Modeling pancreatic pathophysiology using genome editing of adult stem cell-derived and induced pluripotent stem cell (iPSC)-derived organoids

Abstract

In recent years, organoids have become a novel in vitro method to study gastrointestinal organ development, physiology, and disease. An organoid, in short, may be defined as a miniaturized organ that can be grown from adult stem cells in vitro and studied at the microscopic level. Organoids have been used in multitudes of different ways to study the physiology of different human diseases including gastrointestinal cancers such as pancreatic cancer. The development of genome editing based on the bacterial defense mechanism clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 has emerged as a laboratory tool that provides the opportunity to study the effects of specific genetic changes on organ development, physiology, and disease. The CRISPR/Cas9 approach can be combined with organoid technology including the use of induced pluripotent stem cell (iPSC)-derived and tissue-derived organoids. The goal of this review is to provide highlights on the development of organoid technology, and the use of this culture system to study the pathophysiology of specific mutations in the development of pancreatic and gastric cancers.

NEW & NOTEWORTHY The goal of this review is not only to provide highlights on the development of organoid technology but also to subsequently use this information to study the pathophysiology of those specific mutations in the formation of malignant pancreatic and gastric cancer.

INTRODUCTION

For many years, cell culture models have been used to demonstrate and research physiology and pathophysiology such as the more common two-dimensional (2-D) cell cultures as well as the newly emerging three-dimensional (3-D) cell cultures (1–4). Although such cell cultures have provided scientists with answers to many physiological questions, there are limitations such as the alteration of many cell signaling networks, metabolic activity, and nuclear architecture among others (5). In recent years, three-dimensional cell models referred to as “organoids” have been extensively used within the scientific community to answer a number of biological questions. Today, the most basic definition of an “organoid” is a three-dimensional cell structure, grown from primary tissue fragments in different types of culture matrices, that resembles a small organ or the patient’s tumor (6). Nevertheless, this definition has changed over time and even varies today between fields. Many of the definitions depend on different methods that are used to construct the organoid itself. For example, in the field of stem cell biology, an organoid is defined as a “3-D structure grown from stem cells and consisting of organ-specific cell types that self-organizes through cell sorting and spatially restricted lineage commitment” (7). In the field of mammary gland biology, “organoids” can be defined as “primary explants of epithelial ducts into 3-D extracellular matrix (ECM) gels” (8). The definition of what an organoid is has also changed over time going from the first experiments regarding three-dimensional cells created by using the Hang Drop method by Harrison in 1906 (9) to modern experiments creating 3-D organoid cells derived from stem cells and patient-derived tissue.

Figure 1 covers an overview of the history of the development of cell culture techniques, organoids, and induced pluripotent stem cells. Early cell culture included methods such as the Hang Drop method reported in 1906 (10), the first test tube cultures in 1926 (11), the watch glass method in 1929 (12), the grid cell culture method in 1954 (13), and the first use of collagen gels in 1958 (14). These methods were later used to develop the culturing of differentiated cells such as the use of collagen gels to culture differentiated mammary epithelial cells in 1977 (15). During the latter half of the 20th century, other materials such as Matrigel (16) and other cell culture methods such as laminin-rich matrix assays between 1989 and 1992 (17, 18) were developed. As cell culture became more advanced throughout time, scientists began to focus on organoid development. Throughout time, stem cells were derived, including mouse and human embryonic stem cells in 1981 and 1998, respectively (19–21), and eventually induced pluripotent stem cell (iPSC) cultures were reported in 2006 (22). As our knowledge expanded, stem cell biology was used to develop methodology to generate organoids from iPSCs (23–25). Since the discovery of the double helical structure of DNA by Watson and Crick in 1953 and the completion of the human genome project in 2003 (26), the discovery of Cas9 led to the first use of CRISPR/Cas9 system in bacterial cells, followed by the use of clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 for the genetic engineering of colonic organoids in 2015 (27–31), paving the way for genome editing of organoids derived from a number of organ tissues and iPSCs for studies of gastrointestinal diseases including pancreatic and gastric cancers.

Figure 1.

Timeline. A brief history of the development of organoids, cell culture methods, stem cells and genetics, specifically related to CRISPR/Cas9, over time from the first cell culture method created in 1906 to the first use of CRISPR/Cas9 in an organoid system in 2015. CRISPR, clustered regularly interspaced short palindromic repeats.

For the current review, we refer to “organoids” as being three-dimensional cell cultures that may be considered miniature organ-like structures that resemble multiple different kinds of organs and their subsequent microenvironments. In this review, the organoids discussed are either patient or epithelial tissue-derived or pluripotent stem cell (PSC)-derived pancreatic organoids. Specifically, the mini-review discusses: 1) the differences between PSC-derived and patient-derived organoids and how either one may be beneficial in studying pancreatic cancer, 2) how to understand organoid technology may be used to identify genetic mutations that may cause pancreatic cancer, and 3) how inducing genetic mutations within organoids may lead to a novel way to study genetically altered cell pathophysiology.

COMPARISON BETWEEN INDUCED PLURIPOTENT STEM CELL-DERIVED ORGANOIDS AND ADULT STEM CELL-DERIVED ORGANOIDS

Adult stem cell-derived (aSC) or “tissue”-derived organoids are generated from organ-restricted adult stem cells that have the capacity to self-organize into structures that mimic the tissue it was derived from (7, 32, 33). These organoids are epithelial in nature and generally lack a mesenchymal or stromal cellular compartment that can be maintained in culture. The lack of a mesenchymal cellular compartment is a disadvantage to their use in studying stromal/epithelial cross talk. However, the advantages of aSC-derived organoids include the rapid generation of cultures, their use in the study of host-pathogen interactions and gastrointestinal physiology, and importantly their use as preclinical models for the study of cancer biology, targeted and immune therapies, and potentially prioritizing patient treatment (34–36) (Fig. 2).

Figure 2.

Complexity, morphology, and applications of organotype cultures. iPSC, induced pluripotent stem cell.

Induced pluripotent stem cells (iPSCs) are cells that can be programmed to develop into almost any organ in the body that is derived from somatic cells. Although similar to embryonic stem cell technology that has been used in the past, iPSCs have certain advantages when used in research such as being able to be made via somatic cells and having the ability to be collected noninvasively (37). In addition, iPSC-derived organoids have certain factors that make them more advantageous than normal patient-derived organoids in certain situations. Although pancreatic iPSC-derived organoids have not yet been studied, multiple experiments pertaining the investigation of gastric, vascular, and even cerebral iPSC-derived organoids have been done along with several other human organs.

Using (33) gastrointestinal model for the comparison of iPSC-derived versus patient-derived organoids, several key differences between the two appear. One of the main differences is the fact that iPSC-derived organoids take much longer to mature than patient-derived organoids. In addition, the maturation process is much more complex for iPSC organoids and requires diligent supervision. Morphologically, patient-derived organoids only contain one epithelial layer and have no mesenchymal component. Conversely, iPSC-derived organoids contain multiple epithelial layers and contain stroma.

The fact that iPSC-derived organoids contain stromal compartments may be extremely advantageous when studying cancer because of a phenomenon that occurs in cells called cross talk. Cross talk refers to the communication between different compartments of a cell. The most important of these cellular compartments pertaining to this review are the epithelial, mesenchymal, and resident immune compartments. Typically, when a cell is in homeostasis, these cells will act separately; however, in stress they will act together. For example, under stress, such as a wound, the immune compartment will enact a proinflammatory response to clear away unneeded material that could potentially be causing the stress. The mesenchymal, or stromal, compartment which includes resident fibroblasts will begin to make proteins for the extracellular matrix, which will provide structural support for healing. The epithelial compartment will become activated and proliferate. However, in the pancreas when reoccurring injury occurs, these compartments can be set into a cycle of chronic pancreatitis that can transform into pancreatic cancer (38). It is important to keep in mind that in pancreatic cancer, it has been seen that the stromal compartment of the cells do not function as normal. One difference observed in pancreatic cancer cells is the presence of signaling ligands Sonic Hedgehog (SHH) (39). The SHH ligand activates surrounding fibroblasts, which, in turn, create vast amounts of extracellular matrix and hyaluronic acid (HA). The HA creates fluid buildup and high interstitial pressure that can lead to the collapse of blood vessels within pancreatic ductal adenocarcinoma cells (PDAc), which can limit nutrient availability and chemotherapy delivery (40). In addition, the stroma produces and stores multiple growth factors that can change stromal composition and potentially cause cancer (41). Because iPSC-derived organoids have stromal compartments, it is possible to observe and regulate the cross talk within the cell to manipulate some of the potential causes of PDAc and/or better administration of cancer targeting agents such as chemotherapy. The diverse cellular nature of gastrointestinal organoids has made this method an important tool for studies of organ development, physiology, and preclinical cancer models for the development of therapeutic strategies. Figure 2 summarizes the major differences in cellular complexity, morphology, applications, and limitations of use of induced iPSC-derived and tissue-derived gastric organoids. Both cultures exhibit experimental strengths and limitations. For example, tissue-derived epithelial organoids can be rapidly generated in less than 2 wk, whereas PSC-derived organoids require up to 4–5 wk to establish. Both cultures have been used for in vitro studies of organ physiology and host-pathogen interactions, whereas PSC-derived organoids are more suitable for elucidating mechanisms of organ development (25, 42–55).

MAJOR GENETIC MUTATIONS RELATED TO PANCREATIC CANCER

Unlike many other cancers whose most common cause is environmental, one of the most common causes of pancreatic ductal adenocarcinoma (PDAC) is speculated to be genetic. Several genetic mutations such as KRAS, TP53, SMAD4, and CDKN2A have been identified in over 50% of patients with PDAC (56) as well as other mutations that occur in a smaller population of patients with PDAC (∼5%–10%), including KDM6A, BCORL1, RBM10, and MLL3 (also known as KMT2C) (57). This review will focus on the four most common PDAC genetic mutations: KRAS, TP53, SMAD4, and CDKN2A. Scientific advancement has allowed scientists to rapidly and efficiently engineer organoids using the CRISPR/Cas9 system that is discussed in the following section. Thus, the relevance of newly identified subtypes of pancreatic ductal adenocarcinoma could potentially be studied using genetically engineered iPSC-derived pancreatic organoids (58). The foundational study by Bailey et al. (58) revealed 32 recurrently mutated genes that aggregate pathways including KRAS, transforming growth factor-β (TGF-β), Wingless-related integration site (WNT), Notch homolog 1, translocation-associated (Drosophila) (NOTCH), Roundabout/SLIT (ROCO/SLIT), signaling, G1/S transition, SWItch/sucrose nonfermentable (SWI-SNF), chromatin modification, DNA repair, and RNA processing. Importantly, the investigators defined four subtypes of disease including: 1) squamous, 2) pancreatic progenitor, 3) aberrantly differentiated endocrine exocrine, and 4) immunogenic that correlated with the histology of the patient’s tumor. Squamous tumors were enriched for TP53 and KDM6A mutations, upregulation of the TP63ΔN transcriptional network, hypermethylation of pancreatic endodermal cell-fate determining genes, and had a poor prognosis. Pancreatic progenitor tumors preferentially expressed genes involved in early pancreatic development such as FOXA2/3, PDX1, and MNX1. Aberrantly differentiated endocrine exocrine tumors exhibited upregulation in KRAS activation, exocrine (NR5A2 and RBPJL), and endocrine differentiation (NEUROD1 and NKX2-2) genes. Finally, immunogenic tumors showed increased immune networks including those pathways involved in immune suppression. Combining organoid and CRISPR/Cas9 genome editing technologies is an opportunity for therapeutic development targeting these newly identified subtypes of pancreatic cancer (58). The relevance and function of the major mutations identified in the pancreatic cancer subtypes are summarized as follows:

KRAS

Kirsten Rat Sarcoma Viral Oncogene Homolog (KRAS) is one of the most common cancer-causing mutations in humans, being present in almost 25% of all tumors. In addition, the KRAS mutation is not only specific to pancreatic cancer but also occurs in many other cancers, notably non-small-cell lung cancer and colon cancer (59). KRAS consists of a family of hereditary point mutations that along with several other genes including the v-Ha-ras Harvey rat sarcoma viral oncogene and the neuroblastoma RAS viral oncogene encode for a group of guanosine triphosphate (GTP)-binding proteins that, when activated, can set off a signaling cascade that can, in turn, regulate cell growth, differentiation, and apoptosis. When a mutation occurs in the KRAS gene, the mutation can cause the GTP-binding proteins to not be able to dephosphorylate to its guanosine diphosphate (GDP) bound or inactive state. Thus, the signaling proteins are stuck in their active state causing more growth to be signaled in the cell than is needed (60). A notable feature of cells with KRAS mutations is their subsequent resistance to chemotherapy; however, the reason for this resistance is still not known. One hypothesis is that KRAS is strongly associated with CD44 variant isoform 9, which is a cancer stem cell marker. Research has been done that shows that CD44 plays a role in making cells resistant to chemotherapy treatment through making the cells resistant to that form of induced apoptosis in non-small-cell lung adenocarcinoma (61). The question of why KRAS mutations make cells resistant to chemotherapy is one of the main reasons why the topics covered in this article are planning to be studied. In addition, it is also hypothesized that the immune microenvironment may play a role in sustaining the KRAS gene mutation (62). Specifically, with the use of iPSC-derived organoids, this topic may be able to be studied in further depth.

TP53

TP53 is a tumor suppressor gene that can stop the growth of and/or trigger apoptosis in cells in which DNA damage has occurred through transcriptional regulation. Along with KRAS, TP53 is one of the leading genetic causes of cancer in general, and occurs in ∼70% of all pancreatic cancers (63). In normal cells, TP53 can halt the cell cycle by producing a protein that can bind to the DNA at specific sequences, thus inhibiting transcription. If the TP53 protein cannot bind to the DNA, then the cell cycle will not be halted and the cell will be able to proliferate, potentially turning into cancer (64). Similarly, to KRAS, the TP53 mutation can be hereditary or be acquired later in life through random mutation.

SMAD4

The SMAD4 gene also creates a tumor suppressor protein that is activated by the TGFβ receptor and moved to the nucleus in the form of the SMAD2/SMAD3-SMAD4 complex. In the nucleus, the complex can activate the expression of genes that halt cell growth (63). Studies have been done that show that during tumorigenesis in pancreatic cancer, point mutations, and amino acid substitutions have been found in the carboxyl end of the MH2 region of the SMAD4 protein as well as the amino end of the MH1 region. The MH1 region of the protein has specific DNA binding properties, whereas the MH2 region interacts with other SMAD proteins and contains a transcriptional activation domain (65). These mutations in the SMAD4 gene make it so that the SMAD2/SMAD3-SMAD4 complex becomes nonfunctioning or cannot be made at all. Therefore, if the complex cannot be made or does not function properly, it cannot halt cell growth like it is supposed to, leading to the overproliferation of cells, and eventually cancer. The SMAD4 gene can be inherited through a germline mutation or can be acquired later in life.

CDKN2A

CDKN2A is another tumor suppressor gene that stops the cell from going into the replication phase (S phase) of the cell cycle by inhibiting the cyclinD-CDK4 and cyclinD-CDK6 complexes. CDKN2A is one of the most common mutations found in patients with pancreatic cancer with inactivation of the gene occurring in almost 98% of patients (66). However, unlike the other mutations discussed, mutations in CDKN2A can occur through many different kinds of mutations. These include homozygous deletion, promoter silencing, and heterozygosity (63). Furthermore, CDKN2A-associated pancreatic cancer has been found to be associated with the occurrence of familial atypical multiple mole melanoma. Therefore, the occurrence of the two cancers together are signs of a hereditary CDKN2A mutation (67).

COMBINING GENOME EDITING WITH ORGANOID TECHNOLOGY FOR THE STUDY OF GASTROINTESTINAL DISEASES

Introducing genetic mutations into pancreatic organoids is a meticulous process that differs depending on the type of organoid the mutation is being introduced into. For example, the methodology for forming and differentiating patient-derived organoids is different than for iPSC-derived organoids. Furthermore, the protocols for introducing genetic mutations into both kinds of pancreatic organoids has not been published before. Therefore, the methodology must be taken from introducing the mutations into organoids derived from other organs, predominantly gastric organoids. The methods for deriving organoids from iPSC cells and patient cells are fairly similar (32) with the exception that for iPSC-derived organoids, the iPSC cells must be created before an organoid can be derived from it. The main question comes with how the genetic mutations specified earlier will be introduced into the organoids. This will be done with the use of CRISPR/Cas 9 after the differentiation of the iPSC cells into pancreatic organoids.

Clustered regularly interspaced short palindromic repeats (CRISPR) is a genetic modification tool that utilizes bacterial RNA-guided DNA cleavage system, originally used to acquire resistance against viruses, to splice out a section of DNA and replace it with foreign DNA (68). The CRISPR/Cas 9 system works by joining mature crRNA, otherwise known as CRISPR-RNA, with a kind of small RNA that is complementary to the wanted CRISPR gene sequence called tracrRNA to form a complex. This complex subsequently guides the Cas 9 protein to the target site. The Cas 9 protein can then cleave the targeted piece of DNA. Once cut, DNA repair mechanisms already existent in the cell can be used to add, delete, or change segments of DNA by replacing the cut segment with a customized segment of DNA (31, 69, 70).

Over the years, CRISPR/Cas9 has been used to introduce genetic mutations in many different kinds of organoids including gastric-derived (71–73), liver-derived (74), and kidney-derived (75) organoids. Experiments using CRISPR in organoid models allows researchers to study how certain genetic mutations physically cause disease and the physiological mechanisms that mutations change in a controlled environment. Furthermore, introducing the genetic mutations mentioned in Major Genetic Mutations Related To Pancreatic Cancer into pancreatic patient-derived and iPSC-derived organoids, changes in the cell physiology of the organoids can be carefully reviewed to not only possibly decipher the mechanism by which the mutation causes cancer but also compare and contrast which cell model would be the best to use. Introducing genetic mutations into pancreatic-derived organoids using CRISPR has not been done before. Thus, determining the best model to use would be extremely beneficial to the scientific community.

CRISPR/Cas9 has not been used before to induce genetic mutations in iPSC-derived pancreatic organoids, there is no definitive guide to how this will be done. However, experiments such as that done in Fig. 3 can be used to aid in the development of such protocols. Figure 3 is an example of the combination of CRISPR/Cas9 genome-editing of human-derived gastric organoids using nucleofection to mimic gastric cancer that is associated with the p53 mutation (32). In this experiment, once organoids had grown from single cells (∼12–14 days) after nucleofection (Fig. 3, A–D), organoids were placed under nutlin-3 selection, an inhibitor of mouse double minute 2 (MDM2) (76, 77) (Fig. 3E). Notably, organoids expressing a deletion of p53 were readily recovered from cultures in medium containing nutlin-3, which degrades p53 (Fig. 3E). In addition, deletion of p53 in normal gastric organoids resulted in significant morphological/dysplastic changes (Fig. 3, C and D) compared with the normal controls (Fig. 3B). The genome-editing knockout of p53 in human-derived organoids is a driver for gastric tumorigenesis as reflected in the xenograft model (Fig. 3, F and G). The p53KO organoids efficiently formed tumors (Fig. 3, F and G) that mimicked adenocarcinoma (Fig. 3H). Immunofluorescence using antibodies specific for human histone and gastric cancer marker CD44v9 showed the expression of these human markers within the mouse xenografts generated with p53KO organoids (Fig. 3I). These data also emphasize the ability to engineer human-derived gastric organoids using CRISPR/Cas9 genome-editing to model gastric cancer in vitro.

Figure 3.

Oncogenic transformation of normal human-derived gastric organoids by genetic deletion of tumor suppressor p53. A: schematic diagram showing the approach taken for genome editing of tissue-derived gastric organoids. Light microscope images of scramble control (B) and p53KO human gastric organoids (C and D). E: light microscope images of scramble control and p53KO human gastric organoids treated with vehicle or Nutlin-3 for 48 h. F: mouse xenografts of scramble control or p53KO human-derived organoids. G: tumor volume over time of scramble control or p53KO gastric organoids injected into NOD scid gamma (NSG mice). Histology (H) and (I) immunofluorescence of tumors isolated from NSG mice injected with p53KO organoids.

Once iPSC-derived pancreatic organoids are established as a good model for studying CRISPR-induced genetic mutations, a study similar to that described in Fig. 3 could be done to study the morphological effects, microenvironmental changes, and pathways that contribute to how mutations such as KRAS, TP53, SMAD4, and CDKN2A contribute to the development of pancreatic ductal adenocarcinoma. In the case of studying these genetic mutations in the pancreas, the iPSC cells would have to first be grown, then differentiated, and then use CRIPR/Cas9 to induce the mutations in the cells. In the future, this could lead the way to the formation of new detection methods and targeted therapies to treat patients with pancreatic cancer.

POTENTIAL CLINICAL IMPLICATIONS

Pancreatic cancer remains one of the deadliest forms of cancer in the world due to its hard to detect nature and the fact that many typical cancer treatments often do not work. The use of patient-derived and iPSC-derived organoids to study how specific genetic mutations affect the physiology and pathophysiology of pancreatic cancer cells is a new, novel mechanism that may be able to pave the path to a more reliable detection and effective treatment for pancreatic cancer. However, the development of the iPSC-derived pancreatic organoid model with CRISPR/Cas9-induced genetic mutations first and foremost needs to be established as an effective method to study genetic mutations associated with PDAC. Once this is done, future directions may include studying the effect of specific genetic mutations on pancreatic tumor microenvironment. The use of organoids to study the tumor microenvironment is a new concept due to the fact that patient-derived organoids do not have the ability to show epithelial-mesenchymal interactions. The use of iPSC-derived organoids presents a way to study tumor microenvironments without the use of in vivo studies, which is beneficial due to the complexity of working with live animals. Furthermore, it is hypothesized that specific genetic mutations may only be induced when a cell is put in a specific microenvironment. By changing the microenvironment of the organoids, one can study what may trigger a specific genetic mutation as well as potential changes in therapeutic response (78–80). Lastly, the use of patient-derived organoids offer a new way to study pancreatic cancer and its associated genetic mutations, but also other of GI cancers and related genetic mutations. Recent advancements in the genetic editing of organoids to study GI cancers includes the use of CRISPR/Cas9 genome engineering to introduce genetic alterations such as BRAF, APC, KRAS, and TP53 in colonic organoids (81, 82), and Alk, Bclaf3, or Prkra in gastric organoids (83).

GRANTS

This work was supported by National Institutes of Health (NIH) National Institute of Allergy and Infectious Diseases Grant 5U19AI11649105 (PIs: Weiss and Wells, Project Leader 1: Zavros) (to N.S.) and NIH National Institute of Diabetes and Digestive and Kidney Diseases Grant R01DK083402-10 (PI: Zavros).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

S.T.H., N.S., and Y.Z. conceived and designed research; N.S. and Y.Z. performed experiments; N.S. and Y.Z. analyzed data; S.T.H., N.S., and Y.Z. interpreted results of experiments; S.T.H., N.S., and Y.Z. prepared figures; S.T.H., N.S., and Y.Z. drafted manuscript; S.T.H., N.S., and Y.Z. edited and revised manuscript; S.T.H., N.S., and Y.Z. approved final version of manuscript.