© 2006 American Thoracic Society doi: 10.1164/rccm.200606-855ED
Fibroblasts in FocusDepartment of Pathology, The University of Michigan, Ann Arbor, Michigan
Department of Pathology, SUNY Upstate Medical Center, Syracuse, New York The term fibroblast foci refers to the discrete, microscopic, subepithelial zones of acute lung injury characteristic of usual interstitial pneumonia (UIP). The term was first applied to human lungs in an ultrastructural description of localized epithelial necrosis and alveolar septal collapse in UIP/IPF (1). Previous histopathologic reports had described these unique markers of acute lung injury in otherwise chronic interstitial lung disease (2, 3), but their potential significance did not emerge until subsequent electron microscopic and immunohistochemical studies (4, 5). In this issue of the Journal (pp. 654–658), Cool and colleagues challenge these now classical observations, declaring that fibroblast foci are "not discrete" and instead comprise a "fibroblast reticulum" (6). They begin by exploring the extent to which fibroblast foci are interconnected rather than isolated lesions using both a complex stereologic method and computer-generated three-dimensional reconstructions. The first approach uses point-counting applied to serial sections to estimate the Euler number (X), a positive integer purported to reflect extent of connectivity when applied to two-dimensional networks in three-dimensional space. Simply stated, the number of connections is proportional to the Euler number, and, therefore, the higher the number the greater the connectivity. Fibroblast foci had high Euler numbers (X = 19–39), and therefore high degrees of connectivity, in analysis of a single block from each of five UIP cases. Interestingly, the findings seem to indicate that fibroblast foci are more highly interconnected than blood vessels (X = 2) or alveolar macrophages, which were decidedly disconnected (X = 0)! In a single case of cryptogenic organizing pneumonia (COP) included for comparison, the characteristic intraluminal plugs of organizing fibroblasts showed a degree of connectivity (X = 26) similar to that seen for fibroblast foci. It is difficult to reconcile this mathematical simplification with a more logical world view, given what seems the inevitable connectivity of blood vessels and macrophages. While the significance of this numerical summary may elude many readers, it is perhaps made more accessible by including computer-generated three-dimensional illustrations in which a single fibroblast focus is depicted as having a more complex, interconnected, sheet-like architecture contiguous with visceral pleura. It is unclear from the available data the extent to which this illustration can be generalized to all fibroblast foci, an important point given the case-to-case variation in profusion or density of fibroblast foci. Fibroblast foci account for between 0 and 3.8% of the cross-sectional area of surgical lung biopsies when analyzed using a simple point-counting technique, and range in number from 2 to over 300 per square centimeter of examined lung tissue (7, 8). It seems reasonable to speculate that the extent to which fibroblast foci are interconnected is proportional to their prevalence as judged in two dimensions, and therefore likely to vary from one case to the next. In addition, application of static techniques to dynamic disease processes inevitably yields only an incomplete depiction, and it thus remains unknown whether fibroblast foci are propagated from pleura into lung, as the authors suggest, or from lung into pleura, a possibility that is equally appealing. The implications of these novel illustrations for pathogenesis are therefore limited, but raise the possibility that fibroblast foci are less discrete than what might have otherwise been imagined. Dr. Cool and colleagues suggest that there is little direct evidence to support the prevailing theory that fibroblast foci represent areas of acute lung injury and offer as an alternative the possibility that they are instead part of "an organized neoplasm," hypothesizing that connectivity might somehow indicate a monoclonal and therefore neoplastic process. There is, in fact, substantial evidence in support of the prevailing theory, including previously cited ultrastructural and immunohistochemical studies showing a combination of epithelial loss and collagen synthesis in fibroblast foci (1, 4, 5). Uhal and colleagues subsequently demonstrated a direct link between epithelial cell apoptosis and myofibroblasts in fibroblast foci (9). Less direct evidence includes increased expression of pro-apoptotic proteins and increased oxidative stress in alveolar epithelium from patients with UIP/IPF (10, 11). Undeterred by what might be legitimately viewed as compelling evidence, Cool and associates apply human androgen receptor gene methylation assays (HUMARA), a standard technique for addressing clonality in tissues from females, to several cases of UIP and demonstrate a pattern of balanced methylation typical of polyclonal processes. These observations and conclusions are consistent with all previously reported results indicating that, whatever else may be said of fibroblast foci, they are non-neoplastic in nature. The observation that at least some fibroblast foci may be less discrete and instead interconnected is interesting, but how it relates to the pathogenesis of lung fibrosis is unclear. The unique approach of Dr. Cool and colleagues adds to pre-existing studies while leaving many important questions for further study. Filling in details concerning the morphology of fibroblast foci has limited value in expanding our understanding of events critical to inciting and maintaining this abnormal response to an as yet unknown injury. It also remains uncertain how this morphologic clarification might impact current therapeutic strategies. Answering these persistent and important questions will no doubt require continued focus on the enigmatic fibroblast, whatever its connections. FOOTNOTES Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
Related articles in AJRCCM:
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
