TOXICOLOGY

AND

APPLIED

PHARMACOLOGY

106,

CONTEMPORARY

355-374

(1990)

ISSUES

IN TOXICOLOGY

Disruption of Cellular Elements and Water in Neurotoxicity: Studies Using Electron Probe X-Ray Microanalysis RICHARD

M. LOPACHIN,

JR., AND ALBERT

J. SAUBERMANN

Department of Anesthesiology, Medical School, SUNY at Stony Brook, Stony Brook, New York 11794-8480; and Department of Applied Science, Brookhaven National Laboratory, Upton, New York 11973

Received January 11. 1990; accepted August 29, 1990 Disruption

of Cellular Elements and Water in Neurotoxicity: Studies using Electron Probe XR. M., AND SAUBERMANN, A. J. (1990). Toxicol. ~ppl. Pharmacol.

RayMicroanalysis. LOPACHIN. 106,

355-374.

0 1990 Academic

Press, hc.

Substantial researchin the field of Toxicology is devoted to identifying specific biochemical mechanisms by which chemicals (e.g., xenobiotics, pesticides,drugs) and diseaseprocesses (e.g., tumor production, diabetic neuropathy) causecellular injury and death. Accumulating evidence indicates that the structural and functional consequencesof a variety of injurious processesmight be mediated by a lossof element (e.g., Na, K, Ca) and water regulation (Macknight, 1984, 1987; Trump et al., 1979, 1989). This concept is being actively explored in various areas of Toxicology and, based on corresponding research,a role for Ca and other elements is implicated in the cytotoxic mechanismsof certain chemicals (Choi, 1987; Kutty et al., 1989; Orrenius, 1985; Pounds and Rosen, 1988; Starke et al., 1986; Trump et al., 1989). Several techniques (e.g., atomic absorption spectrophotometry, ion selective electrodes) have been used to measure levels of ions (or elements) in normal and injured cells. One very promising method is electron probe X-ray microanalysis (EPMA). This quantitative electron microscope technique can correlate cellular morphological structure with elemental composition and water con355

tent. In this commentary we will discussthe application of X-ray microanalysis to studies examining the involvement of elements and water in cell injury and death. Principles of microprobe analysis will be summarized, as well as methods of quantification, sample preparation, and potential technical problems. We will focus on relevant research from the field of neurotoxicology, however, the concepts and methodology described have been applied to other toxicological disciplines (see reviews by Ingram et al., 1989a; Trump et al., 1979, 1989). To provide a background for our discussionof the contributions (both potential and those realized) of EPMA to neurotoxicology, we will begin with a brief discussion of element and water deregulation asa possible mechanistic basis for certain neuropathological conditions. ELEMENT AND WATER REGULATION IN NORMAL AND INJURED NERVE CELLS Under normal steady-state conditions intracellular levels of diffusible ions are tightly regulated by complex processeswhich involve 0041-008X/90

$3.00

Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in any form resewed.

356

LOPACHIN

AND

ion transport across membranes and binding to specific and nonspecific sites (Baker, 1986; McBumey and Neering. 1987; Nicholls, 1986; Stahl, 1986). Figure 1 illustrates some of the major processes involved in nerve cell ion translocation and homeostasis. The importance of elemental regulation to neuronal viability is exemplified by the high percentage (>50%) of total energy expenditure dedicated to maintaining ion gradients (Ritchie, 1967). Much of this energy is consumed by Na KATPase and other ion pumps (e.g., Ca ATPase) located in plasmalemma and organelles (Greene and Lattimer, 1984; Simmons et al., 1982). In addition to nerve cell-based regulatory systems, glial cells also contribute significantly to neuronal ion homeostasis.By a process known as spatial buffering, a localized increase in extracellular Kf (due to pathogenic or neuronal activity) is cleared by perineuronal glial cells. The accumulated K+ (or an equivalent amount) is then releasedby glial cells at a site distal to the area of localized neuronal activity. Here the nerve cell can reaccumulate lost Kf by Na K-ATPase activity (Walz and Hertz, 1983). Not only is spatial buffering important for K+ homeostasis, but it is critical for proper nerve cell function since, if allowed to accumulate, extracellular K+ can cause abnormal nerve cell excitability (Rasminsky, 1980). The multifaceted nature of ion regulation (Fig. 1) in nerve cells provides numerous sites at which toxicants or diseaseprocessesmight act to promote deregulation. For example, inhibition of mitochondrial energy production might compromise elemental homeostasisby secondarily reducing the activities of ATP-dependent membrane ion pumps. Intracellular ion concentrations might be affected by a neurotoxicant that alters the capacity of pump or exchanger proteins to translocate ions (e.g., by changing tertiary structure or binding sites). Since the nerve cell membrane acts asa semiselective permeability barrier to Na+ and other ions (Keynes and Ritchie, 1965) events which alter membrane integrity (e.g., lipid peroxidation) can cause a global lossof trans-

SAUBERMANN

FIG. I Elemental regulation and translocation in normal nerve cells. Ser, smooth endoplasmic reticulum: Nut, nucleus; 1, mitochondrial Ca’+ uptake system; 2, mitochondrial Na+-Ca2+ exchanger; 3, Ca’+-ATPase of Ser; 4, messenger-initiated release of Ca’+; 5, reversible membrane Na’-Ca’+ exchanger; 6, membrane Ca*+-ATPase: 7, NatH+ exchanger; 8, Ca’+-binding cytoplasmic proteins; 9, passive ion movement down electrochemical gradients: 10, voltage-dependent Ca*’ entry; 11, plasmalemma Na+/ K+-ATPase; 12, nodal Na+/K+-ATPase.

membrane gradients. A decreasein either the rate or carrying capacity of axonal transport might diminish axonal delivery of protein constituents (e.g., Na/K-ATPase) which are vital to ion and water regulation. Such alterations in axonal transport (anterograde and retrograde) are associated with exposure to many chemical neurotoxicants (e.g., acrylamide, 2,5-hexanedione) and have been suggestedaspossibleetiologies in certain acquired and inherited neuropathological conditions (e.g., diabetes) (Griffin and Watson, 1988; Jakobsen et al., 1986; Ochs, 1987). Alternatively, membrane ATPase pumps might be perturbed when neurotoxic events disrupt cellular processessuch as phosphoinositide turnover and the phosphorylation of catalytic subunits which regulate or modify pump activity (Ling and Cantley, 1984; Simmons et al., 1986). For example, recent studies of sciatic nerve from streptozocin-diabetic rat have shown that Na/ K-ATPase activity is significantly reduced (Greene et al., 1988). It has been suggested

ION

AND

WATER

DiSRUPTION

(Greene et al., 1988) that this depression of pump activity is related to abnormal polyphosphoinositide metabolism which also occurs in peripheral nerve of diabetic rats (Bell et al., 1982). Finally, the spatial buffering capacity of Schwann cells or astroglia might be affected by a neurotoxic condition and thereby lead to alterations in extracellular K+ levels and neuronal excitability.

IN

NEUROTOXICITY

357

generated by impermeant anionic cytoplasmic solutes is balanced by low membrane permeability to Na+ and by active pump and exchange mechanisms for Na+, K+, Cl-, and protons. Water regulation will, therefore, be disturbed by any process that affects elemental disposition. In addition, changes in metabolism of impermeant organic solutes will alter osmolarity of the cytoplasm and cause secondary changes in water content (Macknight, 1987).

CONSEQUENCES OF ION AND WATER DEREGULATION The proposed cytotoxic consequences (both primary and secondary) of ion and water deregulation are complex, numerous, and diverse (Pounds and Rosen, 1988; Siesjo, 1981; Trump et al., 1979, 1989). Direct or indirect disruption of cellular Ca2+ homeostasis and/ or the messenger role of Ca’+ can inhibit mitochondrial function and axonal transport, promote cytoskeletal breakdown, and alter neurotransmission and receptor-mediated responses (Carafoli, 1987; Keith et al., 1983; Ochs et al., 1986; Pounds and Rosen, 1988; Yamamoto et al., 1983). Elevation of axoplasmic Ca*+ due to ion influx or release from intracellular stores (e.g., mitochondria) can initiate Ca2+-dependent degradative enzyme activity (e.g., neutral proteases, phospholipase n2) and appears to be related to the generation of free radicals by the arachidonic acid cascade (Banik et al., 1987; Braughler, 1988). Injuryinduced loss of cytoplasmic K+ can interfere with protein synthesis and with glycolytic and mitochondrial energy production (Lubin, 1967; McIlwain and Bachelard, 197 1). Subsequent accumulation of K+ in the extraaxonal space of nervous tissue has been linked to abnormal axonal electrophysiological activity and to cytotoxic brain edema (Rasminsky, 1980; Walz, 1988). Control of cell water content is intimately linked to elemental regulation and is described by a modified Gibbs-Donnan pump-leak hypothesis (Macknight, 1984, 1987). According to this proposal, the potential osmotic force

EVIDENCE FOR ALTERED ELEMENTAL HOMEOSTASIS AS A COMPONENT OF NEUROTOXICITY Morphological and electrophysiological evidence suggest that the central-peripheral distal axonopathy caused by acrylamide (ACR) administration might involve alterations in intracellular levels of Ca2+ (Cavanagh and Gysbers, 1983; Goldstein and Fincher, 1986; Jones and Cavanagh, 1984). In addition, voltage clamp studies of proximal sciatic nerves from ACR-intoxicated rats have provided evidence of increased Na’ and K+ permeabilities at nodes of Ranvier (Brismar et al., 1987). The delayed axonopathy caused by organophosphate (OP) exposure might also involve abnormal translocation of element5 across nerve cell membranes. In sciatic nerve of OP-treated hens, electron probe X-ray microanalysis demonstrated significant changes in axoplasmic and mitochondrial levels of elemental Na, K, Cl, and Ca (LoPachin et al., 1988a). Moreover, a role for Ca is implicated by the finding that Ca channel blockers protect animals from certain behavioral and biochemical changes associated with OP poisoning (Dretchen et al., 1986; El-Fawal et al., 1989). Treatment of laboratory animals with ,&?‘-iminodipropionitrile causes proximal axonal ballooning and aberrant “cross talk” among affected spinal cord neurons (Chou and Hartman, 1964; Delio et al., 1987). Delio et al. (1987) suggested that this latter electrophysiological abnormality might be due to swelling-induced paranodal

358

LOPACHIN

AND

demyelination with subsequent exposure of Kf channels and leakage of this ion into the extracellular space. Several lines of evidence suggest that shifts in intracellular ions play an important role in the neurotoxicity caused by heavy metals (e.g., lead, mercury) and specific neuropoisons such as kainate and capsaicin (Audesirk, 1985; Johnson et al., 1986; Komulainen and Bondy, 1987; Korf and Postema, 1984). Nervous tissue ischemia results in cell death and marked increases in tissue or cell levels of Na+, Cll, and Ca2+ in conjunction with a loss of K+ (Cheung et al., 1986; Dienel, 1984; Kass and Lipton, 1982; LoPachin et al., 1989a; Martins et al., 1988). Anoxic nerve cell death appears to be mediated by excess release of the amino acid neurotransmitter, glutamate (Olney et al., 1986). Recent in vitro studies suggest that neuronal death caused by this excitotoxin involves sequential transmembrane fluxes of Na+, Cl-, K+, and Ca2+ (Choi, 1987; Olney et al., 1986; Rothman, 1984). Studies examining the mechanism of axonal injury induced by experimental spinal cord contusion damage represent an area of active research with obvious clinical relevance. Use of atomic absorption spectrophotometry and nuclear magnetic resonance have shown that total spinal cord tissue levels of elemental Ca, Na, Mg, and K were perturbed at several times following injury (Lemke et al., 1987: Vink et al., 1989; Young and Koreh, 1986). Additional studies using ion selective microelectrodes showed a reduction of extracellular Ca2+ which presumably reflected rapid entry of this ion into injured spinal cord axons (Beattie et al., 1989). The functional consequences of spinal cord trauma, and many of the other injury mechanisms described above, can be related directly to Wallerian degeneration of axons (to be discussed). Several studies suggest that Wallerian degeneration is mediated by a sequential change in axonal levels of diffusible elements (LoPachin et al., 1990; Schlaepfer, 1974; Schlote et al., 1981). Finally, electrophysiological studies indicate that experimental diabetic peripheral neuropathy is associated

SAUBERMANN

with altered intraaxonal Na+ and K+ levels presumably due to defective Na/K-ATPase activity (Brismar and Sima, 198 1; Greene and Lattimer, 1984). Thus, a variety of nerve injury mechanisms are associated with intracellular changes not only in Ca but also in the content of other elements (Na, K, Cl). In some cases these changes precede structural and functional alterations which suggests a mechanistic relationship between elemental translocation and the consequences of damage. Our perception of elements and their involvement in the injury-death continuum is evolving from a strictly latent or epiphenomenal role in the mediation of cell death to a more dynamic, primary participation in initial injury mechanisms. The advent of microanalytical methods (e.g., ion selective electrodes, fluorescent dyes) with improved detection and quantification of elements and ions is facilitating this transition. MEASUREMENTS ELEMENTAL

OF NERVE CELL COMPOSITION

Given the potential participation of elements (or ions) in neurotoxicity, how have corresponding measurements been made? Total elemental content (i.e., ionized and bound) has been determined at the tissue or in vitro cell level using atomic absorption spectrophotometry or flame photometry. The resulting information is limited since it does not identify cell-specific differences in elemental composition nor does it indicate the subcellular distribution of an element. Spectrophotometry in conjunction with cell fractionation has been used in an attempt to measure elemental concentrations in cellular fragments (e.g., synaptosomes), organelles (e.g., mitochondria), or regions (e.g., cytoplasm). However, subfractionation of tissues or cells prior to analysis will most likely perturb normal distribution and thereby cause artifactual translocation of elements. The cytoplasmic concentration of an ion has been determined

ION

AND

WATER

DISRUPTION

by ion-selective microelectrodes or ion-specific fluorescent dyes (e.g., fura-2). Although important information is provided by these methods, they are limited to analysis of a single cytoplasmic or extracellular ion. Simultaneous changes in other biological ions or compartments (e.g., mitochondria) cannot be assessed. Cytochemical markers (e.g., pyroantimonate) have also been used to study elemental distribution in nerve cells (Chan et al., 1984; Duce and Keen, 1978). The selectivity of these markers is, however, questionable and chemical fixation used to prepare tissues for this method disrupts the normal distribution of elements (Somlyo, 1985; Van Iren et al., 1979; Wick and Hepler, 1982). If we are to fully understand how shifts in elemental concentrations and water content might be involved in an injury process we need additional information which is not supplied by the methodologies described above. For example, we must know exactly how injury influences both intracellular and extracellular distributions of elements and water. This requires specific identification of the elements involved, respective quantitative changes, and where these changes are taking place (e.g., ECF, SER, mitochondria, cytoplasm). We must also identify the cell type (e.g., neuron vs glia) within which these changes are occurring. Such detailed information can be furnished by electron probe X-ray microanalysis. This electron microscope technique can simultaneously quantify total (i.e., free and bound) concentrations of biologically relevant elements (2 > 10) and water content in cellular morphological compartments. EPMA, therefore, provides a unique opportunity to assess the status of cellular regions (Schwann cell cytoplasm, nodes of Ranvier) and organelles which might play a role in the etiology of cell injury and which are otherwise inaccessible to direct biochemical evaluation. Since EPMA data reflect both localization and quantitation, they represent the detailed and comprehensive information necessary to define the role of elements and water in nerve cell injury. The remainder of this review will

IN

359

NEUROTOXICITY

describe the principles and methodologies of EPMA and will discuss application of this technique to neurobiology and neurotoxicology. Other microanalysis methods are available, e.g., laser microprobe analysis, particleinduced X-ray emission analysis, and electron energy loss spectroscopy. Although these microanalytical techniques have great potential, their use in cell biology has been limited by technical problems, expense, and availability (for further information see Roomans, 1988). Because EPMA measures both free (ionic) and bound elements (i.e., total elemental content), in the following sections the valency of individual elements is not indicated. ELECTRON PROBE MICROANALYSIS

X-RAY

Principles and Methods Electron probe X-ray microanalysis combines the electron-optical imaging capabilities of scanning or transmission electron microscopes with the X-ray detection capacity of energy or wavelength dispersive X-ray detectors. The basic technique was developed in the late 1940s by Guinier (1949) and found immediate application in several areas of the physical sciences: physics, metallurgy, electronics, and geology. In the 1960s the pioneering studies of T. A. Hall and colleagues at Cambridge (see review by Hall, 1989) provided the groundwork for microprobe analysis of biological materials. Beginning with the early publications of Hall, the use of EPMA expanded rapidly in the field of biology and to date this technique has been used to study cellular elemental content in almost every organ system (LeFurgey et al., 1988; Moreton, 198 1). Production of characteristic and continuum X-rays. Only a brief description of the principles of X-ray production and detection will be presented here, for a comprehensive discussion the reader is referred to the following: Chandler, 1977; Goldstein et al., 198 1; Roomans, 1988; Somlyo, 1985. When incident

360

LOPACHIN

AND SAUBERMANN

electrons of an electron microscope beam interact with atoms of the sample, two types of X-rays are produced: characteristic X-rays and continuum radiation. Characteristic X-rays are generated when incident high-energy electrons strike inner shell electrons of atoms (Fig. 2). As a result of this collision, an orbital electron is ejected causing inner shell ionization. This labile condition is stabilized by decay of an outer shell electron into the lower energy inner orbit. This electron transition requires releaseof energy in the form of an X-ray photon. The energy of this photon will be equivalent to the energy difference between the two shells.Furthermore, the energy level of the releasedphoton is characteristic for that element since the atomic number of the element determines the discreet binding energiesof each shell. Continuum radiation, also called bremsstrahlung or white radiation, occurs when incident beam electrons are inelastically scattered by the electromagnetic field of resident atomic nuclei (Fig. 2). As a result of this interaction, the incident electron looses an amount of energy which ranges from zero to the initial ionization energy of primary electrons. Continuum is important for quantitative analysis of elements for two reasons; ( 1) it represents the background on which characteristic X-ray lines or peaks are superimposed and, therefore, determines the minimum peak-to-background ratio for detectable elemental concentrations, and (2) continuum intensity (number of counts) is a function of the total number of all atoms in the analyzed compartment and is therefore a measure of corresponding mass. Detection ofX-rays. Emitted X-ray photons (characteristic or continuum) can be detected using either their wavelength or energy. Thus, two types of X-ray spectrometershave evolved which detect either wavelength (diffraction or wavelength-dispersive detectors) or energy (solid-state or energy-dispersive detectors) of the photon. A wavelength-dispersive spectrometer can detect only one element at a time and requires a high electron dose.This method

ejected electron

FIG. 2. Simplified Bohr model of an atom illustrating the production of charactetistic and continuum X-rays.

has been used mainly in studies of nondiffusible elements of fixed neural tissues(Duckett et al., 1989). Alternatively, an energy-dispersive spectrometer can perform simultaneous multielemental analysis with lower beam currents and is most commonly usedin biological studies.This detector system consistsof a lithium-drifted silicon Si(Li) semiconductor which can collect and count X-rays of different energy levels. The Si(Li) semiconductor is contained within an evacuated sliding tube which is inserted into the specimen chamber of the microscope. The detector is isolated from the microscope chamber by a beryllium window (7-8 pm thick) and can be brought close to the specimen to optimize X-ray collection. Energy-dispersive (ED) spectrometers can be mounted on either transmission (TEM) or scanning electron microscopes (SEM). TEMED spectrometer systems are generally used in studies requiring high spatial resolution (~20 nm). This type of microanalytical system requiresthe addition of a scanning attachment which allows the electron beam to be rastered within chosen cell regions. In studies where low spatial resolution is acceptable, cellular elemental concentrations can be measured using an SEM-ED spectrometer system. To identify ultrastructural compartments for analysis, scanning microscopes can be fitted with transmitted electron detectors which provide scanning-transmission electron microscopy (STEM). In recent years cold specimen stages (- 185“C) have become an integral component of transmission and scanning microanalytical

ION

AND

WATER

DISRUPTION

systems. Development of the cold stage coincided with the use of cryosections for measurements of diffusible elements (to be discussed). QuantiJication and standardization. In this section we will describe methods of quantification and standardization used in our laboratory. Other methods of quantification are available and the reader is referred to: Gupta and Hall, 1978; Rick et al., 1982; Shuman et al., 1976. To measure elemental concentrations and water content, a continuum is first determined for large regions (60 X 60 pm areas) of frozen, hydrated sections. Analyses of elements or individual compartments are not performed in the hydrated state because it is difficult to identify morphological structures. Once regional hydrated continuums have been acquired, the section is freeze-dried in the vacuum column of the microscope by raising the temperature of the cold stage from - 185” to -60°C (for details of this methodology see Saubermann et al., 1977a, 198 la,b; Saubermann and Heyman, 1987; Saubermann and Scheid, 1985; Saubermann, 1988a,b; Zierold, 1982a,b). Frozen hydrated sections can also be freeze-dried in a separate vacuum chamber and then transferred at room temperature into the microscope (e.g., Somlyo et al., 198 1, 1985a). However, due to exposure of the hygroscopic-dried sections to atmosphere there is a risk of partial rehydration and redistribution of cellular elements (Hagler and Buja, 1984; Zierold, 1982a). Morphological compartments (e.g., mitochondria, ECF) of freeze-dried thin sections (i 500 nm) are easily recognized in STEM images (Figs. 3a, 3b, 3c) and measurements are made while the beam is rastered within chosen compartments. Characteristic X-ray counts are then determined over specific peak energy ranges (e.g., Na = 0.96-l. 12 keV), whereas continuum counts are obtained from a region of the energy spectrum devoid of characteristic peaks (e.g., 4.6-6.0 keV). Generally, for quantification of dry weight concentrations (mmol element/kg dry wt) in thin biological sections,

IN

NEUROTOXICITY

361

the continuum normalization method is used (Hall et al., 1973; Saubermann et al., 1981b). Water content (% H20) of morphological compartments is determined in our laboratory by the ratio of continuum counts in the dried and hydrated states. The absolute wet weight concentration (mmol element/kg wet wt) can then be determined by Cx, = Cx,( 1 - % HzO), where Cx, and Cxd are the wet and dry weight concentrations for element x, respectively (Bulger et al., 198 1; Saubermann and Heyman, 1987; Saubermann, 1988b). Modes of EPMA operation. There are two modes of microprobe operation: selective compartment analysis and digital X-ray imaging. For compartment analysis, the operator manually places the beam over a morphological compartment, generally identified in STEM mode, and then initiates collection of X-rays by the ED spectrometer. Collected counts are then processed and converted to mmol element/kg dry or wet wt. These data are inherently limited since this method can only provide mean (&variance) concentrations for a specified anatomical region. How elements and water are topographically distributed across a compartment is not easily revealed by this method. However, such information can be acquired by digital X-ray imaging (Fiori et al., 1988; Gorlen et al., 1984; Ingram et al., 1989b; Saubermann and Heyman, 1987). This is an EPMA technique where placement of the electron beam is controlled by computer (rather than by operator). Thus, the electron beam can be moved point-bypoint across a specimen region in accordance with a preselected pattern (e.g., 64 X 64 point matrix). X-rays collected at each point are used to form a digital image which is superimposable on corresponding electron micrographs and where each point or pixel is fully quantitative with respect to either elemental concentration or water content. Examples of digital images are presented in Fig. 3. As is evident from this figure, digital imaging combines morphological information with elemental

362

LOPACHIN

AND

distribution and, thus, provides an intuitively understandable map expressing the spatial relationship between anatomic structure and elemental distribution. X-ray mapping also offers excellent statistical sampling, and the analysis is independent of operator bias or interpretation of morphology. The use of X-ray maps is particularly important for studies of nerve cell injury since each map can show how damage causes decompartmentalization of both water and elements. This concept is illustrated by recent studies using digital imaging to map the translocation of elements and water in transected rat sciatic nerve axons and in leech ganglia exposed to ouabain (LoPachin et al., 1989a, 1990; Saubermann and Stockton, 1988). Each map of elemental distribution is analogous to micrographs produced by standard electron microscopic techniques and, therefore, can be considered an expression of elemental (chemical) anatomy. Just as studying conventional morphology has led to an understanding of how form is related to function, studies of elemental anatomy using digital imaging will provide a new perspective on this relationship and, in the case of nerve injury, will define the relationship between chemical form and dys-

FIG. 3. Scanning-transmission electron micrographs and respective composite digital images of a node of Ranvier (a) (X5670) and internodal axons (b) (X6480) from rat sciatic nerve. (3~) A small, dark nerve cell body and associated satellite cell from rat dorsal root ganglion (X413 1). The composite digital image in each panel is formed by superimposing the three individual monocolored images for Na, K, and P located to the left of the large image. Each monocolored image is formed using six grey levels according to the color-coded triad in the upper left corner. Therefore, the hue of each pixel from the composite image represents a unique quantitative mixture ofall three elements as indicated by the exploded color cube. When composite images of, for example, normal and transected axons are compared changes in pixel hue reflect changes in quantitative and spatial relationships among the elements displayed. N, nucleus; n, nucleolus; CYTO, cytoplasm; SAT, satellite cell; LL, large light nerve cell.

SAUBERMANN

*( 3a

FIG. 3-Continued 363

364

LOPACHIN

AND

function. Application of digital imaging to neurotoxicology and neurobiology has opened an exciting new area of research.

Preparation of tissue samples and cryosectioning. Although early EPMA studies used tissues prepared by wet chemical methods, more recent studies showed that such fixation promotes artifactual translocation of elements (Moreton, 198 1; Morgan, 1979; Roos and Barnard, 1985; Somlyo, 1985). Therefore, cryopreparative methods have been developed to ensure rapid freezing of tissue with preservation of normal elemental distribution and morphological structures (Parsons et al., 1985; Saubermann, 1980; Saubermann et al., 198 la). Tissue can be rapidly frozen by a variety of methods, e.g., clamp-freezing using liquid nitrogen-cooled polished metal surfaces (LoPachin et al., 1988b), quench-freezing with super-cooled liquids such as Freon- 12 (Saubermann and Scheid, 1985) or slam-freezing on helium-cooled polished metal (Escaig, 1982). Regardless of the freezing process employed, the goals of cryopreparation are to remove heat rapidly and thereby minimize the growth of ice crystals. The development of large ice crystals (1 pm) in poorly frozen tissues is manifest as a lacy appearance of dried cryosections. This type of damage can make cryosectioning difficult, interfere with recognition of morphological compartments, and can cause displacement of elements. Once the tissue is frozen, it can be sectioned on a cryoultramicrotome. Cryosectioning of frozen biological material is a complex and poorly understood process which, in theory, is consistent with the principles of metal machining (Saubermann, 1980, 1989). A great deal of controversy surrounds the choice of ambient microtome temperature (e.g., < - 100” vs - 5 5 “C) at which tissues are sectioned (for details see Frederik and Busing, 198 1; Karp et al., 1982; Saubermann et al., 1977a,b). Controversy is likely to persist in this area until the physical nature of cryosectioning is understood. Once cut, cryosections are transferred from the knife to a specimen grid which is located within the cryochamber. In

SAUBERMANN

our laboratory, the specimen grid is held by a transport device which can be removed from the cryochamber and transferred, under vacuum,‘to a cold stage in the specimen chamber of a scanning microscope. For work with transmission microscopes, sections are placed on grids located on a cold stage which has been inserted into the microtome cryochamber. When full, the cold stage is removed from the cryochamber, transported to the transmission microscope, and inserted. Frozen hydrated sections can then be freeze dried within the vacuum column of the SEM or TEM. Alternatively, some groups dehydrate their sections in a separate vacuum system as discussed earlier. APPLICATION OF MICROPROBE ANALYSIS TO NEUROTOXICOLOGY In this section we will present results from recent X-ray microprobe studies of nervous tissue. Elemental composition and water content of normal nervous tissue will be described first followed by a discussion of how these parameters are influenced by specific neurotoxic conditions.

Analysis of Unjixed, Unstained Frozen Sections from Peripheral Nerves and Ganglia LoPachin et al. (1988b) used microprobe analysis and digital X-ray imaging to measure water content and concentrations (mmol/kg dry and wet wt) of biological elements in frozen sections (200-500 nm) of rat sciatic and tibia1 nerves. In this study, axoplasm, mitochondria, myelin, extraaxonal space (EAS), and Schwann cell cytoplasm were examined. The results demonstrated that each anatomical compartment exhibits characteristic distributions of elements and water. Thus, both wet and dry weight measurements show the almost exclusive localization of K within sciatic nerve axons, P within myelin and, Na and Cl in EAS (Table 1, Figs. 3a, 3b). Steep transmembrane concentration gradients were also observed for

ION AND WATER

DISRUPTION

TABLE 1 DRY AND WET WEIGHT CONCENTRATIONSOF ELEMENTS IN DISTAL SCIATIC NERVE COMPARTMENTS

Element

mmol/kg dry wt

mmol/kg wet wt

Medium Diameter Axoplasm (92 +- 1% water) Na P Cl K Ca

102 rfr 6 402 k 10 688 +- 18 1914 + 54 1.2 f 0.1 Mitochondria

Na P Cl K Ca

122 477 714 1903 0.7

+ + f k If

7+1 28 + 36 * 115 + 0.1 +

2 3 3 0.01

(84 + 2% water) 7 14 28 82 0.1

II 54 64 165 0.05

+-2 f 6 f 8 f 20 + 0.02

112 423 65 83 1.0

+ 9 + 24 + 5 k4 * 0.1

Myelin (40 ?X4% water) Na P Cl K Ca

194 f 680 f 113+5 133 i 1.7 !z

4 11 10 0.4

Schwann cell cytoplasm (85 + 3% water) Na P Cl K Ca

163 803 179 573 1.9

-ir 13 + 32 f 17 + 28 + 0.4

25 119 45 81 0.3

+ 5 f 12 k 3 k5 + 0.01

118 15 103 13 4

*20 III 3 -t 10 + 1 + 0.2

ECF (94 t 2% water) Na P Cl K Ca

1877 208 1607 180 53

+ 39 + 19 + 135 + 22 +- 4.7

Na, Cl, Ca, and K. Other studies showed a similar distribution of elements in frog sciatic nerve (Rick et al., 1976) and in rat sciatic and mouse cochlear nerve axons (Schlote et al., 1981; Wroblewski, 1989). Analysis of axons from proximal and distal sciatic nerve and from tibia1 nerve showed that K and Cl concentrations decreased as a linear function of

IN NEUROTOXICITY

365

proximodistal distance along the nerve. This phenomenon did not appear to be due to corresponding proximodistal changes in water content. The neurophysiological relevance of this finding is unknown, but might be functionally related to other reported biochemical and electrophysiological proximodistal axonal phenomena (Schrama et al., 1987; Robertson and Anderson, 1986). Microprobe analysis of Schwann cell cytoplasm revealed an elemental composition (i.e., high P and K levels) and a water content similar to those of parenchymal cells from liver or kidney (Bond et al., 1989; Somlyo et ai., 1985a). In rats made diabetic (20 weeks)by injection of streptozocin, LoPachin et al. (1989b) showed that the proximodistal decreasing concentration gradients for K and Cl observed in axoplasm and mitochondria of normal sciatic nerve were reversed; in sections from proximal nerve of diabetic rats axoplasmic K and Cl concentrations were reduced while in tibial sections,the levels of theseelementswere increased relative to those of controls. These alterations were evident on both a dry and wet weight basis and were not due to either proximal or distal changes in water content. Schwann cell cytoplasm in both proximal sciatic and tibial nerve exhibited increased dry and wet weight concentrations of P and K. The changesin elemental composition of peripheral axons are highly selective and might be mechanistically related to certain biochemical effects (e.g., protein phosphorylation) which are also altered asa function of distance along the sciatic-tibia1 nerve axis of diabetic rats (Schrama et al., 1987). In other studies related to diabetic neuropathy, Mizisin et al. (1986a,b) measured elemental content of endoneurial fluid collected by a micropipette from sciatic nerve of galactose-intoxicated rats. Fluid was placed on a solid specimen holder by a microdroplet technique (Myers et al., 1983), water was sublimated, and the remaining material was analyzed by EPMA. Results showed that endoneurial Na concentration (meq/liter) in intoxicated rats was nearly doubled compared to that in controls. The authors

366

LOPACHIN

AND SALJBERMANN TABLE 2

CHANGEFINELEMENTALDISTRIBUTION

RELATIVESIGNIF~CANT

OFAXOPLASM

ANDMITOCHONDRIA

FROMSMALLTRANSECTEDFIBERS

8 hr AX0 Na P Cl K Ca

NC 153 56 65 NC

16 hr MIT0 NC 119 NC NC NC

AX0 321 127 71 62 NC

48 hr MIT0 443 NC NC 68 441

AX0 1194 NC 122 19 2325

MIT0 1500 67 NC 16 8236

n Relative significant changes are calculated from experimental data where control = 100, NC, no significant change from control.

interpret these Na values as osmotically significant and responsible for the endoneurial edema which characterizes galactose neuropathy. Wroblewski (1989) examined the distribution of elements in rat sciatic nerve following acute nephrectomy. Under uremic conditions, it was found that axonal S and K contents increased significantly, whereas Na, Cl, and Ca levels remained unaffected. The author suggests that these findings might be related to axoplasmic changes in neurofilament content and transport. EPMA analysis of rat sciatic nerves made hypoxic in situ (15 min) showed marked increases in axoplasmic dry weight concentrations of Na and Cl in conjunction with a loss of K (LoPachin et al., 1989a). Axoplasmic levels of P, S, and Ca were not significantly altered. Mitochondria of hypoxic nerves demonstrated changes in Na, Cl, and K similar to those of axoplasm. However, mitochondrial Ca and P were increased considerably. These alterations in elemental distribution occurred in myelinated axons regardless of their diameter. Hypoxia did not perturb elemental composition of myelin or extraaxonal space. These changes in elemental anatomy resemble alterations observed in axons 16 hr after transection (next paragraph), but are unlike the discreet elemental changes associated with diabetic neuropathy.

When peripheral and central nerve axons are transected distal portions undergo distinct morphological changes collectively known as Wallerian degeneration. The elemental alterations associated with this degeneration have been examined by microprobe analysis (LoPachin et al., 1989a, 1990). During the first 16 hr post-transection, only small myelinated axons exhibited changes in elemental distribution. On both a dry and a wet weight basis, axoplasm and mitochondria of small fibers lost K and Cl, whereas Na levels increased nearly 3-fold (Table 2). Although axoplasmic Ca did not change during this time period, mitochondrial Ca concentrations rose 4.5fold (Table 2). At 48 hr, intraaxonal compartments of all fibers exhibited large gains in Na, Cl, and Ca, while K levels declined to 20-30% of control. In a previous EPMA study, similar generalized changes in elemental composition were detected in axons 36 hr after sciatic nerve crush (Schlote et al., 1981). In Schwann cell cytoplasm, K, Na, and Cl concentrations increased within the first 8 hr and remained elevated throughout the experimental period. Thus, transection of axons initiates a sequence of elemental changes which might play an important role in the degenerative process. Moreover, initial deregulation of Na, K, and Cl might represent necessary, premonitory events which facilitate later Ca entry and subsequent axon disintegration.

ION

AND

WATER

DISRUPTION

Elernentul distribution in cell bodies of peripheral nerve ganglia. Galvan et al. (1984) measured intracellular wet weight concentrations of Na, K, Cl, and P in frozen sections (1 pm) of rat sympathetic ganglia. Results indicate that cytoplasmic and nuclear regions of cell bodies exhibit similar concentrations of Na (3 * 1 mmol/kg wet wt), K (1.55 f 5 mmol/ kg wet wt) and Cl (25 t- 1 mmol/kg wet wt). In contrast, the level of P in cytoplasm was approximately twice that in nucleus (15 3 f 8 vs 82 t 4 mmol/kg wet wt, respectively). Typical of parenchymal cells, K and P were most abundant of the elements assessed. In vitro exposure of ganglia to ouabain ( 1 mM, 1-hr incubation) or to the muscarinic agonist carbachol(l80 PM, 4-min incubation) produced quantitatively similar changes in Na and K wet weight concentrations; intracellular levels of K decreased (50%) while Na levels increased by 30-fold. However, the significance of these pharmacological data is unknown. It is not clear that proper controls were used for the varying incubation periods, nor was the muscarinic antagonist atropine used to determine specificity of the carbachol response. Cell bodies of dorsal root ganglia (DRG) represent a probable site of action for chemical neurotoxicants (e.g., doxorubicin, mercury), and exhibit biochemical and morphological changes in response to chemical-induced (e.g., acrylamide, hexane) and mechanical-induced axonal damage (Choo, 1977; Jones and Cavanagh, 1984; Sterman, 1984; Sterman and Delannoy, 1985). Although the effects of neurotoxic conditions on elemental distribution and water content in DRG cells have not been assessed, this important ganglia has been examined in normal animals by EPMA (LoPachin et al., 1988b). In frozen DRG sections (200-500 nm), cytoplasm, mitochondria, nucleus, and nucleolus from both large, light and small, dark cell bodies had similar levels of Na, S, Cl, and Ca. These compartments were distinguished by water content and levels of K and P. The dry weight concentration of K in the nuclear region was higher than that in the cytoplasm (1389 & 52 vs 1075 ? 30 mmol/

IN

NEUROTOXICITY

367

kg dry wt, respectively), while this regional relationship was reversed for P (6 14 -t 28 vs 743 +- 25 mmol/kg dry wt, respectively). Hydrated measurements showed that nucleus had a higher water content (83 + 5%) compared to that of nucleolus (52 + 4%) or of cytoplasm (55 f 3%). Correspondingly, the digital image of Fig. 3c shows that when considered on a wet weight basis the concentrations of K and P are lower in nucleus and higher in cytoplasm and nucleolus.

Analysis of Unfixed, Unstained Frozen Sections of Central Nervous Tissue Saubermann and Scheid (1985) examined the distribution of elements and water in neurons and glial cells of normal leech (Mucrobdella decora) CNS ganglia. Ganglia were rapidly removed, placed over wooden pegs, and then quench frozen in melting Freon- 12. Each cell type and corresponding anatomical compartments exhibited distinct elemental patterns and water contents, which the authors suggest reflect cell differentiation and morphological specialization. Like other nerve cell bodies, leech neurons contained low wet weight levels of Na (56 If: 4 mmol/kg) but high K (148 +- 5 mmol/kg) and P (195 + 6 mmol/ kg) concentrations. In contrast, glial cells (perineuronal zone) had a relatively high wet weight Na content (110 + 6 mmol/kg) with lower concentrations of K (50 f 3 mmol/kg) and P (77 -t 4 mmol/kg). Neurons and glial cells could also be distinguished by water content (55 + 1% vs 75 + l%, respectively). The authors point out that, when calculated on a millimole/liter basis, the K and Na levels (274 and 104, respectively) are higher than measurements made with ion selective electrodes (135 and 11, respectively) (Deitmer and Schlue, 198 1; Schlue and Deitmer, 1980). This discrepancy between data suggests that in normal neurons a portion of total Na and K content is bound and unavailable to detection by ion selective electrodes. When leech CNS ganglia were incubated in vitro with increasing concentrations of K and

368

LOPACHIN

AND

Cl (4-20 mivr KCI), the dry weight content of IS in both neuron and glial cells rose as a function of extracellular K (Saubermann and Stockton, 1988). However, wet weight K concentration was not altered since cellular water content in both cells increased as a function of rising intracellular dry weight K. The results show that neurons and glia can act as substantial reservoirs for K and yet maintain normal wet weight concentrations by implementing homeostatic adjustments in water content, The homeostatic response was abolished by incubation with oubain (1 mM), but not with furosemide (2 mM>, which suggests that proper functioning af Na/K-ATPase is a necessary component of the accommodation mechanism. The ability of CNS cells, especially glial cells, to participate in ion regulation at the cellular level (see review by Walz and Hertz, 1983) is highly relevant to neurotoxicology. It is possible that certain neurotoxicants perturb ion regulation by central and peripheral glial cells which secondarily affects function and morphology of nerve cells and axons. The use of microprabe analysis can help identify this potential mechanism of actian for a given neurotoxicant. Microprobe analysis of mammalian CNS tissue is difficult due to the freezing barrier imposed by the skull and vertebral column. To circumvent this barrier, Somlyo et al. ( 1985b) used trephining of rat skull to expose the underlying cerebral cortex. The tissue was rapidly cryofixed by pressing a plug of frozen Freon-22 onto the exposed surface. Cryosections (100 nm) were then cut from the frozen area of cortex and analyzed by EPMA. Because of morphological complexity only two general compartments were identified; mitochondria and cytoplasm. Both compartments were similar with respect to dry weight concentrations of Na, P, S, Cl, K, and Mg. However, cytoplasmic concentrations of Ca were significantly higher than those of mitochondria (6.4 1. 1 vs 1.5 3~0.3 mmol/kg dry wt). Similar results were obtained when elemental composition of Purkinje cell nerve terminals was determined in frozen sections ( 100 nm) of

SAUBERMANN

mouse cerebellar cortex (Andrews and Reese? 1986; Andrews et al., 1987; Fiori et al., 1988). The authors of both studies conclude that the low mitochondrial levels of Ca are not consistent with a role of these organelles in nonpathologic regulation of Ca. Research employing microprobe analysis is beginning to characterize the distribution of elements and water in normal CNS cells and their organelles. Such information, along with the development of necessary cryopreparative methods, provides a foundation for future EPMA studies of CNS nerve damage. Microprobe A4nalysis of Fixed and/or Stained Tismes EPMA has been used to qualitatively localize elements in normal, chemically fixed peripheral and central nervous tissue (Chan et tl2., 1984; Duce and Keen, 1978; Ellisman et al., 1979; Hillman and Llinas, 1974; McGraw et al., 1980; Qschman et al., 1974; Siklos et al., 1983; Stoeckel et al., 1975). Microanalysis has also been used in neurotoxicological studies of fixed nervous tissues from capsaicin- or glutamate-treated rats (Jancso et al., 1984), and from patients with either diabetes (Duckett et al., 1977a; King et al., 1988) or brain tumors (Duckett et al., 1989). Also in fixed tissues, cellular localization of metals (e.g., Te, Al) associated with intoxication or certain disease states (e.g., Alzheimer’s disease) has been determined using EPMA (Duckett et a/., 1977b; Galle et ai., 1980; Garruto etal., 1984, 1986; Long et al., 1980; Per1 and Brody, 1980a,b; Yase, 1980). Substantial evidence suggests that conventional chemical fixation promotes elemental translocation (Moreton, 198 1; Morgan, 1979; Roos and Barnard, 1985; Somlyo, 1985). Since this potential artifact was not assessed in the above studies, the corresponding data are difficult to interpret. SUMMARY Regulation of elements and water in nerve cells is a complex, multifaceted process which

ION AND WATER

DISRUPTION

appears to be vulnerable to neurotoxic events. However, much of our knowledge concerning the potential role of elements in nerve cell injury is limited by the relatively gross level of corresponding analyses. If we are to confirm and understand the proposed role, more precise and detailed information is needed. As indicated in this commentary, research employing electron probe microanalysis and digital X-ray imaging has begun to provide this necessary information. Recent EPMA studies of nerve and glial cells in the peripheral and central nervous systems have shown that each cell type and their corresponding morphologic compartments exhibit unique distributions of elements and water. The use of microprobe analysis has allowed us to document precisely how elements and water redistribute in morphological compartments of damaged nerve cells. Accumulating evidence from EPMA studies suggests that, rather than being an epiphenomenon, intracellular changes in diffusible elements might mediate the functional and structural consequences of neurotoxic insult. It is also evident from this research that elements other than Ca might play a pertinent role in the injury response and that changes in intraneuronal elemental composition might develop according to a specific temporal pattern, e.g., transection-induced sequential alterations in axonal K, Na, Cl, and Ca. Therefore, rather than conducting end-point studies, longitudinal investigations are necessary to define the sequential pattern of elemental perturbation associated with a given neurotoxic event. Such research can also help identify the role of individual elements in the injury response. Future microprobe studies should be combined with measurements of ion levels (e.g., using fura- or ion selective electrodes) to provide a comprehensive and dynamic view of elemental deregulation. In addition, parallel biochemical studies should be performed to determine mechanisms of elemental disruption and possible biochemical and metabolic consequences of this disruption. Although evidence presented in this commentary suggests that each type of neurotoxic event produces a

369

IN NEUROTOXICITY

characteristic pattern of decompartmentalization, further work is necessary to confirm this possibility. Finally, based on a presumed involvement of elements in nerve injury, efforts are currently underway in several laboratories to develop appropriate pharmacological therapies for certain chemical- and trauma-induced neuropathological conditions (Dretchen et al., 1986; El-Fawal et al., 1989; Beattie et al., 1989). Information generated by EPMA studies can provide a rational basis for choosing the type of pharmacological agent and the dosing regimen used in each condition. ACKNOWLEDGMENTS Some of the research presented in this review was supported by NIH grants to R.M.L. (ES03830) and A.J.S. (NS21455). The authors thank Chuck Fiori and Carolyn Castiglia, and Drs. Ken Reuhl, Herb Lowndes, Bob Lagasse,and Margaret Foster for their help in the preparation of this commentary.

REFERENCES ANDREW& S. B., AND REESE, T. S. (1986). Intracellular structure and elemental analysis in rapid-frozen neurons. Ann. N. Y. Acad.

Sci. 483, 284-294.

ANDREWS, S. B., LEAPMAN, R. D., LANDIS, D. M. D., AND REESE, T. S. (1987). Distribution of calcium and potassium in presynaptic nerve terminals from cerebellar cortex. Proc. Natl, Acad. Sci. USA 84, 17 13- 17 17. AUDESIRK, G. (1985). Effects of lead exposure on the physiology of neurons. Prog. Neurobiol. 24, 199-23 1. BAKER, F. F. ( 1986). The sodium-calcium exchange system. In Calcium and the Cell, pp. 73-86. Ciba Foundation Symposium. Wiley, New York. BANIK, N. L., HOGAN, E. L., AND HSU, C. Y. (1987). The multimolecular cascade of spinal cord injury: Studies on prostanoids, calcium and proteinases. Neurochem. Pathol.

7, 57-77.

BEATTIE, M. S., STOKES, B. T., AND BRESNAHAN, J. C. (1989). Experimental spinal cord injury: Strategies for acute and chronic intervention based on anatomic, physiological and behavioral studies. h Pharmacological Approaches Injury (D.

to the Treatment

of Brain

and Spinal

Cord

C. Stein and B. A. Sabel, Eds.), pp. 43-74. Plenum, New York. BELL, M. E., PETERSON,R. G., AND EICHBERG,J. (1982). Metabolism of phospholipids in peripheral nerve from rats with chronic streptozotocin-induced diabetes: In-

370

LOPACHIN

AND SAUBERMANN

creased turnover of phosphatidylinositol-4,5-bisphosphate. J. Neuuochcm. 39, 192-200. BOND, M., JARAKI, A., DISCH, C. H., AND HEALY, B. P. ( 1989). Subcellular calcium content in cardiomyopathic hamster hearts in vivo: An electron probe study. Circ. Res. 64, 1001-1012. BRAUGHLER,J. M. (1988). Calcium and lipid peroxidation. In Oxygen Radicals and Tissue Injury (B. Halliwell, Ed.), pp. 99-106, Feder. Amer. Sot. Exptl. Biol., Rockville Pike, MD. BRISMAR, T., AND SIMA, A. A. F. (198 1). Changes in nodal function in nerve fibres of the spontaneously diabetic BB-Wistar rat: Potential clamp analysis. Acta Physiol. Stand. 113,499-506. BRISMAR, T., HILDEBRAND, C., AND TEGNER, R. (1987). Nodes of Ranvier in acrylamide neuropathy: Voltage clamp and electron microscopic analysis of rat sciatic nerve fibres at proximal levels. Brain Res. 423, 135143. BULGER, R. E., BEEUWKES, R., AND SAUBERMANN, A. J. (198 1). Application of scanning electron microscopy to X-ray analysis of frozen-hydrated sections. III. Elemental content of cells in the rat renal papillary tip. J. Ce// Biol. g&274-280. CARAFOLI, E. (1987). Intracellular calcium homeostasis. Atmu. Rev. BiOdJeJn. 56, 395-433. CAVANAGH, J. B., AND GYSBERS, M. F. (1983). Ultrastructural features of the Purkinje cell damage caused by acrylamide in the rat: A new phenomenon in cellular neuropathology. J. Neurocytol. 12, 4 13-437. CHAN, S. Y., OCHS, S., AND JERSILD, R. A. (1984). Localization of calcium in nerve fibers. J. Neurobiol. 15, 89-108. CHANDLER, J. A. (1977). X-ray microanalysis in the electron microscope. in Practical Methods in Electron Microscopy, Vol. 5. Glauert, AM. North-Holland, Amsterdam. CHEUNG, J. Y., BONVENTRE, J. V., MALIS, C. D., AND LEAF, A. (1986). Mechanisms of disease. N. Engl. J. Med. 314, 1670-1676. CHOI, D. W. (1987). ionic dependence of glutamate neurotoxicity. J. Neurosci. 7(2), 369-379. CHOO, E. S. (1977). Toxic effects of adriamycin on the ganglia of the peripheral nervous system: A neuropathological study. J. Neuropathol. Exp. Neural. 36, 907915. CHOU, S. M., AND HARTMAN, H. A. (1964). Axonal lesion and waltzing syndrome after IDPN administration in rats. Acta Neuropathol. 3, 428-450. DEITMER, J. W., AND SCHLUE, W. R. (1981). Measurements of the intracellular potassium activity of Retzius cells in leech central nervous system. J. Exp. Biol. 91, 87-101. DELIO, D. A., GOLD, B. C., AND LOWNDES,H. E. (1987). Cross talk between intraspinal elements during progression of IDPN neuropathy. Toxicol. Appl. Pharrnacol. 90,253-260.

DIENEL, G. A. (1984). Regional accumulation of calcium in postischemic rat brain. J. Neurochem. 43, 9 13-925. DRETCHEN,K. L., BOWLES,A. M., AND RAINES,A. (1986). Protection by phenytoin and calcium channel blocking agents against the toxicity of diisopropylfluorophosphate. Toxicol. Appl. Pharmacol. 83, 584-589. DUCE, I. R., AND KEEN, P. (1978). Can neuronal smooth endoplasmic reticulum function as a calcium reservoir? Neuroscience 3, 837-848. DUCKETT, S., GALLE, P., AND SAID, G. (1977a). Microprobe scanning of normal and pathological human peripheral nerve. Acza Neurupathol. 37, 27 1-272. DUCKETT, S., GALLE, P., ESCOUROLLE,R., POIRIER, J., AND HAUW, J. (1977b). Presence of zinc, aluminum, magnesium in striopalledodentate (SPD) calcifications (Fahr’s disease): Electron probe study. Acta Neuropathol. 38, 7-10. DUCKETT, S., GALLE, P., AND FIORI, C. (1989). The application of electron probe microanalysis with wavelength dispersive X-ray spectrometry to the study of neural tissue. In Metal Ions in Neurology and Psychiatry. A. R. Liss, New York. EL-FAWAL, H. A. N., JORTNER, B. S., AND EHRICH, M. (I 989). Effect of verapamil on organophosphorus-induced delayed neuropathy in hens. Tosicol. Appl. Pharmacof. 97, 500-5 11. ELLISMAN, M. H., FRIEDMAN, P. L., AND HAMILTON, W. J. (I 979). Cytochemical localization of cations in myelinated nerve using TEM, HVEM, SEM and electron probe microanalysis. Scanning Electron Microsc. 2,793-800. ESCAIG,J. (1982). New instruments which facilitate rapid freezing at 83K and 6K. J. Microsc. 126, 221-229. FIORI, C. E.. LEAPMAN, R. D., SWYT, C. R., AND ANDREWS, S. B. (1988). Quantitative X-ray mapping of biological cryosections. Ultrarnicroscopy 24, 237-250. FREDERIK, P. M., AND BUSING, W. M. (198 I). Strong evidence against section thawing whilst cutting on the cryo-ultratome. J. Microsc. 122.217-220. GALLE, P., BERRY,J., AND DUCKETT, S. (1980). Electron microprobe ultrastructural localization of aluminum in rat brain. Acta Neuropathol. 49, 245-247. GALVAN, M., DORGE, A., BECK, F., AND RICK, R. (I 984). Intracellular electrolyte concentrations in rat sympathetic neurones measured with an electron microprobe. Pflugers Arch. 400, 274-279. GARRUTO, R. M., FUKATSU, R., YANAGIHARA, R., GAJDUSEK, D. C., HOOK, G., AND FIORI, C. E. (1984). Imaging of calcium and aluminum in neurofibrillary tanglebearing neurons in Parkinsonism-dementia of Guam. Proc. Nat/. Acad. Sci. USA 81, 1875-l 879. GARRUTO, R. M., S’NYT, C.. YANAGIHARA, R., FIORI, C. E., AND GAJDUSEK, D. C. (1986). Intraneuronal colocalization of silicon with calcium and aluminum in amyotrophic lateral sclerosis and Parkinsonism with dementia ofGuam. N. Engl. J. Med. 315, 71 I-712.

ION AND WATER

DISRUPTION

GOLDSTEIN, B. D., AND FINCHER, D. R. (1986). Paradoxical changes in spinal cord reflexes following the acute administration of acrylamide. Toxicol. Lett. 31, 93-99.

GOLDSTEIN, J. I., NE~EIIJRY.D. E., ECHLIN, P., JOY,D. C., FIORI, C., AND LIFSHIN, E. (1981). Scanning Electron Microscopy and X-Ray Microanalysis. Plenum, New York. GORLEN, K. E., BARDEN, L. K., DEL PRIORE.J. S., FIORI, C. E., GIBSON, C. C., AND LEAPMAN, R. D. (1984). Computerized analytical electron microscope for elemental imaging. Rev. Sci. Instrum. 55, 9 12-92 1. GREENE, D. A., AND LATTIMER, S. A. ( 1984). Impaired energy utilization and Na-K-ATPase in diabetic nerve. Amer. J. Physiol. 246, E3 11 -E3 18. GREENE, D. A., LATTIMER, S. A., AND SIMA, A. A. F. (1988). Are disturbances of sorbitol. phosphoinositide, and Na-K-ATPase regulation involved in pathogenesis of diabetic neuropathy? Diabetes 37, 688-693. GRIFFIN, J. W., AND WATSON; D. F. (1988). Axonal transport in neurological disease. Ann. Neural. 23, 313. GUPTA, B. L., AND HALL, T. A. (1978). Electron microprobe X-ray analysis of calcium. In Calcium Transport and Cell Function (A. Scarpa and E. Carafoh. Eds.), pp. 28-5 1. New York Academy of Sciences New York. HAGLER, H. K.. AND BUJA, L. M. (1984). New techniques for the preparation of thin freeze dried cryosections for X-ray microanalysis. In Science ofBiological Specimen Preparation, pp. 16 1-166. SEM. Chicago. HALL. T. A., ANDERSON.H. C., AND APPLETON, T. (1973). The use of thin specimens for X-ray microanalysis of biology. J. Microsc. 99, 177- 182. HALL, T. A. (1989). The history of electron probe microanalysis in biology. In Electron Probe Microanalysis (K. Zierold, and H. K. Hagler, Eds.), pp. 1-15. SpringerVerlag, New York. HILLMAN, D. E., AND LLINAS, R. (1974). Calcium-containing electron-dense structures in the axons of the squid giant synapse. J. Cell Biol. 61, 146- 155. INGRAM, P., SHELBURNE, J. D., AND ROGGLI, V. I. (1989a). Microprobe Analysis in Medicine. Hemisphere Publishing Corp., New York. INGRAM, P., NASSAR, R., LEFURGEY, A., DAVILLA, S., AND SOMMER, J. R. (1989b). Quantitative X-ray elemental mapping of dynamic physiologic events in skeletal muscle. In Electron Probe Microanalysis. (K. Zierold, and H. K. Hagler, Eds.), pp. 251-264. SpringerVerlag, New York. JAKOBSEN, J., SIDENIUS, P., AND BRAENDGAARD, H. (1986). A proposal for a classification of neuropathies according to their axonal transport abnormalities. J. Neuroi.

Neurosurg.

Psych. 49, 986-990.

JANCSO,G., KARCSU, S., KIRALY. E., SZEBENI,A., TOTH. L.. BACSY, E., Joo, F., AND PARDUCZ,A. (1984). Neurotoxin induced nerve cell degeneration: Possible involvement of calcium. Brain Res. 295, 2 11-2 16.

IN NEUROTOXICITY

371

JOHNSON,J. D., MEISENHEIMER, T. L.. AND ISOM, G. E. (1986). Cyanide-induced neurotoxicity: Role of neuronal calcium. Toxicol. Appl. Pharmacol. 84,464-469. JONES,H. B., AND CAVANAGH, J. B. (1984). The evolution of intracellular responses to acrylamide in rat spinal ganglion neurons. Neuropathol. Appl. Neurobiol. 10, 101-121. KARP: R. D., SILCOX, J. C., AND SOMLYO, A. V. (1982). Cryoultramicrotomy: Evidence against melting and the use of a low temperature cement for specimen orientation. J. Microsc. 125, 157-165. KASS, I. S., AND LIPTON, P. (1982). Mechanisms involved in irreversible anoxic damage to the in vitro rat hippocampal slice. J. Physiol. 332, 459-472. KEITH, C., DIPAOLA, M., MAXFIELD, F. R., AND SHELANSKI, M. L. (1983). Microinjection of Ca-calmodulin causes a localized depolymerization of microtubules. J. Cell Biol. 97, 19 1% 1924. KEYNES, R. D., AND RITCHIE, J. M. (1965). The movements of labelled ions in mammalian non-myeiinated nerve tibres. J. Physiol. 179, 333-367. KING, R. H. M., LLEWELYN, J. G., THOMAS, P. K., GILBEY,S. G., AND WATKINS, P. J. (1988). Perineurial calcification. Neuropathol. Appl. Neurobiol. 14, 105- 123. KOMULAJNEN, H.. AND BONDY, S. C. (I 987). Modulation of levels of free calcium within synaptosomes by organochlorine insecticides. J. Pharmacol. Exp. Ther. 241, 575-581.

KORF; J., AND POSTEMA, F. (1984). Regional calcium accumulation and cation shifts in rat brain by kainate. J. Neurochem. 43, 1052-1060. KUTTY, R. K., SINGH, Y., SANTOSTASI,G., AND KRISHNA, G. (1989). Maitotoxin-induced liver cell death involving loss of cell ATP following influx of calcium. Toxicol. Appl. Pharmacol. 101, l-10. LEFURGEY,A.: BOND, M., AND INGRAM, P. (1988). Frontiers in electron probe microanalysis: Application to cell physiology. UltramicroscopJ~ 24, 185-220. LEMKE, M., DEMEDIUK, P., MCINTOSH, T. K., VINK, R., AND FADEN, A. I. (1987). Alterations in tissue Mg, Na and spinal cord edema following impact trauma in rats. Biochem. Biophys. Res. Commun. 147: 1170-l 175. LING, L., AND CANTLEY, L. (1984). The (Na,K)-ATPase of Friend erythroleukemia cells is phosphorylated near the ATP hydrolysis by an endogenous membrane-bound kinase. J. Biol. Chem. 259, 4089-4095. LONG, J. F., NAGODE, L. A., KINDIG, O., AND Lrss, L. (1980). Axonal swellings of Purkinje cells in chickens associated with high intake of 1.25-(OH),D, and elevated dietary aluminum-including X-ray microanalysis. Nezlroto.xicology 1, 11 1- 120, LOPACHIN, R. M., LAPADULA, D. M., AND ABOU-DONIA, M. B. (I 988a). Organophosphate intoxication alters distribution of elements in chicken peripheral nerve. Neurosci.

Abstr.

14, 775.

LOPACHIN, R. M., LOWERY, J.. EICHBERG, J., KIRKPATRICK, J. B., CARTWRIGHT, J., AND SALJBERMANN,A. J.

372

LOPACHIN

AND SAUBERMANN

(1988b). Distribution of elements in rat peripheral axons and nerve cell bodies determined by X-ray microprobe analysis. J. Neurochem 51, 764-775. LOPACHIN, R. M., LOPACHIN, V. R., AND SAUBERMANN. A. J. (1989a). X-Ray microprobe analysis of subcellular elemental distribution in normal and injured peripheral nerve axons. In Cell Calcium Metabolism (G. Fiskum, Ed.), pp. 479-489. Plenum, New York. LOPACHIN, R. M., LOWERY,J., EICHBERG,J., LOPACHIN, V. R., AND SAUBERMANN,A. J. (1989b). Effectsofinjury on distribution of elements in peripheral nerve axons. In Microbeam Analysis (P. E. Russell, Ed.), pp. 26-30. San Francisco Press, San Francisco. LOPACHIN, R. M., LOPACHIN, V. R., AND SAUBERMANN, A. J. (1990). Effects of axotomy on distribution and concentration of elements in rat sciatic nerve. J. Neurochem. 54, 320-332. LUBIN, M. (1967). Intracellular potassium and macromolecular synthesis in mammalian cells. Nature (London) 4,451-453. MACKNIGHT, A. D. C. (1984). Cellular response to injury. In Edema (N. C. Staub, and A. E. Taylor, Eds.). pp. 489-520. Raven Press, New York. MACKNIGHT, A. D. C. (1987). Volume maintenance in isosmotic conditions. In Current Topics in Membranes and Transport, pp. 3-43. Academic Press, New York. MARTINS, E., INAMURA, K., THEMNER, K., MALMQVIST. K. G., AND SIESJO,B. K. (1988). Accumulation of calcium and loss of potassium in the hippocampus following transient cerebral ischemia: A proton microprobe study. J. Cereb. Blood Flow Metab. 8, 531-538. MCBURNEY. R. N., AND NEERING, I. R. (1987). Neuronal calcium homeostasis. Trends Neurosci. 10, 164-169. MCGRAW, C. F., SOMLYO, A. V., AND BLAUSTEIN, M. P. (1980). Localization of calcium in presynaptic nerve terminals. J. Cell Biol. 85, 228-24 1. MCILWAIN, H., AND BACHELARD, H. S. (197 1). Cell-free cerebral systems:Glycolysis and the pentose phosphate pathway. In Biochemistry and the CNS, pp. 98-169. Churchill Livingstone, New York. MIZISIN, A. P., MYERS, R. R., AND POWELL, H. C. (1986a). Endoneurial sodium accumulation in galactosemic rat nerves. Muscle Nerve 9, 440-444. MIZISIN, A. P., POWELL, H. C., AND MYERS, R. R. (198613). Edema and increased endoneurial sodium in galactose neuropathy. J. Neural. Sci. 74, 35-43. MORETON, R. B. (1981). Electron-probe X-ray microanalysis: Techniques and recent applications in biology. Biol. Rev. 51, 409-46 1. MORGAN, A. J. (1979). Non-freezing techniques of preparing biological specimens for electron microprobe Xray microanalysis. Scanning Electron Microsc. 2, 635648. MYERS, R. R., HECKMAN, H. M., AND POWELL, H. C. (1983). Endoneurial fluid is hypertonic. J. Neuropathol. Exp. Neural. 42, 2 17-224. NICHOLLS, D. G. (1986). Intracellular calcium homeostasis. Brit. Med. Bull. 42, 353-358.

OCHS, S., GAZIRI. J. C. J.. AND JERSILD, R. A. (1986). Metabolic and ionic properties of axoplasmic transport in relation to its mechanism. In Axoplasmic Transport (2. Iqbal, Ed.), pp. 57-68. CRC Press, Boca Raton, FL. OCHS, S. (1987). The action of neurotoxins in relation to axoplasmic transport. Nezrrotoxicology 8, 155-166. OLNEY, J. W., PRICE, M. T., SAMSON, L., AND LABRUYERE, J. (1986). The role of specific ions in glutamate neurotoxicity. Neurosci. Lett. 65, 65-7 1. ORRENIUS, S. (1985). Biochemical mechanisms of cytotoxicity. Trends Pharmacol. Sci. 6, 17-20. OSCHMAN, J. L., HALL, T. A., PETERS,P. D., AND WALL, B. J. (1974). Association of calcium with membranes of squid giant axon. J. Cell Biol. 61, 156-165. PARSONS,K., BELLOTTO, D. J., SCHULZ, W. W., BUJA, L. M., AND HAGLER, H. K. (1985). Towards routine cryoultramicrotomy. EMSA Bull. 14, 49-60. PERL, D. P., AND BRODY, A. R. (1980a). Alzheimer’s disease: X-ray spectrometric evidence of aluminum accumulation in neurofibrillary tangle-bearing neurons. Science 208, 297-299. PERL, D. P., AND BRODY, A. R. (1980b). Detection of aluminum by SEM X-ray spectrometry within neurofibrillary tangle-bearing neurons of Alzheimer’s disease. Neurotoxicology 1, 133-137. POUNDS,J. G., AND ROSEN, J. F. (1988). Cellular Ca homeostasis and Ca-mediated cell processes as critical targets for toxicant action: Conceptual and methodological pitfalls. Toxicol. Appl. Pharmacol. 94, 33 l-34 1. RASMINSKY, M. (1980). Ephaptic transmission between single nerve fibres in the spinal nerve roots of dystrophic mice. J. Physiol. 305, 151-169. RICK, R., DORGE, A., AND TIPPE, A. (1976). Elemental distribution of Na, P, Cl and K in different structures of myelinated nerve of Rana esculenta. Experientia 32, 1018-1019. RICK, R., DORGE, A., AND THURAU, K. (1982). Quantitative analysis of electrolytes in frozen dried sections. J. Microsc. 125, 239-247. RITCHIE, J. M. (1967). The oxygen consumption of mammalian non-myelinated nerve fibres at rest and during activity. J. Physiol. 188, 309-329. ROBERTSON, D. G.. AND ANDERSON, R. J. (1986). Electrophysiologic characteristics of tibia1 and sciatic nerves in the hen. Amer. J. Vet Res. 47, 1378-l 38 1. ROOMANS, G. M. (1988). Introduction to X-ray microanalysis in biology. J. Electron Micros. Tech. 9, 3-17. Roos, N., AND BARNARD, T. (1985). A comparison of subcellular element concentrations in frozen-dried, plastic-embedded, dry-cut sections and frozen-dried cryosections. Ultramicrotomy 17, 335-344. ROTHMAN, S. (1984). Synaptic release of excitatory amino acid neurotransmitter mediates anoxic neuronal death. J. Neurosci. 4, 1884- 189 1. SAUBERMANN, A. J., RILEY, W., AND ECHLIN, P. (1977a). Preparation of frozen hydrated tissue sections for X-ray

ION

AND

WATER

DISRUPTION

microanalysis using a satellite vacuum coating and eansfer system.Scanning Electron Microsc. 2, 347-356. SAUBERMANN, A. J., RILEY, W., AND BEEUWKES, R. (1977b). Cutting work in thick section cryomicrotomy. J. Microsc.

111, 39-49.

SAUBERMANN, A. J. (1980). Application of cryosectioning to X-ray microanalysis of biological tissue. Scanning Electron Microsc. 2,42 I-430. SAUBERMANN, A. J.. ECHLIN, P., PETERS, P. D., AND BEEUWKES,R. (198 la). Application of scanning electron microscopy to X-ray analysis of frozen-hydrated sections. I. Specimen handling techniques. J. Cell Biol. 88, 257-267. SAUBERMANN, A. J., BEEUWKES, R., AND PETERS, P. D. ( 198 1b). Application of scanning electron microscopy to X-ray analysis offrozen-hydrated sections. II. Analysis of standard solutions and artificial electrolyte gradients, J. Cell Biol. 88, 268-273. SAUBERMANN,A. J., AND SCHEID,V. L. (1985). Elemental composition and water content of neuron and glial cells in the central nervous system of the north american medicinal leech (Macrobdella decora). .I. Nezwochem. 44,825-834.

SAUBERMANN, A. J., DOBYAN, D. C., SCHEID,V. L., AND BULGER, R. E. (1986a). Rat renal papilla: Comparison of two techniques for X-ray analysis. Kidney Int. 29, 675-681.

SAUBERMANN, A. J., SCHEID,V. L., DOBYAN, D. C., AND BULGER. R. E. (I 986b). Simultaneous comparison of techniques for X-ray analysis of proximal tubule cells. Kidne), Int. 29, 682-688. SAUBERMANN, A. J., AND HEYMAN, R. V. (1987). Quantitative digital X-ray imaging using frozen hydrated and frozen dried tissue sections. J. Microsc. 146, 169- 182. SAUBERMANN, A. J. (1988a). X-Ray mapping of frozen hydrated and frozen dried cryosections using electron microprobe analysis. Scanning 10, 239-244. SAIJBERMANN, A. J. (1988b). Quantitative electron microprobe analysis of cryosections. Scanning Microsc. 2, 2207-22

18.

SAUBERMANN,A. J., AND STOCKTON, J. D. (1988). Effects af increased extracellular K on the elemental composition and water content of neuron and glial cells in leech CNS. J. Nezrroehem. 51, 1797- l80T. SAUBERMANN, A. J. (1989). X-Ray microanalysis of cryosections using image analysis. In Electron Probe Microamlysis (K. Zierold, and H. K. Hagler, Eds.), pp. 73-85. Springer-Verlag, New York. SCHLAEPFER,W. W. (1974). Calcium-induced degeneration of axoplasm in isolated segments of rat peripheral nerve. Brain Res. 69, 203-2 15. SCHLOTE. W., WOLBURG, H., AND WENDT-GALLITELLI. M. F. (1981). Ionic shifts in myelinated nerve fibers during early stages of Wallerian degeneration. Acta Neurpathol. 7(Suppl.), 3 I-35. XHLUE. W. R., AND DEITMER, 1. W. (1980). EXtmcellUlar potassium in neuropile and nerve cell body region of

IN

NEUROTOXICITY

373

the leech central nervous system. J. Exp. Biol. 87, 2343.

SCHRAMA, L. H., BERTI-MATTERA, L. N., AND EICHBERG, J. (1987). Altered protein phosphorylation in sciatic nerve from rats with streptozocin-induced diabetes. l>iabetes 36, 1254-1260. SHUMAN, H., SOMLYO, A., AND SOMLYO, A. P. (1976). Quantitative electron probe microanalysis of biological thin sections: Methods and validity. Ultramicroscopy 1, 3 17-339.

SIESJO,B. K. (198 1). Ceil damage in the brain: A speculative synthesis. J. Cereb. Blood Flow Metab. 4, 155185. SIKLOS, L., CSILIK, B., AND KNYIHAR-CSILIK, E. (1983). Calcium binding of presynaptic protrusions as revealed by X-ray spectrum averaging in the rat neuromuscular junction. Histochemistry 79, 77-86. SIMMONS, D. A., WINEGRAD, A. I., AND MARTIN, D. B. (1982). Significance of tissue myo-inositol concentrations in metabolic regulation in nerve. Science 217, 848851. SIMMONS, D. A., KERN. E. F. O., WINEGRAD, A. J., AND MARTIN, D. B. (1986). Basal phosphatidylinositol turnover controls aortic Na/K-ATPase activity. J. C/in. Invest. 77, 503-5 13. SOMLYO, A. V., GONZALEZ-SERRATOS,H., SHUMAN, H., MCCLELLAN, G., AND SOMLYO, A. P. (198 1). Calcium release and ionic changes in the sarcoplasmic reticulum of tetanized muscle: An electron-probe study. J. Cell Biol. 90,577-594. SOMLYO, A. P., BOND, M., AND SOIWLYO,A. V. (1985a). Calcium content of mitochondria and endoplasmic reticulum in liver frozen rapidly in vivo. Nature (London) 314,622-625.

SOMLYO, A. P., URBANICS, R.. VADASZ, G., KOVACH, A. G. B., AND SOMLYO, A. V. (1985b). Mitochondrial calcium and cellular electrolytes in brain cortex frozen in situ: Electron probe analysis. Biochern. Biophys. Rex Commun.

132, 1071-1078.

SOMLYO, A. P. (1985). Cell calcium measurement with electron probe and electron energy loss analysis. CeN Calcium 6, 197-2 12. STAHL, W. L. (I 986). The Na,K-ATPase of nervous tissue. Newochem. Int. 8,449-4X. STARKE, P. E., HOEK, J. B., AND FARBER, J. L. (1986). Calcium-dependent and calcium-independent mechanisms of irreversible cell injury in cultured hepatocytes. J. Biol. Chem 261, 3006-3012. SHERMAN, A. B. (1984). Hexacarbon neuropathy: Cell body changes are early, dynamic and specific. Ann. ;I’Pzrrul. 16, 343-348. STERMAN, A. B., AND DELANNOY. M. R. (1985). CelI body responses to axonal injury: Traumatic axotomy versus toxic neuropathy. Exp. Net&. 89,408-419. STOECKEL,M. E., HINDELANG-GERTNER,C., DELLMANN, H. D., PROTE, A., AND STUTINSKY, F. (1975). Subcei-

374 lular localization

LOPACHIN of calcium

AND

in the mouse hypophysis.

Cell Tissue Res. 157, 301-322. TRUMP, B. F., BEREZESKY, I. K., CHANG, S. H.. PENDERGRASS, R. E., AND MERGNER, W. J. (1979). The role of ion shifts in cell injury. Scanning Electron Microsc. 3, l-14. TRUMP, B. F., BEREZESKY. I. K., SMITH, M. W., PHELPS, P. C., AND ELLIGET, K. A. (1989). The relationship between cellular ion deregulation and acute and chronic toxicity. Toxicol. Appl. Pharmacol. 91, 6-22. VAN IREN, F., VAN ESSEN-JOOLEN, L., VANDER DUYN SCHOUTNE, P., BOERS-VANDER SLUIJS, P., AND DE BRUIJN, W. C. (1979). Sodium and calcium localization in cells and tissues by precipitation with antimonate: A quantitative study. Histochemistry 63, 273-294. VINK, R., YUM, S. W., LEMKE, M., DEMEDIUK, P., AND FADEN, A. 1. (1989). Traumatic spinal cord injury in rabbits decreases intracellular free magnesium concentration as measured by 31P NMR. Bruin Res. 490, 144-147. WALZ, W., AND HERTZ, L. (1983). Functional interactions between neurons and astrocytes. II. Potassium homeostasis at the cellular level. Prog. Neurobiol. 20, 133- 183. WALZ, W. (1988). The role of potassium in cytotoxic brain edema. In The Biochemical Pathology ofAstrocytes pp. 3 15-326. A. R. Liss, New York.

SAUBERMANN WICK, S. M., AND HELPER, P. K. (1982). Selective localization of intracellular Ca with potassium antimonate. .J.Histochem. Cytochem. 30, 1190-1204. WROBLEWSKI, R. (1989). In situ elemental analysis and visualization in cryofixed nervous tissues. J. Microsc. 155, 81-l 12. YAMAMOTO, H., FUKUNAGA, K., TANAKA, E., AND MIYAMOTO, E. (I 983). Ca- and calmodulin-dependent phosphorylation of microtubule-associated protein 2 and T factor, and inhibition of microtubule assembly. J Neurochem. 41, 1119-l 125. YASE, Y. (1980). The role of aluminum in CNS degeneration with interaction of calcium. Neurotoxicology 1, 101-109. YOUNG, W., AND KOREH, I. (1986). Potassium and calcium changes in injured spinal cords. Bruin Res. 365, 42-53. ZIEROLD, K. (1982a). Preparation and transfer of ultrathin frozen-hydrated and freeze-dried cryosections for microanalysis in scanning transmission electron microscopy. Scanning Electron Microsc. 3, 12051214. ZIEROLD, K. (1982b). Cryopreparation of mammalian tissue for X-ray microanalysis in STEM. J. Microsc. 125, 149-156.