Search

It is well known that the resolution of a microscope depends critically on the aperture of its objective lens. Achieving wide apertures requires the lens to be extremely close to the sample being imaged, which is often inconvenient. We describe an optical microscope, currently in breadboard form, that achieves a large aperture, and thus high resolution, with a large (theoretically unlimited) working distance, based on the principles of synthetic aperture radar (SAR).

Centractin (Arp 1) is an actin-related protein that shares 53% sequence identity with conventional actin and is predicted to have a similar core structure for nucleotide binding. Biochemical studies of centractin, named for its apparent localization to the centrosome, reveal that it is a member of a stable macromolecular complex, dynactin. The dynactin complex consists of at least 9 polypeptides and has been implicated in activating dynein mediated vesicle transport along microtubules. Ultra-structural analysis, by rotary shadowing electron microscopy with antibody decoration, reveals that within the context of the dynactin complex, centractin forms a 37 nm filament resembling the structure of an actin filament with capping protein and p62 localizing to opposite ends.We have used transient transfection assays to investigate the effects of overexpression of centractin in cells. Mammalian PtK2 cells were transiently transfected with centractin cloned in the pcDNA3 vector under the control of a CMV promoter. The cells were transfected for 24 hours using Ca2+/DNA co-precipitates. The transfection efficiency was as high as 41 % as determined by direct counting of immunostained cells. Cells were fixed at time points ranging from 18 to 24 hours post washing using 1% formaldehyde plus 1% Triton X-100 (5 min), 0.05% glutaraldehyde (5 min), 1 % formaldehyde (5 min), and 0.05% sodium borohydride to block free aldehyde groups (3×5 min). Immunofluorescence employed an affinity-purified, rabbit polyclonal antibody to centractin followed by a Texas Red-labeled secondary antibody. Epifluorescence microscopy revealed that all of the cells overexpressing centractin contained centractin structures with novel morphology. Two distinct patterns were observed consistently: filamentous centractin localized throughout the cytoplasm (Figure la) and perinuclear centractin accumulations (Figure lb). The perinuclear accumulations appear to be highly three dimensional and curved in structure (confirmed by confocal microscopy), and often have associated filamentous structures (Figure lb). The addition of 33 micromolar nocodozole (30 min. at 4°C) to the transfected cells results in thinner centractin filaments (Figure lc).

Stereo pairs of electron microscope images provide valuable visual impressions of the three-dimensional nature of specimens, including biological objects. Beyond this one seeks quantitatively accurate models and measurements of the three dimensional positions and sizes of structures in the specimen. In our laboratory, we have sought to combine high resolution video cameras with high performance computer graphics systems to improve both the ease of building 3D reconstructions and the accuracy of 3D measurements, by using multiple tilt images of the same specimen tilted over a wider range of angles than can be viewed stereoscopically. Ultimately we also wish to automate the reconstruction and measurement process, and have initiated work in that direction.Figure 1 is a stereo pair of 400 kV images from a 1 micrometer thick transverse section of frog skeletal muscle stained with the Golgi stain. This stain selectively increases the density of the transverse tubular network in these muscle cells, and it is this network that we reconstruct in this example.

Stereoscopic pairs of electron microscope images are used for quantitative 3D information. A prerequisite for the measurement is to position the two images correctly relative to each other and with the stereo rotation axis of each image aligned vertically for viewing. Although this alignment procedure is relatively straightforward to perform using prints of the images, it is not straightforwardto do once the images have been digitized directly from the microscope. We have developed an algorithm for determining the parameters needed for orienting digitized images. Four parameters are needed: The displacements Δx and Δy by which one of the images must be translated to be superimposed on the other, and the angles Θ1 and Θr by which the left and right images must be rotated to bring the stereo-rotation axis vertical.The microscopist first uses an interactive routine to identify the coordinates xl(i),yl(i) and xr(i),yr(i) of a number (N) of fiducial points which can easily be recognized on each image.

The Golgi stain is very useful for selective staining of membrane systems in muscle cells, especially for HVEM and IVEM where thick sections can be examined and three-dimensional information obtained. However, the staining intensity and pattern are highly variable from preparation to preparation, among fibers in the same preparation, and within individual fibers. Staining can be absent altogether, or can involve the T-system (T) or the sarcoplasmic reticulum (SR) alone, or both T and SR. Considerable time can be spent sectioning blocks and scanning the sections in the electron microscope, searching for the desired staining pattern, sometimes without success. We have found that examination of the trimmed face of an embedded block under a dissecting microscope with oblique reflected light reveals a pattern of colors in muscle fibers and parts of muscle fibers. The origin of the colors and the reason that the different staining patterns show different colors are not clear. Nevertheless, the colors can be related to the staining pattern that will be seen in the electron microscope in sections cut from the same blocks, thus facilitating trimming of the block to areas of interest, both saving time and reducing the likelihood of missing rare staining patterns or paticularly interesting parts of the embedding.

High and intermediate voltage electron microscopes can be used to get good images of many types of thick biological specimens, including sections of embedded material up to several micrometers thick, whole cells, or isolated cellular components mounted on support films, stereoscopic pairs of images can be obtained by tilting the specimen between exposures, and this requires only relatively small tilt angles, often less than 15°. With a high tilt specimen holder, images covering a wider range of tilt angles can be obtained, increasing the amount of 3D information. Computer graphic methods for extracting quantitative 3D information from such images have been under development and use in this laboratory for several years.Many cellular structures take the form of networks. Examples that we study are transverse tubular networks in muscle, fibrin clots, and the microtrabecular lattice of the cytoplasm. We perform quantitative, 3D reconstruction and analysis of such networks from multiple stereoscopic image sets obtained using a JEOL 4000-EX TEM.

We have designed and constructed a camera system for acquisition of digital images directly from a JEOL 4000-EX 400 kV transmission electron microscope. Our major goal was the ablity to sample and test images for suitability for various forms of image processing and analysis during a session on the microscope, rather than after processing and digitizing films, so that adjustments could be made while the specimens were still in place. Our design parameters were 1) good resolution, in terms of pixels in the image, 2) wide field, so low magnification images could be obtained, 3) no interference with normal operation of the microscope, including the film camera, and 4) minimal modification of the microscope.Cameras positioned below the film camera have narrow fields of view because of increase in image size below the film camera and the small size of CCD targets relative to film. We chose a position above the microscope's fluorescent viewing screen, where there is a convenient port on the 4000-EX, and achieved a field width approximately equal to that on film.

Immunofluorescence studies have helped to elucidate the process of molecular assembly into cross-striated myofibrils in cultured cardiac and skeletal muscle cells by localizing accurately identified proteins. However, our understanding of how such immunofluorescence images are related to fine structural organization is limited. High and intermediate voltage electron microscopes can image relatively thick specimens, including whole cells, and confocal scanning laser fluorescence microscopy provides thin optical section images of similarly thick specimens. It seemed natural for us to combine these two approaches in a correlated electron and light microscopic analysis of myofibrillogenesis in cultured cardiac muscle cells.Cardiac myocytes from 7-9 day chick embryos, cultured for 3-5 days on glass coverslips carrying gold EM grids covered with formvar/carbon films, were fixed with 2% formaldehyde for 3 min., permeabilized with 5% Triton for 30 min., and stained with two antibodies conjugated with FITC and rhodamine. These specimens were first observed by epi-fluorescence and then by confocal laser scanning microscopy (Sarastro, Stockholm, SWEDEN).

We have used 400 kilovolt intermediate voltage electron microscopy (IVEM) to examine thick sections of fibroblasts cultured in collagen gels. In these 3D collagen lattices, the long, narrow pseudopodial extensions that extend out and make contact with the collagen matrix exhibit a complex topography not seen in the processes put out by cells moving on planar substrata. For this reason, sections 1 to 2 microns thick that enclose a whole cell process are more informative of the overall morphology of the interaction between cells and the collagen than are thin sections. To aid the discrimination of topography of cell processes in stereo views of micrographs, some cells were labeled with antibodies and protein A-colloidal gold conjugates. The gold particles provided clear 3D reference points for computer-aided reconstructions of membrane topography from tilt series of IVEM images. Our results confirm that cells that move through collagen lattices lack the well-spread morphology of their counterparts moving on glass. They are generally rather spindly with several long branching anterior pseudopodia. We found that the cell bodies and major pseudopodial processes were cylindrical, as one might expect of cells in a 3D environment, but at the leading edge of advancing pseudopodia there are small flat extensions similar to those seen in cells on glass. This similarity suggests that the lamellipodium is a basic type of protrusive structure used by fibroblasts during locomotion on all types of substratum. The flattened shape of lamellipodia may be part of the mechanism by which cells sense the orientation of fibrillar extracellular matrices during embryonic morphogenesis.

The sections in this article are: 1 Structural Components of Skeletal Muscle Fibers 1.1 Sarcomeres, Striations, and Fibrils 1.2 Membranes 2 Physiological Correlates 2.1 Local Activation Experiments 2.2 Comparison of Slow-Acting and Fast-Acting Muscle Fibers 2.3 Relation of Total Surface Area to Fiber Capacitance 2.4 Glycerol-Shock Experiments 3 Microscopic Methods in Study of Cellular Membrane Structure 3.1 Scanning Electron Microscopy 3.2 High-Voltage Electron Microscopy of Thick Sections 3.3 Freeze-Fracture Electron Microscopy 4 T System 4.1 Definition, Development, and Function 4.2 T-Tubule Networks 4.3 T-Tubule Shapes 5 Sarcoplasmic Reticulum 5.1 Definition, Development, and Function 5.2 Structural Relationship to Myofibrils and Striations 5.3 Form of Sarcoplasmic Reticulum 5.4 Content of Sarcoplasmic Reticulum 5.5 Calcium Movements 6 Intrinsic Membrane Proteins 6.1 Sarcoplasmic Reticulum 6.2 T Tubules 7 Comparative Structure of Sarcoplasmic Reticulum and T System 7.1 Fibers With One Small Dimension and No T System 7.2 Fibers With Low Speed and Small Quantity of Membranes 7.3 Fibers With High Speed and Large Quantity of Membranes 7.4 Variation in Speed and Quantity of Membranes in Mammalian Fibers 7.5 Invertebrate Fibers 8 T-System-Sarcoplasmic Reticulum Couplings 8.1 Basic Form 8.2 Structural Details 8.3 Role of T-Tubular Calcium Current in Excitation- Contraction Coupling 8.4 Functional Mechanism of Coupling Between T Tubules and Sarcoplasmic Reticulum 9 Special Geometry of T System 9.1 Longitudinal T Tubules 9.2 Helicoids 9.3 Structural and Functional Implications

A package of computer programs is described for three-dimensional reconstruction. The package was originally written for the reconstruction of sub-cellular structures from a series of images produced by an electron microscope after serial sectioning, but the method can be applied to all objects for which a spatial description of parallel planes can be generated. The data are collected by manually tracing the outlines of the structures in the successive sections. A highly interactive visual editing program aids in the spatial alignment of the sections. The three-dimensional display program employs a new fast hidden line and transformation algorithm and includes a variety of depth cues to enhance the spatial impression. A system for the real time removal of hidden lines from a dynamically changing scene is introduced.

Stereology provides a theoretical basis for powerful morphometric methods for the estimation of three-dimensional structural parameters from two-dimensional electron micrographs of cells and tissues. These methods assume at the start that one has a sufficiently large set of micrographs containing valid structural data. The task of obtaining from these micrographs the large quantity of data needed to get statistically valid results has been eased in two general ways. Sampling of data in the micrograph can be done rapidly by point and intersection counting methods. An alternate method, planimetry, obtains all the data in the micrograph, but in general is more time-consuming than point and intersection counting. Some of the relative inefficiency of planimetry is compensated when a digital planimeter is coupled with a computer. Areas and lengths can be computed simultaneously as fast as profiles are traced. Furthermore, rapid and numerically accurate compilation and statistical analysis of the data can be done automatically as the planimetry is done, not as a separate step after the data have been obtained.

Three-dimensional structural parameters can be obtained quantitatively from two-dimensional images using the methods of stereology and morphometry. One method of obtaining data for a morphometric analysis is digital planimetry. An easily constructed and inexpensive digital planimeter for morphometry is described. The programs written specifically for this system generate lengths and areas as the operator traces along lines and profiles in a micrograph placed on a digitizing tablet. A number of operational options can be selected by the operator, including measurement of open or closed lines, summation of a series of profiles, automatic closure, etc. The spatial accuracy and speed of the system assures that accuracy is not limited by the device itself but only by the operator's ability precisely to follow contours in the micrograph.

One way to take advantage of the increased penetrating power of the high energy electrons in a high voltage electron microscope for the study of biological cells and tissues is to examine thicker sections or slices of embedded tissues. Whereas a thickness of 0.1 micrometers usually is considered to be the useful limit on section thickness for electron microscopy at 100 kV, one can reasonably expect this limit to be increased ten-fold when using accelerating voltages in the range of 1000 kV. In fact, experience has shown that one can go considerably beyond this in thick ness and still get images that are useful for biological analysis. The examination of slices of tissues up to several micrometers thick offers clear advantages over the study of thinner sections in the elucidation of three-dimensionals tructure. This approach can be contrasted to the study of whole cell mounts, as discussed in other papers in this symposium. Whole cells usually are examined after critical point drying, without embedding. Such specimens, suspended in a vacuum, present little problem with respect to contrast. Embedded and sectioned tissues, however, are suspended in a plastic embedding material, which itself has considerable electron scattering power. Therefore, unless the tissue elements are well stained relative to the surrounding embedding material, their visibility and contrast can be insufficient for useful and interpretable images, especially in thick slices.

An electron microscopic study of internal and surface-connected membrane systems of leg muscle of the crab shows that there are three kinds of surface-connected membrane systems in addition to an intracellular sarcoplasmic reticulum (SR). One is a system of large infoldings of the sarcolemma referred to as clefts. These are longitudinally-oriented, flattened infoldings of both the plasma membrane and the fibrous sheath of the fiber, and were probably seen earlier with the light microscope. Extending into the fiber both from these clefts and from the free fiber surface are two systems of tubules of much smaller caliber, the Z tubules and the A tubules. The Z tubules are located, as their name indicates, near the Z lines of the myofibrils, and are thought to be attached to them mechanically. The A tubules are found in pairs, near the ends of each A band, and are closely bound to the SR in two-part structures called dyads. Local-activation experiments, like those done earlier by Huxley and Taylor, suggest that the A tubules are involved in excitation-contraction coupling; no such experimental suggestion of function exists for the Z tubules.

Knowledge of the thickness of sections is important for proper interpretation of electron micrographs. Therefore, the thicknesses of sections of n-butyl methacrylate polymer were determined by ellipsometry, and correlated with the color shown in reflected light. The results are: gray, thinner than 60 mmicro; silver, 60 to 90 mmicro; gold, 90 to 150 mmicro; purple, 150 to 190 mmicro; blue, 190 to 240 mmicro; green, 240 to 280 mmicro; and yellow, 280 to 320 mmicro. These results agree well with optical theory and with previous published data for thin films. Sections, after cutting, are 30 to 40 per cent shorter than the face of the block from which they were cut. Only a small improvement results from allowing the sections to remain in the collecting trough at room temperature. Heating above room temperature, however, reduces this shortening, with a corresponding improvement in dimensions and spatial relationships in the sections. When the thickness of the section is considered in interpreting electron micrographs instead of considering the section to be two-dimensional, a more accurate interpretation is possible. The consideration of electron micrographs as arising from projections of many profiles from throughout the whole thickness of the section explains the apparent lack of continuity often observed in serial sections. It is believed that serial sections are actually continuous, but that the change in size of structure through the thickness of one section and the consideration of only the largest profile shown in the micrograph can account for the lack of continuity previously observed.

The structure of the urinary bladder of the toad Bufo marinus was studied by light and electron microscopy. The epithelium covering the mucosal surface of the bladder is 3 to 10 microns thick and consists of squamous epithelial cells, goblet cells, and a third class of cells containing many mitochondria and possibly representing goblet cells in early stages of their secretory cycle. This epithelium is supported on a lamina propria 30 to several hundred microns thick and containing collagen fibrils, bundles of smooth muscle fibers, and blood vessels. The serosal surface of the bladder is covered by an incomplete mesothelium. The cytoplasm of the squamous epithelial cells, which greatly outnumber the other types of cells, is organized in a way characteristic of epithelial secretory cells. Mitochondria, smooth and rough surfaced endoplasmic reticulum, a Golgi apparatus, "multivesicular bodies," and isolated particles and vesicles are present. Secretion granules are found immediately under the plasma membranes of the free surfaces of the epithelial cells and are seen to fuse with these membranes and release their contents to contribute to a fibrous surface coating found only on the free mucosal surfaces of the cells. Beneath the plasma membranes on these surfaces is an additional, finely granular component. Lateral and basal plasma membranes are heavily plicated and appear ordinary in fine structure. The cells of the epithelium are tightly held together by a terminal bar apparatus and sealed together, with an intervening space of only 0.02 mµ near the bladder lumen, in such a way as to prevent water leakage between the cells. It is demonstrated in in vitro experiments that water traversing the bladder wall passes through the cytoplasm of the epithelial cells and that a vesicle transport mechanism is not involved. In vitro experiments also show that the basal (serosal) surfaces of the epithelial cells are freely permeable to water, while the free (mucosal) surfaces are normally relatively impermeable but become permeable when the serosal surface of the bladder is treated with neurohypophyseal hormones. The permeability barrier found at the mucosal surface may be represented, structurally, either by the filamentous layer lying external to the plasma membrane, by the intracellular, granular component found just under the plasma membrane, or by both of these components of the mucosal surface complex. The polarity of the epithelial sheet is emphasized and related to the physiological role of the urinary bladder in amphibian water balance mechanisms.

Electron microscopic evidence is presented, from mitochondria in whole cells of toad urinary bladder and from isolated rat kidney mitochondria, indicating that the divalent cations calcium, strontium, and barium are accumulated in granules localized in the mitochondrial matrix. This accumulation occurs under conditions in which divalent ions are present in the medium bathing either whole cells or isolated mitochondria. The evidence indicates that the divalent ions are deposited on, or in a pre-existing granule, possibly in exchange for other ions. It suggests a possible role of the intramitochondrial granules in the regulation of the internal ionic environment of the mitochondrion. Certain biochemical and physiological implications of this phenomenon are discussed.

Mitochondria were isolated from livers of normal rats and rats made thyrotoxic by daily injection of 1 μg/g body weight of 3,5,3'-triiodo-L-thyronine for 9 days. These mitochondria were incubated at 37 C for 1 hr. In addition, mitochondria isolated from normal rats were incubated in the presence of 5X10-5M L-thyroxine. Samples taken from the incubation mixtures at various times were examined by electronmicroscopy. It was shown that control mitochondria incubated without hormone present undergo structural changes before they lose matrix material. These changes consist of some corrugation of outer membranes, and a clumping of the intramitochondrial granules. Later, after about 20 min of incubation, matrix is lost from the mitochondria, and the mitochondrial membranes become diffuse in appearance. Thyroid hormones, given in vivo or added in vitro, eliminate the clumping of intramitochondrial granules, accelerate the loss of matrix, and improve the appearance of mitochondrial membranes as they are seen late i...

The structure of the longitudinal body muscles of Branchiostoma caribaeum has been studied by light and electron microscopy. These muscles are shown to be composed of fibers in the form of flat lamellae about 0.8µ in thickness, more than 100 µ wide, and reaching in length from one intermuscular septum to the next, a distance of about 0.6 mm. Each flat fiber is covered by a plasma membrane and contains a single myofibril consisting of myofilaments packed in the interdigitating hexagonal array characteristic of vertebrate striated muscle. Little or no sarcoplasmic reticulum is present. Mitochondria are found infrequently and have a tubular internal structure. These morphological observations are discussed in relation to a proposed hypothesis of excitation-contraction coupling. It is pointed out that the maximum distance from surface to myofilament in these muscles is about 0.5 µ and that diffusion of an "activating" substance over this distance would essentially be complete in less than 0.5 msec. after its release from the plasma membrane. It is concluded that the flat form of amphioxus muscle substitutes for the specialized mechanisms of excitation-contraction coupling thought possibly to involve the sarcoplasmic reticulum in higher vertebrate muscles.

1. Exposure of isolated liver mitochondria to high doses of gamma rays from a Co60 source causes the level of DNase II activity to increase. Treatment of the mitochondria with sonic vibration causes a further elevation of the activity to a level which is independent of the prior radiation dose. 2. Such increased mitochondrial DNase II activity appears to be due to the "structural damage" of the subcellular particulates caused by the ionizing radiation. Other methods of disrupting the mitochondrial structure also cause increased DNase II activity. A causal relationship between the structural alteration and the increased enzymatic activity is postulated. 3. The DNase II activity appears to be closely associated with the structural elements of the mitochondria and remains associated with the fragments after irradiation. 4. Upon irradiation, the mitochondrial suspension releases ultraviolet-absorbing materials which are probably nucleotide in nature. 5. The possibility of localization of DNase activity in the lysosome fraction of de Duve (15) is discussed. It is felt that DNase II is at least in part a mitochondrial enzyme and that probably the conclusions drawn here would be applicable to any DNase II present in the lysosomes as well. 6. Irradiation of whole liver homogenate causes no increased DNase II activity. The experiments do not provide any information on the presence or action of protective substances in the homogenate.

In 1902, Veratti published a description of a fine reticular network pervading striated muscle fibers that could be revealed by infiltration with osmium and silver, a procedure developed by Golgi and referred to as his black reaction (Veratti, 1902, 1961). Much later it was shown by electron microscopy that Veratti’s “reticulum” corresponds to a network of transversely oriented tubules (T tubules) that can be visualized by electron microscopy (Porter, 1961; Smith, 1961; Peachey, 1965). T tubules are invaginations of the plasma membrane at the surface of the muscle fiber, and their content is continuous with the extracellular space. Thus Veratti used the osmium-silver technique to infiltrate extensions of the extracellular space, rather than to infiltrate the cell’s interior, as is believed to be the case when the same method is used to study neurons (Ramon y Cajal, 1972). We have slightly modified the original black reaction of Golgi in a way that results in a superior preservation of fine structural details for light and electron microscopy. Our results show that this procedure is ideally suited for the visualization of small and large invaginations of the extracellular space into cells, both by light microscopy and by electron microscopy in the range of 100-1000 kV. Test objects used in this study were a variety of types of muscle fibers whose surface topography is greatly complicated by the presence of clefts and various tubular invaginations. The highlights of our results are as follows

The extraction and display of 3-D information using modern computer graphic equipment in the electron microscope is presented. Thick specimens were imaged at 400 kV. Modelling of vectors and the tilting of surfaces is discussed as is the accuracy of reconstructions.

The disposition of surface invaginations (clefts, Z and T tubules) and of the sarcoplasmic reticulum has been examined by electron microscopy at three accelerating voltages (100, 200 and 1000 kV) and by phase-contrast light microscopy in crustacean muscles infiltrated by the "Golgi stain." In long-sarcomere, tonic type fibers, an extensive system of invaginating clefts has been observed, along with both Z and T tubules. Z and T tubules form interconnections with each other, but only T tubules form specific contacts with the sarcoplasmic reticulum, which in these fibers forms an extended and continuously fenestrated network. In short-sarcomere, phasic type fibers, a ladder-like disposition of an abundant T network is found. Z tubules are absent in these fibers. The sarcoplasmic reticulum forms more frequent junctions with flattened areas of T tubules and with clefts, but has less extensive free surfaces than in the long-sarcomere fibers.

One area in cell biology in which considerable advance has been made over the last 25 years is the study of how internal membranes in skeletal-muscle cells [specifically the transverse tubular system (T-system) and sarcoplasmic reticulum (SR)] are involved in the control of the contractile state of the cell. This chapter will review this research and summarize the present state of our understanding in this area (for a broader review of muscle-cell biology over the last 25 years, see Franzini-Armstrong and Peachey, 1982). As will be seen, we now have a quite complete understanding of how the T-system of a variety of types of skeletal-muscle fibers is constructed and how it works. Furthermore, this knowledge from muscle cells, in which contractile and control mechanisms are highly elaborated and relatively easy to study, should turn out to be useful in studies of nonmuscle cells, where one might perhaps expect to find similar, though probably simpler, control systems.

Publisher Summary This chapter focuses on three-dimensional structure of the T-system of skeletal muscle cells. The T-system or transverse tubular system, of a striated muscle cell is a predominantly transversely oriented network of interconnected tubules invaginating into the fiber from many points on its surface. These tubules are believed to provide a route for electrical excitation to spread into the fiber interior to activate contraction. The morphological studies through 1975 defined the T-system as a series of planar, continuous, transverse networks crossing the fiber at regular intervals in registration with the sarcomere banding. More recent morphological studies have revealed several ways in which the T-system deviates from this pattern, either by having a longitudinal component or with some form of interruption in its planar arrangement. This chapter reviews and summarizes these newer findings. Most of the results described in the chapter are based on studies using selective stains to enhance the density of the T-system and high voltage electron microscopy to examine thick preparations, often stereoscopically.

The need for information about the thickness of sections used for electron microscopy is the direct result of electron optics, and arises from the fact that the depth of field is such that the whole thickness of the section is focused in the final image. The electron image is, then, a superposition of many profiles from various levels in the section. Analysis of such images requires knowledge of the depth of the specimen in addition to the lateral dimensions displayed in the image. The effect of specimen depth can be considerable in electron microscope images, where structures with dimensions of a few tens of mμ are often studied in sections with thickness in the range of 100 mμ. Therefore, consideration of section thickness is indeed essential for proper interpretation of electron images of sections.

The sarcoplasmic reticulum of the frog's sartorius muscle was examined by electron microscopy following sequential fixation in glutaraldehyde and osmium tetroxide and embedding in Epon. The earlier results of Porter and Palade on Ambystoma muscle were confirmed in the sartorius. In addition, the transverse tubules were observed to be continuous across the width of the fiber, a set of flat intermediate cisternae was seen to connect the terminal cisternae to the longitudinal tubules in the A band, and the continuous reticulum collar at the center of the A band was found to be perforated by circular and elongated pores (the fenestrated collar). The transverse tubules have a volume about 0.3 per cent of the fiber volume, and a surface area about 7 times the outer cylindrical surface area for a fiber 100 µ in diameter. The terminal cisternae, the intermediate cisternae, and the longitudinal tubules together with the fenestrated collar each have a volume of 4 to 5 per cent of the fiber volume and a surface area 40 to 50 times the outer surface area of a fiber 100 µ in diameter. Some evidence for continuity of the transverse tubules with the fiber surface is presented, but this is thought to be not so convincing as evidence presented by others. The results are discussed in terms of a possible mechanism for a role of the transverse tubules and sarcoplasmic reticulum in excitation-contraction coupling, as suggested by their morphology and a variety of physiological studies. In this scheme, the transverse tubules are thought to be electrically coupled to the terminal cisternae, so that depolarization of the fiber surface spreads inward along the transverse tubules and to the terminal cisternae, initiating the release of a contraction-activating substance.

Publisher Summary This chapter discusses the electrical properties of the transverse tubular system. The electrical properties of nerve and muscle are analyzed in terms of the ionic and capacity currents flowing across the cell surface membrane. Morphological evidence suggests that the tubular system can be represented in frog twitch striated muscle fibers as a set of regular two-dimensional tubular networks in which the tubular diameter is smaller than the mesh and the mesh is smaller than the fiber diameter. Access resistance occurs because of constriction of the tubular lumen at the fiber surface or tortuosity of the tubular path near the surface. Either an increased tubular length or a reduced luminal caliber over a significant length increases the radial resistance near the fiber surface. The chapter also presents a linear and a nonlinear method for obtaining the spatial distribution of potential in the tubular system.

A hypothesis, suggested previously by morphological studies, for impulse conduction from the sarcolemma to the contractile material via the sarcoplasmic reticulum is discussed. The relation of reticulum morphology and cell size to speed of contraction in smooth and striated muscle agrees with the hypothesis and thus supports it. Additional support comes from evidence concerning an unusual morphological relationship between the sarcolemma and contractile fibrils in striated muscle of amphioxus.

Our present understanding of the distribution and morphology of the sarcoplasmic reticulum (SR) in frog slow and twitch muscle fibers has been derived largely from the examination of thin sections by electron microscopy. This conventional approach to the study of an organelle as complex as the SR is limited to a degree by section thickness, and the extraction of three-dimensional information must usually be gathered from an extensive collection of two-dimensional images. The present study represents an alternative approach to the problem of investigating the three-dimensional organization of the SR by utilizing high voltage electron microscopy (HVEM) and examining stereoscopic images of selectively stained 1.0 /μm thick slices of muscle tissue.Slow and twitch fibers from the distal fiber bundles of the frog (Rana pipiens) cruralis muscle were processed for electron microscopy according to the selective SR staining technique (DAB-H2O2 and Os-ferrocyanide) developed by Waugh. Tissue slices from 0.25 to 1.0 μm in thickness were cut on a diamond knife, mounted on grids either with or without plastic support films, and examined using the JEM-1000 microscope at the University of Colorado operating at an accelerating voltage of 1000 kV.

Knowledge of the form, distribution, and quantity of transverse tubules (T-system) in frog skeletal muscle has been useful in understanding the inward spread of the electrical signal that triggers contraction.Utilizing selective staining and high voltage electron stereoscopy, we have re-examined the T-system in frog slow fibers with the hope of extending the findings of earlier structural studies and of providing a more complete morphological foundation for correlation with physiological models.The T-system of slow fibers was selectively stained by soaking distal fiber bundles of frog (Rana pipiens) cruralis muscles in horseradish peroxidase and incubating according to the method of Graham and Karnovsky. Transverse and longitudinal 1.0 μm thick slices were examined at an accelerating voltage of 1000 kV using the JEM-1000 microscope at Boulder, Colorado. Preliminary HVEM observations of cruralis muscle slow fibers have revealed an aplanar network of cylindrical tubules that follow a tortous pathway and exhibit a considerable longitudinally oriented component (Figure 1).

The Intermediate Voltage Electron Microscopy and Biomedical Image Analysis facility at the University of Pennsylvania is a National Resource supported by NIH and provides users with facilities for transmission electron microscopy and digital image processing. The facility comprises a JEOL 4000EX transmission electron microscope with a 360 degree goniometer specimen holder, a VAX 11/750 computer, and a Raster Technologies Inc. Model One/380 Graphics work station with two high resolution 1280x1024 pixel color video display monitors for stereoscopic display. We are using this facility to examine the morphology of fibroblasts migrating through fibrillar collagen gels: these matrices closely model the environments encountered by migratory cells during embryonic morphogenesis.The identification and three dimensional localization of structures in stereoscopic electron micrographs of thick sections of cells can be hampered by diffuse boundaries and by the superposition of details with similar electron density.

The effective study of biological tissues in thick slices of embedded material by high voltage electron microscopy (HVEM) requires highly selective staining of those structures to be visualized so that they are not hidden or obscured by other structures in the image. A tilt pair of micrographs with subsequent stereoscopic viewing can be an important aid in three-dimensional visualization of these images, once an appropriate stain has been found. The peroxidase reaction has been used for this purpose in visualizing the T-system (transverse tubular system) of frog skeletal muscle by HVEM (1). We have found infiltration with lanthanum hydroxide to be particularly useful for three-dimensional visualization of certain aspects of the structure of the T- system in skeletal muscles of the frog. Specifically, lanthanum more completely fills the lumen of the tubules and is denser than the peroxidase reaction product.