Bioscience Imaging Facility
This page provides information intended to help prospective facility users understand the imaging techniques that are available and to prepare for imaging. Users in need of more detailed or project specific advice should contact the director (firstname.lastname@example.org).A quick guide to the capabilities of each technique can be found here.
- Preparing for microscopy
- Fluorescent microscopy techniques
- Other microscopy techniques
- Whole animal imaging techniques
Methods used to prepare biological samples for microscopic observation are almost as diverse as the samples themselves. Here, we offer some generally applicable advice on how to prepare cell cultures for fluorescent microscopy, as these samples are extremely popular.
Choosing the most appropriate fluorescent dyes must be a first consideration when planning a fluorescent microscopy experiment. A fluorophore's brightness (=extinction coefficient times the quantum yield), stability, specificity, possible toxicity, and spectral compatability with other dyes in the sample, as well as with the microscope's hardware, all need to be considered. Most major commercial vendors offer extensive information about all of these dye properties. For example, Invitrogen's website covers organelle specific fluorescent protein constructs and live-cell stains, fixed-cell stains for many cellular structures or processes,secondary antibody conjuguates, as well as a spectra viewer tool that the user to compare dye excitation and emission spectra with user-defined microscope configurations. After choosing dyes and performing the experiment, it is always worthwhile spending some extra time up front to optimize the staining procedure. No imaging instrument can make up for an unnecessarily weak signal or high non-specific background staining.
Resolving fine sub-cellular detail requires high numerical aperature (NA) objective lenses. This means that if you hope to achieve sub-micron resolution, your sample must be mounted under a glass cover slip (#1.5 thickness for our microscoptes). Hig NA lenses also have a limited 'working distance', which means that the sample must be no more than a couple hundred microns thick and apposed to the cover glass. For cell cultures, these requirements are easily achieved by growing the cells directly on substrate-coated glass coverslips.( 'Chambered'cover glasses also work well.)
Fixing cell cultures: After many experiments, cells are be preserved in a fixative prior to (typically) dye labeling. The fixative used can impact the cells' morphological integrity, the efficacy of dye labeling, and the level of autofluorescence. In most cases, fixation in freshly prepared 4% paraformaldehyde (in PBS) for 5-15 min is the best option. Wash away the fix and then proceed with staining. After staining is complete, the sample must be properly mounted on a slide prior to microscopy. Glycerol-based mounting media are least likely to induce morphological artifacts and are optically advantageous. When mounting, use as little mounting media as possible, and seal the edges of the cover slip with nail polish to minimize sample-drift and keep the media from leaking out and mixing with the immersion oil. After mounting, the quality of staining may degrade rapidly, so microscopy should be performed as soon as possible (although some stains can last for weeks or months).
Live cell imaging: For live cell imaging, the culture, while still bathed in media, is mounted on the microscope stage. The BIF has a very convenient, environmentally-controlled Tokai Hit live imaging chamber that works exceptionally well with 'glass-bottomed' 35 mm Petri dishes. The A1R confocal also has a 'perfect focus' system, which compensates for thermal drift and allows imaging even during media exchanges. Live cultures may be stained with 'vital' dyes or may express genetically encoded fusion proteins that can be imaged directly.
Epifluorescence Microscopy: In epifluorescence microscopy, the entire sample (field of view) is illuminated at once with excitation light that spans a small range (or band pass) of wavelengths. The resulting, longer-wavelength emissions are then collected by the objective, passed through an emission filter, and finally detected using a camera. Because all points (small areas) in the sample are excited and detected simultaneously, frame rates are limited only by the shutter speed (or read speed) of the camera. This acquisition speed is an advantage, if high-frame rate live imaging at higher pixel resolutions is desired. Some cameras (such as our EM CCD) are also more sensitive (5-10x) than the PMT detectors used in confocal microscopy (see below), an definite advantage for extremely low light level imaging, such as is encountered with ion sensitive dyes or during single particle tracking experiments. However, the epifluorescence full field illumination/detection regimen (in x,y, and z) is also responsible for the major drawback of epifluorescence microscopy: the predominance of out-of-focus background emissions, or 'flare', which greatly degrade image contrast (and therefore resolution). On our systems, the sample can also only be illuminated by one color of exciting light at a time, which makes multi-color live imaging impossible. Thus, epifluorescence microscopy tends to work very well for thin, single-label specimens that are quickly moving quickly (>1 fps) or that emit very weak fluorescence.
Laser Scanning Confocal Microscopy (LSCM): Confocal microscopy uses a pinhole and the objective lense to create a focused point of laser light, which illuminates one small (diffraction limited: ~200 nm) point in the sample at a time. (Although regions above and below this point are also illuminated as the beam passes through). Light emitted from the illuminated point then travels back through the objective and another pinhole before reaching a detector called a photomultiplier tube (PMT). Since the optical paths of the illuminating and emitted light both contain a pinhole in the same focal plane, the pinholes are said to be ‘confocal’. This arrangement blocks out-of-focus emissions, the bane of epifluorescence microscopy. The point of illumination is then scanned (using rotating mirrors) back-and-forth across the sample in a raster pattern to create an image. Since the image is free of fluorescent background, multiple focal planes (optical sections) acquired at different depths can be combined to produce a 3D image, a capability not (directly) possible with epifluorescence microscopy. A drawback of this point scanning procedure is that the laser beam must be scanned across the entire sample to create an image. This process can take too long (seconds) for some time-lapse applications, especially if there are many pixels (e.g. >1024x1024). (However, a new technology on our A1R confocal called resonant scanning can greatly increase scan speed). Faint signals may also be difficult to detect with this design, because the emitted light must travel through several intermediate optical elements before reaching the PMT, which is itself not extremely sensitive. Confocal microscopy is most suitable for creating 3D images of samples labeled with multiple fluorophores that are <50 um thick. It also works well for imaging multiply-labeled probes in living, moving samples, as long as high frame rates (i.e. >1 fps) are not required.
Laser Scanning Two Photon Microscopy: The optical elements for two photon microscopy are conceptually similar to those used for confocal microscopy, in the sense that a focused laser beam is scanned across a sample and the light emitted from each point passes back thorough the objective to be collected by a PMT. However, the physical mechanism of excitation is very different: Rather than relying on a single photon of shorter wavelength (shorter than the emitted light) to excite a fluorophore molecule, two photons of a wavength ~twice as long as the single photon excitation wavelength mustinteract simultaneously with the fluorophore instead. This '2-photon' mechanism allowslight in the IR region of the spectrum to excite standard fluorophores that emit at visible wavelengths. There are several benefits associated with this methodology: First, the exciting IR light is scattered/reflected by most biological tissues to a lesser degree than is visible light. Second, the probability of two-photon excitation is so low that it only occurs to any reasonable degree within the extremely small point where the laser is mostfocused, which limits photo-bleaching/damage in parts of the sample not being imaged.Third, sinceno out-of-focus emissions generated, no pinhole is needed to block out-of-gocus light so all emissions can be collected by the detectors, including those that have been scattered (so called non-ballistic photons). The combination of these factors allow much deeper imaging into thick specimens (up to ~400 um) than is possible using a confocal. However, two-photon excitation can also lead to unpredictable emissions from endogenous molecules, and (usually) only one fluorophore can be imaged at a time (since only one IR laser is available for excitation). Two-photon microscopy is useful for creating 3D images that extend deep (50-300 um) into living tissues labeled with a small number of fluorophores.
Confocal Reflectance Microscopy: Confocal reflectance microscopy uses the confocal principle (see above) to generate an image based on a combination of the samples micro-topography and local reflectance within the optical section. Since only reflected light is used to create the image, the sample can be completely opaque and no fluorophores are needed. (Negative contrast microscopy can be used to image the boundary of solid objects that are not reflective). This simple method can be used to map the surface of many biological and physical samples, as well as to visualize precipitates (e.g. DAB) or metallic particles (e.g. colloidal gold) in 3D.
Differential Interference Contrast (DIC) Microscopy: DIC microscopy is a wide-field, transmitted light technique that uses special prisms (birefringent crystals) to take two, images of the sample in parallel, one of which is slightly offset (perpendicular to the optical axis) relative to the other. As light passes through the sample, the phase of the light at each point (in each image) is slightly delayeddepending on the local optical properties (refractive index) and thickness of the sample. After passing the sample, the two images are then ‘un-offset’ and recombined such that neighboring points in the sample (separated by the distance of the shift) are overlaid. After overlay, interference converts the sample-induced phase differences (between neighboring points) into differences in intensity (brightness) in the final image. Thus, contrast encodes the spatial derivative of the refractive properties of the sample in the direction of the shift. This technique is useful for label-free visualization of the boundaries and organelles of living cells whose refractive index is different from their surroundings.
Phase Contrast Microscopy: Phase contrast is a wide-field, transmitted light technique that forms an image based on the diffraction that occurs as light passes through a sample. The diffraction is caused by boundaries between sample regions of different refractive index. Most light passes ‘straight’ though a sample, but some becomes bent in all directions (diffracted) and so travels away at an angle. Phase contrast microscopy illuminates the sample obliquely with a hollow cone of light to accentuate collection of the diffracted light. The objective focuses the non-diffracted (‘straight’) and diffracted (‘bent’) light emanating from the same point in the sample to different positions along the optical axis. A special insert within the objective, called a ‘phase plate’ then phase shifts the bent light relative to the unbent light. Once these beams are re-combined, contrast in the final image represents refractive index changes in the sample, as is the case for DIC microscopy (above). However, since the mechanism used to produce the phase contrast image is based on diffraction rather than directly on the samples refractive index, in most cases a phase contrast image has a noticeably different ‘look’ than a DIC image of the same sample and shows strong contrast at edges in all directions of the image plane. Like DIC, this technique is useful for visualizing the components of living cells without the need for an exogenous label.
Bright Field Microscopy: Bright field is the oldest and least complex of all microscopic techniques. Broad spectrum light (often from a halogen lamp) is shown on to the entire sample, and an image is formed based on how the light is absorbed. Traditional colored, histological stains or light-blocking precipitates are often visualized using bright field. Although this is a wide field technique, out-of-focus background is generally not too much of a problem (unless the entire sample is darkly stained), because the stains themselves do not emitted light as is the case in epifluorescence microscopy.(back to top)
Whole Animal Imaging
Proper PACUC approvals are required prior to performing experiments that involve the use of animals.
As with microscopy, molecular probes used in whole animal imagingmust be chosen based on their biological and chemical properties as well as their compatability with the facility's hardware. Typcially, the labeled-probe is injected into the animal immmediatley prior to imaging (usually via the tail vein), and the animal is then anesthetized to limit motion blur during imaging. (The facility provides isoflurane anesthesia).
Labs planning to perform SPECT imaging on animals will need REM approval for the use of radiation prior to beginning experiments. Caution be must be exercised when performing SPECT. Each animal is typically injected with an activity that produces 100-1000x background ionizing radition (gamma rays) near the source (with in 1 m)). Some exposure cannot be avoided when handling the injected animal.
Whole Animal Imaging Techniques
In vivo optical imaging: "In vivo optical imaging" is a technique that uses a large dynamic range camera and low magnification (<1x) optics to image either fluorescent or luminescent signals produced within an organism. In the case of fluorescence, the signal is generated and detected exactly according to the principles of epifluorescence microscopy (above). However,in vivo imaging requires longer exposure times (>1 sec) and longer wavelength dyes (>600 nm), since very little light can penetrate appreciable amounts of tissue (up to ~5 mm). In vivo fluorescentsignals are also easily contaminated with autofluorescence, a broad spectrum intrinsic fluorescence emitted by chemicals in the animal's body. Although simple steps can be taken to remove materials that produce autofluorescence (such as using furless 'nude' mice, shaving the fur, or feeding an 'alfalfa free' diet to eliminate chlorophylls), as possible, fluorophores should also be choosen such that their emission spectra will be strongest where autofluorescence is weakest. In animals, dyes that emitt in the far-red end of the spectrum bestavoid autofluorescence (and also minimize absorbtion and scatter).In plants, chlorophylls present a strong autofluorescence in the ~650-750 nm range that should be avoided. Invitrogen offers a range of long wavelength dyes specifically for in vivo optical imaging.Signal can also be generated via bioluminscent processes such as the luciferase/luciferin raction, elimanting the need for exciting light. Since these signals are very weak, even longer exposure times (>1 min) are needed. In either case, total intensity (signal integrated over space), which is roughly proportional to the amount of light-generating material present, is typcially the quantity of interest. In vivo optical imaging is useful for longitudinally (days or weeks) tracking the size of a population of cells (e.g. tumors) within an organism. This IVIS imagercan also be used as a plate reader or for visualizing the chemiluminescent signals that arise during blotting procedures.
microSPECT: Unlike all of the other techniques discussed above, which form an image usingvisible light, SPECT (single photon emission computed tomography) uses gamma radiation emitted by a radio-labeled tracer (nM concentration) to form an image. Since high energy gamma rays are (mostly) not absorbed or scattered by biological tissues, this technique allows the three-dimensional distribution of a radio-labeled tracer to be imagedat relatively high resolutions (down to 0.4 mm).Typcially, technetium-99m is used as the radio-label, due to its convenient decay properties (half-life = 6 hrs ;energy = 140 keV). For acquistion, our SPECT instrument uses a specially designed multi-pinhole collimator to project the gamma radiation into an array of two-dimensional images, each of which is acquired from a different perspective (angle). After acquisition, a computer algorithm converts all of the 2-D images into a (single) three dimensional image of the tracer's distribution within the organism's body. (Multi-channel imaging is also possible). A 'typical' mouse scan (of intermediate resolution and scan time) may require 1 mCi of activity and a scan time of 30 minutes. Our system is also equipped with a 'gating' mechanism that allows the data collected during different phases of a biological rhythm (e.g. respiration) to be grouped during image reconstruction, a process that defeats motion blur. SPECT imaging is useful for high resolution and quantitative 3D imaging of the spatial distribution of atracer within a living organism. The kinetics (on a time scale of minutes) of tracer binding can also imaged over small regions if higher amounts of radioactivity are employed, and the technique iscompatible with longitudinal studies. Unfortunately, only a few dozen radio-labeled tracers are commercially available, so in many cases a biologist interested in using SPECT will need to collaborate with a nuclear chemistry laboratory to have sodium pertechnetate (a chelatable technetium salt)) chelated to their molecular probe of interest. The purity and specificity of the newly synthesized tracer mustalsobe validated, which requires substantial effort.A basic understanding of the reconstruction algorithm and post-processing techniques are also necessary in order to properly interpret SPECT images.
microCT: X-ray computed tomography (CT or CAT) uses a camera to collect dozens of two-dimensional images from different perspectives of the x-rays that are transmitted through an organism's body. As with SPECT, a computer algorithm then converts the 2-D images into a (single) three dimensional image with ~ 0.1 mm resolution. Although x-ray dense (absorbing) bone is easy to visualize with this technique, soft tissues are not, unless an x-ray dense 'contrast dye' is injected in to the vasculature. This technique is complimentary to SPECT in that it provides an 'intrinsic' image of the organism's body onto which a SPECT signal can be overlaid. Our dual SPECT/CT system can sequentially acquire SPECT and CT images of the same organism during a single imaging run, which ensures that the images acquired in each modality arein nearly perfect register.(back to top)