In 1993, Jon Kayyem was a postdoc with Thomas Meade, a professor at the California Institute of Technology in Pasadena. Kayyem and Meade had an idea: They decided that they ought to be able to use electrochemistry to detect DNA with a hand-held device. Affymetrix, a biomedical research company?, and its chip technology was just gaining steam; Kayyem and Meade were determined to steer clear of direct competition. "We weren't going to compete with them. Ours would be hand-held, as opposed to a clinical reference lab. That was our niche," recalls Meade, now a professor of chemistry at Northwestern University in Evanston, Illinois.

Kayyem had worked with several researchers in an attempt to find a detection system capable of the task. He tried fluorescent dyes. Although they worked in proteins, the dyes fared abysmally with DNA. His adviser, Caltech biology professor Scott Fraser, sent him to Meade, who suggested that electron transfer in the DNA strand might be interfering with the dyes. That set off a new collaboration, and in 1993, the pair found that double-stranded DNA is a better electron conductor than single-stranded DNA, and that the difference is easily detectable.

SBIR: Perfect Match

This convinced them that they had found the technology that would drive their device. Along with Fraser, they founded a new company, Clinical Micro Sensors, which got its start in Kayyem's garage in Pasadena, California. When they began to search for ways to fund the start-up, they came across an SBIR (Small Business Innovation Research) announcement from the U.S. Department of Energy, which was looking for methods to detect environmental pathogens using portable devices. It was a perfect match, because tailoring the hypothetical device for an environmental pathogen would simply be a matter of using a genetic sequence from the pathogen.

Despite a glaring lack of business acumen, the two succeeded in their bid. Meade attributes it to a combination of good technology and good fortune. The companies that would form the vanguard of the heady late '90s were just getting started, and optimism--and money--was readily available.

They were awarded a phase I grant, followed by a phase II. The development ultimately cost significantly more than the amount of the grants, so the company raised private investments for the rest. The search for private money was made easier by the SBIR grant: "It was evidence that an independent third party thought our approach had merit. It also allowed us to get some early results on our approach.

"We founded this company on the basis of an idea and some basic science done at Caltech, but to demonstrate that we could transform the idea into a sensor required some start-up capital, and the SBIR grant was very useful in that regard," says Kayyem.

Like any research project, the R&D hit a number of snags. In the original SBIR grant proposal, Kayyem and Meade had proposed two alternatives to their technology: one with the DNA embedded along with conducting polymers in a gel, and the other with the DNA linked to molecules called "molecular wires" that would in turn be attached to a gold-plated electrode. Most of the SBIR money went toward investigating the gel-based method, which had been used by others to harbor glucose oxidase in a glucose detector.

Several years of research led them to abandon the gel-based method and move to the molecular-wire approach. It worked "the first time we tried it," Kayyem recalls, with a rueful laugh.

Staying on the Right Path Is Vital

In retrospect, he says, the SBIR grant served a very useful purpose. "It helped us to eliminate an approach that wasn't turning out to be very fruitful, and at the same time it helped us to raise additional money on the approach we did decide to pursue. People refer to those things as their mistakes, but I think it's more accurate to say that they help to reduce choices and align the [company's] direction with the right path."

And staying on the right path is vital, Kayyam says. One potential danger of the SBIR program is that you can end up tailoring your business to the needs of the government agencies putting out SBIR announcements. "I think that's the most important error I saw in the early days of the genomic tools sector. The granting agency says it wants a tool to do X, and you think the market wants a tool that does Y. You [apply for the SBIR anyway and] end up wasting time, and that costs you because the competition is driving steadily toward what the market wants. If it's not what you think the market wants, my advice would be to look for other sources of funding for what you think is the right direction for your business."

It's hard to argue with success. In 1999, Motorola bought the company--it remains a Motorola subsidiary--handsomely rewarding Kayyem, Meade, and Fraser for their insight and hard work.

Jim Kling is a freelance science and medical writer based in Bellingham, Washington.