- Email: [email protected]
Occlusion of Pulmonary Vessels by Megakaryocytes after Treatment with Tumour Necrosis Factor-alpha (TNF-α)
J. Comp. Path. 1999 Vol. 120, 235–245
Occlusion of Pulmonary Vessels by Megakaryocytes after Treatment with Tumour Necrosis Factor-alpha (TNF-a) S. Sulkowski, S. Terlikowski∗ and M. Sulkowska Department of Pathological Anatomy and ∗Department of Gynaecology and Septic Obstetrics, Medical University of Bialystok, ul. Waszyngtona 13, PL 15-269 Bialystok 8, Poland Summary Pulmonary thrombosis in the course of shock remains life-threatening, despite advances in diagnosis, prophylaxis and therapy of the disease. Tumour necrosis factor-alpha (TNF-a) is an important mediator of shock. The aim of this study was to analyze the morphological changes in the pulmonary capillary bed in rats after intraperitoneal administration of multiple doses of TNF-a (10 lg TNF-a/24 h for 5 days; biological activity of 2–4×107 U/ mg of protein). Morphological investigations were undertaken by light and transmission electron microscopy with emphasis on pulmonary thrombopoiesis. The study confirmed that the lungs may be an important site of extramedullary thrombopoiesis in the course of shock. The observations also suggested that megakaryocytes shed large fragments of cytoplasm within the pulmonary capillary bed and that megakaryocytes with copious cytoplasm occlude pulmonary vessels. 1999 W.B. Saunders Company Limited
Introduction Tumour necrosis factor-alpha (TNF-a) is an important mediator of shock in man and animals, and plays a significant role in the pathogenesis of adult respiratory distress syndrome (ARDS) (Zabel and Schade, 1994). Given to animals, it produces signs resembling those of septic shock and leads to acute lung injury with respiratory insufficiency and death. The mechanisms of TNFa-induced shock syndrome include increased lung capillary permeability, activation of neutrophils and endothelial cells, and accumulation of inflammatory cells in the pulmonary capillary bed. The toxic action of TNF-a on the lungs is associated also with damage to the surfactant and type II alveolar epithelial cells (Sulkowska et al., 1996, 1997; Sulkowska, 1997). On post-mortem examination, ARDS patients show oedematous lungs with fluid in the alveolar spaces, and capillaries that frequently contain clumps of neutrophils, platelets and fibrin clots. Platelet accumulation in lung capillaries provides favourable conditions for the formation of platelet thrombi (Zabel and Schade, 1994). Pulmonary thrombosis observed in shock is a serious complication which frequently leads to death. Local formation of blood platelets (in the lung capillary bed) may be of significance in the development of pulmonary thrombosis. There are two basic theories of thrombopoiesis: medullary and pulmonary. In the bone marrow, platelets are formed via the 0021–9975/99/030235+11 $12.00/0
1999 W.B. Saunders Company Limited
236
S. Sulkowski et al.
fragmentation of cytoplasmic processes of megakaryocytes (MKs). Wright (1910) was the first to draw attention to the bone marrow as a site of platelet formation. His theory is still commonly accepted. However, in 1893 Aschoff had demonstrated the occurrence of MKs in the kidneys, the liver, the heart, and particularly in the pulmonary vessels. Other studies have revealed that the occurrence of MKs in the vascular bed of the lungs and in the peripheral blood is a physiological phenomenon (Levine et al., 1990, 1993) which may be exaggerated in certain pathological conditions (Hume et al., 1964; Aabo and Hansen, 1978). A large increase in the number of MKs in the lungs is observed in human patients who die of disseminated intravascular coagulation (DIC), acute infections, haemorrhage, certain types of neoplasm, liver failure or shock (Aabo and Hansen, 1978). Breslow et al. (1968) found a considerable increase in the number of MKs and platelets in the peripheral venous blood in post-operative patients. The theory of pulmonary thrombopoiesis, both in physiological conditions and in shock, has been well documented (Levine et al., 1993); there are, however, few reports on morphological evaluation of the megakaryocyteplatelet system within the lung capillary bed. The aim of the present study was to examine the lung capillary bed after TNF-a treatment, with special attention to changes associated with pulmonary thrombopoiesis. Materials and Methods Experimental Animals The study was made with 48 male Wistar rats of 180–220 g body weight. The animals were maintained in a well-lit room at 18–20°C and fed a standard granulated diet containing cysteine 0·55% and methionine 0·55%. All procedures were in strict accordance with guidance on the care and use of laboratory animals and were approved by the local Animal Care Committee. Experimental Design The rats were divided into two main groups, I (experimental) and II (controls), each of 24 animals. Group I rats were given TNF-a intraperitoneally at a dose of 10 lg in 0.5 ml phosphate-buffered saline (PBS), daily for 5 days. Human recombinant TNFa, biological activity 2–4×107 U/mg of protein, was kindly provided by Prof. W. J. Stec, Department of Bioorganic Chemistry, Polish Academy of Sciences, Ło´ dz´, Poland (Tcho´rzewski et al., 1993; Terlikowski et al., 1996, 1997; Sulkowska, 1997). Group II rats were similarly treated with PBS alone. The rats in both main groups were killed (sodium pentobarbital 100 mg, intraperitoneally) in subgroups of six on day 1, 7, 14 or 28 after the second dose of TNF-a or PBS. The subgroups were labelled I(or II)1, I(or II)-7, I(or II)-14 and I(or II)-28. Collection of Samples for Analysis The lungs of all six animals of each subgroup were examined by light microscopy. The lungs were cut horizontally from the edges to the hilum and two samples from each lobe were fixed in 10% neutral buffered formalin (NBF) and paraffin waxembedded to prepare sections for histological analysis. The sections were stained with haematoxylin and eosin (HE), impregnated with silver salts according to Gomori, and then examined for several parameters, including: pulmonary congestion or oedema, inflammatory infiltration and fibrosis, and the presence of megakaryocytes or naked
TNF-a and Pulmonary Thrombopoiesis
237
cellular nuclei (or both) in the lung capillary bed. The histological (and ultrastructural; see below) parameters were graded from +++ to −, as follows: +++, changes (fibrosis of alveolar septa, congestion, oedema, endothelial damage) were found in all animals of a subgroup, and in all or most of the preparations examined; ++, the changes occurred in all animals, but not in all samples examined; +, the changes noted only in some of the animals of a given subgroup; +/−, morphological changes were rare; −, no changes. Identical criteria and grading were used for the semiquantitative evaluation of the numbers of cells (megakaryocytes, platelets, neutrophils, monocytes) and fragments of MKs (naked nuclei and fragments of MK cytoplasm). The lungs of four animals of each experimental subgroup (I-1, I-7, I-14 and I-28) and two animals of each control subgroup (II-1, II-7, II-14 and II-28) were subjected to analysis by transmission electron microscopy (TEM). Samples for TEM were collected immediately after the removal of the lungs from the thorax, before fixation in NBF. Two blocks (1 mm3) were taken from the parahilar and central parts of all lobes of both lungs and fixed in cold 2·5% glutaraldehyde solution and 1% osmium tetroxide. After dehydration in an alcohol-acetone series and embedding in epoxy resin, the blocks were sectioned and contrast was given with lead citrate and uranyl acetate, before examination with an Opton PC 900 transmission electron microscope. Only those samples in which analysis of semithin toluidine blue-stained sections suggested the presence of megakaryocytes or their fragments were subjected to detailed ultrastructural analysis.
Results Histological Observations The results of examination of the lungs are shown in Table 1 and Figs 1–3. The lungs of control animals in all subgroups of group II appeared normal. Destructive and exudative changes dominated in experimental subgroup I-1. The walls of the thickened interalveolar septa were infiltrated with inflammatory cells. Numerous naked nuclei were observed in the lung capillary bed (Fig. 1A, B). As in subgroup I-7 (Fig. 2), MKs (with a distinct cytoplasmic margin), which occluded the lumen of small vessels, were rare. Subgroups I7 and I-14 showed less severe oedematous changes and less pronounced endothelial damage; moreover, the accumulation of monocytes, particularly of neutrophils in the lung capillary bed, was less striking. After 4 weeks (subgroup I-28), tissue rebuilding was noted, with thinning and focal atrophy of the interalveolar septa and dilation of air spaces (Fig. 3). The changes described above corresponded to panlobular emphysema, but changes typical of centrilobular emphysema were also seen. In the vicinity of atrophic changes in the lung tissue, thickening of the interalveolar septa was observed in places and foci of lung parenchymal atelectasis were found nearby. In subgroup I28, naked nuclei and MKs were absent from the lung capillary bed. Ultrastructural Observations In all control subgroups, type II alveolar epithelial cells showed a varying number of typically-structured lamellar bodies. Short microvilli were observed on the surface of the type II cells. Occasionally, the alveolar lumen showed
TEM +/− + ++ ++ +++ + ++ −
LM + ++ n n +++ ++ n −
I-1
+ ++ n n ++ + n −
LM
I-7
+/− + + + ++ + + +
TEM +/− + n n + +/− n +
LM
I-14
− − + − + − +/− ++
TEM
− + n n + +/− n ++
LM
TEM − − +/− − + − − +++
I-28
Occurrence and severity of the changes observed by LM and TEM in the stated rat subgroups
For grading system (+++ to −) see Materials and Methods. MK, megakaryocyte; l.c.b., lung capillary bed; LM, light microscopy; TEM, transmission electron microscopy; n, not done.
MKs with copious cytoplasm in l.c.b. Naked nuclei in l.c.b. Platelets in l.c.b. Fragments of MK cytoplasm in l.c.b. Neutrophils and/or monocytes in l.c.b. Congestion and/or oedema Endothelial damage Fibrosis of alveolar septa
Changes
Table 1 Results of morphological examination of the lungs
− − n n +/− − n −
LM
− − +/− − +/− − − −
TEM
II-1, 7, 14, 28
238 S. Sulkowski et al.
TNF-a and Pulmonary Thrombopoiesis
Fig. 1. Fig. 2. Fig. 3.
239
Naked nuclei of MKs (arrows) observed in subgroup I-1 of TNF-a-treated animals. (A) Thickened alveolar septa infiltrated with inflammatory cells. (B) Oedematous fluid (oe) in the alveolar lumen. HE. ×240 (A); ×320 (B). Megakaryocyte (MK—arrow) with copious cytoplasm and naked nuclei (arrow). Subgroup I-7. HE. ×240. Fragment of emphysema-like lung parenchyma. Subgroup I-28. HE. ×80.
240
Fig. 4.
S. Sulkowski et al.
The vascular lumen is filled with neutrophils (N) and monocytes (M) which adhere to damaged endothelium (arrows). The alveolar lumen shows a fragment of an alveolar macrophage (AM) with numerous secondary lysosomes. Subgroup I-7. TEM. ×3000.
alveolar macrophages containing a small number of secondary lysosomes. The vascular lumen showed erythrocytes and single leukocytes. In experimental subgroup I-1, the thickened interalveolar septa were infiltrated with inflammatory cells (Fig. 4). Desquamated type II cells and macrophages containing lamellar structures and oedematous fluid were found in the alveolar lumina of this subgroup. Numerous neutrophils and monocytes observed in the vascular lumen adhered to damaged endothelium and totally occluded the lumen of tiny vessels. Some of the blood vessels showed fragments of MK cytoplasm and numerous platelets. Naked cellular nuclei, common in light microscopical preparations, were rare in the ultrastructural sections (Fig. 5). In subgroups I-7 and I-14, ultrastructural examination revealed more extensive damage to type II alveolar epithelial cells. However, oedematous changes were less pronounced. The vascular lumen showed focal accumulation of platelets and also neutrophils and monocytes, but these cells were rarely observed simultaneously in the lung capillary bed. Vessels with numerous granulocytes or monocytes sporadically exhibited accumulations of platelets. At the same time blood vessels which showed platelet accumulation or the presence of larger fragments of megakaryocytic cytoplasm exhibited only focal accumulations of granulocytes or monocytes (Fig. 6). As in subgroup I-1, endothelial cells of some of the blood vessels showed features of damage, and granulocytes or monocytes fused with endothelial cells. The features of endothelial damage were relatively rare in blood vessels with accumulations
TNF-a and Pulmonary Thrombopoiesis
Fig. 5. Fig. 6.
Fig. 7.
241
Megakaryocyte nucleus with narrow cytoplasmic margin. By light microscopy this would appear as a naked nucleus. Subgroup I-1. TEM. ×4400. Fragment of megakaryocyte cytoplasm (Mk-f ). In its vicinity single platelets (pl) are visible; they do not adhere to oedematous endothelium (oe). Subgroup I-7. TEM. ×7000.
A megakaryocyte with copious cytoplasm blocks the vascular lumen. Fragments of cells are fused with endothelium in places (arrows). Subgroup I-7. TEM. ×4400.
242
Fig. 8. Fig. 9.
S. Sulkowski et al.
The lumen of a medium-sized vessel is filled with numerous loosely lying platelets (pl). Subgroup I-14. TEM. ×7000. Fragment of fibrotic lung parenchyma with numerous collagen aggregates (c). The vascular lumen shows an undamaged nucleus (arrow) without cytoplasmic margin. Subgroup I-28. TEM. ×7000.
of platelets or MK fragments. In two cases, whole MKs blocked the vascular lumen (Fig. 7). The changes described were less severe in subgroup I-14 than in I-7. In subgroup I-28, pulmonary tissue architecture rebuilding was observed, with thinning and focal atrophy of the interalveolar septa, and dilation of air spaces. In the vicinity of these atrophic changes, thickening of the alveolar septa was seen and focal atelectasis occurred nearby. Ultrastructural examination showed that fibroplastic processes predominated within these fragments of the pulmonary tissue (Fig. 9). The vascular lumina sporadically showed platelet accumulation and naked cellular nuclei, which may have originated from MK disintegration (Figs 8, 9). Discussion Pulmonary vascular injury is a central feature of the adult respiratory distress syndrome (ARDS). In the early stage of ARDS, acute endothelial injury has been observed ultrastructurally within 1 day of the onset of symptoms (Schnells et al., 1980). Histologically, intense haemorrhage and oedema are seen to be associated with intracapillary engorgement, microvascular thrombosis and thromboemboli (Tomashefski et al., 1983). The morphological types of thromboemboli—macrothrombi, large microthrombi and capillary thrombi—have been described by Eeles and Sevitt (1967) in burned and traumatized patients. As in that study, capillary microthrombi are most numerous in the early stage
TNF-a and Pulmonary Thrombopoiesis
243
in ARDS (Tomashefski et al., 1983). Damage to vascular endothelium and platelet accumulation in the lung capillary bed provide favourable conditions for the formation of platelet thrombi observed in a number of lung tissue injuries, particularly in ARDS. Too little attention has been paid, however, to the problem of local (pulmonary) thrombopoiesis and its significance in platelet thrombus formation. Most MKs released from the bone marrow pass in the blood to the lungs. In the terminal segments of the pulmonary microcirculation, the vascular diameter is much smaller than the MK diameter. Trowbridge et al. (1982) and Trowbridge (1988), taking this into account, indicated three possible outcomes, namely MK disintegration, MK deformation, or vessel occlusion by MKs. There has been no previous documentation of lung vessel occlusion by entire MKs with copious cytoplasm. However, our earlier studies on the effect of cyclophosphamide on pulmonary thrombopoiesis confirmed such a possibility (Sulkowski et al., 1998). The present study provides histological and ultrastructural evidence of luminal occlusion by “whole” MKs. However, it should be noted that MKs with copious cytoplasm were rarely observed in the pulmonary vascular lumina. Fragments of MK cytoplasm or naked cellular nuclei were much more common. The study clearly demonstrated MK disintegration and the formation of numerous blood platelets in the lung capillary bed. If such disintegration occurs in vessels with endothelial damage, conditions are provided for thrombus formation. It is of interest that we did not observe typical platelet thrombi, and vessels with large numbers of platelets had only slightly changed endothelium. Possibly the increased release of MKs from the marrow occurred later than endothelial damage. That interpretation would be supported by the presence of neutrophils or monocytes in vessels with severe endothelial damage, which might have caused occlusion (before the increase in platelet count in the lungs). The possibility cannot be excluded, however, that the more pronounced endothelial damage in the vessels with neutrophil accumulation was caused by neutrophils themselves. This point requires further study. It is not clear what happens to the naked nuclei (NN) of megakaryocytes. Most authors have assumed that NN pass to the arterial circulation. Levine et al. (1993) presented illustrations of NN isolated from the arterial and venous blood; the diameter of the NN was estimated as 16.4–22.0 lm in the blood from pulmonary arteries and 16.0–23.9 lm in the aortic blood. Thus, the diameter of NN is much larger than that of the pulmonary capillaries of the interalveolar septa, and NN could not pass through such vessels without considerable deformation. The increased number of NN observed in the lungs in certain shock conditions may reflect the impeded passage of NN through the vessels of the interalveolar septa. Thus, the lung seems to act as a specific filter, capable of trapping some of the NN. This view is also supported by the studies of Levine et al. (1993), who found more NN in blood from the pulmonary artery(9·2/ml) than in blood from the aorta (6·8/ml, P<0·001). What happens to the NN that settle in the lung capillary bed is not known. They may disintegrate and pass through the pulmonary vessels. This is suggested by the
244
S. Sulkowski et al.
oval-shaped and relatively small nuclei of unknown origin sporadically observed in our study. In conclusion, our observations confirm the hypothesis that MKs shed platelets and larger fragments of cytoplasm within the pulmonary capillary bed and provide evidence that MKs with copious cytoplasm can occlude lung vessels. We also believe that the experimental model presented here may be useful for studies of pulmonary thrombopoiesis. References Aabo, K. and Hansen, K. B. (1978). Megakaryocytes in pulmonary vessels. I. Incidence at autopsy, clinicopathological relations especially to disseminated intravascular coagulation. Acta Pathologica Microbiologica Scandinavica, Section A, 86, 285–291. ¨ ber capillare Embolie von riesenkernhaltigen Zellen. Virchow’s Aschoff, L. (1893). U Archiv fu¨r Pathologische Anatomie und Physiologie, 134, 11–26. Breslow, A., Kaufman, R. M. and Lawsky, A. R. (1968). The effect of surgery on the concentration of circulating megakaryocytes and platelets. Blood, 32, 393–401. Eeles, G. H. and Sevitt S. (1967). Microthombosis in injured and burned patients. Journal of Pathology and Bacteriology, 93, 275–293. Hume, R., West, J. T., Malmgren, R. A. and Chu, E. A. (1964). Quantitative observations of circulating megakaryocytes in the blood of patients with cancer. New England Journal of Medicine, 270, 111–117. Levine, R. F., Eldor, A., Shoff, P. K., Kirwin, S., Tenza, D. and Cramer, E. M. (1993). Circulating megakaryocytes: delivery of large numbers of intact, mature megakaryocytes to the lungs. European Journal of Haematology, 51, 233–246. Levine, R. F., Shoff, P., Han, Z. C. and Eldor, A. (1990). Circulating megakaryocytes and platelet production in the lungs. In: Molecular Biology and Differentiation of Megakaryocytes, J. Breton-Gorius, J. Levin, A. T. Nurden and N. Williams, Eds, Wiley-Liss, New York, pp. 41–52. Schnells, G., Voight, W. H., Redl, H., Schlag, G. and Glatzl, A. (1980). Electron microscopic investigation of lung biopsies in patients with post traumatic respiratory insufficiency. Acta Chirurgia Scandinavica, 449, 9–20. Sulkowska, M. (1997). Effect of human recombinant tumour necrosis factor-a and pentoxifyllin on the ultrastructure of type II alveolar epithelial cells in pregnant and non-pregnant rabbits. Journal of Comparative Pathology, 117, 227–236. Sulkowska, M., Sulkowski, S., Nowak, H. F. and Terlikowski, S. (1996). Type II alveolar epithelial cells and free alveolar cells after intratumor TNF-a administration. Histology and Histopathology, 11, 633–640. Sulkowska, M., Sulkowski, S., Terlikowski, S. and Nowak, H. F. (1997). Ultrastructural analysis of pulmonary capillaries after mutein VI rhTNF-a administration into Morris hepatoma. Folia Histochemica et Cytobiologica, 35, 83–84. Sulkowski, S., Sulkowska, M. and Musiatowicz, B. (1998). The effects of cyclophosphamide on pulmonary thrombopoiesis in rats. Histology and Histopathology, 13, 1027–1036. Tcho´rzewski, H., Zeman, K., Paleolog, E., Brennan, F., Feld-Mann, M., Kahan, M., Guga, P., Kwinkowski, M., Szyman´ska, B., Jarosz, J., Parniewski, P. and Kocur, E. (1993). The effects of tumor necrosis factor (TNF) derivatives on TNF receptors. Cytokine, 5, 125–132. Terlikowski, S., Nowak, H. F., Sulkowska, M. and Sulkowski, S. (1996). The effect of local TNF-a administration upon the spontaneous lung metastases in rats with Morris 5123 hepatoma. Neoplasma, 43, 327–332. Terlikowski, S., Sulkowski, S. and Nowak, H. F. (1997). Regression of Morris hepatoma in response to intralesional treatment with tumor necrosis factor muteins. European Cytokine Network, 8, 259–263.
TNF-a and Pulmonary Thrombopoiesis
245
Tomashefski, J. F., Davies, P., Boggis, C., Greene, R., Zapol, W. M. and Reid, L. M. (1983). The pulmonary vascular lesions of the adult respiratory distress syndrome. American Journal of Pathology, 112, 112–126. Trowbridge, E. A. (1988). Pulmonary platelet production: a physical analogue of mitosis? Blood Cells, 13, 451–458. Trowbridge, E. A., Martin, J. F. and Slater, D. N. (1982). Evidence for a theory of physical fragmentation of megakaryocytes, implying that all platelets are produced in the pulmonary circulation. Thrombosis Research, 28, 461–475. Wright, J. H. (1910). The histogenesis of platelets. Journal of Morphology, 21, 263–269. Zabel, P. and Schade, F. U. (1994). Pentoxifylline and tumour necrosis factor-induced lung injury. European Respiratory Journal, 7, 1389–1391.
Received, June 17th, 1998 Accepted, October 12th, 1998
Occlusion of Pulmonary Vessels by Megakaryocytes after Treatment with Tumour Necrosis Factor-alpha (TNF-a) S. Sulkowski, S. Terlikowski∗ and M. Sulkowska Department of Pathological Anatomy and ∗Department of Gynaecology and Septic Obstetrics, Medical University of Bialystok, ul. Waszyngtona 13, PL 15-269 Bialystok 8, Poland Summary Pulmonary thrombosis in the course of shock remains life-threatening, despite advances in diagnosis, prophylaxis and therapy of the disease. Tumour necrosis factor-alpha (TNF-a) is an important mediator of shock. The aim of this study was to analyze the morphological changes in the pulmonary capillary bed in rats after intraperitoneal administration of multiple doses of TNF-a (10 lg TNF-a/24 h for 5 days; biological activity of 2–4×107 U/ mg of protein). Morphological investigations were undertaken by light and transmission electron microscopy with emphasis on pulmonary thrombopoiesis. The study confirmed that the lungs may be an important site of extramedullary thrombopoiesis in the course of shock. The observations also suggested that megakaryocytes shed large fragments of cytoplasm within the pulmonary capillary bed and that megakaryocytes with copious cytoplasm occlude pulmonary vessels. 1999 W.B. Saunders Company Limited
Introduction Tumour necrosis factor-alpha (TNF-a) is an important mediator of shock in man and animals, and plays a significant role in the pathogenesis of adult respiratory distress syndrome (ARDS) (Zabel and Schade, 1994). Given to animals, it produces signs resembling those of septic shock and leads to acute lung injury with respiratory insufficiency and death. The mechanisms of TNFa-induced shock syndrome include increased lung capillary permeability, activation of neutrophils and endothelial cells, and accumulation of inflammatory cells in the pulmonary capillary bed. The toxic action of TNF-a on the lungs is associated also with damage to the surfactant and type II alveolar epithelial cells (Sulkowska et al., 1996, 1997; Sulkowska, 1997). On post-mortem examination, ARDS patients show oedematous lungs with fluid in the alveolar spaces, and capillaries that frequently contain clumps of neutrophils, platelets and fibrin clots. Platelet accumulation in lung capillaries provides favourable conditions for the formation of platelet thrombi (Zabel and Schade, 1994). Pulmonary thrombosis observed in shock is a serious complication which frequently leads to death. Local formation of blood platelets (in the lung capillary bed) may be of significance in the development of pulmonary thrombosis. There are two basic theories of thrombopoiesis: medullary and pulmonary. In the bone marrow, platelets are formed via the 0021–9975/99/030235+11 $12.00/0
1999 W.B. Saunders Company Limited
236
S. Sulkowski et al.
fragmentation of cytoplasmic processes of megakaryocytes (MKs). Wright (1910) was the first to draw attention to the bone marrow as a site of platelet formation. His theory is still commonly accepted. However, in 1893 Aschoff had demonstrated the occurrence of MKs in the kidneys, the liver, the heart, and particularly in the pulmonary vessels. Other studies have revealed that the occurrence of MKs in the vascular bed of the lungs and in the peripheral blood is a physiological phenomenon (Levine et al., 1990, 1993) which may be exaggerated in certain pathological conditions (Hume et al., 1964; Aabo and Hansen, 1978). A large increase in the number of MKs in the lungs is observed in human patients who die of disseminated intravascular coagulation (DIC), acute infections, haemorrhage, certain types of neoplasm, liver failure or shock (Aabo and Hansen, 1978). Breslow et al. (1968) found a considerable increase in the number of MKs and platelets in the peripheral venous blood in post-operative patients. The theory of pulmonary thrombopoiesis, both in physiological conditions and in shock, has been well documented (Levine et al., 1993); there are, however, few reports on morphological evaluation of the megakaryocyteplatelet system within the lung capillary bed. The aim of the present study was to examine the lung capillary bed after TNF-a treatment, with special attention to changes associated with pulmonary thrombopoiesis. Materials and Methods Experimental Animals The study was made with 48 male Wistar rats of 180–220 g body weight. The animals were maintained in a well-lit room at 18–20°C and fed a standard granulated diet containing cysteine 0·55% and methionine 0·55%. All procedures were in strict accordance with guidance on the care and use of laboratory animals and were approved by the local Animal Care Committee. Experimental Design The rats were divided into two main groups, I (experimental) and II (controls), each of 24 animals. Group I rats were given TNF-a intraperitoneally at a dose of 10 lg in 0.5 ml phosphate-buffered saline (PBS), daily for 5 days. Human recombinant TNFa, biological activity 2–4×107 U/mg of protein, was kindly provided by Prof. W. J. Stec, Department of Bioorganic Chemistry, Polish Academy of Sciences, Ło´ dz´, Poland (Tcho´rzewski et al., 1993; Terlikowski et al., 1996, 1997; Sulkowska, 1997). Group II rats were similarly treated with PBS alone. The rats in both main groups were killed (sodium pentobarbital 100 mg, intraperitoneally) in subgroups of six on day 1, 7, 14 or 28 after the second dose of TNF-a or PBS. The subgroups were labelled I(or II)1, I(or II)-7, I(or II)-14 and I(or II)-28. Collection of Samples for Analysis The lungs of all six animals of each subgroup were examined by light microscopy. The lungs were cut horizontally from the edges to the hilum and two samples from each lobe were fixed in 10% neutral buffered formalin (NBF) and paraffin waxembedded to prepare sections for histological analysis. The sections were stained with haematoxylin and eosin (HE), impregnated with silver salts according to Gomori, and then examined for several parameters, including: pulmonary congestion or oedema, inflammatory infiltration and fibrosis, and the presence of megakaryocytes or naked
TNF-a and Pulmonary Thrombopoiesis
237
cellular nuclei (or both) in the lung capillary bed. The histological (and ultrastructural; see below) parameters were graded from +++ to −, as follows: +++, changes (fibrosis of alveolar septa, congestion, oedema, endothelial damage) were found in all animals of a subgroup, and in all or most of the preparations examined; ++, the changes occurred in all animals, but not in all samples examined; +, the changes noted only in some of the animals of a given subgroup; +/−, morphological changes were rare; −, no changes. Identical criteria and grading were used for the semiquantitative evaluation of the numbers of cells (megakaryocytes, platelets, neutrophils, monocytes) and fragments of MKs (naked nuclei and fragments of MK cytoplasm). The lungs of four animals of each experimental subgroup (I-1, I-7, I-14 and I-28) and two animals of each control subgroup (II-1, II-7, II-14 and II-28) were subjected to analysis by transmission electron microscopy (TEM). Samples for TEM were collected immediately after the removal of the lungs from the thorax, before fixation in NBF. Two blocks (1 mm3) were taken from the parahilar and central parts of all lobes of both lungs and fixed in cold 2·5% glutaraldehyde solution and 1% osmium tetroxide. After dehydration in an alcohol-acetone series and embedding in epoxy resin, the blocks were sectioned and contrast was given with lead citrate and uranyl acetate, before examination with an Opton PC 900 transmission electron microscope. Only those samples in which analysis of semithin toluidine blue-stained sections suggested the presence of megakaryocytes or their fragments were subjected to detailed ultrastructural analysis.
Results Histological Observations The results of examination of the lungs are shown in Table 1 and Figs 1–3. The lungs of control animals in all subgroups of group II appeared normal. Destructive and exudative changes dominated in experimental subgroup I-1. The walls of the thickened interalveolar septa were infiltrated with inflammatory cells. Numerous naked nuclei were observed in the lung capillary bed (Fig. 1A, B). As in subgroup I-7 (Fig. 2), MKs (with a distinct cytoplasmic margin), which occluded the lumen of small vessels, were rare. Subgroups I7 and I-14 showed less severe oedematous changes and less pronounced endothelial damage; moreover, the accumulation of monocytes, particularly of neutrophils in the lung capillary bed, was less striking. After 4 weeks (subgroup I-28), tissue rebuilding was noted, with thinning and focal atrophy of the interalveolar septa and dilation of air spaces (Fig. 3). The changes described above corresponded to panlobular emphysema, but changes typical of centrilobular emphysema were also seen. In the vicinity of atrophic changes in the lung tissue, thickening of the interalveolar septa was observed in places and foci of lung parenchymal atelectasis were found nearby. In subgroup I28, naked nuclei and MKs were absent from the lung capillary bed. Ultrastructural Observations In all control subgroups, type II alveolar epithelial cells showed a varying number of typically-structured lamellar bodies. Short microvilli were observed on the surface of the type II cells. Occasionally, the alveolar lumen showed
TEM +/− + ++ ++ +++ + ++ −
LM + ++ n n +++ ++ n −
I-1
+ ++ n n ++ + n −
LM
I-7
+/− + + + ++ + + +
TEM +/− + n n + +/− n +
LM
I-14
− − + − + − +/− ++
TEM
− + n n + +/− n ++
LM
TEM − − +/− − + − − +++
I-28
Occurrence and severity of the changes observed by LM and TEM in the stated rat subgroups
For grading system (+++ to −) see Materials and Methods. MK, megakaryocyte; l.c.b., lung capillary bed; LM, light microscopy; TEM, transmission electron microscopy; n, not done.
MKs with copious cytoplasm in l.c.b. Naked nuclei in l.c.b. Platelets in l.c.b. Fragments of MK cytoplasm in l.c.b. Neutrophils and/or monocytes in l.c.b. Congestion and/or oedema Endothelial damage Fibrosis of alveolar septa
Changes
Table 1 Results of morphological examination of the lungs
− − n n +/− − n −
LM
− − +/− − +/− − − −
TEM
II-1, 7, 14, 28
238 S. Sulkowski et al.
TNF-a and Pulmonary Thrombopoiesis
Fig. 1. Fig. 2. Fig. 3.
239
Naked nuclei of MKs (arrows) observed in subgroup I-1 of TNF-a-treated animals. (A) Thickened alveolar septa infiltrated with inflammatory cells. (B) Oedematous fluid (oe) in the alveolar lumen. HE. ×240 (A); ×320 (B). Megakaryocyte (MK—arrow) with copious cytoplasm and naked nuclei (arrow). Subgroup I-7. HE. ×240. Fragment of emphysema-like lung parenchyma. Subgroup I-28. HE. ×80.
240
Fig. 4.
S. Sulkowski et al.
The vascular lumen is filled with neutrophils (N) and monocytes (M) which adhere to damaged endothelium (arrows). The alveolar lumen shows a fragment of an alveolar macrophage (AM) with numerous secondary lysosomes. Subgroup I-7. TEM. ×3000.
alveolar macrophages containing a small number of secondary lysosomes. The vascular lumen showed erythrocytes and single leukocytes. In experimental subgroup I-1, the thickened interalveolar septa were infiltrated with inflammatory cells (Fig. 4). Desquamated type II cells and macrophages containing lamellar structures and oedematous fluid were found in the alveolar lumina of this subgroup. Numerous neutrophils and monocytes observed in the vascular lumen adhered to damaged endothelium and totally occluded the lumen of tiny vessels. Some of the blood vessels showed fragments of MK cytoplasm and numerous platelets. Naked cellular nuclei, common in light microscopical preparations, were rare in the ultrastructural sections (Fig. 5). In subgroups I-7 and I-14, ultrastructural examination revealed more extensive damage to type II alveolar epithelial cells. However, oedematous changes were less pronounced. The vascular lumen showed focal accumulation of platelets and also neutrophils and monocytes, but these cells were rarely observed simultaneously in the lung capillary bed. Vessels with numerous granulocytes or monocytes sporadically exhibited accumulations of platelets. At the same time blood vessels which showed platelet accumulation or the presence of larger fragments of megakaryocytic cytoplasm exhibited only focal accumulations of granulocytes or monocytes (Fig. 6). As in subgroup I-1, endothelial cells of some of the blood vessels showed features of damage, and granulocytes or monocytes fused with endothelial cells. The features of endothelial damage were relatively rare in blood vessels with accumulations
TNF-a and Pulmonary Thrombopoiesis
Fig. 5. Fig. 6.
Fig. 7.
241
Megakaryocyte nucleus with narrow cytoplasmic margin. By light microscopy this would appear as a naked nucleus. Subgroup I-1. TEM. ×4400. Fragment of megakaryocyte cytoplasm (Mk-f ). In its vicinity single platelets (pl) are visible; they do not adhere to oedematous endothelium (oe). Subgroup I-7. TEM. ×7000.
A megakaryocyte with copious cytoplasm blocks the vascular lumen. Fragments of cells are fused with endothelium in places (arrows). Subgroup I-7. TEM. ×4400.
242
Fig. 8. Fig. 9.
S. Sulkowski et al.
The lumen of a medium-sized vessel is filled with numerous loosely lying platelets (pl). Subgroup I-14. TEM. ×7000. Fragment of fibrotic lung parenchyma with numerous collagen aggregates (c). The vascular lumen shows an undamaged nucleus (arrow) without cytoplasmic margin. Subgroup I-28. TEM. ×7000.
of platelets or MK fragments. In two cases, whole MKs blocked the vascular lumen (Fig. 7). The changes described were less severe in subgroup I-14 than in I-7. In subgroup I-28, pulmonary tissue architecture rebuilding was observed, with thinning and focal atrophy of the interalveolar septa, and dilation of air spaces. In the vicinity of these atrophic changes, thickening of the alveolar septa was seen and focal atelectasis occurred nearby. Ultrastructural examination showed that fibroplastic processes predominated within these fragments of the pulmonary tissue (Fig. 9). The vascular lumina sporadically showed platelet accumulation and naked cellular nuclei, which may have originated from MK disintegration (Figs 8, 9). Discussion Pulmonary vascular injury is a central feature of the adult respiratory distress syndrome (ARDS). In the early stage of ARDS, acute endothelial injury has been observed ultrastructurally within 1 day of the onset of symptoms (Schnells et al., 1980). Histologically, intense haemorrhage and oedema are seen to be associated with intracapillary engorgement, microvascular thrombosis and thromboemboli (Tomashefski et al., 1983). The morphological types of thromboemboli—macrothrombi, large microthrombi and capillary thrombi—have been described by Eeles and Sevitt (1967) in burned and traumatized patients. As in that study, capillary microthrombi are most numerous in the early stage
TNF-a and Pulmonary Thrombopoiesis
243
in ARDS (Tomashefski et al., 1983). Damage to vascular endothelium and platelet accumulation in the lung capillary bed provide favourable conditions for the formation of platelet thrombi observed in a number of lung tissue injuries, particularly in ARDS. Too little attention has been paid, however, to the problem of local (pulmonary) thrombopoiesis and its significance in platelet thrombus formation. Most MKs released from the bone marrow pass in the blood to the lungs. In the terminal segments of the pulmonary microcirculation, the vascular diameter is much smaller than the MK diameter. Trowbridge et al. (1982) and Trowbridge (1988), taking this into account, indicated three possible outcomes, namely MK disintegration, MK deformation, or vessel occlusion by MKs. There has been no previous documentation of lung vessel occlusion by entire MKs with copious cytoplasm. However, our earlier studies on the effect of cyclophosphamide on pulmonary thrombopoiesis confirmed such a possibility (Sulkowski et al., 1998). The present study provides histological and ultrastructural evidence of luminal occlusion by “whole” MKs. However, it should be noted that MKs with copious cytoplasm were rarely observed in the pulmonary vascular lumina. Fragments of MK cytoplasm or naked cellular nuclei were much more common. The study clearly demonstrated MK disintegration and the formation of numerous blood platelets in the lung capillary bed. If such disintegration occurs in vessels with endothelial damage, conditions are provided for thrombus formation. It is of interest that we did not observe typical platelet thrombi, and vessels with large numbers of platelets had only slightly changed endothelium. Possibly the increased release of MKs from the marrow occurred later than endothelial damage. That interpretation would be supported by the presence of neutrophils or monocytes in vessels with severe endothelial damage, which might have caused occlusion (before the increase in platelet count in the lungs). The possibility cannot be excluded, however, that the more pronounced endothelial damage in the vessels with neutrophil accumulation was caused by neutrophils themselves. This point requires further study. It is not clear what happens to the naked nuclei (NN) of megakaryocytes. Most authors have assumed that NN pass to the arterial circulation. Levine et al. (1993) presented illustrations of NN isolated from the arterial and venous blood; the diameter of the NN was estimated as 16.4–22.0 lm in the blood from pulmonary arteries and 16.0–23.9 lm in the aortic blood. Thus, the diameter of NN is much larger than that of the pulmonary capillaries of the interalveolar septa, and NN could not pass through such vessels without considerable deformation. The increased number of NN observed in the lungs in certain shock conditions may reflect the impeded passage of NN through the vessels of the interalveolar septa. Thus, the lung seems to act as a specific filter, capable of trapping some of the NN. This view is also supported by the studies of Levine et al. (1993), who found more NN in blood from the pulmonary artery(9·2/ml) than in blood from the aorta (6·8/ml, P<0·001). What happens to the NN that settle in the lung capillary bed is not known. They may disintegrate and pass through the pulmonary vessels. This is suggested by the
244
S. Sulkowski et al.
oval-shaped and relatively small nuclei of unknown origin sporadically observed in our study. In conclusion, our observations confirm the hypothesis that MKs shed platelets and larger fragments of cytoplasm within the pulmonary capillary bed and provide evidence that MKs with copious cytoplasm can occlude lung vessels. We also believe that the experimental model presented here may be useful for studies of pulmonary thrombopoiesis. References Aabo, K. and Hansen, K. B. (1978). Megakaryocytes in pulmonary vessels. I. Incidence at autopsy, clinicopathological relations especially to disseminated intravascular coagulation. Acta Pathologica Microbiologica Scandinavica, Section A, 86, 285–291. ¨ ber capillare Embolie von riesenkernhaltigen Zellen. Virchow’s Aschoff, L. (1893). U Archiv fu¨r Pathologische Anatomie und Physiologie, 134, 11–26. Breslow, A., Kaufman, R. M. and Lawsky, A. R. (1968). The effect of surgery on the concentration of circulating megakaryocytes and platelets. Blood, 32, 393–401. Eeles, G. H. and Sevitt S. (1967). Microthombosis in injured and burned patients. Journal of Pathology and Bacteriology, 93, 275–293. Hume, R., West, J. T., Malmgren, R. A. and Chu, E. A. (1964). Quantitative observations of circulating megakaryocytes in the blood of patients with cancer. New England Journal of Medicine, 270, 111–117. Levine, R. F., Eldor, A., Shoff, P. K., Kirwin, S., Tenza, D. and Cramer, E. M. (1993). Circulating megakaryocytes: delivery of large numbers of intact, mature megakaryocytes to the lungs. European Journal of Haematology, 51, 233–246. Levine, R. F., Shoff, P., Han, Z. C. and Eldor, A. (1990). Circulating megakaryocytes and platelet production in the lungs. In: Molecular Biology and Differentiation of Megakaryocytes, J. Breton-Gorius, J. Levin, A. T. Nurden and N. Williams, Eds, Wiley-Liss, New York, pp. 41–52. Schnells, G., Voight, W. H., Redl, H., Schlag, G. and Glatzl, A. (1980). Electron microscopic investigation of lung biopsies in patients with post traumatic respiratory insufficiency. Acta Chirurgia Scandinavica, 449, 9–20. Sulkowska, M. (1997). Effect of human recombinant tumour necrosis factor-a and pentoxifyllin on the ultrastructure of type II alveolar epithelial cells in pregnant and non-pregnant rabbits. Journal of Comparative Pathology, 117, 227–236. Sulkowska, M., Sulkowski, S., Nowak, H. F. and Terlikowski, S. (1996). Type II alveolar epithelial cells and free alveolar cells after intratumor TNF-a administration. Histology and Histopathology, 11, 633–640. Sulkowska, M., Sulkowski, S., Terlikowski, S. and Nowak, H. F. (1997). Ultrastructural analysis of pulmonary capillaries after mutein VI rhTNF-a administration into Morris hepatoma. Folia Histochemica et Cytobiologica, 35, 83–84. Sulkowski, S., Sulkowska, M. and Musiatowicz, B. (1998). The effects of cyclophosphamide on pulmonary thrombopoiesis in rats. Histology and Histopathology, 13, 1027–1036. Tcho´rzewski, H., Zeman, K., Paleolog, E., Brennan, F., Feld-Mann, M., Kahan, M., Guga, P., Kwinkowski, M., Szyman´ska, B., Jarosz, J., Parniewski, P. and Kocur, E. (1993). The effects of tumor necrosis factor (TNF) derivatives on TNF receptors. Cytokine, 5, 125–132. Terlikowski, S., Nowak, H. F., Sulkowska, M. and Sulkowski, S. (1996). The effect of local TNF-a administration upon the spontaneous lung metastases in rats with Morris 5123 hepatoma. Neoplasma, 43, 327–332. Terlikowski, S., Sulkowski, S. and Nowak, H. F. (1997). Regression of Morris hepatoma in response to intralesional treatment with tumor necrosis factor muteins. European Cytokine Network, 8, 259–263.
TNF-a and Pulmonary Thrombopoiesis
245
Tomashefski, J. F., Davies, P., Boggis, C., Greene, R., Zapol, W. M. and Reid, L. M. (1983). The pulmonary vascular lesions of the adult respiratory distress syndrome. American Journal of Pathology, 112, 112–126. Trowbridge, E. A. (1988). Pulmonary platelet production: a physical analogue of mitosis? Blood Cells, 13, 451–458. Trowbridge, E. A., Martin, J. F. and Slater, D. N. (1982). Evidence for a theory of physical fragmentation of megakaryocytes, implying that all platelets are produced in the pulmonary circulation. Thrombosis Research, 28, 461–475. Wright, J. H. (1910). The histogenesis of platelets. Journal of Morphology, 21, 263–269. Zabel, P. and Schade, F. U. (1994). Pentoxifylline and tumour necrosis factor-induced lung injury. European Respiratory Journal, 7, 1389–1391.
Received, June 17th, 1998 Accepted, October 12th, 1998