|Home | About | Journals | Submit | Contact Us | Français|
The epidermal growth factor receptor (EGFR) signaling pathway has emerged as a promising target for cancer therapy. EGFR tyrosine kinase inhibitors (TKIs) such as erlotinib have been approved for cancer treatment but have demonstrated very limited activity in breast cancer patients. Clarifying the molecular mechanism underlying resistance to EGFR-TKIs could lead to more effective treatment against breast cancer. We previously reported that the sensitivity of breast cancer cells to erlotinib is partially dependent on p27 and that cytoplasmic localization of p27 is associated with erlotinib resistance. In the present study, we found that erlotinib induces p27 phosphorylation at serine (S) 10, and S10 p27 phosphorylation leads to erlotinib resistance in EGFR-expressing breast cancer. Inhibiting S10 phosphorylation of p27 by knocking down human kinase interacting stathmin (KIS), a nuclear protein that can phosphorylate p27 at S10, led to p27 accumulation in the nucleus and enhanced erlotinib-mediated cytotoxicity. Further, in vivo KIS gene silencing enhanced the antitumor activity of erlotinib in an orthotopic breast cancer xenograft model. KIS depletion also enhanced erlotinib sensitivity in erlotinib-resistant EGFR-expressing triple-negative breast cancer cells. Our study provides a rationale for the development of combinations of erlotinib with KIS inhibition to overcome EGFR-TKI resistance in EGFR-expressing breast cancer.
Epidermal growth factor receptor (EGFR), a member of the ErbB receptor tyrosine kinase family, is frequently overexpressed in human malignant tumors and is known to drive tumor growth, progression, and metastasis (1–4). EGFR overexpression is associated with poor prognosis and reduced overall survival in patients with lung cancer (5, 6). Therefore, the EGFR signaling pathway has emerged as a promising target for cancer therapy, in particular lung cancer. A number of tyrosine kinase inhibitors (TKIs) that target EGFR have been developed and used successfully to treat cancer patients. For example, erlotinib (also known as OSI-774 or Tarceva, Fig. 1A), a small-molecule EGFR TKI that targets the ATP-binding site of EGFR, is used to treat non–small-cell lung cancer and pancreatic cancer (7, 8).
Although EGFR is overexpressed in 20% to 80% of breast cancers (3, 9, 10), erlotinib induces clinical responses in fewer than 10% of women with breast cancer. The molecular mechanism underlying resistance to EGFR TKIs in breast cancer must be better understood if EGFR is to be used successfully as a molecular target for breast cancer treatment.
p27Kip1 (p27), a major downstream molecule in EGFR signaling pathways, is a cyclin-dependent kinase (CDK) inhibitor that negatively regulates cellular proliferation by inhibiting progression through the cell cycle (11). The functions of p27 depend on the expression levels of p27 and its subcellular localization. In breast cancer, low p27 expression seems to be a marker of poor prognosis (12). The phosphorylation status of p27 is known to affect its nuclear-cytoplasmic localization (13–16), which itself is a critical determinant of growth arrest. p27 is phosphorylated at serine (S) 10, threonine (T) 157, T187, and T198. Phosphorylation of p27 at T187 is mediated by cyclin E-CDK2 (17), whereas phosphorylation at T157 and T198 is mediated by Akt (14–16, 18).
Phosphorylation of p27 at S10 is an important event in the regulation of the tumor suppressor function of p27 (19). Dephosphorylation of p27 at S10 inhibits p27 nuclear export and promotes its assembly into cyclin–CDK complexes, thereby inhibiting cell proliferation (13, 20–22). Phosphorylation of p27 at S10 takes place upon mitogenic stimulation by the nuclear protein human kinase interacting stathmin (KIS). KIS binds the C-terminal domain of p27 and phosphorylates it at S10 in vitro and in vivo, resulting in export of p27 from the nucleus to the cytoplasm and thereby regulating cell cycle progression (13, 23). KIS induces proliferation and cell cycle progression through the phosphorylation of p27 in leukemia cells (24). Overexpression of KIS can overcome p27-induced growth arrest, whereas KIS siRNA knockdown can enhance growth arrest by inhibiting S10 p27 phosphorylation (13).
We previously reported that the sensitivity of breast cancer cells to erlotinib is partially dependent on p27 and that cytoplasmic localization of p27 is associated with erlotinib resistance (25). In the present study, we showed that S10 p27 phosphorylation plays an important role in erlotinib sensitivity. Depletion of KIS by siRNA knockdown led to translocation of p27 to the nucleus and enhanced erlotinib’s cytotoxicity in breast cancer cells. More importantly, we found that the combination of erlotinib with KIS gene silencing had more significant antitumor effect than KIS siRNA knockdown alone or erlotinib alone in an orthotopic breast cancer xenograft model.
The breast cancer cell lines BT-474, SK-BR-3, MDA-MB-231 and MDA-MB-468 were purchased from American Type Culture Collection and were grown in Dulbecco’s modified Eagle’s medium/F12 medium supplemented with 10% fetal bovine serum (FBS, GIBCO/BRL), in a humidified atmosphere containing 5% CO2 at 37°C. The inflammatory breast cancer (IBC) cell line SUM149 was obtained from the University of Michigan. Another IBC cell line SUM190 was purchased from Asterand. The SUM149 and SUM190 cells were grown in Ham's F12 medium supplemented with 5% FBS, 5 µg/ml insulin, and 1 µg/ml hydrocortisone. SUM149 cells have been genotyped to ensure their identity by Link Genomics, Inc., Japan. Cells used for experiments were grown in culture for less than 2 months after resuscitation. All cell lines were tested to be mycoplasma free using a MycoAlert Mycoplasma Detection Kit (Lonza Cologne AG). pcDNA3-S10A-p27 (phosphoinhibitory S10A-p27 mutant generated by substituting an alanine at the S10 phosphorylation site) and pcDNA3-S10D-p27 (phosphomimetic S10D-p27 mutant generated by substituting an aspartic acid at the S10 phosphorylation site) plasmids were kindly provided by Dr. Keiichi I. Nakayama (Kyushu University, Japan) (22).
Erlotinib was kindly provided by OSI Pharmaceuticals, Inc. Erlotinib was dissolved in dimethyl sulfoxide (DMSO) as a 5 mM stock solution for in vitro experiments. Erlotinib was suspended in 0.5% methyl cellulose for oral gavage for in vivo animal work. Lapatinib was synthesized and dissolved in DMSO as a 10 mM stock solution as previously described (26).
Immunofluorescence analyses were performed as previously described (27). Rabbit anti-p27 antibody (Santa Cruz Biotechnology) was used as the primary antibody. Fluorescein isothiocyanate-conjugated antibodies (Biosource) were used as secondary antibodies. Cells were counterstained with propidium iodide before being mounted under glass coverslips and analyzed by confocal microscopy (FV300, Olympus).
Western blot were performed as previously described (25, 26). The antibodies used were rabbit anti-p27 antibody (Santa Cruz Biotechnology), rabbit anti-phospho-p27 (S10) antibody (Santa Cruz Biotechnology), rabbit anti-phospho-p27 (T157) antibody (R&D Systems), rabbit anti-phospho-p27 (T187) antibody (Santa Cruz Biotechnology), and mouse anti-β-actin antibody (Sigma-Aldrich).
WST-1 reagent (Roche Applied Science) was used to perform the WST-1 assay. A cell suspension of 4,000 cells/90 µl was seeded into each well of a 96-well plate and cultured overnight, after which 10 µl of erlotinib (or lapatinib) with various concentration was added to the individual wells. After 3 days of erlotinib (or lapatinib) treatment, 10 µl of the ready-to-use WST-1 reagent was added directly into the medium, the plates were incubated at 37°C for 1 h, and absorbance was measured on a plate reader at 450 nm. All experiments were done in triplicate. The percentage cell viability was calculated as (the absorbance of treated well minus the absorbance of cell-free control) / (absorbance of untreated control minus the absorbance of cell-free control) × 100. Median inhibitory concentrations were determined from these calculations.
RNA was extracted by using RNeasy kit (Qiagen). qRT-PCR was performed as described in detail elsewhere (28). The KIS primers used were as follows: upper primer: AAT CCT GGC AGA GGA CAA GTC TT, lower primer: GTA GAA TGT AGC CAC AAC AAA CTT CC.
KIS siRNA (5’-AAGCAGTTCTTG CCGCCAGGA-3’) and nontargeting control siRNA were purchased from Dharmacon Research Inc. RNA interference assay was performed according to the manufacturer’s protocol (Dharmacon Research). Briefly, cells were seeded in 6-well culture plates at 30% confluence in culture medium supplemented with fetal bovine serum. The next day, cells were transfected with siRNA at a final concentration of 100 nM by using Oligofectamine (Invitrogen).
For flow cytometry analysis, cells were plated in 60-mm dishes, cultured overnight, and then treated with or without erlotinib. After 48 h, floating and adherent cells were collected by trypsinization, fixed overnight in 70% ethanol, and resuspended in propidium iodide (25 µg/mL) supplemented with 0.1% RNase A. DNA content was measured with a FACScan flow cytometer (BD Biosciences). These experiments were repeated three times independently. Student’s t test was performed to compare the groups with respect to percentage of cells in G1 phase.
Anchorage-independent growth assay was performed as previously described (29). In brief, cells were treated with control siRNA or KIS siRNA for 48 h. After that, 5000 cells were cultured on a plate containing 0.8% base agar and 0.4% top agar in medium containing 1 µM erlotinib and incubated at 37°C for 21 days. Plates were stained with 0.005% crystal violet for 1 h. Colonies were counted by use of a dissecting microscope. These experiments were done in triplicate.
A total volume of 0.15 mL of BT-474 cell suspension containing 5 × 106 cells with 50% Matrigel was injected into the bottom left mammary fat pad of 8-week-old athymic female nu/nu mice. When well-established tumors measured an average volume of 60 mm3, mice (8 per group) were randomly assigned to six groups to receive the following treatments for 2 weeks: group 1, vehicle control (0.5% methyl cellulose); group 2, control siRNA; group 3, KIS siRNA; group 4, erlotinib; group 5, erlotinib combined with control siRNA; group 6, erlotinib combined with KIS siRNA. Erlotinib (100 mg/kg of body weight) was given orally once a day. For siRNA delivery, intravenous injections were given twice a week. Alexa 555 siRNA was purchased from Qiagen. siRNA for in vivo delivery was incorporated into 1,2-dioleoyl-sn-glycero-3-phosphatidylcholine (DOPC; Avanti Lipids) liposomes. The mixture was hydrated with normal (0.9%) saline at a concentration of 15 µg/mL to achieve the desired dose in 100 µL/injection. Tumor volume was determined weekly by externally measuring the tumors in 2 dimensions using a caliper. Volume (V) was determined by the following equation, where L is length and W is width of the tumor: V = (L × W2) × 0.5. Two-sample t tests were used to compare tumor volumes between the groups. Tumor growth inhibition (%) was calculated as: 1- (the tumor volume change of treatment group / the tumor volume change of vehicle control group).
Statistical analyses were performed using commercially available software (Statview, version 5.0, SAS Institute). The data obtained were analyzed by t test. Statistical significance was defined as P < 0.05.
We previously tested erlotinib sensitivity in breast cancer cell lines expressing various levels of EGFR and found that cytoplasmic localization of p27 is associated with erlotinib resistance in breast cancer cells (25). Among them, EGFR-expressing BT-474 and SK-BR-3 cells have limited sensitivity to erlotinib (IC50, 5.01 µM for BT-474 and 3.98 µM for SK-BR-3) (25). To determine the role of p27 localization in erlotinib cytotoxicity, we tested the expression levels of total p27 and phosphorylated p27 after erlotinib treatment in BT-474 and SK-BR-3 cells. Expression of both total p27 and p27 phosphorylated at S10 was significantly increased in these cells in response to erlotinib (Fig. 1B). Erlotinib did not, however, affect levels of p27 phosphorylated at T157 or T187, suggesting that phosphorylation of p27 at S10 may be the critical signal to export nuclear p27 to the cytoplasm in response to erlotinib (Fig. 1B).
Because S10 p27 phosphorylation leads to p27 cytoplasmic localization, we hypothesized that S10 p27 phosphorylation also leads to erlotinib resistance. To test this hypothesis, we transfected phosphoinhibitory S10A-p27 mutant or phosphomimetic S10D-p27 mutant to BT-474 and SK-BR-3 cells and then tested their sensitivity to erlotinib. We found that in BT-474 cells, p27 translocated to the nucleus after S10A-p27 transfection but remained in the cytoplasm after S10D-p27 transfection by immunofluorescence staining (Fig. 1C). p27 translocation to the nucleus was also detected in SK-BR-3 cells after S10A-p27 transfection (data not shown). We then treated cells with erlotinib and found that S10A-p27-transfected cells were more sensitive to erlotinib and S10D-p27-transfected cells were more resistant to erlotinib than empty vector-transfected cells (Fig. 1D), suggesting that phosphorylation of p27 at S10 leads to erlotinib resistance.
Because KIS is known to phosphorylate p27 at S10, we depleted KIS by KIS siRNA knockdown to inhibit S10 phosphorylation of p27 in BT-474 cells (Fig. 2A). KIS depletion led to strong accumulation of p27 in the nucleus (Fig. 2B). We then performed FACSan analysis to quantify the cell cycle distribution in control and KIS siRNA knockdown cells with and without erlotinib treatment. Even though both types of cells showed G1 cell cycle arrest after erlotinib treatment, more G1-phase cells were detected in KIS siRNA knockdown cells (82.1 ± 3.24%) than in control siRNA knockdown cells (71.4 ± 0.82%) after erlotinib treatment, indicating that KIS siRNA knockdown enhanced erlotinib-induced G1 cell cycle arrest (Fig. 2C).
We then tested whether inhibition of S10 phosphorylation of p27 enhances the anti-proliferative effects of erlotinib. BT-474 cells that were treated with KIS siRNA knockdown were more sensitive to erlotinib than were cells treated with control siRNA knockdown (Fig. 3A). To test whether the phosphorylation status of p27 at S10 plays a role in anchorage-independent growth in response to erlotinib, which may reflect in vivo tumorigenicity, we performed soft agar colony formation assay in BT-474 cells by treating them with KIS siRNA and erlotinib. We found that KIS siRNA-treated cells formed fewer colonies (i.e., were less tumorigenic) than control siRNA-treated cells in response to erlotinib (Fig. 3B). Similar results were obtained with SK-BR-3 cells (data not shown).
Since RNA interference can have off-target effects, we also transfected another siRNA targeting a different region of KIS into breast cancer cells. Depletion of KIS by KIS siRNA #2 also induced p27 translocation to the nucleus and enhanced erlotonib’s anti-proliferative effect in BT-474 cells (Supplementary Fig. S1).
Because inhibiting the phosphorylation of p27 at S10 enhanced erlotinib’s cytotoxicity in vitro, we hypothesized that the combination of erlotinib with KIS siRNA knockdown exhibits a stronger antitumor effect than either treatment alone on breast cancer in vivo. To test this hypothesis, we examined the therapeutic efficacy of combining erlotinib with KIS siRNA knockdown in an orthotopic breast cancer xenograft model established by implanting BT-474 cells in the mammary fat pads of nude mice. We used a well-established method to deliver KIS siRNA to tumors. This is a highly efficient in vivo gene silencing method using neutral liposome (DOPC)–encapsulated siRNA (30, 31). To confirm the efficacy of siRNA-DOPC delivery, we injected Alexa 555–DOPC, a nonsilencing siRNA tagged with the fluorochrome Alexa 555, into a mouse bearing BT-474 tumor and found that Alexa 555–DOPC was taken up deeply by the tumor (Fig. 4A). qRT-PCR showed that tumor treated with KIS siRNA had significantly lower KIS expression than tumor treated with control siRNA (Fig. 4B). Furthermore, we found that p27 accumulated to the nuclei of the tumor cells after KIS siRNA administration, indicating that KIS siRNA administration is functional in vivo (Fig. 4C).
At day 14 after treatment, tumors from vehicle-treated mice grew to an average size of 131 ± 22 mm3. Tumors from mice treated with KIS siRNA alone or erlotinib alone grew to an average size of 102 ± 33 mm3 and 93 ± 23 mm3, respectively, corresponding to a tumor growth inhibition of 37% and 53%, respectively. Compared with these single treatments, the combination of erlotinib with KIS siRNA knockdown was more efficient at inhibiting tumor growth: tumors from mice treated with the combination grew to an average size of 49 ± 13 mm3, corresponding to a 114% inhibition of tumor growth (Fig. 4D). The mice were treated for a total of 4 weeks and similar results were also found at different time points. At days 21 and 28 after treatment, combination of KIS siRNA knockdown and erlotinib exhibited enhanced antitumor activity than erlotinib alone or KIS siRNA knockdown alone, suggesting that KIS depletion enhanced erlotinib’s cytotoxicity in vivo (data not shown).
To determine whether the impact of phosphorylation of p27 at S10 on erlotinib sensitivity is a universal phenomenon in breast cancer, we performed similar experiments in other EGFR-expressing breast cancer cells: SUM149, an inflammatory breast cancer cell line that is sensitive to erlotinib (IC50 = 0.9 µM) (32), and MDA-MB-231 and MDA-MB-468, triple-negative breast cancer cell lines that are very resistant to erlotinib (IC50 > 20 µM for both) (25). In all three cell lines, we found the induction of pS10-p27 by erlotinib and found that KIS depletion enhanced erlotinib’s cytotoxicity, indicating that this effect is not restricted to a few cell lines in breast cancer (Fig. 5A).
We also studied the effect of KIS depletion on the activity of another TKI targeting EGFR. Lapatinib, a dual EGFR/HER2 TKI, was used to treat breast cancer cells with or without KIS siRNA knockdown. First, we tested the anti-EGFR activity of lapatinib in SUM149 cells that express EGFR but not HER2. Use of these cells allowed us to focus on the EGFR-blocking activity of lapatinib. In SUM149 cells, we found that lapatinib treatment increased the expression level of S10-phosphorylated p27 and KIS siRNA knockdown enhanced the anti-proliferative activity of lapatinib, suggesting that other EGFR TKIs can produce the same effects as erlotinib (Fig. 5B). Similar experiments were then performed in breast cancer cell lines that express both EGFR and HER2: BT-474 and SUM190. Interestingly, even though the expression level of S10-phosphorylated p27 increased after lapatinib treatment in both cell lines, KIS siRNA knockdown did not enhance lapatinib’s anti-proliferative activity in either lapatinib-sensitive BT-474 cells (IC50 = 0.046 µM) or lapatinib-resistant SUM190 cells (IC50 = 2.9 µM) (Fig. 5B).
In this study, we provided evidence that p27 cytoplasmic localization is associated with erlotinib resistance in breast cancer. Inhibition of S10 phosphorylation of p27 led to p27 nucleus accumulation and enhanced the anti-proliferative effect of erlotinib against breast cancer. Our findings elucidated the molecular mechanisms that explain why EGFR TKIs have minimum activity in breast cancer. Here we report for the first time to use KIS in vivo siRNA delivery technique in combination with erlotinib to enhance the EGFR TKI sensitivity.
p27 is an important negative regulator of the mammalian cell cycle. The activity of p27 depends on the subcellular localization of p27. Even though phosphorylation status of p27 at T157, T187, and S10 plays a role in p27 localization, we found that only S10 p27 phosphorylation increased in response to erlotinib treatment, indicating that S10 phosphorylation is involved in erlotinib resistance in breast cancer. The effects produced by KIS siRNA knockdown on growth and anchorage-independent growth are only partial. Therefore, other mechanisms may contribute to erlotinib resistance. We speculate that other types of phosphorylated p27 may play roles in erlotinib resistance. Our current study involved testing only three phosphorylation sites of p27 (S10, T157, and T187), but other phosphorylation sites of p27 may be important for regulating p27 localization and stability. For example, p27 can be phosphorylated by the oncogenic kinase Src at tyrosine (Y) 74 and Y88, and this phosphorylation reduces the p27-mediated inhibition of cyclin E-CDK2 (33). Phosphorylation of p27 at T198 by Akt leads to p27 cytoplasmic mislocalization (18). Therefore, currently we are also focusing on other phosphorylation sites on p27 to further elucidate the molecular mechanism of EGFR-TKI resistance.
HER2/ErbB2 is another member of the ErbB family of receptor tyrosine kinases. Overactivation of HER2 plays a critical role in the progression of human breast cancer (34). Targeting of HER2 by anti-HER2 antibodies such as trastuzumab is now widely used in the treatment of patients with HER2-positive breast cancer (35). Even though trastuzumab showed relevant clinical activity against HER2-positive breast cancer, the overall rate of response rates to trastuzumab as a single agent is only 15% to 30% (36). Resistance to trastuzumab therapy remains a significant clinical problem. p27 is also a downstream molecule of the HER2 pathway and is involved in trastuzumab resistance. It was reported recently that activation of MET receptor tyrosine kinase contributes to trastuzumab resistance, as Met protects cells against trastuzumab by abrogating p27 induction (37). By inhibiting Akt, trastuzumab can inhibit KIS expression and therefore inhibit KIS-induced nuclear export of p27 (21). Therefore, our study suggests that one way to overcome trastuzumab resistance is to modulate S10 phosphorylation and subcellular localization of p27.
Lapatinib is a dual TKI that inhibits both EGFR and HER2. We previously reported that the major activity of lapatinib in EGFR/HER2-positive breast cancer cells is through its anti-HER2 activity and not through anti-EGFR activity (26). In this study, our results provided further evidence to support this finding. When we used lapatinib to treat SUM149 cells, which express EGFR but not HER2, we found that EGFR blockage by lapatinib was enhanced by KIS depletion. In contrast, KIS depletion did not enhance lapatinib’s activity in cells that expressed both EGFR and HER2, suggesting that the activity of lapatinib in such cells is mainly through HER2.
In the current study, we tested the role of S10 phosphorylation of p27 in erlotinib sensitivity not only in cells that overexpress both EGFR and HER2, but also in cells that overexpress EGFR but not HER2. EGFR is reported to be overexpressed in > 60% of basal-like breast cancers (38, 39), and EGFR gene amplification is also found in a subset of basal-like tumors (40). Basal-like breast cancer often has a so-called triple-negative phenotype—in other words, it lacks expression of estrogen receptor, progesterone receptor, and HER2 (41, 42). Triple-negative breast cancer is the most aggressive form of primary breast cancer, and the majority of these tumors cannot be managed effectively with existing targeted treatments (trastuzumab and hormonal treatments) (43, 44). EGFR expression is associated with early relapse and poor survival in triple-negative breast cancer, suggesting that EGFR might be a promising target in this type of disease (45). Several studies have reported the use of cetuximab, a humanized monoclonal antibody against EGFR, in treatment of the basal-like tumor type (46). Another EGFR TKI, gefitinib, enhanced response to other chemotherapeutic drugs such as carboplatin and docetaxel (47). Therefore, even though erlotinib induces clinical responses in only a small proportion of breast cancer patients, there is still reason to believe that inhibition of the EGFR pathway can have substantial activity against triple-negative breast cancer. Identifying the molecules involved in and necessary for resistance to EGFR TKIs will enable us to develop clinically relevant therapeutic approaches by making EGFR a relevant target for breast cancer. In this study we tested the impact of p27 phosphorylation on erlotinib’s activity in 2 triple-negative EGFR-expressing breast cancer cells MDA-MB-231 and MDA-MB-468. We found that inhibiting S10 phosphorylation of p27 enhanced the sensitivity of erlotinib to these cells, suggesting that this effect is not restricted to a few cell lines but is a universal phenomenon.
It is still unclear whether there is some relationship between pS10-p27 expression and erlotinib sensitivity in breast cancer cells. To study whether erlotinib sensitivity depends on p27 phosphorylation in breast cancer cells, we tested the basal expression levels of pS10-p27 in a panel of breast cancer cell lines including erlotinib-sensitive cell lines SUM149 and KPL-4 (32), erlotinib-moderate-sensitive cell lines SK-BR-3 and BT-474 (25), and erlotinib-resistant cell lines MDA-MB-231 and MDA-MB-468 (25). The basal expression level of pS10-p27 is low in most of the cell lines and is not correlates with erlotinib sensitivity (data not shown). The underlying mechanism linking EGFR and KIS is also not clear. To investigate whether the KIS expression level is regulated by EGFR activation, we activated EGFR by EGF stimulation in BT-474 and SK-BR-3 cells and then tested KIS expression. We found that KIS expression did not change significantly after EGFR activation (Supplementary Fig. S2). In a future study, we will focus on detecting whether KIS can regulate the EGFR pathway.
Acquired resistance to EGFR TKIs in non–small-cell lung cancer commonly occurs after continuous drug administration. It is reported that MET amplification is involved in this acquired resistance (48). Therefore, targeting of MET may enhance the sensitivity of non–small cell lung cancer to EGFR TKIs. However, even though MET was found to be highly activated in cetuximab-resistant cells, inhibition of MET activity did not sensitize cetuximab-resistant cells to cetuximab. The reason why MET inhibition does not restore cetuximab sensitivity is still unknown. Because MET’s ligand hepatocyte growth factor induces cell cycle progression in medulloblastoma cells in a p27- and Cdk2-dependent manner (49), we speculate that modulating p27 directly may be more powerful than MET inhibition to abrogate the acquired resistance to EGFR TKIs and is thus worthy of prospective clinical investigation.
Overall, our study demonstrated that p27 phosphorylation at S10 plays a critical role in breast cancer sensitivity to erlotinib. Combining EGFR TKIs with siRNA knockdown of KIS, which leads to inhibition of S10 phosphorylation of p27, enhanced erlotinib activity both in vitro and in vivo in breast cancer. We expect this combination therapy to be potentially translatable to clinical use, where it may ultimately improve the efficacy of EGFR-TKIs for women with advanced breast cancer that is resistant to EGFR-TKIs.
We thank OSI Pharmaceuticals, Inc. for providing erlotinib, Dr. Keiichi I. Nakayama (Kyushu University, Japan) for providing pcDNA3-S10A-p27 and pcDNA3-S10D-p27 plasmids, Dr. Elizabeth G. Nabel (National Institutes of Health, Bethesda) for helpful suggestions, Ms. Ping Liu and Mr. Roland Bassett (Department of Biostatistics, The University of Texas M. D. Anderson Cancer Center) for statistical analysis, and Ms. Stephanie P. Deming (Department of Scientific Publications, The University of Texas M. D. Anderson Cancer Center) for editorial assistance.
Grant Support: NIH grant CA123318-01A1 (Naoto T. Ueno) and Susan G. Komen Postdoctoral Fellowship KG091192 (Dongwei Zhang).
Disclosure of Potential Conflicts of Interest: No potential conflicts of interest were disclosed.