Published ahead of print on September 11, 2003, doi:10.1164/rccm.200306-801OC
© 2003 American Thoracic Society Pulmonary Inflammation and Thrombogenicity Caused by Diesel Particles in HamstersRole of HistamineLaboratory of Pneumology (Lung Toxicology) and Center for Molecular and Vascular Biology, K. U. Leuven, Leuven, Belgium Correspondence and requests for reprints should be addressed to Benoit Nemery, M.D., Ph.D., K. U. Leuven, Laboratory of Pneumology, Unit of Lung Toxicology, Herestraat, 49, B-3000 Leuven, Belgium. E-mail: ben.nemery{at}med.kuleuven.ac.be
Short-term increases in particulate air pollution are associated with increased incidence of cardiovascular events. Previously, we showed that intratracheally instilled diesel exhaust particles (DEPs) are prothrombotic. Here, we investigated the time course and the mechanisms. At 1, 6, and 24 hours after instillation of 50 µg DEPs per hamster, the mean size of in vivo-induced and quantified venous thrombosis was increased by 480%, 770%, and 460%, respectively. Platelets activation in blood was confirmed by a shortened closure time in the platelet function analyzer (PFA-100). In bronchoalveolar lavage, neutrophils and histamine levels were increased at all time points. In plasma, histamine was increased at 6 and 24 hours but not at 1 and 3 hours. Pretreatment with a histamine H1-receptor antagonist (diphenhydramine, 30 mg/kg intraperitoneally) abolished the DEP-induced neutrophil influx in bronchoalveolar lavage at all time points. However, diphenhydramine pretreatment did not affect DEP-induced thrombosis or platelet activation at 1 hour, whereas both were markedly reduced at 6 and 24 hours. In conclusion, pulmonary inflammation and peripheral thrombosis are correlated at 6 and 24 hours, but at 1 hour, the prothrombotic effects do not appear to result from pulmonary inflammation but possibly from the blood penetration of DEP-associated components or by DEP particles themselves.
Key Words: air pollution • particles • thrombosis • lung inflammation • histamine Diesel exhaust particles (DEPs), the main component of particulate matter with a diameter of less than 2.5 µm in urban areas, are a major contributor to inhaled particulate matter pollution. Although several epidemiologic studies have reported associations between the concentration of ambient particulate matter and increased cardiopulmonary morbidity and mortality (1–3), the biological plausibility of these consistent observations is only starting to be elucidated. Currently, three lines of particle-related research are being pursued. Thus, inhaled particles have been suggested to impact on the autonomic nervous system, leading to changes in the pattern of breathing, heart rate and heart rate variability (4, 5). Inhaled particles may affect the cardiovascular system through inflammatory mediators (e.g., cytokines) produced in the lungs and released into the circulation (6, 7). In addition, we (8, 9) and others (10, 11) have demonstrated that ultrafine particles (UFPs) with a diameter of less than 0.1 µm may pass from the lungs into the systemic circulation. This may constitute a third alternative and/or a complementary explanation for the acute cardiovascular events after exposure to particles, besides autonomic stimulation and overflow of inflammatory mediators from the lungs into the blood. Histamine is a proinflammatory mediator stored within cytoplasmic granules of mast cells and basophils (12) and is known to play a central role in pathophysiologic processes (13). In addition to producing bronchoconstriction and vasodilation, histamine upregulates P-selectin on endothelial cells (14), directly activates neutrophils (15), induces exocytosis, stimulates interleukin-6 production from human lung macrophages, and modulates the cytokine network (16). Some studies have reported a release of histamine in the respiratory tract after DEP exposure. Recently, we have shown that DEPs cause a dose-related release of histamine in bronchoalveolar lavage (BAL) fluid 1 hour after their intratracheal instillation in hamsters (17). Others have reported an increase in mast cell numbers in the submucosa and elevated BAL histamine levels 6 hours after exposure to DEPs in humans (18). DEPs have also been demonstrated to directly degranulate mast cells and to increase histamine levels and symptom severity in humans (19). Moreover, it has recently been shown that chemical constituents of DEPs induce interleukin-4 production and histamine release by human basophils (20). We have recently demonstrated that DEPs lead to lung inflammation as well as a rapid activation of circulating blood platelets 1 hour after their intratracheal instillation (17). Therefore, the goal of this study was twofold: (1) to evaluate the time course of thrombus enhancement and its relationship to the pulmonary inflammation accompanying exposure to DEP and (2) to investigate the mechanism of this phenomenon by focusing on histamine release in the respiratory tract and the peripheral circulation and the effects of antihistamine pretreatment. Some of the findings presented in this article have been reported previously in the form of abstracts (21, 22).
DEPs (SRM 1650) from the National Institute of Standards and Technology (Gaithersburg, MD) were suspended in saline (NaCl 0.9%) containing Tween 80 (0.1%). To minimize aggregation, particle suspensions were sonicated (Branson 1200, VEL, Leuven, Belgium) for 15 minutes and vortexed immediately (less than 1 minute) before their dilution and before intratracheal administration. Control animals received saline containing Tween 80 (0.1%). The experiments were performed in accordance with protocols approved by the Institutional Animal Care and Research Advisory Committee of the K. U. Leuven. Hamsters (100–110 g) were anesthetized with sodium pentobarbital (60 mg/kg, intraperitoneally). The tracheal zone was shaved and disinfected with ethanol (70%), and the trachea was exposed for the intratracheal administration of 120 µl of vehicle or DEP (50 µg per animal). The animals were then allowed to recover. They were killed with an overdose of sodium pentobarbital at 1, 3 (only for histamine measurements), 6, or 24 hours after intratracheal instillation. Experiments could not be completed on all animals the same day. However, at least one relevant control animal was always included on each experimental day. Results for the 1-hour time point contain data (without antihistaminic pretreatment) from animals included in a previous publication (17). All techniques have been described in previous publications (17, 23) and can be found in the online supplement. Briefly, BAL was performed by cannulating the trachea and lavaging the lungs with 4.5 ml (three times, 1.5 ml) of sterile NaCl 0.9%. BAL cells were counted and identified (Diff-Quik stain). The histamine content was determined by a radioimmunoassay kit (Immunotech, Marseille, France) in BAL and in plasma obtained after centrifugation (1,000 g x 10 minutes, 4°C) of venous blood collected from the abdominal vena cava on ethylenediaminetetraacetic acid. The BAL and plasma samples were stored at -20°C. The technique used to induce and to monitor mural thrombosis has been recently described in detail (23–25). Briefly, a surgically exposed segment of the femoral vein was illuminated with green (540 nm) light for 2 minutes after the intravenous injection of Rose Bengal (20 mg/kg; Sigma, St. Louis, MO), thus producing a free-radical–mediated endothelial lesion with subsequent vascular thrombosis. The kinetics of thrombus generation were monitored during 40 minutes using an online microscope video camera, and thrombosis was quantitated via image analysis. The size of the thrombus is expressed in arbitrary units as the total area under the curve, when the light intensity is plotted against time. Platelets were counted, and platelet function was assessed in the Platelet Function Analyzer PFA-100 (Dade-Behring, Marburg, Germany) (26) using venous blood collected from the abdominal vena cava on hirudin (20 µg/ml) and supplemented with 0.4% citrate. Hamsters were pretreated with a histamine H1 receptor antagonist, diphenhydramine (Sigma), at a dose of 30 mg/kg intraperitoneally (27) at 1 hour before intratracheal instillation of DEP or saline. Data are means ± SEM. Comparisons between groups were performed by unpaired Student's t test or one-way analysis of variance followed by Newman-Keuls multiple range test.
Cells in BAL Fluid In control hamsters, the total cell count was 3.1 ± 0.5 x 103, 2.8 ± 0.4 x 103, and 3.3 ± 0.4 x 103 per ml BAL fluid at 1, 6, and 24 hours, respectively, after saline administration. No evidence of inflammation or bleeding was seen in BAL of control subjects. The intratracheal instillation of DEP resulted in a time-dependent increase in total BAL cell numbers to 86.2 ± 25.4 x 103 (1 hour, p < 0.01), 102.8 ± 33.0 x 103 (6 hours, p < 0.01), and 163.6 ± 29.4 x 103 (24 hours, p < 0.001) per ml of BAL fluid. Figure 1 shows that in control subjects the percentage of polymorphonuclear neutrophils (PMNs) was as low as 1 ± 0.5%, 1 ± 0.7%, and 3 ± 1% at 1, 6, and 24 hours, respectively, with all other cells being macrophages. The intratracheal instillation of DEP led to a significant and progressive PMN influx at 1 (13% of total cell number), 6 (22%), and 24 hours (37%). The remainder of the BAL cells in the DEP group was primarily macrophages; lymphocytes made up only 1–2% of total cells. No other cells were observed, and there was no evidence of intraalveolar bleeding.
Experimental Thrombosis In Vivo The intratracheal instillation of DEP induced a significant increase in the cumulative mass of thrombus generated in vivo in a photochemically injured femoral vein at 1 (480%), 6 (770%), and 24 hours (460%) compared with the thrombi observed in the corresponding control subjects (Figure 2) . The prothrombotic effect was therefore comparable at all time points investigated.
Platelet Function Analysis Ex Vivo Platelet function was analyzed by means of the PFA-100, which measures the ability of blood flowing through a capillary filter coated with collagen and epinephrine to clog the cartridge as a function of platelet activation and aggregation. A significant and comparable shortening of the closure time, reflecting platelet activation, was found in all DEP-exposed groups at 1 (66%), 6 (55%), and 24 hours (67%) compared with their respective control subjects (Figure 3) . There was no difference in platelet counts in blood between control subjects and DEP-exposed groups at all time points investigated (p = NS).
Histamine Determination in BAL and Plasma Whereas in BAL, elevated histamine levels were already detected 1 hour after DEP instillation, in plasma, histamine elevation was considerably slower. Histamine concentrations in BAL fluid increased considerably after the intratracheal instillation of DEP at 1 (sevenfold), 3 (sixfold), 6 (fourfold), and 24 hours (fivefold) (Figure 4A) . In contrast, in plasma (Figure 4B), an enhanced release of histamine was observed at 6 (threefold) and 24 hours (twofold) but not at 1 and 3 hours.
Effect of Diphenhydramine, a Histamine H1 Receptor Antagonist, on Neutrophil Influx in BAL, Thrombosis In Vivo, and Platelet Function Analysis (PFA) Ex Vivo Pretreatment of control hamsters with diphenhydramine did not affect the total cell numbers per milliliter of BAL (3.8 ± 0.8 x 103 at 1 hour, 2.1 ± 0.2 x 103 at 6 hours, and 3.1 ± 0.4 x 103 at 24 hours) nor the percentage of PMNs in BAL nor the degree of thrombosis in vivo nor the PFA ex vivo at all time points, compared with nontreated control subjects. In saline-treated hamsters, the small release of histamine in BAL was not significantly affected by the pretreatment with diphenhydramine (4.3 ± 1.5 nM at 1 hour, 2.0 ± 0.5 nM at 3 hours, 5.4 ± 2.0 nM at 6 hours, and 8.7 ± 2.5 nM at 24 hours). Likewise, histamine levels in plasma remained unaffected (28.0 ± 4.6 nM at 1 hour, 31.5 ± 3.0 nM at 3 hours, 27.4 ± 5.0 nM at 6 hours, and 33.0 ± 5 nM at 24 hours). Similarly, in DEP-treated animals, pretreatment with diphenhydramine did not affect histamine concentrations in BAL (30.0 ± 7.5 nM at 1 hour, 25.7 ± 2.0 nM at 3 hours, 39.5 ± 8.3 nM at 6 hours, and 27.1 ± 4.2 nM at 24 hours) nor in plasma (31.5 ± 3.3 nM at 1 hour, 29.3 ± 8.1 nM at 3 hours, 59.0 ± 6.1 nM at 6 hours, and 53.3 ± 10.4 nM at 24 hours). However, pretreatment with diphenhydramine strongly diminished the DEP-induced increase in pulmonary inflammation. Total BAL cell numbers (per ml) decreased considerably 1 hour (to 5.3 ± 1.6 x 103, p < 0.01), 6 hours (to 17.8 ± 6.9 x 103, p < 0.01), and 24 hours (to 25.9 ± 3.0 x 1 03, p < 0.001) after intratracheal instillation of DEP. At 1 hour, pretreatment of hamsters with diphenhydramine also strongly reduced the DEP-induced influx of PMNs in BAL (to 2%; Figure 5A) . A significant drop in PMN (%) was also recorded at 6 hours (to 11%; Figure 5B) and at 24 hours (to 13%; Figure 5C).
Pretreatment of hamsters with diphenhydramine did not affect the DEP-induced thrombosis enhancement in vivo at 1 hour (Figure 6A) , whereas thrombosis enhancement was significantly reduced at 6 hours (to 143%; Figure 6B) and at 24 hours (to 315%; Figure 6C).
Diphenhydramine did not affect the closure times of untreated control hamster blood (Figure 7) . Diphenhydramine pretreatment did not significantly affect the closure time shortening caused by DEP exposure at 1 hour; that is, it did not prevent platelet activation. However, at 6 and 24 hours, the diphenhydramine pretreatment reduced the closure time shortening caused by DEP exposure. No statistical significance of the closure time shortening caused by DEP exposure versus saline plus diphenhydramine was observed.
We have demonstrated that the intratracheal instillation of DEP leads to a significant thrombotic tendency, activation of circulating blood platelets, as well as lung inflammation, which persist at least 24 hours after DEP deposition in the lung of hamsters. By pretreating the animals with diphenhydramine, a histamine H1 receptor antagonist, we established that histamine is involved in these processes. Moreover, we found that pulmonary inflammation and peripheral thrombosis are associated at 6 and 24 hours but not at 1 hour after DEP exposure. Elevations in ambient pollution have been associated with increased hospitalizations for cardiovascular disease (28), increased incidence of cardiac arrhythmias (29), increased risk of myocardial infarction (30), and increased susceptibility to myocardial ischemia (3). It has been estimated that 69% of the excess deaths attributable to respirable particulate matter are due to cardiovascular disease (31). Epidemiologic studies indicate that the older population and those with preexisting cardiopulmonary disease may be more susceptible (31). In a search for biological plausibility for these consistent epidemiologic associations, we have shown that UFPs can pass from the airways into the systemic circulation in hamsters (8) and humans (9). We have demonstrated that the intravenous administration and the intratracheal instillation of positively charged polystyrene UFPs, taken as a model of UFPs, enhance the extent of thrombosis in the peripheral circulation (24, 25).The observed in vivo prothrombotic tendency results, at least in part, from platelet activation by positively charged amine-polystyrene particles. In addition, we have also shown that an exposure of hamsters to DEP for as little as 1 hour led to lung inflammation and enhanced the risk for arterial and venous thrombosis (17). However, it was not clear whether the peripheral thrombotic tendency resulted from pulmonary inflammation (6, 7) or was due to the systemic passage of particles (8–11) or to a combination of both phenomena. The intratracheal instillation of a bolus of particles could be considered as being nonphysiologic, but this method of delivery has been shown to be a convenient and valid, although admittedly not perfect, mode of administration of foreign compounds into the airways (32). The intratracheal instillation permits the accurate introduction of a range of doses to the lungs within a short time (32). Since our recent studies had focused on the occurrence of dose-dependent (5–500 µg/animal) prothrombotic effects of DEP within 1 hour after their intratracheal instillation (17), we have verified in this study whether these effects persisted beyond this period. Thus, we have purposefully selected a DEP dose of 50 µg per animal, which corresponds to the intermediate dose that produced a significant increase of thrombogenicity, also because it is probably a relevant dose with regard to human exposure (17). We continued to apply the same validated model of acute thrombosis in the hamster that we have used before in these studies (17, 23). In this photochemical injury model, it is possible to cause mild damage to endothelial cells; developing thrombi are platelet rich, and they resemble clinical thrombi, as shown by electron microscopic analysis (33). By combining transillumination and photochemical vessel wall injury, it has become possible to link the degree of vessel wall injury with the intensity of thrombosis. Our results show that the intratracheal instillation of DEP produces a comparable prothrombotic effect that persists up to 24 hours at least. The platelet adhesion and aggregation studies in the platelet function analyzer, a high-shear, stress-dependent system that simulates platelet-based primary hemostasis in vitro (26, 34), revealed platelet activation at all time points investigated, confirming a role for platelets in the observed in vivo thrombosis. The assessment of lung inflammation by BAL fluid analysis showed a neutrophil influx in the alveoli, which significantly increased in a time-dependent manner but without evidence of major alveolar injury such as bleeding (data not shown). Similar findings have been reported in rats at later time points, that is, 6 hours, 24 hours, and 7 days after intratracheal instillation of ultrafine carbon black particles (35). An increase of neutrophils in BAL fluid of rabbit lungs was observed 20 hours after exposure to charged UFPs (36), and an increase of neutrophil numbers in bronchial submucosa and epithelium was found in human subjects 6 hours after exposure to DEP (18). In addition, we found here a marked and time-dependent elevation of histamine levels in BAL after exposure to DEP, an elevation that was also noted in plasma after 6 and 24 hours but not at 1 and 3 hours. This analysis uncovered a slower increase of plasma histamine. Indeed, at 3 hours after exposure to DEP, plasma histamine levels had not yet risen significantly, and thus, this rise occurred between 3 and 6 hours. Histamine exerts a variety of proinflammatory and immunomodulating effects through the interaction with at least three receptors, H1, H2, and H3 (37–39). The pivotal role of histamine in inflammation is supported by the observation that antagonists of the H1 receptor are effective in reducing some of the acute symptoms of inflammatory disorders (40). The immediate actions of histamine on vascular endothelium and on bronchial and vascular smooth muscle cells are mostly mediated by the activation of the H1 receptor (40). It has been shown that diphenhydramine, a histamine H1 antagonist, can inhibit the histamine-induced upregulation of P-selectin on the endothelial cell surface (41, 42) and prevents intimal thickening, whereas cimetidine, a histamine H2 receptor antagonist, was ineffective in a mouse model of thrombosis (27). Detectable concentrations of histamine have been found in platelets (43–46). Interestingly, the concentration of histamine in platelets of patients with peripheral vascular diseases was higher than that in healthy volunteers, suggesting that the higher histamine content may lead to a larger release of histamine at the sites of vascular injury (45). In addition, Saxena and colleagues (46) have suggested that histamine is an intracellular messenger mediating platelet aggregation. Our results demonstrate that DEP induced a time-dependent increase in histamine release in BAL fluid. This is in agreement with the findings of Salvi and colleagues (18) who have shown an increase in the number of mast cells in the submucosa and elevated BAL histamine levels 6 hours after exposure to DEP in humans. It has been demonstrated that DEPs directly degranulate mast cells and increase histamine levels and symptom severity in humans (19). Recently, it has been shown that chemical constituents of DEPs induce interleukin-4 production and histamine release by human basophils (20). We, therefore, hypothesized that histamine plays a key role in the observed pulmonary and peripheral effects. Pretreatment with diphenhydramine (27) did not affect the release of histamine in BAL or in plasma, excluding a feedback mechanism during the production of histamine in BAL and plasma. However, it markedly inhibited the neutrophil influx in BAL at 1, 6, and 24 hours, consistent with a role of histamine in the DEP-induced lung inflammation. It has been reported that histamine can both directly activate equine neutrophils by acting on H1 receptors located on these cells (15) and can stimulate human neutrophil adhesion by acting on H1 receptors located on endothelial cells (47). The use of diphenhydramine thus enabled us to investigate the role of pulmonary inflammation and vascular blood cell inflammation in the peripheral thrombogenecity induced by DEP. In addition, because diphenhydramine does not affect platelet activation directly, it further allowed linking thrombosis to inflammation. At 1 hour after exposure to DEP, the degree of thrombosis in vivo and platelet aggregation ex vivo was not affected by the pretreatment of hamsters with diphenhydramine. This may be related to the lack of histamine release in plasma (Figure 4B), but the fact that lung inflammation was completely inhibited pleads in favor of a dissociation between DEP-induced lung inflammation and thrombogenecity at this time point. The rapid prothrombotic tendency can be attributed to the direct passage of particles (or their constituents) into the circulation, recently demonstrated to occur within 1 hour for UFPs in hamsters (8) and in humans (9) and to be associated with platelet activation (43, 48). In contrast, a significant reduction of the thrombogenicity in vivo was observed at 6 and 24 hours on diphenhydramine pretreatment. This suggests that inflammatory reactions contribute to the late thrombotic tendency after DEP instillation. In view of the slow appearance of histamine in the blood circulation, a role for pulmonary inflammation seems warranted, especially at 6 hours. We conclude that intratracheally instilled DEPs lead to a significant thrombotic tendency, activation of circulating blood platelets, as well as lung inflammation as early as 1 hour and persisting for 24 hours. Pulmonary inflammation and peripheral thrombosis are linked at 6 and 24 hours, but the thrombotic tendency observed 1 hour after DEP exposure does not appear to result from pulmonary inflammation. We interpret this as being compatible with direct platelet activation either by penetration into the circulation of DEP-associated components or by DEP particles themselves.
Supported by the funds of the K. U. Leuven (OT/02/45) and by the Fund for Scientific Research Flanders (G.0165.03). This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Conflict of Interest Statement: A.N. has no declared conflict of interest; B.N. has no declared conflict of interest; P.H.M.H. has no declared conflict of interest; J.V. has no declared conflict of interest; M.F.H. has no declared conflict of interest. Received in original form June 17, 2003; accepted in final form September 5, 2003
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
0.01 compared with the corresponding control, analyzed by the unpaired Student's t test.
