The phosphatidylinositol 3-kinase (PI3K) pathway is commonly activated in breast cancers due to frequent mutations in PIK3CA, loss of expression of PTEN or over-expression of receptor tyrosine kinases. PI3K pathway activation leads to stimulation of the key growth and proliferation regulatory kinase mammalian target of rapamycin (mTOR), which can be inhibited by rapamycin analogues and by kinase inhibitors; the effectiveness of these drugs in breast cancer treatment is currently being tested in clinical trials. To identify the molecular determinants of response to inhibitors that target mTOR via different mechanisms in breast cancer cells, we investigated the effects of pharmacological inhibition of mTOR using the allosteric mTORC1 inhibitor everolimus and the active-site mTORC1/mTORC2 kinase inhibitor PP242 on a panel of 31 breast cancer cell lines. We demonstrate here that breast cancer cells harbouring PIK3CA mutations are selectively sensitive to mTOR allosteric and kinase inhibitors. However, cells with PTEN loss of function are not sensitive to these drugs, suggesting that the functional consequences of these two mechanisms of activation of the mTOR pathway are quite distinct. In addition, a subset of HER2-amplified cell lines showed increased sensitivity to PP242, but not to everolimus, irrespective of the PIK3CA/PTEN status. These selective sensitivities were confirmed in more physiologically relevant three-dimensional cell culture models. Our findings provide a rationale to guide selection of breast cancer patients who may benefit from mTOR inhibitor therapy and highlight the importance of accurately assessing the expression of PTEN protein and not just its mutational status.
Breast cancers frequently harbour molecular/genomic aberrations that increase activation of the phosphatidylinositol 3-kinase (PI3K)–AKT–mammalian target of rapamycin (mTOR) signalling pathway including the (over-) expression of receptor tyrosine kinases (for example HER2, IGF1R), loss of PTEN function or activating PIK3CA mutations (Yuan and Cantley, 2008; Engelman, 2009). Given the pivotal role of this pathway in cell growth, proliferation and survival (Engelman et al., 2006), PI3K, AKT and mTOR have emerged as promising therapeutic targets for subgroups of breast cancer patients (Engelman, 2009; Liu et al., 2009). mTOR is a highly conserved Ser/Thr kinase that integrates nutrient and growth factor signals through two distinct protein complexes: (1) mTORC1, which is inhibited by rapamycin, responsive to growth factors, amino acids, energy and oxygen levels, and regulates translation initiation and protein synthesis through S6K1 and 4E-BP1 and (2) mTORC2, which is rapamycin insensitive and modulates growth factor signalling by phosphorylating various kinases including AKT at serine 473 (Guertin and Sabatini, 2007).
Rapamycin and rapamycin analogues (‘rapalogues’) are potent allosteric inhibitors of mTORC1, which act by forming a complex with the intracellular receptor FKBP12 (Brown et al., 1994; Sabatini et al., 1994). The rapalogues everolimus (RAD001) and temsirolimus (CCI-779) have received Food and Drug Administration (FDA) approval for the treatment of patients with advanced renal cell carcinoma (Atkins et al., 2009; Kwitkowski et al., 2010). In breast cancer, phase II clinical trials revealed that pre-treated patients with locally advanced or metastatic disease had a modest response when receiving rapalogues as single agent (12% and 9.2% complete or partial clinical response to everolimus (Ellard et al., 2009) and temsirolimus (Chan et al., 2005), respectively). Thus, for the introduction of mTOR inhibitors into clinical practice, it is germane to identify markers to guide selection of patients who would benefit from these drugs and to define mechanisms of de novo resistance (that is patients who would not derive any benefit from these targeted agents).
Second-generation mTOR small molecule inhibitors, which unlike rapalogues inhibit both mTORC1 and mTORC2 by directly targeting the ATP-binding site of mTOR, have recently entered clinical trials (Dancey, 2010). Although the clinical benefit of these agents is yet to be determined, in vitro and in vivo pre-clinical models using the mTOR kinase inhibitor PP242 have demonstrated that this compound inhibits proliferation more completely than the mTORC1 inhibitor rapamycin (Feldman et al., 2009; Janes et al., 2010). This emphasises the complexity of the PI3K–AKT–mTOR signalling pathway and the distinct contributions of mTORC1 and mTORC2 to cell proliferation and survival.
We set out to define the molecular determinants of response to inhibitors that target mTOR via different mechanisms in breast cancer cells. To address this, we investigated the effects of pharmacologic inhibition of mTOR using the rapalogue everolimus and the kinase inhibitor PP242 in a panel of 31 breast cancer cell lines, and showed that activating PIK3CA mutations but not PTEN loss of function determine response of breast cancer cells to allosteric mTORC1 and active-site mTORC1/2 inhibitors. This suggests that the functional consequences in the tumour of these two mechanisms of activation of the mTOR pathway are quite distinct and are differentially affected by drugs in clinical usage. In addition, a subset of cell lines harbouring HER2 amplification showed increased sensitivity to PP242 but not everolimus irrespective of the PIK3CA mutation status. The predictive nature of the PIK3CA mutations and HER2 amplification for response to everolimus and/or PP242 was independent of the cell culture environment, as this association was observed in breast cancer cell lines grown in conventional two-dimensional (2D) monolayers as well as in more physiologically relevant three-dimensional (3D) culture models.
PIK3CA mutations determine response to everolimus and PP242 in breast cancer cells
The effects of pharmacologic mTOR inhibition were tested in a panel of 31 breast cancer cell lines grown in 2D cultures. Cells were treated for 72 h with serial dilutions of the allosteric mTORC1 inhibitor everolimus or the mTOR kinase inhibitor PP242. Cell viability was determined by CellTiter-Blue assay, and a wide range of SF50s (that is surviving fraction of 50% relative to untreated cells; Turner et al. (2010)) from 1.7 nM to 6.6 μM was observed (Figure 1). The SF50s of everolimus and/or PP242-sensitive cell lines were in the nanomolar range (Figure 1); however, for many of the breast cancer cell lines studied, the SF50 was not reached when treated with everolimus (SF50s >10 μM; Figure 1a).
To identify genomic alterations linked to PI3K–AKT–mTOR pathway dependency and mTOR inhibitor sensitivity (that is SF50<1 μM), the mutation status of the 31 breast cancer cell lines was correlated with their mTOR inhibitor response. Given the contradictory reports on the pattern of PIK3CA and PTEN mutations in these cell lines in previous publications (Hollestelle et al., 2007; Hu et al., 2009) and the COSMIC database (Forbes et al., 2008), and the possibility of cross-contamination and misidentification of cell lines (ASN-0002, 2010), we re-sequenced in all our breast cancer cell lines the entire PTEN transcript and exons 2, 4–11 and 19–21 of the PIK3CA transcript (Supplementary Table 1; Figure 2a; Table 1), covering the high-frequency mutations in the p85, C2, helical and kinase domains of the PIK3CA gene (Samuels et al., 2004; Gymnopoulos et al., 2007). In addition, as PTEN function is regulated also by post-translational modifications (Wang and Jiang, 2008; Huse et al., 2009; Poliseno et al., 2010), PTEN protein levels were assessed in these cells by western blotting (Figure 2b). Sequencing analysis revealed 10 breast cancer cell lines harbouring activating PIK3CA mutations (BT20, BT474, BT483, HCC1500, HCC1954, HCC202, MCF7, MDA-MB-361, T47D, MDA-MB-453), of which one cell line had a concurrent PTEN mutation (MDA-MB-453) (Table 1). Additional 11 cell lines showed defective PTEN; in six cell lines a mutation in the PTEN transcript was identified (BT549, CAMA1, HCC1569, HCC70, MDA-MB-415, ZR75-1), whereas five cell lines did not express PTEN protein and/or PTEN transcript (HCC1395, HCC1937, HCC38, MDA-MB-436, MDA-MB-468) (Table 1; Figure 2).
Correlation of the PIK3CA and PTEN status revealed that in this setting, the response of breast cancer cells to the allosteric mTORC1 inhibitor everolimus and to the active-site inhibitor PP242 was determined by the presence of activating PIK3CA mutations (Figure 1). Only one out of nine cell lines with mutated PIK3CA (and wild-type PTEN), MDA-MB-361, was resistant to everolimus treatment. Noteworthy was the observation that PP242 treatment induced a more efficient response in a subset of HER2-amplified cell lines (that is HCC1419, SKBR3, HCC1569, MDA-MB-453; Figure 1b) and in the PIK3R1-mutated cell line HS578T, irrespective of the PIK3CA mutation status. In contrast, 10/12 PTEN null cell lines were resistant to everolimus or PP242 treatment, with exception of CAMA1 and ZR75-1 (Figure 1). The cell line with concurrent PTEN and PIK3CA mutations (MDA-MB-453) was resistant to everolimus. Cells with wild-type PIK3CA and functional PTEN were resistant to everolimus or PP242 treatment. Taken together, our findings provide strong circumstantial evidence that activating mutations of PIK3CA but not PTEN loss of function may predict that breast cancer cells will be responsive to allosteric mTORC1 and active-site mTORC1/2 inhibitors.
Everolimus and PP242 elicit a G1 cell-cycle arrest in PIK3CA-mutated cell lines
To test whether everolimus and PP242 exert anti-proliferative and/or apoptotic effects, a panel of cell lines was treated with everolimus, PP242, and etoposide or staurosporine as positive controls in 2D cell cultures. Neither the rapalogue everolimus nor the mTOR kinase inhibitor PP242 induced apoptosis in mTOR inhibitor sensitive cell lines as assessed by poly (ADP-ribose) polymerase cleavage using western blotting (Figure 3a) and by Caspase-3/7 activation using a fluorescent enzymatic assay (Figure 3b). Instead, both everolimus and PP242 induced a G1 cell-cycle arrest in PIK3CA-mutated mTOR inhibitor sensitive cell lines but not in resistant PTEN null or PIK3CA/PTEN wild-type cells (Figure 3c). These results imply that, in this setting, the allosteric mTORC1 inhibitor everolimus and the mTOR kinase inhibitor PP242 elicit cytostatic responses in the breast cancer cell lines studied.
Effects of everolimus and PP242 on signal transduction pathways in breast cancer cells
To understand the response of signal transduction pathways to mTOR inhibition, we examined AKT, mTORC1 downstream effectors, and MAPK pathway activation in a panel of cell lines. As expected, we observed efficient reduction of the mTORC2 substrate S473-AKT phospho-levels upon treatment with the active-site mTORC1/2 inhibitor PP242 (Feldman et al., 2009; Janes et al., 2010) but not with everolimus (Figure 4; Supplementary Figure 1). In addition, continued activation of 4E-BP1 was detected in everolimus but not PP242-treated cells.
Furthermore, we observed a difference in activated ERK1/2 levels between mTOR inhibitor sensitive and resistant breast cancer cell lines (Figures 4a–d; Supplementary Figures 1 and 2). Quantification of the phospho- and total ERK1/2 protein bands (on the same western blot using the Odyssey Infrared Imaging System) revealed that, as a group, both mTOR inhibitor resistant PTEN null and PIK3CA/PTEN wild-type cell lines displayed sustained levels of ERK1/2 activation when treated with everolimus or increased levels of ERK1/2 activation when treated with PP242, as compared with untreated cells. This phenomenon was generally not observed in PIK3CA mutant cell lines, in the outliers (that is PTEN null mTOR inhibitor sensitive CAMA1 and ZR75-1 cells) and in PTEN mutant HER2-amplified PP242 responsive cells (Figure 4; Supplementary Figures 1–3). The relative expression levels of phospho-ERK1/2 (that is phospho/total ERK1/2 ratios between treated and untreated cells) in resistant breast cancer cell lines were significantly higher than those in sensitive cell lines upon everolimus (P=0.0103) or PP242 treatment (P=0.0262; two-tailed heteroscedastic t-test).
These data suggest the possibility that MAPK signalling might contribute to the resistance of breast cancer cells to mTOR inhibitors. We therefore tested the effect of combined mTOR and MEK inhibition on cell viability in 2D and 3D cultures. The MEK inhibitors U0126 (Favata et al., 1998) or AZD6244 (Yeh et al., 2007) were used at concentrations at which PIK3CA/PTEN wild-type and PTEN null cells displayed a decrease in phospho-ERK1/2 levels (that is 1 μM and 10 μM; Supplementary Figure 4a). Simultaneous inhibition of mTORC1 and MEK did not exert a major effect on the viability of PIK3CA/PTEN wild-type or PTEN null cells beyond the effect of the MEK inhibitor alone. The addition of U0126 or AZD6244 to PP242, however, resulted in a substantial decrease in the surviving cell fraction (Supplementary Figures 5 and 6). As expected, combined treatment with the mTORC1/2 inhibitor PP242 and U0126 or AZD6244 resulted in a marked reduction in the levels of phospho-ERK and phospho-AKT, whereas combination of the mTORC1 inhibitor everolimus and U0126 or AZD6244 led to an effective reduction of phospho-ERK levels but not phospho-AKT levels in the breast cancer cells studied (Supplementary Figure 4b). These data suggest that a subset of breast cancer cell lines that lack PTEN function or are PIK3CA/PTEN wild-type may be resistant to mTOR kinase inhibitors in part due to activation of the MAPK pathway.
PIK3CA mutation and HER2 amplification status correlates with response to mTOR inhibitors independent of cell culture environment
Given that 3D cell cultures have been suggested to resemble their in vivo counterparts more closely than conventional 2D monolayer models (Pampaloni et al., 2007; Yamada and Cukierman, 2007), and that drug response of breast cancer cells has been reported to vary according to culture conditions (that is 2D vs 3D) (Serebriiskii et al., 2008; Weigelt and Bissell, 2008; Pickl and Ries, 2009; Weigelt et al., 2010), we assessed whether the extracellular matrix (ECM) and tissue architecture would have an impact on the mTOR inhibitor response of breast cancer cells. A selection of cell lines was plated on top of 3D ECM cultures (that is Matrigel) for 4 days until tumour-like structures were formed, treated for 72 h with serial dilutions of everolimus or PP242 and cell viability was determined using CellTiter-Blue. The response of PIK3CA/PTEN wild-type, PIK3CA-mutated and PTEN null cell lines was remarkably similar between 2D and 3D culture conditions (Figure 5). In a way akin to our findings in 2D, cell lines with functional PI3K/PTEN or PTEN loss were resistant to everolimus and PP242 treatment, whereas PIK3CA mutant cells were sensitive when grown in a 3D culture environment (Figure 5; Supplementary Figure 7). This was also observed for the HER2-amplified PP242-sensitive PIK3CA/PTEN wild-type cell lines, and the outliers CAMA1, a PTEN null mTOR inhibitor sensitive, and MDA-MB-361, a PIK3CA mutant everolimus-resistant cell line (Supplementary Figure 8). Taken together, our results suggest that PIK3CA mutations determine response of breast cancer cells to everolimus and/or PP242 independent of the culture environment.
In this study, we assessed the molecular determinants of response to inhibitors that target mTOR via different mechanisms in a large panel of breast cancer cell lines. We have shown that breast cancer cells harbouring activating PIK3CA mutations but not lacking PTEN function were selectively sensitive to the allosteric mTORC1 inhibitor everolimus and to the active-site mTORC1/2 inhibitor PP242, confirming the observations that mutations in multiple components of the PI3K pathway are not necessarily equivalent in their biological impact (Stemke-Hale et al., 2008; Vasudevan et al., 2009; Dan et al., 2010). Our results corroborate and expand on the previous observations that PTEN loss of function and PIK3CA mutations have different functional effects on PI3K pathway activation in human breast cancers and in breast cancer cell lines (Stemke-Hale et al., 2008; Dan et al., 2010). Here, we demonstrate that the distinct functional effects of lack of PTEN function versus activating PIK3CA mutations seem to have crucial implications for the use of therapies targeting the PI3K–AKT–mTOR pathway in breast cancer.
It has recently been reported that the efficacy of 25 PI3K pathway inhibitors did not correlate with either gain-of-function mutations of PIK3CA or PTEN loss in a panel of 39 human cancer cell lines derived from tumours from various anatomical sites (Dan et al., 2010). On the other hand, PIK3CA mutations were recently reported to predict sensitivity to everolimus in glioblastoma, breast, ovarian, prostate, endometrial and colorectal cancer cells (Di Nicolantonio et al., 2010), to predict response to temsirolimus alone or in combination in patients with advanced cervical, endometrial, ovarian and breast cancer in a phase I clinical trial (Janku et al., 2011), and the PI3K/mTOR dual inhibitor NVP-BEZ235 has been shown to selectively induce cell death in breast cancer cells harbouring PIK3CA mutations (Serra et al., 2008; Brachmann et al., 2009) but not in cells lacking PTEN function (Brachmann et al., 2009). Taken together, these findings and the results presented herein suggest that the predictive nature of PIK3CA mutations for response to everolimus and PP242 may be tissue specific and of particular relevance to breast cancer.
A feedback loop depending on an S6K-PI3K-RAS pathway has been shown to act as a potential mechanism of resistance to mTOR inhibition in metastatic cancer patients and in in vitro and in vivo models, which leads to MAPK pathway activation and cell survival (Carracedo et al., 2008; Brachmann et al., 2009). We detected increased ERK1/2 activation in breast cancer cell lines that lack PTEN function or are wild-type for PIK3CA/PTEN when treated with the active-site inhibitor PP242, and combined inhibition of mTORC1/2 and MEK resulted in decreased viability of these cells. As a group, mTOR inhibitor sensitive PIK3CA-mutated breast cancer cell lines did not display increased MAPK pathway activation upon treatment with everolimus or PP242. There is burgeoning evidence to demonstrate that breast cancer cells harbouring PIK3CA mutations are physiologically dependent on this kinase for the activation of RPS6 and survival, whereas PIK3CA wild-type cell lines do not necessarily require PI3K for the activation of the mTOR pathway (Crowder et al., 2009; Dan et al., 2010). These observations suggest a greater dependency of PIK3CA mutant cell lines to the canonical PI3K–AKT–mTOR pathway, which is consistent with our observations. Although our findings should be interpreted as hypothesis generating and the underlying mechanism for our observations is yet to be fully established, one could posit that these distinct effects on the MAPK pathway according to PIK3CA and PTEN status may be due to (1) the distinct levels of PI3K pathway activation in cells with different patterns of mutations in this pathway (Stemke-Hale et al., 2008; Dan et al., 2010, 2) AKT-independent signal operant within PIK3CA mutant cell lines (Vasudevan et al., 2009, 3) the increased dependency of PTEN null cancers on the PI3K isoform p110β (Torbett et al., 2008; Wee et al., 2008, 4) the complex cross-talk between PTEN and other pathways, including the MAPK pathway (Zhang and Yu, 2010) or (5) importantly, the redundancy and complex feedback regulation in the PI3K–AKT–mTOR pathway (Efeyan and Sabatini, 2010; Zhang and Yu, 2010). Further mechanistic studies to delineate the genotype-dependent impact of mTOR inhibitors on the signalling networks in breast cancer cells are warranted.
It should be noted that the prevalence of PIK3CA mutations in human breast cancers (that is 18–32.5%; Levine et al., 2005; Saal et al., 2005; Stemke-Hale et al., 2008; Kalinsky et al., 2009) is somewhat higher than the proportion of pre-treated metastatic breast patients who showed clinical benefit with everolimus monotherapy (12%) (Ellard et al., 2009). This may be reconciled by the fact that (1) patients with PIK3CA mutant cancers have a better outcome (Kalinsky et al., 2009) and may be under-represented in the group of patients with metastatic disease, (2) ∼30% of breast cancers show PTEN protein loss (Hennessy et al., 2005; Perez-Tenorio et al., 2007; Stemke-Hale et al., 2008), which in contrast to PTEN mutations is not mutually exclusive with PIK3CA mutations (Perez-Tenorio et al., 2007; Stemke-Hale et al., 2008) and that (3) MAPK pathway activation due to other genomic/molecular aberrations may not be uncommon (Adelaide et al., 2007; Roidl et al., 2010), which would result in mTOR inhibitor resistance in a subset of PIK3CA-mutated cancers according to our findings. In addition, it has been suggested that the strength of PI3K–AKT pathway activation may differ in vitro compared with human PIK3CA-mutated breast cancer (Loi et al., 2010). Response rates to everolimus could potentially be further compounded by the reported anti-angiogenic effects of rapalogues (Guba et al., 2002; Del Bufalo et al., 2006); however, this is unlikely given that only 6.7% of pre-treated metastatic breast cancer patients showed response to the bona fide VEGF inhibitor bevacizumab as a single agent (Cobleigh et al., 2003).
In the phase II clinical trial reported by Ellard et al. (2009), PTEN status was evaluated by immunohistochemistry for a possible correlation with everolimus response in patients with recurrent/metastatic breast cancer; however, no statistical association was found. This may be due to the limited number of patients assessed (Ellard et al., 2009) and to the lack of validated methodology and scoring system for immunohistochemical evaluation of loss of functional PTEN. In fact, different antibodies used for immunohistochemical analysis of PTEN provided divergent results, which often did not correlate with PTEN mutations and/or loss of function (Pallares et al., 2005). Thus, methodologies and scoring systems for immunohistochemical analysis of PTEN need to be improved and standardised before its use in the context of clinical trials. Furthermore, recent massively parallel sequencing studies have provided direct evidence to demonstrate that the repertoire of gene mutations in a primary breast tumour and its metastasis differ (Ding et al., 2010). One could speculate that the PTEN status of a primary tumour may therefore not necessarily be identical to that of the metastasis. Our results suggest PIK3CA mutation status rather than PTEN loss of function would constitute a positive predictive marker for response to mTOR inhibitors in breast cancer patients. Given that ∼80% of PIK3CA mutations are located in exons 9 and 20 (Cosmic database; Bachman et al., 2004; Samuels et al., 2004; Forbes et al., 2008) and that sequencing of PIK3CA mutation hot-spots can be reliably performed in archival material (Kalinsky et al., 2009), analysis of the PIK3CA gene status in relation to clinical response to everolimus is warranted.
There is evidence to suggest that HER2-amplified breast cancers may be ‘addicted’ to (that is physiologically dependent on) continued PI3K–AKT pathway activation (She et al., 2008). We observed here that a subset of HER2-amplified breast cancer cells showed increased sensitivity to the mTOR kinase inhibitor PP242, but not to the rapalogue everolimus, independent of the PIK3CA mutation status. PP242, unlike everolimus, also inhibits mTORC2 and its main substrate AKT, which leads to a more effective decrease in PI3K–AKT signalling. Taken together, our findings suggest that at least a subgroup of HER2-amplified breast cancers may harbour ‘PI3K–AKT pathway addiction’ and are consistent with the observation that the PI3K/mTOR dual inhibitor NVP-BEZ235 induces cell death in breast cancer cells with HER2 gene amplifications (Brachmann et al., 2009). The mechanisms leading to PI3K–AKT independence in a subgroup of HER2-amplified breast cancers merits further investigation.
Both cell–ECM interactions and 3D morphology of breast cancer cells have been reported to modulate signalling pathway activation and drug response (Weaver et al., 2002; Serebriiskii et al., 2008; Weigelt and Bissell, 2008). Interestingly, breast cancer cell lines grown in 3D ECM culture models or as floating aggregates without the supply of exogenous ECM (that is poly-HEMA cultures) showed differential response to the HER2 targeting antibodies trastuzumab and/or pertuzumab compared with cells grown in conventional 2D models (Pickl and Ries, 2009; Weigelt et al., 2010). In this study, we assessed the response of a panel of breast cancer cells to mTOR inhibition when grown in 3D ECM models. Similarly to in 2D cultures, cell lines harbouring PIK3CA mutations were selectively sensitive to everolimus and PP242 in 3D growth conditions. These data suggest that the ECM signals upstream of PIK3CA gene mutations tested in this study have a negligible effect on the dependence of breast cancer cells on the PI3K–AKT–mTOR signalling pathway. Importantly, our results suggest that 2D models may be adequate to investigate the genotypic–phenotypic correlations between mutations in components of this signalling pathway and response to mTOR inhibitors.
In conclusion, in this study we demonstrate that breast cancer cell lines with PIK3CA mutations but not lack of PTEN function are selectively sensitive to allosteric mTORC1 and active-site mTORC1/2 inhibitors independent of the cell culture environment. In addition, we showed that a subset of HER2-amplified breast cancer cell lines have increased sensitivity to the mTOR kinase inhibitor PP242 but not to the rapalogue everolimus. Our results provide circumstantial evidence to suggest that PIK3CA mutations rather than PTEN loss of protein expression by immunohistochemical analysis should be investigated as a predictive marker to guide the selection of breast cancer patients who would benefit from mTOR inhibitor therapy in future clinical trials.
Materials and methods
Cell culture and drug treatment
Human breast cancer cell lines were obtained as NCI-ICBP45 kit procured through American Type Culture Collection (ATCC) (ATCC Breast Cancer Cell Panel, Manassas, VA, USA). Cell lines were authenticated by ATCC using short tandem repeat DNA profiling, and each cell culture was examined by light microscopy and compared with images published by ATCC and the Integrative Cancer Biology Program (ICBP; http://icbp.lbl.gov/breastcancer/celllines.php) to verify identity. For a detailed list of the 31 breast cancer cell lines and growth conditions used, see Supplementary Table 2.
For drug treatment in 2D monolayers, cells were inoculated into 96-well microtiter plates in 100 μl at plating densities ranging from 2500 to 25 000 cells/well depending on the doubling time and cell size of individual cell lines. After 24 h, cells were treated with serial dilutions (100 pM to 10 μM) of everolimus (RAD001; LC Laboratories, Woburn, MA, USA), PP242 (gift from K Shokat, UCSF; Sigma-Aldrich, Dorset, UK), U0126 (Cell Signaling Technology, New England Biolabs, Hitchin, UK) or AZD6244 (Axon Medchem, Groningen, The Netherlands). For combination mTOR and MEK inhibition, 10 μM everolimus or 1 μM PP242 were added to serial dilutions (100 pM to 10 μM) of U0126 or AZD6244, respectively. Cell viability was assessed after 72 h of treatment by incubation with CellTiter-Blue (Promega, Southampton, UK) for 1.5 h; fluorescence was read on an EnVision 2102 plate-reader (Perkin-Elmer, Waltham, MA, USA). The drug dose required for survival of 50% of cells relative to untreated cells (surviving fraction 50, SF50 (Turner et al., 2010)) was determined using Graphpad Prism version 5.0c. For drug treatment in 3D cell cultures, cells were seeded in 96-well plates on top of growth factor reduced phenol red-free Engelbreth–Holm–Swarm tumour matrix (Matrigel, BD Biosciences, Bedford, MA, USA) in 100 μl of their respective medium with 5% Matrigel at densities ranging from 3000 to 5000 single cells/well depending on the doubling time of individual cell lines (Lee et al., 2007; Weigelt et al., 2010). Serial dilutions of everolimus or PP242 (100 pM to 10 μM) were added 4 days after plating, and cell viability was assessed after 72 h of treatment using CellTiter-Blue. For combined mTOR and MEK inhibition, cells were treated with serial dilutions (100 pM to 10 μM) of U0126 or AZD6244 alone, or in combination with 10 μM everolimus or 1 μM PP242, respectively, for 120 h given the decreased level of proliferation of cells grown on top of Matrigel.
Western blot analysis and protein quantification
Cells were rinsed with phosphate-buffered saline and lysed in NuPAGE LDS Sample Buffer (Invitrogen, Paisley, UK). Samples were resolved on 4–12% gradient NuPAGE Novex Bis-Tris gels (Invitrogen) and proteins were transferred onto a nitrocellulose membrane (Whatman, Dassel, Germany). Membranes were blocked for 1 h in 5% BSA at room temperature and probed overnight at 4 °C with primary antibody in 1% BSA. Antibodies against poly (ADP-ribose) polymerase (#9532), β-actin (#4967), PTEN (#9188), AKT (#9272), phospho-AKT (Ser473; #9271), S6 Ribosomal Protein (#2217), phospho-S6 Ribosomal Protein (#2211), 4E-BP1 (#9452), phospho-4E-BP1 (Thr37/46; #9459), ERK1/2 (#9102) and phospho-ERK1/2 (Thr202/Tyr204; #9101) were obtained at Cell Signaling Technology, anti-α-tubulin (clone B-5-1-2) was obtained at Sigma-Aldrich (St Louis, MO, USA). After incubation with horseradish peroxidase-conjugated secondary antibody in 5% skimmed milk, proteins were detected using chemiluminescence (GE Healthcare, Little Chalfont, UK).
For quantification of phospho-ERK1/2 and ERK1/2 protein bands, samples were resolved on 4–12% gradient NuPAGE Novex Bis-Tris gels and proteins were transferred onto an immobilon polyvinylidene difluoride fluorescence membrane (Millipore, Billerica, MA, USA). Membranes were dried and blocked for 1 h in Odyssey Blocking Buffer (LI-COR, Cambridge, UK) at room temperature and probed overnight at 4 °C simultaneously with anti-ERK1/2 (#9107) and anti-phospho-ERK1/2 (Thr202/Tyr204; #9101; Cell Signaling Technology) in Odyssey Blocking Buffer. After incubation with IRDye 800CW Goat anti-Mouse IgG (926-32210) and IRDye 680LT Goat anti-Rabbit IgG (926-68021) secondary antibodies in Odyssey Blocking Buffer, membranes were scanned using the Odyssey Infrared Imaging System and analysed and quantified using the 500 Odyssey Software.
Apoptosis and cell-cycle analysis
For cell-cycle analysis, cells were treated after 24 h of seeding in 10 cm dishes with everolimus or PP242 at indicated concentrations or vehicle (dimethyl sulfoxide). After 48 h, cells were harvested, fixed in cold 70% ethanol, stained using propodium iodide (Sigma-Aldrich, Dorset, UK), and DNA content measured on a LSRII flow cytometer (BD Biosciences). Cell-cycle distribution was assessed using FlowJo 9.0.1 software. For analysis of apoptosis induction, cells were treated after 24 h of seeding with everolimus or PP242 at indicated concentrations or vehicle (dimethyl sulfoxide) for 48 h or Staurosporin (positive control; Sigma-Aldrich) for 3 h and analysed by western blotting for poly (ADP-ribose) polymerase cleavage. Alternatively, cells were treated after 24 h of plating in 96-well microtiter plates with everolimus, PP242, vehicle (dimethyl sulfoxide) or Etoposide (positive control; Sigma-Aldrich) at indicated concentrations. After 48 h, cell viability was determined using CellTiter-Blue by incubation with cells for 1.5 h, and apoptosis induction using the Apo-ONE Caspase-3/7 assay (Promega) by incubation of cells for 5 h. Fluorescence was read on an EnVision 2102 plate-reader.
PIK3CA and PTEN mutation analysis
Mutation analysis was performed of the complete coding sequence of PTEN (NM_000314), and of exons 2, 4–11 and 19–21 of PIK3CA (NM_006218). Cells were rinsed with phosphate-buffered saline and total RNA was extracted using RNA-Bee (Ams Biotechnology, Milton, UK) following the manufacturer's instructions. cDNA synthesis of 2 μg total RNA was performed using the SuperScript VILO cDNA Synthesis Kit (Invitrogen). Three and four PCRs were performed for amplification of the PTEN and PIK3CA transcripts, respectively (see Supplementary Table 1 for primer sequences) using the Platinum Taq DNA Polymerase High Fidelity (Invitrogen) according to the manufacturer's instructions. PCR specificity was assessed by agarose gel electrophoresis. For sequence analysis, amplified products were sequenced in both directions with the BigDye Terminator v3.1 using an ABI3730 DNA Analyser (Applied Biosystems, Foster City, CA, USA) (Supplementary Table 1). Sequences were analysed using Mutation Surveyor software (Softgenetics, State College, PA, USA).
Adelaide J, Finetti P, Bekhouche I, Repellini L, Geneix J, Sircoulomb F et al. (2007). Integrated profiling of basal and luminal breast cancers. Cancer Res 67: 11565–11575.
American Type Culture Collection Standards Development Organization Workgroup ASN-000 (2010). Cell line misidentification: the beginning of the end. Nat Rev Cancer 10: 441–448.
Atkins MB, Yasothan U, Kirkpatrick P . (2009). Everolimus. Nat Rev Drug Discov 8: 535–536.
Bachman KE, Argani P, Samuels Y, Silliman N, Ptak J, Szabo S et al. (2004). The PIK3CA gene is mutated with high frequency in human breast cancers. Cancer Biol Ther 3: 772–775.
Brachmann SM, Hofmann I, Schnell C, Fritsch C, Wee S, Lane H et al. (2009). Specific apoptosis induction by the dual PI3K/mTor inhibitor NVP-BEZ235 in HER2 amplified and PIK3CA mutant breast cancer cells. Proc Natl Acad Sci USA 106: 22299–22304.
Brown EJ, Albers MW, Shin TB, Ichikawa K, Keith CT, Lane WS et al. (1994). A mammalian protein targeted by G1-arresting rapamycin-receptor complex. Nature 369: 756–758.
Carracedo A, Ma L, Teruya-Feldstein J, Rojo F, Salmena L, Alimonti A et al. (2008). Inhibition of mTORC1 leads to MAPK pathway activation through a PI3K-dependent feedback loop in human cancer. J Clin Invest 118: 3065–3074.
Chan S, Scheulen ME, Johnston S, Mross K, Cardoso F, Dittrich C et al. (2005). Phase II study of temsirolimus (CCI-779), a novel inhibitor of mTOR, in heavily pretreated patients with locally advanced or metastatic breast cancer. J Clin Oncol 23: 5314–5322.
Cobleigh MA, Langmuir VK, Sledge GW, Miller KD, Haney L, Novotny WF et al. (2003). A phase I/II dose-escalation trial of bevacizumab in previously treated metastatic breast cancer. Semin Oncol 30: 117–124.
Crowder RJ, Phommaly C, Tao Y, Hoog J, Luo J, Perou CM et al. (2009). PIK3CA and PIK3CB inhibition produce synthetic lethality when combined with estrogen deprivation in estrogen receptor-positive breast cancer. Cancer Res 69: 3955–3962.
Dan S, Okamura M, Seki M, Yamazaki K, Sugita H, Okui M et al. (2010). Correlating phosphatidylinositol 3-kinase inhibitor efficacy with signaling pathway status: in silico and biological evaluations. Cancer Res 70: 4982–4994.
Dancey J . (2010). mTOR signaling and drug development in cancer. Nat Rev Clin Oncol 7: 209–219.
Del Bufalo D, Ciuffreda L, Trisciuoglio D, Desideri M, Cognetti F, Zupi G et al. (2006). Antiangiogenic potential of the Mammalian target of rapamycin inhibitor temsirolimus. Cancer Res 66: 5549–5554.
Di Nicolantonio F, Arena S, Tabernero J, Grosso S, Molinari F, Macarulla T et al. (2010). Deregulation of the PI3K and KRAS signaling pathways in human cancer cells determines their response to everolimus. J Clin Invest 120: 2858–2866.
Ding L, Ellis MJ, Li S, Larson DE, Chen K, Wallis JW et al. (2010). Genome remodelling in a basal-like breast cancer metastasis and xenograft. Nature 464: 999–1005.
Efeyan A, Sabatini DM . (2010). mTOR and cancer: many loops in one pathway. Curr Opin Cell Biol 22: 169–176.
Ellard SL, Clemons M, Gelmon KA, Norris B, Kennecke H, Chia S et al. (2009). Randomized phase II study comparing two schedules of everolimus in patients with recurrent/metastatic breast cancer: NCIC Clinical Trials Group IND.163. J Clin Oncol 27: 4536–4541.
Engelman JA . (2009). Targeting PI3K signalling in cancer: opportunities, challenges and limitations. Nat Rev Cancer 9: 550–562.
Engelman JA, Luo J, Cantley LC . (2006). The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet 7: 606–619.
Favata MF, Horiuchi KY, Manos EJ, Daulerio AJ, Stradley DA, Feeser WS et al. (1998). Identification of a novel inhibitor of mitogen-activated protein kinase kinase. J Biol Chem 273: 18623–18632.
Feldman ME, Apsel B, Uotila A, Loewith R, Knight ZA, Ruggero D et al. (2009). Active-site inhibitors of mTOR target rapamycin-resistant outputs of mTORC1 and mTORC2. PLoS Biol 7: e38.
Forbes SA, Bhamra G, Bamford S, Dawson E, Kok C, Clements J et al. (2008). The catalogue of somatic mutations in cancer (COSMIC). Curr Protoc Hum Genet Suppl 57: 10.11.1–10.11.26.
Guba M, von Breitenbuch P, Steinbauer M, Koehl G, Flegel S, Hornung M et al. (2002). Rapamycin inhibits primary and metastatic tumor growth by antiangiogenesis: involvement of vascular endothelial growth factor. Nat Med 8: 128–135.
Guertin DA, Sabatini DM . (2007). Defining the role of mTOR in cancer. Cancer Cell 12: 9–22.
Gymnopoulos M, Elsliger MA, Vogt PK . (2007). Rare cancer-specific mutations in PIK3CA show gain of function. Proc Natl Acad Sci USA 104: 5569–5574.
Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB . (2005). Exploiting the PI3K/AKT pathway for cancer drug discovery. Nat Rev Drug Discov 4: 988–1004.
Hollestelle A, Elstrodt F, Nagel JH, Kallemeijn WW, Schutte M . (2007). Phosphatidylinositol-3-OH kinase or RAS pathway mutations in human breast cancer cell lines. Mol Cancer Res 5: 195–201.
Hu X, Stern HM, Ge L, O'Brien C, Haydu L, Honchell CD et al. (2009). Genetic alterations and oncogenic pathways associated with breast cancer subtypes. Mol Cancer Res 7: 511–522.
Huse JT, Brennan C, Hambardzumyan D, Wee B, Pena J, Rouhanifard SH et al. (2009). The PTEN-regulating microRNA miR-26a is amplified in high-grade glioma and facilitates gliomagenesis in vivo. Genes Dev 23: 1327–1337.
Janes MR, Limon JJ, So L, Chen J, Lim RJ, Chavez MA et al. (2010). Effective and selective targeting of leukemia cells using a TORC1/2 kinase inhibitor. Nat Med 16: 205–213.
Janku F, Tsimberidou AM, Garrido-Laguna I, Wang X, Luthra R, Hong DS et al. (2011). PIK3CA mutations in patients with advanced cancers treated with PI3K/AKT/mTOR axis inhibitor. Mol Cancer Ther (e-pub ahead of print; doi:10.1158/1535-7163.MCT-10-0994).
Kalinsky K, Jacks LM, Heguy A, Patil S, Drobnjak M, Bhanot UK et al. (2009). PIK3CA mutation associates with improved outcome in breast cancer. Clin Cancer Res 15: 5049–5059.
Kwitkowski VE, Prowell TM, Ibrahim A, Farrell AT, Justice R, Mitchell SS et al. (2010). FDA approval summary: temsirolimus as treatment for advanced renal cell carcinoma. Oncologist 15: 428–435.
Lee GY, Kenny PA, Lee EH, Bissell MJ . (2007). Three-dimensional culture models of normal and malignant breast epithelial cells. Nat Methods 4: 359–365.
Levine DA, Bogomolniy F, Yee CJ, Lash A, Barakat RR, Borgen PI et al. (2005). Frequent mutation of the PIK3CA gene in ovarian and breast cancers. Clin Cancer Res 11: 2875–2878.
Liu P, Cheng H, Roberts TM, Zhao JJ . (2009). Targeting the phosphoinositide 3-kinase pathway in cancer. Nat Rev Drug Discov 8: 627–644.
Loi S, Haibe-Kains B, Majjaj S, Lallemand F, Durbecq V, Larsimont D et al. (2010). PIK3CA mutations associated with gene signature of low mTORC1 signaling and better outcomes in estrogen receptor-positive breast cancer. Proc Natl Acad Sci USA 107: 10208–10213.
Pallares J, Bussaglia E, Martinez-Guitarte JL, Dolcet X, Llobet D, Rue M et al. (2005). Immunohistochemical analysis of PTEN in endometrial carcinoma: a tissue microarray study with a comparison of four commercial antibodies in correlation with molecular abnormalities. Mod Pathol 18: 719–727.
Pampaloni F, Reynaud EG, Stelzer EH . (2007). The third dimension bridges the gap between cell culture and live tissue. Nat Rev Mol Cell Biol 8: 839–845.
Perez-Tenorio G, Alkhori L, Olsson B, Waltersson MA, Nordenskjold B, Rutqvist LE et al. (2007). PIK3CA mutations and PTEN loss correlate with similar prognostic factors and are not mutually exclusive in breast cancer. Clin Cancer Res 13: 3577–3584.
Pickl M, Ries CH . (2009). Comparison of 3D and 2D tumor models reveals enhanced HER2 activation in 3D associated with an increased response to trastuzumab. Oncogene 28: 461–468.
Poliseno L, Salmena L, Zhang J, Carver B, Haveman WJ, Pandolfi PP . (2010). A coding-independent function of gene and pseudogene mRNAs regulates tumour biology. Nature 465: 1033–1038.
Roidl A, Foo P, Wong W, Mann C, Bechtold S, Berger HJ et al. (2010). The FGFR4 Y367C mutant is a dominant oncogene in MDA-MB453 breast cancer cells. Oncogene 29: 1543–1552.
Saal LH, Holm K, Maurer M, Memeo L, Su T, Wang X et al. (2005). PIK3CA mutations correlate with hormone receptors, node metastasis, and ERBB2, and are mutually exclusive with PTEN loss in human breast carcinoma. Cancer Res 65: 2554–2559.
Sabatini DM, Erdjument-Bromage H, Lui M, Tempst P, Snyder SH . (1994). RAFT1: a mammalian protein that binds to FKBP12 in a rapamycin-dependent fashion and is homologous to yeast TORs. Cell 78: 35–43.
Samuels Y, Wang Z, Bardelli A, Silliman N, Ptak J, Szabo S et al. (2004). High frequency of mutations of the PIK3CA gene in human cancers. Science 304: 554.
Serebriiskii I, Castello-Cros R, Lamb A, Golemis EA, Cukierman E . (2008). Fibroblast-derived 3D matrix differentially regulates the growth and drug-responsiveness of human cancer cells. Matrix Biol 27: 573–585.
Serra V, Markman B, Scaltriti M, Eichhorn PJ, Valero V, Guzman M et al. (2008). NVP-BEZ235, a dual PI3K/mTOR inhibitor, prevents PI3K signaling and inhibits the growth of cancer cells with activating PI3K mutations. Cancer Res 68: 8022–8030.
She QB, Chandarlapaty S, Ye Q, Lobo J, Haskell KM, Leander KR et al. (2008). Breast tumor cells with PI3K mutation or HER2 amplification are selectively addicted to Akt signaling. PLoS One 3: e3065.
Stemke-Hale K, Gonzalez-Angulo AM, Lluch A, Neve RM, Kuo WL, Davies M et al. (2008). An integrative genomic and proteomic analysis of PIK3CA, PTEN, and AKT mutations in breast cancer. Cancer Res 68: 6084–6091.
Tomlinson GE, Chen TT, Stastny VA, Virmani AK, Spillman MA, Tonk V et al. (1998). Characterization of a breast cancer cell line derived from a germ-line BRCA1 mutation carrier. Cancer Res 58: 3237–3242.
Torbett NE, Luna-Moran A, Knight ZA, Houk A, Moasser M, Weiss W et al. (2008). A chemical screen in diverse breast cancer cell lines reveals genetic enhancers and suppressors of sensitivity to PI3K isoform-selective inhibition. Biochem J 415: 97–110.
Turner N, Lambros MB, Horlings HM, Pearson A, Sharpe R, Natrajan R et al. (2010). Integrative molecular profiling of triple negative breast cancers identifies amplicon drivers and potential therapeutic targets. Oncogene 29: 2013–2023.
Vasudevan KM, Barbie DA, Davies MA, Rabinovsky R, McNear CJ, Kim JJ et al. (2009). AKT-independent signaling downstream of oncogenic PIK3CA mutations in human cancer. Cancer Cell 16: 21–32.
Wang X, Jiang X . (2008). Post-translational regulation of PTEN. Oncogene 27: 5454–5463.
Weaver VM, Lelievre S, Lakins JN, Chrenek MA, Jones JC, Giancotti F et al. (2002). beta4 integrin-dependent formation of polarized three-dimensional architecture confers resistance to apoptosis in normal and malignant mammary epithelium. Cancer Cell 2: 205–216.
Wee S, Wiederschain D, Maira SM, Loo A, Miller C, deBeaumont R et al. (2008). PTEN-deficient cancers depend on PIK3CB. Proc Natl Acad Sci USA 105: 13057–13062.
Weigelt B, Bissell MJ . (2008). Unraveling the microenvironmental influences on the normal mammary gland and breast cancer. Semin Cancer Biol 18: 311–321.
Weigelt B, Lo AT, Park CC, Gray JW, Bissell MJ . (2010). HER2 signaling pathway activation and response of breast cancer cells to HER2-targeting agents is dependent strongly on the 3D microenvironment. Breast Cancer Res Treat 122: 35–43.
Yamada KM, Cukierman E . (2007). Modeling tissue morphogenesis and cancer in 3D. Cell 130: 601–610.
Yeh TC, Marsh V, Bernat BA, Ballard J, Colwell H, Evans RJ et al. (2007). Biological characterization of ARRY-142886 (AZD6244), a potent, highly selective mitogen-activated protein kinase kinase 1/2 inhibitor. Clin Cancer Res 13: 1576–1583.
Yuan TL, Cantley LC . (2008). PI3K pathway alterations in cancer: variations on a theme. Oncogene 27: 5497–5510.
Zhang S, Yu D . (2010). PI(3)king apart PTEN's role in cancer. Clin Cancer Res 16: 4325–4330.
We thank Miriam Molina Arcas, Ralph Fritsch, Elza de Bruin, Charles Swanton (CRUK London Research Institute) and Maryou Lambros and Jorge Reis-Filho (Breakthrough Breast Cancer Centre, London, UK) for helpful discussions, technical advice or critical reading of the manuscript, members of the LRI Equipment Park for sequencing and of the LRI FACS facility for cell-cycle analysis. We thank Morri Feldman and Kevan Shokat (UCSF) for providing PP242. This work was funded by Cancer Research UK.
The authors declare no conflict of interest.
Supplementary Information accompanies the paper on the Oncogene website
About this article
Cite this article
Weigelt, B., Warne, P. & Downward, J. PIK3CA mutation, but not PTEN loss of function, determines the sensitivity of breast cancer cells to mTOR inhibitory drugs. Oncogene 30, 3222–3233 (2011). https://doi.org/10.1038/onc.2011.42
- breast cancer
Feedback activation of STAT3 limits the response to PI3K/AKT/mTOR inhibitors in PTEN-deficient cancer cells
Network and Systems Medicine (2021)
Briefings in Bioinformatics (2021)
Frontiers in Oncology (2021)
Breast Cancer Research (2020)