The Effect of Hyperoxia on Migration of Alveolar Macrophages In Vitro 4 ALVIN L BOWLES, JAMES H. DAUBER,5 and RONALD P. DANIELE6
SUMMARY
There is in vitro evidence to support the notion that directed migration (chemotaxis) is involved in the recruitment of alveolar macrophages in vivo. Because 0 2 is widely used in the treatment of pulmonary diseases, we examined the effect of hyperoxia on migration of guinea pig alveolar macrophages in vitro. Migration was measured in blind-well chambers incubated in either room air or hyperoxia. Af-formyl-methionyl-phenylalanine was used to stimulate random migration and to produce directed migration. Migration was quantified by counting the number of mononuclear cells per oil immersion field that had migrated completely through a polycarbonate filter with 5-^m pores. The average Po 2 in the cell suspensions incubated in room air was 100 mm Hg. In the hyperoxic environments, the average Po 2 at 1 h was 260 mm Hg, whereas at 2 and 3 h, it was 410 and 425 mm Hg, respectively. In 6 separate experiments, there was no significant difference between the mean response to iV-formyl-methionyl phenylalanine in hyperoxia and in room air after 1 h of incubation. After 2 and 3h of incubation, however, the response in hyperoxia was significantly (p < 0.002) lower than that in room air. The decreased response in hyperoxia did not appear to result from loss of viability of responding cells, diminished adherence of cells to the filters, loss of activity of 2V-formyl-methionyl phenylalanine exposed to high Po2, or failure of the cells to exhibit directed migration. Instead, it appeared that hyperoxia decreased the response of alveolar macrophages primarily by impairing random migration.
Introduction
T h e alveolar macrophage plays a crucial role in the lung's defense against inhaled particles (1). For the alveolar macrophage to maintain the sterility of the lower respiratory tract and remove particles deposited in the distal airways and alveolar spaces, the cell must be capable of functions involving migration, phagocytosis, and intracellular killing. Chemotaxis, the process by (Received in original form December 21,1978 and in revised form April 9,1979) 1 From the Cardiovascular-Pulmonary Division of the Department of Medicine and the Department of Pathology, University of Pennsylvania School of Medicine, Philadelphia, Pa. 19104. 2 This work was supported by SCOR grant HL15061 from the National Heart, Lung, and Blood Institute. 3 This work was presented at the 73rd Annual
which the migration of cells is oriented along a chemical gradient of an attracting substance, is believed to be a major mechanism for the recruitment of polymorphonuclear and mononuclear phagocytes to sites of infection and inflammation. Previously, it was believed that the alveolar macrophage responded poorly to most chemotactic factors. However, it has recently Meeting of the American Thoracic Society, Boston, Mass., May 1978. 4 Requests for reprints should be addressed to Ronald P. Daniele, M.D., 875 Maloney Bldg., Hospital of the University of Pennsylvania, 3600 Spruce St., Philadelphia, Pa. 19104. 5 Dr. Dauber is the recipient of Young Investigator Research Award no. 1R23-HL-21134 from the National Heart, Lung, and Blood Institute. 6 Dr. Daniele is the recipient of Research Career Development Award no. 1K04-HL-00210 from the National Heart, Lung, and Blood Institute.
AMERICAN REVIEW OF RESPIRATORY DISEASE, VOLUME 120, 1979
541
542
BOWLES, DAUBER, AND DANIELE
been shown that AT-formyl-methionyl peptides are potent chemoattractants for the guinea pig alveolar macrophage (2). This finding supports the notion that chemotaxis is involved in the recruitment of alveolar macrophages in vivo. Because hyperoxia has been shown to produce early impairment of pulmonary defense mechanisms, such as mucociliary clearance (3), we inquired whether macrophage function would also be altered. Specifically, these studies were designed to distinguish the acute effect of hyperoxia on random migration, stimulated random migration (chemokinesis), and directed migration (chemotaxis) of guinea pig alveolar macrophages.
Methods Animals. Cells used in the chemotactic assay were recovered from healthy outbred Hartley strain male guinea pigs weighing 400 to 600 g. Preparation of alveolar macrophages. Alveolar macrophages were obtained by lung lavage as previously described (2). Cell viability was determined by trypan blue exclusion and consistently exceeded 90%. Total leukocyte counts were made on a Coulter counter (Model Zf, Coulter Electronics, Hialeah, Fla.). Differential cell counts were performed on cytocentrifuge (Shandon Corp., Sewickley, Pa.) smears stained with Diff-Quik® (Harleco, Philadelphia, Pa.). The lung cells were finally suspended in Gey's balanced salt solution (Grand Island Biological Company, Grand Island, N.Y.) containing 2% bovine serum albumin (Sigma Chemical Co., St. Louis, Mo.) (pH = 7.0) to give 1.5 X 106 macrophages/ml. These suspensions were used as the responding cell populations in the migration assay. Hyperoxic environments. The hyperoxic environment was created by bubbling 100% O s through water in a sealed Plexiglas® chamber that contained a rack of chemotactic chambers. Through stoppered ports on the top of the Plexiglas chamber, the medium of the responding cell suspension could be rapidly sampled with capillary tubes. Measurements of Po 2 , pH, and lactic acid dehydrogenase activity were made on fluid obtained by this method. Migration assay. Migration assays were performed in blind-well chambers with a diameter of 4.7 mm (Neuroprobe, Bethesda, Md.). The blind well was filled with 200 ^1 of either fluid containing the chemotactic factor, N-formyl-methionyl-phenylalanine (fmet-phe), or Gey's balanced salt solution alone. To determine the effect of hyperoxia on stimulated migration of the guinea pig alveolar macrophage, optimal concentrations of f-met-phe were used (2). The responding cell suspension of 3 X 105 macrophages/200 fil was placed in the upper compartment. The 2 compartments were separated by a poly-
carbonate filter of 5-fim pore size (Nucleopore Corp., Pleasanton, Ca.). The chambers were incubated in either humidified air or the hyperoxic environment. After incubation, the cells in the upper compartment that had not migrated through the filter were removed. The filters were placed on cover slips, allowed to dry, and then stained with Diff-Quik. Mononuclear cells that had completely migrated through the filter were counted in 20 oil immersion fields (1,000 X magnification). The migratory response was defined as the mean ± SEM number of migrated mononuclear cells/oil immersion field for triplicate filters. In experiments in which the effect of hyperoxia on stimulated random migration (chemokinesis) and directed migration (chemotaxis) was evaluated, various concentrations of f-met-phe were placed in both the lower and upper compartments of the chambers as previously described (2). Oxygen and pH determinations. The Po 2 and pH were measured on samples taken by capillary tubes from the medium of the responding cell suspension during incubation. The determinations were made on an acid-base analyzer pH M-71 (Radiometer, Copenhagen, Denmark). Lactic acid dehydrogenase. Using the same direct capillary technique described previously, samples of the medium of the responding cell suspension were taken for measurement of lactic acid dehydrogenase using BMC reagent in a Union Carbide analyzer. Macrophage adherence. Adherence of the alveolar macrophages was determined by incubating the cells in glass vials in room air or hyperoxia (Po 2 —400 mm Hg) for 3 h, removing nonadherent cells by washing the monolayer with Hank's balanced salt solution, and measuring the protein content of the adherent monolayer as described by Lowry and associates (4).
Results Effect of hyperoxia on stimulated migration. Migration of the alveolar macrophages toward fmet-phe in room air and under hyperoxia is shown in figure 1. After 1 h of incubation, the P o 2 in chambers exposed to hyperoxia was 260 m m Hg, but there was no significant difference between the responses of cells incubated in room air and those incubated under hyperoxia. However, by the second and third hours of incubation, when the P o 2 in the chambers reached average values of 410 and 425 m m Hg, respectively, migration under hyperoxia was significantly less than that occuring in room air (p < 0.001; p < 0.002, for 2h and 3h, respectively). T h e P o 2 in the chambers exposed to room air was relatively constant throughout these experiments and averaged approximately 100 m m Hg.
543
HYPEROXIA AND ALVEOLAR MACROPHAGE MOTILITY
U_
100
O
\
z_)
RO
> —' LJJ
10
60
zoQ
(/) LU
40
a:
I TIME
• — •
ROOM
O —O
HYPEROXIA
2 OF
INCUBATION
AIR
3 (hrs
)
Fig. 1. The migratory response of the alveolar macrophage in room air ( • • ) and under hyperoxia (O O) in 6 separate experiments. The chemotactic response is expressed as the mean ± SEM of the number of mononuclear cells (MNL) that migrated/oil immersion field (OIF).
T o determine whether shorter periods of hyperoxia impaired the migration of alveolar macrophages, we used a modified chemotactic chamber that allowed for more rapid equilibration (< 30 min) of Po 2 in both the upper and lower compartments. The details concerning the design and construction of this 2-chambered apparatus have been described (5). In contrast to what was observed in the standard chambers, migration in these chambers under hyperoxic conditions was significantly decreased compared to that in room air (p < 0.002) after only 1 h of incubation (data not shown). T o study the possibility that short-term exposure of alveolar macrophages to hyperoxia produced long-term impairment of motility, cells were first exposed to hyperoxia for 3 h, and then migration was assayed in room air. There was no difference in the response of these cells compared to that of cells preincubated in room air. When cells were incubated in hyperoxia for more than 24 h, however, there was a significant decrease in their migration toward f-met-phe (63 ± 2 versus 39 ± 3 mononuclear cells/oil immersion field for cells preincubated in room air and hyperoxia, respectively; p < 0.002). Evaluation of factors affecting migration. Hyperoxia may impair macrophage migration toward f-met-phe by (2) killing the macrophages, (2) decreasing adherence, (5) decreasing the potency of f-met-phe, or (4) altering the pH of
the medium of the cell suspensions so that it is no longer in the optimal range. Viability of the cells in the upper compartment of the blind-well chambers was determined by measuring (7) lactic acid dehydrogenase activity in the responding cell medium, and (2) the ability of cells remaining in the upper compartment at the end of incubation to exclude trypan blue. Both methods indicated that there was no difference in the viability of the cells incubated for 3 h in room air or hyperoxia. The effect of high Po 2 on adherence was investigated by allowing macrophages to adhere to flat-bottom glass vials under hyperoxic conditions (Po 2 — 450 mm Hg), removing nonadherent cells, and determining the amount of protein in the adherent monolayer. As shown in figure 2, there was no significant difference (p = 0.6) in the protein content of the adherent monolayers of samples incubated in room air or hyperoxia for 3 h. T o determine whether high Po 2 decreased the activity of f-met-phe, we exposed the chemoattractant to hyperoxia for 24 h and found no loss in chemotactic activity when this sample was compared to a sample maintained in room air for a similar period. Finally, no significant difference (p = 0.27) was found in the pH of the samples incubated in room air and hyperoxia. The pH of the responding cell suspension in both environments ranged from 7.06 to 7.18 during the 3-h incubation period. Effect of hyperoxia on random and directed migration. We have previously shown that f-met-
WITH0U1 FMP 3hr
WITH FMP EXPOSURE
Fig. 2. Protein content of adherent macrophage monolayer in room air (light hatched bars) and in hyperoxia (heavy hatched bars). Cells were allowed to adhere in the presence or absence of iV-formyl-methionylphenylalanine (FMP). Values given represent mean ± SEM ptg of protein for an experiment done in triplicate and are representative of 3 separate experiments.
544
BOWLES, DAUBER, AND DANtELE
phe stimulates random migration (chemokinesis) and directed migration (chemotaxis) of alveolar macrophages (2). Although hyperoxia decreased the migration of alveolar macrophages toward f-met-phe, it cannot be determined from the foregoing experiments whether this was due to a decrease in stimulated random migration, a decrease in directed migration, or both. One method for distinguishing directed migration from stimulated random migration is to establish a series of positive and negative gradients of factors across the filters, as described by Zigmond and Hirsch (6). The results of this type of experiment are shown in table 1 and indicate that both stimulated and unstimulated random migration (values within diagonals) was less in hyperoxia. In addition, because the migratory response of the cells to positive gradients of fmet-phe (values to the right of diagonals) exceeded the response to uniform concentrations, directed migration could be demonstrated in hyperoxia. The magnitude of directed migration, however, was consistently lower in hyperoxia than in room air. T o determine whether this decrease was due to impairment of chemotaxis or to a generalized defect in random motility, the percentage decrease for migration in hyperoxia compared to room air was calculated for each concentration of f-met-phe used to produce stimulated random migration and directed mi-
gration, and a mean value was derived for the decrease in each type of migration. T h e decrease in directed migration expressed in this manner was nearly identical to the decrease in stimulated random migration (mean ± SEM, 26 ± 2 versus 25 ± 2.0%, respectively), suggesting that hyperoxia interferes primarily with random motility rather than with directed migration (chemotaxis).
Discussion
The results of the studies show that concentrations of 0 2 greater than 400 mm Hg decrease the in vitro migratory response of the guinea pig alveolar macrophage. The diminished response did not appear to be due to impairment of macrophage adherence, decreased viability of cells, loss of activity of f-met-phe, or changes in pH in the medium of the responding cell suspensions. Other studies have shown that macrophage adherence is impaired by high concentrations of 0 2 (fraction of inspired 0 2 > 85%) (7). In these studies, however, cells were recovered from animals exposed for 42 h to hyperoxic conditions. In our experiments, cells were incubated in vitro for a maximum of 3 h. In this time period, we found no effect on macrophage adherence. The decreased response to f-met-phe occurring in hyperoxia appeared to be due primarily to
TABLE 1 STIMULATION OF MIGRATION OF ALVEOLAR MACROPHAGES BY N-FORMYL-METHIONYL-PHENYLALANINE IN ROOM AIR AND HYPEROXIA* Concentration Below Filter, mol
Concentration 0
{mol) 0
20 ± 1 \ Room air ^ \ \ l 4 ± 1 Hyperoxia
5 x 10-7
Room air Hyperoxia
5 x 10"6
Room air Hyperoxia
5 x 10-5
Room air Hyperoxia
11 ± l \ 13 ± 2
-
9 ± 1 9 ± 3
5 x 10 7
5 x 10"6
5 x 10"5
—
105 ± 3 77 ±7
—
78 ± 3 57 ± 2
215 ± 6 155 ± 5
49 ± 4 \ N. \37 ± 1
157 ± 4 25 ± 2 \ s 54 ± 6 \ \ 3 9 ± 1 \^30 ± 5 24 ± 2
-
74 ± 2 s 22 ± l\ 16 ± 1 \ 5 9 ± 2
* The values represent the mean number of mononuclear cells that migrated through a polycarbonate filter/oil immersion field in triplicate chambers. The experiment was repeated 3 times with similar results. Values within the diagonals are the responses in uniform concentrations of N-formyl-methionylphenylalanine (f-met-phe) and thus represent stimulated random migration. Values to the right and left of the diagonals are the responses in positive and negative gradients of f-met-phe, respectively, and thus represent directed migration.
HYPEROXIA AND ALVEOLAR MACROPHAGE MOTILITY
impaired random migration, because the decreases in directed migration and stimulated random migration were equivalent. The mechanisms by which hyperoxia impairs alveolar macrophage motility are unclear. Possible explanations are: (2) injury of the cytoskeletal apparatus (microfilaments and/or microtubules) resulting from free radical formation induced by high Po 2 ; (2) an alteration of the putative membrane receptor for formyl-methionyl peptides leading to impairment of the transduction of chemotactic signals; and (3) impairment of the metabolic pathways involved in cell motility (8). T h e results of our studies do not distinguish among these mechanisms. In the intact organism, clinically apparent toxicity usually does not develop until after 24 h of exposure to one atmosphere (9) of 0 2 . Despite the absence of clinically apparent 0 2 toxicity within the first 24 h of hyperoxic exposure, however, toxicity may occur earlier at the cellular level (3). T h e impairment of cell motility after only 1 h of exposure shown in the experiments using the modified chemotactic chambers supports this notion. T h e relatively rapid impairment of macrophage migration by high Po 2 in vitro may be important, because it suggests that alveolar hyperoxia might compromise lung defense mechanisms by interfering with the movement of alveolar macrophages into and within the bronchoalveolar spaces.
545
Acknowledgment The writers thank Ms. Angelica M. Cassizzi for her technical assistance. References 1. Green GM. Amber son lecture: in defense of the lung. Am Rev Respir Dis 1970; 102:691-703. 2. Dauber JH, Daniele RP. Chemotactic activity of guinea pig alveolar macrophages. Am Rev Respir Dis 1978; 117:673-84. 3. Sackner MA, Landa J, Hirsch J, Zapata A. Pulmonary effects of oxygen breathing: a six-hour study in normal men. Ann Intern Med 1975; 82: 40-3. 4. Lowry OH, Rosebrough NJ, Farr AL, Randall JR. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193:265-75. 5. Gorenberg DJ, Daniele RP. The alveolar macrophage: its capacity to act as an accessory cell in mitogen-stimulated proliferation of guinea pig lymphocytes. Cell Immunol 1978; 36:115-27. 6. Zigmond SH, Hirsch JG. Leukocyte locomotion and chemotaxis. J Exp Med 1973; 137:387-410. 7. Wolff LJ, Boxer LA, Allen JM, Behner RL. Selective effect of hyperoxia on alveolar macrophage membrane properties (abstract). 14th Annual Reticuloendothelial Society Meeting, Tucson, Ariz., December 1977. 8. Fisher AB, Diamond S, Mellen S. Effect of 0 2 exposure on metabolism of the rabbit alveolar macrophage. J Appl Physiol 1974; 37:341-5. 9. Clark JM, Lambertsen CJ. Pulmonary oxygen toxicity: a review. Pharmacol Rev 1971; 23:37133.
SUMMARY
There is in vitro evidence to support the notion that directed migration (chemotaxis) is involved in the recruitment of alveolar macrophages in vivo. Because 0 2 is widely used in the treatment of pulmonary diseases, we examined the effect of hyperoxia on migration of guinea pig alveolar macrophages in vitro. Migration was measured in blind-well chambers incubated in either room air or hyperoxia. Af-formyl-methionyl-phenylalanine was used to stimulate random migration and to produce directed migration. Migration was quantified by counting the number of mononuclear cells per oil immersion field that had migrated completely through a polycarbonate filter with 5-^m pores. The average Po 2 in the cell suspensions incubated in room air was 100 mm Hg. In the hyperoxic environments, the average Po 2 at 1 h was 260 mm Hg, whereas at 2 and 3 h, it was 410 and 425 mm Hg, respectively. In 6 separate experiments, there was no significant difference between the mean response to iV-formyl-methionyl phenylalanine in hyperoxia and in room air after 1 h of incubation. After 2 and 3h of incubation, however, the response in hyperoxia was significantly (p < 0.002) lower than that in room air. The decreased response in hyperoxia did not appear to result from loss of viability of responding cells, diminished adherence of cells to the filters, loss of activity of 2V-formyl-methionyl phenylalanine exposed to high Po2, or failure of the cells to exhibit directed migration. Instead, it appeared that hyperoxia decreased the response of alveolar macrophages primarily by impairing random migration.
Introduction
T h e alveolar macrophage plays a crucial role in the lung's defense against inhaled particles (1). For the alveolar macrophage to maintain the sterility of the lower respiratory tract and remove particles deposited in the distal airways and alveolar spaces, the cell must be capable of functions involving migration, phagocytosis, and intracellular killing. Chemotaxis, the process by (Received in original form December 21,1978 and in revised form April 9,1979) 1 From the Cardiovascular-Pulmonary Division of the Department of Medicine and the Department of Pathology, University of Pennsylvania School of Medicine, Philadelphia, Pa. 19104. 2 This work was supported by SCOR grant HL15061 from the National Heart, Lung, and Blood Institute. 3 This work was presented at the 73rd Annual
which the migration of cells is oriented along a chemical gradient of an attracting substance, is believed to be a major mechanism for the recruitment of polymorphonuclear and mononuclear phagocytes to sites of infection and inflammation. Previously, it was believed that the alveolar macrophage responded poorly to most chemotactic factors. However, it has recently Meeting of the American Thoracic Society, Boston, Mass., May 1978. 4 Requests for reprints should be addressed to Ronald P. Daniele, M.D., 875 Maloney Bldg., Hospital of the University of Pennsylvania, 3600 Spruce St., Philadelphia, Pa. 19104. 5 Dr. Dauber is the recipient of Young Investigator Research Award no. 1R23-HL-21134 from the National Heart, Lung, and Blood Institute. 6 Dr. Daniele is the recipient of Research Career Development Award no. 1K04-HL-00210 from the National Heart, Lung, and Blood Institute.
AMERICAN REVIEW OF RESPIRATORY DISEASE, VOLUME 120, 1979
541
542
BOWLES, DAUBER, AND DANIELE
been shown that AT-formyl-methionyl peptides are potent chemoattractants for the guinea pig alveolar macrophage (2). This finding supports the notion that chemotaxis is involved in the recruitment of alveolar macrophages in vivo. Because hyperoxia has been shown to produce early impairment of pulmonary defense mechanisms, such as mucociliary clearance (3), we inquired whether macrophage function would also be altered. Specifically, these studies were designed to distinguish the acute effect of hyperoxia on random migration, stimulated random migration (chemokinesis), and directed migration (chemotaxis) of guinea pig alveolar macrophages.
Methods Animals. Cells used in the chemotactic assay were recovered from healthy outbred Hartley strain male guinea pigs weighing 400 to 600 g. Preparation of alveolar macrophages. Alveolar macrophages were obtained by lung lavage as previously described (2). Cell viability was determined by trypan blue exclusion and consistently exceeded 90%. Total leukocyte counts were made on a Coulter counter (Model Zf, Coulter Electronics, Hialeah, Fla.). Differential cell counts were performed on cytocentrifuge (Shandon Corp., Sewickley, Pa.) smears stained with Diff-Quik® (Harleco, Philadelphia, Pa.). The lung cells were finally suspended in Gey's balanced salt solution (Grand Island Biological Company, Grand Island, N.Y.) containing 2% bovine serum albumin (Sigma Chemical Co., St. Louis, Mo.) (pH = 7.0) to give 1.5 X 106 macrophages/ml. These suspensions were used as the responding cell populations in the migration assay. Hyperoxic environments. The hyperoxic environment was created by bubbling 100% O s through water in a sealed Plexiglas® chamber that contained a rack of chemotactic chambers. Through stoppered ports on the top of the Plexiglas chamber, the medium of the responding cell suspension could be rapidly sampled with capillary tubes. Measurements of Po 2 , pH, and lactic acid dehydrogenase activity were made on fluid obtained by this method. Migration assay. Migration assays were performed in blind-well chambers with a diameter of 4.7 mm (Neuroprobe, Bethesda, Md.). The blind well was filled with 200 ^1 of either fluid containing the chemotactic factor, N-formyl-methionyl-phenylalanine (fmet-phe), or Gey's balanced salt solution alone. To determine the effect of hyperoxia on stimulated migration of the guinea pig alveolar macrophage, optimal concentrations of f-met-phe were used (2). The responding cell suspension of 3 X 105 macrophages/200 fil was placed in the upper compartment. The 2 compartments were separated by a poly-
carbonate filter of 5-fim pore size (Nucleopore Corp., Pleasanton, Ca.). The chambers were incubated in either humidified air or the hyperoxic environment. After incubation, the cells in the upper compartment that had not migrated through the filter were removed. The filters were placed on cover slips, allowed to dry, and then stained with Diff-Quik. Mononuclear cells that had completely migrated through the filter were counted in 20 oil immersion fields (1,000 X magnification). The migratory response was defined as the mean ± SEM number of migrated mononuclear cells/oil immersion field for triplicate filters. In experiments in which the effect of hyperoxia on stimulated random migration (chemokinesis) and directed migration (chemotaxis) was evaluated, various concentrations of f-met-phe were placed in both the lower and upper compartments of the chambers as previously described (2). Oxygen and pH determinations. The Po 2 and pH were measured on samples taken by capillary tubes from the medium of the responding cell suspension during incubation. The determinations were made on an acid-base analyzer pH M-71 (Radiometer, Copenhagen, Denmark). Lactic acid dehydrogenase. Using the same direct capillary technique described previously, samples of the medium of the responding cell suspension were taken for measurement of lactic acid dehydrogenase using BMC reagent in a Union Carbide analyzer. Macrophage adherence. Adherence of the alveolar macrophages was determined by incubating the cells in glass vials in room air or hyperoxia (Po 2 —400 mm Hg) for 3 h, removing nonadherent cells by washing the monolayer with Hank's balanced salt solution, and measuring the protein content of the adherent monolayer as described by Lowry and associates (4).
Results Effect of hyperoxia on stimulated migration. Migration of the alveolar macrophages toward fmet-phe in room air and under hyperoxia is shown in figure 1. After 1 h of incubation, the P o 2 in chambers exposed to hyperoxia was 260 m m Hg, but there was no significant difference between the responses of cells incubated in room air and those incubated under hyperoxia. However, by the second and third hours of incubation, when the P o 2 in the chambers reached average values of 410 and 425 m m Hg, respectively, migration under hyperoxia was significantly less than that occuring in room air (p < 0.001; p < 0.002, for 2h and 3h, respectively). T h e P o 2 in the chambers exposed to room air was relatively constant throughout these experiments and averaged approximately 100 m m Hg.
543
HYPEROXIA AND ALVEOLAR MACROPHAGE MOTILITY
U_
100
O
\
z_)
RO
> —' LJJ
10
60
zoQ
(/) LU
40
a:
I TIME
• — •
ROOM
O —O
HYPEROXIA
2 OF
INCUBATION
AIR
3 (hrs
)
Fig. 1. The migratory response of the alveolar macrophage in room air ( • • ) and under hyperoxia (O O) in 6 separate experiments. The chemotactic response is expressed as the mean ± SEM of the number of mononuclear cells (MNL) that migrated/oil immersion field (OIF).
T o determine whether shorter periods of hyperoxia impaired the migration of alveolar macrophages, we used a modified chemotactic chamber that allowed for more rapid equilibration (< 30 min) of Po 2 in both the upper and lower compartments. The details concerning the design and construction of this 2-chambered apparatus have been described (5). In contrast to what was observed in the standard chambers, migration in these chambers under hyperoxic conditions was significantly decreased compared to that in room air (p < 0.002) after only 1 h of incubation (data not shown). T o study the possibility that short-term exposure of alveolar macrophages to hyperoxia produced long-term impairment of motility, cells were first exposed to hyperoxia for 3 h, and then migration was assayed in room air. There was no difference in the response of these cells compared to that of cells preincubated in room air. When cells were incubated in hyperoxia for more than 24 h, however, there was a significant decrease in their migration toward f-met-phe (63 ± 2 versus 39 ± 3 mononuclear cells/oil immersion field for cells preincubated in room air and hyperoxia, respectively; p < 0.002). Evaluation of factors affecting migration. Hyperoxia may impair macrophage migration toward f-met-phe by (2) killing the macrophages, (2) decreasing adherence, (5) decreasing the potency of f-met-phe, or (4) altering the pH of
the medium of the cell suspensions so that it is no longer in the optimal range. Viability of the cells in the upper compartment of the blind-well chambers was determined by measuring (7) lactic acid dehydrogenase activity in the responding cell medium, and (2) the ability of cells remaining in the upper compartment at the end of incubation to exclude trypan blue. Both methods indicated that there was no difference in the viability of the cells incubated for 3 h in room air or hyperoxia. The effect of high Po 2 on adherence was investigated by allowing macrophages to adhere to flat-bottom glass vials under hyperoxic conditions (Po 2 — 450 mm Hg), removing nonadherent cells, and determining the amount of protein in the adherent monolayer. As shown in figure 2, there was no significant difference (p = 0.6) in the protein content of the adherent monolayers of samples incubated in room air or hyperoxia for 3 h. T o determine whether high Po 2 decreased the activity of f-met-phe, we exposed the chemoattractant to hyperoxia for 24 h and found no loss in chemotactic activity when this sample was compared to a sample maintained in room air for a similar period. Finally, no significant difference (p = 0.27) was found in the pH of the samples incubated in room air and hyperoxia. The pH of the responding cell suspension in both environments ranged from 7.06 to 7.18 during the 3-h incubation period. Effect of hyperoxia on random and directed migration. We have previously shown that f-met-
WITH0U1 FMP 3hr
WITH FMP EXPOSURE
Fig. 2. Protein content of adherent macrophage monolayer in room air (light hatched bars) and in hyperoxia (heavy hatched bars). Cells were allowed to adhere in the presence or absence of iV-formyl-methionylphenylalanine (FMP). Values given represent mean ± SEM ptg of protein for an experiment done in triplicate and are representative of 3 separate experiments.
544
BOWLES, DAUBER, AND DANtELE
phe stimulates random migration (chemokinesis) and directed migration (chemotaxis) of alveolar macrophages (2). Although hyperoxia decreased the migration of alveolar macrophages toward f-met-phe, it cannot be determined from the foregoing experiments whether this was due to a decrease in stimulated random migration, a decrease in directed migration, or both. One method for distinguishing directed migration from stimulated random migration is to establish a series of positive and negative gradients of factors across the filters, as described by Zigmond and Hirsch (6). The results of this type of experiment are shown in table 1 and indicate that both stimulated and unstimulated random migration (values within diagonals) was less in hyperoxia. In addition, because the migratory response of the cells to positive gradients of fmet-phe (values to the right of diagonals) exceeded the response to uniform concentrations, directed migration could be demonstrated in hyperoxia. The magnitude of directed migration, however, was consistently lower in hyperoxia than in room air. T o determine whether this decrease was due to impairment of chemotaxis or to a generalized defect in random motility, the percentage decrease for migration in hyperoxia compared to room air was calculated for each concentration of f-met-phe used to produce stimulated random migration and directed mi-
gration, and a mean value was derived for the decrease in each type of migration. T h e decrease in directed migration expressed in this manner was nearly identical to the decrease in stimulated random migration (mean ± SEM, 26 ± 2 versus 25 ± 2.0%, respectively), suggesting that hyperoxia interferes primarily with random motility rather than with directed migration (chemotaxis).
Discussion
The results of the studies show that concentrations of 0 2 greater than 400 mm Hg decrease the in vitro migratory response of the guinea pig alveolar macrophage. The diminished response did not appear to be due to impairment of macrophage adherence, decreased viability of cells, loss of activity of f-met-phe, or changes in pH in the medium of the responding cell suspensions. Other studies have shown that macrophage adherence is impaired by high concentrations of 0 2 (fraction of inspired 0 2 > 85%) (7). In these studies, however, cells were recovered from animals exposed for 42 h to hyperoxic conditions. In our experiments, cells were incubated in vitro for a maximum of 3 h. In this time period, we found no effect on macrophage adherence. The decreased response to f-met-phe occurring in hyperoxia appeared to be due primarily to
TABLE 1 STIMULATION OF MIGRATION OF ALVEOLAR MACROPHAGES BY N-FORMYL-METHIONYL-PHENYLALANINE IN ROOM AIR AND HYPEROXIA* Concentration Below Filter, mol
Concentration 0
{mol) 0
20 ± 1 \ Room air ^ \ \ l 4 ± 1 Hyperoxia
5 x 10-7
Room air Hyperoxia
5 x 10"6
Room air Hyperoxia
5 x 10-5
Room air Hyperoxia
11 ± l \ 13 ± 2
-
9 ± 1 9 ± 3
5 x 10 7
5 x 10"6
5 x 10"5
—
105 ± 3 77 ±7
—
78 ± 3 57 ± 2
215 ± 6 155 ± 5
49 ± 4 \ N. \37 ± 1
157 ± 4 25 ± 2 \ s 54 ± 6 \ \ 3 9 ± 1 \^30 ± 5 24 ± 2
-
74 ± 2 s 22 ± l\ 16 ± 1 \ 5 9 ± 2
* The values represent the mean number of mononuclear cells that migrated through a polycarbonate filter/oil immersion field in triplicate chambers. The experiment was repeated 3 times with similar results. Values within the diagonals are the responses in uniform concentrations of N-formyl-methionylphenylalanine (f-met-phe) and thus represent stimulated random migration. Values to the right and left of the diagonals are the responses in positive and negative gradients of f-met-phe, respectively, and thus represent directed migration.
HYPEROXIA AND ALVEOLAR MACROPHAGE MOTILITY
impaired random migration, because the decreases in directed migration and stimulated random migration were equivalent. The mechanisms by which hyperoxia impairs alveolar macrophage motility are unclear. Possible explanations are: (2) injury of the cytoskeletal apparatus (microfilaments and/or microtubules) resulting from free radical formation induced by high Po 2 ; (2) an alteration of the putative membrane receptor for formyl-methionyl peptides leading to impairment of the transduction of chemotactic signals; and (3) impairment of the metabolic pathways involved in cell motility (8). T h e results of our studies do not distinguish among these mechanisms. In the intact organism, clinically apparent toxicity usually does not develop until after 24 h of exposure to one atmosphere (9) of 0 2 . Despite the absence of clinically apparent 0 2 toxicity within the first 24 h of hyperoxic exposure, however, toxicity may occur earlier at the cellular level (3). T h e impairment of cell motility after only 1 h of exposure shown in the experiments using the modified chemotactic chambers supports this notion. T h e relatively rapid impairment of macrophage migration by high Po 2 in vitro may be important, because it suggests that alveolar hyperoxia might compromise lung defense mechanisms by interfering with the movement of alveolar macrophages into and within the bronchoalveolar spaces.
545
Acknowledgment The writers thank Ms. Angelica M. Cassizzi for her technical assistance. References 1. Green GM. Amber son lecture: in defense of the lung. Am Rev Respir Dis 1970; 102:691-703. 2. Dauber JH, Daniele RP. Chemotactic activity of guinea pig alveolar macrophages. Am Rev Respir Dis 1978; 117:673-84. 3. Sackner MA, Landa J, Hirsch J, Zapata A. Pulmonary effects of oxygen breathing: a six-hour study in normal men. Ann Intern Med 1975; 82: 40-3. 4. Lowry OH, Rosebrough NJ, Farr AL, Randall JR. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193:265-75. 5. Gorenberg DJ, Daniele RP. The alveolar macrophage: its capacity to act as an accessory cell in mitogen-stimulated proliferation of guinea pig lymphocytes. Cell Immunol 1978; 36:115-27. 6. Zigmond SH, Hirsch JG. Leukocyte locomotion and chemotaxis. J Exp Med 1973; 137:387-410. 7. Wolff LJ, Boxer LA, Allen JM, Behner RL. Selective effect of hyperoxia on alveolar macrophage membrane properties (abstract). 14th Annual Reticuloendothelial Society Meeting, Tucson, Ariz., December 1977. 8. Fisher AB, Diamond S, Mellen S. Effect of 0 2 exposure on metabolism of the rabbit alveolar macrophage. J Appl Physiol 1974; 37:341-5. 9. Clark JM, Lambertsen CJ. Pulmonary oxygen toxicity: a review. Pharmacol Rev 1971; 23:37133.
