Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The PTEN/PI3K/Akt pathway regulates stem-like cells in primary esophageal carcinoma cells

The PTEN/PI3K/Akt pathway regulates stem-like cells in primary esophageal carcinoma cells Resea RCh pape R Cancer Biology & Therapy 11:11, 950-958; June 1, 2011; © 2011 Landes Bioscience The PTEN/PI3K/Akt pathway regulates stem-like cells in primary esophageal carcinoma cells h ongxia Li, Quanli Gao, Liping Guo and s hih h sin Lu* s tate Key Laboratory of Molecular Oncology and Department of etiology and Carcinogenesis; Cancer Institute and h ospital; Chinese a cademy of Medical s ciences and peking Union Medical College; Beijing, China Current a ddress: h enan provincial Tumor h ospital; Zhengzhou, h enan China Key words: esophageal cancer, SP cells, PI3K/Akt pathway Abbreviations: ESCC, esophageal squamous cell carcinoma; SP, side population; NOD/SCID, non-obese diabetic/severe combined immunodeficiency; ABC, ATP-binding cassette; FBS, fetal bovine serum; PBS, phosphate buffer saline; CAMS&PUMC, Chinese Academy of Medical Science and Peking Union Medical College Recent reports have shown that cancer stem cells exist in many malignancies. s ide population (s p) cells are used to enrich cancer stem-like cells in many cell lines and fresh tumor specimens. In this study, we cultured primary esophageal squamous cell carcinoma (es CC) cells from es CC tissue specimens. sp cells from primary es CC cells were more resistant to chemotherapeutic reagents and formed more colonies in vitro than non-s p cells. In addition, xenograft experiments revealed that s p cells were more tumorigenic in vivo. Further results indicated that the pI3K/a kt pathway is essential to s p cells through the regulation of a BCG2 transporter function. Furthermore, pTeN, rather than mTOR, was found to be involved in sp cell regulation in primary es CC cells. These findings reveal that sp cells are enriched for cancer stem-like cells in primary es CC cells and that the pTeN/pI3K/a kt pathway regulates this stem-like population. This study indicates ©2011 Landes Bioscience. that sp cells in primary culture cells from tissue specimens could be a promising model for cancer stem cell research and may help researchers develop novel therapeutic strategies or efficient drugs that target es CC stem-like cells. Do not distribute. 4 7,8 brain tumor Introduction and colon cancer. Surface markers have not been identified in certain tumors, and SP sorting using Hoechst dye Recent evidence has suggested that only a small population of can enrich for cancer stem-like cells due to the ability of these cells cells in a tumor are responsible for the initiation, proliferation, to exclude the dye via membrane-bound pumps. SP cell sorting metastasis and recurrence of tumors as determined by a variety was initially used to identify hematopoietic stem cells; however, 1-3 of in vitro and in vivo assays. These cells have been defined as SP cells have been proven to be enriched in cancer stem-like cells 13-17 “cancer stem cells,” “cancer stem-like cells” or “tumor-initiating in various tumors and tumor cell lines. SP cells are detected cells.” by their ability to efu fl x Hoechst 33342 dye through adenosine Cancer stem-like cells represent a population of self-renewing triphosphate (ATP)-binding cassette (ABC) membrane trans- and drug-resistant cells that can survive treatment and repopu- porters. Indeed, the SP phenotype appears particularly valuable late the tumor. Experiments using hematopoietic and solid malig- for the future identification of specific stem cell markers. 3-11 nancies support tumor hierarchy as a fundamental concept in The mechanism for the SP phenotype involves several ABC tumor biology and suggest that this hierarchy is a novel cellular transporters such as proteins encoded by multi-drug resistant gene target for anticancer drug discovery. Previous reports have identi- 1 (MDR1), multidrug resistant associated protein 1 (MRP1) and fied cancer stem-like cells from fresh tumor specimens, which are BCRP1/ABCG2 (ABCP/MXR/BCRP) with overlapping prop- more representative of the cells in tumors compared with stem- erties in extruding cytotoxic drugs, as well as the fluorescent dye like cells isolated from cell lines. However, there have also been Hoechst, out of the cell. Further exploration on the maintenance 18,19 exciting reports on cancer stem-like cells sorted from cell lines in of the SP population includes ABCG2 or EGFR expression as 20,21 the study of regulation of stemness. In this study, we show that well as Akt and STAT3 pathway activation. primary cells from tumor specimens may be an additional choice Esophageal cancer is the sixth most common cause of cancer- for cancer stem cell research. related deaths worldwide, and its incidence is characterized by Molecular markers of cancer stem-like cells have been identified large geographical variation. ESCC is the most prevalent type 3 9 in some malignancies such as breast cancer, pancreatic cancer, in China, and most ESCC patients die from recurrence of the *Correspondence to: Shih Hsin Lu; Email: shlu@public.bta.net.cn Submitted: 12/09/10; Revised: 02/17/11; Accepted: 03/17/11 DOI: 10.4161/cbt.11.11.15531 950 Cancer Biology & Therapy Volume 11 Issue 11 Resea RCh pape R Resea RCh pape R ©2011 Landes Bioscience. Do not distribute. Figure 1. primary es CC cells from tumor specimen and sp analysis. (a ) primary cells of es CC tissue specimen. The image was obtained on passage 5. The cells were all adherent, possessed the typical epithelial polygonal morphology, and proliferated in a single layer (400x). (B) Representative image of h &e staining of xenograft tumors formed by primary es CC cells (100x). (C) Representative sp analysis of primary es CC cells. The sp fraction in pri- mary es CC cells was 2.4%. Incubation with 5 μM FTC abolished the sp cell fraction. cancer later in their life. Understanding the mechanisms under- ESCC. Adherent cells were visible after 7 days in the primary cul- lying carcinogenesis is essential for the development of novel ture of the tissue specimen. After 10 days, the solid particles were therapeutic strategies that target esophageal malignancies, few removed, and the cells were trypsinized at 90% conu fl ency and reports on esophageal cancer stem cells or the important signal- split 1:3. The cells were typical epithelial-like cells with polygo- ing pathways in this cancer exist. We previously reported that SP nal morphology and proliferated in a single layer (Fig. 1A). All cells are enriched in cancer stem-like cells in ESCC cell lines. of the experiments were performed on the cells within 15 pas- In this study, we established a method to obtain primary ESCC sages. The primary ESCC cells formed tumors when inoculated cells from tumor specimens of ESCC patients and showed that SP into immune-deficient mice, and pathologically, the tumors were cells from primary ESCC cells are enriched for cancer stem-like found to be squamous cell carcinoma (Fig. 1B). cells. Furthermore, our results indicate that the PTEN/PI3K/Akt To determine whether the primary ESCC cells contained SP signaling pathway plays a modulating role in the SP population in cells, we stained the ESCC cells with Hoechst 33342 and per- primary ESCC cells. Identic fi ation of esophageal cancer stem-like formed SP analysis. Representative results of the flow cytometry cells and further elucidation on the signaling pathways may pro- analysis are shown in Figure 1C. The primary ESCC cells con- vide strategies for the development of novel therapeutic strategies tained about 2.4% SP cells. The fraction of SP cells was decreased and efc fi ient drugs that target esophageal cancer stem-like cells. by treatment with fumitremorgin C (FTC), an ABC transporter inhibitor. We obtained similar results with two additional speci- Results mens by primary culture (Sup. Fig. S1). SP cells are more resistant to chemotherapeutic reagents SP cells in primary ESCC cells. Primary ESCC cells were than non-SP cells. We performed drug sensitivity assays with obtained from a surgical specimen of a 49-y-old female patient. sorted SP and non-SP cells that were treated with cisplatin or The pathology of the tumor was diagnosed as well-differentiated bortezomib by MTT assays. Cisplatin is commonly used for the www.landesbioscience.com Cancer Biology & Therapy 951 chemotherapy of ESCC, and bortezomib is a new chemothera- We collected cell lysates from SP and non-SP cells and performed peutic agent that targets the proteasome. After exposure with protein gel blot analyses. We found that the phospho-Akt (Ser473) 1 μg/ml cisplatin, the viability of the SP cells was markedly level in SP cells was much higher than in non-SP cells, while the higher compared to non-SP cells (p < 0.05). The results were total Akt level was similar in both groups (Fig. 2E). This find - similar in the case of bortezomib treatment (p < 0.05 for both 10 ing indicated that the PI3K/Akt pathway was activated in SP cells. and 25 nM). Because SP and non-SP cells from primary ESCC We tested the importance of the PI3K/Akt pathway in SP cells of cells did not differ in their growth rate (Sup. Fig. S2), SP cells primary ESCC cells using specic fi inhibitors including LY294002 were more resistant to 1 μg/ml of cisplatin as well as to 10 and (20 μM), Akt inhibitor IV (500 nM) and Akt inhibitor VIII (500 25 nM of bortezomib than non-SP cells (Fig. 2A). These results nM). After treatment for 3 days, the SP fraction from the cells was demonstrate the apparent chemoresistance of ESCC stem-like determined. The results show that all the inhibitors signic fi antly cells to anticancer drugs that may contribute to tumor recurrence attenuated the formation of the SP fraction of primary ESCC cells and resistance to multiple drugs. compared with DMSO-treated control cells. As shown in Figure SP cells highly express ABC transporter genes. It has been 3A, the percentage of cells in the SP cell fraction decreased after reported that the SP phenotype and the chemoresistance of SP treatment with LY294002, Akt inhibitor IV or Akt inhibitor VIII, cells depend primarily on ABC transporters such as ABCG2/ from 2.5 to 0.8, 0.1 and 0%, respectively. The Akt inhibitors IV BCRP, ABCB1/MDR1 and ABCA3. To investigate whether the and VIII, in particular, were very effective at reducing the number SP cells from primary cells had elevated levels of these genes, we of primary ESCC cells in the SP population. The inhibitory effect performed quantitative real-time PCR of the ABC transporter of these compounds on phopho-Akt was verie fi d by protein gel genes ABCA3, ABCB1, ABCC1 and ABCG2. As previously blotting and is shown in Figure 3B. reported in reference 17, these genes were preferentially expressed To determine whether the PI3K/Akt pathway was important in SP cells compared with non-SP cells (Fig. 2B), indicating that to the colony formation of the primary ESCC cells, we performed the expression of ABC transporter genes was responsible for the soft agar colony formation assays using the PI3K/Akt inhibi- SP and chemoresistance phenotypes from primary ESCC cells. tors. Pretreatment of primary ESCC cells with the inhibitors as We evaluated the role of ABCG2 in mediating drug efu fl x of described above reduced the colony-formation rate of the primary primary ESCC cells. It has been shown that mitoxantrone is a ESCC cells as compared with the DMSO-treated control cells. substrate of ABCG2. I© n the m2 it0 oxan1 tron1 e e fu flL x asa say, tn he id ntra-es Th e cB oloni y fo orms atic on ri ate e of tn he uc ne treat. ed control group was cellular u fl orescence intensity can be measured by flow cytometry 7.42 ± 1.12%; for the LY294002 treated group, the rate was 5.64 and used to determine ABCG2 activity. As shown in Figure 2D, ± 0.87% (p = 0.003); for the Akt inhibitor IV, the rate was 4.82 ± the intracellular fluorescence intensity of SP cells was lower (28.4 0.56% (p = 0); and for the Akt inhibitor VIII, the rate was 6.24 Do not distribute. ± 2.4) than non-SP cells (56.3 ± 9.3, p = 0.007) and unsorted cells ± 0.32% (p = 0.014) (Fig. 3C). (55.5 ± 11.2 p = 0.008), indicating that the drug was pumped out The Akt pathway regulates ABCG2 activity in ESCC pri- more efficiently in SP cells. mary cells. Because it has been shown that PI3K/Akt signaling SP cells have higher colony-formation ability and are more regulates the SP phenotype in glioma cells by regulating ABCG2 tumorigenic than non-SP cells in mice. To determine whether function, we determined whether this mechanism exists in pri- the SP cells from primary ESCC cells were enriched for cancer mary ESCC cells. Mitoxantrone is a substrate of ABCG2 and can stem-like cells, we tested the tumorigenic potential of SP and be detected by flow cytometry as a measure of ABCG2 function. non-SP cells both in vitro and in vivo. Colony formation ability Primary ESCC cells were treated with LY294002 (20 μM), Akt in soft agar assay revealed that approximately 10.6% of 1,500 SP inhibitor IV (500 nM) or Akt inhibitor VIII (500 nM) for 3 days cells were capable of forming visible colonies, which was almost and subsequently assayed for their ability to efu fl x mitoxantrone. two and a half times more than non-SP cells (about 4%) (Fig. The intracellular intensity of DMSO-treated control cells was 2C). Furthermore, in vivo tumorigenic ability was evaluated by comparatively low (mean fluorescence intensity of 59.34 ± 10.69) injecting both SP and non-SP cells, mixed with matrigel matrix, because mitoxantrone was pumped out by the ABCG2 transporter. into NOD/SCID mice. As shown in Table 2, the tumorigenic- Incubation with the PI3 kinase inhibitor LY294002 or the Akt ity of SP cells was significantly higher than that of non-SP cells. inhibitors IV or VIII increased the intracellular fluorescence inten - Seven out of eight mice inoculated with 100 SP cells gave rise sity of primary ESCC cells (mean value of 110.03 ± 5.95, 122.55 ± to tumors, whereas only one out of eight inoculations of 1,000 13.74 and 147.81 ± 17.32, respectively, p < 0.05) (Fig. 3D). Because non-SP cells formed a tumor. These results indicate that as few the expression of the ABC transporters did not change after treat- as 100 SP cells were tumorigenic in NOD/SCID mice, whereas ment with the PI3K/Akt inhibitors (Sup. Fig. S4), the increased ten times the number of non-SP cells generated very few tumors. intracellular fluorescence intensity indicated decreased ABCG2 The pathology results confirmed that the tumors formed by SP function to efu fl x the drug. These results are consistent with the and non-SP cells were all ESCC (Sup. Fig. S3). changes in the SP fraction upon treatment with the PI3K/Akt The PI3K/Akt pathway regulates the SP phenotype and pro- inhibitors. These data reveal that the PI3K/Akt pathway plays an motes the tumorigenicity of primary ESCC cells. As reported, important role in regulation of the SP fraction. PI3K/Akt pathway plays an important role in the maintenance PTEN, but not mTOR, regulates the SP phenotype in ESCC 20,21 of SP cells. Therefore, we determined the expression level of primary cells. PTEN plays an important role in the regulation of components of the PI3K/Akt pathway in ESCC stem-like cells. SP cells in glioma malignancy and breast cancer by negatively 952 Cancer Biology & Therapy Volume 11 Issue 11 ©2011 Landes Bioscience. Do not distribute. Figure 2. sp cells in primary es CC cells. (a ) s ensitivity of sp and non-sp cells to DDp and bortezomib. sp cells were more resistant to the chemothera- peutic reagents upon 48 hours of treatment. *p < 0.05, bars: se M. (B) Relative mRNa expression of the a BC transporters in sp and non-sp cells. The expression level was normalized to Ga pDh . The a BC transporter genes a BCa 3, a BCB1, a BCC1 and a BCG2 were all highly expressed in sp cells. (C) Representative images from colony-formation assay from sp and non-sp cells. Fifteen-hundred freshly sorted cells were seeded in each well of a 6-well plate in soft agar and the images were obtained 3 weeks later. sp cells formed a greater number of colonies. *p < 0.05, t-test. (D) Flow cytom- etry histograms of mitoxantrone efflux from sp, non-sp cells and unsorted cells. The intracellular fluorescence in the sp fraction was lower than in the non-sp and unsorted cell fractions. (e) protein gel blotting of phosphorylated/total a kt, phosphorylated/total mTOR and phosphorylated/total pTeN in sorted sp and non-sp primary es CC cells. www.landesbioscience.com Cancer Biology & Therapy 953 ©2011 Landes Bioscience. Do not distribute. Figure 3. pI3K/a kt inhibitors decrease the sp fraction. (a ) sp analysis of primary es CC cells treated with pI3K/a kt inhibitors. Compared to control cells treated with DMs O, treatment with LY294002, a kt inhibitor IV and a kt inhibitor VIII for 3 days significantly reduced the sp fraction in primary es CC cells. (B) protein gel blotting of total a kt and phospho-a kt in primary cells treated with DMs O, LY294002, a kt inhibitor IV and a kt inhibitor VIII for 3 days. β-actin was used as a loading control. The level of phospho-a kt was decreased by the treatment with the inhibitors. (C) Inhibition of colony formation by the a KT/pI3K inhibitors. Treatment with LY294002, a kt inhibitor IV and a kt inhibitor VIII reduced the number of colonies of primary cells formed in soft agar. *p < 0.05, t-test. (D) Flow cytometry histograms of mitoxantrone efflux from untreated cells and inhibitor treated cells. Blocking pI3K/a kt pathways reduced the ability of a BCG2 to efflux mitoxantrone. 20,21 regulating PI3K/Akt signaling. Therefore, PTEN may also mTOR in the regulation of SP cells in primary ESCC cells, small- play an important role in ESCC cancer stem-like cells. mTOR is molecule inhibitors and RNA interference were used. Primary a downstream regulator of the PI3K/Akt pathway and has been ESCC cells were treated with the mTOR inhibitor rapamycin reported to regulate SP cells in breast cancers. As shown in (5 μM) or the PTEN specic fi inhibitor bpV(pic) (20 μM) for Figure 2E, the level of phospho-PTEN in non-SP cells was higher 3 days and analyzed for the presence of the SP fraction. The SP than in SP cells, while the level of phospho-mTOR in SP and fraction increased after treatment with the PTEN-specic i fi nhibi - non-SP cells was similar. To determine the effect of PTEN and tor bpV(pic) compared to control cells (6.8 vs. 2.6%); however, 954 Cancer Biology & Therapy Volume 11 Issue 11 ©2011 Landes Bioscience. Do not distribute. Figure 4. The ee ff ct of pTeN and mTOR on sp cells in primary es CC cells. (a ) sp analysis of primary es CC cells treated with mTOR and pTeN inhibitors or with knockdown of mTOR and pTeN by RNa interference. Inhibition of pTeN increased the sp fraction while mTOR inhibition did not alter the level of the sp fraction. (B) protein gel blotting of mTOR and pTeN level after RNa interference. Forty-eight hours after transfection with mTOR or pTeN siRNa , the level of mTOR and pTeN were decreased. β-actin was used as a loading control. (C) eff ect of mTOR and pTeN inhibitors on colony formation of primary es CC cells. BpV(pic), a pTeN inhibitor, increased the colony-forming ability of primary cells, whereas rapamycin had no ee ff ct on colony formation. *p < 0.05, t-test. the SP fraction did not change when treated with rapamycin (2.8 with the PTEN siR NA, whereas the SP fraction did not change vs. 2.6%) (Fig. 4A). after transfection with the mTOR siRNA when compared RNA interference was used to confirm the effect of PTEN with the control cells that were transfected with non-silencing and mTOR on the SP phenotype. Forty-eight hours after siRNA. The results are consistent with the results obtained transfection, the cells were collected for SP analysis. siRNA- with the inhibitors. mediated silencing was validated by decreased levels of the The mTOR and PTEN inhibitors were also used to deter- PTEN and mTOR proteins and is shown in Figure 4B. As mine their effect on colony forming ability of primary ESCC shown in Figure 4A, the SP fraction increased after transfection cells. When the cells were pretreated with bpV(pic), the www.landesbioscience.com Cancer Biology & Therapy 955 colony-formation rate of the primary ESCC cells increased com- function. In our study, we demonstrated that PI3K/Akt signal- pared to control cells (9.93 ± 0.20% vs. 7.42 ± 1.12% , p = 0.047). ing pathway was also important for the ESCC SP cell phenotype. After the treatment with rapamycin, the colony formation rate of Using PI3K/Akt inhibitors, we found that they significantly the cells was similar to that of control cells (7.63 ± 1.80% vs. 7.42 reduced the SP cell fraction, which is consistent with the reports ± 1.12%, p = 0.833) (Fig. 4C). on cancer stem-like cells in other malignancies. Furthermore, the tumorigenicity was also reduced upon inhibition of PI3K/ Discussion Akt signaling, which was most likely a result of the inhibition of ESCC stem-like cells. PI3K/Akt signaling activity is vital in the Most cancer stem cells were initially identiefi d in fresh tumor maintenance of the cells in the SP fraction and in maintaining specimens from surgery, but some cancer cell lines have also been the high tumorigenicity of primary ESCC cells. 13-17 shown to be enriched with cancer stem-like cells. It is thought mTOR is emerging as a major downstream target of Akt in the that cancer stem-like cells from cell lines may be a promising regulation of the cellular response to nutrients and other stim- model for cancer stem cell research as a result of their unlimited uli. Loss of PTEN function leads to increased concentrations of supply and ease of handling. Fresh tumor specimens of patients are PIP3, the main in vivo substrate for PTEN, resulting in constitu- not always available for all research groups. In addition, the cells tive activation of downstream components of the PI3K pathway, might be difficult to culture, especially after the digestion, FACS including the Akt and mTOR kinases. Loss of PTEN, a PI3K analysis and sorting, and the results may be invalid. However, inhibitor, increases the proliferation and survival of neural and 34,35 many available cell lines have been established and cultured for hematopoietic stem cells as well as increases the SP fraction 21,30 years, even decades, through numerous passages. Tumor cells in glioma cells and prostate cancer stem-like cells. Inhibition from primary cultures of fresh tissue specimens provide us with a of mTOR with the specific inhibitor rapamycin or with knock - more physiological model than cell lines to study cancer stem-like down by RNA interference did not change the SP fraction or cell biology. We have determined that ESCC tissue specimens can the tumorigenicity of the primary ESCC cells, whereas PTEN be maintained as primary cultures for several weeks to months inhibition increased both the SP fraction and the tumorigenicity to obtain a sufc fi ient number of cells for many cellular biologi - of primary ESCC cells. All these results indicate that PTEN, the cal and molecular analyses. The cells are easier to handle, and loss of which could upregulate the SP fraction, is a key regula- the results obtained f© rom th2 ese c0 ells a1 re m1 o re rL epa rodun cibld e thae n s tor i n tB he mio aints enac nce oi f pe rin mary Ec Se CC s. tem-like cells, while results obtained directly from tissue specimens. In addition, cells mTOR does not appear to play a major role in maintenance of obtained from primary tissue better approximate tumors in vivo the SP fraction. than cell lines. The primary tumor cells retained characteristics Components of PI3K/Akt pathway have emerged as prom- Do not distribute. of the original tumors when cultured in serum-free conditions, ising novel targets for the development of cancer therapy. In which avoids stimulation by serum. Stem-like cells from ESCC medical oncology, chemoresistance is a major hurdle for success- primary cultures may be a promising model for stem cell research ful cancer therapy. Because Akt activation is also linked to drug in ESCC. Additionally, patient specic c fi ancer stem-like cells may resistance in many cancers, targeting this pathway could restore 37,38 be used in clinical research or for treatment in the future. drug sensitivity. The results of the mitoxantrone efu fl x assay We previously showed that SP cells in ESCC cell lines were in primary cells in the presence of above inhibitors indicate that enriched with cancer stem-like cells. In this study, we obtained the function of the ABC transporters is modulated by the PI3K/ primary cells from ESCC tissue specimens and found that the Akt pathway. These data further explain the connections among percentage of SP cells within the tissue specimen was similar to the Akt pathway, stem-like characteristics and therapeutic resis- that found in cell lines. Our results indicate that SP cells in pri- tance, suggesting that activation of this pathway also enhances mary ESCC cells are enriched for cancer stem-like cells, which the ability of ESCC stem-like cells to efu fl x drugs. This model is consistent with the results obtained from ESCC cell lines and is consistent with reports on the ABC transporters regulated other reports on SP cells. by PI3K/Akt activity in hepatocellular carcinoma and glioma 21,31 The PI3K/Akt signaling cascade is frequently disrupted in malignancy. many human cancers, and it is a key player in mediating tumor A good understanding of cancer stem-like cell biology will cell survival and escape from apoptosis. In light of the wide- provide the possibility of developing cancer stem-like cell tar- ranging biological consequences, this pathway is considered a key geted therapy, which will be helpful in efficiently eradicating determining factor in tumor aggressiveness and is an attractive tumors. Blocking cancer stem-like cell survival signaling is one target for therapeutic intervention. It has been reported that of the most efficient ways in stem cell-targeted therapies. We PI3K/Akt signaling is associated with chemoresistance in many demonstrate that the PTEN/PI3K/Akt signaling pathway is 25-28 malignances and regulates the SP cell phenotype via ABCG2 essential for the SP phenotype and tumorigenicity for primary translocation in hematopoietic stem cells. Recent reports have ESCC cells and regulates the function of ABC transporters asso- indicated that activation of the PI3K/Akt signaling pathway ciated with the SP phenotype. These results provide evidence is required for maintaining cancer stem-like cell viability and that PTEN/PI3K/Akt signaling may be a potential target for maintenance in breast cancer, prostate cancer, hepatocellular eliminating cancer stem-like cells in ESCC and may promote 20,21,30,31 carcinoma and brain tumor cell lines. Additionally, the the development of new therapeutic strategies and efc fi ient drugs PI3K/Akt signaling pathway has been found to regulate ABCG2 to target ESCC stem-like cells. 956 Cancer Biology & Therapy Volume 11 Issue 11 Table 1. primer sequences for specific gene amplification Table 2. Tumorigenicity assay of primary sp and non-sp cells in NOD/ s CID mice Gene name Primers Cell number SP non-SP 5'-a Ga aa Ta CG GTG CCG GCT a TC a Ca -3' (forward) a BCa 3 3 -/- 5'-Ca a TGC CCa GCT CTT TCT GCT TCT-3' (reverse) 5 x 10 2/2 2 x 10 4/4 3/4 5'-GCT CCT Ga C Ta T GCC aaa GC-3' (forward) a BCB1 5'-TCT TCa CCT CCa GGC TCa GT-3' (reverse) 3 1 x 10 4/4 1/8 5'-CTG GGC TTa TTT CGG a TC aa -3' (forward) 2 -/- 1 x 10 7/8 a BCC1 5'-TGa a TG GGT CCa GGT TCa TT-3' (reverse) 5'-Ca C CTT a TT GGC CTC a GGa a -3' (forward) a BCG2 culture, the absorption values were determined using a Bio-Rad 5'-CCT GCT TGG aa G GCT CTa TG-3' (reverse) enzyme reader. 5'-GTG Ga C CTG a CC TGC CGT CT-3' (forward) Ga pDh Soft agar colony formation assay. Soft agar colony formation 5'-GGa GGa GTG GGT GTC GCT GT-3' (reverse) was assessed using a modified method as previously described in reference 39. Cells were suspended in 0.3% agar/culture medium Material and Methods and plated at a density of 1,500 cells per well in a 6-well plate that had been previously coated with 0.5% agar. Fresh medium Primary culture of ESCC specimens. ESCC specimens were was added to each well every 3 days. After 3 weeks of incubation, obtained from patients undergoing surgical resection for esopha- colonies were stained with 0.2% p-iodonitrotetrazolium violet geal cancer subsequent to providing informed consent. This (Amersco) and counted. study was approved by the institutional review board of Cancer Xenograft assay in mice. Nonobese diabetic (NOD)/severe Hospital, Chinese Academy of Medical Sciences and Peking combined immunodeficient (SCID) mice were kept in microiso - Union Medical College (CAMS&PUMC). The tissue specimens lator cages according to the guidelines of CAMS&PUMC, and were received within an hour after surgery, cut into small pieces all experiments were approved by the animal care committee of and rinsed with ice-cold calcium/magnesium free phosphate buf- CAMS&PUMC. The freshly sorted cells were injected subcu- fer saline (PBS). The tissue pieces were digested for about 2 h taneously into the left axillary fossa of female mice (3–4 weeks at 37°C in an incubat© or wi2 th 1 m0 g/1 ml c1 oll agL enaa se I (n Sigmd a) ie n s old ) iB n 20i 0 o μl ms atc rigei l me atrn ix (c BDe ) dil. uted in PBS at a 1:1 serum-free DMEM medium (Invitrogen). The digested tissue ratio. The mice were monitored twice a week for palpable tumor was subsequently centrifuged and collected for further culture formation and were sacrificed at 8–12 weeks after transplanta - in serum-free medium (Invitrogen) supplemented with 20 ng/ tion to determine tumor formation. Tumors were fixed in 10% Do not distribute. ml EGF (PeproTech), 20 ng/ml bFGF (PeproTech), penicillin G buffered formalin and paraffin-embedded; then, the slides were (100 units/ml) and streptomycin (100 μg/ml). The medium was prepared for H&E staining. changed every 2 or 3 days until the epithelial cells attached to the Protein gel blot analysis. Cells were harvested and lysed dishes and expanded. Cells were trypsinized at 90% conu fl ency in lysis buffer, and protein gel blot analysis was performed as and split 1:3. previously described in reference 39. Equivalent amounts of Analyzing and sorting of primary ESCC cells by FACS. protein were separated by SDS-PAGE and transferred to poly- The procedure for SP analysis is based on procedures previously vinylidene diu fl oride membranes (Millipore). After blocking, described in reference 16, with modification. Cells were detached the membranes were incubated with the appropriate diluted pri- from plates with trypsin, washed with PBS, resuspended in mary antibodies, including Akt, phospho-Akt (p-Akt, Ser473), serum-free medium at 1 x 10 cells/ml and incubated for 90 min mTOR, phospho-mTOR (p-mTOR, Ser2448), PTEN, phos- at 37°C with Hoechst 33342 (Sigma) at a final concentration of pho-PTEN (p-PTEN, Ser380) (Cell Signaling Technology) and 5 μg/ml either alone or in the presence of 5 μM FTC (Sigma). β-actin (Sigma). The signal of the target protein was detected by After incubation, cells were washed with ice-cold PBS and fil - an enhanced chemiluminescence detection system (Pierce) and tered through a 40 μm cell strainer. Propidium iodide (Sigma) recorded on film in the linear detection range. was added at a final concentration of 1 μ g/ml for 5 min before RNA extraction and quantitative real-time PCR. Total analysis. Cells were sorted using dual-wavelength analysis with RNA was extracted using the TRIzol reagent (Invitrogen) and BD FACSVantage SE equipped with a Coherent Innova 300C reverse transcribed into cDNA using the SuperScript 3 First laser. The SP gate in the flow cytometry analysis was defined as strand cDNA Synthesis Kit (Invitrogen). Primer sequences for the diminished area on the dot plot in the presence of FTC. SP specific gene amplification are shown in Table 1 . Real-time PCR and non-SP cells were collected for further study. was carried out according to the standard protocol of use with TM Sensitivity to chemotherapeutic reagents. Sensitivity to che- the SYBR Premix Ex Taq Perfect Real Time system (Takara, motherapeutic reagents of SP and non-SP cells was assessed using Dalian, China) on an ABI 7300 detector (Applied BioSystems). -ΔΔCt the MTT assay as previously described in reference 17. Briefly, Fold changes in gene expression were calculated using the 2 3 40 2 x 10 cells per well were seeded on 96-well plate and cul- method. GAPDH was used as an internal control. tured overnight. Cisplatin (0.5 or 1 μg/ml) and bortezomib Flow cytometric detection of mitoxantrone efufl x. The (10 or 25 nM) were chosen as representative chemotherapeutic activity of the ABCG2 pump was tested by mitoxantrone efu fl x agents and were added to the medium individually. After 48 h in analysis as previously described in reference 21. Cells were www.landesbioscience.com Cancer Biology & Therapy 957 6 trypsinized and resuspended at 1 x 10 cells/ml. Then, mitoxan- at a concentration of 100 nM, and the cells were collected for trone was added at the final concentration of 200 nM for 1 h at further study 48 h after transfection. Silencing was validated by 37°C, 5% CO . The cells were subsequently washed two times protein gel blot. with ice-cold PBS. The fluorescence of the intracellular mitoxan - Statistical analysis. Statistical software SPSS12.0 was used trone from 10,000-gated events was logarithmically measured at in data processing and for analyzing the significance between a laser-excitation wavelength of 633 nm using a 670 nm bandpass groups with the t-test. p < 0.05 was considered statistically filter on a BD LSR analyzer. significant. RNA interference synthesis and transfection. Small- Acknowledgments interfering RNA (siRNA) and non-silencing control siRNA were synthesized and annealed by Genechem Company. The siRNA This study has been supported by State Key Basic Research sequences used for PTEN and mTOR were: 5'-CAG UAG AGG Programs 973 of China (2009CB521803). AGC CGU CAA ATT-3' and 5'-GAG GAG UCU ACU CGC Note UUC UAU TT-3', respectively. Cell transfections were per- formed using LipofectAMINE 2000 (Invitrogen) according to Supplemental materials can be found at: the manufacture’s instruction. Briefly, siRNA was transfected http://www.landesbioscience.com/journals/cbt/article/15531/ 29. Mogi M, Yang J, Lambert JF, Colvin GA, Shiojima I, 16. Wang J, Guo LP, Chen LZ, Zeng YX, Lu SH. References Identification of cancer stem cell-like side population Skurk C, et al. Akt signaling regulates side population 1. Lapidot T, Sirard C, Vormoor J, Murdoch B, Hoang T, cell phenotype via Bcrp1 translocation. J Biol Chem cells in human nasopharyngeal carcinoma cell line. Caceres-Cortes J, et al. A cell initiating human acute Cancer Res 2007; 67:3716-24. 2003; 278:39068-75. myeloid leukaemia after transplantation into SCID 30. Dubrovska A, Kim S, Salamone RJ, Walker JR, Maira 17. Huang D, Gao Q, Guo L, Zhang C, Jiang W, Li H, mice. Nature 1994; 367:645-8. et al. Isolation and identification of cancer stem-like SM, Garcia-Echeverria C, et al. The role of PTEN/ 2. Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem Akt/PI3K signaling in the maintenance and viability cells in esophageal carcinoma cell lines. Stem Cells Dev cells, cancer and cancer stem cells. Nature 2001; 2009; 18:465-73. of prostate cancer stem-like cell populations. Proc Natl 414:105-11. Acad Sci USA 2009; 106:268-73. 18. Kim M, Turnquist H, Jackson J, Sgagias M, Yan Y, 3. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison Gong M, et al. The multidrug resistance transporter 31. Hu C, Li H, Li J, Zhu Z, Yin S, Hao X, et al. Analysis SJ, Clarke MF. Prospective identification of tumori- of ABCG2 expression and side population identifies ABCG2 (breast cancer resistance protein 1) effluxes genic breast cancer cells. Proc Natl Acad Sci USA 2003; Hoechst 33342 and is overexpressed in hematopoietic intrinsic drug efflux in the HCC cell line MHCC-97L 100:3983-8. and its modulation by Akt signaling. Carcinogenesis stem cells. Clin Cancer Res 2002; 8:22-8. 4. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani 2008; 29:2289-97. 19. Chen JS, Pardo FS, Wang-Rodriguez J, Chu TS, ©2011 Landes Bioscience. J, Hide T, et al. Identification of human brain tumour 32. Bjornsti MA, Houghton PJ. Lost in translation: dys- Lopez JP, Aguilera J, et al. EGFR regulates the side initiating cells. Nature 2004; 432:396-401. population in head and neck squamous cell carcinoma. regulation of cap-dependent translation and cancer. 5. Bonnet D, Dick JE. Human acute myeloid leukemia is Cancer Cell 2004; 5:519-23. Laryngoscope 2006; 116:401-6. organized as a hierarchy that originates from a primitive 20. Zhou J, Wulfkuhle J, Zhang H, Gu P, Yang Y, Deng J, 33. Di Cristofano A, Pandolfi PP. The multiple roles of Do not distribute. hematopoietic cell. Nat Med 1997; 3:730-7. PTEN in tumor suppression. Cell 2000; 100:387-90. et al. Activation of the PTEN/mTOR/STAT3 pathway 6. Collins AT, Berry PA, Hyde C, Stower MJ, Maitland in breast cancer stem-like cells is required for viabil- 34. Groszer M, Erickson R, Scripture-Adams DD, Lesche NJ. Prospective identification of tumorigenic prostate ity and maintenance. Proc Natl Acad Sci USA 2007; R, Trumpp A, Zack JA, et al. Negative regulation cancer stem cells. Cancer Res 2005; 65:10946-51. 104:16158-63. of neural stem/progenitor cell proliferation by the 7. O’Brien CA, Pollett A, Gallinger S, Dick JE. A human Pten tumor suppressor gene in vivo. Science 2001; 21. Bleau AM, Hambardzumyan D, Ozawa T, Fomchenko colon cancer cell capable of initiating tumour growth in EI, Huse JT, Brennan CW, et al. PTEN/PI3K/Akt 294:2186-9. immunodeficient mice. Nature 2007; 445:106-10. pathway regulates the side population phenotype and 35. Zhang J, Grindley JC, Yin T, Jayasinghe S, He XC, 8. Ricci-Vitiani L, Lombardi DG, Pilozzi E, Biffoni M, ABCG2 activity in glioma tumor stem-like cells. Cell Ross JT, et al. PTEN maintains haematopoietic stem Todaro M, Peschle C, et al. Identification and expan- Stem Cell 2009; 4:226-35. cells and acts in lineage choice and leukaemia preven- sion of human colon-cancer-initiating cells. Nature 22. Umar SB, Fleischer DE. Esophageal cancer: epidemi- tion. Nature 2006; 441:518-22. 2007; 445:111-5. ology, pathogenesis and prevention. Nat Clin Pract 36. Chen YL, Law PY, Loh HH. Inhibition of PI3K/Akt 9. Li C, Heidt DG, Dalerba P, Burant CF, Zhang L, Adsay Gastroenterol Hepatol 2008; 5:517-26. signaling: an emerging paradigm for targeted cancer V, et al. Identification of pancreatic cancer stem cells. therapy. Curr Med Chem Anticancer Agents 2005; 23. Syrjanen KJ. HPV infections and oesophageal cancer. J Cancer Res 2007; 67:1030-7. Clin Pathol 2002; 55:721-8. 5:575-89. 10. Schatton T, Murphy GF, Frank NY, Yamaura K, 37. Sinnberg T, Lasithiotakis K, Niessner H, Schittek 24. Mitsiades CS, Mitsiades N, Koutsilieris M. The Akt Waaga-Gasser AM, Gasser M, et al. Identification pathway: molecular targets for anti-cancer drug devel- B, Flaherty KT, Kulms D, et al. Inhibition of PI3K- of cells initiating human melanomas. Nature 2008; AKT-mTOR signaling sensitizes melanoma cells to opment. Curr Cancer Drug Targets 2004; 4:235-56. 451:345-9. cisplatin and temozolomide. J Invest Dermatol 2009; 25. Arlt A, Gehrz A, Muerkoster S, Vorndamm J, Kruse 129:1500-15. 11. Yang ZF, Ho DW, Ng MN, Lau CK, Yu WC, Ngai P, ML, Folsch UR, et al. Role of NFkappaB and Akt/ et al. Significance of CD90 cancer stem cells in human PI3K in the resistance of pancreatic carcinoma cell 38. Abedini MR, Muller EJ, Bergeron R, Gray DA, liver cancer. Cancer Cell 2008; 13:153-66. Tsang BK. Akt promotes chemoresistance in human lines against gemcitabine-induced cell death. Oncogene 12. Goodell MA, Rosenzweig M, Kim H, Marks DF, 2003; 22:3243-51. ovarian cancer cells by modulating cisplatin-induced, p53-dependent ubiquitination of FLICE-like inhibi- DeMaria M, Paradis G, et al. Dye efflux studies sug- 26. Yang X, Fraser M, Moll UM, Basak A, Tsang BK. Akt- gest that hematopoietic stem cells expressing low or tory protein. Oncogene 2010; 29:11-25. mediated cisplatin resistance in ovarian cancer: modula- undetectable levels of CD34 antigen exist in multiple 39. Li LW, Yu XY, Yang Y, Zhang CP, Guo LP, Lu tion of p53 action on caspase-dependent mitochondrial species. Nat Med 1997; 3:1337-45. death pathway. Cancer Res 2006; 66:3126-36. SH. Expression of esophageal cancer related gene 4 (ECRG4), a novel tumor suppressor gene, in esopha- 13. Kondo T, Setoguchi T, Taga T. Persistence of a small 27. Tokunaga E, Kimura Y, Oki E, Ueda N, Futatsugi subpopulation of cancer stem-like cells in the C6 glio- geal cancer and its inhibitory effect on the tumor M, Mashino K, et al. Akt is frequently activated in growth in vitro and in vivo. Int J Cancer 2009; ma cell line. Proc Natl Acad Sci USA 2004; 101:781-6. HER2/neu-positive breast cancers and associated with 125:1505-13. 14. Haraguchi N, Utsunomiya T, Inoue H, Tanaka F, poor prognosis among hormone-treated patients. Int J 40. Livak KJ, Schmittgen TD. Analysis of relative gene Mimori K, Barnard GF, et al. Characterization of a side Cancer 2006; 118:284-9. population of cancer cells from human gastrointestinal expression data using real-time quantitative PCR and 28. Fraser M, Bai T, Tsang BK. Akt promotes cisplatin the 2(-Delta Delta C(T)) Method. Methods 2001; system. Stem Cells 2006; 24:506-13. resistance in human ovarian cancer cells through inhi- 25:402-8. 15. Hirschmann-Jax C, Foster AE, Wulf GG, Nuchtern bition of p53 phosphorylation and nuclear function. Int J Cancer 2008; 122:534-46. JG, Jax TW, Gobel U, et al. A distinct “side population” of cells with high drug efflux capacity in human tumor cells. Proc Natl Acad Sci USA 2004; 101:14228-33. 958 Cancer Biology & Therapy Volume 11 Issue 11 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cancer Biology & Therapy Taylor & Francis

The PTEN/PI3K/Akt pathway regulates stem-like cells in primary esophageal carcinoma cells

Loading next page...
 
/lp/taylor-francis/the-pten-pi3k-akt-pathway-regulates-stem-like-cells-in-primary-mm5mHHtaV0

References (40)

Publisher
Taylor & Francis
Copyright
Copyright © 2011 Landes Bioscience
ISSN
1555-8576
eISSN
1538-4047
DOI
10.4161/cbt.11.11.15531
Publisher site
See Article on Publisher Site

Abstract

Resea RCh pape R Cancer Biology & Therapy 11:11, 950-958; June 1, 2011; © 2011 Landes Bioscience The PTEN/PI3K/Akt pathway regulates stem-like cells in primary esophageal carcinoma cells h ongxia Li, Quanli Gao, Liping Guo and s hih h sin Lu* s tate Key Laboratory of Molecular Oncology and Department of etiology and Carcinogenesis; Cancer Institute and h ospital; Chinese a cademy of Medical s ciences and peking Union Medical College; Beijing, China Current a ddress: h enan provincial Tumor h ospital; Zhengzhou, h enan China Key words: esophageal cancer, SP cells, PI3K/Akt pathway Abbreviations: ESCC, esophageal squamous cell carcinoma; SP, side population; NOD/SCID, non-obese diabetic/severe combined immunodeficiency; ABC, ATP-binding cassette; FBS, fetal bovine serum; PBS, phosphate buffer saline; CAMS&PUMC, Chinese Academy of Medical Science and Peking Union Medical College Recent reports have shown that cancer stem cells exist in many malignancies. s ide population (s p) cells are used to enrich cancer stem-like cells in many cell lines and fresh tumor specimens. In this study, we cultured primary esophageal squamous cell carcinoma (es CC) cells from es CC tissue specimens. sp cells from primary es CC cells were more resistant to chemotherapeutic reagents and formed more colonies in vitro than non-s p cells. In addition, xenograft experiments revealed that s p cells were more tumorigenic in vivo. Further results indicated that the pI3K/a kt pathway is essential to s p cells through the regulation of a BCG2 transporter function. Furthermore, pTeN, rather than mTOR, was found to be involved in sp cell regulation in primary es CC cells. These findings reveal that sp cells are enriched for cancer stem-like cells in primary es CC cells and that the pTeN/pI3K/a kt pathway regulates this stem-like population. This study indicates ©2011 Landes Bioscience. that sp cells in primary culture cells from tissue specimens could be a promising model for cancer stem cell research and may help researchers develop novel therapeutic strategies or efficient drugs that target es CC stem-like cells. Do not distribute. 4 7,8 brain tumor Introduction and colon cancer. Surface markers have not been identified in certain tumors, and SP sorting using Hoechst dye Recent evidence has suggested that only a small population of can enrich for cancer stem-like cells due to the ability of these cells cells in a tumor are responsible for the initiation, proliferation, to exclude the dye via membrane-bound pumps. SP cell sorting metastasis and recurrence of tumors as determined by a variety was initially used to identify hematopoietic stem cells; however, 1-3 of in vitro and in vivo assays. These cells have been defined as SP cells have been proven to be enriched in cancer stem-like cells 13-17 “cancer stem cells,” “cancer stem-like cells” or “tumor-initiating in various tumors and tumor cell lines. SP cells are detected cells.” by their ability to efu fl x Hoechst 33342 dye through adenosine Cancer stem-like cells represent a population of self-renewing triphosphate (ATP)-binding cassette (ABC) membrane trans- and drug-resistant cells that can survive treatment and repopu- porters. Indeed, the SP phenotype appears particularly valuable late the tumor. Experiments using hematopoietic and solid malig- for the future identification of specific stem cell markers. 3-11 nancies support tumor hierarchy as a fundamental concept in The mechanism for the SP phenotype involves several ABC tumor biology and suggest that this hierarchy is a novel cellular transporters such as proteins encoded by multi-drug resistant gene target for anticancer drug discovery. Previous reports have identi- 1 (MDR1), multidrug resistant associated protein 1 (MRP1) and fied cancer stem-like cells from fresh tumor specimens, which are BCRP1/ABCG2 (ABCP/MXR/BCRP) with overlapping prop- more representative of the cells in tumors compared with stem- erties in extruding cytotoxic drugs, as well as the fluorescent dye like cells isolated from cell lines. However, there have also been Hoechst, out of the cell. Further exploration on the maintenance 18,19 exciting reports on cancer stem-like cells sorted from cell lines in of the SP population includes ABCG2 or EGFR expression as 20,21 the study of regulation of stemness. In this study, we show that well as Akt and STAT3 pathway activation. primary cells from tumor specimens may be an additional choice Esophageal cancer is the sixth most common cause of cancer- for cancer stem cell research. related deaths worldwide, and its incidence is characterized by Molecular markers of cancer stem-like cells have been identified large geographical variation. ESCC is the most prevalent type 3 9 in some malignancies such as breast cancer, pancreatic cancer, in China, and most ESCC patients die from recurrence of the *Correspondence to: Shih Hsin Lu; Email: shlu@public.bta.net.cn Submitted: 12/09/10; Revised: 02/17/11; Accepted: 03/17/11 DOI: 10.4161/cbt.11.11.15531 950 Cancer Biology & Therapy Volume 11 Issue 11 Resea RCh pape R Resea RCh pape R ©2011 Landes Bioscience. Do not distribute. Figure 1. primary es CC cells from tumor specimen and sp analysis. (a ) primary cells of es CC tissue specimen. The image was obtained on passage 5. The cells were all adherent, possessed the typical epithelial polygonal morphology, and proliferated in a single layer (400x). (B) Representative image of h &e staining of xenograft tumors formed by primary es CC cells (100x). (C) Representative sp analysis of primary es CC cells. The sp fraction in pri- mary es CC cells was 2.4%. Incubation with 5 μM FTC abolished the sp cell fraction. cancer later in their life. Understanding the mechanisms under- ESCC. Adherent cells were visible after 7 days in the primary cul- lying carcinogenesis is essential for the development of novel ture of the tissue specimen. After 10 days, the solid particles were therapeutic strategies that target esophageal malignancies, few removed, and the cells were trypsinized at 90% conu fl ency and reports on esophageal cancer stem cells or the important signal- split 1:3. The cells were typical epithelial-like cells with polygo- ing pathways in this cancer exist. We previously reported that SP nal morphology and proliferated in a single layer (Fig. 1A). All cells are enriched in cancer stem-like cells in ESCC cell lines. of the experiments were performed on the cells within 15 pas- In this study, we established a method to obtain primary ESCC sages. The primary ESCC cells formed tumors when inoculated cells from tumor specimens of ESCC patients and showed that SP into immune-deficient mice, and pathologically, the tumors were cells from primary ESCC cells are enriched for cancer stem-like found to be squamous cell carcinoma (Fig. 1B). cells. Furthermore, our results indicate that the PTEN/PI3K/Akt To determine whether the primary ESCC cells contained SP signaling pathway plays a modulating role in the SP population in cells, we stained the ESCC cells with Hoechst 33342 and per- primary ESCC cells. Identic fi ation of esophageal cancer stem-like formed SP analysis. Representative results of the flow cytometry cells and further elucidation on the signaling pathways may pro- analysis are shown in Figure 1C. The primary ESCC cells con- vide strategies for the development of novel therapeutic strategies tained about 2.4% SP cells. The fraction of SP cells was decreased and efc fi ient drugs that target esophageal cancer stem-like cells. by treatment with fumitremorgin C (FTC), an ABC transporter inhibitor. We obtained similar results with two additional speci- Results mens by primary culture (Sup. Fig. S1). SP cells are more resistant to chemotherapeutic reagents SP cells in primary ESCC cells. Primary ESCC cells were than non-SP cells. We performed drug sensitivity assays with obtained from a surgical specimen of a 49-y-old female patient. sorted SP and non-SP cells that were treated with cisplatin or The pathology of the tumor was diagnosed as well-differentiated bortezomib by MTT assays. Cisplatin is commonly used for the www.landesbioscience.com Cancer Biology & Therapy 951 chemotherapy of ESCC, and bortezomib is a new chemothera- We collected cell lysates from SP and non-SP cells and performed peutic agent that targets the proteasome. After exposure with protein gel blot analyses. We found that the phospho-Akt (Ser473) 1 μg/ml cisplatin, the viability of the SP cells was markedly level in SP cells was much higher than in non-SP cells, while the higher compared to non-SP cells (p < 0.05). The results were total Akt level was similar in both groups (Fig. 2E). This find - similar in the case of bortezomib treatment (p < 0.05 for both 10 ing indicated that the PI3K/Akt pathway was activated in SP cells. and 25 nM). Because SP and non-SP cells from primary ESCC We tested the importance of the PI3K/Akt pathway in SP cells of cells did not differ in their growth rate (Sup. Fig. S2), SP cells primary ESCC cells using specic fi inhibitors including LY294002 were more resistant to 1 μg/ml of cisplatin as well as to 10 and (20 μM), Akt inhibitor IV (500 nM) and Akt inhibitor VIII (500 25 nM of bortezomib than non-SP cells (Fig. 2A). These results nM). After treatment for 3 days, the SP fraction from the cells was demonstrate the apparent chemoresistance of ESCC stem-like determined. The results show that all the inhibitors signic fi antly cells to anticancer drugs that may contribute to tumor recurrence attenuated the formation of the SP fraction of primary ESCC cells and resistance to multiple drugs. compared with DMSO-treated control cells. As shown in Figure SP cells highly express ABC transporter genes. It has been 3A, the percentage of cells in the SP cell fraction decreased after reported that the SP phenotype and the chemoresistance of SP treatment with LY294002, Akt inhibitor IV or Akt inhibitor VIII, cells depend primarily on ABC transporters such as ABCG2/ from 2.5 to 0.8, 0.1 and 0%, respectively. The Akt inhibitors IV BCRP, ABCB1/MDR1 and ABCA3. To investigate whether the and VIII, in particular, were very effective at reducing the number SP cells from primary cells had elevated levels of these genes, we of primary ESCC cells in the SP population. The inhibitory effect performed quantitative real-time PCR of the ABC transporter of these compounds on phopho-Akt was verie fi d by protein gel genes ABCA3, ABCB1, ABCC1 and ABCG2. As previously blotting and is shown in Figure 3B. reported in reference 17, these genes were preferentially expressed To determine whether the PI3K/Akt pathway was important in SP cells compared with non-SP cells (Fig. 2B), indicating that to the colony formation of the primary ESCC cells, we performed the expression of ABC transporter genes was responsible for the soft agar colony formation assays using the PI3K/Akt inhibi- SP and chemoresistance phenotypes from primary ESCC cells. tors. Pretreatment of primary ESCC cells with the inhibitors as We evaluated the role of ABCG2 in mediating drug efu fl x of described above reduced the colony-formation rate of the primary primary ESCC cells. It has been shown that mitoxantrone is a ESCC cells as compared with the DMSO-treated control cells. substrate of ABCG2. I© n the m2 it0 oxan1 tron1 e e fu flL x asa say, tn he id ntra-es Th e cB oloni y fo orms atic on ri ate e of tn he uc ne treat. ed control group was cellular u fl orescence intensity can be measured by flow cytometry 7.42 ± 1.12%; for the LY294002 treated group, the rate was 5.64 and used to determine ABCG2 activity. As shown in Figure 2D, ± 0.87% (p = 0.003); for the Akt inhibitor IV, the rate was 4.82 ± the intracellular fluorescence intensity of SP cells was lower (28.4 0.56% (p = 0); and for the Akt inhibitor VIII, the rate was 6.24 Do not distribute. ± 2.4) than non-SP cells (56.3 ± 9.3, p = 0.007) and unsorted cells ± 0.32% (p = 0.014) (Fig. 3C). (55.5 ± 11.2 p = 0.008), indicating that the drug was pumped out The Akt pathway regulates ABCG2 activity in ESCC pri- more efficiently in SP cells. mary cells. Because it has been shown that PI3K/Akt signaling SP cells have higher colony-formation ability and are more regulates the SP phenotype in glioma cells by regulating ABCG2 tumorigenic than non-SP cells in mice. To determine whether function, we determined whether this mechanism exists in pri- the SP cells from primary ESCC cells were enriched for cancer mary ESCC cells. Mitoxantrone is a substrate of ABCG2 and can stem-like cells, we tested the tumorigenic potential of SP and be detected by flow cytometry as a measure of ABCG2 function. non-SP cells both in vitro and in vivo. Colony formation ability Primary ESCC cells were treated with LY294002 (20 μM), Akt in soft agar assay revealed that approximately 10.6% of 1,500 SP inhibitor IV (500 nM) or Akt inhibitor VIII (500 nM) for 3 days cells were capable of forming visible colonies, which was almost and subsequently assayed for their ability to efu fl x mitoxantrone. two and a half times more than non-SP cells (about 4%) (Fig. The intracellular intensity of DMSO-treated control cells was 2C). Furthermore, in vivo tumorigenic ability was evaluated by comparatively low (mean fluorescence intensity of 59.34 ± 10.69) injecting both SP and non-SP cells, mixed with matrigel matrix, because mitoxantrone was pumped out by the ABCG2 transporter. into NOD/SCID mice. As shown in Table 2, the tumorigenic- Incubation with the PI3 kinase inhibitor LY294002 or the Akt ity of SP cells was significantly higher than that of non-SP cells. inhibitors IV or VIII increased the intracellular fluorescence inten - Seven out of eight mice inoculated with 100 SP cells gave rise sity of primary ESCC cells (mean value of 110.03 ± 5.95, 122.55 ± to tumors, whereas only one out of eight inoculations of 1,000 13.74 and 147.81 ± 17.32, respectively, p < 0.05) (Fig. 3D). Because non-SP cells formed a tumor. These results indicate that as few the expression of the ABC transporters did not change after treat- as 100 SP cells were tumorigenic in NOD/SCID mice, whereas ment with the PI3K/Akt inhibitors (Sup. Fig. S4), the increased ten times the number of non-SP cells generated very few tumors. intracellular fluorescence intensity indicated decreased ABCG2 The pathology results confirmed that the tumors formed by SP function to efu fl x the drug. These results are consistent with the and non-SP cells were all ESCC (Sup. Fig. S3). changes in the SP fraction upon treatment with the PI3K/Akt The PI3K/Akt pathway regulates the SP phenotype and pro- inhibitors. These data reveal that the PI3K/Akt pathway plays an motes the tumorigenicity of primary ESCC cells. As reported, important role in regulation of the SP fraction. PI3K/Akt pathway plays an important role in the maintenance PTEN, but not mTOR, regulates the SP phenotype in ESCC 20,21 of SP cells. Therefore, we determined the expression level of primary cells. PTEN plays an important role in the regulation of components of the PI3K/Akt pathway in ESCC stem-like cells. SP cells in glioma malignancy and breast cancer by negatively 952 Cancer Biology & Therapy Volume 11 Issue 11 ©2011 Landes Bioscience. Do not distribute. Figure 2. sp cells in primary es CC cells. (a ) s ensitivity of sp and non-sp cells to DDp and bortezomib. sp cells were more resistant to the chemothera- peutic reagents upon 48 hours of treatment. *p < 0.05, bars: se M. (B) Relative mRNa expression of the a BC transporters in sp and non-sp cells. The expression level was normalized to Ga pDh . The a BC transporter genes a BCa 3, a BCB1, a BCC1 and a BCG2 were all highly expressed in sp cells. (C) Representative images from colony-formation assay from sp and non-sp cells. Fifteen-hundred freshly sorted cells were seeded in each well of a 6-well plate in soft agar and the images were obtained 3 weeks later. sp cells formed a greater number of colonies. *p < 0.05, t-test. (D) Flow cytom- etry histograms of mitoxantrone efflux from sp, non-sp cells and unsorted cells. The intracellular fluorescence in the sp fraction was lower than in the non-sp and unsorted cell fractions. (e) protein gel blotting of phosphorylated/total a kt, phosphorylated/total mTOR and phosphorylated/total pTeN in sorted sp and non-sp primary es CC cells. www.landesbioscience.com Cancer Biology & Therapy 953 ©2011 Landes Bioscience. Do not distribute. Figure 3. pI3K/a kt inhibitors decrease the sp fraction. (a ) sp analysis of primary es CC cells treated with pI3K/a kt inhibitors. Compared to control cells treated with DMs O, treatment with LY294002, a kt inhibitor IV and a kt inhibitor VIII for 3 days significantly reduced the sp fraction in primary es CC cells. (B) protein gel blotting of total a kt and phospho-a kt in primary cells treated with DMs O, LY294002, a kt inhibitor IV and a kt inhibitor VIII for 3 days. β-actin was used as a loading control. The level of phospho-a kt was decreased by the treatment with the inhibitors. (C) Inhibition of colony formation by the a KT/pI3K inhibitors. Treatment with LY294002, a kt inhibitor IV and a kt inhibitor VIII reduced the number of colonies of primary cells formed in soft agar. *p < 0.05, t-test. (D) Flow cytometry histograms of mitoxantrone efflux from untreated cells and inhibitor treated cells. Blocking pI3K/a kt pathways reduced the ability of a BCG2 to efflux mitoxantrone. 20,21 regulating PI3K/Akt signaling. Therefore, PTEN may also mTOR in the regulation of SP cells in primary ESCC cells, small- play an important role in ESCC cancer stem-like cells. mTOR is molecule inhibitors and RNA interference were used. Primary a downstream regulator of the PI3K/Akt pathway and has been ESCC cells were treated with the mTOR inhibitor rapamycin reported to regulate SP cells in breast cancers. As shown in (5 μM) or the PTEN specic fi inhibitor bpV(pic) (20 μM) for Figure 2E, the level of phospho-PTEN in non-SP cells was higher 3 days and analyzed for the presence of the SP fraction. The SP than in SP cells, while the level of phospho-mTOR in SP and fraction increased after treatment with the PTEN-specic i fi nhibi - non-SP cells was similar. To determine the effect of PTEN and tor bpV(pic) compared to control cells (6.8 vs. 2.6%); however, 954 Cancer Biology & Therapy Volume 11 Issue 11 ©2011 Landes Bioscience. Do not distribute. Figure 4. The ee ff ct of pTeN and mTOR on sp cells in primary es CC cells. (a ) sp analysis of primary es CC cells treated with mTOR and pTeN inhibitors or with knockdown of mTOR and pTeN by RNa interference. Inhibition of pTeN increased the sp fraction while mTOR inhibition did not alter the level of the sp fraction. (B) protein gel blotting of mTOR and pTeN level after RNa interference. Forty-eight hours after transfection with mTOR or pTeN siRNa , the level of mTOR and pTeN were decreased. β-actin was used as a loading control. (C) eff ect of mTOR and pTeN inhibitors on colony formation of primary es CC cells. BpV(pic), a pTeN inhibitor, increased the colony-forming ability of primary cells, whereas rapamycin had no ee ff ct on colony formation. *p < 0.05, t-test. the SP fraction did not change when treated with rapamycin (2.8 with the PTEN siR NA, whereas the SP fraction did not change vs. 2.6%) (Fig. 4A). after transfection with the mTOR siRNA when compared RNA interference was used to confirm the effect of PTEN with the control cells that were transfected with non-silencing and mTOR on the SP phenotype. Forty-eight hours after siRNA. The results are consistent with the results obtained transfection, the cells were collected for SP analysis. siRNA- with the inhibitors. mediated silencing was validated by decreased levels of the The mTOR and PTEN inhibitors were also used to deter- PTEN and mTOR proteins and is shown in Figure 4B. As mine their effect on colony forming ability of primary ESCC shown in Figure 4A, the SP fraction increased after transfection cells. When the cells were pretreated with bpV(pic), the www.landesbioscience.com Cancer Biology & Therapy 955 colony-formation rate of the primary ESCC cells increased com- function. In our study, we demonstrated that PI3K/Akt signal- pared to control cells (9.93 ± 0.20% vs. 7.42 ± 1.12% , p = 0.047). ing pathway was also important for the ESCC SP cell phenotype. After the treatment with rapamycin, the colony formation rate of Using PI3K/Akt inhibitors, we found that they significantly the cells was similar to that of control cells (7.63 ± 1.80% vs. 7.42 reduced the SP cell fraction, which is consistent with the reports ± 1.12%, p = 0.833) (Fig. 4C). on cancer stem-like cells in other malignancies. Furthermore, the tumorigenicity was also reduced upon inhibition of PI3K/ Discussion Akt signaling, which was most likely a result of the inhibition of ESCC stem-like cells. PI3K/Akt signaling activity is vital in the Most cancer stem cells were initially identiefi d in fresh tumor maintenance of the cells in the SP fraction and in maintaining specimens from surgery, but some cancer cell lines have also been the high tumorigenicity of primary ESCC cells. 13-17 shown to be enriched with cancer stem-like cells. It is thought mTOR is emerging as a major downstream target of Akt in the that cancer stem-like cells from cell lines may be a promising regulation of the cellular response to nutrients and other stim- model for cancer stem cell research as a result of their unlimited uli. Loss of PTEN function leads to increased concentrations of supply and ease of handling. Fresh tumor specimens of patients are PIP3, the main in vivo substrate for PTEN, resulting in constitu- not always available for all research groups. In addition, the cells tive activation of downstream components of the PI3K pathway, might be difficult to culture, especially after the digestion, FACS including the Akt and mTOR kinases. Loss of PTEN, a PI3K analysis and sorting, and the results may be invalid. However, inhibitor, increases the proliferation and survival of neural and 34,35 many available cell lines have been established and cultured for hematopoietic stem cells as well as increases the SP fraction 21,30 years, even decades, through numerous passages. Tumor cells in glioma cells and prostate cancer stem-like cells. Inhibition from primary cultures of fresh tissue specimens provide us with a of mTOR with the specific inhibitor rapamycin or with knock - more physiological model than cell lines to study cancer stem-like down by RNA interference did not change the SP fraction or cell biology. We have determined that ESCC tissue specimens can the tumorigenicity of the primary ESCC cells, whereas PTEN be maintained as primary cultures for several weeks to months inhibition increased both the SP fraction and the tumorigenicity to obtain a sufc fi ient number of cells for many cellular biologi - of primary ESCC cells. All these results indicate that PTEN, the cal and molecular analyses. The cells are easier to handle, and loss of which could upregulate the SP fraction, is a key regula- the results obtained f© rom th2 ese c0 ells a1 re m1 o re rL epa rodun cibld e thae n s tor i n tB he mio aints enac nce oi f pe rin mary Ec Se CC s. tem-like cells, while results obtained directly from tissue specimens. In addition, cells mTOR does not appear to play a major role in maintenance of obtained from primary tissue better approximate tumors in vivo the SP fraction. than cell lines. The primary tumor cells retained characteristics Components of PI3K/Akt pathway have emerged as prom- Do not distribute. of the original tumors when cultured in serum-free conditions, ising novel targets for the development of cancer therapy. In which avoids stimulation by serum. Stem-like cells from ESCC medical oncology, chemoresistance is a major hurdle for success- primary cultures may be a promising model for stem cell research ful cancer therapy. Because Akt activation is also linked to drug in ESCC. Additionally, patient specic c fi ancer stem-like cells may resistance in many cancers, targeting this pathway could restore 37,38 be used in clinical research or for treatment in the future. drug sensitivity. The results of the mitoxantrone efu fl x assay We previously showed that SP cells in ESCC cell lines were in primary cells in the presence of above inhibitors indicate that enriched with cancer stem-like cells. In this study, we obtained the function of the ABC transporters is modulated by the PI3K/ primary cells from ESCC tissue specimens and found that the Akt pathway. These data further explain the connections among percentage of SP cells within the tissue specimen was similar to the Akt pathway, stem-like characteristics and therapeutic resis- that found in cell lines. Our results indicate that SP cells in pri- tance, suggesting that activation of this pathway also enhances mary ESCC cells are enriched for cancer stem-like cells, which the ability of ESCC stem-like cells to efu fl x drugs. This model is consistent with the results obtained from ESCC cell lines and is consistent with reports on the ABC transporters regulated other reports on SP cells. by PI3K/Akt activity in hepatocellular carcinoma and glioma 21,31 The PI3K/Akt signaling cascade is frequently disrupted in malignancy. many human cancers, and it is a key player in mediating tumor A good understanding of cancer stem-like cell biology will cell survival and escape from apoptosis. In light of the wide- provide the possibility of developing cancer stem-like cell tar- ranging biological consequences, this pathway is considered a key geted therapy, which will be helpful in efficiently eradicating determining factor in tumor aggressiveness and is an attractive tumors. Blocking cancer stem-like cell survival signaling is one target for therapeutic intervention. It has been reported that of the most efficient ways in stem cell-targeted therapies. We PI3K/Akt signaling is associated with chemoresistance in many demonstrate that the PTEN/PI3K/Akt signaling pathway is 25-28 malignances and regulates the SP cell phenotype via ABCG2 essential for the SP phenotype and tumorigenicity for primary translocation in hematopoietic stem cells. Recent reports have ESCC cells and regulates the function of ABC transporters asso- indicated that activation of the PI3K/Akt signaling pathway ciated with the SP phenotype. These results provide evidence is required for maintaining cancer stem-like cell viability and that PTEN/PI3K/Akt signaling may be a potential target for maintenance in breast cancer, prostate cancer, hepatocellular eliminating cancer stem-like cells in ESCC and may promote 20,21,30,31 carcinoma and brain tumor cell lines. Additionally, the the development of new therapeutic strategies and efc fi ient drugs PI3K/Akt signaling pathway has been found to regulate ABCG2 to target ESCC stem-like cells. 956 Cancer Biology & Therapy Volume 11 Issue 11 Table 1. primer sequences for specific gene amplification Table 2. Tumorigenicity assay of primary sp and non-sp cells in NOD/ s CID mice Gene name Primers Cell number SP non-SP 5'-a Ga aa Ta CG GTG CCG GCT a TC a Ca -3' (forward) a BCa 3 3 -/- 5'-Ca a TGC CCa GCT CTT TCT GCT TCT-3' (reverse) 5 x 10 2/2 2 x 10 4/4 3/4 5'-GCT CCT Ga C Ta T GCC aaa GC-3' (forward) a BCB1 5'-TCT TCa CCT CCa GGC TCa GT-3' (reverse) 3 1 x 10 4/4 1/8 5'-CTG GGC TTa TTT CGG a TC aa -3' (forward) 2 -/- 1 x 10 7/8 a BCC1 5'-TGa a TG GGT CCa GGT TCa TT-3' (reverse) 5'-Ca C CTT a TT GGC CTC a GGa a -3' (forward) a BCG2 culture, the absorption values were determined using a Bio-Rad 5'-CCT GCT TGG aa G GCT CTa TG-3' (reverse) enzyme reader. 5'-GTG Ga C CTG a CC TGC CGT CT-3' (forward) Ga pDh Soft agar colony formation assay. Soft agar colony formation 5'-GGa GGa GTG GGT GTC GCT GT-3' (reverse) was assessed using a modified method as previously described in reference 39. Cells were suspended in 0.3% agar/culture medium Material and Methods and plated at a density of 1,500 cells per well in a 6-well plate that had been previously coated with 0.5% agar. Fresh medium Primary culture of ESCC specimens. ESCC specimens were was added to each well every 3 days. After 3 weeks of incubation, obtained from patients undergoing surgical resection for esopha- colonies were stained with 0.2% p-iodonitrotetrazolium violet geal cancer subsequent to providing informed consent. This (Amersco) and counted. study was approved by the institutional review board of Cancer Xenograft assay in mice. Nonobese diabetic (NOD)/severe Hospital, Chinese Academy of Medical Sciences and Peking combined immunodeficient (SCID) mice were kept in microiso - Union Medical College (CAMS&PUMC). The tissue specimens lator cages according to the guidelines of CAMS&PUMC, and were received within an hour after surgery, cut into small pieces all experiments were approved by the animal care committee of and rinsed with ice-cold calcium/magnesium free phosphate buf- CAMS&PUMC. The freshly sorted cells were injected subcu- fer saline (PBS). The tissue pieces were digested for about 2 h taneously into the left axillary fossa of female mice (3–4 weeks at 37°C in an incubat© or wi2 th 1 m0 g/1 ml c1 oll agL enaa se I (n Sigmd a) ie n s old ) iB n 20i 0 o μl ms atc rigei l me atrn ix (c BDe ) dil. uted in PBS at a 1:1 serum-free DMEM medium (Invitrogen). The digested tissue ratio. The mice were monitored twice a week for palpable tumor was subsequently centrifuged and collected for further culture formation and were sacrificed at 8–12 weeks after transplanta - in serum-free medium (Invitrogen) supplemented with 20 ng/ tion to determine tumor formation. Tumors were fixed in 10% Do not distribute. ml EGF (PeproTech), 20 ng/ml bFGF (PeproTech), penicillin G buffered formalin and paraffin-embedded; then, the slides were (100 units/ml) and streptomycin (100 μg/ml). The medium was prepared for H&E staining. changed every 2 or 3 days until the epithelial cells attached to the Protein gel blot analysis. Cells were harvested and lysed dishes and expanded. Cells were trypsinized at 90% conu fl ency in lysis buffer, and protein gel blot analysis was performed as and split 1:3. previously described in reference 39. Equivalent amounts of Analyzing and sorting of primary ESCC cells by FACS. protein were separated by SDS-PAGE and transferred to poly- The procedure for SP analysis is based on procedures previously vinylidene diu fl oride membranes (Millipore). After blocking, described in reference 16, with modification. Cells were detached the membranes were incubated with the appropriate diluted pri- from plates with trypsin, washed with PBS, resuspended in mary antibodies, including Akt, phospho-Akt (p-Akt, Ser473), serum-free medium at 1 x 10 cells/ml and incubated for 90 min mTOR, phospho-mTOR (p-mTOR, Ser2448), PTEN, phos- at 37°C with Hoechst 33342 (Sigma) at a final concentration of pho-PTEN (p-PTEN, Ser380) (Cell Signaling Technology) and 5 μg/ml either alone or in the presence of 5 μM FTC (Sigma). β-actin (Sigma). The signal of the target protein was detected by After incubation, cells were washed with ice-cold PBS and fil - an enhanced chemiluminescence detection system (Pierce) and tered through a 40 μm cell strainer. Propidium iodide (Sigma) recorded on film in the linear detection range. was added at a final concentration of 1 μ g/ml for 5 min before RNA extraction and quantitative real-time PCR. Total analysis. Cells were sorted using dual-wavelength analysis with RNA was extracted using the TRIzol reagent (Invitrogen) and BD FACSVantage SE equipped with a Coherent Innova 300C reverse transcribed into cDNA using the SuperScript 3 First laser. The SP gate in the flow cytometry analysis was defined as strand cDNA Synthesis Kit (Invitrogen). Primer sequences for the diminished area on the dot plot in the presence of FTC. SP specific gene amplification are shown in Table 1 . Real-time PCR and non-SP cells were collected for further study. was carried out according to the standard protocol of use with TM Sensitivity to chemotherapeutic reagents. Sensitivity to che- the SYBR Premix Ex Taq Perfect Real Time system (Takara, motherapeutic reagents of SP and non-SP cells was assessed using Dalian, China) on an ABI 7300 detector (Applied BioSystems). -ΔΔCt the MTT assay as previously described in reference 17. Briefly, Fold changes in gene expression were calculated using the 2 3 40 2 x 10 cells per well were seeded on 96-well plate and cul- method. GAPDH was used as an internal control. tured overnight. Cisplatin (0.5 or 1 μg/ml) and bortezomib Flow cytometric detection of mitoxantrone efufl x. The (10 or 25 nM) were chosen as representative chemotherapeutic activity of the ABCG2 pump was tested by mitoxantrone efu fl x agents and were added to the medium individually. After 48 h in analysis as previously described in reference 21. Cells were www.landesbioscience.com Cancer Biology & Therapy 957 6 trypsinized and resuspended at 1 x 10 cells/ml. Then, mitoxan- at a concentration of 100 nM, and the cells were collected for trone was added at the final concentration of 200 nM for 1 h at further study 48 h after transfection. Silencing was validated by 37°C, 5% CO . The cells were subsequently washed two times protein gel blot. with ice-cold PBS. The fluorescence of the intracellular mitoxan - Statistical analysis. Statistical software SPSS12.0 was used trone from 10,000-gated events was logarithmically measured at in data processing and for analyzing the significance between a laser-excitation wavelength of 633 nm using a 670 nm bandpass groups with the t-test. p < 0.05 was considered statistically filter on a BD LSR analyzer. significant. RNA interference synthesis and transfection. Small- Acknowledgments interfering RNA (siRNA) and non-silencing control siRNA were synthesized and annealed by Genechem Company. The siRNA This study has been supported by State Key Basic Research sequences used for PTEN and mTOR were: 5'-CAG UAG AGG Programs 973 of China (2009CB521803). AGC CGU CAA ATT-3' and 5'-GAG GAG UCU ACU CGC Note UUC UAU TT-3', respectively. Cell transfections were per- formed using LipofectAMINE 2000 (Invitrogen) according to Supplemental materials can be found at: the manufacture’s instruction. Briefly, siRNA was transfected http://www.landesbioscience.com/journals/cbt/article/15531/ 29. Mogi M, Yang J, Lambert JF, Colvin GA, Shiojima I, 16. Wang J, Guo LP, Chen LZ, Zeng YX, Lu SH. References Identification of cancer stem cell-like side population Skurk C, et al. Akt signaling regulates side population 1. Lapidot T, Sirard C, Vormoor J, Murdoch B, Hoang T, cell phenotype via Bcrp1 translocation. J Biol Chem cells in human nasopharyngeal carcinoma cell line. Caceres-Cortes J, et al. A cell initiating human acute Cancer Res 2007; 67:3716-24. 2003; 278:39068-75. myeloid leukaemia after transplantation into SCID 30. Dubrovska A, Kim S, Salamone RJ, Walker JR, Maira 17. Huang D, Gao Q, Guo L, Zhang C, Jiang W, Li H, mice. Nature 1994; 367:645-8. et al. Isolation and identification of cancer stem-like SM, Garcia-Echeverria C, et al. The role of PTEN/ 2. Reya T, Morrison SJ, Clarke MF, Weissman IL. Stem Akt/PI3K signaling in the maintenance and viability cells in esophageal carcinoma cell lines. Stem Cells Dev cells, cancer and cancer stem cells. Nature 2001; 2009; 18:465-73. of prostate cancer stem-like cell populations. Proc Natl 414:105-11. Acad Sci USA 2009; 106:268-73. 18. Kim M, Turnquist H, Jackson J, Sgagias M, Yan Y, 3. Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison Gong M, et al. The multidrug resistance transporter 31. Hu C, Li H, Li J, Zhu Z, Yin S, Hao X, et al. Analysis SJ, Clarke MF. Prospective identification of tumori- of ABCG2 expression and side population identifies ABCG2 (breast cancer resistance protein 1) effluxes genic breast cancer cells. Proc Natl Acad Sci USA 2003; Hoechst 33342 and is overexpressed in hematopoietic intrinsic drug efflux in the HCC cell line MHCC-97L 100:3983-8. and its modulation by Akt signaling. Carcinogenesis stem cells. Clin Cancer Res 2002; 8:22-8. 4. Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani 2008; 29:2289-97. 19. Chen JS, Pardo FS, Wang-Rodriguez J, Chu TS, ©2011 Landes Bioscience. J, Hide T, et al. Identification of human brain tumour 32. Bjornsti MA, Houghton PJ. Lost in translation: dys- Lopez JP, Aguilera J, et al. EGFR regulates the side initiating cells. Nature 2004; 432:396-401. population in head and neck squamous cell carcinoma. regulation of cap-dependent translation and cancer. 5. Bonnet D, Dick JE. Human acute myeloid leukemia is Cancer Cell 2004; 5:519-23. Laryngoscope 2006; 116:401-6. organized as a hierarchy that originates from a primitive 20. Zhou J, Wulfkuhle J, Zhang H, Gu P, Yang Y, Deng J, 33. Di Cristofano A, Pandolfi PP. The multiple roles of Do not distribute. hematopoietic cell. Nat Med 1997; 3:730-7. PTEN in tumor suppression. Cell 2000; 100:387-90. et al. Activation of the PTEN/mTOR/STAT3 pathway 6. Collins AT, Berry PA, Hyde C, Stower MJ, Maitland in breast cancer stem-like cells is required for viabil- 34. Groszer M, Erickson R, Scripture-Adams DD, Lesche NJ. Prospective identification of tumorigenic prostate ity and maintenance. Proc Natl Acad Sci USA 2007; R, Trumpp A, Zack JA, et al. Negative regulation cancer stem cells. Cancer Res 2005; 65:10946-51. 104:16158-63. of neural stem/progenitor cell proliferation by the 7. O’Brien CA, Pollett A, Gallinger S, Dick JE. A human Pten tumor suppressor gene in vivo. Science 2001; 21. Bleau AM, Hambardzumyan D, Ozawa T, Fomchenko colon cancer cell capable of initiating tumour growth in EI, Huse JT, Brennan CW, et al. PTEN/PI3K/Akt 294:2186-9. immunodeficient mice. Nature 2007; 445:106-10. pathway regulates the side population phenotype and 35. Zhang J, Grindley JC, Yin T, Jayasinghe S, He XC, 8. Ricci-Vitiani L, Lombardi DG, Pilozzi E, Biffoni M, ABCG2 activity in glioma tumor stem-like cells. Cell Ross JT, et al. PTEN maintains haematopoietic stem Todaro M, Peschle C, et al. Identification and expan- Stem Cell 2009; 4:226-35. cells and acts in lineage choice and leukaemia preven- sion of human colon-cancer-initiating cells. Nature 22. Umar SB, Fleischer DE. Esophageal cancer: epidemi- tion. Nature 2006; 441:518-22. 2007; 445:111-5. ology, pathogenesis and prevention. Nat Clin Pract 36. Chen YL, Law PY, Loh HH. Inhibition of PI3K/Akt 9. Li C, Heidt DG, Dalerba P, Burant CF, Zhang L, Adsay Gastroenterol Hepatol 2008; 5:517-26. signaling: an emerging paradigm for targeted cancer V, et al. Identification of pancreatic cancer stem cells. therapy. Curr Med Chem Anticancer Agents 2005; 23. Syrjanen KJ. HPV infections and oesophageal cancer. J Cancer Res 2007; 67:1030-7. Clin Pathol 2002; 55:721-8. 5:575-89. 10. Schatton T, Murphy GF, Frank NY, Yamaura K, 37. Sinnberg T, Lasithiotakis K, Niessner H, Schittek 24. Mitsiades CS, Mitsiades N, Koutsilieris M. The Akt Waaga-Gasser AM, Gasser M, et al. Identification pathway: molecular targets for anti-cancer drug devel- B, Flaherty KT, Kulms D, et al. Inhibition of PI3K- of cells initiating human melanomas. Nature 2008; AKT-mTOR signaling sensitizes melanoma cells to opment. Curr Cancer Drug Targets 2004; 4:235-56. 451:345-9. cisplatin and temozolomide. J Invest Dermatol 2009; 25. Arlt A, Gehrz A, Muerkoster S, Vorndamm J, Kruse 129:1500-15. 11. Yang ZF, Ho DW, Ng MN, Lau CK, Yu WC, Ngai P, ML, Folsch UR, et al. Role of NFkappaB and Akt/ et al. Significance of CD90 cancer stem cells in human PI3K in the resistance of pancreatic carcinoma cell 38. Abedini MR, Muller EJ, Bergeron R, Gray DA, liver cancer. Cancer Cell 2008; 13:153-66. Tsang BK. Akt promotes chemoresistance in human lines against gemcitabine-induced cell death. Oncogene 12. Goodell MA, Rosenzweig M, Kim H, Marks DF, 2003; 22:3243-51. ovarian cancer cells by modulating cisplatin-induced, p53-dependent ubiquitination of FLICE-like inhibi- DeMaria M, Paradis G, et al. Dye efflux studies sug- 26. Yang X, Fraser M, Moll UM, Basak A, Tsang BK. Akt- gest that hematopoietic stem cells expressing low or tory protein. Oncogene 2010; 29:11-25. mediated cisplatin resistance in ovarian cancer: modula- undetectable levels of CD34 antigen exist in multiple 39. Li LW, Yu XY, Yang Y, Zhang CP, Guo LP, Lu tion of p53 action on caspase-dependent mitochondrial species. Nat Med 1997; 3:1337-45. death pathway. Cancer Res 2006; 66:3126-36. SH. Expression of esophageal cancer related gene 4 (ECRG4), a novel tumor suppressor gene, in esopha- 13. Kondo T, Setoguchi T, Taga T. Persistence of a small 27. Tokunaga E, Kimura Y, Oki E, Ueda N, Futatsugi subpopulation of cancer stem-like cells in the C6 glio- geal cancer and its inhibitory effect on the tumor M, Mashino K, et al. Akt is frequently activated in growth in vitro and in vivo. Int J Cancer 2009; ma cell line. Proc Natl Acad Sci USA 2004; 101:781-6. HER2/neu-positive breast cancers and associated with 125:1505-13. 14. Haraguchi N, Utsunomiya T, Inoue H, Tanaka F, poor prognosis among hormone-treated patients. Int J 40. Livak KJ, Schmittgen TD. Analysis of relative gene Mimori K, Barnard GF, et al. Characterization of a side Cancer 2006; 118:284-9. population of cancer cells from human gastrointestinal expression data using real-time quantitative PCR and 28. Fraser M, Bai T, Tsang BK. Akt promotes cisplatin the 2(-Delta Delta C(T)) Method. Methods 2001; system. Stem Cells 2006; 24:506-13. resistance in human ovarian cancer cells through inhi- 25:402-8. 15. Hirschmann-Jax C, Foster AE, Wulf GG, Nuchtern bition of p53 phosphorylation and nuclear function. Int J Cancer 2008; 122:534-46. JG, Jax TW, Gobel U, et al. A distinct “side population” of cells with high drug efflux capacity in human tumor cells. Proc Natl Acad Sci USA 2004; 101:14228-33. 958 Cancer Biology & Therapy Volume 11 Issue 11

Journal

Cancer Biology & TherapyTaylor & Francis

Published: Jun 1, 2011

There are no references for this article.