It’s a Smad World: Regulation of TGF-β Signaling in the Kidney

Transforming growth factor–β (TGF-β) is generally accepted to play a significant role in renal fibrogenesis. It is present in human glomeruli under conditions associated with increased matrix accumulation (1) and is found in the glomerulus in diabetic nephropathy (2). Intrarenal infusion of the TGF-β1 gene causes glomerular sclerosis (3); conversely, antisense to TGF-β ameliorates experimental nephropathy (4). Preliminary studies have suggested that modulation of TGF-β–stimulated signal transduction has significant potential to correspondingly affect renal fibrogenesis (5). Thus, understanding how TGF-β causes fibrosis and determining how to alter this process are desirable goals for translational research.

The TGF-β family of intercellular mediators is unique among the known growth factors and cytokines because it signals through an initial serine-threonine kinase (6). In contrast, most so-called growth factors signal through a receptor tyrosine kinase. This functional difference suggests that TGF-β family ligands and receptors are evolutionarily distinct from pathways involving such other mediators as epidermal growth factor (EGF), basic fibroblast growth factor (bFGF) and platelet-derived growth factor (PDGF). Upon ligand binding, the TGF-β family receptors activate a unique signal transduction pathway that acts through the Smad family of proteins (7). These second messengers were discovered simultaneously by two communities of scientists. Using expression cloning, investigators studying Drosophila found a gene that mediates embryologic patterning regulated by a bone morphogenetic protein (BMP) homologue called decapentaplegic and termed it MAD for “mothers against decapentaplegic” (8). Those investigating the developmental genetics of Caenorhabditis elegans identified a series of proteins that, when mutated, produced an identical small phenotype in the worms (indicating a common effector pathway for these molecules). They termed this family the Sma proteins (9). Eventually, the groups combined the names, forming the Smad field.

The Smads are divided into three categories. The receptor-activated or pathway-restricted Smads (R-Smads) are activated by relatively specific TGF-β receptor ligands in mammals: Smad1, Smad5, and Smad8 for BMP; Smad2 and Smad3 for TGF-β or activin (Figure 1). In either case, the ligand binds to the type II TGF-β receptor (TβRII), which associates with and phosphorylates the type I receptor (Figure 1). The type I receptor, in turn, phosphorylates the R-Smads. This process activates the R-Smad molecules, causing them to spring open from their inactive “hairpin” configuration. The phosphorylated R-Smads associate to form a heteromultimer (more than one molecule of each Smad may be present) that includes the second type of Smad—the common-partner Smad (Co-Smad), Smad4. The complex is then translocated to the nucleus, where it can regulate target gene transcription.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Schematic diagram of transforming growth factor–β (TGF-β) signal transduction. TGF-β binds to its receptor, which dimerizes to form a heterotetramer. Subsequently, the type II receptor is activated and phosphorylates the type I receptor. The R-Smads (Smad2 and Smad3 in this case; if the ligand and receptor were of the bone morphogenetic protein [BMP] family, the R-Smads would be Smad1, Smad5, and/or Smad8) interact with the receptor and are phosphorylated. They multimerize with Smad4 and form a transcription-regulating complex in the nucleus. The I-Smad (Smad7 in this example) serves as a competitive inhibitor of Smad activity. It should be emphasized that this drawing offers a simplified representation. Additional molecules have been identified that interact with and modify the actions of this pathway.

A third category comprises of the inhibitory Smads (I-Smads). Smad6 and Smad7 are homologs of the R-Smads that bind to the TβRI, but they lack the carboxy-terminal sequences that are essential for activation. Smad6 may also inhibit BMP signaling by competing with Smad4 for Smad1 binding. Thus, the I-Smads appear to function as competitive inhibitors of Smad activation. Numerous additional proteins have been determined to interact with the Smad pathway; these have been reviewed elsewhere (7), and such interactions have been identified in mesangial cells (10).

Originally characterized in transformed cells and analyzed using expression systems, the Smad pathway recently has been characterized in more physiologic conditions. Smads are expressed in the kidney in a variety of cell types (11–13). They are activated in experimental diabetic nephropathy (14) and are stimulated by TGF-β1 to generate mesangial cell collagen production (15,16). These studies suggest that Smads play a central role in collagen accumulation. Thus, as investigators continue to characterize the regulation of this pathway, it is logical to consider how manipulation of this regulation could be applied therapeutically. However, the intracellular location of the Smads raises significant problems in targeting drugs to these proteins.

A significant step in this direction is reported in the article by Chen et al. (17) in this issue of JASN. It reports the examination of the function of the I-Smad, Smad7, in mesangial cell collagen I production. Consistent with previous observations in other systems (18), TGF-β1 treatment of a mouse mesangial cell line stimulates increased expression of Smad7 in a dose-dependent manner. The authors also studied two other indices of TGF-β stimulation, activation of the promoter for the α2 chain of type I collagen (COL1A2) and induction of spermidine uptake. Transfection of the mesangial cells with a plasmid that expresses Smad7 decreased the effects of TGF-β1. These are the first published studies to show that Smad7 activity is an effective mechanism for regulating TGF-β–stimulated collagen expression in mesangial cell fibrogenesis. In addition, the inhibition of polyamine uptake indicates that the effects of Smad7 could extend to other biologic effects of TGF-β in mesangial cells.

What are the implications of these findings? First, altered Smad7 activity could play a role in mesangial cell fibrogenesis. A question remains whether endogenous Smad7 is capable of suppressing the fibrogenic response. The authors’ findings suggest that this is not the case, because TGF-β1 stimulation of endogenous Smad7 expression is insufficient to overcome the simultaneous, stronger stimulation in favor of fibrogenesis. A likely explanation is that basal, and perhaps even stimulated, levels of Smad7 expression serve to prevent low levels of R-Smad activation from mediating TGF-β effects. By this interpretation, endogenous Smad7 expression creates a threshold of stimulation, above which a fibrogenic response occurs. Understanding how Smad7 works is further complicated by a surprising technical point in the authors’ findings. As they note, transfection of Smad7 is successful in only a small proportion of the cells, yet a profound effect is observed on the entire culture. Is there a novel mechanism of effect that is specific for transfected Smad7 and/or involving intercellular rather than intracellular communication?

The present report also suggests that Smad7 could play a role in regulating the progression of glomerular disease. Further studies clearly are needed in this area. To fully understand the cellular signals involved in glomerulosclerosis, a wider net must be cast. For example, what is the role of other fibrogenic factors (e.g., bFGF [19], PDGF [3], and connective tissue growth factor [20]) in progression? How do these interact with TGF-β? In addition, the mesangial cell represents only a small fraction of the renal cell population. What effects do various Smads have in the podocyte (18), the glomerular endothelial cell (21), and the tubulointerstitium (11)?

Finally, Chen et al. raise the possibility that, with an appropriate vector delivery system (a not insignificant caveat!), ectopic Smad7 expression is a means to ameliorate renal fibrosis. Given the uncontrolled environment of clinical disease, the timing and location of various signals will be important in addressing the therapeutic potential of manipulating these pathways. Moreover, the Smad signal transduction pathway, for all its unique characteristics, does not function in an isolated environment. There is a significant amount of “crosstalk” among intracellular pathways that might modify Smad signaling (22). As we are determining how we might be able to manipulate Smad activity by intracellular means (such as overexpression of I-Smads), we also might find therapeutic potential in the study of other intersecting signaling pathways for which pharmacologic modifiers are already available.

Thus, there are many opportunities for understanding Smad signaling and how it relates to clinical disease and therapeutics. The article by Chen et al. is an important step, but we still have much to learn.

• © 2002 American Society of Nephrology