Rosita Winkler

Hedgehog signaling pathway is inactive in colorectal cancer cell lines

The Hedgehog (Hh) signaling pathway plays an important role in human development. Abnormal activation of this pathway has been observed in several types of human cancers, such as the upper gastro‐intestinal tract cancers. However, activation of the Hh pathway in colorectal cancers is controversial. We analyzed the expression of the main key members of the Hh pathway in 7 colon cancer cell lines in order to discover whether the pathway is constitutively active in these cells. We estimated the expression of SHH, IHH, PTCH, SMO, GLI1, GLI2, GLI3, SUFU and HHIP genes by RT‐PCR. Moreover, Hh ligand, Gli3 and Sufu protein levels were quantified by western blotting. None of the cell lines expressed the complete set of Hh pathway members. The ligands were absent from Colo320 and HCT116 cells, Smo from Colo205, HT29 and WiDr. GLI1 gene was not expressed in SW480 cells nor were GLI2/GLI3 in Colo205 or Caco‐2 cells. Furthermore the repressive form of Gli3, characteristic of an inactive pathway, was detected in SW480 and Colo320 cells. Finally treatment of colon cancer cells with cyclopamine, a specific inhibitor of the Hh pathway, did not downregulate PTCH and GLI1 genes expression in the colorectal cells, whereas it did so in PANC1 control cells. Taken together, these results indicate that the aberrant activation of the Hh signaling pathway is not common in colorectal cancer cell lines.

The combined immunodetection of AP-2α and YY1 transcription factors is associated with ERBB2 gene overexpression in primary breast tumors

IntroductionOverexpression of the ERBB2 oncogene is observed in about 20% of human breast tumors and is the consequence of increased transcription rates frequently associated with gene amplification. Several studies have shown a link between activator protein 2 (AP-2) transcription factors and ERBB2 gene expression in breast cancer cell lines. Moreover, the Yin Yang 1 (YY1) transcription factor has been shown to stimulate AP-2 transcriptional activity on the ERBB2 promoter in vitro. In this report, we examined the relationships between ERBB2, AP-2α, and YY1 both in breast cancer tissue specimens and in a mammary cancer cell line.MethodsERBB2, AP-2α, and YY1 protein levels were analyzed by immunohistochemistry in a panel of 55 primary breast tumors. ERBB2 gene amplification status was determined by fluorescent in situ hybridization. Correlations were evaluated by a χ2 test at a p value of less than 0.05. The functional role of AP-2α and YY1 on ERBB2 gene expression was analyzed by small interfering RNA (siRNA) transfection in the BT-474 mammary cancer cell line followed by real-time reverse transcription-polymerase chain reaction and Western blotting.ResultsWe observed a statistically significant correlation between ERBB2 and AP-2α levels in the tumors (p < 0.01). Moreover, associations were found between ERBB2 protein level and the combined high expression of AP-2α and YY1 (p < 0.02) as well as between the expression of AP-2α and YY1 (p < 0.001). Furthermore, the levels of both AP-2α and YY1 proteins were inversely correlated to ERBB2 gene amplification status in the tumors (p < 0.01). Transfection of siRNAs targeting AP-2α and AP-2γ mRNAs in the BT-474 breast cancer cell line repressed the expression of the endogenous ERBB2 gene at both the mRNA and protein levels. Moreover, the additional transfection of an siRNA directed against the YY1 transcript further reduced the ERBB2 protein level, suggesting that AP-2 and YY1 transcription factors cooperate to stimulate the transcription of the ERBB2 gene.ConclusionThis study highlights the role of both AP-2α and YY1 transcription factors in ERBB2 oncogene overexpression in breast tumors. Our results also suggest that high ERBB2 expression may result either from gene amplification or from increased transcription factor levels.

Androgen receptor controls EGFR and ERBB2 gene expression at different levels in prostate cancer cell lines.

EGFR or ERBB2 contributes to prostate cancer (PCa) progression by activating the androgen receptor (AR) in hormone-poor conditions. Here, we investigated the mechanisms by which androgens regulate EGFR and ERBB2 expression in PCa cells. In steroid-depleted medium (SDM), EGFR protein was less abundant in androgen-sensitive LNCaP than in androgen ablation-resistant 22Rv1 cells, whereas transcript levels were similar. Dihydrotestosterone (DHT) treatment increased both EGFR mRNA and protein levels and stimulated RNA polymerase II recruitment to the EGFR gene promoter, whereas it decreased ERBB2 transcript and protein levels in LNCaP cells. DHT altered neither EGFR or ERBB2 levels nor the abundance of prostate-specific antigen (PSA), TMEPA1, or TMPRSS2 mRNAs in 22Rv1 cells, which express the full-length and a shorter AR isoform deleted from the COOH-terminal domain (ARDeltaCTD). The contribution of both AR isoforms to the expression of these genes was assessed by small interfering RNAs targeting only the full-length or both AR isoforms. Silencing of both isoforms strongly reduced PSA, TMEPA1, and TMPRSS2 transcript levels. Inhibition of both AR isoforms did not affect EGFR and ERBB2 transcript levels but decreased EGFR and increased ERBB2 protein levels. Proliferation of 22Rv1 cells in SDM was inhibited in the absence of AR and ARDeltaCTD. A further decrease was obtained with PKI166, an EGFR/ERBB2 kinase inhibitor. Overall, we showed that ARDeltaCTD is responsible for constitutive EGFR expression and ERBB2 repression in 22Rv1 cells and that ARDeltaCTD and tyrosine kinase receptors are necessary for sustained 22Rv1 cell growth.

Identification of HTF (HER2 transcription factor) as an AP-2 (activator protein-2) transcription factor and contribution of the HTF binding site to ERBB2 gene overexpression

The ERBB2 gene is overexpressed in 30% of human breast cancers and this is correlated with poor prognosis. Overexpression of the ERBB2 gene is due to increased transcription and gene amplification. Our previous studies have identified a new cis element in the ERBB2 promoter which is involved in the genes overexpression. This cis element, located 501 bp upstream from the main ERBB2 transcription initiation site, binds a transcription factor called HTF (HER2 transcription factor). We report here the identification of HTF as an AP-2 (activator protein-2) transcription factor. The new cis element is bound by AP-2 with high affinity, compared with a previously described AP-2 binding site located 284 bp downstream. Co-transfection of an AP-2alpha expression vector with a reporter vector containing the newly identified AP-2 binding site in front of a minimal ERBB2 promoter induced a dose-dependent increase in transcriptional activity. We examined the contribution of the new AP-2 binding site to ERBB2 overexpression. For this purpose we abolished the new and/or the previously described AP-2 binding sequence by site-directed mutagenesis. The results show that the two functional AP-2 sites in the first 700 bp of the ERBB2 promoter co-operate to achieve maximal transcriptional activity.

Amplification units and translocation at chromosome 17q and c-erbB-2 overexpression in the pathogenesis of breast cancer

Abstract Hyperplasia without and with atypia is considered to be a precursor lesion for certain breast carcinomas. The cytogenetic events and the molecular pathology involved in the multistep process from normal to invasive carcinoma are unknown. To characterise the sequence of early genetic abnormalities of chromosome 17q and their biological consequences in the pathogenesis of breast cancer, we performed immunohistochemistry on 451 breast tissues including 180 normal breast specimens, 28 hyperplastic lesions without atypia and 44 with atypia, 100 cases of ductal carcinoma in situ (DCIS) and 99 cases of invasive ductal carcinoma. We correlated the overexpression of the c-ErbB-2 protein, the histological and the recently proposed differentiation classification of DCIS with the extent of DCIS. For fluorescence in situ hybridisation (FISH) analysis, different probes spanning the 17q region including the c-erbB-2 gene locus and those which are found adjacent, were used. Reverse painting and comparative genomic hybridisation (CGH) were performed on several breast cancer cell lines. c-ErbB-2 overexpression was observed in only 29% of DCIS and 23% of invasive carcinomas, but not in hyperplastic and normal tissue. c-ErbB-2 overexpression is correlated with poor differentiation in DCIS but not in invasive carcinoma. In DCIS, there was no correlation with the histological subtype classification. The average extent of DCIS is significantly increased from 13.81 mm in c-ErbB-2 negative cases to 29.37 mm in c-ErbB-2 positive cases. The increase was considered to be a possible consequence of the overexpression and is probably due to the previously described motility enhancing effect of the c-ErbB-2 protein. The histological and differentiation classification of DCIS did not correlate with the extent of disease. Using FISH, amplified genes at 17q12, always including the c-erbB-2 gene, were detected in all cases of DCIS and invasive carcinoma with c-ErbB-2 overexpression. The centromeric region and the NF1 locus, which is located between the centromere and c-erbB-2, were not amplified in any of the DCIS and invasive breast carcinomas, but co-amplification of the myeloperoxidase gene was detected in 3/5 DCIS and 1/5 invasive carcinomas with c-ErbB-2 overexpression. In contrast to c-erbB-2, immunohistochemical overexpression of their respective gene products was not observed. FISH, reverse painting and CGH show similar amplified genes with amplified c-erbB-2 in c-ErbB-2 overexpressing SK-BR-3 and BT474 human breast cancer cells. The amplified genes are part of two different amplicons. Extensive modifications of the 17q chromosomal region, caused by translocation, were also observed in these cell lines. It is concluded that the modifications of chromosome 17q, inducing overexpression of c-ErbB-2 protein, occur at the level of transition from hyperplasia to DCIS. They are preserved in invasive carcinoma with overexpression of c-ErbB-2 protein. This had led to the hypothesis that these modifications at 17q may lead to a larger extent of DCIS.

Ku proteins interact with activator protein-2 transcription factors and contribute to ERBB2 overexpression in breast cancer cell lines.

IntroductionActivator protein-2 (AP-2) α and AP-2γ transcription factors contribute to ERBB2 gene overexpression in breast cancer. In order to understand the mechanism by which the ERBB2 gene is overexpressed we searched for novel AP-2 interacting factors that contribute to its activity.MethodsKu proteins were identified as AP-2α interacting proteins by glutathione serine transferase (GST)-pull down followed by mass spectrometry. Transfection of the cells with siRNA, expression vectors and reporter vectors as well as chromatin immunoprecipitation (ChIP) assay were used to ascertain the implication of Ku proteins on ERBB2 expression.ResultsNuclear proteins from BT-474 cells overexpressing AP-2α and AP-2γ were incubated with GST-AP2 or GST coated beads. Among the proteins retained specifically on GST-AP2 coated beads Ku70 and Ku80 proteins were identified by mass spectrometry. The contribution of Ku proteins to ERBB2 gene expression in BT-474 and SKBR3 cell lines was investigated by downregulating Ku proteins through the use of specific siRNAs. Depletion of Ku proteins led to downregulation of ERBB2 mRNA and protein levels. Furthermore, reduction of Ku80 in HCT116 cell line decreased the AP-2α activity on a reporter vector containing an AP-2 binding site linked to the ERBB2 core promoter, and transfection of Ku80 increased the activity of AP-2α on this promoter. Ku siRNAs also inhibited the activity of this reporter vector in BT-474 and SKBR3 cell lines and the activity of the ERBB2 promoter was further reduced by combining Ku siRNAs with AP-2α and AP-2γ siRNAs. ChIP experiments with chromatin extracted from wild type or AP-2α and AP-2γ or Ku70 siRNA transfected BT-474 cells demonstrated Ku70 recruitment to the ERBB2 proximal promoter in association with AP-2α and AP-2γ. Moreover, Ku70 siRNA like AP-2 siRNAs, greatly reduced PolII recruitment to the ERBB2 proximal promoter.ConclusionsKu proteins in interaction with AP-2 (α and γ) contribute to increased ERBB2 mRNA and protein levels in breast cancer cells.

Interplay between KLF4 and ZEB2/SIP1 in the regulation of E-cadherin expression

E-cadherin expression is repressed by ZEB2/SIP1 while it is induced by KLF4. Independent data from the literature indicate that these two transcription factors could bind close to each other in the proximal region of the E-cadherin gene promoter. We have here explored a potential competition between ZEB2 and KLF4 for the binding to the E-cadherin promoter. We show an inverse correlation between ZEB2 expression levels and KLF4 recruitment on the E-cadherin promoter in three breast cancer cell lines and in A431/HA.ZEB2 cells in which ZEB2 expression is induced by doxycycline (DOX). We identified a region of the E-cadherin promoter bound by KLF4 which is necessary for the activation of the E-cadherin promoter activity after KLF4 overexpression. This region is localized between positions -28 and -10 and thus overlaps with one of the ZEB2 binding sites. Deleting the bipartite ZEB2 binding site results in increased KLF4 induced E-cadherin promoter activity. Taken together, our results suggest that E-cadherin expression in cancer cells is controlled by a balance between ZEB2 and KLF4 expression levels.

Cloning of breakpoints in and downstream the IGF2 gene that are associated with overexpression of IGF2 transcripts in colorectal tumours

The human IGF2 gene belongs to a group of imprinted genes clustered on the short arm of chromosome 11, band p15.5. It contains 9 exons and spans over 30 kb. IGF2 mRNA overexpression has been reported in human tumours and in some inherited growth disorders. It was recently demonstrated that IGF2 mRNA overexpression contributes to tumour progression and that loss of parental imprinting as well as altered transcription factors are contributing to this overexpression. We have reported structural alterations in the 3′ region of the IGF2 gene in two colorectal tumours that overexpressed the IGF2 transcript by 200- and 800-fold. We cloned by the vectorette-PCR strategy, genomic DNA fragments containing the breakpoints from these tumours. The sequencing of these fragments positioned the breakpoint 2 kb downstream the IGF2 gene in one tumour, and in exon 9 in the second. Both breakpoints occurred in regions containing repetitive elements: a TGGA repeat we have identified downstream the gene, and the (CA)n repetition in exon 9. We hypothesize that a negative regulatory element, located downstream the IGF2 gene, has been deleted following these structural alterations and leads to IGF2 gene overexpression.