This work and its defense approved by:
Chair: ______
A Family-based Mapping Study of Autosomal Dominant Nonsyndromic
Sensorineural Hearing Loss
A thesis submitted to the
Division of Research and Advanced Studies
of the University of Cincinnati
In fulfillment of the requirements for the degree of
Master of Science in Medical Genetics
Genetic Counseling Program
in the Department of Analytical and Diagnostic Sciences
of the College of Allied Health Sciences
May 18th, 2007
By:
Monica A. Giovanni
Committee Chair: John H. Greinwald, Jr., MD Abstract
Autosomal dominant nonsyndromic hearing loss (ADNSHL) is characterized by postlingual, progressive hearing impairment. This study sought to identify the gene responsible for hereditary nonsyndromic sensorineural hearing loss in a family with multiple generations affected by hearing impairment presenting in the second decade of life. The Affymetrix GeneChip® was used to identify three linkage intervals on chromosomes 4, 10, and 16. The observed hearing loss in this family is not likely due to previously identified deafness-causing genes as no such genes have been reported in the identified intervals. Since preliminary candidate gene sequencing within the regions did not identify any pathogenic mutations, haplotype mapping was employed to further refine the intervals. The intervals on chromosomes 4 and 16 were excluded and the interval on chromosome 10 was narrowed to a 0.4Mb region at 10q22-q23. Future work will employ candidate gene analysis to identify the gene responsible for this family’s hearing impairment.
Keywords: family-based linkage analysis, haplotype mapping, PCR based sequence analysis, autosomal dominant nonsyndromic hearing loss (ADNSHL)
iii iv Acknowledgements
I would like to thank the University of Cincinnati and Cincinnati Children’s hospital
Medical Center Genetic Counseling program for supporting this research. I would also like to extend my sincerest thanks to Dr. John Greinwald for serving as my research mentor and advisor; his expertise, guidance, and instruction was an integral part of my learning experience and the completion of this project. Finally, I would like to thank my thesis committee members Dr. Bill Nichols and Judy Johnson for their time and assistance throughout this process.
v Table of Contents:
List of Tables and Figures…………………………………………………………..……vi
Introduction………………………………………………………………………………..1
Methods……………………………………………………………………………………3
Results…………………………………………………………………….……………….6
Discussion………………………………………………………………………………..10
Bibliography……………………………………………………………………………..19
Figures……………………………………………………………………………………24
Tables…...………………………………………………………………………………..34
Appendices……………………………………………………………………………….35
vi Tables, Figures, and Appendices
Figure 1. Nonconsanguineous family with apparent autosomal dominant hearing loss across four generations
Figure 2. Affymetrix GeneChip® results show linkage interval on Chromosome 4
Figure 3. Ideogram of Chromosome 4 with linkage interval
Figure 4. Affymetrix GeneChip® results show linkage interval on Chromosome 10
Figure 5. Ideogram of Chromosome 10 with linkage interval
Figure 6. Affymetrix GeneChip® results show linkage interval on Chromosome 16
Figure 7. Ideogram of Chromosome 16 with linkage interval
Figure 8. A smaller branch of the family, branch 2, served as the index branch due to the
presence of multiple affected and unaffected individuals and the appearance of autosomal
dominant transmission of hearing loss
Figure 9. Haplotype analysis using chromosome 4 markers excluded this interval due to marker segregation inconsistent with affectation in branch 2 of the family
Figure 10. Haplotype analysis using chromosome 16 markers showed plausible marker segregation creating a very small interval within which the disease-causing gene could be positioned. This segregation pattern can be observed across affected individuals in branch
2 of the family (A), however the proposed genotype did not segregate with hearing loss
when expanded to include branch 4 of the family (B)
Figure 11. Haplotype analysis using chromosome 10 markers showed plausible marker segregation across affected and unaffected individuals in branch 2 of the family
Table 1. Subject Inclusion Criteria
vii Table 2. Subject Exclusion Criteria
Appendix A. Polymorphic markers used to refine the candidate intervals on chromosomes 4, 10, and 16
Appendix B. Candidate genes in the interval of interest on chromosome 10
viii Introduction
Hearing impairment is the most commonly diagnosed sensory deficit. It is estimated that 1 in 1000 children is born deaf or will become profoundly deaf before speech is acquired. Another 1 in 1000 children will become deaf or develop a significant hearing impairment before adulthood (Morton 1991). Roughly 50% of all cases of hearing loss can be attributed to genetic causes (Marazita et al., 1993). Seventy percent of cases of hearing loss with a genetic etiology appear in isolation and are considered nonsyndromic (Sundstrom et al., 1999). The most common mode of inheritance for all hereditary hearing impairment is autosomal recessive, accounting for an estimated 80% of all cases. Autosomal dominant inheritance accounts for roughly 20% of cases and the remaining cases (less than 1%) show X-linked or mitochondrial inheritance (Petersen &
Willems, 2006). Research into nonsyndromic hearing impairment has established that there are at least 50-100 nonsyndromic deafness-causing genes (Zbar et al., 1998).
Autosomal dominant nonsyndromic hearing loss (ADNSHL) exhibits a wide range in age of onset with some individuals presenting with congenital or early onset hearing impairment, while other individuals develop hearing impairment in their forties and fifties which can be misdiagnosed as presbycusis, or age-related hearing loss. In general, ADNSHL is characterized by hearing impairment beginning in the second or third decade of life after the development of normal speech, or postlingually, and is progressive in nature. ADNSHL genes have been localized using extended family pedigrees in which a single deafness-causing gene segregates. ADNSHL genetic loci are termed DFNA (DFN = deafness, A = dominant), followed by a number. The first gene for ADNSHL, DFNA1, was localized in 1992. Since that time, close to 50 loci have been
1 mapped and 20 genes identified for ADNSHL (Yan et al., 2006; Hereditary Hearing Loss
Homepage: http://webh01.ua.ac.be/hhh/). These genes have been found to encode a wide
variety of proteins, including transcription factors, components of the extracellular matrix
and cytoskeleton, ion channels, and proteins of unknown functions (Sundstrom et al.,
1999).
Deafness research has focused on genetic mapping studies in an effort to elucidate the genes responsible for hearing loss. The mapping of recessive genes is often plagued with challenges including finding families large enough for linkage analyses and ascertaining families with some degree of consanguinity (Sundstrom et al., 1999,
Strachan & Read, 2004). Mapping autosomal dominant genes can be less complicated as individuals affected by autosomal dominant conditions are usually seen with a higher frequency in families as compared to recessive conditions, making it easier to ascertain multiple affected individuals in a bloodline (Sundstrom et al., 1999). By definition, a dominant genetic mutation should manifest in an individual heterozygous at that genetic locus (Strachan & Read, 2004). Dominant family studies are often complicated by nonpenetrance, in which some individuals heterozygous for a gene mutation lack symptoms of the disease. Thus, it is possible for an autosomal dominant condition to appear to skip generations such that an individual can have both an affected parent and an affected child without showing any symptoms of the condition themselves (Strachan &
Read, 2004). In ascertaining large families for mapping studies, some degree of nonpenetrance can be expected in a multigenerational pedigree.
This study sought to identify the genetic cause of nonsyndromic sensorineural hearing loss that exhibits apparent autosomal dominant inheritance in a large
2 nonconsanguineous family. In this investigation, linkage studies were performed using
DNA extracted from blood samples of affected and unaffected members of a family with
ADNSHL to establish candidate intervals within which the disease-causing gene was likely located. After we identified three intervals in which linkage appeared likely, we selected three functional candidate genes within the intervals. PCR based sequencing did not identify any disease causing mutations in these three candidate genes. Haplotype mapping techniques were then employed with the aim of refining the intervals of interest for more targeted candidate gene analysis to identify the deafness-causing gene in this family.
Methods
Study Subjects
This study enrolled members of a large, nonconsanguineous family of northern
European descent with no recognized Jewish heritage. Multiple generations in this family experienced progressive hearing impairment with onset in the second decade of life (see Figure 1). The subjects were enrolled at Cincinnati Children’s Hospital Medical
Center based on their family history, age, and hearing status. Institutional review board approval was obtained from Cincinnati Children’s Hospital Medical Center for research involving human participants. Informed consent was obtained from all participants.
Parental consent and participant assent (when appropriate) was obtained for individuals under the age of 18.
Affected family members exhibited hearing loss (as determined by standard air and bone conduction audiometry) with estimated average age of onset in the mid-second
3 decade of life, between the ages of 11 and 20 years. Affected individuals did not have
any diagnosed vestibular dysfunction, associated abnormalities, or history of rubella,
drug use during pregnancy, brain trauma, or meningitis (See Tables 1, 2). Unaffected
family members over the age of 20 were enrolled as control participants. Individuals
under the age of 20 could not definitively be labeled unaffected, as the phenotype could
still be expressed at a later time. Blood samples were obtained from a total of 34
consenting family members spread across three generations: 13 affected individuals and
21 unaffected individuals.
Linkage Analysis
A genomewide linkage screen of the family was undertaken using the GeneChip®
Mapping 10K Xba Array, a high-density single nucleotide polymorphism (SNP) array
(Affymetrix, Inc., Santa Clara, CA). The 10K SNP array provides comprehensive
coverage of the genome with 11,555 biallelic polymorphisms distributed throughout the
genome with the exception of the Y chromosome. The median distance between SNPs is
104kb with a mean genetic distance of 0.31cm (Matsuzaki et al., 2004).
Genomic DNA was isolated from peripheral blood of both affected and
unaffected individuals (n=34) according to Affymetrix protocol (GeneChip Mapping
Assay manual). Genomic DNA was digested with XbaI, ligated to XbaI adaptor, and amplified by PCR. The PCR products were fragmented with DNase I. Fragmentation was confirmed by gel electrophoresis. The fragmented PCR products were end-labeled with biotin and hybridized to the array. After hybridization, the array was washed, stained, and scanned. Genotype assignments were made automatically by GeneChip®
4 DNA Analysis software (Affymetrix, Inc., Santa Clara, CA) (Kennedy et al., 2003, Liu et al., 2003).
Non-parametric Linkage (NPL) analysis was completed. The Affymetrix 10K
SNP data was analyzed using GeneSpring GT software from Agilent Technologies. The
Caucasian population allele frequencies used in the NPL analysis were downloaded from
GeneSpring GT Extras in the Agilent website. Non-parametric linkage scores were calculated with the Spairs statistic (error rate 0.005). Current human genome maps, both genetic and physical, were searched to identify additional polymorphic or single nucleotide polymorphic (SNP) markers to confirm identified areas of linkage
(Unpublished work by Greinwald et al.).
Candidate Gene Analysis
Genes located in the candidate intervals were identified using National Center for
Biotechnology Information (NCBI) maps. Candidate genes were selected based on previous studies that documented cochlea-related functions. These genes included protocadherin 7 (PCDH7), potassium large conductance calcium-activated channel, subfamily M, alpha member 1 (KCNMA1), and annexin A11 (ANXA11).
Previously extracted DNA was diluted to 30ng/μl for PCR amplification (Guo et al., 2004). Primers were designed for amplification of coding and splice site regions of the candidate genes. Sequencing of the candidate genes, including all exons, was undertaken using designed primers. Amplified PCR products were purified and bidirectionally sequenced by the Cincinnati Children’s Hospital Research Foundation
DNA Core facility. Sequences were aligned and compared with sequences obtained from
5 the human genome database (http://www.ncbi.nlm.nih.gov) using the software package
Sequencher (Gene Codes, Corp., Ann Arbor MI). Sequences from affected and unaffected individuals were also compared.
Haplotype Mapping
Haplotype mapping was employed using primers flanking short tandem repeat polymorphic (STRP) markers localized to the identified regions of interest on chromosomes 4, 10, and 16 (see Appendix A). DNA samples were amplified by PCR using the specific STRP primers. Genotyping was performed by sizing fluorescently labeled PCR products using ABI PRISM automated genotyping system (Applied
Biosystems, Foster City, CA). Individual haplotypes were assigned using Cyrillic 2.1
(Family Genetix, Oxford, UK) to define the boundaries of the co-segregating regions.
Haplotypes for affected and unaffected individuals were then compared.
Results
Linkage Analysis
Available individuals from the family (n=34) were subject to whole-genome SNP linkage analysis (see Figure 1). Autosomal dominant inheritance was supported by the pedigree analysis which presented multiple individuals affected with hearing impairment across four generations in the same bloodline. Approximately equal numbers of males and females appear to be affected and male-to-male transmission was observed.
Linkage analysis identified three intervals of interest on chromosomes 4, 10, and
16 yielding NPL scores > 3, with an NPL p-value < 0.01. The candidate interval on
6 chromosome 4 spans a physical distance of 11.1Mb at 4p14-p15 (see Figures 2 and 3); the candidate interval on chromosome 10 spans a distance of 21.6Mb at 10q22-q23 (see
Figures 4 and 5) and the candidate interval on chromosome 16 spans a region of 3.4Mb at
16p13 (see Figures 6 and 7).
Candidate Gene Analysis
PCR based sequencing of the three candidate genes in these intervals, protocadherin 7 (PCDH7) on chromosome 4, potassium large conductance calcium- activated channel, subfamily M, alpha member 1 (KCNMA1) on chromosome 10, and annexin A11 (ANXA11) on chromosome 10, failed to identify a putative mutation.
Therefore these three genes were eliminated as the disease-causing gene in this family.
Haplotype Mapping
Haplotype analysis was initially performed on an index branch of the family
(branch 2) with multiple affected and unaffected individuals and the appearance of autosomal dominant transmission of hearing loss (see Figure 8). We were able to exclude the interval on chromosome 4 as the markers did not segregate with hearing loss
(see Figure 9). However, plausible segregation was observed for the intervals on chromosome 10 and 16, which lead to expansion of the haplotype analysis to include the entire family.
Haplotype mapping refined the candidate interval on chromosome 16 to a 0.4Mb region which segregated with affected individuals in the index branch (see Figure 10a).
The interval’s boundaries were established by STRPs, D16S3392 as the inclusive
7 telomeric boundary and D16S423 as the inclusive centromeric boundary. Individual
IV:19 sets the centromeric boundary due to recombination at 5.9Mb marked by D16S423.
Individual IV:21 sets the telomeric boundary due to recombination at 6.3Mb marked by
D16S3392.
When haplotype analysis of the region of chromosome 16 was expanded to branch 4 of the family, the expanded marker set did not segregate consistently with affected individuals and many unaffected individuals carried the proposed disease- causing genotype (see Figure 10b). While individual II:5, the patriarch of branch 4, carries the proposed disease-causing chromosome and was affected with hearing loss, individuals III:12, III:16, and III:18 carry this chromosome, but were unaffected.
Additionally, individual III:10 is affected with hearing loss, but does not carry the disease-causing chromosome. The linkage interval on chromosome 16 was thus ruled out due to this lack of marker segregation with the disease phenotype.
Preliminary data from the expanded haplotype analysis of chromosome 10 showed plausible segregation. Recombination events defined the proposed disease gene interval of 0.4Mb. The interval on chromosome 10 segregated well with affected individuals in the index branch (see Figure 11) and is defined by known affected individuals. The centromeric boundary of the co-segregating interval is defined by recombination in individual IV:20 at 85.7Mb marked by D10S1658. The telomeric boundary is defined by recombination in individual IV:18 at 86Mb marked by
D10S1774. Haplotype assignment at marker D10S1717 is common to all affected individuals in this branch.
8 Discussion
Microarray-based single nucleotide polymorphism (SNP) genotyping offers an
efficient method for performing genomewide linkage analysis as compared with the
previously used microsatellite markers. SNPs are abundant markers that are evenly
distributed throughout the genome at an estimated rate of 1 in 1,000 base pairs (Hu et al.,
2005). Microsatellites have a higher degree of heterozygosity than do SNPs, but SNPs offer adaptability to high-throughput genotyping, making them a logical and convenient choice for linkage analysis (Matise et al., 2003). The Affymetrix GeneChip® 10K Array provides a platform for high-throughput SNP genotyping for linkage analysis. This type of platform requires relatively small quantities of DNA and offers improved genetic
resolution as compared with conventional microsatellite markers.
While the Affymetrix GeneChip® does utilize 11,555 SNP markers across the
entire genome with a median inter-marker distance of 104KB, markers are not uniformly placed throughout the genome and therefore some regions are underrepresented.
Specifically, underrepresentation of informative SNP markers is noted in regions on chromosomes 16p, 17q, 19p, and 22q (Sellick et al., 2004). These underrepresented regions are in extreme telomeric and pericentromeric regions of the chromosomes.
In this study, linkage analysis identified three regions of interest on chromosomes
4, 10, and 16 allowing for candidate gene and haplotype analysis. There was, however, an observed lack of hybridization on the telomeric end of the interval of 16p (see Figure
6). In order to establish the telomeric boundary of the linkage interval, haplotype mapping was used to define the boundary with STRPs. The use of SNPs in linkage analysis is efficient and useful, but this technology is not a stand-alone method for
9 family-based genetic mapping; haplotype mapping is still necessary to refine the linkage
interval and identify location of crossovers.
Based upon the results of the linkage analysis, we conclude that the observed hearing loss in this family is not due to previously identified deafness-causing genes as
no such genes have been reported in the intervals identified in this study.
Candidate genes were selected in the three identified intervals based on previous studies that documented cochlea-related functions. PCR based sequencing was performed on three candidate genes: Protocadherin 7 (PCDH7), potassium large conductance calcium-activated channel, subfamily M, alpha member 1 (KCNMA1), and annexin A11 (ANXA11).
Protocadherin 7 (PCDH7) is located on chromosome 4. Members of the cadherin family of genes are known to encode integral membrane proteins that mediate calcium- dependent cell-cell adhesion, communication, and signal transduction (Entrez Gene,
http://www.ncbi.nlm.nih.gov). PCDH genes located on other chromosomes have been
established as a cause of autosomal recessive hearing loss. Mutations in PCDH15 have
been identified as a cause of nonsyndromic hearing loss (DFNB23) and Usher syndrome type 1F, an autosomal recessive condition characterized by profound congenital sensorineural hearing loss and retinitis pigmentosa (Ouyang et al., 2005; Alagramam et al., 2005).
Potassium large conductance calcium-activated channel, subfamily M, alpha member 1 (KCNMA1) is located on chromosome 10. KCNMA1 is a large conductance
calcium-activated potassium channel that regulates electrical current. Voltage-activated
potassium channels are important for shaping the receptor potentials of cochlear hair cells
10 (Langer et al., 2003). Mutations in KCNQ4 have been established as a cause of
autosomal dominant nonsyndromic sensorineural hearing loss (DFNA2) (Su et al., 2007).
Mutations in KCNQ1 and KCNE1 have also been identified as the cause of Jervell and
Lange-Nielsen syndrome, an autosomal recessive condition characterized by long QT syndrome and hearing loss (Zehelein et al., 2006).
Annexin A11 (ANXA11) is also located on chromosome 10 in the interval of interest. ANXA11 is a member of the annexin family of calcium-dependent phospholipid- binding proteins which are reported to have a variety of functions including involvement in cell division and proposed calcium- and cell cycle-dependent association with the nuclear envelope (Entrez Gene, http://www.ncbi.nlm.nih.gov).
In this study, preliminary candidate gene analysis was unsuccessful; analysis of
coding and intron-splice sequences did not identify a putative mutation. As a result,
haplotype analysis was employed. Haplotype analysis greatly narrowed the focus and refined the interval within which the disease-causing gene is proposed to be located. The
linkage interval on chromosome 4 was ruled out as no plausible segregation was observed across any of the family branches, leaving the intervals on chromosome 10 and
16.
Haplotype analysis using chromosome 16 microsatellites demonstrated plausible
marker segregation among affected individuals and obligate carriers in branches 1 and 2
of the family; however, branches 3 and 4 did not show marker segregation that was
consistent with hearing loss (data not shown). In branch 3 of the family, six markers
were unable to be assigned as the offspring did not appear to inherit alleles from the
father, raising the question of sample mix-up in the lab or nonpaternity (data not shown).
11 Branch 4 of the family included multiple unaffected individuals who carried the proposed disease-causing chromosome and multiple affected individuals who did not. Based upon
the refinement of the interval and expansion to the entire family, we conclude that the
genetic cause of hearing impairment in this family is unlikely to be located at 16p13.
The interval on chromosome 10 showed segregation consistent with hearing loss
in branch 2 of the family (see Figure 11). This proposed interval defined by branch 2 is
0.4Mb in size and contains 8 genes, one open reading frame, and an uncharacterized
region (see Appendix B).
Future Work
Haplotype analysis will continue on the interval on chromosome 10 with the
remaining branches of the family to confirm marker segregation consistent with hearing
loss. Once the interval is established across all branches of the family, the next step in
this analysis will be a more exhaustive candidate gene approach. Further mutational
screening within the critical region will focus on possible disease-causing genes on
chromosome 10 (see Appendix B). Candidate genes will be selected based upon cochlear
expression and suspected gene function. Two potential functional candidate genes are
protocadherin 21 (PCDH21) and retinal G protein coupled receptor (RGR).
Protocadherin 21 (PCDH21) is a member of the same gene family as PCDH7, the
candidate gene that was sequenced as a potential functional candidate on chromosome 4.
The encoded protein of PCDH21 has a signal peptide, six cadherin repeat domains and a
unique cytoplasmic region (Entrez Gene, http://www.ncbi.nlm.nih.gov). PCDH21 is a non-classical cadherin, as it appears to be expressed only in the mitral and tufted cells in
12 the main and accessory olfactory bulbs of the brain. It is hypothesized that PCDH21 may
have a role in the formation and maintenance of neuronal networks (Nagai et al., 2005).
Retinal G protein coupled receptor (RGR) is a member of the G-protein coupled
receptor 1 family and is known to be expressed in the eye, brain, muscle, and connective
tissue (Entrez Gene, http://www.ncbi.nlm.nih.gov). Genes with cochlear expression are often also expressed in the brain, leading us to believe that RGR may be expressed in the cochlea in addition to the documented expression pattern. In the eye, the protein acts as a photoisomerase to catalyze the conversion of all-trans-retinal to 11-cis-retinal. The encoded protein of RGR is known to be expressed in tissue adjacent to retinal photoreceptor cells, the retinal pigment epithelium and Mueller cells. This gene may be associated with autosomal recessive and autosomal dominant retinitis pigmentosa (Yang
& Fong, 2002). Alternative splicing results in multiple transcript variants encoding different isoforms (Jiang et al., 1995).
Many methods exist for mutational analysis and typically involve screening for genetic changes within or near target genes. Sequencing is a technically easy and efficient method for mutation detection. When sequencing genomic DNA, each exon is typically amplified separately with approximately 20bp of flanking intron (Strachan &
Read, 2004). PCR based sequencing of PCDH21 and RGR will be performed.
Additional sequencing approaches are available if initial sequencing does not identify a mutation. It is possible that mutations could exist in the untranslated region
(UTR) or the promoter region of the gene which are not typically screened in initial sequencing analysis. Sequencing could also miss a large-scale deletion as heterozygous
13 deletions are not detectable by sequencing because the mutant allele gives no PCR product (Strachan & Read, 2004).
Reverse transcriptase polymerase chain reaction (RT-PCR) of isolated RNA can reliably detect aberrant splicing or exon skipping which is otherwise difficult to predict from a DNA sequence change. RT-PCR may improve detection of exon-scale deletions; this method could also be used to identify genetic changes with the exception of a heterozygous deletion of the entire gene. The limitation of RT-PCR is the sensitivity of
RNA; RNA must be carefully handled and processed quickly after acquisition to prevent the degradation of mRNA. This limitation often prevents the use of RT-PCR in large studies with participants from many different locations (Strachan & Read, 2004).
Quantitative PCR is another method for the detection of heterozygous deletions and measures the amount of PCR product accumulated in real time using fluorescent labeling. This method can be used to explore questions of gene dosage (Strachan &
Read, 2004). Multiplex Ligation-dependent Probe Amplification (MLPA) employs a multiplex PCR reaction to detect gene duplications and deletions. MLPA can be used to establish the copy number of up to 45 nucleic acid sequences in one single reaction
(Strachan & Read, 2004).
Quantitative Southern blot analysis can be employed for the detection of complete or partial deletions and more complex rearrangements of the gene by digesting DNA fragments and separating resulting fragments by size. Southern blotting is technically demanding and requires greater quantities of DNA than does PCR, but can detect recombination events, insertions, and deletions (Strachan & Read, 2004).
14 If completed sequencing of all candidate genes (see Appendix B) in this interval
does not identify a mutation, we would consider a different interpretation of the proposed disease-causing interval before employing alternative methods of mutation analysis. We
could consider accepting individual IV:20 as a nonpenetrant individual, rather than
unaffected, as he carries the centromeric portion of the proposed disease-causing
chromosome that is shared with the three affected siblings, individuals IV:15, IV:17, and
IV:18 (see Figure 11). With the acceptance of this interval, we would then obtain new markers centromeric to D10S195 in an effort to obtain a centromeric boundary for the proposed interval.
Ethical Considerations
Advancements in technological ability have lead to increasing genetic research and clinical genetic testing. With the increase in genetic research studies, researchers are confronted with many ethical obligations to the participants, especially in family studies which present challenges to the traditional approach of protecting human subjects in research (Arar et al., 2005). The Belmont Report sets out the basic ethical principles of human research as (1) respect for persons including individual autonomy and increased protection for those with diminished capacities, (2) beneficence with the obligation to do no harm and maximize benefits while minimizing potential harms, and (3) justice with fairness in distribution (The Belmont Report, 1978). Institutional review boards require informed consent for the participation of human subjects in research; however, this approach arguably does not fully protect the rights of individuals participating in family
15 studies. Since participants share a genetic heritage with other participants, information
gained from the research may affect the entire family.
It is reported that participants in family studies often underestimate the social and
cultural risks associated with participation in a research study (Arar et al., 2005). Many
participants do not recognize any risks associated with disclosing information and
willingly volunteer information about relatives’ physical and psychiatric conditions. It
cannot be assumed, however, that all family members have the same degree of
knowledge about familial issues and conditions; individual disclosures may alter other
relatives’ awareness of relatedness, disease, or family dynamics. Participants’
underestimation of familial and group risks raises questions about an individual’s ability
to provide true informed consent.
It is the responsibility of the researchers to protect the rights and welfare of study
participants by recognizing ethical, social, and cultural issues involved such as familial
issues, risk of stigmatization, and the privacy and confidentiality of health information,
among others. In this study, we ensured the rights of individual participants by
discussing the risks and possible benefits of individual participation as well as possible
disclosure of study findings. Efforts were made to provide information regarding the
research study to any interested family member so that they may consider participating.
Researchers discussed the study with groups of family members who were interested in
participating; researchers also talked to relatives at family gatherings to allow individuals
to learn more about the research before consenting to participate. Informed consent was obtained for each participant and additional family members were contacted by already enrolled relatives, not researchers.
16 Communicating research risks, benefits, and ethical considerations are important issues which provide challenges to the research process. In an effort to fully explore these concerns with participants, researchers can utilize the services of research nurses
and genetic counselors as they are trained to perform these vital services. Within genetic
research studies, research nurses and genetic counselors can promote participant
understanding of the research implications to self and other family members, and thereby
obtain a true informed consent In addition, they can serve as an interface between the
researchers and the families for the sharing of information should results be found
(Markel & Yashar, 2004).
Conclusions
Using both linkage and haplotype analysis, we have mapped a novel locus for
autosomal dominant hearing loss to chromosome 10q22-q23 in a 0.4Mb interval between
D10S1658 and D10S1774. This study has utilized advancing technology to localize a candidate gene region responsible for hereditary nonsyndromic hearing loss. Our results
are strengthened by the size of the family, number of affected individuals, and the
willingness of family members to participate. This study demonstrates a significant
refinement of the linkage interval and should facilitate the identification of the genetic
cause of hearing loss in this large family using candidate gene analysis.
17 Bibliography
Alagramam K, Stahl J, Jones S, Pawlowski K, Wright, C. 2005. Characterization of
vestibular dysfunction in the mouse model for Usher syndrome 1F. Journal of the
Association for Research in Otolaryngology 6:106-118.
Arar NH, Hazuda H, Steinbach R, Arar MY, Abboud HE. 2005. Ethical issues associated
with conducting genetic family studies of complex disease. Annals of
Epidemiology 15(9):712-719.
Greinwald JH, Pilipenko V, Guo Y, Choo D. 2000. Identification of Nonsyndromic
Hearing Loss Genes. Approved IRB proposal submitted to Cincinnati Children’s
Hospital Medical Center.
Guo Y, Pilipenko V, Lim LHY Dou H, Johnson L, Srisailapathy CRS, Ramesh A, Choo
DI, Smith RJH, Greinwald JH. 2004. Refining the DFNB17 interval in
consanguineous Indian families. Molecular Biology Reports 31:97-105.
Hu N, Wang C, Hu Y, Yang HH, Giffen C, Tang ZZ, Han XY, Goldstein AM, Emmert-
Buck MR, Buetow KH, Taylor PR, Lee MP. 2005. Genome-wide association
study in esophageal cancer using GeneChip mapping 10K array. Cancer Research
65(7):2542-2546.
Jiang M, Shen D, Tao L, Pandey S, Heller K, Fong HK. 1995. Alternative splicing in
human retinal mRNA transcripts of an opsin-related protein. Experimental Eye
Research 60(4):401-406.
Kennedy GC, Matsuzaki H, Dong S, Liu WM, Huang J, Liu G, Su X, Cao M, Chen W,
18 Zhang J, Liu W, Yang G, Di X, Ryder T, He Z, Surti U, Phillips MS, Boyce-
Jacino MT, Fodor SP, Jones KW. 2003. Large-scale genotyping of complex
DNA. National Biotechnology 21(10):1233-1237.
Langer P, Grunder S, Rusch A. 2003. Expression of Ca2+-activated BK channel mRNA
and its splice variants in the rat cochlea. Journal of Comparative Neurology
455:198-209.
Liu WM, Di X, Yang G, Matsuzaki H, Huang J, Mei R, Ryder TB, Webster TA, Dong S,
Liu G, Jones KW, Kennedy GC, Kulp D. 2003. Algorithms for large-scale
genotyping microarrays. Bioinformatics 19(18):2397-2403.
Marazita ML, Ploughman LM, Rawlings B, Remingtion E, Arnos KS, Nance WE. 1993.
Genetic epidemiological studies of early-onset deafness in the U.S. school-age
population. American Journal of Medical Genetics 46:486-491.
Markel DS & Yashar MB. 2004. The interface between the practice of medical genetics
and human genetic research: what every genetic counselor needs to know. Journal
of Genetic Counseling 13(5):351- 368.
Matise TC, Sachidanandam R, Clark AG, Kruglyak L, Wijsman E, Kakol J, Buyske S,
Chui B, Cohen P, de Toma C, Ehm M, Glanowski S, He C, Heil J, Markianos K,
McMullen I, Pericak-Vance MA, Silbergleit A, Stein L, Wagner M, Wilson AF,
Winick JD, Winn-Deen ES, Yamashiro CT, Cann HM, Lai E, Holden AL. 2003.
A 3.9-centimorgan-resolution human single-nucleotide polymorphism linkage
map and screening set. American Journal of Human Genetics 73(2):271-284.
Matsuzaki H, Loi H, Dong S, Tsai YY, Fang J, Law J, Di X, Liu WM, Yang G, Liu G,
19 Huang J, Kennedy GC, Ryder TB, Marcus GA, Walsh PS, Shriver MD, Puck JM,
Jones KW, Mei R. 2004. Parallel genotyping of over 10,000 SNPs using a one-
primer assay on a high-density oligonucleotide array. Genome Research
14(3):414-425.
Morton NE. 1991. Genetic epidemiology of hearing impairment. Annals of the New
York Academy of Sciences 630:16-31.
Nagai Y, Sano H, Yokoi M. 2005. Transgenic expression of Cre recombinase in
mitral/tufted cells of the olfactory bulb. Genesis 43(1):12-16.
Ouyang XM, Yan D, Du LL, Hejtmancik JF, Jacobson SG, Nance WE, Li AR, Angeli S,
Kaiser M, Newton V, Brown SD, Balkany T, Liu XZ. 2005. Characterization of
Usher syndrome type I gene mutations in an Usher syndrome patient population.
Human Genetics 116(4):292-299.
Petersen M. & Willems P. 2006. Non-syndromic, autosomal-recessive deafness. Clinical
Genetics 69:371-392.
Scott DA, Greinwald JH, Marietta JR, Drury S, Swiderski RE, Vinas A, DeAngelis MM,
Carmi R, Ramesh A, Kraft ML, Elbedour K, Skworak AB, Friedman RA,
Srisailapathy CRS, Verhoeven K, Van Camp G, Lovett M, Deininger PL, Batzer
MA, Morton CC, Keats BJ, Smith RJH, Sheffield VC. 1998. Identification and
mutation analysis of a cochlear-expressed, zinc finger protein gene at the
DFNB7/11 and dn hearing-loss-loci on human chromosome 9q and mouse
chromosome 19. Gene 215:461-469.
Sellick GS, Longman C, Tolmie J, Newbury-Ecob R, Geenhalgh L, Hughes S, Whiteford
20 M, Garrett C, Houlston RS. 2004. Genomewide linkage searches for Mendelian
disease loci can be efficiently conducted using high-density SNP genotyping
arrays. Nucleic Acids Research 32(20):e164.
Strachan T, Read AP. 2004. Human Molecular Genetics 3. London (UK): Garland
Science 405-407.
Su CC, Yang JJ, Shieh JC, Su MC, Li SY. 2007. Identification of novel mutations in the
KCNQ4 gene of patients with nonsyndromic deafness from Taiwan. Audiology &
Neuro-otology 12(1):20-26.
Sundstrom RA, Van Laer L, Van Camp G, Smith RJH. 1999. Autosomal recessive
nonsyndromic hearing loss. American Journal of Medical Genetics 89:123-129.
The Belmont Report: Ethical principles and guidelines for the protection of human
subjects. Bethesda, MD: U.S. National Commission for the Protection of Human
Subjects of Biomedical and Behavioral Research. 1978.
Van Laer L, McGuirt WT, Yang T, Smith RJH, Van Camp G. 1999. Autosomal
dominant nonsyndromic hearing impairment. American Journal of Medical
Genetics 89:167-174.
Yan D, Ke X, Blanton SH, Ouyang XM, Pandya A, Du LL, Nance WE, Liu XZ. 2006. A
novel locus for autosomal dominant non-syndromic deafness, DFNA53, maps to
chromosome 14q11.2-q12. Journal of Medical Genetics 43:170-174.
Yang M & Fong HK. 2002. Synthesis of the all-trans-retinal chromophore of retinal G
protein-coupled receptor opsin in cultured pigment epithelial cells. Journal of
Biological Chemistry 277(5):3318-3324.
Zbar RIS, Ramesh A, Srisailapathy CRS, Fukushima K, Wayne S, Smith RJH. 1998.
21 Passage to India: the search for genes causing autosomal recessive nonsyndromic
hearing loss. Otolaryngology Head Neck Surgery 118:333-337.
Zehelein J, Kathoefer S, Khalil M, Alter M, Thomas D, Brockmeier K, Ulmer HE, Katus
HA, Koenen M. 2006. Skipping of Exon 1 in the KCNQ1 gene causes Jervell and
Lange-Nielsen syndrome. Journal of Biological Chemistry 281(46):35397-35403.
22 23 Figure 2. Affymetrix GeneChip® results show linkage interval on Chromosome 4.
4
0.1 rs967413 11,069,633 rs1105930
NPL p-Value chr4:-26156414 chr4:37226047
0.01
0.001
0.0001
Base pair 25,000,000 30,000,000 35,000,000 40,000,000
Chromosomal Location: 26,150Kb – 37,220Kb
24 Figure 3. Ideogram of Chromosome 4 with linkage interval.
25 Figure 4. Affymetrix GeneChip® results show linkage interval on Chromosome 10.
10
0.1 rs717387 rs764223 NPL p-Value chr10:-77557500 21,624,425 chr10:99181925
0.01
0.001
0.0001
Base pair 80,000,000 90,000,000 100,000,000
Chromosomal Location: 77,557Kb – 99,181Kb
Figure 5. Ideogram of Chromosome 10 with linkage interval.
26 Figure 6. Affymetrix GeneChip® results show linkage interval on Chromosome 16.
16
0.1 3,421,663 NPL p-Value rs2107321 rs2077685 chr16:-2747264 chr16:6168927 0.01
0.001
0.0001
Base pair 0 2,000,000 4,000,000 6,000,000 8,000,000 10,000,000 12,000,000
Chromosomal Location: 2,740Kb – 6,160Kb
Figure 7. Ideogram of Chromosome 16 with linkage interval.
27 Figure 8. A smaller branch of the family, branch 2, served as the index branch due to the
presence of multiple affected and unaffected individuals and the appearance of autosomal
dominant transmission of hearing loss.
Note: Individual IV:18 demonstrated mild bilateral hearing loss, which was noted to be less severe than other affected family members.
28 Figure 9. Haplotype analysis using chromosome 4 markers excluded this interval due to marker segregation inconsistent with affectation in branch 2 of the family.
Markers Physical Location D4S1091 26.2M D4S391 27.2M D4S2279 29.1M D4S2912 31.6M D4S3027 32.4M D4S1587 35.1M D4S2629 36.2M D4S2382 39.7M
29 Figure 10. Haplotype analysis using chromosome 16 markers showed plausible marker segregation creating a very small interval within which the disease-causing gene could be positioned. This segregation pattern can be observed across affected individuals in branch 2 of the family (A), however the proposed genotype did not segregate with hearing loss when expanded to include branch 4 of the family (B). The proposed disease- causing chromosome is shown in yellow.
A Markers Physical Location D16S3024 1.5M D16S3395 1.9M D16S3124 2.3M D16S3070 3.0M D16S3382 3.6M D16S3027 3.9M D16S3134 5.2M D16S423 5.9M D16S3392 6.3M D16S506 6.6M D16S3092 7.4M
30 B
Markers Physical Location D16S3024 1.5M D16S3395 1.9M D16S3124 2.3M D16S3070 3.0M D16S3382 3.6M D16S3027 3.9M D16S3134 5.2M D16S423 5.9M D16S3392 6.3M D16S506 6.6M D16S3092 7.4M
Note: Haplotype for Individual III:14 was recreated based upon the genotype of the individual’s offspring, parents, and siblings.
31 Figure 11. Haplotype analysis using chromosome 10 markers showed plausible marker segregation across affected and unaffected individuals in branch 2 of the family. The proposed disease-causing chromosome is shown in grey.
Markers Physical Location D10S195 77.2M D10S2327 80.3M D10S1777 80.8M D10S1696 83.2M D10S2475 83.7M D10S1786 83.9M D10S246 84.1M D10S551 85.3M D10S1689 85.6M D10S1658 85.7M D10S1717 85.8M D10S1774 86M D10S1698 87.8M D10S1744 88.3M D10S1739 90.8M D10S583 94.3M D10S571 97.1M
32 Table 1. Subject Enrollment Criteria
Inclusion Criteria Family members with nonsyndromic hearing Affected loss with onset in the second and third decade of life
Family members over the age of 20 years with Control no documented hearing loss
Table 2. Subject Exclusion Criteria
Exclusion Criteria Family members with hearing loss caused by Rubella Prenatal drug exposure Affected Brain trauma Meningitis Family members under the age of 11 years
Family members with any hearing loss Control Family members under the age of 20 years
33 Appendix A. Polymorphic markers used to refine the candidate intervals on chromosomes 4, 10, and 16
Marker Physical
Name Location D4S1091 26.2M Marker Physical D4S391 27.2M Name Location D10S195 77.2M D4S2279 29.1M D10S2327 80.3M D4S2912 31.6M D10S1777 80.8M D4S3027 32.4M D10S1696 83.2M D4S1587 35.1M D10S2475 83.7M Chromosome 4 Chromosome D4S2629 36.2M D10S1786 83.9M D4S2382 39.7M D10S246 84.1M D10S551 85.3M Marker Physical D10S1689 85.6M Name Location D10S1658 85.7M
D16S3024 1.5M D10S1717 85.8M D16S3395 1.9M 10 Chromosome D10S1774 86M D16S3124 2.3M D10S1698 87.8M D16S3070 3.0M D10S1744 88.3M D16S3382 3.6M D10S1739 90.8M D16S3027 3.9M D10S583 94.3M D16S3134 5.2M D10S571 97.1M D16S423 5.9M Chromosome 16 Chromosome D16S3392 6.3M D16S506 6.6M D16S3092 7.4M
34 Appendix B. Candidate genes in the interval of interest on chromosome 10
Gene Symbol Gene Name Gene ID Location Function
SH2 domain containing SH2D4B 4B 387694 10q23.1 Unknown Post-translationally regulated isoform of neuregulin 3 expressed in the developing human central nervous system with a role in oligodendrocyte NRG3 Neuregulin 3 10718 10q22-q23 survival
Growth hormone inducible GHITM transmembrane protein 27069 10q23.1 Dermal papilla-derived protein
Chromosome 10 open C10orf99 reading frame 99 387695 10q23.1 Open reading frame
Member of the cadherin superfamily of calcium-dependent cell-cell adhesion molecules; has a signal 10q22.1- peptide, six cadherin repeat domains PCDH21 Protocadherin 21 92211 q22.3 and a unique cytoplasmic region
Leucine rich repeat LRRC22 containing 22 340745 10q23.1 Unknown
Leucine rich repeat LRRC21 containing 21 26103 10q23 Unknown Encodes a putative retinal G-protein coupled receptor; a member of the opsin subfamily of the 7 transmembrane, G-protein coupled receptor 1 family. acts as a photoisomerase to catalyze the Retinal G protein conversion of all-trans-retinal to 11- RGR coupled receptor 5995 10q23 cis-retinal
KIAA1128 KIAA1128 54462 10q23.1 Granule cell antiserum positive 14
35