Caitlin Smith has a B.A. in biology from Reed College, a Ph.D. in neuroscience from Yale University, and completed postdoctoral work at the Vollum Institute.
An important facet of live-cell imaging, or studying live cells or particles under a microscope, is motion tracking—using microscopes and software to follow the position of cells or subcellular structures as they change over time. The technique is central to studies in chemotaxis, cell motility, organelle trafficking and single-molecule tracking, and it isn’t difficult. But it does require some specialized tools. Here we review some new options.
Leica Microsystems
Motion tracking requires a microscope, of course, but not just any scope will do. Particular features can greatly facilitate motion tracking. A big one is stability, or the ability to keep the specimen motionless over time.
Temperature fluctuations can cause destabilization or drift, even at room temperature. Leica Microsystems takes this into account in microscope construction, says Bernd Sägmüller, director of marketing for confocal imaging at Leica. “The inverted microscope Leica DMI6000 B, the best base for live-cell imaging with our widefield, confocal and super-resolution systems, was designed in a way to minimize temperature-related micro-movements of the microscope itself,” Sägmüller says.
For instance, Leica microscopes feature a stabilization tool called Adaptive Focus Control. AFC monitors and adjusts the position of the microscope stage to keep the sample in focus over the course of the experiment. This “becomes especially important when imaging confocal sections, or using the Leica AM TIRF MC [total internal reflection fluorescence, multicolor] system to study cellular membrane activity in up to 200-nm thin layers,” says Sägmüller. Another option to ward off the effects of temperature fluctuations is to use an environmental chamber, which surrounds the specimen on the stage and enables better temperature control. Both the AFC and environmental-chamber conditions can be controlled through Leica’s software.
Leica offers an additional stabilization tool called the Suppressed Motion (SuMo) Stage, designed for its widefield, super-resolution imaging Leica SR GSD 3D (super-resolution, ground state depletion, 3-dimensions) instrument. “The SuMo Stage ensures that the amount of drift is kept well below the localization precision, down to 20 nm, which the system is able to provide over the course of a typical experiment,” says Sebastian Tille, Leica’s director of widefield imaging.
Leica’s HCS A (high-content screening automation) software for automated high-content screening is also useful for researchers doing motion tracking. The software’s Single Object Tracker module controls the microscope stage, enabling researchers to follow single objects. “The stage is moving in [such] a way [that] the desired object remains in the center of the screen,” Tille says. “This is important for many applications, specifically for confocal imaging and widefield applications.”
Nikon Instruments
Like Leica, Nikon Instruments has developed its own imaging software for live-cell imaging. NIS-Elements software includes tools for motion tracking and analysis, as well as acquisition control for Nikon’s entire range of imaging microscopes.
One of the challenges in motion tracking, according to Lynne Chang, senior applications scientist at Nikon Instruments, is “teaching the software to recognize the objects you want to track.” NIS-Elements includes tools that “enable the user to easily adjust the recognition filters to optimize detection” of the objects of interest, says Chang, including manual, semi-automated and fully automated object detection useful for fine-tuning object recognition.
Models of how objects move are also useful for motion-tracking studies. Nikon’s software includes several models, such as constant velocity, circular motion, random motion and combinations thereof, which can help researchers to optimize tracking. “NIS-Elements also provides tracking algorithms that will track splitting objects, such as dividing cells, and objects that appear midway during the time sequence,” says Chang. The software also can filter out objects that behave in a particular way—for example, objects that change in size by a certain amount during an elapsed time. Other practical features include the software’s ability to deal with gaps in a time sequence. “For example, if the object of interest disappears for a short time interval, usually due to movement into a different focal plane, we can set our tracking algorithm to re-detect the object when it appears again and to connect the tracks,” says Chang.
Such tools can help researchers automate tracking. Yet automated motion tracking does pose unique challenges. For instance, results can be compromised when objects of interest show heterogeneity in size, intensity and motion, because it is “difficult to use one set of parameters to simultaneously detect all the objects in the field of view,” says Chang. One solution is to use semi-manual detection and tracking tools to aid the software in tracking. Another challenge is background motion, which occurs in whole-animal experiments. Tracking cells can be difficult when the cells’ movements are masked by the motion of surrounding tissue, for instance when the animal breathes. NIS-Elements can compensate for background motion by tracking and normalizing movements using fiducial markers (markers of physical location in the microscope’s field of view).
Additional software tools
Other microscope manufacturers, such as Olympus and Zeiss, offer software tools for their live-cell imaging equipment that also can analyze cell or particle tracking. These are compatible with, or based on, the MetaMorph® Microscopy Automation and Image Analysis Software, a platform developed by Molecular Devices for automated image acquisition and analysis. The platform provides a number of specific modules that enable users to tailor the software to their individual needs. One of these modules, the MetaMorph Super-Resolution System, can discern object details smaller than 250 nm in real time, according to Molecular Devices’ website.
Zeiss’ confocal microscopes use their own operating software (LSM 510 v4.2) for image acquisition, but users can subsequently analyze the cell-tracking data with MetaMorph. Zeiss’ LSM software also offers a Physiology module for long-term, live-cell, time-lapse imaging.
MetaMorph for Olympus also is based on the MetaMorph platform. This software includes motion-tracking tools for following the movement of tagged particles over time. Another option for Olympus microscope users is the Imaris Imaging software package with the ImarisTrack module, which tracks and analyzes the motion of live cells over time.
Tracking cells during embryogenesis
Sometimes, though, commercial software isn’t sufficient for researchers’ particular needs. In such cases, they must write their own. Howard Hughes Medical Institute investigator Philipp Keller, at the Janelia Farm Research Campus, takes a computational approach to tracking cell movements and cell divisions at the single-cell level during embryogenesis. His goal is to compile developmental "building plans" from live-animal imaging data. “These building plans can be understood as dynamic, digital atlases of cell behavior underlying the development of an animal,” says Keller. His team mainly studies zebrafish, fruit flies and mice, with particular interest in early nervous-system development.
Keller says there are multiple software packages capable of tracking cells in relatively simple data sets, which he describes as having a “small amount of image data, not too many cells and high image quality.” But these are insufficient for the complex data sets he works with, which are “on the order of terabytes of image data per specimen, heterogeneous image quality and physical coverage of entire embryos comprising tens of thousands of cells.” Keller’s group therefore develops its own software and computational tools—and makes them freely available to the research community, as well. (See http://www.janelia.org/lab/keller-lab and http://www.digital-embryo.org/)
Keller says the methods his group develops are free, fast, efficient and easy for nonexperts to use. But, they may not yet be adaptable to all types of experiments. His group is working to make the methods more widely applicable, but given the diversity of data sets, imaging modalities and labeling schemes, it can be difficult to compile a one-size-fits-all toolkit.
Still, with microscopy, software and analysis methods progressing together, motion tracking with live-cell imaging has never been as powerful. Stay tuned for the next generation of live-cell imaging advances coming from a lab near you.
Update (2/25/14): Nature Methods recently published a comparison of several motion-tracking algorithms based upon an open competition in 2012. The abstract concludes, "Although no single method performed best across all scenarios, the results revealed clear differences between the various approaches, leading to notable practical conclusions for users and developers." (Nat Meth, DOI:10.1038/nmeth.2808)