Cucurbitacin I inhibits tumorigenic ability and enhances radiochemosensitivity in nonsmall cell lung cancer-derived CD133-positive cells

Authors

  • Han-Shui Hsu MD, PhD,

    Corresponding author
    1. Institute of Emergency and Critical Care Medicine, National Yang-Ming University, Taipei, Taiwan, Republic of China
    2. Department of Surgery, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China
    • Institute of Emergency and Critical Care Medicine and Institute of Clinical Medicine, National Yang-Ming University School of Medicine, No. 155, Sec. 2, Li-Nong Street, Taipei, Taiwan
    Search for more papers by this author
    • Fax: (011) 886-2-28746193

  • Pin-I Huang MD,

    1. Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan Republic of China
    2. Cancer Center, Taipei Veterans General Hospital, Taipei, Taiwan
    Search for more papers by this author
  • Yuh-Lih Chang PhD,

    1. Institute of Pharmacology, National Yang-Ming University, Taipei, Taiwan, Republic of China
    2. Department of Education and Research, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China
    Search for more papers by this author
  • Ching Tzao MD, PhD,

    1. Division of Thoracic Surgery, Tri-Service General Hospital, National Defense Medical Center; Taipei, Taiwan, Republic of China
    Search for more papers by this author
  • Yi-Wei Chen MD,

    1. Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan Republic of China
    2. Cancer Center, Taipei Veterans General Hospital, Taipei, Taiwan
    Search for more papers by this author
  • Hsin-Chin Shih MD, PhD,

    1. Institute of Emergency and Critical Care Medicine, National Yang-Ming University, Taipei, Taiwan, Republic of China
    2. Department of Emergency Medicine, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China
    Search for more papers by this author
  • Shih-Chieh Hung MD, PhD,

    1. Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan Republic of China
    2. Institute of Pharmacology, National Yang-Ming University, Taipei, Taiwan, Republic of China
    3. Department of Education and Research, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China
    Search for more papers by this author
  • Yu-Chih Chen PhD,

    1. Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan Republic of China
    2. Department of Education and Research, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China
    Search for more papers by this author
  • Ling-Ming Tseng MD,

    1. Department of Surgery, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China
    2. Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan Republic of China
    Search for more papers by this author
  • Shih-Hwa Chiou MD, PhD

    Corresponding author
    1. Institute of Clinical Medicine, National Yang-Ming University, Taipei, Taiwan Republic of China
    2. Institute of Pharmacology, National Yang-Ming University, Taipei, Taiwan, Republic of China
    3. Department of Education and Research, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China
    • Institute of Emergency and Critical Care Medicine and Institute of Clinical Medicine, National Yang-Ming University School of Medicine, No. 155, Sec. 2, Li-Nong Street, Taipei, Taiwan
    Search for more papers by this author

Abstract

BACKGROUND:

Signal transducer and activator of transcription 3 (STAT3) signaling reportedly promotes tumor malignancy and recurrence in nonsmall cell lung cancer (NSCLC). It was demonstrated previously that the STAT3 pathway maintains the tumorigenicity and therapeutic resistance of malignant tumors as well as cancer stem cells (CSCs). The objective of the current study was to investigate the effect of the strong STAT3 inhibitor, cucurbitacin I, in prominin-1 (CD133)-positive lung cancer cells.

METHODS:

CD133-positive and CD133-negative NSCLC-derived cells were isolated from 7 patients with NSCLC. CD133-positive NSCLC cells that were treated with or without cucurbitacin I were evaluated for their expression of phosphorylated STAT3 (p-STAT3), tumorigenicity, stemness properties, and resistance to chemotherapeutic drugs and ionizing radiation.

RESULTS:

Compared with parental or CD133-negative NSCLC cells, CD133-positive NSCLC cells had greater tumorigenicity, greater radioresistance, and higher expression of octamer-binding transcription factor 4 (Oct-4), Nanog homeobox, and sex-determining region Y, box 2 (Sox2) at high p-STAT3 levels. Cucurbitacin I treatment at 100 nM effectively abrogated STAT3 activation, tumorigenic capacity, sphere formation ability, radioresistance, and chemoresistance in CD133-positive NSCLC cells. Microarray data suggested that cucurbitacin I inhibited the stemness gene signature of CD133-positive NSCLC cells and facilitated the differentiation of CD133-positive NSCLC cells into CD133-negative NSCLC cells. It is noteworthy that 150 nM cucurbitacin I effectively blocked STAT3 signaling and downstream survival targets, such as B-cell chronic lymphocytic leukemia/lymphoma 2 (Bcl-2) and Bcl-2-like 1 (Bcl-xL) expression and induced apoptosis in CD133-positive NSCLC cells. Finally, xenotransplantation experiments revealed that cucurbitacin I plus radiotherapy or chemotherapeutic drugs significantly suppressed tumorigenesis and improved survival in NSCLC-CD133-positive-transplanted, immunocompromised mice.

CONCLUSIONS:

Targeting STAT3 signaling in CD133-positive NSCLC cells with cucurbitacin I suppressed CSC-like properties and enhanced chemoradiotherapy response. The potential of cucurbitacin I should be verified further in future anti-CSC therapy. Cancer 2011. © 2011 American Cancer Society.

Lung cancer is 1 of the leading causes of cancer-related deaths worldwide and has high incidence and mortality.1, 2 Radiotherapy and chemotherapy play significant and crucial roles in the clinical treatment of lung cancer to achieve prolonged survival.3, 4 However, a high failure rate and a short median survival are observed in patients with recurrent and intractable lung cancer who receive chemotherapy or radiotherapy.5 Therefore, novel therapies are needed urgently to overcome resistance. Prominin-1 (CD133; PROM1), a 5-transmembrane glycoprotein, recently was identified as an important marker of a subpopulation of cancer stem-like cells (CSCs) in leukemia, brain tumor, retinoblastoma, colon cancer, prostate carcinoma, and hepatocellular carcinoma.6-10 Eramo et al first demonstrated that CD133-positive lung cancer cells have the ability to self-renew and grow indefinitely as a tumor sphere.11 In another study of small cell lung cancer (SCLC), Jiang et al demonstrated that SCLC cells with high CD133 levels had markedly enhanced tumorigenicity associated with the basic helix-loop-helix transcription factor achaete-scute complex homolog 1.12 In our previous study, we demonstrated that nonsmall cell lung cancer (NSCLC)-derived CD133-positive cells displayed higher octamer-binding transcription factor 4 (Oct-4) expression and possessed the ability to self-renew, potentially representing a reservoir of proliferative potential to generate lung cancer cells.13 Consistent with our findings, Bertolini et al also provided evidence that CD133-positive lung tumor cells possessed stemness features and a higher tumorigenic potential than CD133-negative cells in severe combined immunodeficient (SCID) mice.14 A growing body of evidence indicates the tumor initiation ability and chemoradioresistance of CD133-positive cells in lung CSCs. However, the functional significance of the biomolecular pathways that are overexpressed in lung CSCs remains unclear and needs further clarification.

Signal transducer and activator of transcription (STAT), a transcription factor for cytokine signaling, is constitutively activated in numerous cancer types, including prostate cancer, breast cancer, leukemia, multiple myeloma, nasopharyngeal carcinoma, brain tumors, and lung cancer.15-18 Overexpression of activated STAT3 has been detected immunohistochemically in NSCLC specimens.19 Tissue microarray results also indicted that phosphorylated STAT3 (p-STAT3) expression was correlated significantly with vascular endothelial growth factor (VEGF) receptor 1 (VEGFR-1) expression, implying the clinical significance of activated STAT3 in the tumor progression and angiogenesis of lung adenocarcinoma.20 Bromberg et al reported that STAT3 mutations induced cellular transformation and tumor formation in vivo and that activation of STAT3 signaling further inhibited p53 transcriptional activity, fulfilling the definition of an oncogene.21, 22 Oncogenic STAT3 activation leads to the increased expression of downstream genes that suppress apoptosis (B-cell chronic lymphocytic leukemia/lymphoma 2 [Bcl-2]-like 1 [Bcl-xL]), regulate cell cycle progression (p21, c-Myc, and cyclin D1), mediate cellular invasion (matrix metalloproteinase 9 [MMP-9]), and modulate angiogenesis (VEGF).23 However, the mechanism of the STAT3 signaling pathways and the possible therapeutic targets involved in lung CSCs need further investigation.

Cucurbitacin I (also known as JSI-124), a natural cell-permeable triterpenoid compound, belongs to the cucurbitacin family of drugs, which are isolated from various plant families, and other members include cucurbitaceae and cruciferae. Cucurbitacins have been used as folk medicine for centuries because of their anti-inflammatory and analgesic effects. Recent studies have reported that cucurbitacin I potently inhibits cell growth through the selective repression of tyrosine phosphorylation of STAT3 (p-STAT3) in various human cancer cell lines, including lung cancer.24 In view of these findings, we investigated the clinical significance of p-STAT3 in CD133-positive cells that were isolated from patients with NSCLC. We also evaluated whether blocking STAT3 signaling with cucurbitacin I could improve the prognosis for patients with NSCLC by overcoming resistance to radiation and chemotherapeutic drugs. Our results demonstrated that cucurbitacin I inhibited the CSC-like properties and enhanced chemoradiosensitivity in CD133-positive NSCLC cells in vivo.

MATERIALS AND METHODS

Isolation of CD133-Positive Cells

This research followed the tenets of the Declaration of Helsinki, and all samples were obtained after patients provided informed consent. This study was approved by the Institutional Ethics Committee/Institutional Review Board of Taipei Veterans General Hospital. The cells from samples that were dissociated from the patients with NSCLC were labeled with 1 mL CD133/L micromagnetic beads per 1 million cells using the CD133 cell isolation kit (Miltenyi Biotech, Auburn, Calif). Because serum-free medium culture allowed the selection of undifferentiated colon, glioblastoma, or lung cancer stem and progenitor cells,8, 11, 25 CD133-positive cells were cultured in a medium consisting of serum-free Dulbecco modified Eagle medium (DMEM)/F-12 medium (Gibco-BRL, Gaithersburg, Md), N2 supplement (R&D Systems Inc., Minneapolis, Minn), 10 ng/mL human recombinant basic fibroblast growth factor (bFGF0 (R&D Systems Inc.), and 10 ng/mL epidermal growth factor (EGF) (R&D Systems Inc.). To exclude potential contamination by hematopoietic and endothelial cells, samples were examined using antiepithelial-specific antigen fluorescein isothiocyanate (Biomeda). Then, the samples were assessed by flow cytometry for purity.14 The medium was replaced or supplemented with fresh growth factors twice weekly until the cells started to grow and form floating aggregates. Floating sphere cultures were expanded weekly by mechanical dissociation followed by replating of single cells in complete fresh medium.

Irradiation and Clonogenic Assay

Ionizing radiation (IR) was delivered with a cobalt unit (Theratronic International, Inc., Ottawa, Ontario, Canada) at a dose rate of 1.1 grays (Gy) per minute (source-to-surface distance = 57.5 cm). For a clonogenic assay, cells were exposed to different radiation doses (0 Gy, 2 Gy, 4 Gy, 6 Gy, 8 Gy, and 10 Gy). After incubation for 10 days, colonies (> 50 cells per colony) were fixed and stained for 20 minutes with a solution that contained crystal violet and methanol. Cell survival was determined by using a colony-formation assay. The plating efficiency (PE) and survival fraction (SF) were calculated as follows: PE = (colony number/number of inoculated cells) × 100%. SF = colonies counted/(cells seeded × [PE/100]).

Microarray Analysis and Bioinformatics

Total RNA was extracted from cells using the Trizol reagent (Life Technologies, Bethesda, Md), and the Qiagen RNAeasy column (Qiagen, Valencia, Calif) was used for purification. Total RNA was reverse-transcribed with Superscript II RNase H-reverse transcriptase (Gibco-BRL) to generate indocarbocyanine (Cy3)-labeled and indodicarbocyanine (Cy5)-labeled (Amersham Biosciences Co., Piscataway, NJ) combinational DNA (cDNA) probes for the control and treated samples, respectively. The labeled probes were hybridized to a cDNA microarray that contained 10,000 immobilized gene clone cDNA fragments. Fluorescence intensities of Cy3 and Cy5 targets were measured and scanned separately using a GenePix 4000 B Array Scanner (Axon Instruments, Burlingame, Calif). Data analysis was performed using GenePix Pro 3.0.5.56 (Axon Instruments) and GeneSpring GX 7.3.1 software (Agilent Technologies, Palo Alto, Calif). The average linkage distance was used to assess the similarity between the 2 groups of gene expression profiles, as described below. The difference in distance between the 2 groups of sample expression profiles and a third group was assessed by comparing the corresponding average linkage distances (the mean of all pair-wise distances of linkages between the members of the 2 groups). The error of this comparison was estimated by combining the standard errors (the standard deviation of pair-wise linkages divided by the square root of the number of linkages) of the average linkage distances involved. Classic multidimensional scaling was performed using the standard function in the R program (R Development Core Team, Vienna, Austria) to provide a visual impression of how the various sample groups were related.26

Western Blot Assay

The extraction of proteins from cells and Western blot analyses were performed as described previously.27 Fifteen microliters of sample were boiled at 95°C for 5 minutes and separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The proteins were transferred to Hybond-enhanced chemiluminescence (ECL) nitrocellulose paper (Amersham, Arlington Heights, Ill) by using a wet-transfer system. The following primary antibodies were used: rabbit antihuman phospho-STAT3, rabbit antihuman STAT3, rabbit antihuman cleaved poly(adenosine diphosphate-ribose) polymerase, rabbit antihuman cleaved caspase 3, rabbit antihuman CD133, and mouse antihuman survivin (Cell Signaling Technology, Danvers, Mass); rabbit antihuman LC3 (Novus Biologicals, Cambridge, United Kingdom); rabbit antihuman p21 (Santa Cruz Biotechnology, Santa Cruz, Calif); mouse antihuman Bcl-2 and rabbit antihuman Bcl2-associated X protein (Bax) (Upstate Biotechnology, Lake Placid, NY); and mouse anti-beta actin (Chemicon, Temecula, Calif). The reactive protein bands were detected by the ECL detection system (Amersham).

Enzyme-Linked Immunosorbent Assay, Immunofluorescence Staining, and Terminal 2′-Deoxyuridine, 5′-Triphosphate Nick-End Labeling

The activity of caspase-3 was determined using an enzyme-linked immunosorbent assay kit (R&D Systems Inc.) and was quantified at 490 nm (MRX microplate reader; Dynatech Laboratories, Chantilly, Va). Each individual sample was analyzed in triplicate. The protocol for immunofluorescence staining has been described previously.27 Briefly, an avidin-biotin complex method was used for immunofluorescence staining in the spheroid cells. Each slide was treated with antibodies for p-STAT3 and antihuman survivin (Cell Signaling Technology). Immunoreactive signals were detected with a mixture of biotinylated rabbit-antimouse immunoglobulin G and Fluoesave (Calbiochem, La Jolla, Calif). Positive cells were counted in 6 different fields by microscopy. Furthermore, apoptotic cells were identified by using the terminal 2′-deoxyuridine, 5′-triphosphate nick-end labeling (TUNEL) method (In Situ Cell Death Detection Kit, POD; Roche Boehringer Mannheim Corp., Indianapolis, Ind).

In Vitro Cell Invasion Analysis and Soft Agar Assay

The 24-well plate Transwell system with a polycarbonate filter membrane was used (8 μm pore size; Corning, Flintshire, United Kingdom). Cell suspensions were seeded in the upper compartment of the Transwell chamber at a density of 1 × 105 cells in 100 μL of serum-free medium. The opposite surface of the filter membrane facing the lower chamber was stained with Hoechst 33342 for 3 minutes, and migrating cells were observed under an inverted microscope. For the soft agar assay, the bottom of each well (35 mm) in a 6-well culture dish was coated with 2 mL of an agar mixture (10% DMEM [volume/volume], 0.6% fetal calf serum [weight/volume], and agar). After the bottom layer solidified, 2 mL of a top agar-medium mixture (10% DMEM [volume/volume], 0.3% fetal calf serum [weight/volume], and agar) that contained 2 × 104 cells were added, and the plates were incubated at 37°C for 4 weeks. The plates were stained with 0.5 mL 0.005% crystal violet, and the number of colonies was counted using a dissecting microscope.

In Vivo Analysis of Tumor Growth and Metastasis

All procedures that involved animals were conducted in accordance with the institutional animal welfare guidelines of Taipei Veterans General Hospital. In total, 105 CD133-positive and CD133-negative NSCLC cells were injected into the subcutaneous sites of nude mice (BALB/c strain) aged 8 weeks. Cucurbitacin I (1 mg/kg daily for 5 days) was administrated in xenotransplanted nude mice by intraperitoneal injection. In vivo green fluorescent protein (GFP) imaging was performed using an illuminating device (LT-9500 Illumatool TLS; Lightools Research, Encinitas, Calif) equipped with an excitation illuminating source (470 nm) and a filter plate (515 nm). Tumors were localized and dissected with the aid of GFP imaging. Tumor size was measured using calipers, and the volume was calculated according to the following formula: (length × width2)/2. The integrated optical density of green fluorescence intensity was captured and subsequently analyzed using Image Pro+ software (Media Cybernetics, Silver Spring, Md).27

Statistical Analysis

The results are reported as mean ± standard deviation values. Statistical analysis was performed using the Student t test or a 1-way or 2-way analysis of variance followed by the Turkey test, as appropriate. Survival was estimated by using the Kaplan-Meier method and was compared between groups with the log-rank test. P values < .05 were considered statistically significant.

RESULTS

Isolation and Characterization of Lung Cancer-Derived CD133-Oositive Cells

Recent studies have indicated that the expression of CD133 in lung cancer results in high tumorigenicity and a phenotype that is resistant to cytotoxic therapy.14, 28 Our previous results demonstrated that CD133-positive cells isolated from patients with NSCLC exhibit greater chemoradioresistance compared with NSCLC cells of the CD133-negative lineage.13 To further explore other possible characteristics of CSCs in CD133-positive cells from patients with lung cancer, we isolated CD133-positive cells (Fig. 1A) in tissue samples from 7 patients with NSCLC using the magnetic bead method (Table 1). The CD133-positive NSCLC cells that were isolated from these 7 patients (Table 1) were able to form floating spheroid-like bodies in DMEM/F-12 serum-free medium with bFGF and EGF (Fig. 1A). Quantitative reverse transcriptase-polymerase chain reaction results demonstrated that the amounts of stemness gene transcripts (Oct-4, sex-determining region Y, box 2 [Sox-2], and Nanog homeobox) and drug resistant gene transcripts (multidrug resistance protein 1 [MDR-1], ATP-binding cassette subfamily G, member 2 [ABCG2]) in CD133-positive NSCLC cells were higher than those in CD133-negative NSCLC cells (Fig. 1B). In those 7 patients, Patients 1, 2, and 3 with highest percentages of CD133-positive cells (10.4%, 8.3%, and 5.6%, respectively) and the most robust in vivo tumorigenicity were chosen for the experiments described below. The ability to form spheroid-like bodies was significantly higher in CD133-positive NSCLC cells than in CD133-negative NSCLC cells (P < .05) (Fig. 1C). To evaluate the enhancement of tumorigenicity of CD133-positive NSCLC cells, we performed Matrigel/Transwell invasion and soft agar colony formation assays. CD133-positive NSCLC cells derived from Patients 1, 2, and 3 displayed higher invasion activity and enhanced foci formation ability compared with CD133-negative NSCLC cells from the same 3 patients (P < .001) (Fig. 1C). Moreover, xenotransplantation of 104 CD133-negative cells into SCID mice did not lead to tumor formation; however, transplantation of 103 CD133-positive cells from all 7 NSCLC patients generated visible tumors 8 weeks after injection (Table 1). Next, we evaluated the multidrug chemotherapy-resistant abilities of CD133-positive NSCLC cells and CD133-negative NSCLC cells. Compared with CD133-negative cells, CD133-positive cells from Patients 1, 2, and 3 were significantly resistant to cisplatin, doxorubicin, and taxol (P < .01) (Fig. 1D).

Figure 1.

The isolation and characterization of prominin-1 (CD133)-positive (CD133+) and CD133-negative (CD133−) cells from nonsmall cell lung cancer (NSCLC) tissues are illustrated. (A) By using a magnetic bead method, CD133+ cells were isolated from 7 patients with NSCLC and were identified by flow cytometry. Scale bar-100 μm. (B) Messenger RNA (mRNA) expression levels of octamer-binding transcription factor 4 (Oct-4); sex-determining region Y, box 2 (Sox-2); Nanog homeobox; MDR-1, multidrug resistance protein 1 (adenosine triphosphate [ATP]-binding cassette, subfamily B); and ATP-binding cassette, subfamily G, member 2 (ABCG2) in NSCLC-CD133+ and NSCLC-CD133− cells were determined by quantitative reverse transcriptase-polymerase chain reaction analysis. (C) Left: This chart illustrates an evaluation of the ability to form spheroid-like bodies (SB) by various cell groups in serum-free medium with basic fibroblast factor and epidermal growth factor Middle: Tumor foci (soft agar colony) formation by NSCLC-CD133+ cells was increased significantly compared with NSCLC-CD133− cells from Patients 1, 2 and 3 (P > .001). Right: Migration/invasion capabilities of NSCLC-CD133+ cells were increased significantly compared with NSCLC-CD133− cells from Patient 1, 2, and 3 (P > .001). (D) NSCLC-CD133-positive or NSCLC-CD133− cells (103) from Patients 1, 2, and 3 were plated in a 96-well plate and treated with various concentrations of (Left) cisplatin, (Middle) doxorubicin, and (Right) paclitaxel for 24 hours in 10% fetal bovine serum/Dulbecco modified Eagle medium/F-12 medium. The survival rate was determined by using a 3(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Data shown are the mean ± standard deviation values from 3 experiments.

Table 1. Patient Characteristics, Tumor Characteristics, and Treatment Effects of Prominin-1 (CD133)-Negative and CD133-Positive Nonsmall Cell Lung Cancer
       No. of Cells Injected (No./Total No.)a
Patient No.Age, ySexStagePathologyCD133- Positive, %Sphere FormationParentalCD133- NegativeCD133- Positive
  • a

    Cells were injected subcutaneously into nude mice.

175ManIIA, T2bN0M0Adenocarcinoma10.4Yes1000 (0/3)1000 (0/3)1000 (3/3)
       3000 (0/3)3000 (0/3)3000 (3/3)
       10,000 (2/3)10,000 (0/3)10,000 (3/3)
268ManIIB, T3N0M0Adenocarcinoma8.3Yes1000 (0/3)1000 (0/3)1000 (3/3)
       3000 (1/3)3000 (0/3)3000 (3/3)
       10,000 (1/3)10,000 (0/3)10,000 (3/3)
358ManIIIA, T3N1M0Adenocarcinoma5.6Yes1000 (0/3)1000 (0/3)1000 (3/3)
       3000 (0/3)3000 (0/3)3000 (3/3)
       10,000 (1/3)10,000 (1/3)10,000 (3/3)
471WomanIIA, T2aN1M0Adenocarcinoma3.2Yes1000 (0/3)1000 (0/3)1000 (2/3)
       3000 (0/3)3000 (0/3)3000 (1/3)
       10,000 (0/3)10,000 (0/3)10,000 (3/3)
559ManIIIA, T3N2M0Squamous cell carcinoma3.1Yes1000 (0/3)1000 (0/3)1000 (3/3)
       3000 (0/3)3000 (0/3)3000 (3/3)
       10,000 (0/3)10,000 (0/3)10,000 (3/3)
668ManIIIB, T4N2M0Squamous cell carcinoma1.5Yes1000 (0/3)1000 (0/3)1000 (1/3)
       3000 (0/3)3000 (0/3)3000 (2/3)
       10,000 (0/3)10,000 (0/3)10,000 (3/3)
752ManIIB, T2bN1M0Adenocarcinoma0.7Yes1000 (0/3)1000 (0/3)1000 (0/3)
       3000 (0/3)3000 (0/3)3000 (2/3)
       10,000 (0/3)10,000 (0/3)10,000 (2/3)

To further determine the effect of radiation on the tumor growth rate, we used IR doses from 0 Gy to 10 Gy to treat parental CD133-positive and CD133-negative NSCLC cells from Patients 1, 2, and 3. After IR treatment (Fig. 2A), both the survival rate and the number of CD133-positive NSCLC cells were significantly higher than those in parental and CD133-negative NSCLC cells (P < .001). These data demonstrated that CD133-positive NSCLC cells have a higher degree of radioresistance (Fig. 2A). Because several studies have demonstrated that STAT3 activation is related to the malignancy of lung cancer, the correlation of activation status of STAT3 in CD133-negative and CD133-positive NSCLC cells was evaluated further. Immunofluorescence staining demonstrated that CD133-positive NSCLC cells expressed an elevated level of activated STAT3 (p-STAT3-tyrosine 705 [Tyr705]) compared with CD133-negative NSCLC cells or parental cells in Patients 1, 2, and 3 (Fig. 2B). Western blot results also revealed that the level of activated STAT3 (p-STAT3-Tyr705) was higher in CD133-positive NSCLC cells than in CD133-negative NSCLC cells or parental cells from the 3 patients (Fig. 2C). Taken together, these results indicate that CD133-positive cells isolated from patients with NSCLC share the characteristics and properties of CSCs, including tumor-initiating ability and chemoradioresistance as well as increased expression of stemness genes and p-STAT3 compared with CD133-negative and parental cells.

Figure 2.

The detection of phosphorylated signal transducer and activator of transcription 3 (p-STAT3) expression in prominin-1 (CD133)-positive (CD133+) and CD133-negative (CD133−) cells from nonsmall cell lung cancer (NSCLC) tissues is illustrated. (A) To determine the effect of radiation on the tumor growth rate, an ionizing radiation (IR) dose from 0 to 10 grays (Gy) was used to treat NSCLC-CD133+ and NSCLC-CD133− cells from (Left) Patient 1, (Middle) Patient 2, and (Right) Patient 3 (P < .01: NSCLC-CD133+ cells vs NSCLC-CD133− cells). (B) Immunofluorescence staining was used to detect CD133+ (green) and p-STAT3-tyrosine 705 (Tyr705)-positive (red) cells in tumor spheres. DAPI indicates 4′,6-diamidino-2-phenylindole. Scale bars = 100 μm. (C) The results from Western blot analysis show the protein levels of p-STAT3-Tyr705 and total STAT3 in parental cells, NSCLC-CD133+ cells, and NSCLC-CD133− cells from the same 3 patients.

Cucurbitacin I Suppresses Proliferation by Blocking STAT3 Signaling in Lung Cancer

Cucurbitacin I is a selective Janus kinase (JAK)-STAT inhibitor that blocks the tyrosine phosphorylation of STAT3 and JAK2 but not other oncogenic or survival pathways, such as the protein kinase B (Akt), extracellular signal-regulated kinase (ERK), and c-Jun N-terminal kinase (JNK) signaling pathways, in various cancers.24 However, it remains unknown whether cucurbitacin I can inhibit the CSC properties of CD133-positive NSCLC. The viability of CD133-positive NSCLC determined by a 3(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay decreased significantly as the concentrations of cucurbitacin I increased (P < .05) (Fig. 3A). Treatment with cucurbitacin I also interfered significantly with the colony-formation (Fig. 3B) and migration (Fig. 3B) abilities of CD133-positive NSCLC. Next, we analyzed the genomic traits of CD133-positive NSCLC that was treated with or without cucurbitacin I using gene expression microarray analysis.26 The expression pattern of CD133-positive NSCLC more closely resembled the pattern observed in high-grade NSCLC tissues than the patterns observed in low-grade NSCLC or normal lung tissues (Fig. 3C,D); whereas CD133-negative NSCLC cells, cucurbitacin I-treated CD133-positive NSCLC cells, and low-grade NSCLC had profiles that more closely resembled the profile observed in normal lung tissue (Fig. 4A). Notably, microarray analysis indicated that the expression of 1132 probe sets was altered significantly in the cucurbitacin I-treated group compared with the control group when the hierarchical clustering method was used (Fig. 3C). In addition, we used a literature-based network analysis of all MEDLINE records (titles and abstracts) and the Cytoscape open-source bioinformatics software platform to group the target-linkage genes; and the results indicated that interleukins, STAT3, and STAT3-related pathways are involved in CD133-positive NSCLC compared with CD133-negative NSCLC (Fig. 3D). To validate the microarray findings, the effects of cucurbitacin I on STAT3 downstream genes and cell survival-related genes, such as survivin, Bcl-2, Bcl-xL, and Bax, were examined by Western blot analysis. The protein expression of Bcl-2, Bcl-xL, and survivin was down-regulated; whereas Bax production was enhanced by cucurbitacin I treatment (Fig. 3E).

Figure 3.

Cucurbitacin I inhibited cell proliferation of prominin-1 (CD133)-positive (CD133+) nonsmall cell lung cancer (NSCLC) cells. (A) NSCLC-CD133+ cells from Patients 1, 2, and 3 were plated into 24-well plates and incubated for 48 hours with various concentrations of cucurbitacin I. At the end of the treatment, cell viability was determined by using a 3(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. (B) The treatment of NSCLC-CD133+ cells with cucurbitacin I impeded the capability of colony formation and migration. (C) A microarray analysis of cucurbitacin I-treated (Cu-treated) NSCLC-CD133+ cells is shown. This hierarchy heat map depicts the gene expression microarray analysis results (gene tree) of the 987 genes that were expressed differentially in cucurbitacin I-treated NSCLC-CD133+ cells compared with control cells. The time-dependent changes in expression of these 987 genes are presented as a log scale of the expression values provided by GeneSpring GX software (Agilent Technologies, Santa Clara, Calif) Lung ca indicates lung cancer; local, local disease; meta, metastasis. (D) The literature networks are based on the results from indexing using a Natural Language Processing (NLP) regimen of all MEDLINE records (titles and abstracts) for gene and protein names. Lines indicate cocitation in the literature in more than 1 article. The numbers indicate the number of Medline records that contained the query term or 1 of its synonyms at least once. STAT3 indicates signal transducer and activator of transcription 3; JAK2, Janus kinase 2; EPOR, erythropoietin receptor; EPO, erythropoietin; LRPRC, leucine-rich pentatricopeptide repeat-motif containing; TNF, tumor necrosis factor; IL3, interleukin 3; IL6ST, interleukin 6 signal transducer; NM, neutrophil migration; JUN, jun proto-oncogene; FAM48A, family with sequence similarity 48/member A. (E) Phosphorylated STAT3 (p-Stat3); STAT3; survivin; B-cell chronic lymphocytic leukemia/lymphoma 2 (Bcl-2); BCL2-like 1 (Bcl-xL); and Bcl2-associated X protein (Bax) were detected in cucurbitacin I-treated NSCLC-CD133+ cells by Western blot analysis.

Figure 4.

Cucurbitacin I promoted differentiation, increased radiosensitivity, and enhanced ionizing radiation (IR)-induced apoptosis in prominin-1 (CD133)-positive (CD133+) nonsmall cell lung cancer (NSCLC) cells. (A) Principle component analysis demonstrated that cucurbitacin I (Cu.) treatment in NSCLC-CD133+ cells could decrease the gene signature of stem cells. Lung ca. indicates lung cancer; Meta, metastatic disease; local, local disease. (B) Fluorescence-activated cell sorting analysis demonstrated that the proportion of CD133-positive cells were reduced in a dose-dependent manner after cucurbitacin I treatment. (C,D) Cells were treated with cucurbitacin I. Immunofluorescence staining was used to evaluate (C) cleaved caspase 3-positive cells and (D) terminal 2′-deoxyuridine, 5′-triphosphate nick-end labeling (TUNL)-positive cells. Gy indicates grays. (E) To determine the effect of radiation on the tumor growth rate, IR doses from 0 to 10 Gy were used to treat NSCLC-CD133+ cells in combination with vehicle or cucurbitacin I. (F) The colony-formation and the invasion abilities of NSCLC parental and NSCLC-CD133+ cells were examined after treatment with cucurbitacin I, 4 Gy IR, or both. Data shown are the mean ± standard deviation values from 3 experiments.

Cucurbitacin I Promotes Differentiation and Improves Sensitivity to Radiotherapy and Chemotherapeutic Drugs by Blocking STAT3 Signaling in NSCLC

Multidimensional scaling and principle component analysis further indicated that CD133-positive NSCLC more closely resembled high-grade NSCLC tissues than low-grade NSCLC tissues or normal lung tissues (Fig. 4A). In contrast, the multidimensional scaling results indicted that the expression patterns observed in CD133-negative NSCLC, cucurbitacin I-treated CD133-positive NSCLC, and low-grade NSCLC were closer to the pattern observed in normal lung tissues. To examine whether CSC-like properties are suppressed by STAT3 inhibition, CD133-positive NSCLC cells were incubated with 100 nM or 150 nM cucurbitacin I for 24 hours. Fluorescence-activated cell sorting analysis demonstrated that the quantities of CD133 were decreased dramatically in a dose-dependent manner in cucurbitacin I-treated CD133-positive NSCLC cells (Fig. 4B) (P < .001). The results supported microarray data suggesting that cucurbitacin I inhibited the stemness gene signatures of CD133-positive NSCLC cells and facilitated the differentiation of CD133-positive NSCLC cells into CD133-negative NSCLC cells and low-grade NSCLC (Fig. 4A). Furthermore, cleaved caspase 3 (Fig. 4C) and TUNEL (Fig. 4D) staining also revealed that apoptotic signals were positively correlated with the cucurbitacin I concentration. To further investigate the biologic role of STAT3 in the tumorigenicity of CD133-positive NSCLC under radiation treatment, we applied various IR doses from 0 Gy to 10 Gy to vehicle-treated or cucurbitacin I-treated CD133-positive NSCLC cells. Figure 4E reveals that the survival rate of vehicle-treated CD133-positive NSCLC cells was significantly higher than that of 100-nM cucurbitacin I-treated cells (P < .001). To explore the mechanism involved in the cucurbitacin I-mediated radiosensitizing effect against CD133-positive NSCLC cells, the 100-nM or 150-nM cucurbitacin I-treated NSCLC parental cells or CD133-positive NSCLC cells were exposed to 4 Gy IR. The capabilities of colony formation and invasion (Fig. 4F) were attenuated dramatically by 150 nM cucurbitacin I and by 4 Gy IR. In addition, the combination treatment with cucurbitacin I and IR produced a synergistic effect that abrogated these CD133-positive NSCLC cell capabilities. The activity of caspase 3 also increased concomitantly in CD133-positive NSCLC cells that were treated with cucurbitacin I plus IR compared with IR treatment alone (data not shown). These data suggest that the STAT3 pathway maintains the stemness of CD133-positive NSCLC cells and that the inhibition of STAT3 activation further promotes apoptosis and inhibits radioresistance in NSCLC CSCs.

We also investigated the combined treatment effect of cucurbitacin I and chemotherapy in CD133-positive NSCLC cells. Experiments were conducted with cisplatin, doxorubicin, 5-flouroracil (5-FU), or paclitaxel in vehicle-treated or cucurbitacin I-treated, CD133-positive NSCLC cells. The data revealed that the cell survival rate in vehicle-treated CD133-positive NSCLC cells was not decreased significantly by the 4 drugs that were tested (P > .05; data not shown). Conversely, cell survival declined significantly after chemotherapy with the 4 tested drugs in cucurbitacin I-treated, CD133-positive NSCLC cells (P < .01) (Fig. 5A). The sphere formation ability of CD133-positive NSCLC cells obviously was impeded by combined treatment with cucurbitacin I and cisplatin (Fig. 5B, left) or paclitaxel (Fig. 5B, right). These results suggested that cucurbitacin I can greatly increase chemotherapeutic effectiveness in CD133-positive NSCLC cells by improving drug sensitivity.

Figure 5.

Cucurbitacin I improved chemosensitivity and synergistically suppressed tumor sphere formation in prominin-1 (CD133)-positive (CD133+) nonsmall cell lung cancer (NSCLC) cells. (A) NSCLC-CD133+ and NSCLC-CD133-negative (CD133−) cells (103) were plated in 96-well plates and treated with various concentrations of cisplatin, doxorubicin, 5-flouroracil (5-FU), and paclitaxel for 24 hours in 10% fetal bovine serum/Dulbecco modified Eagle medium//F-12 medium. The survival rate was determined by using a 3(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. (B) Spheroid-like body formation was evaluated in parental, NSCLC-CD133− and NSCLC-CD133+ cells that were treated with cucurbitacin I (Cu.) alone or in combination with cisplatin or paclitaxel. Spheroid-like bodies were counted after 1 week. Data shown are the mean ± standard deviation values from 3 experiments.

Effects of Cucurbitacin I on In Vivo Tumorigenicity and Survival of CD133-Positive NSCLC Cells in a Xenotransplanted Animal Model

We further investigated the role of STAT3 signaling and the effects of cucurbitacin I in CD133-positive NSCLC cells in vivo. CD133-positive and CD133-negative NSCLC cells were transfected with a lentiviral vector that contained the GFP gene (transfection efficiency was 80%).25 First, we injected 1 × 105 CD133-positive/GFP NSCLC cells or 1 × 105 CD133-negative/GFP NSCLC cells into the subcutaneous sites of nude mice that received different treatment protocols. First, the subset of nude mice that received CD133-negative/GFP NSCLC cells formed no tumors in the lung region within 8 weeks of xenotransplantation. The subset of mice that received CD133-positive NSCLC cells, but not CD133-negative NSCLC cells, exhibited significant capabilities of invasion and distant metastasis to the lungs (Fig. 6A). It is noteworthy that, cucurbitacin I treatment in CD133-positive/GFP-transplanted mice effectively reduced the number of lung metastases and tumor size in vivo (Fig. 6B,C). Furthermore, the tumorigenic and metastatic capabilities of CD133-positive cells were decreased synergistically with 4 Gy IR or cisplatin in combination with 100 nM cucurbitacin I compared with cisplatin-only or IR-treated only CD133-positive cells (P < .05) (Fig. 6C,D). Kaplan-Meier survival analysis also demonstrated that the survival rate for the NSCLC-CD133-positive/GFP-transplanted group was significantly improved by IR or cisplatin combined with cucurbitacin I compared with the untreated, IR-alone, or cisplatin alone groups (Fig. 6E). Meanwhile, in vivo xenotransplantation analysis revealed that CD133-positive/GFP-transplanted mice that received cucurbitacin I in combination with cisplatin or IR had a better therapeutic response and prolonged survival compared with mice that received cisplatin combined with IR (data not shown). Overall, this in vivo study indicated that the effectiveness of radiotherapy or chemotherapeutic drugs in mice bearing CD133-positive NSCLC tumors can be improved significantly by adding cucurbitacin I treatment.

Figure 6.

Evaluation of the in vivo tumorigenicity of prominin-1 (CD133)-positive (CD133+) nonsmall cell lung cancer (NSCLC) cells and survival in a xenotransplanted animal model are illustrated. (A,B) In total, 1 × 105 NSCLC-CD133+ cells were injected subcutaneously into nude mice. Six mice in each group received daily intraperitoneal injections of either vehicle (10% ethanol) or drug (1 mg/kg cucurbitacin I [Cu.] in 10% ethanol). After 4 weeks, in vivo green fluorescent protein (GFP) imaging revealed that transplanted NSCLC-CD133+/GFP cells grew solid tumors at the injection site. Tumor volumes in NSCLC-CD133+–transplanted mice that received either cucurbitacin I (100 nM) plus ionizing radiation (IR) (4 grays Gy) or cisplatin (Cis.) alone were significantly lower than the tumor volumes in mice that received IR or drug only (P < .01). (C,D) These charts illustrate (C) the number of metastatic foci and (D) the total volume of tumors in the lungs of mice as determined by macroscopic analysis and histologic examination. Cucurbitacin I combined with 4 Gy IR or cisplatin effectively reduced the number of lung metastases and tumor size in NSCLC-CD133+/GFP-transplanted mice (P < .05 for CD133+/GFP vs all other groups). (E) Kaplan-Meier survival analysis also indicated that the cumulative (Cum) survival rate for mice transplanted with NSCLC-CD133+ cells that were treated with 100 nM cucurbitacin I combined with IR or cisplatin was significantly prolonged compared with the cumulative survival of mice that received IR or drug alone (n = 6 in each group).

DISCUSSION

Currently, surgery remains the most effective treatment for NSCLC. Adjuvant chemotherapy or radiotherapy has been used increasingly to decrease disease recurrence and metastasis. However, resistance to chemotherapy or radiotherapy is the major cause of treatment failure. STAT3, a transcription factor that is regulated by various cytokines and growth factors, especially interleukin-6 (IL-6) and EGF, is activated aberrantly in numerous cancer types, including NSCLC. Recent reports indicate that the IL-6/STAT3 signaling pathway may contribute to tumor progression and to the survival of CSCs, including CSCs in glioma and colon cancers.29 Because the persistent activation of STAT3 promotes tumor cell proliferation and survival, contributing to tumor progression and migration, abrogation of STAT3 signaling is emerging as a potential cancer therapy strategy.23 The current results revealed that CD133-positive cells isolated from 7 patients with NSCLC had greater tumorigenicity, chemoresistance, radioresistance, and expression of stemness as the level of p-STAT3 increased (Table 1; Figs. 1, 2). We also observed that 100 nM cucurbitacin I, a specific JAK-STAT inhibitor, efficiently inhibited cell proliferation and induced apoptosis in CD133-positive NSCLC cells (Figs. 3, 4). Our microarray results suggested that 100 nM cucurbitacin I inhibited the stemness gene signatures of CD133-positive NSCLC cells and facilitated the differentiation of CD133-positive NSCLC cells into CD133-negative NSCLC cells (Figs. 3, 4). In addition, we demonstrated that blocking STAT3 signaling by cucurbitacin I significantly suppressed the self-renewal ability, tumorigenicity, and radiochemoresistance of these cells, suggesting that the activation of STAT3 plays a role in maintaining the CSC-like phenotype of CD133-positive NSCLC cells (Figs. 4, 5). Finally, our Kaplan-Meier survival analysis indicated that the treatment effect of cisplatin or 4 Gy IR for CD133-positive NSCLC can be improved significantly by cucurbitacin I (Fig. 6). To our knowledge, this is the first study to demonstrate that the STAT3 axis plays an important role in maintaining CSC-like properties. Furthermore, this is the first study to demonstrate that targeting STAT3 with cucurbitacin I significantly suppresses tumorigenicity and radiochemoresistance in NSCLC-derived CSCs.

Resistance to chemoradiotherapy is the major clinical criterion for characterizing characterize CSCs.30 Recent studies have indicated that the expression of CD133 in lung cancer represents a population of cells with high tumorigenicity and a phenotype that is resistant to cytotoxic therapy.14, 28 Consistent with these findings,14, 28 we have demonstrated that CD133-positive NSCLC cells exhibit greater chemoradioresistance compared with CD133-negative NSCLC cells or parental cells (Figs. 1, 2). Kruser et al demonstrated that panitumumab, a monoclonal antibody against the EGF receptor, can significantly augment the radiosensitivity of NSCLC through the inhibition of the mitogen-activated protein kinase and STAT3 pathways.31 Constitutively activated STAT3 reportedly was detected in several human lung cancer cell lines and in tumor cells that infiltrated the pleurae of patients who had malignant pleural effusion-associated lung adenocarcinoma.32 Several reports have indicated the synergistic effect of cucurbitacin with known chemotherapeutic agents, such as doxorubicin and gemcitabine.33 In the current study, our data revealed that treatment with cucurbitacin I in CD133-positive NSCLC cells effectively improved their chemoresistance to cisplatin, doxorubicin, 5-FU, and paclitaxel (Fig. 5). Our data also suggested that cucurbitacin I significantly induced apoptosis in CD133-positive, NSCLC cells in part through the inactivation of p-STAT3 and its downstream effectors survivin, Bcl-2, and Bcl-xL (Figs. 3, 4). Furthermore, our data support the possibility that inhibition of the STAT3 pathway by cucurbitacin I may significantly block the migratory and metastatic ability of CD133-positive NSCLC cells in both in vitro and in vivo xenotransplanted models (Figs. 4, 6). Indeed, these data were consistent with a recent report that IR enhances the invasiveness of and the events related to the epithelial-mesenchymal transition of lung cancer cells by Bcl-xL accumulation through STAT3, suggesting that STAT3 and Bcl-xL may contribute to the malignancy of lung cancer cells.34 Notably, in vivo xenotransplanted analysis indicated that cucurbitacin I may be a sensitizer that synergistically enhances both radiosensitivity and chemosensitivity in a xenotransplanted model of human lung CSCs (Fig. 6). Taken together, our findings indicate that the antiproliferative, proapoptotic, and radiochemosensitizing effects of cucurbitacin I on CD133-positive NSCLC cells could be applied as a potential strategy to overcome the resistance of highly tumorigenic CD133-positive NSCLC cells in patients with advanced NSCLC. In addition, the actual biomolecular mechanisms of NSCLC CSCs and CD133-positive NSCLC cells remain unclear, and further investigations are warranted into both the targeting of JAK/STAT3 signaling and the radiochemoresistance involved in the CSC-related microenvironment that has a biologic impact on NSCLC recurrence and metastasis.

In conclusion, we have demonstrated that the STAT3 signaling axis may be responsible for the CSC-like properties and radioresistance of CD133-positive NSCLC cells. Cucurbitacin I potentially may attenuate the malignancy of CD133-positive NSCLC cells and may present a potential clinical benefit for the treatment of lung cancer and CSC. Thus, there is a great need to unravel the underlying mechanisms of the STAT3 pathway in CD133-positive NSCLC and to further evaluate the therapeutic possibilities of cucurbitacin I.

CONFLICT OF INTEREST DISCLOSURES

This work was supported by grant 97-3111-B-075-001-MY3 from the National Science Council, Taiwan.

Ancillary