In the hunt for the genetic causes of hearing loss in local populations, researchers at Tel Aviv University have taken up targeted genomic capture and deep sequencing to root out causative mutations.
The group published a description of its capture and massively parallel sequencing method in Genome Biology last week, and is now expanding its initial study of 11 subjects and their families to a further 96.
Eventually, the researchers' goal is to be able to offer population-specific tests that would assay only the deafness genes that correlate with an individual's ethnic background and deafness phenotype, Karen Avraham, the lab's leader and a professor of genetics and biochemistry at Tel Aviv's Sackler Faculty of Medicine, told Clinical Sequencing News. In the meantime, the method screens all of the known genes involved in human deafness, as well as genomic regions that correspond to known mouse genes associated with hearing loss.
Avraham added that Pronto Diagnostics, an Israeli clinical laboratory company, plans to offer testing for hearing loss using a method that is based on her team's work.
"I'm in a research lab. We develop the method and then hope that genetics clinics will then take over," Avraham said. "But, this is something that, if you have the setup in the hospital, it is very doable," she said.
In their proof of concept of the method for investigating hearing loss mutations, Avraham, her coauthor Moien Kanaan from Bethlehem University, and their collaborators used Agilent's SureSelect in-solution capture and the Illumina Genome Analyzer to screen 246 genes, comprised of 82 human protein-coding genes, two human microRNAs, and the human orthologs of 162 genes associated with deafness in mice. The subjects were from 11 different families of Israeli Jewish and Palestinian Arab origin.
Captured DNA libraries from each subject were labeled with a different barcode, and one to two libraries were multiplexed in each lane of the Illumina GA. After paired-end sequencing, median base coverage across the captured targets was 757x to 2,080x with over 95 percent and 92 percent of targeted bases covered by more than 10 or 30 reads respectively, the group reported.
Using the Burrows-Wheeler aligner and MAQ bioinformatics methods, the researchers aligned their reads to the human reference genome sequence. Rare variants were identified by filtering against dbSNP132, data from the 1000 Genomes project, and additional filters, they reported.
Potential variants were validated by Sanger sequencing, and overall, the Tel Aviv team identified critical mutations in six of the 11 subjects and their families, "leading to the identification of causative alleles in 20 additional probands and their families," they wrote.
One gene, TMC1, had "substantial clinical implications" for the Israeli deaf population of Moroccan Jewish ancestry, the researchers reported. After identifying a novel mutation in TMC1 in one of the study's probands, the team went on to find that more than 30 percent of the larger pool of Moroccan Jewish deaf subjects carried the allele.
Based on the results, the group recommends "that all Israeli Jewish probands of Moroccan ancestry be screened" for mutations in TMC1. "We estimate that TMC mutations explain at least 38 percent of inherited hearing loss in the Moroccan Jewish population," they wrote in the paper.
Novel mutations were also identified in other genes known to be involved in hearing loss, but not routinely evaluated, including CDH23, MYO15A, WSF1, and TECTA. Of the six Palestinian families enrolled in the study, a causative mutation — in MYO15A — was found only in one, the authors reported.
A Similar Approach
A group from the University of Iowa has also developed a sequencing-based deafness test using Agilent SureSelect and the Illumina platform, called OtoSCOPE for Otologic Sequence Capture of Pathogenic Exons (CSN 3/22/2011).
The non-profit lab is now testing individuals on a research basis and will launch the test on a clinical basis in January, Eliot Shearer, one of the researchers behind OtoSCOPE, said in an e-mail to Clinical Sequencing News this week.
[ pagebreak ]
The approach that Avraham’s group used "was very similar to the approach we used for OtoSCOPE
in that they targeted all known deafness genes … However they used it in a specific population for the first time," Shearer wrote. "We expect others to use a similar methodology in many populations around the world in the near future. This will help us to better understand deafness and paves the way for personalized
medicine and molecular therapies for deafness," he said.
Avraham noted that another departure from the OtoSCOPE group "was that we also added all the genes known to cause deafness in the mouse. This is important because there is an enormous amount of research showing how similar the mouse and human inner ear are and many of the 60 human genes cause deafness in mice as well."
Shearer said that though only human genes are included in the OtoSCOPE test, the Iowa team also uses the mouse in its work to identify novel deafness genes in humans.
Avraham said that her team's initial study of 11 probands has not yet revealed any mutations in mouse gene orthologs, but her team predicts it will find them in continuing work.
"We have hundreds [more subjects] with their DNA sitting in our freezer," she said. "This was a proof of principle. Now we are conducting another experiment where we are doing 96 samples at a time."
According to Avraham, almost 150 candidate locations for deafness genes are known, but so far only around 60 of those loci have had genes identified in them. The team expects to expand its assay as more deafness-related genes are uncovered. "It’s a very dynamic process and new genes are being discovered every month, so we're not going to be stuck with these 246," she said. The team has already added another "five or six genes" in its continuing assessment of 96 patients, she added.
Avraham said that she hopes this more comprehensive method of genetic analysis will offer answers for patients who have been disappointed by simpler tests. "We would check a few genes and give them the best answer we could and it was usually negative and that was it," she said.
"Now we have massively parallel sequencing, and the beauty is it allows you to sequence many DNA molecules simultaneously … We used to do 700 base pairs per run, now we can do gigabytes per run and an entire genome in 10 days and for a reasonable price."
"In one experiment, we can check all these 250 genes in all our patients… for about $500 per person," she said.
Beyond the allure of a definitive diagnosis, testing for the mutations at the root of hearing loss also offers other benefits. "If you know what the mutation is, particularly in a child," Avraham said, "you can predict if there is going to be another abnormality like blindness or neurological defects," and potentially delay or treat these associated symptoms.
"You can also predict the progression of the hearing loss — is it going to get worse or stay stable — which is very important for decisions on whether to place a cochlear implant. The earlier you put one in the better it works, but if you can predict that the hearing loss is going to stay moderate, you might not want to do it," she said. In the future, knowledge about the cause of hearing loss could also lead to gene therapies for deafness.
"These are just some of the reasons to know," Avraham said.
Moving forward with this method for searching out mutations, Avraham said there are certainly still some areas for improvement. The bioinformatics side is one important area. With analysis taking up to a few months, "it can be tedious," she said. "Today I would say that's probably the bottleneck in this technique."
Also important, she said, is price. "We would like it to be $200 …. On the other hand it used to be $20,000," she said.
While her lab is not intending to do diagnostic testing outside of its research to establish this method, she said that Pronto Diagnostics plans to offer such testing, and that the procedure could be performed by anyone with the technology and personnel on hand.
Avraham said she expects that knowledge of how causative gene mutations relate to populations and phenotypes will eventually make a multi-gene test like her team's unnecessary.
"The simplest thing is if a patient comes into a clinic, and you say you're Moroccan Jewish, or you're Palestinian Arab, or you have a specific malformation in your ear, there is one test we should do and it should give us the answer," she said. "And these one-gene tests are quite inexpensive."
"That's really our goal — to find a genotype-phenotype correlation, so that a doctor will see a patient and have a pretty good idea what gene it is and just test for that gene … And that's not just for deafness, that's for any disease."
Have topics you'd like to see covered in Clinical Sequencing News? Contact the editor at mashford [at] genomeweb [.] com.