In pioneering practical breakthroughs in microscopy, developers find strength in the combination of several technologies in a single instrument.
Advances in microscopy and fundamental science are closely intertwined. Without prior understanding of the basis for research, the tools of microscopy are useless. Without microscopy, an understanding of how materials, chemistry, or life behave(s) at the molecular and atomic level cannot be discovered.
Until recently, optical imaging at a scale where individual cells, molecules, or proteins interact was unachievable. Now, with super-resolution techniques, true nanoscale study is possible. Electron microscopy has also improved, but a significant shift in this market has also been accessibility. Once confined to only the most well-funded laboratories and corporations, tens of thousands of scanning electron microscopes (SEMs) and transmission electron microscopes (TEMs) are now used globally.
Microscopy methods are proliferating, and microscope developers have been compelled to respond with more flexible and efficient products. Imaging platforms that were once standalone—inverted optical, widefield, scanning electron—have been co-opted to create hybrid instruments that boast features like dual-beam imaging, correlative light microscopy, and super-resolution.
Demand for microscopy tools has also created another trend in the marketplace: High-end technologies are now available to a greater number of researchers. At $4.3 billion in 2012, the global microscopy market is expected to experience double-digit growth percentages in the next five years. With the help of software engineers and embedded controls, and the support network offered by vendors themselves, microscopy companies have reached new customer segments, helping to grow the entire industry.
Super-resolution accelerates life science
One of the most compelling techniques used in optical microscopy is single-molecule localization. Localization can be accomplished in several different ways, as vendors have shown in their commercial products; but the ultimate goal is to isolate single fluorophores and determine their location. At the molecular scale, this cannot be accomplished by conventional light microscopy, which is limited to about 250-nm resolution in the x and y direction, and up to 700 nm in the z direction. In fluorescence imaging, all fluoresced molecules light simultaneously when illuminated with excitation. When the number of fluorescent molecules is high, or when multiple fluorophores are used in a sample, molecular interactions in an ensemble of molecules can only be inferred when closer together than the diffraction limit.
Developers have taken several approaches to solving this problem, and Leica Microsystems, Nikon Instruments, and Carl Zeiss MicroImaging have all developed commercially available systems that employ localization routines. For several years, Nikon has showcased its STochastic Optical Reconstruction Microscopy (STORM) technique in its N-STORM super-resolution instrument, which uses specifically engineered fluorescent markers tracked by way of sequentially activated molecules. The PhotoActivated Localization Microscopy (PALM) method developed by Carl Zeiss for its Elyra.P1 microscope functions in a similar way.
More recently, Leica, Wetzlar, Germany, has pioneered an alternate approach that relies less on special markers than on the ability of laser light to control whether a molecule fluoresces or does not. Using a specially designed laser module that employs four high-power diodes, Leica’s SR GSD super-resolution microscope forces all fluorescent molecules in a specimen to switch off for most of the time. At undetermined points of time, individual molecules spontaneously return to the fluorescent state, while others remain dark. This allows the signals of the illuminated molecules to be read and accumulated sequentially by a sensitive and fast camera system. Spatial position of the detected molecules in the specimen can be measured and stored.
Known as the ground state followed by individual molecule return (GSDIM) method, this theory was developed at the laboratory of Professor Stefan Hell of the Max Planck Institute in Goettingen, Germany, and implemented on Leica’s Stimulated Emission Depletion (STED) platform. In 2005, Leica introduced the first super-resolution microscope, the 4Pi, which, for the first time, offered a major improvement in axial resolution and set the stage for STED. Based on the laser scanning method used in confocal imaging, STED yielded further improvements in axial resolution. It ultimately achieved resolutions down to 70 nm with the help of powerful lasers and one of the keys to effective super-resolution: a total internal reflectance fluorescence (TIRF) module that uses four rapidly switching, fully automated lasers operating at four separate wavelengths—405, 488, 561, and 635 nm.
Leica has already marketed a super-resolution microscope based on confocal imaging (the TCS SP8). The SR GSD debuted in 2012 as the first microscope to utilize GSDIM. Working with Hell and his laboratory, Leica’s engineers adapted the existing total internal reflectance fluorescence module used in the TCS STED microscope, with the DMI6000 B inverted research microscope. The high sensitivity requirements of the GSDIM function required the development of a new Suppressed Motion Stage (SuMo) that uses software and electronics to compensate for minute thermal changes in the piezoelectric actuators. Finally, the large data stack created by the instrument necessitated the use of a graphic processing unit embedded in the system.
“The entire procedure of taking images can take up to 20 minutes to process them all and create a single result. Our priority was to create a system stable enough to do this for 20 minutes,” says Sebastian Tille, director of widefield imaging, Life Science Division, Leica Microsystems.
In the SR GSD, GSDIM permits localization of single molecules down to 20 nm, allowing single cell, molecule, and protein images in 3D. As a result of these capabilities, the SR GSD was presented with an R&D 100 Award in 2012. A super-resolution microscope that delivers imaging results quickly and uses standard fluorescent preparation techniques has been attractive to researchers, Tille reports.
“Some customers, when they compare their results from the SR GSD with others, comment on the ease of staining samples with the standard fluorophore. This compatibility with standard fluorophores allows them to start with a standard preparation and optimize their tools for use with super-resolution,” says Tille.
Already, he says, Leica has received word that researchers are developing new fluorophores and dyes to apply to the GSDIM capability. Several journal publications are in progress, Tille reports. New chemistries will allow two- or even three-color imaging to improve resolution and offer the ability to monitor multiple proteins.
Importance of TEM grows
In the February 2013 issue of R&D Magazine, editors spoke with two leading systems biologists, Richard D. Smith and Leroy Hood. Both cited the need for greater throughput in proteomics studies, greater understanding of single-cell and macromolecular complexes, and the need to improve 3D imaging tools. Mass spectrometry is a crucial tool for identifying areas of study interest, but structural information is also desirable.
To achieve the best understanding of protein structure, researchers often look to nuclear magnetic resonance (NMR) and X-ray diffraction (XRD) techniques. High-powered X-rays have led to a host of fundamental discoveries in biology.
However, the high-resolution structures characterized by XRD and NMR are typically of a single protein or a small section of a larger molecular complex. This limitation can slow laboratory research, particularly when a higher-order protein complex is identified as an area of interest for fundamental biology or development of drug-related therapies.
Increasingly utilized in this area of research, cryo-electron transmission microscopy is able to image—in 3D mode—the structure of a complex that interacts with other complexes or nucleic acids. Though researchers sacrifice resolution as compared to XRD or NMR, a more complete picture of the molecular complex can be built. “With NMR and XRD, you can get resolutions that exceed TEM, but they're limited by size. In a cell, proteins act in complexes and work together. To answer the true biological questions, researchers need to understand the real 3D, and sometimes 4D, structures of multiprotein complexes,” says Gerard Geilen, senior global marketing programs manager at FEI Co., Hillsboro, Ore.
Aware of this direction in research, developers at FEI launched a new TEM earlier this year. Building on several existing technologies, the Tecnai Arctica embodies several qualities that TEMs, particularly cryo-TEMs, have rarely been associated with previously: efficiency, throughput, and automation.
According to Geilen, cryo-TEM is an essential technology for two types of imaging. The first is cryo-tomography, which relies on the nanoscale imaging capabilities of TEM to build an accurate 3D map of a cell's interior. So far, TEM is the only available technology capable of forming these detailed, multidimensional images. The second role of cryo-TEM is to conduct single particle analysis (SPA). This method combines tens or hundreds of thousands of nominally identical particle images to derive a high-resolution, low-noise 3D model.
“The structures being studied are very tiny. To build a picture of what's happening you need thousands of images of individual particles. These studies are not just for the sake of finding protein structures, although that is often the case. Often, structure leads to discoveries for drugs. You can identify the consequences of what drugs attach to what structures,” says Geilen. Sometimes the user wants to disrupt the function of the protein; sometimes proteins should be activated.
SPA was born from efforts to improve the types of information available from raw TEM images, which were often compromised by excessive noise. Experiments that combine images of similar symmetries have produced a clearer, higher-resolution image. SPA can provide 3D images of around 3 A, depending on particle symmetry and stability. For cryo-tomography the best obtainable resolution lies around at 3 nm. The Tecnai Arctica boosts the high throughput and high quality output of data, improving 3D resolution.
More significantly for FEI, however, is the effort made in helping researchers accelerate SPA routines. And around the Tecnai Arctica developers have built an automated workflow that includes cryo-sample preparation, data acquisition, analysis, and final visualization.
This extends to vitrification, in which the samples are cooled so rapidly that water molecules do not have time to crystallize, causing damage. FEI accomplishes this with its Vitrobot tool, which keeps physical and mechanical conditions—temperature, humidity, freezing velocity—at constant states during the cryo-fixation process.
The cryo-sample autoloader, combined with automated target identification and low-dose imaging, enable unattended acquisition of large SPA data sets that are practically impossible to acquire with manual methods, says Geilen.
Geilen anticipates the Tecnai Arctica’s use in conjunction with NMR and XRD in many instances. Reconstructions obtained with TEM can be interpreted at a higher resolution by fitting X-ray and NMR models into the electronic microscopy density map, providing a 3D model of the entire macromolecular complex. The capability extends to a fourth dimension by providing detailed information about the movements and dynamics of these complexes over time. FEI has also developed software (Argos) to combine cellular context data from cryo-tomography analyses with SPA results.
SEMs test limits, broaden impacts
The microchip industry is famous for successfully adhering to the law of Intel’s Gordon Moore, who in 1975 revised his famous graph to predict the number of transistors on a given chip would double every two years.
But this effort is getting increasingly difficult as scientists encounter new barriers at the molecular scale, where physical and electrical forces are not as well understood. For decades, semiconductor companies like Intel have relied on scanning electron microscopes (SEMs) to characterize surface features at the various manufacturing nodes: 90 nm, 45 nm, 32 nm, etc. This ability of SEMs to determine surface features in a wide variety of microstructures and materials explains why, according to a 2011 global electron microscopy market analysis by Future Markets Inc., four times as many SEMs are in use by industry than in public or academic research.
The value of SEMs to the semiconductor industry is so high, says Vern Robertson, JEOL USA’s SEM technical sales manager, that much of the SEM market has moved to Asia, where microchip manufacturing is strong. But as Moore’s law drives research efforts deep into the nanoscale realm, they are looking for even better SEM performance.
“They would love to stay with SEM rather than go to TEM or STEM,” says Robertson. “They are looking at the 9-nm node, and you’re not going to do that sort of imaging with a traditional SEM." Part of the reason is that SEMs are much easier to purchase and operate than transmission-type instruments. Sample preparation is simpler, and analysis of biological samples is easier. Currently, two main types of SEM dominate the marketplace. Thermionic SEMs use high-temperature tungsten wire filaments to emit electrons. These have a lower brightness, larger energy spread, and a larger source size. This limits resolving power, but also lowers cost. Field-emission SEMs are ideal for research applications, which deliver electrons at room temperature in high vacuum.
This order is beginning to break down, however, as new technologies transform what a SEM looks like, how it performs, and who uses it.
In 2010, JEOL USA Inc., Peabody, Mass., earned an R&D 100 Award with the ClairScope, a SEM that, for the first time, integrated optical microscope technology with full-featured electron optics in an instrument that can conduct both analyses concurrently, at atmospheric pressure.
Developed in conjunction with the National Institute of Advanced Industrial Science and Technology in Japan, the instrument helped spark a movement toward correlative light microscopy. The advantages are substantial. By offering an open specimen area, the microscope allows users to easily change samples or reagents and quickly return to correlative photon or electron imaging. Thin-film technology protects the sample from electron radiation, reducing the amount of pre-treatment needed for samples. The window, positioned above the inverted electron gun that forms the bottom half of the instrument, is made of 100-nm-thick silicon nitride that allows electron transmission but retains a 1-atmosphere pressure differential.
According to Robertson, the company was driven by a desire to bring two capabilities to the marketplace that had been unavailable to researchers. First, researchers wanted to analyze samples in their native state using a beam instrument.
“Second, for many years, high resolution and completely flexible microscopy technology have been mutually exclusive. That has been a big pressure for us. Can we combine all of these techniques together into one instrument that can analyze to 1 nm at 1 kV?” says Robertson.
Part of an SEMs’ attraction, he says, is breadth of applications. A geologist by training, Robertson has been working with SEM applications at JEOL since 1986, first as a specialist, then as product manager. He says JEOL recognizes the potential value of electron microscopy for many researchers who haven’t had the training, funds, or knowledge to pursue that capability. Which is why, in addition to offering high-end field-emission microscopes like the newly introduced field-emission JSM-7800F, which can achieve 0.8-nm resolution, lower-end instruments like the benchtop NeoScope help introduce new users to electron microscopy.
Designed from the start as a compact instrument, the NeoScope lacks the performance levels of JEOL’s research telescope, but instead features large depth-of-field imaging at up to 60,000X magnification. The accelerating voltages are adjustable, high- and low-vacuum modes are available, and secondary and backscattered electron imaging modes are possible. Electron-dispersive spectroscopy (EDS) is also available for elemental analysis.
Aside from its potential as a value proposition for microscopy users, a factor that has helped position the NeoScope in the marketplace has been the effort to increase accessibility. Electron microscopes, even SEMs, have been infamously resistant to practical use without a high level of training and knowledge of sample preparation. That preconception is rapidly being dispelled, even in high-end TEMs or STEMs. For the NeoScope, JEOL wanted to make it as simple to use as an inverted optical microscope. They introduced a touchscreen interface that mimics common tablet computers. They also introduced a variety of automatic functions and pre-stored recipe files to facilitate use for a variety of sample types. That effort paid off.
“It’s been beyond our wildest dreams. We figured we’d get some interest, but our concern was that it would poach business from tabletops and traditional R&D laboratory SEMs,” says Robertson. “Since its introduction there’s been no drop in business on the R&D side, and the NeoScope seems to have its own new market in performance and price point.”