Published ahead of print on November 19, 2004, doi:10.1164/rccm.200406-788PP
© 2005 American Thoracic Society doi: 10.1164/rccm.200406-788PP
Lung Inflammation and FibrosisAn Alveolar Macrophage–centered Perspective from the 1970s to 1980s
Pennsylvania State University College of Medicine, Hershey, Pennsylvania; and Division of Lung Diseases, National Heart, Lung and Blood Institute, National Institutes of Health, Department of Health and Human Services, Bethesda, Maryland Correspondence and request for reprints should be addressed to Herbert Y. Reynolds, M.D., DLD/NHLBI, Two Rockledge Center, 6701 Rockledge Drive, Bethesda, MD 20892-7952. E-mail: reynoldh{at}mail.nih.gov
Key Words: cytokines–chemokines • effector immune function The alveolar inflammatory reaction, important for amplifying host defenses in the lungs, is in part under the cellular control of the pulmonary alveolar macrophage, which can initiate and modulate the reaction. A chronic, poorly regulated reaction can cause tissue dysfunction including fibrosis. This historical perspective will review the efferent function of alveolar macrophages that elicits inflammation through secretion of several chemokines, examining how some of these were identified from normal macrophages, and later investigated in lung fibrosis, particularly idiopathic pulmonary fibrosis (IPF). The focus is on two chemokines: interleukin-8 (IL-8) and leukotriene B4 (LTB4). HISTORICAL HIGHLIGHTS OF INFLAMMATION Although purulence in tissue has been described since antiquity, systematic study of lung inflammation began about 200 years ago. René T. H. Laënnec, an anatomical pathologist and clinician who correlated physical signs obtained by indirect auscultation (stethoscope) with lung pathology, described in his Traité, published in 1819, a sequence of three tissue phases in lobar pneumonia (1). But he may not have had a microscope with sufficient resolution then to identify the cell types involved. Microscopy was not perfected until the 1850s and used then by Rudolph Virchow to advance his cellular theory of disease (2). In 1882, Elie Metchnikoff observed that the tissue of starfish larvae engulfed foreign particles with mesodermic layer cells which he termed phagocytes (3), thus identifying a reactive inflammatory process. This led to his cell-based theory of immunity that phagocytic inflammation was protective and that phagocytes assisted with tissue repair and provided surveillance for maglinant cells. When William Osler published his textbook of medicine in 1892, he could be more certain in describing specific cellular changes in pneumonia. As Osler wrote, "since the time of Laënnec, pathologists have recognized three stages of the inflamed lung-engorgement, red hepatization and gray hepatization—the exudate and air cells are densely filled with leucocytes" (4). From the late nineteenth century until the 1960s, there was considerable interest in clinical management of pulmonary infections, especially tuberculosis and pneumococcal pneumonia, but scant evidence of lung research about cellular mechanisms that resisted infection or created inflammation. Bacteria sprayed into the lungs of mice were found to disappear rapidly by tissue culture (5), and this quick clearance was considered important for combating infection. But which component(s) in the airways were most effective in the initial inactivation of bacteria was uncertain—mucociliary clearance, lymphatic drainage, and alveolar macrophages were all considered. LUNG HOST DEFENSE: ALVEOLAR MACROPHAGES AND INFLAMMATION Drs. Gareth Green and Edward Kass were perhaps the first to study definitively the role alveolar macrophages play in clearing a low innoculum of bacteria aerosolized into the lungs of mice (6). They found that microbes were contained by alveolar macrophages with rapid killing; no other host defense seemed needed and an influx of polymorphonuclear cells (PMNs) did not occur. This research was presented at the American Thoracic Society meeting in May 1963. This collective work was later synthesized in a memorable presentation for the J. Burns Amberson Lecture "In Defense of the Lung" at the ATS meeting in 1970 in which Dr. Green elaborated on the spectrum of pulmonary host defenses that protect against many environmental factors. The seminal publication (7) that followed had a great impact on focusing research on lung host defenses pertaining to cellular immunity, especially in humans. This coincided with many new concepts evolving about innate immunity, especially cellular immune effects related to lymphocytes and macrophages (8). Several questions emerged: (1) Was macrophage phagocytosis more of a surveillance mechanism that could be overwhelmed by a large bacterial innoculum or more virulent microbe? (2) What did attract PMNs into the alveolar spaces when pneumonitis with tissue inflammation developed clinically? (3) How did opsonic antibodies enhance alveolar macrophage effectiveness? Finally, (4) what made inflammation subside and clear (or made it fester chronically)? Research using in vitro cultured alveolar macrophages might answer some of these issues about initiating and modulating inflammation. Dr. Quentin Myrvik and colleagues, Department of Microbiology, University of Virginia School of Medicine, Charlottesville, Virginia, were the first to publish a method for lavaging normal rabbit lungs to retrieve alveolar macrophages for in vitro study (9). Although considerable research had been done on macrophages, these were usually glycogen-induced cells obtained from the peritoneum; thereafter other studies retrieved alveolar macrophages from the lung (10) to study their metabolism or phagocytic uptake. I had a special privilege, while working in research during medical school, to observe Dr. Myrvik lavage lungs for macrophages, which provided the segué for my involvement in the immunologic fervor sweeping research. In the late 1960s at the National Institutes of Health Clinical Campus in Bethesda, Maryland, an immunologic approach to understand and treat many forms of human disease was widespread, especially in the Laboratory of Clinical Investigation of the National Institute of Allergy and Infectious Diseases, under the direction of Dr. Sheldon M. Wolff (8), where I began as a Clinical Associate (1967–1970). Research there emphasized how the host responded to infection, especially PMN function and inflammation. My research efforts were directed to the respiratory tract, greatly influenced by Dr. Green's perspective (7), and coupled with the problem of Pseudomonas aeruginosa lung infection, a significant clinical problem for immunocompromised, hospitalized patients (11). Our experimental model initially was in rabbits (12, 13) to sample airway secretions and cells, especially alveolar macrophages. Soon we used nonhuman primates (14) and human subjects (15). In the late 1960s to early 1970s, many investigators had begun to wash out human airways to retrieve alveolar macrophages from normal subjects, cigarette smokers, and nonsmokers, with a variety endobronchial tubes anchored in the airways (reviewed in Reference 16). Initial studies by several groups especially affected other investigators (17–21). Dr. Harold Newball and I were very fortunate to use one of the first fiberoptic bronchoscopes available in the United States for our initial bronchoalveolar lavage (BAL) studies to collect fluid and cells for an analysis of immunity components in the normal airways (15), and to study the interaction between Pseudomonas bacteria–specific opsonins–surface cell receptors on alveolar macrophages that could enhance phagocytosis (22, 23). Phagocytosis and microbial killing were afferent functions of macrophages, but did not provide insight into an inflammatory response. Also, subpopulations of BAL lymphocytes were characterized for their role in local lung immunity (24). EXPANDING THE FUNCTION OF ALVEOLAR MACROPHAGES IN LUNG INNATE IMMUNITY Understanding the kinetics of inflammation in the lung became a paradigm for investigating mechanisms that created acute pneumonitis or caused faulty resolution that could induce chronicity and lead to fibrosis and organ dysfunction (25). Stimulating alveolar macrophages became a new research approach to discover what activated macrophages might produce in the way of soluble factors that could diffuse into the local alveolar milieu and interact with other immunity components. This began to expand the concept of an alveolar macrophage to be more than just a phagocyte or scavenger cell and recognized it as an immune effecter cell (26). This led us and others to investigate if substances elaborated by activated macrophages had an effect on responder cells, namely PMNs. PMNs were the important responder cells to assess, for so many are adherent to the endothelium of alveolar capillaries where perhaps 20% of the host's marginated pool of PMNs reside (27). Using BAL fluid (BALF) and respiratory cells obtained from normal rhesus monkeys, substances were identified in BALF with chemotactic activity and produced by alveolar macrophage cells in culture for PMNs (28). Five repeated lung lavages in the same animals over a 72-hour interval produced maximum recovery of PMNs and mononuclear cells in BALF at 4 and 24 hours. At 24 hours, a 10-fold increase in respiratory cells was found with more than 90% as PMNs. In concentrated BALF, two peaks of chemotactic activity were identified in elution patterns from column fractionation: one containing an approximately 15,000-D molecular weight substance that was C5a and a chemoattractant for PMNs and for mononuclear cells; and another of approximately 5,000 D in size, selective for the directed migration of PMNs, that was not a complement fragment. Thus, a major chemoattractant in BALF was C5a, an active fragment as found previously by Ward and Zvaifler (29) from the cleavage of C5 to C5a by enzymes from PMNs. In addition, stimulated alveolar macrophages, challenged with opsonized, heat-killed Staphylococcus albus, synthesized and released a smaller substance, not an apparent complement component, which was selectively chemotactic for PMNs. Hunninghake and colleagues (30), also in the Laboratory of Clinical Investigation, reported on chemotaxins from guinea pig lung cells. They found that alveolar macrophages produced a much smaller size factor also with preferential activity for PMNs. This factor could be generated in vivo after intratracheal inoculation of heat-killed Staphylococci. Gadek and colleagues (31), using alveolar macrophages prepared from guinea pig lungs, extended the spectrum of stimuli that produced the macrophage-derived neutrophil chemotactic factor, indicating that its release in the lung might occur under other noninfectious conditions. Other particulate stimuli and IgG immune complexes were effective. Also, fixation of C3b to a particle would augment release of the chemotactic factor. This raised the important possibility that the alveolar macrophage might respond to many different stimuli to generate an inflammatory response.
The search for chemotaxins was extended to normal human alveolar macrophages (32). But results from in vitro–cultured human macrophages were more complex. Instead of a single factor with chemoattractant activity as found with subhuman primate cells and guinea pig cells, two factors were identified that selectively attracted PMNs but not mononuclear cells. The larger substance (
Further studies of human alveolar macrophages found that metabolites of lipoxygenase pathway of arachidonic acid were produced, especially LTB4 when stimulated by calcium ionophore (35) and leukotriene C4 by IgE (36). Martin and colleagues (37), also studying human alveolar macrophage production of LTB4, found that this lipoxygenase substance itself had little chemoattractant activity for alveolar macrophages and was more potent for blood PMNs and less so for monocytes. Most of the chemotactic activity produced by these cells, when stimulated by calcium flux across the cell suface membrance or with phagocytosis of opsonized particles, was LTB4 (37). Moreover, if interferon- Sylvester and colleagues (39) found that alveolar macrophages, after LPS stimulation, produced neutrophil attractant/activation protein (NAP-1), as did blood monocytes, which had a molecular mass of 8,000 D. To unravel the relationship between these multiple chemokines, Rankin and associates (40) performed a kinetic study. Various stimuli (LPS, calcium ionophore A23187, and zymosan) were used to stimulate human alveolar macrophages during a 24-hour period with multiple samplings of culture supernatants for evidence of specific chemotactic factors. A timed release sequence was established for at least two chemotactic substances for PMNs: LTB4 and NAP-1. The former was released more quickly after LPS stimulation for 3 to 5 hours, and NAP-1 was a later phase factor appearing after 3 to 5 hours but increasing during the 24-hour period. It was considered (40) that NAP-1 was probably the larger chemotactic factor initially found by Merrill and colleagues (32). NAP-1 and interleukin-8 were determined to be the same cytokine and designated IL-8 (41).
In summary, after the initial isolation of alveolar macrophage–derived chemotactic substances from animal cells and soon thereafter from human cells, it took another decade of rather concerted research to completely characterize several of these chemotactic factors. These chemokines enable alveolar macrophages to generate inflammation in the alveoli as part of the host's immunity defense (42). The list of important chemokines expanded further and included IL-8, LTB4, IL-1, and TNF- INHIBITING INFLAMMATION HAS A MACROPHAGE COMPONENT Eliciting inflammation is an effector function of lung host defense, but suppressing and terminating the reaction is part of the control process, too, and also involves macrophages. Failure to resolve inflammation, once its function is completed, and to clean up the process and restore tissue again to its normal histology and physiologic function can leave residual impairment. Persistent, smoldering inflammation often leads to tissue fibrosis. Sibille and colleagues (44) observed that certain portions of fractionated culture supernatants obtained from normal, nonsmoker alveolar macrophages had an inhibitory effect on the random migration and directed chemotaxis of PMNs. A small molecular weight substance (< 1,000 D) was found to have a broad spectrum of inhibitory activity against many stimuli, including IL-1, FMLP, C5a, and LTB4, that usually elicited chemotactic activity for PMNs. Cooper and colleagues (45) isolated and defined this chemotactic inhibitor of BALF that was secreted by alveolar macrophages (45, 46). The surprising observation was that a small-length polypeptide, which is part of the human armamentarium to suppress PMN function, in fact, resembled a natural portion of influenza A nucleoprotein (47), a common respiratory viral pathogen. A short eleven–amino acid peptide human inhibitor was synthesized and designated synthetic neutrophil inhibitor peptide or SNIP (48). Its activity was then assessed in normal humans, after IL-8–induced nasal inflammation had been created. Nasal pre-exposure to SNIP significantly reduced total cells, PMNs, and protein in nasal lavage (maximal effect found about 3 hours after IL-8 exposure). In vitro, SNIP reduced chemotaxis of blood PMNs, increased the percentage of apoptotic PMNs, and was found to bind specifically to PMNs through an intergrin, CR3, and its associated ligand, fibrinogen (48). What is fascinating about these findings is that influenza virus might use its peptide to render PMNs less effective as inflammatory cells and to increase cell death (apoptosis) as a strategy to gain entry into respiratory epithelial cells and create infection. However, the human host seems to have developed a similar mechanism or co-opted it as a possible regulatory function to limit inflammation or help resolve it. Therefore, bidirectional regulation of the inflammatory response is important. After a successful inflammatory response, active suppression should occur with cellular clean up by apoptosis of cells and any residual of active inflammatory byproducts; restoration of normal tissue histology and function hopefully occurs. CONNECTING CHRONIC LUNG INFLAMMATION WITH FIBROSIS As mentioned, persistent smoldering inflammation in tissue can cause structural changes and create fibrosis that becomes a disease process. Such chronicity can be perpetuated by an antigen, microbe, or autoimmune stimulus, or it may reflect inadequate regulation to suppress the reaction (resolution). The opportunity to study—if this connection might be true—was provided by a continuing collaborative effort with the Pulmonary Branch where Dr. Ronald G. Crystal and colleagues were investigating patients with forms of diffuse interstitial lung diseases (49).
Drs. Kazmierowski, Frank, and I were from the Laboratory of Clinical Investigation, NIAID, and collaborated on describing the BALF and cellular profiles of the initial patients studied with IPF and chronic hypersensitivity pneumonitis (50). In 19 patients considered diagnostically to have IPF (7 untreated, 12 on treatment with oral corticosteroids), BALF cellular analysis indicated current inflammation in the distal airways and alveolar space, as cell differential counts revealed an elevated percentage of PMNs ( Because of the progressive clinical nature of IPF, reflecting ongoing alveolar-interstitial fibrosis, the Pulmonary Branch, NHLBI, focused particularly on delineating components involved in the inflammatory and fibrogenic process. A mechanism by which PMNs were attracted to the lung, contributing to the alveolitis of IPF, was found by Hunninghake and colleagues (53). They previously had observed that human alveolar macrophages produced a small molecular weight, partially lipid chemotactic factor (400–600 D in size) (33). They investigated whether IPF alveolar macrophages did so, and this small size factor was identified in patients with IPF. Moreover, the proportion of PMNs in BALF and the release of this factor correlated (53). Immune complexes might stimulate IgG Fc surface receptors on alveolar macrophages to release it.
Continuing the search for mechanisms by which alveolar macrophages from patients with IPF might contribute to inflammation and then fibrosis, Carré and colleagues (54) measured the IL-8 concentration in BALF from patients with IPF. Patients had approximately 58 pg/ml of IL-8, compared with healthy nonsmokers' BALF ( Investigators in the Pulmonary Branch, NIH, and at many other research centers continued to study cells and substances in BALF from IPF patients, probing for mechanisms that contributed to fibrosis and seeking possible targets for new therapeutic agents. Some of the same alveolar macrophage effector functions that initiate and modulate normal lung inflammation were identified in IPF patients and seemed to be related to fibrogenesis. Many of these highlights of research into mechanisms of inflammation and fibrosis were reviewed by Ward and Hunninghake (25). But the research paradigm began to shift again in the 1990's. A reappraisal of the morphologic forms of pulmonary fibrosis had begun in the mid 1980's. Kuhn and colleagues (55) first introduced an integrated concept for patients with IPF-UIP that an alveolar epithelial injury could occur initially, producing organized alveolar exudates with fibroblasts proliferating and the synthesis of connective tissue occurring. Subepithelial fibrogenic foci then developed. This emphasized an abnormal remodeling of lung parenchyma after local injury. Moreover, inflammation in IPF was not a consistent finding in BALF cellular analysis of patients, nor found in the histopathology of lung tissue biopsy when the IPF-UIP pattern was used to rigorously classify disease (56). The reclassification of IPF proposed by Katzenstein and Myers (57) was of great importance, as it included the entity of nonspecific interstitial pneumonia (NSIP) (58) and reemphasized the distinctive features of UIP. This new paradigm of fibrogenesis, based on IPF-UIP lung histopathology, has now redirected basic research from inflammation to the epithelial injury-fibrosis response (59) and prompted revising clinical and treatment concepts (60). Acknowledgments The preparation of the manuscript by Mrs. Susan Crawford, Mr. Jay Olegario, Ms. Desiree Alvarez, and Mr. Cuong Nguyen is appreciated. My close associates created most of the findings, and I am indebted to their creativity, spirit, and research work; many thanks to: Drs. J. A. Kazmierowski, J. E. Pennington*, R. P. Aduan, and Mr. R. E. Thompson (LCI, NIAID, NIH); and Drs. W. W. Merrill*, J. A. Rankin*, Y. Sibille, K. R. Young, J. A. D. Cooper*, R. A. Matthay*, and Mr. G. P. Naegel (Yale University's School of Medicine). The collaboration of Dr. R. G. Crystal and associates in the Pulmonary Branch, NHLBI, was essential. Finally, the guidance and advice of Drs. C. M. Kunin, J. S. Johnson, and S. M. Wolff sustained me in my early research career. I appreciate the review of the manuscript by Ronald G. Crystal, Gary W. Hunninghake, and John I. Gallin. *Denotes American Lung Association nationally supported research awardee. FOOTNOTES This article was the basis for a presentation at a joint ALA/ATS symposium: "Milestones in Advancing Lung Disease Research: An Historical Contrast from Then to Now"; May 24, 2004, ATS International Meeting, Orlando, Florida. Conflict of Interest Statement: H.Y.R. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form June 21, 2004; accepted in final form November 15, 2004 REFERENCES
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
10,000 D) was not C5a, nor did it interact with the PMN receptor for formyl-methionyl-leucyl-phenylalanine. The nature of the smaller factor (< 1,000 D) was not extensively characterized. Hunninghake and colleagues (
was preincubated with normal human alveolar macrophages, before stimulation with aggregated IgG, a much greater amount of LTB4 was produced (
, and undoubtedly others will be found (
