Bleomycin stimulation by the human alveolar
of cytokine secretion macrophage
RONALD K. SCHEULE, RAYMOND C. PERKINS, RAYMOND HAMILTON, AND ANDRIJ HOLIAN Departments of Internal Medicine and Pharmacology, Medical School, Houston, Texas 77225 Scheule, Ronald K., Raymond C. Perkins, Raymond Hamilton, and Andrij Holian. Bleomycin stimulation of cytokine secretion by the human alveolar macrophage. Am. J. Physiol. 262 (Lung Cell. Mol. Physiol. 6): L386-L391, 1992.-- Bleomycin (BLM) is a very effective antineoplastic drug for many gynecologic and urinary tract carcinomas. However, its use, e.g., cumulative dosage, often is limited by the pulmonary fibrosis that it causes. The mechanism by which BLM causes fibrosis is not understood but is proposed to involve the pulmonary macrophage, a central cell in the cytokine network of the lung. To examine the direct effects of this drug on the human alveolar macrophage, we have treated human alveolar macrophages (isolated from normal subjects by bronchoalveolar lavage) with BLM in vitro and examined resultant macrophage secretory products that have importance for inflammatory and fibrotic processes. A 24-h treatment with BLM (0.5400 mU/ ml) was found to result in 1) a concentration-dependent decrease in the ability of the macrophage to produce superoxide anion in response to phorbol 12,13-dibutyrate, 2) an increase in secreted interleukin-lp (IL-ID), and 3) a decrease in intracellular levels of adenosine 3’,5’-cyclic monophosphate. Kinetic studies revealed a time-dependent appearance of BLM-induced cytokines; tumor necrosis factor-a could be detected as early as 4 h after stimulation, followed by IL-lp at 8 h. The secretion of these cytokines was found to precede the release of prostaglandin E,, which became significant only at 24 h. Taken together, the present results imply that the human alveolar macrophage does not contribute to BLM-induced oxidant injury of the lung but that it may contribute to the development of BLM-induced pulmonary fibrosis. interleukinI& tumor necrosis factor-a; adenosine 3’,5’-cyclic monophosphate
superoxide
anion;
(BLM) is a group of glycopeptides isolated from Streptomyces verticiks. BLM-A2 is the major component used clinically for its antineoplastic properties. In the lung, BLM-induced fibrosis limits the usefulness of the drug (20). The mechanism(s) by which BLM causes fibrosis in humans is not yet understood but is likely to involve pulmonary (alveolar and interstitial) 1 macrophages, the resident phagocytic cells of the lung. Due to its ability to secrete a wide array of inflammatory mediators and growth factors (22, 32, 46), the lung macrophage plays a key role in acute inflammatory responses as well as in the chronic fibrotic responses of the lung to injury. In particular, stimulation of human alveolar macrophages (AM) has been shown to result in the production of three multifunctional cytokines, namely interleukin-1 (IL-l) (15, 48), interleukin-6 (IL6) (26), and tumor necrosis factor-a (TNF-a) (2). These three macrophage-derived cytokines have overlapping --------. -----l_l--BLEOMYCIN
l The terms “lung” or “pulmonary” macrophage changeably to mean cells that could be either alveolar interstitial macrophages. L386
1040-0605/92
$2.00
are used intermacrophages or Copyright
The University
of Texas
activities and are important mediators of both inflammatory and immune responses (1) that may contribute to the generation of a fibrotic response. Individually, IL-l and TNF-cu potentiate fibroblast growth and collagen production (11,25). These mitogenic effects on the fibroblast are generally opposed by the negative regulatory effects of the arachidonic acid metabolite prostaglandin E2 (PGE2, 14, 25). BLM-induced fibrosis resembling the human disease process has been produced in a variety of animal models (9, 17, 19, 27, 29, 35, 43), and AM from these models have been examined for their involvement in the fibrotic process. For example, AM isolated from rats after a single intratracheal instillation of BLM were found to secrete elevated levels of IL-1 activity (19). In addition to these studies of in vivo exposed AM, which have the potential of revealing both direct and indirect effects of the drug on the AM, several investigations have examined the direct effects of BLM on the AM by exposing them to the drug in vitro (19). Thus, in one study (9), BLM-treated rat AM were found to secrete increased amounts of a macrophage-derived growth factor (MDGF) for fibroblasts but not IL-l or TNF-cu, whereas in another study (43), hamster AM were found to secrete elevated levels of IL-l activity. Although animal models of BLM-induced fibrosis have led to insights into the human disease processes caused by this drug, no studies to date have reported responses of the human AM to BLM. The present study was conducted to help define the direct effects of BLM on the human AM by exposing these cells to the drug in vitro and characterizing the acute (92% macrophages [as identified by Giemsa staining (Diff-Quik; Scientific Products) and confirmed by esterase staining (30)] of Cytospin preparations. Cells were ~85% viable by trypan blue exclusion. Cells (I X 106/ml) were cultured for up to 24 h in the presence or absence of BLM in N-2-hydroxyethylpiperazine-N’-2ethanesulfonic acid (HEPES) -buffered medium 199 containing 10% heat-inactivated fetal calf serum (FCS; Sigma). BLM concentrations were chosen to span a relevant physiological dose.2 Cells were maintained in suspension by slow end-over-end tumbling in polypropylene tubes at 37°C. Incubations were terminated by pelleting the cells, and supernatants were stored at -80” C until assayed.
Assays Under no condition was BLM found to be cytotoxic (by trypan blue and cell counting), even at the highest concentrations of BLM used (100 mu/ml) after a 24-h incubation. Endotoxin levels in buffers, media, and BLM stock solutions were assessed by the Limulus amoebocyte assay (Sigma); endotoxin levels were in all cases found to be ~20 pg/ml. For a typical experiment using 5 mu/ml BLM, this result translates into a maximum endotoxin level arising from the BLM of ~20 fg/ml. Thus any BLM-dependent results are due to the effects of the drug itself and not endotoxin. Assay for superoxide anion. Superoxide anion production was quantitated by the reduction of cytochrome c and was followed spectrophotometrically by absorbance at 550 nm (18). Data are from cell suspensions maintained in MHS containing 2 mM Ca2+ at 37°C for 1 h at 0.5 x lo6 cells/ml. Phorbol 12,13dibutyrate (PDB) at 1 PM served as a positive stimulant of superoxide anion production. Assays for IL-ID, IL-6, TNF-CU, PGE2, and adenosine 3’,5’cyclic monophosphate (CAMP). The macrophage cytokines ILI& IL-6, and TNF-cu were assayed in cell-free supernatants with the use of radioimmunoassay (RIA) kits (Advanced Magnetics). Lipopolysaccharide at 10 pg/ml was used as a positive control for stimulation of the human AM. Intracellular CAMP was released from pelleted cells by 5% trichloroacetic acid (TCA). Water-saturated diethyl ether was used to extract the TCA from the aqueous samples, which were then dried with a SpeedVac concentrator (Savant Industries, Hicksville, NY) and used for CAMP RIA assays (New England Nuclear, Boston, MA). PGE, was quantitated in supernatants by constructing an RIA using a rabbit anti-PGE antiserum (Sigma; reacts with both PGE, and PGE2) together with [“H(N)] -prostaglandin E2 ( [3H]PGE2, Amersham; 200 Ci/mmol) and authentic PGE, standards (Sigma). This assay had a working range of 20 pg/ml-10 rig/ml. Because PGE2 is the major PGE produced by human AM (31), the PGE detected by this assay will be referred to as PGE2.
ALVEOLAR
L387
MACROPHAGE
(ANOVA) followed wise comparisons.
by a Student-Newman-Keuls
test for pair-
RESULTS
Macrophage-derived m.odulators. Isolated human al.veolar macrophages were exposed to BLM in vitro, and their secretion of superoxide anion and fibroblast-active factors was quantitated. Figure 1 shows that a 24-h incubation of human AM with BLM results in a depression of the ability of PDB to stimulate the cell that is directly related to the concentration of BLM. For example, after an incubation with 5 mu/ml BLM, stimulation with PDB reveals a decrease of ~60% in PDB-stimulated superoxide anion production compared with cells incubated in the absence of drug. Basal superoxide anion production was not significantly altered by the presence of BLM. PGE2 is a modulator secreted by AM with demonstrated growth inhibitory activity for fibroblasts (14). Indirect evidence suggests that IL-6 may have a similar activity (25). While BLM (5 mu/ml) treatment for 24 h stimulated some PGEz secretion by human AM, the amount was small and could not be related to the concentration of BLM outside of error (data not shown). The low level of PGE2 released from BLM-treated cells (200500) pg/ml) was significantly less than the 3,000-4,000 pg/ml released from cells stimulated by LPS over the same time period. Likewise, the BLM-stimulated production of IL-6 over a 24-h period was also minimal (-500 pg/ml) compared with the amount secreted in response to LPS (w 7,000 pg/ml; data not shown). The macrophage-derived cytokines IL-lp and TNF-cu are positive regulators of fibroblast growth (22). Figure 2 demonstrates that IL-lp release from human AM is dependent on the concentration of BLM; significant increases could be detected at 0.5 mU BLM/ml (0.3 pg/ml). The maximal amount of IL-lp released was less than that released by 10 pg/ml LPS, which was at least 10,000 pg IL-lP/ml under these conditions. The amount of TNF-cx secreted in response to BLM (5 mu/ml) at 24
0
0
Statis tics Values are presented as means t SE. The number of individuals whose cells were used for a given experiment is denoted by n in the corresponding figure legend. For each experiment, statistical treatment included a one-way analysis of variance 2 For a typical m2 and a blood blood concentration tration and may environment of this study vary
dose of volume of or may the lung between
35 U/m2, and using an adult surface area of 1.5 of 5 liter, one can calculate an approximate - 10 mU BLM/ml. This is an average concennot reflect the local drug concentration in the macrophage. Concentrations of BLM used in 0.5 and 100 mu/ml.
0.5
5
50
100
IBLMI, ml-J/d Fig. 1. Phorbol l&13-dibutyrate (PDB)-stimulated superoxide anion production after incubation of human alveolar macrophages with varying concentrations of bleomycin (BLM). Cells were incubated for 24 h at 37°C in suspension with indicated concentrations of BLM and then treated for 1 h with either medium (filled bars) or an equal volume of PDB in medium [final concentration of PDB = 1 ,uM, (hatched bars)]. As a positive control, cells were also incubated with lipopolysaccharide (LPS, 10 pg/ml) for 24 h. Superoxide anion production was quantitated from cell supernatants as described in METHODS. *Values significantly different (P < 0.05) from those obtained with cells incubated in absence of BLM, i.e., [BLM] = 0. n = 5 individuals.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
L388
BLEOMYCIN
0
0.5
STIMULATION
5
50
OF HUMAN
100
IBLMI, mU/d
ALVEOLAR
MACROPHAGE
0
10
20
TIME (h>
Fig. 2. Human alveolar macrophage secretion of interleukin-lp (IL-lp) in response to varying concentrations of BLM. Cells in suspension were incubated with BLM for 24 h and IL-lp was quantitated in the cell-free supernatants by radioimmunoassay (RIA) as described in METHODS. *Values significantly different (P < 0.05) from those obtained with cells incubated in absence of BLM, i.e., [BLM] = 0. n = 5.
Fig. 4. Kinetics of BLM-stimulated secretion of IL-lb (A), tumor necrosis factor-a (TNFcr, o), and PGEZ (0) by human alveolar macrophages incubated in presence of 5 mu/ml BLM. Cells in suspension were incubated with BLM, and cell-free supernatants were analyzed for IL-l@, TNF-a, and PGEZ as described in METHODS. Cytokine secretion at each time point by control cells has been subtracted from these data. n = 4.
rapid compared with that of IL-l& Significant secretion of TNF-a is apparent at 4 h, whereas IL-lp secretion becomes significant only after 8-10 h. The secretion of g4ooo TNF-cu reaches maximal levels around 15 h, whereas the 23 a secretion of IL-l@ continues to increase even at 24 h. CQ Finally, the onset of PGEz secretion appears to lag behind 72alo those of both TNF-cu and IL-l@ and becomes significant d only after a 24-h incubation with BLM. In contrast to the concentration dependence of IL-lp secretion (Fig. 2), the relatively low amounts of TNF-a and PGEB secreted at 24 0 24 h precluded a demonstration of their dependence on the Contact Time (h) concentration of BLM (data not shown). Fig. 3. Human alveolar macrophage secretion of IL-ID in response to Intracellular signaling pathways. PGE2 (47) and IL-1 variable initial contact time with 5 mu/ml BLM. After indicated incu(5,41) stimulate adenylate cyclase, and CAMP is involved bation time in presence of BLM, cells were washed twice and placed in fresh medium without BLM for remainder of a 24-h incubation. IL-lp in the regulation of IL-l@ and TNF-cu expression (21, 33, was quantitated in cell-free supernatants by RIA as described in 44). Potential interactions between intracellular signaling METHODS. *Values significantly different (P < 0.05) from those pathways and some of the observed effects of BLM on the obtained with cells incubated in absence of BLM, i.e., [BLM] = 0. human AM were explored by quantitating the levels of n= 5. intracellular CAMP after a 24-h incubation with the drug. h was 900 t 500 pg/ml; for comparison, LPS induced the Figure 5 shows that incubating these cells with BLM resulted in a concentration-dependent decrease in the levsecretion of ~2,400 pg TNF-a/ml. Blur contact time. The above-described dependence of els of intracellular CAMP. A significant decrease in CAMP cytokine release on BLM was noted in cells that had been was evident at a concentration of 0.5 mU BLM/ml and resulted in an intracellular level of CAMP equivalent to maintained in the presence of drug for the entire 24-h incubation. To define a “contact time” with the drug that resulting from incubating the cells with 10 pg/ml that was sufficient to stimulate cytokine release, cells LPS, a known cellular stimulant. To further explore the apparent (inverse) relationship were incubated with BLM for various times, washed free between the concentration-dependent, BLM-induced of drug, and then incubated for the remainder of the 24-h increases in IL-lp and decreases in CAMP, cells were time period in BLM-free medium. The cytokine quantitated was IL-l@ because its secretion at 24 h was the most incubated with 5 mU BLM/ml in the presence or absence significant response of the cells to BLM. Figure 3 shows of the poorly hydrolyzed CAMP analogue N,Y-O-dibuthat cells incubated with BLM for 2-4 h produced almost tyryladenosine 3’,5’-cyclic monophosphate (DBcAMP), the same amount of IL-10 in a 24-h time period as cells and IL-l@ levels were measured after 24 h. The inset in that had been incubated with BLM for the entire 24 h. Fig. 5 shows that, in the presence of DBcAMP, BLM This observation indicates that a 2-4 h contact time of stimulation of IL-l@ secretion was reduced by ~80%. BLM with the cells is necessary to initiate the processes Thus, when CAMP levels are kept artificially high, the that result in IL-l@ secretion but that BLM is not nec- BLM-induced secretion of IL-lp is significantly reduced. essary thereafter. DISCUSSION Cytokine kinetics. Figure 4 illustrates the time-dependent secretion of TNF-cu, IL-l& and PGEB by the human The mechanism(s) by which BLM causes an interstiAM in response to the continuous presence of 5 mu/ml tial pulmonary pneumonitis that can lead to fibrosis is BLM. A significant feature of these data is the different not known but may involve oxidative or fibrogenic potentemporal pattern of these secreted macrophage products. tiating activities of th .e alveolar macrophage. The use of For example, the onset of TNF-cu secretion is relatively supplemental oxygen therapy in patients treated with 1
T
1
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
BLEOMYCIN
0
0.5
5
[BLM],
50
STIMULATION
100
OF HUMAN
LPS
mu/ml
Fig. 5. Intracellular CAMP levels in human alveolar macrophages after 24-h incubation with a variable concentration of BLM. As a positive control, cells were also incubated with LPS (10 pg/ml) for 24 h. Intracellular CAMP was quantitated by RIA as described in METHODS. *Values significantly different (P < 0.05) from those obtained with cells incubated in absence of BLM, i.e., [BLM] = 0. n = 5. Inset: effect of dibutyryl CAMP (DBcAMP) on BLM-induced IL-10 secretion by human alveolar macrophages. Cells were stimulated with 5 mu/ml BLM for 24 h in presence of 2 mM DBcAMP, and IL-lb was quantitated in the supernatant. Control values of IL-1p secretion either in presence or absence of DBcAMP were not significantly different from 0 (30 t 30 pg/ml). *Significantly different (P < 0.05) from value obtained in absence of DBcAMP. n = 4.
BLM has been associated with an increased risk of fibrosis (16), but this conclusion is controversial (13). Several animal studies (37,42,45) have supported a potentiating role for hyperoxia in BLM-induced fibrosis. BLMinduced pulmonary toxicity, especially under hyperoxic conditions, may be a result of the formation of oxygen radicals, inasmuch as the generation of such radicals is thought to be the mechanism of BLM-induced cytotoxicity (38). One immediate response of the AM to some stimulants is the generation of the reactive oxygen species superoxide anion. Superoxide anion is cytotoxic and can also contribute to inflammation in vivo. In the present studies, BLM itself was not a stimulant for superoxide anion production by human AM (data not shown), a result that is consistent with an analogous finding in the pig model (7). Thus these results argue against a direct role for the AM in BLM-induced oxidant injury in humans. However, they do not rule out the possibility of more indirect effects of BLM on the human AM that may contribute to inflammation or fibrosis. One example of an indirect effect of BLM is the finding that a 24-h BLM treatment of human AM suppressed the subsequent ability of the cell to produce superoxide anion when stimulated by PDB (Fig. 1). The results shown in Fig. 1 depict deficits in PDB-stimulated superoxide anion production after a 24-h incubation with BLM. In contrast, l-h incubations of human AM with either 5 or 50 mu/ml BLM did not alter the ability of the cell to respond to submaximal concentrations of PDB by producing superoxide anion (data not shown). In a related animal study (7), short-term (~60 min) incubations of pig AM with BLM were found to result in an increase in the ability of PDB to stimulate superoxide anion. These latter results are in direct contrast to the present findings
ALVEOLAR
MACROPHAGE
L389
and may reflect species-dependent effects of BLM on the alveolar macrophage. The ability of BLM to suppress PDB-stimulated superoxide anion production was found to be directly dependent on the concentration of BLM (Fig. 1). Phorbol esters such as PDB are thought to stimulate superoxide anion production by directly activating protein kinase C (PKC), which then activates the oxidase. Thus the observation that BLM treatment of the AM decreases the subsequent ability of PDB to stimulate superoxide anion production (Fig. 1) implies that PKC or a component(s) of the oxidase has been downregulated or inactivated by BLM in a concentration-dependent manner. The present results demonstrate that BLM induces the secretion of IL-l@ from human AM (Figs. 2-4). Rodent alveolar macrophages have been stimulated by BLM either directly in vitro or by intratracheal instillation in vivo with variable secretion of IL-l& Treatment of hamster AM with BLM in vitro led to the acute (~24 h) release of IL-l activity, as measured by the effects of supernatants on thymidine incorporation into thymocytes (43). Similar results were obtained using rat AM, with significant lymphocyte-activating activity occurring at 0.05-5 hg/ml BLM (19). The present results obtained with human AM are consistent with these animal data in that the secretion of IL-lp in response to BLM was relatively rapid (Fig. 4) and was maximal at a BLM concentration of 5 mu/ml (-3 pg/ml; Fig. 2). We have shown that in the human AM, BLM has concentration-dependent effects on PDB-stimulated superoxide anion production, IL-l/? secretion, and intracellular CAMP levels, viz., a 24-h incubation with BLM caused increases in IL-16 secretion that were paralleled by decreases in both superoxide anion production and intracellular CAMP (Figs. 1,2, and 5). The mechanism(s) by which BLM could cause a decrease in intracellular CAMP are not clear and may involve the direct or indirect downregulation of the cyclase, stimulation of the diesterase, or both. In this regard, it is worth noting that the IL-l receptor (IL-1R) is coupled to adenylate cyclase (6, 41) and that the binding of IL-l@ to the IL-1R has been shown to result in a transient (~1 h) increase in intracellular CAMP levels. Whether a cause-and-effect relationship exists between longer term (~24 h) intracellular CAMP levels and either IL-l@ secretion or PDBstimulated superoxide anion production remains to be investigated, but it is tempting to speculate that at least some of these later events may result from the earlier actions of BLM on CAMP-dependent protein kinase. The more immediate (~1 h) effects of BLM on CAMP levels in the human AM are under investigation. Several studies have shown that agents that can increase intracellular CAMP levels also decrease IL-l@ secretion (3, 24). Suppression of IL-l@ secretion by CAMP appears to be exerted at the level of mRNA translation (23). This inverse relation between CAMP levels and IL-lp secretion is consistent with the present results, in which BLM-induced increases in IL-l@ are mirrored by decreases in intracellular CAMP levels. The fact that the inclusion of DBcAMP blocked BLM-induced IL-lp secretion bv >80%. even in the continued presence of
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
L390
BLEOMYCIN
STIMULATION
OF HUMAN
BLM (Fig. 5 inset), is consistent with the known inhibitory role of CAMP on IL-lp mRNA translation. Two studies in the rat model have pointed out that BLM stimulation of IL-l activity from AM appears to depend on the presence of unknown serum factors (9,19). In preliminary studies designed to investigate the importance of serum in the BLM-stimulated secretion of IL-l@ by human AM, we found a dependence on serum during the first 4 h of cell stimulation (data not shown). That is, cells stimulated for 4 h by BLM in the presence of serum and then incubated for an additional 20 h in the presence of serum produced more (Z-fold) IL-l@ than cells stimulated for the first 4 h with BLM in the absence of serum. One explanation for these results that is also consistent with the fact that only a 4-h contact time with drug was necessary for near-maximal secreted levels of IL-lp (Fig. 3) is that BLM stimulates transcription of the IL-lp gene, but a serum factor(s) is required for efficient translation of the transcript. Several studies of monocytes and macrophages have demonstrated that the secretion of IL-l& TNF-cw, and PGE2 may be mechanistically linked. For example, TNFCYhas been shown to stimulate IL-lp secretion (12), and IL-lp stimulates PGEB secretion (4). Further, PGE2 has been demonstrated to be a negative regulator of both IL-lb (23) and TNF-a (39). Thus the temporal pattern of the appearance of these products secreted from the human AM (Fig. 4) may be the result of an initial BLMinduced stimulation of TNF-a secretion followed by TNF-cu stimulation of IL-l@ and IL-lp stimulation of PGE2, which then downregulates TNF-cu secretion in an autocrine fashion. Such a sequence of events is consistent with recent results in the mouse model showing that BLM-induced fibrosis could be blocked by intravenous injections of anti-TNF-cu antibodies (36). Unequivocal elucidation of the mechanism(s) by which BLM induces the secretion of these cytokines awaits further study, e.g., an in vitro comparison of IL-l@ secretion by BLM-stimulated AM in the presence and absence of anti-TNF-a antibodies. It is of interest to compare the relative amounts of IL-l@, TNF-cu, IL-6, and PGE2 secreted by the human AM in response to BLM and LPS. IL-lp is known to have mitogenic activity for fibroblasts3 (40) and to result in increased fibroblast collagen production (28); TNF-cu has similar properties. On the other hand, PGE2 (14), and possibly IL-6 (25), can downregulate fibroblast growth. Thus one might expect that the balance of these two sets of factors would play an important role in the promotion of fibrogenesis. Stimulation of the human AM by LPS (10 pg/ml) resulted in the production of -15,000 pg IL-7,000 pg IL-G/ml, and lPlm4 - 2,400 pg TNF-a/ml, -3,500 pg PGEJml (see RESULTS). In contrast, BLM (5 mu/ml) stimulation resulted in the secretion of -3,500 -900 pg TNF-a/ml, -500 pg IL-G/ml, and Pg IL-W/ml, -400 pg PGEJml. Thus, compared with an LPS stimulation of the cell, BLM stimulation results in the secretion of much more IL-l@ and TNF-cu relative to IL-6 and PGE2. This shifting of the relative amounts of these two 3 IL-16
is a progression factor for fibroblasts
(40).
ALVEOLAR
MACROPHAGE
sets of regulatory factors toward growth stimulation may play a role in the initiation of BLM-induced fibrosis. Specific binding of BLM to AM has been reported recently in the rat model, where high- and low-affinity sites were found with dissociation constant (Kd values) of -0.5 and 65 PM, respectively (10). The above-noted concentration-dependent effects of BLM on IL-l& CAMP, and superoxide anion production all have identical mean effective concentrations of -2 mU BLM/ml (-1 PM). Thus, whereas it is possible that the effects of BLM on the human AM are a result of the binding of the drug to receptors in the cell plasma membrane followed by the generation of second messengers, it is also possible, given the DNA-binding properties of BLM, that these effects derive from the direct actions of internalized drug on transcription. In summary, the present results demonstrate that BLM has direct acute (524 h) effects on the secretion of proinflammatory modulators and cytokines by the human alveolar macrophage. The inability of BLM to stimulate AM oxygen radical production, together with its inhibition of PDB-stimulated superoxide anion production, would argue against a direct role for the macrophage in BLM-induced lung injury. However, the induction of significant IL-lb secretion by BLM, together with a corresponding near-absence of PGE2 secretion, suggests that the pulmonary macrophage may play a role in the development of the fibrosis that characterizes BLM-induced lung damage. The authors thank Dr. Robert A. Newman for helpful discussions prior to the submission of this manuscript. This work was supported by the Dept. of Pharmacology of the University of Texas Medical School and by National Institute of Environmental Health Sciences Grant ES-04804 and National Institutes of Health Clinical Research Center Grant MOl-RR-02558 to A. Holian. Present address of R. K. Scheule: Genzyme Corp., 1 Mountain Rd., Framingham, MA 01701-9322. Address reprint requests to A. Holian. Received 16 May 1991; accepted in final form 20 September 1991. REFERENCES 1. Akira, S., T. Hirano, T. Taga, and T. Kishimoto. Biology of the multifunctional cytokines: IL-6 and related molecules (IL-l and TNF). FASEB J. 4: 2860-2867, 1990. 2. Becker, S., R. B. Devlin, and J. S. Haskill. Differential production of tumor necrosis factor, macrophage colony stimulating factor, and interleukin 1 by human alveolar macrophages. J. Leukocyte 3.
Biol.
45: 353-361,
1989.
Brandwein, S. R. Regulation of interleukin mouse peritoneal macrophages. J. Biol. Chem.
1 production 261:
by
8624-8632,
1986.
Censini, S., M. Bartalini, A. Tagliabue, and D. Boraschi. Interleukin 1 stimulates production of LTC4 and other eicosanoids by macrophages. LymphoKine Res. 8: 107-114, 1989. 5. Chedid, M., and S. B. Mizel. Involvement of cyclic AMPdependent protein kinases in the signal transduction pathway for interleukin-1. IMoZ. CeZZ. BioZ. 10: 3824-3827, 1990. 6. Chedid, M., F. Shirakawa, P. Naylor, and S. B. Mizel. Signal transduction pathway for IL-l. Involvement of a pertussis toxin-sensitive GTP-binding protein in the activation of adenylate cyclase. J. Immunol. 142: 4301-4306, 1989. 7. Conley, N. S., J. W. Yarbro, H. A. Ferrari, and R. B. Zeidler. Bleomycin increases superoxide anion generation by pig peripheral alveolar macrophages. MoZ. PharmacoZ. 30: 48-52, 4.
1986. 8.
Dauber, J. H., M. D. Rossman, and R. P. Daniele. Bronchoalveolar cell populations in acute sarcoidosis. J. Lab. CZin. Med. 94: 862-871,
1979.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
BLEOMYCIN
STIMULATION
OF HUMAN
9. Denholm, E. M., and S. H. Phan. The effects of bleomycin on alveolar macrophage growth factor secretion. Am. J. PathoZ. 134: 355-363, 1989. 10. Denholm, E. M., and S. H. Phan. Bleomycin binding sites on alveolar macrophages. J. Leukocyte Biol. 48: 519-523, 1990. 11. Dinarello, C. A. Interleukin-1 and its biologically related cytokines. Adu. Immunol. 44: 153-205, 1989. 12. Dinarello, C. A., J. G. Cannon, S. M. Wolff, H. A. Bernheim, B. Beutler, A. Cerami, I. S. Figari, M. A. Palladino, Jr., and J. Y. O’Connor. Tumor necrosis factor (cachectin) is an endogenous pyrogen and induces production of interleukin 1. J. Exp. Med. 163: 1433-1450, 1986. 13. Douglas, M. J., and C. M. L. Coppin. Bleomycin and subsequent anesthesia: a retrospective study at Vancouver General Hospital. Can. Anaesth. Sot. J. 27: 449-452, 1980. 14. Elias, J. A., M. D. Rossman, R. B. Zurier, and R. P. Daniele. Human alveolar macrophage inhibition of lung fibroblast growth. A prostaglandin-dependent process. Am. Reu. Respir. Dis. 131: 94-99, 1985. 15. Elias, J. A., A. D. Schreiber, K. Gustilo, P. Chien, M. D. Rossman, P. J. Lammie, and R. P. Daniele. Differential interleukin 1 elaboration by density fractionated human alveolar macrophages and blood monocytes: relationship to cell maturity. J. Immunol. 135: 3198-3204, 1985. 16. Goldiner, P. L., G. C. Carlon, E. Cvitkovic, 0. Schweizer, and W. S. Howland. Factors influencing postoperative morbidity and mortality in patients treated with bleomycin. Br. Med. J. 1: 1664-1667, 1978. 17. Harrison, J. H., and J. S. Lazo. High dose continuous infusion of bleomycin in mice: a new model for drug-induced pulmonary fibrosis. J. Pharmacol. Exp. Ther. 243: 1185-1194, 1987. 18. Holian, A., and R. P. Daniele. Stimulation of oxygen consumption and superoxide anion production in pulmonary macrophages by N-formyl methionyl peptides. FEBS Lett. 108: 47-50, 1979. 19. Jordana, M., C. Richard, L. B. Irving, and J. Gauldie. Spontaneous in vitro release of alveolar-macrophage cytokines after the intratracheal instillation of bleomycin in rats. Am. Reu. Respir. Dis. 137: 1135-1140, 1988. 20. Jules-Elysee, K., and D. A. White. Bleomycin-induced pulmonary toxicity. Clin. Chest Med. 11: l-20, 1990. 21. Katakami, Y., Y. Nakao, T. Koizumi, N. Katakami, R. Ogawa, and T. Fujita. Regulation of tumor necrosis factor production by mouse peritoneal macrophages: the role of cellular cyclic AMP. Immunology 64: 719-724, 1988. 22. Kelley, J. Cytokines of the lung. Am. Rev. Respir. Dis. 141: 765-788, 1990. 23. Knudsen, P. J., C. A. Dinarello, and T. B. Strom. Prostaglandins posttranscriptionally inhibit monocyte expression of interleukin 1 activity by increasing intracellular cyclic adenosine monophosphate. J. Immunol. 137: 3189-3194, 1986. 24. Knudsen, P. J., C. A. Dinarello, and T. B. Strom. Glucocorticoids inhibit transcriptional and post-transcriptional expression of interleukin 1 in U937 cells. J. Immunol. 139: 4129-4134, 1987. 25. Kohase, M., D. Henriksen-DeStefano, L. T. May, J. Vilcek, and P. B. Sehgal. Induction of beta 2-interferon by tumor necrosis factor: a homeostatic mechanism in the control of cell proliferation. Cell 45: 659-666, 1986. 26. Kotloff, R. M., J. Little, and J. A. Elias. Human alveolar macrophage and blood monocyte interleukin-6 production. Am. J. Respir. Cell Mol. Biol. 3: 497-505, 1990. 27. Kovacs, E. J., and J. Kelley. Secretion of macrophage-derived growth factor during acute lung injury induced by bleomycin. J. Leukocyte Biol. 37: l-14, 1985. 28. Krane, S. M., J. M. Dayer, L. S. Simon, and L. S. Byrne. Mononuclear cell-conditioned medium containing mononuclear cell factor (MCF), homologous with interleukin 1, stimulates collagen and fibronectin synthesis by adherent rheumatoid synovial cells: effects of Prostaglandin E2 and indomethacin. Collagen Relat. Res. 5: 99-117, 1985.
ALVEOLAR
MACROPHAGE
L391
29. Lindenschmidt, R. C., A. F. Tryka, G. A. Godfrey, E. L. Frome, and H. Witschi. Intratracheal versus intravenous administration of bleomycin in mice: acute effects. Toxicol. Appl. Pharmacol. 85: 69-77, 1986. 30. Miller, G. A., and P. S. Monahan. Use of nonspecific esterase stain. In: Methods for Studying Mononuclear Phagocytes, edited by P. J. Edelson, H. S. Koron, and D. 0. Adams. New York: Academic, 1981, chapt. 40, p. 367-374. 31. Morley, J., M. A. Bray, R. W. Jones, D. H. Nugteren, and D. A. van Dorp. Prostaglandin and thromboxane production by human and guinea-pig macrophages and leucocytes. Prostuglandins 17: 729-746, 1979. 32. Nathan, C. F. Secretory products of macrophages. J. Clin. Invest. 79: 319-326, 1987. 33. Ohmori, Y., G. Strassman, and T. A. Hamilton. CAMP differentially regulates expression of mRNA encoding IL-la and ILlp in murine peritoneal macrophages. J. ImmunoZ. 145: 33333339, 1990. 34. Perkins, R. C., R. K. Scheule, and A. Holian. In vitro bioactivity of asbestos for the human alveolar macrophage and its modification by IgG. Am. J. Respir. Cell Mol. Biol. 4: 532-537, 1991. 35. Phan, S. H., and S. L. Kunkel. Inhibition of bleomycin-induced pulmonary fibrosis by nordihydroguiaretic acid. Am. J. Pathol. 124: 343-352, 1986. 36. Piguet, P. F., M. A. Collart, G. E. Grau, Y. Kapanci, and P. Vassalli. Tumor necrosis factor/cachetin plays a key role in bleomycin-induced pneumopathy and fibrosis. J. Exp. Med. 170: 655-663, 1989. 37. Rinaldo, J., R. H. Goldstein, and G. L. Snider. Modification of oxygen toxicity after lung injury by bleomycin in hamsters. Am. Rev. Respir. Dis. 126: 1030-1033, 1982. 38. Sausville, E. A., R. W. Stein, J. Peisach, and S. B. Horwitz. Properties and products of the degradation of DNA by bleomycin and iron(I1). Biochemistry 17: 2746-2754, 1978. 39. Scales, W. E., S. W. Chensue, I. Otterness, and S. L. Kunkel. Regulation of monokine gene expression: prostaglandin E2 suppresses tumor necrosis factor but not interleukin-lcu or ,&mRNA and cell-associated bioactivity. J. Leukocyte Biol. 45: 416-421, 1989. 40. Schmidt, J. A., S. B. Mizel, D. Cohen, and I. Green. Interleukin 1, a potential regulator of fibroblast proliferation. J. Immunol. 128: 2177-2182, 1982. 41. Shirakawa, F., U. Yamashita, M. Chedid, and S. B. Mizel. Cyclic AMP-an intracellular second messenger for interleukin 1. Proc. Natl. Acad. Sci. USA 85: 8201-8205, 1988. 42. Sogal, R. N., A. A. Gottlieb, A. R. Boutros, R. R. Ganapathi, R. R. Tubbs, P. Satariano, P. R. Blakely, and G. J. Beck. Effect of oxygen on bleomycin-induced lung damage. CZevel. Clin. J. Med. 54: 503-509, 1987. 43. Suwabe, A., K. Takahashi, S. Yasui, S. Arai, and F. Sendo. Bleomycin-stimulated hamster alveolar macrophages release interleukin-1. Am. J. PathoZ. 132: 512-520, 1988. 44. Taffet, S. M., K. J. Singhel, J. F. Overholtzer, and S. A. Shurtleff. Regulation of tumor necrosis factor expression in a macrophage-like cell line by lipopolysaccharide and cyclic AMP. Cell Immunol. 120: 291-300, 1989. 45. Tryka, A. F., W. A. Skornik, J. J. Godleski, and J. D. Brain. Potentiation of bleomycin-induced lung injury by exposure to 70% oxygen. Morphological assessment. Am. Reu. Respir. Dis. 126: 1074-1079, 1982. 46. Wahl, S. M. Lymphocyteand macrophage-derived growth factors. Methods Enzymol. 163: 715-731, 1988. 47. Weissmann, G., J. E. Smolen, and H. Korchak. Prostaglandins and inflammation: receptor/cyclase coupling as an explanation of why PGEs and PGI:! inhibit functions of inflammatory cells. Adv. Prostaglandin Thromboxane Res. 8: 1637-1653, 1980. 48. Wewers, M. D., S. I. Rennard, A. J. Hance, P. B. Bitterman, and R. J. Crystal. Normal human alveolar macrophages have limited capacity to release interleukin-1. J. CZin. Invest. 74: 2208-2218, 1984.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
of cytokine secretion macrophage
RONALD K. SCHEULE, RAYMOND C. PERKINS, RAYMOND HAMILTON, AND ANDRIJ HOLIAN Departments of Internal Medicine and Pharmacology, Medical School, Houston, Texas 77225 Scheule, Ronald K., Raymond C. Perkins, Raymond Hamilton, and Andrij Holian. Bleomycin stimulation of cytokine secretion by the human alveolar macrophage. Am. J. Physiol. 262 (Lung Cell. Mol. Physiol. 6): L386-L391, 1992.-- Bleomycin (BLM) is a very effective antineoplastic drug for many gynecologic and urinary tract carcinomas. However, its use, e.g., cumulative dosage, often is limited by the pulmonary fibrosis that it causes. The mechanism by which BLM causes fibrosis is not understood but is proposed to involve the pulmonary macrophage, a central cell in the cytokine network of the lung. To examine the direct effects of this drug on the human alveolar macrophage, we have treated human alveolar macrophages (isolated from normal subjects by bronchoalveolar lavage) with BLM in vitro and examined resultant macrophage secretory products that have importance for inflammatory and fibrotic processes. A 24-h treatment with BLM (0.5400 mU/ ml) was found to result in 1) a concentration-dependent decrease in the ability of the macrophage to produce superoxide anion in response to phorbol 12,13-dibutyrate, 2) an increase in secreted interleukin-lp (IL-ID), and 3) a decrease in intracellular levels of adenosine 3’,5’-cyclic monophosphate. Kinetic studies revealed a time-dependent appearance of BLM-induced cytokines; tumor necrosis factor-a could be detected as early as 4 h after stimulation, followed by IL-lp at 8 h. The secretion of these cytokines was found to precede the release of prostaglandin E,, which became significant only at 24 h. Taken together, the present results imply that the human alveolar macrophage does not contribute to BLM-induced oxidant injury of the lung but that it may contribute to the development of BLM-induced pulmonary fibrosis. interleukinI& tumor necrosis factor-a; adenosine 3’,5’-cyclic monophosphate
superoxide
anion;
(BLM) is a group of glycopeptides isolated from Streptomyces verticiks. BLM-A2 is the major component used clinically for its antineoplastic properties. In the lung, BLM-induced fibrosis limits the usefulness of the drug (20). The mechanism(s) by which BLM causes fibrosis in humans is not yet understood but is likely to involve pulmonary (alveolar and interstitial) 1 macrophages, the resident phagocytic cells of the lung. Due to its ability to secrete a wide array of inflammatory mediators and growth factors (22, 32, 46), the lung macrophage plays a key role in acute inflammatory responses as well as in the chronic fibrotic responses of the lung to injury. In particular, stimulation of human alveolar macrophages (AM) has been shown to result in the production of three multifunctional cytokines, namely interleukin-1 (IL-l) (15, 48), interleukin-6 (IL6) (26), and tumor necrosis factor-a (TNF-a) (2). These three macrophage-derived cytokines have overlapping --------. -----l_l--BLEOMYCIN
l The terms “lung” or “pulmonary” macrophage changeably to mean cells that could be either alveolar interstitial macrophages. L386
1040-0605/92
$2.00
are used intermacrophages or Copyright
The University
of Texas
activities and are important mediators of both inflammatory and immune responses (1) that may contribute to the generation of a fibrotic response. Individually, IL-l and TNF-cu potentiate fibroblast growth and collagen production (11,25). These mitogenic effects on the fibroblast are generally opposed by the negative regulatory effects of the arachidonic acid metabolite prostaglandin E2 (PGE2, 14, 25). BLM-induced fibrosis resembling the human disease process has been produced in a variety of animal models (9, 17, 19, 27, 29, 35, 43), and AM from these models have been examined for their involvement in the fibrotic process. For example, AM isolated from rats after a single intratracheal instillation of BLM were found to secrete elevated levels of IL-1 activity (19). In addition to these studies of in vivo exposed AM, which have the potential of revealing both direct and indirect effects of the drug on the AM, several investigations have examined the direct effects of BLM on the AM by exposing them to the drug in vitro (19). Thus, in one study (9), BLM-treated rat AM were found to secrete increased amounts of a macrophage-derived growth factor (MDGF) for fibroblasts but not IL-l or TNF-cu, whereas in another study (43), hamster AM were found to secrete elevated levels of IL-l activity. Although animal models of BLM-induced fibrosis have led to insights into the human disease processes caused by this drug, no studies to date have reported responses of the human AM to BLM. The present study was conducted to help define the direct effects of BLM on the human AM by exposing these cells to the drug in vitro and characterizing the acute (92% macrophages [as identified by Giemsa staining (Diff-Quik; Scientific Products) and confirmed by esterase staining (30)] of Cytospin preparations. Cells were ~85% viable by trypan blue exclusion. Cells (I X 106/ml) were cultured for up to 24 h in the presence or absence of BLM in N-2-hydroxyethylpiperazine-N’-2ethanesulfonic acid (HEPES) -buffered medium 199 containing 10% heat-inactivated fetal calf serum (FCS; Sigma). BLM concentrations were chosen to span a relevant physiological dose.2 Cells were maintained in suspension by slow end-over-end tumbling in polypropylene tubes at 37°C. Incubations were terminated by pelleting the cells, and supernatants were stored at -80” C until assayed.
Assays Under no condition was BLM found to be cytotoxic (by trypan blue and cell counting), even at the highest concentrations of BLM used (100 mu/ml) after a 24-h incubation. Endotoxin levels in buffers, media, and BLM stock solutions were assessed by the Limulus amoebocyte assay (Sigma); endotoxin levels were in all cases found to be ~20 pg/ml. For a typical experiment using 5 mu/ml BLM, this result translates into a maximum endotoxin level arising from the BLM of ~20 fg/ml. Thus any BLM-dependent results are due to the effects of the drug itself and not endotoxin. Assay for superoxide anion. Superoxide anion production was quantitated by the reduction of cytochrome c and was followed spectrophotometrically by absorbance at 550 nm (18). Data are from cell suspensions maintained in MHS containing 2 mM Ca2+ at 37°C for 1 h at 0.5 x lo6 cells/ml. Phorbol 12,13dibutyrate (PDB) at 1 PM served as a positive stimulant of superoxide anion production. Assays for IL-ID, IL-6, TNF-CU, PGE2, and adenosine 3’,5’cyclic monophosphate (CAMP). The macrophage cytokines ILI& IL-6, and TNF-cu were assayed in cell-free supernatants with the use of radioimmunoassay (RIA) kits (Advanced Magnetics). Lipopolysaccharide at 10 pg/ml was used as a positive control for stimulation of the human AM. Intracellular CAMP was released from pelleted cells by 5% trichloroacetic acid (TCA). Water-saturated diethyl ether was used to extract the TCA from the aqueous samples, which were then dried with a SpeedVac concentrator (Savant Industries, Hicksville, NY) and used for CAMP RIA assays (New England Nuclear, Boston, MA). PGE, was quantitated in supernatants by constructing an RIA using a rabbit anti-PGE antiserum (Sigma; reacts with both PGE, and PGE2) together with [“H(N)] -prostaglandin E2 ( [3H]PGE2, Amersham; 200 Ci/mmol) and authentic PGE, standards (Sigma). This assay had a working range of 20 pg/ml-10 rig/ml. Because PGE2 is the major PGE produced by human AM (31), the PGE detected by this assay will be referred to as PGE2.
ALVEOLAR
L387
MACROPHAGE
(ANOVA) followed wise comparisons.
by a Student-Newman-Keuls
test for pair-
RESULTS
Macrophage-derived m.odulators. Isolated human al.veolar macrophages were exposed to BLM in vitro, and their secretion of superoxide anion and fibroblast-active factors was quantitated. Figure 1 shows that a 24-h incubation of human AM with BLM results in a depression of the ability of PDB to stimulate the cell that is directly related to the concentration of BLM. For example, after an incubation with 5 mu/ml BLM, stimulation with PDB reveals a decrease of ~60% in PDB-stimulated superoxide anion production compared with cells incubated in the absence of drug. Basal superoxide anion production was not significantly altered by the presence of BLM. PGE2 is a modulator secreted by AM with demonstrated growth inhibitory activity for fibroblasts (14). Indirect evidence suggests that IL-6 may have a similar activity (25). While BLM (5 mu/ml) treatment for 24 h stimulated some PGEz secretion by human AM, the amount was small and could not be related to the concentration of BLM outside of error (data not shown). The low level of PGE2 released from BLM-treated cells (200500) pg/ml) was significantly less than the 3,000-4,000 pg/ml released from cells stimulated by LPS over the same time period. Likewise, the BLM-stimulated production of IL-6 over a 24-h period was also minimal (-500 pg/ml) compared with the amount secreted in response to LPS (w 7,000 pg/ml; data not shown). The macrophage-derived cytokines IL-lp and TNF-cu are positive regulators of fibroblast growth (22). Figure 2 demonstrates that IL-lp release from human AM is dependent on the concentration of BLM; significant increases could be detected at 0.5 mU BLM/ml (0.3 pg/ml). The maximal amount of IL-lp released was less than that released by 10 pg/ml LPS, which was at least 10,000 pg IL-lP/ml under these conditions. The amount of TNF-cx secreted in response to BLM (5 mu/ml) at 24
0
0
Statis tics Values are presented as means t SE. The number of individuals whose cells were used for a given experiment is denoted by n in the corresponding figure legend. For each experiment, statistical treatment included a one-way analysis of variance 2 For a typical m2 and a blood blood concentration tration and may environment of this study vary
dose of volume of or may the lung between
35 U/m2, and using an adult surface area of 1.5 of 5 liter, one can calculate an approximate - 10 mU BLM/ml. This is an average concennot reflect the local drug concentration in the macrophage. Concentrations of BLM used in 0.5 and 100 mu/ml.
0.5
5
50
100
IBLMI, ml-J/d Fig. 1. Phorbol l&13-dibutyrate (PDB)-stimulated superoxide anion production after incubation of human alveolar macrophages with varying concentrations of bleomycin (BLM). Cells were incubated for 24 h at 37°C in suspension with indicated concentrations of BLM and then treated for 1 h with either medium (filled bars) or an equal volume of PDB in medium [final concentration of PDB = 1 ,uM, (hatched bars)]. As a positive control, cells were also incubated with lipopolysaccharide (LPS, 10 pg/ml) for 24 h. Superoxide anion production was quantitated from cell supernatants as described in METHODS. *Values significantly different (P < 0.05) from those obtained with cells incubated in absence of BLM, i.e., [BLM] = 0. n = 5 individuals.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
L388
BLEOMYCIN
0
0.5
STIMULATION
5
50
OF HUMAN
100
IBLMI, mU/d
ALVEOLAR
MACROPHAGE
0
10
20
TIME (h>
Fig. 2. Human alveolar macrophage secretion of interleukin-lp (IL-lp) in response to varying concentrations of BLM. Cells in suspension were incubated with BLM for 24 h and IL-lp was quantitated in the cell-free supernatants by radioimmunoassay (RIA) as described in METHODS. *Values significantly different (P < 0.05) from those obtained with cells incubated in absence of BLM, i.e., [BLM] = 0. n = 5.
Fig. 4. Kinetics of BLM-stimulated secretion of IL-lb (A), tumor necrosis factor-a (TNFcr, o), and PGEZ (0) by human alveolar macrophages incubated in presence of 5 mu/ml BLM. Cells in suspension were incubated with BLM, and cell-free supernatants were analyzed for IL-l@, TNF-a, and PGEZ as described in METHODS. Cytokine secretion at each time point by control cells has been subtracted from these data. n = 4.
rapid compared with that of IL-l& Significant secretion of TNF-a is apparent at 4 h, whereas IL-lp secretion becomes significant only after 8-10 h. The secretion of g4ooo TNF-cu reaches maximal levels around 15 h, whereas the 23 a secretion of IL-l@ continues to increase even at 24 h. CQ Finally, the onset of PGEz secretion appears to lag behind 72alo those of both TNF-cu and IL-l@ and becomes significant d only after a 24-h incubation with BLM. In contrast to the concentration dependence of IL-lp secretion (Fig. 2), the relatively low amounts of TNF-a and PGEB secreted at 24 0 24 h precluded a demonstration of their dependence on the Contact Time (h) concentration of BLM (data not shown). Fig. 3. Human alveolar macrophage secretion of IL-ID in response to Intracellular signaling pathways. PGE2 (47) and IL-1 variable initial contact time with 5 mu/ml BLM. After indicated incu(5,41) stimulate adenylate cyclase, and CAMP is involved bation time in presence of BLM, cells were washed twice and placed in fresh medium without BLM for remainder of a 24-h incubation. IL-lp in the regulation of IL-l@ and TNF-cu expression (21, 33, was quantitated in cell-free supernatants by RIA as described in 44). Potential interactions between intracellular signaling METHODS. *Values significantly different (P < 0.05) from those pathways and some of the observed effects of BLM on the obtained with cells incubated in absence of BLM, i.e., [BLM] = 0. human AM were explored by quantitating the levels of n= 5. intracellular CAMP after a 24-h incubation with the drug. h was 900 t 500 pg/ml; for comparison, LPS induced the Figure 5 shows that incubating these cells with BLM resulted in a concentration-dependent decrease in the levsecretion of ~2,400 pg TNF-a/ml. Blur contact time. The above-described dependence of els of intracellular CAMP. A significant decrease in CAMP cytokine release on BLM was noted in cells that had been was evident at a concentration of 0.5 mU BLM/ml and resulted in an intracellular level of CAMP equivalent to maintained in the presence of drug for the entire 24-h incubation. To define a “contact time” with the drug that resulting from incubating the cells with 10 pg/ml that was sufficient to stimulate cytokine release, cells LPS, a known cellular stimulant. To further explore the apparent (inverse) relationship were incubated with BLM for various times, washed free between the concentration-dependent, BLM-induced of drug, and then incubated for the remainder of the 24-h increases in IL-lp and decreases in CAMP, cells were time period in BLM-free medium. The cytokine quantitated was IL-l@ because its secretion at 24 h was the most incubated with 5 mU BLM/ml in the presence or absence significant response of the cells to BLM. Figure 3 shows of the poorly hydrolyzed CAMP analogue N,Y-O-dibuthat cells incubated with BLM for 2-4 h produced almost tyryladenosine 3’,5’-cyclic monophosphate (DBcAMP), the same amount of IL-10 in a 24-h time period as cells and IL-l@ levels were measured after 24 h. The inset in that had been incubated with BLM for the entire 24 h. Fig. 5 shows that, in the presence of DBcAMP, BLM This observation indicates that a 2-4 h contact time of stimulation of IL-l@ secretion was reduced by ~80%. BLM with the cells is necessary to initiate the processes Thus, when CAMP levels are kept artificially high, the that result in IL-l@ secretion but that BLM is not nec- BLM-induced secretion of IL-lp is significantly reduced. essary thereafter. DISCUSSION Cytokine kinetics. Figure 4 illustrates the time-dependent secretion of TNF-cu, IL-l& and PGEB by the human The mechanism(s) by which BLM causes an interstiAM in response to the continuous presence of 5 mu/ml tial pulmonary pneumonitis that can lead to fibrosis is BLM. A significant feature of these data is the different not known but may involve oxidative or fibrogenic potentemporal pattern of these secreted macrophage products. tiating activities of th .e alveolar macrophage. The use of For example, the onset of TNF-cu secretion is relatively supplemental oxygen therapy in patients treated with 1
T
1
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
BLEOMYCIN
0
0.5
5
[BLM],
50
STIMULATION
100
OF HUMAN
LPS
mu/ml
Fig. 5. Intracellular CAMP levels in human alveolar macrophages after 24-h incubation with a variable concentration of BLM. As a positive control, cells were also incubated with LPS (10 pg/ml) for 24 h. Intracellular CAMP was quantitated by RIA as described in METHODS. *Values significantly different (P < 0.05) from those obtained with cells incubated in absence of BLM, i.e., [BLM] = 0. n = 5. Inset: effect of dibutyryl CAMP (DBcAMP) on BLM-induced IL-10 secretion by human alveolar macrophages. Cells were stimulated with 5 mu/ml BLM for 24 h in presence of 2 mM DBcAMP, and IL-lb was quantitated in the supernatant. Control values of IL-1p secretion either in presence or absence of DBcAMP were not significantly different from 0 (30 t 30 pg/ml). *Significantly different (P < 0.05) from value obtained in absence of DBcAMP. n = 4.
BLM has been associated with an increased risk of fibrosis (16), but this conclusion is controversial (13). Several animal studies (37,42,45) have supported a potentiating role for hyperoxia in BLM-induced fibrosis. BLMinduced pulmonary toxicity, especially under hyperoxic conditions, may be a result of the formation of oxygen radicals, inasmuch as the generation of such radicals is thought to be the mechanism of BLM-induced cytotoxicity (38). One immediate response of the AM to some stimulants is the generation of the reactive oxygen species superoxide anion. Superoxide anion is cytotoxic and can also contribute to inflammation in vivo. In the present studies, BLM itself was not a stimulant for superoxide anion production by human AM (data not shown), a result that is consistent with an analogous finding in the pig model (7). Thus these results argue against a direct role for the AM in BLM-induced oxidant injury in humans. However, they do not rule out the possibility of more indirect effects of BLM on the human AM that may contribute to inflammation or fibrosis. One example of an indirect effect of BLM is the finding that a 24-h BLM treatment of human AM suppressed the subsequent ability of the cell to produce superoxide anion when stimulated by PDB (Fig. 1). The results shown in Fig. 1 depict deficits in PDB-stimulated superoxide anion production after a 24-h incubation with BLM. In contrast, l-h incubations of human AM with either 5 or 50 mu/ml BLM did not alter the ability of the cell to respond to submaximal concentrations of PDB by producing superoxide anion (data not shown). In a related animal study (7), short-term (~60 min) incubations of pig AM with BLM were found to result in an increase in the ability of PDB to stimulate superoxide anion. These latter results are in direct contrast to the present findings
ALVEOLAR
MACROPHAGE
L389
and may reflect species-dependent effects of BLM on the alveolar macrophage. The ability of BLM to suppress PDB-stimulated superoxide anion production was found to be directly dependent on the concentration of BLM (Fig. 1). Phorbol esters such as PDB are thought to stimulate superoxide anion production by directly activating protein kinase C (PKC), which then activates the oxidase. Thus the observation that BLM treatment of the AM decreases the subsequent ability of PDB to stimulate superoxide anion production (Fig. 1) implies that PKC or a component(s) of the oxidase has been downregulated or inactivated by BLM in a concentration-dependent manner. The present results demonstrate that BLM induces the secretion of IL-l@ from human AM (Figs. 2-4). Rodent alveolar macrophages have been stimulated by BLM either directly in vitro or by intratracheal instillation in vivo with variable secretion of IL-l& Treatment of hamster AM with BLM in vitro led to the acute (~24 h) release of IL-l activity, as measured by the effects of supernatants on thymidine incorporation into thymocytes (43). Similar results were obtained using rat AM, with significant lymphocyte-activating activity occurring at 0.05-5 hg/ml BLM (19). The present results obtained with human AM are consistent with these animal data in that the secretion of IL-lp in response to BLM was relatively rapid (Fig. 4) and was maximal at a BLM concentration of 5 mu/ml (-3 pg/ml; Fig. 2). We have shown that in the human AM, BLM has concentration-dependent effects on PDB-stimulated superoxide anion production, IL-l/? secretion, and intracellular CAMP levels, viz., a 24-h incubation with BLM caused increases in IL-16 secretion that were paralleled by decreases in both superoxide anion production and intracellular CAMP (Figs. 1,2, and 5). The mechanism(s) by which BLM could cause a decrease in intracellular CAMP are not clear and may involve the direct or indirect downregulation of the cyclase, stimulation of the diesterase, or both. In this regard, it is worth noting that the IL-l receptor (IL-1R) is coupled to adenylate cyclase (6, 41) and that the binding of IL-l@ to the IL-1R has been shown to result in a transient (~1 h) increase in intracellular CAMP levels. Whether a cause-and-effect relationship exists between longer term (~24 h) intracellular CAMP levels and either IL-l@ secretion or PDBstimulated superoxide anion production remains to be investigated, but it is tempting to speculate that at least some of these later events may result from the earlier actions of BLM on CAMP-dependent protein kinase. The more immediate (~1 h) effects of BLM on CAMP levels in the human AM are under investigation. Several studies have shown that agents that can increase intracellular CAMP levels also decrease IL-l@ secretion (3, 24). Suppression of IL-l@ secretion by CAMP appears to be exerted at the level of mRNA translation (23). This inverse relation between CAMP levels and IL-lp secretion is consistent with the present results, in which BLM-induced increases in IL-l@ are mirrored by decreases in intracellular CAMP levels. The fact that the inclusion of DBcAMP blocked BLM-induced IL-lp secretion bv >80%. even in the continued presence of
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
L390
BLEOMYCIN
STIMULATION
OF HUMAN
BLM (Fig. 5 inset), is consistent with the known inhibitory role of CAMP on IL-lp mRNA translation. Two studies in the rat model have pointed out that BLM stimulation of IL-l activity from AM appears to depend on the presence of unknown serum factors (9,19). In preliminary studies designed to investigate the importance of serum in the BLM-stimulated secretion of IL-l@ by human AM, we found a dependence on serum during the first 4 h of cell stimulation (data not shown). That is, cells stimulated for 4 h by BLM in the presence of serum and then incubated for an additional 20 h in the presence of serum produced more (Z-fold) IL-l@ than cells stimulated for the first 4 h with BLM in the absence of serum. One explanation for these results that is also consistent with the fact that only a 4-h contact time with drug was necessary for near-maximal secreted levels of IL-lp (Fig. 3) is that BLM stimulates transcription of the IL-lp gene, but a serum factor(s) is required for efficient translation of the transcript. Several studies of monocytes and macrophages have demonstrated that the secretion of IL-l& TNF-cw, and PGE2 may be mechanistically linked. For example, TNFCYhas been shown to stimulate IL-lp secretion (12), and IL-lp stimulates PGEB secretion (4). Further, PGE2 has been demonstrated to be a negative regulator of both IL-lb (23) and TNF-a (39). Thus the temporal pattern of the appearance of these products secreted from the human AM (Fig. 4) may be the result of an initial BLMinduced stimulation of TNF-a secretion followed by TNF-cu stimulation of IL-l@ and IL-lp stimulation of PGE2, which then downregulates TNF-cu secretion in an autocrine fashion. Such a sequence of events is consistent with recent results in the mouse model showing that BLM-induced fibrosis could be blocked by intravenous injections of anti-TNF-cu antibodies (36). Unequivocal elucidation of the mechanism(s) by which BLM induces the secretion of these cytokines awaits further study, e.g., an in vitro comparison of IL-l@ secretion by BLM-stimulated AM in the presence and absence of anti-TNF-a antibodies. It is of interest to compare the relative amounts of IL-l@, TNF-cu, IL-6, and PGE2 secreted by the human AM in response to BLM and LPS. IL-lp is known to have mitogenic activity for fibroblasts3 (40) and to result in increased fibroblast collagen production (28); TNF-cu has similar properties. On the other hand, PGE2 (14), and possibly IL-6 (25), can downregulate fibroblast growth. Thus one might expect that the balance of these two sets of factors would play an important role in the promotion of fibrogenesis. Stimulation of the human AM by LPS (10 pg/ml) resulted in the production of -15,000 pg IL-7,000 pg IL-G/ml, and lPlm4 - 2,400 pg TNF-a/ml, -3,500 pg PGEJml (see RESULTS). In contrast, BLM (5 mu/ml) stimulation resulted in the secretion of -3,500 -900 pg TNF-a/ml, -500 pg IL-G/ml, and Pg IL-W/ml, -400 pg PGEJml. Thus, compared with an LPS stimulation of the cell, BLM stimulation results in the secretion of much more IL-l@ and TNF-cu relative to IL-6 and PGE2. This shifting of the relative amounts of these two 3 IL-16
is a progression factor for fibroblasts
(40).
ALVEOLAR
MACROPHAGE
sets of regulatory factors toward growth stimulation may play a role in the initiation of BLM-induced fibrosis. Specific binding of BLM to AM has been reported recently in the rat model, where high- and low-affinity sites were found with dissociation constant (Kd values) of -0.5 and 65 PM, respectively (10). The above-noted concentration-dependent effects of BLM on IL-l& CAMP, and superoxide anion production all have identical mean effective concentrations of -2 mU BLM/ml (-1 PM). Thus, whereas it is possible that the effects of BLM on the human AM are a result of the binding of the drug to receptors in the cell plasma membrane followed by the generation of second messengers, it is also possible, given the DNA-binding properties of BLM, that these effects derive from the direct actions of internalized drug on transcription. In summary, the present results demonstrate that BLM has direct acute (524 h) effects on the secretion of proinflammatory modulators and cytokines by the human alveolar macrophage. The inability of BLM to stimulate AM oxygen radical production, together with its inhibition of PDB-stimulated superoxide anion production, would argue against a direct role for the macrophage in BLM-induced lung injury. However, the induction of significant IL-lb secretion by BLM, together with a corresponding near-absence of PGE2 secretion, suggests that the pulmonary macrophage may play a role in the development of the fibrosis that characterizes BLM-induced lung damage. The authors thank Dr. Robert A. Newman for helpful discussions prior to the submission of this manuscript. This work was supported by the Dept. of Pharmacology of the University of Texas Medical School and by National Institute of Environmental Health Sciences Grant ES-04804 and National Institutes of Health Clinical Research Center Grant MOl-RR-02558 to A. Holian. Present address of R. K. Scheule: Genzyme Corp., 1 Mountain Rd., Framingham, MA 01701-9322. Address reprint requests to A. Holian. Received 16 May 1991; accepted in final form 20 September 1991. REFERENCES 1. Akira, S., T. Hirano, T. Taga, and T. Kishimoto. Biology of the multifunctional cytokines: IL-6 and related molecules (IL-l and TNF). FASEB J. 4: 2860-2867, 1990. 2. Becker, S., R. B. Devlin, and J. S. Haskill. Differential production of tumor necrosis factor, macrophage colony stimulating factor, and interleukin 1 by human alveolar macrophages. J. Leukocyte 3.
Biol.
45: 353-361,
1989.
Brandwein, S. R. Regulation of interleukin mouse peritoneal macrophages. J. Biol. Chem.
1 production 261:
by
8624-8632,
1986.
Censini, S., M. Bartalini, A. Tagliabue, and D. Boraschi. Interleukin 1 stimulates production of LTC4 and other eicosanoids by macrophages. LymphoKine Res. 8: 107-114, 1989. 5. Chedid, M., and S. B. Mizel. Involvement of cyclic AMPdependent protein kinases in the signal transduction pathway for interleukin-1. IMoZ. CeZZ. BioZ. 10: 3824-3827, 1990. 6. Chedid, M., F. Shirakawa, P. Naylor, and S. B. Mizel. Signal transduction pathway for IL-l. Involvement of a pertussis toxin-sensitive GTP-binding protein in the activation of adenylate cyclase. J. Immunol. 142: 4301-4306, 1989. 7. Conley, N. S., J. W. Yarbro, H. A. Ferrari, and R. B. Zeidler. Bleomycin increases superoxide anion generation by pig peripheral alveolar macrophages. MoZ. PharmacoZ. 30: 48-52, 4.
1986. 8.
Dauber, J. H., M. D. Rossman, and R. P. Daniele. Bronchoalveolar cell populations in acute sarcoidosis. J. Lab. CZin. Med. 94: 862-871,
1979.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
BLEOMYCIN
STIMULATION
OF HUMAN
9. Denholm, E. M., and S. H. Phan. The effects of bleomycin on alveolar macrophage growth factor secretion. Am. J. PathoZ. 134: 355-363, 1989. 10. Denholm, E. M., and S. H. Phan. Bleomycin binding sites on alveolar macrophages. J. Leukocyte Biol. 48: 519-523, 1990. 11. Dinarello, C. A. Interleukin-1 and its biologically related cytokines. Adu. Immunol. 44: 153-205, 1989. 12. Dinarello, C. A., J. G. Cannon, S. M. Wolff, H. A. Bernheim, B. Beutler, A. Cerami, I. S. Figari, M. A. Palladino, Jr., and J. Y. O’Connor. Tumor necrosis factor (cachectin) is an endogenous pyrogen and induces production of interleukin 1. J. Exp. Med. 163: 1433-1450, 1986. 13. Douglas, M. J., and C. M. L. Coppin. Bleomycin and subsequent anesthesia: a retrospective study at Vancouver General Hospital. Can. Anaesth. Sot. J. 27: 449-452, 1980. 14. Elias, J. A., M. D. Rossman, R. B. Zurier, and R. P. Daniele. Human alveolar macrophage inhibition of lung fibroblast growth. A prostaglandin-dependent process. Am. Reu. Respir. Dis. 131: 94-99, 1985. 15. Elias, J. A., A. D. Schreiber, K. Gustilo, P. Chien, M. D. Rossman, P. J. Lammie, and R. P. Daniele. Differential interleukin 1 elaboration by density fractionated human alveolar macrophages and blood monocytes: relationship to cell maturity. J. Immunol. 135: 3198-3204, 1985. 16. Goldiner, P. L., G. C. Carlon, E. Cvitkovic, 0. Schweizer, and W. S. Howland. Factors influencing postoperative morbidity and mortality in patients treated with bleomycin. Br. Med. J. 1: 1664-1667, 1978. 17. Harrison, J. H., and J. S. Lazo. High dose continuous infusion of bleomycin in mice: a new model for drug-induced pulmonary fibrosis. J. Pharmacol. Exp. Ther. 243: 1185-1194, 1987. 18. Holian, A., and R. P. Daniele. Stimulation of oxygen consumption and superoxide anion production in pulmonary macrophages by N-formyl methionyl peptides. FEBS Lett. 108: 47-50, 1979. 19. Jordana, M., C. Richard, L. B. Irving, and J. Gauldie. Spontaneous in vitro release of alveolar-macrophage cytokines after the intratracheal instillation of bleomycin in rats. Am. Reu. Respir. Dis. 137: 1135-1140, 1988. 20. Jules-Elysee, K., and D. A. White. Bleomycin-induced pulmonary toxicity. Clin. Chest Med. 11: l-20, 1990. 21. Katakami, Y., Y. Nakao, T. Koizumi, N. Katakami, R. Ogawa, and T. Fujita. Regulation of tumor necrosis factor production by mouse peritoneal macrophages: the role of cellular cyclic AMP. Immunology 64: 719-724, 1988. 22. Kelley, J. Cytokines of the lung. Am. Rev. Respir. Dis. 141: 765-788, 1990. 23. Knudsen, P. J., C. A. Dinarello, and T. B. Strom. Prostaglandins posttranscriptionally inhibit monocyte expression of interleukin 1 activity by increasing intracellular cyclic adenosine monophosphate. J. Immunol. 137: 3189-3194, 1986. 24. Knudsen, P. J., C. A. Dinarello, and T. B. Strom. Glucocorticoids inhibit transcriptional and post-transcriptional expression of interleukin 1 in U937 cells. J. Immunol. 139: 4129-4134, 1987. 25. Kohase, M., D. Henriksen-DeStefano, L. T. May, J. Vilcek, and P. B. Sehgal. Induction of beta 2-interferon by tumor necrosis factor: a homeostatic mechanism in the control of cell proliferation. Cell 45: 659-666, 1986. 26. Kotloff, R. M., J. Little, and J. A. Elias. Human alveolar macrophage and blood monocyte interleukin-6 production. Am. J. Respir. Cell Mol. Biol. 3: 497-505, 1990. 27. Kovacs, E. J., and J. Kelley. Secretion of macrophage-derived growth factor during acute lung injury induced by bleomycin. J. Leukocyte Biol. 37: l-14, 1985. 28. Krane, S. M., J. M. Dayer, L. S. Simon, and L. S. Byrne. Mononuclear cell-conditioned medium containing mononuclear cell factor (MCF), homologous with interleukin 1, stimulates collagen and fibronectin synthesis by adherent rheumatoid synovial cells: effects of Prostaglandin E2 and indomethacin. Collagen Relat. Res. 5: 99-117, 1985.
ALVEOLAR
MACROPHAGE
L391
29. Lindenschmidt, R. C., A. F. Tryka, G. A. Godfrey, E. L. Frome, and H. Witschi. Intratracheal versus intravenous administration of bleomycin in mice: acute effects. Toxicol. Appl. Pharmacol. 85: 69-77, 1986. 30. Miller, G. A., and P. S. Monahan. Use of nonspecific esterase stain. In: Methods for Studying Mononuclear Phagocytes, edited by P. J. Edelson, H. S. Koron, and D. 0. Adams. New York: Academic, 1981, chapt. 40, p. 367-374. 31. Morley, J., M. A. Bray, R. W. Jones, D. H. Nugteren, and D. A. van Dorp. Prostaglandin and thromboxane production by human and guinea-pig macrophages and leucocytes. Prostuglandins 17: 729-746, 1979. 32. Nathan, C. F. Secretory products of macrophages. J. Clin. Invest. 79: 319-326, 1987. 33. Ohmori, Y., G. Strassman, and T. A. Hamilton. CAMP differentially regulates expression of mRNA encoding IL-la and ILlp in murine peritoneal macrophages. J. ImmunoZ. 145: 33333339, 1990. 34. Perkins, R. C., R. K. Scheule, and A. Holian. In vitro bioactivity of asbestos for the human alveolar macrophage and its modification by IgG. Am. J. Respir. Cell Mol. Biol. 4: 532-537, 1991. 35. Phan, S. H., and S. L. Kunkel. Inhibition of bleomycin-induced pulmonary fibrosis by nordihydroguiaretic acid. Am. J. Pathol. 124: 343-352, 1986. 36. Piguet, P. F., M. A. Collart, G. E. Grau, Y. Kapanci, and P. Vassalli. Tumor necrosis factor/cachetin plays a key role in bleomycin-induced pneumopathy and fibrosis. J. Exp. Med. 170: 655-663, 1989. 37. Rinaldo, J., R. H. Goldstein, and G. L. Snider. Modification of oxygen toxicity after lung injury by bleomycin in hamsters. Am. Rev. Respir. Dis. 126: 1030-1033, 1982. 38. Sausville, E. A., R. W. Stein, J. Peisach, and S. B. Horwitz. Properties and products of the degradation of DNA by bleomycin and iron(I1). Biochemistry 17: 2746-2754, 1978. 39. Scales, W. E., S. W. Chensue, I. Otterness, and S. L. Kunkel. Regulation of monokine gene expression: prostaglandin E2 suppresses tumor necrosis factor but not interleukin-lcu or ,&mRNA and cell-associated bioactivity. J. Leukocyte Biol. 45: 416-421, 1989. 40. Schmidt, J. A., S. B. Mizel, D. Cohen, and I. Green. Interleukin 1, a potential regulator of fibroblast proliferation. J. Immunol. 128: 2177-2182, 1982. 41. Shirakawa, F., U. Yamashita, M. Chedid, and S. B. Mizel. Cyclic AMP-an intracellular second messenger for interleukin 1. Proc. Natl. Acad. Sci. USA 85: 8201-8205, 1988. 42. Sogal, R. N., A. A. Gottlieb, A. R. Boutros, R. R. Ganapathi, R. R. Tubbs, P. Satariano, P. R. Blakely, and G. J. Beck. Effect of oxygen on bleomycin-induced lung damage. CZevel. Clin. J. Med. 54: 503-509, 1987. 43. Suwabe, A., K. Takahashi, S. Yasui, S. Arai, and F. Sendo. Bleomycin-stimulated hamster alveolar macrophages release interleukin-1. Am. J. PathoZ. 132: 512-520, 1988. 44. Taffet, S. M., K. J. Singhel, J. F. Overholtzer, and S. A. Shurtleff. Regulation of tumor necrosis factor expression in a macrophage-like cell line by lipopolysaccharide and cyclic AMP. Cell Immunol. 120: 291-300, 1989. 45. Tryka, A. F., W. A. Skornik, J. J. Godleski, and J. D. Brain. Potentiation of bleomycin-induced lung injury by exposure to 70% oxygen. Morphological assessment. Am. Reu. Respir. Dis. 126: 1074-1079, 1982. 46. Wahl, S. M. Lymphocyteand macrophage-derived growth factors. Methods Enzymol. 163: 715-731, 1988. 47. Weissmann, G., J. E. Smolen, and H. Korchak. Prostaglandins and inflammation: receptor/cyclase coupling as an explanation of why PGEs and PGI:! inhibit functions of inflammatory cells. Adv. Prostaglandin Thromboxane Res. 8: 1637-1653, 1980. 48. Wewers, M. D., S. I. Rennard, A. J. Hance, P. B. Bitterman, and R. J. Crystal. Normal human alveolar macrophages have limited capacity to release interleukin-1. J. CZin. Invest. 74: 2208-2218, 1984.
Downloaded from www.physiology.org/journal/ajplung by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 12, 2019.
