Introduction
What are organoids? How can organoids benefit research?
The study of human organ development and diseased states is hampered by the inaccessibility of human samples in vivo and intrinsic differences between animal models and human biology. Advances in three-dimensional (3D) cell culture techniques facilitated by a deeper understanding of extracellular matrix (ECM) biology combined with greater knowledge about signalling pathway regulation of stem cell niches and differentiation programmes have enabled the establishment of organoid culture systems. The term ‘organoid’ has previously been used to refer to a range of 3D culture systems which resemble the modelled organ to varying extents. Here we subscribe to the organoid definition coined by Lancaster and Knoblich1 and Huch and Koo2 to define an organoid as an in vitro 3D cellular cluster derived from tissue-resident stem/progenitor cells, embryonic stem cells (ESCs) or induced pluripotent stem cells (iPSCs) capable of self-renewal and self-organisation that recapitulates the functionality of the tissue of origin. Organoids were named ‘Method of the year 2017’ by Nature Methods,3 reflecting the excitement and promise of this rapidly expanding field to provide new experimentally tractable, physiologically relevant models of organ development, human pathologies and paving the way for therapeutic applications.
In the body, cells reside in complex microenvironments and are subject to numerous signalling interactions, including those from soluble factors, mechanical cues and the ECM. These interactions are key in establishing, maintaining and regulating cellular phenotypes and functions. It is now broadly accepted that cells cultured in 3D more closely resemble architectural and functional properties of in vivo tissues compared with cells cultured with two-dimensional (2D) techniques. One reason for this is the generation of cell–cell or cell–ECM interactions in all three dimensions, while in 2D monolayer cultures interactions are limited to the horizontal plane. Cells within a tissue are often exposed to concentration gradients of signalling effector molecules, nutrients and waste products; this is mimicked to an extent in 3D culture systems with the cells at the centre of an aggregate/organoid having less access to factors in the culture medium. Conversely, in 2D monolayers cells are exposed to a uniform concentration of factors due to direct contact with the culture medium. The establishment of more physiological, biochemical and biomechanical microenvironments using 3D techniques can affect cell proliferation, differentiation, morphogenesis, cell migration, mechanoresponses and cell survival.4 In terms of therapeutic applications, this may, in part, explain the failure of 2D cell culture systems to recapitulate drug screening outcomes as seen in vivo.5 Organoids represent a promising model system to bridge the gap between 2D cultures and in vivo mouse/human models. Organoids are more physiologically relevant than 2D culture models, while providing a reductionist model of in vivo biology in which it is possible to manipulate signalling pathways and perform genome editing.
Initiation of organoid culture requires the isolation of stem/progenitor cells, either pluripotent stem cells (PSCs) or tissue-resident stem/progenitor/differentiated cells isolated from embryonic stages or adult tissues (figure 1). The cells of origin for PSC-derived organoids are ESCs or iPSCs, which are then cultured in media supplemented with growth factors in order to mimic the signals that cells are exposed to during embryonic patterning to give rise to the specific tissue.
Liver organoids can be derived from various cells of origin by regulating signalling pathways during in vitro culture. (A) Liver organoids can be formed from tissue-resident cells isolated from biopsies of adult tissues or from embryonic stages during organogenesis. Hepatoblasts (the bipotent embryonic progenitors in vivo which give rise to ductal cells and hepatocytes) can be placed in Matrigel as ECM and generate ductal or hepatocyte organoids depending on the growth factors supplemented in the culture medium. (Bright-field images of mouse embryonic ductal and hepatocyte organoids taken from Prior et al.52) Signalling pathways which are typically modulated to enable organoid formation are listed; the pathways which are essential for different types of liver organoids are in bold. Formation of ductal or hepatocyte organoids from adult tissues requires the isolation of appropriate cells of origin. In order to generate ductal hepatic organoids from adult tissues, ductal fragments or ductal cells can be placed in Matrigel with the optimised media. Formation of adult hepatocyte organoids requires the isolation of mature hepatocytes. (B) Liver organoids can also be generated from pluripotent stem cells (iPSCs and ESCs), usually by a three-stage differentiation process that recapitulates the signalling programmes active during development. iPSCs/ESCs are first directed towards an endodermal fate by exposure to Act A and Wnt. These endoderm cells then progress to a hepatic fate following induction of HGF and FGF signalling. These hepatic progenitors are hepatoblast-like cells. The hepatic progenitors can form hepatocyte-like cells in response to OSM signalling. Conversely, by placing the hepatic progenitors in ECM and modulating FGF, EGF and Act A signalling, ductal organoids can be generated. Act A, Activin A; BMP, bone morphogenetic protein; ECM, extracellular matrix; EGF, epidermal growth factor; ESCs, embryonic stem cells; FGF, fibroblast growth factor; FSK, forskolin; HGF, hepatocyte growth factor; ICM, inner cell mass; iPSCs, induced pluripotent stem cells; OSM, Oncostatin M; TGFbi, transforming growth factor beta inhibitor; TNFa, tumour necrosis factor-alpha.
Embryonic liver organoids as a tool to understand liver development
With the aim of replicating liver development in vitro, several groups have succeeded in differentiating human iPSCs into hepatocyte-like cells following a stepwise differentiation protocol based on several chemical inhibitors.63 An alternative approach was pursued by the Suzuki and Hui labs, which forced the expression of pioneer liver transcription factors (HNF4a and FOXA1,2,3)64 or (HNF4a, GATA4 and HNF1B)65 to induce the direct differentiation of iPSCs into hepatocyte-like cells in vitro. However, these early differentiation approaches were performed in 2D and lacked the 3D information required to form hepatic tissue in vitro. The first attempt to generate 3D liver tissue in culture that recapitulated embryonic liver features was the establishment of embryonic liver bud cultures by Takebe and colleagues,66 in which human iPSC-derived hepatocytes were cultured with mesenchymal stem cells and umbilical cord cells. The resulting liver bud organoids consisted of proliferating hepatoblasts and associated cells that, on transplantation into different mouse sites, developed into hepatic tissue exhibiting mature hepatic features.66 This system has since been refined so that the hepatic endoderm, mesenchymal and endothelial progenitors are all derived from iPSCs.67 Human iPSCs can also be directed via the exposure to specific signalling pathways to form cholangiocyte organoids.68 Due to the careful modulation of culture media necessary to direct PSCs to a specific lineage, a major challenge that organoid models need to resolve is successful coculture of multiple cell types with different media requirements. Recently, hepatobiliary structures containing both hepatocyte and cholangiocyte cells have been derived from human iPSCs,69 70 although these structures do not self-renew. These reports may facilitate refinement of culture media to soon generate self-organising, self-renewing hepatobiliary organoids.
One limitation of iPSC-derived organoids for clinical use is the concern regarding genomic instability due to exposure to reprogramming factors.71 To negate this concern, organoids can be derived from embryonic or adult tissue-resident stem cells. We recently described the isolation of bipotent Lgr5+ embryonic hepatoblasts which retain the capacity to form either hepatocyte or cholangiocyte organoids depending on the culture medium used.52 Similarly, the establishment of hepatocyte organoids derived from bulk human embryonic liver tissue from aborted fetuses has been described.28 Further optimisation of culture conditions may enable the development of hepatobiliary organoids from these tissue-resident stem progenitors. Hepatoblasts serve as the functional stem cell of the liver during development; however, they are not generally thought to persist postnatally. Therefore, in order to recapitulate liver regeneration and diseased states in vitro, it may be beneficial to initiate cultures using cells isolated from primary adult liver tissue.
Liver organoids that recapitulate adult tissue and liver regeneration
Early studies by Michalopoulos and colleagues,72 in which isolated adult rat hepatocytes and other hepatic cells were placed in roller cultures, led to the formation of tissues resembling features of hepatic architecture; however, these cultures only survived for a short period of time. Self-renewing liver organoids demonstrating genetic stability during long-term culture were reported in 2013. Isolated healthy ducts (or Lgr5+ liver cells postdamage induction) self-organised into 3D structures which sustained long-term expansion as adult ductal progenitor cells while retaining the ability to differentiate into functional hepatocyte-like cells in vitro.16 The expansion of adult ductal liver organoids was enabled by use of an optimised culture medium. In addition to the epidermal growth factor and Wnt agonists, which had been shown to be important for organoid derivation from Lgr5+ intestinal stem cells,13 signalling factors important for liver development, HGF and FGF, were supplemented in the culture medium. This protocol was subsequently adapted to generate liver organoids from healthy human primary tissue27 and to generate disease models from primary liver cancer (PLC) (discussed below).73 Recently, the Nusse and Clevers labs have described the long-term culture of primary mouse hepatocytes,28 29 which retain many morphological and functional properties of hepatocytes. The system proposed by Hu et al
28 uses increased R-spondin and FGF signalling to produce hepatocyte organoids that have expression profiles similar to those of hepatocytes after partial hepatectomy. The model demonstrated by Peng et al
29 was based on observations that during liver injury, liver-resident macrophages secrete high levels of inflammatory cytokines, including TNFa, to aid regeneration. Culture with TNFa indeed served to enhance expansion of hepatocytes.29
In addition to liver organoid generation from mouse and human cells, adult ductal organoids have also been generated from rats,74 cats75 and dogs76; for an extended review on liver organoids from other species please see ref 77. Since several liver pathologies progress in a similar manner in cats and dogs as in humans, use of organoids from these species may provide insight to advance human therapies while not being subject to the same level of ethical constraints.
These advances in liver organoid technology are providing models for prenatal development, tissue maintenance and pathologies, which are otherwise intractable processes to study in human. Below we discuss the biomedical applications of liver organoids, including their use for disease modelling (of both monogenic and acquired liver diseases), drug screening, toxicology studies and regenerative medicine (figure 3).
