Morphometric characterisation of the fine structure of human type II pneumocytes.код для вставкиСкачать
THE ANATOMICAL RECORD 243~49-62 (1995) Morphometric Characterisation of the Fine Structure of Human Type II Pneumocytes HEINZ FEHRENBACH, ANDREAS SCHMIEDL THORSTEN WAHLERS, STEFAN w. HIRT, FRANK BRASCH, DORTE RIEMANN, AND JOACHIM RICHTER Abteilung Elektronenmikroskopie, Zentrum Anatomie, Uniuersitat Gottingen, Gottingen J.R.), Klinik fur Thorax-, Herz- und Gefapchirurgie, Medizinische (H.F., A.S., F.B., D.R., Hochschule Hannouer, Hannouer (T.W.), and Klinik fur Herz- und Gefapchirurgie, Christian Albrechts Uniuersitat, Kiel (S.W.H.), Germany ABSTRACT Background: Pulmonary type I1 pneumocytes have been examined by scanning electron microscopy (SEM), transmission electron microscopy (TEM), and morphometry in numerous mammals. Until now, the fine structure of the human type I1 pneumocyte has not been studied by means of morphometry. Methods: Eleven human donor lungs, which could not be made available for a suitable recipient, were preserved with Euro Collins solution (ECS) according to clinical organ preservation techniques. The lungs were fixed via the airways. Systematic random samples were analyzed by SEM, TEM, and classical stereological methods. Results: Type I1 pneumocytes showed normal fine structural characteristics. Morphometry revealed that although inter-individual variation due to some oedematous swelling was present, the cells were in a normal size range as indicated by an estimated mean volume of 763 k 64 pm3. The volume densities were: nucleus 21.9 & 2.2%,mitochondria 5.8 k 0.9%,lamellar bodies 9.8 f 3.6%, and remaining cytoplasmic components 62.4 2.9%of the cell volume. Since the inter-individual variations in the volume densities referred to the cell may, to variable degrees, reflect the variation in the reference space, the volume densities referred to the constant test point system and the respective volume-to-surface ratios were used for interindividual comparisons. These parameters indicate that lamellar bodies were independent of cellular swelling, while mitochondria < nucleus < remaining cytoplasmic components increased in size with increasing cell size. Conclusions:Two to 7.5 hours of cold ischemia following ECS preservation do not deteriorate the fine structure of type I1 pneumocytes of human donor lungs. For reliable assessment of fine structural variations, morphometric parameters are required that are independent of variations in cell size. 0 1995 Wiley-Liss, Inc. * Key words: Human lung, Type I1 pneumocytes, Lamellar bodies, Surfactant, Electron microscopy, Morphometry, Organ preservation, Euro Collins solution The fine structure of pulmonary type I1 pneumocytes of all major components of the alveolar lining material has qualitatively been described in detail both by termed surfactant (Askin and Kuhn, 1971; Chevalier means of scanning (SEM) and transmission electron and Collet, 1972; Voorhout et al., 1993; Young et al., microscopy (TEM) in various mammalian species (e.g., 1993). The surfactant material, which is composed of Stratton, 1978; Takaro et al., 1979; Crapo et al., 1982; about 90-95% lipid, predominantly phospholipids, and Kliewer et al., 1985; Shimura et al., 1986; Vincent and Nadeau, 1987; Ten Have-Opbroek et al., 1988; Shibamot0 et al., 1989; Snyder and Magliato, 1991; Lakritz et al., 1992) including man (e.g., Weibel, 1963; McDouReceived December 14, 1994; accepted April 6, 1995. gall and Smith, 1975; Stratton, 1984; Weibel and TayAddress reprint requests to Dr. Heinz Fehrenbach, Abt. Eleklor, 1988). Autoradiographic and biochemical studies tronenmikroskopie, Zentrum Anatomie, Universitat Gottingen, clearly showed that type I1 pneumocytes are the source Kreuzbergring 36, D-37075 Gottingen, Germany. 0 1995 WILEY-LISS, INC 50 H. FEHRENBACH ET AL. 5 -10% protein (Hawgood, 19911, is intracellularly deposited in special storage organelles termed lamellar bodies, and is excreted upon various stimuli (Wright and Dobbs, 1991; Van Golde et al., 1994). As alveolar surfactant plays a major role in the maintenance of the normal respiratory function of the mammalian lung, pathogenetic alterations of type I1 pneumocytes have been claimed in some instances to be responsible for the development of surfactant related respiratory dysfunction (e.g., Holm and Matalon, 1989; Seeger et al., 1990; Gregory et al., 1991; Uhlig et al., 1995). Recent studies indicate that impairment of the surfactant system might be one of the factors that are involved in limiting the respiratory function achieved after lung transplantation (Klepetko et al., 1990; Veldhuizen et al., 1993; Erasmus et al., 1994). Hence, surfactant replacement therapy in lung transplantation (Novick et al., 1994) and in adult respiratory distress syndrome (Haslam et al., 1994, Heikinheimo et al., 1994) is coming into focus. Although a number of morphometric studies of human lungs have been performed (Weibel, 1963,1979b; Cassan et al., 1974; Gehr et al., 1978; Crapo et al., 1982;Zeltner et al., 1987; Stone et al., 1992a,b),human type II pneumocytes, however, have not yet been studied in detail by means of morphometry. As TEM based morphometry has recently been shown to be a helpful tool in the assessment of human donor lung quality (Fehrenbach et al., 19941, this study aims at the morphometric characterization of the type I1 pneumocyte of human donor lungs. Today, donor lung preservation by means of flush perfusion of the pulmonary artery with cold modified Euro Collins solution is one of the most widely and successfully applied procedures for transplantation purposes allowing periods of ischemia of about 7 hours (e.g., Wahlers et al., 1991). Ischemic stress is well known to result in cellular membrane damage in various organs (e.g., Gutierrez, 1991). Since the cell and its various components may subsequently respond with oedematous swelling to different degrees (Schmiedl et al., 1990, 1993; Schnabel et al., 1991), close attention was given to the investigation of the relationship between parameters characterizing the size of type I1 pneumocytes and parameters characterizing its components, with special emphasis to the surfactant-storing lamellar bodies. Hence, the study presented provides a basis for future studies of alterations of the human surfactant system as may occur, e.g., during organ preservation and ischemia, due to experimental conditions in cell culture systems, or as a consequence of pathogenetic processes. MATERIALS AND METHODS In 11 cases of clinical single lung transplantations performed a t the division of cardiothoracic and vascular surgery of Hannover Medical School, Germany (head of the division: Prof. Dr. H. G. Borst), the contralateral donor lungs were studied by means of electron microscopy provided that they could not be made available for a suitable recipient by The Eurotransplant Foundation Centre, Leiden, The Netherlands. Due to distant organ procurement, donor lungs are generally subjected to a period of ischemia, i.e., the time period during which the organ is excluded from circulation. To minimize the effects of ischemia and to achieve the more favourable condition of cold ischemia, various organ preservation procedures have been developed, which in general include a lowering of the organ's temperature (e.g., Wahlers et al., 1986). In this study, organ preservation of the donor double lung block was performed according to present standard clinical procedures as described earlier (Wahlers et al., 1991). Following pulmonary arterial perfusion with cold modified Euro Collins solution, the donor lungs were stored at low temperature in the same solution until transplantation. Left and right donor lungs were separated immediately before transplantation. While one donor lung was transplanted, the contralateral donor lung was fixed by instillation via the airways to ensure rapid and uniform fixation of the whole organ for subsequent fine structural examination. Thus, the fine structural and morphometric data obtained are representative of the whole organ, and differences resulting from sampling bias as may be seen in autopsy or biopsy specimens (Bachofenand Bachofen, 1990) can be excluded. The period of cold ischemia of the donor lungs studied ranged between 121 and 458 minutes (mean value s.d. of n = 11lungs: 275.5 88.2 minutes). The age of the donors studied was in the range of 17 to 52 years (mean value 2 s.d. of n = 11lungs: 32.3 12.6 years). Six cases (91102; 91/05; 91/06; 91/10; 92/05; 92/06) were characterized by excellent postoperative respiratory function in the respective recipient as well as by absence of macroscopical contusion areas in the morphometrically examined contralateral donor lung. In 3 cases (92102; 92/03; 92/04), macroscopical inspection of the contralateral donor lung showed areas of contusion. However, good to excellent postoperative respiratory function was achieved in the patient receiving the respective twin lung. In another 2 cases (91101; 92/01), the morphometrically examined contralateral donor lung exhibited normal macroscopical aspect, while only mediocre postoperative respiratory function was achieved in the respective recipient. * * * Fixation and Sampling After separation of the double lung block in the operation theatre, the contralateral donor lung was immediately fixed, i.e., approximately at the time of transplantation as described recently (Fehrenbach et al., 1994). Briefly, a mixture of 1.5% glutaraldehyde and 1.5%paraformaldehyde in 0.1 M cacodylate buffer (pH = 7.35; buffer osmolality = 300 mOsm/kg) was instilled via the airways at a pressure of 25 cm H,O. Instillation was continued until the flow of the fixative ceased automatically. After storage in the same fixative for 12.5 to 54.8 hours at 8"C, a standardized systematic sampling procedure was performed following Cruz-Orive and Weibel(1981) and Miiller et al. (1981) to obtain isotropic uniform random samples of each lung for SEM and TEM examination, respectively. Briefly, the lungs were cut into 2 cm thick slices starting apically at a random point. Then, a double-test point system (18 A-points, d, = 12 cm; 40 B-Points, d, = 6 cm) was superimposed at random on the lung slices. Samples were taken at all points that hit parenchymal regions, with A-points determining samples to be processed for SEM, and B-points determining samples to be processed for TEM, respectively. 51 MORPHOMETRY OF HUMAN TYPE I1 PNEUMOCYTES Tissue Processing For SEM, approximately 5 x 5 x 3 mm3 sized samples were processed according to the OTOTO method described by Malick et al. (1975). After dehydration through a graded series of alcohols, specimens were critical point dried, and subsequently observed by means of a DSM 960 (Zeiss, Oberkochen, Germany) omitting metal-sputtering. For TEM, approximately 1mm3 cubes of pulmonary parenchyma were processed according to Fehrenbach et al. (1991). Processing was standardized using a n automated tissue processor Histomat (bio-med, Theres, Germany). Briefly, after several buffer rinses, samples were post-osmicated, stained en bloc with uranyl acetate, dehydrated through a graded series of ethanol, and transferred to Araldite via propylene oxide. Finally, samples were embedded in Araldite at random orientation. Five isotropic ultrathin sections of at least 10 tissue blocks per lung were collected on Formvarcoated 1 x 2 mm slot grids and counter stained with lead citrate by means of a n Ultrostainer (Leica, Hamburg, Germany). Qualitative and morphometric TEM analysis was performed using a Zeiss EM 10A. To study the fine structure of lamellar bodies a t high magnifications, ultrathin sections of approximately 40 nm were examined by zero loss-filtering with a Zeiss CEM 902, which allows for the examination of unstained ultrathin sections at high contrast (Egle et al., 1984). Morphometry For morphometric analysis of type I1 pneumocytes, 5 random samples were studied in each case. Randomly orientated, isotropic ultrathin sections were examined systematically. Starting at a random point outside, the whole section was examined along a grid of uniformly distanced, straight lines. Micrographs were recorded of all the cells of each section that were hit by the lines and that were separated by at least one distance unit. Micrographs were taken at a primary magnification of x 4,000, to ensure that the whole cell was recorded. The final magnification after photographic reproduction was determined by means of a calibration grid to be x 11,450. Calibration was performed during each photographic session. A total of 748 profiles of type I1 pneumocytes were analysed (mean s.d. of n = 11 lungs, 68 2 8.2 cell profiles per lung), which due to the sampling procedure described can be expected to meet the demands of isotropic uniform random (IUR) samples. For morphometric analysis of the cell and its components, a multipurpose coherent test system was used for point and intersection counting (324 points, 162 test lines, length d = 10 mm) according to classical stereological techniques (comprehensive reviews: Weibel, 1979a, 1990; Chang et al., 1991; Bolender et al., 1993). In each case, a minimum of 2,000 points falling on type I1 pneumocytes were counted (mean 2 s.d. of n = 11lungs, 3,855 1,497 points), so that structures accounting for 5%or more of the total cell volume could be determined with a standard error of less than 10% of the mean (Weibel, 1979a: p. 114). The following morphometric parameters were recorded for each tissue sample according to the formulas given in Table 1. * * TABLE 1. Formulas used for calculation of the morphometric parameters Volume densities m C P cell, VV(ce1litest) i= 1 m x pm3/pm3 Ptest m C P compi i=l 2P pm3/pm3 cyto, i=l m CPcompi a i=l mx pm3/pm3 Ptest Surface-to-volume ratios SVR,,) &Xi i=l pm2/pm3 Volume-to-surface ratios VsR,,, 1 pm3/pm2 S,R(x) Cellular mean caliper diameter H k x vsR(cell) Mean cell volume V 4 3 pm3 Cell = total type I1 pneumocyte; comp = compartment (nucleus, lamellar bodies, mitochondria, or remaining cytoplasmic components, respectively); cyto = cytoplasm (total cell without nucleus); test = coherent test system; P,,,, = number of points of the coherent test system ( = 324 points); P, = number of points falling on respective structure (x = cell, comp); I, = number of test line intersections with respective structure (x = cell or comp); d = test line length in micrometer; a = constant, emperically determined to be 4.42; i = individual cell profile; m = number of cell profiles of a given ultrathin section (= tissue block). The cell The volume density of type I1 pneumocytes was determined with reference to the test system Vv(cell/test) A ER G IC L M MV N PC VSR V" Abbreviations Alveolar space Endoplasmic Reticulum Golgi apparatus Interstitial cell Lamellar body Mitochondrium Multivesicular body Nucleus Projection core Volume-to-surface ratio Volume density 52 H. FEHRENBACH ET AL. to obtain a n estimate of the variation in cell size. Additionally, the surface-to-volume ratio S,R(cell) and the volume-to-surface ratio VsR(cell) were determined, parameters which are independent of alterations of the reference space. Based on the qualitative TEM and SEM data (see Figs. 1,2), the cells can be regarded as oblate ellipsoids and, thus, the mean caliper diameter H was calculated according to Weibel (1979a: Table 6.1) with the shape constant k chosen to be 9.026. The mean caliper diameter was used-to obtain a rough estimate of the mean cell volume V according to the formula given in Table 1. several closely adjacent cells. In cases 91/01,92/04 and 92/06, TEM showed a n overall swelling of the cisternae of the endoplasmatic reticulum and, in places, of mitochondria which was accompanied by a clearing of the cytoplasmic matrix (Fig. lb). The cells contained various types of lamellar bodies (Fig. 31, including immature and mature bodies both of the hemispherical as well a s of the multilamellar type, following the terminology of Stratton (1984). While in all cases multivesicular bodies were regularly seen, homogeneous lipid bodies were observed only in a few of the several hundreds of type I1 pneumocytes examined. The nucleus and the cytoplasmic components The volume densities V, of nucleus (nu), lamellar bodies (lb), mitochondria (mi), and all remaining cytoplasmic components (cy) were determined to obtain a n estimate of the fine structural composition of the type I1 pneumocytes. Volume densities were determined with reference to the whole cell as Vv(comp/cell), and with reference to the cytoplasm, i.e., the cell without the nucleus, as V,(comp/cyto). As one has to take into account that these reference spaces might show considerable variations, V,(comp) were also calculated with reference to the test point system a s V,(comp/test). In addition, the volume-to-surface ratios V,R(comp) or in t u r n the surface-to-volume ratios SvR(comp) were determined, parameters which are indicative of the size of the respective structures, and which are independent of variations of the reference space. At the magnification used for morphometric analysis, the type I1 pneumocyte profiles account for only a fraction of the test system. To facilitate comparison and graphic presentation, the discrete values of V,(comp/ test) obtained from a n individual ultrathin section were multiplied with a constant a, which was empirically determined to be a = XPtest/XPcell= 4.42. Thus, values of dimensions comparable to the values of Vv(comp/cell) were obtained. Statistics Data referring to individual tissue samples are given as discrete values obtained according to the formulas given in Table 1. Data referring to individual lungs are given a s mean values f SEM of five tissue samples unless stated otherwise. Data referring to several lungs are given a s mean values f S.D. unless stated otherwise. Correlation between variables was performed according to standard least-squares linear regression. The levels of significance P of the correlation coefficients r are given according to Sachs (1992: Table 193). The condition of P c0.05 is considered to be significant. RESULTS Qualitative Scanning and Transmission Electron Microscopy The type I1 pneumocytes of the eleven human donor lungs studied exhibit the normal fine structural appearance, as seen both by TEM (Fig. 1)and SEM (Fig. 2). In detail, however, inter- as well as intra-individual variations were observed. While most frequently, type I1 pneumocytes were single epithelial cells widely separated from one another, in some cases (91/01, 91/10, 92/01) they could be seen to constitute groups or rows of Morphometry The cell (Fig. 4) The volume density of the type I1 pneumocytes referred to the test system V,(cell/test) varied between the tissue samples of a n individual lung, and considerable differences were noted from one case to another. As the data histogram of all 55 tissue samples analysed largely resembles a gaussian distribution (Fig. 4A), the type I1 pneumocytes studied were considered to constitute a normally distributed population of cells. Correspondingly, the volume-to-surface ratio V,R(cell) as well a s the respective surface-to-volume ratio SvR(cell) showed normally distributed variations. Both parameters were significantly correlated with V,(cell/test). As was shown for V,R(cell), the correlations were significant not only at the level of the tissue samples (Fig. 4B) but also at the level of the organs (Fig. 4C). Notably, highest values of V,R(cell) were observed in both cases showing macroscopical contusions (92/02; 92/04), as well a s in case 91/01, one of the 2 cases (91/01; 92/01) characterized by mediocre postoperative respiratory function in the corresponding recipient. On the basis of V,R(cell), the mean caliper diameter was calculated according to Weibel(1979a) to be 9.4 to 12.6 pm (mean ? s.d. of n = 11lungs: 10.8 f 1.0 pm). The estimated mean volumes of the t pe I1 pneumocytes were in a range between 457 pm2and 1,160 pm3 (mean f SEM of n = 11 lungs: 763 f 64 pm3), which is similar to the mean cell volumes of 613 pm3 to 1,483 pm3 (mean t SEM: 889 +- 101 pm3) reported for 8 normal human lungs (Crapo et al., 1982: Table 2). The nucleus and the cytoplasmic components Due to this considerable variation in the size of the cells, the volume densities of nucleus, lamellar bodies, mitochondria, and remaining cytoplasmic components were given not only with reference to the cell and to the cytoplasm, as V,(comp/cell) and Vv(comp/cyto) (Table 11, respectively but also with reference to the constant test system, a s V,(comp/test). Nucleus (Fig. 5 ) The volume density V,(nu/cell) of the nucleus of the lungs studied ranged between 18.0 and 24.8% of the respective cell volume (mean 2 s.d. of n = 11 lungs: 21.9 t 2.2%). Linear regression analysis did not indicate significant correlations between V,(nu/cell) or V,(nu/cyto) and VsR(cell), respectively (Fig. 5A). However, if V,(nu/test), the nuclear volume density calculated with reference to the test system, or the volumeto-surface ratio VsR(nu), which is independent of the MORPHOMETRY OF HUMAN TYPE I1 PNEUMOCYTES Fig. 1. Fine structural appearance of human type I1 pneumocytes as seen by transmission electron microscopy. The apical surface, facing the gas containing alveolar space, bears numerous microvilli. Characteristic features of the basal surface are cytoplasmic processes which may (arrows) or may not (arrowheads) perforate the basal lamina, and may then be in close association with interstitial cells. At the lateral face, the cell membrane forms junctional complexes with adjacent type I or type I1 pneumocytes. Variations in fine structure 53 are most obvious in the appearance of the cytoplasmic components. a: (case 91/05): nucleus, lamellar bodies, multivesicular bodies, and cisternae of the endoplasmic reticulum as well as the cytoplasmic matrix are of normal appearance, while mitochondria may show partial loss of matrix ( 0 ) . b: (case 91/01):Cisternae of the endoplasmic reticulum are dilated, the basal part of the cell shows a clearing of the cytoplasmic matrix, and disintegration of cristae. H. FEHRENBACH ET AL. 54 Fig. 2. Fine structural appearance of human type I1 pneumocytes as seen by scanning electron microscopy. While in some cells, the apical surface is characterized by the microvilli being restricted to a marginal zone entouring a bald-like central patch, it may be completely studded with microvilli in others (inset). In places, globular protrusions (arrowheads) may be seen, which probably are lamellar bodies ready to be excreted. reference space, were considered, strong positive correlations with V,R(cell) were observed (Fig. 5B, 5C). The relationship between nuclear and cellular parameters was apparent not only a t the level of the tissue samples (not shown) but also a t the level of the organs. Lamellar bodies (Figs. 6, 7) The volume density of the lamellar bodies varied considerably from lung to lung (Fig. 6). In 9 cases, Vv(lb/cell) ranged between 5.3 and 10.3% of the respecs.d. of n = 9 lungs: 8.4 k tive cell volume (mean l.8%), and Vv(lb/cyto) was in the range of 7.0 to 13.6% of the respective cytoplasmic volume (mean 2 s.d. of n = 9 lungs: 10.8 2.3%). In 2 cases, the volume densities reached considerably higher values with Vv(lb/ cell) of 16.5% and Vv(lb/cyto) of 20.8% in case 91/06, and V,(lb/cell) of 16.2% and V,(lb/cyto) of 19.9% in case 92/05. Similar degrees of variation were seen as concerns Vv(lb/test) (range of n = 9 lungs: 6.3 to 10.7%, mean s.d. 9.0 -C 1.4%; case 91/06: 17.5%; case 91/05: 17.4%). Due to this considerable difference as compared to * * * the other cases, which awaits further investigation, case 91/06 and 92/05 were excluded from the correlation analysis concerning lamellar body related parameters. V,(lb/cell) (Fig. 7A) as well as V,(lb/cyto) showed a strong negative correlation with VsR(cell), i.e., the bigger the cells or their respective cytoplasmic volumes were, i.e., the more pronounced they were swollen, the smaller was the lamellar body volume fraction. However, neither V,(lb/test) nor VsR(lb) were significantly correlated with VsR(cell) (Fig. 7B,C). Mitochondria (Fig. 8) The volume density V,(mi/cell) of the mitochondria of the lungs studied ranged between 4.2 and 8.4% of the respective cell volume (mean k s.d. of n = 11lungs: 6.2 1.3%). At the level of the tissue samples, V,(mi/cell) showed a wider range of 2.4 to 10.2% (mean s.d. of all 55 tissue samples: 5.8 1.6%).Linear regression analysis revealed no correlation between V,(mi/cell) or Vv(mi/cyto) and VsR(cell), respectively (Fig. 8A). In contrast, the volume density referred to the test system Vv(mi/test) (Fig. 8B) a s well as the volume-to-surface * * MORPHOMETRY O F HUMAN TYPE I1 PNEUMOCYTES 55 Fig. 3.Lamellar body sub-structure of human type I1 pneumocyte as seen in the zero loss-mode of an energy-filtering transmission electron microscope (Zeiss CEM 902). The ultrathin section was not counter-stained with lead citrate. The lamellar body shown is of the hemispherical type according to Stratton (1984).The concentric lamellae end on the sub-structured projection core material. The limiting membrane (arrows) of the body forms a projection (open arrowhead), which is in close association with cisternae of the endoplasmic reticulum. ber of studies (Campiche et al., 1963; Weibel, 1963; Balis and Conen, 1964; McDougall and Smith, 1975; Gehr et al., 1978; Stratton, 1978; Weibel and Taylor, Remaining cytoplasmic components (Fig. 9) 1988). Morphometric studies of human lungs, however, At the level of the organs, the volume density of the have focussed on the composition of the pulmonary paremaining cytoplasmic components V,(cy/cell), i.e., of renchyma, and special attention has been given to the all the cellular components other than nucleus, lamel- relationship of structure and function at the level of the lar bodies, and mitochondria, ranged between 56.6 and air-blood barrier (Weibel, 1963, 1979b; Cassan et al., 66.7% of the respective cell volume (mean 5 s.d. of n = 1974; Gehr et al., 1978; Crapo et al., 1982; Zeltner et 11 lungs: 62.4 2.9%). At the level of the tissue sam- al., 1987; Stone et al., 1992a). Until now, however, a ples, Vv(cy/cell)ranged between 49.4 and 76.6% (mean detailed morphometric study of the human type I1 t s.d. of all 55 tissue samples: 62.4 k 5.6%).While only pneumocyte and its fine structural components has not a very weak positive correlation of Vv(cy/cell) with yet been presented. The lungs examined in this study were obtained durVsR(cell) (Fig. 9A) was observed, V,(cy/test) showed a strong positive correlation with VsR(cell) (Fig. 9B). ing clinical single lung transplantations from human This relationship was observed not only at the level of donors, whose contralateral lung could not be made the tissue samples (Fig. 9B) but also a t the level of the available for a suitable recipient. Clinical organ preservation of the donor double lung block was performed organs (Fig. 9C). according to standard procedures with Euro Collins soDISCUSSION lution (Wahlers et al., 1991), which has recently been The normal human type I1 pneumocyte has qualita- shown to yield an adequate preservation of the parentively been described in detail by transmission (TEM) chymal fine structure in lungs subjected to 3.5 to 5 andlor scanning electron microscopy (SEMI in a num- hours of cold ischemia (Fehrenbach et al., 1994). In the ratio V,R(mi) (Fig. 8C) were positively correlated with VsR(cell). * 56 H. FEHRENBACH E T AL. present study, very similar conditions were noted, i.e., 2 to 5.5 hours of cold ischemia. In only one lung, a period of 7.5 hours of cold ischemia following ECS preservation was achieved. However, there were no indications that the prolonged duration of ischemia had any detrimental effect to the fine structure of the tvpe I1 " _ pneumocytes. This is in line with recent experimental data from rat lung transplantations that showed surfactant function being impaired only after onset of reperfusion but not after ischemia alone (Erasmus et al., 1994). A rough estimation of the mean volume of the cells, calculated from the volume-to-surface ratio of the cell; yielded a range of 457 pm3 to 1,160 pm3 (mean SEM of n = 11 lungs: 763 ? 64 pm3). Similar mean cell volumes of type I1 pneumocytes, 613 pm3 to 1,483 pm3 (mean SEM: 889 k 101 pm3) have been given for 8 normal human lungs (Crapo et al., 1982). A mean volume of 815 pm3 ( 2 SEM of 144 pm3)has been reported from the morphometric analysis of 4 resected human lung lobes (Stone et al., 1992b). Thus, a gross oedematous swelling of the cells analysed in this study can be excluded. Similar variations in the values reported for the mean type I1 pneumocyte volume are seen comparing different studies of normal adult Spra ue Dawley rats: 440 pm3 (Young et al., 1981), 525 pm8 (Young et al., 19851, 385 pm3 (Randell et al., 19911, and 443 pm3 (Stone et al., 1992b),respectively. Moreover, a comparative study of lungs of 10 mammalian species including man revealed that there was no significant change in the size of type I1 pneumocytes with increasing body weight (Stone et al., 199213). Hence, inter-species differences in the size of type I1 pneumocyte appear to be relatively small as was also indicated by the volume densitv of the cvtodasm referred to the test svstem (Massaro et al., "1975). However, substantial inter-individual differences may be seen. This in line with our study, as indicated by the inter-individual variations in the volume-to-surface ratio of the cells and the volume density of the cells referred to the test system. Apart from numerous reports of alterations during ontogenetic processes or due to specific pathogenetic conditions, the detailed consideration of which is beyond the scope of this paper, comparative studies of the fine structure of type I1 pneumocytes of various mammalian species have revealed qualitative and quantitative inter-species differences as concerns the cytoplasmic components. Qualitative inter-species differences have been reported for example, to be present with respect to the sub-structure of lamellar bodies (Kikkawa and Spitzer, 1969;Pattle et al., 1974; Stratton, 1984).In agreement with Stratton (19781, the lamellar bodies of the human donor lungs studied are of the hemispherical * CELJ LAR PARAM ETERS 10 9 8 7 6 5 4 3 2 1 4 2.0 - r n E a. 0.8 - 0.4 1 1.6 Y 1.2 m > - 12 20 28 y I 0.0 0 1.6 I I 10 I - + 0.427 0.032~ I 20 52 36 44 I I 30 I I I 50 40 r = 0.824. p<O.OOl n E a. * /o! U U m > + y = 0.024~ 0.0.i.P 0 I ' 20 I 25 ' I 30 0.825 ' I 35 Vv ( c e I I/t e st) [%] Fig. 4 f Fig. 4. Morphometric data. A Histogram showing the distribution of the discrete values of the volume density of the cell referred to the test system V,(cell/test) of all the 55 tissue samples analyzed. Each count represents the result of the analysis of all type I1 pneumocyte profiles collected from a n individual ultrathin section, which corresponds to an individual tissue block. The distribution of the discrete values of V,(cell/test), which is a n indicator of the volume of the cells, largely follows the corresponding Gaussian distribution curve. B,C: There is a strong positive correlation of the volume-to-surface ratio of the cell VsR(cell) with Vv(cell/test) not only at the level of tissue samples (B) but also at the level of the organs (C). Values of the lungs of those cases that showed contusion areas (92/02; 92/04) or that were characterized by mediocre postoperative transplant function (91101; 92/01) are indicated. As in Figs 5, 7-9, solid lines represent the linear regression curve, the mathematical description of which is shown in the lower right corner. Dashed lines represent the corresponding 95% confidence interval. The regression coefficients and levels of significance are displayed in the upper left corner. 57 MORPHOMETRY OF HUMAN TYPE I1 PNEUMOCYTES TABLE 2. Volume densities and volume-to-surfaceratios of components of type I1 pneumocytes of various mammalian species (mean values f s.d. unless indicated otherwise) V,(nuicell) VJmiicell) V,(mi/cyto) V,(lbicell) VJlbicyto) (%I (%I (%I (%) (%I 5.8 f 0.9 7.5 f 1.1 9.8 f 3.6 12.6 f 4.4 Species Man 21.9 Baboon 28.9 f 2.4' f 2.2 6.99 f 0.75' Sheep 13.4 f 1.3 9.3 f 1.1 Macaque, old 32 f 3.1' 6.7 i- 0.7' 8.9 t 1.3 9.37 f 1.54' Mongrel dog 22.7 i- 1.5' Mongrel dog 12.4-16.5, Phospholipid stabilisation' OsO,; UAc oso, oso, oso, oso, 8.9 t 3.4' 11.1 f 2.1 Macaque, young VsR(lb) (nm) 179 f 21 242 f 52,3 OsO, Mongrel dog 9.1 f 1.5' 20.5 f 0.3' oso, oso, Mongrel dog 9.4 t 0.Ol2 33.1 f .05' 050, Rabbit Rabbit Rat. CFN-COBS strain Rat, Wistar Rat, Sprague Dawley Rat, Sprague Dawley Rat, Sprague Dawley Rat, Sprague Dawley Rat, Fischer Rat, Fischer 24.4 2 2.3' 236 i- 14.5'~~OsO, 13.6 f 1.7' oso, 10.9 f 1.4' 4.9 f 0.1' 21.9 2 0.P 7.6 f 0.4' 8.9 f 0.2' 8.2 f 0.3' 12.9 f 0.3' 12.7 f 0.7' 26.9 t 0.6' 145 f 0.5' OsO,; UAc OsO,; UAc oso,; 12.8 i- 1.5' 24 t 3 8.4 2 1 1 2 -t 1 OsO,; UAc 25 f 0.8 8.8 2 0.4 9.8 t 0.6 OsO,; UAc 25 Ifr 3 8 f 1 12 t 1 OsO,; UAc 9 f 1 8.4 f 0.33 15 f 1 15.9 0B3 21 2 2 25.9 t 1.63 oso, * Rat, Long-Evans Mouse, C57BLi6J strain OsO,; UAc 18.6 f 0.6' 22.7 f 235 t 92,3 1.6' 229 f Oso, OsO. References This study Fracica et al., 1994 Shimura et al., 1986 Shimura et al., 1986 Lakritz et al., 1992 Massaro and Massaro, 1975 Shepard et al., 1982 DeFouw and Chinard, 1983 DeFouw and Berendsen, 1977 Massaro and Massaro, 1975 Gail et al., 1975 VidiC and Burri, 1981 Ishii et al., 1989 Massaro and Massaro. 1973 Young et al., 1985 Young et al., 1991 Kliewer et al.. 1985 Pinkerton et al., 1990 Balis et al., 1988 Massaro and Massaro, 1975 Massaro and Massaro. 1975 'Postfixation with osmium tetroxide (OsO,) only, or additional en bloc staining with Uranyl acetate (UAc). 'Mean values t SEM. 3Calculated from original values. ,Range of values from n = 5 normal canine lungs. and multilamellar type, and may contain the typical, sub-structured projection core material (see Fig. 3). As described by Shimura et al. (19831,lamellar bodies were seen in close association with the cisternae of the endoplasmic reticulum, the membranes of which have recently been discussed to directly contribute to lamellar body growth (Risco et al., 1994). We also regularly observed multivesicular bodies, which comprise several structurally different types (Williams, 1987),and which may play a role not only in the de novo formation of lamellar bodies (Chevalier and Collet, 1972)but also in the recycling andlor degradation of alveolar surfactant (Young et al., 1993). Neutral lipid bodies, however, which in cells isolated from human lungs were shown to be sites of cyclooxygenaseactivity (Dvorak et al., 1992), were only very rarely seen in our preparations, while they have regularly been observed in canine lungs (Fehrenbach et al., 1991), and were abundantly present in human type I1 pneumocytes cultured on EnglebrethHolm-Swarm tumour matrix (Edelson et al., 1989). An increase in neutral lipid bodies has been reported with age in monkeys (Shimura et al., 1986), and after haemorrhagic shock in dogs (Barkett et al., 1969). Quantitative inter-species differences with respect to lamellar bodies have been reported to be quite small as indicated by the volume densities referred to the cytoplasm and by the surface-to-volume ratios (Massaro and Massaro, 1975). However, there are differences in the morphometric data reported by different authors studying type I1 pneumocytes of the same species (see Table 2). This may at least partly be due to the fact that lamellar bodies contain high amounts of stored phospholipids (Hawgood, 1991: Van Golde et al., 19941, and thus, are highly susceptible to the particular fixation and tissue processing procedure applied (for review see, e.g., Stratton, 1984). The application of en bloc staining with uranyl acetate prior to dehydration and embedment, as was performed in our study, has recently been shown to preserve the sub-structure of lamellar bodies in conventionally processed samples in a way that is close to the results obtained by special lipid-carbohydrate retention techniques (Fehrenbach et al., 1991). Quantitative inter-species differences have additionally been reported with respect to mitochondria of type I1 pneumocytes. Mitochondria1volume density referred to cytoplasm for example has been reported to be pos- 58 H. FEHRENBACH ET AL. NUCLEUS LAMELLAR BODIES 25 bQ Y 30 cI cI n w I U -< n n - t 0 t' y - 5.34x + 15.43 I I I I 1.0 1.2 1.4 1.6 mVr(lb/toot) mVr(lb/odf) aVv(lb/oyto] 20 15 10 5 0 Case y 1.0 3 8 . 0 1 ~-20.29 1.2 1.4 Fig. 6. Morphometric data. Volume densities of lamellar bodies are given with reference to the test point system VJlbitest), the whole cell V,(lb/cell), and the cytoplasm V,(lb/cyto), respectively. Differences between individual lungs may vary considerably depending on the reference space chosen for calculation of the lamellar body volume density (compare, e.g., case 91/01and 91/02).While in most cases, the values of V,(lb/test) and V,(lb/cell) are very similar and lower than the values of V,(lb/cyto), in case 91\01 V,(lb/test) reaches a value nearly twice the value of V,(lb/cell). 1.6 mitochondrial inner membrane and cristae has been reported to be negatively correlated with the lung's oxygen consumption (Massaro and Massaro, 1977). NoE tably, the mean mitochondrial volume density V,(mi/ Y cyto) of 7.5 -+ 1.1% as determined in our study for hu0.8 man type I1 pneumocytes, is very close to the value of n 3 8.3% calculated according to the equation V,(mi/cyto) c = 7.1 + 0.1 x respiratory rate given by Massaro et al. U (1975: Fig. 21, on the basis of a value of 12 breaths per 0.4 Q) minute given for the respiratory rate of a normal hu> man being (Fishman, 1988: Appendix B). Discrepancies in the morphometric data reported by y 0 . 6 1 ~+ 0.07 I I I different authors for a particular species, may also 0.0 arise from differences in the age of the individuals 1.0 1.2 1.4 1.6 0.0 analysed (Shimura et al., 1986), from factors related to circadian rhythms (Ishii et al., 1989), or due to differVsR(cel1) [ p m ] ences in the sampling procedure performed, which may be particularly true if only a fraction of the whole orFig. 5. Morphometric data. Relationships of volume-to-surface ratio gan has been available as, e.g., in studies of autopsy or of the cell V,R(cell) and volume density of the nucleus referred to (A) biopsy specimens (Bachofen and Bachofen, 1990). the whole cell V,(ndcell), (B) the test point system V,(nuitest), and In most studies investigating the fine structure by (C) the volume-to-surface ratio of the nucleus V,R(nu), respectively means of morphometry, the volume densities of the nu(mean values t- SEM of 5 tissue samples per lung). While V,(nu/cell) i s not correlated with V,R(cell) (A), significant positive correlations cleus and the cytoplasmic components are given with are seen between V,R(cell) and V,(nu/test) (B) and V,R(nu) (C) re- reference to the whole cell (VidiC and Burri, 1981; spectively. Kliewer et al., 1985;Young et al., 1985,1991;Pinkerton et al., 1990; Lakritz et al., 1992) and/or with reference itively correlated with the respiratory rate and the to the cytoplasm (VidiC and Burri, 1981; Massaro and lung's oxygen consumption of a given species (Massaro Massaro, 1975; DeFouw and Berendsen, 1977; DeFouw et al., 1975), while the surface-to-volume ratio of the and Chinard, 1983; Shimura et al., 1986; Snyder and n 1.2 a I - 59 MORPHOMETRY OF HUMAN TYPE I1 PNEUMOCYTES LAMELLAR BODIES r r ae n 15 10 M ITOCH 0 N D R IA - 0571. p<O.OOl z ’ <.-E 1OI- 0 5 ’ U 00 5 ’ > -13.97~ y 0 0.0 1.0 I I 1.2 1.4 > + 25.26 I y 0 0.0 1.6 0 8 Q) 0.8 - + 0.002~ I I 1.2 1.6 5.82 I 2.0 I ’ 1 t l5 r 0.552, p<O.OOl n 10 < n Q) v s 5 -5.38~ 0 0.0 + 15.48 y 0 1.0 1.2 1.4 1.6 0.0 0.20 =t U 0.20 5 U m > 1.2 - 0.18 5.28~ I 1.6 1 2.0 r = 0.361, pq0.01 a 0.18 0 0.12 0.16 0.14 0.16 U n v 0.8 - n n E I IL/ 0.00 0.0 v 1 5 + 0.17 > 0.08 0 y = 0.001~ I I I 1 1.0 1.2 1.4 1.6 VsR(cel1) [ p m ] 0.00 L 0.0 0 y 0.8 1.2 0 . 0 2 9 ~ + 0.079 1.6 2.0 VsR(cel1) [ p m ] Fig. 7. Morphometric data. Relationships of volume-to-surface ratio of the cell V,R(cell) and volume density of lamellar bodies referred to (A) the whole cell V,(lb/cell), (B) the test point system V,(lb/test), and (C) the volume-to-surface ratio of the lamellar bodies VsR(lb), respectively (mean values ? SEM of 5 tissue samples per lung). V,R(cell) shows a strong negative correlation with V,(lb/cell), while V,(lb/test) and V,R(lb) are not correlated with V,R(cell). Fig. 8. Morphometric data. Relationships of volume-to-surface ratio of the cell V,R(cell) and volume density of the mitochondria referred to (A) the whole cell V,(mi/cell), (B) the test point system V,(mi/test), and (C) the volume-to-surface ratio of the mitochondria V,R(mi), respectively (each point represents a n individual tissue block). While V,(mi/cell) is not correlated with V,R(cell) (A), weak but significant positive correlations are observed between V,R(cell) and V,(mi/test) (B) and VsR(mi) (C), respectively. Magliato, 1991). However, as with each “density,” i.e., a ratio of two compartments, the observation of alterations is only meaningful if the reference space is largely constant throughout the individual or the experimental groups studied as was pointed out, e.g., by Massaro and Massaro (1983)and Bolender et al. (1993). H. FEHRENBACH ET AL. 60 Variations in the size of cells or their cytoplasm, a s may result from pathogenetic processes (e.g., Balis e t al., 1988; Fracica et al., 19941, will be of considerable influence on the morphometric data obtained. This is clearly illustrated by our study of the relationship of cellular parameters and the volume densities of nucleus, lamellar bodies, mitochondria, and the remaining cytoplasmic components, which were determined with 0 0 reference to the whole cell, the cytoplasm, and the test point system, respectively. On the basis of data referring to these different reference spaces, opposite correlations of the respective volume densities to the volumeto-surface ratio of the cells VsR(cell) were obtained. While, the lamellar body volume densities referred to 40 the cell V,(lb/cell) or to the cytoplasm Vv(lb/cyto) decreased with increasing VsR(cell), the volume density y 8 . 2 7 ~+ 52.48 referred to the test system remained constant. The test I 1 I 0 ' system was the only reference space with constant size, 0.0 0.8 1.2 1.6 2.0 while the cell and the cytoplasm showed considerable variations. Therefore, the decrease in V,(lb/cell) and in 0 V,(lb/cyto) has to be interpreted as a n indication that the lamellar bodies were diluted in a n increased cellular 120 or cytoplasmic volume, respectively, rather than as a n indication of a decrease in stored surfactant. 100 On the other hand, the volume densities of nucleus and mitochondria referred to the cell or to the cyto80 plasm remained more or less constant with increasing 60 VsR(cell), while the volume densities referred to the test system showed a positive correlation with VsR40 (cell), a n index of cell size (V,(nu>: r = 0.633, P <0.001; r = 0.552, P <0.001). Hence, we may conclude Vv(mi): 20 that the larger the cells were, the bigger the nucleus y 7 6 . 0 1 ~ 24.761 and the mitochondria have been. This was also reflected by the positive correlation of VsR(cell) with vs0.0 0.8 1.2 1.6 2.0 R(nu) (r = 0.468, P <0.001) and VsR(mi) (r = 0.361, P <0.01), respectively. As concerns all components of 120 the cell other than nucleus, mitochondria, and lamellar n bodies, subsumed in the remaining cytoplasmic compo100 nents, the volume density referred to the whole cell showed a very weak positive correlation with VsR(cell) n + 80 (r = 0.382, P <0.01), while a strong positive correlacn tion (r = 0.870, P <0.001) was observed as regards the 60 volume density referred to the test system. Based on the correlations of VsR(cell) with the vol40 ume densities referred to the test system of the respecW tive organelles (correlation coefficients: rcy > r,, > 20 rmJ,we may conclude that a n oedematous increase in Y 8 6 . 2 0 ~- 37.03 the size of the type I1 pneumocytes is mainly determined by a n increase in the volume of cytoplasmic com0 ponents other than mitochondria and lamellar bodies, 0.0 1.0 1.2 1.4 1.6 and only secondly by a n increase in nuclear size. Mitochondrial contribution is only of a minor extent, and VsR(cel1) [pm] lamellar bodies do virtually not contribute to the increase in cell size related to a n oedematous swelling of Fig. 9. Morphometric data. Relationships of volume-to-surfaceratio the type I1 pneumocyte. of the cell V,R(cell) and volume density of the remaining cytoplasmic Therefore, if considerable variations in the size of the components (i.e., the whole cell except nucleus, mitochondria, and lamellar bodies) referred to (A) the whole cell V,(cy/cell), and (B, C) cells to be studied have to be expected, a reference space must be chosen, which is independent of these the test point system V,(cy/test). While V,(cy/cell) and V,R(cell) show only a very weak positive correlation, a strong positive correla- variations. This demand is perfectly met by the multition is observed between V,R(cell) and Vv(cy/test). The correlation of purpose test system itself. In addition, parameters V,R(cell) and V.,(cy/test) is significant not only a t the level of the which are independent of variations in the reference tissue samples (values obtained from the analysis of all cell profiles space as, e.g., the volume-to-surface ratio or in t u r n the sub-sampled from an individual ultrathin section) (B), but also at the level of organs (mean values * SEM of 5 tissue samples per lung) (C). surface-to-volume ratio (Schmiedl et al., 1990, 1993), may be used in quantitative descriptions. Although one has to take into account the lungs' spe- REMAl N IN G CYTOPLASM t 2oL, - - - - - - m . . I . ;c 2 g MORPHOMMETRY O F HIJMAN TYPE I1 PNEUMOCYTES cific donor history, the study presented provides a basis for future morphometrical analysis of alterations of the fine structure of human type I1 pneumocytes as may occur, e.g., during extended ischemia, due to differences in organ preservation procedures, due to experimental conditions in cell culture systems, or as a consequence of pathogenetic processes. ACKNOWLEDGMENTS The authors gratefully acknowledge the expert technical assistance of S. Freese, H. Huhn, and A. Gerken. Many thanks are owed to M. Ochs and B. Will for assistance in fixation and sample preparation, and for fruitful discussions. This study was supported in part by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 330 Organprotektion, Teilprojekt B 12). LITERATURE CITED Askin, F.B., and C. Kuhn 1971 The cellular origin of pulmonary surfactant. Lab. Invest., 25t260-268. Bachofen, M., and H. Bachofen 1990 Fixation of human lungs. In: Models of Lung Disease. Microscopy and Structural Methods. J . Gil, ed. Marcel Dekker Inc., New York, pp. 23-36. Balis, J.U., and P.E. Conen 1964 The role of alveolar inclusion bodies in the developing lung. Lab. Invest., 13:1215-1229. Balis, J.U., J.F. Paterson, E.M. Haller, S.A. Shelley, and M.R. Montgomery 1988 Ozone-induced lamellar body responses in a rat model for alveolar injury and repair. Am. J. Pathol., 132:330344. Barkett, V.M., J.J. Coalson, and L.J. Greenfield 1969 Early effects of hemorrhagic shock on surface tension properties and ultrastructure of canine lungs. Bull. Johns Hopkins Hospital, 124537-94. Bolender, R.P., D.M. Hyde, and R.T. Dehoff 1993 Lung morphometry: A new generation of tools and experiments for organ, tissue, cell, and molecular biology. Am. J . Physiol., 265:L521-L548. Campiche, P.M., A. Gautier, E.I. Hernandez, and A. Reymond 1963 An electron microscopic study of fetal development of human lung. Pediatrics, 32:976-994. Cassan, S.M., M.B. Divertie, and A.L. Brown 1974 Fine structural morphometry on biopsy specimens of human lung. 1. Normal lung. Chest, 65.269-274. Chang, L.-Y., R.R. Mercer, K.E. Pinkerton, and J.D. Crapo 1991 Quantifying lung structure. Experimental design and biologic variation in various models of lung injury. Am. Rev. Respir. Dis., 143:625-634. Chevalier, G., and A.J. Collet 1972 In vivo incorporation of choline3H, l e ~ c i n e - ~and H g a l a ~ t o s e - ~in H alveolar type I1 pneumocytes in relation to surfactant synthesis. A quantitative radioautographic study in mouse by electron microscopy. Anat. Rec., 174: 289-310. Crapo, J.D., B.E. Barry, P. Gehr, M. Bachofen, and E.R. Weihel 1982 Cell number and cell characteristics of the normal human lung. Am. Rev. Respir. Dis., 125:740-745. Cruz-Orive, L.-M. and E.R. Weibel 1981 Sampling designs for stereology. J . Microsc., 122.235-257. DeFouw, D.O., and P.B. Berendsen 1977 A morphometric examination of type I1 alveolar epithelial cells in normal and isolatedperfused dog lungs. Am. J . Anat., 150t139-144. DeFouw, D.O., and F.P. Chinard 1983 Morphometric studies of the alveolar epithelium in dog lungs in vivo after increased rates of outward filtration. Anat. Rec., 205:39-45. Dvorak, A.M., E. Morgan, R.P. Schleimer, S.W. Ryeom, L.M. Lichtenstein, and P.F. Weller 1992 Ultrastructural immunogold localization of prostaglandin endoperoxide synthase (cylooxygenase) to non-membrane-bound cytoplasmic lipid bodies in human lung mast cells, alveolar macrophages, type I1 pneumocytes, and neutrophils. J . Histochem. Cytochem., 40:759-769. Edelson, J.D., J.M. Shannon, and R.J. Mason 1989 Effects of two extracellular matrices on morphologic and biochemical properties of human type I1 cells in vitro. Am. Rev. Respir. Dis., 140:13981404 Egle, W., D. Kurz, and A. Rilk 1984 The EM 902, a new analytical TEM for ESI and EELS. Zeiss-Inform. MEM, 3:4-9. Erasmus, M.E., A.H. Petersen, S.B. Oetomo, and J . Prop 1994 The function of surfactant is impaired during reimplantation re- 61 sponse in rat lung transplants. J . Heart Lung Transplant., 13: 791-802. Fehrenbach,H., J . Richter, and Ph.A. Schnabel 1991 Improved preservation of phospholipid-rich multilamellar bodies in conventionally embedded mammalian lung tissue: An electron spectroscopic study. J . Microsc., 162:91-104. Fehrenbach, H., S.W. Hirt, T. Wahlers, Ph.A. Schnabel, A. Haverich, and J . Richter 1994 Euro-Collins flush perfusion in lung preservation: Ultrastructural studies of the Dreservation aualitv of the contralateral donor lung in clinical s&gle lung traksplantation. J . Heart Lung Transplant., 13:l-14. Fishman, A.P. 1988 Pulmonary Diseases and Disorders. McGraw-Hill Book Co., New York. Fracica, P.J., S.P. Caminiti, C.A. Piantadosi, F.G. Duhaylongsod, J.D. Crapo, and S.L. Young 1994 Natural surfactant and hweroxic lung injury in primates. 11. Morphometric analyses. J: Appl. Physiol., 76:lOOZ-1010. Gail, D.B., G.D. Massaro, and D.Massaro 1975 Influence of cycloheximide on the 1un-g. J . Appl. Physiol., 38:623-629. Gehr, P., M. Bachofen, and E.R. Weibell978 The normal human lung: Ultrastructure and morphometric estimation of diffusion capacity. Respir. Physiol., 32:121-140. Gregory, T.J., W.J. Longmore, M.A. Moxley, J.A. Whitsett, C.R. Reed, A.A. Fowler 111, L.D. Hudson, R.J. Maunder, C. Crim, and T.M. Hyers 1991 Surfactant chemical composition and biophysical activity in acute respiratory distress syndrome. J . Clin. Invest., 88:1976-1981. Gutierrez, G. 1991 Hypoxemia and ischemia. Cellular effects. In: The Lung: Scientific Foundations. R.G. Crystal, and J.B. West, eds. Raven Press, New York, pp. 1525-1533. Haslam, P.L., D.A. Hughes, P.D. MacNaughton, C.S. Baker, and T.W. Evans 1994 Surfactant replacement therapy in late-stage adult respiratory distress syndrome. Lancet, 343:1009-1011. Hawgood, S. 1991 Surfactant: Composition, structure, and metabolism. In: The Lung: Scientific Foundations. R.G. Crystal, and J.B. West, eds. Raven Press, New York, pp. 247-261. Heikinheimo, M., M. Hynynen, P. Rautiainen, S. Andersson, M. Hallman, and s. Kukkonen 1994 Successful treatment of ARDS with two doses of synthetic surfactant. Chest, 105:1263-1264. Holm, B.A., and S. Matalon 1989 Role of pulmonary surfactant in the development and treatment of adult respiratory distress syndrome. Anesth. Analg., 69~805-818. Ishii, Y., S. Hasegawa, and Y. Uchiyama 1989 Twenty-four-hour variations in subcellular structures of rat type I1 alveolar epithelial cells: A morphometric study at the electron-microscopic level. Cell Tissue Res., 256:347-353. Kikkawa, Y., and R. Spitzer 1969 Inclusion bodies of type I1 alveolar cells: Species differences and morphogenesis. Anat. Rec., 163: 525-542. Klepetko, W., A. Lohninger, W. Wisser, M.R. Mueller, G. KhunlBrady, A. Windisch, F. Eckersberger, and E. Wolner 1990 Pulmonary surfactant in bronchoalveolar lavage after canine lung transplantation: Effect of L-carnitine application. J. Thorac. Cardiovasc. Surg., 99:1048-1058. Kliewer, M., E.K. Fram, A.R. Brody, and S.L. Young 1985 Secretion of surfactant by rat alveolar type I1 cells: Morphometric analysis and three-dimensional reconstruction. Exp. Lung Res., 9:351361. Lakritz, J., A.J. Weir, A.T. Mariassy, J.Z. Kendall, and C.G. Plopper 1992 Ultrastructural morphometric characterization of alveolar type II cells of perinatal sheep: A comparison with adults. Anat. Embryo]., 186:355-362. Malick, L.E., R.B. Wilson, and D. Stetson 1975 Modified thiocarbohydrazid procedure for scanning electron microscopy: Routine use for normal, pathological, or experimental tissues. Stain Technol., 50:265-269. Massaro, G.D., and D. Massaro 1973 Hyperoxia: A stereologic ultrastructural examination of its influence on cytoplasmic components of the pulmonary granular pneumocyte. J . Clin. Invest., 52:566-570. Massaro, G.D., and D. Massaro 1975 Stereologic evaluation of granular pneumocyte lamellar bodies in different species (38801). Proc. SOC.Exp. Biol. Medicine, 149~333-335. Massaro, G.D., and D. Massaro 1977 Mitochondria of the pulmonary granular pneumocyte in different species. Am. Rev. Respir. Dis., 115:359-361. Massaro, G.D., and D. Massaro 1983 Morphologic evidence that large inflations of the lung stimulate secretion of surfactant. Am. Rev. Respir. Dis., 127~235-236. Massaro, G.D., D.B. Gail, and D. Massaro 1975 Lung oxygen con- 62 H. FEHKENBACH ET AL. sumption and mitochondria of alveolar epithelial and endothelial cells. J . Appl. Physiol., 38:588-592. McDougall, J., and J.F. Smith 1975 The development of the human type I1 pneumocyte. J . Pathol., 115,245-251. Muller, A.E., L.M. Cruz-Orive, P. Gehr, and E.R. Weibel 1981 Comparison of two subsampling methods for electron microscopic morphometry. J . Microsc., 123:35-49. Novick, R.J., R.A.W. Veldhuizen, F. Possmayer, J . Lee, D. Sandler, and J.F. Lewis 1994 Exogenous surfactant therapy in thirty-eight hour lung graft preservation for transplantation. J. Thorac. Cardiovasc. Surg., 108:259-268. Pattle, R.E., G. Gandy, C. Schock, and J.M. Creasey 1974 Lung inclusion bodies: Different ultrastructure in simian and non-simian mammals. Experientia, 30t797-798. Pinkerton, K.E., S.L. Young, E.K. Fram, and J.D. Crapo 1990 Alveolar type I1 cell responses to chronic inhalation of chrysotile asbestos in rats. Am. J . Respir. Cell. Mol. Biol., 3543-552. Randell, S.H., R. Silbajoris, and S.L. Young 1991 Ontogeny of rat lung type I1 cells correlated with surfactant lipid and surfactant apoprotein expression. Am. J . Physiol., 260:L562-L570. Risco, C., C. Romero, M.A. Bosch, and P.P. da Silva 1994 Type I1 pneumocytes revisited Intracellular membranous systems, surface characteristics, and lamellar body secretion. Lab. Invest., 70:407-417. Sachs, L. 1992 Angewandte Statistik, 7th ed. Springer-Verlag,Berlin. Schmiedl, A,, Ph.A. Schnabel, G. Mall, M.M. Gebhard, D.H. Hunnemann, J. Richter, and H.J. Bretschneider 1990 The surface to volume ratio of mitochondria, a suitable parameter for evaluating mitochondrial swelling. Correlations during the course of myocardial global ischaemia. Virchows Arch. [A] 416:305-315. Schmiedl, A,, Ph.A. Schnabel, J. Richter, and H.J. Bretschneider 1993 Close correlations between mitochondria1 swelling and ATP-content in the ischemic canine myocardium. A combined morphometric and biochemical study. Pathol. Res. Pract., 189:342-351. Schnabel, Ph.A., J. Richter, A. Schmiedl, B. Ramsauer, U. Bartels, M.M. Gebhard, G. Mall, and H.J. Bretschneider 1991 The ultrastructural effects of global ischaemia on Purkinje fibres compared with working myocardium: A qualitative and morphometric investigation on the canine heart. Virchows Arch. [A] 418:17-25. Seeger, W., U. Pison, R. Buchhorn, U. Obertacke, and T. Joka 1990 Surfactant abnormalities and adult respiratory failure. Lung, 168(Suppl):891-902. Shepard, J.W., jr., G.F. Dolan, and S.Y. Yu 1982 Factors regulating lamellar body volume density of type I1 pneumocytes in excised dog lungs. J. Appl. Physiol., 53:555-562. Shibamoto, T., T. Kobayashi, and T. Nagata 1989 Platelike and globelike inclusion bodies associated with rough endoplasmic reticulum in alveolar type 2 cells of the sheep. Anat. Rec., 223t49-54. Shimura, S., T. Aoki, and T. Takishima 1983 Connection of lamellar body to intracellular membrane system in type I1 alveolar cells of human lung. Am. Rev. Respir. Dis., I27:493-494. Shimura, S., E.S. Boatman, and C.J. Martin 1986 Effects of ageing on the alveolar pores of Kohn and on the cytoplasmic components of alveolar type I1 cells in monkey lungs. J. Pathol., 148:l-11. Snyder, J.M., and S.A. Magliato 1991 An ultrastructural, morphometric analysis of rabbit fetal lung type I1 cell differentiation in vivo. Anat. Rec., 229:73-85. Stone, K.C., R.R. Mercer, B.A. Freeman, L.-Y. Chang, and J.D. Crapo 1992a Distribution of lung cell numbers and volumes between alveolar and nonalveolar tissue. Am. Rev. Respir. Dis., 146:454456. Stone, K.C., R.R. Mercer, P. Gehr, B. Stockstill, and J.D. Crapo 199213 Allometric relationships of cell numbers and size in the mammalian lung. Am. J. Respir. Cell Mol. Biol., 6:235-243. Stratton, C.J. 1978 The ultrastructure of multilamellar bodies and surfactant in the human lung. Cell Tissue Res., 193,219-229. Stratton, C.J. 1984 Morphology of surfactant producing cells and of the alveolar lining layer. In: Pulmonary Surfactant. B. Robert- son, L.M.G. Van Golde, and J.J. Batenburg, eds. Elsevier Science Publ., Amsterdam, pp. 67-118. Takaro, T., H.P. Price, S.C. Parra, and L.R. Gaddy 1979 Ultrastructural studies of apertures in the interalveolar septum of the adult human lung. Am. Rev. Respir. Dis., 119t425-434. Ten Have-Opbroek, A.A.W., J.A. Dubbeldam, and C.J.M. Otto-Verberne 1988 Ultrastructural features of type I1 alveolar cells in early embryonic mouse lung. Anat. Rec., 221r846-853. Uhlig, S., F. Brasch, L. Wollin, H. Fehrenbach, J . Richter, and A. Wendel 1995 Functional and fine structural changes in isolated rat lungs challenged with endotoxin ex Vivo and in Vitro. Am. J. Pathol.; 146.1235-1247. Van Golde, L.M.G., J.J. Batenburg, and B. Robertson 1994 The pulmonary surfactant system. NIPS, 9:13-20. Veldhuizen, R.A.W., J . Lee, D. Sandler, W. Hull, J.A. Whitsett, J . Lewis, F. Possmayer, and R.J. Novick 1993 Alterations in pulmonary surfactant composition and activity after experimental lung transplantation. Am. Rev. Respir. Dis., 148:208-215. VidiC, B., and P.H. Burri 1981 Quantitative cellular and subcellular changes in the rat type I1 pneumocyte during early postnatal development. Am. Rev. Respir. Dis., 124:174-178. Vincent, R., and D. Nadeau 1987 Cytoplasmic bar-like structures of alveolar type I1 cells: An ultrastructural study in freshly isolated cells from rat lungs. Am. J . Anat., 179:70-78. Voorhout, W.F., T.E. Weaver, H.P. Haagsman, H.J. Geuze, and L.M.G. Van Golde 1993 Biosynthetic routing of pulmonary surfactant proteins in alveolar type I1 cells. Microsc. Res. Tech., 26:366-373. Wahlers, T., A. Haverich, H.G. Fieguth, H.-J. Schafers, T. Takayama, and H.G. Borst 1986 Flush perfusion using Euro-Collins solution vs. cooling by means of extracorporeal circulation in heart-lung preservation. J . Heart Transplant., 5:89-98. Wahlers, T., H.-J. Schafers, J . Cremer, M. Jurmann, S.W. Hirt, H.G. Fieguth, and A. Haverich 1991 Organ preservation for heart-lung and lung transplantation. Thorac. Cardiovasc. Surg., 39t344348. Weibel, E.R. 1963 Morphometry of the Human Lung. Springer Verlag, Heidelberg. Weibel, E.R. 1979a Stereological Methods. Practical Methods for Biological Morphometry, Vols. 1, 2. Academic Press, New York. Weibel, E.R. 1979b Morphometry of the human lung: The state of the art after two decades. Bull. Eur. Physiopath. Respir., 15:9991013. Weibel, E.R. 1990 Morphometry: Stereological theory and practical methods. In: Models of Lung Disease. Microscopy and Structural Methods. J . Gil, ed. Marcel Dekker Inc., New York, pp. 199-252. Weibel, E.R., and C.R. Taylor 1988 Design and structure of the human lung. In: Pulmonary Diseases and Disorders. A.P. Fishman, ed. McGraw-Hill Book Co., New York, pp. 11-60. Williams, M.C. 1987 Vesicles within vesicles: What role do multivesicular bodies play in alveolar type I1 cells?. Am. Rev. Respir. Dis., 135:744-746. Wright, J.R., and L.G. Dobbs 1991 Regulation of pulmonary surfactant secretion and clearance. Annu. Rev. Physiol., 53:395-414. Young, S.E., E.K. Fram, and B.L. Craig 1985 Three-dimensional reconstruction and quantitative analysis of rat lung type I1 cells: A computer-based study. Am. J . Anat., 174:l-14. Young, S.L., S.A. Kremers, J.S. Apple, J.D. Crapo, and G.W. Brumley 1981 Rat lung surfactant kinetics: Biochemical and morphometric correlation. J . Appl. Physiol., 51~248-253. Young, S.L., E.K. Fram, C.L. Spain, and E.W. Larson 1991 Development of type I1 pneumocytes in rat lung. Am. J . Physiol., 260: L113-Ll22. Young, S.L., E.K. Fram, E. Larson, and J.R. Wright 1993 Recycling of surfactant lipid and apoprotein-A studied by electron microscopic autoradiography. Am. J . Physiol., 265tL19-L26. Zeltner, T.B., J.H. Carduff, P. Gehr, J. Pfenninger, and P.H. Burri 1987 The postnatal development and growth of the human lung. I. Morphometry. Respir. Physiol., 67t247-267.