| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Clinical Genomics Group, Royal Brompton Hospital, London, United Kingdom
Correspondence and requests for reprints should be addressed to R. M. du Bois, M.A., M.D., F.R.C.P., Clinical Genomics Group, Royal Brompton Hospital, Sydney Street, London SW3 6NP, UK. E-mail: r.dubois{at}rbht.nhs.uk
ABSTRACT
Scleroderma (systemic sclerosis) is characterized by dermal thickening and is subclassified, on the basis of the pattern of skin involvement, as diffuse or limited cutaneous disease. The lung fibrosis associated with systemic sclerosis is histopathologically nonspecific interstitial pneumonia and occurs to various extents. A key determinant of the development of lung fibrosis is the carriage of the anti–DNA topoisomerase II autoantibody, which is driven by genotype, particularly major histocompatibility complex class II alleles. Epithelial and endothelial cell injury initiates lung pathology and the local milieu expresses all the expected components of an immune/inflammatory chronic process that has fibrosis as an associated feature. However, novel concepts of the pathogenesis include the role of epithelial mesenchymal transition as a source of myofibroblasts following epithelial cell triggering and the concept of fibroblast heterogeneity in terms of both origin and function. Focus on the poles of the pathogenetic process, than on the redundancy of multiple mediator pathways, may provide more translational ideas in terms of elucidating how epithelial triggering initiates the downstream events and how fibroblast proliferation and connective matrix deposition is controlled, together with establishing how the process of autoantibody formation relates to the pathophysiologic events. Although much remains to be learned about the mechanisms of scleroderma-induced lung disease the clarity of the questions is improving.
Key Words: systemic sclerosis • genetics • autoantibody • epithelial mesenchymal transition
"Systemic sclerosis" (SSc) embraces a wide range of clinical phenotypes that present with varying degrees of skin sclerosis as the unifying feature (1). Systemic sclerosis is subdivided clinically, based on the extent of skin involvement. Patients with limited cutaneous systemic sclerosis (lcSSc) have skin disease that does not extend proximal to the elbow and the knee in the limbs, although the face and neck may also be involved. Patients with diffuse cutaneous systemic sclerosis (dcSSc) have skin involvement that extends proximal to the elbow and knee and not uncommonly affects the trunk as well as the neck and face. The American Rheumatism Association (now the American College of Rheumatology, Atlanta, GA) preliminary criteria for the diagnosis of systemic sclerosis (2) require the presence of the major criterion or two or more minor criteria:
There are two major pulmonary manifestations of disease: interstitial fibrosis, which can be sufficiently severe to result in secondary pulmonary hypertension, and pulmonary vascular disease, which has the same clinical and histopathologic features as idiopathic pulmonary arterial hypertension. These two manifestations are generally mutually exclusive although occasionally overlapping features may be observed.
The focus of this review is the mechanism of interstitial fibrosis.
SSC–ASSOCIATED PULMONARY FIBROSIS: FIBROSING ALVEOLITIS IN SYSTEMIC SCLEROSIS
There are no good epidemiologic studies that address the frequency of lung disease, but the prevalence of fibrosis is variably reported to be roughly 80–90% of patients depending on the methodology used to identify fibrotic change and the threshold of amount of disease that is used to define "fibrosis." This definition has been clearly affected by the advent of computed tomography, which allows relatively early disease to be identified noninvasively (3).
The most common histopathologic pattern of disease is nonspecific interstitial pneumonia. In the largest surgical biopsy study to be reported, Bouros and coworkers showed that 78% of a series of 80 lung biopsies had this nonspecific interstitial pneumonia pattern (4).
PREDISPOSITION TO FIBROSING ALVEOLITIS IN SSC
The etiology of SSc is unknown. It is likely a genetically complex disease in which interactions between environmental factors and variants in one or more genes result in various disease phenotypes. Autoantibody expression is the most powerful predictor of likely internal organ involvement. In this regard the anti–DNA topoisomerase II autoantibody (Scl-70) is strongly predictive of fibrosing alveolitis in systemic sclerosis (FASSc), whereas the anti-centromere autoantibody is a negative predictor of lung fibrosis but a positive predictor of lcSSc, which in turn is associated with pulmonary vascular disease and pulmonary hypertension (5).
GENETIC FACTORS
Major Histocompatibility Complex Region
Evidence that supports the role of genetic factors in SSc includes the following: reports of familial SSc, including twin studies; the fact that family members of patients with SSc have an increased likelihood of having serum autoantibodies or other rheumatologic diseases; variable distribution of clinical phenotype in different races; and major histocompatibility complex (MHC) associations with SSc and specific autoantibodies.
The significance of autoantibody expression in terms of functional consequence is unknown, but in one study of sera from 99 patients with SSc, 123 patients with other collagen vascular disease, and 30 control subjects, antifibroblast factors were found to be present in SSc and systemic lupus erythematosus (6). They were more common in SSc by 50%. All SSc antifibroblast factors reacted with anti–DNA topoisomerase II autoantibody and all bound to fibroblasts. Whether these factors are distinct or just different binding sites for the same autoantibody is unclear.
Of particular interest regarding the genetic predisposition to SSc and its clinical subsets is the population of Choctaw Native Americans in Oklahoma, in whom a higher than predicted prevalence of SSc has been found (7); it appears to be predominantly MHC associated. In one study of this population (8) a number of candidate genes were identified in a genome-wide screen as being in the proximity of markers that tracked with disease. These candidate gene regions included the MHC, SPARC (secreted protein, acidic and rich in cysteine), fibrillin-1, and topoisomerase I.
FASSc has been shown to have some genetic associations. Briggs and coworkers (8) reported the association of DR3/DRw52a or anti–Scl-70 antibodies with FASSc. Subsequently, Gilchrist and coworkers (9) reported associations between internal organ expression of disease and autoantibody carriage in a study of 202 patients with SSc and 307 matched control subjects. Anti–DNA topoisomerase II autoantibody was associated with HLA-DPB1*1301 and HLA-DRB1*11 and this autoantibody was, in turn, a strong predictor of lung fibrosis.
Non-MHC Region Candidate Genes
SPARC is a matricellular protein that modulates cell–cell and cell–extracellular matrix interactions. SPARC expression is restricted mainly to sites of tissue remodeling and wound repair and is prominent in fibrotic disorders. Zhou and coworkers reported an association between a polymorphism at position +998 of the SPARC gene and SSc across ethnic subgroups and showed that two other polymorphic loci (positions +1551 and +1922) were linked to pulmonary fibrosis (10). However, these findings could not be repeated by Lagan and coworkers, who studied eight biallelic single-nucleotide polymorphisms: three from the last untranslated exon that were previously described and an additional five novel single-nucleotide polymorphisms: two in the promoter region, one in exon 3, and two in the 3' untranslated region (11). No differences in genotype, allele or haplotype frequency were observed between SSc subgroups of patients with and without lung fibrosis.
On the basis of their previous study using microsatellite markers that flagged a haplotype on chromosome 15 that contained the fibrillin gene, Tan and coworkers studied the association of polymorphisms of the fibrillin gene, a constituent of extracellular microfibrils, in a Choctaw Indian population of patients with systemic sclerosis (12). Five single-nucleotide polymorphisms were identified and the C allele of one of these, located in the 5' untranslated region, was associated with SSc; this allele was also found in two haplotypes that appear to be unique to the Choctaw Indian SSc population. Of these, haplotype 5 was associated with lung fibrosis.
Fibronectin is a high molecular weight glycoprotein whose functions include binding to integrins and extracellular matrix components. Avila and coworkers found an association between fibronectin polymorphisms and FASSc (13).
Many other genetic studies have failed to identify strong associations with lung fibrosis.
PATHOGENETIC PATHWAYS
Ultrastructural studies have shown that endothelial and/or epithelial injury precedes inflammation and fibrosis although the initiating factor is not known (14). The downstream pathophysiologic features that follow the initial injury include immunologic and inflammatory responses together with a profibrogenic microenvironment.
IMMUNOLOGIC FEATURES
The immunologic features of FASSc include the following: CD8+ T-cell predominance in bronchoalveolar lavage studies; some oligoclonality in bronchoalveolar lavage populations; predominant helper T-cell type 2 (Th2) cytokine profile; polytypic immunoglobulins in lung and serum; and the presence in lung tissue of lymphoid follicles with germinal centers and CD4+ T cells of both Th1 and Th2 subsets that express the hallmark cytokines of these subsets in balanced numbers at the mRNA level (15–17).
FEATURES OF INFLAMMATION
In a range of studies, multiple cytokines, chemokines, and other mediators have been described in the lungs of patients with FASSc. Although it is difficult to assign dominance to any of these, and accepting that many of the inflammatory mechanisms that have been triggered are a homeostasis response to those mediator events that have already occurred, a number of these have remained as persistent candidates. These include the following: a predominant Th2 cytokine milieu; chemokine expression including monocyte chemoattractant protein-1, CXCL8 (interleukin-8), and regulated on activation, normal T-cell expressed and secreted (RANTES); and a profibrotic state in which transforming growth factor-ß (TGF-ß), connective tissue growth factor (CTGF), endothelin-1, and thrombin are notable (18–21).
FIBROGENESIS IN FASSc
The hallmark histopathologic features in both skin and lung are mesenchymal cell proliferation and connective tissue matrix deposition. A number of growth factors have been incriminated in the pathogenesis. Fibronectin plays a role in the chemotaxis and matrix attachment of fibroblasts and may contribute to the recruitment and attachment of fibroblasts in lung fibrosis. It has been shown to be up-regulated in the lungs of individuals with FASSc (22).
TGF-ß is secreted in a latent form and is activated in the local milieu after secretion. Major producers include macrophages, fibroblasts, and epithelial and endothelial cells. TGF-ß stimulates extracellular matrix synthesis (23, 24), decreases the synthesis of matrix metalloproteinases, and stimulates the production of protease inhibitors such as tissue inhibitor of metalloproteinase-1. Importantly, TGF-ß also has immunoregulatory functions. TGF-ß has been shown to be expressed in FASSc (25–28). Bronchoalveolar mononuclear cells express severalfold increased TGF-ß gene transcription products compared with controls. In immunohistochemistry studies of surgical lung biopsies, intracellular TGF-ß was demonstrated in alveolar macrophages, bronchial epithelium, and hyperplastic type II pneumocytes.
CTGF is another mediator that has consistently been shown to be expressed in FASSc (29). CTGF is synthesized by fibroblasts and lung fibroblasts from patients with FASSc, and control fibroblasts show enhanced expression of CTGF mRNA on microarray studies after TGF-ß stimulation (30). Until relatively recently, TGF-ß was thought to be the only trigger for the CTGF final common pathway of fibrosis, but it is now known that coagulation pathway components including thrombin can up-regulate CTGF (31). Thrombin levels have been found to be significantly increased in the lungs of patients with SSc compared with control subjects and up-regulate platelet-derived growth factor (PDGF) and PDGF receptor expression (31,32).
MYOFIBROBLASTS: ROLE IN FASSc
TGF-ß stimulates 
-smooth muscle actin synthesis and myofibroblast transdifferentiation (33, 34). During normal tissue remodeling, myofibroblasts undergo apoptosis. TGF-ß plays a role in perpetuating the presence of myofibroblasts at sites of lung injury by triggering the production of myofibroblasts and inhibiting myofibroblast apoptosis, resulting in a fibrotic milieu (35). Given that myofibroblast proliferation and persistence are key features of FASSc, this underscores the multiple potential roles for TGF-ß in fibrogenesis in SSc.
CONCEPTS OF BALANCE
It is arguably too simplistic, however, to assign a dominant role for one or more mediators or growth factors in isolation in the pathophysiology of any chronic disease process and it has become increasingly recognized that a number of imbalances are likely to contribute to the final histopathologic pattern (36). Quite how these evolve, sequentially or in parallel, is unknown but important processes include the relative dominance of Th1 versus Th2 cytokine expression, angiogenesis versus angiostasis, profibrotic versus antifibrotic agents, and procoagulant versus anticoagulant factors.
NOVEL CONCEPTS
In the context of the real difficulties in teasing out factors that are pivotal to the pathology of any disease from those that are present as a consequence of earlier events in the inflammatory cascade and those that are attempting to provide homeostatic control, a minimalistic approach needs to be considered. Given that it is known that epithelial (and endothelial) cell injury occurs in areas of lung before light microscopy shows any sign of inflammation, it is reasonable to conclude that the consequences of this triggering may be important even if the triggering agent is unknown. Thus, exploring the role of the epithelial cell in FASSc is merited as epithelial triggering is at the initiation of the pathogenetic process. A second focus is on fibroblasts, and gaining further insights into their origin and turnover appears to be intuitively important given their profusion in FASSc.
Epithelial Cell Function
In humans, the clearance of technetium-labeled diethylene triamine pentaacetate (99mTc-DTPA) aerosol from the lungs of patients with FASSc has been shown to be predictive of longer term outcome in this disease; more rapid clearance, indicative of a breached epithelial barrier, is associated with more rapid lung function decline subsequently (37). Furthermore, there are other data that support the concept of epithelial cell triggering in FASSc in that circulating levels of KL-6, a mucin-like glycoprotein and the product of lung epithelial cells only (38), are increased in FASSc.
In further support of the epithelial cell injury concept, Denton and coworkers have shown, in a transgenic mouse model that expresses a fibroblast-specific kinase-deficient human type II TGF-ß receptor, that all animals develop skin fibrosis but that only 20–25% will develop sporadic lung fibrosis even though housed with their littermates. Furthermore, the lung fibrosis is associated with adjacent epithelial cell injury (39).
Epithelial–Mesenchymal Transition
In addition to the mediators that the epithelial cell expresses in response to triggering, it may respond in three ways morphologically—by necrosis, apoptosis, or, as has been shown more recently, by transitioning into a mesenchymal cell (epithelial–mesenchymal transition [EMT]). It is not known what determines each of these responses but in a study by Willis and coworkers on rat type II alveolar epithelial cells, EMT was shown to occur in response to exogenous TGF-ß by demonstrating the coexpression of specific epithelial and -smooth muscle actin (40, 41). Furthermore, coexpression of epithelial cell and mesenchymal cell markers was shown by the same authors to occur in the type II cells in histopathologic samples from patients with idiopathic pulmonary fibrosis. Although this has yet to be demonstrated in FASSc, this study raises the intriguing possibility that the initiating events at the epithelium may drive the development of the repair process directly.
Fibroblast Origin
In addition to the concept of EMT, other potential routes of fibroblast accumulation in the lung include transformation of either circulating fibrocytes that are attracted to sites of injury, or of resident fibroblasts, or both, into the hallmark myofibroblast. Again the relative contribution of each of these processes is not known but in the kidney it has been estimated that the proportions are as follows: 15% fibroblasts from bone marrow, 36% from EMT, and about 50% from other sources (42).
But What about the Endothelial Cell?
Given the proximity of the endothelial cell to the epithelial cell at the alveolar level, it is not surprising that it is likely also to be injured along with the epithelial cell as the first event in what will result in lung fibrosis in systemic sclerosis, and this is what is found ultrastructurally (14). Relatively little attention has focused on this compartment but the contribution cannot be ignored, knowing that both a procoagulant and profibrotic environment can be created after endothelial cell stimulation. Particular mention must be given to one of the most potent endothelial cell products, endothelin-1, which is known to be able to transform fibroblasts into a myofibroblast -smooth muscle actin–expressing phenotype.
SUMMARY
What We Know
What We Do Not Know
What We Would Really Like to Know
FOOTNOTES
Supported by the Asmarley Trust and the Raynaud's & Scleroderma Association (UK).
Conflict of Interest Statement: R.M.d.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.
(Received in original form August 4, 2006; accepted in final form March 16, 2007)
REFERENCES
1+ 
/ T-cells in systemic sclerosis patients. Ann N Y Acad Sci 1995;756:382–391.[Abstract]-receptor. Am J Respir Cell Mol Biol 1994;10:405–412.[Abstract]-smooth muscle actin expression in granulation tissue myofibroblasts and in quiescent and growing cultured fibroblasts. J Cell Biol 1993;122:103–111.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
