Morphometry of tubular bodies in endothelial cells in normal stable isolated perfused and edematous dog lungs.код для вставкиСкачать
THE ANATOMICAL HECORD 196295 -300 (1980) Morphometry of Tubular Bodies in Endothelial Cells in Normal, Stable Isolated Perfused, and Edematous Dog Lungs PETER B. BERENDSEN AND DAVID 0. DEFOUW Department of Anatomy, College of Medicine and Dentistry of New Jersey, Newark, New Jersey 07103 ABSTRACT The volume densities and dimensions of tubular bodies (WeibelPalade bodies) in dog pulmonary capillary endothelial cells were examined by means of stratified random sampling and electron microscopic morphometry. The 0.2%. mean volume density of tubular bodies in normal (control) lungs was 0.43 I The mean volume densities of tubular bodies in stable, isolated lungs perfused for ?hhour, 1hour, and 2 hours and in lungs made edematous by increasing hydrostatic pressure or by decreasing oncotic pressure did not vary significantly from that of normal lungs. The mean thickness of the endothelium measured a t the middle of the tubular bodies of normal dog lungs was nearly twice the mean thickness of the overall capillary endothelial cell sample. The mean endothelial thickness across tubular bodies from stable and from edematous isolated perfused lungs did not differ significantly from that of the control group. The mean width of tubular bodies from normal dog lungs was 0.25 I0.06pm and the mean length was 0.81 t 0.61 pm. The mean widths and lengths of tubular bodies from stable and from edematous isolated perfused dog lung endothelial cells did not differ statistically from those of normal dog lungs. Thirty percent of the tubular bodies in the sample were found to be adjacent to a mitochondrion in the same plane of section. Tubular bodies contained both tightly packed and loosely grouped tubules. It is concluded that the tubular bodies in canine pulmonary endothelial cells remain stable during the perfusion of isolated lungs and in oncotic and hydrostatic edema of isolated perfused lungs. Membrane bounded groups of tubules were first reported to be present in endothelial cells by Weibel('64), Weibel and Palade ('641, and by Stehbens ('65). Although Rhodin ('68) has proposed that these structures be termed "specific endothelial granules," they have often been called "Weibel-Palade bodies." Kawamura et al. ('74)have termed them "tubular bodies," the term by which they will be identified in this paper. They have been identified within the endothelial cell cytoplasm of normal blood vessels, ranging in size from capillaries to the aorta in many species (Weibel, '64; Stehbens, '65; Fuchs and Weibel, '66; Cauna and Hinderer, '69; Sengel and Stoebner, '70; Steinsiepe and Weibel, '70; Kojimahara, '77). Tubular bodies have been noted to be retained in cultured umbilical endothelial cells (Takeshige and Fujimoto, '77; Haudenschild et al., '75). Similar structures have been reported in endothelia of humans with various diseases (Ki- 000-3276X/80/1963-0295$01.40 0 1980 ALAN R. LISS, INC. mura et al., '75; Garancis et al., '71; Nieland et al., '72; Haustein, '73; Jerusalem et al., '74; Macadam et al., '75; De Martino et al., '69; Jao et al., '77; Feiner and Gallo, '77; Daniels et al., '74). They are also reported to be present in the endothelia of intracranial blood vessels (Herrlinger et al., '74; Hirano et al., '73; Hirano et al., '74; Kawamura et al., '74; Hirano and Matsui, '75; Gessega and Anzil, '75; Hassoun et al., '78). Morphometric study of rat aortas demonstrated reduction in the number of tubular bodies following incubation in epinephrine. This finding lead Burri and Weibel('68) to suggest that the tubular bodies are related to the production of a procoagulative substance. The isolated, perfused, dog lung is frequently used as a subject €or experimental studies of endothelial transport (e.g., Per1 et al., '75, '76). Received May 3, 1 9 7 9 accepted August 29, 1979. 295 296 PETER B. BERENDSEN AND DAVID 0. DEFOUW The reaction of the tubular body component of pulmonary endothelial cells may affect the results of these experimental procedures. Therefore, this study reports the volume densities of tubular bodies in the isolated perfused lungs as determined by established stereological procedures. MATERIALS AND METHODS The materials for this study were obtained from the preparations previously reported by DeFouw and Berendsen ('78a,b; '79). In brief, four groups of preparations were used. The first group consisted of normal lungs obtained from three adult mongrel dogs (approximately 20 kg body weight) of either sex, killed with a n overdose of pentobarbital (2 g). The trachea and lungs were exposed, removed from the thorax, and immediately fixed by the endotracheal instillation of cold 2% glutaraldehyde in 0.15 M NaHCO,, buffer, pH 7.4, at a pressure of 20 cm H,O, followed by immersion in fixative. The second group consisted of stable isolated perfused lungs prepared from six mongrel dogs. Following stable perfusion for %, 1, or 2 hours, two lung preparations a t each time period were fixed as above with glutaraldehyde. In the third group, edema was produced by perfusion of the isolated lungs with fluid containing decreasing concentrations of macromolecules. In the fourth group, edema was produced by sustained elevation of the outflow pressure. In these last two groups, the lungs were fixed at the end of two hours. For full details, see DeFouw and Berendsen ('78a,b, '79). Stratified random sampling (Weibel, '73) was used as follows. Each pair offixed lungs was cut into twelve longitudinal slices and two tissue blocks were selected from each slice by the use of a numbered overlay and random number table. Each of the 24 blocks was divided in half, with one half prepared for light microscopy and the other half used for electron microscopy. The specimens used for electron microscopy were post fixed in 1% osmium tetroxide in 0.15 M NaHCO, buffer, pH 7.3, for one hour, were then dehydrated in solutions of ethanol of graded concentrationsand then in propylene oxide and embedded in Epon (Luft, '61). Each of the 24 original tissue specimens per lung gave 10-15 Epon blocks and thus provided approximately 240-360 blocks per test lung. From the primary sample of 24 original tissue specimens per lung, 10 were selected randomly and one Epon block from each of these ten was prepared for electron microscopy. This produced ten grids per dog lung. Thin sections (60-9Onm) were cut on LKB or Sorvall MT-2 ultramicrotomes, stained with uranyl acetate and lead citrate (Reynolds, '631, and examined with a Philips EM-300. Ten electron micrographs of each block were taken at x 21,600 by photographing the upper left corners of sequential grid squares. This provided a stratified random sample of 1,500 electron micrographs. Each micrograph was analyzed with a test grid having 168 equidistant points and t h e point counts were recorded for estimating the volume densities of the endothelial cytoplasmic organelles. This stratified procedure produced a sample size sufficient to yield a n expected error of less than 5% (Weibel, '73) in the estimation of stereologic parameters. The significance of differences between means of the morphometric parameters for the normal (control) and stable and edematous isolated perfused lungs were evaluated by a twotailed Student's t-test (Freund, '67). RESULTS The volume densities, i.e., proportion of endothelial cell cytoplasm occupied by microtubules and tubular (Weibel-Palade) bodies in capillary endothelial cells of normal, stable isolated perfused, and edematous isolated perfused dog lungs are given in Table 1.The mean volume densities of tubular bodies and microtubules within normal lungs did not differ significantly from those observed in stable isolated-perfused or in the hydrostatic and oncotic, edematous isolated perfused lungs in this study. The thickness of t h e endothelial cells through the middle of seventy tubular bodies was measured from the luminal to abluminal surfaces (Table 2). TABLE 1. Volume densities of tubular bodies and microtubules' Lung type Normal Stable isolated perfused ?hhour 1 hour 2 hours mean of ?hhour, 1 hour, 2 hours Hydrostatic edema Oncotic edema 'Expressed as %) of cytoplasm Mean & 0.43 2 1 SEM 0.29% 0.63 .t 0.28% 0.46t 0.5% 0.25 i 0.05% 0.45 f 0.19% 0.56 k 0.09% 0.46 .t 0.21% 297 TUBULAR BODIES IN NORMAL AND PERFUSED DOG LUNGS TABLE 2. Measurements of endothelium and tubular bodies’ Normal Thickness of endothelium at tubular bodies (pm) Range (pmi Width of tubular bodies (pm) Range (pm) Length of tubular bodies (pm) Range (pm) Perfus& 0.52 ? 0.16 0.93 0.33 0.25 t 0.06 0.37 0.15 0.81 ? 0.61 2.8 0.33 ~ ~ ~ 0.63 f 0.23 1.16 - 0.33 0.29 f 0.14 0.70 - 0.14 0.80 ? 0.60 2.56 - 0.23 Hydrostatic edema 0.47 ? 0.16 0.70 - 0.42 2 7 ? .04 .33 - .23 .79 t .13 .88 - .65 Oncotic edema 0.56 0.93 ? 0.19 - 0.33 0.29 f 0.08 0.042 - 0.19 1.09 5 0.60 2.19 - 0.47 ‘Mean values t 1 SEM 2Mean of %, 1,2 hr No statistically significant differences were identified among the mean endothelial thicknesses through tubular bodies in lungs of the groups studied. In many micrographs, the sectioned air-blood barrier had a n interstitial compartment which was thicker on one side of the capillary than on the other due to greater amount of interstital matrix, fibers, and cells. The opposite, “thin” side of the air-blood barrier consisted of opposed type 1 epithelial and endothelial cells with a basal lamina between. Tubular bodies in the endothelium adjacent to the “thick’portion did not vary in number or form compared to those within the endothelium of the thin portion. Clusters or groups of tubular bodies were not common. The lengths and widths of the tubular bodies in the normal and experimental groups are given in Table 2. Widths were measured from luminal to abluminal sides of each tubular body. Approximately two-thirds of the tubular bodies had lengths that were more than twice the width (Fig. 2). The remainder were irregularly rounded (Fig. 3). Approximately 30% of the tubular bodies had a mitochondrion in close association (Figs. 2,3,6) in the plane of section. Ten percent of the tubular bodies were near endothelial cell nuclei (Fig. 4). Each tubular body was bounded by an irregular membrane which was often indistinct. This enclosed tubules of 14-18 nm diameter which were usually seen in cross section. Each tubule had a core that was denser than the surrounding matrix. The tubules within some tubular bodies were irregularly spaced (Fig. 5). Most tubular bodies contained tightly packed hexagonally arranged tubular units (Fig. 11, with a center-to-center spacing of approximately 25 run. Microtubules were infrequent in the cytoplasm but were readily identifiable in longitudinal orientation (Fig. 3). Cross sectioned cyto- plasmic microtubules were distinguished from the circular profiles of pinocytotic vesicles by the smaller diameter of the former. DISCUSSION The mean thickness of pulmonary capillary endothelium through the midpoint of the tubular bodies observed in this sample ofpulmonary alveolar capillaries was 0.52 +- 0.16 pm. This is slightly less than twice the mean overall thickness of the endothelium, which we have reported to be 0.30 & 0.02 pm in normal dog lungs and 0.35 0.02 pm during the two hour stable perfusion period (DeFouw and Berendsen, ’78a). Although the two hour perfusion period has been reported to produce a statistically significant increase in overall endothelial thickness (DeFouw and Berendsen, ’78a), the mean increase in thickness through the tubular bodies was not statistically significant. This indicates that the alveolar capillary endothelial cells at the tubular bodies are thicker than elsewhere, and these sites tend to remain stable during perfusion and during experimental edema of isolated lungs. The volume densities of tubular bodies in alveolar capillary endothelial cells reported in this study are similar to those reported in rat capillaries (Fuchs and Weibel, ’66)and slightly higher than those in frog capillaries (Steinsiepe and Weibel, ’70). A much higher volume density of tubular bodies (Burri and Weibel, ’68) has been reported in normal rat aortic endothelial cells. Burri and Weibel (’68) reported a relative decrease in volume density of these structures during incubation of aorta in Ringer’s solution, possibly owing t o endothelial swelling. In this study, the failure of either perfusion for two hours or the production of edema to have statistically significant effects on the volume density of tubular bodies suggests that the tubular bodies are relatively sta- * 298 PETER B. BERENDSEN AND DAVID 0. DEFOUW Fig. 1. Tubular body (arrow) in the endothelium of an isolated dog lung perfused for ?4 hour. X 173,000. Fig. 2. Elongated tubular body (arrow) adjacent to a mitochondrion (m). Oncotic edema. x 173,000. Fig. 3. Rounded tubular body (arrow)with a mitochondrion (m) and a microtubule (mt) adjacent. Stable perfusion for % hour. x 173,000. Fig. 4. Rounded tubular body (arrow) near the nucleus of an endothelial cell. Stable perfusion for ?4 hour. x 173,0130, Fig. 5. Tubular body (arrow) with loose grouping of tubules. Stable perfusion for ?4 hour. x 173,000. Fig. 6 . Tubular body (arrow) with tight packing of tubules. Oncotic edema. x 173,000. TUBULAR BODIES IN NORMAL AND PERFUSED DOG LUNGS ble during the perfusion process under the experimental conditions used here. The mean width of tubular bodies in normal dog pulmonary capillary endothelial cells in this study is in agreement with the width of tubular bodies reported in human skin capillaries (Zelickson, '661, frog vessels (Steinsiepe and Weibel, '70), and in rabbit and human eye capillaries (Matsuda and Sugiura, '70). The mean widths reported here are smaller than those observed in dog aortic endothelial cells (Sun and Ghidoni, '69). Most investigators give lengths of tubular bodies ranging from maxima of 0.7 pm to 3.2 pm (Weibel and Palade, '64; Zelickson, '66; Steinsiepe and Weibel, '70; Matsuda and Sugiura, '70; Macadam et al., '75) in blood vessels of various species. These lengths are comparable to the maxima in the present report. Although the tubular bodies are assumed to be rod-like, a flattened plaque or other shape is possible. Determination of shape must await serial reconstruction. Mitochondria occupy only 2 5 3 . 5 % of the endothelial cytoplasmic volume in capillaries of normal and isolated perfused dog lungs (DeFouw and Berendsen, '78a). Thirty percent of the tubular bodies in this sample were adjacent to mitochondria in the same plane of section. With the 0.4-0.5% volume density of tubular bodies, the probability that mitochondria and tubular bodies would be contiguous on a random basis should be very small. There may be a functional relationship between the two. The 70% of tubular bodies that did not have mitochondria in the plane of section may have had contiguous mitochondria above or below the plane of section. The diameters of tubules within the tubular bodies in this study were similar to those reported in endothelia of various organs by others (Weibel and Palade, '64; Zelickson, '66; Herrlinger et al., '74; Macadam et al., '75). Although some authors have given center-to-center dimensions between tubules (Weibel and Palade, '64), the tubules within the tubular bodies observed in this study varied in compactness. The more loosely grouped tubules were similar to those termed type 1by Sun and Ghidoni ('691, whereas the more closely compacted clusters of tubules were called type 2 by those authors. The former appearance was similar to that of tubular structures seen in human tumor cells reported by Uzman et al. ('711, and endothelia of patients with systemic lupus erythematosus (Garancis et al., '71; Nieland et al., '72; Naustein et al., '73; Jerusalem et al., '74; Macadam et al., '75). Loosely grouped tubules within tubular bodies have been noted within the endoplasmic reticulum of tumor cells (Uzman et al., '741, and 299 an origin of tubular bodies from the Golgi apparatus has been suggested (Sengel and Stoebner, '70; Matsuda and Sugiura, '70; Kojimahara, '77). Tubular bodies have been reported not to react when cytochemicallytested for the presence of acid phosphatase (Lemeunier, et al., '69) and are therefore probably not related to lysosomes. Burri and Weibel('68) have suggested that these structures may be related to the production of a procoagulative substance. This study demonstrates that tubular bodies have stable volume densities and dimensions in stable and edematous isolated perfused lung preparations and suggests that isolated perfused lung preparations may be useful in future studies of their function. ACKNOWLEDGMENTS The authors are grateful to Doctor F.P. Chinard and the late Dr. W. Per1 for their collaboration and to Doctors P. Chowdhury and A. Ritter for their assistance. This work was supported in part by Public Health Service grants HL-19571 and HL12879. LITERATURE CITED Burri, P.H., and E.R. Weibel (1968) Beinflussung einer spezifischen cytoplasmischen Organelle von Endothelzellen durch Adrenelin. Z. Zellforsch., 88:42&440. Cauna, N., and K.H. Hinderer (1969)Fine structure of blood vessels of the human nasal respiratory mucosa. Ann. Otol. Rhinol. Laryngol., 78t866879. Daniels, T.E., R.A. Sylvester, S. Silverman, V. Polando, and N. Tala1 (1974) Tubuloreticular structures within labial salivary glands in Sjogrens syndrome. Arthritis Rheum., 17:593-597. DeFouw, D.O., and P.B. Berendsen (197th)Isolated perfused dog lungs: A comparative stereological comparison with normal dog lungs. Am. J. Anat., 151.337-344, DeFouw, D.O., and P.B. Berendsen (1978b3 Morphological changes in isolated perfused dog lungs after acute hydrostatic edema. Circ. Res., 43:7Z82. DeFouw, D.O., and P.B. Berendsen (1979) A morphometric analysis of isolated perfused dog lungs after acute oncotic edema. Microvasc. Res., 17:9&103. DeMartino, C., L. Accini, G. Andres, and I. Archetti (1969) Tubular structures associated with the endothelial endoplasmic reticulum in glomerular capillaries of Rhesus monkey and nephritic man. Z. Zellforsch., 97:50&511. Feiner, H., and G. Gallo (1977) Ultrastructure in glomerulonephritis associated with cryoglobulinemia. Am. J. Pathol., 88:146162. Freund, J . E . (1967) Modern Elementary Statistics. Prentice-Hall, Inc. Englewood Cliffs, N.J. Fuchs, A., and E.R. Weibel (1966) Morphometrische Untersuchung der Verteilung einer specifischen cytoplasmischen Organelle in Endotbelzellen der Ratte. Z. Zellforsch., 73;l-9. Garancis, J.C., R.A. Komorowski, G.C. Bernhard, and J.V. Straumfjord (1971) Significance of cytoplasmic microtubules in lupus nephritis. Am. J. Pathol., 64:1-12. Gessega, E.C., and A.P. Anzil(1975)Rod shaped filamentous inclusions and other ultrastructural features in a cerebellar astrocytoma. Acta. Neuropathol., 33; 119-127. 300 PETER B. BERENDSEN AND DAVID 0. DEFOUW Hassoun, J., A. Hirano, and H.M. Zimmerman (1978) Fine structure of intercellular junctions and blood vessels in medulloblastoma. Acta. Neuropathol., 33t67-78. Haudenschild, C.C., R.S. Cotran, M.A Gimbrone, and J. Folkman (1975)Fine structure of vascular endothelium i n culture. J. Ultrastruct. Res., 50:2>32. Haustein, U.F. (1973)Tubulare Strukturen in kranken und scheinbar gesunder Nieren bei Chronischen und Viszeralen Lupus erythematodes. Dermatologica, 147:225-240. Herrlinger, H., A.P. A n d , K. Blinzinger, and D. Kromski (1974) Endothelial microtubular bodies in human brain capillaries and venules. J. Anat., 118.206209. Hirano, A,, N.R. Ghatek, and H.M. Zimmerman (1973)Fenestrated blood vessels in craniopharyngioma. Acta. Neuropathol., 26:17 1- 177. Hirano, A,, N.R. Ghatek, N.H. Becker, and H.M. Zimmerman (1974) A comparison of the fine structure of small blood vessels i n intracranial and retroperitoneal malignant lymphomas. Acta. Neuropathol., 27:95-104. Hirano, A,, and T. Matsui (1975) Vascular structures in brain tumors. Hum. Pathol., 6:611-621. Jao, W., M.A. Swerdlow, and J. Lash (1977) Endothelial microtubules in renal biopsies of patients with idiopathic membranous glomerulonephropathy. Lab. Invest., 36: 341. Jerusalem, F., M. Rakusa, A.G. Engel, and R.D. MacDonald (1974)Morphometric analysis of skeletal muscle capillary ultrastmcture in inflammatory myopathies. J. Neurol. Sci., 23t391-402. Kawamura, J., Y. Kamijyo, T. Sunaga, and E. Nelson (1974) Tubular bodies in vascular endothelium of a cerebellar neoplasm. Lab. Invest., 30:35%365. Kimura, A,, K. Tosaka, and T. Nakao (1975)Measels Rash. Arch. Viral., 47:29&307. Kojimahara, M. (1977)Filaments and rod-shaped tubulated bodies in the endothelia of anterior cerebral arteries in young rats. Cell Tiss. Res., 182:506511. Lemeunier, A,, P.H. Burri, and E.R. Weibel(1969) Absence of acid phosphatase activity in specific endothelial organelles. Histochemie, 20: 143-149. Luft, J . H . (1961) Improvements i n epoxy embedding methods. J. Biophys. Biochem. Cytol., 9:40%414. Macadam, R.E., J.M. Vetters, and N.K. Saida (1975) A search for microtubular inclusions in endothelial cells in a variety of skin diseases. Brit. J. Dermatol., 92.175-182. Matsuda, H., and S. Sugiura (1970)Ultrastructure of “tubu- lar body” in the endothelial cells of the ocular blood vessels. Invest. Ophthalmol., 9:91%925. Nieland, N.W., K. Hashimoto, and A.T. Masi (1972) Microtubular inclusions in normal skin of systemic Lupus erythematosis patients. Arthritis Rheum., 15: 195-200. Perl, W., P. Chowhury, and F.P. Chinard (1975) Reflection coefficients of dog lung endothelium to small hydrophilic solutes. Am. J. Physiol., 288:797-809. Perl, W., G. Silverman, A.C. Delea, and F.P. Chinard (1976) Permeability of dog lung endothelium to sodium, diols, amides, and water. Am. J. Physiol., 230t170~1721. Reynolds, E.S. (1963)The use of lead citrate at high pH as a n electron-opaquestain i n electron microscopy.J. Cell Biol., 17t208-212. Rhodin, J.A.G. (1968)Ulstrastructure of mammalian venous capillaries, venules, and small collecting veins. J. Ultrastruct. Res., 25:45>500. Sengel, A., and P. Stoebner (1970) Golgi origin of tubular inclusions in endothelial cells. J. Cell Biol., 44,223-226. Stehbens, W.E. (1965) Ultrastructure of vascular endothelium in the frog. Q. J. Exp. Physiol., 50:37&384. Steinsiepe, K.F., and E.R. Weibel (1970) Electron mikroskopische Untersuchungen an specizifischen Organelles von Endothelzellen des Frosche (Rana temporaria). Z. Zellforsch., 108:106126. Sun, C.N., and J.J. Ghidoni (1969) Membrane-bound mirotubular and crystalline structures in endothelial cells of normal canine aorta. Experientia, 25:301-302. Takeshige, Y., and S. Fujimoto (1977)Ultrastructural modifications of endothelial cells by changes of the vascular function. Bibl. Anat., 16:364369. Uzman, B.G., H. Saito, and M. Kasac (1971) Tubular arrays i n the endoplasmic reticulum in human tumor cells. Lab. Invest., 24:49>498. Weibel, E.R. (1964) Neue cytoplasmische Komponenten von Endothelzellen. Acta. Anat., 59:39&391. Weibel, E.R., and G.E. Palade (1964) New cytoplasmic components in arterial endothelia. J. Cell Biol., 23: 101-112. Weibel, E.R. (1973) Stereological techniques for electron microscopic morphometry. In: Principles and Techniques of Electron Microscopy. M.A. Hyatt, ed. Van Nostrand Reinholdt Co., New York, Vol. 3. pp. 237-296. Zelickson, A S . 1966) A tubular structure in the endothelial cells and pericytes of human capillaries. J. Invest. Dermatol., 46.167-171.