| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
© The Author 2007. Published by Oxford University Press.
ARTICLES |
Novel Cell Culture Technique for Primary Ductal Carcinoma In Situ: Role of Notch and Epidermal Growth Factor Receptor Signaling Pathways
Affiliations of authors: Department of Surgery (GF, NP, NGA, NJB) and Breast Biology Group (GF, RBC, KS), Division of Cancer Studies, Faculty of Medicine and Human Sciences, University of Manchester, Christie Hospital NHS Trust, Manchester, UK; Wellcome Trust Centre for Cell Matrix Research, Faculty of Life Sciences, University of Manchester, Oxford Road, Manchester, UK (KB)
Correspondence to: Nigel J. Bundred, MD, Department of Surgery and Breast Biology Group, Division of Cancer Studies, Faculty of Medicine and Human Sciences, University of Manchester, Christie Hospital NHS Trust, Wilmslow Road, M20 9BX, Manchester, UK (e-mail: bundredn{at}manchester.ac.uk).
 |
ABSTRACT |
|---|
 Top Abstract Context and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
Background: The epidermal growth factor receptor (EGFR) and Notch signaling pathways have been implicated in self-renewal of normal breast stem cells. We investigated the involvement of these signaling pathways in ductal carcinoma in situ (DCIS) of the breast.
Methods: Samples of normal breast tissue (n = 15), pure DCIS tissue of varying grades (n = 35), and DCIS tissue surrounding an invasive cancer (n = 7) were used for nonadherent (i.e., mammosphere) culture. Mammosphere cultures were treated at day 0 with gefitinib (an EGFR inhibitor), DAPT (N-[N-(3,5-difluorophenacetyl-L-alanyl)]-S-phenylglycine t-butyl ester) (a 
-secretase inhibitor), or Notch 4–neutralizing antibody. Mammosphere-forming efficiency (MFE) was calculated by dividing the number of mammospheres of 60 µm or more formed by the number of single cells seeded and is expressed as a percentage. The Notch 1 intracellular domain (NICD) was detected immunohistochemically in paraffin-embedded DCIS tissue from 50 patients with at least 60 months of follow-up. All statistical tests were two-sided.
Results: DCIS had a greater MFE than normal breast tissue (1.5% versus 0.5%, difference = 1%, 95% confidence interval [CI] = 0.62% to 1.25%, P<.001). High-grade DCIS had a greater MFE than low-grade DCIS (1.6% versus 1.09%, difference = 0.51%, 95% CI = 0.07% to 0.94%, P = .01). The MFE of high-grade DCIS treated with gefitinib in the absence of exogenous EGF was lower than that of high-grade DCIS treated with mammosphere medium lacking gefitinib and exogenous EGF (0.56% versus 1.36%, difference 0.8%, 95% CI = 0.33% to 1.4%, P = .004). Increased Notch signaling as detected by NICD staining was associated with recurrence at 5 years (P = .012). DCIS MFE was reduced when Notch signaling was inhibited using either DAPT (0.89% versus 0.51%, difference = 0.38%, 95% CI = 0.2% to 0.6%, P<.001) or a Notch 4–neutralizing antibody (0.97% versus 0.2%, difference = 0.77%, 95% CI = 0.52% to 1.0%, P<.001).
Conclusion: We describe a novel primary culture technique for DCIS. Inhibition of the EGFR or Notch signaling pathways reduced DCIS MFE.
Prior knowledge The epidermal growth factor receptor (EGFR) and Notch signaling pathways have been implicated in self-renewal of normal breast stem cells, but their involvement in ductal carcinoma in situ (DCIS) of the breast is unclear. Study design In vitro study using DCIS-derived epithelial cells. Contribution The authors used a novel method to culture DCIS cells and showed that EGFR is necessary for DCIS growth and self-renewal and that Notch signaling is important for cell survival and/or self-renewal in nonadherent cultures. Implications This nonadherent cell culture method may facilitate discovery of pathways that are important for DCIS propagation and testing of novel inhibitors to prevent DCIS recurrence and progression to invasive disease. Limitations It is not known whether DCIS mammospheres will recapitulate the DCIS lesion from which they are derived when implanted into immunodeficient mice.
|
Ductal carcinoma in situ (DCIS) is a noninvasive breast malignancy that, if untreated, progresses to invasive cancer in 30%–50% of patients (1,2). Even after breast-conserving surgery and radiotherapy, approximately 15%–20% of women diagnosed with DCIS experience a recurrence within 10 years, at which time half of the recurrences are invasive disease (1,3).
Understanding the mechanisms of DCIS growth and progression is important to develop new treatment strategies. To this end, we developed a xenograft model that allows subcutaneous growth of DCIS lesions in nude mice, making it possible to test the efficacy of new DCIS treatments (4). Using this model, we demonstrated (4) that the low–molecular-weight tyrosine kinase inhibitor gefitinib (Iressa) inhibits proliferation and increases apoptosis in epidermal growth factor receptor (EGFR)–positive DCIS.
However, it would be useful to replace this xenograft model with a cell culture approach that would allow the rapid analysis of new inhibitors and a detailed determination of the molecular mechanism by which these inhibitors affect proliferation and apoptosis in DCIS. Although standard adherent tissue culture techniques have been used to derive cell lines from breast tissue, primary normal breast cells and primary invasive breast cancer cells cultured as adherent cells have limited growth potential and are often outgrown by contaminating stromal cells, limiting their use for growth assays (5).
An alternative approach for in vitro propagation of breast tissue is culture of nonadherent mammospheres (6). This technique has been used to isolate normal human mammary stem and progenitor cells (6,7), allowing them to survive anoikis and to self-renew. Under nonadherent mammosphere culture conditions, mammary stem and progenitor cells proliferate in an undifferentiated state and, when subsequently grown in three-dimensional (3-D) matrigel culture, can be induced to differentiate and form branching ducts and lobules containing both luminal epithelial and myoepithelial cells (6,8).
Here we tested the efficiency and reproducibility with which DCIS-derived epithelial cells can be grown from primary breast lesions using both standard adherent tissue culture and nonadherent mammosphere culture. We further examined whether these culture systems can be used to study mechanisms of DCIS growth by investigating the roles of the EGF and ErbB2 receptor signaling pathways shown to be active in DCIS (9) and to affect DCIS growth in our xenograft model (4). EGFR and ErbB2 heterodimerize to initiate tyrosine kinase phosphorylation, downstream signaling, and DCIS growth (10). In addition, we manipulated Notch signaling, which has been implicated in the self-renewal of stem cells in normal human mammospheres (7) and is aberrantly activated in human breast carcinomas (11,12), and investigated Notch expression.
|
Materials and Methods |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
Human Tissues
Tissue samples were obtained from women who had a mammogram that showed widespread microcalcification indicative of DCIS (n = 35) or DCIS surrounding an invasive breast cancer (n = 7) and for whom histopathologic confirmation of the diagnosis of DCIS was obtained. All DCIS tissue samples were obtained from women who underwent therapeutic surgery at the University Hospital of South Manchester or the Christie Hospital NHS Trust between 2003 and 2006 and were subsequently reviewed and graded by a consultant breast pathologist (Dr W. F. Knox, University Hospital of South Manchester NHS Foundation Trust). The nuclear grade of each DCIS lesion was assessed as previously described (13). Approval to remove tissue from pathology samples was granted by the South Manchester Medical Research Ethics Committee. Normal breast tissue was obtained from patients undergoing fibroadenoma excision or breast reduction surgery (n = 15) at the University Hospital of South Manchester. All patients provided written informed consent.
Of the 42 DCIS samples, the first 16 to be obtained were used for adherent culture, three of the first 16 were also cultured in nonadherent conditions, and the next 26 were used for nonadherent culture. Normal breast tissue was also grown in adherent and nonadherent culture (n = 15) as a control.
We also obtained paraffin-embedded tissue from 50 DCIS patients with documented follow-up from University Hospital of South Manchester, of whom 16 had experienced a recurrence by 60 months after surgery. These paraffin-embedded DCIS samples were retrospectively selected and entirely separate from the tissue that was prospectively collected for DCIS culture. All patients had 5 years of follow-up, and recurrence was calculated at 60 months.
Adherent Primary Cell Culture of Normal Breast Tissue and Ductal Carcinoma In Situ
DCIS or normal breast tissue collected at surgery was dissected into 2-mm cubes, which were placed into tissue culture plates containing adherent culture medium (DCIS adherent culture medium: RPMI-1640 medium supplemented with 10% fetal calf serum [Gibco, Paisley, UK], 2 nM L-glutamine [Gibco], 5 µg/mL hydrocortisone [Sigma, Poole, UK], 5 µg/mL insulin [Sigma], 50 ng/mL cholera toxin [Sigma], 10 ng/mL recombinant human EGF [R & D systems, Abingdon, UK], and 5% penicillin [10 000 U/mL]–streptomycin [10 mg/mL] [Sigma]; normal breast tissue adherent culture medium: keratinocyte serum-free medium [Gibco] with addition of keratinocyte serum-free medium supplement [Gibco], which includes human recombinant EGF and bovine pituitary extract , plus penicillin [100 U/mL]–streptomycin [0.1 mg/mL]). The cultures were incubated at 37 °C in 5% CO2 undisturbed for several days, after which the culture medium was changed every 2–5 days, as needed, for 3–4 weeks. The cells were then trypsinized using 0.125% trypsin (Worthington Biochemical Corporation, Lakewood, NJ) to produce a single-cell suspension and depleted of fibroblasts by negative selection with an antifibroblast antibody (Oncogene Research, San Diego, CA) and a CELLection Pan Mouse IgG kit (Dynal Biotech, Brombrrough, UK). The resulting cell populations contained at least 70% luminal epithelial cells, as shown by immunostaining with a fluorescein isothiocyanate (FITC)–conjugated cytokeratin (CK)18 antibody (1 : 100 dilution; Sigma) specific for luminal epithelial cells, a Cy3-conjugated smooth muscle actin antibody (1 : 100 dilution; Sigma) specific for myoepithelial cells, and a Cy3-conjugated vimentin antibody (1 : 100 dilution; Sigma) specific for fibroblasts (data not shown).
Adherent Cell Culture Growth Assay
Primary normal and DCIS cells from passage 1 were seeded in 96-well plates (4 x 103 cells per well) and incubated overnight to allow them to adhere to the substratum. Gefitinib (Astrazeneca, Macclesfield, UK) at final concentration of 1–5000 nM in dimethyl sulfoxide (DMSO; Sigma) from a 10 mM stock made with 100% DMSO or an equivalent amount of DMSO (control) was added to the cells for 5 days. The cells were then fixed, stained with the protein-binding dye sulforhodamine B (Sigma), and absorbance (at 540 nm) was measured as previously described (14). The IC50 values (drug concentrations producing 50% growth inhibition) were generated from a sigmoidal growth curve produced using Origin software (version 7.5, Origin Lab Corporation, Northampton, MA).
To obtain DCIS-conditioned medium, DCIS cells were cultured in 5-cm dishes (5 x 105 cells per dish) for 2 days in DCIS adherent culture medium. The cells were washed three times with RPMI-1640 medium, overlayed with 3 mL of RPMI-1640 medium, and incubated for 2 days. The resulting conditioned medium was filter sterilized (22-µm [pore size] sterile filters) and used immediately or stored at 4 °C for up to 2 days.
Immortalized normal human mammary luminal HB4a cells (a gift from Professor Michael J. O'Hare, University College London) and normal luminal human mammary epithelial cells (HMECs, grown from primary normal tissue described above) were seeded onto 96-well plates at a density of 3 x 103 cells per well (for cell growth assays) or in 5-cm plates at 3 x 105 cells per plate (for immunoblot analysis) and incubated overnight to allow the cells to adhere to the substratum. The cells were then washed three times with RPMI-1640 medium and incubated for 3 days with DCIS-conditioned medium or serum-free RPMI-1640 medium in the presence of 1 µM gefitinib or DMSO (control). The cells in 96-well plates were then stained with sulforhodamine B and absorbance (at 540 nm) was measured as previously described (14). Cells grown in 5-cm plates were lysed in ionic (sodium dodecyl sulphate [SDS]) lysis buffer (50 mM 2-Amino-2-(hydroxymethyl)propane-1,3-diol (Tris) [pH 7.5], 150 mM sodium chloride, 1% Nonidet P-40, 0.5% sodium doxycholate, 1 mM EDTA, 0.1% SDS) containing protease inhibitors and used for immunoblot analysis, as described below. The growth of HB4a cells and HMEC is presented as the average fold change in cell number (compared with DMSO control); all experiments were performed in triplicate.
Enzymatic Digestion of Ductal Carcinoma In Situ and Normal Breast Tissue for Mammosphere Culture
Twenty-nine samples of DCIS tissue containing microcalcification and 15 samples of normal breast tissue collected at surgery were dissected into 3- to 5-mm cubes and digested for 16–18 hours at 37 °C in serum-free Dulbecco's modified Eagle Medium (DMEM; Gibco) containing either 200 or 50 U/mL type I collagenase (Worthington Biochemical Corporation), respectively, and penicillin (100 U/mL)–streptomycin (0.1 mg/mL; Sigma). The enzymatically digested tissue was then filtered sequentially through sterile 100- and 53-µm nylon mesh to obtain a single-cell suspension, which was verified microscopically. The cells were then washed three times in DMEM:F12 medium (Gibco) by centrifugation at 1000g and resuspended in mammosphere culture medium (DMEM:F12 medium supplemented with B27 without vitamin A [diluted 1 : 50; Gibco]), which is a serum replacement that excludes constituents that lead to cell differentiation, and mammary epithelial growth medium (SingleQuot) (Cambrex Bio Science, Nottingham, UK), which contains aliquots of hydrocortisone, insulin, gentamicin/amphotericin-B, recombinant human EGF, and bovine pituitary extract. The bovine pituitary extract aliquot was omitted from the mammosphere culture medium.
Mammosphere Culture of Ductal Carcinoma In Situ and Normal Breast Tissue
Mammospheres were generated from single cells by seeding a very dilute suspension of enzymatically digested DCIS cells (1 cell per 300 µL mammosphere culture medium) onto poly(2-hydroxyethyl methacrylate) (poly-HEMA [Sigma])–coated 96-well plates. For some experiments, digested DCIS or normal cells were seeded at 1000 cells per cm2 on poly-HEMA–coated tissue culture plates or flasks. The following treatments were added to the DCIS cell suspensions at the time of plating: 1 µM gefitinib (an EGFR tyrosine kinase inhibitor) (Iressa, AstraZeneca, Macclesfield, Cheshire, UK); 5 or 10 µM of the -secretase inhibitor, DAPT (N-[N-(3,5-difluorophenacetyl-L-alanyl)]-S-phenylglycine t-butyl ester) (Calbiochem, Nottingham, UK); or 2 µg/mL of a goat anti-human Notch 4–neutralizing polyclonal antibody (clone-N17 without sodium azide; Santa Cruz Biotechnology Inc, Santa Cruz, CA). Controls were 0.001% DMSO (for the gefinitib and DAPT treatments) and goat immunoglobulin G (IgG) (at 2 µg/mL) (for the antibody treatment). DCIS mammospheres of at least 60 µm in diameter (determined by using an eyepiece graticule with crossed scales) were counted on day 3 after plating (normal mammospheres were always counted on day 7 after plating). Mammosphere forming efficiency (MFE) was calculated by dividing the number of mammospheres (
60 µm) formed by the original number of single cells seeded and is expressed as a percentage. For each DCIS sample, every treatment or control was carried out on at least two replicates. When DCIS MFE was evaluated with respect to tumor grade, low-grade DCIS included samples of grade 1 or 2 and high-grade DCIS included samples of grade 3.
To produce a new generation of mammospheres from original mammospheres cultured, mammospheres grown in mammosphere culture medium were collected by centrifugation (300g) on day 3 or day 4 after seeding, counted, and enzymatically dissociated by incubation for 2 minutes at 37 °C in 0.125% trypsin–EDTA, followed by mechanical dissociation in a 200-µL pipette tip to obtain a single-cell suspension. Cell viability was determined by trypan blue dye exclusion before seeding the cells for a new generation mammosphere growth at a density of 1000 cells per cm2. This procedure was repeated to obtain subsequent generations of mammospheres. A mammosphere generation ratio was calculated at each passage by dividing the number of mammospheres generated in one passage by the number of mammospheres generated in the previous passage.
To prepare mammospheres for immunologic analyses or hematoxylin–eosin staining, primary mammospheres grown in mammosphere culture medium for 3 days were centrifuged at 300g for 2 minutes and then fixed in 4% paraformaldyhyde for 10 minutes at room temperature. The mammospheres were washed in phosphate-buffered saline (PBS) and resuspended in 1% high–melting-point agarose (37 °C), which, after hardening, was embedded in paraffin wax. The paraffin-embedded DCIS mammospheres were then be cut into 3-µm sections.
Matrigel Culture of Dissociated Mammospheres
Single-cell suspensions of dissociated DCIS primary mammospheres or normal breast cells from digested tissue (in DCIS adherent culture medium containing 2% growth factor–reduced matrigel [a biologically active matrix material resembling the mammalian cellular basement membrane; BD Bioscience, Oxford, UK]) were seeded at a density of 5000 cells per well into 8-well glass chamber slides containing 50 µL of 100% growth factor–reduced matrigel per well. Cells were incubated at 37 °C for 21 days with replacement of the growth medium containing 2% growth factor–reduced matrigel every 2–3 days to allow acini to form. The acini within the matrigel cultures were fixed and then removed from the glass slide and resuspended in 1% high–melting-point agarose. After hardening, the agarose was embedded in paraffin, and 3-µm thick sections were cut for immunostaining (described below) or for hematoxylin–eosin staining. Hematoxylin-eosin–stained sections of acini and terminal ductal lobular unit (TDLU)–like structures embedded in matrigel and mammospheres derived from single-cell suspensions of DCIS or normal breast were subsequently reviewed by a breast pathologist who had no prior knowledge of the origin of the samples.
Immunoblot Analysis
Lysates from HB4a cells or from frozen samples of DCIS or normal breast were fractionated (50 µg protein per lane) by SDS–polyacrylamide gel electrophoresis and transferred to Hybond nitrocellulose membranes (Amersham, Little Chalfont, UK). The membranes were incubated in Tris-buffered saline (TBS) containing 0.1% Tween 20 and 5% nonfat milk for 1 hour at room temperature to block nonspecific antibody binding, followed by incubation with the primary antibody for 1 hour at room temperature or overnight at 4 °C (anti–Notch 4 and anti–cleaved Notch 1) with gentle shaking. Primary antibodies (dilutions) were rabbit anti-human AKT polyclonal antibody (1 : 1000; Cell Signaling, Hitchin, UK); rabbit anti-human Phospho-AKT (Ser 473) polyclonal antibody (1 : 1000; Cell Signaling); rabbit anti-human ERK polyclonal antibody (1 : 1000; Santa Cruz Biotechnology Inc), mouse anti-human phospho ERK monoclonal antibody (1 : 1000; Cell Signaling), goat anti-human EGFR polyclonal antibody (1 : 1000; Santa Cruz Biotechnology Inc), rabbit anti-human Phospho-EGFR (Tyr 1068) polyclonal antibody (1 : 1000; Cell Signaling); rabbit anti-human Notch 4 (H-225) polyclonal antibody, 1 : 200; Santa Cruz Biotechnology Inc); rabbit anti-human Notch 1–cleaved N-terminal (Notch 1 intracellular domain [NICD]) polyclonal antibody, which detects only the cleaved intracellular (activated) form of Notch 1 (1 : 500; Rocklands, Gilbertsville, PA); rabbit anti-human Hes1 polyclonal antibody (1 : 1000; Calbiochem); mouse anti-human CK18 monoclonal (1 : 1000; Calbiochem); and mouse anti-human actin monoclonal antibody (1 : 1000; Sigma).
The membranes were washed three times in TBS containing 0.1% Tween 20 and then incubated with a horseradish peroxidase (HRP)–conjugated secondary antibody (depending on primary antibody species polyclonal goat anti-rabbit IgG HRP, polyclonal goat anti-mouse IgG HRP, or polyclonal rabbit anti-goat IgG HRP) (1 : 5000 dilution; Dako, Ely, UK) for 1 hour at room temperature. The membranes were washed and immunoreactive proteins were detected by enhanced chemiluminescence with the use of a kit (Pierce, Cramlington, UK).
Immunohistochemistry
Sections (3-µm thick) of paraffin-embedded DCIS tissue were stained for estrogen receptor 
(ER) and ErbB2 and scored as previously described (9). Briefly, slides were dewaxed, rehydrated, and subjected to antigen retrieval with citrate buffer (pH 6.0) for 2.5 minutes in a pressure cooker. Slides were then incubated for 1 hour with 10% normal rabbit serum to block nonspecific binding, then for 1 hour with the primary antibody (mouse anti-human ER monoclonal antibody [1 : 33 dilution, DAKO]) or mouse anti-human c-erbB2 monoclonal antibody (1 : 50 dilution, Novacastra, UK), and then for 1 hour with a biotinylated rabbit anti-mouse secondary antibody (1 : 300 dilution, DAKO). Antigen–antibody complexes were visualized by applying a standard streptavidin–biotin complex (Vector Laboratories, Peterborough, UK) for 30 minutes followed by diaminobenzidene chromogen (DAB, Sigma) for 8 minutes. Sections were counterstained with the nuclear stain, hematoxylin.
We used a recognized ErbB2-scoring system (15) to score ErbB2 staining as follows: 0 (absent), 1 (weak intensity), 2 (moderate intensity) to 3 (strong intensity), and scores of 2 or higher were considered to be ErbB2 positive ER staining scores were calculated as the percentage of positively stained nuclei. ER positivity was defined as 5% or more of stained nuclei, as previously described (9). For each section, a minimum of 1000 cells was scored across randomly selected areas at a magnification of x400 using a grid graticule and cell counter.
NICD staining was also carried out on 3-µm paraffin-embedded tissue sections. Slides were dewaxed, rehydrated, and subjected to antigen retrieval in citrate buffer (pH 6.0) in a PT module (Lab Vision, Suffolk, UK) for 30 minutes at 98 °C. The slides were incubated for 1 hour in PBS with 10% normal goat serum to block nonspecific antibody binding followed by overnight incubation at 4 °C with rabbit anti-human NICD polyclonal antibody (1 : 500 dilution, Rocklands). After washing, the sections were incubated with biotinylated rabbit anti-mouse secondary antibody (1 : 300 dilution, DAKO) for 1 hour, and antigen–antibody complexes were visualized by applying a standard streptavidin–biotin complex (Vector Laboratories) for 30 minutes followed by DAB (Sigma, UK) for 8 minutes. Sections were counterstained with hematoxylin. Control sections were incubated with a rabbit IgG as the primary antibody.
NICD staining was assessed by light microscopy in both the nucleus and the cytoplasm. Nuclear staining was scored according to the percentage of positively stained nuclei; at least 1000 cells were counted for each sample using a grid graticule and a cell counter at x400 magnification. A nuclear NICD score of 0 corresponded to 0% positively stained nuclei, a score of 1 to 1%–10% positively stained nuclei, a score of 2 to 11%–50% positively stained nuclei, and a score of 3 to 51%–100% positively stained nuclei (16). Cytoplasmic staining of NICD was assessed using the degree of staining intensity, and samples were scored 0 if they had no cytoplasmic staining, 1 if they had weak staining, 2 if they had moderate staining, and 3 if they had strong staining. For assessments of nuclear and cytoplasmic NICD staining, scores 2 and 3 were considered to be positive using cutoff points similar to those for ErbB2 (15). An overall NICD score was determined by either one or both locations being scored positively; for example, a nuclear score of 1 or less but a cytoplasmic score of 2 or more would be considered positive. Scoring of immunostaining was evaluated independently by two evaluators who were blinded to other's scores and to the clinical outcome.
Immunofluorescence Microscopy
Immunofluorescence microscopy was performed essentially as described (17). Briefly, adherent HMEC and DCIS cells grown on glass coverslips were fixed in 4% paraformaldehyde for 10 minutes and then in cold methanol for 2 minutes before staining with specific antibodies. When mammospheres or matrigel structures were paraffin embedded, 3-µm thick sections were cut, and slides were processed as described above (see "Immunohistochemistry") before staining with specific antibodies below.
The following primary antibodies (dilutions) were used for immunofluorescence: FITC-conjugated mouse anti-human CK18 monocolonal antibody (1 : 100; Sigma), FITC-conjugated mouse anti-human epithelial specific antigen monoclonal antibody (1 : 100; Abcam, Cambridge, UK), mouse anti-human CK14 monoclonal antibody (1 : 50; Novocastra, Newcastle upon Tyne, UK), mouse anti-human Ki67 monoclonal antibody (1 : 50; Dako), mouse anti-human ER monoclonal antibody (1 : 100; Vector Laboratories), and mouse anti-human cerbB2 monoclonal antibody (1 : 150; Novocastra). For antibodies not conjugated to fluorescent moieties, slides were washed, followed by 1-hour incubation with appropriate application of biotinylated secondary antibodies (1 : 300 dilution, DAKO). To visualize the antigen, streptavidin Cy3 conjugate (1 : 100 dilution, Sigma, UK) was incubated on slides for 30 minutes. FITC-conjugated primary antibodies needed no further visualization; all slides were then washed and counterstained with 4',6-diamidino-2-phenylindole (DAPI)-mounting media (Vector Laboratories).
Statistical Analysis
Differences in factors affecting nonadherent DCIS culture were calculated using Fisher's exact test. The Mann–Whitney U test was used to compare IC50 values. Recurrence-free survival at 5 years according to NICD expression level was evaluated using Kaplan–Meier plots, and differences in recurrence between the NICD-positive and -negative groups were tested with the log-rank test. A P value of .05 or less was considered to be statistically significant. Bonferroni correction was used to adjust for multiple comparisons in tests of differences in mammosphere formation. When differences were statistically significant (P
.05) using the Kruskal–Wallace test, we used the Mann–Whitney U test to test for differences between treatments (the critical value for statistical significance was reduced from .05 to .00833).
All statistical analyses were performed using SPSS software (version 11.5, SPSS Inc, Chicago, IL) under the guidance of the Medical Statistics Department, Christie Hospital NHS Trust, UK. All statistical tests were two-sided.
|
Results |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
Adherent Culture of Ductal Carcinoma In Situ: Expression of ErbB2 and Autocrine/Paracrine Growth Factors
Of the 16 DCIS samples subjected to adherent culture, 12 produced epithelial cell outgrowth, and six of these cultures generated sufficient luminal epithelium for further experimentation. An example of an adherent DCIS culture that produced luminal epithelial cells is shown in Fig. 1, A: DCIS epithelial cells were defined by the presence of CK18 and the ErbB2 receptor. By contrast, luminal HMEC cultured under the same conditions expressed CK18 but not ErbB2. We used the first four of the six adherent DCIS cultures to generate sufficient luminal epithelium to examine the effect of gefitinib on growth, using HMEC (n = 2) as a control. Gefitinib was a more potent inhibitor of the growth of DCIS cultures than of HMEC cultures (mean IC50: 75 versus 198 nM, difference = 123 nM, 95% confidence interval [CI] = 47.6 to 175 nM, P = .008) (Fig. 1, B).
|
To investigate whether the DCIS cultures were secreting growth factors that may be affected by gefitinib, we generated conditioned medium from the last three of the six DCIS cultures to generate sufficient luminal epithelium by adding serum-free DMEM to the cells for 2 days. The conditioned medium from each of the three different adherent DCIS cultures was added independently to both HMECs and HB4a cells in triplicate for 3 days. An increase in the mean growth rate of both HMEC (2.8-fold [95% CI = 1.76- to 3.85-fold]) and HB4a (9.5-fold [95% CI = 8.25- to 10.8-fold]) was observed compared with that of serum-free DMEM which had not been conditioned with adherent DCIS cultures. With the addition of 1 µM gefitinib to the DCIS-conditioned medium, the increase in the mean growth rate of HMECs statistically significantly decreased to –0.5-fold (95% CI = –1.84- to 0.71-fold, P<.001), suggesting not just a reduction in cell growth but also cell death or detachment of the HMECs originally seeded at day 0 and reduced HB4a cell growth to 7.7-fold (95% CI = 6.8- to 8.7-fold, P = .055) compared to DCIS-conditioned medium alone (Fig. 1, C)
We examined whether these effects on growth were associated with changes in the phosphorylation levels of specific signaling molecules by using phosphorylation state–specific antibodies Addition of conditioned medium from three adherent DCIS cultures caused marked increases in the levels of phosphorylated EGFR and phosphorylated mitogen-activated protein (MAP) kinase and a small increase in the level of phosphorylated AKT in HB4a cells compared with HB4a cells cultured in serum-free DMEM alone (Fig. 1, D, lanes 2, 3, and 4). By contrast, conditioned medium containing 1 µM gefitinib failed to elicit phosphorylation of EGFR, ERK, or AKT (Fig. 1, D, lanes 6, 7, and 8). These results suggest that adherent DCIS cultures secrete growth factors that increase proliferation in HMEC and HB4a cells via activation of EGFR and downstream signaling pathways, including the MAP kinase and AKT pathways.
Nonadherent Culture of Ductal Carcinoma In Situ and Normal Breast Cells as Mammospheres
We used nonadherent cell culture to compare DCIS MFE with that of normal human breast tissue. DCIS mammospheres were successfully cultured in 20 (69%) of 29 cases, and all the DCIS mammosphere cultures generated sufficient material for further experimentation. The ER or ErbB2 expression status or histologic grade of the DCIS sample did not statistically significantly affect the ability to form mammospheres (Table 1).
|
To demonstrate that DCIS mammospheres were clonal in origin, single-cell suspensions from primary DCIS tissue taken at surgery were seeded at very low density (i.e., one cell per well) in a 96-well plate. Figure 2, A, shows the formation of a DCIS mammosphere from a single cell for a period of 5 days. DCIS mammosphere culture conditions were optimized for further experiments; seeding DCIS cells at a density of 1000 cells per cm2 generated a similar MFE to that of the same DCIS cells grown from single isolated cells (1.1% versus 0.9%). However, in cultures seeded at the higher density, DCIS mammospheres of at least 60 µm in diameter formed within 3 days (Figs. 2, B and 3, A), instead of 5 days when DCIS mammospheres were formed from isolated cells. The reduction in the time needed to form a DCIS mammosphere of 60 µm may be the result of autocrine/paracrine growth factors accumulating at higher concentrations in higher density cell cultures.
|
|
To characterize the cellular composition of the mammospheres and to confirm that they arose from DCIS cells, sections of paraffin-embedded mammospheres were stained with hematoxylin–eosin and immunostained for epithelial cell markers. All cells grown in matrigel and as mammospheres derived from DCIS had pleomorphic nuclei with more pronounced nucleoli than normal breast cells grown in matrigel and than mammospheres derived from normal breast tissue, which were consistent with the nuclear grade of the malignant cells within the original lesion. Of 200 DCIS mammospheres examined, 90 (45%) contained both luminal (CK18-positive) and myoepithelial (CK14-positive) cell types (Fig. 2, D), 18 (9%) contained luminal cells only, and 24 (12%) contained myoepithelial cells only; 85% of the cells within these 200 DCIS mammospheres were positive for epithelial specific antigen (Fig. 2, C), suggesting that even when the differentiated epithelial markers were not detected, the DCIS mammospheres still contained epithelial specific cells. In cases of high-grade DCIS and where the ErbB2 receptor status of the original tumor was known or in low-grade DCIS, where the tumor was known to be ER
positive, the resulting mammospheres exhibited the corresponding staining patterns (Fig. 2, E and F). Ki67, a marker of active cell proliferation, was expressed in 84% of the 200 DCIS mammospheres examined (Fig. 2, G). We next examined the growth and differentiation of single cells from dissociated DCIS mammospheres in 3-D matrigel cultures, first to assess whether the 3-D structures would have a DCIS-like morphology and second to examine whether cells from DCIS mammospheres have multilineage potential. During 21 days in 3-D matrigel culture, single cells from dissociated DCIS mammospheres formed TDLU-like structures and acini (Fig. 2, H), but both types of structures were disorganized and lacked lumens (Fig. 2, I); they also contained both luminal (CK18-positive) and myoepithelial (CK14-positive) cells (Fig. 2, K). In addition, cells within the disorganized structures showed marked nuclear pleomorphism and pronounced nucleoli (Fig. 2, J), which were consistent with the nuclear grade of the malignant cells within the original lesion. By contrast, cells from dissociated normal breast tissue cultured in the same conditions formed organized TDLU-like structures and acini with hollow lumen and normal tissue architecture (Fig. 2, L–N).
Effect of Histologic Grade on Ductal Carcinoma In Situ Mammosphere Formation
When primary normal HMECs were seeded at 1000 cells per cm2 and grown using the nonadherent culture conditions used for DCIS, normal breast mammospheres of at least 60 µm in diameter were only observed after 7 days of culture. By contrast, DCIS mammospheres larger than 60 µm were prevalent after only 3 days of culture (Fig. 3, A), indicating that they have a higher growth rate than HMEC grown under the same conditions. In light of this finding, all subsequent DCIS mammosphere experiments were seeded at 1000 cells per cm2 and mammospheres of at least 60 µm in diameter were counted on day 3 after plating to minimize the possibility of contamination by mammospheres derived from normal mammary epithelial cells.
Mammospheres derived from normal breast tissue and from DCIS had the capacity to form new generations of mammospheres when passaged as single cells and reseeded under nonadherent culturing conditions. However, DCIS mammospheres were capable of regenerating mammospheres up to six times at an average mammosphere regeneration ratio of 0.85 at each passage. By contrast, normal breast mammospheres produced a maximum of three new generations of mammosphere with a statistically significantly lower average mammosphere regeneration ratio of 0.13 (0.85 versus 0.13, difference = 0.72, 95% CI = 0.56 to 0.86, P<.001).
DCIS mammospheres also had a statistically significantly higher MFE than normal breast mammospheres (1.5% versus 0.5%, difference = 1%, 95% CI = 0.62% to 1.25%, P<.001). In addition, the MFE for high-grade (grade 3) DCIS was statistically significantly greater than that for low-grade (grade 1 or 2) DCIS (1.6% versus 1.09%, difference = 0.51%, 95% CI = 0.07% to 0.94%, P = .01) (Fig. 3, A).
We next examined whether DCIS mammosphere formation was dependent on EGF, which was present in the standard complete medium used for mammosphere culture. Mammospheres were grown for 3 days in complete medium that contained or lacked EGF, at which time DCIS MFE was evaluated with respect to tumor grade (grouped as either low or high grade). The addition of exogenous EGF did not alter the MFE for either group. In the presence of exogenous EGF, the addition of gefitinib statistically significantly decreased the MFE for both low-grade DCIS (1.28% versus 0.8%, difference = 0.48%, 95% CI = 0.15% to 0.78%, P = .003) and high-grade DCIS (1.6% versus 0.75%, difference = 0.85%, 95% CI = 0.48% to 1.1%, P<.001) compared with exogenous EGF alone. The MFE of high-grade DCIS treated with gefitinib in the absence of exogenous EGF was lower than that of high-grade DCIS treated with mammosphere medium lacking gefitinib and exogenous EGF (0.56% versus 1.36%, difference 0.8%, 95% CI = 0.33% to 1.4%, P = .004) (Fig. 3, B). This finding suggests that high-grade DCIS secretes growth factors that regulate mammosphere initiation and/or growth via the EGFR signaling pathways.
Stratification of DCIS samples by ErbB2 expression status revealed that mammosphere cultures grown from ErbB2-positive DCIS had a higher MFE than those grown from ErbB2-negative DCIS (P = .004) (Fig. 3, C). The addition of exogenous EGF to the culture medium did not alter the MFE in either group. However, in the absence of exogenous EGF, the addition of gefitinib reduced the MFE in cultures grown from ErbB2-positive DCIS compared to those grown from mammosphere medium with no exogenous EGF alone, but not those grown from ErbB2–negative samples (P = .0077 and P = .046, respectively; Fig. 3, C). This finding suggests that gefitinib predominantly inhibits the tyrosine kinase activity of the EGFR/ErbB2 heterodimer.
Notch Expression and Activation in Ductal Carcinoma In Situ and Association With Recurrence
The failure of the EGF inhibitor gefitinib to completely inhibit mammosphere formation indicated that other signaling pathways were also promoting mammosphere formation. Because recent data suggested a role for Notch signaling in normal breast tissue (7) and because Notch signaling is aberrantly activated in invasive breast cancers (11), we investigated the activity of the Notch signaling pathway in DCIS tissue. To monitor Notch signaling, we examined accumulation of NICD and Notch 4 (4ICD) intracellular domain and the expression of Hes1, a downstream target of Notch receptor activation, in DCIS and normal breast tissue by immunoblot analysis. NICD and 4ICD are released from the full-length Notch receptor by two proteolytic cleavages after it interacts with ligand; the first cleavage occurs in the extracellular compartment and is mediated by members of the disintegrin and metalloprotease domain protease family (18,19), and the second cleavage is an intramembranous cleavage by -secretase. The levels of NICD and Hes1 were increased in all DCIS samples examined compared with normal breast tissue, suggesting that dysregulation of Notch signaling may be an early event in breast cancer (Fig. 4, A).
|
To obtain further evidence for a possible role of Notch signaling in breast cancer, we used immunohistochemistry to examine NICD expression in paraffin-embedded tissue from a separate group of 50 patients who presented with pure DCIS and underwent breast surgery. During 60 months of follow-up, 16 of these patients experienced a recurrence (median time to recurrence = 14 months, range = 11–60 months) and 34 did not. Positive staining for NICD expression, as opposed to no staining, was associated with shorter median time to recurrence at 60 months after surgery (P = .012, log-rank test) (Fig. 4, B).
Effect of Notch Antagonists on Primary Ductal Carcinoma In Situ Mammosphere Formation
Because expression of activated (i.e., cleaved) Notch was associated with recurrence after breast surgery among DCIS patients with 60 months of follow-up, we examined whether inhibiting Notch cleavage with the -secretase inhibitor DAPT would affect DCIS MFE. DCIS tissue samples (n = 9) treated for 3 days with 5 or 10 µM DAPT had reduced DCIS MFE compared with DCIS samples treated with DMSO (control) (control versus 5 µM DAPT: 0.89% versus 0.67%, difference 0.22%, 95% CI = 0.05 to 0.48, P = .02; control versus 10 µM DAPT: 0.89% versus 0.51%, difference = 0.38%, 95% CI = 0.2% to 0.6%, P<.001; Fig. 5, A). Treatment of DCIS tissue samples (n = 6) with a Notch 4–neutralizing antibody reduced DCIS MFE to a greater extent than DAPT when compared with treatment with the control antibody (0.97% versus 0.2%, difference = 0.77%, 95% CI = 0.52% to 1.0%, P<.001), and two of the six treated DCIS samples were no longer capable of producing mammospheres (Fig. 5, B and C). These results suggest that the Notch receptor signaling pathway is directly involved in the regulation of DCIS mammosphere formation and/or growth.
|
|
Discussion |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
This is the first report, to our knowledge, to describe reproducible DCIS culture in vitro. Previously, we described a xenograft DCIS model in which a role for the EGFR pathway in facilitating epithelial growth was demonstrated in ER
-positive and -negative DCIS (4,13). However, we were unable to determine the mechanism by which EGFR signaling was increased in DCIS or how the pathway regulated tumor growth. We therefore sought to develop novel in vitro methods for investigating the biology of DCIS. Using nonadherent culture, we have isolated cells that can self-renew and proliferate to form floating colonies called mammospheres that can differentiate and form DCIS-like structures when disaggregated and plated into matrigel. The greater MFE of epithelial cells from DCIS than that of cells from normal breast tissue suggests that DCIS has a greater number of stem or progenitor cells than normal tissue. Finally, using this culture system we were able to demonstrate that inhibition of either EGFR or Notch signaling disrupts mammosphere formation, which may reflect a role for these signaling pathways in the growth of DCIS in vivo. Historically, the adherent growth of malignant cells in culture from solid tumors has been problematic because of overgrowth by stromal cells or fibroblasts (5,20), little or no growth of malignant cells (5,21,22), or loss of proliferative capacity (senescence) in culture (23,24). Similar problems arose when we used adherent culture to grow DCIS-derived epithelial cells: although only one-third of the adherent cultures generated in this report provided data, the results nonetheless demonstrated that the EGFR pathway is important for growth, particularly of high-grade DCIS. We next successfully used the mammosphere culture technique, which has been used previously to isolate cells from normal human breast tissue and breast cancers and induce them to form clonal floating colonies (i.e., mammospheres) that can be passaged a few times (6,25). The colony-forming cells may be similar to stem cells that have been isolated from human and murine mammary glands (26–29). It has been suggested that cancers arise from the accumulation of mutations within stem cells that disrupt their very tightly controlled self-renewal and proliferation processes (30,31); this possibility is supported by the higher capacity of CD44-positive/CD24-negative breast cancer cells, which are enriched for stem cells, than CD44-negative/CD24-positive cells to form mammospheres and/or tumors when injected into immunocompromised mice (25,32,33). To our knowledge, no culture technique for DCIS exists; thus, the nonadherent culture technique we describe should be useful for isolating tumor-forming epithelial cells from primary DCIS lesions to allow a better understanding of their growth.
Our finding that primary DCIS tissue contains a small subset of cells with the ability to form mammospheres in a nonadherent culture system suggests the presence of stem cell–like cells. Mammospheres were formed from 69% of primary DCIS lesions; all the mammospheres produced from the lesions could be passaged repeatedly, indicating their capability for self-renewal. The multipotency of these cells was demonstrated by their ability to produce unorganized structures that contained both luminal and myoepithelial cell lineages when grown in matrigel. Compared with normal breast tissue, primary DCIS tissue contained a greater number (1.5% versus 0.5%) of cells with the ability to form mammospheres, and these mammosphere-forming cells had greater self-renewal capacity than normal breast tissue as demonstrated by their ability to produce mammospheres for more generations and their higher mammosphere regeneration ratio than normal breast. Finally, MFE increased with histologic grade, which supports the cancer stem cell hypothesis (30,31) because high-grade DCIS has the highest risk of recurrence (3,9). Similar findings have been observed in brain cancers, the most aggressive clinical samples of which have stem cells with the highest sphere formation and self-renewal capacities (34).
Overall, the mammosphere culture system described here was a more robust in vitro system than adherent culture. Furthermore, our demonstration that EGFR signaling is required for DCIS mammosphere formation, and similarly for the proliferation of DCIS xenografts (4), indicates that this culture technique will be useful for analyzing signaling pathways involved in tumor formation.
Our previous xenograft study (4) showed that EGFR signaling is required for the growth of DCIS tumors, but we were unable to determine how the pathway was activated. The importance of the pathway is also illustrated by the fact that stimulation of EGFR using EGF-supplemented medium is necessary for the growth of most dissociated tissue grown in nonadherent culture (6,32,34,35). Investigation of the activation of EGFR signaling pathway using our adherent in vitro culture system showed that conditioned medium from adherent DCIS cultures increased growth of both the HB4a cell line and primary HMEC, causing a corresponding increase in EGFR and MAP kinase phosphorylation that was inhibited by the EGFR inhibitor, gefitinib. These results indicate that the conditioned medium contains an EGFR ligand. The secretion of this ligand by DCIS-derived epithelial cells would also explain the faster growth of mammospheres when cells were plated at a higher density and the greater sensitivity of high-grade DCIS to gefitinib. High-grade DCIS with ErbB2 activation is associated with early recurrence after surgery (9). The EGFR ligand is likely to promote EGFR/ErbB2 heterodimerization, and although further research is required to establish the identity of the EGFR ligand in conditioned medium, these data indicate the usefulness of this model system for analyzing the pathways involved in the growth of DCIS.
Elevated levels of Notch1 and Jagged1 mRNA have been shown to be associated with poor prognosis in invasive human breast cancer (36), and our previous data showing aberrant activation of Notch signaling in a broad range of invasive breast cancers suggests that this activation may occur early in tumor development (11). Our finding of increased NICD accumulation and Hes1 expression in DCIS compared with normal breast tissue confirms this possibility. Notch activation was associated with recurrence at 5 years after surgery, a finding that suggests both that accumulation of NICD may be a useful prognostic marker for recurrence and that changes in the Notch pathway may be associated with the progression from DCIS to invasive disease. These results also raise the question of whether targeting the Notch pathway would be therapeutically useful for the treatment of DCIS. The ability of both a -secretase inhibitor (DAPT) and a Notch 4–neutralizing antibody to prevent mammosphere formation strongly suggests that this may be the case. Given the known gut toxicity of -secretase inhibitors in rodents (37), further work is needed to determine the relative efficacy of inhibiting individual Notch receptors with antibodies or using RNA interference (37). Previous work has shown that Notch signaling can inhibit apoptosis (38). The loss of mammosphere viability we observed for two of the six DCIS samples cultured in the presence of Notch 4–neutralizing antibody suggests that disrupting Notch signaling may induce cell death in this model. By contrast, a study carried out on normal breast mammospheres cultured with Notch 4 receptor–neutralizing antibody did not show a similar loss of viability, although the self-renewal of colony-forming cells was blocked (7).
Our study has several limitations. The small numbers of low- and intermediate-grade samples used for the nonadherent cell culture experiments may have contributed to the non–statistically significant results seen in Table 1. We have also not yet implanted DCIS mammospheres into NOD/SCID mice to prove that they can recapitulate the same DCIS lesion in vivo; however, studies on brain and colon cancer spheres have shown that number of spheres formed in vitro correlates with tumor grade and that it is the sphere-forming cells that can reproduce tumors following transplantation into NOD/SCID mice (33,39). Therefore, although in vivo work would definitively prove DCIS mammospheres contained DCIS-initiating cells, the sphere culture method has been shown to enrich for tumor-initiating cells in several invasive cancer types including the breast (25,34,39).
In summary, we used a novel method for DCIS culture in vitro and showed that EGFR is necessary for DCIS growth and self-renewal and that Notch signaling is important for cell survival and/or self-renewal in nonadherent culture. These culture methods may be useful for the discovery of pathways that are important for DCIS propagation and for the testing of novel inhibitors to prevent DCIS recurrence and progression to invasive disease.
|
NOTES |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
We thank Spyros Stylianou and Dr Fiona Knox for help and advice. This work was supported by Breast Cancer Campaign grants 2001:201 and 2005MAY21. Breast Cancer Campaign did not have any role in the design of the study; the collection, analysis, or interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication.
|
REFERENCES |
|---|
|
TopAbstractContext and CaveatsMaterials and MethodsResultsDiscussionReferencesNotes |
|---|
(1) Leonard GD, Swain SM. Ductal carcinoma in situ, complexities and challenges. J Natl Cancer Inst (2004) 96:906–20.
(2) Page DL, Dupont WD, Rogers LW, Jensen RA, Schuyler PA. Continued local recurrence of carcinoma 15-25 years after a diagnosis of low grade ductal carcinoma in situ of the breast treated only by biopsy. Cancer (1995) 76:1197–200.[CrossRef][Web of Science][Medline]
(3) Bijker N, Meijnen P, Peterse JL, Bogaerts J, Van Hoorebeeck I, Julien JP, et al. Breast-conserving treatment with or without radiotherapy in ductal carcinoma in situ: ten-year results of European Organisation for Research and Treatment of Cancer randomized phase III trial 10853—a study by the EORTC Breast Cancer Cooperative Group and EORTC Radiotherapy Group. J Clin Oncol (2006) 24:3381–7.
(4) Chan KC, Knox WF, Gee JM, Morris J, Nicholson RI, Potten CS, et al. Effect of epidermal growth factor receptor tyrosine kinase inhibition on epithelial proliferation in normal and premalignant breast. Cancer Res (2002) 62:122–8.
(5) Speirs V, Green AR, Walton DS, Kerin MJ, Fox JN, Carleton PJ, et al. Short-term primary culture of epithelial cells derived from human breast tumours. Br J Cancer (1998) 78:1421–9.[Web of Science][Medline]
(6) Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ, et al. In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes Dev (2003) 17:1253–70.
(7) Dontu G, Jackson KW, McNicholas E, Kawamura MJ, Abdallah WM, Wicha MS. Role of Notch signaling in cell-fate determination of human mammary stem/progenitor cells. Breast Cancer Res (2004) 6:R605–15.[CrossRef][Web of Science][Medline]
(8) Debnath J, Muthuswamy SK, Brugge JS. Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement membrane cultures. Methods (2003) 30:256–68.[CrossRef][Web of Science][Medline]
(9) Barnes NL, Khavari S, Boland GP, Cramer A, Knox WF, Bundred NJ. Absence of HER4 expression predicts recurrence of ductal carcinoma in situ of the breast. Clin Cancer Res (2005) 11:2163–8.
(10) Neve RM, Lane HA, Hynes NE. The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J (2000) 19:3159–67.[CrossRef][Web of Science][Medline]
(11) Stylianou S, Clarke RB, Brennan K. Aberrant activation of notch signaling in human breast cancer. Cancer Res (2006) 66:1517–25.
(12) Pece S, Serresi M, Santolini E, Capra M, Hulleman E, Galimberti V, et al. Loss of negative regulation by Numb over Notch is relevant to human breast carcinogenesis. J Cell Biol (2004) 167:215–21.
(13) Holland PA, Knox WF, Potten CS, Howell A, Anderson E, Baildam AD, et al. Assessment of hormone dependence of comedo ductal carcinoma in situ of the breast. J Natl Cancer Inst (1997) 89:1059–65.
(14) Skehan P, Storeng R, Scudiero D, Monks A, McMahon J, Vistica D, et al. New colorimetric cytotoxicity assay for anticancer-drug screening. J Natl Cancer Inst (1990) 82:1107–12.
(15) Birner P, Oberhuber G, Stani J, Reithofer C, Samonigg H, Hausmaninger H, et al. Evaluation of the United States Food and Drug Administration-approved scoring and test system of HER-2 protein expression in breast cancer. Clin Cancer Res (2001) 7:1669–75.
(16) Leake R, Barnes D, Pinder S, Ellis I, Anderson L, Anderson T, et al. Immunohistochemical detection of steroid receptors in breast cancer: a working protocol. UK Receptor Group, UK NEQAS, The Scottish BreastCancer Pathology Group, and The Receptor and Biomarker Study Group of the EORTC. J Clin Pathol (2000) 53:634–5.
(17) Clarke RB, Howell A, Potten CS, Anderson E. Dissociation between steroid receptor expression and cell proliferation in the human breast. Cancer Res (1997) 57:4987–91.
(18) Brou C, Logeat F, Gupta N, Bessia C, LeBail O, Doedens JR, et al. A novel proteolytic cleavage involved in Notch signaling: the role of the disintegrin-metalloprotease TACE. Mol Cell (2000) 5:207–16.[CrossRef][Web of Science][Medline]
(19) Mumm JS, Kopan R. Notch signaling: from the outside in. Dev Biol (2000) 228:151–65.[CrossRef][Web of Science][Medline]
(20) Brower M, Carney DN, Oie HK, Gazdar AF, Minna JD. Growth of cell lines and clinical specimens of human non-small cell lung cancer in a serum-free defined medium. Cancer Res (1986) 46:798–806.
(21) Ethier SP, Mahacek ML, Gullick WJ, Frank TS, Weber BL. Differential isolation of normal luminal mammary epithelial cells and breast cancer cells from primary and metastatic sites using selective media. Cancer Res (1993) 53:627–35.
(22) Yasunaga Y, Nakamura K, Ewing CM, Isaacs WB, Hukku B, Rhim JS. A novel human cell culture model for the study of familial prostate cancer. Cancer Res (2001) 61:5969–73.
(23) Holt SE, Shay JW, Wright WE. Refining the telomere-telomerase hypothesis of aging and cancer. Nat Biotechnol (1996) 14:836–9.[CrossRef][Web of Science][Medline]
(24) Shay JW, WrightAging WE. When do telomeres matter? Science (2001) 291:839–40.
(25) Ponti D, Costa A, Zaffaroni N, Pratesi G, Petrangolini G, Coradini D, et al. Isolation and in vitro propagation of tumorigenic breast cancer cells with stem/progenitor cell properties. Cancer Res (2005) 65:5506–11.
(26) Shackleton M, Vaillant F, Simpson KJ, Stingl J, Smyth GK, Asselin-Labat ML, et al. Generation of a functional mammary gland from a single stem cell. Nature (2006) 439:84–8.[CrossRef][Medline]
(27) Stingl J, Eirew P, Ricketson I, Shackleton M, Vaillant F, Choi D, et al. Purification and unique properties of mammary epithelial stem cells. Nature (2006) 439:993–7.[Medline]
(28) 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) 277:443–56.[CrossRef][Web of Science][Medline]
(29) Sleeman KE, Kendrick H, Ashworth A, Isacke CM, Smalley MJ. CD24 staining of mouse mammary gland cells defines luminal epithelial, myoepithelial/basal and non-epithelial cells. Breast Cancer Res (2006) 8:R7.[CrossRef][Medline]
(30) Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem cells, cancer, and cancer stem cells. Nature (2001) 414:105–11.[CrossRef][Medline]
(31) Wang JC, Dick JE. Cancer stem cells: lessons from leukemia. Trends Cell Biol (2005) 15:494–501.[CrossRef][Web of Science][Medline]
(32) 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) 100:3983–8.
(33) Phillips TM, McBride WH, Pajonk F. The response of CD24(-/low)/CD44+ breast cancer-initiating cells to radiation. J Natl Cancer Inst (2006) 98:1777–85.
(34) Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, et al. Identification of human brain tumour initiating cells. Nature (2004) 432:396–401.[CrossRef][Medline]
(35) Collins AT, Berry PA, Hyde C, Stower MJ, Maitland NJ. Prospective identification of tumorigenic prostate cancer stem cells. Cancer Res (2005) 65:10946–51.
(36) Reedijk M, Odorcic S, Chang L, Zhang H, Miller N, McCready DR, et al. High-level coexpression of JAG1 and NOTCH1 is observed in human breast cancer and is associated with poor overall survival. Cancer Res (2005) 65:8530–7.
(37) Van Es JH, van Gijn ME, Riccio O, van den Born M, Vooijs M, Begthel H, et al. Notch/gamma-secretase inhibition turns proliferative cells in intestinal crypts and adenomas into goblet cells. Nature (2005) 435:959–63.[CrossRef][Medline]
(38) Leong KG, Karsan A. Recent insights into the role of Notch signaling in tumorigenesis. Blood (2006) 107:2223–33.
(39) Ricci-Vitiani L, Lombardi DG, Pilozzi E, Biffoni M, Todaro M, Peschle C, et al. Identification and expansion of human colon-cancer-initiating cells. Nature (2007) 445:111–5.[CrossRef][Medline]
Manuscript received September 26, 2006; revised January 31, 2007; accepted February 27, 2007.
Related Articles in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2007 99: 577.
J Natl Cancer Inst 2007 99: 1284-1285.
This article has been cited by other articles:
|
|
 |
|
  O. Meurette, S. Stylianou, R. Rock, G. M. Collu, A. P. Gilmore, and K. Brennan Notch Activation Induces Akt Signaling via an Autocrine Loop to Prevent Apoptosis in Breast Epithelial Cells Cancer Res., June 15, 2009; 69(12): 5015 - 5022. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Mercier, M. C. Casimiro, J. Zhou, C. Wang, C. Plymire, K. G. Bryant, K. M. Daumer, F. Sotgia, G. Bonuccelli, A. K. Witkiewicz, et al. Genetic Ablation of Caveolin-1 Drives Estrogen-Hypersensitivity and the Development of DCIS-Like Mammary Lesions Am. J. Pathol., April 1, 2009; 174(4): 1172 - 1190. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Magnifico, L. Albano, S. Campaner, D. Delia, F. Castiglioni, P. Gasparini, G. Sozzi, E. Fontanella, S. Menard, and E. Tagliabue Tumor-Initiating Cells of HER2-Positive Carcinoma Cell Lines Express the Highest Oncoprotein Levels and Are Sensitive to Trastuzumab Clin. Cancer Res., March 15, 2009; 15(6): 2010 - 2021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. D. Meng, C. C. Shelton, Y.-M. Li, L.-X. Qin, D. Notterman, P. B. Paty, and G. K. Schwartz {gamma}-Secretase Inhibitors Abrogate Oxaliplatin-Induced Activation of the Notch-1 Signaling Pathway in Colon Cancer Cells Resulting in Enhanced Chemosensitivity Cancer Res., January 15, 2009; 69(2): 573 - 582. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Miele, N. Takebe, and S. P. Ivy The Cancer Stem Cell Hypothesis, Embryonic Signaling Pathways, and Therapeutics: Targeting an Elusive Concept ASCO Educational Book, January 1, 2009; 2009(1): 145 - 156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Sang, H. A. Coller, and J. M. Roberts Control of the Reversibility of Cellular Quiescence by the Transcriptional Repressor HES1 Science, August 22, 2008; 321(5892): 1095 - 1100. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. W. Lee, C. M. Raskett, I. Prudovsky, and D. C. Altieri Molecular Dependence of Estrogen Receptor-Negative Breast Cancer on a Notch-Survivin Signaling Axis Cancer Res., July 1, 2008; 68(13): 5273 - 5281. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Li, M. T. Lewis, J. Huang, C. Gutierrez, C. K. Osborne, M.-F. Wu, S. G. Hilsenbeck, A. Pavlick, X. Zhang, G. C. Chamness, et al. Intrinsic Resistance of Tumorigenic Breast Cancer Cells to Chemotherapy J Natl Cancer Inst, May 7, 2008; 100(9): 672 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Rizzo, C. Osipo, K. Foreman, J. Yun, T. Golde, B. Osborne, and L. Miele Rational Targeting of Notch in Breast Cancer Am. Assoc. Cancer Res. Educ. Book, April 12, 2008; 2008(1): 491 - 495. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. F. O'Neill, S. Urs, C. Cinelli, A. Lincoln, R. J. Nadeau, R. Leon, J. Toher, C. Mouta-Bellum, R. E. Friesel, and L. Liaw Notch2 Signaling Induces Apoptosis and Inhibits Human MDA-MB-231 Xenograft Growth Am. J. Pathol., September 1, 2007; 171(3): 1023 - 1036. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||