A Synthetic Quest for Better Medicine and Biological Insight.
Gene therapy has always seemed like a simple enough idea. After all, many genetic diseases arise because people carry broken or dysfunctional versions of one gene or another. If there were a way to add a properly working copy of those genes back in, the associated diseases should go away, or at least get a whole lot better.
In reality, though, “gene therapy has largely failed for the last 25 years, with some high profile negative clinical trial results,” says Charles Gersbach, an assistant professor in the Department of Biomedical Engineering and a member of the IGSP Interactome.
Those failures are not for lack of trying, of course. As of the beginning of this year, there were more than 1,700 clinical trials for gene therapies in countries all over the world for all kinds of conditions, single-gene disorders, cancer and cardiovascular disease among them. Some therapies work – but not often enough and not well enough.
Part of the trouble comes out of the methods used to replace mutant genes, relying on viruses to package and deliver extra, good gene copies into the genome in more or less random locations. In some cases, those replacements have seemed to work, but only temporarily. In others, the therapeutic genes have landed themselves in places where they can cause other problems, including cancer. Gersbach has compared it to fixing a broken headlight on a car by adding a new one that may or may not point in the right direction.
As an engineer and synthetic biology aficionado, he has a different approach to the problem. Synthetic biology can mean different things to different people, but in general it’s oriented around the idea espoused by the Registry of Standard Biological Parts that genetic parts (e.g. genes, promoters, protein domains, and the like) can be isolated and assembled in novel ways to build devices and systems, applying engineering principles to the growing “parts list” emerging from genomic and other comprehensive ‘omics approaches. Gersbach still makes special deliveries into cells, but his goal really isn’t to end up with anything extra at all.
“Gene therapy has been around, but it hasn’t worked well because of the limitations of adding genes,” Gersbach says. “What we’d rather do is to correct the mutation, to take the mutated sequence and restore it back to the functional, healthy sequence.”
It isn’t gene therapy so much as it is genome editing.
If you had to pick a single word to describe what sets Gersbach’s approach apart, it would probably have to be “control.” When thinking about something as big and complex as the human genome, control isn’t easy to come by.
One example of the tools in Gersbach’s toolbox is something he and his former colleagues at The Scripps Research Institute call programmable integrases. Built from existing parts fashioned through evolution to bind and manipulate DNA, those integrases are synthetic enzymes that allow for accurate modification of the human genome and its three billion base pairs with remarkable sequence specificity.
In a 2009 report in Proceedings of the National Academy of Sciences, Gersbach and Scripps’ Carlos Barbas III found they could use their integrases to integrate new genes into the human genome with better than 98 percent accuracy. Those synthetic enzymes could also be programmed and reprogrammed to land on particular sites and to carry out particular tasks.
“Once you target a given DNA sequence with these enzymes, you can tag any function you want,” Gersbach says. “You can turn genes on or off, change the DNA structure, or change the gene sequence entirely.”
The accuracy and versatility of such programmable enzymes could make them ideal for a wide variety of applications, from basic biological research to biotechnology. They offer a degree of biological control that the researchers think could prove defining for medicine in the post-genomic era.
Cut and Paste
Gersbach’s lab is focusing particular attention on a version of synthetic enzyme designed to literally and very precisely cut and paste genome sequences.
“We can now change DNA at target sites in any way we want,” Gersbach says. “We can add things, swap things out…so let’s go correct genetic diseases.”
His lab is on the path to doing exactly that in the case of Duchenne muscular dystrophy, one of the most prevalent genetic diseases. The disease traces to mutant copies of a single gene called dystrophin, which is important to muscle fibers’ stability and connection to their surroundings. Without dystrophin, muscles gradually become weaker.
So far, they’ve shown in the lab that they can use their cut and paste technique to correct dystrophin in muscle cells taken from human patients. They are now testing whether those corrected cells will repair and regenerate damaged muscle after they are implanted into mice.
Gersbach says they are sequencing the genomes of those cells to make sure there aren’t any unexpected, off-target effects of the synthetic enzymatic activity. The enzyme they are using could ultimately be delivered to intact human muscle with a virus, just as most gene therapies now are. The good news is that the expression of those corrective enzymes needs to last only a couple of days before it goes away completely.
In traditional gene therapy, that kind of fleeting effect is a problem, but in this case “that could work,” Gersbach says. Two or three days could be all one needs, or even wants, to reprogram the target and make permanent genomic change.
Get With the Program
Gersbach is also applying his synthetic tools to efforts in stem cell biology and regenerative medicine aimed to reprogram one cell type, often skin cells, into other cell types, including muscle, bone, neural or heart cells. Others have shown they can do just this, using cocktails with a small handful of ingredients inspired by developmental biology. But again, the engineer in Gersbach doesn’t leave him satisfied with that solution to the problem.
“You can find the genes responsible for making neurons and add them to adult skin cells to turn them into neurons, and maybe they do,” he says. “But it doesn’t work that well. Maybe they look like neurons, but they aren’t exactly a neuron. They don’t all turn into neurons or they don’t stay neurons. Well, those factors x, y and z are just what Nature handed to you. They evolved to make neurons during development as an embryo, but what you want to do is take skin cells and make a bunch of neurons to inject into a Parkinson’s patient. Maybe x, y and z aren’t the best option there. Maybe we can do better with synthetic engineering.”
It might make sense to start with those developmental factors, but then tweak them in ways that make them more specific. One challenge with transcription factors, proteins whose job is to turn on other genes, is that they tend to interact within cells in various and complicated ways.
“If I want to turn on neuron genes, I want something that does only that – better targeting for a more predictable effect,” Gersbach says.
Last year, Gersbach received a $1.5 million, five-year, New Innovator Award from the National Institutes of Health in support of his efforts to reprogram cells for regenerative medical purposes. The awards are specifically aimed to young investigators with “big picture” ideas intended to inspire “giant leaps” forward in biomedical research.
Back to Basics
Despite the very applied focus of Gersbach’s lab, he recognizes that there is plenty of opportunity for the synthetic tools he develops and uses to shed light on real biology. As one example, Gersbach has struck up a collaboration with the IGSP’s Greg Crawford, who is a player in an international effort called The Encyclopedia of DNA Elements (ENCODE) Consortium. The goal of ENCODE is to build a comprehensive parts list of functional elements in the human genome.
“I’m trying to identify regulatory elements, but there is no good way to determine precisely what it is they are doing,” Crawford says.
That’s where the genome editing technology comes in. Instead of correcting sequences as Gersbach does in the context of gene therapy, he can also strategically insert typos into human genomes at will.
“Charlie can basically remove or edit chunks of the genome and figure out what happens,” Crawford says. “That’s going to be powerful for understanding what different portions of the genome are really doing.”
There is plenty to figure out. Each individual cell type is run by something on the order of 100,000 regulatory elements. It is those elements that make the difference between one cell type and another, even as they carry the very same genome, and Crawford wants to understand exactly how that works. The sheer size of the problem at hand means that Gersbach and Crawford are putting their heads together to come up with ways to make the genome editing technique feasible on a much larger scale.
“This research has a lot of implications for genome biology,” Crawford says. “There just aren’t many ways to manipulate the genome within cells, and Charlie has a way to do this. It’s a really important tool, and I think there are many groups on campus that could find ways to take advantage of it.”
Fellow Duke synthetic biologist and biomedical engineer Lingchong You is considering ways to partner with Gersbach too. He is certainly no stranger to the benefits of synthetic biology for answering basic biological questions. His research often leads him in unexpected directions, from the study of cancer genes and pathways, to synthetic gene circuits that work in similar ways, to discoveries about the mechanisms underlying clinically relevant problems such as antibiotic resistance, and back to synthetic gene circuits again.
“I think this is an underappreciated aspect, even within the field,” You says. “You can get stuck with a kind of linear thinking,” by focusing your attention narrowly on one potential practical application or another.
Biology and Physics faculty member Nick Buchler, who is also a member of the IGSP’s budding group of synthetic biologists, isn’t one who suffers from that kind of linear thinking either. His interest is in the genetic circuits responsible for the roughly 24-hour circadian clock.
Buchler’s work suggests those complex oscillatory behaviors can arise any time duplicate genes produce pairs of nearly identical but antagonistic proteins. Rather than testing his ideas in natural biological systems, he studies them in the laboratory by assembling synthetic circuits from the simplest components.
“By focusing on simple synthetic systems, we might get a glimpse of what the rules are,” Buchler has said. “Patterns could emerge that tell us how these networks might have evolved in nature over millions of years.”
Clearly, synthetic biology approaches and advances can build a better understanding of the natural world. Sometimes, it can be hard to even tell the difference between the two.
“It’s difficult to draw the line between what’s natural and what’s synthetic,” You says. “Once you introduce a gene circuit, it becomes part of a living thing. Technically, to me, there is no difference.”
Param Sidhu and Morgan Howell, both rising seniors at the North Carolina School of Science and Mathematics, are members of the 2012 Duke International Genetically Engineered Machine (iGEM) team under the mentorship of synthetic biologists Charles Gersbach and Nick Buchler. Their goal is to create a kit to enable light-controlled gene expression in yeast. This fall, the team will travel to present their work in competition with dozens of other teams from around the country. The Massachusetts-based iGEM Foundation is dedicated to education and competition, advancement of synthetic biology, and the development of open community and collaboration.