Automated, 3-D, submicron-resolution imaging of C. elegans
David L. Shenkenberg
The tiny, transparent roundworm, Caenorhabditis elegans, has proved a useful research model. Consisting of fewer than 1000 cells, it is relatively simple to study, yet 60 percent to 80 percent of human and C. elegans genes are derived from common evolutionary ancestors. Furthermore, C. elegans research has garnered two Nobel Prizes in physiology and medicine in the past five years, one for studies of organogenesis and programmed cell death and the other for demonstration of the RNA interference mechanism.
The worms can enable studies of other genetic pathways shared with humans, such as those that lead to Alzheimer’s disease and to other neurodegenerative disorders. They also can serve as useful models for preclinical testing of the effectiveness and safety of drugs and of RNA interference, which has emerged as a therapeutic regime since the prizewinning study.
Figure 1. Researchers have developed microfluidic devices that automatically process the roundworm C. elegans, which is shown in this figure.
The immense and growing popularity of C. elegans as a model organism for research and drug development has led to the creation of automatic devices for studying fluorescently labeled worm specimens. These instruments include microplate readers and flow cytometers adapted for use with the organisms. However, neither of the devices can provide images of animals at the cellular level. Instead, microplate readers give measurements of average fluorescence levels, and flow cytometers provide one-dimensional readouts of fluorescence across the axis of the nematode’s body. Neither technique can immobilize individual worms, and motion makes it impossible to acquire fluorescence images at cellular resolution.
A microfluidic solution
To address the limitations of micro-plate readers and flow cytometers, researchers from MIT in Cambridge, Mass., have developed a microfluidic device that automatically captures and immobilizes individual worms for imaging with high-resolution microscopy systems and for subsequently sorting them by fluorescence profile. They have employed the device and a microscopy system to image individual worms in 3-D with subcellular resolution.
For total automation, one can use the instruments with readily available automated microscopes, said principal investigator Mehmet Fatih Yanik. He also pointed out that fluorescence images can show how individual genes affect the function and location of cells, which microplate readers and flow cytometers cannot do because they do not produce images.
Figure 2. This diagram illustrates the microfluidic device that automatically captures and immobilizes the worms for imaging and sorting. Reprinted with permission of PNAS.
The microfluidic device uses suction to capture and immobilize single worms in a chamber (Figure 2). Fluid washes away the other worms, and valves seal the lone worm in the compartment. The system then applies a series of suctions to hold the worm in a straight line for imaging.
Yanik’s group also has developed a second microfluidic device (Figure 3) that directs compounds, such as potential drugs and small-interfering RNA molecules, one by one from standard multiwell plates to the immobilized worms in an array of chambers. In this way, the device facilitates testing the safety and efficacy of potential drugs. The microfluidic devices can be integrated on a single chip in different configurations to perform various assays. These devices are detailed in the Aug. 28 issue of PNAS.
Figure 3. A second microfluidic device relays compounds, such as potential drugs and small-interfering RNA molecules, to immobilized individual worms. Reprinted with permission of PNAS.
For imaging with the worm in the microfluidic device, the researchers employed a Nikon microscope and a Princeton Instruments CCD camera, which Yanik said was used for high-speed imaging at high resolution. The camera enabled image acquisition faster than one frame for every 100 ms, and it was cooled to increase the signal-to-noise ratio. Yanik said that it was important for the camera to have a large CCD chip for large fields of view so that they could view the entire organism, and to have small pixel sizes to achieve high spatial resolution. For the same reasons, they chose Nikon objectives with low magnification — 10× to 20× — yet with high numerical apertures of 0.25 to 0.45. They used wide-field fluorescence excitation with a UV light source. The setup enabled excitation and imaging of the entire organism simultaneously at high speed, unlike point-scanning techniques using lasers.
The researchers produced 3-D images of C. elegans with submicron resolution. They could see subcellular features such as the axons of touch neurons, which are similar to human nerve cells that govern the sense of touch.
Although the investigators have demonstrated proof of principle, they plan to make improvements and to test the devices more extensively, Yanik said. “We want to make them reliable, like a Toyota.” They also want to make them simple to apply, so that users can operate the devices “without a PhD sitting next to them.” He has received the NIH Director’s New Innovator Award — $2.5 million in funding — to further develop the technology.
As an added bonus for companies, such microfluidic chips could be mass-producible and disposable. Yanik said that, because many labs already have automated microscopes, all they must do is buy the chips. He estimates that manufacturing will cost a few hundred dollars apiece at most, at least one order of magnitude cheaper than the competing flow cytometry technology. He has approached some companies regarding licensing opportunities.
Contact: Mehmet Fatih Yanik, MIT, Cambridge, Mass.; e-mail: email@example.com.
MORE FROM PHOTONICS MEDIA