专题2 细胞生物学研究方法

For well over 100 years, all microscopes were constrained by a fundamental limitation: that a given type of radiation cannot be used to probe structural details much smaller than its own wavelength. A limit to the resolution of a light microscope was therefore set by the wavelength of visible light, which ranges from about 0.4 μm (for violet) to 0.7 μm (for deep red). In practical terms, bacteria and mitochondria, which are about 500 nm (0.5 μm) wide, are generally the smallest objects whose shape we can clearly discern in a standard light microscope;

Because of its wave nature, light does not follow the idealized straight ray paths that geometrical optics predicts. Instead, light waves travel through an optical system by many slightly different routes, like ripples in water, so that they interfere with one another and cause optical diffraction effects. If two trains of waves reaching the same point by different paths are precisely in phase, with crest matching crest and trough matching trough, they will reinforce each other so as to increase brightness. In contrast, if the trains of waves are out of phase, they will interfere with each other in such a way as to cancel each other partly or entirely (Figure 9–3).

The limiting separation at which two objects appear distinct—the so-called limit of resolution—depends on both the wavelength of the light and the numerical aperture of the lens system used. The numerical aperture affects the light-gathering ability of the lens and is related both to the angle of the cone of light that can enter it and to the refractive index of the medium the lens is operating in; the wider the microscope opens its eye, so to speak, the more sharply it can see (Figure 9–5). The refractive index is the ratio of the speed of light in a vacuum to the speed of light in a particular transparent medium. For example, for water this is 1.33, meaning that light travels 1.33 times slower in water than in a vacuum.

Under the best conditions, with violet light (wavelength = 0.4 μm) and a numerical aperture of 1.4, the basic light microscope can theoretically achieve a limit of resolution of about 0.2 μm, or 200 nm. Some microscope makers at the end of the nineteenth century achieved this resolution, but it is routinely matched in contemporary, factory-produced microscopes. Although it is possible to enlarge an image as much as we want—for example, by projecting it onto a screen—it is not possible, in a conventional light microscope, to resolve two objects in the light microscope that are separated by less than about 0.2 μm; they will always appear as a single object.

It is important, however, to distinguish between resolution and detection. If a small object, below the resolution limit, itself emits light, then we may still be able to see or detect it. Thus, we can see a single fluorescently labeled microtubule even though it is about 10 times thinner than the resolution limit of the light microscope. Diffraction effects, however, will cause it to appear blurred and at least 0.2 μm thick (see Figure 9–14). In a similar way, we can see the stars in the night sky, even though their diameters are far below the angular resolution of our unaided eyes: they all appear as similar, slightly blurred points of light, differing only in their color and brightness.

There are many ways in which contrast in a specimen can be generated (Figure 9–6). While fixing and staining a specimen can generate contrast through color (Figure 9–6A), microscopists have always been challenged by the possibility that some components of the cell may be lost or distorted during specimen preparation. The only certain way to avoid the problem is to examine cells while they are alive, without fixing or freezing. For this purpose, light microscopes with special optical systems are especially useful. In the normal bright-field microscope, light passing through a cell in culture forms the image directly. Another system, dark-field microscopy, exploits the fact that light rays can be scattered in all directions by small objects in their path.

If oblique lighting from the condenser is used, which does not directly enter the objective, unstained objects in a living cell can scatter the rays, some of which then enter the objective to create a bright image against a black background (Figure 9–6B). When light passes through a living cell, the phase of the light wave is changed according to the cell’s refractive index: a relatively thick or dense part of the cell, such as a nucleus, slows the light passing through it. The phase of the light, consequently, is shifted relative to light that has passed through an adjacent thinner region of the cytoplasm (Figure 9–6C). The phase-contrast microscope and, in a more complex way, the differential-interference-contrast microscope increase these phase differences to produce amplitude differences, or contrast, when the sets of waves recombine, thereby creating an image of the cell’s structure.

Both types of light microscopy are widely used to look at living cells (see Movie 17.2). Figure 9–7 compares images of the same cell obtained by four kinds of light microscopy. Phase-contrast, differential-interference-contrast, and dark-field microscopy make it possible to watch the movements involved in such processes as mitosis and cell migration. Because many cellular motions are too slow to be seen in real time, it is often helpful to make time-lapse videos in which the camera records successive frames separated by a short time delay, so that when the resulting picture series is played at normal speed, events appear greatly speeded up.

Digital imaging systems have also circumvented two fundamental limitations of the human eye: the eye cannot see well in extremely dim light, and it cannot perceive small differences in light intensity against a bright background. To increase our ability to observe cells in these difficult conditions, we can attach a sensitive digital camera to a microscope. These cameras detect light by means of high-sensitivity complementary metal-oxide semiconductor (CMOS) sensors, similar to those now found in digital cameras and smartphones. Such image sensors can count individual photons and are many times more sensitive than the human eye and can detect 100 times more intensity levels. It is therefore possible to observe cells for long periods at very low light levels, thereby avoiding the damaging effects of prolonged bright light (and heat). Such sensitive detectors are especially important for viewing fluorescent molecules in living cells, as explained later.

Looking at individual living cells in culture is relatively easy, but most cells are found in complex tissues and organs, and this forces another trade-off when we want to look at them. Because most tissue samples are too thick for their individual cells to be examined directly at high resolution, they are often cut into very thin transparent slices, or sections. To preserve the cells within the tissue they must first be treated with a fixative. A common fixative is glutaraldehyde, which forms covalent bonds with the free amino groups of proteins, cross-linking them so they are stabilized and locked into position.

Because tissues are generally soft and fragile, even after fixation, they need to be either frozen or embedded in a supporting medium before they can be sectioned. The usual embedding media are waxes or resins. In liquid form, these media both permeate and surround the fixed tissue before being hardened (by cooling or by polymerization) to form a solid block, which is readily sectioned with a microtome. This is a machine with a sharp blade, usually of steel or glass, which operates like a meat slicer (Figure 9–8). The sections (typically 0.5–10 μm thick) are then laid flat on the surface of a glass microscope slide.

There is little in the contents of most cells (which are 70% water by weight) to impede the passage of light rays. Thus, most cells in their natural state, particularly if fixed and sectioned, are almost invisible in an ordinary light microscope. We have seen that cellular components can be made visible by techniques such as phase-contrast and differential-interference-contrast microscopy, but these methods tell us almost nothing about the underlying chemistry.

There are three main approaches to working with thin tissue sections that reveal differences in the types of molecules that are present. First, and traditionally, sections can be stained with organic dyes that have some specific affinity for particular subcellular components. The dye hematoxylin, for example, has an affinity for negatively charged molecules and therefore reveals the general distribution of DNA, RNA, and acidic proteins in a cell (Figure 9–9). The chemical basis for the specificity of many dyes, however, is not known, although they are used widely in hospital laboratories.

Fluorescent molecules absorb light at one wavelength and emit it at another, longer wavelength (Figure 9–10A and B). If we illuminate such a molecule at its absorbing wavelength and then view it through a filter that allows only light of the emitted wavelength to pass, it will glow against a dark background. Because the background is dark, even a minute amount of the glowing fluorescent dye can be detected. In contrast, the same number of molecules of a nonfluorescent stain, viewed conventionally, would be practically indiscernible because the absorption of light by molecules in the stain would result in only the faintest tinge of color in the light transmitted through that part of the specimen.

The fluorescent dyes used for staining cells are visualized with a fluorescence microscope. This microscope is similar to an ordinary upright or inverted light microscope except that the illuminating light, from a very powerful source, is passed through two sets of filters—one to filter the light before it reaches the specimen, and one to filter the light obtained from the specimen. The first filter passes only the wavelengths that excite the particular fluorescent dye, while the second filter blocks out this light and passes only those wavelengths emitted when the dye fluoresces (Figure 9–10C).

A versatile and widely used technique is to couple fluorescent dyes to antibody molecules, which then serve as highly specific and versatile staining reagents that bind selectively to the particular macromolecules they recognize in cells or in the extracellular matrix. Two fluorescent dyes that have been commonly used for this purpose are fluorescein, which emits an intense green fluorescence when excited with blue light, and rhodamine, which emits a deep red fluorescence when excited with green-yellow light (Figure 9–12). By coupling one antibody to fluorescein and another to rhodamine, the distributions of different molecules can be compared in the same cell; the two molecules are visualized separately in the microscope by switching back and forth between two sets of filters, each specific for one dye.

Antibodies are proteins produced by the vertebrate immune system as a defense against infection (discussed in Chapter 24). They are unique among proteins in that they are made in billions of different forms, each with a different binding site that recognizes a specific target molecule (or antigen). The precise antigen specificity of antibodies makes them powerful tools for the cell biologist. When chemically coupled to fluorescent dyes, antibodies are invaluable for locating specific molecules in cells by fluorescence microscopy (Figure 9–14). When labeled with electron-dense particles such as colloidal gold spheres, they are used for similar purposes in the electron microscope (discussed later).

All of the fluorescent molecules discussed so far are made outside the cell and then artificially introduced into it. But the use of genes encoding protein molecules that are themselves inherently fluorescent also enables the creation of organisms and cell lines that make their own visible tags and labels, without the introduction of foreign molecules. These cellular exhibitionists display their inner workings in glowing fluorescent color. Foremost among the fluorescent proteins used for these purposes by cell biologists is the green fluorescent protein (GFP), isolated from the jellyfish Aequorea victoria.

This protein is encoded by a single gene, which can be cloned and introduced into cells of other species. The freshly translated protein is not fluorescent, but within an hour or so (less for some alleles of the gene, more for others) some of the amino acids undergo a self-catalyzed post-translational modification to generate an efficient fluorochrome, shielded within the interior of a barrel-like protein, which will now fluoresce green when illuminated appropriately with blue light (Figure 9–16).

One of the simplest uses of GFP is as a reporter molecule, a fluorescent probe to monitor gene expression. A transgenic organism can be made with the GFP-coding sequence placed under the transcriptional control of the promoter belonging to a gene of interest, giving a directly visible readout of the gene’s expression pattern in the living organism (Figure 9–17). In another application, a peptide location signal can be added to the GFP to direct it to a particular cell compartment, such as the endoplasmic reticulum or a mitochondrion (see Figure 9–25B), lighting up these organelles so they can be observed in the living state.

The GFP DNA-coding sequence can also be inserted at the beginning or end of the coding sequence for another protein, yielding a chimeric product consisting of that protein with a new GFP domain attached. In many cases, this GFP fusion protein behaves in the same way as the original protein, directly revealing its location and activities by means of its genetically encoded fluorescence (Figure 9–18).

First, interactions between one protein and another can be monitored by Förster resonance energy transfer, also called fluorescence resonance energy transfer but both abbreviated to FRET. In this technique, two molecules of interest are each labeled with a different fluorochrome, chosen so that the emission spectrum of one fluorochrome, the donor, overlaps with the absorption spectrum of the other, the acceptor. If the two proteins interact in such a way as to bring their fluorochromes into very close proximity (closer than about 5 nm), one fluorochrome, when excited, can transfer energy from the absorbed light directly (by resonance, nonradiatively) to the other. Thus, when the complex is illuminated at the excitation wavelength of the first fluorochrome, fluorescent light is produced at the emission wavelength of the second (Figure 9–19).

The FRET can be measured by quantifying the reduction of the donor fluorescence in the presence of the acceptor. The efficiency of FRET is inversely proportional to the sixth power of the distance between the donor and acceptor molecules and so is extremely sensitive to small changes in distance. The genes encoding GFP and related fluorescent proteins can be engineered to produce protein variants, usually with one or more amino acid changes, that fluoresce only weakly under normal excitation conditions but can be induced to fluoresce either more strongly or with a color shift (for example, from green to red) by activating them with a strong pulse of light at a different wavelength in a process called photoactivation.

Another way to exploit GFP fused to a protein of interest is known as fluorescence recovery after photobleaching (FRAP). Here, a strong focused beam of light from a laser is used to extinguish the GFP fluorescence in a specified region of the cell, after which one can analyze the way in which remaining unbleached fluorescent protein molecules move into the bleached area as a function of time. This technique, like photoactivation, can deliver valuable quantitative data about a protein’s kinetic parameters, such as diffusion coefficients, active transport rates, or binding and dissociation rates from other proteins (Figure 9–20).

Many biosensors use two connected fluorescent proteins that can be brought close enough together to undergo Förster resonance energy transfer (see Figure 9–19). Bringing them together, or indeed moving them apart, is a connecting sensor module. The sensor is usually a protein or protein domain that undergoes a large conformational change on binding to the target molecule. The general principle used to construct a genetically encoded biosensor is shown in Figure 9–22. Measuring the ratio of the intensities of light emitted by the two fluorescent proteins in the biosensor provides a quantitative measure of the concentration of the target molecule of interest. Many hundreds of such biosensors have been created. Some can monitor and measure small molecules in living cells, such as Ca2+, cAMP, IP3, NADPH (and hence redox state), H+ ions (and hence pH), and neurotransmitters such as acetylcholine and glutamate. Others can measure the activity of kinases, phosphatases, active caspases, and even temperature.

The fluorescence emitted from the illuminated material is collected at a suitable light detector and used to generate an image. A pinhole aperture is placed in front of the detector, at a position that is confocal with the illuminating pinhole; that is, precisely where the rays emitted from the illuminated point in the specimen come to a focus. Thus, the light from this point in the specimen converges on this aperture and enters the detector. By contrast, the light emitted from regions of the specimen that are out of focus is also out of focus at the pinhole aperture and is therefore largely excluded from the detector. To build up a two-dimensional image, data from each point in the plane of focus are collected sequentially by scanning across the field from one side to the other in a regular pattern of pixels and are displayed on a computer screen.

The variations on light microscopy we have described so far are all constrained by the classic diffraction limit to resolution described earlier; that is, to about 0.2 μm, or 200 nm (see Figure 9–5). Yet many cellular structures—from nuclear pores and ribosomes to nucleosomes and clathrin-coated pits—are much smaller than this and so are unresolvable by conventional light microscopy. However, several approaches are now available that bypass the limit imposed by the diffraction of light, and some can now successfully resolve objects as small as 10 nm, a remarkable, twentyfold improvement. The first of these so-called superresolution approaches, structured illumination microscopy (SIM), is a fluorescence imaging method with a resolution of about 100 nm, or twice the resolution of conventional bright-field microscopy. SIM overcomes the diffraction limit by using a grated or structured pattern of light to illuminate the sample.

Because the excitation spot is blurred according to the point spread function, fluorescent molecules that are closer than about 200 nm will be imaged as a single blurred spot. One approach to increasing the resolution is to switch all the fluorescent molecules at the periphery of the blurry excitation spot back to their ground state or to a state where they no longer fluoresce in the normal way, leaving only those at the very center to be recorded. This can be done in practice by adding a second, very bright laser beam that wraps around the excitation beam like a torus. The wavelength and intensity of this second beam are adjusted so as to switch the fluorescent molecules off everywhere except at the very center of the point spread function, a region that can be as small as 20 nm across (Figure 9–30).

If a single fluorescent molecule is imaged, it appears as a circular blurry disc about 200 nm across, but if sufficient photons have contributed to this image, then the precise mathematical center of the disc-like image, and therefore the position of that fluorescent molecule, can be determined very accurately, often to within a few nanometers (Figure 9–31). But the problem with a specimen that contains a large number of adjacent fluorescent molecules, as we saw earlier, is that they each contribute blurry, overlapping point spread functions to the image, making the exact position of any one molecule impossible to resolve. Another way around this limitation is to arrange for only a very few, clearly separated molecules to actively fluoresce at any one moment. The exact position of each of these can then be computed, before subsequent sets of molecules are examined.

By switching the fluorophores off and on sequentially in different regions of the specimen as a function of time, all the superresolution imaging methods described above allow the resolution of molecules that are much closer together than the 200-nm diffraction limit. In STED, the locations of the molecules are determined by using optical methods to define exactly where their fluorescence will be on or off. In PALM and STORM, individual fluorescent molecules are switched on and off at random over a period of time, allowing their positions to be accurately determined. PALM and STORM techniques have depended on the development of novel fluorescent probes that exhibit the appropriate switching behavior. STORM originally relied on photoswitchable dyes, while PALM used photoswitchable fluorescent proteins, but the general principle is the same for both.

A new specimen preparation technique, called expansion microscopy (ExM), does exactly that. The process starts by staining the fixed sample with fluorescent labels such as antibodies that target the molecules of interest. The labeled specimen is then treated with a chemical cross-linker and incubated with acrylate and acrylamide monomers. These monomers then polymerize to form a polyelectrolyte gel that simultaneously incorporates the cross-linked labels. With the labels covalently cross-linked to the polymer gel, and locked in their original relative positions, cellular material in the sample, predominantly proteins that might hinder subsequent expansion, is then carefully digested away. The gel containing the labeled specimen is now gently swollen by removing the buffer salts with water, so that it expands equally in all directions by between 4 and 10 times (Figure 9–34A).

This problem can be solved by the use of a special optical technique called total internal reflection fluorescence (TIRF) microscopy. In a TIRF microscope, laser light shines onto the cover-slip surface at the precise critical angle at which total internal reflection occurs (Figure 9–38A). Because of total internal reflection, the light does not enter the sample, and the majority of fluorescent molecules are not, therefore, illuminated. However, electromagnetic energy does extend, as an evanescent field, for a very short distance beyond the surface of the cover slip and into the specimen, allowing just those molecules in the layer closest to the surface to become excited. When these molecules fluoresce, their emitted light is no longer competing with out-of-focus light from the overlying molecules and can now be detected.

Light microscopy is limited in the fineness of detail that it can reveal. Microscopes using other types of radiation—in particular, electron microscopes—can resolve much smaller structures than is possible with visible light. This higher resolution comes at a cost: specimen preparation for electron microscopy is complex and it is harder to be sure that what we see in the image corresponds precisely to the original living structure. It is possible, however, to use very rapid freezing to preserve structures faithfully for electron microscopy.

The formal relationship between the diffraction limit to resolution and the wavelength of the illuminating radiation (see Figure 9–5) holds true for any form of radiation, whether it is a beam of light or a beam of electrons. With electrons, however, the limit of resolution is very small. The wavelength of an electron decreases as its velocity increases. In an electron microscope with an accelerating voltage of 100,000 V, the wavelength of an electron is 0.004 nm. In theory, the resolution of such a microscope should be about 0.002 nm, which is 100,000 times that of the conventional light microscope.

Because the aberrations of an electron lens are considerably harder to correct than those of a glass lens, however, the practical resolving power of modern electron microscopes is, even with careful image processing to correct for lens aberrations, about 0.05 nm (0.5 Å) (Figure 9–39). This is because only the very center of the electron lenses can be used, and the effective numerical aperture is tiny. Furthermore, problems of specimen preparation, contrast, and radiation damage have generally limited the normal effective resolution for biological objects to 1 nm (10 Å). This is nonetheless about 200 times better than the resolution of the light microscope.

In overall design, the transmission electron microscope (TEM) is similar to an inverted light microscope, albeit much larger (Figure 9–40). The source of illumination is a filament or cathode that emits electrons at the top of a cylindrical column about 2 m high. Because electrons are scattered by collisions with air molecules, air must first be pumped out of the column to create a vacuum. The electrons are then accelerated from the filament by a nearby anode and allowed to pass through a tiny hole to form an electron beam that travels down the column. Magnetic coils placed at intervals along the column focus the electron beam, just as glass lenses focus the light in a light microscope.

The specimen is put into the vacuum, through an airlock, into the path of the electron beam. As in light microscopy, the specimen can be stained—in this case, with electron-dense material. Some of the electrons passing through the specimen are scattered by structures stained with the electron-dense material; the remainder are focused to form an image. The image can be observed on a monitor or is typically recorded with a sensitive CMOS electron detector. Because the scattered electrons are lost from the beam, the dense regions of the specimen show up in the image as areas of reduced electron flux, which look dark.

In the early days of its application to biological materials, the electron microscope revealed many previously unimagined structures in cells. But before these discoveries could be made, electron microscopists had to develop new procedures for embedding, cutting, and staining tissues. Because the specimen is exposed to a very high vacuum in the electron microscope, living tissue is usually killed and preserved by chemical fixation. As electrons have very limited penetrating power, the fixed tissues normally have to be cut into extremely thin sections (25–100 nm thick, about 1/200 the thickness of a single cell) before they are viewed. This is achieved by dehydrating the specimen, permeating it with a monomeric resin that polymerizes to form a solid block of plastic, then cutting the block with a fine glass or diamond knife on a special microtome.

The best current approaches to this problem depend on rapid freezing. If an aqueous system is cooled fast enough and to a low enough temperature, the water and other components in it do not have time to rearrange themselves or crystallize into ice. Instead, the water is supercooled into a rigid but noncrystalline state—a “glass”—called vitreous ice. This rapid freezing is usually performed by plunging the sample into a coolant such as liquid ethane or by cooling it at very high pressure.

A compromise is to rapidly freeze the tissue, replace the water with organic solvents, embed the tissue in plastic resin, and finally cut sections. This approach, called freeze substitution, stabilizes and preserves the tissue in a condition very close to its original living state (Figure 9–42). Molecules in all kinds of thin sections can be labeled to identify and localize them. We have seen earlier how antibodies can be used in conjunction with fluorescence microscopy to localize specific macromolecules. An analogous method—immunogold electron microscopy—can be used in the electron microscope. The usual procedure is to incubate a thin section first with a specific primary antibody, and then with a secondary antibody to which a colloidal gold particle has been attached. The gold particle is electron-dense and can be seen as a black dot in the electron microscope (Figure 9–43).

Biological tissues are composed mainly of atoms of very low atomic number (primarily carbon, oxygen, nitrogen, and hydrogen). To make them more readily visible, tissues are often impregnated (before or after sectioning) with the salts of heavy metals such as uranium, lead, and osmium. The degree of impregnation, or “staining,” with these salts will vary for different cell constituents. Lipids, for example, tend to stain darkly after osmium fixation, revealing the location of cell membranes (see, for example, Figure 12–2 or Figure 12–15). Alternatively, if isolated molecules are “shadowed” by platinum or other heavy metals evaporated from a heated filament, macromolecules such as DNA or large proteins can be visualized with high contrast in the electron microscope (Figure 9–44A).

Negative staining is a similar approach that also allows fine detail to be seen in isolated molecules or macromolecular machines. In this technique, the molecules are supported on the thin film of carbon on a grid and mixed with a solution of a heavy-metal salt such as uranyl formate or acetate. After the sample has dried, a very thin film of metal salt covers the carbon film everywhere except where it has been excluded by the presence of an adsorbed macromolecule. Because the macromolecule allows electrons to pass through it much more readily than does the surrounding heavy-metal stain, a reverse or negative image of the molecule is created.

The specimen to be examined is usually either fixed, dried, and coated with a thin layer of heavy metal or alternatively rapidly frozen and then transferred to a cooled specimen stage for coating and direct examination in the microscope (Figure 9–45). The specimen is scanned with a very narrow beam of electrons. The quantity of electrons scattered or emitted as this primary beam bombards each successive point of the metallic surface is measured and builds up an image on a computer screen. Often an entire plant part or small animal can be put into the microscope with very little preparation (Figure 9–46).

Only surface features can be examined, however, and in most forms of SEM, the resolution attainable is not very high (about 10 nm). As a result, the technique is usually used to study whole cells and tissues rather than subcellular organelles (see Movie 21.3). However, very-high-resolution SEMs have been developed with a bright, coherent, field-emission gun as the electron source. As resolution in the SEM depends not on the wavelength of the electron beam but on the size of the electron spot that is scanned across the specimen, this type of SEM can produce images that rival the resolution possible with a negatively stained specimen in a TEM (Figure 9–47).

Because of the large depth of field of electron microscopes, all the parts of the three-dimensional specimen are in focus, and the resulting image is a projection (a superimposition of layers) of the structure along the viewing direction. The lost information in the third dimension can be recovered if we have views of the same specimen but from many different directions. The computational methods for this technique are widely used in medical CT scans. In a CT scan, the imaging equipment is moved around the patient to generate the different views. In electron microscope (EM) tomography, the specimen holder is tilted in the microscope, which achieves the same result. The specimen is usually tilted to a maximum of 60° in every direction, and in this way we can arrive at a three-dimensional reconstruction, in a chosen standard orientation, by combining different views of a single object.

As we saw earlier (p. 567), noise is important in light microscopy at low light levels, but it is a particularly severe problem for electron microscopy of unstained macromolecules. A protein molecule can tolerate a dose of only a few hundreds of electrons per square nanometer without damage, and this dose is orders of magnitude below what is needed to define an image at atomic resolution. The solution is to obtain images of many identical molecules—perhaps hundreds of thousands of images of individual particles—and combine them to produce an averaged image, revealing structural details that are hidden by the noise in the original images. This procedure is called single-particle reconstruction (Panel 9–1).

Cryo-electron microscopy (cryoEM) depends crucially on very rapidly freezing the aqueous specimen to form vitreous ice, which does not allow ice crystals to form and therefore does not damage the specimen. A very thin (about 100 nm) film of an aqueous suspension of purified macromolecular complex is prepared on a microscope grid and is then rapidly frozen by being plunged into a coolant. A special sample holder keeps this hydrated specimen at –160°C in the vacuum of the microscope, where it can be viewed directly without fixation, staining, or drying. Unlike negative staining, in which what we see is the envelope of stain exclusion around the particle, cryoEM produces an image from the macromolecular structure itself.

Electron microscopy, however, also has some very clear additional advantages over x-ray crystallography (discussed in Chapter 8) as a method for macromolecular structure determination. First, it does not require crystalline specimens. Second, it can deal with extremely large complexes—structures that may be too large or too variable to crystallize satisfactorily; for example, membrane proteins. Third, it allows the rapid analysis of different conformations of protein machines; for example, the different states of the F1 ATPase proton pump shown in Figure 14–31. Fourth, the glycosylation patterns and mobile loops on the surface of proteins, which are often impossible to see in x-ray structures, are more readily resolved in cryoEM structures. And fifth, only a minute amount of sample is required compared with that needed to make crystals.

We have seen that superresolution light microscopy can be used to very accurately locate specific molecules within a cell. A major disadvantage, however, of all fluorescence imaging techniques is that it is only the tagged molecules that are imaged—their cellular context remains invisible. When fluorescence imaging is combined, however, with looking at the same specimen in the electron microscope, this correlative light microscopy and electron microscopy technique, or CLEM, can allow specific target molecules to be examined in their full cellular context.

Imaging itself involves several trade-offs to be considered. An improvement in any one parameter—image contrast, resolution, signal-to-noise ratio, specimen damage by photons or electrons, the depth of specimen that can be imaged, or the speed of image recording—will inevitably require a sacrifice in one or more of the others, and understanding these trade-offs will help determine which approach is best for the cell biology problem being tackled.

Typically, we gain access to these proteins by obtaining large numbers of cells and physically breaking them open. Unicellular organisms, such as bacteria and yeast, are easy to produce in large amounts in the laboratory and are rich sources of the proteins involved in fundamental cell processes. But a deep understanding of human proteins in specific cell types requires human cells, or at the very least cells from a mammal. Specific animal tissues can be a useful solution, but these tend to be composed of a heterogeneous mixture of cell types. To obtain as much information as possible about specific cell types in a tissue, biologists have developed ways of dissociating cells from tissues and separating them according to type. These manipulations result in a relatively homogeneous population of cells that can then be analyzed—either directly or after their number has been greatly increased by allowing the cells to proliferate in culture.

Given appropriate surroundings, most animal cells can live, multiply, and even express differentiated properties in a culture dish. The cells can be watched continually under the microscope or analyzed biochemically, and the effects of adding or removing specific molecules, such as hormones or growth factors, can be systematically explored. Experiments performed on cultured cells are sometimes said to be carried out in vitro (literally, “in glass”) to contrast them with experiments using intact organisms, which are said to be carried out in vivo (literally, “in the living organism”). These terms can be confusing, however, because they are often used in a very different sense by biochemists. In the biochemistry lab, in vitro refers to reactions carried out in a test tube in the absence of living cells, whereas in vivo refers to any reaction taking place inside a living cell, even if that cell is growing in culture.

The tissue can then be teased apart into single cells by gentle agitation. Unlike yeast, most tissue cells are not adapted to living suspended in fluid and require a solid surface on which to grow and divide. For cell cultures, this support is usually provided by the surface of a plastic culture dish. Cells vary in their requirements, however, and many do not proliferate or differentiate unless the culture dish is coated with materials that cells adhere to, such as polylysine or extracellular matrix components. Cultures prepared directly from the tissues of an organism are called primary cultures. These can be made with or without an initial fractionation step to separate different cell types. In most cases, cells in primary cultures can be removed from the culture dish and recultured repeatedly in so-called secondary cultures; in this way, they can be repeatedly subcultured (passaged) for weeks or months.

Embryonic stem cells are an important cell type isolated from the early mammalian embryo. As described in Chapter 22, these cells are pluripotent; that is, they have the potential to differentiate into any cell type in the body. When cultured in the presence of the appropriate extracellular signaling factors and nutrients, stem cells can be directed to differentiate into a wide range of specific cell types. Under some conditions, it is even possible to stimulate these cells to assemble into three-dimensional multicellular structures that are miniature versions of certain organs, such as the gut. These organoids provide a powerful tool for the analysis of tissue function (see Chapter 22). Cell culture is not limited to animal cells. When a piece of plant tissue is cultured in a sterile medium containing nutrients and appropriate growth regulators, many of the cells are stimulated to proliferate indefinitely in a disorganized manner, producing a mass of relatively undifferentiated cells called a callus.

The cell cultures obtained by disrupting tissues tend to suffer from a problem—eventually the cells die. Most vertebrate cells stop dividing after a finite number of cell divisions in culture, a process called replicative cell senescence (discussed in Chapter 17). Normal human fibroblasts, for example, typically divide only 25–40 times in culture before they stop. In these cells, the limited proliferation capacity reflects a progressive shortening and uncapping of the cell’s telomeres, the repetitive DNA sequences and associated proteins that cap the ends of each chromosome (discussed in Chapter 5). Human somatic cells in the body have turned off production of the enzyme, called telomerase, that normally maintains the telomeres, which is why their telomeres shorten with each cell division. Human fibroblasts can often be coaxed to proliferate indefinitely by providing them with the gene that encodes the catalytic subunit of telomerase; in this case, they can be propagated as an immortalized cell line.

Some human cells, however, cannot be immortalized by this trick. Although their telomeres remain long, they still stop dividing after a limited number of divisions because culture conditions cause excessive stimulation of cell proliferation, which activates a poorly understood protective mechanism that stops cell division—a process sometimes called culture shock. To immortalize these cells, one has to do more than introduce telomerase.

Unlike human cells, most rodent cells do not turn off production of telomerase, and therefore their telomeres do not shorten with each cell division. Therefore, if culture shock can be avoided, some rodent cell types will divide indefinitely in culture. In addition, rodent cells often undergo spontaneous genetic changes in culture that inactivate their protective mechanisms, thereby producing immortalized cell lines. Cell lines can often be most easily generated from cancer cells, but these cultures—referred to as transformed cell lines—differ from those prepared from normal cells in several ways. Transformed cell lines often grow without attaching to a surface, for example, and they can proliferate to a much higher density in a culture dish. Similar properties can be induced experimentally in normal cells by transforming them with a tumor-inducing virus or chemical. The resulting transformed cell lines can usually cause tumors if injected into a susceptible animal.

Transformed and nontransformed cell lines are extremely useful in cell research as sources of very large numbers of cells of a uniform type, especially because they can be stored in liquid nitrogen at –196°C for an indefinite period and retain their viability when thawed. It is important to keep in mind, however, that cell lines nearly always differ in important ways from their normal progenitors in the tissues from which they were derived. Some widely used cell lines are listed in Table 8–1. Different lines have different advantages; for example, the PtK epithelial cell lines derived from the rat kangaroo remain flat during mitosis (unlike many other cell types), allowing the mitotic apparatus to be readily observed in action.

Antibodies are often produced by inoculating animals with the protein of interest and subsequently isolating the antibodies specific to that protein from the serum of the animal. However, only limited quantities of antibodies can be obtained from a single inoculated animal, and the polyclonal antibodies produced will be a heterogeneous mixture of antibodies that recognize a variety of different antigenic sites on the protein. Moreover, antibodies specific for the antigen will constitute only a fraction of the antibodies found in the serum. An alternative technology, which allows the production of an unlimited quantity of identical antibodies and greatly increases the specificity and convenience of antibody-based methods, is the production of monoclonal antibodies by hybridoma cell lines.

B lymphocytes normally have a limited life span in culture, but individual antibody-producing B lymphocytes from an immunized mouse, when fused with cells derived from a transformed B lymphocyte cell line, can give rise to hybrids that have both the ability to make a particular antibody and the ability to multiply indefinitely in culture (Figure 8–3). These hybridomas are propagated as individual clones, each of which provides a permanent and stable source of a single type of monoclonal antibody. Each type of monoclonal antibody recognizes a single type of antigenic site; for example, a particular cluster of five or six amino acid side chains on the surface of a protein. Their uniform specificity makes monoclonal antibodies much more useful than conventional antisera for many purposes.

An important advantage of the hybridoma technique is that monoclonal antibodies can be made against molecules that constitute only a minor component of a complex mixture. In an ordinary antiserum made against such a mixture, the proportion of antibody molecules that recognize the minor component would be too small to be useful. But if the B lymphocytes that produce the various components of this antiserum are made into hybridomas, it becomes possible to screen individual hybridoma clones from the large mixture to select one that produces the desired type of monoclonal antibody and to propagate the selected hybridoma indefinitely so as to produce that antibody in unlimited quantities. In principle, therefore, a monoclonal antibody can be made against any protein in a biological sample. Once an antibody has been made, it can be used to localize the protein in cells and tissues, to follow its movement, and to purify the protein to study its structure and function.

Monoclonal antibodies are not just useful research tools but are valuable as treatments for a number of human diseases. Certain cancers, for example, can be treated by intravenous infusion of monoclonal antibodies that bind and inhibit signaling receptors on the cancer cell surface, thereby reducing proliferation of the tumor cells. In other cases, monoclonal antibodies that bind specific cell-surface immune regulators can promote immunological attack of cancer cells.

The challenge of isolating a single type of protein from the thousands of other proteins in a cell is formidable but must be overcome to study protein function in vitro. As we shall see later in this chapter, recombinant DNA technology can enormously simplify this task by engineering cells to produce large quantities of a given protein, thereby making its purification much easier. Whether the source of the protein is an engineered cell or a natural tissue, a purification procedure usually starts with subcellular fractionation to reduce the complexity of the material and is then followed by purification steps of increasing specificity.

To purify a protein, it must first be extracted from inside the cell. Cells can be broken up in various ways: they can be subjected to osmotic shock or ultrasonic vibration, forced through a small orifice, or ground up in a blender. These procedures break many of the membranes of the cell (including the plasma membrane and endoplasmic reticulum) into fragments that immediately reseal to form small closed vesicles. If carefully carried out, however, the disruption procedures leave organelles such as nuclei, mitochondria, the Golgi apparatus, lysosomes, and peroxisomes largely intact.

The different components of the homogenate must then be separated. Such cell fractionations became possible only after the commercial development in the early 1940s of an instrument known as the preparative ultracentrifuge, which rotates extracts of broken cells at high speeds (Figure 8–4). This treatment separates cell components by size and density: in general, the largest objects experience the largest centrifugal force and move the most rapidly. At relatively low speed, large components such as nuclei sediment to form a pellet at the bottom of the centrifuge tube; at slightly higher speed, a pellet of mitochondria is deposited; and at even higher speeds and with longer periods of centrifugation, first vesicles and then ribosomes can be collected (Figure 8–5). All of these fractions are impure, but many of the contaminants can be removed by resuspending the pellet and repeating the centrifugation procedure several times.

When centrifuged, the various components in the mixture move as a series of distinct bands through the solution, each at a different rate, in a process called velocity sedimentation (Figure 8-6A). For the procedure to work effectively, the bands must be protected from convective mixing, which would normally occur whenever a denser solution (for example, one containing organelles) finds itself on top of a lighter one (the salt solution). This is achieved by augmenting the solution in the tube with a shallow gradient of sucrose prepared by a special mixing device. The resulting density gradient—with the dense end at the bottom of the tube—keeps each region of the solution denser than any solution above it, and it thereby prevents convective mixing from distorting the separation. When sedimented through sucrose gradients, different cell components separate into distinct bands that can be collected individually. The relative rate at which each component sediments depends primarily on its size and shape—normally being described in terms of its sedimentation coefficient, or S value.

Present-day ultracentrifuges rotate at speeds of up to 80,000 rpm and produce forces as high as 500,000 times gravity. These enormous forces drive even small macromolecules, such as tRNA molecules and simple enzymes, to sediment at an appreciable rate and allow them to be separated from one another by size. The ultracentrifuge is also used to separate cell components on the basis of their buoyant density, independently of their size and shape. In this case, the sample is sedimented through a steep density gradient that contains a very high concentration of sucrose or cesium chloride. Each cell component begins to move down the gradient as in Figure 8–6A, but it eventually reaches a position where the density of the solution is equal to its own density.

A series of distinct bands is thereby produced in the centrifuge tube, with the bands closest to the bottom of the tube containing the components of highest buoyant density (Figure 8–6B). This method, called equilibrium sedimentation, is so sensitive that it can separate macromolecules that have incorporated heavy isotopes, such as 13C or 15N, from the same macromolecules that contain the lighter, common isotopes (12C or 14N). In fact, the cesium-chloride method was developed in 1957 to separate the labeled from the unlabeled DNA produced after exposure of a growing population of bacteria to nucleotide precursors containing 15N; this classic experiment provided direct evidence for the semiconservative replication of DNA

Proteins are most often fractionated by column chromatography, in which a mixture of proteins in solution is passed through a column containing a porous gel matrix. Different proteins are retarded to different extents by their interaction with the matrix, and they can be collected separately as they flow out of the bottom of the column

Many types of matrices are available. Ion-exchange columns (Figure 8–9A) are packed with small beads that carry either a positive or a negative charge, so that proteins are fractionated according to the arrangement of charges on their surface. Hydrophobic columns are packed with beads from which hydrophobic side chains protrude, selectively retarding proteins with exposed hydrophobic regions. Gel-filtration columns (Figure 8–9B), which separate proteins according to their size, are packed with tiny porous beads: molecules that are small enough to enter the pores linger inside successive beads as they pass down the column, while larger molecules remain in the solution flowing between the beads and therefore move more rapidly, emerging from the column first. Besides providing a means of separating molecules, gel-filtration chromatography is a convenient way to estimate their size.

Affinity chromatography (Figure 8–9C) takes advantage of the biologically important binding interactions that occur on protein surfaces. If a substrate molecule is covalently coupled to an inert matrix such as a polysaccharide bead, the enzyme that operates on that substrate will often be specifically retained by the matrix and can then be eluted (washed out) in nearly pure form. Likewise, short DNA oligonucleotides of a specifically designed sequence can be immobilized in this way and used to purify DNA-binding proteins that normally recognize this sequence of nucleotides in chromosomes. Alternatively, specific antibodies can be coupled to a matrix to purify protein molecules recognized by the antibodies (called immunoaffinity chromatography). Because of the great specificity of all such affinity columns, 1000- to 10,000-fold purifications can sometimes be achieved in a single pass.

Because most individual proteins represent less than 1/1000 of the total cell protein, it is usually necessary to use several different types of columns in succession to attain sufficient purity, with affinity chromatography being the most efficient (Figure 8–10). Imperfections in the matrices (such as cellulose), which cause an uneven flow of solvent through the column, limit the resolution of conventional column chromatography. Special chromatography resins (usually silica-based) composed of tiny spheres (3–10 μm in diameter) can be packed with a special apparatus to form a uniform column bed. Such high-performance liquid chromatography (HPLC) columns attain a high degree of resolution. In HPLC, the solutes are passed through the column at high pressure and equilibrate very rapidly with the interior of the tiny spheres, and so solutes with different affinities for the matrix are efficiently separated from one another even at very fast flow rates. HPLC is therefore the method of choice for separating many proteins and small molecules.

Studies of organelles and other large subcellular components isolated in the ultracentrifuge have contributed enormously to our understanding of the functions of different cell components. Experiments on mitochondria and chloroplasts purified by centrifugation, for example, demonstrated the central function of these organelles in converting energy into forms that the cell can use. Similarly, resealed vesicles formed from fragments of rough and smooth endoplasmic reticulum (microsomes) have been separated from each other and analyzed as functional models of these compartments of the intact cell.

Immunoprecipitation is a useful variation on the theme of affinity chromatography. Specific antibodies that recognize the protein to be purified are attached to small agarose beads. Rather than being packed into a column, as in affinity chromatography, a small quantity of the antibody-coated beads is simply added to a protein extract in a test tube and mixed in suspension for a short period of time—thereby allowing the antibodies to bind the desired protein. The beads are then collected by low-speed centrifugation, and the unbound proteins in the supernatant are discarded. This method is commonly used to purify small amounts of enzymes from cell extracts for analysis of enzymatic activity or for identification of associated proteins. As we describe later in this chapter, immunoprecipitation also provides a method to identify DNA or RNA sequences recognized by specific proteins.

Using the recombinant DNA methods discussed later in this chapter, any gene can be modified to produce its protein with an extra amino acid sequence that provides a specific recognition tag, so as to make subsequent purification of the protein simple and rapid. Often the recognition tag is an antigenic determinant, or epitope, which can be recognized by a highly specific monoclonal antibody. The antibody can then be used to purify the protein by affinity chromatography or immunoprecipitation (Figure 8–11). Other types of tags are specifically designed for protein purification. For example, a repeated sequence of the amino acid histidine binds to certain metal ions, including nickel and copper. If genetic-engineering techniques are used to attach a short string of histidines to one end of a protein, the slightly modified protein can be retained selectively on an affinity column containing immobilized nickel ions. Metal affinity chromatography can thereby be used to purify the modified protein from a complex molecular mixture.

For example, the experiments to decipher the mechanisms of protein synthesis began with a cell homogenate that could translate RNA molecules to produce proteins. Fractionation of this homogenate, step by step, produced in turn the ribosomes, tRNAs, and various enzymes that together constitute the protein-synthetic machinery. Once individual pure components were available, each could be added or withheld separately to define its exact role in the overall process. A major goal for cell biologists is the reconstitution of every biological process in a purified cell-free system. Only in this way can we define all of the components needed for the process and control their concentrations, which is required to work out their precise mechanism of action.

Proteins usually possess a net positive or negative charge, depending on the mixture of charged amino acids they contain. An electric field applied to a solution containing a protein molecule causes the protein to migrate at a rate that depends on its net charge and on its size and shape. The most useful application of this property is sodium dodecyl sulfate polyacrylamide-gel electrophoresis (SDS-PAGE). It uses a highly cross-linked gel of polyacrylamide as the inert matrix through which the proteins migrate. The gel is prepared by polymerization of monomers; the pore size of the gel can be adjusted so that it is small enough to retard the migration of the protein molecules of interest. The proteins are dissolved in a solution that includes a powerful negatively charged detergent, sodium dodecyl sulfate, or SDS (Figure 8–12).

(see Figure 8–12) is usually added to break any disulfide linkages in the proteins, so that all of the constituent polypeptides in multisubunit proteins can be analyzed separately. What happens when a mixture of SDS-solubilized proteins is run through a slab of polyacrylamide gel? Each protein molecule binds large numbers of the negatively charged detergent molecules, which mask the protein’s intrinsic charge and cause it to migrate toward the positive electrode when a voltage is applied. Proteins of the same size tend to move through the gel with similar speeds because (1) their native structure is completely unfolded by the SDS, so that their shapes are the same, and (2) they bind the same amount of SDS and therefore have the same amount of negative charge.

Larger proteins, with more charge, are subjected to larger electrical forces but also to a larger drag. In free solution, the two effects would cancel out, but, in the mesh of the polyacrylamide gel, which acts as a molecular sieve, large proteins are retarded much more than small ones. As a result, a complex mixture of proteins is fractionated into a series of discrete protein bands mostly according to their mass (Figure 8–13). The major proteins are readily detected by staining the proteins in the gel with a dye such as Coomassie blue. When small amounts of protein are present and more sensitive methods are required, gels can be treated with a silver stain, which will detect as little as 10 ng of protein in a band.

Because different proteins can have similar sizes, shapes, masses, and overall charges, most separation techniques such as SDS polyacrylamide-gel electrophoresis or ion-exchange chromatography cannot typically separate all the proteins in a cell or even in an organelle. In contrast, two-dimensional gel electrophoresis, which combines two different separation procedures, can resolve up to 2000 proteins in the form of a two-dimensional protein map. In the first step, the proteins are separated by their intrinsic charges. The sample is dissolved in a small volume of a solution containing a nonionic (uncharged) detergent, together with β-mercaptoethanol and the denaturing reagent urea. This solution solubilizes, denatures, and dissociates all the polypeptide chains but leaves their intrinsic charge unchanged.

The polypeptide chains are then separated in a pH gradient by a procedure called isoelectric focusing, which takes advantage of the variation in the net charge on a protein molecule with the pH of its surrounding solution. Every protein has a characteristic isoelectric point, the pH at which the protein has no net charge and therefore does not migrate in an electric field. In isoelectric focusing, proteins are separated electrophoretically in a narrow tube of polyacrylamide gel in which a gradient of pH is established by a mixture of special buffers. Each protein moves to a position in the gradient that corresponds to its isoelectric point and remains there (Figure 8–15). This is the first dimension of two-dimensional polyacrylamide-gel electrophoresis.

A specific protein can be identified after its fractionation on a polyacrylamide gel by exposing all the proteins on the gel to a specific antibody that has been labeled with a radioactive isotope or a fluorescent dye. This procedure is normally carried out after transferring all of the separated proteins in the gel onto a sheet of nitrocellulose paper or nylon membrane. Placing the membrane against the gel and driving the proteins out of the gel with a strong electric current transfers the protein onto the membrane. The membrane is then soaked in a solution of labeled antibody to reveal the protein of interest. This method of detecting proteins is called Western blotting, or immunoblotting (Figure 8–17).

Most proteins in a cell are subunits of larger complexes, and knowledge of the size and shape of these complexes often leads to insights regarding their function. This information can be obtained in several ways. Sometimes, a complex can be directly visualized using electron microscopy, as described in Chapter 9. A complementary approach relies on the hydrodynamic properties of a complex; that is, its behavior as it moves through a liquid medium. Usually, two separate measurements are made. One measure is the velocity of a complex as it moves under the influence of a centrifugal field produced by an ultracentrifuge (see Figure 8–6A). The sedimentation coefficient (or S value) obtained depends on both the size and the shape of the complex and does not, by itself, convey especially useful information.

Mass spectrometry exploits this principle to separate ions according to their mass-to-charge (m/z) ratio. It is an enormously sensitive technique. It requires very little material and is capable of determining the precise mass of intact proteins and of peptides derived from them by enzymatic or chemical cleavage. Masses can be obtained with great accuracy, often with an error of less than one part in a million. Mass spectrometry is performed using complex instruments with three major components (Figure 8–18A). The first is the ion source, which transforms tiny amounts of a peptide sample into a gas containing individual charged peptide molecules. These ions are accelerated by an electric field into the second component, the mass analyzer, where electric or magnetic fields are used to separate the ions on the basis of their mass-to-charge ratios. Finally, the separated ions collide with a detector, which generates a mass spectrum containing a series of peaks representing the masses of the molecules in the sample.

One of the most common ion sources depends on a technique called matrix-assisted laser desorption ionization (MALDI). In this approach, the proteins in the sample are first cleaved into short peptides by a protease such as trypsin. These peptides are mixed with an organic acid and then dried onto a metal or ceramic slide. A brief laser burst is directed toward the sample, producing a gaseous puff of ionized peptides, each carrying one or more positive charges. In many cases, the MALDI ion source is coupled to a mass analyzer called a time-of-flight (TOF) analyzer, which is a long chamber through which the ionized peptides are accelerated by an electric field toward a detector. Their mass and charge determine the time it takes them to reach the detector: large peptides move more slowly, and more highly charged molecules move more quickly.

Instead, MS/MS typically involves an electrospray ion source, which produces a continuous thin stream of peptides that are ionized and accelerated into the first mass analyzer. The mass analyzer is typically either a quadrupole or ion trap, which employs large electrodes to produce oscillating electric fields inside the chamber containing the ions. These instruments act as mass filters: the electric field is adjusted over a broad range to select a single peptide ion and discard all the others in the peptide mixture. In tandem mass spectrometry, this single ion is then exposed to an inert, high-energy gas, which collides with the peptide, resulting in fragmentation, primarily at peptide bonds. The second mass analyzer then determines the masses of the peptide fragments, which can be used by computational methods to determine the amino acid sequence of the original peptide and thereby identify the protein from which it came.

Because most proteins in the cell function as part of complexes with other proteins, an important way to begin to characterize the biological role of an unknown protein is to identify all of the other proteins to which it specifically binds. A key method for identifying proteins that bind to one another tightly is co-immunoprecipitation. A target protein is immunoprecipitated from a cell lysate using specific antibodies coupled to beads, as described earlier. If the target protein is associated tightly enough with another protein when it is captured by the antibody, the partner precipitates as well and can be identified by mass spectrometry. This method is useful for identifying proteins that are part of a complex inside cells, including those that interact only transiently; for example, when extracellular signal molecules stimulate cells (discussed in Chapter 15).

For many decades, the primary technique for protein structural analysis has been x-ray crystallography. X-rays, like light, are a form of electromagnetic radiation, but they have a much shorter wavelength, typically around 0.1 nm (the diameter of a hydrogen atom). If a narrow beam of parallel x-rays is directed at a sample of a pure protein, most of the x-rays pass straight through it. A small fraction, however, is scattered by the atoms in the sample.

The position and intensity of each spot in the x-ray diffraction pattern contain information about the locations of the atoms in the crystal that gave rise to it. Computer-assisted computational methods process the diffraction pattern to generate a three-dimensional electron-density map. Interpreting this map—translating its contours into a three-dimensional structure—can be a laborious procedure. Largely by trial and error, the sequence and the electron-density map are correlated by computer to give the best possible fit. The reliability of the final atomic model depends on the resolution of the original crystallographic data: 0.5-nm resolution might produce a low-resolution map of the polypeptide backbone, whereas a resolution of 0.15 nm allows all of the non-hydrogen atoms in the molecule to be reliably positioned.

Nuclear magnetic resonance (NMR) spectroscopy has been widely used for many years to analyze the structure of small molecules, small proteins, or protein domains. Unlike x-ray crystallography, NMR does not depend on having a crystalline sample. It simply requires a small volume of concentrated protein solution that is placed in a strong magnetic field; indeed, it is the main technique that yields detailed evidence about the three-dimensional structure of molecules in solution. Certain atomic nuclei, particularly hydrogen nuclei, have a magnetic moment or spin; that is, they have an intrinsic magnetization, like a bar magnet. When exposed to a strong magnetic field in an NMR experiment, the spin of these nuclei aligns with the magnetic field, but it can be forced into a misaligned, excited state by radiofrequency (RF) pulses of electromagnetic radiation. As the excited hydrogen nuclei return to their aligned state, they emit RF radiation, which can be measured and displayed as a spectrum.

Over time, different groups of biologists have focused on studying a few chosen species, which allows their knowledge and research tools to be pooled to gain a deeper understanding than could be achieved if their efforts were spread over many different organisms. Although the list of these representative, model organisms is continually expanding, a few stand out in terms of the breadth and depth of information that has been accumulated about them over the years—knowledge that has been essential for our understanding of how all cells work. In this section, we examine some of these organisms and review the benefits that each offers to the study of cell biology and, in many cases, to the promotion of human health. We begin with a discussion of some especially powerful strategies that scientists have developed to understand the cell, and we shall see how these approaches dictated the choice of model organisms.

The standard laboratory strain E. coli K-12 has a genome of approximately 4.6 million nucleotide pairs contained in a single circular molecule of DNA that codes for about 4300 different kinds of proteins (Figure 1–38). In molecular terms, we probably have a more complete understanding of E. coli than of any other living organism. Most of our understanding of the fundamental mechanisms of life—for example, how cells replicate their DNA or how they decode the instructions represented in the DNA to direct the synthesis of specific RNAs and proteins—initially came from studies of E. coli and its viruses. This is because the basic genetic mechanisms have turned out to be highly conserved throughout evolution and are essentially the same in our own cells as in E. coli.

The molecular and genetic complexity of eukaryotes is daunting, and biologists need to concentrate their limited resources on a small number of selected model organisms to unravel this complexity. To analyze the internal workings of the eukaryotic cell without the additional problems of multicellular development, it makes sense to use a single-cell species that is as simple as possible. The popular choice for this role of minimal model eukaryote has been the yeast Saccharomyces cerevisiae (Figure 1–39)—the same species that is used by brewers of beer and bakers of bread. S. cerevisiae is a small, single-cell member of the kingdom of fungi and, in terms of its genome sequence, much more closely related to animals than to plants (see Figure 1–35). It is robust and easy to grow in a simple nutrient medium. Like other fungi, it has a tough cell wall, is relatively immobile, and possesses mitochondria but not chloroplasts. When nutrients are plentiful, it grows and divides about every hundred minutes.

In addition to these features, the yeast has a further property that makes it a convenient organism for genetic studies: its genome, by eukaryotic standards, is exceptionally small (see Table 1–2), yet it suffices for all the basic tasks that every eukaryotic cell must perform. Mutants are available for every gene, and thus the consequence of missing each gene—one by one—can be observed under any environmental condition using the high-throughput procedures described in Chapter 8. Over the past 50 years, extensive studies of yeast cells carried out by many laboratories have provided keys to crucial “eukaryotic-only” processes. These include the cell-division cycle (the critical chain of events by which the nucleus and all the other components of a cell are duplicated and parceled out to create two daughter cells from one) and meiosis (the process through which an organism’s reproductive cells are formed).

Out of the nearly 400,000 known species of flowering plants, molecular biologists have chosen to concentrate their efforts on a small weed in the cabbage family, the common wall cress Arabidopsis thaliana (Figure 1–41), which can be grown indoors in large numbers and produces thousands of offspring per plant after 8–10 weeks. Arabidopsis has a total genome size of approximately 135 million nucleotide pairs, about 10 times the size of the yeast genome (see Table 1–2). Work on Arabidopsis has provided a deep understanding of numerous key features of plants, including the mechanisms that cause flower development and its coordination with the seasons, the ability to grow toward sunlight, cell-to-cell signaling by hormones, and the special type of innate immune system that plants use to ward off pathogens.

Five species have emerged as the foremost model organisms for molecular, cell, and developmental biological studies. In order of increasing body size, they are the nematode worm Caenorhabditis elegans, the fly Drosophila melanogaster, the zebrafish Danio rerio, the mouse Mus musculus, and the human, Homo sapiens. Genome sequences from many different individuals within each species have been determined. Caenorhabditis elegans (Figure 1–42) is a small, harmless relative of the eelworm that attacks crops.

C. elegans develops with clockwork precision from a fertilized egg cell into an adult worm with exactly 959 body cells (plus a variable number of egg and sperm cells)—an unusual degree of regularity for animal development. We now have a minutely detailed description of the sequence of events by which this development occurs, as the cells divide, move, and change their character according to strict and predictable rules (see Figure 21–42). The genome of about 100 million nucleotide pairs codes for about 20,000 proteins, and many mutants and other tools are available for testing gene functions. Although the worm has a body plan very different from our own, the conservation of biological mechanisms has been sufficient for the worm to be a model for many of the developmental and cell-biological processes that occur in the human body.

The fruit fly Drosophila melanogaster (Figure 1–43) has been used as a model for animal genetic studies for longer than any other organism; in fact, the foundations of classical genetics were built to a large extent on studies of this insect. Nearly 100 years ago, for example, the fly provided definitive proof that genes—the abstract units of hereditary information at the time—are carried on chromosomes, whose behavior had been closely followed with the light microscope during eukaryotic cell division but whose function was at first unknown.

Frogs have long been used to study the early steps of embryonic development in vertebrates. Because their eggs are big, easy to manipulate, and fertilized outside of the animal, the subsequent development of the early embryo can be easily followed (Figure 1–45). Xenopus laevis, the African clawed frog, continues to be an important model organism (Movie 1.6 and see Movie 21.1). Although the species is poorly suited for genetic analysis, cytoplasm isolated from unfertilized Xenopus eggs has the remarkable ability to recapitulate the formation of cellular structures and organelles in a test tube. These egg extracts allow powerful biochemical approaches to study such fundamental processes as the cell division cycle, described in Chapter 17.

The mouse, being small, hardy, and a rapid breeder, has become the foremost model organism for experimental studies of mammalian molecular cell biology. Many naturally occurring mutations are known, often mimicking the effects of corresponding mutations in humans to a remarkable extent (Figure 1–48). Moreover, methods have been developed to test the function of any chosen mouse gene or of any noncoding portion of the mouse genome by artificially creating mutations in the relevant part of the gene or genome, as we explain in Chapter 8. Just one made-to-order mutant mouse can provide a wealth of information for the cell biologist. It reveals the effects of the chosen mutation in various contexts, simultaneously testing the action of the gene in the many different types of cells in the body that could in principle be affected.