of the cell.
The Scanning Electron Microscope
Whereas the image in a transmission electron microscope is formed by electrons transmitted through the specimen, in the scanning electron microscope it is formed from electrons that are reflected back from the surface of a specimen as the electron beam scans rapidly back and forth over it. These reflected electrons are detected and used to generate a picture on a display monitor. The scanning electron microscope operates over a wide magnification range, from 10 times to 100 000 times, and has a wide depth of focus. The images created give an excellent impression of the three‐dimensional shape of objects (Figure 1.7). The scanning electron microscope is therefore particularly useful for providing topographical information on the surfaces of cells or tissues. Modern instruments have a resolution of about 1 nm.
A major advance in light microscopy was the development of the fluorescence microscope. Fluorescent molecules emit light when they are illuminated with light of a shorter wavelength. Familiar examples are the hidden signature in bank passbooks, which is written in fluorescent ink that glows blue (wavelength about 450 nm) when illuminated with ultraviolet light (UV) (wavelength about 360 nm), and the whitener in fabric detergents that causes your white shirt to glow blue when illuminated by the ultraviolet light in a club. The fluorescent dye Hoechst 33342 has a similar wavelength dependence: it is excited by UV light and emits blue light. However, it differs from the dyes used in ink or detergent in that it binds tightly to the DNA in the nucleus and only fluoresces when so bound. Figure 1.13a shows the optical path through a microscope set up to look at a preparation stained with Hoechst. White light from an arc lamp passes through an excitation filter that allows only UV light to pass. This light then strikes the heart of the fluorescence microscope: a special mirror called a dichroic mirror that reflects light of wavelengths shorter than a designed cutoff but transmits light of longer wavelength. To view Hoechst, we use a dichroic mirror of cutoff wavelength 400 nm, which therefore reflects the UV excitation light down through the objective lens and onto the specimen. Any Hoechst bound to DNA in the preparation will emit blue light. Some of this will be captured by the objective lens and, because its wavelength is greater than 400 nm, will not be reflected by the dichroic mirror but will instead pass through. An emission filter, set to pass only blue light, cuts out any scattered UV light. The blue light now passes to the eye or camera in the usual way. Image (b) shows a field of cells cultured from rat brain after staining with Hoechst. Only the nuclei are seen, as bright ovals.
Although some of the structures and chemicals found in cells can be selectively stained by specific fluorescent dyes such as Hoechst, others are most conveniently revealed by using antibodies. In this technique an animal (usually a mouse, rabbit, or goat) is injected with a protein or other molecule of interest. The animal's immune system recognizes the chemical as foreign and generates antibodies that bind to (and therefore help neutralize) the chemical. Some blood is then taken from the animal and the antibodies purified. The antibodies can then be labeled by attaching a fluorescent dye. Images (c) and (d) show the same field of brain cells but with the excitation filter, dichroic mirror, and emission filter changed so as to reveal in (c) a protein called ELAV that is found only in neurons; then in (d) an intermediate filament protein (page 228) found only in glial cells. The antibody that binds to ELAV is labeled with a fluorescent dye that is excited by blue light and fluoresces green. The antibody that binds to the glial filaments is labeled with a dye that is excited by green light and fluoresces red. Because these wavelength characteristics are different, the location of the three chemicals – DNA, ELAV, and intermediate filament – can be revealed independently in the same specimen. Panel (e) shows the three images superimposed.
The technique just described is primary immunofluorescence and requires that the antibody to the chemical of interest be labeled with a dye. Only antibodies to chemicals that many laboratories study are so labeled. In order to reveal other chemicals, scientists use secondary immunofluorescence. In this approach, a commercial company injects an animal (e.g. a goat) with an antibody from another animal (e.g. a rabbit). The goat then makes “goat anti‐rabbit” antibody. This, the secondary antibody, is purified and labeled with a dye. All the scientist has to do is make or buy a rabbit antibody that binds to the protein of interest. No further modification of this specialized primary antibody is necessary. Once the primary antibody has bound to the specimen and excess antibody rinsed off, the specimen is then exposed to the fluorescent secondary antibody that binds selectively to the primary antibody. Viewing the stained preparation in a fluorescence microscope then reveals the location of the chemical of interest. The same dye‐labeled secondary antibody can be used in other laboratories or at other times to reveal the location of many different proteins because the specificity is determined by the unlabeled primary antibody.
Increasing the Resolution of Fluorescence Microscopes
The resolution and precision of fluorescence microscopes have steadily improved with time. In 1979 a team in Amsterdam invented the confocal light microscope that scanned a point of excitation light across the specimen to markedly reduce the contribution of out‐of‐focus light. In 1987 a team in Cambridge developed a prototype of a commercial system and within a couple of years all major microscope manufacturers began offering confocal light microscopes.
A series of