Abstract
The cancer stem cell (CSC) hypothesis is an evolving concept of oncogenesis that has recently gained wide acceptance. By definition, CSCs exhibit continuous proliferation and self-renewal, and they have been proposed to play significant roles in oncogenesis, tumor growth, metastasis, chemoresistance, and cancer recurrence. The reprogramming of cancer cells using induced pluripotent stem cell (iPSC) technology is a potential strategy for the identification of CSC-related oncogenes and tumor-suppressor genes. This technology has some advantages for studying the interactions between CSC-related genes and the cancer microenvironment. This approach may also provide a useful platform for studying the mechanisms of CSCs underlying cancer initiation and progression. The present review summarizes the recent advances in cancer cell reprogramming using iPSC technology and discusses its potential clinical use and related drug screening.
- Cancer cell reprogramming
- cancer stem cell
- drug screening
- induced pluripotent stem cell
- regenerative medicine
- clinical application
- review
Cancer stem cells (CSCs) have been recognized as a small subset of cells within a tumor that are endowed with stem cell-like properties, including the abilities of self-renewal, pluripotency, cancer generation and drug resistance (1-7). The primary strategy used for inducing CSCs is to enrich the cells using classical stem cell markers such as CD13, CD24, CD44, CD47, CD90 and CD133, followed by other techniques including side-population analysis, sphere formation, and so on (8-11). This cell population is then transplanted into immunodeficient SCID mice to examine its in vivo tumorigenic potential (7-9). Such cells are examined further according to their cancer markers such as WNT, Notch, Hedgehog, transforming growth factor β, epithelial–mesenchymal transition (EMT)/mesenchymal–epithelial transition (MET) signaling proteins, and epigenetic factors (12-15). Putative CSC subpopulations that are capable of initiating tumor development at a lower cell number are tested for self-renewal capacity using serial dilutions of cells to identify the CSCs. In addition to these classical techniques, some techniques, such as reprogramming, are now a research focus, although the driver and the passenger mutation are present in the genome (12-15).
Current cancer cell-reprogramming techniques such as somatic cell nuclear transfer (16) and the generation of induced pluripotent stem cells (iPSCs) (17-19), are used to identify oncogenes, anti-oncogenes and epigenomes. The breakthrough came in 2006, when Takahashi and Yamanaka introduced the concept of iPSCs by generating stem cells with properties related to those of embryonic stem cells (ESCs) (17, 18). The success in reprogramming a somatic cell into a stem cell-like state has led to the idea of reprogramming malignant cells back to their original state well before oncogenic transformation occurs. The generation of iPSCs from cancer cells may provide tools for exploring the mechanisms of tumor initiation and progression in vitro, for studying the plasticity of cancer cells and origin of CSCs, and for achieving cancer type-specific drug discovery (Figure 1).
However, these reprogramming methods remain a challenge because of two problems: the cancer-specific epigenetic state and the chromosomal aberrations or genetic mutations present in cancer cells. The epigenetic memory of the original cell type is critical for reprogramming and is related to the inefficient reprogramming that is caused by a failure to reset the epigenome to an ESC-like state (20). The epigenetic state attempts to reprogram cancer cells that may have produced incomplete resetting of the cancer-associated epigenome because of tumor heterogeneity and further accumulation of oncogenic mutations.
In 2014, the first-in-human clinical trial of iPSC-based cell therapy was conducted. A Japanese elderly woman with exudative age-related macular degeneration received implantation of a retinal pigment epithelial cell sheet that had been differentiated from iPSCs generated from fibroblasts from her own skin. Although this sheet did not improve the patient's vision, it did halt disease progression (21, 22). In 2015, in a second clinical trial using such sheet, the genetic mutations invoIved were identified. However, there was no clear confirmation that these mutations could lead directly to advanced effects of diseases (23, 24). In order to advance iPSC-based novel therapies, it is critical to determine how and when these mutations occur and whether they actually lead to harmful effects. In the above trials, the patients were elderly individuals; thus, the occurrence of mutations might have been facilitated. In order to avoid these difficulties, Yamanaka's group used human leucocyte antigen (HLA)-matched young patients in subsequent trials of iPSC-based cell therapy, which is expect to be more successful (22).
Therefore, cancer cell reprogramming is currently limited to certain cancer types and cancer-specific markers in the epigenome; this impedes successful reprogramming. Moreover, the underlying mechanisms have not been fully elucidated. Thus, further elucidation of these issues may help prevent these alterations. Nevertheless, we expect that this iPSC-based technology and therapy will be a breakthrough in the prevention of cancer generation and progression. In this review, we summarize the features of the iPSC-like cells derived from human cancer for cell therapy and discuss both their merits and demerits regarding clinical and pharmaceutical applications.
The Bilateral Character of Cancer-specific iPSC-Iike Cells
The difficulties encountered in the reprogramming of cancer cells include cancer-specific genetic mutations, chromosomal rearrangements, accumulation of DNA damage, and reprogramming-triggered cellular senescence (25-27). Despite these obstacles, many studies have reported the generation of iPSCs from cancer cells, as summarized in Table I. This has covered a range of cancer cells, including melanoma (28, 29), prostatic (28), gastrointestinal (30), chronic myeloid leukemia (CML) (31), lung (32), breast (33), glioblastoma (34), and sarcoma (35).
The initial success in this field consisted in the successful reprogramming of colon metastatic cells and PC-3 prostate metastatic cells via the expression of intronic miR-302 (28). Subsequently, Miyoshi et al. performed a series of reprogramming studies with different methods using 20 gastrointestinal cancer cell lines, and obtained successful results for eight of them (30). Transduction by a combination of retroviral- or lentiviral-based Octamer-binding protein 4 (OCT4), SRY (sex determining region Y)-box 2 (SOX2), Krüppel-like factor (KLF4), Cellular myelocytomatosis viral oncogene homolog (c-MYC) (OSKM), B cell lymphoma-2 (BCL2), Kirsten rat sarcoma viral oncogene homolog (kRAS), Lin 28 homolog (LIN28), Nanog homeobox (NANOG), then transforming growth factor was added, and shRNA for tumor-suppressor genes for each cell line was used initially to obtain iPSC-like cells that re-expressed NANOG. The eight cell lines from which iPSCs were generated were derived from cholangiocellular carcinoma (HuCCT-1), colorectal (DLD1, HT29), hepatocellular (PLC), gastric (TMKN45), esophageal (YE10), and pancreatic (MIAPPaCa-2, PAV-1). The resultant iPSC-like cells were less tumorigenic as compared with their parental cell lines. Similarly, Noguchi et al. found that PANC1 cells were easily reprogrammed, while three other cell lines, MIAPaCa-2, PSN-1, and AsPC-1, were not (36). Iskender et al. also reported the generation of iPSCs derived from bladder carcinoma T24 cells, but another bladder carcinoma cell line, HTB-9, could not be induced to reprogram (37). Thus, the success of the generation of iPSCs from cancer cells seems to be cell-type specific. This is one of the problems with this technology.
Another problem encountered in this field of research is the lower efficiency of cancer cell reprogramming. This low efficiency in iPSC generation from cancer cells suggests the presence of multiple mechanisms that might be involved in the regulation of reprogramming (37, 38). Mathieu et al. reported that reprogramming factors and Hypoxia-inducible factor 1 alpha (HIF1α) accelerated the induction of iPSCs from the A549 lung carcinoma cell line, suggesting that reprogramming is enhanced by a cumulative effect of environmental hypoxia (32). Moreover, Mohyeldin et al. showed that low oxygen levels promote the self-renewing capacity of stem cells (39). Hypoxia activated the expression of stemness genes, such as OKM and NANOG, and stem-cell-associated miRNAs, in different cancer cell lines that shared an overlapping gene expression signature with human ESC lines (32).
Schematic representation of use of cancer-derived induced Pluripotent stem cell (iPSC)s for cancer biology. Tissue resources may be used to develop human cancer specific iPSCs lines and generate cancer stem cell (CSC)s mechanistic studies of cancer remodeling and drug screening or develop cell-based therapy of human cancer using genomic editing and induced differentiation from organoid 3D cells.
Hematopoietic malignancy with chromosome rearrangement is another challenging issue; disease-specific iPSCs possessing the genetic abnormalities of hematological malignancies would provide an efficient platform for studying pathogenesis. Carette et al. generated iPSCs derived from the CML cell line KBM7 carrying the fusion gene of the breakpoint cluster region protein (BCR)-breakpoint cluster region protein (BCR) (BCR–ABL) via defined factors OSKM (31). An acute myeloid leukemia (AML) mouse model was also generated by retroviral overexpression encoded the human mixed-lineage leukemia-AF9 (MLL–AF9) fusion gene in hematopoietic cells from transgenic mice that carried deoxycycline-inducible four OSKM genes. Upon deoxycycline addition, the MLL–AF9-expressing leukemia cells were efficiently converted into iPSCs that were capable of forming teratomas and producing chimeras. Most of the chimeric mice developed AML spontaneously (40).
Moreover, reports of the generation of iPSCs generated from human primary malignant cells are scarce and are limited to cancers such as leukemia (41-44) and pancreatic cancer (45). Hu et al. (41) used the transgene-free iPSC technology to express OSKM, NANOG, LIN28, and Simian vacuolating virus 40 large T antigen (SV40 LT) genes in primary human lymphoblasts from a BCR–ABL-positive patient with CML (41). Kumano et al. produced iPSCs from samples from imatinib-sensitive patients with CML that became resistant to imatinib despite the expression of the BCR–ABL oncogene (42). Gandre-Babbe et al. (43) and Kotini et al. (44) reported that the reprogramming of cancer cells was feasible despite the presence of genomic alterations in the parental cells, and that iPSCs derived from patients with juvenile myelomonocyte leukemia and with myelodysplatic syndrome recapitulated the disease-associated phenotype. Similarly, pancreatic-cancer-derived iPSC-like cells were also successfully generated from a parental pancreatic ductal adenocarcinoma with a kRASG12D mutation (45). Therefore, a similar concept is found in the reprogramming of normal somatic cells, in which reprogramming-induced multiple genetic/epigenetic abnormalities did not interfere with the differentiation capacity of the resulting iPSCs (46-48). Although expression of reprogramming genes was found to be successful in various primary patient samples of hematological malignancies, Liu et al. reported that NOTCH1-induced T-acute lymphoblastic leukemia could not be reprogrammed into a pluripotent state (40). Therefore, the reprogramming of cancer cells needs to be optimized for each cancer type. For example, Utikal et al. showed that the R545 melanoma cell line could be reprogrammed into iPSCs by introducing OCT4-KLF4-c-MYC (OKM), without ectopic SOX2 requirement (29). The resultant iPSCs were used to generate higher-degree chimeric mice that exhibited competent germline transmission. In contrast, Oshima et al. showed the induction of CSC features in colon cancer cells upon the introduction of OCT4-SOX2-KLF4 (OSK), and found that a subset of colon cancer cells gained cancer properties expressed defined colon CSC markers but not to teratomas in vivo (49). This different reprogramming might be caused by heterogeneity in plasticity or epigenesis.
Summary of studies of reprogramming of human cancer cells to induce pluripotent stem cells (iPSC). Recent reports also demonstrated the success in reducing the tumorigenicity of cancer cells (51, 80-82), or not changing the plasticity of cancer cells (80, 81, 83-85), even exhibiting strong cancerous features such as cancer stem cells (81), even these heterogeneous outputs are due to the status of p53 (86). Modified from the Table in Izgi et al. (87).
Interestingly, Jaenisch's group examined reprogramming activity using nuclear transplantation techniques (transplantation of nuclei from melanoma, leukemia, lymphoma, and breast cancer cells into enucleated oocytes) (50). In fact, all nuclei from primary leukemia and lymphoma cells cannot be reprogramed. A modest percentage of the transplanted nuclei from all cancer cells and transplanted tumors were reprogrammed, and the surviving oocytes went on to develop into blastocysts. By contrast, only blastocysts derived from the melanoma yielded ESCs, indicating that not all cancer genomes can be epigenetically reprogrammed to full pluripotency using the nuclear transplantation. Moreover, chimeras were generated only by using the melanoma nuclear-transferred ESCs. However, the chimeras developed earlier and exhibited higher expansion into tumor cells compared with the original nucleus-donor mouse model. These studies indicate that reprogramming of a primary tumor cells is more difficult in mouse models, and that further technological progress is needed to be able to generate reliable iPSC models of cancer.
Epigenetic Remodeling of Cancer Cells
Epigenetic changes in cancer cells result in reduced or increased aggressive phenotypes of partially reprogrammed iPSCs or iPSC-like cells. Regarding DNA methylation, Moore et al. reported that iPSCs from cancer cells exhibited distinct hypomethylation of the densely methylated regions of the genome, which are specific for cancer cells (51). Stricker et al. showed that glioblastoma cell lines derived from patients with high aneuploidy exhibited erasure of cancer-specific DNA methylation and could be reprogrammed (34). Moreover, reprogramming antagonized the DNA methylations that are significant for non-small cell lung cancer (NSCLC) cell lines, and differentiation of NSCLC-derived iPSCs in vitro did not restore the tumor-specific epigenetic modification (52). iPSCs from glioblastoma-derived neural stem cells exhibited reduced ability to infiltrate into surrounding tissues, suggesting suppression of their aggressive character upon reprogramming (34). Zhang et al. reprogrammed cells from three osteosarcoma, two liposarcoma, and a sarcoma of unknown origin, which altered the epigenetic feature of oncogenes (35). Tumor-suppressor genes render cells with a less aggressive tumor phenotype. However, some studies have suggested the acquisition of sensitivity to anticancer agents in the reprogramming of iPSCs, which is not necessarily an indicator of repression of the malignancy, but shows increased drug sensitivity compared with the parental cells (30). The reactivation of some tumor-suppressor genes, such as p16Ink4a, in iPSCs might lead to increasing chemosensitivity as well as the repressing proliferation and invasiveness in reprogrammed cancer cells (30). Although not all cancer reprogramming studies have analyzed the tumorigenic potential or drug responsiveness of the resulting iPSCs (31, 42, 53), the results of studies contradict the outcomes of Miyoshi et al. (30).
The tumor-suppressor gene products are known to play a critical role in reprogramming to generate iPSCs (54); however, more evidence is needed to draw a conclusion in terms of the role of the suppressor proteins in cancer-specific reprogramming. BMI1 in polycomb repressive complex 1 was demonstrated to increase reprogramming efficiency by replacing the function of KLF4 and c-MYC (55, 56). Another member of the polycomb repressive complex 2, EZH2, is also critical for reprogramming; forced expression of EZH2 enhances, while knockdown of EZH2 impairs, the generation of iPSCs (57-59). The epigenetic study of cancer cells exhibited aberrant epigenetic regulation of the p53-Inhibitors of cyclin-dependent kinase (INK) family network. Thus, the absence or reduced expression of p53 and p21CIP1 favored the generation of iPSCs (60-62). Epigenetic silencing of tumor-suppressor genes, such as though aberrant methylation of the p16INK4A promoter, has been shown to be reversed by reprogramming (63). Therefore, the reprogramming of cancer cells and overcoming of the barriers to pluripotency remain to be solved.
Common epigenetic processes might be involved in reprogramming and in the development of certain cancer types. In fact, global changes in epigenetic modifications that occur in normal cells and cancer cells were demonstrated to be bidirectional rather than unidirectional. Therefore, the application of reprogramming techniques to cancer cells might promote our understanding of the cancer-specific epigenome and elucidate the overlapping mechanisms shared by cancer-initiating and pluripotent cells.
Is it possible to reprogram cells via modification of the epigenetic state? A few authors have reported such reprogramming in the mouse. Hou et al. showed that a combination of small molecules is sufficient for pluripotency and is dispensable for reprogramming in mouse somatic cells (64). Growing evidence suggests that a combination of small molecules in mouse cells could compensate for exogenous reprogramming factors and generated iPSC-like cells with expression profiles and epigenesis similar to those of ESCs (65). Thus, for certain cell types, epigenomic editing could replace the ectopic expression of transcription factors for reprogramming, while for most cell types, the overcoming of epigenetic obstacles warrants a combination of mechanisms induced by forced expression of reprogramming factors and other modifications.
Reprogramming efficiency has been shown to be improved upon treatment with small molecules including inhibitors of DNA methyltransferase, histone deacetylase, WNT signal modulators, modulators of cell senescence, and metabolism (66). Cell origin thoroughly affects reprogramming efficiency, as iPSC induction does not reset the epigenetic memory completely, and the memory of the donor cell may be retained in the iPSCs (67). Incomplete reprogramming with inherited epigenetic memory generates iPSCs that have a tendency to differentiate toward the original lineage (68). Although reprogramming efficiency was further increased after combinational treatment with small molecules, transcription factors, and signaling pathway regulators, efforts should be focused on elucidating the mechanisms that direct terminally differentiated cells to erase their somatic epigenetics and gain pluripotency (69-71). Moreover, some of these observations were made in the mouse, not in humans. Thus, human iPSCs from human cancer cells need to be examined further, and additional information regarding the molecular relationship between epigenetic control and reprogramming should be collected.
Potential Application in Biomedical Research
Human cancer-derived iPSCs can be used to preserve unique genotypes by banking cells that can be differentiated into many cell types. The cancer-derived iPSC model is used for studying the mutation of cancer-related genes and epigenetic alterations in the genome in order to understand the molecular mechanisms underlying tumorigenesis in humans. The use of iPSC technologies has both advantages and disadvantages compared with traditional approaches using cancer cell lines and animal models. High-throughput drug screening using patient-specific iPSCs has been receiving growing attention. Chemotherapy takes a huge toll on patients with cancer because of its undesirable side-effects. A differentiated cytotoxicity screen could lead to the development of drugs that are more specific to their target cells.
Efforts to harness the merits of iPSC technology have been carried out for various neurological disorders (72) and diabetic cardiomyopathy (73). Current recombinant technologies enable precise genomic manipulation in diseased cells. For example, the efficiency of iPSC production can be improved through use of techniques including episomal plasmids, lentivirus-, adenovirus-, or sendai virus-mediated gene transfers. Moreover, the feasibility of genetic manipulation in iPSCs has been demonstrated using several technologies such as knockdown, knockout, and gene correction using homologous recombination, combined with genome-editing tools such as zinc-finger nucleases, Transcription activator-like effector nuclease (TALEN)s, and the Clustered regularly interspaced short palindromic repeats (CRISPER)- CRISPR associated protein 9 (CAS9) system (74, 75). Genetically defective cells could be corrected in vitro and reintroduced into patients. The autologous transplantation approach has been shown to be effective in principle using a humanized mouse model of sickle cell anemia (76). Human iPSCs are a potential source of cells for tissue reconstruction in the long term (77). Saki et al. reported that transplanted hematopoietic precursor cells can be generated from iPSCs, potentially offering new cell sources for cell reconstitution in patients with hematological cancer after treatment (78). Recently, AML patient-derived dermal fibroblasts were reprogrammed into normal iPSCs that did not carry any chromosomal aberrations of the patient's bone marrow cells, and they differentiated into normal hematopoietic progenitor cells (78). The HLA-matched iPSC sources at the iPSC bank of Kyoto University or the RIKEN Cell Bank are now prepared for clinical use (http://www.cira.kyoto-u.ac.jp/e/research/stock.html and http://cell.brc.riken.jp/en/, respectively). The use of three-dimensional (3D) organoid technologies to engineer tissues, such as stomach, small intestine, colon, pancreas and liver, are expected to bring about great advances regarding how we can model human disorders, perform drug screenings, and engineer replacement tissues or organs (79). Human organoid cultures are useful for studies in regenerative medicine and for the therapeutic screening of drugs and small molecules. These engineered 3D tissues can replace intact tissues in the cancer research because they are histologically and functionally more faithful to their in vivo counterparts.
Conclusion
The generation of patient-specific iPSCs from various tissues is revolutionizing the way in which we approach human disease modeling, novel drug development, and autologous/allogenic cell therapy of disorders. In particular, cancer iPSCs offer a new paradigm in cancer modeling and tissue regeneration. Cancer-derived iPSCs may enhance our understanding of the features of tumorigenesis, the effects of microenvironments, and how epigenetic events contribute to the development of various cancer types. This information could be expected to enable the establishment of drug screening platforms, the development of more targetable and less toxic therapies, and effective tissue reconstitution. The study of the reprograming of cancer cells and efforts to harness the versatility of iPSCs for cancer remodeling and for screening effective drugs should contribute to further progress in our understanding of cancer biology.
Acknowledgements
This work was supported in part by MOST 104-2320-B-037-033-My2, and MOST 104-2314-B-037-002, MOST 104-2314-B-037-043 from the Ministry of Science and Technology; NHRI-EX106-10416S1, from the National Health Research Institutes in Taiwan; and KMU-TP105G01, KMU-TP105E21, and KMU-DT106006 from Kaohsiung Medical University in Taiwan. The Authors also thank Wen-Hsin Lin for drawing the Figure and preparing the Table.
Footnotes
This article is freely accessible online.
Competing Interests
The Authors declare that they have no competing interests.
- Received May 19, 2017.
- Revision received May 29, 2017.
- Accepted June 2, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
