1. General Introduction

(Ueli Aebi and Andreas Engel, MSB Biozentrum, and Markus Dürrenberger, IEM Biozentrum)

As the telescope is the instrument of choice to explore the macrocosm, the microscope is the instrument of choice to view the microcosm including whole cells, bacteria, sub-cellular organelles, supramolecular assemblies, and individual protein molecules.

While there are a myriad of methods to study cells and their components, microscopes have the distinct advantage that they provide us with a direct image of the object of interest and not just with an 'abstract representation', e.g., a diffraction pattern, a spectroscopic signal or some other physical measurement, that has first to be decoded or interpreted on the basis of a model or some assumptions, before we know what the object actually looks like. As a consequence, light and electron microscopes have been, and still are among the most important tools to study the structure and function of cells and their constitutents. Thus, the illustrations in most biology textbooks are dominated by light and electron micrographs.

A typical animal cell is 10 to 20 mm in diameter, or about 5x smaller than the smallest object that can be directly seen with the naked eye. Animal cells are not only tiny, but they are also colorless and translucent; consequently, the discovery of their main internal features depended on the development of a variety of stains that provided sufficient contrast to render them visible in the light microscope. Likewise, introduction of the far more powerful electron microscope required the development of new techniques for preserving and staining cells before the full complexities of their internal fine structure could emerge. To this day, microscopy depends as much on specimen preparation techniques as on the performance of the microscope itself.

Fig. 1-1: Sizes of cells and their components drawn on a logarithmic scale, indicating the range of readily resolvable objects in the light and electron microscope.

1.1 Light Microscopy

Some of the landmarks in the development of light microscopy are outlined in Table 1:

Table 1-1 Some Important Discoveries in the history of Light Microscopy

1611 Kepler suggested a way of making a compound microscope.

1655 Hooke used a compound microscope to describe small pores in sections of cork that he called "cells".

1674 Leeuwenhoek reported his discovery of protozoa. He saw bacteria for the first time 9 years later.

1833 Brown published his microscopic observations of orchids, clearly describing the cell nucleus.

1838 Schleiden and Schwann proposed the cell theory, stating that the nucleated cell is the unit of structure and function in plants and animals.

1857 Kolliker described the mitochondria in muscle cells.

1876 Abbé analyzed the effects of diffraction on image formation in the microscope and showed how to optimize microscope design.

1879 Flemming described with great clarity chromosome behavior during mitosis in animal cells.

1881 Retzius described many animal tissues with a detail that has not been surpassed by any other light microscopist. In the next two decades he, Cajal, and other histologists developed staining methods and laid the foundations of microscopic anatomy.

1882 Koch used aniline dyes to stain microorganisms and identified the bacteria that cause tuberculosis and cholera. In the following two decades, other bacteriologists, such as Klebs ans Pasteur, identified the causative agents of many other diseases by examining stained preparations under the microscope.

1886 Zeiss made a series of lenses, to the design of Abbé, that enabled microscopists to resolve structures at the theoretical limits of visible light.

1898 Golgi first saw and described the Golgi apparatus by staining cells with silver nitrate.

1924 Lacassagne and collaborators developed the first autoradiographic method to localize radioactive polonium in biological specimens.

1930 Lebedeff designed and built the first interference microscope. In 1932, Zernike invented the phase-contrast microscope. Thes two developments allowed unstained living cells to be seen in detail for the first time.

1941 Coons used antibodies coupled to fluorescent dyes to detect cellular antigens.

1952 Nomarski devised and patented the system of differential interference contrast for the light microscope that still bears his name.

1981 Allen and Inoué perfected video-enhanced contrast light microscopy.

In general, radiation of a given wavelength cannot be used to probe structural details much smaller than its own wavelength: this represents a fundamental limitation of both light and electron microscopes. The ultimate limit of resolution of a light microscope is, therefore, set by the wavelength of visible light, which ranges from about 0.4 mm (for violet) to about 0.7 mm (for deep red). In practical terms, bacteria and mitochondria, which are about 0.5 mm wide, are generally the smallest objects that can be seen clearly in the light microscope; details smaller than this are obscured by effects resulting from the wave nature of light .

Because of its wave nature, light does not follow exactly the idealized straight ray paths predicted by geometrical optics. Instead, light waves travel through an optical system by a variety of slightly different routes, so that they interfere with one another and thereby cause diffraction phenomena. The interaction of light with an object will change the phase relationship of the light waves in a way that produces complex interference effects. No amount of refinement of the lenses used in a microscope can overcome this limitation imposed by the wave -like nature of light.

The limiting separation at which two objects can still be seen as distinct - the so-called resolution limit - depends on both the wavelength of the light and the numerical aperture of the lens system used. Under optimal conditions, with violet light (wavelength l = 0.4 mm) and a numerical aperture of 1.4, a resolution limit of just under 0.2 mm can theoretically be obtained in the light microscope. Although it is possible to magnify an image as much as one wants, it is never possible to resolve two objects in the light microscope that are separated by less than about 0.2 mm.

1.1.1. Living cells can be visualized in a phase-contrast (PC) or differential-interfer ence-contrast (DIC) microscope

When light passes through a living cell, the phase of the light wave is changed according to the cell's refractive index: light passing through a relatively thick or dense part of the the cell, such as the nucleus, is retarded and its phase consequently shifted relative to light that has passed through an adjacent thinner region of the cytoplasm . Both the phase-contrast microscope and Nomarski's differential-interference-contrast microscope exploit the interference effects produced when these two sets of waves recombine, thereby creating an image of the cell's structure.

Fig. 1-2: A fibroblast in tissue culture visualized with four types of light microscopy. The image in (A) was obtained by the simple transmission of light through the cell, a technique known as bright-field microscopy. (B) Phase-contrast microscopy; (C) differential-interference-contrast microscopy; and (D) dark-field microscopy.

A simpler way to see some of the features of a living cell is to observe the light that is scattered by its various components. In the dark-field microscope , the illuminating rays of light are directed to one side, so that only scattered light enters the objective lens of the microscope. Consequently, the cell appears as an illuminated object against a dark background. Images of the same cell obtained by 4 different kinds of light microscopy are shown in Fig. 1-2.

One of the great advantages of phase-contrast, differential-interference-contrast, and dark -field microscopy is that they make it possible to watch the movements involved in such processes as mitosis and cell migration.

1.1.2. Specific structures can be located in cells by fluorescence microscopy

The fluorescence microscope is similar to an ordinary light microscope except that the illuminating light, from a powerful mercury or xenon lamp, is passed through 2 sets of filters - one to intercept the light before it reaches the specimen and one to filter the light emitted from the specimen upon excitation by the input beam. The first filter is selected so that it only allows those the wavelengths that excite the particular fluorescent dye to pass, while the second filter blocks out this light and passes only those wavelengths emitted by the fluorescent dye.

Fig. 1-3: Organization of the actin filaments in a resting cultured fibroblast visualized by fluorescence microscopy. The distinct actin filament bundles ('stress' fibers) representing part of the cell's cytoskeleton have been labeled specifically with the mushroom poison phalloidin that has been made flurescent by coupling it covalently to the fluorochrome rhodamine.

Fluorescence microscopy is most often used to detect specific proteins or molecules that have been made fluorescent by coupling them covalently to a fluorescent dye. For example, fluorescent dyes can be coupled to antibody molecules, which serve as highly specific and versatile staining reagents by binding selectively to other macromolecules on the surface of a living cell or inside a micro-injected or fixed cell. Two commonly used fluorescent dyes are fluorescein, which emits an intense yellow-green fluorescence when excited with blue light, and rhodamin, which emits a deep red fluorescence when excited with green-yellow light. By coupling one molecule to fluorescein and another to rhodamine, the distributions of different molecules can be followed simultaneously in the same cell by switching back and forth between two sets of filters, each specific for one dye. A fluorescence micrograph is shown in Fig. 1-3, revealing the rhodamin -phalloidin labelled actin stress fiber system in a resting cultured fibroblast.

1.1.3. Seeing beyond the diffraction-limited resolution by video-enhanced light microscopy

In recent years, video cameras in concert with image processing have circumvented two fundamental limitations of the human eye and thus have had a major impact on light microscopy: (1) the eye cannot see in extremely dim light; and (2) the eye cannot perceive small differences in light intensity against a bright background. The first of these limitations can be overcome by attaching a highly light-sensitive video camera (e.g., a SIT- or a CCD-camera) to a microscope. It is then possible to observe cells for long periods at very low light levels, thereby avoiding the damaging effects of prolonged bright light - and heat - exposure. Such image-intensification systems are especially important for viewing fluorescent molecules in living cells.

Because images captured by video cameras are in electronic form, they can be readily digitized, fed into a computer, and processed in various ways to extract latent information. Such image processing makes it possible to compensate for various optical faults so as to attain the theoretical limit of resolution. Moreover, by linking video systems to image processors, the signal-to-noise ratio, and thus the contrast can be greatly enhanced, e.g., by frame accumulation, thereby overcoming the eye's limitations to detect small contrast differences. Although this processing also enhances the effects of random background irregularities, this 'noise' can be removed by electronically subtracting an image of a blank area of the field, so-called background subtraction. Small transparent objects then become visible that were previously impossible to distinguish from the background.

The high contrast attainable by computer-assisted differential-interference-contrast microscopy makes it even possible to see objects whose dimensions are significantly below the diffraction -limited resolution of the microscope, e.g., single microtubules which are 25 nm in diameter (see Fig. 1-4) .

Fig. 1-4: Light-microscope images of unstained microtubules that have been visualized by video-enhanced differential-interference-contrast microscopy. (A) shows the original un-processed video-image, and (B) displays the final result after electronic 'frame accumulation' and 'background subtraction', processing steps that greatly enhance the contrast and reduce the noise. Although the microtubules are only 25 nm in diameter, their apparent width is given by the diffraction-limited resolution of the otical system, i.e. about 200 nm.

In fact, individual microtubules - and even single actin filaments with a diameter of only 9 nm - can also be seen in a fluorescence microscope if they are fluorescently labeled. In both cases, however, the unavoidable diffraction effects badly blur the image, so that the microtubules appear at least 0.2 mm wide - instead of 25 nm, thus making it impossible to distinguish a single microtubule from a bundle of several microtubules on the basis of its apparent width.

1.1.4. Cells can be 'optically sectioned' by confocal imaging

Seldom has the introduction of a new instrument generated as instant an excitement among biologists as the laser-scanning confocal microscope . With this new microscope one can (i) slice incredibly clean, thin optical sections out of thick fluorescent specimens (see Fig. 1-5, a-e), (ii) view specimen planes parallel to the line of sight , (iii) penetrate deep into light-scattering tissues, (iv) gain impressive 3-D views at very high resolution (see Fig. 1-5f), and (v) improve the precision of microphotometry.

The emergence of this microscope stems from several roots, including light microscopy, confocal imaging, video and scanning microscopy, and coherent or laser-illuminated optics. In the confocal microscope , the conventional microscope condenser is replaced by a lens identical to the objective lens. The field of illumination is limited by a pinhole, positioned on the optical axis of the microscope. A reduced image of this pinhole is projected onto the specimen by the 'condenser'. The field of view is also restricted by a second - or exit - pinhole in the image plane placed confocally to the illuminated spot in the specimen and to the first pinhole. Instead of trans-illuminating the specimen with a separate condenser and objective lens (e.g., in bright -field, phase-contrast or differential-interference-contrast), the confocal microscope may also be used in the epi-illuminating mode (e.g., fluorescence or reflection), thus making the single objective lens serve both as the condenser and objective lens.

Fig. 1-5: 3-D confocal fluorescence microscopy of mitotic cells. (a-e) are single confocal optical sections recorded from one and the same mitotic spindle, stained for microtubules. (f) is a through focus projection computed from the series of confocal optical sections shown in (a-e). The corresponding DNA can be imaged with the spindle apparatus by acridine orange labeling. The composite images (g-i) are simultaneous labeling of tubulin and DNA in anaphase (g), late anaphase (h) and telophase (i).

With either mode of confocal illumination, the specimen is scanned by a point of light by either one of the following three mechanisms:

(1) In the stage scanning confocal microscope the specimen is moved over short distances in a raster-like pattern.

(2) In the tandem scanning confocal microscope (TSM), holes on a portion of an opaque spinning disk (a 'Nipkow disk') in front of the light-source collector lens, are imaged onto the specimen by the objective lens. Each point of light, reflected or scattered by the specimen thus illuminated, is focused by the same objective lens onto a centro-symmetric portion of the Nipkow disk. The pinholes at this region exclude the light originating from points in the specimen not illuminated by the first set of pinholes, thus giving rise to confocal illumination.

(3) In the confocal laser scanning microscope (CLSM) a focused spot of a laser beam is made to scan the specimen field as a 'flying spot' microscope.

The laser scanning confocal microscope is currently the most commonly employed in biological investigations and is especially useful for fluorescent imaging. Careful filtering and selection of wavelengths allows dual-labeled specimens to be imaged. Application of acousto-optical scanning devices for moving the fine laser beam across the specimen promises to increase the scanning speed of this system.

In summary, confocal microscopy is a powerful new tool for biologists to examine cellular structure and function, and it complements conventional light and electron microscopy. CLSM and TSM offer a convenient means of examining both living and fixed specimens labeled with multiple fluorochromes to determine the 3-D organization of various cellular components. New advances in microprocessor, memory, and mass-storage technology have produced powerful computers which can rapidly display and analyze large 3-D data sets. The potential for this combination of technologies is enormous, especially when considering 3-D volume renderings of living cells and tissues over time . The application of these new technologies will, no doubt, permit the biologist to observe previously unseen phenomena and explore the subtle relationships between cellular structure and function .

1.2. Electron Microscopy

The relationship between the limit of resolution and the wavelength of the illuminating light holds true for any form of radiation, whether is is a beam of light or a beam of electrons. With electrons the limit of resolution can be made very small, e.g., 0.004 nm with 100-kV electrons. However, because the abberations of an electron lens are considerably harder to correct than those of a glass lens, the practical resolving power of most modern EMs is, at best, 0.1 nm or 1 Å.

Table 1-2 Major Events in the Development of the Electron Microscope and Its Applica tions to Cell Biology

1897 J.J. Thomson announced the existence of negatively charged particles, later termed e lectrons.

1924 de Broglie proposed that a moving electron has wavelike properties.

1926 Busch proved that is was possible to focus a beam of electrons with a cylindrical magnetic lens, laying the foundations of electron optics.

1931 Ruska and colleagues built the first transmission electron microscope.

1935 Knoll demonstrated the feasibility of the scanning electron microscope; three years later a prototype instrument was built by Von Ardenne.

1939 Siemens produced the first commercial transmission electron microscope.

1944 Williams and Wyckoff introduced the metal shadowing technique.

1945 Porter, Claude and Fullam used the electron microscope to examine cells in tissue culture after fixing and staining them with OsO4.

1948 Pease and Baker reliably prepared thin sections (0,1 to 0,2 mm thick) of biological material.

1952 Palade, Porter and Sjöstrand developed methods of fixation and thin sectioning that enabled many intracellular structures to be seen for the first time. In one of the first applications of these techniques, H.E. Huxley showed that skeletal muscle contains overlapping arrays of protein filaments, supporting the "sliding filament" hypothesis of muscle contraction.

1953 Porter and Blum developed the first widely accepted ultramicrotome, incorporating many features introduced by Claude and Sjöstrand previously.

1956 Glauert and associates showed that the epoxy resin Araldite was a highly effective embedding agent for electron microscopy. Luft introduced another embedding resin, Epon, five years later.

1957 Robertson described the trilaminar structure of the cell membrane, seen for the first time in the electron microscope.

1957 Freeze-fracture techniques, initially developed by Steere, were perfected by Moor and Mühlethaler. Later (1966), Branton demonstrated that freeze-fracture allows the interior of the membrane to be visualized.

1959 Singer used antibodies coupled to ferritin to detect cellular molecules in the electron micro scope.

1959 Brenner and Horne developed the negative staining technique, invented four years previously by Hall, into a generally useful technique for visualizing viruses, bacteria and protein fila ments.

1963 Sabatini, Bensch and Barrnett introduced glutaraldehyde (usually followed by OsO 4) as a fixative for electron microscopy.

1965 Cambridge Instruments produced the first commercial scanning electron microscope.

1968 de Rosier and Klug described techniques for the reconstruction of three-dimensional struc tures from electron micrographs.

1975 Henderson and Unwin determined the first structure of a membrane protein by computer -based reconstruction from electron micrographs of unstained samples.

1979 Heuser, Reese, and colleagues developed a high-resolution, deep-etching technique based on very rapid freezing.

Furthermore, problems of specimen preparation, contrast , and radiation damage effectively limit the typical resolution for biological specimens to about 2 nm - still about 100x better than the resolution of the light microscope.

1.2.1. Design of the electron microscope

As illustrated in Fig. 1-6, in overall design, the transmission electron microscope (TEM) is similar to that of a light microscope. Since electrons, the illumination of an EM, are strongly scattered by their collisions with air molecules, the EM has to be run under high-vacuum conditions.

1.2.2. Preparation of biological material for electron microscopy

Since any biological specimen will be exposed to a very high vacuum in the EM, there is no possibility of viewing it in the living, wet state . Tissues and whole cells are usually preserved by chemical fixation - first with glutaraldehyde, which covalently crosslinks protein molecules to each other, followed by osmium tetroxide, which binds to and stabilizes lipid bilayers as well as proteins. Because electrons have a very limited penetrating power , the fixed tissues or cells normally have to be cut into thin sections - typically 30-100 nm thick - before they can be viewed in the TEM. This is achieved by dehydrating the specimen and permeating it with a monomeric resin that polymerizes to form a solid block of plastic. The block is then cut with a fine glass or diamond knife on a device called microtome. The resulting thin sections, free of water and other volatile solvents, are finally placed on a 3-mm diameter circular metal grid for viewing in the electron microscope. An example of such a thin section is shown in Fig. 1-7.

Fig. 1-6: Principal features of a light microscope, a transmission electron microscope (TEM), and a scanning electron microscope (SEM), drawn to emphasize the similarities of overall design. The TEM and SEM require that the specimen be placed in a high -vacuum environment.

Fig. 1-7: Traditional thin section of a cultured fibroblast that was quick-frozen by slamming it onto a liquid helium -cooled copper block, and then freeze-substituted. Within the 'stress fibers' (SF; see also Fig. 1-3), individual actin filaments (see also Fig. 8) are clearly resolved. The unaggregated, wider filaments (MT) are microtubules, and the small dark particles (R) represent ribosomes.

Contrast in the EM depends on the atomic number of the atoms in the specimen: the higher the atomic number, the more electrons are scattered, and the greater is the contrast. Hence, to make biomacromolecules - that are composed mainly of carbon, oxygen, nitrogen, and hydrogen - visible, they are usually impregnated - or stained - with heavy metal salts containing osmium, uranium, or lead.

In some cases, specific macromolecules can be located in thin sections by labeling techniques adapted from light microscopy. Certain enzymes in cells can be detected by incubating the specimen with a substrate whose reaction leads to the local deposition of an electron-dense precipitate. Alternatively, antibodies can be coupled - instead of to a fluorescent probe as used in light microscopy (see Fig. 1-3) - to an indicator enzyme (usually peroxidase) or to an electron-dense marker (usually colloidal gold spheres), and then used to locate the macromolecules that the antibodies recognize.

1.2.3. Negative staining allows supramolecular assemblies and individual macromol ecules to be viewed at high resolution

Although individual macromolecules, such as DNA or proteins, can be visualized in the EM if they are shadowed with a heavy metal to provide contrast (see A2.4. and Fig. 1-9), finer detail can often be gathered by negative staining. Here macromolecules or supramolecular assemblies are adsorbed from their suspension onto a thin carbon support film and then briefly washed with an aqueous solution of a heavy metal salt (e.g., 1% uranyl acetate or 2% Na-phosphotungstate), before the sample is allowed to dry. A very thin film of metal salt now covers the support film everywhere except where it has been excluded by the presence of an adsorbed macromolecule or supramolecular structure. Because the macromolecule allows the electrons to pass much more readily than does the surrounding heavy metal film, a reversed or negative image of the molecule is created, hence the name 'negative staining'. Negative staining is particularly useful for viewing large supramolecular assemblies such as viruses or ribosomes, but also for seeing the subunit structure of protein filaments, e.g., actin filaments (see Fig. 1-8).

Regardless of the method employed, a single protein molecule gives only a weak and ill -defined image in the EM. Efforts to get better information by increasing the electron dose to record an image with a better signal-to-noise ratio are self-defeating because, due to the high radiation sensitivity of biological material, they damage and disrupt the object under examination. Therefore to visualize the molecular details, it is necessary to combine the information obtained from many molecules in such a way as to average out the noise and random errors in the individual particle images. This is possible for ordered supramolecular structures such as viruses or protein filaments (see Fig. 1-8), in which the individual subunits are arranged in a regular fashion relative to each other. Often proteins can also be induced to form regular arrays (e.g., helical structures or 2-D crystalline arrays) in vitro . Given an electron micrograph of either type of ordered protein array, image processing techniques can be used to compute the average image of an individual protein molecule. If projection images viewing the protein molecule from different angles are available, a 3-D reconstruction of the molecule can be computed (see Fig. 1-8, inset).

Fig. 1-8: Electron micrograph of negatively stained 'synthetic' actin filaments. Actin filaments are double-helical polymers consisting of two linear strands of 42-kD actin subunits that are helically twisted around each other. The inset reveals a 3-D volume-rendered representation of a filament 3-D reconstruction.

1.2.4. Metal shadowing reveals surface features at high resolution in the transmis sion electron microscope

As illustrated in Fig. 1-9, the TEM can be used to study the surface of a specimen at high resolution, allowing individual biomacromolecules to be visualized. To achieve this, a thin film (e.g., 0.1-0.5 nm) of a heavy metal such as platinum or tungsten is evaporated - either unidirectionally or by rotating the specimen stage at an angle relative to the evaporation source - onto the air- or freeze-dried specimen. If the metal is evaporated unidirectionally onto the specimen, then a coating that, depending on the surface topography of the specimen, is thicker in some places than in others is obtained - a process known as shadowing because a shadow effect is generated that gives the image a 3-D appearance.

Some specimens coated in this way are thin enough or small enough for the electron beam to penetrate them directly; this is the case for individual macromolecules (see Fig. 1-9), viruses, and cell walls. For thicker specimens the organic material of the cell or tissue must be digested away after shadowing so that only the thin metal replica of the surface of the specimen is left. Usually, the replica is reinforced with a film of carbon so that it can be placed on a grid and examined in the TEM.

Fig. 1-9: Electron micrograph of a mixture of myosin (M) and nuclear lamin (L) dimers after glycerol spraying/rotary metal shadowing with plati num. Both molecules are composed of two globular heads linked to a common rod-like tail, approximately 100 nm long in the case of myosin and 52 nm in the case of nuclear lamin.

1.2.5. Freeze-fracture and freeze-etch electron microscopy allow one to look into the cell interior

Two methods that employ metal replicas have been particularly useful in cell biology. One of these, freeze-fracture EM, provides a means of visualizing the interior of cell membranes. Cells are frozen, and then the frozen block is cracked with a knife blade. The fracture plane often passes through the hydrophobic middle of lipid bilayers, thereby exposing the interior of cell membranes. The resulting fracture faces are shadowed with platinum, the organic material is dissolved away, and the replicas are floated off, put on an supporting grid and viewed in the EM. As shown in Fig. 1-10, such replicas are studded with small bumps, called intramembrane particles, which represent membrane proteins or protein complexes.

Another important and related replica method is freeze-etch EM, which can be used to examine either the exterior or interior of cells. In this technique, the cells are again frozen at low temperature (e.g., -196°C), and the frozen block is cracked with a knife blade. But now the ice level is lowered around the cells - and to a lesser extent within the cells - by the sublimation of water in a vacuum as the temperature is raised, a process called freeze-drying. The parts of the cell exposed by this etching process are then shadowed with platinum as before to generate a metal replica.

1.2.6. 3-D images of surfaces can be obtained by scanning electron microscopy (SEM)

Thin sections are effectively 2-D slices of tissue or cells and therefore fail to convey the 3-D arrangement of cellular components. Although the third dimension can be reconstructed from hundreds of serial sections, this is a lengthy and tedious process.

Fortunately, there are more direct means to obtain a 3-D image. One is to examine a specimen in a scanning electron microscope (SEM). As illustrated schematically in Fig. 1-6, whereas the TEM uses the electrons that have passed through the specimen to form an image, the SEM uses the the electrons that are back-scattered or emitted (i.e., secondary electrons) from the specimen surface. The specimen to be examined in the SEM is fixed, dried (e.g., freeze-dried or critical point dried), and coated (e.g., sputtered) with a thin layer of metal to render the specimen surface conductive. The specimen is then scanned with a narrow beam of electrons . The quantity of electrons scattered or emitted as the primary beam bombards each successive point of the metallic specimen surface is measured by an electron detector and used to modulate the intensity of a second electron beam - which moves in synchrony with the primary beam - and forms an image on a television screen. As shown in Fig. 1-11, in this way, a highly enlarged and topographic image of the surface as a whole is built up.

Fig. 1-10: Freeze-fracture electron micrograph of the thylakoid membranes from the chloroplast of a plant cell. These membranes, which carry out photo -synthesis, are stacked up in multiple layers. The plane of the fracture has moved from layer to layer, passing through the middle of each lipid bilayer and exposing transmembrane proteins that have sufficient bulk in the interior of the bilayer to cast a shadow and show up as intra-membrane particles in this platinum replica.

The SEM technique provides a tremendous depth of focus ; moreover, since the amount of electron scattering depends on the angle of the specimen surface relative to the incident electron beam, the image has highlights and shadows that give it a 3-D appearance (see Fig. 1-11). Only surface features can be examined, however, and in most SEMs the resolution attainable is not very high, typically 10 nm with an effective magnification of up to 20,000x. As a result, the SEM is primarily used to study whole cells and tissues rather than sub-cellular organelles.

Fig. 1-11: SEM image of an 'ugly bug' (a louse fly of an alpine bird, the wallglider) formed by the secondary electrons emitted from the specimen's surface.

1.3. Scanning Probe Microscopy (SPM)

The invention of the scanning tunneling microscope (STM) in the early 1980s by G. Binnig and H. Rohrer has initiated an exciting series of novel local probe microscopes that image the surfaces of conducting as well as insulating solids with atomic resolution. These novel local-probe microscopies provide complementary structural and chemical information neither accessible by conventional microscopies nor by X-ray diffraction.

With a SPM one generates scanning probe micrographs by moving a nm-sized sensor (the 'probe') along an x,y-raster over the sample surface and writing the sensor signal to a storage device such as a television screen. Depending on the sensor and mode of operation, micrographs provide information on the topography, the electronic structure , and the mechanical or thermal properties of solid surfaces. The resolution is determined by the sharpness of the sensing tip, which can ultimately consist of a single atom, and by the precision of the scanning device.

Fig. 1-12: Surface topography of metal-coated bacteriophage T4 polyheads recorded by STM (a,c) and by TEM (b,d). Raw data, topograph recorded in the STM (a), and shadowgraph in the TEM (b). Surface reliefs (c,d) computed (c) from STM topograph in (a), and(d) from TEM shadowgraph in (b).

As illustrated in Fig. 1-12, clearly, these possibilities make SPMs particularly attractive for biological research because they may ultimately permit investigations of biological surfaces at molecular resolution in their native environment.

While the application of the STM in material sciences has progressed substantially, breakthroughs in biological sciences have been more sparse. These breakthroughs include intensely debated high-resolution (1 nm in a few cases) images of DNA, and a number of low-resolution (2-3 nm) topographs of air-dried uncoated supramolecular assemblies. On the other hand, a series of promising results have been obtained recently using the atomic force microscope (AFM), indicating that molecular resolution with biological material in physiological buffers is feasible.

It appears likely that the potentials of SPMs will eventually become routinely exploitable, and that SPM will play a decisive role in the direct assessment of structure-function relationships of biological systems. Two key aspects beyond refinement of the instruments, however, need to be considered seriously: (1) Biological macromolecules and their supramolecular assemblies are very fragile structures requiring dedicated sample preparation procedures to be worked out. (2) More than commonly anticipated, direct comparison with results from conventional imaging methods such as EM is absolutely crucial to enhance our understanding of SPM images.