Phagocytosis and ATP Levels in Alveolar Macrophages during Acute Hypoxia Sandra K. Leeper-Woodford and John W. Mills Departments of Physiology and Anatomy, Dartmouth Medical School, Hanover, New Hampshire
Pulmonary alveolar macrophages (PAM) function as phagocytes of inhaled particulate matter and microorganisms at the air-tissue interface of lung alveoli. Changes in cellular ATP concentrations ([ATP]) and phagocytic function during acute hypoxia may be important in conditions associated with low alveolar O2 • We proposed that acute hypoxia would decrease phagocytosis and reduce [ATP] in freshly isolated PAM. Phagocytic function (fluorescent microscopic technique determining percent phagocytosis in live cells) was monitored by recording uptake and retention of glutaraldehyde-fixed red blood cells (GRBC) in isolated rabbit PAM during acute incubations in air (20% O2 ) or hypoxia (1.7% O2 ) . Macrophage [ATP] were determined spectrophotometrically. Acute hypoxia for 30 to 150 min decreased phagocytic function 30 to 56 % in PAM without significantly affecting cell adherence and viability. Pre-exposure of PAM to hypoxia before addition of GRBC resulted in an even greater reduction in phagocytosis (97 % decrease by 30 min), and recovery of phagocytic function occurred 60 to 90 min after returning PAM to air. The cellular retention of phagocytosed GRBC (percentage of PAM with GRBC and number of GRBC/PAM) was reduced 30 % by 1 h of hypoxia. Compared with [ATP] of PAM in air, [ATP] of PAM exposed to hypoxia were reduced 55 and 35% at 30 and 60 min, respectively. Compared with [ATP] of cells with GRBC in air at 0 and 30 min, PAM with GRBC in hypoxia for 30 min had, respectively, 61 and 40% lower [ATP]. By 60 min with GRBC, PAM [ATP] in air and hypoxia were similar but were 50% lower than [ATP] at time O. These data suggest that acute hypoxia may alter cellular [ATP], inhibit phagocytosis, and reduce retention of phagocytosed particles in freshly isolated rabbit PAM.
With its location in the oxygen-rich environment of lung airways, the pulmonary alveolar macrophage (PAM) may utilize primarily oxidative phosphorylation for energy production in activities such as phagocytosis (1-5) because inhibitors of oxidative metabolism depress phagocytosis in PAM but not in other monocytes (6-9). The relative activities of enzymes in oxidative phosphorylation and glycolysis in PAM seem best suited for environments that are well oxygenated (6-9). Simon and co-workers (9) and Butterick and colleagues (6) measured the key glycolytic enzymes, pyruvate kinase and phosphofructokinase, as well as the terminal electron transport chain enzyme, cytochrome oxidase, in rabbit and guinea pig PAM and found that the lung cells had greater O2 utilization, higher activities of cytochrome oxidase, and lower glycolytic enzyme activities than did peritoneal macrophages. PAM may therefore have high rates of aerobic energy production and utilization when compared
(Received in original form April 22, 1991 and in revised form September 23, 1991) Address correspondence to: Sandra K. Leeper-Woodford, Ph. D., Assistant Professor, School of Medicine, Mercer University, Macon, GA 31207. Abbreviations: ATP concentration(s), [ATP]; Dulbecco's minimal essential medium, DMEM; glutaraldehyde-fixed red blood cells, GRBC; Krebs Ringer bicarbonate, KRB; pulmonary alveolar macrophage(s), PAM; phosphate-buffered saline, PBS; trichloroacetic acid, TCA. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 326-334, 1992
with other similar cells and may depend primarily on cytochrome-linked respiration and energy provided by oxidative phosphorylation for maximal particle ingestion (2, 3, 6,7). A major part of the O 2 required for cellular metabolism is used for production of high-energy phosphate bonds, or ATP, by the process of oxidative phosphorylation (10, 11). Steady-state concentrations of ATP ([ATP]) provide cells with sensitive mechanisms to control and regulate functional activities, and a balance exists between energy-utilizing and energy-generating processes (5-7, 10-12). Both the ATPgenerating processes and the ATP-requiring processes can be slowed or accelerated in the normally functioning cell when conditions in the cell change (5-7, 10-12). The complex series of events leading to engulfment of particles utilizes ATP derived from either aerobic oxidation of trichloroacetic acid (TCA) cycle intermediates or from anaerobic glycolysis (4, 13-17). The contractile apparatus of the cell, including actin and myosin, is thought to be the primary consumer of ATP during phagocytosis (15, 17). However, how much ATP may be required for the phagocytic process has not been clearly outlined from past investigations (10-12). The high-energy compounds, such as ATp, may be reduced with exposure of cells to low O 2 conditions because increased demand is placed on anaerobic metabolism, a less efficient system for providing ATP (10). With a decline in ATP production in the cell during hypoxia, one might expect functional activities such as phagocytosis to decrease (10, 11,
Leeper-Woodford and Mills: Acute Hypoxia and Phagocytosis in Pulmonary Macrophages
18, 19). Although earlier studies (1,8, 18, 19) have suggested that alterations in PAM phagocytosis may occur with hypoxic exposure, defined studies on phagocytic function and cellular [ATP] during periods of acute hypoxia are lacking. Studies investigating alveolar macrophages exposed to hypoxia have tested cells hours or days after isolation from lungs, or have used macrophages from long-term cell cultures (1, 5-9, 11). Because PAM normally reside in a relatively aerobic environment (1-3,5,6, 8, 9), investigators examining the effects of low O2 levels on PAM must be particularly aware of how the hypoxic cell culture environment itself may alter functional activities in these cells (5-7). We therefore designed our studies to examine the effects of acute hypoxia on phagocytosis and other functional parameters in PAM isolated for less than 24 h. We proposed that acute hypoxic exposure decreases [ATP] and particle phagocytosis in freshly isolated rabbit PAM. Macrophage phagocytic function and [ATP] were monitored in vitro during periods of acute hypoxia by observing PAM uptake and retention of glutaraldehyde-fixed red blood cells (GRBC) and by determining the [ATP] in PAM exposed to low O2 • In addition, we tested the viability and adherence ability of PAM during acute hypoxia and the recovery of phagocytic function in these cells after exposure to hypoxia.
Materials and Methods Cell Isolation Procedures PAM were obtained by lung lavage from New Zealand white rabbits (2 to 5 kg) according to the method of Myrvik and associates (20). In each rabbit, the lung washings removed by the lavage procedure were centrifuged (400 x g, 10 min), pooled, washed, and resuspended in phosphate-buffered saline (PBS). Macrophages in each population were identified by cell size and shape, and viability of the cells was determined by their ability to exclude trypan blue dye (21, 22). Purification and cultivation of the alveolar macrophages was done according to Edelson and Cohn (22) by suspending the cells in Dulbecco's minimal essential medium (DMEM) with 10% fetal bovine serum (GIBCO, Grand Island, NY) and placing the suspension (1 x 106 cells/dish) on glass cover slips in 35-mm plastic culture dishes (Corning Glass Works, Corning, NY). Cells were incubated (95% air/5% CO 2 , 37° C) for 16 to 20 h to allow adherence of the alveolar macrophages in the culture dishes. Nonadherent cells and DMEM were then decanted, and the adherent cells were rinsed twice with warm (37° C, pH 7.4) Krebs Ringer bicarbonate (KRB). The experiments that followed the above procedures were completed within 24 h of PAM isolation from the rabbits. The nonspecific esterase method according to Koski and co-workers (23) and the acid phosphatase enzyme staining technique using the Sigma Diagnostic Kit for cytologic demonstration of phosphatase acid in leukocytes (Sigma Chemical Co., St. Louis, MO) were both used periodically to determine that > 98 % macrophages were present in the isolated, adhered cell populations. Incubation of PAM in Air and Hypoxia DMEM over the adhered PAM was replaced with KRB equilibrated with either 95 % air/5 % CO2 or 95 % N2/5 %
327
CO2 , and the dishes with cells were placed in humidified modular incubation chambers (Flow Laboratories, McLean, VA). The chambers were sealed and flushed with the appropriate gas mixture for 10 min. The sealed modular chambers were incubated (37° C) for specific periods of time. In all experiments, exposure of cells to 95% air/5% CO2 is referred to as air exposure, whereas hypoxic exposure refers to cells in an environment of 95 % N2/5 % CO2 , The percentage of O2 and CO 2 in the modular incubation chambers was periodically tested using Beckman gas analyzers. In the chambers incubated with air, the O2 level was 20.0 ± 0.0% and CO2 was 3.9 ± 0.4 %, whereas in the hypoxic chambers the O2 was 1.7 ± 0.4% and CO2 was 4.3 ± 0.3%. Phagocytic Assay The method used to measure percent phagocytosis of inert particles by PAM is a modified version of the procedure of Loike and Silverstein (24). This method enables one to overcome the difficulty in determining which particles are adhered to macrophage surfaces and which are actually phagocytosed. GRBC were used as the inert particle to be phagocytosed by PAM. GRBC fluoresce a green color at about 585 nm when excited at 490 nm (see Figure 1). When 0.4 % trypan blue dye is in contact with GRBC, the fluorescent color is converted from green to red-orange (see Figures 1 and 2). Thus, with exposure to trypan blue, which does not penetrate live cells (21, 22), extracellular GRBC fluoresce red-orange, whereas GRBC that have been phagocytosed and are inside viable PAM fluoresce green (see Figures 1 and 2). GRBC were prepared by incubating at 4° C a 50% suspension (vol/vol) of ox red blood cells (Sterile Bovine BloodAlsevers; Kroy Medical, Inc., Wilfer Laboratories, Stillwater, MN) with 20 vol of2.5% glutaraldehyde (EM Grade; Polysciences, Warrington, PA) in PBS for 17 to 20 h until the red cells fluoresced a bright green. The prepared GRBC were stored as a 10% suspension (vol/vol) in PBS at 4° C until use (24). Effects of Hypoxia on Phagocytosis Two milliliters of fresh KRB equilibrated with either air or hypoxia was layered over adherent PAM, and 100 JLI of the GRBC suspension was added to each dish containing PAM. The GRBC-to-PAM ratio was 100:1. Mixing of the GRBC with the KRB solution was done at the appropriate time by gently tipping the chambers and dishes back and forth. After incubation, the following procedure was used to determine the percentage of macrophages that had phagocytosed GRBC: cells on the coverslips were exposed for 50 s to 100 JLl of 0.4 % trypan blue; the coverslip was rinsed in PBS and immediately mounted on a microscope slide. Cells were examined with a Zeiss photomicroscope using bright-field as well as fluorescent microscopy at an excitation wavelength of 490 nm and an emission spectrum above 520 nm. All cells were examined with 156x and 390x total magnification using light microscopy to count the number of PAM per coverslip and to determine their viability by trypan blue exclusion. Random areas of each coverslip were then examined using both light microscopy and fluorescence optics at 625 x total magnification. Photomicrographs were taken on Ek-
328
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
tachrome 35 mm, ASA 400 film (Eastman Kodak Co., Rochester, NY). The percentage of PAM with phagocytosed GRBC was defined as the number of PAM containing green-fluorescing GRBC divided by the total number of PAM per field observed. Cells containing at least one green GRBC were counted as PAM that had phagocytosed the GRBC particles. Duplicate coverslip samples from each rabbit PAM population were examined, and approximately 100 PAM were counted in random fields in all experiments at each of the time points observed. These studies were done using blind determinations of number of PAM per field and percent phagocytosis of these cells in each condition. In all experiments, PAM with GRBC were examined at time 0, and no phagocytosis was observed in the cells at this point. The number of PAM adhering to coverslips and the viability of these cells after exposure to air or hypoxia were recorded in the same field where phagocytosis was determined. Effects of Hypoxia on [ATP] in PAM In a separate series of experiments, the effects of hypoxia on [ATP] in PAM were measured spectrophotometrically using the Sigma ATP Diagnostic Kit (Sigma). In this ATP assay procedure, by measuring the decrease in absorbance at 340 nm that occurs when NADH is oxidized to NAD, a determination of the amount of ATP originally present in the cell sample is obtained (Sigma ATP Diagnostic Kit). Because PAM in the experiments presented here were adhered to small culture dishes and had to be scraped off the dishes before the assays were performed, several dishes of cells had to be pooled to obtain the necessary number of PAM for the appropriate levels of ATP needed for detection using this assay (5 to 100 1I-M/dl, Sigma ATP Diagnostic Kit). Preliminary studies determined that at least 4 X 106 macrophages were needed for each assay point in these experiments, and PAM from each population of isolated rabbit macrophages were scraped from the bottom of the culture dishes and counted with a Coulter Counter (Coulter Electronics, Hialeah, FL) to determine the number of dishes needed to yield 4 x lQ6 cells per assay point. After exposing the macrophages to air or hypoxia for 30 or 60 min, KRB was removed, PAM were scraped from the dishes into 1 ml of cold 12% teA, and the cold teA-cell mixture was centrifuged (3,000 rpm, 5 min) to obtain a clear supernatant. The ATP assay was performed on the clear supernatant in the following way: 1 rnl of 3-phosphoglycerate solution, 1.5 rnl of water, and 0.5 rnl of the clear supernatant from the centrifuged teA mixture were pipetted into a vial containing 0.3 mg NADH. The vial was capped, inverted, and decanted into a cuvette. The initial absorbance of the cell supernatant was read at 340 nm versus a water reference in a spectrophotometer (Gilford Instrument). The enzymes glyceraldehyde phosphate dehydrogenase and 3-phosphokinase were then pipetted into the cuvette, and the total contents were inverted to mix. The cuvette was replaced in the spectrophotometer, and absorbance readings were continued until a minimum absorbance reading was reached. Calculations were done by first subtracting the final minimum absorbance value from the initial absorbance value, i.e., change in A = initial A - final A. [ATP] per cell was calculated as follows:
Cell sample ATP (1I-mol/dl) Macrophage [ATP] (umol/cell)
= change =
in A
X
0.98.
change in A x 0.98 total cells
.
In
sample
The factor 0.98 was derived as follows: 3.04 0.98 = - - - 6.22 x 0.5 3.04 = total volume of liquid in cuvette in milliliters; 6.22 = millimolar absorptivity at 340 nm; 0.50 = clear supernatant sample volume in cuvette in milliliters. All cell samples were treated in the same way and measurements of [ATP] per macrophage were done at 0-, 30-, and 60-min time points, after exposure to air or hypoxia, without GRBC or with GRBC added at the 0 time point. Effects of Pre-exposure to Hypoxia on PAM Phagocytosis and Recovery of Phagocytic Function in Air The following methods were used in experiments to determine if exposing PAMto hypoxia 30 min before GRBC were added influenced the percentage of PAM phagocytosing GRBC. PAM were prepared and allowed to adhere to coverslips as described earlier. Small wells made from yellow plastic pipette tips (USAllO; USA/Scientific Plastics, Ocala, FL) were firmly placed with the flat bore side down onto the surface of each coverslip with PAM. One well was put into each dish. A total of 10011-1 of 10% GRBC suspension in PBS was pipetted into each well without disturbing the well position. Two milliliters of fresh KRB was then added to each of these culture dishes so that wells containing GRBC were not disturbed. Depositing GRBC into wells in this manner allowed for isolation of the GRBC particles from most of the PAM. Any leakage of the brown GRBC suspension into the clear KRB solution could be observed, and if this occurred the dish was discarded. The prepared dishes were placed in the modular incubation chambers. After sealing, the chambers were flushed with air or hypoxic gas and the cells incubated (370 C) for 30 min. After the 30-min incubation, the sealed modular chambers were gently rocked so that the wells with GRBC tipped to the side and allowed the GRBC suspension to escape and mix evenly with the KRB solution over the PAM. Incubation (3r C) of the cells was then continued in the sealed chambers for 30 min. The modular chambers were then opened, coverslips removed, and percent phagocytosis by PAM determined. The area of each coverslip that had been under the plastic well was determined by observing a dark ring of adhered GRBC over the PAM, and this area was not used for analysis. Periodically, PAM phagocytosis of GRBC was checked at 0 and 30 min to determine that the wells kept the GRBC suspension isolated from macrophages until the time of mixing. A series of experiments were designed to determine whether PAM exposed to hypoxia could recover phagocytic ability when returned to air. After 30 or 150 min of pre- . exposure to air or hypoxia, GRBC were mixed with PAM and cells that had been exposed to hypoxia were returned to air and incubated (37 C) for 30, 60, or 90 min. Percent phagocytosis was determined in all PAM. 0
Leeper-Woodford and Mills: Acute Hypoxia and Phagocytosis in Pulmonary Macrophages
Effects of Hypoxia on Retention of Phagocytosed Particles Loading of GRBC into macrophages was done by incubating (37 0 C) PAM in fresh DMEM with 100 pJ of GRBC in air for 60 min. The GRBC-loaded PAM were gently rinsed, layered with KRB, placed in either air or hypoxia, and incubated for another 60 min. Percent phagocytosis of GRBC by PAM was then determined using two different methods. The fluorescent method was used as described earlier to examine whole, live PAM. In addition, parallel sets of PAM were immediately fixed with 2.5 % glutaraldehyde, scraped from the dishes, postfixed in a 2 % aqueous solution of osmium tetroxide for 30 min, dehydrated through ethanol, and embedded in freshly prepared Spurr's embedding medium (Spurr-Low Viscosity Embedding Media Kit #14300; Electron Microscopy Sciences, Fort Washington, PA). Onemicron serial sections of the embedded cells were stained with toluidine blue stain and examined by light microscopy. More than 300 cells were examined in each condition in each of three rabbits. The number of PAM in each field and percentage of those PAM with more than two phagocytosed GRBC as well as the number of GRBC inside the fixed, sectioned macrophages were determined. In these experiments in which PAM were preloaded with GRBC, distinguishing one versus two phagocytosed GRBC was difficult; therefore, only PAM with more than two GRBC were counted as phagocytic. Results of the two methods, i.e., live fluorescent versus fixed stained techniques, were then compared. Statistical Analysis Results are presented as mean ± SEM, and n equals the number of rabbits used in each experiment. Duplicate samples of each population of rabbit PAM were used at every time point. Each rabbit supplied PAM for both control and experimental conditions in each individual experiment. In all experiments in which percent values were used, the means, SEM, and all statistical comparisons were computed after transforming percent values to arcsine values using arcsine transformation tables (25). ANOVA, one-way and paired t tests were used to determine the significant differences within and between groups, and significance levels in each experiment were determined according to the Bonferroni analysis method (26). This method was used to place limits on the chance of one or more type I errors occurring within each statistical test that was used within a family of related hypothesis tests (26). The levels of significance were between P < 0.01 and P < 0.0025 in all experiments.
Results Effects of Hypoxia on Phagocytosis Thirty minutes after incubation with GRBC, percent phagocytosis was the same in PAM incubated in either air or hypoxia, whereas at 60, 90, and 150 min, PAM incubated in hypoxia showed less percent phagocytosis of GRBC than cells in air (Figure 3). In addition, progressively lower percent phagocytosis was observed at successive time points in PAM incubated in hypoxia, whereas during the same time period, PAM in air maintained high percent phagocytosis values (Figure 3). The photomicrographs in Figures 4 and 5 illustrate how
329
PAM with GRBC appear after 60 min of incubation in air or hypoxia, respectively. Although many green-fluorescing GRBC were visible inside PAM incubated in air (Figure 4), PAM exposed to hypoxia exhibited few phagocytosed GRBC (Figure 5). In addition, lower numbers of PAM with phagocytosed GRBC were observed after hypoxia (Figure 5). The many red-fluorescing nonphagocytosed GRBC near and on the surfaces of the PAM in hypoxia indicated that cell contact with GRBC particles occurred, but that uptake of GRBC particles was less under hypoxic conditions (Figure 5). Observations were also made on GRBC-loaded PAM that died and were thus unable to exclude trypan blue in order to see how trypan blue reacted with GRBC particles inside nonviable PAM (Figure 6). Note that some GRBC retained their green fluorescence even after trypan blue was no longer excluded from nonviable PAM (Figure 6). The nucleus was stained by trypan blue and appears dark in dead PAM (Figure 6). This further demonstrates that the fluorescent method used is an adequate way to monitor phagocytosis in live cells, as the loss of cell viability (dark nucleus) appears before GRBC fluorescence is affected by trypan blue (Figure 6). Effects of Hypoxia on [ATP] in PAM The PAM incubated in air showed a rise in [ATP] per cell from 0 to 30 min of incubation (Figure 7). By 60 min, however, this rise in [ATP] was somewhat attenuated and [ATP] after 60 min in air were similar to those of cells at the 0 time point (Figure 7). The [ATP] of PAM exposed to hypoxia did not show a rise from 0 to 30 min like that observed in air (Figure 7). PAM incubated in hypoxia had [ATP] at both 30 and 60 min that were similar and no different than those at time 0 (Figure 7). Thus, when compared with the respective values in air, the [ATP] of PAM in hypoxia were significantly different at 30 and 60 min because of the increase in [ATP] of PAM in air rather than to a decrease in [ATP] of cells in hypoxia (Figure 7). Effects of Hypoxia on [ATP] in PAM during Phagocytosis The results in Figure 8 illustrate whether [ATP] of PAM incubated in air or hypoxia. are affected if GRBC are added when cells are placed in the respective environments. When compared with [ATP] observed in cells at time 0, PAM incubated with GRBC in either air or hypoxia had lower [ATP] at 60 min, whereas only the PAM incubated in hypoxia had lower [ATP] at 30 min (Figure 8). The [ATP] per PAM incubated in hypoxia rose during the 30- to 60-min time period, whereas PAM in air had [ATP] that were similar at the 30- and 60-min time points (Figure 8). When compared with the respective values of PAM in air, PAM incubated with GRBC in hypoxia for 30 min had [ATP] that were lower, whereas, at 60 min, PAM with GRBC in either air or hypoxia had similar [ATP] (Figure 8). Effects of Pre-exposure to Hypoxia on PAM Phagocytosis and Recovery of Phagocytic Function in Air There was only mimimal phagocytosis by PAM in either air or hypoxia (both air and hypoxia = 2.0 ± 2.0%) at 30 min, the time of GRBC mixing, indicating that GRBC remained isolated in the plastic wells during the pre-exposure time
330
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
Figure 1. Pulmonary alveolar macrophages (PAM) viewed with fluorescent microscopy after 60 min of incubation with glutaraldehydefixed red blood cells (GRBC). Note the many green GRBC particles (G) in or around PAM (P) and that it is difficult to distinguish those GRBC inside PAM from those outside PAM. Bar = 20 J.'m. Figure 2. PAM viewed with fluorescent microscopy after 60 min of incubation with GRBC and a 50-s exposure to trypan blue. Trypan blue is excluded from live PAM, and GRBC that have been phagocytosed are not exposed to trypan blue and remain green (G). The many red GRBC particles (R) are those that are not phagocytosed and have thus been exposed to trypan blue, which shifts GRBC fluorescence from green to red. A diffuse light green fluorescence is observed in live PAM (P) without any GRBC. Nonviable PAM (D) appear red. Bar = 20 J.'m. Figure 4. PAM incubated with GRBC in air for 60 min, as viewed with fluorescent microscopy following exposure to trypan blue. Note many green GRBC particles (G) in PAM, indicating these have been phagocytosed. The red PAM is a dead or dying cell (D). Bar = 20 J.'m.
Figure 5. PAM incubated with GRBC in hypoxia for 60 min as viewed with fluorescent microscopy following exposure to trypan blue. When compared with Figure 4, note the fewer green GRBC (G) within each PAM and the decreased number of PAM that contain phagocytosed GRBC (G). Many red, nonphagocytosed GRBC (R) are near or adhered to PAM in hypoxia, indicating these GRBC have not been phagocytosed. Bar = 20 J.'m. Figure 6. A nonviable PAM that had phagocytosed GRBC while it was alive but which is now unable to exclude trypan blue, as viewed with fluorescent microscopy. Note the dark nucleus (arrowhead) indicating that cell death had occurred, and the green GRBC (G) as well as red GRBC (R) within and around PAM. Magnification as in Figure 4.
(Figure 9). When PAM were incubated for 30 min in air or hypoxia, allowed to mix with GRBC in the sealed modular chambers and incubated in air or hypoxia until 60 min, PAM in hypoxia showed lower percent phagocytosis at 60 min than cells incubated in air (Figure 9). Phagocytosis of GRBC by
PAM preincubated 30 min in hypoxia then incubated with GRBC in air for another 30 min had percent phagocytosis values midway between those of PAM incubated exclusively in hypoxia or air (Figure 9). After 30 min with GRBC in air, PAM preincubated in
Leeper-Woodford and Mills: Acute Hypoxia and Phagocytosis in Pulmonary Macrophages
331
20
90 80
•
Air
a
Hypoxia
!!! 70
V>
•
g 50
a
~ 60
Air Hypoxia
~ 40
:J: Cl-
30
.
•
;I!.
•:I
§
10
o
20
40
60
80
100
120
20
140
30
40
50
60
TIME (MIN)
TIME (MIN)
Figure 3. Percent phagocytosis of GRBC by PAM versus time in air (closed circles) or hypoxia (open circles). Modular chambers were sealed until percent phagocytosis was determined at each time point. Values are mean ± SEM (n = 4 to 8 rabbits, > 400 PAM per point). * P < 0.0025, air versus hypoxia at respective time points; :j: P < 0.0025, air, 30 min versus air at successive time points; § P < 0.0025, hypoxia, 30 min versus hypoxia, 150 min.
hypoxia for 150 min showed lower percent phagocytosis than PAM preincubated 150 min in air (Figure 10). However, after 60 to 90 min with GRBC in air, PAM preincubated 150 min in either air or hypoxia had similar values of percent phagocytosis (Figure 10). Note that the PAM in air for 30 min in Figure 10 had lower percent phagocytosis than PAM in air at 30 min in Figure 3. This may indicate that PAM preincubated for 150 min in the non-nutrient KRB medium in Figure 10 have a slower rate of phagocytosis than cells just removed from the nutrient-rich DMEM in Figure 3. PAM Viability and Adherence during Hypoxia Viability and adherence determinations were made on every PAM examined in these studies. Cell viability in all experiments was > 90 % at all time points after exposure to air or hypoxia. Adherence of PAM to coverslips was statistically similar after incubation in air or hypoxia for 90 to 150 min (PAM per field, air = 45 ± 8 and 47 ± 9; hypoxia = 36 ± 5 and 43 ± 6, 90 and 150 min, respectively; P = NS, air versus hypoxia). Although hypoxic cells were slightly less adherent, the dramatically lower percent phagocytosis observed in PAM exposed to hypoxia in these experiments was probably not due to loss of PAM viability or to significant changes in adherence of PAM to coverslips.
Figure 8. ATP levels of PAM with GRBC in air (closed circles) or hypoxia (open circles). Values are mean ± SEM (n = 4 rabbits, 1.6 x 107 PAM assayed per time point). With the Bonferroni method, level of significance is P < 0.01; * P < 0.01, air versus hypoxia, 30 min; t P < 0.01, air, 0 versus 60 min; :j: P < 0.01, air, 0 min versus hypoxia, 30 and 60 min; and § P < 0.01, hypoxia, 30 versus 60 min.
Effects of Hypoxia on Retention of Phagocytosed GRBC Previously, we had found that phagocytosis was similar in PAM exposed to air or hypoxia for 30 min, with a steady decline in the number of PAM containing phagocytosed particles occurring between 30 and 150 min (Figure 3). In our subsequent experiments, we observed that PAM loaded with GRBC for 60 min in air (percent phagocytosis = 80.0 ± 3.0 %) and then exposed for 60 min to hypoxia had decreased phagocytosis when compared with PAM exposed to air for the same time (Figures 11 to 13). In Figure 11, percent phagocytosis of GRBC in PAM was determined in whole, live cells using the fluorescent method or by counting cells with GRBC in fixed, sectioned, and stained PAM. Both methods yielded similar results, i.e., PAM with phagocytosed GRBC had lower percent phagocytosis after 60 min in hypoxia (Figure 11). In sectioned and stained PAM, the number of GRBC per PAM was lower in PAM exposed to hypoxia than in those incubated in air (Figure 12). Figure 13 illustrates the distribution of number of GRBC per PAM. These curves demonstrate that the percentage of PAM containing zero, one, or two GRBC was higher in cells exposed to hypoxia, whereas 80 en
70
0
60
en
•
30
.....
Air
25
:J:
30
n,
E >e 20
•
5 0
Hypoxia 30 min, Air 30 min
0
IO
0
Hypoxia
~
;I!. 20 10
::< 15 ~ u
0
•
Air
a
10
20
30
• 40
50
60
TIME(MIN)
Figure 7. ATP levels of PAM exposed to air (closed circles) or hypoxia (open circles). Values are mean ± SEM (n = 4 rabbits, 1.6 X 107 PAM assayed per time point). With the Bonferroni method, level of significance is P < 0.01; * P < 0.01, hypoxia, 30 and 60 min compared with respective times in air; and t P < 0.01, air, 0 versus 30 min.
0
10
20
30
40
50
60
TIME (MIN)
Figure 9. Percent phagocytosis of GRBC by PAM in air (closed circles) or hypoxia (open circles). After pre-exposing PAM for 30 min to air or hypoxia, GRBC were mixed with PAM and incubations continued in the respective gases. In addition, percent phagocytosis by PAM pre-exposed to hypoxia for 30 min then incubated 30 min with GRBC in air is shown (closed triangles). Values are mean ± SEM (n = 6 rabbits, > 600 PAM per point). * P < 0.013, hypoxia versus air; :j: P < 0.013 air versus hypoxia/air; and § P < 0.013, hypoxia versus hypoxia/air; all at 60-min time point.
332
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
•
Air
a
80
4
Hypoxia
en 70
~ 60
3
t;
2
to-
o
•
5
90
50
~
40 5: 30 ;fl. 20
o
10
Air
o V~-----L_-L------'-_---'---_'-----'---_.L---L--' -150 0
10
20
30
40
50
60
70
80
90
TIME (MIN)
Figure 10. Percent phagocytosis of GRBC by PAM in air following pre-exposure to air (closed circles) or hypoxia (open circles). After 150 min of incubation of PAM in air or hypoxia, GRBC were added to PAM and preparations incubated in air. Values are mean ± SEM (n = 5 to 7 rabbits, > 500 PAM per point). * P < 0.01, air versus hypoxia, 30 min; :j: P < om, air versus air and § P < 0.01, hypoxia versus hypoxia at successive time points.
PAM incubated in air were more likely to have larger numbers of GRBC in each PAM (Figure 13).
Discussion Previous investigators using cultured alveolar macrophages have suggested that PAM phagocytic function is altered by low O2 levels (1, 5-9, 11). Oren and co-workers (8) found that cultured alveolar macrophages incubated for 1 h in 100% N2 phagocytosed less polystyrene or starch particles than did monocytes. Cohen and Cline (1) found that isolated, cultured human alveolar macrophages preincubated in 3 % O2 for 1 h phagocytosed 39% less heat-killed Candida albicans. In our studies, acute hypoxia (1.7% O2) altered phagocytic function in rabbit PAM by decreasing uptake of GRBC particles by the freshly isolated pulmonary cells (Figures 3, 4, 5, and 9). The time at which GRBC were FLUORESCENT
(f)
v;
l2 ~
u 0
~
«
:r::
Q.
Pulmonary alveolar macrophages (PAM) function as phagocytes of inhaled particulate matter and microorganisms at the air-tissue interface of lung alveoli. Changes in cellular ATP concentrations ([ATP]) and phagocytic function during acute hypoxia may be important in conditions associated with low alveolar O2 • We proposed that acute hypoxia would decrease phagocytosis and reduce [ATP] in freshly isolated PAM. Phagocytic function (fluorescent microscopic technique determining percent phagocytosis in live cells) was monitored by recording uptake and retention of glutaraldehyde-fixed red blood cells (GRBC) in isolated rabbit PAM during acute incubations in air (20% O2 ) or hypoxia (1.7% O2 ) . Macrophage [ATP] were determined spectrophotometrically. Acute hypoxia for 30 to 150 min decreased phagocytic function 30 to 56 % in PAM without significantly affecting cell adherence and viability. Pre-exposure of PAM to hypoxia before addition of GRBC resulted in an even greater reduction in phagocytosis (97 % decrease by 30 min), and recovery of phagocytic function occurred 60 to 90 min after returning PAM to air. The cellular retention of phagocytosed GRBC (percentage of PAM with GRBC and number of GRBC/PAM) was reduced 30 % by 1 h of hypoxia. Compared with [ATP] of PAM in air, [ATP] of PAM exposed to hypoxia were reduced 55 and 35% at 30 and 60 min, respectively. Compared with [ATP] of cells with GRBC in air at 0 and 30 min, PAM with GRBC in hypoxia for 30 min had, respectively, 61 and 40% lower [ATP]. By 60 min with GRBC, PAM [ATP] in air and hypoxia were similar but were 50% lower than [ATP] at time O. These data suggest that acute hypoxia may alter cellular [ATP], inhibit phagocytosis, and reduce retention of phagocytosed particles in freshly isolated rabbit PAM.
With its location in the oxygen-rich environment of lung airways, the pulmonary alveolar macrophage (PAM) may utilize primarily oxidative phosphorylation for energy production in activities such as phagocytosis (1-5) because inhibitors of oxidative metabolism depress phagocytosis in PAM but not in other monocytes (6-9). The relative activities of enzymes in oxidative phosphorylation and glycolysis in PAM seem best suited for environments that are well oxygenated (6-9). Simon and co-workers (9) and Butterick and colleagues (6) measured the key glycolytic enzymes, pyruvate kinase and phosphofructokinase, as well as the terminal electron transport chain enzyme, cytochrome oxidase, in rabbit and guinea pig PAM and found that the lung cells had greater O2 utilization, higher activities of cytochrome oxidase, and lower glycolytic enzyme activities than did peritoneal macrophages. PAM may therefore have high rates of aerobic energy production and utilization when compared
(Received in original form April 22, 1991 and in revised form September 23, 1991) Address correspondence to: Sandra K. Leeper-Woodford, Ph. D., Assistant Professor, School of Medicine, Mercer University, Macon, GA 31207. Abbreviations: ATP concentration(s), [ATP]; Dulbecco's minimal essential medium, DMEM; glutaraldehyde-fixed red blood cells, GRBC; Krebs Ringer bicarbonate, KRB; pulmonary alveolar macrophage(s), PAM; phosphate-buffered saline, PBS; trichloroacetic acid, TCA. Am. J. Respir. Cell Mol. BioI. Vol. 6. pp. 326-334, 1992
with other similar cells and may depend primarily on cytochrome-linked respiration and energy provided by oxidative phosphorylation for maximal particle ingestion (2, 3, 6,7). A major part of the O 2 required for cellular metabolism is used for production of high-energy phosphate bonds, or ATP, by the process of oxidative phosphorylation (10, 11). Steady-state concentrations of ATP ([ATP]) provide cells with sensitive mechanisms to control and regulate functional activities, and a balance exists between energy-utilizing and energy-generating processes (5-7, 10-12). Both the ATPgenerating processes and the ATP-requiring processes can be slowed or accelerated in the normally functioning cell when conditions in the cell change (5-7, 10-12). The complex series of events leading to engulfment of particles utilizes ATP derived from either aerobic oxidation of trichloroacetic acid (TCA) cycle intermediates or from anaerobic glycolysis (4, 13-17). The contractile apparatus of the cell, including actin and myosin, is thought to be the primary consumer of ATP during phagocytosis (15, 17). However, how much ATP may be required for the phagocytic process has not been clearly outlined from past investigations (10-12). The high-energy compounds, such as ATp, may be reduced with exposure of cells to low O 2 conditions because increased demand is placed on anaerobic metabolism, a less efficient system for providing ATP (10). With a decline in ATP production in the cell during hypoxia, one might expect functional activities such as phagocytosis to decrease (10, 11,
Leeper-Woodford and Mills: Acute Hypoxia and Phagocytosis in Pulmonary Macrophages
18, 19). Although earlier studies (1,8, 18, 19) have suggested that alterations in PAM phagocytosis may occur with hypoxic exposure, defined studies on phagocytic function and cellular [ATP] during periods of acute hypoxia are lacking. Studies investigating alveolar macrophages exposed to hypoxia have tested cells hours or days after isolation from lungs, or have used macrophages from long-term cell cultures (1, 5-9, 11). Because PAM normally reside in a relatively aerobic environment (1-3,5,6, 8, 9), investigators examining the effects of low O2 levels on PAM must be particularly aware of how the hypoxic cell culture environment itself may alter functional activities in these cells (5-7). We therefore designed our studies to examine the effects of acute hypoxia on phagocytosis and other functional parameters in PAM isolated for less than 24 h. We proposed that acute hypoxic exposure decreases [ATP] and particle phagocytosis in freshly isolated rabbit PAM. Macrophage phagocytic function and [ATP] were monitored in vitro during periods of acute hypoxia by observing PAM uptake and retention of glutaraldehyde-fixed red blood cells (GRBC) and by determining the [ATP] in PAM exposed to low O2 • In addition, we tested the viability and adherence ability of PAM during acute hypoxia and the recovery of phagocytic function in these cells after exposure to hypoxia.
Materials and Methods Cell Isolation Procedures PAM were obtained by lung lavage from New Zealand white rabbits (2 to 5 kg) according to the method of Myrvik and associates (20). In each rabbit, the lung washings removed by the lavage procedure were centrifuged (400 x g, 10 min), pooled, washed, and resuspended in phosphate-buffered saline (PBS). Macrophages in each population were identified by cell size and shape, and viability of the cells was determined by their ability to exclude trypan blue dye (21, 22). Purification and cultivation of the alveolar macrophages was done according to Edelson and Cohn (22) by suspending the cells in Dulbecco's minimal essential medium (DMEM) with 10% fetal bovine serum (GIBCO, Grand Island, NY) and placing the suspension (1 x 106 cells/dish) on glass cover slips in 35-mm plastic culture dishes (Corning Glass Works, Corning, NY). Cells were incubated (95% air/5% CO 2 , 37° C) for 16 to 20 h to allow adherence of the alveolar macrophages in the culture dishes. Nonadherent cells and DMEM were then decanted, and the adherent cells were rinsed twice with warm (37° C, pH 7.4) Krebs Ringer bicarbonate (KRB). The experiments that followed the above procedures were completed within 24 h of PAM isolation from the rabbits. The nonspecific esterase method according to Koski and co-workers (23) and the acid phosphatase enzyme staining technique using the Sigma Diagnostic Kit for cytologic demonstration of phosphatase acid in leukocytes (Sigma Chemical Co., St. Louis, MO) were both used periodically to determine that > 98 % macrophages were present in the isolated, adhered cell populations. Incubation of PAM in Air and Hypoxia DMEM over the adhered PAM was replaced with KRB equilibrated with either 95 % air/5 % CO2 or 95 % N2/5 %
327
CO2 , and the dishes with cells were placed in humidified modular incubation chambers (Flow Laboratories, McLean, VA). The chambers were sealed and flushed with the appropriate gas mixture for 10 min. The sealed modular chambers were incubated (37° C) for specific periods of time. In all experiments, exposure of cells to 95% air/5% CO2 is referred to as air exposure, whereas hypoxic exposure refers to cells in an environment of 95 % N2/5 % CO2 , The percentage of O2 and CO 2 in the modular incubation chambers was periodically tested using Beckman gas analyzers. In the chambers incubated with air, the O2 level was 20.0 ± 0.0% and CO2 was 3.9 ± 0.4 %, whereas in the hypoxic chambers the O2 was 1.7 ± 0.4% and CO2 was 4.3 ± 0.3%. Phagocytic Assay The method used to measure percent phagocytosis of inert particles by PAM is a modified version of the procedure of Loike and Silverstein (24). This method enables one to overcome the difficulty in determining which particles are adhered to macrophage surfaces and which are actually phagocytosed. GRBC were used as the inert particle to be phagocytosed by PAM. GRBC fluoresce a green color at about 585 nm when excited at 490 nm (see Figure 1). When 0.4 % trypan blue dye is in contact with GRBC, the fluorescent color is converted from green to red-orange (see Figures 1 and 2). Thus, with exposure to trypan blue, which does not penetrate live cells (21, 22), extracellular GRBC fluoresce red-orange, whereas GRBC that have been phagocytosed and are inside viable PAM fluoresce green (see Figures 1 and 2). GRBC were prepared by incubating at 4° C a 50% suspension (vol/vol) of ox red blood cells (Sterile Bovine BloodAlsevers; Kroy Medical, Inc., Wilfer Laboratories, Stillwater, MN) with 20 vol of2.5% glutaraldehyde (EM Grade; Polysciences, Warrington, PA) in PBS for 17 to 20 h until the red cells fluoresced a bright green. The prepared GRBC were stored as a 10% suspension (vol/vol) in PBS at 4° C until use (24). Effects of Hypoxia on Phagocytosis Two milliliters of fresh KRB equilibrated with either air or hypoxia was layered over adherent PAM, and 100 JLI of the GRBC suspension was added to each dish containing PAM. The GRBC-to-PAM ratio was 100:1. Mixing of the GRBC with the KRB solution was done at the appropriate time by gently tipping the chambers and dishes back and forth. After incubation, the following procedure was used to determine the percentage of macrophages that had phagocytosed GRBC: cells on the coverslips were exposed for 50 s to 100 JLl of 0.4 % trypan blue; the coverslip was rinsed in PBS and immediately mounted on a microscope slide. Cells were examined with a Zeiss photomicroscope using bright-field as well as fluorescent microscopy at an excitation wavelength of 490 nm and an emission spectrum above 520 nm. All cells were examined with 156x and 390x total magnification using light microscopy to count the number of PAM per coverslip and to determine their viability by trypan blue exclusion. Random areas of each coverslip were then examined using both light microscopy and fluorescence optics at 625 x total magnification. Photomicrographs were taken on Ek-
328
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
tachrome 35 mm, ASA 400 film (Eastman Kodak Co., Rochester, NY). The percentage of PAM with phagocytosed GRBC was defined as the number of PAM containing green-fluorescing GRBC divided by the total number of PAM per field observed. Cells containing at least one green GRBC were counted as PAM that had phagocytosed the GRBC particles. Duplicate coverslip samples from each rabbit PAM population were examined, and approximately 100 PAM were counted in random fields in all experiments at each of the time points observed. These studies were done using blind determinations of number of PAM per field and percent phagocytosis of these cells in each condition. In all experiments, PAM with GRBC were examined at time 0, and no phagocytosis was observed in the cells at this point. The number of PAM adhering to coverslips and the viability of these cells after exposure to air or hypoxia were recorded in the same field where phagocytosis was determined. Effects of Hypoxia on [ATP] in PAM In a separate series of experiments, the effects of hypoxia on [ATP] in PAM were measured spectrophotometrically using the Sigma ATP Diagnostic Kit (Sigma). In this ATP assay procedure, by measuring the decrease in absorbance at 340 nm that occurs when NADH is oxidized to NAD, a determination of the amount of ATP originally present in the cell sample is obtained (Sigma ATP Diagnostic Kit). Because PAM in the experiments presented here were adhered to small culture dishes and had to be scraped off the dishes before the assays were performed, several dishes of cells had to be pooled to obtain the necessary number of PAM for the appropriate levels of ATP needed for detection using this assay (5 to 100 1I-M/dl, Sigma ATP Diagnostic Kit). Preliminary studies determined that at least 4 X 106 macrophages were needed for each assay point in these experiments, and PAM from each population of isolated rabbit macrophages were scraped from the bottom of the culture dishes and counted with a Coulter Counter (Coulter Electronics, Hialeah, FL) to determine the number of dishes needed to yield 4 x lQ6 cells per assay point. After exposing the macrophages to air or hypoxia for 30 or 60 min, KRB was removed, PAM were scraped from the dishes into 1 ml of cold 12% teA, and the cold teA-cell mixture was centrifuged (3,000 rpm, 5 min) to obtain a clear supernatant. The ATP assay was performed on the clear supernatant in the following way: 1 rnl of 3-phosphoglycerate solution, 1.5 rnl of water, and 0.5 rnl of the clear supernatant from the centrifuged teA mixture were pipetted into a vial containing 0.3 mg NADH. The vial was capped, inverted, and decanted into a cuvette. The initial absorbance of the cell supernatant was read at 340 nm versus a water reference in a spectrophotometer (Gilford Instrument). The enzymes glyceraldehyde phosphate dehydrogenase and 3-phosphokinase were then pipetted into the cuvette, and the total contents were inverted to mix. The cuvette was replaced in the spectrophotometer, and absorbance readings were continued until a minimum absorbance reading was reached. Calculations were done by first subtracting the final minimum absorbance value from the initial absorbance value, i.e., change in A = initial A - final A. [ATP] per cell was calculated as follows:
Cell sample ATP (1I-mol/dl) Macrophage [ATP] (umol/cell)
= change =
in A
X
0.98.
change in A x 0.98 total cells
.
In
sample
The factor 0.98 was derived as follows: 3.04 0.98 = - - - 6.22 x 0.5 3.04 = total volume of liquid in cuvette in milliliters; 6.22 = millimolar absorptivity at 340 nm; 0.50 = clear supernatant sample volume in cuvette in milliliters. All cell samples were treated in the same way and measurements of [ATP] per macrophage were done at 0-, 30-, and 60-min time points, after exposure to air or hypoxia, without GRBC or with GRBC added at the 0 time point. Effects of Pre-exposure to Hypoxia on PAM Phagocytosis and Recovery of Phagocytic Function in Air The following methods were used in experiments to determine if exposing PAMto hypoxia 30 min before GRBC were added influenced the percentage of PAM phagocytosing GRBC. PAM were prepared and allowed to adhere to coverslips as described earlier. Small wells made from yellow plastic pipette tips (USAllO; USA/Scientific Plastics, Ocala, FL) were firmly placed with the flat bore side down onto the surface of each coverslip with PAM. One well was put into each dish. A total of 10011-1 of 10% GRBC suspension in PBS was pipetted into each well without disturbing the well position. Two milliliters of fresh KRB was then added to each of these culture dishes so that wells containing GRBC were not disturbed. Depositing GRBC into wells in this manner allowed for isolation of the GRBC particles from most of the PAM. Any leakage of the brown GRBC suspension into the clear KRB solution could be observed, and if this occurred the dish was discarded. The prepared dishes were placed in the modular incubation chambers. After sealing, the chambers were flushed with air or hypoxic gas and the cells incubated (370 C) for 30 min. After the 30-min incubation, the sealed modular chambers were gently rocked so that the wells with GRBC tipped to the side and allowed the GRBC suspension to escape and mix evenly with the KRB solution over the PAM. Incubation (3r C) of the cells was then continued in the sealed chambers for 30 min. The modular chambers were then opened, coverslips removed, and percent phagocytosis by PAM determined. The area of each coverslip that had been under the plastic well was determined by observing a dark ring of adhered GRBC over the PAM, and this area was not used for analysis. Periodically, PAM phagocytosis of GRBC was checked at 0 and 30 min to determine that the wells kept the GRBC suspension isolated from macrophages until the time of mixing. A series of experiments were designed to determine whether PAM exposed to hypoxia could recover phagocytic ability when returned to air. After 30 or 150 min of pre- . exposure to air or hypoxia, GRBC were mixed with PAM and cells that had been exposed to hypoxia were returned to air and incubated (37 C) for 30, 60, or 90 min. Percent phagocytosis was determined in all PAM. 0
Leeper-Woodford and Mills: Acute Hypoxia and Phagocytosis in Pulmonary Macrophages
Effects of Hypoxia on Retention of Phagocytosed Particles Loading of GRBC into macrophages was done by incubating (37 0 C) PAM in fresh DMEM with 100 pJ of GRBC in air for 60 min. The GRBC-loaded PAM were gently rinsed, layered with KRB, placed in either air or hypoxia, and incubated for another 60 min. Percent phagocytosis of GRBC by PAM was then determined using two different methods. The fluorescent method was used as described earlier to examine whole, live PAM. In addition, parallel sets of PAM were immediately fixed with 2.5 % glutaraldehyde, scraped from the dishes, postfixed in a 2 % aqueous solution of osmium tetroxide for 30 min, dehydrated through ethanol, and embedded in freshly prepared Spurr's embedding medium (Spurr-Low Viscosity Embedding Media Kit #14300; Electron Microscopy Sciences, Fort Washington, PA). Onemicron serial sections of the embedded cells were stained with toluidine blue stain and examined by light microscopy. More than 300 cells were examined in each condition in each of three rabbits. The number of PAM in each field and percentage of those PAM with more than two phagocytosed GRBC as well as the number of GRBC inside the fixed, sectioned macrophages were determined. In these experiments in which PAM were preloaded with GRBC, distinguishing one versus two phagocytosed GRBC was difficult; therefore, only PAM with more than two GRBC were counted as phagocytic. Results of the two methods, i.e., live fluorescent versus fixed stained techniques, were then compared. Statistical Analysis Results are presented as mean ± SEM, and n equals the number of rabbits used in each experiment. Duplicate samples of each population of rabbit PAM were used at every time point. Each rabbit supplied PAM for both control and experimental conditions in each individual experiment. In all experiments in which percent values were used, the means, SEM, and all statistical comparisons were computed after transforming percent values to arcsine values using arcsine transformation tables (25). ANOVA, one-way and paired t tests were used to determine the significant differences within and between groups, and significance levels in each experiment were determined according to the Bonferroni analysis method (26). This method was used to place limits on the chance of one or more type I errors occurring within each statistical test that was used within a family of related hypothesis tests (26). The levels of significance were between P < 0.01 and P < 0.0025 in all experiments.
Results Effects of Hypoxia on Phagocytosis Thirty minutes after incubation with GRBC, percent phagocytosis was the same in PAM incubated in either air or hypoxia, whereas at 60, 90, and 150 min, PAM incubated in hypoxia showed less percent phagocytosis of GRBC than cells in air (Figure 3). In addition, progressively lower percent phagocytosis was observed at successive time points in PAM incubated in hypoxia, whereas during the same time period, PAM in air maintained high percent phagocytosis values (Figure 3). The photomicrographs in Figures 4 and 5 illustrate how
329
PAM with GRBC appear after 60 min of incubation in air or hypoxia, respectively. Although many green-fluorescing GRBC were visible inside PAM incubated in air (Figure 4), PAM exposed to hypoxia exhibited few phagocytosed GRBC (Figure 5). In addition, lower numbers of PAM with phagocytosed GRBC were observed after hypoxia (Figure 5). The many red-fluorescing nonphagocytosed GRBC near and on the surfaces of the PAM in hypoxia indicated that cell contact with GRBC particles occurred, but that uptake of GRBC particles was less under hypoxic conditions (Figure 5). Observations were also made on GRBC-loaded PAM that died and were thus unable to exclude trypan blue in order to see how trypan blue reacted with GRBC particles inside nonviable PAM (Figure 6). Note that some GRBC retained their green fluorescence even after trypan blue was no longer excluded from nonviable PAM (Figure 6). The nucleus was stained by trypan blue and appears dark in dead PAM (Figure 6). This further demonstrates that the fluorescent method used is an adequate way to monitor phagocytosis in live cells, as the loss of cell viability (dark nucleus) appears before GRBC fluorescence is affected by trypan blue (Figure 6). Effects of Hypoxia on [ATP] in PAM The PAM incubated in air showed a rise in [ATP] per cell from 0 to 30 min of incubation (Figure 7). By 60 min, however, this rise in [ATP] was somewhat attenuated and [ATP] after 60 min in air were similar to those of cells at the 0 time point (Figure 7). The [ATP] of PAM exposed to hypoxia did not show a rise from 0 to 30 min like that observed in air (Figure 7). PAM incubated in hypoxia had [ATP] at both 30 and 60 min that were similar and no different than those at time 0 (Figure 7). Thus, when compared with the respective values in air, the [ATP] of PAM in hypoxia were significantly different at 30 and 60 min because of the increase in [ATP] of PAM in air rather than to a decrease in [ATP] of cells in hypoxia (Figure 7). Effects of Hypoxia on [ATP] in PAM during Phagocytosis The results in Figure 8 illustrate whether [ATP] of PAM incubated in air or hypoxia. are affected if GRBC are added when cells are placed in the respective environments. When compared with [ATP] observed in cells at time 0, PAM incubated with GRBC in either air or hypoxia had lower [ATP] at 60 min, whereas only the PAM incubated in hypoxia had lower [ATP] at 30 min (Figure 8). The [ATP] per PAM incubated in hypoxia rose during the 30- to 60-min time period, whereas PAM in air had [ATP] that were similar at the 30- and 60-min time points (Figure 8). When compared with the respective values of PAM in air, PAM incubated with GRBC in hypoxia for 30 min had [ATP] that were lower, whereas, at 60 min, PAM with GRBC in either air or hypoxia had similar [ATP] (Figure 8). Effects of Pre-exposure to Hypoxia on PAM Phagocytosis and Recovery of Phagocytic Function in Air There was only mimimal phagocytosis by PAM in either air or hypoxia (both air and hypoxia = 2.0 ± 2.0%) at 30 min, the time of GRBC mixing, indicating that GRBC remained isolated in the plastic wells during the pre-exposure time
330
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
Figure 1. Pulmonary alveolar macrophages (PAM) viewed with fluorescent microscopy after 60 min of incubation with glutaraldehydefixed red blood cells (GRBC). Note the many green GRBC particles (G) in or around PAM (P) and that it is difficult to distinguish those GRBC inside PAM from those outside PAM. Bar = 20 J.'m. Figure 2. PAM viewed with fluorescent microscopy after 60 min of incubation with GRBC and a 50-s exposure to trypan blue. Trypan blue is excluded from live PAM, and GRBC that have been phagocytosed are not exposed to trypan blue and remain green (G). The many red GRBC particles (R) are those that are not phagocytosed and have thus been exposed to trypan blue, which shifts GRBC fluorescence from green to red. A diffuse light green fluorescence is observed in live PAM (P) without any GRBC. Nonviable PAM (D) appear red. Bar = 20 J.'m. Figure 4. PAM incubated with GRBC in air for 60 min, as viewed with fluorescent microscopy following exposure to trypan blue. Note many green GRBC particles (G) in PAM, indicating these have been phagocytosed. The red PAM is a dead or dying cell (D). Bar = 20 J.'m.
Figure 5. PAM incubated with GRBC in hypoxia for 60 min as viewed with fluorescent microscopy following exposure to trypan blue. When compared with Figure 4, note the fewer green GRBC (G) within each PAM and the decreased number of PAM that contain phagocytosed GRBC (G). Many red, nonphagocytosed GRBC (R) are near or adhered to PAM in hypoxia, indicating these GRBC have not been phagocytosed. Bar = 20 J.'m. Figure 6. A nonviable PAM that had phagocytosed GRBC while it was alive but which is now unable to exclude trypan blue, as viewed with fluorescent microscopy. Note the dark nucleus (arrowhead) indicating that cell death had occurred, and the green GRBC (G) as well as red GRBC (R) within and around PAM. Magnification as in Figure 4.
(Figure 9). When PAM were incubated for 30 min in air or hypoxia, allowed to mix with GRBC in the sealed modular chambers and incubated in air or hypoxia until 60 min, PAM in hypoxia showed lower percent phagocytosis at 60 min than cells incubated in air (Figure 9). Phagocytosis of GRBC by
PAM preincubated 30 min in hypoxia then incubated with GRBC in air for another 30 min had percent phagocytosis values midway between those of PAM incubated exclusively in hypoxia or air (Figure 9). After 30 min with GRBC in air, PAM preincubated in
Leeper-Woodford and Mills: Acute Hypoxia and Phagocytosis in Pulmonary Macrophages
331
20
90 80
•
Air
a
Hypoxia
!!! 70
V>
•
g 50
a
~ 60
Air Hypoxia
~ 40
:J: Cl-
30
.
•
;I!.
•:I
§
10
o
20
40
60
80
100
120
20
140
30
40
50
60
TIME (MIN)
TIME (MIN)
Figure 3. Percent phagocytosis of GRBC by PAM versus time in air (closed circles) or hypoxia (open circles). Modular chambers were sealed until percent phagocytosis was determined at each time point. Values are mean ± SEM (n = 4 to 8 rabbits, > 400 PAM per point). * P < 0.0025, air versus hypoxia at respective time points; :j: P < 0.0025, air, 30 min versus air at successive time points; § P < 0.0025, hypoxia, 30 min versus hypoxia, 150 min.
hypoxia for 150 min showed lower percent phagocytosis than PAM preincubated 150 min in air (Figure 10). However, after 60 to 90 min with GRBC in air, PAM preincubated 150 min in either air or hypoxia had similar values of percent phagocytosis (Figure 10). Note that the PAM in air for 30 min in Figure 10 had lower percent phagocytosis than PAM in air at 30 min in Figure 3. This may indicate that PAM preincubated for 150 min in the non-nutrient KRB medium in Figure 10 have a slower rate of phagocytosis than cells just removed from the nutrient-rich DMEM in Figure 3. PAM Viability and Adherence during Hypoxia Viability and adherence determinations were made on every PAM examined in these studies. Cell viability in all experiments was > 90 % at all time points after exposure to air or hypoxia. Adherence of PAM to coverslips was statistically similar after incubation in air or hypoxia for 90 to 150 min (PAM per field, air = 45 ± 8 and 47 ± 9; hypoxia = 36 ± 5 and 43 ± 6, 90 and 150 min, respectively; P = NS, air versus hypoxia). Although hypoxic cells were slightly less adherent, the dramatically lower percent phagocytosis observed in PAM exposed to hypoxia in these experiments was probably not due to loss of PAM viability or to significant changes in adherence of PAM to coverslips.
Figure 8. ATP levels of PAM with GRBC in air (closed circles) or hypoxia (open circles). Values are mean ± SEM (n = 4 rabbits, 1.6 x 107 PAM assayed per time point). With the Bonferroni method, level of significance is P < 0.01; * P < 0.01, air versus hypoxia, 30 min; t P < 0.01, air, 0 versus 60 min; :j: P < 0.01, air, 0 min versus hypoxia, 30 and 60 min; and § P < 0.01, hypoxia, 30 versus 60 min.
Effects of Hypoxia on Retention of Phagocytosed GRBC Previously, we had found that phagocytosis was similar in PAM exposed to air or hypoxia for 30 min, with a steady decline in the number of PAM containing phagocytosed particles occurring between 30 and 150 min (Figure 3). In our subsequent experiments, we observed that PAM loaded with GRBC for 60 min in air (percent phagocytosis = 80.0 ± 3.0 %) and then exposed for 60 min to hypoxia had decreased phagocytosis when compared with PAM exposed to air for the same time (Figures 11 to 13). In Figure 11, percent phagocytosis of GRBC in PAM was determined in whole, live cells using the fluorescent method or by counting cells with GRBC in fixed, sectioned, and stained PAM. Both methods yielded similar results, i.e., PAM with phagocytosed GRBC had lower percent phagocytosis after 60 min in hypoxia (Figure 11). In sectioned and stained PAM, the number of GRBC per PAM was lower in PAM exposed to hypoxia than in those incubated in air (Figure 12). Figure 13 illustrates the distribution of number of GRBC per PAM. These curves demonstrate that the percentage of PAM containing zero, one, or two GRBC was higher in cells exposed to hypoxia, whereas 80 en
70
0
60
en
•
30
.....
Air
25
:J:
30
n,
E >e 20
•
5 0
Hypoxia 30 min, Air 30 min
0
IO
0
Hypoxia
~
;I!. 20 10
::< 15 ~ u
0
•
Air
a
10
20
30
• 40
50
60
TIME(MIN)
Figure 7. ATP levels of PAM exposed to air (closed circles) or hypoxia (open circles). Values are mean ± SEM (n = 4 rabbits, 1.6 X 107 PAM assayed per time point). With the Bonferroni method, level of significance is P < 0.01; * P < 0.01, hypoxia, 30 and 60 min compared with respective times in air; and t P < 0.01, air, 0 versus 30 min.
0
10
20
30
40
50
60
TIME (MIN)
Figure 9. Percent phagocytosis of GRBC by PAM in air (closed circles) or hypoxia (open circles). After pre-exposing PAM for 30 min to air or hypoxia, GRBC were mixed with PAM and incubations continued in the respective gases. In addition, percent phagocytosis by PAM pre-exposed to hypoxia for 30 min then incubated 30 min with GRBC in air is shown (closed triangles). Values are mean ± SEM (n = 6 rabbits, > 600 PAM per point). * P < 0.013, hypoxia versus air; :j: P < 0.013 air versus hypoxia/air; and § P < 0.013, hypoxia versus hypoxia/air; all at 60-min time point.
332
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 6 1992
•
Air
a
80
4
Hypoxia
en 70
~ 60
3
t;
2
to-
o
•
5
90
50
~
40 5: 30 ;fl. 20
o
10
Air
o V~-----L_-L------'-_---'---_'-----'---_.L---L--' -150 0
10
20
30
40
50
60
70
80
90
TIME (MIN)
Figure 10. Percent phagocytosis of GRBC by PAM in air following pre-exposure to air (closed circles) or hypoxia (open circles). After 150 min of incubation of PAM in air or hypoxia, GRBC were added to PAM and preparations incubated in air. Values are mean ± SEM (n = 5 to 7 rabbits, > 500 PAM per point). * P < 0.01, air versus hypoxia, 30 min; :j: P < om, air versus air and § P < 0.01, hypoxia versus hypoxia at successive time points.
PAM incubated in air were more likely to have larger numbers of GRBC in each PAM (Figure 13).
Discussion Previous investigators using cultured alveolar macrophages have suggested that PAM phagocytic function is altered by low O2 levels (1, 5-9, 11). Oren and co-workers (8) found that cultured alveolar macrophages incubated for 1 h in 100% N2 phagocytosed less polystyrene or starch particles than did monocytes. Cohen and Cline (1) found that isolated, cultured human alveolar macrophages preincubated in 3 % O2 for 1 h phagocytosed 39% less heat-killed Candida albicans. In our studies, acute hypoxia (1.7% O2) altered phagocytic function in rabbit PAM by decreasing uptake of GRBC particles by the freshly isolated pulmonary cells (Figures 3, 4, 5, and 9). The time at which GRBC were FLUORESCENT
(f)
v;
l2 ~
u 0
~
«
:r::
Q.
