Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
All sciences are constrained by certain immutable laws. Thou shalt not travel faster than light. Noble gases shall not covet their neighbor’s electrons. For biologists, the law might read: Two points may not be resolved if they are closer together than half the wavelength of light used to view them. Less catchy, perhaps, but just as unbreakable. Or so we thought.
Researchers have now devised various approaches to circumvent the so-called “diffraction limit” of light, the aforementioned law whose practical meaning was that biologists could never distinguish two objects closer than about 200 nm apart—at least not with a light microscope. (They could, of course, with electron microscopes). The resulting super-resolution microscopy methods have fundamentally transformed researchers’ ability to probe the inner lives of the cell and earned their inventors the 2014 Nobel Prize in Chemistry.
“The diffraction limit of the microscope is right where things start getting interesting in biology—the width of a condensed chromosome, the width of a mitochondrion, the complex of proteins at the [cell] surface, the size of a caveolin or clathrin pit, bundles of microtubules or actin, the size of an endocytic vesicle. Those sorts of structures are just beyond our reach with a diffraction-limited microscope, and now with super-resolution [they] come into view,” explains Paul Goodwin, science director for cellular imaging and analysis tools at GE Healthcare Life Sciences, one of the companies that has commercialized super-resolution systems.
Here, we review the current state of the super-resolution microscopy market.
A super-resolution primer
Super-resolution methods come basically in three flavors. PALM (photoactivated localization microscopy), STORM (stochastic optical reconstruction microscopy) and GSDIM (ground-state depletion microscopy followed by individual molecule return, also called direct or dSTORM) are localization-based methods: They create images by imaging a few sparse fluorophores, turning them off, activating a new set of fluors and repeating. The methods differ in the types of fluorescent molecules they use and the mechanism for activating and deactivating them, but in all cases the final image is a composite of many hundreds or thousands of images collected over extended periods.
Localization methods offer lateral resolutions about 10 times better than conventional microscopy, down to about 20 nm. But because so many frames must be collected, the imaging itself is slow, complicating live-cell applications.
STED (stimulated emission depletion) microscopy uses two lasers that raster over a sample, as in laser-scanning confocal microscopy. One laser activates fluorophores in a diffraction-limited volume, and the second laser deactivates the fluors at the periphery of the spot, thereby shrinking the activated area and enhancing resolution to about 30 nm.
SIM (structured illumination) microscopy takes advantage of Moiré interference. If you look through a fine screen, as in a screen door, you’ll notice the screen seems to disappear, and what is beyond it is clear. But, look through two different screens, and they’ll be clearly visible. The principle of SIM is that by projecting patterned light onto a sample and reading the resulting signal, it’s possible to back-calculate the interference pattern that produced it—that is, the biological sample itself.
“If you know what one pattern is, you can derive what the other one is, i.e., the distribution of fluorophores in the sample,” Goodwin explains.
The result, at least for GE Healthcare’s DeltaVision OMX SIM-based instruments, is “a little better than twice the lateral resolution of a conventional fluorescence microscope,” Goodwin says.
Leica Microsystems
Leica Microsystems’ super-resolution portfolio includes both GSDIM and STED systems. According to product manager Peter Laskey, the GSDIM system was upgraded to provide 3D imaging in 2013 and is slated to receive a scientific CMOS (sCMOS) camera upgrade this July.
“sCMOS cameras offer the ability to acquire data dramatically faster, significantly increasing system throughput,” Laskey says. “Localization microscopy can be time-consuming. Now, instead of waiting anywhere from three to 20 minutes per image, you can get a good image in less than a minute, and in some cases even sub-10 seconds.”
Among other features, the Leica SR GSD 3D system includes a special drift-suppressing SuMo stage, which decouples the objective lens from the microscope nosepiece attaching it to the stage. When users collect raw data over several minutes using a conventional microscope and stage, the objective will move relative to the sample, sometimes by many tens of nanometers, Laskey says. “The SuMo stage can reduce that drift below 20 nm, so the drift is below the resolution of the system.”
The Leica TCS SP8 STED 3X has also been updated recently, with the ability to apply STED in three dimensions, use a wider assortment of fluorophores and better tune the point spread function. An upright microscope stage will be released “very soon,” according to application manager Jan Schröder.
Nikon Instruments
Nikon Instruments offers two super-resolution systems, N-SIM and N-STORM. According to Christopher O’Connell, senior product manager, advanced biosystems, researchers traditionally use the former for live-cell applications and the latter for “detailed ultrastructural studies,” as STORM imaging typically has been slower than SIM. A soon-to-be-released product update could change that equation, O’Connell says.
Slated for release this summer, N-STORM 4.0 will feature a brighter laser and new Hamamatsu sCMOS camera. “Purely from a camera frame-rate standpoint, [the sCMOS] is five times faster” than the existing EMCCD (electron-multiplying charge coupled device), O’Connell says.
According to O’Connell, N-STORM users can collect approximately1-mm-thick, three-dimensional datasets by using an optical trick. Unlike the N-SIM, the N-STORM collects only a single plane of data. But by employing a special lens, the system can determine a fluorophore’s axial position by measuring its ellipticity. “We can basically map the shape of all molecules in a frame to extract the z coordinate,” he explains. (Other localization-based systems use similar strategies to acquire 3D datasets. Leica’s SR GSD 3D, for instance, uses an astigmatic lens to induce ellipticity, according to Laskey.)
Carl Zeiss Microscopy
Zeiss offers a single super-resolution platform, the ELYRA, which is available in three configurations: PALM (ELYRA P.1), SIM (ELYRA S.1) and dual (ELYRA PS.1). However, the company has most recently developed a new detector module for its laser scanning confocal instruments that also technically breaks the diffraction barrier.
The so-called Airyscan detector offers 140-nm lateral (x-y) resolution and 400-nm resolution in the axial direction, says Joseph Huff, product marketing manager for laser scanning and super-resolution microscopy. “It gives you five-fold the spatial information compared to a traditional confocal.”
As Huff explains, confocal microscopes use pinhole apertures to block out-of-focus light. The smaller the pinhole, the better the resolution. Yet at the same time, signal suffers. Typically, this signal is detected with a point detector, such as a photomultiplier tube. The Airyscan detector uses 32 detectors arranged “in a compound-eye fashion” to “oversample” the data, says Huff. As a result, it yields “the efficiency of a 1.25 Airy unit pinhole with the resolution of 0.2 Airy units.” (An Airy unit refers to the appearance of a fluorescent point when imaged in a microscope: a bright point surrounded by concentric rings. Each ring is one Airy unit.)
“In the context of live-cell imaging, this is a huge benefit,” Huff says, providing faster imaging with lower illumination. “You get a temporal resolution increase as well as the ability to keep the laser exposure down, which makes cells even happier.”
According to Huff, the Airyscan is included in the new Zeiss LSM 800 and 880 confocals; users of the earlier LSM 710 and 780 systems can upgrade for about $175,000.
GE Healthcare Life Sciences
GE Healthcare Life Sciences acquired instrumentation vendor Applied Precision in 2011. Applied Precision sells 3D SIM systems under its DeltaVision brand, including the DeltaVision OMX and new, more compact DeltaVision OMX SR, set to launch at the end of June.
According to Goodwin, DeltaVision systems are true 3D SIM instruments. Rather than collecting a series of two-dimensional images and combining them to create a 3D dataset, “the problem is solved as a true 3D solution.”
The OMX SR lets users image in either 2D or 3D mode; the former, Goodwin says, “is exceptionally fast.” Using total internal reflection illumination, that 2D mode “generates a structured illumination field, but only in a TIRF volume.” This accelerates imaging, because only nine frames per image are required, instead of the usual 15.
According to Goodwin, microscopy balances four key parameters: resolution, speed, viability and depth. These may be thought of as the vertices of a tetrahedron, he explains, and they are perpetually in tension. “To go fast at high resolution and depth, you need a tremendous amount of light, which affects viability. To maintain viability, you cannot have speed and resolution and depth at the same time.”
Structured illumination, Goodwin continues, represents one possible compromise to that problem. But other compromises are possible, too. Users may struggle to identify the best system for their needs, but this much, at least, is clear: Researchers need no longer compromise on the diffraction limit itself. In microscopy, that law no longer applies.