- Research article
- Open Open Peer Review
Estrogen receptor, progesterone receptor, interleukin-6 and interleukin-8 are variable in breast cancer and benign stem/progenitor cell populations
BMC Cancervolume 14, Article number: 733 (2014)
Estrogen receptor positive breast cancers have high recurrence rates despite tamoxifen therapy. Breast cancer stem/progenitor cells (BCSCs) initiate tumors, but expression of estrogen (ER) or progesterone receptors (PR) and response to tamoxifen is unknown. Interleukin-6 (IL-6) and interleukin-8 (IL-8) may influence tumor response to therapy but expression in BCSCs is also unknown.
BCSCs were isolated from breast cancer and benign surgical specimens based on CD49f/CD24 markers. CD44 was measured. Gene and protein expression of ER alpha, ER beta, PR, IL-6 and IL-8 were measured by proximity ligation assay and qRT-PCR.
Gene expression was highly variable between patients. On average, BCSCs expressed 10-106 fold less ERα mRNA and 10-103 fold more ERβ than tumors or benign stem/progenitor cells (SC). BCSC lin-CD49f−CD24−cells were the exception and expressed higher ERα mRNA. PR mRNA in BCSCs averaged 10-104 fold less than in tumors or benign tissue, but was similar to benign SCs. ERα and PR protein detection in BCSCs was lower than ER positive and similar to ER negative tumors. IL-8 mRNA was 10-104 higher than tumor and 102 fold higher than benign tissue. IL-6 mRNA levels were equivalent to benign and only higher than tumor in lin-CD49f−CD24−cells. IL-6 and IL-8 proteins showed overlapping levels of expressions among various tissues and cell populations.
BCSCs and SCs demonstrate patient-specific variability of gene/protein expression. BCSC gene/protein expression may vary from that of other tumor cells, suggesting a mechanism by which hormone refractory disease may occur.
Breast cancer treatment options are based partially upon immunohistochemical staining of tissue specimens for the expression of hormone receptors. Expression of estrogen and progesterone receptors leads to specific therapeutic strategies, including tamoxifen and aromatase inhibitors. These strategies have been followed for decades. The data at a 15 year endpoint indicate that 5 years of tamoxifen therapy will reduce the disease recurrence rate 11.8% and the mortality rate 9.8% . These data are encouraging and support continued use of traditional tamoxifen therapy, but the fact that approximately 30% of patients still relapse indicates research to improve outcomes is warranted.
One hypothesis as to why disease recurs in the presence of tamoxifen therapy is that the bulk of the estrogen receptor positive tumor cells are destroyed by treatment, but tumor initiating cells that are negative for estrogen receptor expression persist. Tumor initiating cells, or cancer stem cells, represent a small percentage of cells that make up breast tumors but have the ability to induce growing tumors in immunodeficient mice . Al-Hajj and colleagues demonstrated that as few as 1000 CD44high/CD24low cells isolated from human breast cancer could develop a tumor in immunodeficient mice . However, CD44high/CD24low cells may not be the universal breast cancer stem cell profile, as mammospheres from a pleural effusion lacking CD44high/CD24low cells, and CD49flow/CD24high cells from the infiltrating ductal carcinoma cell line (HCC 1954) could also generate tumors in immunodeficient mice [4, 5]. Furthermore, the CD44high/CD24low cancer stem cell phenotype was shown to be similar to the bipotent progenitor cell phenotype CD49fhigh/MUC1neg, with CD44 and CD49f being widely distributed among mammary epithelial cells and expressed by both luminal restricted and bipotent progenitors . Thus, data generated using CD44high/C24low and CD49flow/CD24high sorted cell populations suggest that mammary repopulating units and/or bipotent progenitor cells may be functioning as cancer stem cells in tumors.
Recent studies suggest that measuring estrogen receptor (ER) and progesterone receptor (PR) gene expression in individual intra- and extra- tumoral cells generates additional clinically relevant information. Aktas and colleagues demonstrated that in 77% of their patients with ER positive tumors (ERpos), circulating tumor cells were negative for ER gene expression . Heterogeneity of hormone expression is well documented in breast cancers  but a detailed correlation of the receptor status of tumor cell subpopulations and clinical impact has yet to be completed. Studies suggest that ER gene expression is low in human CD44/CD24  and mouse CD49f/CD24  sorted cell populations.
Protein expression of ER and PR in tumor samples was historically measured using ligand binding assays [11, 12]. The development of monoclonal antibodies led to utilization of enzyme immunoassays . Advancements in embedding, sectioning and antigen retrieval in tumor specimens contributed to immunohistochemistry becoming the current standard for clinical evaluation of biopsy and tumor specimens . These methods measure ER or PR in whole fixed tumor samples and thereby prohibit the study of live cells. The study presented herein, in contrast, is the first to measure the gene and protein expression of ER and PR in uncultured CD49f/CD24 stem and progenitor sorted cell populations (BCSCs) from freshly isolated benign breast tissue or human invasive ductal carcinomas. The proximity ligation assay for detecting protein expression has been used for years [15, 16], but this study represents the first use of this technology in breast cancer stem/progenitor cells.
A growing body of research indicates that pro-inflammatory cytokines can facilitate tumor growth and metastasis [17, 18]. Interleukin-6 (IL-6) is a key factor in regulating estrogen activity through stimulation of aromatase, steroid sulphatase and 17β-hydroxysteroid dehydrogenase [19, 20]. Studies have also demonstrated a positive correlation between IL-6 and ERα expression in breast tumors in a manner thought to be stem cell mediated [21, 22]. In contrast, Interleukin-8 (IL-8) was shown to have an inverse correlation with ERα expression in breast tumors, and IL-8 increases the invasive potential of breast cancer cells [23, 24]. These data suggest that IL-6 and IL-8 pro-inflammatory cytokines may affect tamoxifen response or aromatase inhibition through modulation of hormone activity. Thus, to further delineate the role that stem cells may play in tumor progression through the evasion of hormone-based therapies, IL-6 and IL- 8 gene and protein expression were measured and correlated with ER and PR expression in BCSC.
Benign and malignant tissue procurement and cell culture
This study was approved by the Oregon Health & Science University institutional review board. Benign and malignant specimens, clinical data and consent to publish clinical details from patients included in this study, were obtained with informed written consent in accordance with an IRB approved protocol. Twenty-nine invasive ductal carcinomas were obtained at the time of mastectomy or lumpectomy prior to neoadjuvant treatment. Thirteen pathologically confirmed benign breast tissue specimens were obtained from reduction mammoplasty. ER and PR tumor status were obtained from pathological evaluation of biopsy specimens according to ASCO guidelines . MCF10A (ATCC, CRL-10317) and breast cancer cell lines, MCF7 (ATCC, HTB-22), T47D (ATCC, HTB-133) and HCC1806 (ATCC, CRL-2335) were authenticated by ATCC and confirmed through morphological examination and growth curve analysis. Cell lines were maintained as recommended by ATCC.
Collection of breast cancer stem/progenitor cells (BCSCs)
All specimens were minced and digested in mammary epithelial cell-specific medium containing 1× collagenase/hyaluronidase (Epicult, StemCell Technologies). Cell lines were cultured in Roswell Park Memorial Institute (RPMI)-1640 supplemented with 10% serum and 0.05% Gentamicin. Approximately 106 cells were labeled with monoclonal antibodies against human CD45-FITC, CD31-FITC, CD24-PE, CD49f-PE-Cy 5, and CD44-PE-Cy7. Isotype control testing excluded nonspecific binding. Surface antibody labeling and collection by discriminatory gating were used to remove CD31+/CD45+ endothelial cells and leukocytes (lineage negative; linneg) and to collect four linneg populations of benign and malignant SCs: CD49f+CD24+ (PP), CD49f+CD24−(PM), CD49f−CD24+(MP), and CD49f−CD24−(MM). CD44 expression was measured.
PCR amplification of genetic material
Gene expression in BCSCs and benign stem cells (SCs) was determined by quantitative real-time PCR using Taqman low density array (TLDA) technology (Life technologies, Carlsbad, CA). RNA was isolated using the Qiagen Mini RNeasy kit (Qiagen, Valencia, CA). cDNA was produced using random hexamers (Superscript III First Strand Kit, Invitrogen). An average of 50 ng of cDNA, 15 μl TaqMan’s PreAmp Master Mix (2×) (Applied Biosystems) and 7.5 μl of TaqMan custom PreAmp Pool (Applied Biosystems) were combined. cDNA was amplified for 14 cycles (95°C 10 min, 95°C 15 s, 60°C 4 min). Pre-amplified cDNA was utilized as per manufacturer’s protocol using custom TLDA cards on the Viia7 Real-Time PCR system. Data were included in the analyses if the endogenous control 18S rRNA had a Ct value of 28 or less and triplicate values were within 0.5 Ct of each other. Delta Ct (dCt) values were calculated by subtracting the 18S rRNA Ct value from the target Ct value. Thus, dCt values are inversely related to gene expression (i.e. negative dCt values indicate high levels of gene expression).
Preparation of protein lysates and the proximity ligation assay (PLA)
Given the rarity of BCSCs and the small size of some breast cancers, traditional western blot analysis of protein expression was not possible in this study. As an alternative approach, proximity-dependent DNA ligation assays (PLA) were utilized to detect protein expression [15, 16]. PLAs were conducted according to manufacturer’s protocol (PLA, Life Technologies, Carlsbad, CA) with the following modifications. Approximately 50,000 cells were lysed in 100 μl total volume and serially diluted. For sorted cell populations with less than 50,000 cells available, lysis volume was reduced to 50 μl. Samples were run in triplicate. IL-6 and IL-8 antibody probes (IL-6 BAF206; IL-8 BAF391, biotinylated polyclonal goat, R&D Systems) were made as per manufacturer’s protocol (Life Technologies). ERα and PR antibodies (ERα, AF5715; PR-AF5415; sheep polyclonal, R&D Systems, Minneapolis, MN) were biotinylated using Biotin-XX Microscale Protein labeling kit (B30010, Life Technologies). ERβ antibody (S2015, polyclonal rabbit, Epitomics, Burlingame, CA) was desalted before biotinylation. Amplification was performed (ABI Viia7 RT-PCR system), and dCt values were calculated by subtracting the sample Ct from the no protein control Ct. In contrast to gene expression analyses, a positive dCt value correlates with an increase in protein detection over background.
Statistical analyses were conducted in the form of two-tailed Student’s t-Test with p ≤ 0.05 values considered significant. Pairing was utilized when comparing sorted cells to the tissue or tumor of origin. Unpaired analyses with unequal variance were performed when comparing tumors or tumor sorted cells to benign tissues or benign sorted cells.
Estrogen receptor gene expression in tumors correlated with pathological IHC analyses
Table 1 lists the ER and PR status of the breast cancers included in this study. Tumor hormone status was determined as part of the routine diagnostic testing for all breast tumor biopsies by immunohistochemical (IHC) staining of paraffin-embedded tissue samples as per ASCO guidelines (14). Estrogen receptor alpha (ERA) mRNA was measured in tumors for which a pathologic ER status was known. In tumors determined ER positive by IHC (ERpos), detection of ERA expression was ten-fold higher than in benign tissue (Figure 1). Moreover, detection of estrogen receptor beta (ERB) mRNA was more than 10-fold less in ERpos tumors than benign tissue. ER negative (ERneg) tumors exhibited similar levels of ERA and ERB compared to benign tissue.
CD44 expression is highest in CD49+CD24+cells
Four linneg cell populations were collected from each benign tissue or tumor sample. The linneg sorted cell populations were CD49f+CD24+(PP), CD49f+CD24−(PM), CD49f−CD24+(MP), and CD49f−CD24−(MM). Measurement of CD44 expression indicated that in 80% of tumors, CD49f+CD24+(PP) cell populations were greater than 75% CD44 positive (51-99%, Figure 2A); in contrast, CD49f+CD24−(PM), CD49f−CD24+(MP), and CD49f−CD24−(MM) cell populations exhibited a range of CD44 expression: PM (range: 11-84%), MP (range: 20-92%), and MM (range: 11-84%). Even with this range of expression, CD44 levels detected in these three BCSC populations were significantly lower than CD44 levels in CD49f+CD24+(PP) cell populations. Benign CD49f+CD24+(PP) cells were significantly less positive for CD44 expression than BCSCs (62-100% p = 0.036, Figure 2B). CD44 levels in benign CD49f+CD24−(PM), CD49f−CD24+(MP), and CD49f−CD24−(MM) cells also exhibited a range of CD44 expression: PM (range: 30-83%), MP (range: 2-89%) and MM (range: 9-85%), but again were significantly lower than CD44 levels in benign CD49f+CD24+(PP) cells.
Estrogen receptor gene expression was variable in human BCSCs, but highest in CD49f−CD24−cells
Detection of ERA and ERB mRNA in sorted cell populations isolated from ERpos tumors, ERneg tumors and benign tissues is presented in Figure 3. The delta Ct (dCt) data, which are inversely correlated with expression, indicate that ERA and ERB expression levels in BCSCs and benign SCs were highly variable between patient samples (dCt range: −4 to 20, Figure 3A, B). dCt data were analyzed to generate fold change (Rq) comparisons between BCSCs and the tumor of origin. In ERpos tumors, 70% (17/22) of BCSCs expressed 10-106 fold less ERA than tumor of origin (Figure 3B), while 57% of BCSCs expressed 10-103 fold more ERB than tumor of origin (Figure 3F). In ERneg tumors 50% of BCSCs expressed 103 fold less ERA and ERB than tumor of origin (Figure 3B, F). When compared to benign tissue or benign SCs, ERA expression was 10-105 fold lower in 72% of BCSCs from ERpos tumors and 10-104 fold lower in 85% of BSCSs from ERneg tumors (Figure 3C, D). Seventy-three percent of BCSCs from ERpos and 85% from ERneg tumors, expressed 10-105 fold less ERB than benign tissue (Figure 3G). But when compared to benign SCs, BCSC ERB expression was higher in 50% of ERpos and 30% of ERneg tumors (Figure 3H). Of note, the CD49fneg populations were the exception in which detection of ERA was higher in CD49f−CD24−(MM) BCSC than tumor regardless of tumor status (Figure 3B), and detection of ERB was higher in both CD49f−CD24+(MP) and CD49f−CD24−(MM) populations compared to tumor of origin and compared to benign SCs (Figure 3F and H).
PR gene expression did not correlate with ER expression
PR gene expression levels in BCSCs, benign SCs and tumor or tissue of origin are shown in Figure 4. Estrogen is a transcriptional activator of progesterone receptor (PR) ; therefore, the presence of functional ER protein is expected to correlate with increased levels of PR message. In this study, PR expression was generally similar between benign tissue and tumors regardless of ER status. Detection of PR was significantly higher in benign tissue than in seven BCSC and two benign SC populations (Figure 4A). PR in 85% of BCSCs from ERpos tumors was 10–10,000 fold less and PR in 62% of BCSCs from ERneg tumors was about 100 fold less than in tumor of origin (Figure 4B). Detection of PR in 90% of BCSCs from ERpos and ERneg tumors was 10–10,000 fold lower than in benign tissue (Figure 4C). Comparison of PR expression between BCSCs and benign SC reveals more similarly in levels of expression than those seen for ERA (Figure 4D).
IL-6 and IL-8 genes were differentially expressed in BCSCs
Finally, because the presence of IL-6 and IL-8 in tumor cells may be surrogate markers for ER activation [21, 23], IL-6 and IL-8 mRNA levels (IL-6, IL-8) were examined (Figure 5). Experiments reveal that IL-6 expression was comparable with 18S rRNA in benign tissues and ERneg tumors (Figure 5A). ERpos tumors exhibited a range of IL-6 expression (dCt -2 to 22) that was usually lower than 18S (Figure 5A). When compared to tumor of origin IL-6 expression was significantly elevated (10-106 fold) in the CD49f−CD24−(MM) population and a majority of CD49f+CD24−(PM) populations, while significantly decreased an average 100 fold in the CD49f+CD24+(PP) populations when compared to tumor (Figure 5B). When compared to benign tissue, IL-6 expression was 10 fold greater in the CD49f−CD24−(MM) population, but significantly lower in the CD49f−CD24+(MP) (1-104 fold) and CD49f+CD24+(PP) populations (5-106 fold) (Figure 5C). Interestingly, a bimodal expression pattern was observed in the CD49f+CD24−(PM) population, with six specimens exhibiting about 10 fold increased expression and six specimens exhibiting 104 fold decreased expression compared to benign tissue (Figure 5C). When BCSCs were compared benign SCs, IL-6 was elevated 20 fold in the CD49f−CD24−(MM) population, but a range of expression was detected in CD49f+CD24+ (PP) (104 to 10−4 fold) and CD49f−CD24+(MP) (102.5 to 10−2 fold) populations. A bimodal pattern of expression was again observed in the CD49f+CD24−(PM) populations (20 fold vs. 10−4 fold) (Figure 5D).
IL-8 expression was variable in benign tissue and tumor samples (dCt range: −10 to 20) (Figure 5E). On average, more IL-8 was detected in sorted cells than in whole tumor or tissue. IL-8 expression in benign SCs and BCSCs from ERpos tumors was variable while BCSCs from ERneg tumors exhibited consistently higher levels of IL-8 mRNA than 18S. Fold change analyses revealed significantly elevated levels of mRNA expression when compared to tumor of origin in the CD49f−CD24−(MM) population (10-105 fold increases). Compared to tumor, IL-8 levels were on average 10 fold higher in CD49f+CD24−(PM) cells and 5 fold higher in CD49f−CD24+(MP) cells. CD49f+CD24+(PP) cells exhibited highly variable (10−5 - 104) IL-8 expression (Figure 5F). IL-8 expression was 20–100 fold higher in BCSCs than in benign tissue for most samples (CD49f−CD24−(MM) population, p < 0.05) (Figure 5G). Finally, when BCSCs were compared to benign SCs, a 100 fold increase in IL-8 expression in the CD49f−CD24−(MM) and CD49f−CD24+(MP) populations, a 20 fold increase in CD49f+CD24−(PM) cells, and a broader range of expression in the CD49f+CD24+(PP) population (10−2-102) were observed (Figure 5H).
Protein expression was determined by proximity ligation assay (PLA)
Protein expression was determined for ER, PR, IL-6 and IL-8 in freshly isolated BCSCs and benign SC (Figure 6) and compared to gene expression data. Breast cancer cell lines MCF7 and T47D were used as positive controls for ERα, ERβ, and PR, and as a negative control for IL-8neg (Figure 6A). HCC1806 cells served as a negative control for ERα, ERβ, and PR and as a positive control for IL-8. IL-6 was not detected in MCF7, T47D or HCC1806, but was expressed in a tissue control (84 M). ERα, ERβ, PR, IL-6 and IL-8 expression were determined by serial titration of cell lysates followed by PLA. Titrations of cell lysates established the relationship between cell number and protein detection. Sensitivity of detection was determined for each probe. The region in between the lowest positive value detected and the highest level of background detected was defined as equivocal. This area is indicated by a grey box in Figure 6C-F. Values above the box were considered positive for expression while values below the box were considered negative. In agreement with western blot studies , PLA indicated that MCF7 and T47D expressed ERα, and that MCF7 cells had higher amounts of ERα than T47D cells (Figure 6A).
The method of cell isolation influences PLA results
In contrast to expectations, levels of ERα protein in ERpos tumor lysates and unsorted benign tissue lysates were both comparable to no-protein controls when measured following overnight digestion with collagenase/hyaluronidase ([27, 28] (Figure 6B). Detection of IL-6 and IL-8 in these enzyme treated lysates provided evidence that these samples as a whole are not degraded. When lysates were made from tissue that had been pulverized in liquid nitrogen, higher levels of detection of ERα and PR protein were achieved, i.e. levels comparable with those found in MCF7 lysates (Figure 6B). Detection of ERα and PR were not above background in ERneg tumor lysates whether pulverized or enzyme treated. Interestingly, IL-6 and IL-8 levels were lower in pulverized lysates than in enzyme treated lysates. The data further indicate that the method of lysate preparation may influence the results obtained.
ERα, ERβ and PR protein expression in tumors, benign tissue, SCs and BCSCs
Overall, ERα levels were not significantly different between BCSCs obtained from ERpos, ERneg tumors or benign tissues. However, among BCSCs obtained from ERpos tumors, ERα levels varied from background levels (dCt ≈ 1.0) to levels similar to those detected in ERpos tumors and MCF7 cells (dCt ≈ 3.0-6.0) (Figure 6C). ERα levels in benign SCs were less than benign tissues. The ERα levels in BCSCs from ERneg tumors were comparable to or less than expression detected in ERneg tumors and HCC1806 cell line.
Two different sets of antibodies were used as probes to detect ERβ protein (see Methods), but neither set successfully detected ERβ in the positive control MCF7 nor T47D cells [29–31] (Additional file 1: Figure S1A). In whole tumor or tissue lysates, detection of ERβ was comparable to or less than that found in HCC1806. Given the lack of validation of these probes in positive cell lines, ERβ protein levels could not be completely quantified when measured by PLA. However, detection of ERβ was above no protein control background levels in BCSC and benign SC. Also, BCSCs and benign SCs contained higher levels of ERβ than whole tumor or benign tissue (Additional file 1: Figure S1B).
Significantly higher levels of PR protein were found in benign SCs compared to BCSCs (Figure 6D). Specifically, benign CD49f+CD24+(PP), CD49f−CD24+(MP), and CD49f−CD24−(MM) populations were significantly higher than their ERpos BCSC counterparts, and benign CD49f−CD24+(MP) cells expressed significantly more PR than ERneg CD49f−CD24+(MP) cells. While there were no significant differences in PR expression between ERpos and ERneg BCSCs, more PR protein was detected in ERpos than ERneg tumors.
IL-6 and IL-8 cytokine protein expression in tumors, benign tissue, SCs and BCSCs
IL-6 levels were also high in benign tissue and SCs, but in contrast to PR, detection of IL-6 was higher in CD24neg cells rather than in CD24poscells (Figure 6E). IL-6 protein in ERpos CD49f+CD24+(PP) cells was slightly higher than in CD49f+CD24−(PM) cells while expression in CD49f−CD24+(MP) cells was significantly higher than that in CD49f−CD24−(MM) cells. ERneg BCSCs also expressed less IL-6 than ERneg tumor, but similarity of expression between 112 T tumor and BCSCs precluded significance. Published studies suggest that IL-6 expression correlates with ER expression . However, in our gene expression studies more IL-6 was detected in ERneg tumors than in ERpos tumors, and protein expression was similar in ERpos and ERneg BCSCs (Figures 5 and Figure 6E).
Akin to IL-6 and PR, the highest levels of IL-8 protein were consistently detected in benign tissue and SCs, but comparable levels were also found in some ERneg tumors and corresponding BCSCs (Figure 6F). These levels were similar to or greater than those detected in the positive control cell line HCC 1806 (Figure 6A). Patient to patient variation precluded statistical significance for most comparisons, but significantly less IL-8 was detected in all ERneg BCSCs compared to ERneg whole tumor samples. The IL-8 protein data was in agreement with the mRNA data in that there was great variation in levels of expression within the patient population and that, on average, cells from ERneg tumors had more IL-8 than cells from ERpos tumors.
The findings that breast cancer tumors contain a subpopulation of cells that are not effectively targeted by chemotherapeutic agents and radiation has led to cellular and molecular analyses of benign and cancerous breast tissues [32–34]. In this study we compared the gene and protein expression of ERα and ERβ, PR, IL-6 and IL-8 in cells isolated from invasive ductal carcinomas and benign breast tissue specimens. These data reveal variable levels of hormone receptors and cytokine expression which may explain the inconsistent response of breast cancers to hormone therapies and suggest a mechanism by which some patients experience recurrent disease whereas others achieve long term remission.
The identification and classification of stem and progenitor cell lineages in breast cancer remains under development. Al-Hajj and colleagues focused on cells with the profile CD44+/CD24−, Wicha and colleagues added in the ALDH marker, and Clarke and colleagues isolated cells based on p21CIP1 and Msi-1 expression [3, 35, 36]. In this study we separated cells by CD49f/CD24 expression and measured CD44 expression ( and Figure 2). We found CD49f+CD24+ cells to be primarily CD44+, while all other populations exhibited a range of CD44 expression. Despite variation in stem cell isolation strategies, studies from multiple laboratories report that BCSCs express very little ERA compared to the tumor of origin or to breast cancer cell lines [9, 10, 38, 39]. Data presented here are innovative and expand the field in that we measured gene and protein expression of ERα and ERβ, PR IL-6 and IL-8 in uncultured CD49f/CD24 BCSCs from individual human invasive ductal carcinomas.
A limitation of this study is that we were unable to study BCSC gene and protein expression in the same tumor sample. The rarity of BCSCs and a tumor specimen size on average of 0.2 mg precluded the study of gene and protein expression in the same patients. The average number of cells collected for linneg FACS populations CD49f+CD24+ (PP) and CD49f−CD24+ (MP) was 40,098 and 30,491, respectively. The PP and MP populations were below 50,000 cells on average which required that all FACS cells were used for PLA. Interestingly, in a cell dilution analysis of the sample 102 T we were able to detect IL-8 protein in as few as 5 sorted cells. While this demonstrates the potential sensitivity of this assay, this was not the norm for any other protein. Thus, we could not directly correlate mRNA and protein expression between the same samples, rather gene and protein expression comparisons were made by averaging the results of study populations. We could, however, compare gene and protein expression between BCSCs and their tumor of origin, as well as SCs with their benign tissues of origin.
With the technical limitations in mind, the analysis of these data led to several important conclusions. The variability of gene and protein expression observed in this study reinforces that breast cancers are biologically complex. When data is presented in averages patient-to-patient variability is masked. As we approach the age of personalized cancer care, identifying significant differences between breast cancers will facilitate superior targeted treatment.
Gene expression averages demonstrated low levels of ERA and ERB in BCSCs (Figure 3, bars), while individual data points reveal the range of expression observed between patients (Figure 3, symbols). We detected the highest levels of ERA in ERpos tumors and CD49f−CD24−(MM) cells; ERB was more variable in these populations. Similar to gene expression, ERα protein expression is also varied in sorted cells from ERpos tumors. Interestingly, ERα protein expression in the CD49f−CD24−(MM) population is lower than that detected in the other BCSC populations. Gene and protein expression studies were not conducted in the same cells but the high levels of ERA gene expression and low levels of ERα protein expression in the CD49f−CD24−(MM) population suggest that ERα protein expression in BCSCs may be subject to post-transcriptional regulation as has been demonstrated in cell lines [40–42].
ERneg tumors also exhibited variable ERA and ERB expression, but little ERα protein was detected. Low levels of ERβ protein were detected in ERneg tumors, but it is hard to determine the relevance of this finding, as we could not detect ERβ in reportedly positive cell lines. Some studies report detection of basal levels of ERβ protein in MCF7 and T47D cell lines while others state that ERβ is only present in MCF7 and T47D upon induction [29–31]. We were not able to detect ERβ in MCF7 or T47D by PLA, but in each of these studies, including ours, different antibodies were used.
We cannot yet correlate differences in BCSC gene and protein expression with clinical outcome, as we lack long term patient follow-up. However, given what is known regarding IHC ER staining of breast cancers and treatment response, we suggest that there will be a correlation of patient outcome with BCSC ER status. Our data suggest that BCSCs with PLA values above 2.5 are likely ERpos. They may therefore respond to hormonal treatments in a similar fashion as breast cancer receiving an ERpos IHC evaluation by ASCO guidelines . Examination of BCSC ER by PLA reveals that most ERpos cancers contain BCSCs that do not express ER protein. Individual data points reveal that there is a range of expression between tumors, but that ER expression is negligible in most BCSCs (Figure 6C). The discrepancy of hormone status between various BCSCs and the tumor may have serious therapeutic implications. In theory, treatments targeted to ERpos cells would not affect the ERneg BCSCs in the tumor. This may result in ineffective eradication of BCSCs and the means for tumor recurrence. The patients in this study will be followed to determine the clinical outcomes associated with variable BCSC hormone receptor expression.
In general, PR expression was low in BCSCs and lower than that in tumors and benign tissues. Again, there was variability in expression between patients as well as in the correlation of ER to PR expression. Interestingly, when hormone receptors were compared between BCSC and SC, only PR was significantly higher in expression. The difference in PR protein expression in BCSCs from ERpos tumors versus ERneg tumors was not statistically different. We found no correlation between ERα and PR protein expression. It may be that while expressed, the level of ER activity varies between breast cancers or other untested co-activators such as HER4 are low . We cannot comment further on this, as we did not study the activity of ER or expression of HER4. However, these data agree with other expression studies in mice and humans which also report low levels of PR expression in CD44high/CD24high sorted cells [9, 10].
IL-6 is found in ER positive tumors and is thought to synergize with estrogen to increase ER transcriptional activity [19, 21], while IL-8 is inversely correlated with ER expression . IL-6 has been implicated in maintaining a feedback loop between cancer stem cells and non-stem cancer cells through induction of epithelial-mesenchymal transition , and IL-8 has been implicated in BCSC self-renewal . Thus, increased IL-6 and decreased IL-8 could indicate better responses to tamoxifen or aromatase inhibitors. In this study IL-6 expression was detected in both ERpos and ERneg tumors. Contrary to other studies , IL-6 expression was slightly higher in ERneg tumors. In ERpos tumors, the CD49f−CD24−(MM) population exhibited the highest gene expression but the lowest protein detection suggesting post-transcriptional regulation of IL-6 in BCSCs as demonstrated in HeLa cells . IL-6 protein was highest in benign tissue and CD24neg cells. Studies have shown that ERpos tumors were responsive to IL-6 therapy due to low autocrine levels of IL-6, while ERneg tumors were not responsive potentially due to high autocrine levels of IL-6 . Thus, higher levels of IL-6 found in benign SCs may protect them from unwanted side effects of IL-6 therapy. However, similar to ER PR data, the IL-6 data suggest that tamoxifen and aromatase inhibitors would likely target the largest tumor BCSC population represented by CD49f−CD24−(MM), but not the scarce stem/progenitor populations represented by CD49f+CD24+(PP), CD49f+CD24−(PM) or CD49f−CD24+(MP) cells.
IL-8 gene and protein expression were highly variable in benign tissue and tumors, and in both benign SC and BCSC populations. IL-8 levels were consistent with an inverse correlation between IL-8 expression and ER tumor status. IL-8 levels were higher in benign SC and BCSC populations than benign tissue or whole tumor and highest in the CD49f−CD24−(MM) population. IL-8 protein was lower in benign SCs and BCSCs than in tissue/tumor of origin, but significant levels were still detected. IL-8 has been implicated in regulating the epithelial-mesenchymal transition , and blocking IL-8 signaling selectively depletes ADLH+ stem cells . Thus targeting IL-8 positive BCSCs may benefit patients with high IL-8 levels. In addition the inverse correlation between IL-8 and ERα expression could provide a level of diagnostic confirmation.
Estrogen and progesterone receptors and cytokines IL-6 and IL8 gene and protein expression in tumors and BCSCs among patients was highly variable. In addition, the data presented here indicate that the gene and protein expression of BCSCs may vary from that of other cells within a tumor. Because BCSCs are a rare population of cells within a tumor, they are not accurately tested by random sampling of whole tumor specimens . Thus, from a clinical perspective, determining the gene and protein status of directly isolated BCSCs from each patient tumor may prove to be critical for informed care management.
Breast cancer stem/progenitor cells
- Benign SC:
Benign breast stem/progenitor cells
Estrogen receptor alpha gene expression
Estrogen receptor alpha protein expression
Estrogen receptor beta gene expression
Estrogen receptor beta protein expression
- ERpos :
ER positive tumor by IHC
- ERneg :
ER negative tumor by IHC
Progesterone receptor gene expression
Progesterone receptor protein expression
Interleukin-6 gene expression
Interleukin-6 protein expression
Interleukin-8 gene expression
Interleukin-8 protein expression.
Davies C, Godwin J, Gray R, Clarke M, Cutter D, Darby S, McGale P, Pan HC, Taylor C, Wang YC, Dowsett M, Ingle J, Peto R: Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: patient-level meta-analysis of randomised trials. Lancet. 2011, 378 (9793): 771-784.
Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ, Wicha MS: In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes Dev. 2003, 17 (10): 1253-1270. 10.1101/gad.1061803.
Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF: Prospective identification of tumorigenic breast cancer cells. Proc Natl Acad Sci U S A. 2003, 100 (7): 3983-3988. 10.1073/pnas.0530291100.
Grimshaw MJ, Cooper L, Papazisis K, Coleman JA, Bohnenkamp HR, Chiapero-Stanke L, Taylor-Papadimitriou J, Burchell JM: Mammosphere culture of metastatic breast cancer cells enriches for tumorigenic breast cancer cells. Breast Canc Res: BCR. 2008, 10 (3): R52-10.1186/bcr2106.
Pommier SJ, Quan GG, Christante D, Muller P, Newell AE, Olson SB, Diggs B, Muldoon L, Neuwelt E, Pommier RF: Characterizing the HER2/neu status and metastatic potential of breast cancer stem/progenitor cells. Ann Surg Oncol. 2010, 17 (2): 613-623. 10.1245/s10434-009-0730-z.
Raouf A, Zhao Y, To K, Stingl J, Delaney A, Barbara M, Iscove N, Jones S, McKinney S, Emerman J, Aparicio S, Marra M, Eaves C: Transcriptome analysis of the normal human mammary cell commitment and differentiation process. Cell Stem Cell. 2008, 3 (1): 109-118. 10.1016/j.stem.2008.05.018.
Aktas B, Muller V, Tewes M, Zeitz J, Kasimir-Bauer S, Loehberg CR, Rack B, Schneeweiss A, Fehm T: Comparison of estrogen and progesterone receptor status of circulating tumor cells and the primary tumor in metastatic breast cancer patients. Gynecol Oncol. 2011, 122 (2): 356-360. 10.1016/j.ygyno.2011.04.039.
Lee M, Lee CS, Tan PH: Hormone receptor expression in breast cancer: postanalytical issues. J Clin Pathol. 2013, 66 (6): 478-484. 10.1136/jclinpath-2012-201148.
Shipitsin M, Campbell LL, Argani P, Weremowicz S, Bloushtain-Qimron N, Yao J, Nikolskaya T, Serebryiskaya T, Beroukhim R, Hu M, Halushka MK, Sukumar S, Parker LM, Anderson KS, Harris LN, Garber JE, Richardson AL, Schnitt SJ, Nikolsky Y, Gelman RS, Polyak K: Molecular definition of breast tumor heterogeneity. Cancer Cell. 2007, 11 (3): 259-273. 10.1016/j.ccr.2007.01.013.
Asselin-Labat ML, Shackleton M, Stingl J, Vaillant F, Forrest NC, Eaves CJ, Visvader JE, Lindeman GJ: Steroid hormone receptor status of mouse mammary stem cells. J Natl Cancer Inst. 2006, 98 (14): 1011-1014. 10.1093/jnci/djj267.
Jensen EV, Desombre ER, Hurst DJ, Kawashima T, Jungblut PW: Estrogen-receptor interactions in target tissues. Arch Anat Microsc Morphol Exp. 1967, 56 (3): 547-569.
McGuire WL: Estrogen receptors in human breast cancer. J Clin Invest. 1973, 52 (1): 73-77. 10.1172/JCI107175.
Pasic R, Djulbegovic B, Wittliff JL: Comparison of sex steroid receptor determinations in human breast cancer by enzyme immunoassay and radioligand binding. J Clin Lab Anal. 1990, 4 (6): 430-436. 10.1002/jcla.1860040608.
Hammond ME, Hayes DF, Dowsett M, Allred DC, Hagerty KL, Badve S, Fitzgibbons PL, Francis G, Goldstein NS, Hayes M, Hicks DG, Lester S, Love R, Mangu PB, McShane L, Miller K, Osborne CK, Paik S, Perlmutter J, Rhodes A, Sasano H, Schwartz JN, Sweep FC, Taube S, Torlakovic EE, Valenstein P, Viale G, Visscher D, Wheeler T, Williams RB, et al: American Society of Clinical Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer (unabridged version). Arch Pathol Lab Med. 2010, 134 (7): e48-e72.
Fredriksson S, Dixon W, Ji H, Koong AC, Mindrinos M, Davis RW: Multiplexed protein detection by proximity ligation for cancer biomarker validation. Nat Methods. 2007, 4 (4): 327-329.
Fredriksson S, Gullberg M, Jarvius J, Olsson C, Pietras K, Gustafsdottir SM, Ostman A, Landegren U: Protein detection using proximity-dependent DNA ligation assays. Nat Biotechnol. 2002, 20 (5): 473-477. 10.1038/nbt0502-473.
Korkaya H, Liu S, Wicha MS: Breast cancer stem cells, cytokine networks, and the tumor microenvironment. J Clin Invest. 2011, 121 (10): 3804-3809. 10.1172/JCI57099.
Chin AR, Wang SE: Cytokines driving breast cancer stemness. Mol Cell Endocrinol. 2014, 382 (1): 598-602. 10.1016/j.mce.2013.03.024.
Speirs V, Kerin MJ, Walton DS, Newton CJ, Desai SB, Atkin SL: Direct activation of oestrogen receptor-alpha by interleukin-6 in primary cultures of breast cancer epithelial cells. Br J Cancer. 2000, 82 (7): 1312-1316.
Reed MJ, Purohit A: Breast cancer and the role of cytokines in regulating estrogen synthesis: an emerging hypothesis. Endocr Rev. 1997, 18 (5): 701-715. 10.1210/edrv.18.5.0314.
Fontanini G, Campani D, Roncella M, Cecchetti D, Calvo S, Toniolo A, Basolo F: Expression of interleukin 6 (IL-6) correlates with oestrogen receptor in human breast carcinoma. Br J Cancer. 1999, 80 (3–4): 579-584.
Sansone P, Storci G, Tavolari S, Guarnieri T, Giovannini C, Taffurelli M, Ceccarelli C, Santini D, Paterini P, Marcu KB, Chieco P, Bonafè M: IL-6 triggers malignant features in mammospheres from human ductal breast carcinoma and normal mammary gland. J Clin Invest. 2007, 117 (12): 3988-4002. 10.1172/JCI32533.
Freund A, Chauveau C, Brouillet JP, Lucas A, Lacroix M, Licznar A, Vignon F, Lazennec G: IL-8 expression and its possible relationship with estrogen-receptor-negative status of breast cancer cells. Oncogene. 2003, 22 (2): 256-265. 10.1038/sj.onc.1206113.
Lin Y, Huang R, Chen L, Li S, Shi Q, Jordan C, Huang RP: Identification of interleukin-8 as estrogen receptor-regulated factor involved in breast cancer invasion and angiogenesis by protein arrays. Int J Canc J Int Canc. 2004, 109 (4): 507-515. 10.1002/ijc.11724.
Rokicki J, Das PM, Giltnane JM, Wansbury O, Rimm DL, Howard BA, Jones FE: The ERalpha coactivator, HER4/4ICD, regulates progesterone receptor expression in normal and malignant breast epithelium. Mol Cancer. 2010, 9: 150-10.1186/1476-4598-9-150.
Lu M, Mira-y-Lopez R, Nakajo S, Nakaya K, Jing Y: Expression of estrogen receptor alpha, retinoic acid receptor alpha and cellular retinoic acid binding protein II genes is coordinately regulated in human breast cancer cells. Oncogene. 2005, 24 (27): 4362-4369. 10.1038/sj.onc.1208661.
Emerman JT, Fiedler EE, Tolcher AW, Rebbeck PM: Effects of defined medium, fetal bovine serum, and human serum on growth and chemosensitivities of human breast cancer cells in primary culture: inference for in vitro assays. In Vitro Cell Dev Biol: J Tissue Cult Assoc. 1987, 23 (2): 134-140. 10.1007/BF02623594.
Stingl J, Caldas C: Molecular heterogeneity of breast carcinomas and the cancer stem cell hypothesis. Nat Rev Cancer. 2007, 7 (10): 791-799. 10.1038/nrc2212.
Chen JQ, Russo PA, Cooke C, Russo IH, Russo J: ERbeta shifts from mitochondria to nucleus during estrogen-induced neoplastic transformation of human breast epithelial cells and is involved in estrogen-induced synthesis of mitochondrial respiratory chain proteins. Biochim Biophys Acta. 2007, 1773 (12): 1732-1746. 10.1016/j.bbamcr.2007.05.008.
Power KA, Thompson LU: Ligand-induced regulation of ERalpha and ERbeta is indicative of human breast cancer cell proliferation. Breast Cancer Res Treat. 2003, 81 (3): 209-221. 10.1023/A:1026114501364.
Lindberg K, Strom A, Lock JG, Gustafsson JA, Haldosen LA, Helguero LA: Expression of estrogen receptor beta increases integrin alpha1 and integrin beta1 levels and enhances adhesion of breast cancer cells. J Cell Physiol. 2010, 222 (1): 156-167. 10.1002/jcp.21932.
Phillips TM, McBride WH, Pajonk F: The response of CD24(−/low)/CD44+ breast cancer-initiating cells to radiation. J Natl Cancer Inst. 2006, 98 (24): 1777-1785. 10.1093/jnci/djj495.
Fillmore CM, Kuperwasser C: Human breast cancer cell lines contain stem-like cells that self-renew, give rise to phenotypically diverse progeny and survive chemotherapy. Breast Canc Res: BCR. 2008, 10 (2): R25-10.1186/bcr1982.
Conley SJ, Gheordunescu E, Kakarala P, Newman B, Korkaya H, Heath AN, Clouthier SG, Wicha MS: Antiangiogenic agents increase breast cancer stem cells via the generation of tumor hypoxia. Proc Natl Acad Sci U S A. 2012, 109 (8): 2784-2789. 10.1073/pnas.1018866109.
Clarke RB, Spence K, Anderson E, Howell A, Okano H, Potten CS: A putative human breast stem cell population is enriched for steroid receptor-positive cells. Dev Biol. 2005, 277 (2): 443-456. 10.1016/j.ydbio.2004.07.044.
Charafe-Jauffret E, Ginestier C, Iovino F, Tarpin C, Diebel M, Esterni B, Houvenaeghel G, Extra JM, Bertucci F, Jacquemier J, Xerri L, Dontu G, Stassi G, Xiao Y, Barsky SH, Birnbaum D, Viens P, Wicha MS: Aldehyde dehydrogenase 1-positive cancer stem cells mediate metastasis and poor clinical outcome in inflammatory breast cancer. Clin Canc Res: Offic J Am Assoc Canc Res. 2010, 16 (1): 45-55. 10.1158/1078-0432.CCR-09-1630.
Pommier SJ, Hernandez A, Han E, Massimino K, Muller P, Diggs B, Chamberlain E, Murphy J, Hansen J, Naik A, Vetto J, Pommier RF: Fresh Surgical Specimens Yield Breast Stem/Progenitor Cells and Reveal Their Oncogenic Abnormalities. Ann Surg Oncol. 2011, 19 (2): 527-535.
Morimoto K, Kim SJ, Tanei T, Shimazu K, Tanji Y, Taguchi T, Tamaki Y, Terada N, Noguchi S: Stem cell marker aldehyde dehydrogenase 1-positive breast cancers are characterized by negative estrogen receptor, positive human epidermal growth factor receptor type 2, and high Ki67 expression. Cancer Sci. 2009, 100 (6): 1062-1068. 10.1111/j.1349-7006.2009.01151.x.
Charafe-Jauffret E, Ginestier C, Iovino F, Wicinski J, Cervera N, Finetti P, Hur MH, Diebel ME, Monville F, Dutcher J, Brown M, Viens P, Xerri L, Bertucci F, Stassi G, Dontu G, Birnbaum D, Wicha MS: Breast cancer cell lines contain functional cancer stem cells with metastatic capacity and a distinct molecular signature. Cancer Res. 2009, 69 (4): 1302-1313. 10.1158/0008-5472.CAN-08-2741.
Kos M, Denger S, Reid G, Gannon F: Upstream open reading frames regulate the translation of the multiple mRNA variants of the estrogen receptor alpha. J Biol Chem. 2002, 277 (40): 37131-37138. 10.1074/jbc.M206325200.
Ishii H, Kobayashi M, Munetomo A, Miyamoto T, Sakuma Y: Novel splicing events and post-transcriptional regulation of human estrogen receptor alpha E isoforms. J Steroid Biochem Mol Biol. 2013, 133: 120-128.
Kobayashi M, Ishii H, Sakuma Y: Identification of novel splicing events and post-transcriptional regulation of human estrogen receptor alpha F isoforms. Mol Cell Endocrinol. 2011, 333 (1): 55-61. 10.1016/j.mce.2010.12.003.
Iliopoulos D, Hirsch HA, Wang G, Struhl K: Inducible formation of breast cancer stem cells and their dynamic equilibrium with non-stem cancer cells via IL6 secretion. Proc Natl Acad Sci U S A. 2011, 108 (4): 1397-1402. 10.1073/pnas.1018898108.
Dhamija S, Kuehne N, Winzen R, Doerrie A, Dittrich-Breiholz O, Thakur BK, Kracht M, Holtmann H: Interleukin-1 activates synthesis of interleukin-6 by interfering with a KH-type splicing regulatory protein (KSRP)-dependent translational silencing mechanism. J Biol Chem. 2011, 286 (38): 33279-33288. 10.1074/jbc.M111.264754.
Dethlefsen C, Hojfeldt G, Hojman P: The role of intratumoral and systemic IL-6 in breast cancer. Breast Cancer Res Treat. 2013, 138 (3): 657-664. 10.1007/s10549-013-2488-z.
Fernando RI, Castillo MD, Litzinger M, Hamilton DH, Palena C: IL-8 signaling plays a critical role in the epithelial-mesenchymal transition of human carcinoma cells. Cancer Res. 2011, 71 (15): 5296-5306. 10.1158/0008-5472.CAN-11-0156.
Ginestier C, Liu S, Diebel ME, Korkaya H, Luo M, Brown M, Wicinski J, Cabaud O, Charafe-Jauffret E, Birnbaum D, Guan JL, Dontu G, Wicha MS: CXCR1 blockade selectively targets human breast cancer stem cells in vitro and in xenografts. J Clin Invest. 2010, 120 (2): 485-497. 10.1172/JCI39397.
Donovan CA, Pommier RF, Schillace R, O'Neill S, Muller P, Alabran JL, Hansen JE, Murphy JA, Naik AM, Vetto JT, Pommier SJ: Correlation of breast cancer axillary lymph node metastases with stem cell mutations. JAMA Surg. 2013, 148 (9): 873-878. 10.1001/jamasurg.2013.3028.
The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2407/14/733/prepub
The authors would like to thank the patients who agree to be part of this study, and Dr. Megan Troxell and the department of Pathology at OHSU for breast tissue specimens. This study was supported by the Janet E. Bowen Foundation.
This work was funded The Grand Chapter of Oregon, Order of the Eastern Star, the Janet E. Bowen Foundation, and the Avon Foundation for Women.
The authors declare that they have no competing interests.
RS: Conception and design, collection and/or assembly of data, data analysis and interpretation, and manuscript writing. AS: manuscript writing. RP: Provision of study material or patients. SO: Collection and/or assembly of data. PM: Collection and/or assembly of data. AN: Provision of study material or patients. JH: Provision of study material or patients. SP: Conception and design, data analysis and interpretation, financial support, writing and final approval of manuscript. All authors read and approved the final manuscript.