Sp1 and Smad proteins cooperate to mediate transforming growth factor-beta 1-induced alpha 2(I) collagen expression in human glomerular mesangial cells.

The mechanism(s) by which Smads mediate and modulate the transforming growth factor (TGF)-beta signal transduction pathway in fibrogenesis are not well characterized. We previously showed that Smad3 promotes alpha2(I) collagen gene (COL1A2) activation in human glomerular mesangial cells, potentially contributing to glomerulosclerosis. Here, we report that Sp1 binding is necessary for TGF-beta1-induced type I collagen mRNA expression. Deletion of three Sp1 sites (GC box) between -376 and -268 or mutation of a CAGA box at -268/-260 inhibited TGF-beta1-induced alpha2(I) collagen promoter activity. TGF-beta1 inducibility was also blocked by a Smad3 dominant negative mutant. Chemical inhibition of Sp1 binding with mithramycin A, or deletion of the GC boxes, inhibited COL1A2 activation by Smad3, suggesting cooperation between Smad3 and Sp1 in the TGF-beta1 response. Electrophoretic mobility shift assay showed that Sp1 and Smads form complexes with -283/-250 promoter sequences. Coimmunoprecipitation experiments demonstrate that endogenous Sp1, Smad3, and Smad4 form complexes in mesangial cells. In a Gal4-LUC reporter assay system, Sp1 stimulated the TGF-beta1-induced transcriptional activity of Gal4-Smad3, Gal4-Smad4 (266), or both. Using the transactivation domain B of Sp1 fused to the Gal4 DNA binding domain, we show that, in our system, the transcriptional activity of this Sp1 domain is not regulated by TGF-beta1, but it becomes responsive to this factor when Smad3 is coexpressed. Finally, combined Sp1 and Smad3 overexpression induces marked ligand-independent and ligand-dependent promoter activity of COL1A2. Thus, Sp1 and Smad proteins form complexes and their synergy plays an important role in mediating TGF-beta1-induced alpha2(I) collagen expression in human mesangial cells.

TGF-beta1 activates MAP kinase in human mesangial cells: a possible role in collagen expression.

BACKGROUND: Although the pathogenic relevance of transforming growth factor-beta (TGF-beta) to glomerular sclerosis has been established, the intracellular mechanisms by which TGF-beta induces extracellular matrix accumulation are not fully understood. We examined whether the mitogen-activated protein (MAP) kinase pathway is involved in TGF-beta1-induced collagen expression by cultured human mesangial cells. METHODS: The activation of MAP kinase pathways by TGF-beta1 was assessed by immunoblot with anti-phospho-ERK or -JNK antibodies and by transfection of plasmids expressing pathway-specific transcription activators fused to the DNA-binding domain of GAL4, as well as a GAL4 response element-luciferase reporter gene. The role of MAP kinase was assessed using biochemical inhibitors and transiently expressed dominant negative mutant constructs. The effects on TGF-beta1-induced alpha1(I) collagen expression were evaluated by Northern blot and by activation of a transiently transfected alpha1(I) promoter-luciferase reporter construct. RESULTS: ERK and JNK phosphorylation occurred 30 minutes and one hour, respectively, after TGF-beta1 treatment. A biochemical blockade of the ERK pathway inhibited TGF-beta1-induced alpha1(I) collagen expression. A dominant negative mutant of ERK1 but not of JNK decreased alpha1(I) gene promoter activation. Activation of the TGF-beta-responsive p3TP-Lux construct was partially inhibited by cotransfection of an ERK1 dominant negative mutant. CONCLUSION: These data indicate that MAP kinase pathways can be activated by TGF-beta1 in mesangial cells and that the ERK MAP kinase plays a role in TGF-beta-stimulated collagen I expression. Because we have shown previously that SMADs mediate TGF-beta1-stimulated collagen I expression, our findings raise the possibility of interactions between the MAP kinase and the SMAD pathways.

The transforming growth factor-beta/SMAD signaling pathway is present and functional in human mesangial cells.

BACKGROUND: Transforming growth factor-beta (TGF-beta) signals through a unique set of intracellular proteins, called SMADs, that have been characterized mainly in transient overexpression systems. Because several models of glomerulosclerosis suggest a role for TGF-beta in the extracellular matrix accumulation, we sought to characterize the role of SMAD proteins in mediating TGF-beta1 responses in a more physiological system using nontransformed human mesangial cells. METHODS: Endogenous SMAD expression and its modulation by TGF-beta1 were evaluated by Western and Northern blot analyses. Phosphorylation of Smad2 and Smad3 was determined by both phospholabeling and immunoblot. SMAD function and its role in type I collagen transcription were investigated in cotransfection experiments using promoter-luciferase reporter gene constructs. RESULTS: Cultured human mesangial cells express Smad2, Smad3, and Smad4 proteins. TGF-beta1 down-regulated Smad3 mRNA and protein expression, respectively, after 4 and 24 hours of treatment, whereas Smad2 and Smad4 were less affected. Both Smad2 and Smad3 were phosphorylated in response to TGF-beta1 beginning at 5 minutes, with maximal phosphorylation at 15 minutes, and decreasing phosphorylation by 2 hours. Smad2/3 and Smad4 coimmunoprecipitate only after TGF-beta1 treatment. The activity of a transiently transfected, TGF-beta-responsive construct, p3TP-Lux, was stimulated 3.6-fold by TGF-beta1. Overexpressed wild-type Smad3 increased basal luciferase activity, which was further stimulated by TGF-beta1. A dominant negative mutant form of Smad3 lacking the C-terminal serine phosphoacceptor sites (Smad3A) inhibited TGF-beta1-induced luciferase activity. TGF-beta1 also increased the activation of an alpha2(I) collagen promoter-luciferase reporter construct transfected into mesangial cells. This activation was inhibited by cotransfection with the Smad3A mutant. CONCLUSIONS: Smad2, Smad3, and Smad4 are present and activated by TGF-beta1 in human mesangial cells. The SMAD pathway is functional in these cells and appears to be involved in TGF-beta1-induced type I collagen gene transcription. These findings raise the possibility that SMAD signaling plays a role in glomerular matrix accumulation.

Regulation of human mesangial cell collagen expression by transforming growth factor-beta1.

Transforming growth factor (TGF)-beta1 has been implicated in glomerular extracellular matrix accumulation. Since the spectrum and mechanism of changes in collagen turnover have not been fully characterized, we evaluated effects of TGF-beta1 on collagen expression by human mesangial cells. TGF-beta1 induced increased alpha1(I), alpha1(III), and alpha1(IV) collagen mRNA expression. Greater mRNA expression of matrix metalloproteinase (MMP)-2 was compensated by increased tissue inhibitor of metalloproteinases (TIMP)-2 mRNA. There was no change in TIMP-1 or membrane-type MMP mRNA expression, whereas MMP-1 mRNA decreased. Types I and IV collagen protein accumulated in both the cell layer and medium. Changes in collagen mRNA and protein occurred within 4 and 8 h, respectively. MMP-2 and TIMP-1 and -2 activities showed little change. Cycloheximide markedly decreased collagen detection within 4 h and reversed late, but not early, changes in alpha1(I) collagen mRNA. In this system, increased synthesis may be more significant than degradation for collagen accumulation, but collagen is short-lived in culture. Diverse TGF-beta1 actions on collagen turnover may be either immediate or mediated through synthesis of regulatory molecules.

Increased expression of extracellular matrix proteins and decreased expression of matrix proteases after serial passage of glomerular mesangial cells.

The cellular events causing pathological extracellular matrix (ECM) accumulation in vivo are not well understood. Prolonged serial passage of several cell types in culture leads to increased production of extracellular matrix (ECM) proteins, but the mechanism for these putative fibrotic changes is not known. Here, human fetal glomerular mesangial cells were subjected to serial passage (P) in culture and the expression of ECM proteins, proteases and protease inhibitors was comprehensively evaluated. From P11 through P14, a series of phenotypic changes occurred. Steady-state expression of mRNA for alpha 1 chains of type III and type IV (but not type I) collagen, and for laminin beta 1 and gamma 1, increased 2- to 8-fold, while expression of mRNA for interstitial collagenase (MMP-1) and gelatinase A (MMP-2) virtually ceased. Expression of tissue-type plasminogen activator (tPA) mRNA also decreased markedly. Expression of mRNA for the tissue inhibitor of metalloproteinases (TIMP)-1, and of the smaller of two mRNA species for the PA inhibitor PAI-1, ceased by P14. There was a switch in expression of the two species of TIMP-2 mRNA: whereas the ratio of signal intensity comparing the 3.5 kb mRNA species to the 1.0 kb species was 5:1 up to P11, it was reversed (1:5) at P14 and later. Serial passage also led to changes in protein expression, with increased type IV collagen and laminin, but decreased interstitial collagenase and gelatinase A. The cells showed a progressive increase in staining for type IV collagen. These findings define the appearance of a matrix-accumulating phenotype in later-passage mesangial cells. Matrix expansion in vivo has been associated with increased transforming growth factor (TGF)-beta synthesis; the cells were found to show at least 5-fold increased expression of TGF-beta 1 mRNA from P8 to P16. However, treatment of P9 or P10 cells with graded doses of TGF-beta 1 increased expression of both collagen IV and gelatinase A mRNA and did not alter the ratio of signal intensity for TIMP-2 mRNA species. Thus, assumption of a matrix-accumulating phenotype by these cultured fetal glomerular mesangial cells is not accelerated by exogenous TGF-beta. These data describe an in vitro model of mesangial cell matrix turnover in which matrix accumulation could result from a concerted increase in ECM synthesis and decrease in ECM degradation.

The tilapia prolactin I gene: evolutionary conservation of the regulatory elements directing pituitary-specific expression.

To study the elements involved in the pituitary specific transcriptional regulation of the tilapia prolactin I gene (tiPRL I), we have cloned and entirely sequenced a 3.4-kb genomic fragment immediately upstream from the first exon. In footprinting experiments, three tilapia sequences are protected from DNase I digestion by rat pituitary extracts (base pair coordinates -643 to -593, -160 to -111, and -73 to -46). Computer analysis of the nucleotide sequence reveals significant homology to mammalian binding sites for Pit-1, a transcription factor that is known to mediate pituitary-specific expression of the PRL genes in mammals. The tiPRL I 5'-flanking sequences can direct transient expression of a linked luciferase reporter gene in transfected rat pituitary cell lines and tilapia pituitary primary cell cultures. Transient expression experiments with 5'-deletion mutants reveal three regulatory regions. Two have a stimulatory effect on transcription and one an inhibitory effect. Electrophoretic mobility-shift assays (EMSA) demonstrate that the rat Pit-1 factor specifically binds to tilapia DNA sequences. Several such tilapia Pit-1 binding sites mediate activation of a linked heterologous promoter in transfected rat and tilapia pituitary cells. As evidenced by EMSA, a Pit-1-like protein is present in tilapia pituitary extracts. All these data point to a high conservation of the molecular mechanisms involved in pituitary-specific expression of the PRL genes in vertebrates.

Structure of the tilapia (Oreochromis mossambicus) prolactin I gene.

The tilapia (Oreochromis mossambicus) prolactin-I (PRL-I) gene has been cloned and sequenced. Its transcript (3,677 bases long) begins with a guanine and is organized in five exons and four introns like the other known prolactin genes. Analysis of the 1,555-bp 5'-flanking region suggests that pituitary-specific expression of the gene could be regulated through a trans-factor related to the mammalian pituitary-specific factor Pit-1. Two potential binding sites for such a factor were found in the first intron, suggesting a possible regulatory role for this region. Moreover, two potential Z-DNA regions are located at positions -837 to -812 and -246 to -179 from the transcription start site. These two regions could play an important role in the regulation of PRL gene expression.