NEW YORK, March 22 (GenomeWeb News) - Call it reality TV for the cell-based assay crowd. Amnis, a Seattle startup, has developed a system that it says can image thousands of living cells in real time, while they are moving and interacting with one another, and meanwhile takes hundreds of measurements of the cells for later analysis.
"There is nobody else in the world that can image cells in flow," said Jack Ball, a 30-year biotech veteran who joined the company as CEO in February. "We take about 200 measurements per cell at 100 cells per second," he said. These measurements include parameters such as the size of the cells or the nucleus size, or dividing vs. not dividing cells.
The company announced last week an early access installation of the system, the ImageStream 100, at the University of Washington's Cell Systems Initiative-a joint project of the UW school of Medicine and the school of engineering, which focuses on cell-based systems biology.
The device, which is being donated to the CSI, is actually the fifth beta instrument to be installed at the CSI by Amnis, as part of a four-year collaboration between the two parties, according to Robert Franza, director of the CSI and a professor of Bioengineering at UW. "The instrument is a crucial part of how we are studying molecular events in living cells, particularly studies in which cells are coming into contact with each other," Franza said. "It allows us to do substantial quantifications of numerous parameters, and have images of cells in a high-throughput mode." (Click here for pictures of the system and the CSI's description of how it's using the device)
The ImageStream system combines CCD-camera based microscopy and a flow cytometer, according to Amnis' website. To illuminate the cell, it has both UV and 488 nm lasers, which can be aimed at the cells in flow, while a brightfield illuminator is aimed at them from a perpendicular angle. The image is then focused and goes through a spectral decomposition element, which can generate up to six different images of the cell in different colors as well as bright field and dark field. (See a diagram here.)
To track the motion of cells, the device uses a technology the company calls time-delay integration, which electronically pans the detector. TDI "is the critical core patent that gives us the ability to [collect the multiple images of the cell], Ball said. "We collect images pixel by pixel, as the cell passes for up to 500 pixels. [Then] we end up getting the sum of 500 pixels of image that effectively stops the motion of the cell."
These images are stored, although the users can decide to throw them away if they are not in focus. A common criticism of these new-generation imaging systems has been the volume of the data they generate, and Ball admitted that the ImageStream "does generate some pretty large data files -- 50 to 60 mb per experiment." But he dismissed this as a non-issue, given the low cost of data storage these days.
On the informatics end, users can create a scatter plot using the data points representing any of the 200 parameters that the system measures. "One parameter might be the size of the nucleus related to the size of the cell. Another might be color in a certain channel vs. color in another channel in the nucleus," Ball said. "What makes the software unique is that you can actually see the cells." Rather than just looking at data points, the user can click on them to see images of the cells and visually verify the classification that the system has performed, Ball said.
The system can do fluorescence in situ hybridization assays in combination with its imaging, according to Ball. This feature enables users to determine, for example, what's happening in the nucleus of a living cell in suspension, he said. These capabilities to image cells and derive so many measurements at the molecular level, Ball said, "will drive a new paradigm in systems biology."
Still On the Ball
Ball came to Amnis from Molecular Probes -- which Invitrogen acquired last August -- where he was chief commercial officer. Before that, he was senior vice president at Orchid Biosciences, and held numerous positions during 13 years at Amersham Pharmacia Biotech.
When asked why he chose to take the CEO position at Amnis, replacing the company's founder and acting CEO, Harold Kawaguchi, Ball said he "was excited by the stage the company is at." Instead of the usual startup with a new idea, Amnis is "a company that has a product that is in early release, early access phase, and will soon be in the market," he said.
In addition to the UW, Amnis currently has one commercial early access customer for its ImageStream: an unnamed "major Southern California biotech," according to Ball. The company will formally launch the system in the fourth quarter. It will cost $250,000, he said.
While Ball said the company does not have any direct competition, he acknowledged that a form of competition might come from manufacturers of flow cytometers and automated microscopes, as well as Cellomics, with its high-content screening system or Amersham (which makes the InCell Analyzer).
But he said the ImageStream is different than a high-content screening system, in that it does not image cells on a 96-or 384-well plate, and "gives you much more information about what goes on with a cell population." Additionally, while high-content screening requires that the cells be able to grow on a plate, limiting screening to adherent cell lines, the ImageStream can image non-adherent cells such as immune cells and certain stem cells, Ball said.
He also sounded the familiar theme of complementarity rather than competitiveness: the imager, he said, could be used as a follow up to high-content screening, to determine the mechanism of action of a compound in cells, for example.
The challenge for Amnis will be to convince core labs in the private sector that they need this type of ultra-high-content screening device: Ball said he is currently working to build a sales and marketing team to do just that.
Additionally, he said, the company is looking to raise another $10 million in the near future, in addition to the $13 million it has raised so far, to fund these commercialization efforts, and grow from the current headcount of 26 to 35, or more, employees.
Meanwhile, Franza and his team at CSI will continue to tinker with the current system. With the ImageStream 100, he said, "we are able to do things that we were not able to do before." But the system could be even more sensitive or have a larger throughput, and work in different environments, he said. "Scientists, typically, are never satisfied."