314 Diabetes Volume 65, February 2016

Jimmy Masjkur,1 Steven W. Poser,1 Polyxeni Nikolakopoulou,1 George Chrousos,2 Ronald D. McKay,3 Stefan R. Bornstein,1 Peter M. Jones,4 and Andreas Androutsellis-Theotokis1,5,6

Endocrine Development and Regeneration: Noncanonical Ideas From Neural Stem Cell Biology

Diabetes 2016;65:314–330 | DOI: 10.2337/db15-1099

Loss of -producing pancreatic islet b-cells is a regeneration (1–4). Such plasticity in the adult brain is hallmark of . Several experimental para- exhibited only by endogenous neural stem cell (NSC) digms demonstrate that these cells can, in principle, be populations. How NSCs manage to maintain this plas- regenerated from multiple endogenous sources using ticity is an intensely studied question that has led to the signaling pathways that are also used during pancreas identification of unusual (noncanonical) signal trans- development. A thorough understanding of these path- duction pathways that help them manage the cellular ways will provide improved opportunities for therapeutic properties that define them as stem cells, namely the intervention. It is now appreciated that signaling path- ability to self-renew and the potential to differentiate “ ” “ ” ways should not be seen as on or off but that the into mature cell types. Part of this noncanonical ma- degree of activity may result in wildly different cellular chinery is a novel branch of the Notch signaling path- outcomes. In addition to the degree of operation of a way that has pronounced consequences in vitro and in signaling pathway, noncanonical branches also play im- vivo (5). It is logical to wonder whether other tissues portant roles. Thus, a pathway, once considered as “off” or “low” may actually be highly operational but may be also use these signaling pathways. using noncanonical branches. Such branches are only Plasticity in the pancreas is demonstrated by multiple now revealing themselves as new tools to assay them examples of trans- and dedifferentiation among various are being generated. A formidable source of noncanon- cell types reported; these have been experimentally PERSPECTIVES IN DIABETES ical signal transduction concepts is neural stem cells induced, whether by pharmacological or genetic manipu- because these cells appear to have acquired unusual lation or by the implementation of damage and regener- signaling interpretations to allow them to maintain their ation models (1,6–12). An impressive amount of work unique dual properties (self-renewal and multipotency). has been done identifying the molecular mechanisms that We discuss how such findings from the neural field can control this plasticity. The patterns that emerge indicate provide a blueprint for the identification of new molec- that a variety of signaling molecules and transcription ular mechanisms regulating pancreatic biology, with a factors act during development to both suppress and in- focus on Notch, Hes/Hey, and hedgehog pathways. duce fates as well as to maintain proper function of the adult cell types, including the Notch and hedgehog path- ways and a number of transcription factors, including To the neuroscientist, the pancreas can seem like a pancreatic and duodenal 1 (Pdx1), neurogenin highlyplasticorganwhosecellsareabletoundergovast 3 (Ngn3), and the hairy and enhancer of split-1 (Hes/Hey) changes in the context of homeostatic control and family (1,13–42).

1Department of Internal Medicine III, Technische Universität Dresden, Dresden, Corresponding author: Andreas Androutsellis-Theotokis, andreas.theotokis@ Germany uniklinikum-dresden.de. 2 First Department of Pediatrics, University of Athens Medical School and Aghia Received 6 August 2015 and accepted 2 November 2015. Sophia Children’s Hospital, Athens, Greece J.M. and S.W.P. contributed equally to this work. 3Lieber Institute for Brain Development, Baltimore, MD 4Diabetes Research Group, Division of Diabetes & Nutritional Sciences, King’s © 2016 by the American Diabetes Association. Readers may use this article as College London, London, U.K. long as the work is properly cited, the use is educational and not for profit, and 5Center for Regenerative Therapies Dresden, Dresden, Germany the work is not altered. 6Department of Stem Cell Biology, Centre for Biomolecular Sciences, Division of Cancer and Stem Cells, School of Medicine, University of Nottingham, Nottingham, U.K. diabetes.diabetesjournals.org Masjkur and Associates 315

A binary view of signal transduction (where a partic- Notch signaling is critical in the maintenance of pancre- ular pathway is either on or off) seems to be incomplete atic progenitors (24,35,47,48). Specifically, loss of function as emerging evidence points toward several examples of the Notch ligand Delta-like 1 (Dll1) results in premature where low activity of a particular pathway (e.g., Notch endocrine differentiation and depletion of the pancreatic and hedgehog) leads to vastly different developmental and progenitors (24,35). cellular outcomes. But a simply quantitative distinction Conversely, overexpression of the Notch intracellular between low and high activity of a signaling pathway domain prevents differentiation and traps cells in a may also not be sufficient. A cell that is exhibiting low progenitor state (48). This appears to be due to caNotch activity of a canonical pathway may actually also exhibit signaling because similar phenotypes involving pancreatic high activity in noncanonical pathways that are only now growth arrest are observed with genetic mouse models becoming revealed. For example, whereas canonical Notch deficient in recombining binding suppressor of (caNotch) signaling is often assessed by the amount of hairless (RBPJ-k), a transcriptional mediator of caNotch Hes1 expression (a caNotch pathway target ), it is signaling (14) or Hes1 (24,35,46,49). Apart from Hes1, possible that the cell exhibits high expression of, for Notch signaling also regulates Sox9, a example, Hes3, an indirect, noncanonical target of that, like Hes1, has multiple developmental roles in many Notch (noncaNotch) with distinct outcomes in NSCs tissues (2,46,50–54). The molecular mechanisms that (43). Low activity of a signaling pathway, as assessed by control Hes1 are very important for transcriptional con- measurements of its canonical functions, does not pre- trol and the responsiveness of pancreatic progenitors to cludethatotherbranchesofthesamepathwayarenot mitogenic support from the surrounding mesenchyme. In highly active. Noncanonicalbranchesmaybeunidenti- one example, Hes1 opposes expression of p57, which nor- fiedsimplybecausewehavenotyetdesignedassaysfor mally induces cell cycle exit (55). Notch activity is neces- them. sary for the cells to respond to fibroblast growth factor NSCs are a treasure trove of noncanonical functions, (FGF) 10, a mitogen expressed by the mesenchyme at perhaps because they are required to fit multiple proper- ;E9.5–11.5 (56). In fact, inhibition of Notch cleavage ties into their signal transduction machinery (e.g., their by a g-secretase inhibitor renders cells nonresponsive to self-renewal and cell fate–decision abilities), forcing them the proliferative effects of FGF10 (57–59). to come up with unique molecular solutions. Some of Later in development, these progenitors become more these have already proven important in pancreatic func- restricted in their differentiation potential and separate tion. We will focus on lessons learned about noncaNotch into two cell types: unipotent acinar progenitor “tip cells” and hedgehog signaling in NSCs as a paradigm for their and bipotent “trunk epithelial” cells that can generate potential roles in b-cell development and function. First, endocrine and duct cells. High caNotch signaling, as we discuss specific aspects of pancreatic development on assessed by the involvement of RBPJ-k and Hes1, pro- which light can be shed by our emerging understanding of motes the trunk fate (47,48,60–64). High caNotch signal- noncanonical signaling in NSCs. We do not aim to provide ing in trunk cells activates Hes1 and Sox9 (a repressor and a comprehensive review of the developmental processes an activator of Ngn3, respectively). On the one hand, of the pancreas because this has been meticulously done Hes1 activity apparently dominates this fight, leading to elsewhere (1–4). low Ngn3 levels and, subsequently, specification to the ductal fate. Low caNotch activity, on the other hand, OVERVIEW OF PANCREAS DEVELOPMENT AND leads to induction of only Sox9, leading to high Ngn3 INVOLVEMENT OF KEY PATHWAYS levels and the acquisition of the endocrine fate (3,65). In this section we present a brief overview of specific The interpretation of high versus low Notch activity aspects in pancreas development (Fig. 1). We focus only on is an important and recurring theme throughout this particular pathways that are of direct relevance to the con- perspective. cepts discussed here. At approximately mouse embryonic In this article, we discuss data acquired from different day 8.5 (E8.5), a prepancreatic region is specified in the gut developmental stages of the endocrine pancreas and also endoderm (44). For pancreatic specification to occur and from the exocrine pancreas; these observations point toward pancreatic markers to be induced, sonic hedgehog (Shh) novel regulatory signaling networks that may be of signif- expression must be suppressed from the presumptive icant interest to pancreas biology. However, additional pancreatic endoderm (31,33,34). Subsequently, during future work will be necessary to fully elucidate at which early pancreatic development (;E9.5; primary transition developmental stages these networks may be operational phase), the forms, which contains precur- and with what precise consequences. sor cells (multipotent pancreatic precursors [MPCs]) that are able to generate differentiated cells from all three ma- NSCs POINT US TOWARD NONCANONICAL jor lineages of the adult pancreas (endocrine, duct, and SIGNALING PATHWAYS THAT ALSO CONTROL acinar). These pancreatic progenitors express Pdx1, Sox9, PANCREATIC ISLET CELLS and Hes1, a direct transcriptional target of caNotch sig- In this section, we present work from the NSC field dem- naling with multiple developmental roles (25,29,45,46). onstrating alternative interpretations of common signal 316 Neural Stem Cell Clues for Pancreas Biology Diabetes Volume 65, February 2016

Figure 1—A simplified diagram of pancreatic development. In early pancreatic development, pancreatic progenitor cells coexpress Pdx1, Sox9, and Hes1. Low caNotch activity promotes the tip cell fate, which later generates acinar cells. In contrast, high caNotch activity promotes the trunk cell fate, which coexpresses Sox9 and Hes1. In trunk cells, high caNotch activity promotes the generation of ductal cells that maintain Sox9 and Hes1 coexpression. In contrast, low caNotch activity generates the endocrine precursor fate. These cells lose Sox9 and Hes1 expression and induce the expression of Ngn3. They are able to generate all cell types of the pancreatic islet, including b-cells, which maintain low expression of Sox9 and Hes1. Adapted from Shih et al. (2).

transduction pathways, and in subsequent sections, we Special Cellular Requirements Force Special Signal address and speculate how these pathways may con- Transduction Interpretations tribute to the explanation of open questions in pancre- NSCs appear to have acquired an inventive mechanism to atic biology. fit both their self-renewal and multipotency properties diabetes.diabetesjournals.org Masjkur and Associates 317 into the existing signal transduction machinery. For tissues, stem cell and progenitor growth, self-renewal, example, they have allocated some of these different and differentiation; these functions are of direct conse- properties to different phosphorylation sites of the same quence to development, regeneration, and cancer (14,50,60). molecule, such as those on signal transducer and activator The Notch family of consists of plasma membrane– of transcription 3 (STAT3). STAT3 has two phosphor- spanning receptors that, upon activation by ligands (also ylation sites, one on the tyrosine residue (STAT3- membrane spanning) from adjacent cells, undergo a se- Tyr) at amino acid position 705 (numbering is for the ries of proteolytic cleavages that release the intracellular mouse protein) and one on the serine residue at position domain of the into the cytoplasm. This subse- 727 (STAT3-Ser) (66). On the one hand, tyrosine phos- quently translocates to the nucleus where it interacts phorylation leads to the induction of differentiation, with other such as the transcriptional regulator ending their stem cell state (67–69). Serine phosphory- RBPJ-k (14). A well-studied target gene in many tissues lation, on the other hand, is important for their growth is the transcription factor Hes1, a member of the Hes/ and, therefore, self-renewal (5). In stark contrast, many Hey gene family of basic helix-loop-helix (bHLH) tran- other cell types that are only capable of self-renewal scriptional repressors (50). The involvement of RBPJ-k and therefore do not have issues of such choices (e.g., and the induction of Hes1 are often used as indicators of astrocytes), use both phosphorylations for growth Notch activation. RBPJ-k and Hes1 involvement typically and survival (with, typically, the tyrosine site being denote the operation of caNotch signaling (70–72) (Fig. 2), much more important than the serine site). In NSCs, a although RBPJ-k–independent functions of Notch noncanonical branch of the is a have also been reported (but also Notch-independent major activator of STAT3-Ser phosphorylation (5). RBPJ-k functions). The noncaNotch signaling branch that leads to STAT3- Notch Signaling in Neural Development Ser phosphorylation also involves a number of other Notch signaling is a major regulator in brain develop- intracellular signaling components (e.g., phosphoinositide ment and the regulation of NSCs. The Notch signaling 3-kinase [PI3K], Akt, mammalian target of rapamycin pathway is involved in a variety of biological process- [mTOR], etc.) as well as the transcription factor Hes3, a es, including cell fate specification of many different member of the Hes/Hey gene family like Hes1 and Hes5

Figure 2—caNotch and noncaNotch signaling branches control NSC self-renewal and differentiation. Notch receptor activation in NSCs can lead to canonical signaling induction that involves the release of the intracellular domain of Notch receptors into the cytoplasm, its association with RBPJ-k and other proteins, and the induction of the transcription of genes, including Hes1. Hes1 promotes differentiation to the glial fate by activating the transcription of genes specific to the glial lineage. In addition, Hes1 can function in the cytoplasm where it mediates the interaction between JAK and STAT3, leading to STAT3-Tyr phosphorylation, which also promotes glial differentiation by activating the transcription of genes specific to the glial lineage. Notch receptor activation can also lead to the activation of a noncanonical signaling branch that involves the sequential activation of PI3K, Akt, mTOR, and STAT3-Ser phosphorylation. This branch leads to in- duction of Hes3, an indirect target of Notch and a member of the same gene family as Hes1. Hes3 promotes the expression of Shh, a morphogen in neural development and mitogen of NSCs. This noncanonical branch promotes the survival of NSCs, and activation in vitro and in vivo increases NSC number. JAK activity inhibits the induction of Hes3 expression (through mechanisms that are not well un- derstood), demonstrating that the canonical and noncanonical pathways may compete for dominance. 318 Neural Stem Cell Clues for Pancreas Biology Diabetes Volume 65, February 2016

(5,54,73). Therefore, Hes3 is also downstream of Notch breast cancer (81). Recent studies have implicated signaling but is not a direct transcriptional target like Hes1 Hes3 (still largely in a correlative manner) in different and Hes5; instead, Notch receptor activation leads to Hes3 reprogramming paradigms (mouse embryonic fibroblasts induction via Akt, mTOR, and STAT3-Ser phosphorylation. to induced pluripotent cells and direct adult to induced Notch ligands that lead to the induction of Hes3 expression NSCs) by demonstrating Hes3 expression regulation dur- greatly increase NSC numbers in vitro and in vivo (5,74). ing reprogramming and by correlating Hes3 transduction These two Notch signaling branches (ca/Hes1 and of adult cells with successful reprogramming to induced nonca/Hes3) have opposing functions in NSCs. Hes1 NSCs (82,83). induces differentiation to the glial fate by directly activat- These observations challenge the question of how ing prodifferentiation genes and indirectly by promoting widespread the operation of this pathway may be in the interaction of Janus kinase (JAK) with STAT3, which different cell types in the body, including in those of the further promotes gliogenic differentiation (66–68,75). This pancreas. After all, STAT3-Ser phosphorylation is down- involves the phosphorylation of STAT3 by JAK on the stream of several pathways known to promote b-cell tyrosine residue. Hes3, in contrast, is induced by several proliferation in various developmental stages of the pan- factors that lead to STAT3-Ser phosphorylation in the creas, including FGF signaling (84,85), which has been absence of STAT3-Tyr phosphorylation (5). shown to intercept Notch signaling components and to Therefore, the Hes/Hey family of transcription factors regulate cell proliferation and self-renewal in both NSC has multiple functions in the nucleus and cytoplasm, and pancreatic systems (58,86), several cell surface re- including transcriptional repression, passive repression ceptors, Akt, and mTOR. It is possible that Hes3 may be (through protein-protein interactions), and the mediation amediatorofsomeofthefunctionsofthesemolecules. of signal transduction. Elucidating these highly complex This would also raise the possibility that, being a bHLH functions for all members of this gene family is likely to factor, Hes3 may interfere with the actions of other contribute to our detailed understanding of a number of bHLH factors of importance to pancreatic development processes where these genes are involved. Because Hes/ and function such as Ngn3 (5,73,87,88). In another par- Hey proteins interact with each other, typically leading to allel between brain and pancreas, Ngn3 has important the repression of their transcriptional activity, studying functions in the development of the brain (including the expression patterns, subcellular localization, and hippocampal and hypothalamic neurons and glial pro- function of all family members together in different genitors), islet cell specification, and b-cell regeneration biological systems may be useful. (18,89–91). Hes3 itself is a functional mediator of the non- Early data from mice lacking Hes3 corroborate some of canonical action of Notch signaling because NSCs from these ideas because these mice exhibit phenotypes of adult Hes3-null mice fail to respond to Notch activa- direct relevance to pancreatic function; specifically, tion that normally promotes survival (76). They also whereas otherwise apparently normal, these mice exhibit fail to respond to insulin, another stimulator of Akt, a more pronounced loss of pancreatic b-cells, develop di- mTOR,STAT3-Ser,andHes3(Fig.3A). Hes3 also pro- abetes faster (43), and show impaired regeneration (as motes Shh expression, which is a mitogen for NSCs (5). assessed by b-cell marker expression) after streptozotocin Therefore, Hes3 is an indirect target of Notch and rep- (STZ)-induced pancreatic damage (Fig. 3B–D). Further, resents the operation of a noncanonical branch of Notch after damage they fail to induce Ngn3 expression in the that involves a number of intracellular signaling com- exocrine pancreas compartment 5 months after partial ponents, including STAT3 when it is phosphorylated STZ damage compared with wild-type controls (92). These on the serine residue. In other words, this noncanonical data invite further studies on the possible developmental branch of Notch is an input to the STAT3-Ser/Hes3 sig- roles of Hes3 and its regulators and mediators. naling pathway. Why Did We Miss It? A Generalizable, Noncanonical Molecular Pathway? The lack of an obvious phenotype in the Hes3-null mice Deleting Hes3 in mice that already lack the Hes/Hey (88) and the absence of Hes3 probes in some DNA micro- gene family members Hes1 and Hes5 leads to precocious arrays may have hindered research focus on this gene differentiation of the NSC population during brain because these have often favored other Hes/Hey gene development (77). Aspects of this pathway are not lim- family members that are commonly used to assay caNotch ited to NSCs but can be found in a number of plastic cell signaling. An additional difficulty in studying the function types, including putative cancer stem cells from glioblas- of this gene arises from its regulation at the molecular toma multiforme patients and chromaffinprogenitor level. Hes3 exists in two isoforms generated by distinct cellsfromthebovineadrenal medulla (78,79). STAT3- promoters (87). The coding region in the Hes3-null mouse Ser has been shown to drive carcinogenesis in models of has been replaced by the lacZ gene (knock-in), and this prostate cancer (80). Hes3 expression in vivo has been was placed under the control of the “a” promoter, which implicated as an indicator of the efficacy of a g-secretase produces the full-length, DNA-binding isoform. There- inhibitor (that blocks Notch signaling) in models of fore, despite the widespread use of this mouse line, it is diabetes.diabetesjournals.org Masjkur and Associates 319

Figure 3—Damage and regeneration paradigms reveal Hes3-related phenotypes. A: Hes3-null (Hes3–/–) mice have no obvious phenotype, and adult NSC (aNSC) cultures can be established from the subventricular zone lining the lateral ventricles of the brain. However, whereas wild-type (wt) aNSCs respond to activation of the STAT3-Ser/Hes3 signaling axis (induced by addition of Notch ligands or insulin to the culture medium) by vast increases in cell number, Hes3–/– mice are largely nonresponsive. These observations suggest Hes3 roles in the context of simulated regeneration. B: Similarly, whereas blood glucose levels and gross pancreas morphology are normal in Hes3–/– mice, when mice are challenged through STZ treatments (low-dose STZ, daily injections for 5 days), Hes3–/– mice suffer a greater loss of b-cells compared with wt controls. These results suggest a role of Hes3 in the protection of during toxic insults. C: Mice subjected 320 Neural Stem Cell Clues for Pancreas Biology Diabetes Volume 65, February 2016 limited because expression information is provided on only Ptf1a control or the control of other genes helping to one of the two isoforms. Being passive repressors (i.e., re- limit ectopic pancreas generation. Hes3 and Hes5, genes pressors of transcription factor activity by protein-protein with established roles in neural development and spec- interactions), the “b” isoform may be hugely important, ification (5,74,82,83,95,96), are also expressed in the even though it does not have DNA binding capability. A developing stomach and small intestine, respectively role of the b isoform in regeneration may be supported by (24,97), so they could be candidates as cocontrollers of work in the developing brain demonstrating that the b iso- pancreas organogenesis. Another interesting member of form is preferentially expressed in neural progenitors of the the Hes/Hey gene family is Hes6, a proneural gene im- developing brain (87). plicated in glioma progression (98) and prostate cancer aggression, possibly by opposing the hedgehog signaling DO “ALTERNATIVE” HES/HEY GENES OPERATE IN inhibitor (via Gli1, at least) suppressor of fused (99), PANCREATIC CELLS? withrolesinpancreaticisletcellswhereitopposes Early reports demonstrated the operation of the STAT3- Hes1 expression, thus promoting the mature b-cell phe- Ser/Hes3 signaling axis in adult pancreatic cells in vitro notype (100). and in vivo (43). It may be worth speculating how this NSC Biology Suggests the Involvement of pathway may intercept other established signaling path- Noncanonical Signaling Pathways ways involved in pancreas development and function, in- Specifically for Hes3, however, the data from NSC biology cluding the mitogenic FGF pathway, Notch (canonical via may argue against a compensatory role for Hes1 deletion. Hes1, noncanonical via Hes3), and the hedgehog pathway. This may suggest that other Hes/Hey genes or bHLH genes These testable hypotheses may provide a useful blueprint compensate for Hes1 loss. Assessing the roles of these for the integration of multiple signaling pathways in pan- signaling pathways in pancreatic development and adult creas biology. pancreatic cells will be highly valuable. Putting together Early Clues for the Involvement of Alternative Hes/Hey observations from the brain and the pancreas, one may Genes in Pancreas Biology suggest a molecular interaction decreasing its stability Early experiments with genetically engineered mice sug- and inhibiting promoter activity (21,101), leading to gested that complex interactions among members of the largely mutually exclusive Hes1 and Ngn3 expression Hes/Hey gene family profoundly affect pancreatic devel- patterns. 1 opment. In Hes1-null mice, ectopic pancreas formation is This is based on a number of observations: ) isolated observed in areas of the developing primitive stomach, adult pancreatic islets from Hes3-null mice have increased 2 , and bile duct (93). Hes1 inactivation induced Hes1 mRNA levels (43); ) whereas Hes1 opposes Ngn3 the miss-expression of pancreas-specific transcription fac- expression (21), Hes3 seems to be required for the in- tor 1a (Ptf1a), a bHLH that is important in the formation creased Ngn3 expression in the exocrine pancreas after 3 of the exocrine pancreas and the spatial organization of the STZ-induced damage (92); and ) the pancreatic pheno- endocrine pancreas in these regions (94). Using Hes1-null types in the Hes1- and Hes3-null mice are very different mice that were crossed with a Cre/loxP reporter mouse, (Hes1-null: hindered development [24]; Hes3-null: no allowing the tracing of Ptf1a cells, it was demonstrated obvious phenotype in undamaged animals [88]). that ectopic Ptf1a-expressing cells had acquired multipo- Hes3, therefore, may allow for regeneration-induced tent pancreatic progenitor properties in the Hes1-null mechanisms to promote the induction of Ngn3 expression D mouse that differentiated into cells of the pancreatic line- (Fig. 3 ). How Hes6 may intercept this network is not yet age, including insulin-producing cells with characteristic well understood; however, as we discuss in a later section, nuclear Pdx1 expression. A model where Hes1 suppresses Hes6 may activate components of the Shh pathway while Ptf1a was thus suggested as a means of restricting pan- suppressing Hes1 expression. creas formation in the appropriate location. However, caNotch Levels Differentially Regulate the Balance Hes1 expression is much broader than the pattern of Between Sox9 and Hes1 ectopicpancreasformationin the Hes1-null mice, sug- In trunk cells, caNotch signaling is an important regulator gesting that additional mechanisms could contribute to of Sox9 and Hes1 (102). The levels of activation of Sox9

to the STZ regimen described can be placed under normal conditions for months to assess their regeneration potential. After 5 months, wt mice regenerate most cells of the pancreatic islet, although their ability to regulate blood glucose levels may still be compromised. In contrast, Hes3–/– mice are less efficient at regenerating the number of b-cells, suggesting a role of Hes3 in the regeneration of the pan- creatic islet after damage. D: Examples of islets after the experimental procedure described in C. Experimental details for panels B-D: low-dose STZ treatment was performed as previously described (43), but the mice were allowed to survive for 5 months after the last injection (n = 6 mice per each of the 4 groups). Immunolabeling images were acquired after 4% paraformaldehyde fixation and sectioning; counting was from 30 islet sections per group. The percentage of Pdx1+ cells per islet (i.e., per number of DAPI+ nuclei in the islet) is as follows: wt: 79.21 6 10.9; wt-STZ: 35.8 6 12.3; Hes32/2: 76.2 6 13.7; Hes32/2-STZ: 23.7 6 20.9. A t-test comparison between the wt-STZ and Hes32/2-STZ values showed a significant difference (P < 0.05). diabetes.diabetesjournals.org Masjkur and Associates 321 and Hes1 determine whether Ngn3 will be subsequently Pdx1 (107,108). (In a seeming paradox, Sox9 also induces induced (leading to endocrine commitment) or sup- Hes1 expression [46], perhaps as a means of limiting its pressed (leading to ductal commitment). These levels action on Ngn3.) appear to be regulated by highly complex interactions In support of a role of caNotch signaling opposing among Sox9, Hes1, and Ngn3 (Sox9 induces Hes1 and Ngn3 expression, mice deficient in the Notch receptor Ngn3 expression; Hes1 opposes Ngn3 expression) as well family ligand Dll1 and RBPJ-k exhibit reduced Hes1 ex- as the levels of caNotch signaling (high caNotch leads pression (35). In concert with these observations, Hes1 to the induction of both Sox9 and Hes1; low caNotch inhibits the expression of Ngn3 (15), Hes1-null mice ex- signaling leads to the preferential induction of Sox9) hibit increased Ngn3 activity (24), and Hes1 inhibition (Fig. 4). induces the redifferentiation of expanded human pancre- More specifically, to demonstrate the dose-dependent atic b-cell–derived cells (109). Identifying new regulators effects of Notch activity, whole embryonic (E12.5) of Hes1 and Sox9 may improve our understanding of pancreas explants were cultured under different concen- pancreas development. trations of a g-secretase inhibitor to oppose Notch re- ceptor activation, and it was demonstrated that Sox9 Possible Involvement of noncaNotch Branches transcription requires lower levels of Notch activity In contrast to the caNotch signaling pathway, the than Hes1 transcription (51). These data are supported noncaNotch branch involves the indirect (likely via Akt, by reports that the expression of Ngn3 and Hes1 is mu- mTOR, and STAT3-Ser phosphorylation, as shown in tually exclusive in the developing pancreas (27,35), as is NSCs [5]) induction of Hes3, which then suppresses Hes1 the expression of Sox9 and Ngn3 at the endocrine pro- transcription (the mechanism for this is not yet known), genitor stage (46,103–105). Furthermore, Sox9 occupancy potentially releasing Ngn3 from Hes1-induced inhibition. regions on the Ngn3 gene promoter have been identified The contribution of such a putative mechanism in devel- (104,106). In addition, Sox9 may also regulate Ngn3 via opment is not yet demonstrated because no detailed

Figure 4—caNotch and noncaNotch branches regulating the expression of key pancreatic genes. In this diagram, we fuse data from different pancreatic cellular systems (pancreatic progenitors, mature b-cells) to create a generalized molecular model of how caNotch and noncaNotch branches may interact; we suggest that this model can be used as a blueprint to address key open questions in the regulation of pancreas development. caNotch signaling (involving RBPJ-k) can lead to two established cellular outcomes: high caNotch activity in trunk cells leads to the induction of Sox9 and Hes1. Sox9 is an activator of Ngn3, and Hes1 is a repressor of Ngn3. Overall, this leads to loss of Ngn3 expression and the acquisition of the ductal cell fate. In contrast, low caNotch activity leads to the induction of only Sox9 (and not Hes1), leading to Ngn3 expression and the acquisition of the endocrine precursor fate. Hes3 may intercept this regulatory system given its roles in adult b-cell lines, in the damaged and regenerating pancreas, in Ngn3 induction in vivo, and in opposing Hes1 expression in pancreatic islets. Hes3, possibly under the control of upstream regulators as described in Fig. 2, may contribute to the suppression of Hes1 under low caNotch conditions, ensuring that only Sox9 will be induced (and not Hes1), leading to Ngn3 induction and subsequent endocrine fate specification. Hes6 may also play a role in inhibiting Hes1 in this system. 322 Neural Stem Cell Clues for Pancreas Biology Diabetes Volume 65, February 2016 studies have been performed, although in the adult (46), high versus low FGF10 signaling may also prefer- Hes3-null mouse, no gross morphological phenotypes entially stimulate the Hes1 (59) versus the Hes3 Notch are obvious (43,88). Perhaps, this mechanism is partic- signaling branch. ularly active during regeneration because Hes3-null mice During early pancreatic development, FGF10, derived exhibit increased damage and reduced regeneration of from mesenchymal cells, leads to Sox9 induction via its b-cells after STZ damage (43). A technical difficulty in receptor, Fgfr2b. Sox9 promotes the expression of separating caNotch from noncaNotch effects is that, at Fgfr2b, helping to maintain the receptivity of MPCs to least in NSC cultures, both pathway branches are sensi- FGF10 (53). Pdx1 also promotes Fgfr2b expression, and tive to g-secretase inhibition (5). this may be mediated by Sox9, because Sox9 and Fgfr2b Regardless, the possibility that noncaNotch is also expression are lost simultaneously in Pdx1-deficient involved in pancreas development and function prompts pancreata. Therefore, Sox9 is regulated by both the the question of whether low Notch is really simply low Notch and FGF10 pathways. How these apparently dis- Notch or actually a different equilibrium between caNotch parate pathways may regulate the same gene is not (Hes1-mediated) and noncaNotch (e.g., Hes3-mediated) well understood. activity. But is there evidence, even circumstantial, in line The STAT3-Ser/Hes3 signaling axis may possibly with this possibility? contribute because it provides a molecular mechanism that connects the Notch signaling pathway with classic Integrating FGF10 Signaling mitogenic intracellular signals (e.g., Akt, mTOR), leading The MPC mitogen FGF10 (56) could provide an entry to the induction of Hes3 or still unidentified mediators. point to Hes3; in fact, it may provide a Notch receptor– Whether this molecular mechanism is operational in independent means of activating both Hes1 (a caNotch MPCs is not yet reported (Fig. 5). target) and Hes3 (a noncaNotch target). We speculate There are clues suggesting complex molecular mecha- that, in a manner similar to high versus low Notch signaling nisms by which FGF10 may be connected to caNotch and

Figure 5—FGF10 may intercept both caNotch and noncaNotch. FGF10 can induce classic mitogenic signal transduction pathways, including PI3K, Akt, and mTOR, but it can also activate JAK-STAT signaling. We speculate that the choice may be influenced by the levels of receptor activation. The PI3K branch may lead, via STAT3-Ser, to a number of events that lead to the induction of Ngn3: Hes3 induction, which may then suppress Hes1, and Sox9 induction. In contrast, the JAK-STAT branch may lead to Hes1 induction, which opposes Ngn3 transcription. Sox9 also induces Hes1, and this, we speculate, may be a molecular mechanism to limit the degree of Ngn3 activation via Sox9. Sox9, possibly via Pdx1, positively regulates the expression of Fgfr2b and maintains receptiveness to FGF10. diabetes.diabetesjournals.org Masjkur and Associates 323 noncaNotch signaling in MPCs: genetic perturbations of Obvious developmental deficits in Hes3-null mice are FGF10 or its receptor Fgfr2b show that FGF10 signaling not known, although detailed studies are still lacking. This leads to hyperplasia, the maintenance of the immature does not mean that Hes3 does not have developmental cell state (high Ptf1a,lowPdx1,lowNkx6.1state), roles, because other factors (possibly Hes6, for example) expression of Hes1, low Dll1, and low Ngn3 (58,59). At may functionally compensate for the loss of Hes3 in these first sight, these data may argue against the involvement genetic mouse models. However, the current data point of Hes3 downstream of FGF10 signaling because Hes3 toward a clearer role for Hes3 in differentiated, insulin- activation may be expected to result in low Hes1 and producing cells: In MIN6 cells, Hes3 overexpression high Pdx1, at least based on data from mouse insulinoma inducesPdx1expression,andHes3RNAinterference cell lines and the adult pancreata from Hes3-null mice (43). opposes growth, insulin expression, and insulin sensitiv- However, FGF10 signaling can also activate the JAK-STAT ity. In line with the operation of the STAT3-Ser/Hes3 pathway, which is a strong inhibitor of Hes3 as well as an signaling axis (which in NSCs is opposed by JAK activity), inducer of Hes1 in other immature cell types (110). Sim- these cells can be efficiently cultured in conditions that ilarly, in NSC cultures, the cytokine ciliary neurotrophic suppress JAK-STAT signaling, thereby supporting STAT3- factor can be converted from a prodifferentiation signal Ser phosphorylation and Hes3 expression (43). These re- suppressing Hes3 transcription to a mitogen promoting sults are mirrored by observations reported in adult mice Hes3 transcription simply by cotreatment with a JAK in- where antibody- and PCR-based techniques demonstrate hibitor (78). These observations raise the intriguing possi- Hes3 expression in islets; immunofluorescence experi- bility that FGF10 may, perhaps under different levels ments also suggest the expression of Hes3 in adult human of activation or cellular and developmental contexts, in- islets. Further, Hes3–knock-in reporter mice confirm Hes3 tercept caNotch or noncaNotch signaling with complex promoter activity in adult mouse islets and demonstrate downstream results. promoter activation after STZ-induced damage. The func- tional significance of pancreatic Hes3 is suggested by ob- Neural/Pancreas Progenitor Cell Equivalence? servations in the adult Hes3-null mice showing increased Identifying the cell types in the developing pancreas that sensitivity to STZ damage and impaired regeneration, al- may use aspects of the STAT3-Ser/Hes3 signaling axis will though no stark phenotypes are obvious in the uninjured require substantial experimentation. Issues to resolve adult Hes3-null mice. include identifying which cell type expresses Hes3 (or a Therefore, the STAT3-Ser/Hes3 signaling axis may functional equivalent) and in which cell types STAT3 is not be a molecular mechanism that determines a par- phosphorylated on the serine residue and not on the ticular progenitor cell type but a signaling module that tyrosine residue. As noted earlier, an integrated elucida- is used by a variety of plastic cell types in the pancreas, tion of the function of all members of the Hes/Hey gene both during development and in the adult. Modeling family may be required. the operation of this signaling pathway in culture may Few detailed studies on Hes3 expression during devel- provide new opportunities in basic science and drug opment exist (24,73,88,93). Detailed analyses of adult discovery (discussed later). tissues have demonstrated Hes3 expression in putative NSCs; such observations during development may be dif- WHO CONTROLS Shh? fi fi cult to make by low-magni cation inspection (e.g., Shh plays important roles in pancreatic development whole-embryo immunostaining or in situ hybridization) + and function, but the mechanisms that regulate its because Hes3 cells may be few and Hes3 expression can expression are not fully understood. Similar to what we fi be low. Therefore, future studies may shed rst light on discussed for Notch signaling and possibly also for FGF Hes3 expression at different developmental stages of the signaling, Shh signaling shows dosage-dependent effects, pancreas. intercepts both caNotch and noncaNotch signaling pathway Is it possible then that this system is operational in branches (e.g., Hes5 and Hes3) (5,112), and NSC biology bipotent trunk cells? One may speculate that in the may contribute to our understanding of Shh regulation in decision between the endocrine and ductal commitment, the pancreas. Hes3 could help bias a progenitor to take the endocrine route by opposing Hes1 expression. There is no evidence, Shh Signaling in Neural and Pancreatic Development to our knowledge, that Hes3 regulates Sox9 itself. In neural development, different levels of Shh result in It may be interesting to study whether tip cells different cellular outcomes, with low Shh levels more express Hes3, because they exhibit low caNotch signal- prone to promoting self-renewal/proliferation and high ing and because adult Hes3-null mice exhibit pheno- Shh more prone to morphogenetic/cell specification types in acinar cells (the inability to induce Ngn3 after decisions (112). The dosage-dependent Shh effects may partial STZ damage [43]). Tip cells express Sox9 (111), be linked to the dosage-dependent Notch effects. In but there is currently no evidence that this may be me- fact, there is direct evidence that Shh intercepts both diated by Hes3; Sox9 induction may be a consequence of caNotch (e.g., Hes5) and noncaNotch signaling (e.g., low caNotch signaling. Hes3) (5,112). 324 Neural Stem Cell Clues for Pancreas Biology Diabetes Volume 65, February 2016

Early pancreatic development and pancreatic pro- induction in several cellular systems, providing a Notch- genitor maintenance requires suppression of Shh and independent mechanism of inducing Hes1 (126). There- activation of caNotch signaling (involving RBPJ-k and fore, hedgehog signaling in this in vivo system may be Hes1). At ;E9.5, the Shh receptor Ptch1 is still not perturbed such that it preferentially leads to Hes1 induc- expressed in the pancreatic (31). However, tion at the expense of other target genes, suppressing it appears ;1 day later, albeit at low amounts; after Pdx1 expression. Overall, therefore, there are strong in- birth, Ptch1 expression increases and is widespread in dications that hedgehog signaling may both promote and the islets and ducts (113). Therefore, although hedgehog suppress Pdx1 expression. We propose a model, fusing signaling must be suppressed for pancreatic specifica- data from both the NSC and pancreas fields to consolidate tion, it is possible that it is required for optimal devel- these observations. opment starting soon after specification, even if at As mentioned above, Shh mediates the nuclear locali- relatively low levels. Indeed, genetic deletion of the zation of Pdx1 in cultured insulinoma cells (120) (Fig. 6A). Shh mediator Smo in the pancreas epithelium induces Hes3 overexpression also induces Pdx1 expression and delayed expansion of the early pancreatic epithelium and nuclear localization in insulinoma cells, and chromatin delayed b-cell mass development (113). Several other immunoprecipitation on-chip data suggest the direct studies demonstrate a role of hedgehog signaling in the regulation of the Pdx1 promoter as a putative mecha- growth of the pancreas (31,33,34,114–118). These seem- nism (Fig. 6B) (43,127). In NSCs, Hes3 overexpression ingly conflicting studies may be consolidated by taking promotes Shh expression (Fig. 6C) (5). A hedgehog sig- into account the different mechanisms of hedgehog sig- naling branch may oppose Pdx1 expression through in- naling modulation, such as ligand overexpression or Smo creased Hes1 expression in cells that overexpress Gli2 deletion. It also suggests that hedgehog signaling opposes and lack primary cilia (122) (Fig. 6D). Put together, these the establishment of pancreas organ boundaries in the observations suggest the possible (and testable) involve- while subsequently promoting the expansion of ment of noncaNotch signaling components in the regu- early pancreatic epithelium starting at ;E12.5 (113). lation of Shh and Pdx1 (Fig. 7). Adding to the context-dependent complexity of hedgehog These varied observations raise a number of questions: signaling, the actions of this pathway may oppose prolif- eration of pancreatic epithelial cells at around midgesta-  Could Hes3 be both an inducer of Shh signaling activity tion. We will address the possibility that noncanonical as well as a compensator for the lack of Shh during signaling pathways may mediate some of the pleiotropic early pancreatic development and perhaps also during effects of Shh. regeneration? Effects of Shh on Pancreatic Function  If Gli2 does not induce Pdx1 expression, is there an- In adult b-cells, conclusions on the roles of Shh are other Gli member that does (128)?  complex, partly because of the different biological sys- Are there Gli factors that preferentially lead to Hes1 tems (cell lines in vitro, in vitro vs. in vivo) used and the and others to Pdx1 expression?  different experimental approaches used to modulate How important is the inhibitory effect of primary cilia Shh (pharmacological or genetic manipulation of differ- in Pdx1 regulation?  ent components of the Shh pathway). However, despite Given the positive role of Sox9 in the formation of differences, these studies show important roles of Shh primary cilia (51) and the negative role of primary cilia on adult b-cells. b-Cell lines, the Shh signaling small- in the mediation of hedgehog signaling (122), how sig- nificant might this Sox9/hedgehog signaling cross talk molecule inhibitor cyclopamine, and ectopic expression be in terms of development and function? of Shh were used to show that hedgehog signaling pro-  How much of the inhibitory effect of the hedgehog motes insulin secretion and content and insulin pro- branch on Pdx1 is mediated by Hes1? moter activity (119,120). In a similar experimental system, Shh was also demonstrated to provide protective properties when cells were stressed by the addition of CAN WE MODEL DIFFERENT SIGNALING STATES proinflammatory cytokines (121). IN CULTURE? In vivo, a somewhat different story arises on the role Answering these questions will require a more thorough of Shh on adult b-cells. A study that used a different understanding of the many states that pancreatic cells can approach to activate hedgehog signaling (using an active adopt during development and disease. It would be a great version of Gli2, a mediator of hedgehog signaling in advantage to researchers if these different cellular/ b-cells devoid of primary cilia that normally negatively signaling states could be modeled in vitro, contributing regulate hedgehog signaling [122]) showed that increased to our search for alternative growth pathways that may Shh signaling correlated with increased expression of the eventually lead to new therapies. precursor markers Hes1 and Sox9, both direct targets that Again, there are valuable lessons from NSC biology are normally excluded from b-cells (123–126). It should that can be applied to b-cell research. MIN6 cells are a be noted that Gli1 and Gli2 overexpression leads to Hes1 commonly used transformed mouse insulinoma cell line diabetes.diabetesjournals.org Masjkur and Associates 325

(129). Despite the caveats with transformed cell lines, high Hes1 state that does not model the in vivo state of they have proven a valuable tool to study evoked insulin adult b-cells. release. Still, these cells exhibit low nuclear Pdx1 expres- We recently showed that indeed MIN6 cells can be sion, relative to b-cells in vivo, suggesting that they do cultured in defined (serum-free conditions) even in the not model the in vivo state of b-cells perfectly. They also presence of a JAK inhibitor, which suppresses STAT3- exhibit Hes1 expression, which is normally absent from Tyr phosphorylation (43). Hes3 is expressed under these b-cells. They are typically cultured in serum-containing conditions, and a polyclonal antibody against Hes3 dem- medium, a common strategy to provide ample yet un- onstrates nuclear localization. Nuclear Pdx1 expression defined factors required for cell maintenance and pro- incidence also increases, providing potential access to liferation. NSC biology reveals that serum, a potent JAK/ its gene targets and representing more accurately b-cells STAT pathway activator, suppresses Hes3 expression in vivo. Pdx1 expression is regulated by Hes3, because and, even more, nuclear Hes3 localization (5,43). This overexpression of Hes3 in these cells invariably leads to might explain why common insulinoma cell line culture high expression of nuclear Pdx1 (43). The Hes3-Pdx1 re- conditions maintain the cells in a low Hes3, low Pdx1, lation may be direct because Hes3 has been demonstrated

Figure 6—Shh and mediators of caNotch and noncaNotch signaling intercept to regulate cell outcome. A: In insulinoma cell lines, Shh overexpression promotes Pdx1 expression and nuclear localization. B: In insulinoma cell lines, Hes3 overexpression promotes Pdx1 expression and nuclear localization; the mechanism may involve the activation of promoter regions of the Pdx1 gene. C: In primary fetal rodent NSC cultures, Hes3 overexpression promotes Shh expression; the mechanism is not yet elucidated. D: In genetically engineered mice, pancreatic cells that have increased Gli2 expression and lack primary (1ary) cilia (which normally oppose aspects of hedgehog signaling) exhibit increased Hes1 expression and decreased Pdx1 expression. 326 Neural Stem Cell Clues for Pancreas Biology Diabetes Volume 65, February 2016

Figure 7—Notch and Shh signaling pathways intercept to regulate the expression of key pancreatic transcription factors. In this diagram, we fuse together observations from different cell types to create a generalized model of how Shh and Notch signaling may intercept in previously unexplored ways. Hes3 directly regulates Pdx1 expression in MIN6 cells. Shh also directly does this, suggesting that in these cell systems Shh and Hes3 share common functions. In NSCs, Hes3 overexpression leads to Shh expression induction, suggesting that in addition to the parallel functions of Shh and Hes3, Hes3 may also be a regulator of Shh. However, this possibility is yet unexplored in the pancreatic system, to our knowledge. It is not unlikely, however, because another member of the Hes/Hey family of genes, Hes6, does regulate Shh signaling in prostate cancer cells, opposes Hes1 action in culture, and promotes Pdx1 expression. In insulinoma cell line cultures, Shh pathway activation increases insulin secretion and content and promotes cell survival after inflammatory cytokine–induced stress. Hes3 has similar roles in MIN6 cells. A particular branch of the Shh signaling pathway, however, suggests more complex roles of hedgehog signaling. In a genetic mouse model where Gli2, a transcriptional mediator of hedgehog signaling, is overexpressed and where cells lack primary (1ary) cilia (which normally contribute to the suppression of hedgehog signaling), increased expression of the precursor markers Hes1 and Sox9 was observed. This apparent discrepancy may be explained by the fact that Gli2 promotes Hes1 expression, demonstrating how Shh signaling intercepts with caNotch signaling. Hes1, in turn, suppresses Pdx1 expression; for example, Hes1-null mice exhibit ectopic Pdx1 expression and pancreas formation. Hes6 opposes Hes1 expression. Taken together, Notch, hedgehog, and Hes/Hey signaling intercept in canonical and noncanonical ways to regulate the development and function of the pancreas. This complex signaling network offers many opportunities for manipulation and study and argues for an extensive elucidation of additional aspects of this network, including a better understanding of other Hes/Hey genes and Gli factors in this process.

to bind to Pdx1 gene promoter regions in MIN6 cells the nucleus and Hes3 RNA interference does not signif- (127). Hes3 RNA interference opposes cell growth icantly affect cell growth. In contrast, Hes3 manipulation and insulin release (43). Switching the same cells be- has powerful roles in serum-free, defined culture condi- tween culture conditions induces reversible changes tions, where Hes3 is allowed in the nucleus. These results showcasing the bidirectional ability of the cells to pro- give a glimpse to strategies that may allow locking cells in mote their growth via distinct signaling pathways. different states to allow their study. Whether Shh manip- fi Such approaches may assist in experimental studies ulation affects cells in de ned conditions is not yet known. fi by locking b-cells in culture in particular signaling It will be important to de ne cell states using distinct culture conditions and reevaluate the effects of the states representing specific aspects of their develop- perturbation of canonical and noncanonical signaling ment and function. pathway components. It is possible that Shh and Hes3 signaling have similar outcomes to cells at different developmental stages or CONCLUSIONS signaling states. Despite the similar effects of hedgehog The protection and regeneration of insulin-producing and Hes3 signaling manipulation, the effects of hedgehog pancreatic islet b-cells of the endocrine pancreas is a ma- signaling were reported using cell culture conditions that jor focus of diabetes research (3,4). Compared with other place cells in a signaling state that is nonresponsive to adult tissues, the endocrine pancreas exhibits consider- Hes3 manipulation. Specifically, the effects of hedgehog able plasticity in that many stimuli, including age, nutri- signaling modulation were seen in commonly used serum- tion, pregnancy, insulin sensitivity, excessive caloric containing culture systems where Hes3 is excluded from intake, and various paradigms of damage, can affect diabetes.diabetesjournals.org Masjkur and Associates 327 b-cell proliferation and alter b-cell mass, at least in ex- 4. Jurczyk A, Bortell R, Alonso LC. Human b-cell regeneration: progress, perimental rodent models. Understanding the signal hurdles, and controversy. Curr Opin Endocrinol Diabetes Obes 2014;21:102– transduction pathways behind these phenomena will pro- 108 vide potentially new therapeutic avenues. Clues are pro- 5. Androutsellis-Theotokis A, Leker RR, Soldner F, et al. Notch signalling regulates stem cell numbers in vitro and in vivo. Nature 2006;442:823–826 vided by reports of experimental interventions that 6. Cohen DE, Melton D. Turning straw into gold: directing cell fate for re- b manipulate -cell mass using treatments with known generative medicine. Nat Rev Genet 2011;12:243–252 downstream signaling pathways. These include hypergly- 7. Graf T. Historical origins of and reprogramming. Cell cemia, incretins, Nodal (a TGF-b family member), vascular Stem Cell 2011;9:504–516 endothelial growth factor, Wnt pathway activators, and 8. Cherry AB, Daley GQ. Reprogramming cellular identity for regenerative g-aminobutyric acid, for example (4). Additional clues are medicine. Cell 2012;148:1110–1122 provided by reports pointing out the importance of tran- 9. Horb ME, Shen CN, Tosh D, Slack JM. Experimental conversion of liver to scription factors in the specification of the b-cell fate, pancreas. Curr Biol 2003;13:105–115 including Pdx1, MafA, and Ngn3 (130). 10. Afelik S, Chen Y, Pieler T. Combined ectopic expression of Pdx1 and Ptf1a/ It is important to elucidate in detail how these p48 results in the stable conversion of posterior endoderm into endocrine and – signaling pathways regulate their downstream transcrip- exocrine pancreatic tissue. Genes Dev 2006;20:1441 1446 tion factors to provide a deeper understanding of pancre- 11. Jarikji ZH, Vanamala S, Beck CW, Wright CV, Leach SD, Horb ME. Dif- ferential ability of Ptf1a and Ptf1a-VP16 to convert stomach, duodenum and liver atic development and function and to give more precise to pancreas. Dev Biol 2007;304:786–799 direction to drug discovery programs. Recent evidence 12. Zhou Q, Brown J, Kanarek A, Rajagopal J, Melton DA. In vivo re- suggests that the view that a signaling pathway is on/off programming of adult pancreatic exocrine cells to beta-cells. Nature 2008;455: or high/low may be too simplistic because it does not 627–632 account for the operation of alternative (noncanonical) 13. Arda HE, Benitez CM, Kim SK. Gene regulatory networks governing pan- branches of these pathways that may be highly active creas development. Dev Cell 2013;25:5–13 while canonical branches may be suppressed. In an 14. Artavanis-Tsakonas S, Rand MD, Lake RJ. Notch signaling: cell fate example of cross-disciplinary approach, the NSC field may control and signal integration in development. Science 1999;284:770–776 provide a template to evaluate the involvement of such 15. Mansouri A. Development and regeneration in the endocrine pancreas. noncanonical signaling pathway branches in pancreatic ISRN Endocrinol 2012;2012:640956 development and function, providing new ideas that may 16. Gradwohl G, Dierich A, LeMeur M, Guillemot F. neurogenin3 is required for the development of the four endocrine cell lineages of the pancreas. Proc Natl help understand and manipulate better the plasticity of Acad Sci U S A 2000;97:1607–1611 this organ. 17. Schwitzgebel VM, Scheel DW, Conners JR, et al. Expression of neuro- genin3 reveals an islet cell precursor population in the pancreas. Development 2000;127:3533–3542 Funding. This work was partly supported by the Helmholtz Alliance 18. Rukstalis JM, Habener JF. Neurogenin3: a master regulator of pancreatic Imaging and Curing Environmental Metabolic Diseases (ICEMED), through the islet differentiation and regeneration. Islets 2009;1:177–184 Helmholtz Association Initiative and Networking Fund grant 051_40001, Deutsche 19. Kopp JL, Dubois CL, Hao E, Thorel F, Herrera PL, Sander M. Progenitor cell Forschungsgemeinschaft grant SFB 655 “Cells Into Tissues” Project A24, and domains in the developing and adult pancreas. Cell Cycle 2011;10:1921–1927 Deutsche Forschungsgemeinschaft Clinical Research Unit grant KFO 252. 20. Baeyens L, Lemper M, Leuckx G, et al. Transient cytokine treatment in- Duality of Interest. No potential conflicts of interest relevant to this article duces acinar cell reprogramming and regenerates functional mass in were reported. diabetic mice. Nat Biotechnol 2014;32:76–83 Author Contributions. J.M. and P.N. collected, assembled, analyzed, 21. Lee JC, Smith SB, Watada H, et al. Regulation of the pancreatic pro- and interpreted the data and revised the manuscript. S.W.P. collected, endocrine gene neurogenin3. Diabetes 2001;50:928–936 assembled, analyzed, and interpreted the data; revised the manuscript; and 22. Lynn FC, Skewes-Cox P, Kosaka Y, McManus MT, Harfe BD, German MS. provided a critical revision for important intellectual content. G.C., R.D.M., and MicroRNA expression is required for pancreatic islet cell genesis in the mouse. S.R.B. analyzed and interpreted the data, revised the manuscript, and provided a Diabetes 2007;56:2938–2945 critical revision for important intellectual content. P.M.J. analyzed and interpreted 23. Bar Y, Russ HA, Knoller S, Ouziel-Yahalom L, Efrat S. HES-1 is involved in the data, wrote the manuscript, revised the manuscript, and provided a critical adaptation of adult human beta-cells to proliferation in vitro. Diabetes 2008;57: revision for important intellectual content. A.A.-T. conceived the topic, drafted the 2413–2420 article, wrote the manuscript, and interpreted and analyzed data. All authors 24. Jensen J, Pedersen EE, Galante P, et al. Control of endodermal endocrine approved the final manuscript. A.A.-T. is the guarantor of this work and, as such, development by Hes-1. Nat Genet 2000;24:36–44 had full access to all the data in the study and takes responsibility for the integrity 25. Kageyama R, Ohtsuka T, Tomita K. The bHLH gene Hes1 regulates dif- of the data and the accuracy of the data analysis. ferentiation of multiple cell types. Mol Cells 2000;10:1–7 References 26. Lammert E, Brown J, Melton DA. Notch during pancreatic 1. Puri S, Folias AE, Hebrok M. Plasticity and dedifferentiation within the organogenesis. Mech Dev 2000;94:199–203 pancreas: development, homeostasis, and disease. Cell Stem Cell 2015;16: 27. Jensen J, Heller RS, Funder-Nielsen T, et al. Independent development of 18–31 pancreatic alpha- and beta-cells from neurogenin3-expressing precursors: a role 2. Shih HP, Wang A, Sander M. Pancreas organogenesis: from lineage de- for the notch pathway in repression of premature differentiation. Diabetes 2000; termination to morphogenesis. Annu Rev Cell Dev Biol 2013;29:81–105 49:163–176 3. Carolan PJ, Melton DA. New findings in pancreatic and intestinal endocrine 28. Zhang YQ, Mashima H, Kojima I. Changes in the expression of transcription development to advance regenerative medicine. Curr Opin Endocrinol Diabetes factors in pancreatic AR42J cells during differentiation into insulin-producing Obes 2013;20:1–7 cells. Diabetes 2001;50(Suppl. 1):S10–S14 328 Neural Stem Cell Clues for Pancreas Biology Diabetes Volume 65, February 2016

29. Ahlgren U, Jonsson J, Edlund H. The morphogenesis of the pancreatic 52. Seymour PA, Sander M. Historical perspective: beginnings of the beta-cell: mesenchyme is uncoupled from that of the pancreatic epithelium in IPF1/PDX1- current perspectives in beta-cell development. Diabetes 2011;60:364–376 deficient mice. Development 1996;122:1409–1416 53. Seymour PA, Shih HP, Patel NA, et al. A Sox9/Fgf feed-forward loop 30. Stoffers DA, Zinkin NT, Stanojevic V, Clarke WL, Habener JF. Pancreatic maintains pancreatic organ identity. Development 2012;139:3363–3372 agenesis attributable to a single nucleotide deletion in the human IPF1 gene 54. Imayoshi I, Kageyama R. bHLH factors in self-renewal, multipotency, and coding sequence. Nat Genet 1997;15:106–110 fate choice of neural progenitor cells. Neuron 2014;82:9–23 31. Apelqvist A, Ahlgren U, Edlund H. Sonic hedgehog directs specialised 55. Georgia S, Soliz R, Li M, Zhang P, Bhushan A. p57 and Hes1 coordinate mesoderm differentiation in the intestine and pancreas. Curr Biol 1997;7: cell cycle exit with self-renewal of pancreatic progenitors. Dev Biol 2006;298: 801–804 22–31 32. Stolow MA, Shi YB. Xenopus sonic hedgehog as a potential morphogen 56. Bhushan A, Itoh N, Kato S, et al. Fgf10 is essential for maintaining the during embryogenesis and thyroid hormone-dependent metamorphosis. Nucleic proliferative capacity of epithelial progenitor cells during early pancreatic or- Acids Res 1995;23:2555–2562 ganogenesis. Development 2001;128:5109–5117 33. Hebrok M, Kim SK, Melton DA. Notochord repression of endodermal sonic 57. Miralles F, Lamotte L, Couton D, Joshi RL. Interplay between FGF10 and hedgehog permits pancreas development. Genes Dev 1998;12:1705–1713 Notch signalling is required for the self-renewal of pancreatic progenitors. Int 34. Kim SK, Melton DA. Pancreas development is promoted by cyclopamine, a J Dev Biol 2006;50:17–26 hedgehog signaling inhibitor. Proc Natl Acad Sci U S A 1998;95:13036–13041 58. Hart A, Papadopoulou S, Edlund H. Fgf10 maintains notch activation, 35. Apelqvist A, Li H, Sommer L, et al. Notch signalling controls pancreatic cell stimulates proliferation, and blocks differentiation of pancreatic epithelial cells. – differentiation. Nature 1999;400:877 881 Dev Dyn 2003;228:185–193 36. Gu G, Dubauskaite J, Melton DA. Direct evidence for the pancreatic 59. Norgaard GA, Jensen JN, Jensen J. FGF10 signaling maintains the pan- lineage: NGN3+ cells are islet progenitors and are distinct from duct progenitors. creatic progenitor cell state revealing a novel role of Notch in organ development. – Development 2002;129:2447 2457 Dev Biol 2003;264:323–338 37. Jonsson J, Carlsson L, Edlund T, Edlund H. Insulin-promoter-factor 1 is 60. Afelik S, Jensen J. Notch signaling in the pancreas: patterning and cell fate required for pancreas development in mice. Nature 1994;371:606–609 specification. Wiley Interdiscip Rev Dev Biol 2013;2:531–544 38. Offield MF, Jetton TL, Labosky PA, et al. PDX-1 is required for pancreatic 61. Esni F, Ghosh B, Biankin AV, et al. Notch inhibits Ptf1 function and acinar outgrowth and differentiation of the rostral duodenum. Development 1996;122: cell differentiation in developing mouse and zebrafish pancreas. Development 983–995 2004;131:4213–4224 39. Stoffers DA, Ferrer J, Clarke WL, Habener JF. Early-onset type-II diabetes 62. Magenheim J, Klein AM, Stanger BZ, et al. Ngn3(+) endocrine progenitor mellitus (MODY4) linked to IPF1. Nat Genet 1997;17:138–139 cells control the fate and morphogenesis of pancreatic ductal epithelium. Dev Biol 40. Ahlgren U, Jonsson J, Jonsson L, Simu K, Edlund H. b-Cell-specific in- 2011;359:26–36 activation of the mouse Ipf1/Pdx1 gene results in loss of the b-cell phenotype 63. Horn S, Kobberup S, Jørgensen MC, et al. Mind bomb 1 is required and maturity onset diabetes. Genes Dev 1998;12:1763–1768 for pancreatic b-cell formation. Proc Natl Acad Sci U S A 2012;109:7356– 41. Gannon M, Ables ET, Crawford L, et al. Pdx-1 function is specifically re- 7361 quired in embryonic beta cells to generate appropriate numbers of endocrine cell 64. Schaffer AE, Freude KK, Nelson SB, Sander M. Nkx6 transcription factors types and maintain glucose homeostasis. Dev Biol 2008;314:406–417 and Ptf1a function as antagonistic lineage determinants in multipotent pancreatic 42. Gao T, McKenna B, Li C, et al. Pdx1 maintains b cell identity and function progenitors. Dev Cell 2010;18:1022–1029 by repressing an a cell program. Cell Metab 2014;19:259–271 65. Pan FC, Wright C. Pancreas organogenesis: from bud to plexus to gland. 43. Masjkur J, Arps-Forker C, Poser SW, et al. Hes3 is expressed in the adult – pancreatic islet and regulates gene expression, cell growth, and insulin release. Dev Dyn 2011;240:530 565 66. Levy DE, Darnell JE Jr. Stats: transcriptional control and biological impact. J Biol Chem 2014;289:35503–35516 – 44. Slack JM. Developmental biology of the pancreas. Development 1995;121: Nat Rev Mol Cell Biol 2002;3:651 662 1569–1580 67. Bonni A, Sun Y, Nadal-Vicens M, et al. Regulation of gliogenesis in the 45. Imayoshi I, Isomura A, Harima Y, et al. Oscillatory control of factors de- central nervous system by the JAK-STAT signaling pathway. Science 1997;278: – termining multipotency and fate in mouse neural progenitors. Science 2013;342: 477 483 1203–1208 68. Rajan P, McKay RD. Multiple routes to astrocytic differentiation in the CNS. – 46. Seymour PA, Freude KK, Tran MN, et al. SOX9 is required for maintenance J Neurosci 1998;18:3620 3629 of the pancreatic progenitor cell pool. Proc Natl Acad Sci U S A 2007;104:1865– 69. Johe KK, Hazel TG, Muller T, Dugich-Djordjevic MM, McKay RD. Single 1870 factors direct the differentiation of stem cells from the fetal and adult central 47. Afelik S, Qu X, Hasrouni E, et al. Notch-mediated patterning and cell fate nervous system. Genes Dev 1996;10:3129–3140 allocation of pancreatic progenitor cells. Development 2012;139:1744–1753 70. Andersen P, Uosaki H, Shenje LT, Kwon C. Non-canonical Notch signaling: 48. Murtaugh LC, Stanger BZ, Kwan KM, Melton DA. Notch signaling controls emerging role and mechanism. Trends Cell Biol 2012;22:257–265 multiple steps of pancreatic differentiation. Proc Natl Acad Sci U S A 2003;100: 71. Tanigaki K, Honjo T. Two opposing roles of RBP-J in Notch signaling. Curr 14920–14925 Top Dev Biol 2010;92:231–252 49. Fujikura J, Hosoda K, Iwakura H, et al. Notch/Rbp-j signaling prevents 72. Souilhol C, Cormier S, Tanigaki K, Babinet C, Cohen-Tannoudji M. RBP- premature endocrine and ductal cell differentiation in the pancreas. Cell Metab Jkappa-dependent notch signaling is dispensable for mouse early embryonic 2006;3:59–65 development. Mol Cell Biol 2006;26:4769–4774 50. Kageyama R, Ohtsuka T, Kobayashi T. The Hes gene family: repressors 73. Lobe CG. Expression of the helix-loop-helix factor, Hes3, during embryo and oscillators that orchestrate embryogenesis. Development 2007;134: development suggests a role in early midbrain-hindbrain patterning. Mech Dev 1243–1251 1997;62:227–237 51. Shih HP, Kopp JL, Sandhu M, et al. A Notch-dependent molecular circuitry 74. Androutsellis-Theotokis A, Rueger MA, Park DM, et al. Targeting neural initiates pancreatic endocrine and ductal cell differentiation. Development 2012; precursors in the adult brain rescues injured dopamine neurons. Proc Natl Acad 139:2488–2499 Sci U S A 2009;106:13570–13575 diabetes.diabetesjournals.org Masjkur and Associates 329

75. Kamakura S, Oishi K, Yoshimatsu T, Nakafuku M, Masuyama N, Gotoh Y. 94. Krapp A, Knöfler M, Ledermann B, et al. The bHLH protein PTF1-p48 is Hes binding to STAT3 mediates crosstalk between Notch and JAK-STAT sig- essential for the formation of the exocrine and the correct spatial organization of nalling. Nat Cell Biol 2004;6:547–554 the endocrine pancreas. Genes Dev 1998;12:3752–3763 76. Androutsellis-Theotokis A, Rueger MA, Mkhikian H, Korb E, McKay RD. 95. Imayoshi I, Kageyama R. The role of Notch signaling in adult neurogenesis. Signaling pathways controlling neural stem cells slow progressive brain disease. Mol Neurobiol 2011;44:7–12 Cold Spring Harb Symp Quant Biol 2008;73:403–410 96. Basak O, Taylor V. Identification of self-replicating multipotent progenitors 77. Hatakeyama J, Bessho Y, Katoh K, et al. Hes genes regulate size, shape in the embryonic nervous system by high Notch activity and Hes5 expression. Eur and histogenesis of the nervous system by control of the timing of neural stem J Neurosci 2007;25:1006–1022 cell differentiation. Development 2004;131:5539–5550 97. Schröder N, Gossler A. Expression of Notch pathway components in fetal 78. Park DM, Jung J, Masjkur J, et al. Hes3 regulates cell number in cultures and adult mouse small intestine. Gene Expr Patterns 2002;2:247–250 from glioblastoma multiforme with stem cell characteristics. Sci Rep 2013;3: 98. Haapa-Paananen S, Kiviluoto S, Waltari M, et al. HES6 gene is selectively 1095 overexpressed in glioma and represents an important transcriptional regulator of 79. Masjkur J, Levenfus I, Lange S, et al. A defined, controlled culture system glioma proliferation. Oncogene 2012;31:1299–1310 for primary bovine chromaffin progenitors reveals novel biomarkers and modu- 99. Carvalho FL, Marchionni L, Gupta A, et al. HES6 promotes prostate cancer lators. Stem Cells Transl Med 2014;3:801–808 aggressiveness independently of Notch signalling. J Cell Mol Med 2015;19: – 80. Qin HR, Kim HJ, Kim JY, et al. Activation of signal transducer and activator 1624 1636 of transcription 3 through a phosphomimetic serine 727 promotes prostate 100. Ball AJ, Abrahamsson AE, Tyrberg B, Itkin-Ansari P, Levine F. HES6 re- tumorigenesis independent of tyrosine 705 phosphorylation. Cancer Res 2008; verses nuclear reprogramming of insulin-producing cells following cell fusion. – 68:7736–7741 Biochem Biophys Res Commun 2007;355:331 337 81. Zhang CC, Pavlicek A, Zhang Q, et al. Biomarker and pharmacologic 101. Qu X, Afelik S, Jensen JN, et al. Notch-mediated post-translational control of Ngn3 protein stability regulates pancreatic patterning and cell fate commit- evaluation of the g-secretase inhibitor PF-03084014 in breast cancer models. ment. Dev Biol 2013;376:1–12 Clin Cancer Res 2012;18:5008–5019 102. Seymour PA. Sox9: a master regulator of the pancreatic program. Rev 82. Cassady JP, D’Alessio AC, Sarkar S, et al. Direct lineage conversion of Diabet Stud 2014;11:51–83 adult mouse liver cells and B lymphocytes to neural stem cells. Stem Cell Rep 103. Kopp JL, von Figura G, Mayes E, et al. Identification of Sox9-dependent 2014;3:948–956 acinar-to-ductal reprogramming as the principal mechanism for initiation of 83. Salewski RP, Buttigieg J, Mitchell RA, van der Kooy D, Nagy A, Fehlings pancreatic ductal adenocarcinoma. Cancer Cell 2012;22:737–750 MG. The generation of definitive neural stem cells from PiggyBac transposon- 104. Lynn FC, Smith SB, Wilson ME, Yang KY, Nekrep N, German MS. Sox9 induced pluripotent stem cells can be enhanced by induction of the NOTCH coordinates a transcriptional network in pancreatic progenitor cells. Proc Natl signaling pathway. Stem Cells Dev 2013;22:383–396 Acad Sci U S A 2007;104:10500–10505 84. Miralles F, Czernichow P, Ozaki K, Itoh N, Scharfmann R. Signaling 105. Jennings RE, Berry AA, Kirkwood-Wilson R, et al. Development of the through fibroblast growth factor receptor 2b plays a key role in the de- human pancreas from foregut to endocrine commitment. Diabetes 2013;62: velopment of the exocrine pancreas. Proc Natl Acad Sci U S A 1999;96: 3514–3522 6267–6272 106. Seymour PA, Freude KK, Dubois CL, Shih HP, Patel NA, Sander M. 85. Hart AW, Baeza N, Apelqvist A, Edlund H. Attenuation of FGF signalling in A dosage-dependent requirement for Sox9 in pancreatic endocrine cell formation. mouse beta-cells leads to diabetes. Nature 2000;408:864–868 Dev Biol 2008;323:19–30 86. Faux CH, Turnley AM, Epa R, Cappai R, Bartlett PF. Interactions between 107. Dubois CL, Shih HP, Seymour PA, et al. Sox9-haploinsufficiency causes fibroblast growth factors and Notch regulate neuronal differentiation. J Neurosci glucose intolerance in mice. PLoS One 2011;6:e23131 – 2001;21:5587 5596 108. Oliver-Krasinski JM, Kasner MT, Yang J, et al. The diabetes gene Pdx1 87. Hirata H, Ohtsuka T, Bessho Y, Kageyama R. Generation of structurally and regulates the transcriptional network of pancreatic endocrine progenitor cells in functionally distinct factors from the basic helix-loop-helix gene Hes3 by alter- mice. J Clin Invest 2009;119:1888–1898 fi – native rst exons. J Biol Chem 2000;275:19083 19089 109. Bar Y, Russ HA, Sintov E, Anker-Kitai L, Knoller S, Efrat S. Redifferentiation 88. Hirata H, Tomita K, Bessho Y, Kageyama R. Hes1 and Hes3 regulate of expanded human pancreatic b-cell-derived cells by inhibition of the NOTCH maintenance of the isthmic organizer and development of the mid/hindbrain. pathway. J Biol Chem 2012;287:17269–17280 – EMBO J 2001;20:4454 4466 110. Neradugomma NK, Subramaniam D, Tawfik OW, et al. Prolactin signaling 89. Liu Y, Wu Y, Lee JC, et al. Oligodendrocyte and astrocyte development in enhances colon cancer stemness by modulating Notch signaling in a Jak2- rodents: an in situ and immunohistological analysis during embryonic develop- STAT3/ERK manner. Carcinogenesis 2014;35:795–806 ment. Glia 2002;40:25–43 111. Lioubinski O, Müller M, Wegner M, Sander M. Expression of Sox tran- 90. Simon-Areces J, Membrive G, Garcia-Fernandez C, Garcia-Segura LM, scription factors in the developing mouse pancreas. Dev Dyn 2003;227:402–408 Arevalo MA. Neurogenin 3 cellular and subcellular localization in the developing 112. Shikata Y, Okada T, Hashimoto M, et al. Ptch1-mediated dosage- and adult hippocampus. J Comp Neurol 2010;518:1814–1824 dependent action of Shh signaling regulates neural progenitor development at 91. Pelling M, Anthwal N, McNay D, et al. Differential requirements for neu- late gestational stages. Dev Biol 2011;349:147–159 rogenin 3 in the development of POMC and NPY neurons in the hypothalamus. 113. Lau J, Hebrok M. Hedgehog signaling in pancreas epithelium regulates Dev Biol 2011;349:406–416 embryonic organ formation and adult beta-cell function. Diabetes 2010;59:1211– 92. Poser SW, Chenoweth JG, Colantuoni C, et al. Concise review: re- 1221 programming, behind the scenes: noncanonical neural stem cell signaling 114. Nakayama S, Arakawa M, Uchida T, et al. Dose-dependent requirement of pathways reveal new, unseen regulators of tissue plasticity with therapeutic patched homologue 1 in mouse pancreatic beta cell mass. Diabetologia 2008;51: implications. Stem Cells Transl Med 2015;4:1251–1257 1883–1892 93. Fukuda A, Kawaguchi Y, Furuyama K, et al. Ectopic pancreas formation in 115. Kawahira H, Scheel DW, Smith SB, German MS, Hebrok M. Hedgehog Hes1-knockout mice reveals plasticity of endodermal progenitors of the gut, bile signaling regulates expansion of pancreatic epithelial cells. Dev Biol 2005;280: duct, and pancreas. J Clin Invest 2006;116:1484–1493 111–121 330 Neural Stem Cell Clues for Pancreas Biology Diabetes Volume 65, February 2016

116. Kawahira H, Ma NH, Tzanakakis ES, McMahon AP, Chuang PT, Hebrok M. 124. Bien-Willner GA, Stankiewicz P, Lupski JR. SOX9cre1, a cis-acting regu- Combined activities of hedgehog signaling inhibitors regulate pancreas devel- latory element located 1.1 Mb upstream of SOX9, mediates its enhancement opment. Development 2003;130:4871–4879 through the SHH pathway. Hum Mol Genet 2007;16:1143–1156 117. Bitgood MJ, McMahon AP. Hedgehog and Bmp genes are coexpressed at 125. Wall DS, Mears AJ, McNeill B, et al. Progenitor cell proliferation in the many diverse sites of cell-cell interaction in the mouse embryo. Dev Biol 1995; retina is dependent on Notch-independent Sonic hedgehog/Hes1 activity. J Cell – 172:126 138 Biol 2009;184:101–112 118. Champeris Tsaniras S, Jones PM. Generating pancreatic beta-cells from 126. Wall DS, Wallace VA. Hedgehog to Hes1: the heist of a Notch target. Cell embryonic stem cells by manipulating signaling pathways. J Endocrinol 2010; Cycle 2009;8:1301–1302 – 206:13 26 127. Sachdeva MM, Claiborn KC, Khoo C, et al. Pdx1 (MODY4) regulates 119. Thomas MK, Rastalsky N, Lee JH, Habener JF. Hedgehog signaling regu- pancreatic beta cell susceptibility to ER stress. Proc Natl Acad Sci U S A 2009; lation of insulin production by pancreatic beta-cells. Diabetes 2000;49:2039–2047 106:19090–19095 120. Thomas MK, Lee JH, Rastalsky N, Habener JF. Hedgehog signaling reg- 128. ZeRuth GT, Takeda Y, Jetten AM. The Krüppel-like protein Gli-similar 3 ulation of homeodomain protein islet duodenum homeobox-1 expression in (Glis3) functions as a key regulator of insulin transcription. Mol Endocrinol 2013; pancreatic beta-cells. Endocrinology 2001;142:1033–1040 – 121. Umeda H, Ozaki N, Mizutani N, et al. Protective effect of hedgehog sig- 27:1692 1705 naling on cytokine-induced cytotoxicity in pancreatic beta-cells. Exp Clin Endo- 129. Miyazaki J, Araki K, Yamato E, et al. Establishment of a pancreatic crinol Diabetes 2010;118:692–698 beta cell line that retains glucose-inducible insulin secretion: special ref- 122. Cervantes S, Lau J, Cano DA, Borromeo-Austin C, Hebrok M. Primary cilia erence to expression of glucose transporter isoforms. Endocrinology 1990; regulate Gli/Hedgehog activation in pancreas. Proc Natl Acad Sci U S A 2010;107: 127:126–132 10109–10114 130. Bernardo AS, Hay CW, Docherty K. Pancreatic transcription factors and 123. Landsman L, Parent A, Hebrok M. Elevated Hedgehog/Gli signaling causes their role in the birth, life and survival of the pancreatic beta cell. Mol Cell En- beta-cell dedifferentiation in mice. Proc Natl Acad Sci U S A 2011;108:17010–17015 docrinol 2008;294:1–9