| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Division of Therapeutics and Molecular Medicine, University of Nottingham, Nottingham, United Kingdom
Correspondence and requests for reprints should be addressed to Ian P. Hall, D.M., F.R.C.P., Division of Therapeutics and Molecular Medicine, University of Nottingham, Nottingham, NG7 2UH, UK. E-mail: ian.hall{at}nottingham.ac.uk
ABSTRACT
Myofibroblasts are mesenchyme-derived cells responsible for tissue repair after injury. Resident populations of myofibroblasts are present throughout the lung. In addition, it is likely that myofibroblast progenitors (fibrocytes) can migrate to the lung from the circulation during injury. The relationship and interdependencies among myofibroblasts, fibroblasts, and myocytes within the airway wall remain poorly understood. Myofibroblasts are likely to be present in primary culture systems derived from airway wall tissue. The phenotyping of cells in such cultures is complicated by the lack of specific markers for these cell types. Important responses including migration, synthetic function, and the regulation of matrix, in the normal and asthmatic airway previously considered to be driven by airway myocytes may in fact at least in part be due to responses of myofibroblast populations.
Key Words: airway myofibroblasts • fibroblasts • airway myocytes
The airway wall contains a number of cell types derived from mesenchymal cells, of which the most important contributors to airway remodeling are the airway smooth muscle cells, fibroblasts, and myofibroblasts. Of these three cell types, the myofibroblast is far less well understood than the other two. Myofibroblasts are contractile cells possessing morphologic and biochemical features intermediate between fibroblasts and smooth muscle cells (1). They are present in all tissues in the body and play an important role in remodeling after inflammation or injury (2). The key properties of myofibroblasts are as follows: (1) the ability to secrete matrix factors that alter the local environment within the tissue (a feature they share with fibroblasts), (2) their ability to multiply after injury, (3) their potential to differentiate into fibroblast and myocyte lineages, and (4) their ability to secrete growth factors and play a vital role in organogenesis or morphogenesis (3).
The source of myofibroblasts in the airway wall remains unclear: There is a large resident myofibroblast cell population, particularly under the basement membrane in the major airways (subepithelial myofibroblasts) and other resident myofibroblasts scattered throughout the submucosa. It has generally been assumed that these myofibroblasts are long-term resident cells derived from progenitor stem cell populations present within the airways. However, more recently, it has been suggested that at least some mesenchyme-derived cells can seed to the airway from the systemic circulation (4). Cells with mesenchymal features, known as fibrocytes, can be isolated in small numbers from the peripheral blood of both normal subjects and patients with asthma, and once placed into culture, these cells have the ability to adhere, multiply (5, 6), and develop a spindle-shaped morphology (7). As such, it is possible that circulating fibrocytes may contribute to airway wall remodeling in diseases such as asthma and also to the development of interstitial fibrosis in fibrosing lung conditions (8), but whether or not this happens in vivo and, if it does, to what extent this contributes to the overall expansion of myofibroblast populations is far from clear. Certainly, the number of such circulating cells is much smaller than the number of cells resident within the airway wall and, therefore intuitively, it seems likely that the majority of the response to inflammation and injury is mediated by resident rather than circulating cells, although during chronic inflammatory response, the contribution of circulating cells could theoretically be greater.
Another key issue that remains to be fully resolved is the relationship between mesenchyme-derived cell populations (9). Fibroblasts and airway smooth muscle have long been considered to be terminally differentiated cell types incapable of altering their phenotype, and each derived from myofibroblast progenitors. However, when airway myocytes are placed into primary culture from both animal and human lung biopsies, a population of cells with typical features of myofibroblasts is obtained, suggesting that at least in vitro airway myocytes may be capable of dedifferentiating into airway myofibroblasts (9). The other explanation for this phenomenon is that there are a small number of progenitor stem cells present within airway myocyte bundles that are capable of rapid division to create this population in primary culture, although such cells have not to date been identified in biopsies from human lung.
Similarly, the ability of fibroblasts to differentiate/dedifferentiate into myofibroblasts has been described in some studies, although the underlying processes have not been studied in detail. For example, in one study, increases in cell Smad3 activation led to up-regulation of 
-smooth muscle actin expression, leading the authors to consider that fibroblast to myofibroblast transformation had occurred (10). Human fibroblasts are known to exhibit heterogeneity with reference to the expression of Thy-1 on their cell surface and it has also been demonstrated that fibroblast cell populations expressing Thy-1 (Thy1+) differentiated into the myofibroblastic phenotype post-treatment with a stimulator such as transforming growth factor (TGF)-β (11). Studies using cultured corneal fibroblasts have shown that culture density also modulates the cellular phenotype. Corneal fibroblasts differentiated to a "myofibroblastic" phenotype when cultured at low density in serum-enriched media but maintained their phenotype in high-density seedings. These results suggested that loss of cell–cell contact is an important determinant of myofibroblast differentiation in addition to the regulatory effects of TGF-β (12). Some of the results of these studies have been previously reported in the form of an abstract (27).
DEFINING CHARACTERISTICS OF MESENCHYME-DERIVED POPULATIONS WITHIN THE AIRWAY WALL
Airway Smooth Muscle
Airway smooth muscle cells are characterized when terminally differentiated by the presence of a typical smooth muscle contractile apparatus (13). This is characterized by the presence of myosin heavy chain and -smooth muscle actin arranged in contractile filaments and a range of other contractile proteins, including high-molecular-weight caldesmon, vimentin, and desmin. These cells in vivo also express key G-protein–coupled receptors, including the muscarinic M3 receptor, the histamine H1 receptor, the Cys-leukotriene 1 receptor, and the β2-adrenoceptor (reviewed in Reference 14). Signaling of contraction is achieved by coupling of these receptors to signaling cascades, which allow the influx of intracellular calcium: Key ion channels that control contraction and relaxation of airway smooth muscle include large conductance Ca2+ activated potassium channels (BKCa) (15), small and intermediate conductance potassium channels (16), and a range of non–voltage-dependent calcium channels including TRP (transient receptor potential) channels 1, 3, 4, and 6 (17). When airway smooth muscle is placed in tissue culture conditions and cell numbers expand, there is rapid (within one passage) loss of some of these key components of contractile and relaxant signaling (13). In particular, M3 receptor expression and BKCa expression is rapidly lost, although β2-adrenoceptor and H1 receptor expression persists over repeated passages, and there is some residual potassium current carried probably in part by intermediate conductance Ca2+ activated potassium channels (13, 18, 19). Myosin heavy chain is gradually lost over repeated passage, which at least in part accounts for the lack of ability of cultured airway smooth muscle to contract rapidly in primary culture (20).
Fibroblasts
Classically, fibroblasts are not believed to contain organized contractile elements, although they are capable of slow contraction due to cytoskeletal rearrangement and also are able to migrate to regions of injury to secrete matrix factors. Markers believed to be specific for fibroblasts include fibroblast surface protein (21), the cell surface marker Thy-1 (22), and D7-FIB (a fibroblast-specific antigen) (23, 24), but there has been no systematic characterization of airway wall fibroblasts in terms of their phenotype.
Myofibroblasts
Fibrocytes present in the circulation can be identified by cell surface expression of CD45, CD34, collagen I, and CXCR4 (8); however, there are no data on the expression of this marker on resident myofibroblast populations in the airway wall. Given that there is likely to be a progenitor stem cell population within the airway wall, one would predict that more primitive markers might be present on these cells. Unfortunately, there are no specific markers for myofibroblasts per se, which has in part accounted for the difficulty in truly defining the populations of cells present, especially after cell culture of airway wall biopsies.
EXPRESSION PROFILING OF MESENCHYME-DERIVED CELL POPULATIONS
In an attempt to resolve issues around the characterization of the mesenchyme-derived cell populations in the airway, we have recently undertaken extensive immunocytochemical studies using a range of cell populations to try and identify specific markers for different cell populations. To undertake this work, we profiled the expression of a number of potential cell lineage–specific markers in (1) human airway myocytes from both early and late passage numbers, (2) a well-established fibroblast cell line (MRC5), (3) a myofibroblast-like population derived from human fetal lung, and (4) a commercially available human bronchial smooth muscle cell population. The results of this work are summarized in Table 1.
|
-smooth muscle actin and myosin heavy chain as detected by Western blot), and some qualitative differences in the distribution of, for example, -smooth muscle actin, which is believed to demonstrate a myofibrillar staining pattern in airway myocytes. The proportion of cells exhibiting this feature was quite diverse (Figure 1). Almost all the human fetal lung myofibroblasts expressed a filamentous staining pattern (>95%), whereas the same feature was observed in about 60 to 70% of the commercially available bronchial smooth muscle cells. This proportion was comparatively lower in high passage human airway myocytes (
20%) and low passage human airway myocytes (30%). In fibroblasts, -smooth muscle actin revealed diffuse irregular cytoplasmic staining with no filamentous pattern. Immunocytochemical studies revealed characteristic filamentous staining for smooth muscle myosin in human airway myocytes, whereas fibroblasts showed diffuse low-intensity staining with absence of any filaments. In fibroblasts, immunocytochemical studies for fibroblast surface protein was characterized by a punctate staining pattern defined by intense perinuclear fluorescence. Other cell types were observed to be weakly but uniformly stained throughout their long axis. Monoclonal fibroblast surface protein antibody recognizes a series of proteins present on the surface of human fibroblasts and other fibroblast cell lines and also on tissue macrophages and peripheral blood monocytes (21). Western blot analysis revealed the presence of -smooth muscle actin in all cell populations in keeping with immunocytochemistry findings (Figure 2). In these experiments, note that apparent quantitative differences in -smooth muscle actin expression may be related to seeding density or protein loading. At present, therefore, we believe that care must be taken in the interpretation of immunocytochemical staining in cell populations derived from the airway wall in terms of interpretation of the cell types present: Certainly, our data suggest that the common practice of using -smooth muscle actin expression as a marker by itself of smooth muscle phenotype is likely to be misleading given that this protein is expressed in all the cell lineages we studied.
|
|
One other approach that can potentially be used to clarify the relationship between different cell populations is the use of single-cell cloning techniques. Again, we have recently used this approach to determine whether different lineages of airway mesenchymal cells can be derived from early passage myocyte cultures using both dilution cloning and cell sorting approaches to produce single cell cultures (26). Surprisingly, we identified that many single cells present within early human airway myocyte cultures derived from explants have the potential to create expanded cell populations, suggesting that there are a large number of progenitor stem cells present within these cultures. These populations differ to some extent both in their doubling time and their ability to continue multiplying, but we have not to date seen major differences in the expression of potential lineage markers from these populations derived from single cells.
CONTROL OF PROLIFERATION AND APOPTOSIS
The number of individual cells within a cell population is controlled by the balance between proliferation and apoptosis. Although the number of airway myocytes needs to be kept constant over time, numbers of fibroblast and myofibroblast cells need to expand and contract after inflammatory injury. Therefore, these cells maintain a proliferative potential both in vivo and in culture conditions. This is accompanied by a marked resistance to apoptosis. A wide range of inflammatory mediators that have been found in inflamed airways have been shown to be mitogens for airway myofibroblasts (and also for cells derived from airway myocyte populations) of which the most important are probably platelet-derived growth factor (PDGF), epidermal growth factor (EGF); thrombin, histamine, and tryptases (27). As discussed above, whether or not the airway myocyte is capable of proliferation in its differentiated state is unclear and it remains possible that the cells derived from airway myocyte tissue that undergo expansion are relatively more primitive myofibroblast progenitors.
Airway myofibroblasts are extremely resistant to apoptosis induced through mechanisms such as serum deprivation. The probable explanation of this phenomenon is that the cells receive a very strong survival signal, which appears to be driven by matrix interaction with integrin receptors on the cell surface (28).
Airway myofibroblasts express a range of integrin receptors, such as 1, 3, 4, 5, v, and ß1, of which the most important are probably 5 and β1 because these exert a strong survival signal in airway myofibroblasts (28, 29). The range of integrin receptors expressed appears to be similar to those receptors described from very early passage airway myofibroblast cultures, again suggesting that these cells may share many characteristics. We have recently assessed integrin expression in the range of cell types described above. Flow cytometry results revealed that 1 integrins were positive in less than 5% of early passage human airway myofibroblasts but were expressed at higher levels (about 80%) in fibroblasts. This proportion increased with increasing human airway myofibroblast passage number (30%). 2 Integrins were maximally expressed on the surface of fibroblasts and late passage human airway myofibroblasts. Approximately 50% of human airway smooth muscle cells and commercially available bronchial smooth muscle cells expressed 2 integrins on their surface. These results were also confirmed by immunocytochemical studies, in which the investigators looked at the qualitative expression of these integrins. Both 1 and 2 integrins were expressed on all cell types, with the 2 integrin subtype being strongly expressed on all cells.
In addition to matrix factors providing a strong survival signal for airway myofibroblasts, the interaction between matrix and cell surface integrins also influences downstream signaling. For example, β2-adrenoceptor–mediated cyclic AMP formation is modulated by integrin interaction with the extracellular matrix, with heightened responses being seen after exposure to fibronectin in comparison to collagen V or laminin (30). The complex interactions between the cell and surrounding matrix are further complicated by the ability of the cell to influence the immediate matrix environment not only through secretion of new matrix factors but also by modulation of existing matrix via production of matrix metalloproteinases (MMPs) such as MMP-2 (31).
SYNTHETIC FUNCTIONS
There is an extensive literature demonstrating that cells derived from airway myocyte cultures have a synthetic phenotype when maintained in cell culture in vitro (reviewed in Reference 27). There are far fewer data on the ability of airway myocytes in vivo to synthesize proinflammatory cytokines and chemokines. There are no data available specifically describing synthetic functions of airway myofibroblasts in vivo, although myofibroblasts in other tissues have been shown to have significant synthetic function. These data suggest that the synthetic function of myofibroblast cells within the airway is part of a relatively dedifferentiated phenotype more akin to a myofibroblast than a terminally differentiated airway myocyte, and lend weight to the concept that cultured airway myocytes have a myofibroblast phenotype rather than a terminally differentiated airway myocyte phenotype.
CELL MIGRATION
Classic migration studies using cells derived from airway wall biopsies demonstrate chemotaxis to a range of mediators such as PDGF (32). This is perhaps one of the strongest bits of evidence to suggest that the majority of cells studied in cell culture experiments are of the myofibroblast phenotype. Airway myocytes in the airway wall are present within bundles of airway smooth muscle, and morphometric studies have demonstrated that it is these bundles that increase in size in allergic airway inflammation in both sensitized rodent models after allergen challenge and in ex vivo biopsies from asthmatic lung. To date, the development of new airway smooth muscle bundles under the subepithelial basement membrane has not been described. These data taken together suggest that the migratory response of mesenchymal cells in the airway is designed to allow the movement of airway myofibroblasts to regions of inflammation within the airway wall to synthesize matrix and/or to expand cell numbers to remodel the airway wall. In practice, this migratory response has been very difficult to study in vivo using classical cell trafficking approaches because of the lack of specific markers.
CONCLUSIONS
Historically, airway myocytes, myofibroblasts, and fibroblasts within the airway wall have been considered to be separate cell populations derived from progenitor mesenchymal stem cells. The extensive work over the last 10 years using cultured cell populations derived from airway smooth muscle or airway wall biopsies suggests that these populations may be much more fluid and capable of both differentiation and also dedifferentiation. Key issues that remain to be determined are whether or not a terminally differentiated airway myocyte is able to dedifferentiate to an airway myofibroblast or whether or not airway smooth muscle bundles contain a small number of progenitor cells from which airway myofibroblasts arise. This has led to significant problems with the phenotypic definition of cell populations used for cell biology studies in the literature. In practice, cultures derived from the human airway wall are a valuable source of nontransformed airway cells, expressing signaling cascades of fundamental importance to airway biology, and the study of these is undoubtedly likely to be more physiologically relevant than the use of many transformed cell systems. However, the lack of clear phenotypic markers of differentiation that can reliably be used to fully distinguish between different cell populations creates difficulties. This is a particular problem for recent studies in which cell populations derived from asthmatic and nonasthmatic airway wall biopsies have been grown by a number of groups, with differences in proliferative and signaling responses being described by some but not all groups. Without fully understanding the phenotype of the cells being examined in these studies, the physiologic importance of these findings remains unclear.
The fundamental role of an airway myocyte is to control airway tone by contraction and relaxation. However, cells derived from airway myocytes fail to contract after expansion in culture and the surrogates used to attempt to study contraction (e.g. twisting magnetic cytometry, cell shortening measured by microscopy, or contraction of collagen gels) have relatively slow time constants and therefore probably measure cytoskeletal rearrangement rather than true contraction. Indeed, it is difficult to see how a cell population with marked down-regulation of myosin heavy chain would be able to contract in a classical smooth muscle cell–like manner. In conclusion, therefore, it appears that cells derived from mesenchymal cell populations in the airway wall should be considered to be a more fluid and dynamic population than perhaps has previously been believed. These cells still provide extremely valuable physiologic models but overinterpretation of data generated using these models should be avoided. The solution to providing integrated in vitro models therefore depends on the combination of cell culture systems together with ex vivo tissue culture model systems, such as lung slice preparations, that allow short- to medium-term in vitro responses to be measured in a more physiologically relevant setting.
FOOTNOTES
Work in the authors' laboratory is supported by the Medical Research Council and Asthma UK.
Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
(Received in original form June 15, 2007; accepted in final form August 9, 2007)
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
