Wolffe’s expertise was crucial in creating artificial transcription factors—proteins that bind to specific DNA motifs to switch genes on and off—that can drive cells to particular developmental fates. Pabo had shown that by mixing and matching individual zinc finger units, researchers could design a new hybrid transcription factor to recognize a specific DNA sequence. By the time Urnov arrived, Sangamo had designer zinc finger proteins to show this approach was feasible, setting the stage to test their efficacy in human cells.
Then in May 2001, tragedy struck. While attending a conference in Rio de Janeiro, Wolffe was struck by a bus and killed while out running one morning. “Alan was the hub of the spokes… and then he’s gone.” Pabo, the chair of Sangamo’s scientific advisory board, acted as the research team leader until Gregory assumed the role two years later.
Just as Sangamo was preparing to enter the clinic, using a novel ZFP to treat nerve damage in diabetes patients by switching on the VEGF gene, news arrived of another setback. In the summer of 2002, just months after Alain Fischer and colleagues had published exciting gene therapy trial results,9 reports emerged that one of his patients had developed leukemia. The virus carrying the therapeutic gene had inserted itself into the patient’s genome, causing cancer. Following the 1999 Jesse Gelsinger tragedy, gene therapy trials were put on hold or abandoned.
By this time, however, a glimmer of light for zinc fingers had emerged, offering the possibility of not merely replacing a disease gene with a healthy copy, or switching on a silent gene, but actually editing and fixing the mistake at the DNA level. A team at Johns Hopkins had devised a means to modify zinc fingers to target them to genes of interest.
In 1978, Hamilton “Ham” Smith, a 6’5” biochemist at Johns Hopkins University, received the call almost every scientist dreams about: he had won the Nobel Prize. Smith, never comfortable in social situations, survived the scrum of press photographers and well-wishers, and a mild case of Imposter Syndrome. Even his mother was surprised: when she heard the news on the car radio, she turned to her husband and said: “I didn’t know there was another Hamilton Smith at Hopkins.”10
Smith’s Nobel was awarded for his serendipitous discovery of restriction endonucleases, a large family of bacterial enzymes that recognize and cut specific DNA sequences or motifs. Genetic engineers turned these enzymes into the catalysts of the recombinant DNA revolution. “Everything about modern biology, from the idea of determining a DNA sequence to the idea of recombinant DNA to DNA fingerprinting, it all starts with restriction enzymes,” said geneticist David Botstein.11 By the mid-1990s, thousands of restriction enzymes had been catalogued, shipped commercially around the world in polystyrene buckets of dry ice. But their usefulness as a scalpel for precision gene editing was limited. These enzymes cut DNA at very short recognition sites, typically only four to six base pairs. While those motifs might only occur a few times in the tiny genome of a virus, they crop up thousands of times scattered across the human genome.
Smith often discussed the idea of engineering artificial enzymes that could be more selective in cutting DNA with his students, including in 1986 a visiting chemist named Srinivasan “Chandra” Chandrasegaran. Years later, Chandra set out to engineer a chimeric restriction enzyme, a new kind of nuclease. Flicking through the enzyme catalogue offered by New England Biolabs, Chandra and his colleague Jeremy Berg settled on the amusingly named FokI from Flavobacterium okeanokoites. Like a Star Wars TIE fighter seeking the thermal exhaust vent, FokI scans the DNA in search of a specific landmark—a sequence of five bases, GGATG. But once it settles on the DNA, the actual cutting is carried out by a different domain of the enzyme about ten bases downstream. As the two domains were separate, Chandra reckoned he could alter the target parameters by tethering a different DNA recognition domain to the cutting site.
Chandra published his “hybrid restriction enzyme” breakthrough in 1996.12 His team fused the DNA-cutting domain of FokI with zinc-finger domains that supplied the specificity. “In theory,” Chandra wrote, “one can design a zinc finger for each of the sixty-four possible triplet codons, and, using a combination of these fingers, one could design a protein for sequence-specific recognition of any segment of DNA.” These zinc-finger nucleases (ZFNs) could be programmed to latch onto any DNA sequence that would serve all manner of applications. Interestingly, Chandra’s choice has stood the test of time. “Like the fact that a [soccer] match lasts ninety minutes or the QWERTY keyboard starts with the letter Q, it is widely accepted,” says Urnov. “People haven’t seen the need to evolve beyond that.”
Chandra was in no doubt that his chimeric nucleases—“a new type of molecular scissors”—could transform gene therapy: in 1999 he said his goal was to excise a gene mutation and replace it neatly with its normal counterpart. Ethical issues aside, he wrote, “gene therapy will be routinely used in clinical practice, signifying a paradigm shift in the treatment of human disease.”13
Chandra teamed up with Dana Carroll, a biochemist at the University of Utah, who wanted to customize a ZFN to engineer a mutation in a classic animal model such as the fruit fly. If done right, the Drosophila cells would turn from brown to yellow. Carroll’s colleague saw the yellow bristles down the microscope.14 “If I were you, I’d be pretty excited,” he told his boss. By 2002,15 Carroll’s group had demonstrated the ability to engineer DNA in living organisms, the first use of ZFNs not merely to modulate the expression of certain genes, but actually to change their DNA sequence. Carroll’s development of ZFNs coupled with the editogenic insights from Jasin and colleagues laid the foundation for genome editing in humans.16
From 1997, Sangamo’s headquarters was in a building in Point Richmond, shared with Pixar, the animation studio behind Toy Story and A Bug’s Life later acquired by Disney. When Pixar moved to a larger headquarters in Emeryville, Sangamo expanded into the space. For three years, Urnov and Holmes shared an office that was formerly Pixar’s screening room. Urnov says their partnership was akin to Lennon and McCartney, before conceding that might be a bit of a stretch. Assisting the Sangamo team was Matthew Porteus, a physician-scientist at Stanford who had trained with David Baltimore. He’d also been inspired by Carroll’s ZFN papers and wanted to get them to work in human cells. Porteus developed an assay using the green fluorescent protein that could report successful gene targeting using ZFNs.17
Sangamo’s young musketeers were on a mission and there was no time for failure. “Nothing creates a sense of urgency like being on Nasdaq,” says Urnov. Over the next few years, Sangamo figured out how to turn good ZFNs into effective gene editors. There were multiple disease targets—sickle cell disease, hemophilia, and severe combined immunodeficiency (SCID). (Urnov and Holmes even dabbled with editing the CCR5 gene.) With the French gene therapy setback in everyone’s minds, Sangamo began looked to repair the genetic glitch in those SCID patients—a mutation in the gene for the interleukin-2 gamma receptor (IL2Rζ).
One day, Urnov was reading the results skipping off a lab instrument called a phosphoimager. It looked like “we’d achieved a one-in-five efficiency of gene editing. Efficiency like this happens spontaneously in about one in a million cells.” Urnov shared the results with Holmes. “If this is real, we’ve just entered into a new era!” Holmes concurred, already planning the next experiments to pressure test the result while an exhilarated Urnov paced around the room. “Extraordinary claims demand extraordinary evidence. We both knew nobody would believe us!” Urnov recalls. They kept the results to themselves, while secretly running every control experiment that they could think of. Sangamo’s expertise was starting to pay off. “We’d finally built the fast engine in the car with the superb tires, a super-aerodynamic frame, and a super-flat racing track.”