| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Cancer Therapy: Preclinical |
Impairs Growth and Survival in Neuroblastoma CellsAuthors' Affiliations: 1 Division of Clinical Chemistry and Biochemistry and 2 Department of Oncology, University Children's Hospital Zurich, Zurich, Switzerland; and 3 Department of Pediatric Oncology and Hematology, University Hospital of Essen, Essen, Germany
Requests for reprints: Alexandre Arcaro, Division of Clinical Chemistry and Biochemistry, University Children's Hospital Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland. Phone: 41-1-266-7640; Fax: 41-44-266-7169; E-mail: Alexandre.Arcaro{at}kispi.uzh.ch.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Experimental Design: The expression pattern and functions of class IA PI3K isoforms were investigated in tumor samples and cell lines. Effects on cell survival and downstream signaling were analyzed following down-regulation of p110
or p110
in SH-SY5Y and LA-N-1 cells by means of RNA interference.
Results: Overexpression of the catalytic p110 and regulatory p85 isoforms was detected in a panel of primary neuroblastoma samples and cell lines, compared with normal adrenal gland tissue. Although down-regulation of either p110 or p110 led to impaired cell growth, reduced expression of p110 also had a selective effect on the survival of SH-SY5Y cells. Decreased levels of p110 were found to induce apoptosis and lead to lower expression levels of antiapoptotic Bcl-2 family proteins. SH-SY5Y cells with decreased p110 levels also displayed reduced activation of ribosomal protein S6 kinase in response to stimulation with epidermal growth factor and insulin-like growth factor-I.
Conclusions: Together, our data reveal a novel function of p110 in neuroblastoma growth and survival.
A better understanding of the biology of neuroblastoma will potentially lead to the identification of novel therapeutic targets, which in turn could facilitate the development of new drugs for neuroblastoma. A promising field of investigation is to target receptor tyrosine kinase (RTK) signaling to some of their downstream mediators such as phosphoinositide 3-kinase (PI3K), Akt, and the mammalian target of rapamycin (mTOR). Polypeptide growth factors have indeed been shown to play a key role in neuroblastoma biology. Insulin-like growth factor (IGF) signaling has been extensively studied in the context of neuroblastoma proliferation, survival, and motility (4–7). Several potential antitumor approaches involving the IGF-I system have been reported in neuroblastoma (8, 9). Moreover, inhibition of platelet-derived growth factor receptor and c-Kit signaling with imatinib mesylate was reported to impair growth in neuroblastoma cell lines (10). Neurotrophins such as brain-derived neurotrophic factor also play an important role in neuroblastoma chemoresistance by binding to the Trk receptor family (11).
In view of the fact that neuroblastoma express a variety of different RTKs, it remains unclear whether targeting individual receptors will provide a successful therapeutic strategy. An alternative approach would be to identify downstream signaling molecules essential for transmitting the proliferative and survival message of several different RTKs. PI3K represents such a molecule, in view of its crucial role in controlling cell proliferation, survival, and motility/metastasis downstream of many different RTKs (12–14). PI3Ks are an enzyme family comprising eight catalytic isoforms in humans, with different substrate specificities, regulatory mechanisms, and tissue distribution (12, 14). The importance of PI3K signaling in human cancer was first shown by the observation that mutations in the tumor suppressor gene PTEN occur frequently in human tumors. PTEN is a phosphatase that antagonizes the action of PI3K by dephosphorylating the D-3 position of polyphosphoinositides (14, 15). Moreover, recent reports have described activating mutations in the PIK3CA gene encoding the catalytic p110 isoform of PI3K in a variety of human cancers, including breast, colon, and ovarian cancer, as well as medulloblastoma (16). In neuroblastoma, brain-derived neurotrophic factor was shown to protect the tumor cells from chemotherapy-induced apoptosis via the PI3K pathway (17). IGF-I signaling via PI3K was shown to be required for neuroblastoma differentiation, cytoskeletal rearrangements (18), as well as angiogenesis and vascular endothelial growth factor expression (19). PI3K signaling is also activated by epidermal growth factor (EGF) in neuroblastoma cells and contributes to cell proliferation by this growth factor (20). Thus, targeting the PI3K/Akt/mTOR/S6 kinase (S6K) pathway may represent an attractive novel approach to develop therapies for neuroblastoma.
In the present report, we have evaluated the expression of PI3K isoforms in primary human neuroblastoma samples and cell lines. Moreover, we have investigated whether targeting distinct PI3K isoforms could impair growth and survival of neuroblastoma cell lines. Our results show for the first time that the class IA PI3K p110 is overexpressed in a subset of neuroblastoma samples and plays a crucial role in the growth and survival of neuroblastoma cells.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
, p110β, p110, poly(ADP)ribose polymerase, PTEN, Akt/PKB, Erk1/2, Santa Cruz Biotechnology; S6 protein, 4EBP1 and phosphospecific antibodies for Akt/PKB (Ser473; Thr308), Erk1/2 (Thr202/Tyr204), S6 protein (Ser235/236; Ser240/244), 4E-BP1 (Thr37/46), Cell Signaling Technology; β-actin, Sigma-Aldrich; p110 (clone U3A), generous gift from Dr. A. Klippel (Atugen AG, Berlin, Germany). Analysis of proteins involved in apoptosis was done using the Pro-Survival and the Pro-Apoptosis Bcl-2 Family Antibody Sampler Kit (Cell Signaling). LY294002, rapamycin, EGF, platelet-derived growth factor, and IGF-I were obtained from Calbiochem. Determination of PI3K-related gene expression in human primary neuroblastoma samples by cDNA microarray analysis. The expression of PI3K-related genes was determined in tumors from 68 neuroblastoma patients by Affymetrix U95A array analysis. The patient cohort and data normalization procedures have been described elsewhere (21). Correlation of PI3K-related gene expression with molecular and clinical variables was determined using the stats package included in R2.2.4 Visualization of gene expression was accomplished using Spotfire 8.1.
Primary neuroblastoma samples. Ethical approval to use residual tissue was obtained. RNAlater-preserved tumor tissue was available from the Swiss Pediatric Oncology Group tumor bank from neuroblastoma patients diagnosed between January 2003 and December 2005 at the University Children's Hospital of Zurich (n = 14), the Children's Hospital Luzern (n = 3), the University Children's Hospital Bern (n = 1), and the University Children's Hospital Basel (n = 1). The selection of the tumors for the study was based on the availability of a sufficient quantity of tumor tissue to perform RNA isolation. All diagnoses were confirmed by histologic assessment of the tumor specimen obtained at surgery. An overview of the tumor characteristics is given in Supplementary Fig. S1.
Protein extraction from tumor samples. Tumor tissue was disrupted with a sterile disposable tissue grinder (Sage Products, Inc.). Protein extracts were obtained using the PARIS Kit (Ambion) according to the manufacturer's instructions.
Isolation of RNA from tumor samples and reverse transcription-PCR. Tumor tissue was disrupted as described above and homogenized in guanidinium isothiocyanate-containing buffer. Total RNA was isolated using the RNeasy kit (Qiagen, Inc.) according to the manufacturer's protocol. Total RNA (3 µg) from each tumor sample was converted into cDNA using the SuperScript First-Strand Synthesis System for PCR according to manufacturer's instructions (Invitrogen Life Technologies). mRNA expression of four target genes and 18S (internal control gene) was measured in tumor samples and cell lines by TaqMan Assay-on-Demand Gene Expression products (Applied Biosystems). Normal human adrenal gland tissue was used as a reference. The following primers were used (gene assay ID): PIK3R1-Hs00236128_m1; PIK3CA-Hs00180679_m1; PIK3CB-Hs00178872_m1; PIK3CD-Hs00192399_m1; eukaryotic 18S rRNA-Hs99999901_s1. Three replicates were run for each sample in a 96-well format plate. Gene expression assays consisted of a FAM dye-labeled TaqMan MGB probe and two PCR primers. The thermal cycling conditions consisted of an initial denaturation step at 95°C for 10 min and a 50-cycle countdown at 95°C for 15 s and 60°C for 1 min. Each sample was normalized on the basis of its 18S rRNA content. Relative mRNA expression levels were calculated using the comparative threshold cycle (CT) method (22).
Cell culture. Human neuroblastoma cell lines were kindly provided by Dr. Brodeur (Children's Hospital of Philadelphia, Philadelphia, PA). The cells were grown in RPMI (Life Technologies/Invitrogen) supplemented with 10% (v/v) FCS and penicillin/streptomycin/L-glutamine, and passaged every 3 to 5 days by trypsinization.
Stable transfectants. SH-SY5Y and LA-N-1 cells were stably transfected with the murine ecotropic receptor (EcoR) using Lipofectamine (Invitrogen) according to the manufacturer's protocol. Seventy-two hours posttransfection, cells were diluted in medium containing G418 (0.8 mg/mL). In parallel, retroviral plasmid constructs encoding short hairpin RNA (shRNA) specifically targeting p110 or p110 (23) were transiently transfected into packing cells. The supernatant was diluted with one part RPMI containing G418 (final concentration: 0.8 mg/mL), as well as polybrene (final concentration, 8 µg/mL). The diluted supernatant was then added to cells stably expressing EcoR. Twenty-four hours postinfection, SH-SY5Y and LA-N-1 cells were split in medium containing puromycin (0.5 µg/mL). Single colonies were picked and expanded in selective medium. Protein down-regulation was confirmed by Western blot analysis. All experiments were done with two clones from each transfection.
Transient transfection. Cells were transiently transfected using the Amaxa Nucleofector Device according to the optimized protocol provided for SH-SY5Y cells. The following constructs were used: pRS, pRS-PIK3CA, pRS-PIK3CD (23), pcDNA3 (Invitrogen), and pcDNA-S6K AK (9).
Cell proliferation. Neuroblastoma cells were seeded in 96-well plates at a density of 7,500 per well and grown for 72 h in RPMI containing low (1%) or high (10%) serum. Alternatively, cells were treated with growth factors or inhibitors as indicated. Cell proliferation was analyzed by the CellTiter 96 AQueous Cell Proliferation Assay (Promega) according to the manufacturer's instructions.
Apoptosis. Neuroblastoma cells were seeded in 96-well plates at a density of 20,000 per well. Basal caspase-3/7 activity was measured after 24 h using the Caspase-Glo 3/7 Assay (Promega) according to the manufacturer's instructions.
Cell cycle distribution. Cell cycle distribution was analyzed by means of propidium iodide staining and fluorescence-activated cell sorting analysis. Cells were seeded in a six-well plate at a density of 6.5 x 105 and incubated in RPMI containing low (1%) or high (10%) serum for 24 h. Cells were collected using 0.5% trypsin and resuspended in cold PBS. One tenth volume of 10x propidium iodide solution [500 µg/mL propidium iodide, 10 mg/mL sodium citrate and 1% (v/v) Triton X-100] was added and cells were acquired in a flow cytometer within half an hour.
Growth factor stimulations. Cells were grown to confluency in a six-well plate and starved overnight in RPMI containing 0.5% FCS. Cells were maintained in serum-free RPMI for 1 h in the presence or absence of inhibitors as indicated and were then stimulated with the indicated growth factors for 10 min. Cellular lysates were prepared as described below.
Western blotting. Cellular lysates were prepared as described (24) and normalized using a bicinchoninic acid protein assay (Pierce). Cell lysates were separated by SDS-PAGE, transferred to a hydrophobic polyvinylidene difluoride membrane (Hybond-P; Amersham Biosciences), and immunoblotted with the indicated antibodies before chemiluminescent detection (ECL Western blotting detection reagents; Amersham Biosciences).
Statistical analysis. Spearman's rank correlation and the exact Wilcoxon's rank-sum test were used for the analysis of PI3K gene expression in primary neuroblastoma samples. For experiments on cell lines, the statistical significance of differences between groups was assessed with ANOVA using Bonferroni's multiple comparison test; P < 0.01 was indicated with a double asterisk.
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
) and PIK3CD (encoding the catalytic subunit p110) genes was found to be increased >2-fold in 10 of 19 neuroblastoma samples when compared with normal adrenal gland tissue (Fig. 1A, left). In contrast, mRNA expression of the catalytic p110 and p110β isoforms was not increased (Fig. 1A, left). In this panel of tumor samples, expression of p85 and p110 was significantly correlated (Spearman's rank correlation, P < 0.0001), indicating that the p85/p110 heterodimer is overexpressed in neuroblastoma. Expression of p85 and p110 was significantly higher in children under the age of 1 year (median values, 2.48 for p85 and 5.17 for p110) than in patients older than 1 year old (median values, 0.64 for p85 and 0.95 for p110; exact Wilcoxon's rank-sum test, P = 0.0121 for p85; P = 0.0025 for p110). Furthermore, the expression of p85 and p110 was found to be significantly lower in neuroblastoma samples with MYCN amplification (median values, 2.42 for p85 and 4.63 for p110) than without amplification (median values, 0.81 for p85 and 0.75 for p110; exact Wilcoxon's rank-sum test, P = 0.0339 for p85; P = 0.0339 for p110). In contrast, no correlation was found between p85 and p110 expression and tumor stage, 1p status, or progression. Thus, p85 and p110 expression was found to be increased in neuroblastoma samples from children under the age of 1 year and with no MYCN amplification.
|
Reanalysis of cDNA microarray data of an independent panel of 68 primary neuroblastoma samples also revealed striking variations in the expression levels of the class IA PI3K regulatory subunit p85 and the catalytic subunit p110. Expression of the p85 and p110 subunits was significantly higher in samples from children under the age of 1 year compared with patients older than 1 year old (Fig. 1B), confirming the results of the quantitative reverse transcription-PCR analysis (Fig. 1A).
We next investigated the expression of PI3K isoforms in a set of eight representative neuroblastoma cell lines to validate the findings obtained in tumor samples. In this panel of cell lines, p85 was overexpressed in five of eight samples, whereas the catalytic isoforms p110, p110β, and p110 showed increased expression in four of eight, six of eight, and one of eight cell lines, respectively (Fig. 1C, left). In line with the findings in patient samples, cell lines harboring MYC amplification showed decreased levels of PIK3CD mRNA when compared with adrenal tissue (Fig. 1C, left). As with the patient samples, mRNA levels were largely found to be predictive of protein levels (Fig. 1C, right).
The expression of PI3K isoforms was next investigated at the protein level in a panel of eight neuroblastoma cell lines. All neuroblastoma cell lines included in this study expressed the p85, p110, and p110β isoforms, whereas the expression of p110 was more variable, being highest in SH-SY5Y cells (Fig. 2A
). Together, these data show that the expression levels of PI3K isoforms are altered in neuroblastoma samples and cell lines at the mRNA and/or protein level, and that the expression of the catalytic p110 isoform is increased in neuroblastoma.
|
EGF and IGF-I stimulated growth of SH-SY5Y (2- and 2.5-fold) and LA-N-1 (2- and 2.2-fold) cells, whereas platelet-derived growth factor was less potent at promoting a growth response (Fig. 2C). The pharmacologic PI3K inhibitor LY294002 reduced basal growth of SH-SY5Y (66.5% inhibition) and LA-N-1 (42.8% inhibition) cells. Treatment with the PI3K inhibitor also reduced EGF- and IGF-I–stimulated growth in SH-SY5Y (44.1% and 27.6% inhibition, respectively) and LA-N-1 cells (Fig. 2C). However, LY294002 as a single agent did not significantly affect the extent of growth factor–stimulated cell growth in either SH-SY5Y or LA-N-1 cells (Fig. 2C). Recently, a similar observation was reported in neuroblastoma cell lines, where the combination of both LY294002 and the mTOR inhibitor rapamycin was required to inhibit IGF-I–induced proliferation (9). Thus, various growth factors, including EGF and IGF-I, stimulate growth of neuroblastoma cells via the PI3K/Akt pathway.
Impact of shRNA-mediated down-regulation of p110 and p110 on neuroblastoma cell proliferation and apoptosis. To gain insight into the individual functions of class IA PI3K isoforms in neuroblastoma cell responses, SH-SY5Y and LA-N-1 cells were stably transduced with short hairpin RNA (shRNA) constructs targeting p110 (hereinafter p110low) or p110 (hereinafter p110low). Specific down-regulation of target gene expression was verified by Western blot analysis (Fig. 3A
). The growth of SH-SY5Y p110low cells was significantly reduced (53.1% and 44.1% inhibition) when the cells were cultivated in medium containing low (1%) serum, but not high (10%), serum (Fig. 3B). This effect was less pronounced in LA-N-1 cells (Fig. 3B), correlating with the differential expression of p110 in these cell lines. Although an effect of p110 down-regulation was also observed in SH-SY5Y cells (low serum, no inhibition or 21.3% inhibition; high serum, no inhibition or 21.2% inhibition), the effect was stronger in LA-N-1 cells (low serum, 22.1% and 66.4% inhibition; high serum, 11.7% and 43.4%, inhibition).
|
low or p110low cells. A significant (2.2-fold) increase in caspase-3/7 activity was observed in SH-SY5Y p110low cells (Fig. 3C). In contrast, a comparable response was not observed in SH-SY5Y p110low cells (Fig. 3C). Analysis of cell cycle distribution revealed a decrease in the S phase in SH-SY5Y p110low and p110low cells, both in low serum (61.8% decrease in both cases) and high serum (71.4% and 59.2% decrease). The increase in cell death under low serum conditions was higher in SH-SY5Y p110low cells (319.4% increase) than in SH-SY5Y p110low cells (64.5% increase), confirming the results of the caspase assay (Fig. 3C). Together, these results show that the p110 isoform plays a major role in SH-SY5Y cell growth and survival under suboptimal culture conditions. In contrast, a comparable role was not apparent for p110 in these cells, although this isoform appeared to play a more important function in LA-N-1 cells, which express only very low levels of p110.
In view of the increased apoptosis observed in SH-SY5Y p110low cells, we analyzed the expression of proapoptotic and antiapoptotic Bcl-2 family proteins, which are key regulators of apoptosis. The levels of Bcl-2 and Bcl-XL were elevated in SH-SY5Y cells grown in high serum, compared with low serum conditions (Figs. 3D and 5B). When compared with control cells, SH-SY5Y p110low cells displayed reduced expression of the antiapoptotic proteins Bcl-2 and Bcl-XL (Figs. 3D and 5B). In contrast, the levels of Bax were unaffected (Fig. 3D). Thus, p110 seems to contribute to the control of the expression of antiapoptotic Bcl-2 family proteins, which may affect its function in neuroblastoma cell survival.
|
and p110 on neuroblastoma cell responses to growth factors. We next investigated the involvement of class IA PI3K isoforms in polypeptide growth factor signaling in neuroblastoma cells. The ability of EGF and IGF-I to stimulate the growth of SH-SY5Y cells was significantly reduced in both SH-SY5Y p110low and p110low cells (EGF, 6.4% and 40.3%, 23.4%, and 16.2%, respectively; IGF-I, 27.5% and 41.5%, 42.1% and 49.1%, respectively; Fig. 4A
). In contrast, the ability of EGF and IGF-I to stimulate the growth of LA-N-1 p110low and p110low cells was not significantly impaired (Fig. 4A).
|
and p110 on the activation of early signaling mediators of RTKs was then investigated in SH-SY5Y and LA-N-1 cells. Activation of Akt by EGF and IGF-I was unaltered in SH-SY5Y and LA-N-1 cells (Fig. 4B; Supplementary Fig. S2). In contrast, activation of the mTOR/S6K pathway by EGF and IGF-I was impaired in both SH-SY5Y p110low and p110low cells, as assessed by phosphorylation of the S6 protein (Fig. 4B). Although the effect of p110 down-regulation on signal transduction was expected, the observation that decreased p110 expression leads to a comparable effect on its own was surprising. Activation of Erk1/2 by EGF and IGF-I was unaffected by the PI3K shRNAs (data not shown). Thus, p110 plays an important role in activation of the mTOR/S6K pathway in SH-SY5Y cells, which correlates with its involvement in neuroblastoma cell growth. However, activation of Akt by growth factors is insensitive to down-regulation of a single PI3K isoform in these cells. This is not unexpected, as previous studies have reported that PI3K isoforms from other classes contribute to Akt activation (24). In both LA-N-1 p110low and p110low cells, basal activation of Akt and S6K was reduced (Supplementary Fig. S2). However, in LA-N-1 cells, phosphorylation of the S6 protein was hardly affected by down-regulation of p110 (Supplementary Fig. S2). Although decreased p110 expression led to a stronger inhibition of downstream signaling, no apparent effect on the proliferative response was observed, suggesting that other pathways may compensate for reduced signal transduction in this cell line (Fig. 4A; Supplementary Fig. S2).
Constitutive activation of the S6K pathway abrogates the effect of p110 shRNA on SH-SY5Y cell growth. Our previous data had shown a marked effect of p110 down-regulation by shRNA on activation of the mTOR/S6K pathway in SH-SY5Y cells, contributing to cell growth. To confirm this model, SH-SY5Y p110low cells were transiently transfected with an activated mutant of S6K1. Activated S6K promoted growth of SH-SY5Y cells and was able to partially rescue the growth of SH-SY5Y p110low cells (Fig. 5A
). However, the induction of apoptosis induced by the p110 shRNA was only marginally impaired by S6K transfection, as assessed by cleavage of poly(ADP-ribose) polymerase (Fig. 5B). Moreover, transient transfection of activated S6K did not restore the levels of Bcl-2 and Bcl-XL in SH-SY5Y p110low cells (Fig. 5B). Together, these results show that the PI3K p110 isoform controls neuroblastoma cell growth via the S6K pathway. In addition, it also seems to play a role in controlling the levels of Bcl-2 family proteins, which seems not to involve S6K or Akt.
Rapamycin mimics the effects of p110 down-regulation. The data presented above highlighted the importance of signaling via the mTOR/S6K pathway in the regulation of neuroblastoma cell survival and the ability to respond to growth factors. To confirm these findings, SH-SY5Y and LA-N-1 cells were treated with rapamycin, a mTOR inhibitor, and cellular responses were investigated. As expected, phosphorylation of the S6 protein was abrogated in the presence of rapamycin, both in low (1%) and high (10%) serum conditions (Fig. 6A
). Treatment of SH-SY5Y and LA-N-1 cells with rapamycin led to a dose-dependent inhibition of basal cell proliferation [SH-SY5Y, 29.2% (20 ng/mL) and 56.6% (100 ng/mL) inhibition; LA-N-1: 37.3% (20 ng/mL) and 61.8% (100 ng/mL) inhibition; Fig. 6B]. Cotreatment of cells with rapamycin attenuated the proliferative response to EGF [SH-SY5Y, 34.7% (20 ng/mL) and 55.8% (100 ng/mL) inhibition; LA-N-1, 32.5% (20 ng/mL) and 44.4% (100 ng/mL) inhibition] and IGF-I [SH-SY5Y, 32.5% (20 ng/mL) and 53% (100 ng/mL) inhibition; LA-N-1, 27.7% (20 ng/mL) and 42.1% (100 ng/mL) inhibition; Fig. 6B]. Furthermore, phosphorylation of the S6 protein was found to be absent upon stimulation with EGF or IGF-I in cells pretreated with rapamycin (Fig. 6C). These findings emphasize the importance of intact mTOR/S6K signaling in neuroblastoma cell survival.
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
plays a crucial role in neuroblastoma cell growth and survival and that the expression of p110 is increased at the mRNA and protein level in a subset of primary neuroblastoma samples and cell lines. In addition, our data reveal a selective expression of p85/p110 in neuroblastoma samples from children under the age of 1 year. Our results also reveal a negative correlation between PIK3CD gene expression and MYCN amplification. Expression of PIK3CD was recently shown to be significantly lower in neuroblastoma samples with loss of heterozygosity at 1p36 (29). A recent study also described a correlation between decreased PIK3CD expression, 1p deletion, and poor clinical outcome (30). Loss of heterozygosity at 1p36 was associated with amplification of the MYCN oncogene (31). Reduced expression of PIK3CD in neuroblastoma samples may thus correlate with poor prognosis, due to its association with age at diagnosis (>1 year), LOH at 1p36, and/or MYCN amplification, because these events are prognostic markers of poor outcome in neuroblastoma (1, 2, 31, 32). N-Myc was shown to be able to replace IGF/PI3K signaling in medulloblastoma formation (33). Moreover, c-Myc expression had a negative effect on the activation of PI3K/Akt signaling (34). Thus, it is conceivable that MYCN-amplified neuroblastoma display decreased PIK3CD expression, either because of (a) loss of heterozygosity at 1p36 (29), (b) a direct negative effect of MYCN amplification on PIK3CD expression, or (c) because they are less dependent on p85/p110 signaling. Collectively, these data indicate that the class IA p110 isoform may play a role in the development of neuroblastoma in very young patients (<1 year old) or in a subset of tumors that do not harbor loss of heterozygosity at 1p36 and/or MYCN amplification.
The expression of p110 was previously shown to be predominantly restricted to leukocytes in normal tissues (35). However, different human cancer cell lines were also shown to express this PI3K isoform, indicating that aberrant expression of p110 in tumors may contribute to the malignant properties of the cancer cells (24, 36). In support of this notion, the ability of p110 to induce transformation of chicken fibroblasts was recently shown (37). In breast cancer cell lines, a selective function of p110 in cell migration was also described (36). In acute myeloid leukemia, recent reports have shown that p110 plays a major role in activation of Akt by RTKs (FLT3), cell proliferation, and chemoresistance (38, 39). We show here that the class IA PI3K isoforms p110 and p110 do not have overlapping functions in neuroblastoma cell responses. Indeed, in SH-SY5Y cells that display increased PIK3CA and PIK3CD expression, p110 seemed to play a major role in controlling neuroblastoma cell growth and survival under limiting growth conditions. In contrast, in LA-N-1 cells with only very low p110 expression, down-regulation of p110 by shRNA impaired cell growth and Akt activation. The effect of p110 down-regulation by shRNA on SH-SY5Y growth correlated with an impairment of the activation of the mTOR/S6K pathway. Surprisingly, Akt activation by EGF and IGF-I was unaffected by p110 down-regulation, under conditions where the mTOR/S6K pathway was inhibited. Thus, activation of Akt and mTOR/S6K have different sensitivities to class IA PI3K down-regulation. It is conceivable that other PI3K isoforms may compensate for p110 or p110 in growth factor–stimulated Akt activation (24). Transfection of an activated mutant of S6K was sufficient to partially rescue the growth defect of SH-SY5Y p110low cells under low serum conditions. However, p110 seemed to have an additional function in maintaining the levels of antiapoptotic Bcl-2 family proteins in neuroblastoma cells, which did not involve S6K or Akt. The observation that the expression levels of antiapoptotic Bcl-2 family proteins were higher in SH-SY5Y under high serum conditions than in low serum indicates that growth factors contribute to the regulation of the levels of Bcl-2 family proteins in these cells. Previous reports have documented a role for IGF-I, which is present in the FCS, in maintaining Bcl-2 (40) and Bcl-XL (41) expression levels. In view of the data presented here showing a role for p110 in IGF-I signaling in neuroblastoma cells, it can be postulated that IGF-IR/p110 contributes to regulation of the expression of Bcl-2 family proteins. Because Akt and S6K were not involved in this pathway, the simplest explanation for these observations is that mTOR-mediated phosphorylation of 4E-BP1 controls translational activation of the expression of Bcl-2 and/or Bcl-XL in neuroblastoma cells, which is supported by previous findings in other systems (42). Down-regulation of p110 by shRNA had a less pronounced effect on the growth of SH-SY5Y cells under low serum conditions, although activation of S6K was impaired. In view of the observation that p110 down-regulation did not induce apoptosis in SH-SY5Y cells, in contrast to p110, it can be speculated that Bcl-2 family proteins are selectively controlled by the p110 isoform.
Together, our results show that p110 contributes to neuroblastoma cell growth and survival by regulating the activation of the mTOR/S6K pathway and the expression levels of antiapoptotic Bcl-2 family proteins.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
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: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).
M.A. Grotzer is a member of the Swiss Pediatric Oncology Group.
Received 3/30/07; revised 8/29/07; accepted 10/ 8/07.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
and vascular endothelial growth factor expression induced by insulin-like growth factor-I in neuroblastoma cells. Cancer Res 2005;65:4775–81., a novel phosphoinositide 3-kinase in leukocytes. Proc Natl Acad Sci U S A 1997;94:4330–5.. Cancer Res 2003;63:1667–75.
, and - isoforms of class I phosphoinositide 3-kinase. Proc Natl Acad Sci U S A 2006;103:1289–94. isoform in phosphoinositide 3-kinase activation and cell proliferation in acute myeloid leukemia. Blood 2005;106:1063–6. isoform of PI 3-kinase inhibits AML cell proliferation and survival and increases the cytotoxic effects of VP16. Oncogene 2006;25:6648–59.[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  E. A. Papakonstanti, O. Zwaenepoel, A. Bilancio, E. Burns, G. E. Nock, B. Houseman, K. Shokat, A. J. Ridley, and B. Vanhaesebroeck Distinct roles of class IA PI3K isoforms in primary and immortalised macrophages J. Cell Sci., December 15, 2008; 121(24): 4124 - 4133. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
