| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Laboratory of Cell Regulation and Carcinogenesis (Drs. Roberts, Piek, Ashcroft, and Flanders) and Radiation Oncology Branch (Dr. Mitchell), National Cancer Institute, Bethesda, MD; and Department of Medicine (Dr. Böttinger), Albert Einstein College of Medicine, Bronx, NY.
Correspondence to: Anita B. Roberts, PhD, Chief, Laboratory of Cell Regulation and Carcinogenesis, National Cancer Institute, Bldg 41, Room C629, 41 Library Dr, MSC 5055, Bethesda, MD 20892-5055; e-mail: Robertsa{at}dce41.nci.nih.gov
Abstract
Transforming growth factor (TGF)-ß plays a central role in fibrosis, contributing both to the influx and activation of inflammatory cells, as well as to activation of fibroblasts to elaborate extracellular matrix. In the past few years, new insight has been gained into signal transduction pathways downstream of the TGF-ß receptor serine-threonine kinases with the identification of a family of evolutionarily conserved Smad proteins. Two receptor-activated Smad proteins, Smad2 and Smad3, are phosphorylated by the activated TGF-ß type I receptor kinase, after which they partner with the common mediator, Smad4, and are translocated to the nucleus to where they participate in transcriptional complexes to control expression of target genes. We have shown in wound healing studies of mice null for Smad3, that loss of this key signaling intermediate interferes with the chemotaxis of inflammatory cells to TGF-ß as well as with their ability to autoinduce TGF-ß. Moreover, studies with mouse embryo fibroblasts null for Smad3 show that TGF-ß–dependent induction of c-Jun and c-Fos, important in induction of collagen as well as in autoinduction of TGF-ß, is mediated by Smad3. Based on these observations, we hypothesize that loss of Smad3 will confer resistance to fibrosis and result in reduced inflammatory cell infiltrates, reduced autoinduction of TGF-ß, important to sustain the process, and reduced elaboration of collagen. Preliminary observations in a model of radiation-induced fibrosis confirm this hypothesis and suggest that inhibitors of Smad3 might have clinical application both to improve wound healing and to reduce fibrosis.
Key Words: autoinduction • fibrosis • Smad proteins • transforming growth factor-ß • wound healing
(CHEST 2001; 120:43S–47S)
Abbreviations: AP = activating protein; MAP = mitogen-activated protein; R = receptor-activated Smads; TGF = transforming growth factor
Transforming Growth Factor-ß Plays a Critical Role in Fibrotic Disease
Transforming growth factor (TGF)-ß is known to play a central role in fibrotic diseases, including cirrhosis, chronic hepatitis, glomerulonephritis, and pulmonary fibrosis, among others.1 All aspects of the disease process have been shown to be regulated by TGF-ß, including the initial inflammatory phase in which infiltrating inflammatory cells and macrophages set the stage for the subsequent fibrotic phase in which activated fibroblasts and myofibroblasts contribute to pathogenetic accumulation of matrix.2 These multifunctional effects of TGF-ß are consistent with the observation that nearly all cell types express TGF-ß receptors and are capable of secreting TGF-ß ligand.3 TGF-ß is the most potent chemotactic factor known for both macrophages and fibroblasts, suggesting that it is important not only in activation of these cells to elaborate cytokines and extracellular matrix, respectively, but also for their recruitment to sites of inflammation. Moreover, no other growth factor has as many different activities contributing to matrix accumulation as does TGF-ß, since it results not only in increased synthesis of the matrix proteins themselves, but also results in enhanced secretion of protease inhibitors and reduced secretion of proteases, so as to favor accumulation of extracellular matrix proteins.
Animal models as well as clinical data confirm the key role of TGF-ß in fibrosis. Thus, overexpression of TGF-ß driven by an albumin promoter results in elevated levels of plasma TGF-ß and leads to fibrosis in both liver and kidney.4 Induction of glomerulonephritis in rats correlates with markedly enhanced expression of TGF-ß in the glomerulus; repeated injections lead to progressive disease characterized by pathogenetic expansion of the mesangial matrix. Either antibodies to TGF-ß or stable expression of decorin or of a soluble TGF-ß receptor, each of which sequesters TGF-ß, prevents the glomerular changes.5 Intratracheal injection of bleomycin leads to pulmonary inflammation characterized by activated alveolar macrophages that secrete active TGF-ß and subsequently to TGF-ß–dependent collagen synthesis and fibrosis.6 Intranasal infusion into mice of an adenoviral construct expressing active TGF-ß also gives rise to a progressive and ultimately lethal pulmonary fibrosis.7 Clinically, TGF-ß has also been shown to be expressed at elevated levels in both epithelial cells and alveolar macrophages in patients with idiopathic pulmonary fibrosis,8 and elevated levels of plasma TGF-ß after chemotherapy in breast cancer patients have been shown to be prognostic for lethal complications involving liver and lung fibrosis.9 These multifaceted effects of TGF-ß in promoting fibrosis in a variety of tissues have led to the suggestion that therapies aimed at reducing the expression or activity of TGF-ß might be efficacious in treatment or prevention of fibrotic disease.1
Smad Proteins Mediate a Short-Circuit Signaling Pathway Downstream of TGF-ß Superfamily Receptors
TGF-ß signals through a set of transmembrane receptor serine threonine kinases unique to the larger superfamily of TGF-ß–related proteins.10 The active heterotetrameric receptor complex is formed by binding of ligand to a type II receptor, which then results in recruitment and activation of the type I receptor kinase. The last 4 years has seen an explosion of research focused on identification of the cytoplasmic signaling elements downstream of the receptors. Based on homology with Mad and Sma proteins identified by genetic screens as being downstream of key TGF-ß family members in Drosophila melanogaster and Caenorhabditis elegans, respectively, a family of mammalian orthologs of these proteins has been called Smad proteins.11 To date, eight different mammalian Smad proteins have been described that fall into three distinct groupings based both on their structure and function. Receptor-activated Smads (R-Smads) are directly phosphorylated on a C-terminal serine motif by the activated type I receptor kinase. Of the five known R-Smads, Smads 2 and 3 signal predominantly from TGF-ß and activin receptors, whereas Smads 1, 5, and 8 mediate signals predominantly from the set of bone morphogenetic proteins. Following phosphorylation, each of these R-Smads must partner with the common mediator Smad4, which lacks the C-terminal phosphorylation motif, and this R-Smad/Smad4 complex is then translocated into the nucleus where it activates transcription of target genes. Two other proteins, Smad6 and Smad7, are called inhibitory Smads, based on their ability to block Smad-dependent signal transduction. Mitogen-activated protein (MAP) kinase pathways also contribute to transduction of signals from TGF-ß.12 Data suggest complex interactions of this pathway with the Smad pathway attributed to MAP kinase-dependent phosphorylation of R-Smads in their middle linker region and with resultant synergistic and antagonistic effects, depending both on the particular MAP kinase pathway and its level of activation in particular cells.13 14
Many lines of evidence suggest that the two R-Smads important for mediating signals from TGF-ß have distinct effects on regulation of target gene activity. Smad3 binds to DNA directly by interaction of its Mad homology 1 domain (defined by its homology with the Drosophila Mad protein) with a core CAGA sequence, called the Smad binding element.15 In contrast, an insertion of 30 amino acids in the Mad homology 1 domain of Smad2 prevents its direct binding to DNA.16 Instead, Smad2 activates transcription indirectly by binding to other DNA-binding transcription factors, such as forkhead activin signal transducer 1, a member of a winged-helix family of DNA-binding proteins. Since the affinity of an R-Smad for the Smad binding element is quite low, Smad proteins are thought to partner with other DNA-binding proteins to achieve high-affinity DNA binding and the selective activation of target genes characteristic of the different TGF-ß superfamily ligands. Nevertheless, the different DNA binding characteristics of Smad2 and Smad3 seem to define sets of genes that uniquely require one or the other of these two R-Smads. Not surprisingly then, mice in which the Smad2 or Smad3 genes have been deleted by homologous recombination also have dramatically different phenotypes. Targeted deletion of the Smad2 gene results in early embryonic lethality,17 whereas mice null for Smad3 are viable, surviving 1 to 8 months, and dying eventually of opportunistic infections resulting from defects in T-cell activation and mucosal immunity.18
Study of mouse embryo fibroblasts derived from Smad2 or Smad3 null mice also show distinct patterns of gene induction by TGF-ß (Fig 1) . In particular, TGF-ß–dependent induction of components of the activating protein (AP)-1 complex, including c-fos, c-jun, and jun-B are selectively dependent on Smad3.19 20 21 Many genes contain AP-1 binding sites in their regulatory regions, including collagen aI(1),22 collagen 7A(1),23 and TGF-ß1,24 and activation of AP-1 complexes has been implicated in many normal cellular processes as well as in fibrogenesis and oncogenesis.25
|
We have used wound healing as a model in which to study effects of loss of Smad3 in vivo, since this model involves the interaction of several different cell types in the context of a specific tissue architecture and a temporally defined sequence of events after wounding. Topical application of TGF-ß improves the healing of a variety of animal models of impaired healing,26 suggesting to us that deletion of one of the downstream pathways of TGF-ß signaling might impair wound healing. To our surprise, mice null for Smad3 show accelerated healing of cutaneous incisional wounds with markedly enhanced re-epithelization and reduced influx of inflammatory cells and accumulation of matrix.27 Comparison of various parameters of healing in Smad2- and Smad3-heterozygous mice showed that these effects were specific for Smad3 and, moreover, that the effects of Smad3 were dependent on the gene dosage, with values for the heterozygous mice lying midway between those of wild-type and null littermates. The latter is consistent with the nonenzymatic nature of Smad signaling pathway resulting in signal transduction that is not amplified and therefore strictly dependent on the concentration of any particular Smad protein present in the cell. Study of primary macrophages isolated from bone marrow of Smad3-null mice, showed them to be impaired not only in their chemotactic response to TGF-ß, but also in their ability to autoinduce TGF-ß.27 Importantly, the reduced amount of matrix protein found in wounds in Smad3-null mice was not due to a primary effect on the fibroblasts, but rather on the reduced levels of TGF-ß in the wound bed assumed to result from the markedly reduced numbers of macrophages and their inability to autoinduce TGF-ß. Injection of either TGF-ß or wild-type monocytes in Smad3-null wounds elevated levels of fibronectin in the wound bed. Neither of these treatments interfered with the enhanced epithelization of the Smad3-null mice, demonstrating that loss of Smad3 abrogated the inhibitory effects of TGF-ß on this process.
Parallels Between Wound Healing and Fibrosis Suggest That Inhibition of Smad3 May Confer Resistance to Fibrosis
Many elements of fibrotic disease are shared with those of wound
healing, including the inflammatory, angiogenic, and fibrogenic
components.28
In normal wound healing, however, the
process resolves, possibly by reduction of Smad3 levels in later stages
of wound healing. Since the loss of Smad3 interfered with the effects
of TGF-ß on chemotaxis of macrophages and on
autoinduction,27
and since induction of both type I
collagen22
and type VII collagen23
and
components of the AP-1 complex have been shown to be dependent on Smad3
(Fig 1)
, we hypothesized that mice null for Smad3 would be resistant to
fibrosis. Indeed, our preliminary investigations of the effects of loss
of Smad3 on radiation-induced fibrosis suggest that the absence of
Smad3 confers protection from the epidermal hyperplasia and progressive
fibrosis resulting from a single dose of radiation to the leg of
Smad3-null mice and littermate control mice. We are currently
attempting to determine whether early signaling events implicated in
response of cells to radiation, and including activation of nuclear
factor-
B, AP-1, and MAP kinase pathways will be altered in tissues
of Smad3-null mice and whether alterations in these early response
patterns can be shown to block subsequent progressive fibrosis.
Published data29
show that irradiation of either pig skin
or human dermal fibroblasts results in rapid (6 h) elevation of the
protein levels of c-Fos and c-Jun, accompanied by increased levels of
AP-1 activity, with a shift from Jun homodimers in the basal state to
Jun-Fos heterodimers following irradiations. This could have
significance for the mechanism of protection in Smad3-null mice, since
TGF-ß–dependent induction of c-Fos in mouse embryo fibroblasts was
shown to require Smad3. Among other gene targets, this increased AP-1
activity has been shown to result in enhanced binding to a
high-affinity AP-1 binding site in the TGF-ß1 promoter
implicated in autoinduction,24
suggesting that inhibition
of Smad3 might interfere with the cascade of events leading to
progressive fibrotic sequelae (Fig 2)
.
|
These new insights into TGF-ß signaling pathways, and specifically into events regulated by the Smad3 pathway, suggest that development of Smad3 pathway inhibitors might have broad clinical application, especially for the improvement of wound healing and for inhibition of fibrosis. To establish the proof of principle, it will be necessary to test the effects of blocking Smad3 production in wild-type cells, possibly by use of an antisense approach. Ultimately, one would like to design a cell-permeable, small-molecule inhibitor either by combinatorial chemistry or possibly by high-throughput screening of natural product libraries. In fibrosis, this could be especially effective in situations in which the outcome is predictable, such as following radiation therapy or possibly in patients with high plasma levels of TGF-ß known to result in lethal fibrotic complications.9 This type of approach leads us into the new arena of molecular medicine where our ability to identify specific pathways draws us away from empiricism toward rational drug design based on understanding of the molecular mechanisms of disease pathogenesis.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
