Introduction
Breast cancer is the major contributor to cancer incidence and mortality in women in the Western world. Although the genetic and environmental factors that lead to the initiation of breast cancer remain unclear, it is known that exposure to estrogens plays a crucial role in the development and progression of this disease [
1]. It has been proposed that the causative link between estrogen and breast cancer is due to its potent mitogenic and antiapoptotic effects [
2]. However, it is not fully understood how these effects are mediated at the molecular level. Such insight may provide clues to the mechanisms of estrogen-induced mitogenesis and cell survival, or resistance to endocrine therapies, or identify potential novel therapeutic targets for breast cancer, particularly in the settings of endocrine insensitivity and resistance. Thus, the identification and characterization of estrogen target genes is a major research priority.
The majority of breast cancers (about 75%) are estrogen receptor (ER)-positive, and estrogen is a potent mitogen for human breast cancer cells
in vitro. The proliferation of ER-positive MCF-7 breast cancer cells in culture is inhibited by antiestrogens, and this effect is reversed by estrogen. Estrogen and antiestrogens regulate cell cycle entry and rates of progression during early G
1 phase [
3‐
5], and this is effected by modulation of G
1 cyclin gene expression and activation of cyclin-dependent kinases 2 and 4 [
6,
7]. In addition, there is now evidence of converging activation of downstream estrogen signaling through crosstalk with growth factor-activated tyrosine kinase receptors [
8]. Thus, there are compelling data suggesting that estrogen can mediate its growth effects by influencing the expression and function of genes critical to cell proliferation, by both 'genomic' and 'nongenomic' (cytoplasmic signaling) mechanisms [
9].
One of the earliest transcriptional responses to estrogen is increased
MYC expression [
10]. Myc is a nuclear transcription factor that exhibits high-affinity and site-specific DNA-binding activity when complexed with its cellular partner Max, and it is rate-limiting for cell cycle progression through G
1 phase [
11], mediated in part through its effects on activation of cyclin-dependent kinases [
12,
13]. Inhibition of c-Myc expression abrogates estrogen-stimulated breast cancer cell proliferation [
14], and blocks cell cycle progression leading to a G
1 arrest [
15]. Estrogen-regulated induction of
MYC may play a critical role in the initiation of breast tumorigenesis, because
MYC was the first mammary oncogene to be demonstrated by transgenesis [
16]. These data strongly implicate c-Myc as an important mediator of the mitogenic function of estrogen, with a potential role in the initiation and progression of breast cancer. This concept is supported by studies demonstrating that Myc over-expression confers resistance to antiestrogens
in vitro [
17,
18], and that inducible expression of c-Myc can replace estrogen in reinitiating cell cycle progression in antiestrogen-arrested breast cancer cells [
12].
Because c-Myc can mimic the effects of estrogen on cell cycle progression in MCF-7 cells [
12], we examined the transcriptional response to estrogen and to inducible c-Myc to identify novel targets of both estrogen and c-Myc in breast cancer cells (Musgrove EA, Sergio CM, Butt AJ, Sutherland RL; unpublished data). Here, we report an initial characterization of one such gene, namely HBV pre-S2 trans-regulated protein 3 (
HSPC111). These studies reveal that
HSPC111 is a direct transcriptional target of Myc, which is localized in the nucleolus and is over-expressed in several common cancers. Furthermore, elevated expression of HSPC111 is associated with reduced survival in breast cancer patients.
Materials and methods
Breast cancer cell lines and tissue samples
The human breast cancer cell line, MCF-7, was routinely maintained in RPMI-1640 medium supplemented with 10% fetal calf serum, 10 μg/ml insulin and 2.92 mg/ml glutamine under standard conditions. The human breast cancer mRNA samples utilized in this study have previously been described [
19].
Quantitative real-time PCR
Total RNA was isolated using the RNAeasy kit (Qiagen Pty Ltd, Victoria, Australia) from cells pretreated with ICI 182780 (7α-[9-(4,4,5,5,5-pentafluoropentylsulfinyl) nonyl] estra-1,3,5,(10)-triene-3,17β-diol), which was a kind gift from Dr Alan Wakeling (Astra-Zeneca Pharmaceuticals, Alderly Park, Cheshire, UK), and then treated with 17β-estradiol, zinc, or vehicle. Quantitative real-time PCR was performed using the ABI Prism 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA, USA) using Taq-Man® probes for HSPC111 (Applied Biosytems). Data analyses were performed using the ΔCt method with RPLP0 (Applied Biosystems) as internal loading control. Fold changes in gene expression were calculated relative to the 0 hours time point. For correlation experiments, total RNA from a panel of breast cancer cell lines was isolated and quantitative real-time PCR was performed using Taq-Man® probes for MYC and HSPC111. Correlation was performed using standard linear regression analysis.
Immunoblot analysis
Cell lysates were collected as described previously [
6]. Antibodies used were HSPC111 (see below) or V5 (Invitrogen Life Technologies Inc., Carlsbad, CA, USA). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH; Ambion, Austin, TX) or actin (Sigma, St Louis, MO, USA) was used as loading control.
Constructs
The sequence between -799 and +43 base pairs (bp) of the HSPC111 promoter was amplified by nested PCR from MCF-7-derived genomic DNA. The resulting 842 bp fragment was cloned into pGL3-Basic reporter construct (Promega, Madison, WI, USA).
Luciferase reporter assays
MCF-7 cells were transfected using Lipofectamine 2000 (Invitrogen) with luciferase reporter construct, renilla luciferase reporter construct, pRLSV40 (Promega), and either the c-Myc expression plasmid pCDNA3.1-cMyc or pcDNA3.1. Transfected cells were stimulated with increasing concentrations of zinc (up to 80 μmol/l) for 6 hours before harvesting. Luciferase activity was assayed 24 hours after transfection using the Dual-Luciferase Reporter Assay System (Promega) and normalized to renilla luciferase activity. All values are relative to the activity of the pGL3-Basic reporter.
Electrophoretic mobility shift assays (EMSAs)
The sequences of the oligonucleotides used to investigate the three putative c-Myc binding sites in the HSCP111 promoter were as follows: HSPCsite1(TOP): 5'-CTAGGAGGCCCATGTGTCGCTG-3' ; HSPCsite1(BOT): 5'-CTAGCAGCGACACATGGGCCTC-3' ; HSPCsite2(TOP): 5'-CTAGGGCTCACACCTGTAATCC-3' ; HSPCsite2(BOT): 5'-CTAGGGATTACAGGTGTGAGCC-3' ; HSPCsite3(TOP): 5'-CTAGGCGGATCACCTGAGGTCA-3' ; HSPCsite3(BOT): 5'-CTAGTGACCTCAGGTGATCCGC-3' ; CAD(TOP): 5'-CTAGGTTAGCCACGTGGACCGA-3' ; and CAD(BOT): 5'-CTAGTCGGTCCACGTGGCTAAC-3'. The annealed oligonucleotides were radiolabeled with [α-32P]dCTP using Klenow fragment. Electrophoretic mobility shift assays were performed using nuclear extracts from MCF-7 cells. Equal amounts of nuclear extracts were incubated with the radiolabeled oligonucleotides following standard protocols, resolved on a 5% acrylamide gel and visualized by autoradiography. Competition assays were performed using 100-fold excess of competitor unradiolabeled oligonucleotides. The following oligonucleotides were used as nonspecific competitor oligonucleotides: 5'-CTAGTCTACTCCACTGCTGTCTATC-3' and 5'-CTAGGATAGACAGCAGTGGAGTAGA-3'.
Chromatin immunoprecipitation assays
Chromatin immunoprecipitation (ChIP) assays were performed on chromatin from MCF-7/MycWT cells using a ChIP Assay Kit (Upstate Biotechnology, Millipore Corp. Billerica, MA, USA), following the manufacturer's instructions. Complexes were immunoprecipitated with c-Myc antibodies (9E10, C-33; Santa Cruz Biotechnology, Santa Cruz, CA, USA) or a nonspecific PICK-1 antibody (Santa Cruz Biotechnology). The oligonucleotides used to detect the putative c-Myc binding sites in the HSPC111 promoter were as follows: HSPC111-ChIP P1: 5'-GAGTTTATTAAGCAGGGGAGTGGAG-3' ; HSPC111-CHIP P2: 5'-CCGCAGAAATGATTCCAAAACC-3', for site 1; HSPC111-CHIP P3: 5'-GTTGGTCAGGCTGGTCTTGAAC-3' ; HSPC111-ChIP P4: 5'-CGGACTTTGGAGTGGTGCTTAG-3', for site 3. For the analysis using quantitative real-time PCR, the following oligonucleotides were used: HSPC111-QPCR P1: 5'-TCCGCAGAAATGATTCCAAAA-3' ; and HSPC111-QPCR: P2 5'-AAGGGTCACTTCCTCCCCAG-3'.
Stable transfection
Full-length HSPC111 cDNA was generated by reverse transcription PCR from MCF-7 cells, and cloned into pDONR221 (Invitrogen). Constructs were recombined with the Gateway destination vector pcDNA3.1/nV5-pDEST (amino-terminal V5 fusion; Invitrogen), and then transfected into MCF-7 using Fugene-6 transfection reagent (Roche Applied Science, Indianapolis, IN, USA). Clones (MCF/HSPC) were selected and expanded in the presence of Geneticin (800 μg/ml; Invitrogen). Cells transfected with a pcDNA3.1/nV5-pDEST-LacZ vector were used as control (MCF/LacZ). MCF-7 cells inducibly expressing c-Myc wild-type (MCF/MycWT) or empty vector control cells were generated as previously described [
12].
HSPC111-specific antibody production
Amino-terminal HIS-tagged HSPC111 was expressed in Escherichia coli BL21 (DE3) pLysS. Cultures were lyzed and proteins elutions were pooled for polyclonal antibody production in rabbits. Initial bleeds were purified using a protein A column and optimized for immunoblot analysis. Antibody specificity was confirmed in MCF/HSPC-NV5 cells by comparison with V5-tagged protein detected by immunocytochemisty and immunoblotting. The antibody detected both endogenous and V5-tagged HSPC-111 protein.
Cell proliferation and S phase analysis
HSPC111-expressing cells and LacZ controls were plated at 1 × 105 (day 0) and subsequently harvested and counted up to day 5. Exponentially growing MCF-7 cells expressing HSPC111 or LacZ controls were treated with 1 μmol/l 4-hydroxytamoxifen (Sigma), 10 nmol/l ICI 182780 or vehicle (ethanol) for 48 hours. Cells were harvested and S phase was analyzed by propidium iodide staining and flow cytometry.
Small interfering RNA
Small interfering (si)RNAs (siMyc17: D-003282-17-0050; siHSPC2: D-016096-02-0050; siHSPC4: D-016096-04-0050; siCONTROL RISC-Free siRNA: D-001220-01-20; and siRNA nontargeting control 2: D-001210-02-20) were purchased from Dharmacon (Lafayette, CO, USA) and transfected using Lipofectamine 2000 (Invitrogen). For estrogen 'rescue' experiments, cells were pretreated with ICI 182780 (10 nmol/l) at 24 hours after transfection and 48 hours later were treated with vehicle (ethanol) or 17β estradiol (100 nmol/l).
Immunofluorescence
Parental MCF-7 cells or those expressing V5-tagged HSPC111 were stained with anti-HSPC111, anti-V5 (Invitrogen), anti-nucleophosmin or anti-fibrillarin (Santa Cruz Biotechnology) antibodies and DAPI (4,6-diamidino-2-phenylindole), and were visualized using confocal microscopy.
Sucrose density gradient fractionation
Nuclear extracts from exponentially growing MCF-7 cells were separated by sucrose density gradient fractionation as described previously [
20]. The gradients were analyzed through a UV monitor for continuous measurement of the absorbance at 254 nm and fractions collected. For immunoblot analysis, proteins from each fraction were precipitated with cold trichloroacetic acid at a final concentration of 10%.
Survival analyses
Datasets from two breast cancer cohorts using two different methodologies to analyze global gene expression were accessed. The first (referred to as the Uppsala cohort) is a group of 236 breast cancer patients [
21] whose tumor RNA was analyzed using Affymetrix Genechip
® (Affymetrix Inc., Santa Clara, CA, USA) HGU133A and B microarrays (NCBI GEO accession GSE3494; files were GSE3494-GPL96_series_matrix.txt.gz [HG U133A] and GSE3494-GPL97_series_matrix.txt.gz [HGU133B]). The second (from The Nederlands Kanker Instituut and designated the NKI cohort [
22]) contained 295 cases that were assessed using Rosetta NKI spotted oligonucleotide arrays [
23]. Datasets from both published series had complete data for clinicopathological variables and ER, progesterone receptor and HER2/neu status, as well as disease-specific survival. Univariate and multivariate analyses were performed as previously described [
24] to assess the association of
HSPC111 and
MYC expression with survival using Statview 5.0 Software (Abacus Systems, Berkeley, CA, USA).
P < 0.05 was considered statistically significant. The outcome variables were assessed as time to event, which was defined as the difference between the time of diagnosis and the time of death from breast cancer. Kaplan-Meier analysis was used for univariate analysis and to plot survival curves. Cox proportional hazards models were used to estimate hazard ratio (and its 95% confidence interval [CI]) associated with each risk factor and covariate and were also used for multivariate analyses.
Discussion
Although it is now well established that the mitogenic effects of estrogen play a pivotal role in the initiation and progression of breast cancer, how these effects are mediated at the molecular level remains to be fully elucidated. The transcription factor c-Myc is a prominent player in the response of breast cancer cells to estrogen, mimicking the effects of estrogen on cell cycle progression [
12] and conferring resistance to antiestrogens
in vitro [
14,
15,
43]. Thus, identification and characterization of key downstream effectors of estrogen and Myc action will not only provide a greater insight into estrogen effects on mitogenesis and survival, but could also lead to an enhanced understanding of the mechanisms governing endocrine resistance [
8].
In a search for estrogen-target genes that are regulated secondarily to estrogen's induction of c-Myc, we identified a novel gene of unknown function, namely
HSPC111, which was among the most highly regulated estrogen and Myc target genes in our model [
12] (Musgrove EA, Sergio CM, Butt AJ, Sutherland RL; unpublished data).
HSPC111 is rapidly (within 3 hours) upregulated in response to treatment with estrogen (about threefold) and induction of Myc (about fourfold). However, the response to estrogen is abrogated in the presence of Myc siRNA, providing strong evidence that estrogen stimulates HSPC111 expression via its well documented upregulation of Myc. This conclusion is further supported by our demonstration of functional E-boxes in the
HSPC111 promoter, and Myc-responsive promoter activity, identifying
HSPC111 as a direct transcriptional target of Myc. Although gene expression profiling has recently identified
HSPC111 as a target of estrogen [
44] and Myc [
45], this is the first report demonstrating that estrogen's effects on HSPC111 are dependent upon a direct transcriptional activation by Myc.
Although HSPC111 is a previously uncharacterized protein, it is known to reside in the nucleolus [
46]. In an attempt to elucidate a cellular role for HSPC111, we further investigated its subcellular localization. The nucleolus is the center of ribosomal biosynthesis and assembly [
29]. HSPC111 did not colocalize with either NPM/B23 or fibrillarin, both of which are known to play a role in ribosomal biosynthesis. However, sucrose density fractionation demonstrated that HSPC111 is part of a RNA-dependent complex sedimenting in the 40 to 80S region, which also contains preribosomal ribonucleoprotein particles [
29]. In addition to driving cell division, Myc plays a crucial role in controlling cell growth and protein synthesis [
47]. Thus, the acute transcriptional regulation of HSPC111 by Myc may represent part of a coordinated stimulation of ribosome biogenesis [
47], occurring concurrently with its stimulation of cell proliferation. However, whether HSPC111 has a role in the ribosomal biosynthesis pathway is not clear from these studies and requires further investigation.
Recent studies have emphasized an important link between nucleolar function, in particular ribosomal biogenesis, and cell cycle control, and several genes coordinately regulate both processes. For example, disruption of the nucleolar PeBoW complex, consisting of Pes1, Bop1 and WDR12, blocks both rRNA processing and cell cycle progression [
48,
49]. Given the proliferative role of Myc in our model and our data suggesting HSPC111 interacts with RNA in the nucleolus, we questioned whether HSPC111 might play a role in Myc's effects on cell cycle progression. However, modulation of HSPC111 expression had no effect on cell proliferation end-points. We detected no effect of constitutive HSPC111 expression on proliferation, and although it is possible that the level of over-expression achieved was not sufficient for a detectable increase in proliferation rate, HSPC111 expression was not required for cell cycle progression, and neither was its downregulation required for antiestrogen inhibition of proliferation. These data are supported by Schlosser and coworkers [
45], who identified
HSPC111 as a Myc target gene in the human B-cell line P493-6 under conditions in which Myc induces cell growth but not cell proliferation [
50]. Furthermore, they suggested that, even if HSPC111 does play a role in rRNA synthesis, there is either an element of functional redundancy in its role or it is not rate-limiting for cell cycle progression. Indeed, although adequate cell growth is essential for proliferation, it is not sufficient, requiring additional proliferative signals for cell cycle progression to proceed [
51]. These data emphasize the complexity of the Myc phenotype, even within the relative restrictions of our model system, and support the concept that the coordinated regulation of multiple effector genes is required to recapitulate Myc functions [
52].
To address further a potential role of HSPC111 in cancer, we initially identified a strong positive correlation between
MYC mRNA and both HSPC111 mRNA and protein in breast cancer cell lines, raising the possibility that HSPC111 expression might be a useful surrogate marker of Myc over-expression in breast cancer. However, the relationship at the mRNA level was less robust in primary breast cancer (r
2 = 0.19 versus 0.60 for primary cancer versus cell lines). An extension of this analysis to published datasets from a number of other cancers identified that elevated expression of HSPC111 was a feature of several cancers including those of the breast, prostate, ovary, testis, liver, colon, and pancreas [
31,
41], but this was not always associated with
MYC over-expression. Thus, HSPC111 over-expression appears to be a common feature of many cancers but its relationship to aberrant Myc function, which is only contributed in part by
MYC mRNA levels, remains to be elucidated. More importantly, HSPC111 over-expression was a strong predictor of an adverse outcome in two cohorts of breast cancer patients on univariate analysis and remained significant in a multivariate model in the Uppsala cohort. These effects were independent of
MYC mRNA over-expression, which is in support of our conclusions from other cancers. Whether HSPC111 over-expression is functionally associated with disease progression remains an open question. The data presented here failed to support a role in cell proliferation or endocrine sensitivity, but other aspects of the biology of tumor progression require further investigation. It is well established that aberrant cell growth (increased/dysregulated ribosome biogenesis and protein synthesis) are common features of cancer. Because our preliminary data point to nucleolar localization of HSPC111 in association with ribonucleoproteins, it may be either functionally associated with these processes or a marker of aberrant cell growth regulation in general. In any event, further investigation of the normal physiological role of HSPC111 in nucleolar function and the functional consequences of overexpression on cellular growth control are warranted.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
AJB, RLS, and EAM conceived the study, participated in its design, coordination and interpretation, and drafted the manuscript. AJB additionally generated cell lines over-expressing HSPC111. CMS performed analyses of HSPC111 expression, sucrose density gradient fractionation, and analyses of cells following HSPC111 over-expression or knock-down. CKI performed analyses of HSPC111 transcriptional regulation, participated in sucrose density fractionation, measured gene expression in breast cancers, and helped to draft the manuscript. LRA performed immunolocalization experiments, participated in HSPC111 knock-down experiments, and measured gene expression in breast cancer cell lines. CMM analyzed HSPC111 expression in breast cancers. AJR generated and characterized the HSPC111 antibody. MN and TP participated in the design and interpretation of sucrose density fractionation. AVB examined HSPC111 association with disease outcome and helped to draft the manuscript. All authors read and approved the final version.