当前位置: 首页 > 医学版 > 期刊论文 > 内科学 > 内分泌学杂志 > 2005年 > 第3期 > 正文
编号:11168235
Gene Expression Profiling of Leiomyoma and Myometrial Smooth Muscle Cells in Response to Transforming Growth Factor-?
     Department of Obstetrics and Gynecology, University of Florida College of Medicine, Gainesville, Florida 32610

    Address all correspondence and requests for reprints to: Dr. Nasser Chegini, Department of Obstetrics and Gynecology, University of Florida, Box 100294, Gainesville, Florida 32610. E-mail: cheginin@obgyn.ufl.edu.

    Abstract

    Altered expression of the TGF-? system is recognized to play a central role in various fibrotic disorders, including leiomyoma. In this study we performed microarray analysis to characterize the gene expression profile of leiomyoma and matched myometrial smooth muscle cells (LSMC and MSMC, respectively) in response to the time-dependent action of TGF-? and, after pretreatment with TGF-? type II receptor (TGF-?RII) antisense oligomer-blocking/reducing TGF-? autocrine/paracrine actions. Unsupervised and supervised assessments of the gene expression values with a false discovery rate selected at P 0.001 identified 310 genes as differentially expressed and regulated in LSMC and MSMC in a cell- and time-dependent manner by TGF-?. Pretreatment with TGF-?RII antisense resulted in changes in the expression of many of the 310 genes regulated by TGF-?, with 54 genes displaying a response to TGF-? treatment. Comparative analysis of the gene expression profile in TGF-?RII antisense- and GnRH analog-treated cells indicated that these treatments target the expression of 222 genes in a cell-specific manner. Gene ontology assigned these genes functions as cell cycle regulators, transcription factors, signal transducers, tissue turnover, and apoptosis. We validated the expression and TGF-? time-dependent regulation of IL-11, TGF-?-induced factor, TGF-?-inducible early gene response, early growth response 3, CITED2 (cAMP response element binding protein-binding protein/p300-interacting transactivator with ED-rich tail), Nur77, Runx1, Runx2, p27, p57, growth arrest-specific 1, and G protein-coupled receptor kinase 5 in LSMC and MSMC using real-time PCR. Together, the results provide the first comprehensive assessment of the LSMC and MSMC molecular environment targeted by autocrine/paracrine action of TGF-?, highlighting potential involvement of specific genes whose products may influence the outcome of leiomyoma growth and fibrotic characteristics by regulating inflammatory response, cell growth, apoptosis, and tissue remodeling.

    Introduction

    TGF-? IS A multifunctional cytokine and key regulator of cell growth and differentiation, inflammation, apoptosis, and tissue remodeling (1, 2, 3, 4, 5). Although under normal physiological conditions, the expression and autocrine/paracrine actions of TGF-? are highly regulated, alterations in TGF-? and TGF-? receptor expression and their signaling mechanisms often result in various pathological disorders, including fibrosis (1, 2, 3, 4, 5). Leiomyoma is a benign uterine tumor characterized by features typical of fibrotic disorder. We have previously identified altered expression of TGF-? isoforms (TGF-?1, -?2, and -?3) and TGF-? receptors (types I, II, and III) in leiomyoma and their isolated smooth muscle cells (LSMC) compared with normal myometrium (6, 7, 8, 9). Recently, we also demonstrated that leiomyoma and LSMC express elevated levels of Smads, components of the TGF-? receptor signaling pathway, compared with myometrium and myometrial smooth muscle cells (MSMC) (9, 10). TGF-? regulates its own expression and the expression of Smad in LSMC and MSMC, and through downstream signaling from this and MAPK pathways regulates the expression of c-Fos, c-Jun, fibronectin, collagen, and plasminogen activator inhibitor 1 in these cells (7, 8, 11). Additionally, data from our laboratory and others have demonstrated the ability of TGF-? to regulate LSMC and MSMC cell growth (12, 13, 14, 15).

    Because leiomyoma growth is dependent on ovarian steroids, GnRH analog (GnRHa) therapy and, most recently, selective estrogen and progesterone receptor modulators have been used for their medical management. We demonstrated that GnRHa therapy markedly down-regulates TGF-? and TGF-? receptor expression and alters the expression and activation of Smads in leiomyoma as well as LSMC (6, 8, 9). We have also shown that TGF-? expression in LSMC and MSMC is inversely regulated by ovarian steroids compared with their antagonists, ICI-182780, ZK98299, and RU486 (8). In addition, we have shown that other cytokines, such as granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-13, and IL-15, which promote myofibroblast transition, granulation tissue formation, and inflammatory response, respectively, may mediate their actions either directly or indirectly through induction of TGF-? expression in LSMC and MSMC (7, 9, 16). From these observations we proposed that the TGF-? system serves as a major autocrine/paracrine regulator of fibrosis in leiomyoma (6, 7, 8, 9, 10, 11, 12, 17). We have provided evidence reflecting the molecular environments directed by GnRHa therapy in leiomyoma and myometrium as well as by GnRHa direct action in LSMC and MSMC (18, 19). In the present study we evaluated the underlying differences between molecular responses directed by TGF-? autocrine/paracrine actions in LSMC and MSMC and after interference with these actions using TGF-? type II receptor (TGF-?RII) antisense oligomer treatment. Because TGF-? receptor expression is targeted by GnRHa in leiomyoma and myometrium, we also evaluated the gene expression profiles in response to TGF-?RII antisense treatment and GnRHa-treated LSMC and MSMC to identify the genes whose expression are the specific targets of these treatments. Using this approach we identified several differentially expressed and regulated genes targeted by TGF-? and validated the expression of 12 genes in LSMC and MSMC in response to the time-dependent action of TGF-? using real-time PCR.

    Materials and Methods

    All materials used for this study, including isolation of leiomyoma and myometrial cells, are identical and were previously described in detail (19). Prior approval was obtained from the University of Florida institutional review board for the experimental protocol of this study.

    To determine the effect of TGF-?1 on global gene expression in LSMC and MSMC, the cells were cultured in six-well plates at an approximate density of 106 cells/well in DMEM-supplemented medium containing 10% fetal bovine serum. After reaching visual confluence, often after 2–3 d, the cells were washed in serum-free medium and incubated for 24 h under serum-free, phenol red-free conditions (10, 11). The cells were then treated with 2.5 ng/ml TGF-?1 (R&D System, Inc., Minneapolis, MN) for 2, 6, and 12 h. To further profile the autocrine/paracrine action of TGF-?1 on gene expression in LSMC and MSMC, the cells were cultured as described above and treated with 1 μM TGF-?RII antisense or sense oligonucleotides for 24 h as previously described (10, 11). The cells were washed and then treated with TGF-?1 (2.5 ng/ml) for 2 h. Parallel experiments using untreated cells were used as controls, including an additional control for TGF-?RII antisense and sense experiments.

    Total cellular RNA was isolated from LSMC- and MSMC-treated and untreated controls and subjected to microarray analysis; a detailed description of all procedures was previously provided (19). To maintain standards and allow for comparative analysis, GeneChips in this study were used and simultaneously processed, and their gene expression values were subjected to global normalization and transformation with the GeneChips used in the other study (19). After these unsupervised assessments, the coefficient of variation was calculated for each probe set across all chips used in this and the other study (19), and the selected gene expression values of this study were independently subjected to supervised learning, including statistical analysis in R programming environment and ANOVA with false discovery rate selected at P 0.001 (19). The genes identified in these cohorts were analyzed for functional annotation and visualized using Database for Annotation, Visualization, and Integrated Discovery (DAVID) software with integrated GoCharts as described in detail previously (19). After the analysis, we selected 12 of the differentially expressed and regulated genes, including 10 identified and validated in leiomyoma and myometrium from untreated and GnRHa-treated cohorts as well as LSMC and MSMC treated in vitro with GnRHa (19), for validation in response to TGF-? time-dependent action using real-time PCR. They include IL-11, EGR3 (early growth response 3), TIEG (TGF-?-inducible early gene response), TGIF (TGF-?-induced factor), CITED2 (cAMP response element binding protein-binding protein/p300-interacting transactivator with ED-rich tail), Nur77, p27, p57, Gas-1 (growth arrest-specific 1), and GPRK5 (G protein-coupled receptor kinase). In addition, the expression of Runx1 and Runx2, transcription factors that interact with TGF-? receptor signaling pathways (20), was validated in LSMC and MSMC in response to TGF-? and GnRHa action as well as in leiomyoma and myometrium from GnRHa-treated and untreated cohorts. Detail descriptions of the materials and methods for real-time PCR as well as data analysis were provided in the previous study (11, 19).

    Results

    Gene expression profiles of leiomyoma and matched myometrium cells in response to TGF-?1

    In this study we performed microarray analysis to further characterize the molecular environment of LSMC and MSMC directed by TGF-? autocrine/paracrine actions. Using the same cell preparations and culture conditions as those described to study GnRHa action (19), LSMC and MSMC were treated with TGF-?1 (2.5 ng/ml) for 2, 6, and 12 h, and total RNA was isolated and subjected to microarray analysis. After unsupervised and supervised learning of array data (19), the gene expression values for this study were independently subjected to statistical analysis in R programming and ANOVA. At a false discovery rate selected at P 0.001, we identified 310 genes, or 2.46% of the genes on the array, as differentially expressed and regulated in response to the time-dependent action of TGF-? in LSMC and MSMC. As illustrated in Fig. 1, hierarchical clustering analysis separated these genes into distinctive clusters with sufficient difference in their patterns to allow each cohort to cluster into their respective subgroup. The genes were separated into five clusters in response to the time-dependent action of TGF-? in LSMC and MSMC, with genes in clusters A and B displaying a late response, genes in cluster D displaying an early response, and genes in clusters C and E showing biphasic regulatory behaviors (Fig. 1). Additional analysis of the variance-normalized mean gene expression values divided the genes into six clusters, each displaying a different level of response to time-dependent action of TGF-?, with overlapping behavior between LSMC and MSMC with the exception of genes in clusters E and F (Fig. 2).

    FIG. 1. Hierarchical clustering analysis of 310 genes identified in LSMC (f) and MSMC (m) in response to TGF-?1 (2.5 ng/ml) treatment for 2, 6, and 12 h or in the untreated control (C). The genes were identified after unsupervised and supervised analyses of the expression values, statistical analysis in the R programming environment, and ANOVA with a false discovery rate selected at P 0.001. Each column represents data from a single time point using two independent cell cultures, with shades of red and green indicating up- or down-regulation of a given gene according to the color scheme shown below. Genes represented by rows were clustered according to their similarities in expression patterns for each treatment and cell type. The dendrogram showing similarity of gene expression among the treatments/cells is shown on top of the overview image, and relatedness of the arrays is denoted by the distance to the node linking the arrays. The gene tree shown at the left of the image corresponds to the degree of similarity (Pearson correlation) of the pattern of expression for genes across the experiments. The clustering depicts five groups of genes, designated A–E, and their zoomed images are presented in A–E. Genes that appear more than once are represented by multiple clones on arrays. (Figure continues on next page.)

    FIG. 2. A k-means clustering analysis of genes in LSMC (f) and MSMC (m) in response to time-dependent action of TGF-?, as described in Fig. 1. The gene expression values in these cohorts were combined and subjected to k-means clustering that grouped the genes into six clusters (A–F) based on similarity of expression over the three time points and an untreated control. The rows represent the genes, and the columns represent the samples, with shades of red and green indicating up- or down-regulation of a given gene; genes are clustered according to their similarities in expression patterns. The line graphs display the SD from the mean (y-axis) for each cluster in MSMC and LSMC in response to TGF-? time-dependent action for 2, 6, and 12 h (x-axis) compared with the untreated control (Ctrl).

    Comparative analysis of gene expression profiles of LSMC and MSMC in response to TGF-? action with their corresponding leiomyoma and myometrium (tissues) from the untreated group (19) revealed a substantial variability among their profiles (data not shown). However, gene ontology assessment and division into functional categories indicated that the majority of the genes (60–70%) are involved in transcriptional regulation and metabolism, cell cycle regulation, extracellular matrix and adhesion molecules, and signal transduction and transcription factors (19) (Fig. 3). The time-dependent action of TGF-? on the expression profile of a selective group of these genes in the above clusters representing transcription factors, growth factors, cytokines, signal transduction pathways, extracellular matrix (ECM)/adhesion molecules, etc., in LSMC and MSMC are presented in Fig. 4. As illustrated, the expressions of these genes are regulated by TGF-?1 in LSMC and MSMC, displaying both overlapping and differential patterns of expression. The profile of IL-11 was not included in the growth factor/cytokine group in LSMC and MSMC, because the relative expression values were too high for graphic presentation (see Fig. 7).

    FIG. 3. Gene ontology assessment and division of genes identified in LSMC and MSMC in response to TGF-? treatment into similar functional categories illustrated as bar graphs, with the percentage of the total number of genes in each group shown in the front of each bar. A and B, Gene ontology assessment for LSMC and MSMC treated with TGF-? (A) and pretreatment with TGF-?RII antisense (B) for 24 h, followed by TGF-? treatment as indicated in Materials and Methods.

    FIG. 4. The expression profile of a group of genes representing transcription factors (row 1), growth factors/cytokines/polypeptide hormones/receptors (row 3), intracellular signal transduction pathways (rows 3 and 4), cell cycle (row 5), oncogenes/tumor suppressers (row 6), and cell adhesion/ECM/cytoskeletons (row 7) in response to the time-dependent action of TGF-? in LSMC and MSMC. Values on the y-axis represent an arbitrary unit derived from the mean gene expression value for each factor after supervised analysis, statistical analysis in R programming environment, and ANOVA as described in Fig. 1, with gene expression values for the untreated controls (Ctrl) set at 1. (Figure continues on next page.)

    FIG. 7. Comparative analysis of the expression profile of 12 genes identified as differentially expressed and regulated in response to time-dependent action of TGF-?1 in LSMC and matched MSMC by microarray and real-time PCR. Values on the y-axis represent an arbitrary unit derived from the mean expression value for each gene, and values on the x- axis represent the time course of TGF-? (2.5 ng/ml) treatment (2, 6, and 12 h), with untreated control (Crtl) gene expression values set at 1. Total RNA isolated from these cells was used for both microarray analysis and real-time PCR, validating the expression of IL-11, EGR3, CITED2, Nur77, TIEG, TGIF, p27, p57, Gas-1, and GPRK5.

    Gene expression profiles of LSMC and MSMC in response to TGF-? after pretreatment with TGF-?RII antisense

    To further evaluate the autocrine/paracrine action of TGF-? in leiomyoma and myometrial microenvironments, LSMC and MSMC were pretreated with TGF-?RII antisense oligomers to block/reduce TGF-? receptor signaling. After pretreatments, the cells were treated with or without TGF-? for 2 h, and their total RNA was subjected to microarray analysis. Based on the same data analysis described above with a false discovery rate of P 0.001, we identified 54 genes whose expression was targeted by TGF-?1 (2.5 ng/ml for 2 h) in LSMC and MSMC after pretreatment with TGF-?RII antisense compared with the 310 genes regulated by TGF-?. Interestingly, pretreatment with TGF-?RII antisense blocked or altered the expression of many genes known to be the target of TGF-? action, including those validated in our study. Hierarchical cluster analysis distinctively separated these genes into three clusters, with each cohort separated into their respective subgroups (Fig. 5). The genes in clusters A and C displayed different responses to TGF-?RII antisense pretreatment, whereas genes in cluster B showed overlapping behavior in LSMC and MSMC (Fig. 5). However, there was an overlapping pattern between the gene expression profiles in TGF-?RII sense- and antisense-treated cells that could be due to the inability of antisense pretreatment to block all of the combined actions of autocrine/paracrine and exogenously added TGF-?. Gene ontology and division into similar functional categories indicated that the majority of these genes are involved in transcriptional regulation and metabolism, cell cycle regulation, extracellular matrix and adhesion molecules, and transcription factors (Figs. 3 and 4).

    FIG. 5. Hierarchical clustering analysis of gene expression values in untreated and TGF-?-treated LSMC and MSMC after pretreatment with TGF-?RII antisense or sense oligomers. The cells were cultured in serum-free, phenol-red free medium for 24 h, washed, and treated with TGF-?RII antisense or sense oligomers for an additional 24 h. The cells were then washed and treated with TGF-? (2.5 ng/ml) for 2 h, with untreated cells serving as controls. Supervised analysis of the gene expression values and statistical analysis in R programming and ANOVA identified 54 genes at a false discovery rate of rate of P 0.001, with expression levels discriminated among the treatment groups and the untreated control. Each column represents data from a single time point using two independent cell cultures (f314 and f316 for LSMC and m314 and m316 for MSMC), with shades of red and green indicating up- or down-regulation of a given gene according to the color scheme shown below. Genes represented by rows were clustered according to their similarities in expression patterns for each treatment and cell type. The dendrogram showing similarity of gene expression among the treatments/cells is shown on top of the overview image, and relatedness of the arrays is denoted by the distance to the node linking the arrays. The gene tree shown at the left of the image corresponds to the degree of similarity (Pearson correlation) of the pattern of expression for genes across the experiments. The clustering depicts three groups of genes (A–C), and their zoomed images are presented in A–C. Genes appearing more than once are represented by multiple clones on arrays.

    Comparative analysis of gene expression profiles in response to TGF-?RII antisense and GnRHa treatments in LSMC and MSMC

    Because GnRHa alters the expression of TGF-? and TGF-? receptors in leiomyoma and myometrium as well as in LSMC and MSMC, we compared the gene expression profile of TGF-?RII antisense-treated vs. GnRHa-treated LSMC and MSMC, searching for common genes whose expressions are affected by these treatments. Based on the same data analysis described above with a false discovery rate selected at P 0.001, we identified 222 genes as differentially expressed and regulated in LSMC and MSMC in response to TGF-?RII antisense- and GnRHa-treated cells (19) (Tables 1 and 2). Hierarchical clustering analysis separated these genes into four clusters displaying different patterns of regulation, allowing their separation into respective subgroups (Fig. 6). The genes in clusters A, B, and D displayed different responses to TGF-?RII antisense and GnRHa treatments, with genes in cluster C showing overlapping behavior in LSMC and MSMC (Fig. 6).

    TABLE 1. Categorical list of differentially expressed genes in LSMC treated with GnRHa (2–12 h) compared with TGF-? type II receptor antisense-pretreated cells

    TABLE 1A. Continued

    TABLE 1B. Continued

    TABLE 2. Categorical list of differentially expressed genes in MSMC treated with GnRHa (2–12 h) compared with TGF-? type II receptor antisense-pretreated cells (continued on next page)

    TABLE 2A. Continued

    FIG. 6. Hierarchical clustering analysis of gene expression in LSMC and MSMC pretreated with TGF-?RII antisense for 24 h, followed by TGF-? treatment for 2 h (f 314, f316, m314, and m316 antisense), GnRHa-treated cells for 2 h (f-314G, f316G, m314G, and m316G), and untreated control (C). Supervised analysis of the gene expression values and statistical analysis in R programming and ANOVA identified 222 genes with a false discovery rate of rate of P 0.001, whose expression levels discriminated among the treatment groups and the untreated control. The clustering depicts four groups of genes (A–D), and their zoomed images are presented in A–D. Genes appearing more than once are represented by multiple clones on arrays.

    Verification of gene transcripts in TGF-?-treated LSMC and MSMC

    Using real-time PCR, we validated the expression of 12 genes in response to time-dependent action of TGF-? in LSMC and MSMC (Figs. 7 and 8). They include IL-11, TIEG, TGIF, EGR3, CITED2, Nur77, p27, p57, GAS-1, and GPRK5, whose expression was also validated in leiomyoma and matched myometrium from untreated and GnRHa-treated cohorts as well as in LSMC and MSMC treated in vitro with GnRHa as previously described (19). In addition, we verified the expression of Runx1 and Runx2 in leiomyoma and myometrium as well as in LSMC and MSMC in response to time-dependent actions of TGF-? and GnRHa (Fig. 8). As illustrated, TGF-?, in a time-dependent manner, differentially regulated the expression of these genes in LSMC and MSMC, with a pattern of expression displaying significant overlap between real-time PCR and microarray analysis (Figs. 7 and 8). However, the expression values of GPRK5 and Runx2 genes in microarray analysis of LSMC and MSMC and of Runx2 in leiomyoma and myometrium did not meet the standard of analysis used in our study. However, Runx2 mRNA was detectable by real-time PCR in leiomyoma and myometrium from untreated and GnRHa-treated tissues at low levels compared with Runx1, and GnRHa therapy seemed to increase its expression in these tissues (Fig. 8). The expression of Runx1 and Runx2 not only was the target of TGF-? regulatory action, but was also regulated by GnRHa in a cell- and time-dependent manner in LSMC and MSMC in vitro (Fig. 8).

    FIG. 8. Comparative analysis of the gene expression profiles of Runx1 and Runx2 in leiomyoma (LM) and matched myometrium (MM) from untreated subjects (un-Trt) and women who received GnRHa therapy (GnRHa-Trt) as well as in LSMC and MSMC in response to the time-dependent action (2, 6, and 12 h) of GnRHa (0.1 μM) as described in detail previously (19 ) and in response to the time-dependent (2, 6, and 12 h) action of TGF-?1 (2.5 ng/ml) determined by real-time PCR. In microarray analysis, Runx2 expression was not included because its expression value did not reach the study standard. Values on the y-axis represent an arbitrary unit derived from the mean expression value for each gene, and values on the x-axis represent the time course of TGF-? and GnRHa treatments, with untreated control (Crtl) gene expression values set at 1. Total RNA isolated from these cells was used for both microarray analysis and real-time PCR validation.

    We verified the expression of IL-11, TIEG, TGIF, EGR3, CITED2, Nur77, p27, p57, and Gas-1 by Western blotting and their cellular distribution using immunohistochemistry in leiomyoma and myometrium (19). These findings provide additional support for real-time PCR data showing that the products of these genes are expressed in leiomyoma and myometrium. We are currently investigating time- and dose-dependent regulation of these genes in response to TGF-?.

    Discussion

    By extending our previous work on the role of TGF-? in leiomyoma, in the present study we provided the first example of gene expression fingerprints of LSMC and MSMC in response to TGF-? action. We also characterized the molecular environment of these cells after pretreatment with TGF-? type IIR (TGF-?RII) antisense as a tool to interfere with the autocrine/paracrine action of TGF-? isoforms, and comparatively assessed their expression profiles with GnRHa-treated cells, which also inhibits TGF-? receptor expression in these cells (6, 8). Since the aim of our study was to capture the early and late autocrine/paracrine action of TGF-? in these cells, we selected a treatment strategy based on our previous observations reflecting TGF-? time-dependent regulation of c-Fos, c-Jun, fibronectin, collagen, and plasminogen activator inhibitor-1 expression (11). Promoters of these genes are known to contain TGF-? response elements (20, 21) and are regulated in LSMC and MSMC through TGF-? receptor activation of Smad and MAPK pathways (3, 10, 11). Our study design is also consistent with other microarray studies profiling gene expression in response to TGF-? action in human dermal fibroblasts, the HaCaT keratinocyte cell line, and NMuMG, a mouse mammary gland epithelial cell line, in which the cells were treated for 1, 2, 6, and 24 h, displaying a Smad-mediated regulation of selected number of genes (22, 23, 24).

    Cluster and Tree-View analysis revealed a considerable similarity in overall gene expression patterns between LSMC and MSMC in response to TGF-? action, with sufficient difference allowing their separation into respective subgroups. The genes in these clusters displayed different regulatory response to TGF-? action in a cell- and time-specific manner, with genes in clusters A and B displaying a late response, genes in cluster D displaying early responsiveness, and genes in clusters C and E showing a biphasic regulatory behavior. These results suggest that the same factors and/or mechanisms coregulate the expression of these genes in each cluster, possibly due to the presence of common regulatory elements in their promoters. Whether the expression profile of these genes in LSMC and MSMC respond differently to varying concentrations of TGF-? or other TGF-? isoforms is not established. However, the concentration of TGF-? used in this and our other studies as well as those reported by others examining the effect of TGF-? on the expression of other genes (10, 11, 12, 13, 22, 23, 24) is comparable with level of TGF-? produced by these cells, although LSMC produces more TGF-?1 than MSMC (7, 8). Moreover, based on the expression profile of TGF-? isoforms in leiomyoma, we have previously proposed that TGF-?1 and TGF-?3 play more critical roles in leiomyoma (7), and in vitro studies have indicated a higher growth response to TGF-?1 (Chegini, N., unpublished observations) and TGF-?3 in LSMC compared with MSMC (14, 15). However, TGF-? isoforms mediate their actions through TGF-?RII, and alterations in the TGF-? receptor system may serve as a more accurate indicator of their overall actions in these and other cell types. We have shown that leiomyoma overexpresses TGF-?RII compared with myometrium (6, 9), and pretreatment of LSMC with TGF-?RII antisense oligomers and/or neutralizing antibodies prevented TGF-? receptor-mediated actions (8, 10).

    These observations as well as identification of specific genes whose expression exhibited sensitivity to pretreatment with TGF-?RII antisense, among them genes containing TGF-? regulatory response elements in their promoters, provide additional support for the idea that TGF-? receptors mediated signaling in regulating the overall expression of these genes in LSMC and MSMC and possibly in leiomyoma and myometrium. The lack of response of other TGF-?-targeted genes to TGF-?RII antisense pretreatment could be due to the inability of antisense to block all autocrine/paracrine actions as well as exogenously added TGF-?. However, the expression of these genes may also be regulated as a consequence of overexpression of TGF-? receptor and/or their altered intracellular signaling. Interestingly, activin receptor-like kinases (ALK) ALK1 and ALK5, which serve as TGF-?RI and are activated by TGF-?RII, have been shown to regulate the expressions of different genes in endothelial cells in response to TGF-? action (25). We have identified the expression of all components of the TGF-? receptor system, including ALK5 and Smads in leiomyoma and myometrium as well as LSMC and MSMC (9, 10). However, TGF-?-mediated action through ALK1 could result in the regulation of a different set of genes not involving ALK5. In addition, an alteration in Smad expression has also been considered to influence the outcome of several disorders targeted by TGF-?, including tissue fibrosis (2).

    Gene ontology revealed that the majority of the genes targeted in response to TGF-? treatment of LSMC and MSMC are functionally associated with cellular metabolism, cell growth regulation (cell cycle and apoptosis), cell and tissue structure (ECM, adhesion molecules, and microfilaments), signal transduction, and transcription factors. Despite differences in their expression profiles, there was a substantial degree of similarity in functional annotation among the genes identified at tissue (leiomyoma and myometrium) (19) and cellular (LSMC and MSMC) levels in response to TGF-?1. The differences between gene expression profiles at tissue and cellular (LSMC/MSMC) levels in response to TGF-? could be due to the contributions of other cell types to the gene pool and the influence of other autocrine/paracrine regulators on the overall gene expression at the tissue level. Previous studies from our laboratory and others have reported the expression of a few other genes targeted by TGF-? action in LSMC and MSMC (10, 11, 17). However, to our knowledge this is the first example of a large scale gene expression profiling of these cells in response to TGF-?. We validated the expression of several of these genes in response to time-dependent action of TGF-? in LSMC and MSMC, including the expression of 10 genes validated in leiomyoma/myometrium as well as in LSMC/MSMC in response to GnRHa treatment (19). Because detailed discussion of the genes identified in this study is beyond the scope of this manuscript, we focused on genes whose expression was confirmed and present only some major aspects of their possible function in leiomyoma and myometrium.

    We demonstrated that LSMC express a significantly higher level of IL-11 compared with MSMC and a major target of TGF-? regulatory action. Although the biological significance of IL-11 expression in leiomyoma and myometrial environments and the consequence of its overexpression in leiomyoma await investigation, IL-11, alone, or through interaction with TGF-?, is considered to play a critical role in the progression of fibrotic disorders (26, 27, 28). Equally, other members of the IL family, IL-4 and IL-13, and their interactions with TGF-? are reported to influence the outcome of tissue fibrosis (29). We have identified IL-13 expression in leiomyoma and discovered that exposure of LSMC to IL-13 up-regulates the expression of TGF-? and TGF-?RII in LSMC, suggesting a direct and/or indirect regulatory function for IL-13 in mediating events leading to the progression of tissue fibrosis in leiomyoma (16). Other cytokines in this category, including IL-4, IL-6, IL-8, IL-15, IL-17, TNF-, and GM-CSF, are also expressed in leiomyoma and myometrium (16, 17, 18, 19). These cytokines are classified as type 1/type 2-related subsets, and predominance toward type II direction is considered to result in inflammatory/immune responses leading to the progression of tissue fibrosis (27, 28, 29). A recent report has further elaborated the participation of IL-11, TGF-?, and transcription factor EGR1 in tissue fibrosis, through a mechanism involving regulation of balance between the rate of cellular apoptosis and inflammatory response (30). We have previously identified EGR1 among the differentially expressed genes in leiomyoma and myometrium (18) and here demonstrated the expression of EGR2 and EGR3 in these tissues (19) and regulation of EGR3 in response to TGF-? action in LSMC and MSMC. Elevated expression and preferential phosphorylation of EGR1 lead to regulation of target genes whose products are involved in apoptosis as well as angiogenesis and cell survival, including IL-2, TNF-, Flt-1, Fas, Fas ligand, cyclin D1, p15, p21, p53, platelet-derived growth factor-A, angiotensin II-dependent activation of platelet-derived growth factor and TGF-?, vascular endothelial growth factor, tissue factor, 5-lipoxygenase, intercellular adhesion molecule-1, fibronectin, urokinase-type plasminogen activator, and matrix metalloproteinase-1 (MMP-1) (31, 32, 33, 34, 35, 36). The expression of many of these genes has been documented in myometrium and leiomyoma and is known to be the target of TGF-? regulatory action (1, 3, 18, 20, 21). EGR1 also acts as a transcriptional repressor of TGF-?RII through direct interaction with specificity protein-1 and Ets-like ets-related transcription factor sites in the proximal promoter of the gene (36). Transfection of EGR1 expression vector into a myometrial cell line expressing low levels of EGR1 resulted in a rapid growth inhibition of these cells (37). To our knowledge, our study is the first to report a regulatory function of TGF-? on EGR3 expression not only in LSMC and MSMC, but in any other cell type. Based on our previous and present observations, we propose that there is a local inflammatory response mediated through individual and combined actions of TGF-?, IL-11, and IL-13 as well as a regulatory function of TGF-? on EGRs expression, resulting in the local expression of genes whose products promote apoptotic and nonapoptotic cell death, further enhancing an inflammatory reaction that orchestrates various events leading to the progression of fibrosis in leiomyoma.

    Additional genes identified as differentially expressed and regulated by TGF-? autocrine/paracrine action in LSMC and MSMC in this functional category include TGIF, TIEG, CITED2, Nur77, Runx1, and Runx2. These transcription factors possess key regulatory functions in the expression of a wide range of genes in response to various stimuli, specifically TGF-?. The expression of TGIF, TIEG, CITED2, and Nur77 is highly regulated in LSMC and MSMC, and with the exception of CITED2, TGF-? transiently increased their expression in a time-dependent manner. TGIF is a transcriptional corepressor that directly associates with Smads and inhibits Smad-mediated transcriptional activation by competing with p300 for Smad association (38, 39). CITED2, induced by multiple cytokines, growth factors, and hypoxia, also interacts with p300 and functions as a coactivator for the transcription factor activating protein-2 (40). CITED2-mediated action is reported to result in down-regulation of MMP-1 and MMP-13 through interactions with CBP/p300 and other transcription factors, such as c-Fos, Ets-1, nuclear factor-B, and Smads, that control MMP promoter activities (41, 42). TGF-? targets the expression of these transcription factors and MMPs in many cell types, including LSMC and MSMC (2, 5, 11, 42); thus their differential regulation and interactions with CITED2 and TGIF may regulate the outcome of TGF-? actions in leiomyoma involving cell growth, inflammation, apoptosis, and tissue turnover. Unlike TGIF, TIEG is rapidly induced by TGF-? and enhances TGF-? actions through Smad2/3 activation (43, 44, 45). However, TIEG has no effect on gene transcription in the absence of Smad4 or due to overexpression of Smad7, although it is capable of increasing Smad2/3 activity in the absence of Smad7 (43, 46). We showed that TGF-? induced a rapid, but transient, expression of TIEG in LSMC and MSMC, and we recently demonstrated the expression of Smad2/3, Smad4, and Smad7 and their differential regulation by TGF-? in these cells (10, 11). Based on these observations, we also propose that TGF-?, through a mechanism involving TGIF, TIEG, and Smads, self-regulates its own autocrine/paracrine action in leiomyoma/myometrium. Estrogen has also been shown to increase TIEG expression in breast tumor cell (43, 47). Because estrogen, a major growth-promoting factor for leiomyoma, induces TGF-? expression in LSMC and MSMC (7, 8), either estrogen directly or through estradiol-induced TGF-? may regulate TIEG expression in leiomyoma. TIEG is also reported to trigger apoptotic cell programs by a mechanism involving the formation of reactive oxygen species (45), often created as a result of a local inflammatory response. Whether TGF-?-induced TIEG through the above mechanism results in apoptotic response in leiomyoma is not known; however, the formation of reactive oxygen species may enhance the local inflammatory response, serving as an additional mediator of tissue fibrosis in leiomyoma.

    Nur77 regulates the expression of a group of genes whose products are involved in cell cycle regulation, differentiation, apoptosis, and malignant transformation (48, 49). We provided evidence that Nur77 is the target of the regulatory action of TGF-? in LSMC and MSMC, with a pattern of expression resembling that observed in leiomyoma and myometrium, respectively (18, 19). Although the nature and functional significance of Nur77 expression in leiomyoma and its regulation by TGF-? are unknown, malignant transformation in leiomyoma is rare, suggesting that Nur77 may function as a regulator of the cell cycle in leiomyoma and myometrium. In addition to Nur77, we discovered that the expression of various genes functionally associated with cell cycle regulation and apoptosis is influenced by the autocrine/paracrine action of TGF-?, and the balance of their expression may be a critical factor in leiomyoma growth and regression. Additional transcription factors whose expression was the target of TGF-? action in LSMC and MSMC are Runx1 and Runx2. This family of transcriptional factors, consisting of Runx1 to Runx3, is an integral component of signaling cascades mediated by TGF-? and bone morphogenetic proteins regulating various biological processes, including cell growth and differentiation, hemopoiesis, and angiogenesis (20, 46, 50). We provide the first evidence for regulatory action of GnRHa therapy and GnRHa direct action on Runx1 and Runx2 expression in leiomyoma, myometrium, as well as LSMC and MSMC, with GnRHa significantly inhibiting their expression, specifically in MSMC. Although Runx2 is expressed at low levels in leiomyoma and myometrium, Runx1 and Runx2 expression in LSMC and MSMC displayed a rapid response to TGF-? action in vitro, with Runx1 showing a significantly higher response. TGF-? activation of Smad and MAPK cascades regulates the expression of Runx2; however, interaction with Smad3 causes repression of Runx2 and downstream transcription activation of specific genes (20, 46). We recently reported that TGF-? and GnRH activate the MAPK pathway (11), and GnRHa alters TGF-?-activated Smad in LSMC and MSMC (10), a signaling cascade that may regulate the expression of Runx1 and Runx2 in these cells. Differential regulation of Runx1 and Runx2 by TGF-? and GnRHa implies their potential biological implication, specifically in regulating TGF-? action in the leiomyoma microenvironment. This is particularly interesting because estrogen is also reported to enhance Runx2 activity through a mechanism involving TGF-?RI gene promoter, which contains several Runx binding sequences (51). Together, the identification of these and several other key transcription factors in LSMC and LSMC and their regulation by TGF-?, serving as integral components of inflammatory, cell cycle, and apoptotic processes, support our hypothesis that a regulatory balance between these events is a key factor in the progression of fibrosis mediated by TGF-? in leiomyoma.

    The balance between cell proliferation and apoptosis is critical to tissue homeostasis and central to leiomyoma growth and regression. Because both positive and negative signals determine the outcomes of these events, we identified several genes in this category in our previous and current study as differentially expressed and regulated in leiomyoma and myometrium, as well as in LSMC and MSMC in response to TGF-?. Our primary focus here was on p27Kip1, p57Kip2, and Gas-1 expression, because of their regulation by GnRHa (19). We found that TGF-? suppressed the expression of these genes in LSMC and in a biphasic fashion, accompanied by suppression of Gas-1 expression in MSMC. TGF-? is known to regulate the expression of several cell cycle regulatory proteins, including p27, which binds cyclin-dependent kinase (CDK), and by inhibiting the catalytic activity of the CDK-cyclin complex, regulate cell cycle progression and apoptosis (52). However, TGF-? regulation of p57 expression is limited (20, 21, 53), and available data suggest that TGF-? enhances p57 degradation through the ubiquitin-proteasome pathway and Smad-mediated signaling (54). TGF-?-induced p57 degradation, CDK2 activation, and cell proliferation are blocked by proteasome inhibitors or overexpression of Smad7 (54, 55, 56, 57). TGF-? also induces cell growth by influencing c-Myc expression and activation of G1, G2, CDK, and cyclins, and their inhibitors p15IN4b and p21 (20, 21, 46), and we identified them among TGF-?-targeted genes in LSMC and MSMC (18, 19). To our knowledge, our study is the first to demonstrate Gas-1 expression in human uterine tissue and its regulation by TGF-?. Gas-1 acts as a negative regulator of the cell cycle and has been positively correlated with the inhibition of endothelial cell apoptosis and the integrity of resting endothelium (58). Gas-1 is reported to suppress the growth and tumorigenicity of human tumor cells, and overexpression of c-Myc and murine double-minute clone 2 protein or a p53 mutation inhibits Gas-1-mediated action (59, 60, 61). Estrogen has also been reported to regulate Gas-1 in rat uterus (62). Because TGF-?s stimulate DNA synthesis, but not cell division, in LSMC and MSMC, p27, p57, Gas-1, and other cell cycle regulators may influence TGF-? action on leiomyoma cell growth late in S to M phases of the cell cycle transition.

    We also identified several genes functionally belonging to signal transduction pathways as the target of TGF-? action in LSMC and MSMC. Among them are members of family of Ras/Rho, Smads, MAPK, protein tyrosine kinase 2, S100 calcium-binding proteins, LIM protein and LIM domain kinase 2, serine/threonine kinase 17 (apoptosis-inducing), focal adhesion kinase 2, signal transducers and activators of transcription, etc. Although Smad and MAPK pathways are recruited and activated by TGF-? receptors, including in LSMC and MSMC, components of other pathways are not known to be targeted by TGF-?. However, many growth factors, cytokines, chemokines, polypeptide hormones, and adhesion molecules expressed by LSMC and MSMC, either alone or through cross-talk with TGF-? receptor signaling, may activate various components of the other pathways, although only the expression and activation of a few have been demonstrated in these cells. Because GPRK5 expression was detected in leiomyoma and myometrium and was the target of GnRHa action in LSMC and MSMC (19), we found that GPRK5 expression is regulated by TGF-?. The biological implication of GPRK5 and its regulation by TGF-? in LSMC and MSMC is unclear; however, GPKs serve as negative regulators of G protein-coupled receptor-mediated action through the generation of second messengers, such as cAMP and calcium/calmodulin, and down-regulation of their activity (desensitization) (63, 64, 65). Activation of calcium/calmodulin is reported to alter Smad function, influencing the outcome of TGF-?’s action (46). This result suggests that GPRK may act as a downstream regulator of TGF-? receptor signaling, possibly through modulation of protein kinase C, MAPK, and/or calmodulin and hence influencing TGF-? action in leiomyoma.

    Tissue remodeling is a critical event in the progression of fibrotic disorders and modulation of ECM, adhesion molecules, and protease expression, and phenotypic changes toward a myofibroblastic phenotype are essential components of this process (1, 2, 66, 67, 68, 69). In this and our previous study we identified the expression of several genes in this category in leiomyoma/myometrium and LSMC/MSMC, including fibronectin, collagens, decorin, versican, desmin, vimentin, fibromodulin, several members of the integrin family, desmoplakin, extracellular matrix protein 1, porin, SPARC (secreted protein acidic and rich in cysteine)-like 1, syndecan 4, endothelial cell-specific molecule 1, as well as MMPs, TIMPs, and ADAMs (A disintigran and metalloprotease) (18, 19). We have previously demonstrated the expression of fibronectin, vimentin, collagen, fibromodulin, MMPs, and TIMPs in leiomyoma and myometrium and their regulation by TGF-? through the activation of MAPK (11, 42, 70). Of particular interest is the elevated expression of decorin, vimentin, and fibromodulin in leiomyoma, because of their ability to bind TGF-? and control TGF-? autocrine/paracrine action, a mechanism considered to regulate the outcome of tissue fibrosis (1, 5, 71, 72). Because leiomyoma is believed to derive from transformation of myometrial connective tissue fibroblast and/or smooth muscle cells, the expression of vimentin in leiomyoma/LSMC implies that these cells have adopted a myofibroblastic characteristic. Although granulation tissue myofibroblasts are derived from local fibroblasts, other cell types, including smooth muscle cells, have the potential to acquire a myofibroblastic phenotype (30, 66, 67, 68). These cells express various cytokines, including GM-CSF, IL-11, and TGF-?, of which GM-CSF is considered to participate in fibroblast transformation into myofibroblasts and to enhance their TGF-? expression (66, 67, 68). We have shown that GM-CSF regulates TGF-? expression in LSMC, and their interaction with other cytokines, such as IL-11 and IL-13, may play a key role in events leading to leiomyoma formation and the outcome of fibrosis (7, 11, 16). IL-11, either alone or through induction by TGF-?, alters ECM turnover in myofibroblasts, resulting in the progression of tissue fibrosis (30, 73). Despite the importance of tissue turnover in the pathophysiology of leiomyoma, few data are currently available about the extent of ECM expression and differences that may exist compared with myometrium that contribute to the fibrotic character of leiomyoma.

    In conclusion, as a continuation of work with TGF-?, we provided the first large scale example of a gene expression profile of LSMC and MSMC, identifying a specific cluster of genes whose expression is targeted by the autocrine/paracrine action of TGF-?. We validated the expression of a selective number of these genes functionally recognized to regulate inflammatory response, angiogenesis, cell cycle, apoptotic and nonapoptotic cell death, and ECM matrix turnover, events central to leiomyoma pathobiology. Based on the present and our previous work with TGF-?, we propose that the individual and combined actions of TGF-? with other profibrotic cytokines, such as IL-11, IL-13, and GM-CSF, orchestrate local inflammatory responses mediated through and influenced by the expression of genes whose products regulate the above processes, providing an environment leading to the progression of fibrosis.

    References

    Blobe GC, Schiemann WP, Lodish HF 2000 Role of transforming growth factor ? in human disease. N Engl J Med 342:1350–1358[Free Full Text]

    Flanders KC 2004 Smad3 as a mediator of the fibrotic response. Int J Exp Pathol 85:47–64

    Schnaper HW, Hayashida T, Hubchak SC, Poncelet AC 2003 TGF-? signal transduction and mesangial cell fibrogenesis. Am J Physiol 284:F243–F252

    Clancy RM, Buyon JP 2003 Clearance of apoptotic cells: TGF-? in the balance between inflammation and fibrosis. J Leukocyte Biol 74:959–960

    Olman MA, Matthay MA 2003 Transforming growth factor-? induces fibrosis in immune cell-depleted lungs. Am J Physiol 285:L522–L526

    Dou Q, Zhao Y, Tarnuzzer RW, Rong H, Williams RS, Schultz GS, Chegini N 1996 Suppression of transforming growth factor-? (TGF-?) and TGF-? receptor messenger ribonucleic acid and protein expression in leiomyomata in women receiving gonadotropin-releasing hormone agonist therapy. J Clin Endocrinol Metab 81:3222–3230

    Chegini N, Tang XM, Ma C 1999 Regulation of transforming growth factor-?1 expression by granulocyte macrophage-colony-stimulating factor in leiomyoma and myometrial smooth muscle cells. J Clin Endocrinol Metab 84:4138–4143

    Chegini N, Ma C, Tang XM, Williams RS 2002 Effects of GnRH analogues, ‘add-back’ steroid therapy, antiestrogen and antiprogestins on leiomyoma and myometrial smooth muscle cell growth and transforming growth factor-? expression. Mol Hum Reprod 8:1071–1078

    Chegini N, Luo X, Ding L, Ripley D 2003 The expression of Smads and transforming growth factor ? receptors in leiomyoma and myometrium and the effect of gonadotropin releasing hormone analogue therapy. Mol Cell Endocrinol 209:9–16

    Xu J, Luo X, Chegini N 2003 Differential expression, regulation, and induction of Smads, transforming growth factor-? signal transduction pathway in leiomyoma, and myometrial smooth muscle cells and alteration by gonadotropin-releasing hormone analog. J Clin Endocrinol Metab 88:1350–1361

    Ding L, Xu J, Luo L, Chegini N 2004 Gonadotropin releasing hormone and transforming growth factor ? activate MAPK/ERK and differentially regulate fibronectin, type I collagen, and PAI-1 expression in leiomyoma and myometrial smooth muscle cells. J Clin Endocrinol Metab 89:5549–5557

    Tang XM, Dou Q, Zhao Y, McLean F, Davis J, Chegini N 1997 The expression of transforming growth factor-?s and TGF-? receptor mRNA and protein and the effect of TGF-?s on human myometrial smooth muscle cells in vitro. Mol Hum Reprod 3:233–240

    Arici A, Sozen I 2000 Transforming growth factor-?3 is expressed at high levels in leiomyoma where it stimulates fibronectin expression and cell proliferation. Fertil Steril 73:1006–1011

    Lee BS, Nowak RA 2001 Human leiomyoma smooth muscle cells show increased expression of transforming growth factor-?3 (TGF?3) and altered responses to the antiproliferative effects of TGF-?. J Clin Endocrinol Metab 86:913–920

    Arici A, Sozen I 2003 Expression, menstrual cycle-dependent activation, and bimodal mitogenic effect of transforming growth factor-?1 in human myometrium and leiomyoma. Am J Obstet Gynecol 188:76–83

    Ding L, Luo X Chegini N 2004 The expression of IL-13 and IL-15 in leiomyoma and myometrium and their influence on TGF-? and proteases expression in leiomyoma and myometrial smooth muscle cells and SKLM, leiomyosarcoma cell line. J Soc Gyncol Invest 11:319A

    Chegini N 2000 Implication of growth factor and cytokine networks in leiomyomas. In: Hill J, ed. Cytokines in human reproduction. New York: Wiley & Sons; 133–162

    Chegini N, Verala J, Luo X, Xu J, Williams RS 2003 Gene expression profile of leiomyoma and myometrium and the effect of gonadotropin releasing hormone analogue therapy. J Soc Gynecol Invest 10:161–171

    Luo X, Ding L, Xu J, Williams RS, Chegini N 2005 Leiomyoma and myometrial gene expression profiles and their response to gonadotropin releasing hormone analogue therapy. Endocrinology 146:1074–1096

    Miyazono K, Maeda S, Imamura T 2004 Coordinate regulation of cell growth and differentiation by TGF-? superfamily and Runx proteins. Oncogene 23:4232–4237

    Moustakas A, Pardali K, Gaal A, Heldin C-H 2002 Mechanisms of TGF-? signaling in regulation of cell growth and differentiation. Immunol Lett 82:85–91

    Verrecchia F, Chu ML, Mauviel A 2001 Identification of novel TGF-?/Smad gene targets in dermal fibroblasts using a combined cDNA microarray/promoter transactivation approach. J Biol Chem 276:17058–17062

    Zavadil J, Bitzer M, Liang D, Yang YC, Massimi A, Kneitz S, Piek E, Bottinger EP 2001 Genetic programs of epithelial cell plasticity directed by transforming growth factor-?. Proc Natl Acad Sci USA 98:6686–6691

    Xie L, Law BK, Aakre ME, Edgerton M, Shyr Y, Bhowmick NA, Moses HL 2003 Transforming growth factor ?-regulated gene expression in a mouse mammary gland epithelial cell line. Breast Cancer Res 5:R187–R198

    Ota T, Fujii M, Sugizaki T, Ishii M, Miyazawa K, Aburatani H, Miyazono K 2002 Targets of transcriptional regulation by two distinct type I receptors for transforming growth factor-? in human umbilical vein endothelial cells. J Cell Physiol 193:299–318

    Kuhn C, Homer RJ, Zhu Z, Ward N, Elias JA 2000 Morphometry explains variation in airway responsiveness in transgenic mice overexpressing interleukin-6 and interleukin-11 in the lung. Chest 117:260S–262S

    Zhu Z, Lee CG, Zheng T, Chupp G, Wang J, Homer RJ, Noble PW, Hamid Q, Elias JA 2001 Airway inflammation and remodeling in asthma. Lessons from interleukin 11 and interleukin 13 transgenic mice. Am J Respir Crit Care Med 164:S67–S70

    Chakir J, Shannon J, Molet S, Fukakusa M, Elias J, Laviolette M, Boulet LP, Hamid Q 2003 Airway remodeling-associated mediators in moderate to severe asthma: effect of steroids on TGF-?, IL-11, IL-17, and type I and type III collagen expression. J Allergy Clin Immunol 111:1293–1298

    Wynn TA 2004 Fibrotic disease and the TH1/TH2 paradigm. Nat Rev Immunol 4:583–594

    Lee CG, Cho SJ, Kang MJ, Chapoval SP, Lee PJ, Noble PW, Yehualaeshet T, Lu B, Flavell RA, Milbrandt J, Homer RJ, Elias JA 2004 Early growth response gene 1-mediated apoptosis is essential for transforming growth factor ?1-induced pulmonary fibrosis. J Exp Med 200:377–389

    Thiel G, Cibelli G 2002 Regulation of life and death by the zinc finger transcription factor Egr-1. J Cell Physiol 193:287–292

    Khachigian LM 2004 Early growth response-1: blocking angiogenesis by shooting the messenger. Cell Cycle 3:10–11

    Nagamura-Inoue T, Tamura T, Ozato K 2001 Transcription factors that regulate growth and differentiation of myeloid cells. Int Rev Immunol 20:83–105

    Liu C, Rangnekar VM, Adamson E, Mercola D 1998 Suppression of growth and transformation and induction of apoptosis by EGR-1. Cancer Gene Ther 5:3–28

    Liu C, Yao J, de Belle I, Huang RP, Adamson E, Mercola D 1999 The transcription factor EGR-1 suppresses transformation of human fibrosarcoma HT1080 cells by coordinated induction of transforming growth factor-?1, fibronectin, and plasminogen activator inhibitor-1. J Biol Chem 274:4400–4411

    Baoheng D, Chenzhong F, Kent KC, Bush Jr H, Schulick AH, Kreiger K, Collins T, McCaffrey TA 2000 Elevated Egr-1 in human atherosclerotic cells transcriptionally represses the transforming growth factor-type II receptor. J Biol Chem 275:39039–39047

    Shozu M, Murakami K, Segawa T, Kasai T, Ishikawa H, Shinohara K, Okada M, Inoue M 2004 Decreased expression of early growth response-1 and its role in uterine leiomyoma growth. Cancer Res 64:4677–4684

    Chen F, Ogawa K, Nagarajan RP, Zhang M, Kuang C, Chen Y 2003 Regulation of TG-interacting factor by transforming growth factor-?. Biochem J 371:257–263

    Wotton D, Knoepfler PS, Laherty CD, Eisenman RN, Massague J 2001 The Smad transcriptional corepressor TGIF recruits mSin3. Cell Growth Differ 12:457–463

    Tien ES, Davis JW, Vanden Heuvel JP 2004 Identification of the CREB-binding protein/p300-interacting protein CITED2 as a peroxisome proliferator-activated receptor coregulator. J Biol Chem 279:24053–24063

    Yokota H, Goldring MB, Sun HB 2003 CITED2-mediated regulation of MMP-1 and MMP-13 in human chondrocytes under flow shear. J Biol Chem 278:47275–47280

    Ma C, Chegini N 1999 Regulation of matrix metalloproteinases (MMPs) and their tissue inhibitors in human myometrial smooth muscle cells by TGF-?1. Mol Hum Reprod 5:950–954

    Johnsen SA, Subramaniam M, Janknecht R, Spelsberg TC 2002 TGF-? inducible early gene enhances TGF-?/Smad-dependent transcriptional responses. Oncogene 21:5783–5790

    Cook T, Urrutia R 2000 TIEG proteins join the Smads as TGF-?-regulated transcription factors that control pancreatic cell growth. Am J Physiol 278:G513–G521

    Ribeiro A, Bronk SF, Roberts PJ, Urrutia R, Gores GJ 1999 The transforming growth factor ?1-inducible transcription factor TIEG1, mediates apoptosis through oxidative stress. Hepatology 30:1490–1497

    Shi Y, Massague J 2003 Mechanisms of TGF-? signaling from cell membrane to the nucleus. Cell 113:685–700

    Sorbello V, Fuso L, Sfiligoi C, Scafoglio C, Ponzone R, Biglia N, Weisz A, Sismondi P, De Bortoli M 2003 Quantitative real-time RT-PCR analysis of eight novel estrogen-regulated genes in breast cancer. Int J Biol Markers 18:123–129

    Rajpal A, Cho YA, Yelent B, Koza-Taylor PH, Li D, Chen E, Whang M, Kang C, Turi TG, Winoto A 2003 Transcriptional activation of known and novel apoptotic pathways by Nur77 orphan steroid receptor. EMBO J 22:6526–6536

    Castro-Obregon S, Rao RV, del Rio G, Chen SF, Poksay KS, Rabizadeh S, Vesce S, Zhang XK, Swanson RA, Bredesen DE 2004 Alternative, nonapoptotic programmed cell death: mediation by arrestin 2, ERK2, and Nur77. J Biol Chem 279:17543–17553

    Levanon D, Groner Y 2004 Structure and regulated expression of mammalian RUNX genes. Oncogene 23:4211–4219

    McCarthy TL, Chang WZ, Liu Y, Centrella M 2003 Runx2 integrates estrogen activity in osteoblasts. J Biol Chem 278:43121–43119

    Reed SI 2003 Ratchets and clocks: the cell cycle, ubiquitylation and protein turnover. Nat Rev Mol Cell Biol 4:855–864

    Kim SJ, Letterio J 2003 Transforming growth factor-? signaling in normal and malignant hematopoiesis. Leukemia 17:1731–1737

    Nishimori S, Tanaka Y, Chiba T, Fujii M, Imamura T, Miyazono K, Ogasawara T, Kawaguchi H, Igarashi T, Fujita T, Tanaka K, Toyoshima H 2001 Smad-mediated transcription is required for transforming growth factor-?1-induced p57 (Kip2) proteolysis in osteoblastic cells. J Biol Chem 276:10700–10705

    Yokoo T, Toyoshima H, Miura M, Wang Y, Iida KT, Suzuki H, Sone H, Shimano H, Gotoda T, Nishimori S, Tanaka K, Yamada N 2003 p57Kip2 regulates actin dynamics by binding and translocating LIM-kinase 1 to the nucleus. J Biol Chem 278:52919–52923

    Brown KA, Roberts RL, Arteaga CL, Law BK 2004 Transforming growth factor-? induces Cdk2 relocalization to the cytoplasm coincident with dephosphorylation of retinoblastoma tumor suppressor protein. Breast Cancer Res 6:R130–R139

    Kawaguchi K, Oda Y, Saito T, Yamamoto H, Takahira T, Tamiya S, Iwamoto Y, Tsuneyoshi M 2004 2004 Decreased expression of transforming growth factor-? II receptor is associated with that of p27KIP1 in giant cell tumor of bone: a possible link between transforming growth factor-? and cell cycle-related protein. Hum Pathol 35:61–68

    Spagnuolo R, Corada M, Orsenigo F, Zanetta L, Deuschle U, Sandy P, Schneider C, Drake CJ, Breviario F, Dejana E 2004 Gas1 is induced by VE-cadherin and vascular endothelial growth factor and inhibits endothelial cell apoptosis. Blood 103:3005–3012

    Gartel AL, Shchors K 2003 Mechanisms of c-myc-mediated transcriptional repression of growth arrest genes. Exp Cell Res 283:17–21

    Lee TC, Li L, Philipson L, Ziff EB 1997 Myc represses transcription of the growth arrest gene gas1. Proc Natl Acad Sci USA 94:12886–12891

    Ferrero M, Cairo G 1993 Estrogen-regulated expression of a growth arrest specific gene (gas-1) in rat uterus. Cell Biol Int 17:857–862

    Evdokiou A, Cowled PA 1998 Growth-regulatory activity of the growth arrest-specific gene, GAS1, in NIH3T3 fibroblasts. Exp Cell Res 240:359–367

    Luo J, Benovic JL 2003 G protein-coupled receptor kinase interaction with Hsp90 mediates kinase maturation. J Biol Chem 278:50908–50914

    Miyagawa Y, Ohguro H, Odagiri H, Maruyama I, Maeda T, Maeda A, Sasaki M, Nakazawa M 2003 Aberrantly expressed recoverin is functionally associated with G-protein-coupled receptor kinases in cancer cell lines. Biochem Biophys Res Commun 300:669–673

    Krasel C, Dammeier S, Winstel R, Brockmann J, Mischak H, Lohse MJ 2001 Phosphorylation of GRK2 by protein kinase C abolishes its inhibition by calmodulin. J Biol Chem 276:1911–1915

    Gabbiani G 2003 The myofibroblast in wound healing and fibrocontractive diseases. J Pathol 200:500–503

    Phan SH 2002 The myofibroblast in pulmonary fibrosis. Chest 122:286S–289S

    Shephard P, Hinz B, Smola-Hess S, Meister JJ, Krieg T, Smola H 2004 Dissecting the roles of endothelin, TGF-? and GM-CSF on myofibroblasts differentiation by keratinocytes. Thromb Haemost 92:262–274

    Gauldie J, Sime PJ, Xing Z, Marr B, Tremblay GM 1999 Transforming growth factor-? gene transfer to the lung induces myofibroblasts presence and pulmonary fibrosis. Curr Top Pathol 93:35–45

    Dou Q, Tarnuzzer RW, Williams RS, Schultz GS, Chegini N 1997 Differential expression of matrix metalloproteinases and their tissue inhibitors in leiomyomata: a mechanism for gonadotrophin releasing hormone agonist-induced tumour regression. Mol Hum Reprod 3:1005–1014

    Levens E, Luo X, Ding L, Williams RS, Chegini N 2004 Differential expression of fibromodulin and Abl-interactor 2 in leiomyoma and myometrium and regulation by gonadotropin releasing hormone analogue (GnRHa) therapy. Fertil Steril 82(Suppl 2):S88

    Chakravarti S 2002 Functions of lumican and fibromodulin: lessons from knockout mice. Glycoconj J 19:287–293

    Bamba S, Andoh A, Yasui H, Makino J, Kim S, Fujiyama Y 2003 Regulation of IL-11 expression in intestinal myofibroblasts: role of c-Jun AP-1- and MAPK-dependent pathways. Am J Physiol 285:G529–G538(Xiaoping Luo, Li Ding, Ji)