Cancer Letters 357 (2015) 429–437

Contents lists available at ScienceDirect

Cancer Letters j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / c a n l e t

Mini-review

Pancreatic cancer stem cells: New insight into a stubborn disease Han-xiang Zhan a, Jian-wei Xu a, Dong Wu a, Tai-ping Zhang b,*, San-yuan Hu a,** a

Department of General Surgery, Qilu Hospital, Shandong University, Jinan, Shandong Province, 250012, China Department of General Surgery, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100730, China b

A R T I C L E

I N F O

Article history: Received 18 October 2014 Received in revised form 30 November 2014 Accepted 2 December 2014 Keywords: Cancer stem cells Pancreatic cancer microRNA Chemotherapy resistance Metastasis

A B S T R A C T

Resistance to conventional therapy and early distant metastasis contribute to the unsatisfactory prognosis of patients with pancreatic cancer. The concept of cancer stem cells (CSCs) brings new insights into cancer biology and therapy. Many studies have confirmed the important role of these stem cells in carcinogenesis and the development of hematopoietic and solid cancers. Recent studies have shown that CSCs regulate aggressive behavior, recurrence, and drug resistance in pancreatic cancer. Here, we review recent advances in pancreatic cancer stem cells (PCSCs) research. Particular attention is paid to the regulation mechanisms of pancreatic cancer stem cell functions, such as stemness-related signaling pathways, microRNAs, the epithelial-mesenchymal transition (EMT), and the tumor microenvironment, and the development of novel PCSCs targeted therapy. We seek to further understand PCSCs and explore potential therapeutic targets for pancreatic cancer. © 2014 Elsevier Ireland Ltd. All rights reserved.

Introduction Pancreatic ductal adenocarcinoma (PDAC) is the fourth leading cause of cancer death in the United States and has low overall survival [1,2]. Resection is the only radical treatment for this aggressive malignancy. Unfortunately, only 15–20% of patients can be resected, due to the low rate of early diagnosis. Resistance to chemotherapy and radiation therapy also contributes to the dismal prognosis [3–5]. Only 9.4% of patients with metastatic pancreatic cancers acquired partial response, and 34.5% experienced disease progression during chemotherapy with gemcitabine [3]. During the past decades, the carcinogenesis and pathogenesis of pancreatic cancer (PC) have been thoroughly described. However, the overall survival of patients with pancreatic cancer has not significantly improved. An in-depth investigation of the mechanisms of carcinogenesis, progression, and drug resistance that is based on the theory of cancer stem cells (CSCs) might help to improve pancreatic cancer prognosis. CSCs have been confirmed in hematopoietic and solid tumors, which have the capacity of self-renewal and can differentiate into other cell types of cancer cells [6]. These types of cells promote tumor growth, invasion, metastasis, and therapeutic resistance [7–9]. Pancreatic cancer stem cells (PCSCs) were first identified by Li et al. in

* Corresponding author. Tel.: +86-10-65296007; fax: +86-10-65124875. E-mail address: [email protected] (T. Zhang) ** Corresponding author. Tel.: +86-531-82166351; fax: +86-531-82166009. E-mail address: [email protected] (S. Hu) http://dx.doi.org/10.1016/j.canlet.2014.12.004 0304-3835/© 2014 Elsevier Ireland Ltd. All rights reserved.

2007 [10]. Later studies revealed that PCSCs could self-renew, differentiate, and divide asymmetrically, like normal stem cells; the critical role of PCSCs in regulating pancreatic cancer progression, metastasis, and drug resistance has since been confirmed [11,12]. However, some critical questions still remain to be answered, including the identification of specific PCSCs, the signaling pathways involved in regulating PCSCs, and the regulatory roles of PCSCs in pancreatic cancer. In this article, we reviewed the recent advances regarding the regulating roles and mechanisms of CSCs in pancreatic cancer, paying particular attention to the metastasis, drug resistance, and prognostic functions of these cells. Cancer stem cells Human cancer cells are composed of a heterogeneous population of cells and are characterized by unlimited proliferation and resistance to conventional therapy. However, only a small population of cancer cells survives and proliferates under high doses of chemotherapy drugs or radiation. In immunodeficient mice, few cancer cells can form xenografts, which have the capacity to selfrenew, have tumor-initiating potential, and can recapitulate the cellular heterogeneity of the original tumor. This small subset of cells was named CSCs, or “tumor-initiating” cells [6,13,14]. The first identification of CSCs was reported by Bonnet (1997) in acute myeloid leukemia [15]. The authors found that leukemic stem cells (CD34+CD38−) possessed differentiative and proliferative capacities and the potential for self-renewal. However, bulk cancer cells could not differentiate into other subpopulations of cancer cells and had limited self-renewal. In chronic myeloid

430

H. Zhan et al./Cancer Letters 357 (2015) 429–437

leukemia patients, STI571 (imatinib) killed almost all dividing cells, but Lin−CD34+ cells remained viable even under treatment with growth factors and imatinib [16]. Then, CSCs were identified in a variety of other cancers, including breast cancer [17], melanoma [18], brain tumor [19], pancreatic cancer [10], colon cancer [20], ovarian cancer [21], and prostate cancer [22]. These CSCs have been defined by their ability to self-renew and differentiate. Self-renewal can maintain the cells’ ability to survive long-term, whereas the differentiation of CSCs results in the heterogeneity of all cancer cells and histological recapitulation of the original tumor. CSCs are usually isolated and identified by flow cytometry using cell surface markers, such as CD34, CD24, CD44, CD133, ESA, ALDH, c-Met, and EpCAM [14,23]. These cell surface markers vary in each type of cancer. There is no universal cell surface marker that can identify CSCs from different types of cancer cells, causing confusion regarding which subpopulation cells are truly CSCs. Abel et al. [24] postulated that this confusion may stem from several sources. Highly passaged cell lines do not have the hierarchies observed in primary tumors. Limited studies, lack of standardization of digestion techniques, flow cytometry analyses, and the antibodies used may contribute to the varied findings regarding cell surface markers in CSCs. A combination of different markers may purify the CSC phenotype. CD44+c-Met+ provided robust selection of pancreatic CSCs [25], and a further enrichment of ovarian CSCs was observed in ALDH+CD133+ ovarian cancer cells. As few as 11 ALDH+CD133+ cells isolated from human ovarian cancer can form xenografts in immunodeficient mice [26]. CSCs may have heterogeneous and biologically distinct subpopulations. This phenomenon was confirmed in both hematopoietic and solid tumors. These subsets regenerated the phenotypic and functional heterogeneity of the parental tumor. Moreover, different CSC subgroups are related to one another, and can also interconvert each other [27]. However, the molecular mechanisms of these subgroups and interactions are poorly understood. Because of their capacity to self-renew and differentiate, CSCs are confirmed to be involved in invasion, metastasis, and drug resistance in malignant tumors. CSCs are more resistant to conventional chemotherapy and radiation, and are important drivers of distant metastasis and recurrence. CSCs targeted therapy has been shown to improve chemosensitivity and prevent invasion and metastasis in cancer cell lines and preclinical in vivo studies [12,28]. Several signaling pathways are involved in regulating the functions of CSCs. Different CSCs and normal stem cells share key regulation genes and pathways, such as Sonic Hedgehog, Wnt/βcatenin, Notch, Hippo, c-myc, and Bmi-1 [24,28,29]. The Hedgehog (Hh) signaling pathway controls a number of genes involved in the determination of cell fate and stemness features. Hh signaling is active in various cancer stem cells. The signaling pathway can increase the tumor-initiating population, and contribute to cell migration, growth, and survival. SMO inhibitors can inhibit the CSCpromoting effects. In addition, Hh signaling has been shown to promote tumor metastasis and recurrence by regulating the core genes involved in the epithelial-to-mesenchymal transition (EMT) process [30]. Wnt/β-catenin and the Notch pathway also play key roles in stem cell self-renewal in normal and cancer stem cells. These pathways or genes may provide candidate targets for future cancer therapies.

Discovery and identification of PCSCs PCSCs were first described by Li et al. in 2007 [10]. The authors used a xenograft model originating from cells from a human patient with pancreatic adenocarcinoma. They found that CD44+CD24+ESA+ pancreatic cancer cells accounted for only 0.2–0.8% of all pancreatic cancer cells, but had a 100-fold higher tumorigenic potential than nontumorigenic cancer cells. As few as 100 cells were able to form tumors in half of the mice. Only one in twelve mice developed a tumor following injection with 10,000 CD44−CD24−ESA− cells. Meanwhile, the tumors formed by the CD44+CD24+ESA+ cells shared the same histological features as the original tumors. These tumor-initiating cells were also observed in pancreatic cancer cell lines. CD44+CD24+ cells were 20fold more tumorigenic than CD44−CD24−cells, and could form tumors identical to unsorted PANC-1 cells. Hermann et al. [31] reported that CD133+ pancreatic cancer cells had CSC properties, just a few months after CD44+CD24+ESA+ cells were identified. CD133+ pancreatic cancer cells are more tumorigenic than CD133− cells. CD133+ subpopulation cells are highly resistant to gemcitabine chemotherapy. The authors concluded that a subset of CD133+ CXCR4+ cancer stem cells was essential for tumor metastasis. Interestingly, the authors showed that CD44+CD24+ESA+ subpopulation partially overlapped with the CD133+ population. The results from a study aimed at revealing the relationship between CD133 expression and the clinical and pathological features of PCs suggested that CD133 expression in pancreatic cancer was significantly associated with lymphatic metastasis, vascular endothelial growth factor-C (VEGF-C) expression, and poor long-term survival [32]. c-Met is a receptor of the tyrosine kinase family and is stimulated by hepatocyte growth factor (HGF) to mediate normal organ development; it is a marker of normal mouse pancreatic stem and progenitor cells [24,33,34]. Previous studies confirmed that the level of c-Met is increased in pancreatic cancer and is associated with invasion, metastasis, and chemoresistance [35]. Li et al. (2011) confirmed c-Met as a cell surface marker of PCSCs. The researchers reported that c-MetHigh CD44+ had the capability for self-renewal and the highest tumorigenic potential of all cell populations studied. c-Met inhibitor XL184, alone or combined with gemcitabine, inhibited tumor growth and reduced the population of CSCs [25]. Cabozantinib, a c-Met inhibitor, can downregulate CSCs markers CD133, SOX2, and c-Met, which results in improved gemcitabine sensitivity and induces apoptosis [36]. The aldehyde dehydrogenase-1 (ALDH1) activity assay is also used to identify pancreatic CSCs. Kim et al. reported that ALDH1High alone was sufficient for efficient tumor initiation, while ALDH1High CD133+ subpopulation cells had enhanced tumorigenic potential [37]. ALDH1 is also a functional marker of PCSCs. Kahlert et al. reported that low expression of ALDH1A1 was a prognostic marker for poor survival in pancreatic cancer [38]. The level of ALDH1A1 was significantly higher in the gemcitabine-resistant MIA PaCa-2 cell line. Knockdown of ALDH1A1 combined with gemcitabine significantly increased apoptosis and decreased cell viability [39]. Although many molecules have been reported to be markers of PCSCs (Table 1), there are no generally accepted markers to define PCSCs. This indicates a significant bottleneck in PCSCs research. Further studies are needed to define and sort the subgroup of stemlike cells in pancreatic cancer.

Pancreatic cancer stem cells (PCSCs) Signaling pathways involved in regulating PCSCs characteristics In pancreatic cancer, a small subset of “tumor-initiating” cancer cells has been identified. This subset has been named PCSCs. Recent evidence has demonstrated that PCSCs are responsible for the progression and relapse of cancer, as well as its resistance to chemotherapy and radiation. A thorough understanding of PCSCs and its regulating mechanisms may provide a potential novel target therapy for pancreatic cancer.

Thoroughly understanding the signaling pathways that regulate the maintenance and epigenetics of PCSCs is important for the progress of further study and development of novel therapy. As mentioned before, PCSCs and normal tissue stem cells seem to share the same key regulatory genes and signaling pathways, such as Notch, Sonic Hedgehog, Wnt/β-catenin, PI3K/AKT, c-Myc, Bim-1, and FOXM1.

H. Zhan et al./Cancer Letters 357 (2015) 429–437

Table 1 Cell surface markers of pancreatic cancer stem cells.

f

Cell surface markers

Percentage (%)

Tumorigenic potential

First Author

Year

CD44/CD24/ ESA CD133

0.2–0.8

Li [10]

2007

1.09–3.21

Hermann [31]

2007

c-Met ALDH

2–16 16c

100 cells; 6/12 (50%) 10,000cells; 1/12(8.3%a) 500 cells; 80–90% 106 cells; Nullb 100 cells; 7/20 (35%) 100 cells; 6/6 (100%d) 100 cells; 3/7 (42.8%e) 100 cells; 0/8f 100 cells; 0/21g

Li [25] Kim [37]

2011 2011

a CD44−/CD24−/ESA−; b CD133−; c ALDHhigh; d ALDHhigh/CD133−; e ALDHhigh/CD133+; ALDHlow/CD133+; g ALDHlow/CD133−.

The Notch signaling pathway plays a key role in the development and homeostasis of most tissues, and is also one of the important pathways involved in CSCs self-renewal and survival [40,41]. Niess et al. reported that the Notch, Wnt, and EGFR signaling pathways associated with tumor stem cells were altered in the side population (SP) pancreatic cancer cell line L3.6pl [42]. In a recent study, inhibition of the Notch pathway by gamma secretase inhibitor (GSI) or Hes1 shRNA in pancreatic cancer cells reduced the percentage of CSCs and tumorosphere formation. GSI can also inhibit tumor growth and reduce the CSCs population in NOD/SCID mice [43]. Bao et al. reported that overexpression of Notch-1 led to the induction of the EMT phenotype and also increased the formation of pancreatospheres, consistent with expression of CSCs surface markers CD44 and EpCAM. GSI IX can suppress the EMT and pancreatic tumor initiating CD44+EpCAM+ cells by targeting the Notch pathway [44]. Hedgehog (Hh) signaling is confirmed to be a pivotal pathway in embryonic patterning and growth control [45,46]. It is activated in various CSCs, including PCSCs [30,46]. Inhibition of the Hh pathway results in the decreased self-renewal of PCSCs and reverse drug resistance. Numerous agents have been confirmed to inhibit PCSCs pluripotency via suppression of Hh signaling. One such agent is sulforaphane, an active compound in cruciferous vegetables, which can inhibit Shh-Gli signaling to suppress the expression of PCSCs pluripotency maintaining factors Nanog and Oct-4 [47]. Similar results were observed with epigallocatechin-3-gallate, an active compound in green tea [48], wasabi compounds 6-MITC [49], and arsenic trioxide [50]. In a recent study, GANT-61, a Gli transcription factor inhibitor, was reported to inhibit PCSCs pluripotency-maintaining factors Nanog, Oct4, Sox-2, and c-Myc. GANT-61 also suppressed EMT by inhibiting N-cadherin and transcription factors, including Snail, Slug, and ZEB1 [51]. These studies confirmed the importance of Hh signaling for selfrenewal and stemness maintenance of pancreatic CSCs. The Wnt/β-catenin pathway is another crucial signaling pathway involved in the regulation of CSCs. A previous study reported that Wnt antagonist sFRP4 could induce apoptosis, decrease sphere formation, and revere the expression of EMT markers in squamous cell carcinoma CSCs of the head and neck [52]. However, few studies have focused on the Wnt/β-catenin pathway and PCSCs, and the molecular mechanism is still unclear. Vaz et al. demonstrated that levels of β-catenin and Oct3/4 were elevated in pancreatic CSCs. Knockdown of pancreatic differentiation 2 (PD2) in CSCs resulted in reduced cell viability and enhanced apoptosis [53]. FOXM1, a member of the forkhead box transcription factor superfamily, is a key cell cycle regulator. It has been confirmed that FOXM1 is overexpressed in pancreatic cancer, and can stimulate stem cell-like characteristics in pancreatic cancer cells. Furthermore, overexpression of FOXM1 leads to decreased expression of tumor inhibitor miRNA, resulting in the acquisition of EMT phenotype, increased sphere-forming capacity, and CD44 and EpCAM expression

431

[54]. FOXM1 also has a close relationship with important signaling pathways that were previously mentioned to regulate the functions of pancreatic CSCs [55]. The signaling pathways involved in the regulation of PCSCs are summarized in Table 2. The regulatory roles of miRNAs in PCSCs MicroRNAs (miRNAs) are a family of highly conserved small noncoding RNAs that target protein-coding mRNAs at the posttranscriptional level. They have important roles in regulating carcinogenesis and progression in malignant tumors, and have been confirmed as biomarkers that can be used for early diagnosis, prediction of prognosis, and the identification of therapeutic targets [14,59]. Based on previous studies, a number of microRNAs take part in the invasion, metastasis, and drug resistance of pancreatic cancer. miR-21 has been identified to be overexpressed in pancreatic cancer tissues [60]. Moriyama et al. reported that miR-21 was associated with cell proliferation, invasion, and chemoresistance in pancreatic cancer [61]. The PI3K/AKT signal pathway was involved in the miR-21-mediated invasion behavior and gemcitabine resistance of pancreatic cancer. Expression of miR-21 was correlated with clinical outcomes of patients with pancreatic cancer [62]. In later studies, numerous miRNAs, such as miR-200c, miR-146a, miR-20a, miR-126, miR-150, and miR-96, were confirmed to be upregulated or downregulated in pancreatic cancer tissues or cell lines; these microRNAs are targeted in many genes/pathways to regulate molecular behaviors of pancreatic cancer [63]. Previous studies showed that deregulated expression of miRNAs may play a key role in the regulation of CSCs characteristics and functions. miR-34 was confirmed to be directly regulated by p53, and acted as tumor-suppressor microRNA [64]. Ji et al. reported that upregulation of miR-34 significantly inhibited cell proliferation and invasion, induced apoptosis, and sensitized the cells to chemotherapy and radiation. The authors also found high levels of Notch/ Bcl-2 and loss of miR-34 in MiaPaCa2 pancreatic cancer cells were enriched with CSCs subpopulations. miR-34 restoration led to an 87% reduction of the tumor-initiating cells population, which suggested that miR-34 may play an important role in pancreatic cancer stem cell self-renewal [65]. Lu et al. reported that expression of miR-200a was decreased in CD44 + CD24 + ESA + cells in pancreatic cancer cell line PANC-1. Overexpression of miR-200a induced decreased levels of EMT markers, such as N-cadherin, ZEB1, and vimentin. miR-200a also inhibited cell migration and invasion in CSCs. The authors concluded that miR-200a inhibited the EMT process of pancreatic cancer stem cells [66]. Hasegawa et al. demonstrated that miR-1246 can increase tumor-initiating potential in gemcitabine-resistant PANC-1 cells. miR-1246 was associated with chemoresistance and CSClike properties by targeting CCNG2 [67].

Table 2 Signaling pathways involved in the regulation and target therapy of PCSCs. Biology behaviors

Signaling pathways

Target population

Target agents

Proliferation

Notch Hedgehog

ALDH+ CD44+CD24+ESA+ ALDH+

c-Met ALK4 CXCR4 DR5

c-MetHigh CD44+ CD133+ CD133+ CD44+CD24+ESA+ ALDH+ CD133+ CD133+ c-Met+, CD133+

GSI [43] IPI-269609,GDC-0449 BMS-663513 Cyclopamine [30,46] XL184 [25] SB431542 [56] AMD3100 [57] DR5 agonist [58]

Invasion and metastasis

Drug resistance

CXCR4 PD2 [53] c-Met

AMD3100 [57] Cabozantinib [36]

432

H. Zhan et al./Cancer Letters 357 (2015) 429–437

Zinc finger E-box binding homeobox 1 (ZEB-1), a EMT promoter, can inhibit expression of the miR-200 family and miR-203, which can target stem cell factors, such as Sox2, Bim-1 and Klf4. This inhibition results in the activation of EMT and maintenance of stemness in pancreatic and colorectal cancer. The ZEB1-miR-200 feedback loop might be a potential therapeutic target for pancreatic cancer [68]. DCAMKL-1 is a putative pancreatic stem cell marker [63,69]. Sureban et al. found that DCAMKL-1 was overexpressed in a KRAS transgenic mouse model of pancreatic cancer and in human pancreatic adenocarcinoma tissue. In the AsPC-1 human pancreatic cancer cell line, knockdown of DCAMKL-1 by RNAi can upregulate expression of an EMT-inhibitor microRNA, miR-200a, which results in downregulation of ZEB1 and ZEB2 with upregulation of E-cadherin. The authors also reported that Notch-1, an important CSCs regulating gene, was a downstream target of miR-144, and that DCAMKL-1 regulated Notch-1 at a post-transcriptional level [70]. In a subsequent study, this research group also confirmed other

tumor suppressor miRNAs (miR-145, let-7a) involved in the regulation network of pancreatic cancer pluripotency [71]. The regulatory functions of microRNAs targeted in PCSCs are summarized in Fig. 1. PCSCs and EMT EMT is an essential embryonic process during which polarized epithelial cells convert into motile mesenchymal cells. This process plays a critical role in embryonic development, injury, wound healing, inflammation, fibrosis, and tumorigenesis [72,73]. Cancer-related EMT is not only involved in the invasion and metastasis of malignant tumors, but also in tumor initiation and early tumor development [73]. A link between the EMT process and the acquisition of molecular and functional traits of CSCs has been confirmed. The EMTlike cellular phenotype has a greater likelihood of developing resistance to chemotherapy or radiation therapy and a higher potential for metastasis. The EMT process not only leads to the gain

Fig. 1. Regulatory functions of microRNAs targeted in PCSCs.

H. Zhan et al./Cancer Letters 357 (2015) 429–437

433

Fig. 2. Crosstalk between EMT, microRNAs, and PCSCs.

of the mesenchymal phenotype, but also induces the expression of stem cell markers [74]. EMT transcription factors, including SNAIL1, TWIST, ZEB1, ZEB2, FOXQ1, FOXA1, and SOX9, can induce the differentiated mammary epithelial cell surface marker profile (CD44low/CD24high) to that characterizing cancer stem cells (CD44+/ CD24−/low) and form spheres in breast cancer cells [75,76]. Tumors with high expression of EMT transcription factors are enriched with CSCs. Genes and signaling pathways associated with invasion and metastasis are also overexpressed in CSCs. EMT-like cells and CSCs share similar molecular characteristics; this phenomenon has been observed in breast cancer, lung cancer, bladder cancer, colorectal cancer, and pancreatic cancer [73,74]. ZEB-1 can repress the members of the miR-200 family, resulting in activation of EMT and maintenance of stemness in pancreatic cancer [68]. TWIST1 can stimulate the expression of Bim-1, which promotes the EMT process and maintains self-renewal by repression of E-cadherin and p16INK4a [77]. Nevertheless, the relationship between EMT and CSCs is complex. Puisieux et al. concluded that CSCs can either be embedded in the epithelial mass of primary and metastases sites, or linked to EMT and motility in invading and disseminating growth-arrested tumor cells. EMT might dynamically regulate the stemness and cell plasticity [73]. The EMT process also has great influence on the renewal and survival of pancreatic cancer stem cells. HIF-1α can induce enhanced autophagy and EMT, resulting in enriched stem-like cells (CD133+) and a greater ability to metastasize under intermittent hypoxia [78]. EMT transcription factor SNAIL1 was highly expressed in the pancreatic CSChigh cell line Panc-1. Repression of SNAIL1 resulted in the inhibition of CSC marker ALDH expression and decreased sphereand colony-forming capacity [79]. Wang et al. reported that activation of the platelet-derived growth factor D (PDGF-D) signaling pathway could promote invasion and metastasis in pancreatic ductal

adenocarcinoma by the concomitant acquisition of EMT phenotype and induction of CSCs [80]. Numerous microRNAs contribute to the EMT and generation and maintenance of pancreatic CSCs, including the miR-200 family, miR-203, miR-21, miR-34, miR-365, miR-144, and let-7 [63]. The crosstalk between EMT, microRNAs, and pancreatic CSCs is summarized in Fig. 2. The effects of the tumor microenvironment on PCSCs In traditional cancer development theory, cell-autonomous changes and the local microenvironment of the primary tumor are compared to “seed and soil”. In pancreatic cancer, cancer cells and the surrounding pancreatic cells, such as stellate cells, panendothelial cells, and infiltrating immune cells, make up the tumor microenvironment. These interactions are theorized to promote tumor invasion and metastasis, maintain the population of PCSCs, and contribute to chemotherapy resistance. Studies of the hypoxic microenvironment, the CSCs niche, the role of the EMT in tumor progression, and the pre-metastatic niche have produced interesting data on a variety of solid tumors, including pancreatic tumors [2,81,82]. Coaction of pancreatic cancer cells and stromal cells promotes invasive growth of cancer cells and builds a specific microenvironment which further manipulates the malignant properties of cancer cells. Pancreatic stellate cells (PSCs) play a key role in the progression of fibrosis within the cancer tissue, and also influence cancer cell function [82]. Initial and metastatic tumors possess few PCSCs, which undergo self-renewal and differentiate into bulk tumor cells. Growth factors secreted by either stromal cells or other PCSCs can stimulate PCSCs to self-renew. Meanwhile, PCSCs can also activate stromal cells via growth factors and SHH. PCSCs can be attracted

434

H. Zhan et al./Cancer Letters 357 (2015) 429–437

to sites of metastasis by factors produced by distant stromal cells, such as SDF-1. In addition, the tumor microenvironment plays an important role in chemotherapy resistance and disease recurrence [24]. The regulation mechanisms of the tumor microenvironment on PCSCs are complex and merit further study. The regulatory roles of PCSCs in pancreatic cancer invasion and metastasis Metastasis is the major cause of death for patients with pancreatic cancer. There is currently no radical treatment for metastatic pancreatic cancer. An increasing number of studies have implicated CSCs in cancer invasion and metastasis in several unrelated cancers [83,84]. A specific set of surface markers and genes in PCSCs have been associated with the occurrence of distant metastasis. Hermann et al. found that a subset of CD133 + PCSCs that expressed chemokine receptor 4 was more invasive and mediated the formation of liver metastases in an animal model [31]. Rasheed et al. found that ALDH expression was upregulated in metastatic lesions compared to the primary tumors, indicating a pivotal role for these cells in mediating distant metastasis [85,86]. Li et al. indicated that c-Met+ PCSCs were also able to mediate invasion and metastasis in vivo [25]. The exact relationship between metastasis and PCSCs remains unclear. PCSCs may be associated with an increased ability to invade, the invasive front of tumors, and EMT type cells. A subpopulation of CD133+/CXCR4+ cells was identified. These cells revealed a high migratory activity toward gradients of the CXCR4 ligand SDF-1 [31], the migration of which could be repressed by antiCXCR4 neutralizing antibodies and the small molecule inhibitor AMD3100 [57] in vitro. In an in vivo study, CD133+/CXCR4+ CSCs were located in the invasive front of pancreatic tumors, suggesting a more invasive and metastatic phenotype of pancreatic CD133+/CXCR4+ CSCs [31]. Other results collectively show that CXCR4 is implicated in the metastasis of breast cancer [87] and pancreatic cancer [88]. As mentioned before, EMT is an important biological transformation process for morphogenesis, embryogenesis, and cancer progression, and contributes to invasion and metastasis [73]. It is likely that the CSCs can acquire a migrating phenotype during EMT in primary neoplasms, which would allow them to spread to distant sites. Results from a recent study performed by Zhang et al. indicated that EMT correlated with CD24+CD44+ and CD133+ cells in pancreatic cancer. EMT may induce pancreatic cancer stem-like cells; different degrees of EMT may induce different amounts of CD24+CD44+ and CD133+ cells [89]. Li et al. concluded that metastasis may be promoted by CSCs or EMT-type cells. The two cell types share similar properties, such as molecular characteristics, resistance to treatment, and propensity for invasion [74]. It was proposed that the EMT-inducer ZEB1 links activation of EMT and maintenance of stemness in one cell by suppressing expression of stemnessinhibiting miRNAs, thus promoting metastasis [90]. However, the exact association between EMT and CSCs remains to be completely explained. PCSCs and drug resistance PDAC has a grim prognosis. The majority of patients already have either locally advanced or metastatic cancer when they are diagnosed, while the therapeutic options for most patients eventually become restricted to an intensive regimen of systemic chemotherapy [91]. Most therapies for PC are aimed at cytoreduction of differentiated tumor cells. These therapies, however, do not affect PCSCs, which can then re-establish tumors after traditional treatment. Clinical data showed that PDAC developed resistance to standard treatments such as gemcitabine, resulting in a mere six months of median overall survival [92].

Growing evidence has demonstrated that CSCs are resistant to conventional chemotherapy [74]. Despite the unclear mechanism, CSCs are thought to be inherently resistant to chemotherapy and radiation. The potential explanation for this may include improved DNA repair capacity, increased ALDH oxidation of aldehydes/ detoxification, resistance to apoptosis, low mitotic rate, increased tolerance of DNA damage, and radiotherapy resistance via a DNA damaged checkpoint/increased repair capacity [93–95]. Studies have also shown that CSCs in the pancreas [31,96] are responsible for chemotherapy and radiation resistance. It has been demonstrated that while CD133+ CSCs stopped proliferating under the influence of the gemcitabine, they did not undergo apoptosis. Thus, the cancer (stem) cell pool was immediately repopulated as soon as gemcitabine was withdrawn. The more differentiated CD133 negative cells, which represented the vast majority of the tumor cell population, became apoptotic after the application of gemcitabine. Thus, it is increasingly obvious that only the more differentiated tumor cells could be affected by traditional therapy, potentially resulting in a selection process for CSCs [97–99]. Lee et al. found that CD44+CD24+ESA+ cells were resistant to gemcitabine and radiotherapy [93]. Treatment with gemcitabine eliminated the bulk of cells, but left an enriched population of CD44+ cells and a small subset of CD44+CD24+ESA+ cells, which could reconstitute the neoplasm and propagate metastasis [96]. In addition, side population (SP) cells are enriched in cells displaying CSC properties. SP was confirmed to be resistant to gemcitabine treatment. Whole-genome expression profiling of the SP revealed that the multidrug transporters ABCG2 and ABCA9 were highly expressed, both of which can mediate drug resistance [100]. miRNAs regulate various molecular mechanisms of cancer progression that are associated with CSCs. Hasegawa et al. recently identified miR-1246 as the miRNA that regulated clinical properties, such as high refractoriness to chemotherapy and tumor recurrence of PCSCs [67]. They designed self-assembling, gemcitabine-conjugated cationic copolymers for co-delivery of a tumor suppressor miRNA-205. This miRNA could transfect and reverse chemoresistance, invasion, and metastasis in gemcitabineresistant pancreatic cancer cells. In vivo studies in a pancreatic cancer xenograft model showed significant inhibition of tumor growth, reduced cell proliferation, and increased apoptosis in animals treated with miR-205 and gemcitabine [92]. PCSCs and targeted therapy An increasing number of investigations have focused on designing novel therapeutic strategies to target CSCs and EMT-type cells in order to increase drug sensitivity and inhibit cancer cell invasion and metastasis. CSCs have clear therapeutic implications. These cells may be inherently resistant to chemotherapy and radiation. Furthermore, traditional cytotoxic therapy may actually increase the proportion of CSCs. Targeting CSCs alone or in combination with treatment of bulk cancer cells may produce a better therapeutic response [101]. As previously mentioned, CSCs might share cellular pathways, such as Hedgehog, Notch, Wnt, BMI-1, and PTEN, that are required for the regulation of normal stem cells [86]. Targeting these pathways holds promise in cancer therapy. A number of pre-clinical studies have shown that blocking the Hedgehog pathway in animal models abrogates the PCSCs’ ability to metastasize and decreases tumorigenicity [30,86]. IPI-926, IPI269609, GDC-0449, BMS-663513, and cyclopamine are frequently used Hedgehog inhibitors [30,46]. Feldmann studied IPI-26909 in both in vivo and in vitro models [102]. In vitro studies demonstrated that downregulation of Hedgehog resulted in decreased cell migration, colony formation, and expression of ALDH. In an orthotropic xenograft model, treatment with IPI-269609 decreased ALDH expression with little effect on tumor volume, but

H. Zhan et al./Cancer Letters 357 (2015) 429–437

completely blocked metastasis. Cyclopamine inhibits CSCs, abrogates risk of metastatic disease, and suppresses EMT in animal models; it could reduce tumor bulk and abrogate metastasis when used in combination with gemcitabine [46,103]. Notch signaling, which is under the regulatory control of ZEB1 and miRNA-200, has also been implicated in the regulation of PCSCs and, therefore, might be a therapeutic target against CSCs [104–106]. Inhibition of a key Notch gene (Hes1) with γ secretase inhibitor or shRNA decreased sphere formation in vitro and tumor growth in vivo [43]. c-Met is another potential CSC target. Inhibition of c-Met with XL184 decreased tumor sphere formation in CD44+CD24+ESA+ cells [25]. The c-Met inhibitor cabozantinib also inhibits stem cell function. Pancreatic cells treated with cabozantinib had decreased sphere formation and increased apoptosis [36]. Several studies have found that inhibition of CXCR4 with AMD3100 or chloroquine decreased stem cell migration and metastasis in a mouse model [31,107]. Other potential stem cell-mediated treatments include inhibition of the transforming growth factor-beta (TGF-β) family and stimulation of death receptor 5 (DR5). Inhibition of a Nodal or Activin receptor, both of which are members of the TGF-β superfamily, could essentially eliminate CSCs self-renewal and in vivo tumor growth [56]. DR5 is overexpressed in PCSCs. Administering a DR5 agonist to mice harboring pancreatic cancer xenografts results in a reduction of PCSCs and enhanced antitumor activity when combined with gemcitabine [58]. In addition, targeting the ABC receptor in the side population may impact the CSCs fraction [42]. A multimodal therapy consisting of cyclopamine, rapamycin (a natural mTOR inhibitor), and gemcitabine (CRG) represents a promising approach [99,108]. While this triple therapy has been demonstrated to be highly effective against PCSCs, combining chemotherapy may bear many potential side effects. Among these, the most relevant and dangerous risk is bone marrow depression, followed by subsequent leukopenia and a highly increased chance of infection. Interestingly, Mueller et al. showed that this regimen was well tolerated by mice [108]. Further safety studies are required before the clinical application of CRG as a novel therapy against PC. Clinical trials of Hedgehog, Notch, and Wnt inhibitors are ongoing in a variety of malignancies [86]. As our understanding of PCSCs expands, additional targeted therapeutic interventions will be forthcoming. PCSCs and prognosis The functional properties of PCSCs provide a potential explanation for clinical observations, such as metastasis, resistance to chemotherapy, and cancer relapse. CSCs are posited to be the pivotal element that drives the natural history of disease progression. Thus, their frequency or functional potential could be correlated with clinical outcomes. A vast number of studies have been performed to validate whether CSCs can serve as predictive biomarkers. Expression of CSCs markers has been linked to diminished survival in many cancers, including colorectal, breast, prostate, ovarian, small bowel, and pancreatic cancer [86,109–111]. Side population and PCSCs markers, such as CD133, CD44/CD24/ESA, and ALDH, have been associated with poorer prognosis [85,86,112,113]. CD133 expression in PC is associated with EMT and lymphatic invasion, and is an independent prognostic factor [113]. Double positive CD44+CD24+ expression has been closely associated with poor clinical outcomes [112]. Increased expression of ALDH, based on immunohistochemistry in primary pancreatic tumors from patients who underwent resection for localized disease, was associated with worse median survival (14 vs. 18 months, hazard ratio of death = 1.28, P = .05) [85]. Side population gene expression profiles showed that expression levels of ABCB1 and CXCR4 were

435

correlated with poorer prognosis [114]. In addition, Chen et al. demonstrated that the IL-8/CXCR1 axis was associated with PCSCs, and positive CXCR1 expression was correlated with lymph node metastasis and poor survival in patients with PDAC [115]. PCSCs could be a useful prognostic factor for pancreatic cancer. Conclusion and future perspective CSCs have been confirmed to be involved in carcinogenesis, progression, and drug resistance in pancreatic cancer. Many signaling pathways and factors are involved in regulating the characteristics of PCSCs, including the Notch, Hedgehog, and Wnt/β-catenin signaling pathways, microRNAs, the tumor microenvironment, and the EMT process, which constitutes a complex crosstalk network to regulate PCSC phenotypes, heterogeneity, and functions. In addition, PCSCs could serve as new therapeutic targets and prognostic factors in pancreatic cancer. Nevertheless, some questions still merit further investigation. The primary issue is how to define and sort the subgroup cells possessing stem cell characteristics in pancreatic cancer. Patient-derived xenografts planted on SCID mice can effectively maintain the features of primary tumors from patients. These xenografts might be used to define and sort PCSCs. The regulation mechanisms of PCSCs, especially those of the tumor microenvironment, also must be further explored in in vitro and in vivo studies. The mechanisms of how CSCs escape from immune surveillance and then spread to targeted organs deserve further investigation. The value of PCSCs in predicting chemosensitivity and radiosensitivity in pancreatic cancer is also worthy of further study. PCSCs might also be potential targets for the treatment of metastasis. In addition, identification of genomics, proteomics, and ncRNA profiles is necessary to disclose the origin and maintenance mechanisms of PCSCs, and may help us to discover novel therapeutic targets. Acknowledgments This work was supported by the China Postdoctoral Science Foundation (2013M531606) and the Shandong Provincial Natural Science Foundation, China (ZR2013HQ049). Conflict of interest The authors declare that there are no conflicts of interest. References [1] R. Siegel, J. Ma, Z. Zou, A. Jemal, Cancer statistics, CA Cancer J. Clin. 64 (2014) (2014) 9–29. [2] H. Oettle, Progress in the knowledge and treatment of advanced pancreatic cancer: from benchside to bedside, Cancer Treat. Rev. 40 (2014) 1039–1047. [3] T. Conroy, F. Desseigne, M. Ychou, O. Bouche, R. Guimbaud, Y. Becouarn, et al., FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer, N. Engl. J. Med. 364 (2011) 1817–1825. [4] J.W. Chiu, H. Wong, R. Leung, R. Pang, T.T. Cheung, S.T. Fan, et al., Advanced pancreatic cancer: flourishing novel approaches in the era of biological therapy, Oncologist 19 (9) (2014) 937–950. [5] D.D. Von Hoff, T. Ervin, F.P. Arena, E.G. Chiorean, J. Infante, M. Moore, et al., Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine, N. Engl. J. Med. 369 (2013) 1691–1703. [6] M.F. Clarke, J.E. Dick, P.B. Dirks, C.J. Eaves, C.H. Jamieson, D.L. Jones, et al., Cancer stem cells – perspectives on current status and future directions: AACR Workshop on cancer stem cells, Cancer Res. 66 (2006) 9339–9344. [7] J.E. Visvader, G.J. Lindeman, Cancer stem cells: current status and evolving complexities, Cell stem cell 10 (2012) 717–728. [8] E. Fessler, F.E. Dijkgraaf, E.M.F. De Sousa, J.P. Medema, Cancer stem cell dynamics in tumor progression and metastasis: is the microenvironment to blame? Cancer Lett. 341 (2013) 97–104. [9] X. Wang, Y. Zhu, Y. Ma, J. Wang, F. Zhang, Q. Xia, et al., The role of cancer stem cells in cancer metastasis: new perspective and progress, Cancer Epidemiol. 37 (2013) 60–63.

436

H. Zhan et al./Cancer Letters 357 (2015) 429–437

[10] C. Li, D.G. Heidt, P. Dalerba, C.F. Burant, L. Zhang, V. Adsay, et al., Identification of pancreatic cancer stem cells, Cancer Res. 67 (2007) 1030–1037. [11] J.A. Castellanos, N.B. Merchant, N.S. Nagathihalli, Emerging targets in pancreatic cancer: epithelial-mesenchymal transition and cancer stem cells, Onco. Targets Ther. 6 (2013) 1261–1267. [12] J. Xia, C. Chen, Z. Chen, L. Miele, F.H. Sarkar, Z. Wang, Targeting pancreatic cancer stem cells for cancer therapy, Biochim. Biophys. Acta 2012 (1826) 385–399. [13] M. Signore, L. Ricci-Vitiani, R. De Maria, Targeting apoptosis pathways in cancer stem cells, Cancer Lett. 332 (2013) 374–382. [14] J.A. Leal, M.E. Lleonart, MicroRNAs and cancer stem cells: therapeutic approaches and future perspectives, Cancer Lett. 338 (2013) 174–183. [15] D. Bonnet, J.E. Dick, Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell, Nat. Med. 3 (1997) 730–737. [16] S.M. Graham, H.G. Jorgensen, E. Allan, C. Pearson, M.J. Alcorn, L. Richmond, et al., Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro, Blood 99 (2002) 319–325. [17] M. Al-Hajj, M.S. Wicha, A. Benito-Hernandez, S.J. Morrison, M.F. Clarke, Prospective identification of tumorigenic breast cancer cells, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 3983–3988. [18] N.Y. Frank, A. Margaryan, Y. Huang, T. Schatton, A.M. Waaga-Gasser, M. Gasser, et al., ABCB5-mediated doxorubicin transport and chemoresistance in human malignant melanoma, Cancer Res. 65 (2005) 4320–4333. [19] S.K. Singh, C. Hawkins, I.D. Clarke, J.A. Squire, J. Bayani, T. Hide, et al., Identification of human brain tumour initiating cells, Nature 432 (2004) 396–401. [20] C.A. O’Brien, A. Pollett, S. Gallinger, J.E. Dick, A human colon cancer cell capable of initiating tumour growth in immunodeficient mice, Nature 445 (2007) 106–110. [21] S.A. Bapat, A.M. Mali, C.B. Koppikar, N.K. Kurrey, Stem and progenitor-like cells contribute to the aggressive behavior of human epithelial ovarian cancer, Cancer Res. 65 (2005) 3025–3029. [22] J. Miki, B. Furusato, H. Li, Y. Gu, H. Takahashi, S. Egawa, et al., Identification of putative stem cell markers, CD133 and CXCR4, in hTERT-immortalized primary nonmalignant and malignant tumor-derived human prostate epithelial cell lines and in prostate cancer specimens, Cancer Res. 67 (2007) 3153–3161. [23] J.P. Medema, Cancer stem cells: the challenges ahead, Nat. Cell Biol. 15 (2013) 338–344. [24] E.V. Abel, D.M. Simeone, Biology and clinical applications of pancreatic cancer stem cells, Gastroenterology 144 (2013) 1241–1248. [25] C. Li, J.J. Wu, M. Hynes, J. Dosch, B. Sarkar, T.H. Welling, et al., c-Met is a marker of pancreatic cancer stem cells and therapeutic target, Gastroenterology 141 (2011) 2218–2227 e2215. [26] I.A. Silva, S. Bai, K. McLean, K. Yang, K. Griffith, D. Thomas, et al., Aldehyde dehydrogenase in combination with CD133 defines angiogenic ovarian cancer stem cells that portend poor patient survival, Cancer Res. 71 (2011) 3991– 4001. [27] V.R. Penchev, Z.A. Rasheed, A. Maitra, W. Matsui, Heterogeneity and targeting of pancreatic cancer stem cells, Clin. Cancer Res. 18 (2012) 4277–4284. [28] M.L. O’Connor, D. Xiang, S. Shigdar, J. Macdonald, Y. Li, T. Wang, et al., Cancer stem cells: a contentious hypothesis now moving forward, Cancer Lett. 344 (2014) 180–187. [29] E. Proctor, M. Waghray, C.J. Lee, D.G. Heidt, M. Yalamanchili, C. Li, et al., Bmi1 enhances tumorigenicity and cancer stem cell function in pancreatic adenocarcinoma, PLoS ONE 8 (2013) e55820. [30] H. Onishi, M. Katano, Hedgehog signaling pathway as a new therapeutic target in pancreatic cancer, World J. Gastroenterol. 20 (2014) 2335–2342. [31] P.C. Hermann, S.L. Huber, T. Herrler, A. Aicher, J.W. Ellwart, M. Guba, et al., Distinct populations of cancer stem cells determine tumor growth and metastatic activity in human pancreatic cancer, Cell Stem Cell 1 (2007) 313–323. [32] S. Maeda, H. Shinchi, H. Kurahara, Y. Mataki, K. Maemura, M. Sato, et al., CD133 expression is correlated with lymph node metastasis and vascular endothelial growth factor-C expression in pancreatic cancer, Br. J. Cancer 98 (2008) 1389–1397. [33] P. Michieli, M. Mazzone, C. Basilico, S. Cavassa, A. Sottile, L. Naldini, et al., Targeting the tumor and its microenvironment by a dual-function decoy Met receptor, Cancer Cell 6 (2004) 61–73. [34] M. Herreros-Villanueva, A. Zubia-Olascoaga, L. Bujanda, c-Met in pancreatic cancer stem cells: therapeutic implications, World J. Gastroenterol. 18 (2012) 5321–5323. [35] D. Delitto, E. Vertes-George, S.J. Hughes, K.E. Behrns, J.G. Trevino, c-Met signaling in the development of tumorigenesis and chemoresistance: potential applications in pancreatic cancer, World J. Gastroenterol. 20 (2014) 8458–8470. [36] C. Hage, V. Rausch, N. Giese, T. Giese, F. Schonsiegel, S. Labsch, et al., The novel c-Met inhibitor cabozantinib overcomes gemcitabine resistance and stem cell signaling in pancreatic cancer, Cell Death Dis. 4 (2013) e627. [37] M.P. Kim, J.B. Fleming, H. Wang, J.L. Abbruzzese, W. Choi, S. Kopetz, et al., ALDH activity selectively defines an enhanced tumor-initiating cell population relative to CD133 expression in human pancreatic adenocarcinoma, PLoS ONE 6 (2011) e20636. [38] C. Kahlert, F. Bergmann, J. Beck, T. Welsch, C. Mogler, E. Herpel, et al., Low expression of aldehyde dehydrogenase 1A1 (ALDH1A1) is a prognostic marker for poor survival in pancreatic cancer, BMC Cancer 11 (2011) 275.

[39] H.Q. Duong, J.S. Hwang, H.J. Kim, H.J. Kang, Y.S. Seong, I. Bae, Aldehyde dehydrogenase 1A1 confers intrinsic and acquired resistance to gemcitabine in human pancreatic adenocarcinoma MIA PaCa-2 cells, Int. J. Oncol. 41 (2012) 855–861. [40] C. Karamboulas, L. Ailles, Developmental signaling pathways in cancer stem cells of solid tumors, Biochim. Biophys. Acta 2013 (1830) 2481–2495. [41] I. Espinoza, R. Pochampally, F. Xing, K. Watabe, L. Miele, Notch signaling: targeting cancer stem cells and epithelial-to-mesenchymal transition, Onco. Targets Ther. 6 (2013) 1249–1259. [42] H. Niess, P. Camaj, A. Renner, I. Ischenko, Y. Zhao, S. Krebs, et al., Side population cells of pancreatic cancer show characteristics of cancer stem cells responsible for resistance and metastasis, Targeted Oncol. (2014). [43] E.V. Abel, E.J. Kim, J. Wu, M. Hynes, F. Bednar, E. Proctor, et al., The Notch pathway is important in maintaining the cancer stem cell population in pancreatic cancer, PLoS ONE 9 (2014) e91983. [44] B. Bao, Z. Wang, S. Ali, D. Kong, Y. Li, A. Ahmad, et al., Notch-1 induces epithelial-mesenchymal transition consistent with cancer stem cell phenotype in pancreatic cancer cells, Cancer Lett. 307 (2011) 26–36. [45] M.M. Cohen Jr., The hedgehog signaling network, Am. J. Med. Genet. A 123A (2003) 5–28. [46] J. Xie, C.M. Bartels, S.W. Barton, D. Gu, Targeting hedgehog signaling in cancer: research and clinical developments, Onco. Targets Ther. 6 (2013) 1425– 1435. [47] S.H. Li, J. Fu, D.N. Watkins, R.K. Srivastava, S. Shankar, Sulforaphane regulates self-renewal of pancreatic cancer stem cells through the modulation of Sonic hedgehog-GLI pathway, Mol. Cell. Biochem. 373 (2013) 217–227. [48] S.N. Tang, J. Fu, D. Nall, M. Rodova, S. Shankar, R.K. Srivastava, Inhibition of sonic hedgehog pathway and pluripotency maintaining factors regulate human pancreatic cancer stem cell characteristics, Int. J. Cancer 131 (2012) 30–40. [49] Y.J. Chen, Y.C. Huang, T.H. Tsai, H.F. Liao, Effect of wasabi component 6-(methylsulfinyl)hexyl isothiocyanate and derivatives on human pancreatic cancer cells, Evid. Based Complement. Alternat. Med. 2014 (2014) 494739. [50] J.B. Han, F. Sang, J.J. Chang, Y.Q. Hua, W.D. Shi, L.H. Tang, et al., Arsenic trioxide inhibits viability of pancreatic cancer stem cells in culture and in a xenograft model via binding to SHH-Gli, Onco. Targets Ther. 6 (2013) 1129–1138. [51] J. Fu, M. Rodova, S.K. Roy, J. Sharma, K.P. Singh, R.K. Srivastava, et al., GANT-61 inhibits pancreatic cancer stem cell growth in vitro and in NOD/SCID/IL2R gamma null mice xenograft, Cancer Lett. 330 (2013) 22–32. [52] S. Warrier, G. Bhuvanalakshmi, F. Arfuso, G. Rajan, M. Millward, A. Dharmarajan, Cancer stem-like cells from head and neck cancers are chemosensitized by the Wnt antagonist, sFRP4, by inducing apoptosis, decreasing stemness, drug resistance and epithelial to mesenchymal transition, Cancer Gene Ther. 21 (9) (2014) 381–388. [53] A.P. Vaz, M.P. Ponnusamy, S. Rachagani, P. Dey, A.K. Ganti, S.K. Batra, Novel role of pancreatic differentiation 2 in facilitating self-renewal and drug resistance of pancreatic cancer stem cells, Br. J. Cancer 111 (2014) 486– 496. [54] B. Bao, Z. Wang, S. Ali, D. Kong, S. Banerjee, A. Ahmad, et al., Over-expression of FoxM1 leads to epithelial-mesenchymal transition and cancer stem cell phenotype in pancreatic cancer cells, J. Cell. Biochem. 112 (2011) 2296–2306. [55] M. Quan, P. Wang, J. Cui, Y. Gao, K. Xie, The roles of FOXM1 in pancreatic stem cells and carcinogenesis, Mol. Cancer 12 (2013) 159. [56] E. Lonardo, P.C. Hermann, M.-T. Mueller, S. Huber, A. Balic, I. Miranda-Lorenzo, et al., Nodal/activin signaling drives self-renewal and tumorigenicity of pancreatic cancer stem cells and provides a target for combined drug therapy, Cell Stem Cell 9 (2011) 433–446. [57] S. Singh, S.K. Srivastava, A. Bhardwaj, L.B. Owen, A.P. Singh, CXCL12-CXCR4 signalling axis confers gemcitabine resistance to pancreatic cancer cells: a novel target for therapy, Br. J. Cancer 103 (2010) 1671–1679. [58] N.V. Rajeshkumar, Z.A. Rasheed, E. Garcia-Garcia, F. Lopez-Rios, K. Fujiwara, W.H. Matsui, et al., A combination of DR5 agonistic monoclonal antibody with gemcitabine targets pancreatic cancer stem cells and results in long-term disease control in human pancreatic cancer model, Mol. Cancer Ther. 9 (2010) 2582–2592. [59] P. Ceppi, M.E. Peter, MicroRNAs regulate both epithelial-to-mesenchymal transition and cancer stem cells, Oncogene 33 (2014) 269–278. [60] M. Dillhoff, J. Liu, W. Frankel, C. Croce, M. Bloomston, MicroRNA-21 is overexpressed in pancreatic cancer and a potential predictor of survival, J. Gastrointest. Surg. 12 (2008) 2171–2176. [61] T. Moriyama, K. Ohuchida, K. Mizumoto, J. Yu, N. Sato, T. Nabae, et al., MicroRNA-21 modulates biological functions of pancreatic cancer cells including their proliferation, invasion, and chemoresistance, Mol. Cancer Ther. 8 (2009) 1067–1074. [62] E. Giovannetti, N. Funel, G.J. Peters, M. Del Chiaro, L.A. Erozenci, E. Vasile, et al., MicroRNA-21 in pancreatic cancer: correlation with clinical outcome and pharmacologic aspects underlying its role in the modulation of gemcitabine activity, Cancer Res. 70 (2010) 4528–4538. [63] H. Cheng, S. Shi, X. Cai, J. Long, J. Xu, C. Liu, et al., microRNA signature for human pancreatic cancer invasion and metastasis, Exp. Ther. Med. 4 (2012) 181–187. [64] M. Agostini, R.A. Knight, miR-34: from bench to bedside, Oncotarget 5 (2014) 872–881. [65] Q. Ji, X. Hao, M. Zhang, W. Tang, M. Yang, L. Li, et al., MicroRNA miR-34 inhibits human pancreatic cancer tumor-initiating cells, PLoS ONE 4 (2009) e6816. [66] Y. Lu, J. Lu, X. Li, H. Zhu, X. Fan, S. Zhu, et al., MiR-200a inhibits epithelialmesenchymal transition of pancreatic cancer stem cell, BMC Cancer 14 (2014) 85.

H. Zhan et al./Cancer Letters 357 (2015) 429–437

[67] S. Hasegawa, H. Eguchi, H. Nagano, M. Konno, Y. Tomimaru, H. Wada, et al., MicroRNA-1246 expression associated with CCNG2-mediated chemoresistance and stemness in pancreatic cancer, Br. J. Cancer (2014). [68] U. Wellner, J. Schubert, U.C. Burk, O. Schmalhofer, F. Zhu, A. Sonntag, et al., The EMT-activator ZEB1 promotes tumorigenicity by repressing stemnessinhibiting microRNAs, Nat. Cell Biol. 11 (2009) 1487–1495. [69] R. May, S.M. Sureban, S.A. Lightfoot, A.B. Hoskins, D.J. Brackett, R.G. Postier, et al., Identification of a novel putative pancreatic stem/progenitor cell marker DCAMKL-1 in normal mouse pancreas, Am. J. Physiol. Gastrointest. Liver Physiol. 299 (2010) G303–G310. [70] S.M. Sureban, R. May, S.A. Lightfoot, A.B. Hoskins, M. Lerner, D.J. Brackett, et al., DCAMKL-1 regulates epithelial-mesenchymal transition in human pancreatic cells through a miR-200a-dependent mechanism, Cancer Res. 71 (2011) 2328–2338. [71] S.M. Sureban, R. May, D. Qu, N. Weygant, P. Chandrakesan, N. Ali, et al., DCLK1 regulates pluripotency and angiogenic factors via microRNA-dependent mechanisms in pancreatic cancer, PLoS ONE 8 (2013) e73940. [72] J.P. Thiery, H. Acloque, R.Y. Huang, M.A. Nieto, Epithelial-mesenchymal transitions in development and disease, Cell 139 (2009) 871–890. [73] A. Puisieux, T. Brabletz, J. Caramel, Oncogenic roles of EMT-inducing transcription factors, Nat. Cell Biol. 16 (2014) 488–494. [74] Y. Li, D. Kong, A. Ahmad, B. Bao, F.H. Sarkar, Pancreatic cancer stem cells: emerging target for designing novel therapy, Cancer Lett. 338 (2013) 94– 100. [75] S.A. Mani, W. Guo, M.J. Liao, E.N. Eaton, A. Ayyanan, A.Y. Zhou, et al., The epithelial-mesenchymal transition generates cells with properties of stem cells, Cell 133 (2008) 704–715. [76] F. Vesuna, A. Lisok, B. Kimble, V. Raman, Twist modulates breast cancer stem cells by transcriptional regulation of CD24 expression, Neoplasia 11 (2009) 1318–1328. [77] M.H. Yang, D.S. Hsu, H.W. Wang, H.J. Wang, H.Y. Lan, W.H. Yang, et al., Bmi1 is essential in Twist1-induced epithelial-mesenchymal transition, Nat. Cell Biol. 12 (2010) 982–992. [78] H. Zhu, D. Wang, L. Zhang, X. Xie, Y. Wu, Y. Liu, et al., Upregulation of autophagy by hypoxia-inducible factor-1alpha promotes EMT and metastatic ability of CD133+ pancreatic cancer stem-like cells during intermittent hypoxia, Oncol. Rep. (2014). [79] W. Zhou, R. Lv, W. Qi, D. Wu, Y. Xu, W. Liu, et al., Snail contributes to the maintenance of stem cell-like phenotype cells in human pancreatic cancer, PLoS ONE 9 (2014) e87409. [80] Z. Wang, S. Ali, S. Banerjee, B. Bao, Y. Li, A.S. Azmi, et al., Activated K-Ras and INK4a/Arf deficiency promote aggressiveness of pancreatic cancer by induction of EMT consistent with cancer stem cell phenotype, J. Cell. Physiol. 228 (2013) 556–562. [81] S. Takao, Q. Ding, S. Matsubara, Pancreatic cancer stem cells: regulatory networks in the tumor microenvironment and targeted therapy, J. Hepatobiliary Pancreat Sci. 19 (2012) 614–620. [82] S. Hamada, A. Masamune, T. Shimosegawa, Alteration of pancreatic cancer cell functions by tumor-stromal cell interaction, Front. Physiol. 4 (2013) 318. [83] S.T. Boyle, M. Kochetkova, Breast cancer stem cells and the immune system: promotion, evasion and therapy, J. Mammary Gland Biol. Neoplasia 19 (2014) 203–211. [84] S. Chen, E.H. Huang, The colon cancer stem cell microenvironment holds keys to future cancer therapy, J. Gastrointest. Surg. 18 (2014) 1040–1048. [85] Z.A. Rasheed, J. Yang, Q. Wang, J. Kowalski, I. Freed, C. Murter, et al., Prognostic significance of tumorigenic cells with mesenchymal features in pancreatic adenocarcinoma, J. Natl Cancer Inst. 102 (2010) 340–351. [86] Z.A. Rasheed, W. Matsui, Biological and clinical relevance of stem cells in pancreatic adenocarcinoma, J. Gastroenterol. Hepatol. 27 (Suppl. 2) (2012) 15–18. [87] Z. Liang, X. Bian, H. Shim, Inhibition of breast cancer metastasis with microRNA-302a by downregulation of CXCR4 expression, Breast Cancer Res. Treat. 146 (2014) 535–542. [88] B. Shen, M.Q. Zheng, J.W. Lu, Q. Jiang, T.H. Wang, X.E. Huang, CXCL12-CXCR4 promotes proliferation and invasion of pancreatic cancer cells, Asian Pac. J. Cancer Prev. 14 (2013) 5403–5408. [89] Y. Zhang, J. Wei, H. Wang, X. Xue, Y. An, D. Tang, et al., Epithelial mesenchymal transition correlates with CD24+CD44+ and CD133+ cells in pancreatic cancer, Oncol. Rep. 27 (2012) 1599–1605. [90] A. Diaz-Lopez, G. Moreno-Bueno, A. Cano, Role of microRNA in epithelial to mesenchymal transition and metastasis and clinical perspectives, Cancer Manag. Res. 6 (2014) 205–216. [91] S. Kruger, M. Haas, S. Ormanns, S. Bachmann, J.T. Siveke, T. Kirchner, et al., Translational research in pancreatic ductal adenocarcinoma: current evidence and future concepts, World J. Gastroenterol. 20 (2014) 10769–10777.

437

[92] A. Mittal, D. Chitkara, S.W. Behrman, R.I. Mahato, Efficacy of gemcitabine conjugated and miRNA-205 complexed micelles for treatment of advanced pancreatic cancer, Biomaterials 35 (2014) 7077–7087. [93] C.J. Lee, J. Dosch, D.M. Simeone, Pancreatic cancer stem cells, J. Clin. Oncol. 26 (2008) 2806–2812. [94] V. Richard, M.G. Nair, T.R. Santhosh Kumar, M.R. Pillai, Side population cells as prototype of chemoresistant, tumor-initiating cells, Biomed. Res. Int. 2013 (2013) 517237. [95] M. Cojoc, K. Mabert, M.H. Muders, A. Dubrovska, A role for cancer stem cells in therapy resistance: cellular and molecular mechanisms, Semin. Cancer Biol. (2014). [96] S.P. Hong, J. Wen, S. Bang, S. Park, S.Y. Song, CD44-positive cells are responsible for gemcitabine resistance in pancreatic cancer cells, Int. J. Cancer 125 (2009) 2323–2331. [97] P.C. Hermann, S.M. Trabulo, B. Sainz Jr., A. Balic, E. Garcia, S.A. Hahn, et al., Multimodal treatment eliminates cancer stem cells and leads to long-term survival in primary human pancreatic cancer tissue xenografts, PLoS ONE 8 (2013) e66371. [98] D. Wang, H. Zhu, Y. Zhu, Y. Liu, H. Shen, R. Yin, et al., CD133(+)/CD44(+)/ Oct4(+)/Nestin(+) stem-like cells isolated from Panc-1 cell line may contribute to multi-resistance and metastasis of pancreatic cancer, Acta Histochem. 115 (2013) 349–356. [99] E. Lonardo, P.C. Hermann, C. Heeschen, Pancreatic cancer stem cells – update and future perspectives, Mol. Oncol. 4 (2010) 431–442. [100] A. Van den Broeck, L. Gremeaux, B. Topal, H. Vankelecom, Human pancreatic adenocarcinoma contains a side population resistant to gemcitabine, BMC Cancer 12 (2012) 354. [101] T.L. Fitzgerald, J.A. McCubrey, Pancreatic cancer stem cells: association with cell surface markers, prognosis, resistance, metastasis and treatment, Adv. Biol. Regul. 56 (2014) 45–50. [102] G. Feldmann, V. Fendrich, K. McGovern, D. Bedja, S. Bisht, H. Alvarez, et al., An orally bioavailable small-molecule inhibitor of Hedgehog signaling inhibits tumor initiation and metastasis in pancreatic cancer, Mol. Cancer Ther. 7 (2008) 2725–2735. [103] G. Feldmann, S. Dhara, V. Fendrich, D. Bedja, R. Beaty, M. Mullendore, et al., Blockade of hedgehog signaling inhibits pancreatic cancer invasion and metastases: a new paradigm for combination therapy in solid cancers, Cancer Res. 67 (2007) 2187–2196. [104] C. Moyret-Lalle, E. Ruiz, A. Puisieux, Epithelial-mesenchymal transition transcription factors and miRNAs: “Plastic surgeons” of breast cancer, World J. Clin. Oncol. 5 (2014) 311–322. [105] E. Karamitopoulou, Role of epithelial-mesenchymal transition in pancreatic ductal adenocarcinoma: is tumor budding the missing link? Front. Oncol. 3 (2013) 221. [106] X.M. Ding, MicroRNAs: regulators of cancer metastasis and epithelialmesenchymal transition (EMT), Chin. J. Cancer 33 (2014) 140–147. [107] A. Balic, M.D. Sorensen, S.M. Trabulo, B. Sainz Jr., M. Cioffi, C.R. Vieira, et al., Chloroquine targets pancreatic cancer stem cells via inhibition of CXCR4 and hedgehog signaling, Mol. Cancer Ther. 13 (2014) 1758–1771. [108] M.T. Mueller, P.C. Hermann, J. Witthauer, B. Rubio–Viqueira, S.F. Leicht, S. Huber, et al., Combined targeted treatment to eliminate tumorigenic cancer stem cells in human pancreatic cancer, Gastroenterology 137 (2009) 1102–1113. [109] C. Coco, G.F. Zannoni, E. Caredda, S. Sioletic, A. Boninsegna, M. Migaldi, et al., Increased expression of CD133 and reduced dystroglycan expression are strong predictors of poor outcome in colon cancer patients, J. Exp. Clin. Cancer Res. 31 (2012) 71. [110] S. Piscuoglio, F.S. Lehmann, I. Zlobec, L. Tornillo, W. Dietmaier, A. Hartmann, et al., Effect of EpCAM, CD44, CD133 and CD166 expression on patient survival in tumours of the ampulla of Vater, J. Clin. Pathol. 65 (2012) 140–145. [111] R.C. Langan, J.E. Mullinax, M.T. Raiji, T. Upham, T. Summers, A. Stojadinovic, et al., Colorectal cancer biomarkers and the potential role of cancer stem cells, J. Cancer 4 (2013) 241–250. [112] Y. Ohara, T. Oda, M. Sugano, S. Hashimoto, T. Enomoto, K. Yamada, et al., Histological and prognostic importance of CD44(+)/CD24(+)/EpCAM(+) expression in clinical pancreatic cancer, Cancer Sci. 104 (2013) 1127–1134. [113] Q. Ding, Y. Miyazaki, K. Tsukasa, S. Matsubara, M. Yoshimitsu, S. Takao, CD133 facilitates epithelial-mesenchymal transition through interaction with the ERK pathway in pancreatic cancer metastasis, Mol. Cancer 13 (2014) 15. [114] A. Van den Broeck, H. Vankelecom, W. Van Delm, L. Gremeaux, J. Wouters, J. Allemeersch, et al., Human pancreatic cancer contains a side population expressing cancer stem cell-associated and prognostic genes, PLoS ONE 8 (2013) e73968. [115] L. Chen, J. Fan, H. Chen, Z. Meng, Z. Chen, P. Wang, et al., The IL-8/CXCR1 axis is associated with cancer stem cell-like properties and correlates with clinical prognosis in human pancreatic cancer cases, Sci. Rep. 4 (2014) 5911.