Skip to Content

A Family Mystery, Solved by a Genome

Physicians can now use DNA sequencing to uncover the causes of rare genetic disorders.
August 25, 2010

More than a dozen people who have had their genomes sequenced stand on stage in an R&D center near Boston. Billed as the last time all such people might fit in one room before the technology moves into the mainstream, the event doesn’t quite include the whole group: actress Glenn Close and South African archbishop Desmond Tutu, among others, didn’t make it. But those who did include James Watson, codiscoverer of the structure of DNA; Harvard historian Henry Louis Gates Jr.; entrepreneur Esther Dyson; and a smattering of leaders from gene-sequencing companies.

Leaning against a wall at one end of the stage is James Lupski, a pediatrician, clinical geneticist, and scientist at Baylor College of Medicine. Unlike many of the others, Lupski wasn’t interested in sequencing as a way to trace his ancestry or determine his future likelihood of developing some ailment. Instead, he had hoped to solve a medical mystery that affects him in the most personal way: the cause of a genetic disorder, called Charcot-Marie-Tooth disease, that struck him and several of his siblings as teenagers, severely weakening the muscles in their legs and feet. After a quarter-century searching for the gene responsible, the 53-year-old scientist finally found it by scouring his own genome, combing through the billions of DNA building blocks represented by the letters A, T, C, and G. It marks the very first time that whole-genome sequencing–determining the exact order of all the letters in an individual’s DNA–has identified the mutation to blame for a specific case of a genetic disease.

Since the human genome was first sequenced a decade ago, scientists have discovered thousands of genetic variations linked to different diseases. Until very recently, however, sequencing an individual genome cost millions of dollars, making it an impractical way to search for the cause of a particular person’s genetic disorder. Now the cost of sequencing has fallen so dramatically that it’s becoming realistic to do just that. A genetic test for inherited nerve diseases, which costs about $15,000, screens for only a limited number of genes. But now sequencing is available to consumers for $20,000 and provides the entirety of a person’s genetic information. Searching through it can reveal genes and pathways whose role in a disorder scientists may never have suspected. This could help illuminate the more than a thousand rare genetic disorders for which scientists have been unable to pinpoint a specific culprit. And it could contribute to a new way of thinking about the role heredity plays in many common ailments, such as diabetes and Alzheimer’s.

In Lupski’s view, the frontiers opened by whole-genome sequencing will be to traditional Mendelian genetics what ­Einstein’s discoveries were to Newtonian mechanics. “Newton wasn’t wrong,” he says. “We were just expanding our understanding to include relativity.” Mendel wasn’t wrong either: single genes and certain diseases do follow Mendelian patterns of inheritance. But, he says, “there are genetic modifiers and new mutations that ‘Mendelism’ perhaps did not anticipate.” Just as Newtonian physics is a special case of Einsteinian relativity, Mendelian inheritance is one piece of a more complete picture that will be revealed in the genome.

To bring that picture into better focus, Lupski is now sequencing the genomes of several patients with rare neurological disorders of unknown cause. The results are unlikely to have much immediate impact on their care: there are no effective treatments for their diseases right now. But the technology could offer new insight into those diseases and guide the way to future therapies.

The Hunt

At Baylor’s campus in Houston, the Human Genome Sequencing Center spans three floors of the building next to Lupski’s office. The center, one of three nationally funded labs in the United States, became a key player in the Human Genome Project in the late 1990s. In 2007, its director, Richard Gibbs, invited Lupski to help sequence the genome of James Watson. The high-profile project, completed later that year, was a technical accomplishment because it employed a new generation of cheap sequencing technologies. But it also highlighted the limitations of the approach: they’d identified mutations in genes whose function they knew little about. What’s more, Watson had no diseases that the researchers could try to trace to a gene. So Gibbs offered to sequence Lupski’s genome.

When Lupski was first diagnosed with Charcot-Marie-Tooth as an adolescent in New York in the 1960s, the tools of human genetics were still rudimentary, and no disease genes had yet been identified. Physicians relied on particular symptoms and patterns of family inheritance to diagnose genetic disorders. In Lupski’s case, three of his seven siblings developed muscular symptoms similar to his own, suggesting that they were suffering from a recessive genetic disease. The affected siblings had apparently inherited two mutant copies of an unknown gene, one from their mother and one from their father.

In 1983, while studying for both a medical degree and a PhD at New York University, Lupski picked up a copy of the journal Nature in which geneticists reported for the first time that they had identified the approximate location of a gene responsible for a human disease, in this case the neurological disorder Huntington’s disease. (It would take another decade to find the gene within the target region.) Lupski decided that he could follow the example of the Huntington’s researchers, who had studied large families in which the disease was prevalent, to search for the genetic cause of his own disorder. That decision, which he now laughs off as naïve, would trigger a decades-long quest.

The strategy that Lupski and other gene hunters used in the 1980s was to build large family trees of relatives afflicted with a disorder. The scientists would then screen family members’ DNA for genetic markers–specific sequences, found at spots on the genome known to vary from person to person–present only in those with the disease. If all the affected family members carried a particular marker and none of the unaffected family members did, the researchers hypothesized that the disease-causing variation was somewhere near that marker. Scientists needed to study large families in order to rule out markers that were inherited in this pattern by chance; the more subjects in a sample, the easier it is to distinguish a relevant signal from genetic noise. Large families were especially important in studying diseases like Charcot-Marie-Tooth, which has such variable symptoms that individual cases can be misdiagnosed. Once they had identified a likely marker, scientists would laboriously sift through the DNA in that area of the genome, looking for candidate genes and mutations in them.

Lupski’s efforts paid off in 1991, when he and his coworkers discovered the first genetic variation linked to Charcot-Marie-Tooth. A gene on chromosome 17, at a spot involved in producing the fatty insulation that covers nerve fibers, turned out to be duplicated in some people with the disorder. It was the first time a disease had been linked to a variation in the structure of DNA, rather than to a change in a single letter or some other simple alteration in sequence. (These mutations, now known as copy-number variations, have since been implicated in a wide array of diseases, including schizophrenia and autism.)

Over the next 17 years, Lupski’s lab would identify a number of other genetic variations tied to the disease. Yet Lupski never found any of the newly discovered mutations in his own DNA. Then in 2008 came Gibbs’s offer–an opportunity to examine every gene simultaneously. After Lupski and his team sifted through the roughly 90 gigabases of raw data generated by sequencing his genome, they identified approximately three million spots where his DNA differed from the reference sequence created by the Human Genome Project. They homed in on those variations found only in genes previously linked to Charcot-Marie-Tooth or other nerve disorders.

Finally, they found two mutations in a gene called SH3TC2, one inherited from each parent. With that anomaly in sight, the researchers defrosted a set of DNA samples that Lupski had collected 25 years earlier and sequenced the gene in his siblings, his parents, and his late grandparents. He and all of his affected siblings turned out to carry both mutations, while the

unaffected family members carried either one or none.

Mendel Revisited

The variant of Charcot-Marie-Tooth disease that Lupski suffers from is a Mendelian disorder, meaning that it is caused by mutations in a single gene. (Many other diseases–typically more common ones, such as diabetes and heart disease–are triggered by a combination of complex genetic and environmental factors.) Some Mendelian diseases, known as dominant disorders, affect people who inherit just one copy of the mutant gene. For so-called recessive diseases, such as Lupski’s, it takes two defective copies to do the damage. This concept has dominated the study of human genetics for decades. But as more people have their genomes sequenced, and researchers and physicians begin to look more closely at the genes linked to specific disorders, it’s becoming clear that Mendelian genetics isn’t black and white. Genetic variations once thought to follow Mendelian rules may in fact behave in a more subtle and complicated way.

A catalogue of genetic mistakes: James Lupski has spent the last 25 years searching for the genetic basis of a number of inherited disorders. The bookshelves of his Baylor office are filled with data on different mutations.

For instance, analysis has revealed that Lupski carries two copies of mutations in each of four genes linked to other Mendelian disorders. According to traditional thinking, he should suffer from all four. But he does not. The findings may turn out to be an error in sequencing, but more likely, they suggest that these mutations don’t work the way researchers have assumed. Now the researchers are being forced to conclude that mutations in genes linked to Mendelian diseases don’t always guarantee those disorders.

It’s also turning out that carriers of recessive diseases–those who have inherited a single copy of the disease-linked mutation–may not be wholly unscathed, as Mendelian theory says they should be. Previous studies have shown that people with a single copy of the defective gene that causes cystic fibrosis are more likely than people with two copies of the normal gene to suffer from chronic sinusitis and pancreatitis. The DNA from Lupski’s family fits a similar pattern.

Twenty years ago, when he was collecting the family DNA samples, Lupski had his relatives undergo a common test for neurological disease in which physicians attach electrodes to the upper arm and measure the speed of an electrical signal sent down to the wrist. Thanks to his new genetic insights, Lupski now realizes that this test revealed nerve impairment in siblings who did not have Charcot-Marie-Tooth but had inherited a single defective gene from their mother. This type of dysfunction is linked to carpal tunnel syndrome, a common disorder often caused by repetitive hand movements. The findings hint that a single mutant copy of this gene makes people more susceptible.

If the variant is indeed linked to carpal tunnel syndrome, it probably explains only a tiny percentage of cases; the genetic defect involved in the Lupskis’ disease is rare. But the finding illustrates how studies of rare genetic diseases may also shed light on more common ones. Perhaps a number of different mutations, each one rare on its own, can all give rise to the set of relatively common symptoms that characterize carpal tunnel syndrome.

This notion falls in line with a shift in thinking among geneticists. Until recently, the role of genes in common diseases like Alzheimer’s and type 2 diabetes was thought to be very different from the one they play in Mendelian diseases. The predominant theory was that these disorders were triggered by a number of common genetic variations, each individually exerting a relatively minor effect. Over the last five years, scientists have used microarray chips designed to quickly and cheaply detect a million or more of the most common genetic variations in hundreds of thousands of people with a variety of diseases.

But the effort has failed to identify most of the genetic basis for many diseases. So scientists are increasingly concluding that the common-variant hypothesis is wrong, and that rare variants play an important role in common diseases. If so, the best way for scientists to understand the genetics of common diseases will be to take the same approach they are now using to study rare

disorders. They will need to sequence the entire genomes of patients and their families.

Better Medicine

On the second Tuesday of every month, Lupski meets with other clinical geneticists at Baylor to discuss challenging cases. Short video clips of children with a range of strange and disturbing disorders are projected on a screen in the front of the room. One boy has widely set eyes, each a different color, and hearing loss in one ear. One toddler won’t put anything in his mouth and must use a gastric feeding tube. Three brothers suffer from mental retardation of unknown cause. For parents whose children’s mysterious disorders haven’t been identified by traditional genetic testing, genome sequencing might finally bring a diagnosis, and years of medical testing could come to an end.

In a few lucky cases, it could lead to treatment. Last year, researchers at Yale University probably saved the life of a five-month-old infant in Turkey who’d been admitted to the hospital with the catch-all diagnosis of “failure to thrive.” Physicians suspected a kidney disorder. But by sequencing his exome, the portion of the genome that codes for proteins, researchers discovered a genetic mutation linked to congenital chloride diarrhea. This rare disorder can be treated by simply replacing the body’s lost salt.

In most cases, we’re still years away from cures or drugs for the genetic disorders uncovered by sequencing. Still, understanding the genetic causes of a disease is a first step to identifying its molecular mechanisms, which in turn will help researchers develop treatments. And sequencing can greatly speed up the search for disease genes. Once Lupski identified the first genetic variation for Charcot-Marie-Tooth in 1991, researchers used genetic engineering to re-create that mutation in mice and then used those animals to test potential treatments. A drug that emerged from this research is now in clinical trials for Charcot-Marie-Tooth patients who have the duplication that Lupski discovered (it turns out that about 70 percent of sufferers do). He hopes this success can be repeated for rarer disease-linked variations like his own. Another lab has already developed a mouse with a mutation in the SH3TC2 gene.

As the price of sequencing continues to plummet, Lupski believes, genetically guided diagnosis will spur a major transition in medicine by helping to spotlight the complex genetics behind both rare and common diseases.

“At one point, if you had a cough, the doctor said you had pneumonia,” he says. “Now we can distinguish between bacterial and viral pneumonias, and prescribe the right drug for the right type.”

Emily Singer is Technology Review’s Senior Editor for Biomedicine.

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.