Systemic hypoxia produces an inflammatory response characterized by increases in reactive O2 species (ROS), venular leukocyte-endothelial adherence and emigration, and vascular permeability. Inflammation is typically initiated by mediators released from activated perivascular cells that generate the chemotactic gradient responsible for extravascular leukocyte accumulation. These experiments were directed to study the possible participation of mast cells in hypoxia-induced microvascular inflammation. Mast cell degranulation, ROS levels, leukocyte adherence and emigration, and vascular permeability were studied in the mesenteric microcirculation by using intravital microscopy of anesthetized rats. The main findings were 1) activation of mast cells with compound 48/80 in normoxia produced microvascular effects similar, but not identical, to those of hypoxia; 2) systemic hypoxia resulted in rapid mast cell degranulation;3) blockade of mast cell degranulation with cromolyn prevented or attenuated the hypoxia-induced increases in ROS, leukocyte adherence/emigration, and vascular permeability; and 4) mast cell degranulation during hypoxia was prevented by administration of the antioxidant lipoic acid and of nitric oxide. These results show that mast cells play a key role in hypoxia-induced inflammation and suggest that alterations in the ROS-nitric oxide balance may be involved in mast cell activation during hypoxia.
- mast cell degranulation
- reactive oxygen species
- leukocyte adherence/emigration
- vascular permeability
systemic hypoxia produces a rapid inflammatory response in the mesenteric, pial, and skeletal muscle microcirculations (8). This response includes increased leukocyte-endothelial adhesive interactions in postcapillary venules (29), increased vascular permeability, and leukocyte emigration to the perivascular space (28). Lipid inflammatory mediators such as platelet activating factor (4) and leukotriene B4(24) are involved in this response. Participation of reactive O2 species (ROS) is supported by the observation of a hypoxia-induced increase in the fluorescence intensity of dihydrorhodamine (DHR), an oxidant-sensitive probe, and by the ability of antioxidants to attenuate both the increased DHR fluorescence and the microvascular inflammatory response (27). The exact mechanism responsible for the increase in ROS is not clear; however, lipid inflammatory mediators could contribute to the development of oxidative stress during hypoxia.
Inflammatory mediators released from cells activated by a wide range of stimuli are responsible for the microvascular changes of inflammation (9). These involve increased leukocyte endothelial-adhesive interactions and emigration of leukocytes to the perivascular space. Leukocyte emigration occurs along a chemotactic gradient that is established by activated cells present in the interstitium. The nature of these extravascular cells and the specific stimulus responsible for the hypoxic response are not known. Mast cells are abundant in the microcirculatory beds in which the inflammatory effects of hypoxia have been demonstrated (15) and are known to degranulate and release various inflammatory mediators, including leukotriene B4 and platelet activating factor, in response to several stimuli (4,24). Mast cells have been shown to play a role in leukocyte recruitment into tissue after ischemia and reperfusion (10, 16), sepsis (17), allergic reactions (30), and other conditions (12). On the basis of this evidence, we hypothesized that mast cells could be involved in establishing the chemotactic gradient in systemic hypoxia.
The objective of the present study was to test the hypothesis that mast cells are a key component of the hypoxia-induced microvascular inflammation. To this end, experiments were designed to address the following questions: 1) Does mast cell degranulation under normoxic conditions produce responses similar to those of hypoxia?2) Does hypoxia cause mast cell degranulation? 3) If so, does prevention of mast cell degranulation during hypoxia block the microvascular inflammation? Alterations in the ROS-NO balance are known to play a role in hypoxia-induced inflammation; accordingly, a legitimate question in this respect is 4) Do changes in the ROS-NO balance contribute to mast cell stimulation in hypoxia? The data obtained support a central role for mast cells in the microvascular inflammatory response to hypoxia.
All surgical and experimental procedures involving animals received prior approval from the Animal Care and Use Committee at the University of Kansas Medical Center. The University of Kansas is fully accredited by the American Association for the Accreditation of Laboratory Animal Care. Guidelines established by the National Institutes of Health and the Public Health Service Policy on the humane use and care of laboratory animals were followed at all times.
Male Sprague-Dawley rats (Sasco, Omaha, NE) weighing 200–300 g were fasted overnight and anesthetized with urethane (1.5 g/kg im). A polyethylene catheter (PE-50) was then inserted into the right common carotid artery to collect blood samples and measure systemic arterial blood pressure (Micro-Med, Louisville, KY). A catheter (PE-50) was also inserted into the right jugular vein to infuse lactated Ringer solution (2 ml/h) and administer drugs. A tracheotomy was performed with the use of polyethylene tubing (PE-240).
The animal's temperature was maintained at 36–37°C during all procedures by using a homeothermic blanket system connected to an intrarectal temperature probe (Harvard Apparatus, Natick, MA). The abdomen was opened along the midline by using a radiocautery (Harvard Apparatus). The animal was then positioned on a Plexiglas sheet on top of the stage of a Zeiss Axiovert inverted microscope. A section of the small intestine was carefully removed from the abdomen and positioned over a glass coverslip on the Plexiglas sheet to view the mesenteric venules. The mesentery was covered with Saran wrap to prevent drying of the tissue and to minimize the effect of ambient O2 on the mesenteric venules.
Adhesive Interactions of Circulating Leukocytes in Mesenteric Venules
Mesenteric venules were selected for experiments by use of the following criteria: 1) straight, unbranched vessels at least 100 μm in length; 2) vessel diameters of 20–40 μm;3) no adjacent vessels within 100 μm of the venule; and4) fewer than two adherent leukocytes within a 100-μm segment of the venule. Images of the mesenteric venules (×40 objective) were recorded on a digital Sony video recorder with a time-date generator (Panasonic, Osaka, Japan) by means of a video camera. Venular diameter was measured by using a video caliper (Microcirculation Research Institute, College Station, TX). An optical Doppler velocimeter (Microcirculation Research Institute) was used to measure center-line red blood cell velocity in venules. Average red blood cell velocity was calculated as center-line velocity/1.6 (5). Venular wall shear rate, which represents the physical force generated at the vessel wall due to the movement of blood, was calculated as 8 × (average red blood cell velocity/venular diameter) (11). The extent of leukocyte adherence within mesenteric venules was assessed during playback of the videotapes. The total number of adherent leukocytes was determined in each minute by counting the number of leukocytes that remained stationary for >30 s (29).
Measurement of Microvascular ROS Levels
The oxidant-sensitive probe DHR (27) was used to quantitate ROS levels within mesenteric venules by use of an intensified charge-coupled device (ICCD) camera (Hamamatsu Photonics, Shizouka, Japan). Recordings of the DHR fluorescence were made during brief intervals (∼15 s) to avoid light-induced activation of the probe. The fluorescence intensity was later measured during playback of videotapes by using image analysis software (NIH Image 1.62). The intensity of the fluorescent signal was measured in five adjacent circles of 5-μm diameter along the vessel and averaged to obtain a single estimate of the fluorescent signal during each experimental period. The same field of view was maintained throughout the experiment. Values for fluorescence during hypoxia and normoxic recovery were expressed relative to the value observed during the normoxic control period, defined as 100%.
Measurement of Mast Cell Degranulation
Ruthenium red (0.001%) was added to the superfusate, and images of the mesenteric microcirculation were then recorded every 2 min throughout the experiment. The images were converted to digitized grayscales and then phase inverted (14). The relative light intensity of each mast cell within the field of view was measured, and the degree of mast cell activation was calculated as the ratio of experimental to control intensities.
Measurement of Leukocyte Emigration and Vascular Permeability
Leukocyte emigration was assessed from video recordings by counting the number of leukocytes in an area defined as 100 μm along the venule by 40 μm away from the vessel. Emigration was then expressed as the number of extravascular leukocytes per 4 × 103 μm2. Five venules were analyzed in each rat, and the results were then averaged to obtain an estimate of leukocyte emigration for each animal.
FITC-labeled bovine albumin (50 mg/kg iv) was injected into conscious rats 30 min before examination of the mesenteric microcirculation. By use of an ICCD camera, fluorescence intensity was recorded at an excitation wavelength of 420–490 nm and an emission wavelength of 520 nm. Duration of fluorescence recordings was <15 s in a given area. The fluorescence intensity in three separate venules and surrounding areas was measured in each rat by using NIH Image 1.62, and the values were averaged. The vascular permeability index was then calculated as the ratio of extravascular to intravascular fluorescence intensities (28).
Drugs and Chemicals
Compound 48/80, cromolyn sodium, ruthenium red, lipoic acid (thioctic acid), FITC-albumin, PBS, and urethane were purchased from Sigma Chemical (St. Louis, MO). (Z)-1-[N-(3-aminopropyl)-N-4-(3-aminopropylammonio)butyl-amino] diazen-1-ium-1,2-diolate [spermine NONOate (SNO), Cayman Chemical, Ann Arbor, MI] was dissolved in PBS (Sigma Chemical) at pH 8.5. The rate of decomposition of this NO donor is both pH and temperature dependent (18) and is relatively stable at pH 8.5 (28). At pH 7.4, SNO spontaneously dissociates releasing NO. DHR was obtained from Molecular Probes (Eugene, OR).
Effects of compound 48/80 and cromolyn on mast cell degranulation under normoxic conditions.
In all protocols, experiments began after a 30-min postsurgery stabilization period. The animals spontaneously breathed through a two-way valve (2384 series, Hans Rudolph, Kansas City, MO), which was attached to the tracheal tube.
This experimental protocol consisted of a 10-min control period, followed by superfusion of ruthenium red over the mesentery for 15 min, followed by superfusion of compound 48/80 for 10 min. Compound 48/80 was removed from the superfusate, and the experiment continued for another 10 min. In a separate group of animals, the same protocol was followed, except that the mast cell stabilizer cromolyn was administered as a bolus (40 mg/kg iv) at the beginning of the experiment followed by a continuous infusion (0.8 mg · kg−1 · h−1). The extent of mast cell degranulation was measured every 2 min throughout the experiments as described above.
Does mast cell degranulation in normoxia produce microvascular responses similar to those induced by systemic hypoxia?
The protocol of these experiments consisted of an initial 10-min normoxic control period in which the animals breathed room air, a 20-min equilibration period after injection of DHR (10 μg/kg iv), and a 10-min hypoxic period (10% O2-90% N2) followed by a 10-min normoxic recovery period. Recordings of DHR fluorescence were made at the end of the normoxic, hypoxic, and normoxic recovery periods.
An additional series of experiments was carried out to study leukocyte-endothelial adherence as described above. This protocol consisted of a 10-min normoxic period and a 10-min hypoxia period, followed by a 10-min normoxic recovery period.
In a third series of experiments, after a control period of 10 min, the mesentery was superfused with compound 48/80 for an additional 10 min. Compound 48/80 was removed, and the microcirculation was examined for an additional 10 min. Animals breathed room air throughout the entire experiment. DHR fluorescence and leukocyte adherence were measured in separate groups of rats.
Does systemic hypoxia induce mast cell degranulation?
This experimental protocol consisted of a 10-min normoxic period and a 10-min hypoxia period, followed by a 10-min normoxic recovery period. Ruthenium red was superfused over the mesentery in these experiments to assess the degree of mast cell activation as described above.
Does prevention of mast cell degranulation block the microvascular response to systemic hypoxia?
In a first series of experiments, the protocol described for the hypoxia experiments was used to determine the effect of the mast cell stabilizer cromolyn on hypoxia-induced increase in ROS levels, mast cell degranulation, and leukocyte-endothelial adherence. Cromolyn was administered as described above. Microvascular ROS levels or mast cell degranulation and leukocyte adherence were measured in separate groups of animals.
A second set of experiments assessed the role of mast cells on hypoxia-induced leukocyte emigration and increased vascular permeability. On the day before the experiment, the rats were anesthetized with xylazine-ketamine (2 mg/kg body wt im), and PE-50 catheters were placed in a jugular vein and a carotid artery and tunneled to the back of the neck. On the next morning, the animals were randomly assigned to protocols designed to determine leukocyte emigration and vascular permeability as described above. The rats were placed in a chamber in which 10% O2 was continuously circulated. After 3.5 h, FITC-albumin was injected intravenously; at 4 h of hypoxia, the rats were quickly anesthetized (urethane, 1 g/kg iv) and prepared for intravital microscopy. Vascular permeability and leukocyte emigration were measured as described above. To minimize exposure of the animals to room air, the measurements were completed within 15 min of induction of anesthesia. Parallel experiments were conducted in rats exposed to room air as well as in rats given cromolyn throughout the hypoxia period.
Do changes in the ROS-NO balance contribute to mast cell stimulation in hypoxia?
After the normoxic period, SNO (100 μM superfusion) or lipoic acid (2 mg/kg iv) were administered to animals beginning 20 min before hypoxic challenge. Rats then breathed 10% O2 for 10 min, followed by a 10-min normoxic recovery period. Mast cell activation was measured as described above.
Data are presented as means ± SE. Analysis of variance with Bonferroni's test for multiple comparisons was used to compare differences between groups (Statistix 4.0, Analytical Software, St. Paul, MN). Values of P < 0.05 were considered to be statistically significant.
Effects of Compound 48/80 and Cromolyn on Mast Cell Degranulation Under Normoxic Conditions
Compound 48/80 was used in these experiments to activate mast cells. Figure 1 shows a photograph of the mesenteric microcirculation treated with ruthenium red to visualize mast cell degranulation, before [top left, normoxia (NX)] and 10 min after topical application of compound 48/80 (top right, NX + C48/80). The extravascular degranulated mast cells are indicated by the large arrowheads. Ten minutes after application of compound 48/80, mast cell degranulation index increased from a control value of 1.0 to 1.53 ± 0.06 (n = 6; P < 0.05), indicating rapid and intense mast cell degranulation. Pretreatment with the mast cell stabilizer cromolyn had no effect on mast cell degranulation in normoxia (bottom left, NX + cromolyn) and totally blocked the increase in ruthenium red intensity produced by compound 48/80 (bottom right, mast cell degranulation index 1.02 ± 0.04, n = 6, P > 0.05 NX + cromolyn vs. NX + cromolyn + C48/80), indicating that cromolyn effectively prevents mast cell degranulation. The small arrows inside the vessel shown at top right point to leukocytes adhering to the endothelium. No adherent leukocytes are observed in the remaining photographs. Specifically, the lack of adherent leukocytes in the bottom right photograph illustrates the fact that cromolyn blocks the leukocyte-endothelial adherence as well as the mast cell degranulation induced by compound 48/80 (see Fig. 5).
Does Mast Cell Degranulation in Normoxia Produce Microvascular Responses Similar to Those Induced by Systemic Hypoxia?
Systemic hypoxia results in increased ROS levels in the mesenteric microcirculation (25, 26, 28). Ten minutes after a reduction in inspired O2, DHR fluorescence intensity increased significantly to 166 ± 18% of the normoxic control (Fig. 2). Ten minutes after topical administration of compound 48/80 under normoxic conditions, DHR fluorescence intensity increased to 143 ± 11% of control (Fig.2), a value that is not significantly different from that observed during hypoxia. After 10 min of recovery from hypoxia, DHR fluorescence intensity had returned to a value not significantly different from that of the normoxic control (Fig. 2); on the other hand, DHR fluorescence intensity remained significantly elevated above control 10-min after compound 48/80 superfusion ended (Fig. 2).
Figure 3 shows that administration of compound 48/80 in normoxia is followed by an increase in leukocyte adherence that is of similar magnitude and time course as that produced by systemic hypoxia. The only difference between compound 48/80 administration and systemic hypoxia is that leukocyte-endothelial adherence tended to decrease in the posthypoxia period, whereas it remained elevated after removal of compound 48/80 from the superfusate.
The data obtained in this section indicate that degranulation of mast cells using compound 48/80 mimics the increases in ROS and leukocyte adherence induced by systemic hypoxia.
Does Systemic Hypoxia Induce Mast Cell Degranulation?
Figure 4 illustrates the effects of hypoxia and compound 48/80 on mast cell degranulation as evidenced by ruthenium red uptake. The topmost row of Fig. 4 shows photographs of a mesenteric venule immediately before and every 2 min after the onset of hypoxia, whereas the panel below shows enlargements of the mast cell indicated by the arrow in the photographs above. It is clear that systemic hypoxia produces rapid and sustained mast cell degranulation, as indicated by the increase in ruthenium red intensity occurring after inspired Po 2 was decreased. A similar pattern of mast cell degranulation is observed after administration of compound 48/80 (Fig. 4, third and fourth rows). The time course and intensity of mast cell degranulation elicited by both hypoxia and compound 48/80 were similar (Fig. 5, P> 0.05, n for each group = 6). Cromolyn administration completely abolished the increase in ruthenium red intensity secondary to systemic hypoxia (Fig. 5, ●, n = 5) and to administration of compound 48/80 (Fig. 5 , ▴,n = 6). In summary, hypoxia causes increases in mast cell degranulation that are similar in time course and intensity to those produced by administration of compound 48/80 and that are blocked by administration of cromolyn.
Does Prevention of Mast Cell Degranulation Block the Microvascular Response to Systemic Hypoxia?
Figure 6 shows that 10 min of hypoxia result in an increase in DHR fluorescence intensity to 154 ± 12% above control (P < 0.05, n = 6). This increase is reversible, and, after 10 min of normoxia, DHR fluorescence intensity has decreased significantly from the hypoxic levels (Fig. 6). Pretreatment with cromolyn significantly limits the hypoxia-induced increase in DHR fluorescence intensity to 114 ± 5% of control (Fig. 6; P > 0.05, n = 6), indicating that prevention of hypoxia-induced mast cell degranulation abolishes the increase in microvascular ROS levels associated with hypoxia.
Cromolyn also significantly inhibits hypoxia-induced leukocyte adherence (Fig. 3 ), as well as the leukocyte adherence secondary to compound 48/80 administration (Figs. 1 and 3). This observation links mast cell degranulation to the increase in leukocyte-endothelial interactions that accompanies systemic hypoxia.
Conscious rats exposed to hypoxia for 4 h showed a significant increase in leukocyte emigration to the perivascular space, from 2.4 ± 0.4 in the normoxic controls (Fig.7, n = 5) to 11.5 ± 1.6 emigrated leukocytes per 4 × 103μm2 in the hypoxic rats (Fig. 7, P < 0.05 hypoxia vs. normoxia, n = 6). Cromolyn attenuated the increase in leukocyte emigration associated with 4 h of hypoxia to 4.9 ± 0.6 emigrated leukocytes per 4 × 103 μm2 (Fig. 7, P < 0.05 hypoxia + cromolyn vs. hypoxia).
Interestingly, prevention of mast cell degranulation only partially attenuated the increase in vascular permeability that follows systemic hypoxia in conscious rats. The ratio of perivascular to intravascular FITC-albumin fluorescence intensity, a vascular permeability index, increased from a normoxic value of 0.12 ± 0.03 (Fig.8, n = 5) to a value of 0.99 ± 0.07 after 4 h of hypoxia in the untreated rats (Fig.8, P < 0.05 hypoxia vs. normoxia, n = 6). In contrast, cromolyn-treated rats showed a permeability index of 0.68 ± 0.08 (Fig. 8, P < 0.05 hypoxia + cromolyn vs. hypoxia, P < 0.05 hypoxia + cromolyn vs. control, n = 6), which, although significantly lower than that of the untreated hypoxic rats, was significantly higher than that of the normoxic controls.
In summary, the data obtained in this section indicate that prevention of mast cell degranulation either blocks or attenuates the microvascular responses to systemic hypoxia.
Do Changes in the ROS/NO Balance Contribute to Mast Cell Stimulation in Hypoxia?
Figure 9 illustrates the effects of administration of lipoic acid and of the NO donor SNO on mast cell degranulation. No mast cell degranulation is observed during normoxia, as evidenced from the lack of uptake or ruthenium red (Fig. 9,top left). Ten minutes of hypoxia (Fig. 9; top right) resulted in an average degranulation index of 1.90 ± 0.12 [P < 0.05 hypoxia (HX) vs. NX, n= 6]. Treatment with lipoic acid did not influence mast cell degranulation under normoxic conditions (Fig. 9, middle left) and effectively blocked hypoxia-induced mast cell degranulation (middle right, average mast cell degranulation index: 1.03 ± 0.09; P > 0.05, NX + lipoic acid vs. HX + lipoic acid, n = 6). Treatment with the NO donor had effects similar to those of lipoic acid, as shown in the bottom two panels; average mast cell degranulation index during hypoxia was 1.09 ± 0.07 (P > 0.05, NX + SNO vs. HX + SNO, n = 6). The small arrows inside the vessel in the top right photograph indicate adherence leukocytes. No leukocyte adherence was observed in any other photograph. The absence of leukocyte adherence during hypoxia after treatment with lipoic acid and with SNO confirm our laboratory's previous observations on the effects of these agents on hypoxia-induced leukocyte adherence (24, 27, 29).
The main findings of this study are 1) interventions that induce mast cell degranulation produced microvascular responses similar but not identical in time course and magnitude to those elicited by systemic hypoxia; 2) systemic hypoxia resulted in mast cell degranulation; 3) blockade of the hypoxia-induced mast cell degranulation prevented or attenuated the microvascular responses; and 4) mast cell degranulation appears to be dependent on the ROS-NO balance within the microcirculation. These results point out the involvement of mast cells in the microcirculatory changes of acute systemic hypoxia. To our knowledge, this is the first demonstration of a link between mast cell degranulation and hypoxia-induced microvascular inflammation.
Mast cell degranulation induced by compound 48/80 resulted in increased levels of ROS and leukocyte-endothelial adherence. The time course and intensity of these responses were not different from those produced by systemic hypoxia; however, whereas both ROS and leukocyte adherence returned toward control values during the normoxic recovery period, they remained elevated after compound 48/80 was removed from the mesenteric surface. There are several possible explanations for this difference, the simplest being that removal of compound 48/80 from the surface does not immediately reduce its concentration in the tissue and therefore its effect is maintained. It is also possible that compound 48/80 and hypoxia do not have identical effects on the mechanisms of ROS generation and leukocyte-endothelial interactions, which may involve actions independent of mast cell stimulation, but this possibility cannot be answered from the present experiments. At any rate, the similarity of the initial responses to hypoxia and compound 48/80 administration provide one of the pieces of evidence linking mast cells with the inflammatory response to systemic hypoxia.
A second piece of evidence supporting a causal relationship between mast cell activation and hypoxia-induced microvascular inflammation was the demonstration that systemic hypoxia produces rapid mast cell degranulation (Figs. 4 and 5). The time course of mast cell degranulation corresponds closely to that of the increase in leukocyte-endothelial adherence during systemic hypoxia and provides support for a link between these two phenomena. Furthermore, mast cell degranulation induced by compound 48/80 promotes leukocyte-endothelial adhesive interactions with a time course similar to that of hypoxia-induced mast cell degranulation (Fig. 3). Accordingly, activation of mast cells, whether produced by hypoxia or compound 48/80, is rapidly followed by increased leukocyte-endothelial adherence. These two events are highly correlated, as illustrated in Fig. 10, which shows leukocyte adherence plotted as a function of the degree of mast cell degranulation, as measured by ruthenium red intensity. Each point represents the average values obtained during the control period and every 2 min after the onset of hypoxia or after application of compound 48/80. Interestingly, for a given level of mast cell activation, leukocyte adherence was significantly greater after compound 48/80 administration than during hypoxia, as indicated by the significant difference in the slopes relating leukocyte adherence and mast cell degranulation. This suggests that, for a given degree of mast cell activation, the leukocyte-endothelial adhesive interactions are more forceful with compound 48/80 administration than with hypoxia. The reason for this difference is not clear. One possibility is that compound 48/80 stimulates cells other than mast cells that further contribute to increased leukocyte-endothelial adhesive interactions. The observation that administration of the specific mast cell stabilizer cromolyn has the same effect on leukocyte-endothelial adherence induced by compound 48/80 and by hypoxia (Fig. 3) argues against this possibility. An alternative explanation could be the liberation of anti-inflammatory substances during hypoxia. Adenosine, for instance, is released during hypoxia (20) and has been shown to attenuate leukocyte-endothelial adherence in ischemia-reperfusion (1) and other models of inflammation. Although compound 48/80 has been reported to liberate adenosine from activated mast cells (19), hypoxia could result in the production of additional amounts of this agent from other sources.
A final piece of evidence supporting a critical role of mast cells in hypoxia-induced microvascular inflammation is the demonstration that blockade of degranulation with the mast cell stabilizer cromolyn significantly modifies the microvascular responses to hypoxia. Cromolyn prevented the mast cell degranulation produced by hypoxia; this was associated with a reduction in ROS levels, a complete blockade of leukocyte adherence and emigration, and a significant, albeit small, reduction in vascular permeability.
The increase in ROS during systemic hypoxia observed in these experiments confirms our laboratory's previous observations (24,25, 27) and those of others (2, 6, 26). Systemic hypoxia is accompanied by a reversible increase in DHR fluorescence intensity that is blocked by antioxidants; this, together with the inhibitory effect of antioxidants on leukocyte-endothelial interactions and vascular permeability, supports a role for ROS in the microvascular response to hypoxia. NO administration also blocks the hypoxia-induced increase in DHR fluorescence as well as the increased leukocyte-endothelial adhesive interactions. This suggests that an increase in ROS, or an alteration in the ROS-NO balance, is a key event in the microvascular changes induced by hypoxia (25). A link between mast cell degranulation and ROS increase in hypoxia is evidenced by the facts that the mast cell stabilizer cromolyn blocked the hypoxia-induced increase in ROS and that mast cell activation with compound 48/80 resulted in elevated ROS levels as well as in increased leukocyte-endothelial adherence. The facts that hypoxia-induced mast cell degranulation is accompanied by increased ROS levels and that blockade of mast cell degranulation attenuates the ROS increase as well as the microvascular response suggest that activation of mast cells occurs early in the inflammatory cascade elicited by hypoxia.
The ability of cromolyn to block the hypoxia-induced increase in leukocyte adherence supports a role of activated mast cells in this phenomenon. In general, adherence of leukocytes to the endothelium is followed by emigration to the perivascular space along a chemotactic gradient generated by the presence of activated cells outside the vessel. Cromolyn blocked the leukocyte emigration in conscious rats exposed to 4 h of hypoxia. This could be due to the inhibitory effect of cromolyn on leukocyte adherence (Fig. 3). In addition, cromolyn could have prevented the development of the chemotactic gradient necessary for leukocyte emigration, suggesting that activated mast cells have a role in generating this gradient.
Cromolyn attenuated, but did not totally prevent, the increase in vascular permeability of systemic hypoxia, indicating the participation of factors in addition to mast cell activation. A direct effect of hypoxia on vascular permeability is suggested by the observation that cultured endothelial cells show an increase in permeability in response to reductions in Po 2 (21). Formation of gaps between endothelial cells has been shown to be necessary for both the increase in permeability and the leukocyte emigration during inflammation (9). The dissociation between increases in vascular permeability and leukocyte emigration when hypoxia-induced mast cell degranulation was prevented (Figs. 7 and8) indicate that increased permeability is a necessary but not sufficient factor in the process of leukocyte emigration. The results obtained here support the possibility that activation of mast cells contributes to generate the chemotactic gradient necessary for leukocyte recruitment into tissue during systemic hypoxia.
Although the results show that mast cells are activated in hypoxia, the mechanism underlying this response is unclear. The ability of lipoic acid and of the NO donor to prevent mast cell degranulation during hypoxia points to an involvement of the ROS-NO balance in mast cell activation. These results are consistent with previous observations of blockade of mast cell activation by antioxidants (16, 23) and NO (7, 22), as well as enhanced degranulation of mast cells after inhibition of NO synthase (3,13, 22). Indirect evidence for a participation of ROS in mast cell degranulation is provided by the DHR fluorescence images of the microcirculation obtained in this study (Fig.11). It is clear that hypoxia and compound 48/80 administration are accompanied by increases in ROS levels at the vessel wall, where leukocyte-endothelial adhesive interactions take place. However, close inspection of these images shows discrete areas of increased fluorescence in the extravascular compartment. Fluorescence intensity at these sites is attenuated by cromolyn, suggesting that fluorescence originates either in activated mast cells or in cells that generate ROS after mast cell activation in response to hypoxia or compound 48/80 (Fig. 11).
It is clear that further research is needed to determine the mechanism of mast cell activation during hypoxia. If changes in the mast cell ROS-NO balance are early events in the initiation of the hypoxic inflammatory response, as suggested by the present experiments, a possible sequence of events would involve an initial increase in mast cell ROS levels, which would trigger mast cell activation, release of inflammatory mediators, and further elevation of ROS levels and depletion of NO within the microcirculation. This scenario would explain the ability of antioxidants and the mast cell stabilizer to block mast cell activation as well as the microvascular inflammatory response. This would imply that ROS participate in signal transduction during hypoxia, a role suggested by a recent series of studies (2, 6, 26).
In summary, the results presented here strongly support the participation of mast cells in the microvascular effects of systemic hypoxia and provide additional evidence on the inflammatory nature of this response. This issue is highly relevant because an inflammatory component has been proposed for the various acute conditions occurring at altitude, including acute mountain sickness and high-altitude cerebral edema, which are characterized by abnormal body fluid balance and increased vascular permeability. Although the early microvascular inflammation of hypoxia appears to be a deleterious response, it is interesting that prolonged exposure to hypoxia in rats leads to acclimatization of the vascular endothelium, with the inflammatory lesion resolving and the animals becoming resistant to even more severe hypoxia (29). It could be hypothesized that the early microvascular inflammatory response initiates the cascade of events that leads to acclimatization of vascular endothelial function. In this light, it is important to note that acute mountain sickness and high-altitude cerebral edema do not occur in acclimatized individuals.
This work was supported by National Heart, Lung, and Blood Institute Grants HL-39443 (to N. C. Gonzalez) and HL-64195 (to J. G. Wood).
Address for reprint requests and other correspondence: J. G. Wood, Dept. of Molecular & Integrative Physiology, Univ. of Kansas Medical Center, 3901 Rainbow Blvd., Kansas City, KS 66160 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
September 20, 2002;10.1152/japplphysiol.00637.2002
- Copyright © 2003 the American Physiological Society