Human Alveolar Macrophage and Blood Monocyte Interleukin-6 Production Robert M. Kotloff, Jacqueline Little, and Jack A. Elias Pulmonary Division, Department of Internal Medicine, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania

Interleukin-6 (IL-6) modulates a number of processes relevant to host immunity and inflammation. We investigated the capacity of the human alveolar macrophage to elaborate IL-6 in response to lipopolysaccharide (LPS), recombinant interleukin-l (rIL-l), and recombinant tumor necrosis factor (rTNF), and compared macrophage IL-6 production to that of blood monocytes and lung fibroblasts. Unstimulated and TNF-stimulated alveolar macrophages and monocytes produced little or no detectable IL-6. In contrast, macrophages and monocytes produced large amounts of IL-6 in response to LPS and monocytes produced lesser but readily detectable amounts in response to rIL-l. Monocytes and alveolar macrophages differed significantly in their capacity to produce IL-6, with macrophages making more IL-6 in response to LPS and less IL-6 in response to rIL-l than autologous blood monocytes. Monocytes aged in vitro produced little detectable IL-6 in response to LPS or rIL-1, suggesting that differences in cell maturity may account for the diminished capacity of the alveolar macrophage to produce IL-6 in response to IL-l but not its enhanced capacity to produce IL-6 in response to LPS. Mononuclear phagocytes and lung fibroblasts also differed in their ability to produce IL-6. Lung fibroblasts produced more IL-6 in response to rIL-1 and less IL-6 in response to LPS than monocytes and macrophages. In addition, monocytes and macrophages elaborated electrophoretically identical IL-6 moieties that differed from those produced by lung fibroblasts. These differences could be at least partially attributed to differences in sialylation and/or glycosylation. These studies demonstrate that the human alveolar macrophage is a potent producer of IL-6 and that IL-6 production by alveolar macrophages differs in several important ways from that of blood monocytes and lung fibroblasts. Production of IL-6 by alveolar macrophages may be an important mechanism by which local and systemic inflammatory processes are regulated.

Interleukin-6 (IL-6) is a pleiotropic cytokine that serves as an important regulator of inflammation and immunity. It induces the proliferation ofthymocytes and T cells (1,2), promotes the differentiation of cytolytic T cells (3), stimulates the production of IgG, IgM, IgA, and IgE by B cells (4-6), stimulates hepatocyte production of acute-phase proteins (7), and functions as an endogenous pyrogen (8). Support for the concept that IL-6 is a central mediator of inflammation comes from studies demonstrating elevated levels of IL-6 in the serum of normal volunteers after endotoxin administra(Received in original form March 16, 1990 and in revised form May 31, 1990) Address correspondence to: Jack A. Elias, M.D., Pulmonary and Critical Care Medicine, Yale University School of Medicine, 105 LCI, P.O. Box 3333, 333 Cedar Street, New Haven, CT 06510. Dr. Kotloff's present address is: Pulmonary and Critical Care Division, Temple University Hospital, Philadelphia, PA 19140. Abbreviations: bovine serum albumin, BSA; fetal calf serum, FCS; granulocyte/macrophage colony-stimulating factor, GM-CSF; Hanks' balanced salt solution, HBSS; interferon-gamma, IFN-'Y; interleukin-l, IL-l; interleukin-6, IL-6; lipopolysaccharide, LPS; macrophage colony-stimulating factor, M-CSF; protein A-coupled Sepharose 4B, PAS; recombinant, r; tumor necrosis factor, TNF. Am. J. Respir. Cell Mol. BioI. Vol. 3. pp, 497-505, 1990

tion (9), the serum of patients with sepsis (10, 11), the cerebrospinal fluid of patients with meningitis (10, 12), the synovial fluid of patients with septic and rheumatoid arthritis (10, 13, 14), and the skin of patients with psoriasis (15). The alveolar macrophage is a major effector cell in the lung with the capacity to elaborate a wide array of cytokines and eicosanoids (16). Alveolar macrophage-initiated inflammatory processes provide a first line of defense against foreign antigens, playa central role in wound healing and anabolism, and have been implicated in the pathogenesis of inflammatory and fibrotic lung disorders (17-19). Alveolar macrophages are derived largely from blood monocytes (20). However, the common origin of these cells does not result in functional identity as alveolar macrophages and blood monocytes differ in their capacity for antibody-dependent cytotoxicity (21), their ability to support lymphocyte proliferation (22), and their ability to produce E series prostaglandins (23) and leukotrienes (24). Studies from this and other laboratories have demonstrated that monocytes and alveolar macrophages also differ in their ability to produce the inflammatory cytokines, interleukin-1 (lL-1) (25-27), and tumor necrosis factor (TNF) (28, 29). Blood monocytes are well-established producers of IL-6 (30-33). Although IL-6 production by alveolar macrophages has been demonstrated

498

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 3 1990

in the rat (34), the ability of human alveolar macrophages to produce IL-6 has not been characterized. The present studies were undertaken to characterize the ability of the human alveolar macrophage to elaborate IL-6. They demonstrate that the human alveolar macrophage is a potent producer of IL-6, that alveolar macrophages and blood monocytes differ in their capacity to produce IL-6, and that these differences can be only partially attributed to differences in cell maturity. These studies also demonstrate that monocytes and alveolar macrophages produce electrophoretically identical IL-6 moieties, that these moieties differ from those elaborated by cytokine-stimulated human lung fibroblasts, and that differences in sialylation and/or glycosylation can at least partially account for these differences.

Materials and Methods Recombinant Human Cytokines and Anti-cytokine Antibodies Recombinant (r) IL-la (sp act, 4 X 107 U/mg protein) was obtained from Dr. Peter Lomedico (Hoffman-LaRoche, Nutley, NJ), rIL-I/3 (sp act, 6 x 107 U/mg protein) was obtained from Dr. Philip L. Simon and colleagues (Smith Kline and French, Gulph Mills, PA), and rTNF (5 x 107 U/mg) was obtained from Dr. H. Michael Shepard (Genentech, San Francisco, CA). Recombinant granulocyte/macrophage colony-stimulating factor (rGM-CSF; 9.3 x 1{)6 U/mg protein) and recombinant macrophage colony-stimulating factor (rMCSF; 8 x 1{)5 U/mg protein) were obtained from the Genetics Institute (Cambridge, MA). Recombinant interferon gamma (rIFN-'Y) (1.4 X 108 IU/mg) was obtained from Hoffman-LaRoche. rIL-6 and monospecific polyclonal antiserum against rIL-6 were obtained courtesy of Drs. Pravinkumar Sehgal and Lester May (Rockefeller University, New York, NY). All recombinant cytokines were prepared in Escherichia coli. Alveolar Macrophages Lung mononuclear cells were obtained from normal, nonsmoking volunteers by bronchoalveolar lavage employing standard techniques as previously described (35). In brief, fiberoptic bronchoscopy was performed and the right middle lobe was lavaged with six 50-ml aliquots of 0.9% sterile saline solution. The fluid was collected by gentle suction, and the cellular component recovered by centrifugation (600 x g, 10 min). The resulting pellet was washed twice in calcium- and magnesium-free Hanks' balanced salt solution (HBSS) (GIBCO, Grand Island, NY) and resuspended in complete medium (Dulbecco's modified Eagle's medium [DMEM] [GIBCO] with nonessential amino acids [GIBCO] and penicillin and streptomycin [Flow Laboratories, McLean, VA]) supplemented with 10% heat-inactivated fetal calf serum (FCS) (GIBCO). The cells that were recovered were > 90 % viable as assessed by trypan blue dye exclusion and were> 90 % alveolar macrophages as assessed by modified Giemsa staining. These cells were further purified by a l-h adherence to serum-pretreated tissue culture dishes (Falcon; Becton Dickinson, Lincoln Park, NJ). The resulting adherent cell population was comprised of> 98 % alveolar macrophages as assessed by modified Giemsa and nonspecific esterase stains.

Blood Monocytes Autologous blood monocytes were obtained using a modification of previously described techniques (36). Briefly, peripheral blood mononuclear cells were obtained by venipuncture and Ficoll-Hypaque centrifugation (LSM; Organon Teknika Corp., Durham, NC). These cells were washed twice in HBSS, resuspended at a concentration of 8 x 106 cells/ml in complete medium supplemented with 10% FCS, adhered for 1 h, and washed to remove nonadherent cells. The adherent cell population, representing approximately 25 to 40% of the total cells initially plated, was comprised of > 90 % monocytes as judged by latex bead ingestion and nonspecific esterase staining. In some experiments, adherent monocytes were allowed to mature in vitro for periods up to 8 d prior to their use. This was achieved by incubating these cells at 370 C in 5 % CO2 and air in complete medium supplemented with 20 % FCS. Medium was removed and fresh medium added every 3 d. The viability of the adherent cells remained > 95 % throughout the incubation period. Human Lung Fibroblasts Strain CCL-201, CCL-202, and CCL-210 normal human adult lung fibroblasts were obtained from the American Type Culture Collection (Rockville, MD). They were grown to confluence in complete medium supplemented with 10% FCS prior to use in experiments. Preparation of Supernatants Adherent blood monocytes and alveolar macrophages were used at a concentration of 2 x 1()6 cells/ml and lung fibroblasts were used as a confluent monolayer. They were incubated in serum-free complete medium with or without the addition of varying concentrations of lipopolysaccharide (LPS) (phenol-extracted from Escherichia coli strain 0111: B4; Sigma Chemical Co., S1. Louis, MO) or cytokine in 5 % CO2 and air at 370 C for up to 72 h. Under these culture conditions, cell viability was consistently > 95 %. Supernatants were then aspirated, cleared of cellular debris by centrifugation (400 x g), and stored at -70 0 C. Measurement of IL-6 in Cell Supernatants Previous studies have demonstrated a good correlation between IL-61evels measured by radioimmunoassay (RIA) and IL-6 bioactivity as assessed in a B cell hybridoma proliferation assay (13, 37). Thus, we employed a solid-phase RIA, representing a modification of the technique described by Haverty and associates (38), to measure the IL-6 in cellular supernatants. Duplicate 50-1-'1 aliquots of test samples or controls were added to individual wells of a flexible polyvinyl chloride 96-well microtiter plate (Dynatech Laboratories, Alexandria, VA). After an overnight incubation at 4 0 C, the wells were emptied and unbound surfaces blocked with 200 I-'l/well of 4 % bovine serum albumin (BSA) (Sigma) in phosphate buffer (Na2P04 , pH 8) for 1 h at room temperature. The wells were again emptied, and 50 I-'l/well of a 1:200 dilution (in phosphate buffer) of the primary antiserum (rabbit anti-rIL-6) was added. The specificity of this antibody is well documented and its use well reported (1, 33). After a 2-h incubation at room temperature, the plate was washed 3

Kotloff, Little, and Elias: Alveolar Macrophage/Blood Monocyte IL-6 Production

120

~

BM

----.-

AM

5 min. The resulting supernatants were analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDSPAGE) using a 12% polyacrylamide/2.7% bisacrylamide gel with a 5% polyacrylamide/2.7% bisacrylamide stacking gel. After fixing, the gel was incubated in Fluorohance (Research Products, Mount Pleasant, IL), dried, and analyzed by autoradiography.

100

lil Qj 0

80

CD

0

til

.s CD

...J

499

60

Deglycosylation and Desialylation 40

20

24

48

72

INCUBATION TIME (h)

Figure 1. Kinetics of LPS-induced IL-6 production. Blood monocytes (BM) and alveolar macrophages (AM) were incubated at 37° C in the presence of LPS (25 p,g/ml) for the noted periods of time, and their supernatants assayed for IL-6.

times with 0.1% BSA and [I2SI]anti-rabbit IgG antibody (sp act, 20 JLCi/mg; Amersham, Arlington Heights, IL) was added (lOS cpm/well in 50-JLI aliquots). After a final2-h incubation at room temperature, the plate was washed thoroughly with tap water and the bound radioactivity was counted. The IL-6 concentrations in the test samples were determined by comparison to a standard curve prepared with rIL-6 that was run for each assay. Using this technique, as little as 0.5 ng/ml of IL-6 could be reliably detected and no cross-reactivity occurred with rlL-la, rlL-IJ3, rGM-CSF, rM-CSF, tTNF, and rIFN-'Y.

To assess IL-6 N-glycosylation, supernatants were prepared from monocytes, macrophages, and fibroblasts stimulated in the presence and absence of 5 JLg/ml tunicamycin (Sigma). The IL-6 in the resulting supernatants was immunoprecipitated and analyzed by SOS-PAGE. Enzymatic removal of sialic acid and O-glycosylated moieties was performed using techniques previously described by Gross and coworkers (40). Briefly, (3sS]methionine-labeled supernatants from LPS-stimulated monocytes and rlL-l-stimulated fibroblasts were dialyzed at 4 0 C for 24 h against 50 mM sodium acetate buffer, pH 5.5, and 1 mM CaCI 2, and subsequently incubated with neuraminidase (Genzyme Corp., Boston, MA) at a concentration of 500 mU/ml for 16 h at 37 0 C. The neuraminidase-treated media was then further dialyzed for 24 h at 4 0 C against 20 mM sodium phosphate buffer, pH 6.5, and subsequently incubated for 16 h at 37 0 C with 25 mU/ml of endo-a-N-acetylgalactosaminidase (O-glycanase; Genzyme Corp.). At the end of this incubation period, IL-6 was immunoprecipitated and analyzed by SOS-PAGE. Statistics All results are expressed as mean ± SEM. Statistical comparisons were done using Student's t test with a probability value of ~ 0.05 considered to be significant.

Results Measurement of TNF in Supernatants Quantitative determination of supernatant TNF levels was achieved with use of a sandwich enzyme immunoassay (Biokine; T Cell Sciences, Cambridge, MA). Immunoprecipitation Monocytes, macrophages, and fibroblasts were incubated for 24 h in methionine-free DMEM (GIBCO) supplemented with 100 JLCi/ml of (3sS]methionine (sp act, > 800 Ci/mmol; Amersham) in the presence or absence of LPS (25 JLg/ml) or rIL-la (2.5 to 25 ng/ml). At the end of the incubation, the supernatants were collected and protease inhibitors were added to a final concentration of 3 mM phenylmethanesulfonyl fluoride, 5 mM N-ethylmaleimide, 5 mM EDTA, and 2 mM p-aminobenzamidine hydrochloride. The labeled IL-6 moieties were then immunoprecipitated using techniques previously described (39). Samples for immunoprecipitation were precleared with FCS and a 20 % solution of protein A-coupled Sepharose 4B (PAS) (Pharmacia; Piscataway, NJ). A 1:200 dilution of polyclonal antiserum against rlL-6 was added and the samples were agitated overnight at 4 0 C. PAS was again added and the solution was gently rocked for 1 h at 4 0 C and was subsequently subjected to centrifugation (10,000 X g,S min). The pellet was extensively washed, resuspended in 1X Laemmli buffer, and boiled at 100 0 C for

IL-6 Production by LPS-stimulated Monocytes and Alveolar Macrophages Monocytes and alveolar macrophages incubated in complete medium alone for up to 72 h did not produce detectable IL-6. In contrast, LPS stimulated alveolar macrophage and monocyte IL-6 production in a dose-dependent fashion (data not shown). The dose response for monocytes and macrophages was similar, with detectable levels of IL-6 being seen with concentrations of LPS as low as 100 pg/ml and maximum IL-6 production occurring with doses of LPS ~ 1 JLg/ml. Virtually identical results were obtained with LPS doses of 1, 10, and 25 JLg/ml. The kinetics ofIL-6 production in these two cell populations was similar, with detectable levels of IL-6 seen after as little as 4 hand near-maximal levels seen after 24 to 48 h of LPS stimulation (Figure 1). Importantly, however, maximally stimulated alveolar macrophages produced more IL-6 than maximally stimulated autologous blood monocytes (87.0 ± 11.9 ng/tO' cells/24 h versus 21.4 ± 2.9 ng/lQ6 cells/24 h; P < 0.01) (Figure 1). Qualitatively similar results were obtained after 12, 24, 48, and 72 h of incubation, demonstrating that this differential response was not attributable to differences in the kinetics of IL-6 production by these two cell types (Figure 1). In addition, supernatants from LPS-stimulated monocytes and macrophages did not differ in their ability to degrade exogenous IL-6 (data not

500

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 3 1990

20

15

lil

""

G> o

__ ------------1I

I?"/"

U>

0

"

f------

10 I

Cl

I

s

I I

I I

CD

I

~

I I I I

-- -------------------0+-..........--,...-----.------.-------.-o

12

24

36

48

INCUBATION TIME (h)

Figure 2. Kinetics of rIL-l-induced IL-6 production. Blood monocytes (open circles) and alveolar macrophages (closed circles) were incubated at 37° C in the presence of 25 ng/ml ofrIL-la (solid line) or 10 ng/ml of rIL-ll3 (dashed line) for the noted periods of time, and the IL-6 in the resulting supernatants was assayed.

shown), suggesting that the differences in monocyte and macrophage IL-6 levels were largely attributable to differential production rather than differential degradation of IL-6. IL-6 Production by rIL-l-stimulated Blood Monocytes and Alveolar Macrophages Blood monocytes and alveolar macrophages were cultured in the presence of varying concentrations of rIL-la or rIL-II3, and the levels of IL-6 in their supernatants quantitated. Both rIL-l moieties stimulated monocyte and alveolar macrophage IL-6 production in a dose-dependent fashion (data not shown). The highest levels of monocyte- and macrophagederived IL-6 were seen with 25 ng/ml rIL-Ia and 10 ng/ml rIL-113 (the highest doses tested). The stimulatory effects of rIL-l fell off rapidly at lower concentrations, with minimal IL-6 produced by either cell type in response to doses of rILIo ~ 2.5 ng/ml or doses of rIL-113 ~ 1 ng/ml. In all cases, the stimulatory effects of rIL-la and rIL-113 were near-maximal by 24 h of incubation (Figure 2). Importantly, monocytes produced more IL-6 than alveolar macrophages when maximally stimulated with either rIL-l moiety (Figure 2). In the presence of 25 ng/ml rIL-la, monocytes produced 11.5 ± 1.4 ng IL-6/1Q6 cells/24 h while alveolar macrophages produced only 2.8 ± 0.7 ng IL-6/lQ6 cells/24 h (P < 0.05). In response to 10 ng/ml rIL-II3, monocytes elaborated 13.9 ± 1.4 ng!lQ6 cells/24 h compared with only 1.9 ± 0.8 ng! lQ6 cells/24 h elaborated by alveolar macrophages (P < 0.05). This differential response of blood monocytes and alveolar macrophages could not be attributed to differences in the kinetics of IL-6 production inasmuch as similar results were seen at 12-, 24-, and 48-h periods of incubation (Figure 2). Because blood monocytes and alveolar macrophages elaborate IL-6 in response to doses of LPS as low as 100 pg/ml, and recombinant cytokines contain variable levels of endotoxin contamination, studies were undertaken to determine the degree to which rIL-l-induced IL-6 production was due to contaminating endotoxin. Blood monocytes and alveolar

macrophages were cultured for 24 h in the presence of rIL-l with or without the addition of polymyxin B, an agent capable of neutralizing the effects of endotoxin (41). At 10 j.tg/ml, polymyxin completely abolished the ability of LPS to stimulate monocyte and macrophage IL-6 production (data not shown). In the presence of polymyxin, rIL-la-induced IL-6 production by monocytes diminished by approximately 30 % while that of macrophages was totally eliminated (Figure 3). Similar results were obtained when monocytes and macrophages were incubated with rIL-113 in the presence and absence of polymyxin (data not shown). These effects were not due to polymyxin-induced alterations in cell viability or to interference by polymyxin with the ability of the RIA to detect IL-6 (data not shown). IL-l is heat-labile and can be neutralized by boiling whereas endotoxin is heat-stable (41). Heating rIL-la for 20 min at 95° C caused a 70% reduction in its ability to stimulate monocyte IL-6 production while having no effect on its ability to stimulate macrophage IL-6 production (Figure 3). Similar results were obtained when rIL-113 was heat-inactivated. When viewed in combination, these studies demonstrate that the stimulatory effects of rIL-l on monocyte IL-6 production are largely cytokine-mediated whereas the stimulatory effects on macrophage IL-6 production are largely if not exclusively attributable to low-level endotoxin contamination. Additionally, they demonstrate that the differential response of monocytes and macrophages to rIL-l persists even when the effects of contaminating endotoxin are negated. IL-6 Production by rTNF-stimulated Blood Monocytes and Alveolar Macrophages TNF and IL-l have overlapping spectra of bioactivity (42).

12

~ IL-1

•

10

ri

'0

a;

IL-1 + PMX IL-1 + Heating

8

u

CD

....0 a

..:.. lD

~ 2

8M

AM

Figure 3. Effect of polymyxin B (PMX) and heat treatment on rIL-la-stimulated IL-6 production by blood monocytes (BM) and alveolar macrophages (AM). Cells were incubated for 24 h at 37° C with rIL-la (25 ng/ml) with or without PMX (10 JLg/ml), and the supernatants were assayed for IL-6. For heat treatment, rIL-la was heated at 95° C for 20 min prior to incubation. Cells incubated with PMX alone did not produce detectable quantities ofIL-6. N.D. = none detected.

Kotloff, Little, and Elias: Alveolar Macrophage/Blood Monocyte IL-6 Production

TABLE 1

Effect of in vitro maturation on monocyte production of IL-6 and TNF* IL-6 t (ng/l()6 cells) Days in Culture Exp. 1 0 4 7 Exp.2 0 4

TNFJ: (ng/l()6 cells)

IL-l

LPS

LPS

36.4 4.0

ND§

91.5 4.4 6.9

56.6 62.3 95.0

9.1

46.2

ND

ND

33.3 227.6

* Monocytes were incubated for 24 h with tIL-l a (25 ng/ml) or LPS (25 p,g/ ml). They were used immediately after harvesting (day 0) or after incubation for 4 to 7 d in complete medium supplemented with 20 % FCS. t IL-6 in supernatants as determined by RIA. Supernatants from unstimulated cells did not contain detectable IL-6. :j: TNF in supernatants as determined by enzyme-linked immunosorbent assay. § ND = none detectable. Thus, we investigated whether TNF stimulated monocyte and alveolar macrophage IL-6 production. Blood monocytes and alveolar macrophages incubated in the presence of rTNF at concentrations of 1 to 200 ng/ml for periods of 4 to 72 h elaborated little IL-6 (data not shown). rTNF also did not alter the ability of rIL-1 to stimulate IL-6 production by either cell type (data not shown). Effect of In Vitro Maturation on Monocyte IL-6 Production The majority of alveolar macrophages are derived from blood monocytes (20). Because maturation is known to regulate monocyte cytokine production (25,26,29), studies were undertaken to determine whether the differences in IL-6 production that were noted could be attributed to differences in cell maturity. This was done by comparing the IL-6 produced by freshly isolated blood monocytes and monocytes that had been incubated in vitro for 4 to 7 d. Freshly harvested blood monocytes elaborated large amounts of IL-6 in response to LPS and lesser but readily detectable quantities in response to rIL-1. Within 96 h of in vitro incubation, these cells had acquired a macrophage-like appearance with enhanced cytoplasm and eccentric nuclei. These monocytederived macrophages produced little detectable IL-6 in response to either stimulus (Table 1). Monocytes cultured for 7 d before stimulation were similarly unable to produce soluble IL-6 in response to LPS and rIL-1. In contrast, these cells elaborated increased amounts of TNF in response to LPS (Table 1), demonstrating that they did not have a generalized impairment in protein synthesis or loss of LPS responsiveness. These results suggest that differences in cell maturity may account, at least in part, for the different amounts of IL-6 produced by rIL-1-stimulated monocytes and macrophages. In contrast, the enhanced ability of the alveolar macrophage to elaborate IL-6 in response to LPS cannot be accounted for in a similar manner. Comparison of Mononuclear Phagocyte and Lung Fibroblast IL-6 Previous work from this and other laboratories has demon-

501

strated that fibroblasts are potent producers of IL-6 and that the IL-6 that they produce is polymorphic (1, 7, 33). Thus, to further understand monocyte/macrophage IL-6 production, we compared alveolar macrophage, blood monocyte, and lung fibroblast IL-6 production quantitatively and qualitatively. Blood monocytes and alveolar macrophages produced significantly more IL-6 in response to LPS than did lung fibroblasts (Table 2). Overall, maximally stimulated monocytes and macrophages produced 21.4 ± 2.9 ng IL-6/1Q6 cells/24 hand 87.0 ± 11.9 ng IL-6/l06 cells/24 h, respectively, while fibroblasts elaborated only 1.3 ± 0.6 ng IL-6/1Q6 cells/24 h when incubated with an identical dose of LPS (P < 0.05 comparing monocytes and fibroblasts, and P < 0.001 comparing macrophages and fibroblasts). In contrast, lung fibroblasts produced more IL-6 than monocytes or macrophages when maximally stimulated with rIL-1 (Table 2). Overall, fibroblasts incubated with 2.5 ng/ml rIL-1cx produced 140.1 ± 2.7 ng IL-6/106 cells/24 h while monocytes and alveolar macrophages incubated with 25 ng/ml rIL-cx produced 11.5 ± 1.4 ng IL-6/1Q6 cells/24 h and 2.8 ± 0.7 ng IL-6/106 cells/24 h, respectively (P < 0.001 comparing fibroblasts and monocytes and fibroblasts and alveolar macrophages). Lung fibroblasts were also more sensitive to the effects of rll.-I« than monocytes and macrophages since doses of rIL-1cx as low as 0.025 ng/ml stimulated fibroblast IL-6 production while doses ~ 2.5 ng/ml were required to stimulate monocyte and macrophage IL-6 production (Table 2). In order to structurally characterize the IL-6 moieties produced by mononuclear phagocytes and human lung fibroblasts, the IL-6 in their supernatants was analyzed by immunoprecipitation and SDS-PAGE. Monocytes and macrophages stimulated with either rIL-1 or LPS produced identical IL-6 moieties (Figure 4). The predominant form appeared at approximately 24.6 kD and a second less intense band was noted at approximately 29.1 kD. In some preparations, a third band was noted at approximately 22.3 kD. Interestingly, the IL-6 species elaborated by monocytes and macrophages migrated differently than the IL-6 moieties produced by CCL-202 lung fibroblasts (Figure 4). The predominant moiety produced by these fibroblasts was slightly smaller, appearing at approximately 24.0 kD, and the minor band was slightly larger, appearing at approximately 29.8 kD. Inter-

TABLE 2

Comparison of IL-6 production by blood monocytes (BM), alveolar macrophages (AM), and lung fibroblasts (F) IL-6 (ng/1()6 cells)* Stimulus"

IL-la

LPS

0.025 ng/ml 0.25 ng/ml 2.5 ng/ml 25 ng/ml 25 jLg/ml

BM

AM

ND:j: ND 1.4 ± 0.2 11.5 ± 1.4 21.4 ± 2.9

ND ND 1.3 ± 0.2 2.8 ± 0.7 87.0 ± 11.9

F 3.0 48.2 140.1 86.3 1.3

± ± ± ± ±

0.1 1.6 2.7 1.5 0.6

* IL-6 in supernatants as determined by RIA (mean ± SEM). Supernatants from unstimulated BM, AM, and F did not contain detectable IL-6. t BM, AM, and F were incubated for 24 h at 37° C in the presence of the noted concentrations of rIL-la or LPS. :j: ND = none detectable.

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 3 1990

502

Stimulant

kD

-41-30-25-

-14Cell

8M

AM

8M 8M

F

B

A

Figure 4. Immunoprecipitation and SDS-PAGE of metabolically labeled IL-6 elaborated by blood monocytes (BM), alveolar macrophages (AM), and human lung fibroblasts (F). Cells were incubated in the presence of [35S]methionine (100 p.Ci/ml) and either LPS (25 p.g/ml) or rIL-la (25 ng/ml) for 24 h at 37° C, and the supernatants immunoprecipitated with polyclonal antiserum against IL-6. (A) Comparison of BM and AM IL-6. (B) Comparison of BM and F IL-6.

mittently, we detected a third fibroblast-elaborated IL-6 moiety. This moiety was smaller than its corresponding monocyte/macrophage product, appearing at approximately 21.8 kD. Similar size differences were noted when CCL-201 and CCL-210 human lung fibroblasts were compared to monocytes and macrophages (data not shown).

1 Cell type

8M

2 F

3 8M

4 F

Studies were undertaken to determine if the electrophoretic differences between mononuclear phagocyte and lung fibroblast IL-6 were due to differential glycosylation and/or sialylation. In these studies, monocytes were stimulated with LPS while fibroblasts were stimulated with rfl.-L In order to determine the contribution of N-glycosylation, we compared blood monocyte- and fibroblast-derived IL-6 produced in the presence and absence of tunicamycin, an agent that inhibits the addition of N-linked oligosaccharides. This resulted in the disappearance of the high-molecular-weight moieties but did not shift the position of the predominant low-molecular-weight forms ofIL-6 produced by these cells (Figure 5, left panel). The contributions of O-glycosylation and sialylation were investigated by serially treating cell supernatants with neuraminidase, to remove sialic acid residues, and O-glycanase (endo-c-V-acetylgalactosaminidase), an enzyme that removes O-glycosylated residues. These enzymes were used serially because the activity of O-glycanase requires prior removal of sialic acid residues. This treatment increased the mobility of, and diminished but did not entirely negate the differences between, the low-molecular-weight IL-6 moieties produced by these cells (Figure 5, right panel). Definitive statements concerning the effect of these enzymes on the higher-molecular-weight IL-6 moieties could not be made due to the loss of intensity and definition of these bands during enzymatic digestion. When viewed in combination, these studies demonstrate that the predominant low-molecular weight IL-6 moieties produced by mononuclear phagocytes and fibroblasts are not N-glycosylated and that their molecular weight differences can be at least partially accounted for by differential sialylation and/or O-glycosylation.

Discussion In an effort to better understand the processes regulating immune and inflammatory events within the lung, we characterized the ability of the human alveolar macrophage to elaborate IL-6 in response to a number of biologically rele-

5 F

6 8M

7 F

8 8M

+ +

+ +

kD

-30-25-

-14 Tunicamycin Neuraminidase O-glycanase

+

+

Figure 5. Deglycosylation of blood monocyte (BM) and lung fibroblast (F) IL-6. [35S]methioninelabeled supernatants of LPS-stimulated BM and rIL-l-stimulated F were used in these experiments. (Left panel) To test for the presence of N-linked carbohydrates, IL-6 was immunoprecipitated from supernatants of BM and F stimulated in the absence (lanes 1 and 2) or presence (lanes 3 and 4) oftunicamycin (5p.g/ml). (Right panel) To test for the presence of sialic acid and O-linked carbohydrate moieties, IL-6 was immunoprecipitated from untreated BM and F supernatants (lanes 5 and 6) and from supernatants treated sequentially with neuraminidase and O-glycanase (lanes

7 and 8).

Kotloff, Little, and Elias: Alveolar Macrophage/Blood Monocyte IL-6 Production

vant stimuli and compared macrophage IL-6 production to that of blood monocytes and lung fibroblasts. These studies indicate that the human alveolar macrophage is an important source ofIL-6 in the lung. They also demonstrate that human alveolar macrophages produce more IL-6 in response to LPS and less IL-6 in response to rIL-l than do autologous blood monocytes and that these differences cannot be entirely accounted for by differences in the maturational state of these cells. Furthermore, these studies demonstrate that mononuclear phagocytes and lung fibroblasts produce IL-6 in response to different stimuli, with fibroblasts responding predominantly to IL-l and monocytes and macrophages responding predominantly to LPS. Lastly, they demonstrate that alveolar macrophages and blood monocytes elaborate electrophoretically identical IL-6 moieties that differ from those elaborated by lung fibroblasts. Despite their common origin, alveolar macrophages and blood monocytes manifest a number of important functional differences. They differ in their ability to support antigenstimulated lymphocyte proliferation (22), in their capacity for antibody-dependent cell cytotoxicity (21), and in their ability to produce prostaglandins (23), leukotrienes (24), soluble IL-113 (25, 26), and TNF (28, 29). In this study, we compared the ability of blood monocytes and alveolar macrophages to produce IL-6. Like others, we noted that LPS and rIL-l stimulate monocyte IL-6 production (31-33). Importantly, our studies expand on these previous studies by demonstrating for the first time that the human alveolar macrophage can produce IL-6 and that monocytes and alveolar macrophages differ in their capacity to produce this cytokine. Specifically, we have shown that alveolar macrophages make more IL-6 in response to LPS and less IL-6 in response to IL-l than do blood monocytes. Studies from this (25) and other laboratories (26, 29) have demonstrated that the differing capacity ofmonocytes and alveolar macrophages to produce IL-l and TNF are due, at least in part, to differences in cell maturity. To determine if cell maturity is an important determinant of monocyte and alveolar macrophage IL-6 production, we compared the IL-6 production of freshly harvested monocytes to that of monocytes matured in vitro. With the incubation conditions that were employed in these studies, monocytes acquired both physical and functional characteristics of tissue macrophages, since they enlarged, acquired eccentric nuclei, and produced increased amounts of TNF. In accord with previous studies by Bauer and coworkers (31), we noted that in vitro maturation decreased the capacity of monocytes to produce IL-6 in response to rIL-l, thus reproducing the difference observed between monocytes and alveolar macrophages. This suggests that cell maturity may be an important regulator of ILl-induced IL-6 production by monocytes and macrophages. However, in vitro maturation also decreased the ability of monocytes to produce IL-6 in response to LPS. This suggests that the heightened capacity of the alveolar macrophage to produce IL-6 in response to LPS is not solely a function of cell maturity, and that other aspects of the in vivo environment of the lung may be involved in the acquisition of this phenotype. We have previously demonstrated that alveolar macrophages produce less soluble IL-113 than do autologous blood monocytes and that among alveolar macrophages, the smaller,

503

denser, more monocyte-like cells produce more soluble IL-ll3 than the less dense, more tissue macrophage-like cells (25). In contrast, these studies show that LPS-stimulated alveolar macrophages produce more IL-6 than do autologous blood monocytes. Similarly, LPS-stimulated alveolar macrophages elaborate more TNF than do monocytes (28, 29). These findings suggest that resident alveolar macrophages in the normal lung have differentiated to a state that allows them to produce large amounts of IL-6 and TNF and only limited quantities of soluble IL-ll3. It is tempting to speculate that this pattern of cytokine production may allow for differing levels of pulmonary inflammation. Studies from this and other laboratories have demonstrated that IL-l interacts in a synergistic fashion with IL-6 (1, 5) and TNF (37, 42) in regulating a variety of inflammatory events. In the normal lung, insults that stimulate resident alveolar macrophages would result in inflammatory responses that reflect the ability of these cells to produce large amounts of IL-6 and TNF and only limited quantities of IL-l. In contrast, insults that cause an influx of monocytes into the lung, as occurs in sarcoidosis (43) and experimental models of pulmonary inflammation (44), would lead to enhanced local production of IL-l. The synergistic interaction of IL-l with IL-6 and TNF would serve to further amplify local inflammatory events. IL-6 is produced by a wide array of cells in response to a wide array of stimuli. However, it is becoming increasingly apparent that cells differ in their capacity to respond to specific stimuli. Our studies demonstrate that LPS is a potent inducer of IL-6 production by alveolar macrophages and blood monocytes and a weak inducer of lung fibroblast IL-6 production. In contrast, IL-l is a potent stimulator of fibroblast IL-6 production while only weakly stimulating mononuclear phagocytes. These findings suggest that macrophages and fibroblasts are serving different but equally important roles in the pulmonary inflammatory response. The alveolar macrophage appears to be best suited to respond in a primary fashion to invading organisms and other noxious insults by producing large amounts of IL-6 and TNF and more limited quantities of IL-l. These cytokines individually and in combination serve to initiate local and systemic inflammatory processes, including T cell proliferation, immunoglobulin production by B cells, fever, and acute-phase protein production. The IL-l and TNF produced by alveolar macrophages and recruited blood monocytes can then secondarily stimulate fibroblasts to elaborate IL-6. This response would serve to nonspecifically amplify local and systemic inflammatory events. This fibroblast-based amplification may be an important determinant of the severity and chronicity of pulmonary inflammation. Studies from a number of laboratories have demonstrated IL-6 moieties in the range of 20 to 30 kD (31, 33, 40), 43 to 45 kD (9, 14), and 60 to 70 kD (10). The biologic significance of this observed heterogeneity remains uncertain. Our studies demonstrate that blood monocytes and human alveolar macrophages produce identically sized IL-6 isoforms, with a predominant moiety appearing at approximately 24.6 kD and a minor form appearing at approximately 29.1 kD. We also demonstrated that the IL-6 moieties produced by monocytes and macrophages migrate differently when subjected to SDS-PAGE than those produced by human lung fibroblasts. These observations superficially

504

AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 3 1990

differ from those of May and associates, who reported that human blood monocytes and human skin fibroblasts produce identical IL-6 isoforms (33). Interestingly, we have also found that human skin fibroblasts (strain CRL-1224; American Type Culture Collection) and monocytes produce electrophoretically identical IL-6 moieties (R. M. Kotloff and 1. A. Elias, unpublished observation). This suggests that strainspecific and/or organ-specific differences in fibroblast IL-6 may exist. Previous studied by May and associates (33, 45) and Gross and colleagues (40) have demonstrated that fibroblast- and monocyte-derived IL-6 is extensively glycosylated and sialylated. Our studies extend these observations by suggesting that differential glycosylation and/or sialylation is at least partially responsible for the electrophoretic differences between the predominant low-molecular-weight forms of IL-6 produced by mononuclear phagocytes and fibroblasts. The role that other post-translational modifications such as differential phosphorylation (46) play in producing these differences will require additional investigation. In summary, the present study demonstrates that human alveolar macrophages are potent producers of IL-6 and that the production of this cytokine by alveolar macrophages differs in several important ways from that of blood monocytes and human lung fibroblasts. Although the in vivo significance of alveolar macrophage IL-6 production is uncertain, it is tempting to speculate that elaboration of IL-6 by alveolar macrophages may playa role in the pathogenesis of acute and chronic inflammatory lung disorders. Production of IL-6 in response to bacterial products such as LPS may contribute to the inflammatory and acute-phase responses that characterize bacterial pneumonias. Additionally, abnormal or excessive IL-6 production by activated alveolar macrophages may contribute to the T cell proliferation, polyclonal B cell activation, fever, and elevation of circulating acute-phase proteins found in patients with diseases such as sarcoidosis (47). Acknowledgments: The writers thank Margaret Reynolds and Vicky Lentz for their expert technical assistance; Ronald Collman, M.D., for his assistance with the TNF enzyme-linked immunosorbent assay. Milton Rossman, M.D., for his donation of bronchoalveolar lavage samples; Mary McDevitt, for excellent secretarial support; and the scientists who generously contributed cytokines and antibodies. This work was supported by the following grants from the National Institutes of Health: HL-36708, HL-41216 (Dr. Elias), and Research Training Grant HL-07000 (Dr. Kotloff), Dr. Elias is an RJR Nabisco Scholar and a Career Investigator of the American Lung Association and Philadelphia-Montgomery County Lung Association. This work was presented in part at the Annual Meeting of the American Thoracic Society, Cincinnati, Ohio, 1989.

References 1. Elias, J. A., G. Trinchieri, J. M. Becket al. 1989. A synergistic interaction of IL-6 and IL-l mediates the thymocyte-stimulating activity produced by recombinant IL-l-stimulated fibroblasts. J. Immunol. 142:509-514. 2. Lotz, M., F. Jirik, P. Kabouridis et al. 1988. B cell stimulating factor 2/interleukin 6 is a costimulant for human thymocytes and T lymphocytes. J. Exp. Med. 167:1253-1258. 3. Okada, M., M. Kitahara, S. Kishimoto, T. Matsuda, T. Hirano, and T. Kishimoto. 1988. IL-6/BSF-2 functions as a killer helper factor in the in vitro induction of cytotoxic T cells. J. Immunol. 141:1543-1549. 4. Muraguchi, A., T. Hirano, B. Tang et al. 1988. The essential role of B cell stimulatory factor 2 (BSF-2/IL-6) for the terminal differentiation of B cells. J. Exp. Med. 167:332-344. 5. Kunimoto D. Y., R. P. Nordan, and W. Strober. 1989. IL-6 is a potent cofactor of IL-l in IgM synthesis and of IL-5 in IgA synthesis. J. Immunol. 143:2230-2235. 6. Vercelli, D., H. H. Jabara, K. Arai, T. Yokota, and R. S. Geha. 1989. Endogenous interleukin 6 plays an obligatory role in interleukin 4-dependent human IgE synthesis. Eur. J. Immunol. 19:1419-1424.

7. Gauldie, J., C. Richards, D. Harnish, P. Lansdorp, and H. Baumann. 1987. Interferon /32/B-cell stimulatory factor type 2 shares identity with monocyte-derived hepatocyte-stimulating factor and regulates the major acute phase protein response in liver cells. Proc. Natl. Acad. Sci. USA 84:7251-7255 8. Helle, M., J. P. J. Brakenhoff, E. R. DeGroot, and L. A. Aarden. 1988. Interleukin 6 is involved in interleukin I-induced activities. Eur. J. Immunol. 18:957-959. 9. Fong, Y., L. L. Moldawer, M. Marano et al. 1989. Endotoxemia elicits increased circulating /3TIFN/IL-6 in man. J. Immunol. 142:2321-2324. 10. Helfgott, D. C., S. B. Tatter, U. Santhanam et al. 1989. Multiple forms of IFN-/32/IL-6 in serum and body fluids during acute bacterial infection. J. Immunol. 142:948-953. 11. Waage, A., P. Brandtzaeg, A. Halstensen, P. Kierulf, and T. Espevik. 1989. The complex pattern of cytokines in serum from patients with meningococcal septic shock. J. Exp. Med. 169:333-338. 12. Waage, A., A. Halstensen, R. Shalaby, P. Brandtzaeg, P. Kierulf, and T. Espevik. 1989. Local production of tumor necrosis factor a, interleukin 1, and interleukin 6 in meningococcal meningitis: relation to the inflammatory response. J. Exp. Med. 170:1859-1867. 13. Hirano, T., T. Matsuda, M. Turner et al. 1988. Excessive production of interleukin 6/B cell stimulatory factor-2 in rheumatoid arthritis. Eur. J. Immunol. 18:1797-1801. 14. Bhardwaj, N., U. Santhanam, L. L. Lau et al. 1989. IL-6/IFN-/32 in synovial effusions of patients with rheumatoid arthritis and other arthritides: identification of several isoforms and studies of cellular sources. J. Immunol. 143:2153-2159. 15. Grossman, R. M., J. Krueger, D. Yourish et al. 1988. Interleukin 6 is expressed in high levels in psoriatic skin and stimulates proliferation of cultured human keratinocytes. Proc. Natl. Acad. Sci. USA 86:6367-6371. 16. Fels, A. O. S., and Z. A. Cohn. 1986. The alveolar macrophage. J. Appl. Physiol. 60:353-369. 17. Hunninghake, G. W. 1984. Release of interleukin-l by alveolar macrophages of patients with active pulmonary sarcoidosis. Am. Rev. Respir. Dis. 129:569-572. 18. Bitterman, P. B., S. Adelberg, and R. G. Crystal. 1983. Spontaneous release of the alveolar macrophage-derived growth factor in the interstitial lung disorders. J. Clin. Invest. 72: 1801-1813. 19. Crystal, R. G., P. B. Herman, S. I. Rennard, A. J. Hance, and B. A. Keogh. 1984. Interstitial lung diseases of unknown cause: disorders characterized by chronic inflammation of the lower respiratory tract. N. Engl. J. Med. 310:154-166,235-244. 20. Thomas, E. D., R. E. Ramberg, G. E. Sale, R. S. Sparkes, and D. W. Golde. 1976. Direct evidence for a bone marrow origin of the alveolar macrophage in man. Science 192:1016-1018. 21. LeMarbre, P., J. Hoidel, R. Vasella, and J. Rinehart. 1980. Human pulmonary macrophage tumor cell cytotoxicity. Blood 55:612-617. 22. Ferro, T. J., J. A. Kern, J. A. Elias, M. Kamoun, R. P. Daniele, and M. D. Rossman. 1987. Alveolar macrophages, blood monocytes, and density-fractionated alveolar macrophages differ in their ability to promote lymphocyte proliferation to mitogen and antigen. Am. Rev. Respir. Dis. 135:682-687. 23. Elias, J. A., T. J. Ferro, M. D. Rossman et al. 1987. Differential prostaglandin production by unfractionated and density-fractionated human monocytes and alveolar macrophages. J. Leukocyte Bioi. 42:114-121. 24. Balter, M. S., G. B. Toews, and M. Peters-Golden. 1989. Different patterns of arachidonate metabolism in autologous human blood monocytes and alveolar macrophages. J. Immunol. 142:602-608. 25. Elias, J. A., A. D. Schreiber, K. Gustilo et al. 1985. Differential interleukin 1 elaboration by unfractionated and density fractionated human alveolar macrophages and blood monocytes: relationship to cell maturity. J. Immunol. 135:3198-3204. 26. Wewers, M. D., S. I. Rennard, A. J. Hance, P. B. Bitterman, and R. G. Crystal. 1984. Normal human alveolar macrophages obtained by bronchoalveolar lavage have a limited capacity to release interleukin-l. J. Clin. Invest. 74:2208-2218. 27. Bernaudin, J. F., K. Yamauchi, M. D. Wewers, M. J. Tocci, V. J. Ferrans, and R. G. Crystal. 1988. Demonstration by in situ hybridization of dissimilar IL-l/3 gene expression in human alveolar macrophages and blood monocytes in response to lipopolysaccharide. J. Immunol. 140: 3822-3829. 28. Martinet, Y., K. Yamauchi, and R. G. Crystal. 1988. Differential expression of the tumor necrosis factor/cachectin gene by blood and lung mononuclear phagocytes. Am. Rev. Respir. Dis. 138:659-665. 29. Rich, E. A., J. R. Panuska, R. S. Wallis, C. B. Wolf, M. L. Leonard, and J. J. Ellner. 1989. Dyscoordinate expression of tumor necrosis factoralpha by human blood monocytes and alveolar macrophages. Am. Rev. Respir. Dis. 139:1010-1016. 30. Aarden, L. A., E. R. De Groot, O. L. Schaap, and P. M. Lansdorp. 1987. Production ofhybidoma growth factor by human monocytes. Eur. J. Immunol. 17:1411-1416. 31. Bauer, J., U. Ganter, T. Geiger et al. 1988. Regulation of interleukin-6 expression in cultured human monocytes and monocyte-derived macro-

Kotloff, Little, and Elias: Alveolar Macrophage/Blood Monocyte IL-6 Production

phages. Blood 72: 1134-1140. 32. Navarro, S., N. Debili, J. Bernaudin, W. Vainchenker, and J. Doly. 1989. Regulation of the expression of IL-6 in human monocytes. J. lmmunol. 142:4339-4345. 33. May, L. T., J. Ghrayeb, U. Santhanameta!' 1988. Synthesis and secretion of multiple forms of 13rinterferon/B-cell differentiation factor 2/hepatocyte-stimulating factor by human fibroblasts and monocytes. J. Bio!. Chem. 263:7760-7766. 34. Jordana, M., C. Richards, L. B. Irving, andJ. Gauldie. 1988. Spontaneous in vitro release of alveolar-macrophage cytokines after the intratracheal instillation of bleomycin in rats. Am. Rev. Respir. Dis. 137: 1135-1140. 35. Elias, J. A., M. D. Rossman, R. B. Zurier, and R. P. Daniele. 1985. Human alveolar macrophage inhibition of lung fibroblast growth: a prostaglandin-dependent process. Am. Rev. Respir. Dis. 131:94-99. 36. Elias, J. A., P. Chien, K. M. Gustilo, and A. D. Schreiber. 1985. Differential interleukin-I elaboration by density-defined human monocyte subpopulations. Blood 66:298-301. 37. Elias, J. A., and V. Lentz. 1990. Interleukin-I and tumor necrosis factor synergistically stimulate fibroblast interleukin-6 production and stabilize interleukin-6 messenger RNA. J. lmmunol. 145:161-166. 38. Haverty, T. P., C. J. Kelly, W. H. Hines et al. 1988. Characterization of renal tubular epithelial cell line which secretes the autologous target antigen of autoimmune experimental interstitial nephritis. J. Cell Bioi. 107: 1359-1368. 39. Elias, J. A., M. M. Reynolds, R. M. Kotloff, andJ. A. Kern. 1989. Fibroblast interleukin 113: synergistic stimulation by recombinant interleukin 1 and tumor necrosis factor and posttranscriptional regulation. Proc. Nat!' Acad. Sci. USA 86:6171-6175.

505

40. Gross, V., T. Andus, J. Castell, D. Yom Berg, P. C. Heinrich, and W. Gerok. 1989. 0- and N-glycosylation lead to different molecular mass forms of human monocyte interleukin-6. FEBS Lett. 247:323-326. 41. Warner, S. J. C., K. R. Auger, and P. Libby. 1987. Interleukin 1 induces interleukin 11. II. Recombinant human interleukin 1 induces interleukin 1 production by adult human vascular endothelial cells. J. lmmunol. 139: 1911-1917. 42. Dinarello, C. A. 1989. Interleukin-l and its biologically related cytokines. Adv. lmmuno!. 44: 153-205. 43. Hance, A. J., S. Douches, R. J. Winchester, V. J. Ferrans, and R. G. Crystal. 1985. Characterization of mononuclear phagocyte subpopulations in the human lung by using monoclonal antibodies: changes in alveolar macrophage phenotype associated with pulmonary sarcoidosis. J. lmmunol. 134:284-292. 44. Fuchs, H. J., M. Sniezek, andJ. E. Shellito. 1988. Cellular influx and activation increase macrophage cytotoxicity and interleukin-I elaboration during pulmonary inflammation in rats. Am. Rev. Respir. Dis. 138: 572-577. 45. May, L. T., U. Santhanam, S. B. Tatter, J. Ghrayeb, and P. B. Sehgal. 1989. Multiple forms of human interleukin-6. Ann. NY Acad. Sci. 557: 114-121. 46. May, L. T., U. Santhanam, S. B. Tatter, N. Bhardwaj, J. Ghrayeb, and P. B. Sehgal. 1988. Phosphorylation of secreted forms of human 132-interferon/hepatocyte stimulating factor/interleukin-6. Biochem. Biophys. Res. Commun. 152:1144-1150. 47. Thomas, P. D., and G. W. Hunninghake. 1987. Current concepts of the pathogenesis of sarcoidosis. Am. Rev. Respir. Dis. 135:747-760.