Limits...
Down-regulation of HtrA1 activates the epithelial-mesenchymal transition and ATM DNA damage response pathways.

Wang N, Eckert KA, Zomorrodi AR, Xin P, Pan W, Shearer DA, Weisz J, Maranus CD, Clawson GA - PLoS ONE (2012)

Bottom Line: Reduction of HtrA1 levels resulted in the epithelial-to-mesenchymal transition with acquisition of mesenchymal phenotypic characteristics, including increased growth rate, migration, and invasion, as well as expression of mesenchymal biomarkers.Moreover, reduction of HtrA1 expression resulted in activation of the ATM and DNA damage response, whereas overexpression of HtrA1 prevented this activation.Collectively, these results suggest that HtrA1 may function as a tumor suppressor by controlling the epithelial-to-mesenchymal transition, and may function in chemotherapeutic responsiveness by mediating DNA damage response pathways.

View Article: PubMed Central - PubMed

Affiliation: Gittlen Cancer Research Institute & Department of Pathology, College of Medicine, Pennsylvania State University, Hershey, Pennsylvania, United States of America.

ABSTRACT
Expression of the serine protease HtrA1 is decreased or abrogated in a variety of human primary cancers, and higher levels of HtrA1 expression are directly related to better response to chemotherapeutics. However, the precise mechanisms leading to HtrA1 down regulation during malignant transformation are unclear. To investigate HtrA1 gene regulation in breast cancer, we characterized expression in primary breast tissues and seven human breast epithelial cell lines, including two non-tumorigenic cell lines. In human breast tissues, HtrA1 expression was prominent in normal ductal glands. In DCIS and in invasive cancers, HtrA1 expression was greatly reduced or lost entirely. HtrA1 staining was also reduced in all of the human breast cancer cell lines, compared with the normal tissue and non-tumorigenic cell line controls. Loss of HtrA1 gene expression was attributable primarily to epigenetic silencing mechanisms, with different mechanisms operative in the various cell lines. To mechanistically examine the functional consequences of HtrA1 loss, we stably reduced and/or overexpressed HtrA1 in the non-tumorigenic MCF10A cell line. Reduction of HtrA1 levels resulted in the epithelial-to-mesenchymal transition with acquisition of mesenchymal phenotypic characteristics, including increased growth rate, migration, and invasion, as well as expression of mesenchymal biomarkers. A concomitant decrease in expression of epithelial biomarkers and all microRNA 200 family members was also observed. Moreover, reduction of HtrA1 expression resulted in activation of the ATM and DNA damage response, whereas overexpression of HtrA1 prevented this activation. Collectively, these results suggest that HtrA1 may function as a tumor suppressor by controlling the epithelial-to-mesenchymal transition, and may function in chemotherapeutic responsiveness by mediating DNA damage response pathways.

Show MeSH

Related in: MedlinePlus

Cluster Analysis of Whole Genome Gene Expression Data.Genes showing significant changes (and a minimum of 50% change) in expression levels (at p<0.01) in MCF10A/siRNA cell lines vs. controls (3 independent experiments) were analyzed by Cluster Analysis. Gene expression profiles in MCF10A/siRNA3 and siRNA4 cell lines, and in the MCF10A/HtrA1 over-expressing cell lines, were determined vs. the control cell line. The optimal number of clusters was found to be 22, and the centroids for the various individual clusters are indicated. Clusters 19, 6, and 2 showed the greatest changes in expression levels in the MCF10A/siRNA cell lines which were inversely correlated with HtrA1 expression levels. Clusters 12, 21, and 4 showed the greatest changes in expression levels that were positively correlated with HtrA1 expression levels.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3383700&req=5

pone-0039446-g009: Cluster Analysis of Whole Genome Gene Expression Data.Genes showing significant changes (and a minimum of 50% change) in expression levels (at p<0.01) in MCF10A/siRNA cell lines vs. controls (3 independent experiments) were analyzed by Cluster Analysis. Gene expression profiles in MCF10A/siRNA3 and siRNA4 cell lines, and in the MCF10A/HtrA1 over-expressing cell lines, were determined vs. the control cell line. The optimal number of clusters was found to be 22, and the centroids for the various individual clusters are indicated. Clusters 19, 6, and 2 showed the greatest changes in expression levels in the MCF10A/siRNA cell lines which were inversely correlated with HtrA1 expression levels. Clusters 12, 21, and 4 showed the greatest changes in expression levels that were positively correlated with HtrA1 expression levels.

Mentions: In order to understand the full range of phenotypic consequences of modulating HtrA1 levels, we examined changes in genome-wide gene expression, using the MCF10A/HtrA1 over-expressing cell line and 2 of the MCF10A/siRNA cell lines, and the Illumina Human Whole Genome Beadchip assay. We used two methods to analyze the data: gene clustering and signaling pathway analysis. First, 1402 genes were identified for cluster analysis by considering changes in expression levels in the MCF10A/siRNA cells that were greater than 50% compared to the control cell line (at p<0.01). We used the Silhouette measurement to determine the optimal number of clusters. Using this analysis, the optimal number of clusters was determined to be 22 (Silhouette width  = 0.384) (Figure 9). We detected genes whose expression was inversely correlated with HtrA1 expression level, as well as genes whose expression was positively correlated with HtrA1 expression levels. Clusters 19, 6, and 2 were the top three clusters, based on the magnitude of changes in gene expression levels, which showed expression changes inversely correlated with HtrA1 expression level. As examples, cluster 19 (4 genes) contained VIM (2 loci), cluster 6 (21 genes) included FGFR3, IGFBP2, and TNFRSF6B, and cluster 2 (281 genes) included many interesting genes such as LAMB1, RAD21, ATM, HIF1A, FGFRL1, VEGFB, VEGFC, H2AFX, MTA1, and PTPRE, PTPLA, and POLR3GL. Clusters 12, 21, and 4 were the top four clusters positively correlated with HtrA1 expression level. Cluster 12 (10 genes) included CD24, cluster 21 (23 genes) prominently included a number of histone genes, and cluster 4 (22 genes) included PRSS8, KRT15, CLDN7, and CDH1. We confirmed changes in transcript levels for many of the pertinent genes using QPCR, including VIM, CDH1, CLDN1, and ATM (not shown).


Down-regulation of HtrA1 activates the epithelial-mesenchymal transition and ATM DNA damage response pathways.

Wang N, Eckert KA, Zomorrodi AR, Xin P, Pan W, Shearer DA, Weisz J, Maranus CD, Clawson GA - PLoS ONE (2012)

Cluster Analysis of Whole Genome Gene Expression Data.Genes showing significant changes (and a minimum of 50% change) in expression levels (at p<0.01) in MCF10A/siRNA cell lines vs. controls (3 independent experiments) were analyzed by Cluster Analysis. Gene expression profiles in MCF10A/siRNA3 and siRNA4 cell lines, and in the MCF10A/HtrA1 over-expressing cell lines, were determined vs. the control cell line. The optimal number of clusters was found to be 22, and the centroids for the various individual clusters are indicated. Clusters 19, 6, and 2 showed the greatest changes in expression levels in the MCF10A/siRNA cell lines which were inversely correlated with HtrA1 expression levels. Clusters 12, 21, and 4 showed the greatest changes in expression levels that were positively correlated with HtrA1 expression levels.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3383700&req=5

pone-0039446-g009: Cluster Analysis of Whole Genome Gene Expression Data.Genes showing significant changes (and a minimum of 50% change) in expression levels (at p<0.01) in MCF10A/siRNA cell lines vs. controls (3 independent experiments) were analyzed by Cluster Analysis. Gene expression profiles in MCF10A/siRNA3 and siRNA4 cell lines, and in the MCF10A/HtrA1 over-expressing cell lines, were determined vs. the control cell line. The optimal number of clusters was found to be 22, and the centroids for the various individual clusters are indicated. Clusters 19, 6, and 2 showed the greatest changes in expression levels in the MCF10A/siRNA cell lines which were inversely correlated with HtrA1 expression levels. Clusters 12, 21, and 4 showed the greatest changes in expression levels that were positively correlated with HtrA1 expression levels.
Mentions: In order to understand the full range of phenotypic consequences of modulating HtrA1 levels, we examined changes in genome-wide gene expression, using the MCF10A/HtrA1 over-expressing cell line and 2 of the MCF10A/siRNA cell lines, and the Illumina Human Whole Genome Beadchip assay. We used two methods to analyze the data: gene clustering and signaling pathway analysis. First, 1402 genes were identified for cluster analysis by considering changes in expression levels in the MCF10A/siRNA cells that were greater than 50% compared to the control cell line (at p<0.01). We used the Silhouette measurement to determine the optimal number of clusters. Using this analysis, the optimal number of clusters was determined to be 22 (Silhouette width  = 0.384) (Figure 9). We detected genes whose expression was inversely correlated with HtrA1 expression level, as well as genes whose expression was positively correlated with HtrA1 expression levels. Clusters 19, 6, and 2 were the top three clusters, based on the magnitude of changes in gene expression levels, which showed expression changes inversely correlated with HtrA1 expression level. As examples, cluster 19 (4 genes) contained VIM (2 loci), cluster 6 (21 genes) included FGFR3, IGFBP2, and TNFRSF6B, and cluster 2 (281 genes) included many interesting genes such as LAMB1, RAD21, ATM, HIF1A, FGFRL1, VEGFB, VEGFC, H2AFX, MTA1, and PTPRE, PTPLA, and POLR3GL. Clusters 12, 21, and 4 were the top four clusters positively correlated with HtrA1 expression level. Cluster 12 (10 genes) included CD24, cluster 21 (23 genes) prominently included a number of histone genes, and cluster 4 (22 genes) included PRSS8, KRT15, CLDN7, and CDH1. We confirmed changes in transcript levels for many of the pertinent genes using QPCR, including VIM, CDH1, CLDN1, and ATM (not shown).

Bottom Line: Reduction of HtrA1 levels resulted in the epithelial-to-mesenchymal transition with acquisition of mesenchymal phenotypic characteristics, including increased growth rate, migration, and invasion, as well as expression of mesenchymal biomarkers.Moreover, reduction of HtrA1 expression resulted in activation of the ATM and DNA damage response, whereas overexpression of HtrA1 prevented this activation.Collectively, these results suggest that HtrA1 may function as a tumor suppressor by controlling the epithelial-to-mesenchymal transition, and may function in chemotherapeutic responsiveness by mediating DNA damage response pathways.

View Article: PubMed Central - PubMed

Affiliation: Gittlen Cancer Research Institute & Department of Pathology, College of Medicine, Pennsylvania State University, Hershey, Pennsylvania, United States of America.

ABSTRACT
Expression of the serine protease HtrA1 is decreased or abrogated in a variety of human primary cancers, and higher levels of HtrA1 expression are directly related to better response to chemotherapeutics. However, the precise mechanisms leading to HtrA1 down regulation during malignant transformation are unclear. To investigate HtrA1 gene regulation in breast cancer, we characterized expression in primary breast tissues and seven human breast epithelial cell lines, including two non-tumorigenic cell lines. In human breast tissues, HtrA1 expression was prominent in normal ductal glands. In DCIS and in invasive cancers, HtrA1 expression was greatly reduced or lost entirely. HtrA1 staining was also reduced in all of the human breast cancer cell lines, compared with the normal tissue and non-tumorigenic cell line controls. Loss of HtrA1 gene expression was attributable primarily to epigenetic silencing mechanisms, with different mechanisms operative in the various cell lines. To mechanistically examine the functional consequences of HtrA1 loss, we stably reduced and/or overexpressed HtrA1 in the non-tumorigenic MCF10A cell line. Reduction of HtrA1 levels resulted in the epithelial-to-mesenchymal transition with acquisition of mesenchymal phenotypic characteristics, including increased growth rate, migration, and invasion, as well as expression of mesenchymal biomarkers. A concomitant decrease in expression of epithelial biomarkers and all microRNA 200 family members was also observed. Moreover, reduction of HtrA1 expression resulted in activation of the ATM and DNA damage response, whereas overexpression of HtrA1 prevented this activation. Collectively, these results suggest that HtrA1 may function as a tumor suppressor by controlling the epithelial-to-mesenchymal transition, and may function in chemotherapeutic responsiveness by mediating DNA damage response pathways.

Show MeSH
Related in: MedlinePlus