| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Pulmonary and Critical Care Medicine Division, Virginia Commonwealth University, Richmond, Virginia
Correspondence and requests for reprints should be addressed to Norbert F. Voelkel, M.D., Sanger Building, 7th Floor, Room 7024, P.O. Box 980456, 1101 E. Marshall Street, Virginia Commonwealth University, Richmond, VA 23298. E-mail: nvoelkel{at}mcvh-vcu.edu
Art at times reflects mirages northern lights ... feasts of gods abysses takes on history too with mixed results attempts to domesticate it to give it human meaning.
—Zbigniew Herbert (1)
Many of the summarizers of previous Aspen Lung Conferences pointed out that a summary of these 3 full days of conference was a challenge—in fact, an impossible task. So why do we insist on this tradition of the Aspen Conference founding fathers? Do we enjoy a rhetorical exercise or is there a higher, not just artful, purpose? The reader is kindly asked to judge for her- or himself. As a veteran of these conferences I have benefited from the examples of former summarizers, and one particular role model comes to mind: the late Jack Reeves, who summarized the 30th Aspen Conference (2) and enjoyed impossible tasks. Having been called on to substitute for the real summarizer, 2 days before the start of this year's Thomas L. Petty Aspen Conference on Lung Injury and Repair, by the unusually cheerful chair of this 50th annual meeting, I saw three options: to decline, to feel utterly inadequate, or to take it personally. It's the third option that I accepted. Although an impossible task, "the summary" is an attempt at a contextual synthesis—in the good tradition of German Idealism, an attempt to define where we are and how to move from here on forward and also, to a degree, to crystallize what we really know.
The organizers of this 50th annual meeting, Drs. David Riches, Greg Cosgrove, and Stephen Frankel, deserve our gratitude for choosing outstanding State of the Art speakers to focus our interest on five topics: mechanisms of lung injury; resolution of lung injury; epithelium, mesenchyme, and matrix; stem cells and alveolar repair mechanisms of fibrosis; and the genetic control of lung injury, repair, and fibrosis.
DEFINITION OF LUNG INJURY
My private holistic view of lung injury is that of an acute or chronic departure of the integrated system that we call lung from homeostatic maintenance of its structure and function. What causes this departure is most often inflammation, or something leading to it, as has been eloquently illustrated by two of the State of the Art speakers, Tom Martin and Michael Matthay, discussing acute lung injury (3–6). Inflammation can be seen in terms of informatics, as a flow of information (7), or as the language favored by evolution to alert multicellular organisms to dangers of invasion, but also to pass on signals during wound healing, transplant rejection, and carcinogenesis. Apoptosis and cell proliferation are sentences of this language, and oxidants, cytokines, proteases, and peptides are the words of this language.
Acute Lung Injury
Acute lung injury can either be catastrophic and fatal, or resolve with the passage of time and supportive care; or it can be chronic and progressive, ultimately ending with fibrotic scars or emphysematous holes.
Thomas Martin took us to the intersection between mechanical forces and cellular biology when he presented compelling data demonstrating in ventilator-induced lung injury that it is the delivered volume, not the pressure, that causes the injury, and that tissue stretch sensitizes to the effects of endotoxin, enhancing cytokine output by activating the CD14 TLR4 (Toll-like receptor-4) coreceptor on macrophages. Soluble Fas ligand multimers form under the influence of oxidants increasing its bioactivity, leading to apoptosis of the stretch-sensitive alveolar epithelial cells. William Altemeier showed in mice that injurious mechanical ventilation activates the transcription of proinflammatory and proapoptotic genes. Further insightful and complementary data were provided by George Su, who showed that the integrin 
vβ5, which colocalizes with vβ3 at focal adhesion sites of pulmonary endothelial cells, was critically involved in the control of ischemia/reperfusion- and vascular endothelial growth factor–induced lung injury. Indeed, vβ5 antibodies inhibit the formation of actin stress fibers ex vivo and ventilator-induced lung injury in mice. Michael Shasby examined fundamental issues of epithelial cell barrier function and explained that the type 2 protease–activated receptor disables the association between E-cadherin and β-catenin by phosphorylating tyrosines 489 and/or 654 of β-catenin. Lorraine Ware (6) shed light on the mechanism of intraalveolar fibrin deposition (clot formation) by showing release of tissue factor containing microparticles in a model of stimulated lung epithelial cells, as well as release of plasminogen activator inhibitor-1 from these cells, providing the concept that the alveolar unit actively participates in clot formation and resolution.
Michael Matthay reviewed progress in the treatment of acute lung injury, pointing out that low tidal volume ventilation does reduce nonpulmonary organ failure, proposing that lung-directed therapies may reduce overall patient mortality and that restrictive fluid management shortens the overall duration of mechanical ventilation. Starting with his concept of alveolar edema resolution by treatment of the lung with β-adrenergic agonists, he introduced experimental intratracheal delivery of cells (cell-based therapy). In mice the intratracheal instillation of bone marrow–derived CD11b–CD45– mesenchymal stem cells—but not fibroblasts—decreased tumor necrosis factor-, increased IL-10, and improved survival. Treatment of perfused human lungs (4 hours after endotoxin challenge) with mesenchymal stem cells prevented alveolar flooding and preserved the integrity of the alveolar spaces (8, 9).
These provocative experimental results raise many questions. For example, where do the instilled cells land in the lungs? What do they do in such a short time? And to what cells?
Ventilator-induced lung injury appears presently to occupy many of the best minds in the field of acute lung injury, yet the vocabulary—or language—of ventilator-induced injury is likely only one of several vocabularies of acute lung injury.
RESOLUTION OF LUNG INJURY
We would like to understand how lung injury resolves, but we are still having trouble coming up with the right questions. One such question could be as follows: What is the difference between the fibroproliferative phase of adult respiratory distress syndrome, which often resolves spontaneously, and the nonresolving fibrosis in idiopathic pulmonary fibrosis (IPF)? Where in the injury/resolution continuum is the "point of no return" (Figure 1)? Can we extrapolate from fibrosis of the liver? In the liver, as David Brenner told us, removal of the stimulus is followed by resolution of injury. Are micro-RNAs, which play an important role during lung development, involved (Brigid Hogan)? But Dr. Hogan was unable to identify any lung-specific micro-RNAs. Is inflammation resolution (perhaps like conflict resolution) to some degree a learned, acquired skill, but more effective in the young and less effective in the aged? Active inflammation termination programs have been proposed by other investigators (10). Apparently regulatory T lymphocytes are part of this resolution program, perhaps in part by inducing granulocyte apoptosis (F. D'Alessio) and—at least in rodents—also multipotent epithelial progenitor cells (E. Rawlins). Robert Strieter in his State of the Art lecture proposed that, in addition to lung epithelial and endothelial cell damage, the loss of the basement membrane was the "point of no return," that transforming growth factor (TGF)-β was a necessary—but not sufficient—factor moving the tissue along the path to fibrosis and that CD3+ lymphocytes, dendritic cells, and alternatively activated macrophages were found at the sites of lung injury. He sees roles of different phenotypes of macrophages in both proline and collagen synthesis and in myofibroblast apoptosis, and assigned circulating fibrocytes (11) a role in the pathogenesis in fibrosis; they may be mobilized by granulocyte-macrophage colony–stimulating factor and turn (differentiate?) in the lung into lipofibroblasts. IPF may (also) be a systemic disease. Surely the cell–cell and cytokine mediators involved in fibrosis and fibrosis resolution are complex. To highlight a few of these: prostaglandin E2 enhances or inhibits IL-1β–triggered fibroblast growth via different prostaglandin receptors (K. White), and the transcription factor Nrf2 (nuclear factor erythroid 2–related factor-2), which determines antioxidant enzyme gene expression and is involved in alveolar type II cell proliferation control (N. Reddy).
|
Geoff Laurent reminded us that (at least in rodent lungs) about 10% of the collagen turns over every day, setting the stage for the role of lung matrix as a "dynamic remodeler" rather than a set of rigid structural elements. He agreed with Dr. Strieter that we should not side with epithelial cells versus inflammatory cells when it comes to the pathogenesis of pulmonary fibrosis. We must consider both. He developed his concept of fibrosis based on the activation of cascades of interacting oxidants and proteases and layers of cytokines, coagulation factors, and changes induced after mechanosensing of cells and by hypoxia, and he left room for the participation of stem cells. Fibroblasts respond to thrombin and up-regulate the expression of connective tissue growth factor; again, prostaglandin E2 is usually antifibrotic but patient fibroblasts (phenotype switch) are resistant and—when stimulated—do not make prostaglandin E2.
Philip Simonian reported a new hypersensitivity pneumonitis/fibrosis model triggered by Bacillus subtilis infection; here the lung lesions were characterized by influx of CD4+ and CD8+ T lymphocytes, and although 
T cells were among the infiltrating cells, –/– mice were not protected but T-cell receptor knockout mice were.
Matrix metalloproteinases (MMPs) digest components of the extracellular matrix, they are involved in many physiologically relevant processes of tissue remodeling, but they also are involved in organ fibrosis and tumor growth. Derek Radisky, using mouse mammary epithelial cells, demonstrated that stromelysin-1 (MMP-3) stimulates epithelial—mesenchymal transition, cytoskeletal alterations, genomic instability, and aberrant mitosis (12–14). This sequence of events, which is also driven by O2– production, may indeed have relevance for IPF, as nearly half of those affected develop cancer.
Diane Krause, Barry Stripp, and Luis Ortiz reported on bone marrow–derived hematopoietic and mesenchymal stem cells and on "conditionally differentiated" basal cells. Collectively, their experimental data indicate that after massive lung injury only 0.1% of the lung epithelial cells are bone marrow derived, that a mesenchymal stem cell subpopulation antagonizes IL-1–stimulated lymphocyte proliferation and ameliorates bleomycin-induced lung injury when administered early in disease development; apparently in the mouse airway cells can be activated to regenerate denuded epithelium and perhaps even alveolar cells. Dan Weiss showed ex vivo that keratinocyte growth factor and retinoic acid induced a surfactant protein D–expressing lung phenotype. In certain areas, stem cell research in the lung is moving forward but is still at an early stage. At present it appears that a consensus is building: mesenchymal stem cells may have profound paracrine effects, true engraftment of bone marrow–derived precursor cells may be a rare event, and whether there is activation of alveolar cell growth during airway epithelial cell repair remains an open question. Hiroshi Kubo reported on increased numbers of CD133+CD45–PECAM– cells in human fibrotic lung tissue but found decreased numbers of these bone marrow–derived precursor cells in patients with emphysema.
Myofibroblasts, originally acknowledged for their role in wound healing, can arise from lung epithelial cells via transdifferentiation. They are involved in extracellular matrix deposition, release inflammatory mediators, and induce epithelial call apoptosis (15, 16). Sem Phan addressed the question: How does one get a critical mass of myofibroblasts? Or, how does the myofibroblastic "focus" form? He offered a phenotype characterization of the bone marrow–derived cells and it is becoming clear that the myofibroblastic focus will remain a topic of research for years to come; it is clear that the mechanisms of transdifferentiation of these multiple myofibroblast progenitors (fibroblasts, smooth muscle cells, epithelium, and bone marrow–derived fibrocytes) and the mechanisms leading to these cellular agglomerates are not understood. The lysophosphatidic acid receptor may be involved (J. Hagood). How central is TGF-β and its persistent antiapoptotic actions in all of this?
LESSONS FROM THE LIVER: MECHANISMS OF FIBROSIS
The organizers need to be praised for inviting David Brenner to talk about liver fibrosis; he provided a convincing concept of the pathogenesis of liver fibrosis (17) in that reactive oxygen species, angiotensin II, leptin, and TGF-β were shown to be involved in the activation of hepatic stellate cells (HSCs). Apparently, the activated HSC turns into a myofibroblast and both TGF-β receptor type I kinase inhibition and TLR4 knockout prevent fibrosis. He also confirmed that liver fibrosis is in principle reversible and that increased expression of collagenase (MMP-1) and decreased expression of tissue inhibitor of MMP-1 (which inhibits HSC apoptosis) are associated with the resolution of liver fibrosis. The lessons from liver fibrosis for pulmonary fibrosis may be as follows: if we knew the nature of the pathogenic stimuli (in the liver: alcohol, viral infection, Cu2+, schistosomiasis, and lipids) then removal of the stimulus, for example, by treating a viral infection, might lead to resolution of fibrosis. How critical are active oxygen species in the pathogenesis of pulmonary fibrosis? (See References 18–20 for a discussion.)
GENETIC CONTROL OF LUNG INJURY REPAIR AND FIBROSIS
Of course, genetic control of lung injury repair and fibrosis is about single-nucleotide polymorphisms, genomewide association studies, candidate genes, and epigenetics. That there must be a genetic susceptibility to lung injury and fibrosis is apparent from our knowledge of familial pulmonary fibrosis, familial sarcoidosis, and a now quarter-century–old observation that only one-tenth of patients at risk develop adult respiratory distress syndrome (21). Two of the State of the Art speakers, David Schwartz and Joe Garcia, covered a wide array of concepts, methods, and data. At the time of this writing data concerning more than 350 families of patients with interstitial fibrosis have been collected; approximately 82% of these patients have IPF, and risk factors include male sex and cigarette smoking. Both association and candidate gene studies are underway. Dr. Schwartz also highlighted that epigenetic studies are now underway, bringing first results. Although Conrad Waddington proposed the concept of an epigenetic landscape in 1957 (22), ideas regarding the epigenetic control of phenotypes had been entertained before Waddington's work (see Arthur Koestler's The Case of the Midwife Toad [23]), and the study of potentially stable and heritable changes in gene expression, or cellular phenotypes independent of base pairing of DNA, is increasingly making progress as we understand more about DNA methylation, histone deacetylation, phosphorylation, and ubiquitination. We now know that many growth-promoting genes are activated via hypomethylation (24, 25) whereas tumor suppressor gene silencing has been linked to promoter hypermethylation. Because epigenetic plasticity may be a property of stem cells, a diet that changes the DNA methylation status may alter the way the animal, and the lung, respond to environmental stimuli. David Schwartz showed just that. A high-methylation diet (folate supplementation?) in mice caused blood eosinophilia and increased airway resistance, traits that were passed on to their offspring. Another highlight of this conference was the report by Mary Armanios on telomere mutations in patients with familial IPF. The phenotype of these patients was characterized by short stature, aplastic anemia, and a gray forelock; they had a haploinsufficiency in human telomerase reverse transcriptase (hTERT); hTERT mutations led to telomere shortening and bone marrow failure.
We can now rephrase the statements of several conference speakers and ask: to what degree is lung fibrosis a systemic disease and how exactly is lung homeostasis controlled by the bone marrow?
Finally, Dr. J. Garcia provided an impressive overview of his group's efforts to take results of acute lung injury patient whole-genome scan data all the way to a functional assessment of GADD-45 (growth arrest and DNA damage–inducible protein-45) and PBEF (pre–B-cell colony-enhancing factor), examining cultured endothelial cells and several murine models. His investigations led him to discover connections between sphingosine 1-phosphate and hyaluran receptors and to consider cortactin, myosin light chain kinase, and migration inhibition factor. Where to go from here? To system analysis and multigeneic mapping?
CONCLUSIONS
Progress? Yes! But there is no real conclusion, just the next step. There are still the old questions: how does a granuloma form? What kind of organization takes place in "organizing pneumonia"? When will we stop hoping that the bleomycin mouse model might be a model of IPF? For the better part of a week the masters and apprentices of the "glass bead game" came together and played—and what a beautiful game it was.
FOOTNOTES
Conflict of Interest Statement: N.F.V. has received lecture fees and grant support for preclinical research.
(Received in original form August 30, 2007; accepted in final form October 9, 2007)
REFERENCES
. J Biol Chem 2007;282:22910–22920.
| ||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
