Stem Cell Reports Resource Analysis of Differentiation Protocols Defines a Common Pancreatic Progenitor Molecular Signature and Guides Refinement of Endocrine Differentiation Agata Wesolowska-Andersen, 1 Rikke Rejnholdt Jensen, 5 Marta Pe ´rez Alca ´ntara, 1 Nicola L. Beer, 2 Claire Duff, 2 Vibe Nylander, 2 Matthew Gosden, 4 Lorna Witty, 1 Rory Bowden, 1 Mark I. McCarthy, 1,2,3 Mattias Hansson, 5 Anna L. Gloyn, 1,2,3 and Christian Honore 5, * 1 Wellcome Centre Human Genetics, University of Oxford, OX3 7BN Oxford, UK 2 Oxford Centre for Diabetes, Endocrinology and Metabolism, University of Oxford, OX3 7LE Oxford, UK 3 NIHR Oxford Biomedical Research Centre, Churchill Hospital, OX3 7LE Oxford, UK 4 The MRC Weatherall Institute of Molecular Medicine, University of Oxford, OX3 9DS Oxford, UK 5 Stem Cell R&D, Novo Nordisk A/S, 2760 Ma ˚løv, Denmark *Correspondence: clfh@novonordisk.com https://doi.org/10.1016/j.stemcr.2019.11.010 SUMMARY Several distinct differentiation protocols for deriving pancreatic progenitors (PPs) from human pluripotent stem cells have been described, but it remains to be shown how similar the PPs are across protocols and how well they resemble their in vivo counterparts. Here, we evaluated three differentiation protocols, performed RNA and assay for transposase-accessible chromatin using sequencing on isolated PPs derived with these, and compared them with fetal human pancreas populations. This enabled us to define a shared tran- scriptional and epigenomic signature of the PPs, including several genes not previously implicated in pancreas development. Further- more, we identified a significant and previously unappreciated cross-protocol variation of the PPs through multi-omics analysis and demonstrate how such information can be applied to refine differentiation protocols for derivation of insulin-producing beta-like cells. Together, our study highlights the importance of a detailed characterization of defined cell populations derived from distinct differenti- ation protocols and provides a valuable resource for exploring human pancreatic development. INTRODUCTION Human pluripotent stem cells (hPSCs) have tremendous potential for modeling human diseases in vitro as well as for regenerative medicine in degenerative diseases. How- ever, the realization of both these applications of hPSCs is dependent on the ability to derive the relevant cell lineages from hPSCs by directed differentiation. In the context of pancreas development, studies in mice have demonstrated that exocrine, ductal, and endocrine lineages all derive from multipotent pancreatic progenitor (PP) cells, defined by co-expression of several transcription factors (TFs), including PDX1, NKX6.1, PTF1a, and SOX9 (Larsen and Grapin-Botton, 2017). Despite noteworthy dif- ferences in human pancreas development compared with mouse (Jennings et al., 2015; Nair and Hebrok, 2015), hu- man PPs express a similar core network of TFs, including PDX1 and NKX6.1 (Petersen et al., 2018). When trans- planted into immunocompromised mice, the hPSC- derived PPs are able to give rise to all lineages of the pancreas (Kelly et al., 2011; Kroon et al., 2008; Rezania et al., 2012, 2013), supporting their similarity to multipo- tent PPs observed during development. Knowledge gained from rodent models of pancreas development facilitated many of the advancements in differentiation protocols. For example, retinoic acid and fibroblast growth factor signaling are indispensable for the specification and expansion of PPs during development (Bhushan et al., 2001; Molotkov et al., 2005), and the majority of current differentiation protocols include agonists of these signaling pathways. However, there are also notable differ- ences in protocols reported to differentiate hPSCs to PPs. For example, bone morphogenetic protein (BMP) signaling has been shown to promote a liver fate choice rather than pancreas development (Wandzioch and Zaret, 2009), and thus several protocols include BMP inhibitors during differ- entiation. However, a recent report argued for the exclu- sion of BMP inhibitors, since these were shown to promote a premature endocrine differentiation at the expense of PDX1/NKX6.1-positive PPs (Russ et al., 2015). There is also no consensus on inclusion of other pathway modula- tors, such as epidermal growth factor (EGF) or protein ki- nase C (PKC) agonists, in the differentiation protocols (Nostro et al., 2015; Rezania et al., 2014; Russ et al., 2015). As hPSC-derived PPs are often defined by co-expression of a limited set of genes (e.g., PDX1 and NXK6.1), it re- mains unclear how similar the PP cells derived from various differentiation protocols are and how well they represent embryonic development and subsequent devel- opment of more mature cell types of the pancreatic islet. 138 Stem Cell Reports j Vol. 14 j 138–153 j January 14, 2020 j ª 2019 The Author(s). This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).