Molecular Cancer Research Folkman CR Podcast
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Molecular Cancer Research 5, 847-861, August 1, 2007. Published Online First August 1, 2007;
doi: 10.1158/1541-7786.MCR-06-0427
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
1541-7786.MCR-06-0427v1
1541-7786.MCR-06-0427v2
5/8/847    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Streicher, K. L.
Right arrow Articles by Ethier, S. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Streicher, K. L.
Right arrow Articles by Ethier, S. P.

Signaling and Regulation

Activation of a Nuclear Factor {kappa}B/Interleukin-1 Positive Feedback Loop by Amphiregulin in Human Breast Cancer Cells

Katie L. Streicher1, Nicole E. Willmarth1,4, Jose Garcia5, Julie L. Boerner1,2, T. Gregory Dewey5 and Stephen P. Ethier1,3

1 Breast Cancer Program, Karmanos Cancer Institute and Departments of 2 Pharmacology and Department of 3 Pathology, Wayne State University School of Medicine, Detroit, Michigan; 4 Cellular and Molecular Biology Graduate Program, University of Michigan, Ann Arbor, Michigan; and 5 Keck Graduate Institute of Applied Life Sciences, Claremont, California

Requests for reprints: Stephen P. Ethier, Barbara Ann Karmanos Cancer Institute, 4100 John RDetroit, MI 48201. Phone: 313-576-8613; Fax: 734-647-9480. E-mail: ethier{at}karmanos.org


   Abstract
Top
 Notes
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 References
 
We have recently shown that an amphiregulin-mediated autocrine loop is responsible for growth factor–independent proliferation, motility, and invasive capacity of some aggressive breast cancer cells, such as the SUM149 breast cancer cell line. In the present study, we investigated the mechanisms by which amphiregulin activation of the epidermal growth factor receptor (EGFR) regulates these altered phenotypes. Bioinformatic analysis of gene expression networks regulated by amphiregulin implicated interleukin-1{alpha} (IL-1{alpha}) and IL-1ß as key mediators of amphiregulin's biological effects. The bioinformatic data were validated in experiments which showed that amphiregulin, but not epidermal growth factor, results in transcriptional up-regulation of IL-1{alpha} and IL-1ß. Both IL-1{alpha} and IL-1ß are synthesized and secreted by SUM149 breast cancer cells, as well as MCF10A cells engineered to express amphiregulin or MCF10A cells cultured in the presence of amphiregulin. Furthermore, EGFR, activated by amphiregulin but not epidermal growth factor, results in the prompt activation of the transcription factor nuclear factor–{kappa}B (NF-{kappa}B), which is required for transcriptional activation of IL-1. Once synthesized and secreted from the cells, IL-1 further activates NF-{kappa}B, and inhibition of IL-1 with the IL-1 receptor antagonist results in loss of NF-{kappa}B DNA binding activity and inhibition of cell proliferation. However, SUM149 cells can proliferate in the presence of IL-1 when EGFR activity is inhibited. Thus, in aggressive breast cancer cells, such as the SUM149 cells, or in normal human mammary epithelial cells growing in the presence of amphiregulin, EGFR signaling is integrated with NF-{kappa}B activation and IL-1 synthesis, which cooperate to regulate the growth and invasive capacity of the cells. (Mol Cancer Res 2007;5(8):847–62)


   Introduction
Top
 Notes
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 References
 
The epidermal growth factor (EGF) receptor (EGFR), or erbB1, is a transmembrane protein (1) possessing intrinsic tyrosine kinase activity. There are several EGF family ligands that can bind and activate EGFR, including epidermal growth factor (EGF; ref. 2), amphiregulin (AR; ref. 3), heparin-binding EGF (HB-EGF; ref. 4), transforming growth factor (TGF)-{alpha} (5), epiregulin (6), betacellulin (7), and epigen (8). Ligand binding facilitates dimerization of EGFR, which activates downstream pathways known to be involved in cell growth, proliferation, differentiation, and migration (reviewed in ref. 9). AR was originally purified from the conditioned media of MCF-7 breast cancer cells treated with the tumor promoter phorbol 12-myristate 13-acetate (10). An AR/EGFR autocrine loop has been implicated in cancer progression based on studies using colorectal cancer, pancreatic cancer, and hepatocellular carcinoma cells (11-13). Our laboratory recently discovered that SUM149 inflammatory breast cancer cells have an AR/EGFR autocrine loop that is required for their proliferation, suggesting that an AR/EGFR autocrine loop also plays a role in breast cancer progression (14).

It is apparent from both previous literature and work done in our laboratory that AR activation of EGFR can generate different biological effects on cells and tissues compared with other EGF family ligands. For example, AR but not TGF-{alpha}, was able to induce a spindle-like morphology and a relocalization of E-cadherin in Madin-Darby canine kidney cells (15). Also, using targeted knockout mice, Luetteke et al. reported that specific loss of AR, but not EGF or TGF-{alpha}, severely stunted ductal outgrowth in the mammary gland (16). Additionally, we have shown that normal MCF10A human mammary epithelial cells exhibit an increased level of cell motility and invasion when stimulated with AR versus EGF (14). Understanding this difference between AR signaling and signaling induced by other EGF family ligands is important to provide more insight into how an AR/EGFR autocrine loop can play a critical role in breast cancer progression.

Our previous studies on the differential effects of EGF versus AR on cell motility and invasion led us to the proinflammatory cytokine interleukin-1 (IL-1). By using microarray expression analysis, we found that AR overexpression in the normal human mammary epithelial cell line MCF10A up-regulates the expression of several genes involved in cell motility and invasion compared with MCF10A cells growing in EGF. Among the genes that were most highly up-regulated by AR were the cytokines IL-1{alpha} and IL-1ß (14).

The IL-1 cytokines (IL-1{alpha} and IL-1ß) activate the IL-1RI and are usually secreted only in response to infection or injury. However, IL-1 overexpression, as an autocrine growth factor, has been observed in some cancers (17, 18). IL-1 signaling exerts its effects by regulating the expression of a number of proinflammatory proteins, including growth factors, adhesion molecules, chemokines, and tissue-degrading enzymes (19-22). IL-1, therefore, has been implicated in the regulation of cell proliferation and differentiation, as well as cell motility and invasion. An increased level of IL-1 usually correlates with tumor invasiveness and poor prognosis in cancer patients (reviewed in ref. 23). IL-1 is a downstream target gene of the transcription factor nuclear factor–{kappa}B (NF-{kappa}B) and also a potent inducer of NF-{kappa}B activity, thus permitting an autoregulatory feedback loop (24, 25). The NF-{kappa}B proteins [NF-{kappa}B1 (p50 and its precursor p105), NF-{kappa}B2 (p52 and its precursor p100), RelA (p65), RelB, and c-Rel] are typically sequestered in the cytoplasm and inhibited by I{kappa}B, which masks their nuclear localization signal. Upon phosphorylation of I{kappa}B by the I{kappa}B kinase, I{kappa}B is selectively degraded, and consequently NF-{kappa}B proteins are released and translocated into the nucleus to activate NF-{kappa}B responsive genes (26, 27). IL-1 acts as a potent activator of NF-{kappa}B because it induces phosphorylation and, thus, degradation of I{kappa}B. An IL-1/NF-{kappa}B positive feedback loop has been observed in pancreatic cancer and may also play a role in breast cancer (25).

In this report, we sought to investigate more thoroughly this connection between AR and IL-1 to determine whether an IL-1/NF-{kappa}B feedback loop is triggered by AR activation of EGFR. We show that SUM149 inflammatory breast cancer cells, which have an AR/EGFR autocrine loop, are also secreting significant amounts of IL-1{alpha} and IL-1ß and have increased NF-{kappa}B activity, all of which requires EGFR activation. Additionally, we discovered that by overexpressing AR in MCF10A cells or simply growing them in the presence of exogenous AR, an IL-1/NF-{kappa}B feedback loop is initiated. Furthermore, inhibition of this feedback loop significantly blocks AR-induced cell proliferation. However, the IL-1/NF-{kappa}B feedback loop is not observed when cells are stimulated with EGF instead of AR. Thus, IL-1 signaling is required for AR-induced cell proliferation, suggesting a unique mechanism by which an AR/EGFR autocrine loop can contribute to breast cancer progression.


   Results
Top
 Notes
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 References
 
Activation of EGFR by AR Up-Regulates IL-1
Altered EGFR signaling in breast cancer contributes to increased tumor proliferation and progression. Specifically, we have shown a self-sustaining autocrine loop between EGFR and its ligand AR in the SUM149 breast cancer cell line that is essential for cell growth and contributes to increased migration and invasion (14). To understand the differential effects of EGFR signaling after activation by EGF or AR, we used two different bioinformatic strategies to identify genes regulated by EGFR activation due to each ligand. First, we used a computational strategy developed by Dewey and coworkers (28-30) that assembles time-series data into phenomenological networks indicative of the specific biological phenomena regulated by EGFR. The networks obtained using this approach have a scale-free topology, and the hub genes (genes with the highest level of connectivity) in the network are the genes whose expressions are most profoundly affected by blocking EGFR signaling. To map the gene expression networks regulated by AR-stimulated EGFR signaling in SUM149 cells and EGF-stimulated EGFR signaling in MCF10A cells, EGFR activation was inhibited in each cell line with 1 µmol/L CI-1033 for times ranging from 4 to 48 h. mRNA expression was analyzed using Affymetrix U-133a microarrays, and networks were generated using the methods developed by Dewey and coworkers (28-30). Table 1 shows the top 20 hub genes regulated by EGFR signaling and their levels of connectivity in the AR-stimulated SUM149 and EGF-stimulated MCF10A cells. Complete analysis of these networks in the two cell lines will be described in detail in other publications. For the purposes of the present studies, we observed that IL-1{alpha} and IL-1ß represented major hub genes specific to the EGFR network in SUM149 cells, whereas these genes were not part of the EGFR network in the MCF10A cells with EGF-stimulated EGFR. In a separate, more traditional experimental approach, MCF10A cells, overexpressing AR, were found to express dramatically higher levels of IL-1{alpha} and IL-1ß than MCF10A cells growing in the presence of EGF (14). The identification of overexpressed IL-1{alpha} and IL-1ß in two different cell lines with AR-stimulated EGFR using two independent analyses strongly suggests an important role for AR in mediating the activation of IL-1{alpha}/ß in these cells. Therefore, we designed a series of experiments to further investigate the role of AR-stimulated IL-1 synthesis in the biology of cells responding to AR as an EGFR ligand.


View this table:
[in this window]
[in a new window]

  		Table 1. IL-1{alpha} and IL-1ß Are among the Top Hubs Connected to EGFR in the SUM149 Cell Network

 
Amphiregulin-Induced EGFR Regulates IL-1
To investigate the regulation of IL-1 by AR, we measured mRNA and protein expression levels for both IL-1{alpha} and IL-1ß in SUM149 cells, MCF10A cells engineered to overexpress AR (MCF10A AR) cells, and MCF10A cells grown without EGF in the presence of exogenous AR (MCF10A+AR) and compared them to levels in MCF10A cells grown in the presence of EGF. Figure 1(A and B) shows that all cell lines in which EGFR was activated with AR expressed higher mRNA and secreted protein levels of both IL-1{alpha} and IL-1ß than MCF10A cells, in which EGFR was activated with EGF. To confirm that IL-1 expression induced by AR depends on EGFR activation, we inhibited EGFR activation in our panel of cell lines using the small molecule EGFR kinase inhibitor Iressa, which effectively inhibits EGFR phosphorylation at a dose of 0.5 µmol/L (inset in Fig. 1). Figure 1(A and B) confirms that blocking EGFR activity in SUM149, MCF10A AR, and MCF10A+AR cells resulted in dramatic decreases in mRNA and secreted protein levels of both IL-1{alpha} and IL-1ß but had no effect in MCF10A cells cultured in the presence of EGF.


Figure 1
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 1. EGFR regulates IL-1 mRNA and secretion in cells with AR-stimulated EGFR. IL-1{alpha} and IL-1ß mRNA expression (A) and secreted protein levels (B) were measured in MCF10A AR, MCF10A+AR, SUM149, and MCF10A cells by quantitative, real-time PCR and ELISA, respectively. EGFR activation was inhibited by 0.5 µmol/L of the kinase inhibitor Iressa. Columns, means of three independent experiments; bars, SD. Fold change relative to MCF10A control. (An inset confirming the inhibition of EGFR activation by treatment with 0.5 µmol/L Iressa shows an EGFR immunoprecipitation followed by phosphotyrosine and EGFR Western blots in control and Iressa-treated SUM149 cells).

 
Although IL-1{alpha} and IL-1ß bind to the same receptors, IL-1ß is solely active as a secreted protein, whereas IL-1{alpha} is active both as an intracellular precursor and as a secreted protein (31, 32). Therefore, we measured levels of intracellular IL-1{alpha} in our panel of cell lines under control and Iressa-treated conditions. We found similar patterns as those observed for secreted protein, which is that IL-1{alpha} intracellular protein was considerably up-regulated in cells with AR-stimulated EGFR relative to MCF10A cells and this up-regulation was abrogated after EGFR inhibition (data not shown). Together, these data confirm the results obtained from the network analysis and show a distinct role for AR-activated EGFR in regulating the expression of IL-1.

IL-1 Signaling Is Required for Proliferation of Cells with AR-Activated EGFR
Having confirmed that AR-activated EGFR regulates the expression of IL-1, we next investigated the functional role of the IL-1 pathway on cell proliferation. To inhibit IL-1 signaling, we used recombinant IL-1 receptor antagonist (IL-1ra), which binds the same receptor as IL-1{alpha} and IL-1ß but does not transduce a signal (33-35). A 10-fold to 100-fold molar excess of IL-1ra will effectively block IL-1 signaling and decrease IL-1 secretion (33). To insure the use of an appropriate concentration of IL-1ra, we exposed SUM149 and MCF10A cells to 1, 5, 10, and 20 ng/mL IL-1ra and measured cell proliferation. The data in Fig. 2A show that inhibition of IL-1 signaling in SUM149 cells dose-dependently decreased cell proliferation in SUM149 cells but had no effect in MCF10A cells grown in the presence of EGF (Fig. 2A). The growth inhibitory effect of IL-1ra on SUM149 cells was maximal at 10 ng/mL; therefore, we used this concentration for our remaining experiments (Fig. 2A). Figure 2B shows that cells with overexpression of AR (MCF10A AR), as well as cells grown in the presence of exogenous AR (MCF10A+AR), were also potently growth-inhibited by IL-1ra (Fig. 2B). The observation that MCF10A cells growing with EGF do not depend on IL-1 for growth shows that AR alters downstream EGFR signaling, leading to increases in IL-1{alpha} and IL-1ß expression that influence cell proliferation.


Figure 2
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 2. IL-1 regulates growth of cells with AR-stimulated EGFR and alters response to exogenous IL-1{alpha}. A. SUM149 and MCF10A cells were incubated with various concentrations of IL-1ra to block IL-1 signaling. Cell counts were determined on days 1 and 8 using a Coulter counter. B. MCF10A AR, MCF10A+AR, SUM149, and MCF10A cells were treated with 10 ng/mL IL-1ra. Cell counts were determined on days 1 and 8 using a Coulter counter. C. SUM149, MCF10A AR, MCF10A+AR, and MCF10A cells were treated with 50 pg/mL recombinant IL-1{alpha}, 0.5 µmol/L Iressa, or 0.5 µmol/L Iressa followed by 50 pg/mL recombinant IL-1{alpha}. Cell counts were determined on days 1 and 8 using a Coulter Counter. Columns, means of three independent experiments; bars, SD.

 
The results described above show that AR regulates cell proliferation, at least in part through EGFR-mediated up-regulation of IL-1. However, IL-1 has also been shown to stimulate proliferation of cancer cells independent of EGFR activation (36). To investigate the cellular response to increased IL-1 in our panel of cell lines without EGFR activation, we examined the ability of SUM149, MCF10A AR, MCF10A+AR, and MCF10A cells to respond to IL-1 in the presence of Iressa. In these experiments, cells were treated with 50 pg/mL IL-1{alpha}, 0.5 µmol/L Iressa, or 0.5 µmol/L Iressa + 50 pg/mL IL-1{alpha}. The 50 pg/mL as dose of recombinant IL-1{alpha} was chosen based on the secreted levels of this cytokine measured in SUM149 cells (Fig. 1B). Figure 2C shows that all cell types exhibited minimal increases in proliferation in the presence of IL-1{alpha}, suggesting that IL-1 is not a potent stimulator of proliferation in these cells under normal growth conditions (Fig. 2C). Interestingly, however, the addition of IL-1{alpha} after EGFR inhibition with Iressa resulted in a significant proliferative response of SUM149, MCF10A AR, and MCF10A+AR cells, but not MCF10A cells (Fig. 2C). The ability of cells with AR-stimulated EGFR to exhibit a mitogenic response to IL-1{alpha} in the presence of Iressa could represent a mechanism of resistance to EGFR tyrosine kinase inhibitors, particularly if stromal cells also express and secrete IL-1.

NF-{kappa}B Binding Activity Is Increased after AR-Induced Activation of EGFR
The results described above validated the link between EGFR, AR, and IL-1 suggested by the network and expression array analyses and suggested that IL-1 is an important regulator of breast cancer cell growth. Therefore, we did a series of experiments to determine if NF-{kappa}B is active in AR-stimulated cells and to examine the relationship between IL-1 expression and NF-{kappa}B activation. Accordingly, we measured NF-{kappa}B DNA binding in our panel of cell lines relative to MCF10A control cells cultured in the presence of EGF. To confirm the importance of EGFR signaling in the activation of NF-{kappa}B, experiments were done in the presence or absence of Iressa. Figure 3A shows increased binding of the p50 and p65 NF-{kappa}B subunits in the SUM149 breast cancer cells, as well as the MCF10A AR and MCF10A+AR cells, relative to MCF10A cells. Importantly, inhibition of EGFR activity with Iressa completely abrogated the increased p50 and p65 DNA binding observed in cells with AR-stimulated EGFR but had no effect on NF-{kappa}B DNA binding in MCF10A cells (Fig. 3A). To determine if the effects of AR on NF-{kappa}B DNA binding extend beyond SUM149 and MCF10A AR cells, we also used an MCF7 model with inducible EGFR expression. MCF7 cells were exposed to 2 µg/mL doxacyclin for 24 h to induce EGFR followed by exposure to EGF or AR for the indicated time points. As shown in Fig. 3B, AR was able to induce NF-{kappa}B DNA binding similar to that observed in MCF10A+AR, whereas EGF was unable to induce NF-{kappa}B activation, as seen in MCF10A cells.


Figure 3
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 3. AR-stimulated EGFR increases DNA binding of the p50 and p65 NF-{kappa}B subunits. Nuclear extracts were collected from the following cells. A. MCF10A AR, MCF10A+AR, SUM149, and MCF10A cells vehicle-treated or treated with 0.5 µmol/L Iressa for 24 h. B. MCF7 cells engineered to inducibly overexpress EGFR after 24-h stimulation with 2 µg/mL doxacyclin plus 4 nmol/L EGF or 4 nmol/L AR for the indicated time points. C. MCF10A cells incubated for 1 h with bioequivalent doses of each of the EGFR ligands: betacellulin (BTC; 2 nmol/L), TGF-{alpha} (4 nmol/L), HB-EGF (2 nmol/L), epiregulin (EPR, 2 nmol/L), epigen (50 nmol/L), AR (4 nmol/L), or EGF (2 nmol/L). A nonradioactive EMSA was used to measure the DNA binding of the p50 and p65 subunits of NF-{kappa}B in these samples. Columns, means of three independent experiments; bars, SD. A positive control tumor necrosis factor–stimulated HeLa cell extract, a negative control, a competitor probe, and a control after treatment of SUM149 cells with 10 µmol/L of the NF-{kappa}B inhibitor parthenolide for 8 h were included in each run to determine the efficacy of the EMSA. Data for these controls were pooled across all EMSA experiments and presented as the means ± SD.

 
To determine the specificity of the effects of AR on NF-{kappa}B DNA binding, we incubated MCF10A cells for 1 h with bioequivalent doses of each of EGFR ligands, betacellulin (2 nmol/L), TGF-{alpha} (4 nmol/L), HB-EGF (2 nmol/L), epiregulin (2 nmol/L), epigen (50 nmol/L), AR (4 nmol/L), EGF (2 nmol/L), as determined by the ligand concentration, yielding similar growth in MCF10A cells with 2 nmol/L EGF. Figure 3C shows that only AR was able to increase DNA binding of the p65 and p50 NF-{kappa}B subunits, which confirms that AR is critical for mediating the observed alterations in EGFR downstream signaling. These results provide further support for the notion that the relationship between EGFR and IL-1 is distinct in cells with AR-stimulated EGFR compared with cells with EGF-stimulated EGFR.

AR-Activated EGFR Signaling Generates a Positive Feedback Loop Involving IL-1 and NF-{kappa}B
After confirming that NF-{kappa}B DNA binding was decreased after EGFR inhibition, we investigated the role of NF-{kappa}B in regulating EGFR-mediated effects on IL-1. To inhibit NF-{kappa}B DNA binding, we incubated our panel of cell lines with various concentrations of the NF-{kappa}B inhibitor parthenolide and collected nuclear extracts. Parthenolide inhibits NF-{kappa}B by two mechanisms: by inhibiting I{kappa}B kinase complex and preventing the degradation of I{kappa}B-{alpha} and I{kappa}B-ß, which is necessary for translocation of NF-{kappa}B to the nucleus (37), as well as the specific alkylation of cysteine residues within NF-{kappa}B subunits (38). In the presence of 10 µmol/L parthenolide, NF-{kappa}B DNA binding was dramatically reduced in SUM149, MCF10A AR, and MCF10A+AR cells (Fig. 3). Although 5 µmol/L and 20 µmol/L doses of parthenolide were also tested (data not shown), the 10 µmol/L dose was chosen for future experiments because it inhibited NF-{kappa}B DNA binding to levels similar to that seen in MCF10A control cells, and no additional effect was seen with higher concentrations.

We treated SUM149, MCF10A AR, MCF10A+AR, and MCF10A cells with 10 µmol/L parthenolide and collected RNA and conditioned medium. Both IL-1{alpha} and IL-1ß mRNA expression and secreted protein levels were measured by QPCR and ELISA, respectively. The results shown in Fig. 4A showed a role of NF-{kappa}B in the transcriptional up-regulation of IL-1. Figure 4B shows that NF-{kappa}B inhibition decreased the secreted protein levels of IL-1{alpha} and IL-1ß in all cells except MCF10A, and a similar effect was also observed for intracellular IL-1{alpha} protein (data not shown). Together, these results point to a role for NF-{kappa}B in regulating the IL-1 pathway downstream of AR-stimulated EGFR.


Figure 4
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 4. AR-stimulated EGFR increases IL-1 mRNA and secretion in an NF-{kappa}B–dependent manner. IL-1{alpha} and IL-1ß mRNA expression (A) and secreted protein levels (B) were measured in MCF10A AR, MCF10A+AR, SUM149, and MCF10A cells by quantitative, real-time PCR and ELISA, respectively. EGFR activation was inhibited by 10 µmol/L of the NF-{kappa}B inhibitor parthenolide. Columns, means of three independent experiments; bars, SD. Fold change relative to MCF10A control.

 
Because IL-1 is also a well-characterized activator of NF-{kappa}B (21, 23, 32, 34, 39) and an IL-1/NF-{kappa}B positive feedback loop has been previously shown to affect the development and progression of other cancers (25, 40), we investigated the role of AR in the induction of a similar feedback loop in breast cancer cells. We incubated our panel of cell lines with 10 ng/mL IL-1ra for 8 h, then nuclear extracts were collected and evaluated for NF-{kappa}B DNA binding. Figure 5 shows that NF-{kappa}B DNA binding was decreased after inhibition of IL-1 signaling in all cells with AR-mediated EGFR activation. Together, our data illustrate that the activation of EGFR signaling by AR generates an active IL-1/NF-{kappa}B positive feedback loop, such that disrupting either IL-1 or NF-{kappa}B results in loss of activity of the other.


Figure 5
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 5. EGFR activation by AR generates a positive feedback loop between IL-1 and NF-{kappa}B. Nuclear extracts from MCF10A AR, MCF10A+AR, SUM149, and MCF10A cells vehicle-treated or treated with 10 ng/mL IL-1ra for 8 h were collected. A nonradioactive EMSA was used to measure the DNA binding of the p50 and p65 subunits of NF-{kappa}B in these samples. Columns, means of three independent experiments; bars, SD. A positive control tumor necrosis factor–stimulated HeLa cell extract, a negative control, a competitor probe, and a control after treatment of SUM149 cells with 10 µmol/L of the NF-{kappa}B inhibitor parthenolide for 8 h were included in each run to determine the efficacy of the EMSA. Data for these controls were pooled across all experiments and presented as the means ± SD.

 
Our data are consistent with a model where AR-stimulated EGFR activates NF-{kappa}B and leads to the transcriptional up-regulation of IL-1, as well as a model in which IL-1 up-regulation is induced before NF-{kappa}B activation. Therefore, we designed experiments to understand which of these models accurately describes the effects of AR on downstream EGFR signaling. We evaluated changes in NF-{kappa}B DNA binding, as well as IL-1{alpha} and IL-1ß mRNA expression, after a change in ligand from exogenous EGF to AR. Figure 6A shows that NF-{kappa}B DNA binding increases within 1 h of the withdrawal of EGF and addition of exogenous AR, and this initial up-regulation is not affected by inhibition of IL-1 signaling. Only at later time points that coincide with the up-regulation of IL-1 does its inhibition have any effect on NF-{kappa}B, which is consistent with previous data regarding a feedback loop between these two molecules. Figure 6B shows the results for IL-1{alpha} and IL-1ß QPCR at multiple time points after the addition of exogenous AR in the presence and absence of the NF-{kappa}B inhibitor parthenolide. This figure further supports data in Fig. 6A by showing that IL-1{alpha} is not transcriptionally up-regulated by AR until ~12 h after AR stimulation, and this up-regulation is dependent upon NF-{kappa}B activation. Interestingly, IL-1ß transcriptional up-regulation is also NF-{kappa}B–dependent; however, it does not occur until 24 h after AR exposure, suggesting that IL-1{alpha} may also have a role in regulating IL-1ß expression (Fig. 6B). Taken together, these data show that NF-{kappa}B activation precedes IL-1 transcriptional up-regulation and the development of a feedback loop that is required for proliferation of cells with AR-stimulated EGFR (Fig. 7 ). This IL-1/NF-{kappa}B interaction induced specifically by AR may represent an important regulatory pathway in a particular subset of breast cancers that could be targeted for therapy.


Figure 6
View larger version (22K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 6. AR-stimulated EGFR transcriptionally up-regulates IL-1 due to increased NF-{kappa}B activation. A. Nuclear extracts from MCF10A, SUM149, and MCF10A cells that were grown in the presence of 20 ng/mL AR for the indicated time points were collected and used in a nonradioactive EMSA to measure DNA binding of the p50 and p65 subunits of NF-{kappa}B. Samples were treated also with 10 ng/mL IL-1ra for the indicated time points. Columns, means of three independent experiments; bars, SD. Fold change relative to MCF10A control. B. IL-1{alpha} and IL-1ß mRNA expressions were measured in MCF10A, SUM149, and MCF10A cells that were grown in the presence of 20 ng/mL AR (+AR) for the indicated times by quantitative, real-time PCR. Samples were treated with parthenolide for 4 h at the initial time when MCF10A cells began growing in the presence of exogenous AR, then parthenolide was removed, and for the remainder of the indicated time, cells were grown in +AR medium. Columns, means of three independent experiments; bars, SD. Fold change relative to MCF10A control.

 

Figure 7
View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
  		FIGURE 7. Model of the alteration of EGFR downstream signaling due to AR. AR stimulation of EGFR generates a self-sustaining AR/EGFR autocrine loop in which binding of the ligand AR activates EGFR (EGFR*), which in turn increases the message levels and protein secretion of its ligand AR. Activation of EGFR by AR also up-regulates NF-{kappa}B DNA binding, leading to increased IL-1 mRNA expression and secretion, activation of the IL-1 pathway (IL-1R*), and the generation of an NF-{kappa}B/IL-1–positive feedback loop. Both the AR/EGFR autocrine loop and the NF-{kappa}B/IL-1–positive feedback loop are critical for the proliferation of cells with AR-stimulated EGFR.

 

   Discussion
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 References
 
The EGF family ligand AR is overexpressed in ~50% of all human breast carcinomas, and expression of AR is usually higher in invasive breast carcinomas than ductal carcinoma in situ or normal mammary epithelium (41-43). Additionally, expression of antisense AR in a transformed human breast epithelial cell line reduced tumorigenicity of those cells in mice, suggesting that AR plays a very relevant role in breast cancer progression (44). An AR/EGFR autocrine loop has been found to contribute to the progression of several cancers, including non–small cell lung cancer, hepatocellular carcinoma, and colon cancer (11, 13). In agreement, our laboratory has recently shown that the SUM149 inflammatory breast cancer cell line depends on a functional AR/EGFR autocrine loop for proliferation and that AR increases both cellular invasion and motility of human mammary epithelial cells (14).

AR may play a unique role in cancer progression as it has been shown previously to induce differential biological effects through the EGFR compared with other EGF family ligands. For example, the use of knockout mice showed that AR, but not EGF or TGF-{alpha}, is required for ductal outgrowth in the developing mouse mammary gland (16). In addition, AR, but not TGF-{alpha}, was shown to induce actin rearrangement due to relocalization of E-cadherin in Madin-Darby canine kidney cells (15). Our laboratory has also discovered that AR signaling differs from EGF signaling. MCF10A-immortalized human mammary epithelial cells that overexpress AR or are simply grown in exogenous AR show increased motility and an increased ability to invade compared with MCF10A cells growing in EGF (14). Amphiregulin has a significantly lower affinity for EGFR than EGF and is unable to induce phosphorylation of ErbB-2 in the absence of high levels of EGFR (45, 46). Both ligands are able to induce ErbB-3 phosphorylation to a similar extent, but AR does this in an EGFR-dependent, ErbB-2–independent manner, which is in contrast to the effects observed with EGF (45, 46). The differential effects of AR and EGF on EGFR family members and the possibility that binding of AR to heparin sulfate proteoglycans could promote the formation of dimers between EGFR and ErbB-3 may help to explain the AR-specific alterations in downstream signaling. Although the promotion of EGFR/ErbB3 heterodimers may not be a critical mechanism of AR effects in the SUM149 cells because these cells only have active EGFR, this could play a role in a broader mechanism of AR action in breast cancer cells that express ErbB-3 and should be characterized in future experiments.

The differential effects of EGF and AR signaling on gene expression are not well defined. Therefore, it was important to find genes that may be specifically regulated by AR activation of EGFR in breast cancer cells. Our network analysis identified IL-1 downstream of the EGFR in SUM149 breast cancer cells, which suggested a possible connection between EGFR signaling and IL-1 in the context of AR. The literature showing a connection between IL-1 and EGFR signaling pathways is not extensive. A link between IL-1 and AR was found previously; however, that study found that IL-1{alpha} up-regulated AR expression in cervical cancer cells (22). Our studies presented here are the first to indicate that AR-induced EGFR leads to increased expression of both IL-1{alpha} and IL-1ß. In addition, blocking the IL-1 pathway using IL-1ra in the presence of AR almost completely blocks proliferation, suggesting that AR-induced proliferation is dependent upon IL-1 signaling. The fact that this effect is observed in the SUM149 breast cancer cells, which we know have an AR/EGFR autocrine loop, suggests that AR up-regulation of IL-1 could occur in other breast cancers with AR/EGFR autocrine loops.

There is recent evidence that suggests a connection between IL-1 and breast cancer. Overexpression of IL-1 in breast cancer cells is associated with poorly differentiated and aggressive breast tumors that have an adverse prognosis (47, 48). IL-1 overexpression has been identified in invasive breast cancers (23, 49), with the highest expression found in ER/PR– cancers, a subset that often exhibits EGFR overexpression (36, 47-50). There are also studies showing that IL-1 signaling increases the expression of matrix metalloproteinase-9, E-selectin, and integrin-1 (19, 20, 51) and enhances tumor cell motility and invasion (50, 52). Thus, it seems that up-regulation of IL-1 is a mechanism by which an AR/EGFR autocrine loop can specifically contribute to the progression of aggressive, invasive breast cancer.

In SUM149 cells with AR-stimulated EGFR, exogenous IL-1 stimulates proliferation even in the absence of EGFR activation; however, this effect is not observed in MCF10A cells with EGF-stimulated EGFR. This specific effect of IL-1 on cancer cells is consistent with a previous work, in which IL-1{alpha} stimulated proliferation of cervical cancer cells but not normal cervical cells (22). The effect of IL-1 on cell proliferation has interesting implications for breast cancer progression if IL-1 is present in the tumor microenvironment. Given that IL-1 is produced both by tumor cells and the surrounding stroma (23, 31, 32, 34, 36), the ability of tumor cells to proliferate in response to IL-1 after EGFR inhibition could represent a mechanism of resistance to EGFR inhibitors. Therefore, inhibiting IL-1 signaling with IL-1ra may be useful in combination with other EGFR inhibitors or other treatment strategies in a subset of breast cancers, in which EGFR activation is driven by AR.

Our data show that AR-stimulated EGFR up-regulates IL-1{alpha} and IL-1ß through the rapid activation of NF-{kappa}B. This is consistent with previous literature showing that EGFR overexpression plays a role in the constitutive activation of NF-{kappa}B observed in pancreatic cancer (53), HNSCC (40), smooth muscle cells (54), and several ER–, EGFR+ breast cell lines (54, 55). However, previous research has not fully characterized the role of ligand specificity in regulating that activation of NF-{kappa}B by EGFR. The EGFR ligand HB-EGF has been shown to inhibit NF-{kappa}B activation in intestinal epithelial cells in a phosphoinositide-3 kinase–dependent manner (56), and TGF-{alpha} can induce NF-{kappa}B in the vascular wall in response to stress (57). Furthermore, EGF has been implicated in NF-{kappa}B activation in fibroblasts due to interactions between Grb7 and NF-{kappa}B–inducing kinase (58) and in the ER– breast cancer cell lines MDA-MB-231 and MDA-MB-435 cells, although a mechanism describing these effects was not explored (59). However, we are the first to report the involvement of the NF-{kappa}B pathway, specifically downstream of AR-stimulated EGFR in breast cancer. Our findings of a specific regulation of NF-{kappa}B by AR in breast cancer cells are in contrast with other studies on HB-EGF and TGF-{alpha} in intestinal epithelial cells and vascular endothelium, respectively (56, 57). Additionally, our data also are in contrast with work from Biswas et al., showing that EGF activates NF-{kappa}B in ER– breast cancer cell lines (59). Although the reason for this difference is unknown, it is possible that the cells used in this study also produce AR or the addition of EGF leads to increased AR secretion that could be responsible for the observed effects on NF-{kappa}B activation. A number of mechanisms have been proposed to explain the activation of NF-{kappa}B by EGFR (54, 55, 60). Specifically, Habib et al. have shown that EGFR interacts directly with the key NF-{kappa}B signaling proteins, NF-{kappa}B–inducing kinase, and RIP (TNFR-interacting protein) to initiate signaling through I{kappa}B{alpha} and localize NF-{kappa}B to the nucleus in MDA-MB-468 breast cancer cells (54). Other studies implicate phosphoinositide-3 kinase in EGFR-mediated NF-{kappa}B activation (55, 59, 60). The mechanism of NF-{kappa}B activation after AR-stimulated EGFR was not specifically examined in this report but deserves further study.

The identification of NF-{kappa}B as a critical component linking EGFR to IL-1 is important because the activation of this transcription factor induces chemotactic genes, growth factors, and matrix metalloproteinases, which are responsible for metastasis, cell proliferation, and progression (61). We showed that the NF-{kappa}B subunits p65 and p50 exhibit increased DNA binding in cells with AR-stimulated EGFR, but not those with EGF-stimulated EGFR. The p65 and p50 subunits of NF-{kappa}B are most active in epithelial cells (60), and increased DNA binding of these subunits are most often associated with ER– breast cancer; therefore, it is not surprising that the binding activity of these subunits are increased in SUM149 breast cancer cells (62). Currently, treatment of ER/PR–, EGFR+ breast cancers is difficult, as they are typically very aggressive and have not responded well to targeted therapies (59, 60, 63, 64). Interestingly, previous work has shown the regression of EGFR+/ER– breast tumor xenografts after NF-{kappa}B inhibition without deleterious side effects (60), as well as the ability of NF-{kappa}B and EGFR inhibitors in combination to synergistically block proliferation at concentrations that were ineffective when used individually (60). Although inhibition of NF-{kappa}B, specifically in breast cancers with AR-stimulated EGFR, has yet to be fully explored, NF-{kappa}B may be a useful biomarker in EGFR+ breast cancer for identifying when combination therapy may be most appropriate.

The induction of an IL-1/NF-{kappa}B–positive feedback loop potentiates cancer progression in multiple models (49, 55, 60, 65). Previous research in pancreatic cancer and HNSCC suggests that EGFR overexpression is activating an NF-{kappa}B/IL-1 autocrine loop in these cancer cells that is required for cell growth (25, 40). In keratinocytes, the regulation of IL-1 by NF-{kappa}B also requires EGFR activation, which is consistent with data presented in this manuscript. AR has been shown to be an autocrine factor in keratinocytes, so it is possible that AR-stimulated EGFR is driving this IL-1/NF-{kappa}B feedback loop in the same way as we show in SUM149 breast cancer cells (66). Sustained increases in IL-1 and NF-{kappa}B have the ability to alter matrix metalloproteinase activity and the production of other proangiogenic and invasion-inducing factors like IL-6 and IL-8 (40, 52, 67). Accordingly, an active autocrine loop involving IL-1, EGFR, and NF-{kappa}B would have the power to induce the rapid growth, invasion, and angiogenesis seen in aggressive breast cancer. Our data show clearly that the prompt activation of NF-{kappa}B, induced by simply changing the EGFR ligand from EGF to AR, is required for the transcriptional up-regulation of IL-1 and sufficient to induce generation of this feedback loop involving IL-1 and NF-{kappa}B. None of the aforementioned studies examined the specific effects of AR on IL-1 and NF-{kappa}B, although the differential EGFR signaling resulting from AR activation could be contributing to the low efficacy of EGFR inhibitors in the clinic.

SUM149 cells were developed from a patient with locally advanced inflammatory breast cancer. Our studies have pointed to a role for AR-activated EGFR in the aggressive growth, motile and invasive phenotype of these cells, which is consistent with the aggressive nature of the patient's disease. Indeed, other studies in breast cancer, pancreas cancer, and other tumor types suggest a role for AR in cancer with aggressive clinical features (11-13, 68). At variance with these observations are recent results reported by Kenny and Bissell, who showed that AR mRNA expression in breast cancer, as detected using gene arrays, did not strongly correlate with EGFR expression, but rather correlated with estrogen receptor expression and a good prognosis (69). At the present time, it is not possible to reconcile these disparate observations, as the two data sets were obtained under completely different conditions. The different results do suggest, however, that simple correlative approaches using AR expression at the message level will not be sufficient to understand the biology of AR in human breast cancer. The results of our experiments reported here and elsewhere indicate that, when HBC cells use AR as an autocrine ligand, EGFR accumulates on the cell surface and EGFR signaling results in expression and activation of IL-1 and NF-{kappa}B. Improving our understanding of AR's role in breast cancer will require measurement of all of these biological features associated with AR/EGFR signaling to determine their role in the progression of specific subsets of breast cancer.

Breast cancer is the most commonly diagnosed cancer in women, and EGFR overexpression correlates with a poor prognosis in breast cancer (68). Although there are EGFR inhibitors that are currently being used in the clinic, it is apparent that more research needs to be done to address EGFR inhibitor resistance. AR activation of the IL-1 pathway can have clinical implications for EGFR inhibitor resistance based on the fact that inhibitors of EGFR activity may be relatively ineffective if the IL-1 pathway is also activated and inducing cell proliferation. Additionally, AR induces gene expression changes that are considerably different from EGF and therefore AR could be targeted therapeutically. Alternatively, IL-1 and AR could play potentially important roles as biomarkers for EGFR-positive breast cancer.


   Materials and Methods
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
References
 
Network Analysis
Gene expression networks from SUM149 and MCF10A cells were determined from an analysis of global gene expression time series data. MCF10A and SUM149 cells were cultured to ~75% confluence and exposed to EGFR inhibitor CI-1033. RNA was isolated from cells at 0, 4, 8, 12, 16, 24, and 48 h after addition of drug, and corresponding gene expression levels were determined using Affymetrix U-133a and U-133b microarrays. Microarrays were analyzed using standard procedures. For filtering purposes, only genes showing a fold change in its time series of >1.7 were considered for further analysis. Cubic spline interpolation was used to calculate expression levels at 4-h time intervals. To calculate the network associated with each time series, a linear finite difference model was used as described previously (28-30). This model assumes that the gene expression levels of a given gene at a given time depend upon the expression levels of other genes at a single previous time. A discrete linear response model is used to generate a transition matrix calculated from the experimental data using matrix inversion procedures (28-30). Unitless variables are generated from these matrices, which represent phenomenological variables that show how the expression level of a gene at one point influences the expression level of another at a later time point. Biological systems are generally nonlinear, and it is not presumed that such a simple linear response model will fully capture the underlying causal network of gene expression. However, this model is designed as a data-mining tool to capture the phenomenological influence of one expression level on another. The resulting networks have a scale-free topology and display a hub-and-spoke pattern that follows a power law distribution. Two network variables, clustering coefficient and the characteristic path length, were determined from the networks and were used to determine overall connectivity of the genes, which also determines the genes identified as hubs. We identified hub genes because a strong relationship has been shown between a molecule's hub status and its importance in maintaining appropriate cellular function (70). Appropriate biological assays were used to validate the results of the network analysis.

Cell Lines and Cell Culture Conditions
The MCF10A human mammary epithelial cell line is cultured in SFIHE medium [Ham's F-12 medium supplemented with 0.1% bovine serum albumin, fungizone (0.5 µg/mL), gentamicin (5 µg/mL), ethanolamine (5 mmol/L), HEPES (10 mmol/L), transferrin (5 µg/mL), 3,3,'5-triiodo-L-thyronine (10 µmol/L), selenium (50 µmol/L), hydrocortisone (1 µg/mL), insulin (5 µg/mL), and 10 ng/mL EGF]. SUM149 cells were maintained in 5% IH (Ham's F-12 with 5% fetal bovine serum, supplemented with insulin, hydrocortisone, fungizone, and gentamicin at the same concentrations as for MCF10A cells). Complete culture conditions for both cell lines were as described previously (71). MCF10A AR cells were grown in the same culture media as MCF10A cells but without EGF (SFIH). MCF10A+AR cells were grown in the same media as MCF10A AR cells with addition of 20 ng/mL recombinant AR (R&D Systems; SFIHA). Other EGFR ligands were used at the following doses: betacellulin, 2 nmol/L; TGF-{alpha}, 4 nmol/L; HB-EGF, 2 nmol/L; EREG, 2 nmol/L; epigen, 50 nmol/L; AR, 4 nmol/L; EGF, 2 nmol/L. All cells were cultured at 37°C in a humidified incubator containing 10% CO2 and were maintained free of Mycoplasma.

MCF7 breast cancer cells stably overexpressing EGFR under control of the Tet-ON vector were a generous gift from by Dr. Julie Boerner and generated as previously described (72). To generate these cells, a pTRE2 vector encoding wt-EGFR was transfected together with a pBabe-puro vector for antibiotic selection of stably expressing clones. Clonal isolates were recloned until cell lines that inducibly expressed EGFR constructs in 80% to 90% of the cells (as determined by immunofluorescence) were obtained. This cell line is maintained in DMEM with 10% fetal bovine serum, 1 mmol/L sodium pyruvate, 100 µg/mL G418, and 10 µg/mL puromycin. EGFR expression in induced by the addition of 2 µg/mL doxacyclin for 24 h.

Nuclear Extracts and Whole-Cell Lysates
Cells were rinsed twice with ice-cold HBSS (Life Technologies) and then lysed on ice with a buffer consisting of Tris-HCl (50 mmol/L, pH 8.5), NaCl (150 mmol/L), 1% NP40 (ICN Biomedical, Inc.), EDTA (5 mmol/L) supplemented with sodium orthovanadate (5 mmol/L), phenylmethsulfonyl fluoride (50 µg/mL), aprotinin (20 µg/mL), and leupeptin (10 µg/mL). Lysates were spun at 14,000xg at 4°C for 10 min and then analyzed for protein using the Bradford method (Bio-Rad Laboratories).

Nuclear extracts were isolated using the NE-PER extraction kit according to manufacturer's instructions (Pierce). Briefly, cell pellets were lysed with hypotonic lysis buffer, including DTT and protease inhibitors. Cells were incubated on ice for 15 min before adding 10% IGEPAL CA-630 solution to the swollen cells. Cells were centrifuged for 30 s at 11,000xg, and the supernatant was transferred to a fresh tube (cytoplasmic fraction). The remaining pellet was resuspended in extraction buffer, including DTT and protease inhibitors. Tubes were vortexed for 15 s every 10 min for 40 min then centrifuged for 5 min at 21,000xg. Supernatant was transferred to new tube and stored at -70°C.

IL-1{alpha}/ß Enzyme–Linked Immunosorbent Assay
MCF10A and SUM149 cells were untreated or treated for 24 h with 0.5 µmol/L Gefitinib/Iressa (AstraZeneca Pharmaceuticals), 10 ng/mL IL-1ra (Cell Sciences), or 10 µmol/L NF-{kappa}B inhibitor parthenolide (Alexis Biochemicals). Conditioned medium and whole-cell lysates from each of these treatment groups were collected to evaluate the secreted and cellular protein levels of IL-1{alpha}, respectively. Conditioned medium from these treatment groups was also analyzed for secreted levels of IL-1ß.

IL-1{alpha}/ß ELISAs were done using commercially available DuoSet ELISA development kits (R&D Systems) following the manufacturer's instructions. Briefly, high-binding ELISA 96-well plates were coated with IL-1{alpha} or IL-1ß capture antibody overnight at room temperature. Absorbance at 450 nmol/L minus the absorbance at 550 nmol/L was measured on a VERSAmax microplate reader (Molecular Devices Corp.). A standard curve of known concentrations of recombinant IL-1{alpha} or IL-1ß was generated for each ELISA by plotting the log of the IL-1{alpha} or IL-1ß concentration versus the log of the absorbance reading and used to quantify the concentration of IL-1{alpha} or IL-1ß in each sample. Cells were lysed and nuclei were counted with a Coulter Counter (Beckman Coulter) for normalization.

Assessment of Monolayer Growth
Cells were seeded into six-well plates at 3.5 x 104 per well in SFIHE (plus 2% fetal bovine serum to allow attachment) or 5% IH or SFIH media. The next day, plating medium was removed, and cells were treated with SFIH (MCF10A AR), SFIHE (MCF10A), SFIHA (MCF10A+AR), or 5% IH (SUM149) ± 10 ng/mL IL-1ra, 0.5 µmol/L Iressa, or recombinant IL-1{alpha} for 7 days, with fresh treatment added everyday. The number of cells was determined by counting isolated nuclei with a Coulter counter 7 days after treatment. A plating efficiency was done 24 h after plating to determine the number of attached cells per well. All experiments were done in triplicate and repeated at least twice.

NF-{kappa}B Transcription Factor Assay
To evaluate the DNA binding of the p50 and p65 subunits of NF-{kappa}B, we used a transcription factor assay kit combining an electrophoretic mobility shift assay with an ELISA assay according to manufacturer's instructions (Chemicon International). Briefly, a double-stranded biotinylated oligonucleotide containing the flanked DNA binding consensus sequence for NF-{kappa}B (5'-GGGACTTTCC-3') is mixed with nuclear extract in the provided transcription factor assay buffer. During this incubation, the active form of NF-{kappa}B in the nuclear extract binds to its consensus sequence. The extract/probe/buffer mixture is then directly transferred to a strepavidin-coated 96-well plate. The active NF-{kappa}B protein immobilized on the biotinylated double-stranded oligonucleotide capture probe binds to the strepavidin plate well, and any inactive, unbound material is washed away. The bound NF-{kappa}B transcription factor subunits, p50 and p65, are detected with specific primary antibodies, a rabbit anti–NF-{kappa}B p50 and a rabbit anti–NF-{kappa}B p65. A highly sensitive horseradish peroxidase–conjugated secondary antibody is used for colorimetric detection that is read in a spectrophotometric plate reader. To insure specific NF-{kappa}B binding, a positive control (tumor necrosis factor–stimulated HeLa cell extract), nonspecific double-stranded oligonucleotide, and a specific competitor double-stranded oligonucleotide are included in each assay.

Q-Reverse Transcription–PCR Reactions
RNA was extracted from SUM149, MCF10A, MCF10A+AR, and MCF10A AR cells using the Qiagen RNeasy kit. RNA was converted into cDNA via a reverse transcription reaction using random hexamer primers. IL-1{alpha} and IL-1ß probes were ordered from Applied Biosystems Assays-by-Design service. A glyceraldehyde-3-phosphate dehydrogenase primer set was used as a control. RNA (2 µg) was used for the reverse transcription–PCR reaction, and the product was diluted at 1:12. Q–reverse transcription–PCR was done in 25-µL reactions in 96-well plates using the Taqman Universal PCR Master Mix (Applied Biosystems). Reactions were done twice, in replicates of three or four, using the Bio-Rad iQ5 real-time PCR machine (Bio-Rad Laboratories). Cycles to threshold values for IL-1{alpha} and IL-1ß were normalized to values for glyceraldehyde-3-phosphate dehydrogenase then compared with IL-1{alpha} and IL-1ß expression in MCF10A cells. Control wells containing PCR master mix and primers without sample cDNA emitted no fluorescence after 40 cycles. Relative expression data was calculated as described by Livak and Schmittgen (73). Briefly, average values were determined for number of cycles in each reaction to achieve a threshold of fluorescence. The average numbers of cycles necessary for the glyceraldehyde-3-phosphate dehydrogenase reaction were subtracted from these values, followed by subtraction of the average cycle numbers in a control cell line, in this case MCF10A. The fold difference was determined by raising 2 to the negative power of the calculated difference.


   Notes
Top
 Notes
Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 References
 
Grant support: NIH National Cancer Institute grant R01 CA100724-05 and Department of Defense Breast Cancer Program grant W81XWH-06-1-0405.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: K.L. Streicher and N.E. Willmarth contributed equally to this work.

Views and opinions of, and endorsements by the author(s) do not reflect those of the U.S. Army of the Department of Defense.

Received 12/19/06; revised 5/ 9/07; accepted 5/23/07.


   References
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 References
 

  1. Alexandroff AB, Jackson AM, Chisholm GD, James K. Cytokine modulation of epidermal growth factor receptor expression on bladder cancer cells is not a major contributor to the antitumour activity of cytokines. Eur J Cancer 1995;31A:2059–66.
  2. Carpenter G, Cohen S. Epidermal growth factor. J Biol Chem 1990;265:7709–12.[Free Full Text]
  3. Shoyab M, Plowman GD, McDonald VL, Bradley JG, Todaro GJ. Structure and function of human amphiregulin: a member of the epidermal growth factor family. Science 1989;243:1074–6.[Abstract/Free Full Text]
  4. Higashiyama S, Abraham JA, Miller J, Fiddes JC, Klagsbrun M. A heparin-binding growth factor secreted by macrophage-like cells that is related to EGF. Science 1991;251:936–9.[Abstract/Free Full Text]
  5. Derynck R. The physiology of transforming growth factor-{alpha}. Adv Cancer Res 1992;58:27–52.[Medline]
  6. Toyoda H, Komurasaki T, Uchida D, et al. Epiregulin. A novel epidermal growth factor with mitogenic activity for rat primary hepatocytes. J Biol Chem 1995;270:7495–500.[Abstract/Free Full Text]
  7. Shing Y, Christofori G, Hanahan D, et al. Betacellulin: a mitogen from pancreatic ß cell tumors. Science 1993;259:1604–7.[Abstract/Free Full Text]
  8. Strachan L, Murison JG, Prestidge RL, Sleeman MA, Watson JD, Kumble KD. Cloning and biological activity of epigen, a novel member of the epidermal growth factor superfamily. J Biol Chem 2001;276:18265–71.[Abstract/Free Full Text]
  9. Wells A. EGF receptor. Int J Biochem Cell Biol 1999;31:637–43.[CrossRef][Medline]
  10. Shoyab M, McDonald VL, Bradley JG, Todaro GJ. Amphiregulin: a bifunctional growth-modulating glycoprotein produced by the phorbol 12-myristate 13-acetate-treated human breast adenocarcinoma cell line MCF-7. Proc Natl Acad Sci U S A 1988;85:6528–32.[Abstract/Free Full Text]
  11. Castillo J, Erroba E, Perugorria MJ, et al. Amphiregulin contributes to the transformed phenotype of human hepatocellular carcinoma cells. Cancer Res 2006;66:6129–38.[Abstract/Free Full Text]
  12. Funatomi H, Itakura J, Ishiwata T, et al. Amphiregulin antisense oligonucleotide inhibits the growth of T3M4 human pancreatic cancer cells and sensitizes the cells to EGF receptor-targeted therapy. Int J Cancer 1997;72:512–7.[CrossRef][Medline]
  13. Johnson GR, Saeki T, Gordon AW, Shoyab M, Salomon DS, Stromberg K. Autocrine action of amphiregulin in a colon carcinoma cell line and immunocytochemical localization of amphiregulin in human colon. J Cell Biol 1992;118:741–51.[Abstract/Free Full Text]
  14. Willmarth NE, Ethier SP. Autocrine and juxtacrine effects of amphiregulin on the proliferative, invasive, and migratory properties of normal and neoplastic human mammary epithelial cells. J Biol Chem 2006;281:37728–37.[Abstract/Free Full Text]
  15. Chung E, Graves-Deal R, Franklin JL, Coffey RJ. Differential effects of amphiregulin and TGF-{alpha} on the morphology of MDCK cells. Exp Cell Res 2005;309:149–60.[CrossRef][Medline]
  16. Sternlicht MD, Sunnarborg SW, Kouros-Mehr H, Yu Y, Lee DC, Werb Z. Mammary ductal morphogenesis requires paracrine activation of stromal EGFR via ADAM17-dependent shedding of epithelial amphiregulin. Development 2005;132:3923–33.[Abstract/Free Full Text]
  17. Ito R, Kitadai Y, Kyo E, et al. Interleukin 1{alpha} acts as an autocrine growth stimulator for human gastric carcinoma cells. Cancer Res 1993;53:4102–6.[Abstract/Free Full Text]
  18. Kawakami Y, Nagai N, Ota S, Ohama K, Yamashita U. Interleukin-1 as an autocrine stimulator in the growth of human ovarian cancer cells. Hiroshima J Med Sci 1997;46:51–9.[Medline]
  19. Andersen K, Maelandsmo GM, Hovig E, Fodstad O, Loennechen T, Winberg JO. Interleukin-1{alpha} and basic fibroblast growth factor induction of matrix metalloproteinases and their inhibitors in osteosarcoma cells is modulated by the metastasis associated protein CAPL. Anticancer Res 1998;18:3299–303.[Medline]
  20. Nguyen M, Corless CL, Kraling BM, et al. Vascular expression of E-selectin is increased in estrogen-receptor-negative breast cancer: a role for tumor-cell-secreted interleukin-1{alpha}. Am J Pathol 1997;150:1307–14.[Abstract]
  21. Wolf JS, Chen Z, Dong G, et al. IL (interleukin)-1{alpha} promotes nuclear factor-{kappa}B and AP-1-induced IL-8 expression, cell survival, and proliferation in head and neck squamous cell carcinomas. Clin Cancer Res 2001;7:1812–20.[Abstract/Free Full Text]
  22. Woodworth CD, McMullin E, Iglesias M, Plowman GD. Interleukin 1{alpha} and tumor necrosis factor {alpha} stimulate autocrine amphiregulin expression and proliferation of human papillomavirus-immortalized and carcinoma-derived cervical epithelial cells. Proc Natl Acad Sci U S A 1995;92:2840–4.[Abstract/Free Full Text]
  23. Apte RN, Voronov E. Interleukin-1-a major pleiotropic cytokine in tumor-host interactions. Semin Cancer Biol 2002;12:277–90.[CrossRef][Medline]
  24. Hiscott J, Marois J, Garoufalis J, et al. Characterization of a functional NF-{kappa}B site in the human interleukin 1ß promoter: evidence for a positive autoregulatory loop. Mol Cell Biol 1993;13:6231–40.[Abstract/Free Full Text]
  25. Niu J, Li Z, Peng B, Chiao PJ. Identification of an autoregulatory feedback pathway involving interleukin-1{alpha} in induction of constitutive NF-{kappa}B activation in pancreatic cancer cells. J Biol Chem 2004;279:16452–62.[Abstract/Free Full Text]
  26. Ghosh S, Karin M. Missing pieces in the NF-{kappa}B puzzle. Cell 2002;109 Suppl:S81–96.[CrossRef][Medline]
  27. Li Q, Verma IM. NF-{kappa}B regulation in the immune system. Nat Rev Immunol 2002;2:725–34.[CrossRef][Medline]
  28. Bhan A, Galas DJ, Dewey TG. A duplication growth model of gene expression networks. Bioinformatics 2002;18:1486–93.[Abstract/Free Full Text]
  29. Dewey TG. From microarrays to networks: mining expression time series. Drug Discov Today 2002;7:S170–5.[Medline]
  30. Wu X, Dewey TG. From microarray to biological networks: Analysis of gene expression profiles. Methods Mol Biol 2006;316:35–48.[Medline]
  31. Basu A, Krady JK, Levison SW. Interleukin-1: a master regulator of neuroinflammation. J Neurosci Res 2004;78:151–6.[CrossRef][Medline]
  32. Dinarello CA. Interleukin-1. Cytokine Growth Factor Rev 1997;8:253–65.[CrossRef][Medline]
  33. Arend WP, Malyak M, Guthridge CJ, Gabay C. Interleukin-1 receptor antagonist: role in biology. Annu Rev Immunol 1998;16:27–55.[CrossRef][Medline]
  34. Braddock M, Quinn A. Targeting IL-1 in inflammatory disease: new opportunities for therapeutic intervention. Nat Rev Drug Discov 2004;3:330–9.[CrossRef][Medline]
  35. Subramaniam S, Stansberg C, Cunningham C. The interleukin 1 receptor family. Dev Comp Immunol 2004;28:415–28.[CrossRef][Medline]
  36. Apte RN, Krelin Y, Song X, et al. Effects of micro-environment- and malignant cell-derived interleukin-1 in carcinogenesis, tumour invasiveness and tumour-host interactions. Eur J Cancer 2006;42:751–9.[Medline]
  37. Hehner SP, Hofmann TG, Droge W, Schmitz ML. The antiinflammatory sesquiterpene lactone parthenolide inhibits NF-{kappa}B by targeting the I{kappa}B kinase complex. J Immunol 1999;163:5617–23.[Abstract/Free Full Text]
  38. Garcia-Pineres AJ, Castro V, Mora G, et al. Cysteine 38 in p65/NF-{kappa}B plays a crucial role in DNA binding inhibition by sesquiterpene lactones. J Biol Chem 2001;276:39713–20.[Abstract/Free Full Text]
  39. Stylianou E, Saklatvala J. Interleukin-1. Int J Biochem Cell Biol 1998;30:1075–9.[CrossRef][Medline]
  40. Bancroft CC, Chen Z, Yeh J, et al. Effects of pharmacologic antagonists of epidermal growth factor receptor, PI3K and MEK signal kinases on NF-{kappa}B and AP-1 activation and IL-8 and VEGF expression in human head and neck squamous cell carcinoma lines. Int J Cancer 2002;99:538–48.