Kevin Jarrell has been a gene engineer ever since his days as an undergraduate at Ohio State University. But he never did think much of the laborious manner through which insertion, deletion, gene modification, and recombination were accomplished. “It’s more like craftwork than a high-throughput, automated engineering system,” he says of the technologies that haven’t changed much since they were invented in the late ’70s and early ’80s.
So he set out to automate these processes and usher them into the high-throughput world of genomics. Jarrell had been working on improving gene engineering methods through his postdoc at Harvard and later at Boston University. His idea was to change the basic reactions necessary for these engineering processes, leading to reactions that were more amenable to automation.
The main problem Jarrell faced was that DNA generated by PCR “is blunt ended,” he says. “It doesn’t have the sticky-ended tails on it that gene engineers like to build things.” His initial step was to build molecules with RNA tails that could be joined together with ligase. Eventually, he moved on to using methyl residues to replace the RNA and act as a terminator. “That gives DNA strands with tails of known length and sequence,” he explains. “You can make recombinant DNA molecules using ligation and cloning.” That basic concept can also be used for insertion, deletion, and site-directed mutagenesis, he says.
Armed with the ideas, Jarrell finally left Boston University to start up a company and work on gene engineering full-time. With funding from angel investors, he and Temple Smith co-founded Modular Genetics, based in Woburn, Mass., in June of 2000 and the company really got going by early 2002, he says. Because his research spanned so many years, Jarrell says, patents on the work are owned by both Harvard and Boston University.
So far, Modular Genetics is making progress. “We’ve shown that we can build genes from scratch greater than 2 KB in length using this system,” he says. The company has worked with customers on projects involving site-directed mutagenesis to repair as many as 1,000 genes in a library, Jarrell adds. Clients come from all areas, including government labs, academia, and big pharma. Because small projects tend to be significantly more expensive on a per-gene basis, the 10-employee company actively seeks out customers in need of high capacity and tends to avoid small-volume researchers.
It’s too early to say if Jarrell’s gamble will pay off, but he’s grown more optimistic over the years. Thanks to an NIH initiative “to clone at least one copy of each human gene” as well as more demand for altered genes in general, he says, “there’s a big appetite for both assembled genes and the creation of particular variants.”
— Meredith Salisbury