Oncogenicity potential was a defining characteristic of v-ets, the first of the large family of ETS transcription factors to be cloned. v-ets was discovered as an oncoprotein fusion with gag and myb in the E26 avian erythroblastosis virus (1). There are now more than 25 ETS family members defined by a DNA binding or ETS domain, which is highly conserved among family members, and binds to the core DNA binding motif C/A GCA A/T. In addition, a subset of ETS family members have a highly conserved PNT domain that, at least in some contexts, acts an as oligomerization motif (2).
ETS family members play an important role in mammalian development. For example, targeted disruption of the Fli-1 gene results in hemorrhage from the dorsal aorta to the lumen of the neural tube and brain ventricles on embryonic day 11.0 (E11.0; Ref. 3). Deficiency of Ets2 causes day E8.5 embryonic lethality, with defects in extraembryonic tissue gene expression and function and failure of ectoplacental cone proliferation (4). Loss of function of ETV6/TEL2 causes E9.5 embryonic lethality with severe defects in yolk sac angiogenesis and apoptosis of neural crest-derived ganglia (5). In addition, ETS family members play important roles in hematopoietic development. Disruption of Ets-1 causes defects in natural killer cell development (6), loss of function of PU.1 causes defects in myelopoiesis as well as B-lymphopoiesis (7, 8, 9), and loss of ETV6/TEL causes inability for the transition of definitive hematopoietic cells from the liver to the bone marrow (10).
In addition to playing an important role in mammalian development, ETS family members have been directly implicated in the pathogenesis of a spectrum of human cancers; ETS involvement in human malignancy is notable for pleiotropic structural and functional contributions to the contributions to the malignant phenotype (Fig. 1). For example, there is a spectrum of cancers that include soft tissue sarcomas and leukemias in which balanced reciprocal translocations result in fusion proteins that contain the ETS DNA binding domain. Examples include the EWS-FLI1 and EWS-ERG (11) fusion associated with Ewing’s sarcoma and the TLS/ERG (12) and ETV6/MN1 (13) fusion associated with human leukemia. In addition, there is a broad spectrum of hematological malignancies in which the PNT oligomerization motif is involved in chromosomal translocations with a diverse group of partners that include transcription factors, tyrosine kinases, and genes of unknown function. For example, the ETS translocation variant 6 (also known as TEL) gene is known to be involved in >40 different translocations in human hematological malignancies. In most of these cases, balanced reciprocal chromosomal translocations result in expression of a fusion protein in which the PNT domain is fused in-frame to the respective partners. Examples of TEL fusions to tyrosine kinases include TEL/PDGFβR (14, 15), TEL/TRKc (16, 17, 18, 19), TEL/ABL (20, 21, 22), and TEL/JAK2 (23, 24, 25, 26), among others. TEL may also be fused to transcription factor partners such as AML1 (27, 28, 29, 30, 31), EVI1 (32), and others. In these examples, there is strong epidemiological support for a direct role in transformation of cells. Although these diseases are rare, with an incidence of approximately 1 per 100,000 individuals/year, the chromosomal translocations are highly conserved and in most cases result in expression of identical in-frame fusions. In addition, in some cases it has been possible to demonstrate that the ETS fusion protein is both necessary and sufficient for disease pathogenesis. For example, the TEL/PDGFβR, TEL/TRKc, TEL/ABL, and TEL/JAK2 fusion proteins are each capable of causing myeloproliferative disease in murine bone marrow transplant models of leukemia. These observations have prompted further analysis of downstream targets that may contribute to malignant transformation of cells. For example, TEL/JAK2 and other TEL tyrosine kinase fusion molecules activated STAT5, a member of the STAT family of latent cytoplasmic transcription factors. In the case of TEL/JAK2, activation of STAT5 is both necessary and sufficient to transform hematopoietic cells. STAT5, in turn, is known to transactivate a number of genes that regulate cellular proliferation and survival, such as oncostatin M, Bcl-Xl, cyclin D1, and Pim1 (26). Thus, direct or indirect transcriptional targets of the ETS fusion proteins contributes directly to the transformed phenotype of the cells.
Loss of function of ETS family members may also contribute to the pathogenesis of human cancers. For example, in TEL/AML1 leukemias, there is nearly invariable deletion of the residual TEL allele (27, 28, 29, 30, 31, 33, 34). Thus, as a consequence of translocation and deletion, there is no functional TEL present in leukemic cells, suggesting that TEL may have tumor suppressor function. Recently, Mueller et al. (35) have reported the presence of mutations in PU.1 associated with human leukemias, indicating that the loss of function of PU.1 may also contribute to the pathogenesis of disease.
In the current report, Davidson et al. (see this issue, pp. 551–557) have provided new and important clinical therapeutic insights into another role of ETS family members in human cancer. In addition to direct contributions to the transformed phenotype, aberrant overexpression of ETS family members may influence metastatic potential and response to therapy of certain human epithelial tumors. Davidson et al. assayed ETS-1 expression using mRNA in situ hybridization from 66 primary ovarian carcinomas and metastatic lesions obtained from 41 patients diagnosed with advanced stage ovarian carcinoma. ETS-1 expression was detected in 42 and 33%, respectively, of carcinoma and stromal cells, respectively. There was a statistically significant correlation between ETS-1 and VEGF expression in carcinoma and stromal cells, basic fibroblast growth factor in carcinoma cells, and membrane-type-1 matrix metalloproteinase. Furthermore, ETS-1 expression was statistically significant more common in both carcinoma and stromal cells in tumors of short-term survivors, and in univariate analysis, ETS-1 expression in both tumor and stromal cells correlated with poor survival. The patient numbers are small, and there are several puzzling aspects of the observations made. For example, one would expect that if, as suggested, the genes characterized are surrogate markers for metastatic potential, then metastatic lesions should have higher levels and frequency of expression of these surrogates. Although this did not appear to be the case in this study, additional studies may clarify the role of these expressed genes in the pathogenesis of disease and response to therapy.
The study also raises interesting questions about molecular pathogenesis of these tumors. It is known, for example, that overexpression of ETS family members alone may be sufficient to transform mammalian cells in culture. Thus, it is plausible that the overexpression of ETS-1 in ovarian cancer not only activates transcription of genes required for potentiation of metastasis but contributes directly to the transformation of cells. This study also suggests that it may be of value to analyze the expression of other ETS family members in solid tumors. In addition, it will be important to determine the molecular mechanisms whereby ETS-1 or other family members are overexpressed. Are these downstream effectors of transformation mediated by aberrant trans-acting elements, or are their cis-acting mutations that confer are responsible for the overexpression of ETS-1. Finally, these data, with the limited spectrum of target genes analyzed, suggest that efforts to identify global gene expression signatures for epithelial tumors characterized by ETS-1 overexpression may identify additional target genes that are potential prognostic factors or targets for therapy.
Taken together, this analysis has identified a new potential marker for prognosis in ovarian carcinomas and provides indirect support for the hypothesis that ETS-1 expression regulates expression of proteinases and angiogenic factors that impact clinical behavior and response to therapy. The study provides fertile new ground for investigation of the molecular basis of the correlation between ETS-1 expression and genes that may potentiate the malignant phenotype, as well as testing these observations in cell culture and vertebrate models of disease pathogenesis.
These studies also provide a cogent reminder that the ability to transform an epithelial cell is only a part of the spectrum of biological activities that a transforming oncoprotein confers on the malignant phenotype. Further investigations of the spectrum of genes that are expressed in tumor cells using sophisticated global expression analysis may shed additional insights on transcriptional regulatory networks that contribute to disease pathogenesis.
The abbreviations used are: TEL, translocation ETS leukemia; PDGFβR, platelet-derived growth factor β receptor.
Diverse mechanisms for contributing to the cancer phenotype by ETS family members. These include fusion of the PNT oligomerization motif to a spectrum of partners, including tyrosine kinases such as PDGFβR, ABL, JAK2, and TRKC, as well as transcription factors such as AML1. In addition, the ETS, or DNA binding domain, is fused to partners that are thought to confer transactivating function, such as EWS, in soft tissue sarcomas and in some leukemias. There is evidence that loss of function of ETS family members may contribute to transformation, including the loss of function of both TEL (ETV6) alleles in TEL/AML1 pediatric acute lymphoblastic leukemias and loss of function mutations of PU.1 in acute myeloid leukemia. Finally, there is evidence, as described by Davidson et al. in this report, that overexpression of ETS family members and transactivation of their downstream target genes contribute to the malignant phenotype.
Diverse mechanisms for contributing to the cancer phenotype by ETS family members. These include fusion of the PNT oligomerization motif to a spectrum of partners, including tyrosine kinases such as PDGFβR, ABL, JAK2, and TRKC, as well as transcription factors such as AML1. In addition, the ETS, or DNA binding domain, is fused to partners that are thought to confer transactivating function, such as EWS, in soft tissue sarcomas and in some leukemias. There is evidence that loss of function of ETS family members may contribute to transformation, including the loss of function of both TEL (ETV6) alleles in TEL/AML1 pediatric acute lymphoblastic leukemias and loss of function mutations of PU.1 in acute myeloid leukemia. Finally, there is evidence, as described by Davidson et al. in this report, that overexpression of ETS family members and transactivation of their downstream target genes contribute to the malignant phenotype.
RSS Feeds