Total Pageviews

Saturday, September 10, 2011

Zoo of super resolution microscopy.

Microscope, one of the most popular optical instruments, has been paving the way of biological science for the past three hundred years. With the aid of the microscope, detailed observations of sub-cell size resolution were made possible. This, in turn, accelerated our understanding of the biology in an unprecedented way. Three hundred years have passed; we now arrived at a new cross road -- While triumphing on the universe of biology, a desire to develop microscopes with specificities and better resolutions is creating another revolution.

Specificities problems are less optical relevant. It is like painting different organelles of the cell with different colors. To do so, scientists use fluorescent dyes to attach to different organelles or encode them directly into the genetic codes of the proteins. So we can differentiate what they are and where they are. Scientists are quite good in doing so.

Resolution is another story. It is a barrier imposed by fundamental physics. In other words, the enemy of a microscope is diffraction, which prevents how well you can resolve two points on the focal plane. Same principle also applies to how tight you can focus a collimated beam. Using the traditional microscope, you cannot have resolution better than hundreds of nanometers if visible light is used. The axial resolution is not much better. As a result, no matter how small the particle in the focal plane is (in this case, the fluorescent dye), you would always observe a blob with some sizable volume. How do achieve better resolution? What kind of tricks scientists can play to break the diffraction limit?

For me, the first milestone in super resolution is called FIONA (Fluorescence Imaging with One Nanometer Accuracy). What a lovely name! In a nutshell, it fits the fluorescent signal with a Gaussian function. By doing so, it finds the center of the dye theoretically. Just like finding a center of the blob in the example we gave above. This method is generally adopted in modern microscopy since it localizes the location of the dye in the lateral plane quite well. There is a caveat though -- you cannot have too many dyes in focal point. This is just going to screw up your fitting.

Same mathematical manipulation does not work satisfactory in axial direction. In addition to multi-photon microscopy which aims on attacking this problem, there are other neat techniques existent. The way to get around it is modifying and mixing the experimental setup with other optical phenomena. The most eye-catching technique to me is the research led by professor H. Hess in HHMI. By putting a three-way beam splitter, the florescent signal from the dye in the focal plane would interfere with itself and generate different interference pattern depending on how far the dye is offset from the true focal point. This method achieved tens of nm of axial resolution. What impresses me the most is the feeling I have when trying to understand the diagram of the experimental layout. Suddenly, you realize, the imagination to advance optical science is unlimited.

Figure 1. The optical layout for interference microscopy. Courtesy of G. Shtengel, et al. in PNAS 106 9 3125 (2009). 

Other neat ways emanate from bright minds also. One way to do so is to create bizarre spatial beam profiles at the focal plane. By putting an annular apodization mask, the work from the research group led by professor E. Betzig in HHMI created the Bessel beam profile at the focal plane. Combined with structural illumination, they created a focal spot, which has z resolution of less than 300 nm.  Another research group led by professor W. Moerner in Stanford used spatial liquid modulator (SLM) to create helix beam profile at the focal plane. With the help of de-convolution algorithm, they could localize a fluorescent dye with ~ 20 nm z resolution in a total of 2 micron depth. This is truly amazing.

Focusing the light into a tiny spot is not the only solution. In the branch of developmental and embryonic biology, the speed of taking the image with decent resolution is of prime concern. To solve this problem, scientists used a century old technique (using a tube lens to focus the light into a light sheet) with the modern spice of fluorescent labeling in the genetic level. Generated light sheet at the focal point excites a plane of fluorescent proteins in one shot. Research led by professor P. Keller (now in HHMI also) in European Molecular Biology Laboratory (EMBL) used this technique to elucidate the developmental process of a zebra fish embryo in the first 24 hours. A full digital documentation of the embryo is resulted. Optical science is delving very deep to search the origin of nature, isn’t it! A good website of light sheet microscopy can be found here!

Figure 2. Light scanned microscopy. The laser beam illuminated the sample from the side and excited the fluorescent proteins in the plane. Courtesy of  P. Keller et al. in Science 322 1065 (2008).

Coincidentally I found a very intuitive and interactive website for super resolution microscopy. It is like a power shot to everyone who is interested in knowing more in modern microscopy.

The opinions expressed herein are those of the author and do not represent the Optical Society of America (OSA) or any OSA affiliate.

No comments:

Post a Comment