The cause of systemic sclerosis is unknown. Vascular injury (manifested by cutaneous ulceration and telangiectasia) and altered cellular and humoral immunity (manifested by highly specific antinuclear autoantibodies in the serum and mononuclear inflammatory cell infiltrates in many organs) are characteristics of the disease. In later stages, diffuse tissue fibrosis becomes prominent, and excessive collagen in affected organs progressively impairs their function [1]. The accumulation of collagen in affected tissues results primarily from overproduction of the substance by resident stromal fibroblasts. This has been shown by in vitro studies of collagen biosynthesis in cultured fibroblasts from affected skin, immunohistochemical staining of affected tissues, and in situ hybridization studies with cloned complementary DNA [2-4]. These studies showed increased expression of type I, III, VI, and VII collagen genes in affected skin from patients with systemic sclerosis compared with unaffected skin from patients with systemic sclerosis or skin from healthy persons. Normal fibroblasts can modulate their collagen production in response to the dynamic requirements of the extracellular matrix during development, differentiation, and repair. Changes in the complex regulatory mechanisms that determine the net amount of collagen produced may account for overproduction in systemic sclerosis fibroblasts. Recent evidence indicates that gene transcription is a major point for regulating production of collagen in particular cell types or in response to particular signals [5]. Consequently, there is great interest in understanding how the transcriptional activity of collagen genes is regulated.

The expression of eukaryotic genes is controlled primarily by short DNA sequences in a region of the gene called the promoter, located immediately 5' from the start site of transcription. The promoter region binds RNA polymerase II, the enzyme responsible for transcribing protein-coding genes. Many eukaryotic genes share common DNA sequences (motifs) in their promoters. Mutations within these motifs dramatically reduce the level of the transcription of the corresponding genes. Additional sequences important for transcription, called enhancers, may be located 5' from the promoters, beyond the 3' end of the genes, or within the first intron. Compared with conventional promoter sequences, enhancers are larger and may have several distinct motifs. Both promoter and enhancer elements contain protein-binding sites that are specifically recognized by cellular transcription factors. A complex interaction of transcription factors binding to enhancers and to promoters results in modulation of transcriptional activity (Figure 1).

View larger version (29K): [in this window] [in a new window]   Figure 1. A simplified view of the regulation of collagen gene transcription, the process that leads to production of collagen from the initial template of the collagen gene. Transcription of the collagen gene is initiated by binding of the transcription initiation complex, which contains RNA polymerase II and various initiation factors, to the DNA (upper panel). This basal level of transcription is markedly stimulated by transforming growth factor-ß, which may cause binding of transcriptional regulatory factors (shaded ovals) to upstream promoter elements and to upstream or intronic enhancer DNA sequences. Binding of the transcriptional factors may result in looping of DNA, allowing distant transcriptional regulatory factors to contact the transcription initiation complex (lower panel).

Analysis of the genes coding for the 1 and 2 chains of type I collagen, the main structural component of the extracellular matrix, showed characteristic CCAAT nucleotide sequences within 100 base pairs 5' from the start site of transcription. These DNA sequences are recognition sites for CCAAT-binding factor, a recently characterized heteromeric protein that can stimulate transcription of the type I collagen genes in vitro [6]. Additional cis elements that positively or negatively affect transcription have been identified in the upstream region, as well as within the first intron of several collagen genes, and appear to be potential recognition sites for as-yet unidentified trans-acting factors. Transcriptional activity of collagen genes may be regulated not only by the interaction of cis elements and trans-acting DNA-binding factors but also by the chromatin structure and the methylation status of the genes [7].

In the skin, resident fibroblasts are potential targets for extracellular signaling molecules that can modulate how collagen genes are expressed. Important physiologic modulators include hormones such as glucocorticoids and various cytokines. Because the development of fibrosis in systemic sclerosis and other fibrosing disorders appears to be preceded by infiltration of affected tissues with mononuclear cells (primarily macrophages and T lymphocytes), inflammatory cell-derived cytokines are candidate-signaling molecules that account for the up-regulation of fibroblast collagen production in these disorders. Of the many cytokines that have been shown to influence collagen production in vitro, transforming growth factor-ß is emerging as crucial in tissue repair and fibrosis. This multifunctional cytokine, which has three structurally and functionally similar isoforms, is related to a large superfamily of homologous proteins whose members mediate key events in growth and development [8]. Transforming growth factor-ß 1, the best-studied isoform, is a homodimer of two 112-amino acid polypeptides derived from the proteolytic cleavage of larger inactive precursor molecules.

Recent studies have shed light on the biological activities of transforming growth factor-ß involved in connective tissue turnover and the mechanisms of these effects. For example, it has been shown in vitro that transforming growth factor-ß stimulates human fibroblasts to produce type I collagen [9, 10]. This response appears to involve the induction of collagen gene promoter activity and is associated with stimulation of DNA binding by nuclear proteins in the target fibroblasts [11, 12]. The identity and function of the collagen promoter-binding proteins activated by transforming growth factor-ß are still unknown. Because transforming growth factor-ß causes both sustained stimulation of collagen production [10] and autoinduction of its own synthesis [13], a brief exposure of fibroblasts to transforming growth factor-ß may result in persistent alteration in their biosynthetic phenotype. Exogenous administration of transforming growth factor-ß in vivo has been shown to accelerate the healing of incision wounds [14], and endogenous transforming growth factor-ß (produced by targeted gene transfer) has been shown to cause glomerulosclerosis when expressed in the kidney [15] and medial hyperplasia when expressed in arteries [16]. On the other hand, antagonists of transforming growth factor-ß prevent fibrosis. For instance, neutralizing anti-transforming growth factor-ß antibody inhibited scar formation in healing dermal wounds [17] and prevented the development of carotid intimal hyperplasia after balloon angioplasty [18].

Currently, only circumstantial evidence implicates transforming growth factor-ß in the formation of tissue fibrosis in systemic sclerosis. Transforming growth factor-ß is abundant in platelets and inflammatory cells, cell types that have been shown to be activated in early systemic sclerosis. Expression of transforming growth factor-ß has been shown by immunohistochemical methods and by in situ hybridization in the skin of patients with early systemic sclerosis but not in that of healthy controls [4, 19-21]. The expression of transforming growth factor-ß in tissues before excessive collagen accumulation suggests a role for transforming growth factor-ß in the development of fibrosis in systemic sclerosis, but the lack of an animal model for this disease currently precludes a direct definition of the role of transforming growth factor-ß. Furthermore, it must be pointed out that fibroblasts in tissues undergoing inflammation and remodeling are probably sequentially or simultaneously exposed to numerous cytokines. These cytokines may interact with each other and have been shown to modulate target cell responsiveness to other cytokines by various mechanisms. It is therefore the total cytokine context that determines how a given cell responds to a specific signal [22].

In light of the rapid progress being made in understanding the biology of physiologic and pathologic tissue repair processes, the importance of delineating the molecular mechanisms of fibroblast signaling and regulation of collagen gene expression in systemic sclerosis has become clear. How receptor and ligand interact, how the intracellular pathways are activated after signaling, and how the DNA-binding nuclear transcriptional factors are induced by transforming growth factor-ß and other cytokines may soon be determined. Each of these steps is a potential target for highly specific and perhaps nontoxic therapy. Novel antifibrotic strategies will involve agents that directly inhibit fibroblast collagen production [23], down-regulate the production [24] and decrease the availability [25] of fibrogenic cytokines, or interfere with signaling and transcriptional activation by these cytokines. Molecular genetic approaches targeted at modulation of the increased transcriptional activity of collagen genes (for example, with the antimitotic agent mithramycin [26; unpublished observations]) may provide potent therapies for systemic sclerosis and other diseases that are accompanied by pathologic tissue fibrosis.