Fluorescence Microscopy

Fluorescent dyes are a critical component of the systems described below because they allow for specific and sensitive determination of the localization of molecules (i.e., proteins and DNA) within the cell. Simply, fluorescent dyes are themselves molecules that are able to absorb light of one wavelength and then emit light of another, longer wavelength. For fluorescence microscopy, these dyes are usually bound to the cellular molecule under study. By illuminating a dye-labeled specimen with light matching the excitation spectrum of the dye, and then collecting the emitted light, it is possible to visualize only the location of the dye molecules (providing specificity). Figure 1.5 includes a schematic of an epifluorescence microscope like the one used to collect the image data used below and shows how the exciting light and emitted light are separated. Another important feature of fluorescent dyes is that they are visualizable in relatively low numbers (providing sensitivity). Although the systems described below are not so sensitive, specially constructed microscope systems have been able to visualize single dye molecules [8].

  

Figure 1.5: A schematic of the light path in an epifluorescence microscope. Excitation light (left) is passed through the excitation filter and then reflected off the dichroic beam splitter. This light then passes through the objective to illuminate the specimen. Light emitted from the specimen (right) is collected by the objective and then passed through the dichroic and an emission filter (not shown) before being collected via an eyepiece or camera (also not shown). Reprinted, with permission of the publisher, from Fluorescence Microscopy of Living Cells in Culture: Part A. Fluorescent Analogs, Labeling Cells, and Basic Microscopy, Vol. 29 of Methods in Cell Biology, ©1989 by Academic Press, Inc.

The specificity of fluorescent dyes is partially a secondary characteristic, however. That is, each dye must be targeted and bound to a particular cellular molecule in order to work as a specific marker. One way to accomplish this targeting is to conjugate the dye to an antibody. Antibodies are proteins produced by the immune systems of higher organisms and are capable of binding very specifically and strongly to their target molecule (the antigen), usually a protein. Humans, for instance, are producing between 10⁷ and 10⁹ different antibodies at any given time.

In general, antibodies to be used for immunofluorescence are obtained by first injecting a mouse with a sample of the protein to be labeled. After allowing time for the mouse to mount an immune response to the injected, foreign protein, antibody-producing cells are collected from the mouse's spleen. Since these cells will not grow indefinitely in vitro, they are fused with tumor cells so that the resulting hybrid cells both produce antibodies and proliferate indefinitely (such a cell line is called a hybridoma). From among all of the cells that successfully fuse, individual cells are then cloned (i.e., allowed to proliferate in their own dish). Because a mouse will have immune cells producing a variety of antibodies at any given time, each of the clones is then screened to determine whether it produces antibodies against the desired protein. After all of these steps, one is left with a hybridoma that produces many copies of the same antibody molecule. The antibodies produced from such a hybridoma are called monoclonal because the cells producing the antibodies are all derived from a single clone. Figure 1.6 depicts the major steps in this process.

  

Figure 1.6: A schematic representation of the process used to generate monoclonal antibodies against a protein of interest. Reprinted from MOLECULAR CELL BIOLOGY by Lodish et al. ©1986, 1990, 1996 by Scientific American Books, Inc. Used with permission by W.H. Freeman and Company.

Once monoclonal antibodies are available for the desired protein, there are two general approaches to attaching a fluorescent dye to each antibody. The first approach is to chemically conjugate the dye molecules directly to the antibodies (direct immunofluorescence - see Figure 1.7). While this method allows the fluorescent labeling of a specimen to be accomplished in a single step, it also requires each kind of antibody to be conjugated with dye in separate steps. A second approach relies on the fact that all antibodies produced by a mouse have a common region in their protein structure (the F_c region). It is therefore possible using another kind of animal, say a goat, to generate antibodies against the F_c region of mouse antibodies. The fluorescent dye can then be conjugated to a large quantity of this secondary antibody. The now fluorescent secondary antibody is then used to label each of the primary antibodies which in turn label their antigenic protein. This method of indirect immunofluorescence (see Figure 1.7) was used for most of the antibody labeling in the work described below. Both direct and indirect immunofluorescence are able to utilize the high specificity of antibody-antigen binding and the sensitivity of fluorescence microscopy.

  

Figure 1.7: Schematic representation of indirect (left) and direct (right) immunofluorescence.

After the cells have been labeled appropriately, they are taken to a fluorescence microscope (see Figure 1.5) for imaging. One important aspect of a fluorescence microscope is the effective resolution of the image it is able to create (i.e., the smallest distance two fluorescent objects must be separated by in order for them to be resolved). Analysis of both the transverse (in the image plane) and axial resolution of the microscope are detailed below.

No matter how well designed, a microscope objective cannot collect all of the light emitted by a fluorescent specimen. Because of this, and because the wavelength of the fluorescence is finite, high frequency information is lost and the transverse resolution of a fluorescence microscope is finite. Theoretically, the transverse resolution is defined by the Rayleigh criterion which says that the minimum distance (D) between two point sources in the specimen, such that they can still be resolved, is equal to the radius (r) of the Airy disk for that microscope system. Simply, an Airy disk is the output of the microscope system corresponding to a point source at the input. The radius of the Airy pattern is defined by

$$r=\frac{1.22\lambda}{2\mathrm{NA}}$$ (1.1)

where $\lambda$ is the wavelength of the emitted light in air and NA is the numerical aperture of the lens ( $\mathrm{NA}=n\sin\theta$ where n is the refractive index of the medium between the specimen and the lens, and $\theta$ is the half-cone angle of light captured by the objective) [9, p. 31]. Following this analysis for $\lambda=520$nm, and NA=1.3, the approximate transverse resolution (D) of the microscope system used in this work is 200 nm.

Given that the data below were collected as stacks of images by changing the focus after each slice, the axial resolution of the system is also important. The axial resolution is determined by the depth of field of the microscope, and is defined by

$$\Delta_z=\frac{2\lambda n}{NA^2}$$ (1.2)

where $\lambda$, n, and NA are as defined above. For $\lambda=520$nm, NA=1.3, and n=1.5, $\Delta_z\approx 920$ nm, or about 5 times worse than the transverse resolution.

For comparison, the cells used below are $\approx 100-150\mu$m across.