Introduction
Cancer cells with an inherent or acquired capability to resist induction of apoptosis at some point(s) in the signal cascade pathway leading to cell death generally tend to be resistant to chemotherapy or radiotherapy. The serine–threonine protein kinase Akt has received much interest in recent years because it suppresses apoptosis induced by chemotherapy or radiotherapy through interaction with several critical molecules that regulate or execute apoptosis. For instance, after activation, Akt could do the following: it phosphorylates the proapoptotic protein Bcl-2 partner, Bad, which binds to and blocks the activity of Bcl-x, a factor in cell survival [
1]; it inactivates caspase-9, which initiates the caspase cascade leading to apoptosis [
2]; it represses the forkhead transcription factor FKHRL-1, which regulates the expression of the apoptosis-inducing Fas ligand [
3]; and it phosphorylates IκB, thereby promoting the degradation of IκB and increasing the activity of the nuclear factor κB (NFκB) [
3,
4].
The kinase activity of Akt is triggered after the interaction of its pleckstrin homology domain with the lipid second messenger phosphatidylinositol 3,4,5-trisphosphate, which is generated by phosphoinositide 3-kinase (PI3-K). This interaction recruits Akt from the cytoplasm to the inner cytoplasmic membrane, where Akt undergoes conformational changes and is phosphorylated by the phosphatidylinositol-dependent kinases. The activated Akt is then relocated to the cytoplasm and may be transported further to the nucleus, phosphorylating a wide spectrum of substrates including the molecules mentioned above that are involved in the regulation of cell survival. PI3-K itself is activated by multiple mechanisms, including the activation of growth factor receptor tyrosine kinases [
5,
6] and G protein-coupled receptors [
7,
8], integrin-mediated cell adhesion [
7,
8], and the actions of oncogene products such as Ras [
9,
10] and hormones such as estrogen [
11‐
13]. By controlling the levels of lipid second messengers, PI3-K regulates various cellular processes, including growth, differentiation, survival, migration and metabolism [
14,
15].
We have recently shown that expression of a constitutively active Akt, or an increased activity of the human epidermal growth factor receptor-2 (HER2)/PI3-K/Akt or Ras/PI3-K/Akt pathway, leads to multidrug or radiation resistance in human breast cancer cells [
16‐
18]. In those studies we assessed the sensitivity to chemotherapy (including doxorubicin) or radiotherapy of breast cancer cells that contain a higher level of Akt activity due to the overexpression of HER2, constitutively active Ras or constitutively active Akt. To expand our previous studies, we report here a differential pattern of responses of breast cancer cell lines in terms of Akt phosphorylation and activity as a result of treatment with doxorubicin. Depending on the cell types, treatment of breast cancer cells with doxorubicin may trigger a transient phosphorylation and activation of Akt. This therapeutic intervention-triggered activation of Akt depends on an inherent activity of PI3-K, and the capability of the response is potentiated after the expression of Akt upstream regulators including HER2, HER3 or the focal adhesion kinase (FAK), but not by all the signals that are known to affect Akt activity, an example of which is the estrogen-mediated signal. Deprivation of the effect of estrogen did not alter the responsiveness of MCF7 cells to doxorubicin-induced Akt phosphorylation. Our data suggest that the therapeutic intervention-triggered activation of Akt might contribute to the resistance of breast cancer cells to doxorubicin. These results provide further experimental evidence that justifies targeting the PI3-K/Akt pathway to enhance the efficacy of breast cancer chemotherapy or radiotherapy.
Materials and methods
Cell lines and cell cultures
Eight breast cancer cell lines used in this study (MCF7, MDA468, SKBR3, MDA157, MDA231, MDA361, BT474 and T47D) were originally purchased from American Type Culture Collection (Manassas, VA, USA). The cells were grown and routinely maintained in Dulbecco's modified Eagle's medium/F12 medium supplemented with 10% fetal bovine serum (FBS), 2 mM glutamine, 100 U/ml penicillin and 100 μg/ml streptomycin. MCF7HER2 cells were described previously [
19]. All cells were grown in a 37°C incubator supplied with 5% CO
2 and 95% air.
Western blot antibodies and other reagents
Antibodies directed against Akt, Ser473-phosphorylated Akt1 (p-Akt), Ser21/9-phosphorylated glycogen synthase kinase-3 (GSK3), Ser136-phosphorylated Bad and anti-HER2 monoclonal antibody were obtained from Cell Signaling Technology (Beverly, MA, USA). Anti-HER3 antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-His tag monoclonal antibody was ordered from Upstate Biotechnology (Charlottesville, VA, USA), as was the anti-FAK antibody that recognizes both FAK and FAK-related non-kinase (FRNK), a dominant-negative mutant of FAK [
20,
21].
The humanized anti-HER2 monoclonal antibody trastuzumab (Herceptin) was made by Genentech (San Francisco, CA, USA). PI3-K-specific inhibitor LY294002 was obtained from CalBiochem (San Diego, CA, USA), and the estrogen receptor (ER) antagonist ICI 182,780 was purchased from Tocris (Ballwin, MO, USA). Doxorubicin (Adriamycin) was ordered from the pharmacy of MD Anderson Cancer Center. All other reagents were purchased from Sigma-Aldrich (St Louis, MO, USA).
cDNA and transient expression
The pcDNA3 expression construct containing HER3 was provided by Dr Xiaofeng Le (MD Anderson Cancer Center), and the expression constructs of FAK and FRNK (pCMV-Myc) were kindly provided by Dr Thomas Parsons (University of Virginia, Charlottesville, VA, USA). Transient transfection was performed with the FuGENE 6 transfection kit, in accordance with instructions provided by the manufacturer (Roche Diagnostic, Indianapolis, IN, USA).
Western blot analysis and Akt kinase assay
Western blot analysis and Akt kinase assay were performed as described previously [
16,
19].
Cytoplasmic and nuclear fractionation
The method for cytoplasmic and nuclear fractionation was adopted from the literature [
22,
23] with minor modifications. In brief, pellets containing 2 × 10
7 cells were resuspended into 800 μl of buffer A (50 mM NaCl, 10 mM HEPES pH 8.0, 500 mM sucrose, 1 mM EDTA, 0.5 mM spermidine, 0.15 mM spermine, 0.2% Triton X-100, 1 mM phenylmethylsulphonyl fluoride, 2 mM Na
3VO
4, 25 μg/ml leupeptin, 25 μg/ml aprotinin). After incubation on ice for 10 min, the cells were homogenized with 10 strokes in a Dounce homogenizer. A small aliquot of the cell homogenates was then examined under a microscope to confirm that more than 98% of cells were lysed. After brief centrifugation of the cell homogenates at 4°C, the supernatant (cytoplasmic fraction) was collected and the pellet was washed twice with 400 μl of buffer B (50 mM NaCl, 10 mM HEPES pH 8, 25% glycerol, 0.1 mM EDTA, 0.5 mM spermidine, 0.15 mM spermine) and then resuspended in 150 μl of buffer C (350 mM NaCl, 10 mM HEPES pH 8.0, 25% glycerol, 0.1 mM EDTA, 0.5 mM spermidine, 0.15 mM spermine) with gentle rocking for 30 min at 4°C [
22]. After centrifugation, the supernatant (nuclear fraction) was collected. The amounts of protein in the cytoplasmic and nuclear fractions were determined with the Bradford method (Bio-Rad, Hercules, CA, USA).
Ionizing radiation
Cells grown on Petri dishes were irradiated with γ-rays from a high-dose-rate
137Cs unit (4.5 Gy/min) at room temperature (25 – 27°C), as described previously [
17,
19]. After irradiation, the cells were harvested by trypsinization.
Discussion
In our present study we found that the activity of Akt, an important signal molecule that promotes cell survival and confers cellular resistance to chemotherapy and radiotherapy as shown by us [
16,
18,
19] and others [
34,
35], was transiently elevated in a subset of breast cancer cell lines as a result of exposure to doxorubicin, a chemotherapeutic agent commonly used to treat patients with breast cancers. Activation of Akt in MCF7 cells after exposure to doxorubicin was reported earlier, but the mechanism was not explored in detail [
34,
35]. We noted here that, in comparison with resting (non-stimulated) cells, in which most Akt was found in the cytoplasm, exposure of the cells to doxorubicin or ionizing radiation led to a relocation of Akt to the nucleus. It is noteworthy that several antiapoptotic substrates of Akt are nuclear proteins. This subcellular translocation of Akt is important for cells to overcome the death signals initiated by treatment with doxorubicin or ionizing radiation. Taken together with our previous results [
16‐
18], the present results suggest that doxorubicin-triggered activation of Akt has a role in the resistance of breast cancer cells to this drug and that the same might apply to radiotherapy.
Because the overall cellular sensitivity of breast cancer cells to chemotherapy or radiotherapy is attributed to multiple intrinsic and extrinsic factors, such as p53 status, Bcl-2/Bax levels, expression of multiple drug resistance proteins, and hypoxic status, a caveat is that our data do not necessarily imply that one group of breast cancer cells showing increases in the level of p-Akt after chemotherapy or radiotherapy would absolutely be more chemoresistant or radioresistant than the another group of breast cancer cells without showing such a response. Rather, the data indicate that the activation and phosphorylation of Akt triggered by chemotherapy or radiotherapy contribute to the overall cellular sensitivity to these conventional therapies.
Several questions remain to be fully answered. First, why was Akt activation after treatment with doxorubicin found in only some of the breast cancer cell lines we tested? Apparently, cells must be equipped with certain molecular components that enable them to react to signals induced by chemotherapy or radiotherapy. We found that the drug-triggered activation of Akt depends on the activity of PI3-K, which can be activated by several known pathways, some of which we have explored in the present study. Which pathway is activated depends on the genetic context and functional status of the signal transduction network in individual cell types. In our study, MCF7 cells transiently expressing a high level of HER2 potentiated the response of the cells to the doxorubicin-induced activation of Akt. This result is consistent with those shown recently by us [
16,
19] and others [
36‐
38] indicating that HER2 expression in breast cancer cells might render them more resistant to chemotherapy or radiotherapy.
However, a high level of HER2 expression alone might not be sufficient to mediate this response. For example, we detected no change in the level of p-Akt in BT474 breast cancer cells after treatment with doxorubicin, even though they expressed a high level of HER2. SKBR3, another breast cancer cell line that expresses high levels of HER2, even showed a reduced level of p-Akt after treatment with doxorubicin. Expression of a transient transfected HER3 in the SKBR3 cells prevented this decline, indicating that heterodimerization and crosstalk between HER2 and HER3 might be important in mediating the downstream pathway that leads to Akt activation in breast cancer cells after treatment with doxorubicin. This might explain the negative findings from a recent clinical study reporting that HER2 overexpression does not seem to predispose to locoregional recurrence for breast cancer patients treated with neoadjuvant doxorubicin-based chemotherapy, mastectomy and radiotherapy [
39].
A second question is what molecular executioner leads to the activation of Akt after chemotherapy or radiotherapy. Are any soluble factors or non-secreted membrane-bound ligands involved, or is the PI3-K/Akt pathway activated directly and autonomously? In our study, we demonstrated that several different mechanisms, two of which are the expression of HER2 and of FAK, might enhance the doxorubicin-induced activation of Akt. Each mechanism activates PI3-K but does so through different ligands. Interference with these pathways by the anti-HER2 monoclonal antibody trastuzumab or by a dominant-negative mutant FAK (FRNK) abolished the drug-triggered activation of Akt mediated by HER2 and FAK, respectively. An interesting finding from our studies is that not all stimuli that lead to PI3-K activation enhance the drug-triggered activation of Akt. For example, abnormal estrogen exposure is associated with an increased risk of breast cancer, and estrogen is known to activate Akt via a non-nuclear estrogen-signaling pathway involving the direct interaction of ER with PI3-K [
40].
The ER isoform ERα binds to the p85α regulatory subunit of PI3-K in a ligand-dependent manner. Stimulation with estrogen increases ERα-associated PI3-K activity, leading to the activation of Akt. This interaction between ER and p85α is independent of gene transcription and does not involve phosphotyrosine adapter molecules or Src-homology domains of p85α [
40]. We found that the ER antagonist ICI 182,780 blocked estrogen-induced Akt activation in the ER-positive MCF7 cells but did not affect doxorubicin-induced Akt activation. Depletion of estrogen from the culture medium did not affect the doxorubicin-induced activation of Akt either. These data suggest that estrogen-induced signals, whether dependent on ER or not, are not involved in the pathway that enhances the doxorubicin-induced activation of Akt.
In fact, this atypical activation of Akt seems not to be limited to doxorubicin or ionizing radiation. We have observed that treatment of MCF7 cells with several different drugs (paclitaxel, 5-flurouracil and gemcitabine) that act through different mechanisms can also induce Akt phosphorylation, although the response and the timing and dose required for this effect varied between the drugs tested (data not shown). Cellular stress such as hypoxia and ultraviolet radiation has been reported by others to induce PI3-K-dependent Akt activation [
41‐
43]. Thus, inherent properties of individual cell types, rather than specific cell death signals, might determine whether Akt is activated after cells are exposed to stresses. Cancer cells with functional aberrations, such as overexpression of HER family members or increased cell adhesion potential, are probably more capable than noncancerous cells of activating Akt as a defensive mechanism against external detrimental stimuli, which justifies a novel approach of targeting the PI3-K/Akt for chemosensitization or radiosensitization.
In summary, doxorubicin might cause a PI3-K-dependent increase of Akt activity in breast cancer cells. Together with other recent results of ours [
16‐
18], the present observations suggest that clinical benefits in treating patients with breast cancer could be obtained with appropriate combinations of novel Akt inhibitors and conventional chemotherapeutic drugs or ionizing radiation.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
The authors' contributions to this research are reflected in the order shown, with the exception of ZF, who supervised all aspects of this research and prepared the manuscript. XL did most Western blot analyses and sample preparations; YL contributed the experiments of cell transfection and some Western blot analyses; KL performed radiation of the cells; BL participated in the overall design of experiments and data interpretation. All authors read and approved the final manuscript.