Churning out highly repetitive polypeptides is like trekking across a featureless plain—the dearth of landmarks makes it easy for would-be synthesizers to lose their way. To help synthesizers arrive at their desired destinations, Duke University scientists are assuming the role of pathfinders. More precisely, the Duke scientists have created a computer program that shows how repetitive polypeptides can be encoded by nonrepetitive DNA. Essentially, the program enlivens the DNA landscape, helping synthetic tools, such as the polymerase chain reaction (PCR), stay on track.
“Synthesizing and working with highly repetitive polypeptides is a very challenging and tedious process, which has long been a barrier to entering the field [of synthetic biomaterials],” said Ashutosh Chilkoti, Ph.D., the chair of the biomedical engineering department at Duke. “But with the help of our new tool, what used to take researchers months of work can now be ordered online by anyone for about $100 and the genes received in a few weeks, making repetitive polypeptides much easier to study.”
The new tool mentioned by Dr. Chilkoti was unveiled January 4 in the journal Nature Materials, in an article entitled, “Combinatorial codon scrambling enables scalable gene synthesis and amplification of repetitive proteins.” According to this article, the tool is a codon-scrambling algorithm. The algorithm enables the PCR-based gene synthesis of repetitive proteins by exploiting the codon redundancy of amino acids. Basically, the algorithm finds the least-repetitive synonymous gene sequence.
Nature has 61 codons that code for 20 amino acids, meaning there are multiple codons that yield a given amino acid. Because synthetic biologists can get the same amino acid from multiple codons, they can avoid troublesome DNA repeats by swapping in different codons that achieve the same effect. The challenge is finding the least repetitive genetic code that still makes the desired polypeptide or protein.
According to the Duke scientists, this challenge resembles a well-known problem in combinatorial optimization—the traveling salesman problem. The classic question is, given a map with a set of cities to visit, what is the shortest route possible that hits every city exactly once before returning to the original city?
The scientists reported that they addressed this problem in the polypeptide synthesis–context via De Bruijn graphs and a modern mixed integer linear program solver. “As experimental proof of the utility of this approach,” the authors of the Nature Materials article wrote, “we use it to optimize the synthetic genes for 19 repetitive proteins, and show that the gene fragments are amenable to PCR-based gene assembly and recombinant expression.”
This part of the Duke scientists’ work was carried out by Nicholas Tang, a doctoral candidate in Dr. Chilkoti’s laboratory. Tang created a laundry list of 19 popular repetitive polypeptides that are currently being studied in laboratories around the world. After passing the codes through the program, he sent them for synthesis by commercial biotechnology outfits—a task that would be impossible for any one of the original codes.
When Tang received his DNA samples, each was introduced into a living cell to produce the desired polypeptide as hoped. “He made 19 different polymers from the field in one shot,” stated Dr. Chilkoti. “What probably took tens of researchers years to create, he was able to reproduce in a single paper in a matter of weeks.”
Dr. Chilkoti and Tang are now working to make the new computer program available online for anybody to use through a simple web form, opening a new area of synthetic biology for all to explore.
“This advance really democratizes the field of synthetic biology and levels the playing field,” concluded Tang. “Before, you had to have a lot of expertise and patience to work with repetitive sequences, but now anyone can just order them online. We think this could really break open the bottleneck that has held the field back and hopefully recruit more people into the field.”