From gene expression to sequencing, from proteomics to point-of-care diagnostics, life sciences is the first frontier of nanotechnology
By Aaron J. Sender
In March, hundreds of scientists and investors filled the auditorium of the Ronald Reagan Building in Washington, DC, for a day-long discussion on the next small thing: nanotechnology. “Nanoscience has gone from a gleam in the eye to commercial promise in a very, very short time, just a few years,” says National Science Foundation director Rita Colwell, whose agency sponsored the symposium called Small Wonders.
The topic of discussion was breakthroughs in manipulating materials as small as a nanometer — one-billionth of a meter. It would take 100,000 nanometer-sized particles arranged side by side to reach the width of a human hair.
Futurists predict the influence of nanotech in the next century will be at least as profound as the industrial revolution. But many of the first practical applications of nanotech will likely be in biological research, including genomic and proteomic analysis.
That’s firstly because nano is the scale of life: Nanoscale molecular components perform the body’s sophisticated tasks. Proteins, for example, range from two to 15 nanometers. “So nanotechnology has the prospect of doing things on the scale where the action’s really going on,” says UC Berkeley physical chemist Paul Alivisatos.
And second, in the life science research market, producing and selling just a few products that can get a customer to market or publication first can have a lot of value. Even if the science is not fully mature, “you are selling to a sophisticated research audience who can use these materials,” says Joel Martin, cofounder of nanotech company Quantum Dot. On the other hand, applications such as computing and networking, for instance, have a much longer-term payoff because they must be ready for a consumer audience, which requires large-scale production and integration with other parts. “If you’re going to make optical repeaters, you better be able to ship millions of them at penny margins,” says Martin.
Quantum Dot is one of several new companies racing to exploit the special optical and electrical properties of nanoscale materials and build the next generation of smaller, faster, more sensitive tools for gene expression, protein analysis, genotyping, and labeling. Others include Nanosphere, NanoInk, Nanoplex Technologies, and Nanofluidics.
Researchers are just beginning to understand the properties of nanoscale materials — too large to be fully in the domain of quantum mechanics, yet too small to completely obey the laws of classical physics. How exactly stuff this size behaves is a subject of intense study. And without nanobiotech materials on the market yet, it’s still far from clear whether they will offer a significant advantage in the real world.
But nanotech could be the next area in the genomics industry to pique investor interest and capture government dollars. The US government will put $679 million into nanotech in 2003 as part of its National Nanotechnology Initiative led by the NSF — a 17 percent increase from 2002. With Colwell’s background in microbiology, it’s a good bet that a chunk of that will go toward life science. The endeavor has gone international as the European Commission agreed in December to collaborate with the NSF. And investors are beginning to recognize that small tech will be big business. “Efforts have matured to the point where we now think that we can do all of our future new investments in nanotech and have more than enough to do for the rest of our working lives,” says Charles Harris, CEO of venture capital firm Harris & Harris.
Little spheres of influence
Chad Mirkin, tall, slender, and boyishly good looking, stands out among the crowd of scientists exploring the exhibits just outside the auditorium at the NSF symposium. The founder of two nanotechnology companies and director of the Institute for Nanotechnology at Northwestern University, he is one of the most sought-after speakers at the event as attendees vie for a few minutes of his time to chat about possible collaborations. A DOD scout, looking for portable devices for soldiers to detect biological threats in the field, enthusiastically checks out a working prototype of a DNA detection system the size of a large toaster developed by Nanosphere, a company in Northbrook, Ill., that Mirkin founded in 2000. Eventually the instrument will be the size of a Palm Pilot, Mirkin says.
“Let me tell you what Nanosphere’s mantra is,” Mirkin says. “Replace fluorescence, eliminate PCR, and enable point-of-care diagnostics.” Because the company’s long-term goal is to place the diagnostic system in the doctor’s office, it sees getting around PCR as a necessity. “A doctor will never use PCR. It just won’t happen,” says Mirkin.
In one application, 13-nanometer gold particles with oligonucleotides attached to them detect hybridized sequences. If DNA in a sample binds to probes on a glass slide, it brings a gold particle with it. The chips can then be developed like a photograph and read with an ordinary $60 flatbed scanner available at Staples. In fact, the prototype admired by the DOD scout is little more than a souped-up scanner with analysis software included. Current fluorescence-based chips on the market require complex confocal microscopes that cost more than $60,000.
The silver in the photographic developing solution reacts with gold and amplifies the probe signal by as much as 100,000. As a result, the probes can detect a few double-stranded DNA molecules in a sample, which, according to Mirkin, may make expensive and time-consuming PCR amplification a thing of the past. When the chips are run through the scanner, perfect matches between oligo probes and DNA in the sample appear as gray dots. The darker the dot, the more target DNA there is.
Nanosphere is now working on an instrument that, instead of visualizing the nanoparticles to detect the presence of a particular sequence, will use the particles to bridge a gap between two nanoelectrodes and emit an electrical signal.
For some reason not clearly understood by Mirkin or his colleagues, the nanoprobes are also 100,000 times more selective than fluorescence-based detection, says Mirkin. “It turns out that the particles have different hybridization properties than normal DNA.” The DNA denatures over a very narrow temperature range, reducing the chance of false positives. “They actually have much sharper transitions, much tighter binding constants,” says Mirkin.
At first, Northwestern didn’t see any commercial value in patents for tiny gold particles. “The university actually released the patents to us. They didn’t think that we could do this,” says Mirkin. Having no business experience, he crossed the campus to Northwestern’s Kellogg business school and found a small-business expert, whose last name was just a SNP away from his own: Barry Merkin.
Merkin agreed to assign his students to create a business plan for Nanosphere. “He picked the four best people he had. They worked with me for about three to six months and put together a decent business plan,” says Mirkin. “And that led to initial investment.”
First came $250,000 from Applied Biosystems for proof of concept experiments. Now Nanosphere has raised $9.5 million, has a staff of 27, and plans to release the first generation of its DNA detection instrument by early 2003. And Northwestern is back in the game with an equity investment in return for access to all new related technology.
“Nanosphere is focusing on a diagnostic system in every doctor’s office in the world, and eventually one in every medicine cabinet,” says Mirkin.
Earlier this year, NanoInk, another company based on technology out of Mirkin’s lab, raised $3.2 million in initial VC investment. “We can put an entire gene chip in the area that makes up a single spot in a conventional microarray,” he says. Using a technique called dip pen nanolithography, NanoInk can print DNA, proteins, or anything else for that matter, on a scale of 15 nanometers. To put that in perspective, the features on Affymetrix’s state-of-the-art chips are 18 microns, or 18,000 nanometers. “It’s the world’s smallest spotter,” says Mirkin.
Dots Don’t Fade Away
Instead of using tiny gold particles to detect DNA and proteins, Quantum Dot exploits a quantum mechanical property of nano-sized semi-conductors: as the material changes size, it changes colors. The molecular structures of conventional organic fluorescent dyes determine their color. To make a new color requires a completely new molecule.
Victorian-era glass and paint makers were making quantum dots, nanocrystals of zinc and cadmium sulfides and selenides, back in the 19th century without knowing it. “When people started using cadmium-based paints in the 1880s, the color was different every time they made paint,” says Moungi Bawendi, a physical chemist at MIT. “What they were making was very small particles of semiconductors. And the color change was because they were making different size particles.” The makers of Tiffany lamps didn’t know that the signature colors were emitted by nanoparticles either.
In the early 1980s Louis Brus — then at Bell Labs, now at Columbia University — experimented with nanocrystal semiconductors in solution, “and every time he did a prep he would get a different color,” say Bawendi. Two of Brus’s postdocs, Bawendi and Alivisatos, continued working on quantum dots in their own labs at MIT and UC Berkeley, respectively. They later founded Quantum Dot with Martin.
“With the quantum dots we can manufacture colors to our specifications. We can make quantum dots with a different color every two nanometers along the visual spectrum,” says Martin, who recently resigned as the company’s CEO to join Forward Ventures. (“I’m a startup guy,” he says.)
For example, at 2.3 nanometers the particles give off blue light, at 4.2 nanometers they are green, at 4.8 they emit yellow light, and at 5.5 they fluoresce red.
And unlike fluorescent dyes, quantum dots don’t fade with time. “They are permanent. They don’t photobleach,” says Martin.
The wide range of colors allows for dozens of DNA or protein samples to be studied at once. Quantum Dot further expands the number of possible parallel assays by inserting different color combinations into beads, creating distinct spectral bar codes.
Qbeads, as the company calls them, act as fluid microarrays. Because the beads are dissolved in solution and have a greater surface area than planar arrays, the probes hybridize target DNA in as fast as five minutes, instead of overnight. “You can just pipette your array from one place to the next with robots or with hand-held pipetters,” says VP of business development Andy Watson, previously a cofounder of the Sanger Centre. There are also significant manufacturing advantages. “When we make a single bead type, we make enough for more than a million different assays,” says Watson. “And the data quality one receives from our bead system is typically much better than a microarray, because we have at least 20 different replicates of essentially the same bead. So you get much more confidence in the data that’s produced.” And because the beads are randomly distributed in solution, there are “no spatial effects caused by the particular bead being in a particular place,” he says.
GlaxoSmithKline tested the Qbead platform for genotyping last year. “We were given a set of randomized samples that GSK had sent out to be genotyped,” says Martin. “We were able to hit a 100 percent accuracy on the genotypes.” The test did not turn into a deal. Martin claims that’s only because the company decided not to pursue that market. “We’re keeping that on the back burner as an application of the Qbead technology,” he says. “Although genotyping will undoubtedly be important in the future, it has a relatively small market today.”
Meanwhile Quantum Dot is looking to hook up with an instrument maker to develop and market hardware in which to deliver the system. “We plan to announce within in the next few months deals for commercialization of the bead analysis technology,” says Watson. The 50-employee company, based in Hayward, Calif., will focus its internal efforts on generating its own drug discovery IP in collaborations with pharmaceutical companies, instead of marketing technology. Quantum Dot has raised $37.5 million so far. “We have the majority of that left,” says Watson.
Inspired by autumn chores
SurroMed, a Mountain View, Calif., biomarker discovery company, will also use a nanoscale barcode strategy to multiplex biological assays in its spinoff Nanoplex Technologies. “The world doesn’t know about this yet,” says nanobarcode inventor and SurroMed CTO Michael Natan. At press time he expected the first round of funding to close late last month.
“I essentially came up with the invention while raking leaves,” says Natan, then a Penn State researcher. Perhaps the various bright colors of the foliage inspired him. But instead of varying sizes to create different colors, the way Quantum Dot does, Nanoplex’s nanobarcodes are 30 nanometers and cylindrically shaped with alternating stripes of different metals including silver, gold, and platinum. Under an ordinary microscope the particles appear as a barcode,
light and dark bands of varying widths. “Instead of reading by difference in contrast the way you read a barcode, say, on a can of Coke, we read by the difference in reflectivity of adjacent lines,” says Natan.
Nanoplex expects to raise $12.5 million in its first round, and SurroMed will incubate it and retain 50 percent of the company. “We’re already oversubscribed,” says Natan. Applications he is considering include multiplexed surface plasmon resonance for protein-protein interaction experiments, genome-wide SNP scoring, and gene expression. “What if you wanted to look at five different samples on the spot on a GeneChip? How would you do that? You can’t,” he says. “This will allow you to get five, 10, maybe even 20 colors at once.”
Aside from greater flexibility and faster hybridization than planar arrays, nanobarcode particles also have the advantage of scalability. “It’s just as easy to make a large bucket of particles as a small bucket of particles,” says Natan.
Shrinking fluidics, supershrunk sequencers
Another area where nanotech is finding its way into commercial genomics is taking microfluidics down to a new level: nanofluidics. For example, Stephen Turner is not yet shopping around for funding to commercialize such a technology he developed with Cornell University’s Harold Craighead when he was a grad student in his lab. Instead, his company, Nanofluidics, is still exploring the market for just the right applications. “Gene expression is an area we are contemplating,” he says. “But we’re reluctant to get into it because there are so many companies already taking a serious attack on that market,” says Turner.
Because nanofluidic channels are the same scale as DNA, there is no wiggle room for even a single strand. The DNA has to bump up against the wall surface where all kinds of optical and electronic measurements can be made. Turner has already published data showing that his nanofluidic technology can separate long strands of DNA faster and more efficiently than current electrophoresis techniques. “Electrophoresis is the closest competitor and it’s quite slow,” says Turner. “It can take dozens of hours or even days.” Nanofluidics can size the fragments in minutes.
Nanotech also promises to alter the way DNA is sequenced. Craighead’s lab is also working on techniques for single- molecule nanosequencing as the DNA strands zip past an optical reader.
Agilent is also getting in on the nanosequencing act. It is codeveloping a technology with Harvard’s Daniel Branton that would sequence a long strand of DNA by shuttling it through a 10-9 meter hole at a rate of a million bases per second. Branton says the nanopore sequencer, which would sequence an entire genome in less than two hours, may be ready in as soon as two years.
What’s more exciting about nanotechnology than what’s been done already is what’s yet to come. At the NSF symposium, it is obvious that the field is still in its infancy, perhaps at the stage where DNA research was in the first half of last century. For instance, Angela Belcher of University of Texas, Austin, spoke of her work screening semiconductors against libraries of proteins to find proteins with high affinity, perhaps one day to fuse them into organic-inorganic circuits. “It’s hard to say what the range of application could be,” says UC Berkeley’s Alivisatos. “But it just seems like such an intriguing idea. It just seems like it’s going to turn out to be important somehow.”
Though most breakthroughs presented at the meeting are in basic science with no apparent immediate applications, the tone is, “Wow, something important is happening here. Let’s see how it plays out.”
As Mirkin says, “It’s a field that’s screaming for new developments.”