Commentary Cellscience Reviews Vol.2 No.3 ISSN 1742-8130

Loss of PAX-2 Methylation: The turning point of endometrial cells to the dark side of proliferation

Filippo Acconcia & Rakesh Kumar

Cancer is a degenerative disease that arises from the clonal expansion of a single transformed cell into a mass of uncontrolled proliferating cells. During the last decades, a large body of scientific efforts has led to formulate a simple multistep process for the development of cancer (carcinogenesis): initiation, promotion and progression. Despite the simplicity of the scientific model, each step hides a plethora of specific molecular mechanisms, often interlaced, which allow a somatic cell to change in a cancer cell. Paradigmatic is the definition that mutations occur within genes which encode for proteins required in the regulation of the cell cycle progression, thereby modifying the balance between the growth-inhibitory programs and proliferative networks, thus allowing the cell to escape its physiologic growth restraints (Lowe et al., 2004). In addition to the DNA damage within these oncogenes and tumor-suppressor genes, abnormal changes in the epigenetic cellular information (e.g., DNA methylation) can also support clonal evolution in human cancers (Feinberg & Tycko, 2004). Thus the development and progression of cancer, including endometrial cancer, does not occur as the result of a single specific event.

Endometrial cancer is the most common found within the female genital tract, and 80% of all endometrial carcinomas are of the ‘endometrioid’ type. Most endometrioid carcinomas (type I) are well differentiated and arise on a background of endometrial hyperplasia (Amant et al., 2005). Among the various risk factors which have been identified for the development of endometroid carcinomas, the unopposed estrogenic stimulation and the long-term use of tamoxifen in the treatment of ER-positive breast cancer are the leading causes for the development of endometrial hyperplasia that may eventually result in type I endometrial cancer (Amant et al., 2005). Tamoxifen (Tam), like the 17β-estradiol (E2), binds to the estrogen receptors (ERα and ERβ) and can oppose or even mimic the effects of E2 (e.g., cell proliferation and migration) depending on the target tissue (Acconcia et al., 2005a).

Figure 1. Proposed model for endometrial carcinogenesis. Normal endometrial cells undergo transformation following loss of methylation. Hypomethylation dependent PAX-2 reactivation drives E2- and Tam-mediated cell proliferation and tumor growth via ERα (Wu et al., 2005), and may lead to E2- and Tam-induced invasion and metastasis (Acconcia et al., 2005a). For details, see please the text.

In a recent paper published in Nature, Wu et al. (2005) applied an elegant approach to demonstrate the critical role of E2 and Tam in endometrial carcinogenesis. Using endometrial epithelial cells purified from human type I endometroid carcinomas for global human genome microarray analysis, the authors were able to identify 35 genes regulated both by E2 and Tam and to define one particular gene under the control of the ERα signaling (i.e. PAX-2) which was responsible for the ligand-induced proliferation of endometrial cancer cells and the growth of endometrial tumors. In addition, by comparing normal endometrial epithelial cells versus cancer endometrial epithelial cells, these authors also demonstrated that the ability of E2 and Tam to stimulate endometrial carcinogenesis is linked to the hypomethylation-driven reactivation of the PAX-2 promoter.

These novel findings shed new light upon the effect of Tam within endometrial cells. The accepted model supports a mixed agonistic and antagonistic activity of Tam towards ER. This notion is based upon the ability of Tam to induce conformational changes within ER that allow the recruitment of specific sets of cofactors. The differential recruitment of coactivators or corepressors to the ER in different cell types is thought to be one such possible mechanism for E2- and Tam-dependent carcinogenesis (Marino et al., 2005). This concept is now changing since, in addition to being a partial agonist of ERα signaling by mimicking the effects of E2, Tam possesses a distinct and specific genomic activity which account for the transcriptional regulation of unique target genes. Thus, Tam-mediated gene transcription per se might play a critical role in Tam-induced endometrial carcinogenesis (Wu et al., 2005).

The finding that E2 and Tam activate PAX-2 transcription suggests that distinct extracellular stimuli could converge through the same pathway toward specific effector proteins (e.g., transcription factors) that function as master regulators of particular physiological processes. For example, the transcription factor Snail is activated by different signals and regulates endothelial to mesenchimal transition (EMT, Schlessinger & Hall, 2004). In this context, PAX-2 can be also regarded as a master regulator of endometrial cell proliferation. PAX-2 is a transcription factor that belongs to the subgroup II of the paired box gene family. The PAX gene family plays a critical role in the process of tissue development and cellular differentiation within embryos, and acts by promoting cell proliferation and migration (Robson et al., 2006). PAX proteins of subgroup II are highly susceptible to deregulated expression and thus contribute to cancer development. PAX-2 up-regulation has been associated with cancers of the female genital tract, and results in increased proliferation and gain of invasive and metastatic potential (Robson et al., 2006).

The PAX-2 promoter region does not contain any ERα binding motif, and the E2- and Tam-induced PAX-2 transcriptional activation is mediated by the activation of Sp-1 and NFκB transcription factors by ERα but not by ERβ (Wu et al., 2005). These findings highlight the importance of the ERα in the proliferative networks of endometrial carcinogenesis. Studies from in vitro and in vivo model systems have defined that ERα mediates the induction of cell proliferation, whereas ERβ inhibits proliferation and induces apoptosis (Weihua et al., 2003; Cheng et al., 2004). Although the endometrial cells express both isoforms of the ER, only the E2- or Tam-activated ERα can be recruited to the PAX-2 promoter to activate it (Wu et al., 2005), sustaining the idea that the balance between ERα and ERβ expression dictates the proliferative outcome of the cell (i.e. a Ying/Yang relationship) (Matthews & Gustafsson, 2003; Acconcia et al., 2005b).

On the other hand, the Sp-1- and NFκB-dependent activation of the PAX-2 promoter suggests a non-classical mode of action of E2 and Tam within endometrial cells (Nilsson et al., 2001). Remarkably, E2 and Tam induce actin cytoskeleton remodeling, cell motility and migration of endometrial cancer cells through ERα via a non-classical mechanism (Acconcia et al., 2005a). Thus, in addition to cell proliferation, PAX-2 might mediate the effects of E2 and Tam on endometrial cell migration. In this respect Wu et al., (2005) reported nebulette and cadherin-8 to be up-regulated by E2 and Tam in endometrial cancer cells. Notably these two cytoskeleton proteins have a potential role in cell motility (Moncma & Wang, 1995; Kido et al., 1998).

Within a physiological setting (i.e., normal endometrium), the PAX-2 promoter is methylated on specific CpG islands. Upon DNA methylation, an inhibitor complex containing MeCP, mSin3A and HDAC is recruited onto the PAX-2 promoter. Consequently, ERα is not associated with the PAX-2 promoter, which is in turn insensitive to E2 and Tam. On the other hand, in endometrial cancer cells, the PAX-2 promoter is hypomethylated. The loss of this signal results in the dissociation of the inhibitor complex upon and within the E2-and Tam-induced recruitment of ERα to the PAX-2 promoter (Wu et al., 2005).

Epigenetic variations such as DNA methylation are critical within the development of cancer. Among the established DNA methylation patterns, hypermethylation has long been considered to be critical for cell proliferation, and indeed an important role for methylation in the inhibition of tumor-suppressor genes in cancer cells has been reported (Feinberg & Tycko, 2004). However, oncogenes such as HRAS or cyclin D2 are activated by promoter hypomethylation in tumor cells. Remarkably, cancers of the uterus have a high frequency of hypomethylated sites (Feinberg & Tycko, 2004). Consistent with this model, Wu et al., (2005) reported that PAX-2 de-methylation in endometrial cancer cells allows PAX-2 to become sensitive to the proliferative actions of E2 and Tam. However, the authors did not show any effect of either E2 or Tam on the DNA methylation pattern of either normal or transformed endometrial cells.

Unopposed estrogenic stimulation and the long-term use of tamoxifen have been recognized as risk factors for the development of endometroid carcinomas (Amant et al., 2005). However these molecules promote, but do not initiate cancer progression. In fact, a loss of methylation appears to be a prerequisite for the transformation of a normal endometrial cell into a cancer cell. These changes may lead to PAX-2 reactivation and to an increased overall sensitization of gene expression and, in turn, to enhanced cell proliferation (Wu et al., 2005). The over-expression of invasion-associated proteins (i.e., PAX-2, nebulette, cadherin-8) (Wu et al., 2005), as well as the activation of specific signal pathways directly regulating actin cytoskeleton remodeling (Acconcia et al., 2005a) can further dictate the E2- and Tam-mediated invasive and metastatic behavior of endometrial cancer cells. These findings suggest that PAX-2 may be a critical regulatory molecule during the progression of endometrial cancers, and its activity status may potentially be used for predictive and diagnostic purposes.

References

Acconcia, F., Barnes, C.J., and Kumar, R. (2005a) Estrogen and Tamoxifen Induce Cytoskeletal Remodeling and Migration in Endometrial Cancer Cells. Endocrinology Dec 8; [Epub ahead of print].

Acconcia, F., Totta, P., Ogawa, S., Cardillo, I., Inoue, S., Leone, S., Trentalance, A., Muramatsu, M., and Marino, M. (2005b) Survival versus apoptotic 17beta-estradiol effect: role of ER alpha and ER beta activated non-genomic signaling. J. Cell. Physiol. 203, 193-201.

Amant, F., Moerman, P., Neven, P., Timmerman, D., Van Limbergen, E., and Vergote, I. (2005) Endometrial cancer. Lancet 366, 491-505.

Cheng, J., Lee, E.J., Madison, L.D., and Lazennec, G. (2004) Expression of estrogen receptor beta in prostate carcinoma cells inhibits invasion and proliferation and triggers apoptosis. FEBS Lett. 566, 169-172.

Feinberg, A.P., and Tycko, B. (2004) The history of cancer epigenetics. Nat. Rev. Cancer 4, 143-153.

Kido, M., Obata, S., Tanihara, H., Rochelle, J.M., Seldin, M.F., Taketani, S., and Suzuki, S.T. (1998) Molecular properties and chromosomal location of cadherin-8. Genomics 48, 186-194.

Lowe, S.W., Cepero, E., and Evan, G. (2004) Intrinsic tumor suppression. Nature 432, 307-315.

Marino, M., Acconcia, F., and Ascenzi, P. (2005) Estrogen receptor signalling: bases for drug actions. Curr. Drug. Targets Immune Endocr. Metabol. Disord. 5, 305-314.

Matthews, J., and Gustafsson, J.A. (2003) Estrogen signaling: a subtle balance between ER alpha and ER beta. Mol. Interv. 3, 281-292.

Moncman, C.L., and Wang, K. (1995) Nebulette: a 107 kD nebulin-like protein in cardiac muscle. Cell Motil. Cytoskeleton. 32, 205-225.

Robson, E.J., He, S.J., and Eccles, M.R. (2006) A PANorama of PAX genes in cancer and development. Nat. Rev. Cancer 6, 52-62.

Schlessinger, K., and Hall, A. (2004) GSK-3beta sets Snail's pace. Nat. Cell Biol. 6, 913-915.

Weihua, Z., Andersson, S., Cheng, G., Simpson, E.R., Warner, M., and Gustafsson, J.A. (2003) Update on estrogen signaling. FEBS Lett. 546, 17-24.

Wu, H., Chen, Y., Liang, J., Shi, B., Wu, G., Zhang, Y., Wang, D., Li, R., Yi, X., Zhang, H., Sun, L., and Shang, Y. (2005) Hypomethylation-linked activation of PAX2 mediates tamoxifen-stimulated endometrial carcinogenesis. Nature 438, 981-987.