Marlys H. Gee

Role of intrapulmonary release of eicosanoids and superoxide anion as mediators of pulmonary dysfunction and endothelial injury in sheep with intermittent complement activation.

In 30 anesthetized sheep, we show that repeated bolus injections of autologous zymosan-activated plasma produce pulmonary hypertension, hypoxemia, intrapulmonary thromboxane release, pulmonary leukostasis, and sustained increases in lung lymph flow and protein clearance. Studies with platelet-rich plasma demonstrated that addition of zymosan-activated plasma does not induce platelet aggregation or thromboxane release. We studied the role of cyclooxygenase products as mediators of these pathophysiological responses by pretreating sheep with either meclofenamate (4 mg/kg) or ibuprofen (12.5 mg/kg). Both drugs inhibited thromboxane release and hypoxemia. Ibuprofen, but not meclofenamate, reproducibly attenuated the hypertensive responses and the increases in lymph flow and protein clearance. Neither drug prevented pulmonary leukostasis. These results demonstrate that cyclooxygenase products mediate the development of complement-induced hypoxemia but are not sole mediators of pulmonary hypertension or increases in vascular permeability. Furthermore, ibuprofen has anti-inflammatory actions, not shared by meclofenamate, which enhance the effectiveness of this drug. Since activated leukocytes release reactive oxygen metabolites, we treated sheep with superoxide dismutase (2800 U/kg per hour) to determine the role of superoxide anions in these responses. This treatment significantly attenuated the increases in lung lymph flow and protein clearance. The results suggest that multiple mediators, which may originate from activated leukocytes sequestered in the pulmonary circulation, contribute to the pathophysiological changes seen with intermittent complement activation. Cyclooxygenase products of arachidonic acid contribute to the hypertension and are solely responsible for the hypoxemia. Reactive oxygen metabolites are important mediators of the complement-induced increases in lung vascular permeability.

The relationship between pulmonary perivascular cuff fluid and lung lymph in dogs with edema

Abstract We studied the relationship between pulmonary perivascular cuff fluid and lymph in dogs with alloxan-induced edema. Evans blue dye was injected intravenously into edematous dogs. We allowed the dye-albumin complex to equilibrate for 15 to 60 min. Morphological examination of freeze-dried lung sections showed the presence of dye in lymphatic vessels during every time period studied. We found less dye in perivascular fluid cuffs relative to lymphatic vessels at all times. The amount of dye contained in perivascular cuff fluid varied as a function of equilibration time and vessel size. In dogs with equilibration times of less than 30 min there was significantly less dye in perivascular fluid cuffs of large vessels relative to those of small vessels. There was no difference in cuff color between small and large vessels in dogs with equilibration times of 45 to 60 min. Both small and large vessels showed significantly less dye at 15- to 30-min than at 45- to 60-min equilibration times. We conclude that perivascular cuff fluid is a sequestered pool that mixes with the microvascular filtrate by diffusion of solute and water through the interstitial space. The diffusion distance is determined by cuff diameter and proximity of the cuff to exchange vessels.

Effect of lung inflation on perivascular cuff fluid volume in isolated dog lung lobes.

Isolated, degassed dog lung lobes were suspended in a sealed container. We filled the lobes with saline to a lobe fluid pressure of 20 cm H2O. The pressure inside the container was adjusted so the lobes were inflated to transpulmonary pressures (PL) of 2 to 20 cm H2O. After 15 min of fluid inflation, the lobes were frozen in liquid nitrogen. We measured the volume of perivascular cuff fluid relative to the blood volume by a point counting procedure. In addition, we measured lobe hemoglobin content, wet weight, and dry weight. We found that perivascular cuff fluid volume increased as PL increased. There were significant (p < 0.05) increases in cuff fluid volume as PL doubled from 5 to 10 and from 10 to 20 cm H2O. The largest increases in cuff volume/cm H2O increase in PL were observed in the pressure range from 2 to 10 cm H2O. Previous investigators demonstrated that radial traction exerted by the expanding lung parenchyma results in an increase in volume of the extraalveolar vessels with lung inflation. We suggest that the same mechanism explains the measured increase in perivascular cuff fluid volume occurring with lung inflation. The relationship between perivascular interstitial fluid volume and transpulmonary pressure suggests that the compliance of the extraalveolar interstitial space decreased as the interstitial volume increased. The maximum perivascular cuff fluid volume measured in these experiments was 2.6 ml/g dry blood-free lung, which suggests that, in pulmonary edema, a substantial volume of fluid can accumulate in the perivascular interstitial space. We conclude that a decrease in interstitial fluid pressure around extraalveolar vessels with static lung inflation resulted in an increased perivascular cuff fluid volume.

In vivo labeling of neutrophils using a fluorescent cell linker.

To study neutrophil circulation time and trafficking in vivo requires labeling the cells so that their movement can be followed temporally and spatially. Labeling procedures used to date, however, have relied on ex vivo separation and labeling methods, an undesired consequence of which may be neutrophil activation. Moreover, the labeled cells preferentially stick in the pulmonary circulation for several hours upon reinjection into the host. Therefore, we devised an alternate labeling procedure that relied on intra‐arterial infusion of the fluorescent phagocytic cell linker PKH26. We infused PKH26 (5.5 × 10‐6 M for 1 h) into six awake sheep and measured systemic and pulmonary hemodynamics and lung lymph dynamics continuously for 3–4 h after the infusion. We collected simultaneous arterial and venous blood samples hourly for 4 h, and again 24 h after the infusion ended, for white blood cell and neutrophil differential counts and to identify labeled cells in fresh smears by fluorescence and differential interference contrast (DIC) microscopy. Infusion of PKH26 had minimal and transient physiologic effects on systemic and pulmonary artery pressure, lung lymph flow, and leukocyte counts. Labeled cells were present in venous blood after the infusion for at least 24 h. DIC microscopy observation of the blood smears indicated that the fluorescently labeled cells were exclusively neutrophils. Unstimulated superoxide anion release from neutrophils isolated 2 and 24 h after PKH26 infusion was not different from baseline release. Phorbol myristate acetate‐stimulated release was not affected either. The labeled neutrophils responded to chemoattractants by migrating to extravascular sites. Our results indicate that neutrophils can be selectively labeled in vivo, after which trafficking of the labeled neutrophils can be determined.

Biochemical analysis of the activation of adherent neutrophils in vitro

The role played by neutrophil oxidative responses in host defense and injury is an area of active investigation. In order to study neutrophil responses in vitro, methods are required for cell purification, enumeration, and quantification of activation responses, which mimic the in vivo situation as closely as possible. In this communication (and its companion paper, Albertine et al., 1988) improved methods for all of these tasks are described and applied to investigate neutrophil structure-function relationships in vitro and in vivo. Human neutrophils were purified by using a series of platelet-poor plasma-Percoll gradients (51, 62, 76 and 80% in Percoll). This modification of previously published procedures results in consistently successful neutrophil purification and has allowed us to purify neutrophils from bronchoalveolar lavage fluid as well as blood. Activation of human and sheep neutrophils (superoxide anion production) was quantitated by the reduction of ferricytochrome c using a microtiter plate reader to measure the increase in absorbance at 550 nm from adherent neutrophils. Adherence of neutrophils was quantitated by measurement of LDH in cells lysed with Triton X-100 using a new method which uses readily available commercial reagents and can quantitate the LDH content of as few as 5000 neutrophils (or the LDH released from 5% of 100,000 neutrophils). Assay conditions for superoxide anion were optimized, limitations both in assay design and instruments used to measure OD were explored and enumerated, and these methods were used to quantitate sheep and human neutrophil activation responses. Using methods described in Albertine et al. (1988) for fixing neutrophils in microtiter wells after assay of their functional capacity, we have studied the same cells functionally and morphologically. We have used these techniques to study blood and alveolar neutrophils from a patient with acute respiratory failure. His alveolar neutrophils displayed 67% of the activation response as peripheral neutrophils (4.31 +/- 0.12 nmol superoxide released per 250,000 neutrophils at 60 min vs. 6.38 +/- 0.18 in blood, P less than 0.01) and structural changes which suggested previous activation in vivo. These studies demonstrate that similar morphological changes are observed in neutrophils activated with phorbol myristate acetate in vitro, as are observed in cells which have been activated by pathophysiologic processes in vivo.

Prostaglandin E1 prevents increased lung microvascular permeability during intravascular complement activation in sheep.

Prostaglandin E1 (PGE1) inhibits a variety of functions of activated neutrophils including respiratory burst, release of leukotriene B4, and adherence to endothelial cells. To determine if PGE1 alters the pathophysiology of complement-induced lung vascular injury, experiments were conducted in anesthetized sheep with lung lymph fistulas given a 1-hour infusion of zymosan-activated plasma. PGE, (30 ng/min/kg) or its saline vehicle was infused intravenously for 90 minutes beginning 30 minutes before the infusion of activated plasma. PGE, had no effect on leukocyte count, the initial hypoxemia and thromboxane A2 release, or the development of acute pulmonary hypertension. However, PGE, prevented steady-state increases in lung lymph flow that in vehicle-treated sheep signaled an increase in lung microvascular permeability. Furthermore, extraction of PGE1 by pulmonary endothelial cells was unaffected by the infusion of activated plasma. We propose that PGE1 prevented the increase in lung vascular permeability by inhibiting adherence of activated neutrophils to endothelial cells.

Dialysis-Induced Hypoxemia: Membrane Dependent and Membrane Independent Causes

Hypoxemia during hemodialysis may result from several differing processes. We initially studied patients undergoing standard acetate hemodialysis. At 15 minutes of dialysis, leukopenia (primarily neutropenia), a decline of platelet count, and hypoxemia occurred, but without a significant change in mean minute ventilation. Complement activation (V/A ratios of C5a greater than 1.0) persisted throughout dialysis. Leukocyte count returned to baseline by one hour. To separate the effects of solute and/or gas fluxes from those of blood-membrane interaction we studied changes in Po2, WBC, C5a, TxB2, and PGI2 during a period of blood membrane interaction without dialysis, and during subsequent acetate dialysis. Patients were studied with both polyacrylonitrile (PAN) and cuprophan membranes containing different priming solutions during membrane contact alone. Despite leukopenia and complement activation, hypoxemia failed to occur during membrane contact alone. At 15 minutes of subsequent acetate dialysis, significant hypoxemia occurred with both membranes. However, the degree of hypoxemia was twice as great with a cuprophan membrane primed with acetate (18.6 +/- 3.3 mm Hg) compared with air or bicarbonate (9.1 +/- 1.4 and 7.0 +/- 2.0 mm Hg, respectively), or compared with PAN (8 +/- 2.8 mm Hg). Changes in thromboxane B2, PGI2, and C5a did not correlate with changes in Po2. We conclude that there are two major components to dialysis related hypoxemia. One is membrane independent, and may relate to the metabolic effects of acetate or to dialyzer CO2 loss. The remaining portion is membrane dependent, occurring with cuprophan, but not with PAN, and is conditioned by an acetate dependent interaction between blood and membrane.

Aggregation and thromboxane synthesis and release in isolated sheep neutrophils and lymphocytes in response to complement stimulation

In an attempt to determine possible cellular sources of thromboxane synthesis and release during intravascular complement activation, we prepared pure, viable fractions of neutrophils and lymphocytes from sheep blood. Lymphocytes were also isolated from thoracic duct lymph. Blood neutrophils and lymphocytes release thromboxane and the cells aggregate when challenged with zymosan-activated plasma. Blood lymphocytes release thromboxane in greater quantity than do neutrophils with the complement stimulus. Lymph lymphocytes, when stimulated with zymosan-activated plasma, did not release thromboxane or aggregate. These data suggest that both blood neutrophils and lymphocytes are likely sources of complement-initiated thromboxane synthesis and release in sheep.

Autoregulation of human neutrophil activation in vitro: regulation of phorbol myristate acetate-induced neutrophil activation by cell density.

Phorbol myristate acetate (PMA, 10‐7 M) activation of adherent neutrophils (PMNs) led to a markedly attenuated release of superoxide anion (O2 ‐) per cell when PMNs were activated at high density (2.85 fmol O2 ‐/PMN at 2 million in 0.1 ml) in comparison with cells activated at low cell density (12.0 fmol O2 ‐/PMN at 250,000 in 0.1 ml). This “autoregulatory” phenomenon was not due to a defect in the superoxide anion assay employed, to a differential adherence of neutrophils at high vs. low density, or to substrate (cytochrome c) or cell stimulus (PMA) limitation. It was associated with an inhibition of apparent NADPH oxidase activity and a leftward shift (toward a lower level of activation) in the activation profile of PMNs (as determined by FACS analysis using PMNs preloaded with 2′7′‐dichlorofluorescin diacetate in which H2O2 production results in the production of the fluorescent product 2′7‐dichlorofluorescein intracellularly). Other aspects of the neutrophil activation response including arachidonic acid mobilization, phospholipid metabolism, and perhaps phosphatidylinositol turnover were also attenuated when PMNs were activated at high cell density. Studies with cells in solution, cells treated with cycloheximide, and cells treated with 4,4′‐diisothiocyanostilbene‐2,2′‐disulfonic acid suggest that PMN contact with a surface, neutrophil protein synthesis, and an increased surface expression of the heterodimer CD11b/CD18 on PMNs all were not required for autoregulation. Finally, morphometric and morphologic examination of PMNs activated at low vs. high density revealed histologic and structural correlates associated with the attenuated PMN activation response of cells triggered at high cell density. We conclude that multiple structural and functional aspects of the PMN activation response are modulated by cell density and suggest that this property is important both in the physiologic control of neutrophil activation and in the design of in vitro assays of the neutrophil activation response.

Superoxide anion release from blood and bone marrow neutrophils is altered by endotoxemia.

In vivo endotoxin infusion produces neutrophil-mediated acute lung injury and increases superoxide anion release from phorbol myristate acetate (PMA)-stimulated blood neutrophils collected 18-24 hours after the infusion. Because the turnover time of circulating blood neutrophils is only 6-8 hours, it was hypothesized that the prolonged increase in superoxide anion release from peripheral blood neutrophils is associated with increased superoxide anion release from bone marrow neutrophils. To test this hypothesis, two doses of Escherichia coli endotoxin (5.0 and 0.5 micrograms/kg) were infused into chronically instrumented awake sheep. Blood and bone marrow neutrophils were collected 24 hours after the infusion, and superoxide anion release from unstimulated and PMA-stimulated neutrophils was measured in vitro. Endotoxin infusion produced an increase in pulmonary microvascular permeability, in intravascular activation (degranulation) of blood neutrophils, and in circulating blood neutrophils 24 hours after the infusion. High-dose endotoxin (5.0 micrograms/kg; n = 4) increased superoxide anion release from unstimulated peripheral blood neutrophils (2.25 +/- 0.38 times baseline [p less than or equal to 0.05]) and from peripheral blood neutrophils stimulated with 10(-9) M PMA in vitro (1.46 +/- 0.55 times baseline). Low-dose endotoxin (0.5 micrograms/kg; n = 5), on the other hand, did not alter superoxide anion release from peripheral blood neutrophils. Bone marrow neutrophils could not be isolated reproducibly after high-dose endotoxin because of leukoaggregation. Bone marrow neutrophils were isolated after low-dose endotoxin infusion. Stimulation of these cells with 10(-9) M PMA in vitro resulted in a two- to fourfold increase above control release (p less than or equal to 0.05). Increased superoxide anion release from both peripheral blood and bone marrow neutrophils occurred in the absence of circulating endotoxin, as measured by a Limulus assay. These results show that the prolonged increase in superoxide anion release from peripheral blood neutrophils is associated with an increase in the superoxide anion release from bone marrow neutrophils. Furthermore, the recruitment of affected bone marrow neutrophils into peripheral blood may explain the increased superoxide anion release from blood neutrophils 24 hours after endotoxin infusion.