A research team has identified mitochondrial disease-causing mutations by specifically sequencing the genes contributing to the organelle's function, demonstrating the diagnostic potential of this "MitoExome" sequencing approach.
"Because the majority of mitochondrial diseases are due to mutations in proteins localized to the organelle, [the MitoExome] is a high-quality set of candidate genes for these disorders," co-corresponding author Vamsi Mootha, a human genetics and systems biology researcher affiliate with the Broad Institute, Harvard Medical School, and Massachusetts General Hospital, told Clinical Sequencing News.
"We basically developed an exome capture reagent for all of the genes encoding the mito-proteome," Mootha said.
The researchers believe the approach could be useful in diagnosing other mitochondrial conditions, since it is more cost efficient than whole-genome sequencing and provides information on both mitochondrial and nuclear DNA. In addition, it does not rely on information about an individual's family history.
In a proof-of-principle study published in the current issue of Cell Metabolism, the American and Australian group described how they employed targeted sequencing of mitochondria-related genes to find disease-causing mutations in two children with a mitochondrial condition called Leigh syndrome. The search uncovered mutations in a gene called MTFMT, which codes for an enzyme related to protein translation in the mitochondria.
With the finding, MTFMT joins a growing list of genes that have been implicated in Leigh syndrome. Past studies have shown that the condition can stem from mutations in roughly 40 different mitochondria-related genes, some found in the maternally inherited mitochondrial genome and others coded by the nuclear genome.
"Leigh syndrome is sort of a microcosm of mitochondrial disorders in general," co-corresponding author David Thorburn, a pediatrics and mitochondria researcher with the University of Melbourne and Australia's Murdoch Children's Research Institute, told CSN. "In one sense, they're sort of a perfect test case for next-generation sequencing techniques, because there are so many genes in which mutations can cause mitochondrial dysfunction."
Mitochondria produce fuel for the body in the form of adenosine triphosphate, or ATP, via electron transfer in the oxidative phosphorylation process. In addition to their energy production duties, mitochondria also contain pathways involved in cellular processes such as homeostasis and apoptosis.
The mitochondrial genome houses genes for some proteins used by the organelle, including those coding for transfer RNAs, ribosomal RNAs, and 13 of the 90 proteins involved in oxidative phosphorylation. But many proteins contributing to mitochondrial processes are housed within the cell's nuclear genome, the researchers explained.
Mutations in dozens of genes in the mitochondrial genome and some 80 nuclear genes can lead to mitochondrial diseases, Thorburn said, which means diagnosing these conditions and tracking down culprit mutations can be tricky.
"Overall, there are more than 120 different, known disease genes," he said. "So trying to get a diagnosis in these [mitochondrial] disorders, even in one clinical syndrome like Leigh syndrome, is very complicated by this huge genetic complexity and diversity."
To try to get a handle on this complexity, researchers have started cataloging and characterizing mitochondrial proteins produced by both the nuclear and mitochondrial genomes.
In a study published in Cell in 2008, for example, Mootha and Broad Institute researcher Steven Carr led a team of researchers from the US and Australia, including Thorburn, who used mass spectrometry-based proteomics to catalog 1,098 mouse mitochondrial proteins in a compendium they called the MitoCarta.
For the current study, researchers examined sequences coding for the human version of this mitochondrial protein catalog in an effort to find mutations behind Leigh syndrome, the most common mitochondrial disease diagnosed in childhood, in individuals who also had OXPHOS deficiency, a shortage of proteins involved in oxidative phosphorylation in the mitochondria.
Children with Leigh syndrome are born healthy, but start experiencing progressive neurodegeneration at around six months of age, Thorburn explained. That can lead to problems with growth, movement, and, in some cases, seizures, heart problems, or death. The condition is usually diagnosed through neuroimaging, which detects characteristic changes in the brain stem and basal ganglia of the brain.
If genetic mutations causing the disease are identified, it's usually through a laborious process that starts with determining which mitochondrial complexes are active in muscle or liver biopsy samples from affected individuals, followed by targeted sequencing of candidate genes.
In an effort to streamline and simplify this process, Thorburn, Mootha, and their colleagues used the MitoExome sequencing approach to look as many candidate genes as possible in one go.
"The idea with the MitoExome study is that we're sequencing the genes encoding all the known mitochondrial proteins as a subset of the whole exome," Thorburn said, "reasoning that virtually all of the genes associated with mitochondrial disease are going to be represented in that number of a little over a thousand genes encoding the MitoExome."
[ pagebreak ]
Using the Illumina Genome Analyzer, the team sequenced 4.1 million bases of MitoExome DNA, representing the mitochondrial genome and the coding sequences from 1,381 nuclear genes, in two unrelated children with Leigh syndrome and OXPHOS deficiency.
The MitoExome sequences, which were captured using in-solution hybridization using RNA baits, were selected based largely on MitoCarta genes identified in the 2008 mouse study.
"[The MitoCarta study] provided us with a gene list, if you will," Mootha said. "Thanks to the advances in sequencing technology, we can now capture all of those genes and sequence them."
The team is not alone in looking at methods for targeted sequencing of mitochondrial genes. In April, researchers at the Stanford Genome Technology Center published a paper in PNAS describing a targeted sequencing assay for genes encoding mitochondrial proteins (CSN 4/19/2011).
While the Stanford assay, which uses long padlock probes, captured 524 mitochondrial genes, the researchers affiliated with the Leigh syndrome study looked at 1,013 genes designated as mitochondrial genes in the MitoCarta database, 21 genes that appeared to be related to mitochondrial genes, and another 347 genes suspected of participating in mitochondrial processes.
By comparing these MitoExome sequences with those in the human reference genome, the team tracked down roughly 700 SNPs, small insertions, and small deletions for each patient.
After tossing out the variants that appeared to be harmless based on their frequency in public databases, the researchers were left with just 20 or so suspicious variants per person. They narrowed the variants down even further by looking at the predicted consequences and inheritance patterns for each genetic change.
"We make the underlying assumption that variants must be rare because if they weren't rare then the prevalence of the disease would be much higher. The second assumption we make is that the particular variant must have an impact on protein function," Mootha explained.
In both patients, the genetic search led to mutations in MTFMT, an enzyme that formylates a transfer RNA that shuttles the amino acid methionine in the mitochondrion. Within animal cells, this methionine-tRNA not only transfers methionine but also initiates translation. The tRNA is formylated by MTFMT prior to translation initiation.
Through a series of follow-up experiments in patient fibroblast cells, the researchers found evidence that the MTFMT mutations led to lower-than-usual levels of formylated methionine-tRNA in the individuals' mitochondria that, in turn, hindered efficient protein translation in the mitochondria.
"[W]e have used MitoExome sequencing to identify MTFMT as a gene underpinning combined OXPHOS deficiency associated with Leigh syndrome," the researchers wrote. "More generally, this study demonstrates how MitoExome sequencing can reveal insights into basic biochemistry and the molecular basis of mitochondrial disease."
'One Stop Shop'
Those involved in the Leigh syndrome study say the same approach holds promise for diagnosing other mitochondria-related conditions, since it is cheaper than whole-genome or whole-exome sequencing, provides information on both mitochondrial and nuclear DNA, and does not rely on information about an individual's family history.
"In essence, what we're trying to generate is kind of a 'one stop shop' for looking at genes known to be involved in mitochondrial disorders and a huge range of candidates," Thorburn said.
The deep coverage of the mitochondrial DNA provided by MitoExome sequencing also aids in mitochondrial genome assembly and, in some cases, points to deletions larger than those usually visible with high-throughput sequencing alone, he noted.
Although the sequencing technology is fairly straightforward at this point, challenges remain in systematically analyzing and categorizing sequence variants to narrow in on relevant mutations.
"The key part of our approach is the bioinformatics strategy and getting down to a manageable number of variants to prioritize and then being able to validate that by experimental approaches," Thorburn said. "That's really still the challenge for bringing these sorts of studies into the clinic."
At the moment, he estimates that the MitoExome sequencing approach would uncover disease-causing genes in about half of OXPHOS cases. Nevertheless, for many individuals with mitochondrial disease, detailed experimental follow-up will probably still be needed to sift through the genetic changes identified in the MitoExome data.
"In a clinical sense, this is essentially ready to be used in a diagnostic context," Thorburn said, "but it does require quite extensive experimental follow-up as well as the sequencing."
As more genome sequence data and analysis tools become available as a result of large sequencing studies such as the 1000 Genomes Project, Thorburn predicts that it will become easier to interpret and analyze sequence data to find clinically relevant variants.
While Mootha said he and his colleagues might eventually do some whole-genome sequencing studies of mitochondrial disorders, the researchers are currently working to apply their MitoExome sequencing approach to a much larger group of individuals.
"There are a lot of applications of exome sequencing in general to individual cases or to individual families," Mootha said, "but we have some ongoing studies right now that will apply this to entire cohorts."
Have topics you'd like to see covered in Clinical Sequencing News? Contact the editor at anderson [at] genomeweb [.] com.