JOURNAL OF ELECTRON MICROSCOPY TECHNIQUE 18:38-49 (1991)

Application of Confocal Scanning Laser Microscopy in Experimental Pathology GARY J. SMITH, C. ROBERT BAGNELL, WILLIAM E. BAKEWELL, KURT A. BLACK, THOMAS W. BOULDIN, TODD S. EARNHARDT, GARY E.R. HOOK, AND KATHERINE B. PRYZWANSKY Department ofPathology, (G.J.S., C.R.B., K.A.B., T.W.B., T.S.E., K.B.P.) and Curriculum in Toxicology, University of North Carolina, Chapel Hill, North Carolina (W.E.B.), and National Institute Of Environmental Health Sciences, Research Triangle Park, North Carolina 27709 (G.E.R.H.)

KEY WORDS

Apoptosis, Surfactant apoprotein-A, Blood-nerve barrier, Pinocytosis

ABSTRACT Confocal scanning laser microscopy (CSLM) represents a n exciting new tool for scientific disciplines which focus on mechanistic studies such as experimental pathology. Enhanced resolution in the specimen plane and rejection of out-of-focus fluorescence flare allow analysis of specific nucleic acid sequences, enzymes, structural macromolecules, and cellular homeostasis utilizing fluorescent probes. Four different experimental applications are discussed which utilize CSLM to evaluate pathological processes a t the subcellular, cellular, and tissue levels. Programmed cell death, or apoptosis, is a natural process of significance both during development and as a response to toxic stimuli. CSLM-imaging of nuclei of human B lymphoblastoid cells following exposure to a monofunctional alkylating agent suggests that the degradation of chromatin characteristic of apoptosis may occur in asymmetric patterns. Surfactant apoprotein-A is the major non-serum protein component of pulmonary surfactant and is essential for the extracellular function of surfactant. CSLM of alveolar type I1 cells suggests that apoprotein-A is present in both the cytoplasm, predominantly in lamellar bodies, and in the nucleus. The tumor promoter, phorbol myristate acetate, rapidly stimulated the formation of vacuoles in human neutrophils. CSLM using Lucifer Yellow a s a probe suggests that cylindrical vacuoles are formed by fluid-phase pinocytosis. The blood-nerve barrier (BNB) in peripheral nerves may be a n important target during toxininduced neuropathies. Ricin-induced permeability of the BNB in the rat was rapidly visualized by CSLM as leakage of fluorescein isothiocynate (F1TC)-dextran into the endoneurial compartment. INTRODUCTION Experimental pathology is a scientific discipline which focuses on the mechanisms of induction and progression of disease, or abnormal biology, as models for understanding normal biology. Confocal scanning laser microscopy (CSLM) represents a n exciting new tool with rapidly expanding potential for application to this field. CSLM promises to enhance dramatically the application of the molecular probes developed by recombinant DNA technology and molecular immunology, and the metabolic probes developed for flow microfluorimetry, to the study of cellular biology. Labelling of substrates, cellular proteins, or nucleic acids with specific f luorescently-labelled tracer molecules allows multiparameter evaluation of morphological localization and of macromolecular relationships in fixed cells or tissues, a s well as the evaluation of vital function and molecular assembly in viable cells (Taylor and Wang, 1989; Wang and Taylor, 1989). The CSLM in the Department of Pathology at The University of North Carolina has been utilized for a wide range of experimental studies of the subcellular localization of antigens, nuclear structure and DNA content in normal and carcinogen-exposed cells, cytoskeletal organization in cells transfected with oncogenes, mechanisms of pinocytosis, breakdown of the blood-nerve barrier (BNB), fungal accumulation a t clathrin-coated pits, karyotypic

0 1991 WILEY-LISS, INC

analysis, localization and translocation of receptors, and the discovery of a previously unknown structure in the ear of the bat. CSLM offers many advantages over conventional epifluorescence microscopy, including: 1) greater spatial resolution in the x-y or specimen plane, particularly in the non-coherent fluorescent format; 2) optical sectioning of a specimen in the axial plane by rejection of out-of-focus information from above and below the optical plane, and of extraneous fluorescence due to glare and scatter outside of the optical plane; and 3) superior contrast and resolution in the digital images (Brackenhoff et al., 1989; Rober-Nicoud et al., 1988; Sheppard, 1987; Shotten, 1989). Interfacing of the CSLM with a n image processing and display system for filtering, signal ratio imaging, and three-dimensional (3-D) display of serial optical sections produces a powerful system for morphological analysis and macromolecular localization. For a review of the optical and performance characteristics of confocal microscopes and modes of digital image processing and display, see

Received February 26, 1990; accepted in revised form J u n e 20. 1990 Address reprint requests to Dr. Gary J . Smith, Department of Pathology, CB#7525, University of North Carolina, Chapel Hill, NC 27599-7525.

CONFOCAL MICROSCOPY IN EXPERIMENTAL PATHOLOGY

The Handbook of Biological Confocal Microscopy (Pawley, 1989). However, the majority of investigators with applications for CSLM, which are concerned only with determining the subcellular localization of a n antigen amid the haze of fluorescence, or with producing a 3-D representation of a structure specifically labelled with a fluorescent probe, do not require such esoteric digital image processing. The basic image analysis and display systems integrated with the commercially available CSLMs provide the ease of operation, rapid image formation, digital image collection for morphometric analysis, and multiple display formats to meet the demands of the majority of potential users while requiring little training on the instrumentation (Pawley, 1989; Shotten, 1989). Unprocessed images of confocal optical sections generally provide adequate information concerning subcellular localization of macromolecules, and the rapid presentation of video loops or stereo pairs of sequential serial sections affords powerful visual aids for evaluation of 3-D relationships. In the current work, applications of confocal microscopy to experimental pathology are presented in four experimental systems, ranging from the subcellular to the tissue level of organization; digital images as collected by the CSLM were utilized without postprocessing to illustrate the variety of potential applications for this technology. INSTRUMENTATION A Zeiss laser scan microscope (LSM) equipped with argon ion and helium-neon lasers was utilized for all studies. The Zeiss LSM is fitted with 1 0 (NA ~ 0.31, 20 x (NA 0.51, 40 x (NA 0.75), 63 x (NA 1.40, oil), and 1 0 0 ~(NA 1.30, oil) objectives. Cells or structures of interest were located utilizing the helium-neon laser (633 nm) to prevent photobleaching, and digital images of fluorescence in response to excitation with the argon Laser (488 nm) were collected at a scan rate of 2 seconds per image. Digital images were archived on a hard disk and were transferred subsequently to 35 mm film using a Polaroid Freeze-Frame. CHARACTERIZATION OF CHROMATIN ALTERATIONS OCCURRING DURING CARCINOGEN-INDUCED CELL DEATH Cell death is a n important process not only in disease and toxic injury but also in normal embryonic development, involution of endocrine-dependent tissues, and the immune response (Martz and Howell, 1989; Shier, 1988; Walker et al., 1988; Wyllie et al., 1980). At least two pathways of cell death, necrosis and apoptosis, are thought to exist (Walker et al., 1988; Wyllie et al., 1980). Necrosis, which occurs after death of the organism or is induced by toxic insult, is characterized typically by mitochondria1 and cellular swelling. During the early phase of necrosis, alteration in nuclear chromatin is relatively minor and reversible; however, during later stages, after lysis of cellular organelles and membranes has occurred, karyolysis is often apparent. In constrast, apoptosis, or programmed cell death, is involved in normal tissue development and homeostasis, and also is induced by hormone withdrawal from

39

sensitive cells, cell damage caused by cytotoxic T lymphocytes, and DNA damage caused by radiation. Apoptosis is characterized by the condensation of nuclear chromatin into large discrete masses that abut the nuclear membrane prior to the loss of cell membrane integrity. Recently we have been studying mechanisms of cell death induced by carcinogenic chemicals in a human lymphoblastoid strain of B cell origin, T5-1 (Black et al., 1989a,b). Apoptosis is a frequent mode of cell death in lymphoid cells (Walker e t al., 1988; Wyllie et al., 19801, and our previous studies suggest that monofunctional alkylating carcinogens induce this process in T51 cells (Black e t al., 1989a,b). DNA histograms obtained during cell cycle analysis of populations of T5-1 cells exposed to N-methyl-N’-nitro-N-nitrosoguanidine (MNNG) reveal a subpopulation of cells possessing the markedly reduced DNA content (Black e t al., 1989a) characteristic of cells undergoing apoptosis (Afanas’ev et al., 1986; Compton e t al., 1988). Consequently, we have utilized CSLM to evaluate the changes in nuclear and chromatin structure in T5-1 cells undergoing programmed cell death after exposure to MNNG. Asynchronously proliferating cultures of T5-1 cells were treated with 30 ng MNNGiml (Balck et al., 1989a). At various times after treatment, aliquots of cultures were fixed in 70% ethanol a s described by Black et al. (1989b), and cells were resuspended to a concentration of 1.0 x lo6 cellsiml in a buffer consisting of 0.1 M NaC1,ll.O mM glucose, 5.6 mM Na,HPO,, 5.4 mM KCl, 0.4 mM Ca(NO&, and 0.4 mM MgSO, at pH 7.4 (DeLuca et al., 1983). RNAse A was added to 50 pgiml, and the cells were incubated at 37°C for 30 minutes. Propidium iodide (PI; 40 pgiml) was added to stain DNA, and the samples were incubated a t 37°C for a n additional 30 minutes. (RNAse-treatment of cells ensures that observed fluorescence is attributable specifically to DNA [Arndt-Jovin and Jovin, 198911. Samples were centrifuged at 200g, and the cells resuspended to a concentration of 1 x lo7 cellsiml in the above buffer containing 40 pg PIlml. An aliquot (10 p1) of the cell suspension was added to a n equal volume (10 pl) of polyvinyl alcohol mounting medium on a glass microscope slide and covered gently with a glass coverslip, and the PI-stained nuclei were imaged using the 63 x plan-apochromat objective lens (NA 1.4). Figure 1 presents a possible sequence for the apoptotic decay of the nuclear chromatin. Confocal optical sections were taken through the center of nuclei of T5-1 cells exposed to either the vehicle (0.05% dimethyl sulfoxide (DMSO), Fig. 1A) or 30 ng MNNGiml (Fig. 1BH). MNNG-treatment caused a profound disruption in the structure of nuclear chromatin compared with the vehicle-treated cell (Fig. 1A) where chromatin fluorescence was distributed throughout the nucleus with focal areas of increased intensity, especially in areas immediately adjacent to the nuclear membrane. In nuclei of MNNG-treated cells the extent of chromatin destruction ranges from subtle, focal reductions in fluorescence and perturbation of chromatin fine structure (Fig. 1B,C), which are difficult to observe by conventional fluorescence microscopy, to a nucleus essentially devoid of chromatin (Fig. 1H). The nuclei appear to

40

G.J. SMITH ET AL

Fig. 1. Confocal images through the center of PI-stained nuclei of MNNG-treated T5-1 human lymphoblastoid cells. A Vehicle, 0.05% DMSO-treated cell. B H : MNNG-treated cells. Bar = 10 pm.

swell rapidly a s the initial signs of chromatin degradation become apparent. Chromatin becomes localized near the margin of the nucleus as DNA degradation progresses, forming small, brightly stained clumps which fuse to form the larger clumps of condensed chromatin typical of nuclei undergoing apoptosis (Fig. 1DG). Stereoscopic image pairs constructed from serial optical sections of nuclei from a DMSO-treated T5-1 cell and from several MNNG-treated cells are shown in Figure 2. The chromatin of a control nucleus (Fig. 2A) is relatively evenly distributed throughout the nucleus, the nuclear margin is intact and regularly shaped, and the interior of the nucleus is obscured by staining of superior sections. The nuclei of MNNGtreated cells range from containing focal areas devoid of chromatin, but maintaining a n intact membrane, to nuclei that are essentially empty of chromatin except for multiple clumps of condensed chromatin a t the margin of the nucleus. The asymmetry of chromatin degradation is better appreciated by viewing a single nucleus as four different stereoscopic image pairs, reconstructed using the same set of optical sections but from different perspectives (Fig. 3). Figures 3A and 3C were constructed using all of the optical sections, but in opposite order. The remaining chromatin is not distributed evenly throughout the nucleus and appears to be more highly concentrated in the proximal pole in Fig. 3C. Figures 3B and 3D are views from the center of the nucleus toward the opposite poles. This presentation appears more informative because there is less obscuration of information by chromatin near the nuclear membrane, and it clearly shows that loss of chromatin did not occur equally in all parts of the nucleus. Future studies will address whether the various states of chromatin condensation represent a sequence

in a single pathway or multiple avenues of cell death, whether the asymmetric distribution of chromatin is a random or patterned event, and whether chromatin clumping occurs along or is influenced by components of the nucleoskeleton.

CYTOPLASMIC AND NUCLEAR DISTRIBUTION OF SURFACTANT APOPROTEIN-A IN ALVEOLAR TYPE I1 CELLS Pulmonary surfactant is a heterogeneous complex of lipids and proteins that serves to stabilize the distal airways at low lung volumes. Although lipids account for approximately 90% of the surfactant complex, the remaining lo%, consisting of proteins, is of critical importance to the extracellular functioning of the complex in the alveoli of the lungs (King, 1985). Both lipid and protein constituents of surfactant are made and secreted by alveolar type I1 cells (Askin and Kuhn, 1971; Harwood et al., 1975). The major protein of pulmonary surfactant that is specific for the complex is surfactant apoprotein-A (SP-A) (Whitsett e t al., 1985a, 1985b). Many functions have been ascribed to this protein including assisting with the spreading of the surfactant complex at the airiextracellular fluid interface, recycling of surfactant, the construction of tubular myelin figures, and opsonin activity (Hawgood et al., 1987; Hook et al., 1986; LaForce et al., 1973). In truth, however, the functions of SP-A are obscure although there is no doubt that the protein is absolutely essential for the proper functioning of the surfactant complex. Controversy exists regarding the distribution of SPA within the type I1 cell. Determination of the subcellular distribution of SP-A could provide information

Fig. 2. Stereoscopic image pairs of PI-stained nuclei reconstructed from confocal optical sections of a DMSO-treated or MNNG-treated T5-1 cells. Images were reconstructed from 20 consecutive confocal images taken at 0.6 pm increments in the z axis. A: DMSO-treated cell. l3-E: MNNG-treated cells. Bar = 10 wm.

42

G.J. SMITH ET AL.

Fig. 3. Multiple stereoscopic image pairs of a single nucleus from an MNNG-treated T5-1 cell reconstructed from optical sections taken at 0.6 (rm intervals. A,C: Stereoscopic image pairs of the complete nucleus viewed from opposite poles constructed from 20 optical sec-

tions. B,D: Stereo image pairs constructed from 10 sequential optical sections, starting from the center of the cell and viewing toward the proximal pole in A and C, respectively. Bar = 10 (rm.

regarding the manner in which this protein is secreted. Exclusive localization in the lamellar bodies (subcellular storage sties of surfactant) would suggest that the release of SP-A by type I1 cells could be via a regulated pathway; whereas, a diffuse distribution might be con-

sistent with a constitutive secretory process. Since SPA contributes to the surfactant complex, it seems reasonable to expect its presence in the lamellar bodies of the cell, and some reports have proposed such a n association (Williams and Benson, 1981).However, it is not

CONFOCAL MICROSCOPY IN EXPERIMENTAL PATHOLOGY

43

Fig. 4. Conventional immunofluorescence image of type I1 cells isolated from the lungs of rats revealing the subcellular distribution of SP-A. x 750.

clear whether the lamellar bodies are the exclusive location of SP-A or other sites might also be significant. Our initial studies of SP-A localization within type I1 cells by conventional immunofluorescence microscopy did not provide adequate details regarding the distribution of SP-A because of interference from SP-A distributed in regions of the cell above and below the plane of interest. For this reason we turned to the CSLM. Type I1 cells were isolated from the lungs of normal rats (Miller and Hook, 19881, and processed for immunofluorescence using the procedure of Liley et al. (1987). Cells were fixed in 2% formaldehyde containing 0.2% glutaraldehyde buffered with 1% sodium cacodylate and washed with 5% bovine serum albumin containing 0.3%Triton X-100 (vol./vol.) in phosphate-buffered saline (BSAiTritonlPBS). The fixed cells were incubated overnight at 4°C in a 1:1,000 dilution of rabbit anti-rat SP-A or preimmune serum and washed twice with BSAITritoniPBS. A fluorescein isothiocyanate (F1TC)-labelled goat anti-rabbit conjugate at a 1:60 dilution was used to visualize the binding of the primary antibody. After two additional washes with BSAiTritonlPBS the specimens were mounted on glass slides in a n antifade reagent which consisted of diazobicyclooctane in 4% N-propylgallate with Tris-buffered glycerol. Immunohistochemical localization of SP-A in isolated type I1 cells by conventional fluorescence microscopy revealed punctate deposits of the protein in the cytoplasm (Fig. 4). Questions regarding the distribution of SP-A within the endoplasmic reticulum, or its association with the nuclear membrane and matrix could not be answered. Compared with conventional fluorescence microscopy, CSLM allows us to image more clearly the distribution of SP-A throughout the alveolar type I1 cell. As shown in Figure 5, SP-A is located primarily in discrete cytoplasmic deposits which probably correspond to the lamellar bodies, stor~~~

Fig. 5. SP-A immunofluorescence of a n isolated type I1 cell obtained with the confocal microscope. Fluorescence was not detected when the cells were treated with preimmune serum. Bar = 5 km.

age sites of pulmonary surfactant within the type I1 cell. However, some diffuse cytoplasmic fluorescence is also present, probably within the endoplasmic reticulum of the cell. Fluorescence was not detected in isolated type I1 cells treated with preimmune serum and examined by CSLM (data not shown). Unexpectedly, diffuse fluorescence was observed within the nucleus of the type I1 cell. The shallow depth of field of the confocal format, and its ability to reject fluorescence outside of the focal plane, revealed that SP-A appears to be associated with the nuclear membrane of the cell, and areas of the nuclear matrix appeared to contain focal areas of fluorescence. Consequently, the nuclear localization of SP-A was evaluated further by examining nuclei isolated from type I1 cells. As shown in Figure 6, SP-A appears to be located on the nuclear membrane with lesser amounts within the nucleus itself. Diffuse fluorescence is present in the nuclear matrix, but is relatively reduced in the nucleolus. Focal patches of fluorescence are also present in the nuclear matrix. Figure 7 shows the distribution of SP-A within four individual confocal optical sections of a single nucleus isolated from a type I1 cell. Figures 7A and 7B show that SP-A is located in patches on the nuclear membrane. These patches appear to be arranged around central regions, relatively deficient in SP-A, which could be nuclear pores. Optical sections through central portions of the nucleus (Fig. 7C,D) again reveal relatively intense patchy fluorescence of the nuclear membrane with diffuse fluorescence within the nuclear matrix. Punctate areas of fluorescence are also seen within the nucleus. However, at this time we do not know why SP-A might be present in the nucleus of the type I1 cell.

44

G.J. SMITH ET AL.

Fig. 6 . Confocal image of SP-A immunofluorescence of nuclei isolated from type I1 cells. Bar = 10 pm.

LUCIFER YELLOW LABELLING OF ENDOCYTIC VESICLES IN PHORBOL ESTER-STIMULATED NEUTROPHILS The tumor promoter, phorbol myristate acetate (PMA), stimulates rapid development of intracellular vacuoles in neutrophils. Electron dense tracer studies suggest that the vacuole membrane originates from the plasma membrane of the neutrophil (White and Estensen, 1974). To determine if these vacuoles are formed by fluid-phase pinocytosis, we used Lucifer Yellow as a fluorescent tracer. Lucifer Yellow is a n excellent probe for studying fluid-phase pinocytosis because it is not toxic, is not degraded, and its uptake is inhibited a t 4°C (Swanson e t al., 1985). This fluorescent probe has been used to investigate fluid-phase pinocytosis in several other cellular systems (Malouf and Wilson, 1986; Miller et al., 1983; Riezman, 1985; Swanson et al., 1987). Human neutrophil monolayers were isolated from peripheral blood and cultured onto glass coverslips in Gey's balanced salts containing 10% human type AB serum as previously described (Pryzwansky et al., 1979). Monolayers were incubated with Lucifer Yellow ( 2 mgiml) in the presence or absence of PMA (20 ngiml) from 5 to 30 minutes a t 37°C. Cells were washed with Gey's balanced salts, fixed for 10 minutes at 4°C with 1% paraformaldehyde in PBS, washed in PBS, mounted in glycero1:PBS (9:1), and imaged by CSLM using the 63 x plan-apochromat objective lens (NA 1.4). Neutrophils have a low rate of constitutive pinocytosis, since Lucifer Yellow did not accumulate in neutrophils incubated for 30 minutes without PMA. However, in the presence of PMA, a rapid uptake of Lucifer Yellow was observed. Within 10 minutes of stimulation with PMA, Lucifer Yellow was observed in neutrophils within small pinocytotic vesicles, principally localized near the cell margin (not shown). These vesicles became larger and more prominent after 30 minutes. Lucifer Yellow was always found within discrete compartments, and did not diffuse into the cytoplasm. Uptake

of Lucifer Yellow was inhibited a t 4"C, indicating that this endocytic process is temperature dependent. CSLM was utilized to resolve the Lucifer Yellow containing endocytic compartments of PMA-treated neutrophils. A series of 19 optical sections were taken at 0.2 pm increments through a neutrophil which had been incubated with PMA for 30 minutes in the presence of Lucifer Yellow. Representative sections are shown in Figure 8. Lucifer Yellow staining was observed within the entire cell thickness. However, more vesicles containing Lucifer Yellow were observed in sections proximal to the plane of cell adherence. These endocytic compartments which stained for Lucifer Yellow resemble vacuoles. An average of approximately 15 vacuoles per cell were observed in neutrophils following incubation with PMA for 30 minutes. Vacuoles varied in size and depth and became noticeably larger with increased time of exposure to PMA. Three-dimensional reconstruction of the serial optical sections demonstrated that the vacuoles appear cylindrical (Fig. 91, although 3-D reconstructions of serial sections are elongated along the z axis due to the geometry of the voxels (Fine et al., 1988; Shotten, 1989). To illustrate the distribution of vacuoles within the body of the cell, a composite fluorescence image built by projecting all of the confocal optical sections onto a single plane was superimposed on the differential interference contrast (DIC) image. The overlay shown in Figure 10 demonstrates that after 30 minutes incubation with PMA, vacuoles were predominantly clustered near the nuclear cleft, or cytocenter. Phagosomes have also been observed to translocate with time from peripheral zones to the cytocenter (Hoffstein, 1980). In summary, PMA stimulates fluid-phase pinocytosis in human neutrophils. Our studies suggest that extracellular fluid is first internalized in vesicles derived by invagination of the plasma membrane. Vacuoles are then formed by fusion of pinosomes with other pinosomes or with other intracellular vesicles such as endosomes. With time, the vacuoles translocate from peripheral zones to the cytocenter. This translocation of vacuoles is similar to the movement of phagosomes to the cytocenter after phagocytosis (Hoffstein, 1980). Unlike thioglycollate-stimulated mouse macrophages, neutrophil lysosomes (azurophil granules) do not fuse with vacuoles (White and Estensen, 1974; K.B. Pryzwansky, personal observation), and a network of tubular lysosomes is not observed. However, studies suggest that specific granules, which are stimulated to exocytose in PMA-treated neutrophils, may fuse with these intracellular vacuoles (White and Estensen, 1974). Stereomicroscopy of CSLM sections demonstrates that vacuoles in neutrophils are cylindrical and frequently span 2-3 pm. We plan to investigate living cells using CSLM to determine if these cylindrical vacuoles interconnect since fixation has been shown to disrupt tubular lysosomes (Swanson et al., 1985). CSLM analysis provided information regarding size, shape, number, and distribution of Lucifer Yellow containing endocytic compartments, and promises to be a useful tool for identifying structural and functional components of endocytic pathways in other cells. Thus, important information regarding the pathways and fate

CONFOCAL MICROSCOPY I N EXPERIMENTAL PATHOLOGY

45

Fig. 7. Optical sections through a nucleus isolated from a type I1 cell. A Patches of fluorescence on the bottom of the nuclear membrane. B Perinuclear fluorescence 0.2 +m above that shown on A. C: Optical section through the middle of the nucleus. D: Optical section 0.2 pm above section shown in (C). Bar = 5 wm.

of the endocytic process, and the mechanism of intracellular movement of various organelles, including vesicular transport, can be obtained on large numbers of cells without the time consuming procedures required for electron microscopy and reconstruction of electron micrographs of serial thin-sections of only a few cells.

ASSESSMENT OF BNB ABNORMALITIES The BNB of the peripheral nervous system serves to restrict the passage of water-soluble substances from the blood into the nervous system. This structural barrier closely regulates the composition of the interstitial fluid surrounding the individual nerve fibers. The integrity of the BNB is compromised in many neuropathies and following nerve trauma (Olsson, 1984), suggesting that BNB breakdown may play a role in the pathogenesis of neuropathy. Alterations of the BNB

lead to nerve edema, which in t u r n causes increased endoneurial pressure and decreased endoneurial blood flow (Low et al., 1985; Myers et al., 1982; Tuck et al., 1984). Breakdown of the BNB also facilitates passage of potentially neurotoxic substances from the blood into the nerve. Several macromolecular tracers have been employed to assess morphologically BNB abnormalities. Among the most popular are horseradish peroxidase (HRP), fluorescently-tagged or radiolabelled albumin, and fluoresceinated dextrans (FITC-dextrans) (Olsson, 1984). It has been our experience that the FITC-dextrans offer several advantages over other macromolecular tracers. FITC-dextrans are more sensitive indicators of BNB alterations than HRP and are less expensive and easier to handle than [12gI]-albumin(Bouldin et d . , 1989, 1990). Furthermore, the FITC-dextran technique per-

Fig. 8. Fluid-phase pinocytosis of Lucifer Yellow by neutrophils stimulated with 20 ngiml of PMA for 30 minutes at 37°C. A: DIG image. A series of 19 optical sections were taken at 0.2 pm increments beginning at the top of the cell. Shown are B: optical sections at 0.4 pm; C: 1.4 Fm; D: 2.4 pm; E: 3.0 pm; and F: 3.8 pm. Numerous

vacuoles containing Lucifer Yellow are resolved a t various depths within the cell. Note that some vacuoles are quite large and span almost the entire cell thickness (small arrows, B-El, while other vacuoles are found only within an area of 0.5 pm or less (large arrow, E). Bar = 10 pm.

CONFOCAL MICROSCOPY IN EXPERIMENTAL PATHOLOGY

47

Fig. 9. Stereoscopic image pair reconstructed from 19 optical sections of Lucifer Yellow fluorescence of the neutrophils shown in Figure 8. Vacuoles vary in size and depth and appear as small cylinders. Bar = 10 pm.

Fig. 10. A composite image produced by projecting the 19 optical sections of Lucifer Yellow immunofluorescence (black) onto the single plane of the DIC image of the neutrophils containing the fluorescence. This overlay demonstrates that most of the vacuoles which contain Lucifer Yellow (Fig. 9) are in the vicinity of the nuclear hoff, or cleft (arrows). Bar = 10 pm.

mits a much more rapid assessment of BNB permeabil- these BNB alterations, we intravenously injected a 4,000 molecular weight (MW) FITC-dextran into anesity than either the HRP or [1251]-albumintechniques. We have recently studied the temporal course of thetized rats at various times after ricin exposure. The BNB breakdown in ricin-induced neuropathy in the rat tracer was allowed to circulate for 5 minutes before 1 using FITC-dextrans and CSLM (Bouldin et al., 1990). cm long segments of peripheral nerve were removed. In this particular model, breakdown of the BNB can be The nerves were flash frozen in embedding medium (to demonstrated with FITC-dextrans within 48 hours of inhibit translocation of tracer) and cryosectioned at 10 exposure to ricin, a neurotoxic lectin. To demonstrate pm. Sections of nerve were then assessed by CSLM for

G.J. SMITH ET AL.

48

Fig. 11. Confocal image of a longitudinal frozen section of sciatic nerve from rat 12 hours after ricin exposure. x 70.

Fig. 12. Confocal image of a longitudinal frozen section of sciatic nerve from rat 48 hours after ricin exposure. x 150.

the presence of FITC-dextran within the endoneurial compartment of the nerve as a n indicator of BNB breakdown. Figure 11is a confocal image of a longitudinal frozen section of sciatic nerve a t 12 hours after ricin exposure. FITC-dextran was injected intravenously 5 minutes prior to flash freezing the nerve. The presence of brightly fluorescent extravascular FITC-dextran in the epineurial tissue around the nerve, but not within the endoneurial compartment, indicates a n intact BNB. Figure 12 shows a longitudinal frozen section of sciatic nerve at 48 hours after ricin exposure. Brightly fluorescent FITC-dextran is now diffusely present in the endoneurial compartment, indicating breakdown of the BNB. The FITC-dextran tracer and CSLM are a n excellent combination for assessing the sequential changes in the permeability of the BNB over the course of ricin neuropathy. Future studies will more fully utilize the capabilities of CSLM. One such proposal involves injecting tracers that have been tagged with different f luorochromes. This should permit recognition of a graded breakdown of the BNB to different MW tracers. Additionally, other fluorescent probes such as Lucifer Yellow and the voltage sensitive dye RH414 have been shown to be useful for study by CSLM of the structure, interaction, and electrical activity of living neurons (Carlsson and Aslund, 1987; Fine et al., 1988).

Askin, F.B., and Kuhn, C. (1971) The cellular origin of pulmonary surfactant. Lab. Invest., 25260-268. Black, K.A., McFarland, R.D., Grisham, J.W.,andSmith,G.J. (1989a) Cell cycle perturbation and cell death after exposure of a human lymphoblastoid cell strain to N-methyl-N’-nitro-N-nitrosoguanidine. Am. J . Pathol., 13453-61. Black, K.A., McFarland, R.D., Grisham, J.W., and Smith, G.J. (1989b) S phase block and cell death in human lymphoblasts exposed to benzo(a)pyrene diol epoxide or N-acetoxy-2-acetylaminofluorene. Toxicol. Appl. Pharmacol., 97:463-472. Bouldin, T.W., Earnhardt, T.S., and Goines, N.D. (1990) Sequential changes in the permeability of the blood-nerve harrier over the course of ricin neuronopathy in the rat. Neurotoxicology, In Press. Bouldin, T.W., Earnhardt, T.S., Goines, N.D., and Goodrum, J . (1989) Temporal relationship of blood-nerve barrier breakdown to the metabolic and morphologic alterations of tellurium neuropathy. Neurotoxicology, 10:79-90. Brakenhoff, G.J., van der Voort, H.T.M., van Spronsen, E.A., and Nanninga, N. (1989)Three-dimensional imaging in fluorescence by confoca1,scanning microscopy. J . Microsc. 153:151-159. Carlsson, K., and Aslund, N. (1987) Confocal imaging for 3-D digital microscopy. Appl. Optics, 26:3232-3238. Compton, M.M., Haskill, J.S., and Cidlowski, J.A. (1988) Analysis of glucocorticoid actions on rat thymocyte deoxyribonucleic acid by fluorescence-activated flow cytometry. Endocrinology, 122:21582164. DeLuca, J.G., Weinstein, L., and Thilly, W.G. (1983) Ultraviolet lightinduced mutation of diploid human lymphoblasts. Mutat. Res., 107: 347-370. Fine, A., Amos, W.B., Durbin, R.M., and McNaughton, P.A. (1988) Confocal microscopy: Applications in neurobiology. Trends Neurosci., 11:346-351. Harwood, J.L., Desai, R., Hext, P., Tetley, T., and Richards, R. (1975) Characterization of pulmonary surfactant from ox, rabbit, rat and sheep. Biochem. J., 151:707-714. Hawgood, S., Benson, B.J., Schilling, J., Damm, D., Clements, J.A., and White, R.T. (1987) Nucleotide and amino acid sequences of pulmonary surfactant protein SP 18 and evidence for cooperation between SP 18 and SP 28-36 in lipid adsorption. Proc. Natl. Acad. Sci. USA, 84:66-70. Hoffstein, S.A. (1980) Intra- and extracellular secretion from polymorphonuclear leukocytes. In: Cell Biology of Inflammation. G. Weissmann ed. Amsterdam, Elsevier North-Holland, pp. 387-435. Hook, G.E.R., Gilmore, L.B., and Talky, F.A. (1986) Dissolution and reassembly of tubular myelin-like multilamellated structures from the lungs of patients with pulmonary alveolar proteinosis. Lab. Invest., 55194-208. King, R.J. (1985) Composition and metabolism of the aaoliuouroteins ofpulmonary surfactant. Ann. Rev. Physiol., 47:775178k

ACKNOWLEDGMENTS These studies were supported by NIH-CA-24144, NIH-(24-42765, NIH-ES-07126, NIH-ES-01104, and ACS-CH-401. REFERENCES Afanas’ev, ’J.N., Korol’, B.A., Mantsygin, Y.A., Nelipovich, P.A., Pechatnikov, V.A., and Umansky, S.R. (1986) Flow cytometry and biochemical analysis of DNA degradation characteristic of two types of‘ cell death. FERS Lett., 194:347-350. Arndt-Joviti. D.J., and Jovin, T.M. (1989) Fluorescence labeling and microscopy of D N A . Methods C,ell Biol., 30:417-448.

I

CONFOCAL MICROSCOPY IN EXPERIMENTAL PATHOLOGY LaForce, F.M., Kelley, W.J., and Huber, G.L. (1973) Inactivation of Staphylococci by alveolar macrophages with preliminary observations on the importance of alveolar lining material. Am. Rev. Respir, Dis., 108:784-790. Liley, H.G., Hawgood, S., Wellenstein, G.A., Benson, B., White, T.R., and Ballard, P.L. (1987) Surfactant protein of molecular weight 28,000-36,000 in cultured human fetal lung: Cellular localization and effect of dexamethasone. Mol. Endocrinol., 1:205-215. Low, P.A., Nukada, H., Schmelzer, J.D., Tuck, R.R., and Dyck, P.J. (1985) Endoneurial oxygen tension and radial topography in nerve edema. Brain Res., 341:147-154. Malouf, N.N., and Wilson, P.E. (1986) Proliferation of the surfaceconnected intracytoplasmic membranous network in skeletal muscle disease. Am. J. Pathol., 125:358-368. Martz, E., and Howell, D.M. (1989) CTL: Virus control cells first and cytolytic cells second? DNA fragmentation, apoptosis and the prelytic halt hypothesis. Immunol. Today, 10:79-86. Miller, B.E., and Hook, G.E.R. (1988) Isolation and characterization of hypertrophic type I1 cells from the lungs of silica treated rats. Lab. Invest., 58:565-575. Miller, D.K., Grifiths, E., Lenard, J., and Firestone, R.A. (1983) Cell killing by lysosomotropic detergents. J . Cell Biol., 97:18411851. Myers, R.R., Mizisin, A.P., Powell, H., and Lampert, P.W. (1982) Reduced nerve blood flow in hexachlorophene neuropathy. Relationship to elevated endoneurial fluid pressure. J . Neuropathol. Exp. Neurol., 41:391-399. Olsson, Y. (1984) Vascular permeability in the peripheral nervous system. In: Peripheral Neuropathy, 2nd Ed., Vol. 1. P.J. Dyck, P.K. Thomas, E.H. Lambert, and R. Bunge, eds. Saunders, Philadelphia, pp. 579-597. Pawley, J. (1989) The Handbook of Biological Confocal Microscopy. IMR Press, University of Wisconsin, Madison, Wisconsin, 194 PP. Pryzwansky, K.B., MacRae, E.K., Spitznagel, J.K., and Cooney, M.H. (1979) Early degranulation of human neutrophils: Immunocytochemical studies of surface and intracellular phagocytic events. Cell, 18:1025-1033. Riezman, H. (1985) Endocytosis in yeast: Several of the yeast secretory mutants are defective in endocytosis. Cell, 4O:lOOl-1009. Robert-Nicoud, M., Arndt-Jovin, D.J., Schormann, T., and Jovin, T.M.

49

(1988) 3-D imaging of cells and tissues using CSLM and digital processing. Eur. J. Cell Biol., 48: (Suppl. 25) 49-52. Sheppard, C.J.R. (1987) Scanning optical microscopy. Adv. Optic. Elect. Microsc. 1O:l-98. Shier, W.T. (1988) Why study mechanisms of cell death? Methods Achiev. Exp. Pathol., 13:l-17. Shotten, D.M. (1989) Confocal scanning optical microscopy and its applications for biological specimens. J . Cell Sci., 94:175-206. Swanson, J.. Burke. E.. and Silverstein. S.C. (1987) Tubular Ivsosomes accompany stimulated pinocytosis in macrophages. J. eel1 Biol.. 104:1217-1222. Swanson, J.A., Yirinec, B.D., and Silverstein, S.C. (1985) Phorbol esters and horseradish peroxidase stimulate pinocytosis and redirect the flow of pinocytosed fluid in macrophages. J. Cell Biol., 100:851-859. Taylor, D.L., and Wang, Y.-L. (1989) Fluorescence microscopy of living cells in culture. In: Methods In Cell Biology, Vol. 30. Academic Press, San Diego, 478 pp. Tuck, R.R., Schmelzer, J.D., and Low, P.A. (1984) Endoneurial blood flow and oxygen tension in the sciatic nerves of rats with experimental diabetic neuropathy. Brain, 107:935-950. Walker, N.I., Harmon, B.V., Gobe, G.C., and Kerr, J.F.R. (1988) Patterns of cell death. Methods Achiev. Exp. Pathol., 13:18-54. Wang, Y-L., and Taylor, D.L. (1989) Fluorescence microscopy of living cells in culture. In: Methods In Cell Biology, Vol. 29. Academic Press, San Diego, 313 pp. White, J.G., and Estensen, R.D. (1974) Selective labilization of specific granules in polymorphonuclear leukocytes by phorbol my&tate acetate. Am. J. Pathol., 7545-60. Whitsett, J.A., Ross, G.M., Weaver, T., Rice, W., Dion, C., and Hull, W. (1985a) Glycosylation and secretion of surfactant-associated glycoprotein A. J. Biol. Chem., 260:15273-15279. Whitsett, J.A., Weaver, T., Hull, W., Ross, G., and Dion, C. (198513) Synthesis of surfactant-associated glycoprotein A by rat type I1 epithelial cells. Primary translation products and post-translational modification. Biochem. Biophys. Acta, 828:162-171. Williams, M.C., and Benson, B.S. (1981) Immunocytochemical localization and identification of the major surfactant protein in adult rat lungs. J . Histochem. Cytochem., 29291-305. Wyllie, A.H., Kerr, J.F.R., and Currie, A.R. (1980) Cell death: The significance of apoptosis. Int. Rev. Cytol., 68:251-306.