Jonathan J. Li

Estrogen mediates Aurora-A overexpression, centrosome amplification, chromosomal instability, and breast cancer in female ACI rats

Estrogens play a crucial role in the causation and development of sporadic human breast cancer (BC). Chromosomal instability (CIN) is a defining trait of early human ductal carcinoma in situ (DCIS) and is believed to precipitate breast oncogenesis. We reported earlier that 100% of female ACI (August/Copenhagen/Irish) rats treated with essentially physiological serum levels of 17β-estradiol lead to mammary gland tumors with histopathologic, cellular, molecular, and ploidy changes remarkably similar to those seen in human DCIS and invasive sporadic ductal BC. Aurora-A (Aur-A), a centrosome kinase, and centrosome amplification have been implicated in the origin of aneuploidy via CIN. After 4 mo of estradiol treatment, levels of Aur-A and centrosomal proteins, γ-tubulin and centrin, rose significantly in female ACI rat mammary glands and remained elevated in mammary tumors at 5–6 mo of estrogen treatment. Centrosome amplification was initially detected at 3 mo of treatment in focal dysplasias, before DCIS. At 5–6 mo, 90% of the mammary tumor centrosomes were amplified. Comparative genomic hybridization revealed nonrandom amplified chromosome regions in seven chromosomes with a frequency of 55–82% in 11 primary tumors each from individual rats. Thus, we report that estrogen is causally linked via estrogen receptor α to Aur-A overexpression, centrosome amplification, CIN, and aneuploidy leading to BC in susceptible mammary gland cells.

Induction of chromosone aberrations in Syrian hamster renal cortical cells by various estrogens

Estrogens, both natural and synthetic, have been implicated in carcinogenesis at different organ sites in a variety of animals, including man, for more than six decades. However, the molecular mechanism(s) involved in the carcinogenic action of estrogens still remains both controversial and elusive. Cytogenetic damage in the hamster kidney has been studied after in vivo treatment with either potent or weak estrogens for varying periods. Compared to age-matched untreated control, diethylstilbestrol (DES) treatment resulted in significant increases in the number of chromatid gaps and breaks, chromosome breaks, and endoreduplicated cells in hamster renal cortical cells. These chromosomal aberrations (CA) were cumulative with continued hormone exposure from 1.0 to 5.0 months. However, chromosome exchanges as a result of the breaks were not elevated. After 5.0 months of hormone treatment, potent estrogens such as 17 beta-estradiol and Moxestrol exhibited similar frequencies of CA in the hamster kidney to that found for DES, whereas weak estrogens such as 17 alpha-estradiol and beta-dienestrol exhibited CA frequencies that were not significantly different from untreated levels. Ethinylestradiol treatment for a similar period resulted in significant increases in chromatid gaps, although these did not evolve into increases in either chromatid or chromosome breaks, and in a rise in endoreduplicated cells. These results raise the possibility that the CA generated after estrogen treatment may be involved in renal tumorigenic processes.

Ploidy differences between hormone- and chemical carcinogen–induced rat mammary neoplasms: Comparison to invasive human ductal breast cancer*

To ascertain differences between solely hormone– and chemical carcinogen–induced murine mammary gland tumors (MGTs), a direct comparison of their ploidy status was assessed. Nuclear image cytometry (NIC) was used to evaluate ploidy in ductal carcinoma in situ (DCIS) and MGTs induced solely by 17β‐estradiol (E2) in female A‐strain Copenhagen Irish hooded gene rats (ACI) and E2 plus testosterone propionate in male Noble rats. These results were compared to ploidy data from primary MGTs induced by two synthetic carcinogens, 7,12‐dimethylbenz[a]antracene and nitrosomethylurea in female Brown Lewis Norway rats and an environmental carcinogen, 6‐nitrochrysene, in female Sprague‐Dawley rats. Both DCIS and primary MGTs induced solely by hormones were highly aneuploid (> 84%), whereas MGTs induced by either synthetic or environmental carcinogens were primarily diploid (> 85%). Examination of 76 metaphase plates obtained from eight individual E2‐induced ACI female rat MGTs revealed the following consistent chromosome alterations: gains in chromosomes 7, 11, 12, 13, 19, and 20 and loss of chromosome 12. On Southern blot analysis, six of nine ACI female rat primary E2‐induced MGTs (66%) exhibited amplified copy numbers (range: 3.4–6.9 copies) of the c‐myc gene. Fluorescence in situ hybridization (FISH) analysis of these MGTs revealed specific fluorescent hybridization signals for c‐myc (7q33) on all three homologs of a trisomy in chromosome 7. NIC analysis of 140 successive nonfamilial sporadic invasive human ductal breast cancers (BCs) showed an aneuploid frequency of 61%, while 31 DCISs revealed a 71% aneuploid frequency. These results clearly demonstrate that the female ACI rat E2‐induced MGTs more closely resemble invasive human DCIS and ductal BC in two pertinent aspects: they are highly aneuploid compared with chemical carcinogen–induced MGTs and exhibit a high frequency of c‐myc amplification.

Aurora A and B overexpression and centrosome amplification in early estrogen-induced tumor foci in the Syrian hamster kidney: Implications for chromosomal instability, aneuploidy, and neoplasia

Estrogen-induced Syrian hamster tumors in the kidney represent a useful model to gain insight into the role of estrogens in oncogenic processes. We provided evidence that early tumor foci in the kidney arise from interstitial ectopic uterine-like germinal stem cells, and that early tumor foci and well-established tumors are highly aneuploid (92-94%). The molecular mechanisms whereby estrogens mediate this process are unclear. Here, we report that estrogen treatment induced significant increases in Aurora A protein expression (8.7-fold), activity (2.6-fold), mRNA (6.0-fold), and Aurora B protein expression (4.6-fold) in tumors, compared with age-matched cholesterol-treated kidneys. Immunohistochemistry revealed that this increase in Aurora A and B protein expression was essentially confined to cells within early and large tumor foci at 3.5 and 6 months of estrogen treatment, respectively. Upon estrogen withdrawal or coadministration of tamoxifen for 10 days, a 78% to 79% and 81% to 64% reduction in Aurora A and B expression, respectively, were observed in primary tumors compared with tumors continuously exposed to estrogens. These data indicate that overexpressed Aurora A and B in these tumors are under estrogen control via estrogen receptor alpha. Aurora A coenriched with the centrosome fraction isolated from tumors in the kidney. Centrosome amplification (number and area/cell) was detected in early tumor foci and large tumors but not in adjacent uninvolved or age-matched control kidneys. Taken together, these data indicate that persistent overexpression of Aurora A and B is under estrogen control, and is coincident with centrosome amplification, chromosomal instability, and aneuploidy, and represent an important mechanism driving tumorigenesis.

Estrogen-Induced Aurora Kinase-A (AURKA) Gene Expression is Activated by GATA-3 in Estrogen Receptor-Positive Breast Cancer Cells

Aurora-A is a proto-oncogenic mitotic kinase that is frequently overexpressed in human epithelial malignancies including in breast and ovarian cancers. The mechanism of transcriptional upregulation of Aurora-A in human breast cancer is not yet elucidated. We report herein that Aurora-A transcription is positively regulated by GATA-3 in response to estrogen in estrogen receptor α (ERα)-positive cells. Transient expression of aurora-A promoter deletion mutants in luciferase constructs identified a GATA binding sequence motif as a functional regulatory element in ERα-positive breast cancer cells. Electrophoretic mobility shift assay identified the binding of regulatory proteins to the GATA element. Anti-GATA-3 antibody generated a supershifted complex. Recruitment of GATA-3 to the aurora-A promoter was verified by chromatin immunoprecipitation analysis with GATA-3 antibody. Ectopic expression of GATA-3 resulted in elevated expression of Aurora-A in both ERα-positive and negative cells while siRNA-mediated silencing led to downregulation of endogenous Aurora-A in ERα-positive cells. Estrogen treatment of ERα-positive cells induced increased Aurora-A expression with enhanced recruitment of GATA-3 to the aurora-A promoter. Finally, in the ACI rat model of estrogen-induced breast cancer, known to be associated with elevated Aurora-A expression, we observed increased expression of GATA-3 in preinvasive and invasive mammary epithelial cells exposed to prolonged estrogen treatment and in developing breast tumors. These results demonstrate a direct positive role of estrogen in regulating Aurora-A expression through activation of the ERα-GATA-3 signaling cascade and suggest that this pathway may be critical in the origin of estrogen-stimulated sporadic breast cancer.

Serum and tissue levels of estradiol during estrogen-induced renal tumorigenesis in the Syrian hamster

The estrogen-induced renal tumor in the hamster has emerged as a major animal model in hormonal carcinogenesis. However, a fundamental aspect of this experimental model has as yet not been investigated. In the present study, comparisons between the serum and tissue 17 beta-estradiol (E2) levels in cyclic female hamsters and corresponding hormone levels in E2-treated castrated male hamsters have been made. Data is provided concerning the concentration of estrogenic hormones in the serum and target tissue typically required to elicit renal tumorigenesis in this species. Serum E2 levels in the cyclic female hamster average 79 pg/ml on days 1-2 and 311 pg/ml on days 3-4, attaining a maximum of 358 pg/ml on day 4 of the cycle. Elevation in uterine, renal and hepatic E2 tissue levels during days 3-4 of the cycle reflect increases in serum E2 levels which were 3.0-, 2.0-, and 2.6-fold higher when compared to day 1 of the cycle in these tissues. As expected, serum E2 levels of untreated castrated male hamsters did not appreciably vary over a 6 month period of aging and averaged about 32 pg/ml. Under conditions which produced essentially 100% renal tumor incidence, a rapid rise in serum E2 levels, averaging 71.0-fold higher than untreated castrated levels, was seen. A steady state serum E2 level of 2400 to 2700 pg/ml was maintained from 45-180 days of continuous estrogen treatment. Compared to kidneys of untreated hamsters, renal E2 levels in E2-treated hamsters rose only on average 5.4-fold between 15-180 days of hormone exposure. Serum levels of E2-treated hamsters were 5.7- to 8.0-fold higher than those observed in cyclic female hamsters on days 3 and 4. However, at these higher E2-treated serum levels there was no apparent effect either on weight loss or mortality of the animals.

Comparative genomic hybridization of estrogen-induced ectopic uterine-like stem cell neoplasms in the hamster kidney: nonrandom chromosomal alterations.

Karyotype and comparative genomic hybridization (CGH) analyses were performed to identify nonrandom/consistent chromosomal alterations in solely estrogen (E)‐induced primary ectopic uterine‐like stem cell tumors in the kidney (EULTK) of the Syrian hamster, using a criterion of ≥20% frequency for nonrandom occurrence. Employing this criterion, EULTK karyotype analysis showed consistent gains in chromosomes 3, 6, 11, 14, 16, 20, and 21. Consistent trisomies were seen in all of these nonrandomly gained chromosomes. Only chromosomes 3 and 6 exhibited appreciable tetrasomies. Chromosome losses were observed consistently in chromosomes 7, 12, 17, and 19. Employing the same criterion, CGH analysis of primary EULTKs showed nonrandom amplified sequences at 1pa1–a4, 2cen–pter, 3pa1–a4, 6qb2–b4, 20qa1–a4, 21qa1–a2, Xqa3–qter and regional consistent losses at 1qc1‐qter, 2qb1–c1, 3qa2–a7, 11qb5‐qter, 15qa2–a5, 18qa2–a4, and 21pa. Moreover, 88% of the EULTKs examined exhibited amplification of the 6qb2–b4 region, where c‐myc resides. The data presented lend credence to the supposition that chromosomal instability (CIN) is elicited by the upstream overexpression and subsequent amplification of c‐myc by Es in multipotential interstitial uterine stem cells present in the kidney, thus leading to neoplastic development.

Specific Overexpression of Cyclin E·CDK2 in Early Preinvasive and Primary Breast Tumors in Female ACI Rats Induced by Estrogen

Overexpressed Aurora A, amplified centrosomes, and aneuploidy are salient features of estrogen-induced mammary preinvasive lesions and tumors in female August–Copenhagen Irish (ACI) rats. Intimately involved in these events are cyclins and their associated cyclin-dependent kinase (CDK) partners. Cyclin E1·CDK2 overexpression plays an important dual role in late G1/S phase of the cell cycle in cancer cells. It increases DNA replication providing growth advantage to cancer cells and facilitates aberrant centrosome duplication, generating chromosomal instability and aneuploidy leading to tumor development. Presented herein, a 24.0- and 45.0-fold elevation in cyclin E1 and CDK2 was found in 17β-estradiol (E2)-induced ACI rat mammary tumors (MTs), respectively. Cyclin E·CDK2 positive staining was confined to the large round cells found within focal dysplasias, ductal carcinomas in situ, and invasive MTs. Co-immunoprecipitation and in vitro kinase activity of these tumors revealed that these cell cycle entities are functional. When mammary tissue derived from untreated normal, E2-induced hyperplasia and primary tumors were normalized to cyclin E1 levels, low molecular weight (LMW) cyclin E1 forms (33- and 45-kDa) were detected in all of these tissue groups. Moreover, increasing concentrations of protease inhibitor in tissue lysates resulted in a marked reduction of LMW forms, indicating that the presence of cyclin E1 LMW forms can be markedly reduced. Significant increases in cyclin E1 mRNA (2.1-fold) were detected in primary ACI rat E2-induced breast tumors, and quantitative real-time polymerase chain reaction revealed a 20% amplification of the cyclin E1 gene (CCNE1). Collectively, these results support the involvement of cyclin E1·CDK2 in centrosome overduplication during each stage of E2-induced mammary tumorigenesis.

Expression of selected aurora a kinase substrates in solely estrogen-induced ectopic uterine stem cell tumors in the syrian hamster kidney

Sustained over-expression of Aurora A (AurA), centrosome amplification, chromosomal instability, and aneuploidy are salient features that occur in high frequency in human breast premalignant stages and in primary ductal breast cancer (BC), as well as in 17beta-estradiol (E2)-induced oncogenesis in animal models. We have reported that AurA/B protein expression increases 8.7- and 4.6-fold, respectively, in primary E2-induced male Syrian hamster uterine stem cell-like tumors of the kidney (EUTK) when compared with cholesterol-treated control kidneys. Upon a 10-day E2-withdrawal or coadministration of tamoxifen citrate, a 78-79% and 81-64% reduction in AurA/B protein expression, respectively, were observed in primary tumors when compared with tumors from animals continuously exposed to E2. These data indicate that AurA/B expression is regulated by estrogens via estrogen receptor alpha. To determine whether this E2-induced over-expression of the Aur kinases may contribute to the alterations observed during oncogenesis via their phosphorylation of specific substrates, we analyzed the protein expression of histone H3 and targeting protein for Xklp2 (TPX2). Histone H3 and TPX2 were significantly over-expressed 3.7- and 1.6-fold, respectively, in E2-induced tumors when compared with cholesterol-treated control kidney samples. Immunohistochemistry revealed that TPX2 protein expression was essentially confined to tumor foci cells. Collectively, these data indicate that over-expression of AurA/B is under estrogen control and that the deregulation of Aur kinase protein substrates is implicated in eliciting the alterations observed during oncogenesis.

Estrogen Receptor-mediated Genomic Instability in the Syrian Hamster Kidney: A Critical Event in Hormonal Oncogenesis

Perhaps one of the most unique early hormonal events in estrogen (E) oncogenesis is the elicitation of genomic instability as a consequence of estrogen receptor (ER-α)-mediated cell proliferation. The focus of the studies presented herein is to provide evidence for this contention, and for a multi-stage sequence of events leading to tumor development, employing the Syrian hamster E-induced and -dependent renal neoplasm model.