The term correlative microscopy (CM) refers to the pairing of data sets derived from two different imaging techniques using the same specimen. For example, the most common type of CM is correlated light and electron microscopy (CLEM), in which a specimen is imaged by light microscopy (LM)—often using fluorescent markers—before being imaged by electron microscopy (EM). CM may include modalities such as scanning electron microscopy (SEM), transmission electron microscopy (TEM), focused ion beam SEM (FIB-SEM), atomic force microscopy, magnetic resonance imaging, secondary ion mass spectroscopy, traction force microscopy, and scanning small-angle x-ray scattering.
Regardless of which modalities are used, CM takes advantage of the strengths of two techniques, correlating their observations to learn more than from either method alone. For example, CLEM correlates functional, dynamic events within living cells observed by LM, to ultrastructural images in EM. Such a feat creates challenges—moving a specimen between instruments, using software to overlay the correlated images, contamination between modalities, large data sets, to name a few—that scientists are tackling in hot pursuit of answers to their own particular questions. Here’s a look at some technologies helping scientists grapple with these CM challenges.
The physical and software connections
Most correlative microscopy experiments require the specimen to be transferred physically from one instrument to another—creating challenges for overlaying both data sets. “It is possible to correlate images from different microscopes manually, but it is very tedious and difficult without the help of software and hardware tools,” says Scott Olenych, product marketing manager for life science automation at Carl Zeiss Microscopy. Examples from Zeiss include ZEN Shuttle & Find, a software module and a sample holder designed to shuttle between light and scanning electron microscopes for correlative microscopy. The sample holder facilitates physical positioning of sample between microscopes. The software recognizes fiducial markers on the sample holder to calibrate positions of the regions of interest, so that images from both modalities can be properly aligned.
The Zeiss ZEN Connect software module also simplifies the correlation process by organizing and overlaying images from multiple modalities. “You can correlate and navigate on images from ZEISS light microscopes and electron microscopes, [as well as] images imported from other devices and instruments,” says Olenych. In addition, the Zeiss ZEN Correlative Array Tomography software module can put together hundreds of serial LM and EM images—of sample slices cut with an ultramicrotome, for example—into one data set for 3D correlative microscopy.
Integrated CLEM systems
An alternative to shuttling the sample from one microscope to another is an integrated instrument. Integrated CM platforms are now available for CLEM, and typically consist of an SEM instrument equipped with optical capabilities.
Integrated CLEM systems have the advantage of the sample remaining stationary during the entire experiment. “Our instruments allow the user to navigate seamlessly from an optical to an SEM image,” says Natasha Erdman, product manager at JEOL. “When the sample is introduced into the scanning electron microscope, an optical image is taken [inside the SEM] and your entire navigation and correlation stems from that image.” JEOL’s platforms also incorporate the software into the instrument. “This is a total solution that integrates the software and hardware in a single tool,” Erdman adds.
Researchers often use a LM image, which typically resolves a few hundred nm, as a navigation map to find an area of interest. To zoom in further, they use SEM for resolution of a few nm, and further still with TEM to look at single atoms, with resolution of angstroms. Recently JEOL worked with a research group headed by Simon Watkins, a professor at the University of Pittsburgh’s Center for Biologic Imaging, to develop a method for bridging the resolution gap between LM and 3D SEM, so that cellular mechanisms can be studied without limits of resolution or scale imposed by imaging modality.
That intermediate scale of around a couple of microns, somewhere at the nexus of LM and EM, can be important biologically, yet tricky to image, says Erdman: “So that’s where we’re developing techniques with them to try to correlate what they see by fluorescence on the hundreds of microns scale, down to several hundreds of nanometers scale in the SEM.”
Delmic offers two integrated platforms for correlative microscopy. The SECOM can be retrofitted to an existing SEM, and is compatible with standard instruments from major SEM manufacturers. The DELPHI is a complete system that integrates a tabletop SEM with an inverted fluorescence microscope, along with a software interface. “The precise and automated overlay of light and EM images offers high accuracy correlation between cellular ultrastructure and function, removing operator bias in recording this, which is particularly powerful when there is no prior knowledge of fluorophore localization,” says Sangeetha Hari, applications team lead at Delmic. Hari notes that the large data sets of high-resolution images acquired by CM are challenging, but predicts that greater automation will enable “the application of CM in life sciences and biomedical imaging.”
Cryo-EM for CM
Software can assemble 3D images from serial EM images of sections of a sample that is either embedded in resin, or fixed in a vitreous state by rapid freezing at cryopreservation temperatures (also known as serial blockface EM). Cryo-EM of vitrified samples can be advantageous when it’s important to avoid the fixation, staining, and dehydrating involved in resin embedding, which may better preserve certain structural features depending on the nature of the specimen.
Leica Microsystems offers a range of products to support cryo sample preparation and manipulation, cryo-LM, cryo FIB-SEM, and cryo-TEM. Their new ARTOS 3D ultramicrotome slices samples into hundreds of serial sections that can be mounted for analysis by LM and EM to build high-resolution 3D maps. Other tools for cryo workflows include Leica’s VCT 500 and ACE 600 instruments, which allow researchers to manipulate cryo samples safely through the correlation microscopy process.
The Leica THUNDER Imager EM Cryo CLEM is a high-resolution light microscope built to image and optimally maintain cryo samples that will subsequently be transferred to an SEM or TEM instrument. “Our products enable experiments that previously were not possible without either complicated home-built solutions or at the high cost of manual labor,” says Robert Kirmse, senior product manager for sample preparation at Leica Microsystems. “With our THUNDER Imager EM Cryo CLEM, for example, we include workflow software that guides the user very efficiently through the process of checking their cryo EM samples and identifying the locations of interest for later retrieval in electron microscopes such as FIB-SEM or TEM.”
New research in 3D CLEM includes different methods to register the z dimension. Hoffman et al. recently correlated 3D super-resolution fluorescent LM images, with focused ion beam scanning EM (FIB-SEM) to resolve multicolor labeling of proteins within the 3D ultrastructure of whole vitrified cells. Wu et al. uses cryo-Airyscan confocal microscopy to determine the z position of fluorescent targets within vitrified cells, then uses FIB-SEM for 3D imaging of the interior of whole cells.
Now, more than ever, correlative microscopy allows researchers to benefit from multiple imaging data sets. “A live-cell imaging experiment can be complemented with cryo light microscopy and then EM, down to molecular resolution,” says Kirmse. “This really connects what we know from a living environment down to single molecules.” Indeed, Olenych thinks that more researchers are seeing the value of combining imaging modalities to gain new perspectives in their research: “This area is growing, and in the future will be much more prevalent.”