(NOTE: This item originally appeared in today’s daily briefing. It’s been expanded and rewritten here.)
DNA-based diagnostics face a fundamental, though hardly insuperable, obstacle: When you’re looking for a rare mutation or other identifying sequence of DNA “letters” (technically known as bases or nucleotides), there are rarely enough matching DNA molecules in your average blood or tissue sample for today’s technology to detect. So many such tests first require technicians to “amplify” DNA in the sample, usually by a process called polymerase chain reaction, or PCR, which essentially “clones” DNA molecules by the millions — an extremely useful process that is nevertheless time-consuming and which requires expensive, specialized equipment.
One company claiming to have built a better DNA mousetrap is Adnavance Technologies, a Vancouver, Canada, startup with an intriguing nanotech method for detecting small quantities of DNA. Adnavance just raised C$3.7 million ($3.7 million) in a second round of funding, with investors including the Working Opportunity Fund, JovInvestment Management and the Business Development Bank of Canada.
As you can tell from its older logo, reproduced to the left, Adnavance chose its name in part to emphasize its focus on DNA. The company, founded in 2002, is developing tests based on a molecular trick that makes selected strands of DNA conduct electricity, making it possible to detect their presence using a microarray of tiny electrodes.
The basic idea is this: Using specific chemical conditions, Adnavance says it’s possible to force metallic ions between the double strands of DNA, where they displace the protons that usually help hold the helical molecules together. The metallic ions then serve much the same purpose as the central wire in a coaxial cable by freely conducting electricity, giving the DNA molecule as a whole an electrical property that can be measured directly.
Adnavance’s method starts by binding “capture” strands of DNA — that is, stretches that will bind to whatever mutated gene or viral sequence someone is looking for — to a gold electrode and “dehybridizing” them into their single-stranded forms. Once introduced to serum from a blood sample, say, those capture strands stick to matching DNA from the sample and resume their double-stranded shape. The system measures the conductivity of the bound DNA molecules, then introduces the metal ions to the solution, which in theory only integrate into capture probes that have bound perfectly to their targets — thus presumably eliminating mismatches. Measuring the difference in conductivity of the metallized strands compared to their predecessors yields a signal that the company says can detect as few as 500 matching DNA molecules. (By contrast, similar existing tests may require anywhere from 5,000 to 1 million target sequences to yield a detectable signal.)
The company’s first product candidate is a test for antibiotic-resistant staphylococcus, which it believes it can get onto the market by 2010. Adnavance believes its test will be simple enough for use in as many as 30,000 clinical laboratories that aren’t currently licensed to carry out existing DNA-based tests.
Oddly enough, Adnavance’s latest funding round is smaller than its first $3.9 million funding in 2005. That’s a little unusual for a company at this stage, and raises the natural question of whether it’s been forced to take a “down round” with a lower valuation than it previously held. An Adnavance representative, however, says that in 2005, the company actually included two other lines of business — focused on hydrogen fuel cells and DNA vaccines — that were spun out prior to the current funding round. So by that logic, the share of funding devoted to the DNA-diagnostic work has presumably risen. I should note, though, that the company’s lengthy release at the time was mostly devoted to its diagnostic work, and mentions the fuel-cell research in just two sentences at the end (and the vaccine not at all).
VB's research team is studying web-personalization... Chime in here, and we’ll share the results.