He was the first human, 1 of 7 billion, found to lack the enzyme. The chances of both parents having the mutation and passing it to their son: 1 in 4 million. Before recent advances in genetics, researchers faced similar odds in finding it.
But find it they did. And it's just the beginning. Now that the Mights know their son's killer, they're chasing a treatment or cure.
"This is a story," wrote Matthew, "about the kind of hope that only science can provide."
'Somber boundary' • Conceived while the Mights were finishing graduate school at Georgia Tech, Bertrand was born jaundiced but healthy in December 2007. He was the family's first grandchild, a serious boy who they imagined was brooding on big ideas.
Matthew set up a college savings fund, joking how he was a liberal parent who "didn't care in which field Bertrand earned his Ph.D."
But at six months, Bertrand, affectionately known as "Buddy," wasn't smiling and his movements were jerky and irregular "jiggly," as the Mights describe it. His only controlled motion was to wring his hands like Dr. Evil, said Cristina. "It was cute, endearing, but worrisome."
The Mights feared he may have autism.
They moved to Salt Lake City, where Matthew joined the computer science faculty at the University of Utah, and took Bertrand to a developmental pediatrician who guessed he had brain damage, possibly from the jaundice.
"I remember walking home from a faculty retreat thinking, 'Oh my God, what did we do?' ... It was a really awful feeling," said Matthew.
An MRI, however, showed Bertrand's brain was normal.
Cristina, who had just earned her MBA, abandoned her plans to help Matthew with a software start-up. Instead, they launched what they called "Operation Diagnose Bertrand," a whirlwind of blood tests, biopsies and MRI, CT and EEG scans.
With every disorder crossed out came a cascade of others most of them degenerative and fatal.
Bertrand's feet, fingers and forearms became so scarred, nurses looked for veins in his head and shoulders, Cristina recalls. And he was losing ground, his ability to sit up and roll over.
Driving Bertrand home from a doctor's appointment one day, Cristina almost crashed. She remembers despairing, "Maybe it would have been better if we had died. I was totally burned out. I couldn't be happy. I couldn't be sad. It was just a numbness that took over."
Doctors came to suspect Bertrand had a recessive disorder caused by two mutant copies of the same gene, one from each parent. In such cases the parents are usually distant cousins.
But Matthew is of Northern European descent, and Cristina's family is Puerto Rican. And the small number of genetic screens available for recessive disorders came back negative.
At 10 months, Bertrand had landed in what Matthew calls "the empty set," a mathematical term for the unknown.
"There's a boundary that many parents of children with rare disorders cross, a point at which you've ruled out all diagnosable illnesses," he said. "It's a somber boundary."
Cristina scoured the Internet for case studies and clinical trials, sharing what she learned in her own blog.
"It [is] difficult feeling so alone, so un-Googleable," she wrote in January 2009. "It has become my personal mission in life to make search terms like 'jiggly baby' result in at least some medically helpful information other than the annoying YouTube videos."
'Baby' bucket list • The first big clinical discovery came shortly after Bertrand's first birthday: oligosaccharides, or chains of simple sugars, were in his urine. This suggested he had a glycoprotein-storage disease, typically caused by the body's inability to produce an enzyme.
Bertrand's cells were unable to recycle waste, which was building up in his cells and destroying them.
This explained his new seizures and why he was symptom-free at birth, since it takes time for waste to build up.
But the finding was a death sentence. He would live another year, maybe two.
Every type of therapy had failed to awaken Bertrand, to stop his decline. Cognitively, Bertrand, who is now nearly 5, is 8 months old.
Cristina had been hoping to find time to make new friends in Utah. Instead, she cleared her calendar for a "baby" bucket list: trips to the zoo, a farm, a beach and horseback riding.
"Screw new friends. I don't need them. Bertrand's my BFF," she wrote in March 2009. "Bertrand is going to have the best next 21 months that any baby has ever had!"
The Mights briefly pursued a bone marrow transplant from expert Joanne Kurtberg at Duke University, hoping a donor's cells would produce Bertrand's unknown missing enzyme the only known treatment for disorders like his.
But a scan showed his brain was losing white matter, the networking structure that allows nerve cells to communicate. It was too late. He was unlikely to survive the procedure.
On May 15, 2009, they closed Bertrand's college account.
"We've come to accept that he's never going to attend college, even if he survives long enough," Matthew wrote on Cristina's blog.
Using the savings for medical expenses, they focused on treating his epilepsy and preserving his quality of life. A high-fat diet with twice-daily hormone injections eliminated Bertrand's seizures, but the hormones put him in a rage, doubled his size and stripped away his immune system. After his second birthday, he nearly died from a respiratory infection.
But in the hospital, Bertrand laughed for the first time.
"It was the most direct sign of [his] humanity we had ever seen," wrote Matthew on his blog.
Bertrand recovered and surprised his therapists with his attempts to sit up and crawl.
Matthew was still stalking a diagnosis, to better care for Bertrand and to consider as they began thinking about having another child. Pursuing a last-ditch idea sequencing the entire family's genome to search for Bertrand's genetic glitch he wrangled a dinner in 2010 with U. geneticist Lynn Jorde.
Sequencing is still fraught with errors, Jorde said in an interview. Everyone has thousands of mutations, some helpful, others benign and some harmful.
"If we analyzed your DNA and you have Type 2 diabetes, you would have 10,000 variants," Jorde said, "and any single one could be the cause of your diabetes."
Matthew estimated that, without other affected family members to narrow the target, the sequencing would cost at least half a million dollars.
"What I had proposed was theoretically possible," said Matthew. "But it was anguishing to know that while the technology already existed, it was far too expensive to hunt for mutations this way."
'Overwhelming evidence' • Two weeks later, the phone rang.
At Duke's Center for Human Genome Variation, researcher David Goldstein wanted to test a new idea for diagnosing the undiagnosable.
He would focus on the "exome," a tiny portion of a genome, the 1.5 percent of human DNA that drives protein production. Misfires in protein production are thought to be responsible for most genetic disorders.
Familiar with Bertrand's case from the transplant attempt, a member of Goldstein's team asked to include him in a pilot study of 12 children with developmental delays and birth defects thought to be genetic, but with symptoms that defied explanation.
Goldstein wanted the hardest cases he could find, and Bertrand was the first to come to mind.
"We thought if we could succeed 10 percent of the time," Goldstein later explained, "that would be good enough" to show the approach works.
About a month after the study began, Cristina learned she was pregnant.
Matthew had become convinced that Bertrand's mutation was new, not inherited, which meant a second child would not be at risk. Cristina, uncertain, worried during the pregnancy about both children. Contemplating having a child who "would enjoy the park and hug me back," she said, felt like a betrayal of Bertrand.
Victoria was born plump and healthy in 2011.
This May, the family flew to North Carolina to hear Duke's results a breakthrough.
Goldstein's team had landed diagnoses for seven of the 12 children, including Bertrand.
The more than 50 percent success rate, Goldstein said, is "overwhelming evidence" of technology's promise for "all patients who fail to get a diagnosis through traditional testing."
"There are about 50,000 live births a year of children who probably should be analyzed this way," he said, "and that ignores the backlog."
Exome sequencing has its limits, warns Goldstein. "The vast majority of genetic diagnoses we obtain will not lead to a treatment for the patient. ... And in developmental cases, it will be the case where the damage is done before you get a diagnosis."
Here's what the Mights now know: Matthew and Cristina each have a different mutation in the same gene, NGLY1, and Bertrand had inherited both.
To claim discovery of a new disease, Duke had to find another Bertrand. But they had double-checked their finding by sending blood samples to Hudson Freeze, a glycobiology expert in La Jolla, Calif. Freeze confirmed Bertrand was producing almost no N-glycanase, an enzyme regulated by the NGLY1 gene.
The enzyme plays a lead role in recycling misfolded proteins, which were accumulating in Bertrand's body. That night in their hotel room, Matthew and Cristina wondered, could replacing the enzyme help him? They emailed Freeze.
At home, a week later, an exultant Matthew wrote about his wife on Mother's Day.
Being a mom means believing the crippled will stand, the mute will speak and the blind will see, he said. "It is about exerting a force of will so strong that, even if for only the briefest of moments, the line between the possible and the impossible bends. It is about exerting that force for so long and with such intensity that, in rare circumstances, that line breaks."
'You're the Columbus' • On May 29, he wrote his blog post trumpeting the unmasking of Bertrand's assassin.
It was like "a social media nuclear bomb went off," Matthew said. It shot to the top of Hacker, Reddit and Gizmodo and other sites and was being retweeted at a rate of every two to three seconds.
In 48 hours it had drawn more than a million views.
"The result," Matthew said, "was an avalanche of information coming back at us from anyone who knew anything about glycobiology or exome sequencing."
Cristina found a synthesized form of N-glycanase, patented by Genzyme. It's been used in labs for two decades and costs very little, but it has not be cleared for use in humans.
Getting federal approval to try it as a treatment would take Genzyme's cooperation; the first step would be to test its impact on Bertrand's cells.
Freeze was willing to do it, but his lab at the California-based medical research institute, Sanford-Burnham, didn't have the staffing or funding. The tab for hiring their own postdoctoral researcher to do the work: about $120,000.
Thanks in part to their blogging, the Mights swiftly raised the money.
Freeze's study of glycans is arcane, even among biologists. Glycans are sugar chains that stick to proteins, sticky molecules that tell cells how to develop into the brain, the nervous system and other major organs. Freeze explores how they are made and what happens when the process goes awry.
In August, the Mights flew to Freeze's lab for their first visit. "We just want them to meet Bertrand to know that he's more than a case number, that there's a name and face behind those cells," Cristina said the night before their flight.
Bertrand's missing enzyme, Freeze explained in an interview, appears to "strip off sugar chains from proteins that the body needs to throw away because they're not folded right or assembled the way they should have been."
But Bertrand can't get rid of such proteins, because they won't fit in a cell's "mulcher," Freeze said. "It's like trying to fit an entire tree in a mulcher. The sugar chains are like bulky branches and flop all over the place."
Injecting a synthesized form of N-glycanase into Bertrand's cells isn't working. There's still hope experimental gene therapies small compounds that read through, or patch over, damaged DNA will work. And Bertrand's body appears to have some ability to discard the proteins and there might be ways to enhance that.
"We may benefit as scientists as much as we are able to help him," acknowledges Freeze. "When you're writing your own book, as he's doing, you're the Columbus. You're going to take people where they've never been before, to new insights and new ideas, and whether they like it or not, to some dead ends."
Armed with Freeze's explanation of the type of stress on Bertrand's cells, Cristina is trying antioxidants, including alpha lipoic acid and medical-grade curcumin.
"My son smells like curry," said Cristina. "But he seems happier. His seizures seem reduced. He's sleeping better."
'A kind of peace' • The Mights are trying to connect with a family in Turkey, recently found by doctors at Yale University to have two children with N-glycanase deficiency. Their son, with symptoms almost identical to Bertrand's, died at age 8. His sister is still well. Doctors at Yale are legally barred from disclosing their identities.
Bertrand is now almost 5. As he headed to preschool on a recent afternoon, Victoria scaled his wheelchair and sat astride his lap for the ride to his school bus.
She's his biggest fan, said Cristina. She'll sometimes wake to see Victoria has left her crib, snatched his iPad from the bookshelf and hopped into his bed to watch videos with him.
Victoria can tease a smile from him, sometimes a grimace.
But, at 18 months, she talks more than he ever has or maybe ever will.
The family's routine is a blur of hand-fed meals, baths and diaper changes. There's the hourly eye drops to lubricate Bertrand's unblinking eyes, and daily in-home therapy sessions to build his motor control.
His condition is still likely fatal.
"But I prefer to think we've finally reached a kind of peace," Cristina blogged in September. "We're comfortable with doing what we can rather than obsessing about what we can't." Maybe it's finally having a diagnosis, or knowing what they've found will benefit other families, she said.
Matthew continues to talk about making "new dents in the boundary of human knowledge." Watching his son loaded onto the bus, he said he's confident there's a 50 percent chance they'll find a treatment.
"We need to do more pushing on the science," he said. "We're almost there."
By the numbers
22,000 » Each person's number of genes, which make up his or her DNA, the genetic blueprint for human life.
More than 3 billion » The number of "base pairs" that determine genetic makeup.
$3,000 » Cost to sequence a family's exome, 2 percent of each member's DNA.
$20,000 » What one family in the Duke study that diagnosed Bertrand spent on dead-end diagnostic tests.
$75,661 » Four years of the Mights' out-of-pocket health expenses.
$3 billion » Cost of the Human Genome Project, begun in 1990 and completed in 2003.
Help find a cure
To donate to the Bertrand Might Research Fund at Sanford Burnham, go to: http://bit.ly/QbCUDO