ALDH-positive lung cancer stem cells confer resistance to epidermal growth factor receptor tyrosine kinase inhibitors

Cancer Letters 328 (2013) 144–151

Contents lists available at SciVerse ScienceDirect

Cancer Letters journal homepage: www.elsevier.com/locate/canlet

ALDH-positive lung cancer stem cells confer resistance to epidermal growth factor receptor tyrosine kinase inhibitors Cheng-Po Huang a,1, Meng-Feng Tsai b,1, Tzu-Hua Chang c, Wei-Chien Tang a, Su-Yu Chen a, Hsiao-Hsuan Lai a, Ting-Yu Lin a, James Chih-Hsin Yang d, Pan-Chyr Yang c, Jin-Yuan Shih c,⇑, Shwu-Bin Lin a,⇑ a

Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, Taipei, Taiwan Department of Molecular Biotechnology, College of Biotechnology and Bioresources, Dayeh University, Changhua, Taiwan Department of Internal Medicine, National Taiwan University Hospital, College of Medicine, National Taiwan University, Taipei, Taiwan d Graduate Institute of Oncology and Cancer Research Center, College of Medicine, National Taiwan University, Taipei, Taiwan b c

a r t i c l e

i n f o

Article history: Received 6 February 2012 Received in revised form 17 July 2012 Accepted 17 August 2012

Keywords: Lung cancer Aldehyde dehydrogenase 1A1 Cancer stem cell Gefitinib resistance Tyrosine kinase inhibitor

a b s t r a c t Molecular targeting therapeutics, such as EGFR tyrosine kinase inhibitors (TKIs), are important treatment strategies for lung cancer. Currently, the major challenge confronting targeted cancer therapies is the development of resistance. Cancer stem cells (CSCs) represent a rare population of undifferentiated tumorigenic cells responsible for tumor initiation, maintenance and spreading. Resistance to conventional chemotherapeutic drugs is a common characteristic of CSCs. However, the issue of whether CSCs contribute to EGFR TKI resistance in lung cancer is yet to be established. In the current study, we explored the association of ALDH1A1 expression with EGFR TKI resistance in lung cancer stem cells. ALDH1A1positive lung cancer cells displayed resistance to gefitinib, compared to ALDH1A1-negative lung cancer cells. Moreover, PC9/gef cells (gefitinib-resistant lung cancer cells) presented a higher proportion of ALDH1A1-positive cells, compared to PC9 cells (gefitinib-sensitive lung cancer cells). Clinical sample studies were consistent with results from cell culture model systems showing that lung cancer cells with resistance to EGFR TKI and chemotherapy drugs contain significantly increased proportions of ALDH1A1positive cells. These findings collectively suggest that ALDH1A1 positivity in cancer stem cells confers resistance to EGFR TKI in lung cancer. Ó 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Lung cancer is the leading cause of cancer-related mortality worldwide, with non-small cell lung cancer (NSCLC) accounting for about 80–85% of all cases. Despite considerable progress in diagnosis and treatment, the overall five-year survival rate of NSCLC patients remains lower than 15% [1,2]. Traditional therapeutic strategies (chemotherapy and radiotherapy) are often associated with unsatisfactory outcomes in lung cancer patients. Thus, inhibition of the epidermal growth factor receptor (EGFR) tyrosine

⇑ Corresponding authors. Addresses: Department of Internal Medicine, National Taiwan University Hospital, College of Medicine, National Taiwan University, 7 Chung-Shan South Road, Taipei 100, Taiwan. Tel.: +886 2 23123456x62905; fax: +886 2 2358 2867 (J.-Y. Shih). Department of Clinical Laboratory Sciences and Medical Biotechnology, College of Medicine, National Taiwan University, 7 ChungShan South Road, Taipei 100, Taiwan. Tel.: +886 2 23123456x66910; fax: +886 2 23711574 (S.-B. Lin). E-mail addresses: [email protected] (J.-Y. Shih), [email protected] (S.-B. Lin). 1 These authors contributed equally to this work. 0304-3835/$ - see front matter Ó 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.canlet.2012.08.021

kinase activity has become an important alternative therapeutic strategy for lung cancer treatment [3–5]. However, potential development of resistance is a significant challenge in the advancement of EGFR-targeted cancer therapies. NSCLC patients with EGFR mutations, specifically, L858R and exon 19 deletions, display an initial clinical response to EGFR tyrosine kinase inhibitors (TKIs). Despite dramatic initial responses to these drugs, the majority of patients ultimately develop drug resistance and relapse. Several clinical studies have indicated that a second-site point mutation of EGFR at position 790 (T790M) and MET gene amplification contribute to acquisition of EGFR TKI resistance [6,7], but the molecular mechanisms leading to drug resistance in the remaining cases are currently unknown. Elucidation of the molecular machinery underlying tumor progression following treatment with EGFR TKIs may facilitate the development of more effective therapeutic interventions. Accumulating evidence supports the concept that the cancer stem cell (CSC), also designated ‘tumor-initiating cell’, is a rare population of undifferentiated tumorigenic cells responsible for tumor initiation, maintenance and metastasis. These cells are capable of self-renewing, differentiating, and maintaining tumor growth

C.-P. Huang et al. / Cancer Letters 328 (2013) 144–151

145

and heterogeneity, and play an important role in tumorigenesis and resistance to therapy [8,9]. CSCs have been isolated from leukemia and several solid cancers. The cells can be expanded in vitro as tumor spheres, while reproducing the original tumor when transplanted in immunodeficient mice [8–11]. Resistance to conventional chemotherapeutic drugs is a common characteristic of CSCs [9,12]. However, no studies to date have focused on the characterization of CSCs contributing to EGFR TKI resistance in lung cancer. The aldehyde dehydrogenase (ALDH) family is cytosolic isoenzyme responsible for oxidizing intracellular aldehydes, thus contributing to the oxidation of retinol to retinoic acid in early stem cell differentiation. Recently, activation of ALDH1A1 has been found in stem cells populations in different cancers, including breast cancer, bladder cacner, head and neck squamous cancer and lung cancer [13–16]. An Aldefluor kit (StemCell Technologies) was designed for optimal identification and isolation of ALDH1A1 caner stem cells using specific ALDH1A1 inhibitor, diethylaminobenzaldehyde (DEBA), and the Aldefluor flow cytometry assay [13–16]. Notably, high expression of ALDH1A1 is reported to be correlated with poor clinical prognosis in breast and lung cancer [13,17]. In the current study, we explored the expression of ALDH1A1, with a view to its application as a potential marker for lung cancer stem cells. Our main aim was to determine the significance of ALDH1A1-positive CSCs in the mediation of EGFR TKI resistance in lung cancer.

a 25 ll reaction mixture. Cycling conditions were used according to the manufacturer’s instructions. Comparative gene expression analysis was performed using the 2(DDCt) method with normalization to the level of internal control gene, GAPDH. The primer sequences used are as follows: Oct4-QR: 50 -TGCTCCAGCTTCTCCT TCTC -30 ; c-Myc-QF: 50 -GGAACGAGCTAAAACGGAGCT-30 , c-Myc-QR: 50 -GGCCTTTT CATTGTTTTCCAACT-30 ; Bmi1-QF: 50 -AAATGCTGGAGAACTGGAAAG-30 , Bmi1-QR: 50 -CTGTGGATGAGGAGACTGC-30 ; Nanog-QF: 50 -ATTCAGGACAGCCCTGATTCTTC-30 , Nanog-QR: 50 -TTTTTGCGACACTCTTCTCTGC-30 ; Nestin-QF: 50 -AGGAGGAGTTGGGTT CTG-30 , Nestin-QR: 50 -GGAGTGGAGTCTGGAAGG-30 ; KLF4-QF: 50 -CCGCTCCATTACC AAGAGCT-30 , KLF4-QR: 50 -ATCGTCTTCCCCTCTTTGGC-30 ; SOX2-QF: 50 -CGAGTGGAA ACTTTTGTCGGA-30 , SOX2-QR: 50 -TGTGCAGCGCTCGCAG-30 ; GAPDH-QF: 50 -GGTATC GTGGAAGGACTCATGAC-30 , GAPDH-QR: 50 -ATGCCAGTGAGCTTCCCGTCAGC-30 .

2. Materials and methods

Animal experiments were approved by the Institutional Animal Care and Use Committee of National Taiwan University. Six-week-old NOD/SCID male mice were obtained from the animal center at the College of Medicine, National Taiwan University, Taipei, Taiwan. Mice were maintained under pathogen-free conditions and a 12-h light/dark cycle, with free access to food and water. All mice were cared for in accordance with institution guidelines. In the tumorigenicity assay, PC9 and PC9/gef cells were dissociated with trypsin and re-suspended in 1XPBS. Cells (106) were subcutaneously injected into the abdomen of mice, and tumor sizes measured weekly. Tumor volume, V, was calculated using the equation: V = W2  L  0.5, where W and L represent tumor width and length, respectively [22].

2.1. Cell cultures Human lung adenocarcinoma cell lines, PC9 (gefitinib-sensitive cells with EGFR mutation), and derivative PC9/gef (gefitinib-resistant cells) [18], were gifts from Dr. Yang CH (Graduate Institute of Oncology and Cancer Research Center, National Taiwan University). Cells were cultured in DMEM (Gibco-Invitrogen, Carlsbad, CA) containing 10% fetal calf serum (FCS), 2 mM L-glutamine, 100 U/ml penicillin and 100 lg/ml streptomycin at 37 °C in a humidified 5% CO2 atmosphere.

2.2. Aldefluor assay and cell sorting Cell populations with high ALDH1A1 enzymatic activity in lung cancer cells were optimized and identified with the Aldefluor kit (StemCell Technologies, Vancouver, BC, Canada) using specific ALDH1A1 inhibitor, diethylaminobenzaldehyde (DEBA), and the Aldefluor flow cytometry assay [16,17]. The cell-permeable enzyme substrate, BODIPY-aminoacetaldehyde (BAAA), is intracellularly retained following conversion by ALDH to the fluorescent BODIPY-aminoacetate. Briefly, 106 cells were re-suspended in 1 ml Aldefluor buffer and 1 ll Aldefluor reagent in the presence or absence of the specific ALDH1A1 inhibitor for 30 min at 37 °C, according to the manufacturer’s protocol. Brightly fluorescent ALDH1A1-positive cells were detected in the green fluorescence channel, FL1, and samples treated with the specific ALDH1A1 inhibitor, DEBA, were used as the control to set the gates defining the ALDH1A1-positive region. After sorting, cells were washed and subsequently used for tumor sphere formation and detection of specific gene expression.

2.3. Tumor sphere formation assay In the sphere formation experiment, ALDH-positive and ALDH-negative cells were sorted from the PC-9 and PC-9/gef cell lines, and cultured in serum-free DMEM–F12 medium (Gibco-Invitrogen, Carlsbad, CA) containing N2 supplement (Gibco-Invitrogen, Carlsbad, CA), 20 ng/ml EGF, 10 ng/ml bFGF (R&D Systems, Minneapolis, MN) and antibiotics. The indicated number of cells was seeded on ultralow attachment plates (Corning, Corning, NY) in triplicate wells. Spherical aggregates comprising more than 16 cells were counted after one week.

2.4. Real-time quantitative reverse transcription-polymerase chain reaction (RT-PCR) Expression of related stem cell genes, such as Oct4, c-Myc, Bmi1, Nanog, Nestin, KLF4, and SOX2, was detected using real-time quantitative RT-PCR. Total RNA was extracted using TRIzol reagent (Gibco-Invitrogen, Carlsbad, CA). The procedures for RNA extraction and RT-PCR have been described previously [19]. Briefly, 5 lg of total RNA was obtained for cDNA synthesis using the Oligo dT primer. Real-time quantitative PCR was carried out using 50–100 ng cDNA and 250 nM specific primer pairs in 2  SYBR Green PCR Master Mix (PE Applied Biosystems, Foster City, CA) in

2.5. Sequencing of EGFR exons 18–21 The tyrosine kinase domain of the EGFR coding sequence, exons 18, 19, 20, and 21 were amplified by PCRs from cDNA, and PCR amplicons were purified as described previously [20,21]. 2.6. Cell viability assay The cell viability assay was performed immediately after sorting. ALDH-positive and ALDH-negative PC9 cells were seeded into 96-well plates (3000 cells/well) and cultured with 10% FBS DMEM/F12 medium. Cell viability was evaluated using the Abacus Cell Proliferation kits (Clontech, Palo Alto, CA), which measured cellular acid phosphatase activity (ACP assay). Absorbance was measured at 410 nm using the SpectraMax M5 Microplate Reader (Molecular Devices, Sunnyvale, CA). 2.7. Tumorigenicity assay

2.8. Malignant pleural fluid isolation and culture This study was approved by the Institutional review board (IRB) of National Taiwan University Hospital (NTUH). Between July 2009 and March 2010, we consecutively collected pleural effusions from patients receiving thoracentesis in the chest ultrasonography examination room of NTUH. Pleural fluids of patients were acquired aseptically in vacuum bottles via thoracentesis. Red blood cells in the specimen were hemolyzed using RBC lysis buffer. The remaining cells were washed twice with PBS and cultured in complete DMEM, as described previously [23]. After 5 days, primary living cells were formed. Primary cells were treated with increasing doses of gefitinib for 72 h, and the proportion of ALDH-positive cells estimated. 2.9. Statistical analysis The student’s t-test was used to compare the mean values of the two groups. Two-sided p-values less than 0.05 were considered significant. All analyses were performed using SPSS software, version 15.0, for Windows (SPSS Inc, Chicago, IL).

3. Results 3.1. Characterization and isolation of ALDH1A1-positive CSCs from PC9 cells Gefitinib-sensitive PC9 lung cancer cells were investigated for ALDH1A1 activity using the specific ALDH1A1 inhibitor (DEBA) and the Aldefluor assay, followed by flow cytometry. The PC9 subpopulations, including ALDH1A1-positive and ALDH1A1-negative cells, were sorted via FACS analysis. The ALDH1A1-positive cell population of PC9 cells was about 6.5% (Fig. 1A). Earlier studies have shown that tumor spheres are enriched in CSCs grown in anchorage-independent and serum-free conditions. Lung tumor spheres exhibit cancer stem cell features and have

146

C.-P. Huang et al. / Cancer Letters 328 (2013) 144–151

Fig. 1. Identification of the ALDH1A1-positive population from PC9 lung cancer cells. ALDH1A1-positive cells were identified and sorted using the specific ALDH1A1 inhibitor (DEBA) and the Aldefluor flow cytometry assay (A). ALDH1A1-positive, but not ALDH1A1-negative cells, formed spheres (B). Gene expression levels in ALDH-negative and positive cells were monitored using real-time quantitative RT-PCR (C). The repopulation ability of the ALDH1A1-positive PC9 cells were detected (D).

high tumorigenic potential. After harvesting ALDH1A1-positive cells derived from PC9 lung cancer cells, we examined the properties of CSCs via sphere formation, and determined the expression levels of CSC-related genes. ALDH1A1-positive and ALDH1A1-negative PC9 cells were plated in the tumor sphere formation assay, and the number and size of tumor spheres analyzed. We observed no significant aggregations of tumor spheres formed by ALDH1A1negative PC9 cells, while 6 ± 1.8 (n = 5) tumor spheres were formed per microscopic field by ALDH1A1-positive PC9 cells. The average diameter of the spheres formed by ALDH1A1-positive PC9 cells was 162 ± 35.1 lm. Results from one of three independent experiments are shown in Fig. 1B. Expression of various biomarkers putatively related to cancer stem cells was investigated in freshly sorted ALDH1A1-positive and ALDH1A1-negative subpopulations of PC9 cells. The gene

levels of Oct4, c-Myc, Bmi1, Nanog, Nestin, KLF4, SOX2 and ABCG2 were measured using real-time quantitative reverse transcriptase PCR. Bmi1, Nanog, Nestin, KLF4, and SOX2 mRNA levels were significantly increased in ALDH1A1-positive cells (Fig. 1C). Specifically, a 3.1-fold increase in Nanog mRNA and 3.2-fold increase in KLF4 mRNA was estimated in ALDH1A1-positive cells, compared to ALDH1A1-negative cells. In contrast, Oct4, c-Myc and ABCG2 mRNA levels were decreased or no significant difference in ALDH1A1-positive, compared to ALDH1A1-negative PC9 cells. These results support the hypothesis that high ALDH1A1 activity acts as a marker of a population enriched for lung cancer stem cells. To evaluate the repopulation ability of the ALDH1A1-positive PC9 CSCs, we cultured the sorted ALDH1A1-positive PC9 CSCs in DMEM medium with 10% fetal calf serum (under differentiation condition) for 4–14 days. After each culture period, the proportion

C.-P. Huang et al. / Cancer Letters 328 (2013) 144–151

147

Fig. 2. ALDH1A1-positive cells derived from PC9 lung cancer cells are resistant to EGFR-TKI and anti-cancer chemotherapeutic drugs. The cell viability assay indicated that ALDH-positive cells show resistance to gefitinib (A), cisplatin (B), etoposide (C) and fluorouracil (D), whereas ALDH-negative cells are sensitive to those drugs. The ALDH1A1positive PC9 cells after 14 days of culture (under attachment condition) were evaluated for response to EGFR TKI gefitinib (E).

of ALDH1A1-positive and ALDH1A1-negative subpopulation was analyzed. The results indicated that the proportion of ALDH1A1positive and ALDH1A1-negative subpopulation were 28.6% and 71.4% after 7 days of culture. With an additional 7 days of culture, the proportion of these two subpopulations was 9.4% and 90.6%, respectively. The ALDH1A1-positive PC9 CSCs regenerated both ALDH1A1-positive and ALDH1A1-negative subpopulation and tended to reconstruct ALDH-negative subpopulation (Fig. 1D).

EGFR TKI gefitinib. As shown in Fig. 2E, incubating of the ALDH1A1-positive PC9 CSCs in DMEM medium with 10% fetal calf serum for two weeks were evaluated more sensitive to EGFR TKIs, when compared to the ALDH1A1-positive PC9 CSCs examined just after sorting.

3.2. ALDH1A1-positive CSCs from PC9 cells are resistant to gefitinib and chemotherapeutic drugs

Next, we examined whether TKI drug-resistant lung cancer cells are enriched in the CSC population using the human lung cancer cell lines, PC9 (gefitinib-sensitive) and PC9/gef (gefitinib-resistant). PC9 cells expressing mutant EGFR with a deletion in exon 19 represent gefitinib-sensitive NSCLC [18]. PC9/gef cells were selected from parental PC9 that had been continuously exposed to increasing concentrations of gefitinib. Previously, we reported IC50 values of 0.041 lM and >5 lM for gefitinib in PC9 and PC9/gef cells, respectively. Cell viability remained unaltered in PC9/gef in response to increasing doses of gefitinib from 0.025 to 5 lM [18]. To investigate the differences in tumorigenic potential between PC9 and PC9/gef cells, 1  105 cells were subcutaneously injected into NOD-SCID mice, which were monitored for tumor development. As shown in Fig. 3A, PC9/gef xenograft tumors were significantly larger than those of PC9 xenograft after 40 days. PC9/gef cells exhibited significant tumor-forming capacity, whereas the tumor growth rate of PC9 cells was significantly slower. The expression levels of related stem cell genes and markers in PC9 and PC9/gef cells were additionally evaluated using quantitative RT-PCR. Stem cell-related genes, Bmi1, Nanog, and SOX2, were

Cancer stem cells are involved in carcinogenesis, invasion and metastasis, and play a key role in chemo- and radiotherapy resistance. However, the issue of whether CSCs contribute to EGFR TKI resistance in lung cancer is currently unclear. Accordingly, we investigated the responses of ALDH1A1-positive PC-9 CSCs to TKI and chemotherapeutic drugs. ALDH1A1-positive and ALDH1A1-negative cell populations derived from PC9 cells were treated with different doses of gefitinib and anti-cancer chemotherapeutic drugs, including cisplatin, etoposide and fluorouracil. Cell viability was evaluated with the Abacus Cell Proliferation (ACP) assay, and resistance quantified as a percentage of cell viability after drug exposure for 72 h. The cell viability assay revealed that ALDH1A1-positive PC9 cells were significantly more resistant to therapy than ALDH1A1-negative cells receiving 1 lM gefitinib, 50 lM cisplatin, 5 lM etoposide or 50 lM fluorouracil (Fig. 2A–D). We also evaluated the response of the ALDH1A1-positive PC9 CSCs after two weeks of culture (under attachtment condition) to

3.3. PC9/gef, gefitinib-resistant lung cancer cells, present a significantly higher ALDH1A1-positive cell population than PC9 cells

148

C.-P. Huang et al. / Cancer Letters 328 (2013) 144–151

Fig. 3. An enriched ALDH1A1-positive population in PC9/gef cells expressed the CSC related genes. PC9/gef cells exhibited higher tumor growth rate than PC9 cells in the tumorigenesis assay in the SCID mouse (A). The CSC-related genes were analyzed in PC9 and PC9/Gef cells (B). Comparing to PC9 cells, PC9/gef cells contained a greater population of ALDH1A1-positive cells (C).

expressed at higher levels in PC9/gef cells, compared to PC9 cells (Fig. 3B), suggestive of an enriched cancer stem cell population. Aldefluor flow cytometry assay data showed that the ALDH1A1positive cell populations constitute about 6.5% of PC9 cells and 37.8% of PC9/gef cells (Fig. 3C). Cancer stem cell populations were clearly enriched in TKI drug-resistant PC9/gef cells. Based on these findings, we propose that ALDH1A1-positive cell populations play a crucial role in gefitinib resistance in our cell model. We analyzed the EGFR exons 18–21 by sequencing. The EGFR mutation in both of the ALDH1A1-positive and ALDH1A1-negative subpopulations was the same deletion in exon 19. There were no other alterations, such as T790M detected in these two subpopulations. These results suggested that the kinase domain of EGFR were conserved between ALDH1A1-positive and ALDH1A1-negative subpopulations, and the mechanisms of resistance to gefitinib in ALDH1A1-positive PC9 subpopulation were not related to a secondary mutation T790M in EGFR. 3.4. ALDH1A1-positive cell populations are increased in lung cancer samples displaying resistance to EGFR TKI or chemotherapy To apply our findings to the clinic, malignant pleural effusions were collected from 10 patients before anti-cancer treatment

(treatment naïve), 6 with acquired resistance to first-line EGFR TKI, 6 with resistance to chemotherapy, and 4 that failed to respond to EGFR TKI and chemotherapy. Primary cells were collected and isolated from malignant pleural effusions of lung cancer patients, washed with PBS and re-suspended in DMEM supplemented with 10% fetal calf serum. After one week of culture, we evaluated the proportion of ALDH1A1-positive primary cells. The primary cells from lung cancer patient’s before anti-cancer treatment contained an ALDH1A1-positive cell population of about 10.5% (Fig. 4A). The ALDH1A1-positive subpopulation of primary cells was also independently cultured in 96-well plates with sphere culture medium. The sphere formed after one week, and increased for another two weeks (Fig. 4A). To further determine whether gefitinib treatment augments the proportion of ALDH1A1-positive cells in lung cancer, spheres from primary cells of lung cancer patients were treated with gefitinib (0–100 nM) for 72 h, and the proportion of ALDH1A1-positive cells measured. Interestingly, the proportions of ALDH1A1-positive cells were significantly increased upon treatment with gefitinib at a concentration 100 nM (Fig. 4B). Consistently, studies on clinical samples disclosed that in lung cancer cells displaying resistance to EGFR-TKI and chemotherapy, the proportions of ALDH1A1-positive cells were significantly increased by 2- and 3-fold, respectively (Fig. 4C).

C.-P. Huang et al. / Cancer Letters 328 (2013) 144–151

149

Fig. 4. EGFR TKI and chemotherapy treatments enhance the ALDH1A1-positive cell population in a clinical lung cancer sample. A population of ALDH1A1-positive cells was identified in malignant pleural effusions (MPE) from treatment naïve lung cancer patient. Spheres formed after one week of culture, and the size and number of spheres increased after three weeks (A). Following gefitinib treatment, the ALDH1A1-positive population was enriched in a dose-dependent manner (B). MPE from EGFP TKI, chemotherapy and TKI + Chemo were enriched in ALDH1A1-positive cells to differential degrees, compared to nil (MPE from treatment naïve patients) (C).

4. Discussion Molecular-targeted therapy drugs can interfere with and block specific molecular pathways involved in cancer cell growth and progression. Activated EGFR, one of the major targets for cancer therapy, is effectively blocked by EGFR-TKI drugs [3,4]. However, development of resistance to EGFR-TKI drugs continues to critically limit the long-term control of cancer using this strategy [6,7]. Thus, the generation of more effective therapeutic interventions based on the molecular mechanisms underlying the development of tumor resistance to EGFR TKI drugs is an urgent requirement. Several investigations have shown that continued activation of the phosphatidylinositol 3 Kinase (PI3K)/Akt network is sufficient to confer resistance to EGFR TKI [7,22–24]. MET amplification leads

to gefitinib resistance in lung cancer, as it is mediated via activation of erbB3 signaling and restoration of the PI3K/Akt pathway [7]. Gefitinib-sensitive NSCLC cell lines specifically utilize erbB-3 to activate the PI3K/Akt pathway [26]. In fact, several models of acquired resistance demonstrate continued signaling along the PI3K pathway, despite TKI treatment [24–26]. The PI3K/Akt signaling pathways play key roles in widely divergent physiological processes, including cell cycle progression, differentiation, survival, motility, and autophagy [27]. Recent evidence has linked PI3K/ Akt signaling with CSC functions, both in solid and hematological tumors [28]. Additionally, epithelial-mesenchymal transition (EMT) has been recently shown to be involved in resistance to EGFR TKIs. Accumulating evidence supports the theory that ectopic expression of E-

150

C.-P. Huang et al. / Cancer Letters 328 (2013) 144–151

cadherin enhances gefitinib sensitivity in NSCLC cells, which possess a mesenchymal phenotype [29–31]. Silencing of the EMT regulator, Slug, in gefitinib-resistant cells can restore gefitinibinduced apoptosis [18]. The EMT program contributes to normal and transformed epithelial cells with stem cell properties, including the ability to self-renew and efficiently initiate tumors [32]. Moreover, TKI drugs, including imatinib, nilotinib, and dasatinib, show potency against differentiated chronic myeloid leukemia (CML) cells, but are not as effective for quiescent primitive CML stem cells [33–35]. These findings imply that cancer stem cells contribute to EGFR-TKI resistance in lung cancer. Resistance to radiation therapies and conventional chemotherapeutic drugs is a common characteristic of cancer stem cells [9,12,36,37]. However, EGFR-TKI drug resistance of cancer stem cells in lung cancer has not been described in the literature. Data from the current study showed that ALDH1A1-positive cancer stem cells promote EGFR-TKI resistance in lung cancer. Additional clinical studies revealed significantly increased proportions of ALDH1A1-positive cells in lung cancer cells displaying resistance to EGFR-TKI and chemotherapy drugs. Specific populations of cancer stem cells exist in many human tumors, including lung cancer. Subpopulations of the lung cancer-initiating cells have been identified, which exhibit increased tumorigenic potential and greater resistance to chemotherapeutic drugs [16,38,39]. ALDH1A1 is reported to participate in stem cell biology and anti-cancer chemotherapy drug resistance in lung cancer. Lung cancer subpopulations displaying high ALDH1A1 activity (ALDH1A1-positive cells) show greater tumorigenic potential than ALDH1A1-negative cells [16]. High expression of ALDH1A1 in lung cancer cells is positively associated with the stage and grade of tumors, and inversely related to patient survival. ALDH1A1 activity is therefore utilized as a stem cell marker, and a potential prognostic factor and therapeutic target for lung cancer [14,40]. Using this approach, we have established that ALDH1A1-positive lung cancer stem cells participate in the development of EGFR-TKI drug resistance. Cancer stem cells are involved in carcinogenesis, invasion and metastasis, and play a key role in chemo- and radiotherapy resistance [41–43]. However, the mechanisms underlying cancer stem cell contribution to TKI resistance in lung cancer remain unclear. Identification of the differential responses of cancer stem cells and other cells to EGFR-TKI and conventional chemotherapeutic drugs may provide a better insight into the effects of anti-cancer medications. In the current study, EGFR-TKI sensitivity was evaluated following isolation of ALDH1A1-positive and ALDH-negative cells derived from PC9 gefitinib-sensitive lung cancer cells. Our results indicate that the ALDH1A1-positive PC9 cells display significantly higher resistance to EGFR-TKI and conventional chemotherapeutic drugs, compared to ALDH1A1-negative PC9 cells. To further ascertain whether the cancer stem cell population is enriched in TKI drug-resistant lung cancer cells, the human lung cancer cell line model (including PC9 gefitinib-sensitive cells and PC9/ gef gefitinib-resistant cells) was employed. PC9/gef cells displayed greater tumor-forming capacity, resistance to EGFR-TKI, and putative stem cell-related gene expression, compared to PC9 cells. A 5to 6- fold enrichment of cancer stem cell populations was also detected in TKI drug-resistant PC9/gef cells, relative to PC9 cells. Thus, it appears that ALDH1A1-positive cell populations play a crucial role in gefitinib resistance in this cell model. ABCG2 is a member of the ATP-binding cassette transporter family and is known to contribute to multi-drug resistance in cancer chemotherapy [44,45]. Recently, studies have indicated that ABCG2 may be also involved in cancer cells TKI resistance [46,47]. We also detected the expression level of stem cell marker ABCG2 in ALDH1A1-positive and ALDH-negative PC9 subpopulations using real-time quantitative reverse transcriptase PCR. The results shown that the

ABCG2 expression level were no difference between ALDH1A1-positive and ALDH1A1-negative PC9 subpopulations, and the mechanisms of resistance to gefitinib in ALDH1A1-positive PC9 subpopulation were not related to elevated ABCG2 expression. However, we still cannot exclude the possibility that other members of ABC transporter family play a role in resistance to gefitinib of ALDH1A1-positive PC9 subpopulation. Accumulating evidence has shown that malignant pleural effusions (MPE) serve as a useful model to study hierarchical progression of cancer and/or intratumoral heterogeneity [23]. We collected and isolated primary cells from MPEs in lung cancer patients. Consistent with our hypothesis, clinical studies indicated that lung cancer patients receiving EGFR-TKI or chemotherapy drugs displayed significantly increased proportions of ALDH1A1positive cells. Higher proportions of ALDH1A1-positive cells were additionally detected in lung cancer patients administered a combination of EGFR-TKI and chemotherapy drugs. The repopulation ability is another stem cell characteristic in addition to the high tumorigenic ability. In this study, we also indicated that the ALDH1A1-positive PC9 CSCs regenerated both ALDH1A1-positive and ALDH1A1-negative subpopulation and tended to reconstruct ALDH1A1-negative subpopulation. The ALDH1A1-positive cell population constituted 6.5% and 37.8% in gefitinib-sensitive and gefitinib-resistant PC9 cells, respectively, a 6-fold difference (Fig. 3C). In clinical specimens study, the primary cells isolated from malignant pleural effusions of lung cancer patients were cultured in 10% FCS-supplemented DMEM medium for one week before the flow analysis of ALDH1A1 positivity. That is the reason why the proportion of ALDH1A1-positive cells from gefitinib-resistant patients was only 2-fold higher than those from treatment-naïve patients (Fig. 4C). In summary, we have provided evidence that ALDH1A1-positive lung cancer cells display significant resistance to EGFR-TKI (gefitinib) and anti-cancer chemotherapeutic drugs (cisplatin, etoposide and fluorouracil), compared to ALDH1A1-negative lung cancer cells. Our findings suggest that ALDH1A1-positive cancer stem cells promote both EGFR-TKI and chemotherapy resistance in lung cancer. Thus, strategies aimed at killing ALDH1A1-positive lung cancer stem cells may present a more reliable approach to overcome EGFR-TKI resistance and develop effective treatments for lung cancer.

Acknowledgments This study was supported from Grants 98-2628-B-002-087MY3, 98-2314-B-002-117-MY3 and 100-2320-B-002-077 (National Science Council, Taiwan), and Grants 99S-1316 and 100M1722 (National Taiwan University Hospital, Taiwan). We would like to acknowledge the service provided by the Cell Sorting Core Facility of the First Core Laboratory, National Taiwan University College of Medicine.

References [1] J.R. Molina, P. Yang, S.D. Cassivi, S.E. Schild, A.A. Adjei, Non-small cell lung cancer: epidemiology, risk factors, treatment, and survivorship, Mayo Clin. Proc. 83 (2008) 584–594. [2] P. Yang, Epidemiology of lung cancer prognosis: quantity and quality of life, Methods Mol. Biol. 471 (2009) 469–486. [3] T.J. Lynch, D.W. Bell, R. Sordella, S. Gurubhagavatula, R.A. Okimoto, B.W. Brannigan, P.L. Harris, S.M. Haserlat, J.G. Supko, F.G. Haluska, D.N. Louis, D.C. Christiani, J. Settleman, D.A. Haber, Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib, New Engl. J. Med. 350 (2004) 2129–2139. [4] J.G. Paez, P.A. Jänne, J.C. Lee, S. Tracy, H. Greulich, S. Gabriel, P. Herman, F.J. Kaye, N. Lindeman, T.J. Boggon, K. Naoki, H. Sasaki, Y. Fujii, M.J. Eck, W.R. Sellers, B.E. Johnson, M. Meyerson, EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy, Science 304 (2004) 1497–1500.

C.-P. Huang et al. / Cancer Letters 328 (2013) 144–151 [5] R.S. Herbst, M. Fukuoka, J. Baselga, Gefitinib–a novel targeted approach to treating cancer, Nat. Rev. Cancer 4 (2004) 956–965. [6] W. Pao, V.A. Miller, K.A. Politi, G.J. Riely, R. Somwar, M.F. Zakowski, M.G. Kris, H. Varmus, Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the egfr kinase domain, PLoS Med. 2 (2005) e73. [7] J.A. Engelman, K. Zejnullahu, T. Mitsudomi, Y. Song, C. Hyland, J.O. Park, N. Lindeman, C.M. Gale, X. Zhao, J. Christensen, T. Kosaka, A.J. Holmes, A.M. Rogers, F. Cappuzzo, T. Mok, C. Lee, B.E. Johnson, L.C. Cantley, P.A. Janne, Met amplification leads to gefitinib resistance in lung cancer by activating erbb3 signaling, Science 316 (2007) 1039–1043. [8] B.T. Tan, C.Y. Park, L.E. Ailles, I.L. Weissman, The cancer stem cell hypothesis: a work in progress, Lab. Invest. 86 (2006) 1203–1207. [9] B.B. Zhou, H. Zhang, M. Damelin, K.G. Geles, J.C. Grindley, P.B. Dirks, Tumourinitiating cells: challenges and opportunities for anticancer drug discovery, Nat. Rev. Drug Discov. 8 (2009) 806–823. [10] D. Bonnet, J.E. Dick, Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell, Nat. Med. 3 (1997) 730– 737. [11] C.A. O’Brien, A. Pollett, S. Gallinger, J.E. Dick, A human colon cancer cell capable of initiating tumour growth in immunodeficient mice, Nature 445 (2007) 106– 110. [12] M. Dean, T. Fojo, S. Bates, Tumour stem cells and drug resistance, Nat. Rev. Cancer 5 (2005) 275–284. [13] C. Ginestier, M.H. Hur, E. Charafe-Jauffret, F. Monville, J. Dutcher, M. Brown, J. Jacquemier, P. Viens, C.G. Kleer, S. Liu, A. Schott, D. Hayes, D. Birnbaum, M.S. Wicha, G. Dontu, ALDH1 is a marker of normal and malignant human mammary stem cells and a predictor of poor clinical outcome, Cell Stem Cell 1 (2007) 555–567. [14] Y. Su, Q. Qiu, Zhang, Z. Jiang, Q. Leng, Z. Liu, S.A. Stass, F. Jiang, Aldehyde dehydrogenase 1 A1-positive cell population is enriched in tumor-initiating cells and associated with progression of bladder cancer, Cancer Epidemiol. Biomarkers Prev. 19 (2010) 327–337. [15] M.R. Clay, M. Tabor, J.H. Owen, T.E. Carey, C.R. Bradford, G.T. Wolf, M.S. Wicha, M.E. Prince, Single-marker identification of head and neck squamous cell carcinoma cancer stem cells with aldehyde dehydrogenase, Head Neck. 32 (2010) 1195–1201. [16] F. Jiang, Q. Qiu, A. Khanna, N.W. Todd, J. Deepak, L. Xing, H. Wang, Z. Liu, Y. Su, S.A. Stass, R.L. Katz, Aldehyde dehydrogenase 1 is a tumor stem cell-associated marker in lung cancer, Mol. Cancer Res. 7 (2009) 330–338. [17] X. Li, L. Wan, J. Geng, C.L. Wu, X. Bai, Aldehyde Dehydrogenase 1A1 possesses stem-like properties and predicts lung cancer patient outcome. J. Thorac. Oncol., 2012, May 21 (Epub ahead of print). [18] T.H. Chang, M.F. Tsai, K.Y. Su, S.G. Wu, C.P. Huang, S.L. Yu, Y.L. Yu, C.C. Lan, C.H. Yang, S.B. Lin, C.P. Wu, J.Y. Shih, P.C. Yang, Slug confers resistance to the epidermal growth factor receptor tyrosine kinase inhibitor, Am. J. Respir. Crit. Care Med. 183 (2011) 1071–1079. [19] J.Y. Shih, M.F. Tsai, T.H. Chang, Y.L. Chang, A. Yuan, C.J. Yu, S.B. Lin, G.Y. Liou, M.L. Lee, J.J. Chen, T.M. Hong, S.C. Yang, J.L. Su, Y.C. Lee, P.C. Yang, Transcription repressor slug promotes carcinoma invasion and predicts outcome of patients with lung adenocarcinoma, Clin. Cancer Res. 11 (2005) 8070–8078. [20] T. Mitsudomi, T. Kosaka, H. Endoh, Y. Horio, T. Hida, S. Mori, S. Hatooka, M. Shinoda, T. Takahashi, Y. Tatabe, Mutations of the epidermal growth factor receptor gene predict prolonged survival after gefitinib treatment in patients with non-small-cell lung cancer with postoperative recurrence, J. Clin. Oncol. 23 (2005) 2513–2520. [21] S.G. Wu, Y.L. Chang, Y.C. Hsu, J.Y. Wu, C.H. Yang, C.J. Yu, M.F. Tsai, J.Y. Shih, P.C. Yang, Good response to gefitinib in lung adenocarcinoma of complex epidermal growth factor receptor (EGFR) mutations with the classical mutation pattern, Oncologist 13 (2008) 1276–1284. [22] T.F. Conway Jr, M.S. Sabel, M. Sugano, J.G. Frelinger, N.K. Egilmez, F. Chen, R.B. Bankert, Growth of human tumor xenografts in SCID mice quantified using an immunoassay for tumor marker protein in serum, J. Immunol. Methods 233 (2000) 57–65. [23] S.K. Basak, M.S. Veena, S. Oh, G. Huang, E. Srivatsan, M. Huang, S. Sharma, R.K. Batra, The malignant pleural effusion as a model to investigate.intratumoral.heterogeneity in lung cancer, PLoS One 4 (2009) e5884. [24] F. Yamasaki, M.J. Johansen, D. Zhang, S. Krishnamurthy, E. Felix, C. Bartholomeusz, R.J. Aguilar, K. Kurisu, G.B. Mills, G.N. Hortobagyi, N.T. Ueno, Acquired resistance to erlotinib in A-431 epidermoid cancer cells requires down-regulation of MMAC1/PTEN and up-regulation of phosphorylated Akt, Cancer Res. 67 (2007) 5779–5788. [25] R. Bianco, I. Shin, C.A. Ritter, F.M. Yakes, A. Basso, N. Rosen, J. Tsurutani, P.A. Dennis, G.B. Mills, C.L. Arteaga, Loss of PTEN/MMAC1/TEP in EGF receptorexpressing tumor cells counteracts the antitumor action of EGFR tyrosine kinase inhibitors, Oncogene 22 (2003) 2812–2822.

151

[26] J.A. Engelman, P.A. Jänne, C. Mermel, J. Pearlberg, T. Mukohara, C. Fleet, K. Cichowski, B.E. Johnson, L.C. Cantley, ErbB-3 mediates phosphoinositide 3kinase activity in gefitinib-sensitive non-small cell lung cancer cell lines, Proc. Natl. Acad. Sci. USA 102 (2005) 3788–3793. [27] T.L. Yuan, L.C. Cantley, PI3K pathway alterations in cancer: variations on a theme, Oncogene 27 (2008) 5497–5510. [28] A.M. Martelli, C. Evangelisti, F. Chiarini, C. Grimaldi, J.A. McCubrey, The emerging role of the phosphatidylinositol 3-kinase/Akt/mammalian target of rapamycin signaling network in cancer stem cell biology, Cancers 2 (2010) 1576–1596. [29] B.A. Frederick, B.A. Helfrich, C.D. Coldren, D. Zheng, D. Chan, P.A. Bunn Jr., D. Raben, Epithelial to mesenchymal transition predicts gefitinib resistance in cell lines of head and neck squamous cell carcinoma and non-small cell lung carcinoma, Mol Cancer Ther. 6 (2007) 1683–1691. [30] S. Thomson, E. Buck, F. Petti, G. Griffin, E. Brown, N. Ramnarine, K.K. Iwata, N. Gibson, J.D. Haley, Epithelial to mesenchymal transition is a determinant of sensitivity of non-small-cell lung carcinoma cell lines and xenografts to epidermal growth factor receptor inhibition, Cancer Res. 65 (2005) 9455– 9462. [31] R.L. Yauch, T. Januario, D.A. Eberhard, G. Cavet, W. Zhu, L. Fu, T.Q. Pham, R. Soriano, J. Stinson, S. Seshagiri, Z. Modrusan, C.Y. Lin, V. O’Neill, L.C. Amler, Epithelial versus mesenchymal phenotype determines in vitro sensitivity and predicts clinical activity of erlotinib in lung cancer patients, Clin Cancer Res 11 (2005) 8686–8698. [32] A. Singh, J. Settleman, EMT, cancer stem cells and drug resistance. an emerging axis of evil in the war on cancer, Oncogene 29 (2010) 4741–4751. [33] S.M. Graham, H.G. Jørgensen, E. Allan, C. Pearson, M.J. Alcorn, L. Richmond, T.L. Holyoake, Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro, Blood 99 (2002) 319–325. [34] H.G. Jørgensen, E.K. Allan, N.E. Jordanides, J.C. Mountford, T.L. Holyoake, Nilotinib exerts equipotent antiproliferative effects to imatinib and does not induce apoptosis in CD34+ CML cells, Blood 109 (2007) 4016–4019. [35] H. Konig, M. Copland, S. Chu, R. Jove, T.L. Holyoake, R. Bhatia, Effects of dasatinib on SRC kinase activity and downstream intracellular signaling in primitive chronic myelogenous leukemia hematopoietic cells, Cancer Res. 68 (2008) 9624–9633. [36] N.Y. Frank, T. Schatton, M.H. Frank, The therapeutic promise of the cancer stem cell concept, J. Clin. Invest. 120 (2010) 41–50. [37] J. Wang, T.P. Wakeman, J.D. Lathia, A.B. Hjelmeland, X.F. Wang, R.R. White, J.N. Rich, B.A. Sullenger, Notch promotes radioresistance of glioma stem cells, Stem Cells 28 (2010) 17–28. [38] E.L. Leung, R.R. Fiscus, J.W. Tung, V.P. Tin, L.C. Cheng, A.D. Sihoe, L.M. Fink, Y. Ma, M.P. Wong, Non-small cell lung cancer cells expressing CD44 are enriched for stem cell-like properties, PLoS One 5 (2010) e14062. [39] J.M. Sung, H.J. Cho, H. Yi, C.H. Lee, H.S. Kim, D.K. Kim, A.M. Abd El-Aty, J.S. Kim, C.P. Landowski, M.A. Hediger, H.C. Shin, Characterization of a stem cell population in lung cancer A549 cells, Biochem. Biophys. Res. Commun. 371 (2008) 163–167. [40] J.P. Sullivan, M. Spinola, M. Dodge, M.G. Raso, C. Behrens, B. Gao, K. Schuster, C. Shao, J.E. Larsen, L.A. Sullivan, S. Honorio, Y. Xie, P.P. Scaglioni, J.M. DiMaio, A.F. Gazdar, J.W. Shay, I.I. Wistuba, J.D. Minna, Aldehyde dehydrogenase activity selects for lung adenocarcinoma stem cells dependent on notch signaling, Cancer Res. 70 (2010) 9937–9948. [41] T. Borovski, F. De Sousa, E. Melo, L. Vermeulen, J.P. Medema, Cancer stem cell niche: the place to be, Cancer Res. 71 (2011) 634–639. [42] R. Perona, B.D. López-Ayllón, J. de Castro Carpeño, C. Belda-Iniesta, A role for cancer stem cells in drug resistance and metastasis in non-small-cell lung cancer, Clin. Transl. Oncol. 13 (2011) 289–293. [43] J.P. Sheehan, M.E. Shaffrey, B. Gupta, J. Larner, J.N. Rich, D.M. Park, Improving the radiosensitivity of radioresistant and hypoxic glioblastoma, Future Oncol. 6 (2010) 1591–1601. [44] M.T. Kuo, Roles of multidrug resistance genes in breast cancer chemoresistance, Adv. Exp. Med. Biol. 608 (2007) 23–30. [45] K. Takara, T. Sakaeda, K. Okumura, An update on overcoming MDR1-mediated multidrug resistance in cancer chemotherapy, Curr. Pharm. Des. 12 (2006) 273–286. [46] C. Ozvegy-Laczka, T. Hegedus, G. Várady, O. Ujhelly, J.D. Schuetz, A. Váradi, G. Kéri, L. Orfi, K. Német, B. Sarkadi, High-affinity interaction of tyrosine kinase inhibitors with the ABCG2 multidrug transporter, Mol. Pharmacol. 65 (2004) 1485–1495. [47] Y.J. Chen, W.C. Huang, Y.L. Wei, S.C. Hsu, P. Yuan, H.Y. Lin, I.I. Wistuba, J.J. Lee, C.J. Yen, W.C. Su, K.Y. Chang, W.C. Chang, T.C. Chou, C.K. Chou, C.H. Tsai, M.C. Hung, Elevated BCRP/ABCG2 expression confers acquired resistance to gefitinib in wild-type EGFR-expressing cells, PLoS One 6 (2011) e21428.