New microscopes reveal cells with unprecedented clarity
BETHESDA, Md. – Two new microscopes, both the first of their kind, could help biologists observe how the brain develops and viruses attack, say researchers at the National Institutes of Health (NIH). The first captures small, fast-moving organisms at an unprecedented rate and double the spatial resolution of a conventional microscope; the second displays large-cell samples in 3-D while decreasing the amount of harmful light exposure to the cells.
The first, 10 to 100 times faster than traditional technologies, also enables cell components that were once quite blurry to become sharply defined.
“It’s always helpful to look at smaller and smaller things,” said Dr. Hari Shroff, lab chief of the section on High Resolution Optical Imaging at NIH’s National Institute of Biomedical Imaging and Bioengineering (NIBIB). “Looking at a fixed cell at high resolution can tell you where different parts of the cell are at any given moment, but because much of biology depends on the movement of very small proteins finding each other and interacting, we really needed to look at how things move in a live cell.”
Shroff and research fellow Dr. Andrew York found an answer to these problems with their new instant linear structured illumination microscopy (iSIM), described in a paper published in Nature Methods (doi: 10.1038/nmeth.2687
). Building on traditional SIM technology, the iSIM allows real-time, 3-D superresolution imaging of small, rapidly moving structures such as individual blood cells moving through a live zebra fish embryo.
“What we’ve essentially done is eliminate the need for extensive computer processing by creating a better microscope at every stage of data gathering,” Shroff said. “Before, we relied on computer software and algorithms to do things like sort through hundreds of images, eliminate out-of-focus light and combine the individual images together. Now, we can do most of that optically with the microscope itself.”
(a) MSIM image of microtubules stained with Alexa Fluor 488 in a fixed cell. Scale bar: 5 μm. (b) Magnification of the boxed region in (a); MPSS = multifocal-excited, pinholed, scaled and summed images. MSIM = MPSS and deconvolved image. (c) Magnification of respective regions in (b); scale bars: 1 μm. (d) Plots of intensity along the respectively colored lines in (b). Full-width half-maximum values are: wide field, 299 nm; MPSS, 224 nm; and MSIM, 145 nm. Courtesy of Dr. Hari Shroff, NIBIB/NIH.
This means that researchers can see the images instantly, instead of waiting hours or sometimes days, and the data takes only about 1 percent of the hard-drive space as that produced by previous microscopes.
The second microscope, described in a paper published online in Nature Biotechnology (doi:10.1038/nbt.2713
), builds on selective plane illumination microscopy (SPIM), which uses a thin beam of light to illuminate only the single plane being imaged. This prevents the biological sample from being overexposed to light. Traditionally, SPIM microscopes rotate the sample so that they can clearly see all the dimensions, but this severely limits the imaging speed and can increase the damage done to the cells from light exposure because of the many extra images taken at multiple angles. As a result, the ability to capture fast cellular motion is lost.
To combat this, Shroff and NIBIB staff scientist Dr. Yicong Wu developed a dual-view SPIM (diSPIM) microscope with two separate detection cameras. The cameras are set at a 90° angle to capture perpendicular views of the sample, resulting in undistorted 3-D images. Because only two views are acquired, the microscope can still capture events at very high speed – high enough to image very fast moving viruses. Also, with relatively simple modifications, traditional single-camera SPIM microscopes can be converted into the dual-camera diSPIM. The real challenge in developing this technology was to find a way to combine the two disparate images from the two cameras, which required the creation of a new postprocessing software algorithm.