Why use a confocal microscope?
The most important feature of a confocal microscope is the capablity of isolating and collecting a plane of focus from within a sample, thus eliminating the out of focus "haze" normally seen with a fluorescent sample. Fine detail is often obscured by the haze and cannot be detected in a non-confocal, fluorescent microscope.
The confocal microscope has a stepper motor attached to the fine focus, enabling the collection of a series of images through a three dimensional object. These images can then be used for a two or three dimensional reconstruction.
Double and triple labels can be collected with a confocal microscope. Since these images are collected from an optical plane within the sample, precise colocalizations can be performed.
A plane of focus within a specimen is defined by the optics of the microcope. In a fluorescent microscope a small part of a sample may be in focus but you look at the entire object (i.e. you view what is in focus as well as what is out of focus).
With the confocal microscope, the z-resolution, or optical sectioning thickness, depends on a number of factors: the wavelength of the excitation/emission light, pinhole size, numerical aperture of the objective lens, refractive index of components in the light path and the alignment of the instrument.
This figure depicts the effect of the pinhole, or iris diaphragm, on the thickness of the optical plane that is collected. The pinhole and focal plane in the sample are at conjugate planes of focus.
The small pinhole opening in the diagram on the left enables data collection from a thin optical plane within the specimen. Points that are out of the plane of focus (red) will have a different secondary focal plane thus, most of the data is deflected.
Although some of the out-of-focus light enters the photomultiplier tube (PMT) in the figure on the left, the intensity is too dim to be visualized. All of the data at the plane of focus is collected (blue). In this manner, the confocal microscope can collect only the data from within the focal plane.
The larger pinhole opening in the figure on the right allows both in-focus and out-of-focus data to be collected.
Fluorescence, Reflectance or Transmission?
Confocal microscopy is most commonly used for detecting fluorescent labels but can also be used in a reflectance mode for DAB or immunogold labeling. A transmitted image can be collected along with the fluorescent or reflectance images but, a transmitted image is not confocal.
Which confocal microscope should you use?
The Keck Center has two confocal microscopes - a BioRad MRC600 and a Leica TCS NT/SP.
The BioRad MRC600 is on an Optiphot (upright) Nikon microscope. Objectives available are 2X, 10X and 20X dry lenses, and 60X and 100X oil immersion lenses. The two photomultiplier tubes enable collection of two images simultaneously. This system is most useful for collecting single or double labels with fixed tissue. Laser lines available for excitation are 488nm, 568nm and 647nm.
The Leica TCS SP/NT is on an DMIRBE inverted microscope. Objectives available are a 5X and 10X dry, 25X, 40X and 100X oil and a 63X water immersion lens. Three photomultiplier tubes for fluorescence or reflectance and 1 for transmission allow for the acquisition of four images simultaneously. A non-confocal DIC (differential interference contrast) image can be collected simultaneously with a confocal image.
The Leica system is good for live imaging although we currently don't have a perfusion chamber or heated stage. Visible laser lines are 457nm (weak), 488nm, 568 nm and 633 nm. Variable emission ranges can be selected. The Leica is also capable of two photon excitation. UV excitable dyes can be collected with the 2-photon laser and combined with confocal images. If you plan to use Dapi or Hoechst as a nuclear label you will need to use the Leica 2-photon equipment.
Confocal or 2-photon microscopy?
View the links in the section on 2-photon microscopy for a more detailed explanation.
Confocal microscopy and two-photon microscopy are both used to achieve optical sectioning but they do so in different ways. With confocal microscopy, the size of the variable pinhole in front of the detector (photomultiplier tube) determines, in part, the thickness of the optical section. In two photon microscopy, optical sectioning results from the fact that the probability of a two photon event occuring (i.e. excitation) happens only at the focal plane where there is an extremely high photon density. As a result, in 2-photon imaging, excitation occurs only at the plane of focus. Conversely, excitation (and bleaching) occurs throughout a significant portion of the sample depth with the confocal. This is illustrated by M. Cannell amd C.Soeller at: http://www.sghms.ac.uk/depts/pharmacology/cellbiophysicsgroup/2P_Theory.html
Therefore, 2-photon imaging is particularly useful for live imaging of thick samples. Photodamage at the focal plane will still occur, as with the confocal - but there isn't damage above and below the plane of focus.
Two photon imaging is also useful with UV excitable dyes in live cells as the excitation is achieved with infrared light so the cells are never exposed to the more damaging UV excitation.
The infrared radiation used for excitation in 2-photon imaging penetrates into tissue more efficiently than shorter wavelengths.
In the Keck Center, the Leica two photon system must be used for achieving optical sectioning with UV excitable fluorophores. If optical sectioning isn't required, then we have three CCD systems that can be used for video imaging.
The best starting point is to follow a protocol from a lab getting good results using the same or similar material. You still need to experiment with times and concentrations for fixation and labeling. Controls are essential.
A good, basic reference is Introduction to Immunocytochemistry by J.M.Polak and S.Van Noorden.
Good tissue preservation is generally a trade-off with preserving antigenicity. Fixation must adequately stabilize antigenic sites, and other tissue components, for subsequent treatment. If fixation is inadequate, the fluorescence in your sample can diffuse and disappear over (a possibly very short) time. Ideally one would compare a fixed sample to a living specimen so you know what artifacts you have induced by fixation. All processing induces artifacts...we can only try to minimize them.
Several factors must be considered when deciding on a fixation protocol: composition of the fixative, pH, osmolarity, temperature, time and method (perfusion or immersion) can all affect the outcome.
The most common fixative is buffered 4% paraformaldehyde used at 4oC. Paraformaldehyde is prepared fresh from powder or purchased in vials sealed with inert gas. Commercial formalin contains methanol (6-15%) and other impurities that may affect fixation.
Tissue needs to be submerged in the fixative (often following perfusion) and gently agitated. Cells and tissue remain osmotically active after fixation in aldehydes. Check for excessive swelling or shrinkage in the tissue.
Fixation times vary with the type of specimen and the accessibility of the epitope. Cell cultures are generally fixed for a minimum of 30 minutes. Tissue is fixed for a minimum of one to four hours (or longer). Use the longest time possible that still results in good antibody labeling. (Para)formaldehyde penetrates rapidly but fixes slowly.
Glutaraldehyde induces autofluorescence in your tissue. Samples fixed with glutaraldehyde must subsequently be treated with sodium borohydride.
Methanol and acetone are not good fixatives for preserving most structures. They precipitate proteins. Tissue will be shrunken and distorted.
The excitation and emission spectra of the fluorophore(s) used must coincide with the filters on our systems.
The Keck Center BioRad MRC 600 is equipped with a krypton-argon laser. The major lines of the laser are 488, 568 and 647 nm. - corresponding to fluorescein, rhodamine and Cy5. The system can be used for double labels with 488 and 568 excitation.
The lasers on the Keck Center Leica TCS SP have excitation lines at 457(weak), 488, 568 and 633 nm. Fluorescent emissions are collected using adjustable slits. Multiple labels can be collected simultaneously or sequentially.
Check the entire spectrum, not just the maxima, for overlap between multiple labels.
BioRad has a web page at http://fluorescence.bio-rad.com that allows the selection of up to four fluorophores to view on on one plot. Filters on the system can also be added to the plot. The Imaging Facility at the University of Arizona: http://www.mcb.arizona.edu/IPC/spectra_page.htm also has a web page where you can view the entire spectra and the overlap of the excitations and emissions of specific fluorophores.
Molecular Probes makes the spectra for many of their fluorophores available on their website: http://www.probes.com
1. longer wavelengths penetrate further into a thick sample
2. shorter wavelengths result in better resolution
3. shorter wavelengths result in a thinner optical section
4. autofluorescence is usually worse with 488nm. excitation.
Cells and tissue are usually permeabilized with Triton x or saponin for
labeling with antibodies. This step can affect your results. Membrane
extraction can occur with Triton. Saponin is "gentler". As with all steps,
time, concentration and temperature are important.
An avidin-biotin protocol is often necessary to boost the fluorescent signal.
Fluorescence from well-fixed tissue can often be detected in samples that are several months old and kept in the dark at 4oC.
Controls are essential.
Laboratories has a "Technical Center" about immunolabeling
A good resource for information is the archives of the confocal users group http://listserv.acsu.buffalo.edu/archives/confocal.html. The following are excerpts from postings to the group:
The mountant you use under the coverslip is important for several reasons:
1. Fluorescein is sensitive to pH (absorption/emission maximal at pH 8.5).
2. The mountant must have an additive to prevent photobleaching.
3. The refractive index of the mountant should be as close to that of your (fixed) tissue as possible (1.515). In most cases, do not use an aqueous mountant with fixed tissue
EXAMPLES OF REFRACTIVE INDICES:
A mountant of 9 parts glycerol; 1 part PBS and 1-3% n-propyl gallate works well. Dissolve the n-propyl gallate in the PBS, then add the glycerol.
The working distance of a lens is the space between the front element of the objective and the top of the coverslip. Any space between the coverslip and the tissue (*) adds to the working distance since a manufacturers figure for the working distance of a lens presumes that the tissue is just under the coverslip.
Working distances for some of our Nikon lenses:
Note the relatively short working distances. Pap pens, rubber cement, or any other substance used to create a well for immunostaining must be completely removed before coverslipping as any remnants left on the slide can elevate the coverslip. The resulting increase in the distance between the front element of the objective lens and the tissue will prevent focusing within the sample if this distance exceeds the working distance of the lens. For the same reason, don't use too much mountant under the coverslip. The wells in depression slides are usually too deep to allow focusing above 10X or 20X.
A coverslip placed directly on cells or tissue will compress the sample. To prevent compression, elevate the coverslip with a spacer as close as possible to the size of the tissue. Coverslips are too thick to be used as spacers for most samples. The thickness of a #1 coverslip is approximately 140 micrometers and a # 1 1/2 coverslip is about 160 micrometers. Cells cultured on coverslips are ideal since they can be flipped over, mounted on the slide and the thickness of the spacer is immaterial.
Thickness of the coverslip is important for optimal image quality. The coverslip thickness an objective lens is designed for is engraved on the lens. Most of the objective lenses in the Keck Center are designed to be used with a # 1 1/2 coverslip and will have "0.17" engraved on the objective lens (=0.17mm. or 170 micrometers). A variation in the thickness of 0.01 mm. with higher magnification lenses can make a difference in image quality.
If distance between the tissue and the coverslip is unavoidable, then a #1 coverslip might be necessary in order to focus a short working distance lens - at the expense of resolution.
A slide is usually sealed around the coverslip with nail polish. This prevents evaporation of the mountant and stabilizes the coverslip for use with immersion lenses. The coverslip needs to be stabilized for use with immersion objective lenses (60X, 100X on the BioRad and 25X, 40X and 100X on the Leica).
On the Leica confocal/2P system the coverslip should be at least 4 mm. from the long edges of the slide due to the slide holder design.
Additional information on sample preparation can be found at http://www.itg.uiuc.edu/publications/techreports/99-006
Preparation of buffered 4% paraformaldehyde fixative from powder:
Mix 4.0 grams of paraformaldehyde powder with 50ml. of 0.2M buffer. Heat at 60oC while stirring.
Add 1N NaOH drop by drop, stirring continuously, to clear (depolymerize) the solution.
Dilute to 100ml with dH20 to make a final concentration of 4% paraformaldehyde in 0.1M buffer.
Concentration of fixative or buffer may vary with different tissue types.
Heat the solution under a hood to avoid breathing fumes.