The present protocol describes the reprogramming of Pancreatic Ductal Adenocarcinoma (PDAC) and normal pancreatic ductal epithelial cells into induced pluripotent stem cells (iPSCs). We provide an optimized and detailed, step-by-step procedure, from preparing lentivirus to establishing stable iPSC lines.
The generation of induced pluripotent stem cells (iPSCs) using transcription factors has been achieved from almost any differentiated cell type and has proved highly valuable for research and clinical applications. Interestingly, iPSC reprogramming of cancer cells, such as pancreatic ductal adenocarcinoma (PDAC), has been shown to revert the invasive PDAC phenotype and override the cancer epigenome. The differentiation of PDAC-derived iPSCs can recapitulate PDAC progression from its early pancreatic intraepithelial neoplasia (PanIN) precursor, revealing the molecular and cellular changes that occur early during PDAC progression. Therefore, PDAC-derived iPSCs can be used to model the earliest stages of PDAC for the discovery of early-detection diagnostic markers. This is particularly important for PDAC patients, who are typically diagnosed at the late metastatic stages due to a lack of reliable biomarkers for the earlier PanIN stages. However, reprogramming cancer cell lines, including PDAC, into pluripotency remains challenging, labor-intensive, and highly variable between different lines. Here, we describe a more consistent protocol for generating iPSCs from various human PDAC cell lines using bicistronic lentiviral vectors. The resulting iPSC lines are stable, showing no dependence on the exogenous expression of reprogramming factors or inducible drugs. Overall, this protocol facilitates the generation of a wide range of PDAC-derived iPSCs, which is essential for discovering early biomarkers that are more specific and representative of PDAC cases.
Pancreatic ductal adenocarcinoma (PDAC) is one of the most fatal malignancies, and early diagnosis remains challenging due to the asymptomatic nature of the disease. The majority of PDAC patients are diagnosed at the advanced metastatic stage when very limited treatment options are available1,2. This is mainly due to the lack of reliable biomarkers for the earlier stages, such as those that could be conveniently detected as proteins released into the bloodstream.
PDAC can disseminate very early during its progression, and a better prognosis has been linked to early cancer detection when PDAC is localized in the pancreas3. However, less than a tenth of PDAC patients are diagnosed with a favorable prognosis, allowing for surgical resection. Nonetheless, those few with resectable tumors are also prone to tumor recurrence within 12 months4.
In the past five decades, remarkable improvements have been made in surgical techniques, patient care, and treatment modalities5,6. However, the 5-year survival rate in surgically resected PDAC patients has barely risen to 17%. Nonetheless, this is still better than that in non-resected patients, which has remained almost unchanged (0.9%)4,7. Chemotherapy is the only other alternative PDAC treatment. Yet, this option is very limited as the great majority of PDAC patients exhibit strong resistance to chemotherapy medications such as Gemcitabine7,8. Other drugs, such as Erlotinib, are only available to a small group of PDAC patients with specific mutations, most of whom show Erlotinib resistance9. The adverse side effects associated with chemotherapy in most PDAC patients are yet another disadvantage of this treatment10. Recently, promising strategies have shown that immune checkpoint inhibitors (ICIs) and small molecule kinase inhibitors (SMKIs) can be effective in treating PDAC, but durable responses to these targeted therapies remain limited to a minority of patients11,12. Overall, the discovery of PDAC-specific early biomarkers can pave new avenues for early diagnosis and treatment.
PDAC develops from pancreatic intraepithelial neoplasms (PanIN) precursor lesions that result from non-invasive pancreatic duct epithelial proliferations13,14. While the formation of PanIN is initiated by oncogene mutations such as KRAS, additional genetic and epigenetic alterations are required for the progression to PDAC. It has been projected that the progression of PanIN through the different stages into invasive PDAC takes about 10 years13,15,16,17. This timeframe provides a great opportunity to benefit from early PDAC diagnosis. Therefore, extensive research has been carried out to establish tumor xenograft animal models and organoid cultures to study PDAC progression18,19,20,21. These models have been very useful for studying the invasive stages of PDAC, although not the transition from the early PanIN phases. It is, therefore, important to develop experimental models that can recapitulate the early progression of PanIN stages to enable the discovery of early detection biomarkers.
Reprogramming somatic cells into induced pluripotent stem cells (iPSCs) using the four transcription factors OCT4, SOX2, KLF4, and c-MYC (OSKM) has illustrated the extent of cellular plasticity22. Cancer cell plasticity has been well-documented, and reprogramming human cancer cells into iPSCs has been successfully used to reset cells to their original cellular state, removing many of the epigenetic insults that have accumulated during cancer progression23,24,25,26,27,28,29. The possibility of using this reprogramming strategy to manipulate cancer cell identity has, therefore, presented great promise in treating cancer30,31. Indeed, we have previously shown that the differentiation of iPSCs derived from PDACs can recapitulate PDAC progression through the early PanIN stages32. By identifying genes and pathways specific to the early-to-intermediate stages of PDAC, candidate biomarkers were identified that can be clinically used for early PDAC diagnosis32,33. However, the biomarkers discovered using a single iPSC line showed limited coverage in the majority of PDAC patients32. The challenges of generating iPSC lines from other PDAC patients have halted the ability to discover more reliable biomarkers. This is due to many technical factors, including the heterogeneity of OSKM delivery, as only a small portion of human primary PDAC cells contained all four factors and responded successfully to reprogramming. Here, a detailed protocol is presented for reprogramming primary PDAC cells using a more efficient and consistent dual lentiviral delivery of OSKM.
All experimental protocols were approved by the OHSU Institutional Review Board. All methods were carried out in accordance with relevant guidelines and regulations. All animal works for PDX tumors were performed with the OHSU Institutional Animal Use and Care Committee (IACUC) approval. This protocol was tested in Primary PDAC cells from patient-derived xenograft (PDX), BxPc3 cell line exhibiting epithelial morphology that was isolated from the pancreas tissue of a 61-year-old female patient with adenocarcinoma, the H6C7 immortalized epithelial cell line derived from normal human pancreatic duct epithelial, and primary human fibroblasts derived from skin biopsy of healthy individuals. Human PDAC specimens were obtained under the Oregon Pancreas Tissue Registry study (IRB00003609). Informed consent was obtained from all subjects. Human primary fibroblasts were derived in RBiomedical, Edinburgh, UK, from skin samples from anonymous donors undergoing routine surgery at the Edinburgh Royal Infirmary, Little France, UK, under their consent and ethical approval (09/MRE00/91).All lentivirus work has been carried out under Class 2 research activity (GM207/16.6) and approved by the Health and Safety Department at the University of Edinburgh and notified to the HSE competent authority of the Scottish Government. All reprogramming experiments using human pancreatic cells were carried out under the ethical approval of the School of Biological Sciences ethics committee at the University of Edinburgh (reference # asoufi-0002).
1. Preparation of lentiviruses
2. Reprogramming lentivirus transduction
3. Lentivirus transduction of PDAC
4. Lentivirus transduction of BxPc3 cells
5. Lentivirus transduction of H6c7 cells infection
6. Lentivirus transduction of hFib cells
7. Preparation of iMEF feeder cells
8. Transferring the infected cells onto the iMEF feeder layer
9. Passaging of iPSC colonies
10. Live staining of iPSC colonies with TRA-1-60
Representative images displaying the morphology of iPSC colonies derived from PDAC, BXPc3, H6C7, and hFib cells are shown in Figure 1. PDAC-iPSC colonies started to form on Day 25 of reprogramming. Robust iPSC colonies with a more established ESC-like morphology were identified on Day 40 of reprogramming (Figure 1). Similarly, the formation of BxPc3-iPSCs began on Day 23 and became more established by Day 35. H6C7-iPSC formation was similar to PDAC-iPSCs and started to become established on Day 45. hFib-iPSC colonies began to form on Day 15 of reprogramming.
Figure 1: Representative images of iPSC colonies. The images show established hESC-like morphology derived from (A) PDAC, (B) BxPc3, (C) H6c7, and (D) hFib. The days of reprogramming are indicated above each image. Scale bars = 100 µm. Please click here to view a larger version of this figure.
Clonal iPSC lines were established from each reprogramming of PDAC, BxPc3, H6C7, and hFib cells. All established iPSC lines stained positive for TRA-1-60, an undifferentiated hESC cell surface marker, confirming their reprogramming to pluripotency (Figure 2).
Figure 2: Representative images of clonal iPSC lines. The images are derived from (A) PDAC, (B) BxPc3, (C) H6c7, and (D) hFib showing ESC-like morphology (top panels) and positive staining for TRA-1-60 (bottom panel). Scale bars = 100 µm. Please click here to view a larger version of this figure.
To facilitate the use of iPSC reprogramming for studying cancer progression, a robust protocol has been established for reprogramming pancreatic cancer cells. Reprogramming cancer cells into pluripotency has proven to be very challenging thus far, as only a few studies have successfully generated iPSCs from cancer cells32,36,37,38,39,40,41,42,43,44,45,46. Most of these studies used immortalized cancer cell lines to generate iPSC lines, not primary patient-derived cells36,37,38,40,42,43,44,46. For instance, reprogramming four different liver cancer cell lines was attempted using the retroviral introduction of OSKM factors, but only a single cell line was successfully reprogrammed41. However, the generated liver cancer iPSC line showed a loss of stemness after a few passages, highlighting the high resistance of cancer cells to OSKM reprogramming41. Previously, attempts were made to reprogram PDAC cells using lentiviral delivery of OSKM individually; however, only a single iPSC line was generated, dependent on exogenous OSKM expression and therefore not fully reprogrammed37. Another study used episomal vectors to deliver OSKM factors without genomic integration but only managed to generate a single iPSC clone from PDAC39. The limited success in reprogramming cancer cells adds to the many challenges hampering the use of iPSC technology in cancer research.
Here, the generation of iPSCs from primary PDAC samples derived from two different patients and one established PDAC cell line (BxPc3) is explained. Moreover, iPSCs have also been generated from H6c7, a pancreatic ductal epithelial cell line. To our knowledge, this is the first report of successfully derived stable iPSC lines from BxPc3 and H6c7. Following this protocol, iPSCs have also been successfully generated from primary human fibroblasts derived from healthy individuals, expanding the applicability of the method beyond cancer research.
One of the key elements behind the success of the protocol is the use of bicistronic lentiviral vectors to co-express OS and KM factors. These vectors contain the internal ribosome entry site 2 (IRES2) to express OCT4 and SOX2 in one vector and KLF4 and cMYC in the other. Multiple studies using monocistronic vectors where each reprogramming factor was delivered individually have shown that the uptake of each vector is different, affecting OSKM stoichiometry and reprogramming efficiency47. Using bicistronic vectors can help mitigate this issue. Moreover, using IRES in lentiviral vectors has shown a significant increase in reprogramming efficiency48. In this lentivirus system, OS expression is driven by the EF1a promoter, while KM expression is under the control of a CMV promoter. It is known that the CMV promoter can be subjected to highly efficient silencing by DNA methylation and histone deacetylation, unlike EF1a49,50. Therefore, the early silencing of KM before OS during reprogramming may be a key factor for the success of the protocol. This is consistent with previous studies showing the importance of OSKM expression dynamics during reprogramming51,52,53,54,55. Thus, the bicistronic vector is also more advantageous than the polycistronic vector, where all OSKM factors are expressed from a single promoter56. Other factors that contribute to the success of the protocol include the doses of lentivirus infection and the number of infected cells used for reprogramming, which were customized for each cell type.
In summary, an optimized method for reprogramming primary PDAC cells along with other cell lines and normal cells is presented. This method will help expand the use of iPSC reprogramming to model cancer progression and, in this case, discover early PDAC biomarkers.
The authors have nothing to disclose.
A.S and J.K would like to thank Cancer Research UK and OHSU for funding (CRUK-OHSU Project Award C65925/A26986). A.S is supported by an MRC career development award (MR/N024028/1). A.A is funded by a Ph.D. scholarship (Scholarship ref. 1078107040) from King Abdulaziz City for Science and Technology. J.K is funded by MRF New Investigator Grant (GCNCR1042A) and Knight CEDAR grant (68182-933-000, 68182-939-000). We thank Prof Keisuke Kaji for kindly providing the reprogramming vector pSIN4-EF1a-O2S and pSIN4-CMV-K2M. For the purpose of open access, the author has applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising from this submission.
2-Mercaptoethanol (50 mM) | Thermo Fisher | 31350010 |
Alexa Fluor 488 anti- human TRA-1-60-R | BioLegend | 330613 |
Bovine Pituitary Extract (BPE) | Thermo Fisher | 13028014 |
BxPc3 | ATCC | CRL-1687 |
Cholera Toxin from Vibrio cholerae | Merck | C8052-1MG |
Collagen, Type I solution from rat tail | Merck | C3867 |
Completed Defined K-SFM | Thermo Fisher | 10744-019 |
Corning Costar TC-Treated Multiple Well Plates | Merck | CLS3516 |
Corning syringe filters | Merck | CLS431231 |
Corning tissue-culture treated culture dishes | Merck | CLS430599 |
Day Impex Virkon Disinfectant Virucidal Tablets | Thermo Fisher | 12328667 |
Dulbecco′s Phosphate Buffered Saline (PBS) | Merck | D8537 |
Fetal Calf Serum (FCS) | Thermo Fisher | 10270-106 |
Fugene HD Transfection Reagent | Promega | E2312 |
Gelatin solution, Type B, 2% in H2O | Merck | G1393-100ML |
Glasgow Minimum Essential Media (GMEM) | Merck | G5154 |
Human EGF Recombinant Protein | Thermo Fisher | PHG0311 |
Human FGF-basic (FGF-2/bFGF) (154 aa) Recombinant Protein, PeproTech | Thermo Fisher | 100-18B |
Human Pancreatic Duct Epithelial Cell Line (H6c7) | Kerafast | ECA001-FP |
iMEF feeder cells | iXcells Biotechnologies | 10MU-001-1V |
Keratinocyte Serum Free Media (KSFM) | Thermo Fisher | 17005-042 |
KnockOut DMEM | Thermo Fisher | 10829018 |
KnockOut serum Replacement | Thermo Fisher | 10828028 |
L-Glutamine (200 mM) | Thermo Fisher | 25030-024 |
MEM Non-Essential Amino Acids Solution (100x) | Thermo Fisher | 11140050 |
Millex-HP 0.45 μM syringe Filter Unit (Sterile) | Merck | SLHP033RS |
Opti-MEM Reduced Serum Medium | Thermo Fisher | 31985062 |
pMDG | AddGene | 187440 |
Polybrene (Hexadimethrine bromide) | Merck | H9268-5G |
pSIN4-CMV-K2M | AddGene | 21164 |
pSIN4-EF2-O2S | AddGene | 21162 |
psPAX2 | AddGene | 12260 |
pWPT-GFP | AddGene | 12255 |
RPMI 1640 Medium (ATCC modification) | Thermo Fisher | A1049101 |
Sodym Pyruvate | Thermo Fisher | 11360-039 |
Sterile Syringes for Single Use (60 mL) | Thermo Fisher | 15899152 |
TrypLE Express Enzyme (1x), phenol red | Thermo Fisher | 12605036 |
UltraPure 0.5M EDTA, pH 8.0 | Thermo Fisher | 15575020 |
Y-27632 (Dihydrochloride) | STEMCELL Technologies | 72304 |