Clinical & Experimental Metastasis

, Volume 29, Issue 5, pp 493–509

Luminal breast cancer metastasis is dependent on estrogen signaling

  • Vidya Ganapathy
  • Whitney Banach-Petrosky
  • Wen Xie
  • Aparna Kareddula
  • Hilde Nienhuis
  • Gregory Miles
  • Michael Reiss
Research Paper

DOI: 10.1007/s10585-012-9466-4

Cite this article as:
Ganapathy, V., Banach-Petrosky, W., Xie, W. et al. Clin Exp Metastasis (2012) 29: 493. doi:10.1007/s10585-012-9466-4


Luminal breast cancer is the most frequently encountered type of human breast cancer and accounts for half of all breast cancer deaths due to metastatic disease. We have developed new in vivo models of disseminated human luminal breast cancer that closely mimic the human disease. From initial lesions in the tibia, locoregional metastases develop predictably along the iliac and retroperitoneal lymph node chains. Tumors cells retain their epithelioid phenotype throughout the process of dissemination. In addition, systemically injected metastatic MCF-7 cells consistently give rise to metastases in the skeleton, floor of mouth, adrenal glands, as well as in the lungs, liver, brain and mammary fat pad. We show that growth of luminal breast cancer metastases is highly dependent on estrogen in a dose-dependent manner and that estrogen withdrawal induces rapid growth arrest of metastatic disease. On the other hand, even though micrometastases at secondary sites remain viable in the absence of estrogen, they are dormant and do not progress to macrometastases. Thus, homing to and seeding of secondary sites do not require estrogen. Moreover, in sharp contrast to basal-like breast cancer metastasis in which transforming growth factor-β signaling plays a key role, luminal breast cancer metastasis is independent of this cytokine. These findings have important implications for the development of targeted anti-metastatic therapy for luminal breast cancer.


Luminal breast cancerMetastasisEstrogenTransforming growth factor-β



Epithelial-to-mesenchymal transitions


Transforming growth factor-β


Estrogen receptor α


TGF-β type II receptor gene




Estrogen receptor α gene


Progesterone receptor gene


Glyceraldehyde 3-phosphate dehydrogenase gene


Bioluminescence imaging




Tartrate resistant acid phosphatase


Analysis of variance




Estrogen withdrawal


Mammary fat pad


TGF-β response gene signature

Supplementary material

10585_2012_9466_MOESM1_ESM.tif (91.6 mb)
Supplemental Fig. 1. Local tumor growth following intratibia tumor cell injection. MCF-7-derived bone tropic MCF-7-5624 cells were injected into tibiae of nude mice. Tumor growth was monitored in vivo using microCT and BLI. Tumor lesions retained ERα expression, and were characterized by a highly epithelioid phenotype, as demonstrated by expression of cytokeratins (pan-CK) and membrane associated E-cadherin. Moreover, these lesions induced a predominantly osteoblastic response of the surrounding bone, as shown by orange G and phloxine positivity (a measure of new bone formation), while there was little to no evidence of osteolytic activity (TRAP negativity). Comparison was made with the SCP2 bone tropic subclone of the basal-like ERα-negative human breast cancer line, MDA-MB-231. When injected into the tibia, SCP2 cells also gave rise to bone lesions with a phenotype that was quite distinct from that of MCF-7-5624: None of the cells expressed ERα or PR (not shown). In addition, SCP2-induced lesions were associated with significant osteolysis, as evidenced by strong TRAP positivity. Thirdly, SCP2-derived tumors were distinctly less epithelioid as pan-cytokeratin expression was significantly weaker than in MCF-7-5624 lesions and they failed to express E-cadherin on. Supplementary material 1 (TIFF 93765 kb)
10585_2012_9466_MOESM2_ESM.tif (196.5 mb)
Supplementary Fig. 2. Locoregional tumor cell dissemination via retroperitoneal lymphatic channels. A. BLI demonstrated that tumors that arose following intra-tibia inoculation of MCF-7-5624A-GF cells disseminated regionally via the iliac and retroperitoneal lymph node chain. B. Solid cords of tumor cells could be seen in local lymphatic channels in the vicinity of the initial tibia lesions. C. Solid cords of tumor cells are also seen to fill retroperitoneal lymphatic channels and lymph nodes. Ly Ch: Lymphatic channel; Ly No: Lymph node. Supplementary material 2 (TIFF 201227 kb)
10585_2012_9466_MOESM3_ESM.tif (90.7 mb)
Supplementary Fig. 3. Luminal breast cancer cells disseminate as cohesive epithelial clusters. Metastatic MCF-7-5624A-GF cell deposits in the lymphatic vessels, heart and lungs all strongly expressed E-cadherin at the cell membrane, indicating that these tumor cells retain cohesiveness throughout the process of dissemination. Supplementary material 3 (TIFF 92846 kb)
10585_2012_9466_MOESM4_ESM.tif (85.5 mb)
Supplementary Fig. 4. 17β-estradiol stimulates migration of luminal breast cancer cells. Confluent cultures of ERα-positive luminal human breast cancer cells were wounded using a sterile plastic pipet tip. A. The resulting wound was photographed at multiple sites and the wound area calculated using Image J. Tumor cells migrate collectively to close the gap. B. Treatment with 17β-estradiol resulted in a marked acceleration of wound closure compared to treatment with vehicle only (p=0.0045, t test with Welch correction). Supplementary material 4 (TIFF 87528 kb)
10585_2012_9466_MOESM5_ESM.tif (90.5 mb)
Supplementary Fig. 5. Histopathology of systemic metastatic lesions. Representative examples of MCF-7-5624A-GF-derived metastases to the skeleton, adrenal glands, central nervous system, lungs, muscle, lymph nodes and mammary gland (Photomicrographs 200x). Supplementary material 5 (TIFF 92705 kb)
10585_2012_9466_MOESM6_ESM.tif (90.8 mb)
Supplementary Fig. 6. Tumor dormancy in estrogen-deficient animals. Ovariectomized mice were inoculated systemically with MCF-7-5624A-GF cells. At 10 weeks post-inoculation, a small number of metastatic lesions in adrenal glands and mammary fatpads were detectable by BLI (Baseline) (see Figure 4). We then introduced E2 pellets into these animals (Figure 4B). Subsequently, several additional metastases appeared in the skeleton and adrenal glands, indicating that tumor cells had initially seeded those areas and had remained viable but dormant, presumably because of a lack of estrogen. Supplementary material 6 (TIFF 92933 kb)
10585_2012_9466_MOESM7_ESM.xls (124 kb)
Supplemental Table 1. Genes differentially expressed between MCF-7 parental cells and metastatic MCF-7-5624A-GF cells. Gene expression profiling experiments were run in triplicate on the Affymetrix Human Exon 1.0 ST exon microarray platform (1.4 million probes). Using GeneSpring GX 11.5.1 (Agilent Technologies, Inc., Santa Clara, CA, USA), raw exon expression signals were combined and summarized with ExonRMA16 (RMA) using all transcripts (28,829 transcript clusters from RefSeq and full-length GenBank mRNAs). The data were quantile normalized with baseline transformation by the median of all samples. Furthermore, the normalized expression signals were averaged between biological replicates. Gene expression data were first filtered by 10th percentile cut-off, resulting in removal of genes with low signal. The 336 genes shown to be differentially expressed between MCF-7-5624A-GF and MCF-7 parental cells were identified by looking for a significant fold change of >2.0 and an unpaired t test p value with Benjamini Hochberg FDR correction of <0.05. Supplementary material 7 (XLS 124 kb)

Copyright information

© Springer Science+Business Media B.V. 2012

Authors and Affiliations

  • Vidya Ganapathy
    • 1
  • Whitney Banach-Petrosky
    • 1
  • Wen Xie
    • 1
  • Aparna Kareddula
    • 1
  • Hilde Nienhuis
    • 2
  • Gregory Miles
    • 3
  • Michael Reiss
    • 1
  1. 1.Division of Medical Oncology, Department of Internal MedicineUMDNJ-Robert Wood Johnson Medical School and The Cancer Institute of New JerseyNew BrunswickUSA
  2. 2.University Medical Center GroningenGroningenNetherlands
  3. 3.Bioinformatics Shared Resource, Cancer Institute of New JerseyNew BrunswickUSA