Permalink
Cannot retrieve contributors at this time
Fetching contributors…
Cannot retrieve contributors at this time
| <?xml version="1.0" encoding="UTF-8"?><!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.1d1 20130915//EN" "JATS-archivearticle1.dtd"><article article-type="research-article" dtd-version="1.1d1" xmlns:xlink="http://www.w3.org/1999/xlink"><front><journal-meta><journal-id journal-id-type="nlm-ta">elife</journal-id><journal-id journal-id-type="hwp">eLife</journal-id><journal-id journal-id-type="publisher-id">eLife</journal-id><journal-title-group><journal-title>eLife</journal-title></journal-title-group><issn publication-format="electronic">2050-084X</issn><publisher><publisher-name>eLife Sciences Publications, Ltd</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="publisher-id">00940</article-id><article-id pub-id-type="doi">10.7554/eLife.00940</article-id><article-categories><subj-group subj-group-type="display-channel"><subject>Research article</subject></subj-group><subj-group subj-group-type="heading"><subject>Developmental biology and stem cells</subject></subj-group><subj-group subj-group-type="heading"><subject>Human biology and medicine</subject></subj-group></article-categories><title-group><article-title>Expansion and conversion of human pancreatic ductal cells into insulin-secreting endocrine cells</article-title></title-group><contrib-group><contrib contrib-type="author" id="author-5374"><name><surname>Lee</surname><given-names>Jonghyeob</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="other" rid="par-1"/><xref ref-type="other" rid="par-3"/><xref ref-type="fn" rid="con1"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" equal-contrib="yes" id="author-5375"><name><surname>Sugiyama</surname><given-names>Takuya</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="fn" rid="equal-contrib">†</xref><xref ref-type="fn" rid="con2"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" equal-contrib="yes" id="author-5376"><name><surname>Liu</surname><given-names>Yinghua</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="fn" rid="equal-contrib">†</xref><xref ref-type="fn" rid="con3"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5377"><name><surname>Wang</surname><given-names>Jing</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="other" rid="par-1"/><xref ref-type="fn" rid="con4"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5378"><name><surname>Gu</surname><given-names>Xueying</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="fn" rid="con5"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5379"><name><surname>Lei</surname><given-names>Ji</given-names></name><xref ref-type="aff" rid="aff2"/><xref ref-type="fn" rid="con6"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5380"><name><surname>Markmann</surname><given-names>James F</given-names></name><xref ref-type="aff" rid="aff2"/><xref ref-type="fn" rid="con7"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5381"><name><surname>Miyazaki</surname><given-names>Satsuki</given-names></name><xref ref-type="aff" rid="aff3"/><xref ref-type="fn" rid="con10"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5382"><name><surname>Miyazaki</surname><given-names>Jun-ichi</given-names></name><xref ref-type="aff" rid="aff3"/><xref ref-type="fn" rid="con11"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5383"><name><surname>Szot</surname><given-names>Gregory L</given-names></name><xref ref-type="aff" rid="aff4"/><xref ref-type="fn" rid="con8"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" id="author-5384"><name><surname>Bottino</surname><given-names>Rita</given-names></name><xref ref-type="aff" rid="aff5"/><xref ref-type="fn" rid="con9"/><xref ref-type="fn" rid="conf1"/></contrib><contrib contrib-type="author" corresp="yes" id="author-5334"><name><surname>Kim</surname><given-names>Seung K</given-names></name><xref ref-type="aff" rid="aff1"/><xref ref-type="aff" rid="aff6"/><xref ref-type="corresp" rid="cor1">*</xref><xref ref-type="other" rid="par-1"/><xref ref-type="other" rid="par-2"/><xref ref-type="fn" rid="con12"/><xref ref-type="fn" rid="conf1"/></contrib><aff id="aff1"><institution content-type="dept">Department of Developmental Biology</institution>, <institution>Stanford University School of Medicine</institution>, <addr-line><named-content content-type="city">Stanford</named-content></addr-line>, <country>United States</country></aff><aff id="aff2"><institution content-type="dept">Department of Surgery</institution>, <institution>Massachusetts General Hospital, Harvard Medical School</institution>, <addr-line><named-content content-type="city">Boston</named-content></addr-line>, <country>United States</country></aff><aff id="aff3"><institution content-type="dept">Division of Stem Cell Regulation Research</institution>, <institution>Osaka University Graduate School of Medicine</institution>, <addr-line><named-content content-type="city">Osaka</named-content></addr-line>, <country>Japan</country></aff><aff id="aff4"><institution content-type="dept">UCSF Transplantation Surgery</institution>, <institution>University of California, San Francisco</institution>, <addr-line><named-content content-type="city">San Francisco</named-content></addr-line>, <country>United States</country></aff><aff id="aff5"><institution content-type="dept">Department of Pediatrics, Division of Immunogenetics</institution>, <institution>Children’s Hospital of Pittsburgh, University of Pittsburgh School of Medicine</institution>, <addr-line><named-content content-type="city">Pittsburgh</named-content></addr-line>, <country>United States</country></aff><aff id="aff6"><institution content-type="dept">Department of Medicine, Oncology Division</institution>, <institution>Howard Hughes Medical Institute, Stanford University School of Medicine</institution>, <addr-line><named-content content-type="city">Stanford</named-content></addr-line>, <country>United States</country></aff></contrib-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Rossant</surname><given-names>Janet</given-names></name><role>Reviewing editor</role><aff><institution>University of Toronto</institution>, <country>Canada</country></aff></contrib></contrib-group><author-notes><corresp id="cor1"><label>*</label>For correspondence: <email>seungkim@stanford.edu</email></corresp><fn fn-type="con" id="equal-contrib"><label>†</label><p>These authors contributed equally to this work</p></fn></author-notes><pub-date date-type="pub" publication-format="electronic"><day>19</day><month>11</month><year>2013</year></pub-date><pub-date pub-type="collection"><year>2013</year></pub-date><volume>2</volume><elocation-id>e00940</elocation-id><history><date date-type="received"><day>14</day><month>05</month><year>2013</year></date><date date-type="accepted"><day>08</day><month>10</month><year>2013</year></date></history><permissions><copyright-statement>© 2013, Lee et al</copyright-statement><copyright-year>2013</copyright-year><copyright-holder>Lee et al</copyright-holder><license xlink:href="http://creativecommons.org/licenses/by/3.0/"><license-p>This article is distributed under the terms of the <ext-link ext-link-type="uri" xlink:href="http://creativecommons.org/licenses/by/3.0/">Creative Commons Attribution License</ext-link>, which permits unrestricted use and redistribution provided that the original author and source are credited.</license-p></license></permissions><self-uri content-type="pdf" xlink:href="elife00940.pdf"/><abstract><object-id pub-id-type="doi">10.7554/eLife.00940.001</object-id><p>Pancreatic islet β-cell insufficiency underlies pathogenesis of diabetes mellitus; thus, functional β-cell replacement from renewable sources is the focus of intensive worldwide effort. However, in vitro production of progeny that secrete insulin in response to physiological cues from primary human cells has proven elusive. Here we describe fractionation, expansion and conversion of primary adult human pancreatic ductal cells into progeny resembling native β-cells. FACS-sorted adult human ductal cells clonally expanded as spheres in culture, while retaining ductal characteristics. Expression of the cardinal islet developmental regulators Neurog3, MafA, Pdx1 and Pax6 converted exocrine duct cells into endocrine progeny with hallmark β-cell properties, including the ability to synthesize, process and store insulin, and secrete it in response to glucose or other depolarizing stimuli. These studies provide evidence that genetic reprogramming of expandable human pancreatic cells with defined factors may serve as a general strategy for islet replacement in diabetes.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.001">http://dx.doi.org/10.7554/eLife.00940.001</ext-link></p></abstract><abstract abstract-type="executive-summary"><object-id pub-id-type="doi">10.7554/eLife.00940.002</object-id><title>eLife digest</title><p>Diabetes mellitus is a disease that can lead to dangerously high blood sugar levels, causing numerous complications such as heart disease, glaucoma, skin disorders, kidney disease, and nerve damage. In healthy individuals, beta cells in the pancreas produce a hormone called insulin, which stimulates cells in the liver, muscles and fat to take up glucose from the blood. However, this process is disrupted in people with diabetes, who either have too few pancreatic beta cells (type 1 diabetes) or do not respond appropriately to insulin (type 2 diabetes).</p><p>All patients with type 1 diabetes, and some with type 2, must inject themselves regularly with insulin, but this does not always fully control the disease. Some type 1 patients have been successfully treated with beta cells transplanted from deceased donors, but there are not enough donor organs available for this to become routine. Thus, intensive efforts worldwide are focused on generating insulin-producing cells in the lab from human stem cells. However, the cells produced in this way can give rise to tumors.</p><p>Now, Lee et al. have shown that duct cells, which make up about 30% of the human pancreas, can be converted into cells capable of producing and secreting insulin. Ductal cells obtained from donor pancreases were first separated from the remaining tissue and grown in cell culture. Viruses were then used to introduce genes that reprogrammed the ductal cells so that they acquired the ability to make, process and store insulin, and to release it in response to glucose—hallmark features of functional beta cells.</p><p>As well as providing a potential source of cells for use in transplant or cell conversion therapies for diabetes, the ability to grow and maintain human pancreatic ductal cells in culture may make it easier to study other diseases that affect the pancreas, including pancreatitis, cystic fibrosis, and adenocarcinoma.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.002">http://dx.doi.org/10.7554/eLife.00940.002</ext-link></p></abstract><kwd-group kwd-group-type="author-keywords"><title>Author keywords</title><kwd>diabetes</kwd><kwd>islet</kwd><kwd>pancreas</kwd><kwd>beta-cells</kwd><kwd>insulin</kwd><kwd>conversion</kwd></kwd-group><kwd-group kwd-group-type="research-organism"><title>Research organism</title><kwd>Human</kwd></kwd-group><funding-group><award-group id="par-1"><funding-source><institution-wrap><institution>JDRF</institution></institution-wrap></funding-source><award-id>43-2010-347</award-id><principal-award-recipient><name><surname>Lee</surname><given-names>Jonghyeob</given-names></name><name><surname>Wang</surname><given-names>Jing</given-names></name><name><surname>Kim</surname><given-names>Seung K</given-names></name></principal-award-recipient></award-group><award-group id="par-2"><funding-source><institution-wrap><institution>Howard Hughes Medical Institute</institution></institution-wrap></funding-source><principal-award-recipient><name><surname>Kim</surname><given-names>Seung K</given-names></name></principal-award-recipient></award-group><award-group id="par-3"><funding-source><institution-wrap><institution>Larry L Hillblom Foundation</institution></institution-wrap></funding-source><award-id>2008-D-018-FEL</award-id><principal-award-recipient><name><surname>Lee</surname><given-names>Jonghyeob</given-names></name></principal-award-recipient></award-group><funding-statement>The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.</funding-statement></funding-group><custom-meta-group><custom-meta><meta-name>elife-xml-version</meta-name><meta-value>2</meta-value></custom-meta><custom-meta specific-use="meta-only"><meta-name>Author impact statement</meta-name><meta-value>Cells from the human pancreatic duct can be grown in culture and triggered to become insulin-producing cells, which could potentially be transplanted into patients with diabetes.</meta-value></custom-meta></custom-meta-group></article-meta></front><body><sec id="s1" sec-type="intro"><title>Introduction</title><p>The pancreas is a vital organ with exocrine and endocrine cell functions, and a root of lethal human diseases including diabetes mellitus, pancreatitis, and pancreatic ductal adenocarcinoma. Exocrine acinar cells produce digestive zymogens that are delivered to the intestines by a branching network of exocrine ductal cells that secrete bicarbonate and other products. Pancreatic endocrine functions derive from clusters of epithelial cells (islets of Langerhans) called α-, β-, δ-, and PP-cells that respectively synthesize, store, and secrete the hormones Glucagon, Insulin, Somatostatin, and Pancreatic polypeptide (<xref ref-type="bibr" rid="bib4">Benitez et al., 2012</xref>). Insulin production by islet β-cells is highly regulated: key features of mature β-cells include preproinsulin (<italic>INS</italic>) transcription, proinsulin processing by endo- and exo-peptidases and storage of the proinsulin cleavage products insulin and C-peptide in dense core vesicles. Likewise, cardinal β-cell functions regulate insulin release in response to glucose and other secretagogues, including glucose sensing and metabolism through the enzyme glucokinase, and use of ATP-dependent potassium channels (K<sub>ATP</sub>) and voltage-gated calcium channels to induce insulin exocytosis (reviewed in <xref ref-type="bibr" rid="bib50">Suckale and Solimena, 2010</xref>). Deficiency or malfunctioning of β-cells produces impaired glucose regulation and diabetes mellitus, a disease with autoimmune (type 1, T1DM) and pandemic forms (type 2; <xref ref-type="bibr" rid="bib3">Ashcroft and Rorsman, 2012</xref>). Thus, replacement or regeneration of functional human β-cells is an intensely-sought goal.</p><p>Human islet transplantation can be used to replace β-cell function in T1DM (reviewed in <xref ref-type="bibr" rid="bib57">Vardanyan et al., 2010</xref>), but a shortage of donors currently precludes broad use of human pancreatic islets for β-cell replacement. Because of their expandability and multipotency, human embryonic stem cells (hESCs) and induced pluripotent stem cells (iPSCs) have been explored as sources of replacement insulin-producing cells (reviewed in <xref ref-type="bibr" rid="bib19">Hebrok, 2012</xref>). However, directing the differentiation of these developmentally ‘primitive’ cells through multiple sequential fates into β-cell-like progeny that synthesize, process, store, and secrete insulin while lacking tumorigenic potential has challenged investigators worldwide (<xref ref-type="bibr" rid="bib13">Fujikawa et al., 2005</xref>; <xref ref-type="bibr" rid="bib32">McKnight et al., 2010</xref>; <xref ref-type="bibr" rid="bib10">Cheng et al., 2012</xref>). Moreover, different hESC and iPSC cell lines exhibit significant variability during development into insulin-producing cells (<xref ref-type="bibr" rid="bib38">Nostro and Keller, 2012</xref>). Recent work demonstrated that differentiated cell types in adult organs, including the mouse pancreas, can be experimentally ‘reprogrammed’ into progeny resembling islet cells, suggesting a new strategy for β-cell replacement (<xref ref-type="bibr" rid="bib58">Vierbuchen and Wernig, 2011</xref>). For example, adult mouse pancreatic acinar cells can be converted into insulin-producing cells in vitro and in vivo (<xref ref-type="bibr" rid="bib34">Minami et al., 2005</xref>; <xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>). However, little progress has been made in reprogramming primary human epithelial cells into different cell types, including conversion of pancreatic non-β-cells toward a human β-cell fate (<xref ref-type="bibr" rid="bib58">Vierbuchen and Wernig, 2011</xref>). Thus, systems permitting expansion and genetic modulation of human pancreatic cells could powerfully influence studies of β-cell biology and replacement.</p><p>Pancreatic ducts constitute 30–40% of human pancreas and have been proposed as a potential source of replacement β-cells (<xref ref-type="bibr" rid="bib8">Bouwens and Pipeleers, 1998</xref>; <xref ref-type="bibr" rid="bib7">Bonner-Weir et al., 2004</xref>). During pancreas development, fetal endocrine cells derive from primitive ductal epithelium (reviewed by <xref ref-type="bibr" rid="bib39">Pan and Wright, 2011</xref>; <xref ref-type="bibr" rid="bib40">Pictet and Rutter, 1972</xref>). In addition, some studies have suggested that in adult mice, β-cells may be produced from pancreatic ductal epithelium (<xref ref-type="bibr" rid="bib24">Inada et al., 2008</xref>; <xref ref-type="bibr" rid="bib63">Xu et al., 2008</xref>; <xref ref-type="bibr" rid="bib42">Rovira et al., 2010</xref>). However, recent lineage tracing evidences have not supported this view (<xref ref-type="bibr" rid="bib48">Solar et al., 2009</xref>; <xref ref-type="bibr" rid="bib14">Furuyama et al., 2011</xref>; <xref ref-type="bibr" rid="bib27">Kopp et al., 2011</xref>). In humans, prior studies have suggested that adult human primary ductal cells in heterogeneous cell mixtures may harbor the potential to generate endocrine-like progeny (<xref ref-type="bibr" rid="bib6">Bonner-Weir et al., 2000</xref>; <xref ref-type="bibr" rid="bib21">Heremans et al., 2002</xref>; <xref ref-type="bibr" rid="bib53">Swales et al., 2012</xref>), but interpretation in these studies was limited by the probability of islet cell contamination. Therefore, the potential for conversion of pancreatic ductal cells toward an endocrine fate remains unclear. Moreover, prior studies have revealed only limited proliferative capacity of primary human pancreatic ductal cells in culture (<xref ref-type="bibr" rid="bib41">Rescan et al., 2005</xref>). Thus, despite their relative abundance, multiple practical issues have prevented development of human pancreatic ductal cells as a source of replacement β-cells.</p><p>Here we report that normal human adult pancreatic duct cells can be sorted, clonally expanded, and genetically converted into endocrine cells. Human insulin-producing cells (IPCs) produced from sorted duct cells exhibited hallmark features of functional neonatal β-cells including high-level preproinsulin (<italic>INS</italic>) expression, proinsulin processing and dense-core granule formation. Moreover, secretion of insulin and insulin C-peptide from IPCs is stimulated by glucose and K<sub>ATP</sub> channel stimulants in a calcium-dependent manner. Together these studies reveal a new system for investigating human pancreatic duct cell biology, genetics, and β-cell regeneration.</p></sec><sec id="s2" sec-type="results"><title>Results</title><sec id="s2-1"><title>Purification and expansion of primary CD133<sup>+</sup> human pancreatic ductal cells</title><p>To identify human pancreatic epithelial cells that can be grown and maintained in culture, we systematically screened cell isolation methods and culture conditions with dispersed adult human pancreatic cells obtained from cadaveric donors without known pancreatic cancer, diabetes mellitus, or other pancreatic diseases (<xref ref-type="table" rid="tbl1">Table 1</xref>). With primary cells plated at low density, we observed formation of multicellular epithelial spheres, when cultured in Matrigel with a serum-free culture medium without feeder cells (‘Materials and methods’, <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1A</xref>). The multicellular sphere formation suggested primary cell expansion, so based on this assay we fractionated cells by fluorescence-activated cell sorting (FACS) to isolate and characterize sphere-forming pancreatic cells. A survey of cell surface markers used for fetal mouse pancreatic cell isolation (<xref ref-type="bibr" rid="bib51">Sugiyama et al., 2007</xref>) revealed that antibodies recognizing CD133 enriched sphere-forming cells by four fold, whereas sphere-forming cells were depleted in the CD133<sup>neg</sup> fraction (<xref ref-type="fig" rid="fig1">Figure 1A,B</xref>). Immunohistochemical analysis of the human adult pancreas revealed CD133 expression at the apical portion of duct epithelial cells that co-expressed keratin 19 (KRT19), whereas CD133 was undetectable in islet endocrine cells or acinar cells (<xref ref-type="fig" rid="fig1">Figure 1C</xref>, <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1B</xref>), consistent with prior reports (<xref ref-type="bibr" rid="bib29">Lardon et al., 2008</xref>). We have achieved sphere formation from over 35 consecutive adult donors (<xref ref-type="table" rid="tbl1">Table 1</xref>); thus, the sphere formation of primary adult human pancreatic CD133<sup>+</sup> cells was highly reproducible.<table-wrap id="tbl1" position="float"><object-id pub-id-type="doi">10.7554/eLife.00940.003</object-id><label>Table 1.</label><caption><title>Phenotypes of pancreas donors</title><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.003">http://dx.doi.org/10.7554/eLife.00940.003</ext-link></p></caption><table frame="hsides" rules="groups"><thead><tr><th>Anonymous ID</th><th>Age (year)</th><th>Gender</th><th>Body mass index</th></tr></thead><tbody><tr><td align="char" char=".">1</td><td align="char" char=".">31</td><td>Male</td><td>28.1</td></tr><tr><td align="char" char=".">2</td><td align="char" char=".">52</td><td>Male</td><td>31.6</td></tr><tr><td align="char" char=".">3</td><td align="char" char=".">52</td><td>Male</td><td>Not provided</td></tr><tr><td align="char" char=".">6</td><td align="char" char=".">16</td><td>Female</td><td>20.4</td></tr><tr><td align="char" char=".">9</td><td align="char" char=".">34</td><td>Male</td><td>35.4</td></tr><tr><td align="char" char=".">10</td><td align="char" char=".">50</td><td>Female</td><td>23</td></tr><tr><td align="char" char=".">11</td><td align="char" char=".">32</td><td>Female</td><td>36.2</td></tr><tr><td align="char" char=".">12</td><td align="char" char=".">35</td><td>Male</td><td>45.7</td></tr><tr><td align="char" char=".">13</td><td align="char" char=".">23</td><td>Female</td><td>26.6</td></tr><tr><td align="char" char=".">14</td><td align="char" char=".">51</td><td>Female</td><td>23.3</td></tr><tr><td align="char" char=".">15</td><td align="char" char=".">48</td><td>Male</td><td>36.7</td></tr><tr><td align="char" char=".">16</td><td align="char" char=".">25</td><td>Male</td><td>21.8</td></tr><tr><td align="char" char=".">17</td><td align="char" char=".">63</td><td>Female</td><td>30.9</td></tr><tr><td align="char" char=".">18</td><td align="char" char=".">44</td><td>Male</td><td>24.7</td></tr><tr><td align="char" char=".">19</td><td align="char" char=".">39</td><td>Male</td><td>27.36</td></tr><tr><td align="char" char=".">20</td><td align="char" char=".">44</td><td>Male</td><td>23.5</td></tr><tr><td align="char" char=".">21</td><td align="char" char=".">50</td><td>Female</td><td>31</td></tr><tr><td align="char" char=".">22</td><td align="char" char=".">40</td><td>Female</td><td>26</td></tr><tr><td align="char" char=".">23</td><td align="char" char=".">53</td><td>Male</td><td>31</td></tr><tr><td align="char" char=".">24</td><td align="char" char=".">19</td><td>Female</td><td>20.83</td></tr><tr><td align="char" char=".">25</td><td align="char" char=".">34</td><td>Male</td><td>22.8</td></tr><tr><td align="char" char=".">26</td><td align="char" char=".">55</td><td>Male</td><td>37.7</td></tr><tr><td align="char" char=".">27</td><td align="char" char=".">17</td><td>Female</td><td>31.1</td></tr><tr><td align="char" char=".">28</td><td align="char" char=".">33</td><td>Male</td><td>18.8</td></tr><tr><td align="char" char=".">29</td><td align="char" char=".">48</td><td>Male</td><td>36.6</td></tr><tr><td align="char" char=".">30</td><td align="char" char=".">40</td><td>Female</td><td>28.4</td></tr><tr><td align="char" char=".">31</td><td align="char" char=".">43</td><td>Female</td><td>35.3</td></tr><tr><td align="char" char=".">32</td><td align="char" char=".">47</td><td>Female</td><td>21</td></tr><tr><td align="char" char=".">33</td><td align="char" char=".">48</td><td>Female</td><td>23.3</td></tr><tr><td align="char" char=".">37</td><td align="char" char=".">28</td><td>Male</td><td>24.2</td></tr><tr><td align="char" char=".">40</td><td align="char" char=".">34</td><td>Male</td><td>32.8</td></tr><tr><td align="char" char=".">41</td><td align="char" char=".">22</td><td>Male</td><td>19.6</td></tr><tr><td align="char" char=".">42</td><td align="char" char=".">53</td><td>Female</td><td>22.4</td></tr><tr><td align="char" char=".">44</td><td align="char" char=".">16</td><td>Male</td><td>33.9</td></tr><tr><td align="char" char=".">45</td><td align="char" char=".">54</td><td>Male</td><td>29.6</td></tr><tr><td align="char" char=".">46</td><td align="char" char=".">18</td><td>Male</td><td>21.8</td></tr><tr><td align="char" char=".">48</td><td align="char" char=".">24</td><td>Male</td><td>25.5</td></tr></tbody></table></table-wrap><fig-group><fig id="fig1" position="float"><object-id pub-id-type="doi">10.7554/eLife.00940.004</object-id><label>Figure 1.</label><caption><title>The ductal cell surface marker CD133 enriches sphere-forming cells from dissociated human adult pancreas.</title><p>(<bold>A</bold>) Left panel, FACS plot of the dissociated human adult pancreas stained with (gray) or without (blue) antibodies specific for CD133. Right panel, A schematic of the sphere culture system and a representative sphere after culture. (<bold>B</bold>) Quantification of spheres generated from CD133<sup>+</sup>, CD133<sup>neg</sup>, and unsorted cells. Data are presented as mean ± SEM (n = 4). (<bold>C</bold>) Immunostaining of CD133 (green) with a ductal marker KRT19 (red) and C-peptide (red) in adult human pancreas. (<bold>D</bold>) The gene expression profiles of FACS-sorted human adult pancreatic cells and isolated islets (islet values normalized to 1). Data are presented as mean ± SEM (n = 3). (<bold>E</bold>) Representative immunostaining pictures of sorted cells with KRT19 (green) or C-peptide (green). (<bold>F</bold>) Quantification of cell immunostaining after FACS. ≥7200 cells were counted per staining condition. n.d.= not detected. Scale bars, 50 µm. See also <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1D</xref>.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.004">http://dx.doi.org/10.7554/eLife.00940.004</ext-link></p></caption><graphic xlink:href="elife00940f001"/></fig><fig id="fig1s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.005</object-id><label>Figure 1—figure supplement 1.</label><caption><title>Sorted CD133+ cells originate from pancreatic ducts.</title><p>(<bold>A</bold>) Schematic diagram of the experimental procedure. Dissociated pancreatic cells were embedded and cultured as previously described (<xref ref-type="bibr" rid="bib30">Lawson et al., 2007</xref>). Scale bar, 200 µm. (<bold>B</bold>) Confocal images of CD133 (green) and CPA1 (red) co-staining in adult human pancreas tissue. Scale bar, 20 µm. (<bold>C</bold>) <italic>CEL</italic> expression profiles of FACS-sorted human adult pancreatic cells and isolated islets (islet values normalized to 1). Data are presented as mean ± SEM (n=3). (<bold>D</bold>) Representative immunostaining pictures of sorted cells. Scale bar, 50 µm.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.005">http://dx.doi.org/10.7554/eLife.00940.005</ext-link></p></caption><graphic xlink:href="elife00940fs001"/></fig></fig-group></p><p>To assess the properties of FACS-purified adult pancreatic CD133<sup>+</sup> cells, we performed quantitative reverse transcription PCR (qRT-PCR). This revealed that CD133<sup>+</sup> cells expressed high levels of mRNA encoding ductal markers (<italic>KRT19</italic> and <italic>CAR2</italic>), while mRNAs expressed in acinar (<italic>CPA1</italic> and <italic>CEL</italic>) and endocrine (<italic>CHGA, INS,</italic> and <italic>GCG</italic>) cells were exclusively enriched in the CD133<sup>neg</sup> fraction (<xref ref-type="fig" rid="fig1">Figure 1D</xref>, <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1C</xref>). Immunostaining confirmed that >98% of sorted CD133<sup>+</sup> cells produced KRT19, whereas CD133<sup>+</sup> cells produced no detectable islet hormone (<xref ref-type="fig" rid="fig1">Figure 1E,F</xref>, <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1D</xref>). Thus, FACS efficiently eliminated islet endocrine and acinar cells, and enriched for a population of primary adult pancreatic duct cells that expanded as epithelial spheres in feeder- and serum-free culture.</p></sec><sec id="s2-2"><title>Maintenance of ductal phenotypes by self-renewing human pancreatic CD133<sup>+</sup> cells</title><p>After commencing in vitro cultures, the epithelial spheres from CD133<sup>+</sup> ductal cells attained diameters ranging from 40 to 520 µm in 2 weeks (<xref ref-type="fig" rid="fig1">Figure 1A</xref>, <xref ref-type="fig" rid="fig1s1">Figure 1—figure supplement 1A</xref> and <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1A</xref>). Spheres 350–500 µm in diameter were composed of 1470 ± 310 cells (n = 5); thus, based on evidence of clonal expansion (see below), we calculated that spheres resulted from a minimum of 10 cell divisions in 2 weeks. Sphere epithelium maintained KRT19 protein expression and a polarized monolayer as indicated by apical localization of CD133 (<xref ref-type="fig" rid="fig2">Figure 2A</xref>, <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1A,D</xref>). Neither acinar (CPA1) nor islet endocrine (CHGA and insulin C-peptide) markers were detectable (<xref ref-type="fig" rid="fig3">Figure 3C</xref> and data not shown), suggesting epithelial cells in cultured spheres maintained ductal characteristics.<fig-group><fig id="fig2" position="float"><object-id pub-id-type="doi">10.7554/eLife.00940.006</object-id><label>Figure 2.</label><caption><title>Clonal expansion and passaging of ductal spheres.</title><p>(<bold>A</bold>) Confocal images of 2-week-old spheres immunostained with KRT19, CD133, Ki-67, and Phospho-Histone H3 (all green). Note the apical localization of CD133. Scale bars, 50 µm. (<bold>B</bold>) Representative time-lapse images of sphere formation from single cell (arrowhead). Images taken every 12 hr for 9 days are shown. Arrows point a non-sphere forming cell used as a landmark. (<bold>C</bold>) Representative pictures of spheres after each passage. Scale bars, 100 µm. (<bold>D</bold>) Quantification of cell number in spheres after each indicated passage. Y axis represents fold increase of total cell numbers relative to the one measured in the first ‘generation’ of spheres (G1).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.006">http://dx.doi.org/10.7554/eLife.00940.006</ext-link></p></caption><graphic xlink:href="elife00940f002"/></fig><fig id="fig2s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.007</object-id><label>Figure 2—figure supplement 1.</label><caption><title>Quantification of sphere growth and passaging.</title><p>(<bold>A</bold>) A representative image of human spheres grown for 2 weeks in culture. Note variable sphere sizes. Scale bar, 100 µm. (<bold>B</bold>) Quantification of Ki-67-expressing cells as a percentage of total cell number shown in <xref ref-type="fig" rid="fig2">Figure 2A</xref>. More than 200 cells per slide, from three or more slides per sample were counted. Data are presented as means ± S.D. (<bold>C</bold>) Quantification of the total cell number in each passage of the individual samples shown in <xref ref-type="fig" rid="fig2">Figure 2D</xref>. Y axis represents fold increase of total cell numbers relative to the one measured in the first ‘generation’ of spheres (G1). (<bold>D</bold>) Representative confocal images of G1 and G7 spheres co-immunostained with KRT19 (red) and CD133 (green). Scale bars, 50 µm. (<bold>E</bold>) Quantification of the total cell number of CD133<sup>+</sup> cells in G1 and G7 spheres. Data are presented as means ± S.D.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.007">http://dx.doi.org/10.7554/eLife.00940.007</ext-link></p></caption><graphic xlink:href="elife00940fs002"/></fig></fig-group><fig-group><fig id="fig3" position="float"><object-id pub-id-type="doi">10.7554/eLife.00940.008</object-id><label>Figure 3.</label><caption><title>Neurog3 is sufficient to convert pancreatic ductal spheres into hormone-expressing endocrine-like cells.</title><p>(<bold>A</bold>) Schematic of growth and reprogramming strategies. See ‘Materials and methods’ for details. (<bold>B</bold>) Schematics of adenoviral constructs used. (<bold>C</bold>) Relative mRNA level of Neurog3 targets (<italic>NEUROD1, INSM1</italic>, and <italic>RFX6</italic>), endocrine cell-specific genes (<italic>PAX4, NKX2.2</italic>, and <italic>CHGA</italic>), and pancreatic hormones (<italic>SST</italic> and <italic>GHRL</italic>). Data are presented as mean ± SEM (n ≥ 3). (<bold>D</bold>) Representative confocal images of Ad-RFP-Neurog3 infected spheres after immunostaining with antibodies specific to mouse Neurog3, NEUROD1, NKX2.2, SST, and GHRL. Note that all hormone-positive cells are CHGA-positive. Right: co-staining of SST and GHRL. Scale bar, 20 µm. (<bold>E</bold>) Quantification of the staining results shown in (<bold>D</bold>). Pie graph represents the percentage of the hormone<sup>+</sup> cells. (<bold>F</bold>) A representative FACS plot of dissociated ductal spheres infected with Ad-RFP-Neurog3 adenovirus (red) or uninfected control (gray). Fractions P1 through P5 were sorted based on RFP fluorescence intensity. (<bold>G</bold>) qRT-PCR analysis of fractions P1 through P5 from (<bold>F</bold>). ‘U’ indicates unsorted cells. Analytical duplicates are shown. Data are presented as mean ± SD.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.008">http://dx.doi.org/10.7554/eLife.00940.008</ext-link></p></caption><graphic xlink:href="elife00940f003"/></fig><fig id="fig3s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.009</object-id><label>Figure 3—figure supplement 1.</label><caption><title>Representative confocal images of spheres infected with control virus (Ad-RFP).</title><p>(A) Representative confocal images of control virus (Ad-RFP) infected spheres after immunostaining with antibodies specific to mouse Neurog3, NEUROD1, NKX2.2, SST, and GHRL. Right: immunostaining to detect cells co-expressing SST and GHRL.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.009">http://dx.doi.org/10.7554/eLife.00940.009</ext-link></p></caption><graphic xlink:href="elife00940fs003"/></fig></fig-group></p><p>To assess whether sphere growth was achieved by cell proliferation or by other mechanisms like cell migration and aggregation, we analyzed spheres by immunostaining and time-lapse imaging. Immunohistochemistry revealed the proliferation marker Ki-67 in more than 25% of cells comprising 2-week-old spheres (<xref ref-type="fig" rid="fig2">Figure 2A</xref>, <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1B</xref>; labeling index 26.5 ± 5.1%), data further supported by detection of a second proliferation marker, phospho-histone H3 (<xref ref-type="fig" rid="fig2">Figure 2A</xref>). Time-lapse imaging revealed that spheres arose from single cells (<xref ref-type="fig" rid="fig2">Figure 2B</xref>), providing strong evidence that sphere formation resulted from CD133<sup>+</sup> ductal cell proliferation, rather than through cell migration and aggregation. Enzymatic dispersion of 2-week-old G1 spheres and subsequent culture revealed that the spheres can be passaged up to seven generations (G7, 3 months) and that the total number of cells increased with each generation (<xref ref-type="fig" rid="fig2">Figure 2C,D</xref>, <xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1C</xref>). After G7, ductal cell expansion was not achieved, and the spheres were not formed (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1C</xref> and data not shown), supporting the view that ductal epithelial cells are not immortalized, and consistent with the origin of pancreatic cells from donors without neoplasia.</p></sec><sec id="s2-3"><title><italic>Neurog3</italic> converts pancreatic duct cells into progeny expressing islet hormones</title><p>The endocrine potential of human or mouse pancreatic ductal cells remains controversial. To investigate the potential of purified human pancreatic ductal cells to achieve an endocrine fate, we used an adenovirus-mediated transgenic system. <italic>Neurog3</italic> is a transcription factor necessary and sufficient for pancreatic endocrine cell differentiation in vivo (<xref ref-type="bibr" rid="bib15">Gradwohl et al., 2000</xref>; <xref ref-type="bibr" rid="bib16">Gu et al., 2002</xref>) and, combined with other factors, can induce pancreatic acinar-to-islet cell conversion in mice (<xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>). To test if <italic>Neurog3</italic> expression could respecify human duct cells toward an endocrine fate, we infected cultured spheres as well as primary CD133<sup>+</sup> cells with recombinant adenovirus co-expressing red fluorescent protein and <italic>Neurog3</italic> (Ad-RFP-Neurog3), and assessed changes in gene expression by qRT-PCR (<xref ref-type="fig" rid="fig3 fig4">Figure 3A–C and 4C</xref>). <italic>Neurog3</italic> induced the expression of <italic>NEUROD1, INSM1</italic>, and <italic>RFX6</italic> (<xref ref-type="fig" rid="fig3">Figure 3C</xref>), genes whose mouse homologs are known direct targets of <italic>Neurog3</italic> in pancreas development (<xref ref-type="bibr" rid="bib33">Mellitzer et al., 2006</xref>; <xref ref-type="bibr" rid="bib47">Smith et al., 2010</xref>). Ad-RFP-Neurog3 infection induced expression of the pan-endocrine markers chromogranin A (<italic>CHGA</italic>) and synaptophysin in both primary CD133<sup>+</sup> duct cells and cultured spheres (<xref ref-type="fig" rid="fig3 fig4">Figures 3C and 4C</xref>, and data not shown). Ad-RFP-Neurog3 infection also induced expression of mRNA encoding PAX4 and NKX2.2, transcriptional regulators of pancreatic endocrine cell fate (<xref ref-type="bibr" rid="bib49">Sosa-Pineda et al., 1997</xref>; <xref ref-type="bibr" rid="bib52">Sussel et al., 1998</xref>), and mRNA encoding crucial β-cell factors such as the prohormone processing enzymes <italic>PCSK1</italic> (PC1/3) and <italic>PCSK2</italic> (PC2), K<sub>ATP</sub> channel components <italic>KCNJ11</italic> (KIR6.2) and <italic>ABCC8</italic> (SUR1), and glucokinase (<italic>GCK</italic>) (<xref ref-type="fig" rid="fig4">Figure 4D</xref>). Moreover, Ad-RFP-Neurog3 significantly induced mRNA encoding the pancreatic hormones ghrelin and somatostatin, but not mRNAs encoding insulin, glucagon, PPY or the intestinal hormones cholecystokinin and gastrin (<xref ref-type="fig" rid="fig3 fig4">Figures 3C and 4D</xref>, <xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1A</xref>, and data not shown). These findings support the conclusion that human adult pancreatic ductal cells harbor pancreatic endocrine potential upon induction of Neurog3.<fig-group><fig id="fig4" position="float"><object-id pub-id-type="doi">10.7554/eLife.00940.010</object-id><label>Figure 4.</label><caption><title>Induction of four transcription factors <italic>(Neurog3, MafA, Pdx1, and Pax6)</italic> produces Insulin<sup>+</sup> endocrine cells in pancreatic ductal spheres in vitro.</title><p>(<bold>A</bold>) Schematics of adenoviruses used. (<bold>B</bold>) <italic>INS</italic> qRT-PCR analysis of human spheres infected with control (R = RFP) or a combination of MafA (M), Neurog3 (N), and Pdx1 (P) (MNP) n = 4. (<bold>C</bold>) qRT-PCR analysis of <italic>INS</italic>, <italic>SST</italic>, and <italic>CHGA</italic> with freshly sorted CD133<sup>+</sup> ductal cells infected with adenoviruses encoding Neurog3 or all four genes (4V) (n = 2). (<bold>D</bold>) qRT-PCR analysis of the spheres infected with a combination of adenoviruses. <italic>Pax6</italic> abbreviated as ‘6’, (n ≥ 3). (<bold>E</bold>) qRT-PCR analysis of the spheres infected with 4V minus each indicated factor n = 2. All bar graph data are presented as mean ± SEM with mRNA levels from purified adult human islets normalized to 1. (<bold>F</bold>) Confocal images of infected spheres after staining with antibodies recognizing C-peptide. Note that adenoviruses encoding <italic>Neurog3</italic> (N) and <italic>Pdx1</italic> (P) also express RFP. Scale bar, 20 µm. (<bold>G</bold>) Quantification of the CHGA-, SST-, and C-peptide-immunoreactive cells in the spheres infected with the indicated combination of adenoviruses. Note that the number of C-peptide-positive cells increased in 4V than MNP by 18–20-fold.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.010">http://dx.doi.org/10.7554/eLife.00940.010</ext-link></p></caption><graphic xlink:href="elife00940f004"/></fig><fig id="fig4s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.011</object-id><label>Figure 4—figure supplement 1.</label><caption><title>GCG, PPY, and PAX6 mRNA levels after sphere infection with adenovirus combinations.</title><p>(<bold>A</bold>) qRT-PCR analysis of the spheres infected with a combination of adenoviruses. <italic>PAX6</italic> abbreviated as ‘6’, (n ≥ 3). Note that <italic>PAX6</italic> qRT-PCR probe recognizes both endogenous and exogenous <italic>PAX6</italic> mRNA.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.011">http://dx.doi.org/10.7554/eLife.00940.011</ext-link></p></caption><graphic xlink:href="elife00940fs004"/></fig></fig-group></p><p>Immunostaining confirmed these qRT-PCR findings and demonstrated that only RFP<sup>+</sup> cells produced by Ad-RFP-Neurog3 infection were immunostained with antibodies recognizing NEUROD1, NKX2.2, CHGA, SST or GHRL (<xref ref-type="fig" rid="fig3">Figure 3B,D</xref>, <xref ref-type="fig" rid="fig3s1">Figure 3—figure supplement 1</xref>). We also confirmed that no insulin-, glucagon- or PPY-positive cells were observed by immunostaining (data not shown). While only a subset of cells infected with Ad-RFP-Neurog3 (RFP<sup>+</sup>) expressed CHGA, we noted all GHRL<sup>+</sup> or SST<sup>+</sup> cells co-expressed CHGA (<xref ref-type="fig" rid="fig3">Figure 3D</xref>). Quantification of CHGA<sup>+</sup> and hormone<sup>+</sup> cells revealed that 30% of infected cells (RFP<sup>+</sup>) expressed CHGA. At least 45% of CHGA<sup>+</sup> cells produced SST or GHRL, and less than 2% of CHGA<sup>+</sup> cells expressed both hormones (<xref ref-type="fig" rid="fig3">Figure 3D,E</xref>). Thus, Neurog3 expression efficiently converted primary human ductal cells and cultured ductal epithelial spheres into hormone-expressing cells with cardinal features of endocrine pancreas.</p><p>In mice, <italic>Neurog3</italic> gene dosage can determine commitment between exocrine and endocrine lineages in pancreas development (<xref ref-type="bibr" rid="bib61">Wang et al., 2010</xref>). Therefore, we next assessed the possibility that the 70% of RFP<sup>+</sup> cells infected by Ad-RFP-Neurog3 failing to express CHGA may have achieved inadequate levels of Neurog3 expression. We fractionated cells produced by Ad-RFP-Neurog3 infection by RFP intensity and measured mRNA expression of Neurog3, CHGA, SST and GHRL by qRT-PCR (<xref ref-type="fig" rid="fig3">Figure 3F,G</xref>). We found that cell fractions with the highest levels of RFP expression (‘P4 and P5’, <xref ref-type="fig" rid="fig3">Figure 3F</xref>) had the highest levels of mouse <italic>Neurog3</italic> mRNA, and only these cell fractions produced mRNA encoding CHGA, SST or GHRL (<xref ref-type="fig" rid="fig3">Figure 3G</xref>). These data suggest that relatively high threshold levels of Neurog3 may be necessary and sufficient for directing endocrine differentiation of human pancreatic cells.</p></sec><sec id="s2-4"><title>Conversion of pancreatic duct cells into progeny that produce, process, and store insulin</title><p>The transcription factors <italic>MafA</italic>, <italic>Neurog3,</italic> and <italic>Pdx1</italic> (a combination hereafter summarized as ‘MNP’) were sufficient to convert adult mouse acinar cells into insulin-producing cells (IPCs: <xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>). We constructed three adenoviruses expressing <italic>MafA</italic>, <italic>Neurog3,</italic> or <italic>Pdx1</italic> (see ‘Materials and methods’; <xref ref-type="fig" rid="fig4">Figure 4A</xref>), and infected cultured spheres with this MNP combination. Within 5 days after infection, we reproducibly detected <italic>INS</italic> mRNA induction but at extremely low levels relative to adult human islet controls (0.0035 ± 0.0012% of islet levels; <xref ref-type="fig" rid="fig4">Figure 4B</xref>). Thus, we sought additional factors and discovered that mRNA encoding PAX6<italic>,</italic> an important regulator of mouse pancreatic endocrine cell development (<xref ref-type="bibr" rid="bib44">Sander et al., 1997</xref>), was induced by MNP to only 0.03% of levels in control islets (<xref ref-type="fig" rid="fig4s1">Figure 4—figure supplement 1A</xref>). When combined with <italic>MafA</italic>, <italic>Neurog3,</italic> and <italic>Pdx1</italic> (encoded in four viruses, ‘4V’), <italic>Pax6</italic> induced <italic>INS</italic> expression in primary CD133<sup>+</sup> ductal cells or spheres by over 30-fold relative to MNP (<xref ref-type="fig" rid="fig4">Figure 4A,C,D</xref>). We observed ductal conversion to IPCs with four consecutive, independent donors (<italic>INS</italic>, <xref ref-type="fig" rid="fig4">Figure 4D</xref>). We also detected substantially increased expression of other islet endocrine markers, including <italic>SST, GCK, PCSK1, KCNJ11</italic>, and <italic>ABCC8</italic> (<xref ref-type="fig" rid="fig4">Figure 4D</xref>). Immunohistochemical analyses demonstrated that the number of Insulin<sup>+</sup> cells was increased by 18 to 20-fold in spheres transduced by the four factor combination (4V) compared to the MNP combination (<xref ref-type="fig" rid="fig4">Figure 4F,G</xref>). ELISA studies quantified and confirmed this increase of proinsulin levels, showing that the spheres derived from 4V exposure contained proinsulin levels that averaged 0.7% of those in human islets (<xref ref-type="fig" rid="fig5">Figure 5E</xref>). Systematic removal of individual factors from this four virus combination revealed that omission of <italic>Neurog3</italic> prevented expression of <italic>INS</italic>, <italic>CHGA</italic> or <italic>SST</italic> (<xref ref-type="fig" rid="fig4">Figure 4E</xref>). Omission of virus expressing <italic>MafA</italic> or <italic>Pax6</italic> from this combination significantly reduced <italic>INS</italic> expression (<xref ref-type="fig" rid="fig4">Figure 4E–G</xref>), whereas omission of virus expressing <italic>Pdx1</italic> did not significantly decrease <italic>INS</italic> expression. Thus, <italic>Neurog3</italic>-mediated endocrine cell conversion is required for the production of IPCs as well as other hormone-producing cells from ductal spheres.<fig-group><fig id="fig5" position="float"><object-id pub-id-type="doi">10.7554/eLife.00940.012</object-id><label>Figure 5.</label><caption><title>Induced insulin-secreting cells resemble functional β-cells.</title><p>(<bold>A</bold>) Schematic of adenoviral constructs used. See ‘Materials and methods’ and <xref ref-type="fig" rid="fig5s5">Figure 5—figure supplement 5</xref> for details. (<bold>B</bold>) A schematic diagram of growth, conversion, and maturation procedures. (<bold>C</bold>) qRT-PCR analysis of spheres infected with Ad-eGFP (black) or Ad-4TF (gray) followed by extended culture. Data are normalized to adult human islet samples (red dotted line). (<bold>D</bold>) Representative confocal images of 4TFM spheres immunostained with indicated antibodies. Scale bar, 20 µm. (<bold>E</bold>) Quantification of total proinsulin and C-peptide content in GFP, 4TF, 4TFM spheres and human adult islets (Top). Total protein level (pmol) was normalized by total genomic DNA content (µg). Ratio of proinsulin and C-peptide content is presented as % (Bottom). Sph. = Spheres. (<bold>F</bold>) Representative electron microscopic images of 4TFM spheres. Dotted white line demarks cell boundary between converted, granulated (left) and non-converted (right) cells. Dense core vesicles with different morphology in converted cells are shown in the right panels. Scale bar, 1 µm. (<bold>G</bold>) Human C-peptide secretion assay of 4TFM spheres stimulated by the indicated secretagogues and drugs. Gluc = Glucose, Tol = Tolbutamide, Diaz. = Diazoxide. Data are presented as means ± SEM (n = 2 for Diaz.; n ≥ 3 for all other conditions).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.012">http://dx.doi.org/10.7554/eLife.00940.012</ext-link></p></caption><graphic xlink:href="elife00940f005"/></fig><fig id="fig5s1" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.013</object-id><label>Figure 5—figure supplement 1.</label><caption><title>Phenotypes of induced Insulin-secreting cells.</title><p>(<bold>A</bold>) qRT-PCR analysis of <italic>INS</italic> and <italic>IAPP</italic> with GFP, 4TF or 4TFM spheres. Note that spheres from extended culture (4TFM) had significantly increased mRNA levels of <italic>INS</italic> and <italic>IAPP</italic> (n = 6). (<bold>B</bold>) Quantification of CHGA- and C-peptide-immunoreactive cells in the 4TFM spheres. (<bold>C</bold>) Representative confocal images of 4TFM spheres with NKX6.1 and SST. Note the non-overlapping staining. Blue = DAPI. (<bold>D</bold> and <bold>E</bold>) Representative images of electron microscopy. (<bold>D</bold>) Dense core vesicles found adjacent to the plasma membrane. (<bold>E</bold>) Rare cells contain vesicles with irregular shape, reminiscent of δ-cells. (<bold>F</bold>) Four different culture media used for this study. See ‘Experimental procedures’ for details. (<bold>G</bold>) C-peptide secretion was shown as a percentage of total C-peptide content (n = 4). (<bold>H</bold>) Human C-peptide secretion assay with step increase of glucose concentration (left) and with KCl in the presence or absence of extracellular calcium (right) (n ≥ 3). All bar graphs are presented as means ± SEM.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.013">http://dx.doi.org/10.7554/eLife.00940.013</ext-link></p></caption><graphic xlink:href="elife00940fs005"/></fig><fig id="fig5s2" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.014</object-id><label>Figure 5—figure supplement 2.</label><caption><title>Grafted IPCs survive long term and secrete insulin C-peptide upon glucose stimulation.</title><p>(<bold>A</bold>) Representative confocal images of kidney-transplanted IPCs immunostained with indicated antibodies (HuNu = human nuclei-specific antibody). (<bold>B</bold>) Human insulin levels in serum of an IPC-grafted mouse (ID51) before (fasting) or 30 min after glucose challenge (glucose injection). Data are presented as means ± S.D. (<bold>C</bold>) Representative confocal images of liver-transplanted IPCs immunostained with indicated antibodies. (<bold>D</bold>) Representative confocal images of human islets transplanted in the indicated sites (Kidney or EFP) and immunostained with indicated antibodies. (<bold>E</bold>) Human insulin level in serum of human islet-grafted mice before (gray) or 30 min after glucose challenge (black). Data are presented as means ± SD. Scale bars, 20 µm.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.014">http://dx.doi.org/10.7554/eLife.00940.014</ext-link></p></caption><graphic xlink:href="elife00940fs006"/></fig><fig id="fig5s3" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.015</object-id><label>Figure 5—figure supplement 3.</label><caption><title>Sustained expression of exogenous factors after maturation period.</title><p>(<bold>A</bold>) Schematic of qRT-PCR probes designed against adenoviral constructs used. (<bold>B</bold>) qRT-PCR of spheres infected with Ad-eGFP (eGFP) or Ad-Neurog3-IRES-eGFP and Ad-eGFP-M6P (4TF) with or without extended culture. Note that insulin expression is markedly elevated independent of transgene expression.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.015">http://dx.doi.org/10.7554/eLife.00940.015</ext-link></p></caption><graphic xlink:href="elife00940fs007"/></fig><fig id="fig5s4" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.016</object-id><label>Figure 5—figure supplement 4.</label><caption><title>Conversion of human dermal fibroblasts.</title><p>qRT-PCR of human dermal fibroblasts or spheres infected with Ad-eGFP (GFP) or Ad-Neurog3-IRES-eGFP and Ad-eGFP-M6P (4TFM) and cultured. qRT-PCR probes for <italic>INS</italic> (<bold>A</bold>) and <italic>CHGA</italic> (<bold>B</bold>) were used to assess cell conversion. See ‘Materials and methods’ for detail.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.016">http://dx.doi.org/10.7554/eLife.00940.016</ext-link></p></caption><graphic xlink:href="elife00940fs008"/></fig><fig id="fig5s5" position="float" specific-use="child-fig"><object-id pub-id-type="doi">10.7554/eLife.00940.017</object-id><label>Figure 5—figure supplement 5.</label><caption><title>Protein expression of viral transgenes.</title><p>(<bold>A</bold>) Schematic of adenoviral constructs used. To differentiate viral transgenes from endogenously encoded proteins, MAFA, PAX6, and PDX1 were epotpe-tagged with Myc (N-terminus), HA (C-terminus), and Flag (C-terminus), respectively. (<bold>B</bold>) Representative confocal images of 4TFM spheres with antibodies against mouse Neurog3, Myc, HA, and Flag. Scale bar, 20 µm.</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.017">http://dx.doi.org/10.7554/eLife.00940.017</ext-link></p></caption><graphic xlink:href="elife00940fs009"/></fig></fig-group></p><p>Although ELISA studies readily detected proinsulin production by IPCs in our 4V spheres, we failed to detect processed C-peptide by ELISA (<xref ref-type="fig" rid="fig5">Figure 5E</xref>) or by immunostaining with antibodies recognizing cleaved C-peptide (data not shown). Thus, we sought methods to enhance proinsulin processing in IPCs produced by genetic conversion. For this, we used Ad-<italic>Neurog3</italic>-IRES-eGFP and a second adenovirus constructed to express simultaneously the three transcription factors MAFA, PAX6, and PDX1 (Ad-eGFP-M6P) in cultured G1 spheres (<xref ref-type="fig" rid="fig5">Figure 5A</xref>, referred to as ‘4TF’ combination). Compared to our standard 5 day post-infection culture (4TF), we found that two additional weeks of culture (referred to as ‘4TFM’) resulted in a 10-fold increase of <italic>INS</italic> mRNA expression in spheres (<xref ref-type="fig" rid="fig5">Figure 5B,C</xref>, <xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1A</xref>). We observed conversion to IPCs with five consecutive, independent donors (<italic>INS</italic>, <xref ref-type="fig" rid="fig5">Figure 5C</xref>), demonstrating the robustness of our conversion method. The total number of converted IPCs appeared unchanged after this extended culture compared to 4V cultures (<xref ref-type="fig" rid="fig4">Figure 4G</xref>, <xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1B</xref>), suggesting that <italic>INS</italic> mRNA levels per cell were increased in the 4TFM (4 <bold>t</bold>ranscription <bold>f</bold>actors in two viruses plus <bold>m</bold>aturation period) condition. In addition, mRNA encoding islet amyloid pancreatic polypeptide (IAPP), a β-cell dense core granule component not detectable in standard 4TF conditions, was readily detected in 4TFM spheres (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1A</xref>). Likewise, multiple mRNAs encoding β-cell factors were expressed at levels comparable to those in purified human islets (<xref ref-type="fig" rid="fig5">Figure 5C</xref>), including the transcription factors <italic>NKX2.2</italic> and <italic>NKX6.1</italic>, <italic>GCK</italic>, glucose transporters <italic>SLC2A1</italic> (GLUT1) and <italic>SLC2A2</italic> (GLUT2), <italic>PCSK1</italic>, <italic>PCSK2</italic>, Zinc transporter <italic>SLC30A8</italic>, <italic>KCNJ11</italic>, <italic>ABCC8</italic>, the voltage-gated calcium channel component <italic>CACNA1C</italic>, regulators of Ca<sup>++</sup>-induced insulin exocytosis like <italic>RAB3A</italic>, <italic>SYT3</italic>, and <italic>VAMP2</italic>, and the postulated maturation marker Urocortin 3 (<italic>UCN3</italic>) (<xref ref-type="bibr" rid="bib50">Suckale and Solimena, 2010</xref>; <xref ref-type="bibr" rid="bib5">Blum et al., 2012</xref>). Immunohistochemical analyses corroborated our qRT-PCR analysis, and showed that converted Insulin<sup>+</sup> IPCs did not express other islet hormones (<xref ref-type="fig" rid="fig5">Figure 5D</xref>). Although we were unable to assess endogenous MAFA and PDX1 in cells with virally-expressed exogenous MAFA and PDX1 protein, we readily detected other known β-cell specific markers including NKX6.1, IAPP, and PC1/3 (<xref ref-type="fig" rid="fig5">Figure 5D</xref>). Moreover, Insulin<sup>+</sup> cells, but not other hormone<sup>+</sup> cells, expressed NKX6.1, a transcription factor with expression normally restricted in islets to β-cells (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1C</xref> and data not shown).</p><p>To assess enhanced IPC maturation after extended culture (4TFM), we measured proinsulin and insulin C-peptide by ELISA. Total insulin (proinsulin + C-peptide) levels ranged from 3.4 to 15.2 pmol/µg DNA (<xref ref-type="fig" rid="fig5">Figure 5E</xref>), equal to approximately 9.6% of the total insulin protein level found in human adult islets (<xref ref-type="fig" rid="fig5">Figure 5E</xref>). Moreover, the percentage of insulin C-peptide processing in IPCs was comparable to that found in adult human islets (IPCs 77–92%; human islets 96–97%), indicating that maturation of IPCs during extended culture permitted proinsulin processing (<xref ref-type="fig" rid="fig5">Figure 5E</xref>). Ultrastructural studies by electron microscopy demonstrated round dense-core vesicles (<xref ref-type="fig" rid="fig5">Figure 5F</xref>) resembling those in adult human β-cells, including subsets of immature (light core) and mature (dense or crystallized core) vesicles, and vesicles adjacent to the plasma membrane (<xref ref-type="fig" rid="fig5">Figure 5F</xref>, <xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1D</xref>). Consistent with the detection of <italic>SST</italic> mRNA (<xref ref-type="fig" rid="fig5">Figure 5C,D</xref>), we also observed rare cells with irregular electron-dense granules characteristic of islet δ-cells (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1E</xref>; <xref ref-type="bibr" rid="bib25">Klimstra et al., 2007</xref>).</p></sec><sec id="s2-5"><title>Regulated insulin C-peptide secretion by IPCs</title><p>Native islet β-cells depolarize and secrete insulin and C-peptide in response to glucose and other physiological or pharmacological stimuli, but reconstructing these hallmark functions in progeny of purified primary human non-β-cells has not been previously achieved during in vitro culture. Compared to baseline secretion in media with 0.1 mM glucose, IPCs increased insulin C-peptide secretion by 2.4-fold upon exposure to 2 mM glucose (<xref ref-type="fig" rid="fig5">Figure 5G</xref>). Similar to insulin release by human islet β-cells (<xref ref-type="bibr" rid="bib31">Lupi et al., 1999</xref>), glucose stimulated the secretion of approximately 4% of total insulin C-peptide in IPCs (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1G</xref>). This effect was blocked when the cells were incubated with glucose and Diazoxide, a drug that opens K<sub>ATP</sub> channels and prevents glucose-stimulated insulin secretion (<xref ref-type="fig" rid="fig5">Figure 5G</xref>). However, unlike adult human islet β-cells, the release of insulin by IPCs was not further increased by 11 mM glucose. Islets from fetal or neonatal stages do not show elevated insulin secretion by high level glucose challenge (<xref ref-type="bibr" rid="bib43">Rozzo et al., 2009</xref>), suggesting that IPCs are similar to immature islet β-cells and that further maturation is possible (<xref ref-type="fig" rid="fig5">Figure 5G</xref>). Calcium and voltage-dependent calcium channels are important regulators of normal insulin secretion after K<sub>ATP</sub> channel-mediated membrane depolarization in β-cells (<xref ref-type="bibr" rid="bib20">Henquin, 2005</xref>). When calcium was omitted in secretion buffer, C-peptide secretion stimulated by glucose was abolished, but restored upon calcium addition (<xref ref-type="fig" rid="fig5">Figure 5G</xref>). Insulin C-peptide release by cultured IPCs was also induced by the depolarizing agent potassium chloride (30 mM KCl), an effect reversed by a subsequent wash in media with 4.8 mM potassium ion (<xref ref-type="fig" rid="fig5">Figure 5G</xref>, <xref ref-type="fig" rid="fig5s1">figure 5—figure supplement 1H</xref>). Treatment with tolbutamide, a K<sub>ATP</sub> channel blocker causing membrane depolarization, also stimulated insulin secretion by IPCs, an effect prevented by omission of calcium (<xref ref-type="fig" rid="fig5">Figure 5G</xref>). Together with data showing expression of key regulators of stimulus-secretion coupling, these findings provide strong evidence that IPCs produced by conversion and extended culture in our system develop regulated insulin secretion.</p><p>We examined the stability of the conversion of human ducts into IPCs by long-term transplantation of the converted spheres into specific transplantation sites of NOD <italic>scid</italic> gamma (NSG) mice (<xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2</xref>; <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1A</xref>). Human C-peptide was readily detected in kidney grafts harvested at specific times by immunostaining (8/12 cases, <xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2A</xref>; <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1A</xref>) and by ELISA (9/10 cases, <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1A</xref>) without detectable tumor formation. This also included C-peptide<sup>+</sup> IPCs left in the transplant location beyond 5 months (<xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2A</xref>, d151), suggesting converted IPCs were stable. However, we observed that the total number of grafted C-peptide<sup>+</sup> cells was drastically reduced within 2 weeks after transplantation, likely due to the apoptotic cell death. In three independent IPC transplants, however, we were able to detect circulating human insulin in the serum of host mice, and its level increased following intraperitoneal glucose challenge (<xref ref-type="fig" rid="fig5s2">Figure 5—figure supplement 2B</xref>; <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1B</xref>). Therefore, these data suggest that despite extensive cell death in early stages of transplantation, IPCs can further mature in vivo and release increased levels of insulin in response to acute glucose challenge.</p></sec></sec><sec id="s3" sec-type="discussion"><title>Discussion</title><p>Methods to regenerate lost or injured cells in diseases like diabetes mellitus are the focus of intensive investigations (<xref ref-type="bibr" rid="bib32">McKnight et al., 2010</xref>; <xref ref-type="bibr" rid="bib4">Benitez et al., 2012</xref>). Generation of insulin-producing cells from human stem cell lines like human ES cells (<xref ref-type="bibr" rid="bib12">D’Amour et al., 2006</xref>; <xref ref-type="bibr" rid="bib28">Kroon et al., 2008</xref>) is an important, and oft-cited ‘benchmark’, in efforts to achieve β-cell replacement. However, in these prior reports, progeny of human ES cells developed largely as poly-hormonal cells, most frequently expressing both glucagon and insulin. Moreover such hESC progeny failed to secrete insulin in response to glucose or other secretagogues unless transplanted as progenitors for >2 months in mice (<xref ref-type="bibr" rid="bib38">Nostro and Keller, 2012</xref>). This transplant-based maturation strategy was complicated by tumor formation (<xref ref-type="bibr" rid="bib13">Fujikawa et al., 2005</xref>). Thus, it has remained unknown whether human cells can develop solely in vitro to generate glucose-responsive insulin-secreting progeny without tumorigenicity. Our data indicate that in principal this can be achieved, using a small number of genes in sorted human pancreatic ductal cells that convert them toward an islet fate, including progeny that produce, store, and secrete insulin in response to glucose.</p><p>Conversion of mouse acinar cells into insulin-producing cells using adenoviral delivery of <italic>Neurog3, Pdx1,</italic> and <italic>MafA</italic> was previously reported (<xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>). However, it has remained unknown whether human pancreatic cells can be converted using transgenic methods toward a β-cell fate. We were unable to culture and expand primary human pancreatic acinar cells (<xref ref-type="fig" rid="fig1">Figure 1B</xref> and data not shown); moreover, we found that the combination of these three genes (MNP) was insufficient to reprogram primary or expanded human pancreatic ductal cells toward a β-cell fate, suggesting transgenic conversion may be restricted by species and cell type. Thus, we postulated that additional transcriptional regulators might be needed to promote human ductal conversion toward a β-cell fate. Like Neurog3, MafA, and Pdx1, the transcription factor Pax6 is expressed in both fetal and adult pancreas, and required to achieve appropriately high levels of <italic>Ins</italic> and <italic>Gcg</italic> expression in mouse islet cell development (<xref ref-type="bibr" rid="bib44">Sander et al., 1997</xref>; <xref ref-type="bibr" rid="bib60">Wang et al., 2009</xref>, <xref ref-type="bibr" rid="bib61">2010</xref>; <xref ref-type="bibr" rid="bib39">Pan and Wright, 2011</xref>). Together with the other factors, we found that Pax6 significantly enhanced expression of β-cell markers during ductal reprogramming into β-cells, and was shown as an essential factor for this process. By systematic addition or omission of each transcription factor, we found <italic>PDX1</italic> is not required for IPC formation. Thus, unlike mouse acinar cells (<xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>) and human hepatocytes (<xref ref-type="bibr" rid="bib45">Sapir et al., 2005</xref>), human ductal cells do not require exogenous Pdx1 expression for conversion toward an endocrine fate, for reasons that remain unclear. Our findings are also consistent with recent reports that transgenic adult mouse ductal cells can generate endocrine cells in vivo (<xref ref-type="bibr" rid="bib1">Al-Hasani et al., 2013</xref>).</p><p>We initially attempted to induce spontaneous differentiation of pancreatic ductal cells using systematic variations of culture conditions, but these efforts proved unsuccessful (J Lee, unpublished results). During pancreas development, Neurog3 level surges in a subset of pancreatic progenitors located in primitive ducts, inducing development of endocrine cell fates (<xref ref-type="bibr" rid="bib68">Zhou et al., 2007</xref>; <xref ref-type="bibr" rid="bib36">Miyatsuka et al., 2009</xref>). Therefore, based on this model, we attempted to mimic induction of Neurog3 in human ductal cells using adenoviral overexpression of <italic>Neurog3</italic>. We found that Neurog3 was necessary and sufficient for reprogramming human ductal cells, and that the level of ectopic <italic>Neurog3</italic> mRNA expressed in ductal cells correlated well with the extent of endocrine reprogramming, including expression of islet hormones (<xref ref-type="fig" rid="fig3">Figure 3G</xref>). These findings are reminiscent of studies by Gu et al. showing that reduced <italic>Neurog3</italic> gene dosage in mice leads to respecification of pancreatic endocrine progenitors into ductal and acinar cells (<xref ref-type="bibr" rid="bib61">Wang et al., 2010</xref>). Thus, <italic>Neurog3</italic> functions may be evolutionarily conserved in allocating cells toward an exocrine or endocrine fate (whether in development or experimental cell conversion) in a dosage-dependent manner. Consistent with prior work revealing that Neurog3 attenuates islet cell proliferation (<xref ref-type="bibr" rid="bib35">Miyatsuka et al., 2011</xref>), we did not observe multiple rounds of cell division, an important prerequisite for some de-differentiation events (<xref ref-type="bibr" rid="bib17">Hanna et al., 2009</xref>), during Neurog3-dependent cell conversion. Also, we observed <italic>Neurog3</italic> induction alone can rapidly upregulate endocrine molecular signatures in cultured human ductal cells. Thus endocrine cell conversion described here may involve direct conversion of human ductal cells into endocrine cells, rather than de-differentiation, but additional studies are required to assess this possibility. Our findings, albeit with enforced transcription factor expression in adult cells, indicate that Neurog3 expression is sufficient to induce latent endocrine programs in human adult ductal cells, a capacity not yet clearly demonstrated, to our knowledge.</p><p>We demonstrated robust expansion of purified human ductal cells in 3-dimensional culture. The cells were clonally expanded and serially passaged up to seven generations over 3 months, achieving an increase in cell number calculated to be up to 3,200-fold. By contrast, in prior studies, the maximum duration of sustained culture achieved with primary human pancreatic ductal cells was 5 weeks (<xref ref-type="bibr" rid="bib56">Trautmann et al., 1993</xref>; <xref ref-type="bibr" rid="bib6">Bonner-Weir et al., 2000</xref>; <xref ref-type="bibr" rid="bib41">Rescan et al., 2005</xref>; <xref ref-type="bibr" rid="bib18">Hao et al., 2006</xref>; <xref ref-type="bibr" rid="bib64">Yatoh et al., 2007</xref>; <xref ref-type="bibr" rid="bib22">Hoesli et al., 2012</xref>). Moreover, cultured cells in spheres maintained cardinal features of primary pancreatic ducts such as apical-basal polarity and KRT19 expression up to seven generations (<xref ref-type="fig" rid="fig2s1">Figure 2—figure supplement 1D,E</xref>). Thus, features of our culture system may be useful for studying the genetics and biology of human ductal cells.</p><p>Prior studies have reported that duct-containing fractions from human adult pancreas can form insulin-producing cells in vitro (<xref ref-type="bibr" rid="bib6">Bonner-Weir et al., 2000</xref>; <xref ref-type="bibr" rid="bib18">Hao et al., 2006</xref>; <xref ref-type="bibr" rid="bib21">Heremans et al., 2002</xref>; <xref ref-type="bibr" rid="bib37">Noguchi et al., 2006</xref>; <xref ref-type="bibr" rid="bib26">Koblas et al., 2008</xref>; <xref ref-type="bibr" rid="bib53">Swales et al., 2012</xref>) or after xeno-transplantion in mice (<xref ref-type="bibr" rid="bib64">Yatoh et al., 2007</xref>). However, the possibility of endocrine cell contamination in the initial ductal fraction or feeder/stromal cells used for co-culture was raised by the detection of mRNAs encoding islet cell hormones and other endocrine markers in these and other studies (<xref ref-type="bibr" rid="bib21">Heremans et al., 2002</xref>; <xref ref-type="bibr" rid="bib14a">Gao et al., 2005</xref>). Therefore, it remained elusive whether human pancreatic ducts retained the potential to produce islet endocrine cells in adult. In this report, we used FACS to fractionate CD133<sup>+</sup> ductal cells and used molecular and immunocytological studies to demonstrate complete elimination of cells expressing markers of differentiated endocrine cells (including islet hormones). Therefore, subsequent conversion of these cells into functional endocrine cells provided unequivocal evidence that endocrine cell-free human adult CD133<sup>+</sup> ductal cell fraction can be converted into islet endocrine cells. Centroacinar cells are located at the junction of acini and tip of intercalated ducts (<xref ref-type="bibr" rid="bib11">Cleveland et al., 2012</xref>) and their properties remain poorly understood. These cells express CD133 (<xref ref-type="bibr" rid="bib23">Immervoll et al., 2008</xref>), raising the possibility that our fractionated CD133<sup>+</sup> cells also include centroacinar cells. Based on their relative paucity in the pancreas, it is unlikely that centroacinar cells are the exclusive source of spheres within this CD133<sup>+</sup> fraction, as more than 11% of CD133<sup>+</sup> cells were capable of generating spheres (<xref ref-type="fig" rid="fig1">Figure 1B</xref>). However, because of difficulties performing lineage-tracing experiments with human samples, we cannot exclude the possibility that centroacinar cells may also contribute to the conversion into endocrine cell lineages.</p><p>While expression of <italic>Pax6</italic> along with <italic>Neurog3</italic>, <italic>Pdx1</italic> and <italic>MafA</italic> significantly enhanced expression of <italic>INS</italic> and other β-cell marker genes in converted ductal cells, this transcription factor combination alone was not sufficient to generate mature IPCs. We found that extending the culture period for 2 weeks after viral infection led to maturation of several hallmark β-cell functions, including expression of key β-cell factors, significant increases of <italic>INS</italic> mRNA and protein levels, proinsulin processing, dense-core granule formation, and Insulin secretion in response to glucose or other depolarizing stimuli. We tested four distinct culture media with or without serum for this extended culture, and all media permitted maturation of these β-cell properties in IPCs (<xref ref-type="fig" rid="fig5s1">Figure 5—figure supplement 1F</xref> and see ‘Materials and methods’), indicating that the duration of culture is a key variable for promoting β-cell maturation in vitro. After maturation, the spheres contained an average of 7% total insulin compared to human islet controls, and 7–11% of cells comprising these spheres produced insulin C-peptide. Thus, we calculate that each reprogrammed Insulin<sup>+</sup> cell produced between 49 and 77% of insulin levels observed in native β-cell controls, a comparable level to the IPCs derived from human ES cells (<xref ref-type="bibr" rid="bib12">D’Amour et al., 2006</xref>).</p><p>Is the capacity of human ductal cells to be converted toward endocrine islet fates unique? A prior study by <xref ref-type="bibr" rid="bib45">Sapir et al. (2005)</xref> suggests that human hepatocytes may be induced to express insulin. However, the conversion toward an insulin-producing fate was comparatively poor; resulting cells produced about 10,000-fold lower insulin mRNA level than that of human islets, about 3–4 orders of magnitude lower than from conversion of pancreatic duct spheres. In addition, characteristic dense core vesicles in converted hepatocytes were not observed, indicating insufficient conversion towards β-cells. Here, we also assessed the endocrine potential of primary human dermal fibroblasts, cells successfully ‘reprogrammed’ toward many non-fibroblast fates, including induced pluripotent stem cells (<xref ref-type="bibr" rid="bib54">Takahashi et al., 2007</xref>), but detected no clear evidence of conversion toward an endocrine or β-cell fate (<xref ref-type="fig" rid="fig5s4">Figure 5—figure supplement 4</xref>, see ‘Materials and methods’ for details). Thus, conversion of human adult duct spheres into cells that produce and secrete insulin is singularly robust. Moreover, unlike prior studies of human ES cells that have high variability among ES cell lines used (<xref ref-type="bibr" rid="bib12">D’Amour et al., 2006</xref>; <xref ref-type="bibr" rid="bib28">Kroon et al., 2008</xref>), we demonstrated conversion toward insulin<sup>+</sup> fates by ductal cells from multiple unrelated donors, another feature of the robustness of our methods.</p><p>Expression of factors produced from viral transgenes persisted in Insulin<sup>+</sup> cells for at least 5 months, evidenced by the GFP expression in transplanted insulin-producing cells (<xref ref-type="fig" rid="fig5s2 fig5s3">Figure 5—figure Supplement 2A and 3</xref>). The transgenes delivered by adenovirus do not generally persist in dividing cells (<xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>). We speculate that cell cycle arrest in Insulin<sup>+</sup> cells may be induced by Neurog3 (<xref ref-type="bibr" rid="bib35">Miyatsuka et al., 2011</xref>), thereby preventing dilution of viral transgene-encoded factors. Thus, further studies are needed to investigate how persistent expression of conversion factors like Neurog3 affects maintenance and maturation of endocrine phenotypes in converted cells. Survival of transplanted insulin-secreting cells produced from ductal cells was poor, and reduced yields following transplantation of ductal cells precluded physiological studies in mouse models of diabetes. Promoting survival of transplanted insulin-secreting cells is a general problem for transplant-based islet replacement approaches. Thus, studies of factors that enhance survival of Insulin<sup>+</sup> ductal cell progeny are an important current focus.</p><p>In conclusion, our study provides unique evidence that primary human cells can generate progeny that produce, store and secrete insulin in response to glucose or depolarizing agents, the hallmark features of pancreatic β-cells. We also show that human pancreatic exocrine cells, like in mice (<xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>), can be converted by transgenes toward an endocrine islet-like cell fate. We speculate that gene-based strategies like those described here may be combined with other methods, including culture modulation by growth factors and small molecules (<xref ref-type="bibr" rid="bib62">Warren et al., 2010</xref>), to optimize endocrine differentiation or conversion of diverse cellular sources to advance cell replacement for diabetes. We speculate that our cell culture system may also serve as the foundation to investigate the genetics and pathogenesis of diverse human diseases rooted in pancreatic ductal cells, including pancreatitis, cystic fibrosis, and adenocarcinoma.</p></sec><sec id="s4" sec-type="materials|methods"><title>Materials and methods</title><sec id="s4-1"><title>Cell preparation</title><p>Institutional review board approval for research use of human tissue was obtained from the Stanford University School of Medicine. Human islet-depleted cell fractions were obtained with appropriate consent from healthy, non-diabetic organ donors deceased due to acute traumatic or anoxic death by overnight shipping from the following facilities: Division of Transplantation (Massachusetts General Hospital, MA), UAB Islet Resource Facility (University of Alabama at Birmingham, AL), UCSF Diabetes Center (University of California, San Francisco, CA), Kidney/pancreas transplantation center (University of Pennsylvania, PA), Islet Core of the University of Pittsburgh (Pittsburgh, PA), and Human Islet Isolation Program (The Hospital of the University of Virginia, VA). Donor samples with the age range 16–63 years (mean 38.24 years) used for this study are listed in <xref ref-type="table" rid="tbl1">Table 1</xref>. On receipt, the cell fractions were washed with PBS and cultured with CMRL media (Mediatech, Inc, Manassas, VA) supplemented with 10% heat inactivated fetal bovine serum (FBS, HyClone, Logan, UT), 2 mM GlutaMax (Life Technologies, Grand Island, NY), 2 mM nicotinamide (prepared in PBS, Sigma, St.Louis, MO), and 100 U Penicillin and 100 µg Streptomycin (Pen/Strep, Life Technologies) in a non-coated culture dish at 25.5°C in 5% CO<sub>2</sub> until use. For dissociation, the cell pellet was washed with PBS, trypsinized with 0.05% Trypsin-EDTA solution (Life Technologies) for 5 min, and quenched with 5 vol of FACS buffer (10 mM EGTA, 2% FBS in PBS). Cells were collected by centrifugation and further digested in 1 U/ml dispase solution (Life Technologies) containing 0.1 mg/ml DNaseI in PBS on a nutating mixer at 37°C for 30 min. PBS washing was performed after each enzymatic digestion step. After centrifugation, the cell pellet was resuspended in FACS buffer and passed through a 40-µm-cell strainer. Cell viability and number were assessed using a Vi-Cell analyzer (Beckman Coulter, Fullerton, CA) and the samples exceeding 70% cell viability were used for subsequent antibody staining for FACS.</p></sec><sec id="s4-2"><title>Cell sorting and culture</title><p>Dissociated cells were stained with biotin-conjugated CD133 antibodies (clone AC133 and 293C3, Miltenyi Biotec, Auburn, CA) and then Allophycocyanin-conjugated Streptavidin (eBioscience, San Diego, CA) for 15 min, each at room temperature. Cell pellets were collected by centrifugation and washed with PBS after each staining steps. Propidium Iodide (Life Technologies) staining was used to exclude dead cells. The cells were sorted using a FACSAria II (BD Biosciences, Bedford, MA) and collected in 100% FBS, washed with PBS twice, and resuspended in ice-cold Advanced DMEM/F-12 media (Life Technologies) at a density of 8000 cells/µl. The average percentage of CD133<sup>+</sup> fraction was 32.73% (n = 32). 50 µl of growth factor-reduced Matrigel (BD Biosciences) was then added to 30 µl cell suspension and the mixture was placed around the bottom rim of each well. After solidification at 37°C for 60 min, each well was overlaid with 500 µl of modified crypt culture media (<xref ref-type="bibr" rid="bib46">Sato et al., 2009</xref>) comprised of Advanced DMEM/F-12 media supplemented with recombinant human (rh) EGF (50 ng/ml, Sigma), rhR-spondin I (500 ng/ml, R&D systems, Minneapolis, MN), rhFGF10 (50 ng/ml, R&D systems), recombinant mouse Noggin (100 ng/ml, R&D systems), 10 mM Nicotinamide in PBS, and Pen/Strep. The media was changed twice weekly. The spheres were harvested after 2 to 3 weeks for passaging or viral infection. Static and time-lapse images of sphere growth were collected using Zeiss Axiovert 200 inverted microscope and Zeiss Observer.Z1 equipped with a temperature- and CO<sub>2</sub>-controlled chamber using Axiovision (Carl Zeiss, Germany) and MetaMorph (Molecular Devices, Sunnyvale, CA) softwares, respectively. For harvesting spheres, 500 µl of 2 U/ml dispase (Life Technologies) solution containing 0.1 mg/ml DNaseI in PBS was added in each well and the Matrigel was mechanically disrupted by pipetting and incubated at 37°C for 45 min. The released spheres were collected, washed twice with PBS and used for subsequent applications. For passaging spheres, the harvested spheres were trypsinized at 37°C for 5 min followed by quenching with FBS. The dispersed cells were then used for cell counting with a hemocytometer or were plated as described above.</p></sec><sec id="s4-3"><title>Construction of adenoviral vectors</title><p>Ad-eGFP and Ad-RFP control adenoviruses were purchased from Vector Biolabs (Philadelphia, PA). Ad-MafA and Ad-Neurog3-IRES-eGFP were described previously (<xref ref-type="bibr" rid="bib55">Tashiro et al., 1999</xref>). To construct Ad-RFP-Neurog3 and Ad-RFP-Pdx1 adenoviruses, mouse cDNAs for Neurog3 (BC104326) and Pdx1 (BC103581) were purchased from Open Biosystems (Lafayette, CO) and the inserts were obtained by restriction enzyme digestion with EcoR V/BamH I and EcoR V/Msc I, respectively. The inserts were then subcloned into multiple cloning sites of Dual-RFP-CCM shuttle vector (Vector Biolabs) and adenoviruses were constructed by Vector Biolabs. For Ad-eGFP-M6P, human <italic>MAFA</italic> cDNA (gift from M German), <italic>PDX1</italic> (NM_000209; GeneCopoeia, Rockville, MD), and <italic>PAX6</italic> (BC011953; Open Biosystems) were used for PCR amplification with the primers shown in <xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1C</xref> to add T2A, P2A, restriction enzyme sites, and/or tagging proteins (<xref ref-type="fig" rid="fig5s5">Figure 5—figure supplement 5</xref>). A fused construct of MAFA-T2A-PAX6 was generated by PCR with <italic>MAFA</italic> and <italic>PAX6</italic> PCR amplicons as templates. Similarly, PCR products for PAX6 and PDX1 were used to construct PAX6-P2A-PDX1. Next, MAFA-T2A-PAX6, PAX6-P2A-PDX1, and pDual-GFP-CCM vector (Vector Biolabs) were cut with BglII/PstI, PstI/EcoRI, and BglII/EcoRI, respectively, and ligated with NEB quick ligation kit (New England Biolabs, Ipswich, MA) followed by transformation of TOP10 chemically competent cells (Invitrogen, Carlsbad, CA). The construct was then used for generating adenoviruses by Vector Biolabs.</p></sec><sec id="s4-4"><title>Sphere infection and post-infection culture</title><p>Spheres were infected at 37°C in suspension overnight at a multiplicity of infection (MOI) 100 for Ad-MafA and Ad-eGFP-M6P, or MOI 500 for the rest of viruses used. The spheres were then washed twice with culture medium and embedded in Matrigel as described above. The infected spheres were overlayed with sphere growth media without R-spondin I and with 0.33 µM all-trans retinoic acid (Sigma), and cultured for 5 days. For extended culture, the media was replaced with either (1) DMEM with high glucose (Life Technologies) supplemented with 10% FBS (Hyclone) and Pen/Strep (Life Technologies) for 2 weeks (referred as ‘DF’ in <xref ref-type="fig" rid="fig3s1">Figure 3—figure supplement 1F</xref>), (2) DF plus 20 mM KCl and 10 µM R0-28-1675 (glucokinase activator; Axon Ligands) for 2 weeks (referred as ‘DFK’), (3) DF for one week and then DMEM/F-12 media (Life Technologies) supplemented with 0.5 × N2 supplement (Life technologies), 0.5 × B27 (Life technologies), 0.2% BSA (Sigma), 1% ITS supplement (Life Technologies), 10 mM nicotinamide, 10 ng/ml recombinant human basic FGF (R&D systems), 50 ng/ml Exendin-4 (R&D systems), recombinant human BMP-4 (R&D systems) for additional 1 week (referred as ‘Z’; <xref ref-type="bibr" rid="bib66">Zhang et al., 2009</xref>), or (4) DMEM high glucose supplemented with 1 × B27, 55 nM GLP-1, 50 ng FGF10 (R&D Systems), and Pen/Strep for 3 days followed by 5 days with DMEM high glucose supplemented with 1 × B27, 55 nM GLP-1 (Sigma), 10 µM DAPT (Sigma), and Pen/Strep, then for 6 days with CMRL1066 media (Mediatech) supplemented with 1 × B27, 55 nM GLP-1, 50 ng HGF (R&D Systems), 50 ng IGF-1 (R&D Systems), and Pen/Strep (referred as ‘T’; <xref ref-type="bibr" rid="bib55a">Thatava et al., 2011</xref>). The media was replaced every other day unless otherwise noted.</p></sec><sec id="s4-5"><title>cDNA preparation and qRT-PCR analyses</title><p>Total RNA was prepared from sorted cells or cultured spheres with QIAGEN RNeasy micro kit (QIAGEN Sciences, MD), and used for cDNA synthesis using QIAGEN Omniscript RT kit (QIAGEN), according to the manufacturer’s protocol. Relative mRNA level was measured by qRT-PCR of each cDNA in duplicate with gene-specific probe sets (Applied Biosystems, Foster City, CA) with TaqMan Universal PCR Master Mix (Applied Biosystems) and the ABI Prism 7500 detection system (Applied Biosystems). Normalizations across samples were performed using β-actin primers. Information of the primer and probe sets is available upon request.</p></sec><sec id="s4-6"><title>Immunohistochemistry</title><p>For immunohistochemical analyses, cultured spheres were harvested, washed with PBS, mixed with 20 µl of Collagen Gel Kit (Nitta Gelatin, Osaka, Japan), solidified at 37°C for 1 hr, fixed with 4% paraformaldehyde for 2 hr at 4°C, cryoprotected in 30% sucrose solution in PBS overnight, embedded in OCT on dry ice, and sectioned in 8 µm thickness. For sorted cells, the cell suspension was washed once and resuspended with 20 µl of PBS, placed on a Polysine slide (Thermo scientific, Waltham, MA), and waited for 30 min at room temperature (RT) to let the cells sit on the slide glass by gravity. Then the solution was removed carefully and 40 µl of 4% paraformaldehyde was added. After 10 min of incubation at RT, the fixative was removed and the slides were washed with PBS three times for 5 min each. After removal of PBS, the slides were dried at RT for 1 hr and stored at −20°C. For immunostaining transplanted IPCs, grafted organs (kidney, EFP, or liver) were harvested, fixed with 4% paraformaldehyde overnight at 4°C, cryoprotected in 30% sucrose solution in PBS overnight, embedded in OCT on dry ice, and sectioned in 8 µm (kidney and liver) or 40 µm (EFP) thickness. The primary antibodies used were rabbit anti-Amylase (1:1000; Sigma), goat anti-Amylase (sc-12821; 1:200; Santa Cruz Biotechnology, Dallas, TX), CD133 (1:100 each; clone AC133 and 293C3; Miltenyi Biotec, Auburn, CA), rabbit anti-ChromograninA (20085; 1:100; Immunostar, Hudson, WI), mouse anti-ChromograninA (LK2H10; 1:200; Cell Marque, Rocklin, CA), mouse anti-CK19 (KRT19) (M0888; 1:200; DAKO, Carpinteria, CA), rabbit anti-CK19 (319R-15; 1:200; Cell Marque), rabbit anti-CPA1 (1810-0006; 1:100; AbD Serotec, UK), rabbit anti-C-peptide (#4593B; 1:200; Cell Signaling Technology, Danvers, MA), mouse anti-C-peptide (capt) (1:100; Mercodia, Sweden), mouse anti-Flag (F1804; 1:1000; Sigma), goat anti-GHRL (sc-10368; 1:200; Santa Cruz Biotechnology), guinea pig anti-Glucagon (4031-01; 1:200; Linco, Billerica, MA), mouse anti-HA (MMS-101P-1000; 1:1000; Covance), mouse anti-HuNu (MAB1281; 1:200; Millipore, Billerica, MA), mouse anti-IAPP (MCA1126T; 1:200; AbD serotec), rabbit anti-Ki-67 (NCL-Ki67p; 1:100, Leica Microsystems, Germany), rabbit anti-Myc (sc-789; 1:1000; Santa Cruz Biotechnology), mouse anti-NeuroD (sc-46684; 1:10; Santa Cruz Biotechnology), mouse anti-Neurog3 (F25A1B3; 1:4000; DSHB, Iowa City, IA), mouse anti-Nkx2.2 (74.5A5; 1:10; DSHB), mouse anti-Nkx6.1 (F55A10; 1:200; DSHB), rabbit anti-PC1/3 (PCSK1, AB10553; 1:200; Millipore), rabbit anti-phospho-H3 (06-570; 1:500; Millipore), goat anti-PPY (NB100-1793; 1:200; Novus Biologicals, Littleton, CO), rabbit anti-Somatostatin (1:200, DAKO), goat anti-Somatostatin (sc-7819; 1:200; Santa Cruz Biotechnology), goat anti-SUR-1 (sc-5789; 1:50; Santa Cruz Biotechnology). Tyramide signal amplification (Perkin Elmer, Waltham, MA) was used for antibodies against Neurog3, NeuroD, Nkx2.2, Nkx6.1, and PC1/3. Antigen unmasking (H-3300; Antigen Unmasking Solution, Citric Acid Based, Vector Laboratories, Burlingame, CA) was performed for anti-Flag antibody staining. The Neurog3, Nkx2.2, and Nkx6.1 antibodies developed by Dr OD Madsen were obtained from the Developmental Studies Hybridoma Bank (DSHB) developed under the auspices of the NICHD and maintained by The University of Iowa, Department of Biological Sciences, Iowa City, IA 52242. Secondary antibodies used were from Jackson ImmunoResearch (West Grove, PA) or Molecular Probes (Eugene, OR). Stained sections were mounted with VECTASHIELD Mounting Medium with DAPI (Vector Laboratories). Fluorescence images were taken using Zeiss Axio Imager.M1 or Leica SP2 inverted confocal laser scanning microscope.</p></sec><sec id="s4-7"><title>Electron microscopy</title><p>The samples were fixed in Karnovsky’s fixative: 2% Glutaraldehyde (EMS Cat# 16000) and 4% Paraformaldehyde (EMS; Electron Microscopy Sciences, Hatfield, PA) in 0.1 M Sodium Cacodylate (EMS) pH 7.4 for 1 hr at RT then cut, post fixed in 1% Osmium tetroxide (EMS) for 1 hr at RT, washed three times with ultrafiltered water, then en bloc stained for 2 hr at RT or moved to 4°C overnight. The samples were then dehydrated in a series of ethanol washes for 15 min each at 4°C beginning at 50%, 70%, 95%, where the samples are then allowed to rise to RT, changed to 100% two times, followed by Acetonitrile for 15 min. The samples are infiltrated with EMbed-812 resin (EMS) mixed 1:1 with Acetonitrile for 2 hr followed by two parts EMbed-812 to 1 part Acetonitrile for 2 hr. The samples were then placed into EMbed-812 for 2 hr and then placed into molds, and resin filled gelatin capsules with labels were orientated over the cells of interest and placed into 65°C oven overnight. Sections were taken between 75 and 90 nm on a Leica Ultracut S (Leica, Wetzlar, Germany), picked up on formvar/Carbon coated slot grids (EMS Cat#FCF2010-Cu) or 100 mesh Cu grids (EMS). Grids were contrast stained for 15 min in 1:1 saturated UrAcetate (∼7.7%) to 100% ethanol followed by staining in 0.2% lead citrate for 3 to 4 min. JEOL JEM-1400 TEM was used to observe at 120 kV and photos were taken using a Gatan Orius digital camera.</p></sec><sec id="s4-8"><title>C-peptide secretion and content measurement</title><p>C-peptide secretion assay and content measurement were performed as described previously with minor modification (<xref ref-type="bibr" rid="bib9">Chen et al., 2001</xref>). Briefly for secretion assay, media was replaced a day before assay was performed. On the day, each well with matrigel-embedded spheres was incubated with fresh media for 2 hr, washed twice with plain Krebs-Ringer bicarbonate buffer (KRBB), and incubated twice with plain KRBB for 1 hr each for thorough washing. Next, the spheres were incubated consecutively with 400 µl KRBB containing indicated concentrations of glucose (Sigma) with or without 0.5 mM Diazoxide (Sigma), KCl (30 mM, Sigma), or Tolbutamide (0.2 mM, Sigma) for 2 hr each. KRBB without Calcium (No Ca<sup>++</sup>) was prepared by omission of CaCl<sub>2</sub> and addition of 1 mM EGTA (Sigma). Secreted C-peptide level was measured with Human Ultrasensitive C-peptide ELISA kit (Mercodia). For C-peptide content measurement, the spheres were harvested in 1.5 ml microfuge tube, washed with PBS, resuspended with 300 µl of ice-cold TE/BSA buffer (10 mM Tris-HCl, 1 mM EDTA, 0.1% wt/vol BSA, pH 7.0), and sonicated with Bioruptor Sonicator (Diagenode, Denville, NJ). Half of the lysate was used for genomic DNA isolation and quantification with Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen). Same volume of acid alcohol (75% vol/vol ethanol, 2% vol/vol concentrated HCl, 23% vol/vol H<sub>2</sub>O) was added to the rest of lysate to extract C-peptide by rocking overnight at 4°C. The extract was then neutralized with 10 vol of PBS and used for C-peptide ELISA.</p></sec><sec id="s4-9"><title>Transplantation</title><p>Transplantation in kidney capsule, epididymal fat pad (EFP), or in the liver by portal vein injection was performed as previously described (<xref ref-type="bibr" rid="bib28">Kroon et al., 2008</xref>; <xref ref-type="bibr" rid="bib2">Alipio et al., 2010</xref>; <xref ref-type="bibr" rid="bib59">Wang et al., 2011</xref>). For transplantation in kidney or EFP, converted spheres with or without extended culture were harvested and mixed with or without mouse embryonic fibroblasts (<xref ref-type="supplementary-material" rid="SD1-data">Supplementary file 1A</xref>). The spheres were then mixed with matrigel to make a final volume of 10 µl for kidney transplantation or overlayed on pre-wet gelfoam for EFP transplantation. For liver transplantation, single cells produced by trypsinization of harvested spheres were resuspended in 100 µl PBS and injected into the portal vein with a 27 G needle. All animal experiments and methods were approved by the Institutional Animal Care and Use Committee (IACUC) of Stanford University.</p></sec><sec id="s4-10"><title>In vivo glucose-stimulated insulin secretion assay</title><p>Secretion of human Insulin or C-peptide by glucose injection was measured as previously described (<xref ref-type="bibr" rid="bib28">Kroon et al., 2008</xref>). Briefly, transplanted mice were fasted overnight (14–16 hr) and 120 µl of blood was collected from tail into Microvette CB300LH (Sarstedt, Germany) to prepare 50 µl of serum. 3 g/kg glucose was then injected and blood was collected again 30 min after glucose administration. Secreted C-peptide or insulin level was measured with Human Ultrasensitive C-peptide or Insulin ELISA kits (Mercodia).</p></sec><sec id="s4-11"><title>Human adult dermal fibroblast culture and conversion assay</title><p>Human adult dermal fibroblasts (Coriell Institute for Medical Research, Camden, New Jersey, USA) were cultured and maintained as described previously (<xref ref-type="bibr" rid="bib65">Yoo et al., 2011</xref>). The cells were either trypsinized for suspension infection (as was described above for ductal spheres) or infected as adherent cells in six-well plates by direct addition of virus into the culture medium, with Ad-eGFP (GFP) or Ad-eGFP-M6P and Ad-Neurog3-IRES-eGFP (4TFM). The same MOIs used for ductal sphere infection were also used. The suspension-infected cells were harvested the following day and embedded in Matrigel as described above for infected ductal spheres. The culture was maintained for additional 18 days to match the duration of infected ductal sphere maturation. The infected adherent cells were cultured with virus for 48 hr and the media was replaced. The culture was maintained for additional 10 days, passaged in 1:3 ratio due to confluency, re-plated, and cultured additional 7 days to match the duration of infected ductal sphere maturation. In both cases, media was replaced every other day. Three independent experiments were performed for both conditions and each experiment at least in duplicates. RNA isolation, cDNA preparation, and qRT-PCR were performed with primers specific to human <italic>INS</italic>, <italic>CHGA</italic>, and β-actin as described above.</p></sec></sec></body><back><ack id="ack"><title>Acknowledgements</title><p>We thank Drs H-E Hohmeier and Christopher Newgard (Duke University) for helpful discussions and advice, Dr M German (UCSF) for human <italic>MAFA</italic> cDNA, Drs K D’Amour and E Kroon (Viacyte) for advice on EFP transplantation, Ms S Bryant (University of Alabama), Drs A Naji and C Liu (University of Pennsylvania), Dr X Huang (University of Virginia) for human non-islet cell processing, Mr J Perrino and the Stanford Cell Sciences Imaging Facility for transmission electron microscopy and confocal microscopy, Ms E Snyder and Mr J Albright for technical support, and members of the Kim Laboratory for comments on the manuscript.</p></ack><sec sec-type="additional-information"><title>Additional information</title><fn-group content-type="competing-interest"><title>Competing interests</title><fn fn-type="conflict" id="conf1"><p>The authors declare that no competing interests exist.</p></fn></fn-group><fn-group content-type="author-contribution"><title>Author contributions</title><fn fn-type="con" id="con1"><p>JL, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article</p></fn><fn fn-type="con" id="con2"><p>TS, Contributed initial conception and execution of pancreatic sphere culture</p></fn><fn fn-type="con" id="con3"><p>YL, Contributed cell sorting and ductal sphere culture</p></fn><fn fn-type="con" id="con4"><p>JW, Renal subcapsular transplantation and human tissue assessment</p></fn><fn fn-type="con" id="con5"><p>XG, Contributed immunohistochemical and morphometric studies</p></fn><fn fn-type="con" id="con6"><p>JL, Procurement of cadaveric human pancreas and islet-depleted cell preparation</p></fn><fn fn-type="con" id="con7"><p>JFM, Procurement of cadaveric human pancreas and islet-depleted cell preparation</p></fn><fn fn-type="con" id="con8"><p>GLS, Procurement of cadaveric human pancreas and islet-depleted cell preparation</p></fn><fn fn-type="con" id="con9"><p>RB, Procurement of cadaveric human pancreas and islet-depleted cell preparation</p></fn><fn fn-type="con" id="con10"><p>SM, Provision of Ad-MafA and Ad-Neurog3-IRES-eGFP</p></fn><fn fn-type="con" id="con11"><p>J-iM, Provision of Ad-MafA and Ad-Neurog3-IRES-eGFP</p></fn><fn fn-type="con" id="con12"><p>SKK, Conception and design, Analysis and interpretation of data, Drafting or revising the article</p></fn></fn-group><fn-group content-type="ethics-information"><title>Ethics</title><fn fn-type="other"><p>Animal experimentation: This study was performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All of the animals were handled according to approved institutional animal care and use committee (IACUC) protocols (#10160) of the Stanford University. All surgery was performed under anesthesia, and every effort was made to minimize suffering.</p></fn></fn-group></sec><sec sec-type="supplementary-material"><title>Additional files</title><supplementary-material id="SD1-data"><object-id pub-id-type="doi">10.7554/eLife.00940.018</object-id><label>Supplementary file 1.</label><caption><p>(<bold>A</bold>) Summary table of converted sphere transplantation. (<bold>B</bold>) Glucose-stimulated insulin C-peptide secretion in vivo. (<bold>C</bold>) PCR primers used for Ad-GFP-M6P construction. (<bold>D</bold>) Quantification data of cell immunostaining after FACS (<xref ref-type="fig" rid="fig1">Figure 1F</xref>).</p><p><bold>DOI:</bold> <ext-link ext-link-type="doi" xlink:href="10.7554/eLife.00940.018">http://dx.doi.org/10.7554/eLife.00940.018</ext-link></p></caption><media mime-subtype="doc" mimetype="application" xlink:href="elife00940s001.doc"/></supplementary-material></sec><ref-list><title>References</title><ref id="bib1"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Al-Hasani</surname><given-names>K</given-names></name><name><surname>Pfeifer</surname><given-names>A</given-names></name><name><surname>Courtney</surname><given-names>M</given-names></name><name><surname>Ben-Othman</surname><given-names>N</given-names></name><name><surname>Gjernes</surname><given-names>E</given-names></name><name><surname>Vieira</surname><given-names>A</given-names></name><etal/></person-group><year>2013</year><article-title>Adult duct-lining cells can reprogram into beta-like cells able to counter repeated cycles of toxin-induced diabetes</article-title><source>Dev Cell</source><volume>26</volume><fpage>86</fpage><lpage>100</lpage><pub-id pub-id-type="doi">10.1016/j.devcel.2013.05.018</pub-id></element-citation></ref><ref id="bib2"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Alipio</surname><given-names>Z</given-names></name><name><surname>Liao</surname><given-names>W</given-names></name><name><surname>Roemer</surname><given-names>EJ</given-names></name><name><surname>Waner</surname><given-names>M</given-names></name><name><surname>Fink</surname><given-names>LM</given-names></name><name><surname>Ward</surname><given-names>DC</given-names></name><etal/></person-group><year>2010</year><article-title>Reversal of hyperglycemia in diabetic mouse models using induced-pluripotent stem (iPS)-derived pancreatic beta-like cells</article-title><source>Proc Natl Acad Sci USA</source><volume>107</volume><fpage>13426</fpage><lpage>31</lpage><pub-id pub-id-type="doi">10.1073/pnas.1007884107</pub-id></element-citation></ref><ref id="bib3"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Ashcroft</surname><given-names>FM</given-names></name><name><surname>Rorsman</surname><given-names>P</given-names></name></person-group><year>2012</year><article-title>Diabetes mellitus and the beta cell: the last ten years</article-title><source>Cell</source><volume>148</volume><fpage>1160</fpage><lpage>71</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2012.02.010</pub-id></element-citation></ref><ref id="bib4"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Benitez</surname><given-names>CM</given-names></name><name><surname>Goodyer</surname><given-names>WR</given-names></name><name><surname>Kim</surname><given-names>SK</given-names></name></person-group><year>2012</year><article-title>Deconstructing pancreas developmental biology</article-title><source>Cold Spring Harbor Perspect Biol</source><volume>4</volume><fpage>a012401</fpage><pub-id pub-id-type="doi">10.1101/cshperspect.a012401</pub-id></element-citation></ref><ref id="bib5"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Blum</surname><given-names>B</given-names></name><name><surname>Hrvatin</surname><given-names>SS</given-names></name><name><surname>Schuetz</surname><given-names>C</given-names></name><name><surname>Bonal</surname><given-names>C</given-names></name><name><surname>Rezania</surname><given-names>A</given-names></name><name><surname>Melton</surname><given-names>DA</given-names></name></person-group><year>2012</year><article-title>Functional beta-cell maturation is marked by an increased glucose threshold and by expression of urocortin 3</article-title><source>Nat Biotechnol</source><volume>30</volume><fpage>261</fpage><lpage>4</lpage><pub-id pub-id-type="doi">10.1038/nbt.2141</pub-id></element-citation></ref><ref id="bib6"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bonner-Weir</surname><given-names>S</given-names></name><name><surname>Taneja</surname><given-names>M</given-names></name><name><surname>Weir</surname><given-names>GC</given-names></name><name><surname>Tatarkiewicz</surname><given-names>K</given-names></name><name><surname>Song</surname><given-names>KH</given-names></name><name><surname>Sharma</surname><given-names>A</given-names></name><etal/></person-group><year>2000</year><article-title>In vitro cultivation of human islets from expanded ductal tissue</article-title><source>Proc Natl Acad Sci USA</source><volume>97</volume><fpage>7999</fpage><lpage>8004</lpage><pub-id pub-id-type="doi">10.1073/pnas.97.14.7999</pub-id></element-citation></ref><ref id="bib7"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bonner-Weir</surname><given-names>S</given-names></name><name><surname>Toschi</surname><given-names>E</given-names></name><name><surname>Inada</surname><given-names>A</given-names></name><name><surname>Reitz</surname><given-names>P</given-names></name><name><surname>Fonseca</surname><given-names>SY</given-names></name><name><surname>Aye</surname><given-names>T</given-names></name><etal/></person-group><year>2004</year><article-title>The pancreatic ductal epithelium serves as a potential pool of progenitor cells</article-title><source>Pediatr Diabetes</source><volume>5</volume><fpage>16</fpage><lpage>22</lpage><pub-id pub-id-type="doi">10.1111/j.1399-543X.2004.00075.x</pub-id></element-citation></ref><ref id="bib8"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Bouwens</surname><given-names>L</given-names></name><name><surname>Pipeleers</surname><given-names>DG</given-names></name></person-group><year>1998</year><article-title>Extra-insular beta cells associated with ductules are frequent in adult human pancreas</article-title><source>Diabetologia</source><volume>41</volume><fpage>629</fpage><lpage>33</lpage><pub-id pub-id-type="doi">10.1007/s001250050960</pub-id></element-citation></ref><ref id="bib9"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Chen</surname><given-names>H</given-names></name><name><surname>Carlson</surname><given-names>EC</given-names></name><name><surname>Pellet</surname><given-names>L</given-names></name><name><surname>Moritz</surname><given-names>JT</given-names></name><name><surname>Epstein</surname><given-names>PN</given-names></name></person-group><year>2001</year><article-title>Overexpression of metallothionein in pancreatic beta-cells reduces streptozotocin-induced DNA damage and diabetes</article-title><source>Diabetes</source><volume>50</volume><fpage>2040</fpage><lpage>6</lpage><pub-id pub-id-type="doi">10.2337/diabetes.50.9.2040</pub-id></element-citation></ref><ref id="bib10"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Cheng</surname><given-names>X</given-names></name><name><surname>Ying</surname><given-names>L</given-names></name><name><surname>Lu</surname><given-names>L</given-names></name><name><surname>Galvao</surname><given-names>AM</given-names></name><name><surname>Mills</surname><given-names>JA</given-names></name><name><surname>Lin</surname><given-names>HC</given-names></name><etal/></person-group><year>2012</year><article-title>Self-renewing endodermal progenitor lines generated from human pluripotent stem cells</article-title><source>Cell Stem Cell</source><volume>10</volume><fpage>371</fpage><lpage>84</lpage><pub-id pub-id-type="doi">10.1016/j.stem.2012.02.024</pub-id></element-citation></ref><ref id="bib11"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Cleveland</surname><given-names>MH</given-names></name><name><surname>Sawyer</surname><given-names>JM</given-names></name><name><surname>Afelik</surname><given-names>S</given-names></name><name><surname>Jensen</surname><given-names>J</given-names></name><name><surname>Leach</surname><given-names>SD</given-names></name></person-group><year>2012</year><article-title>Exocrine ontogenies: on the development of pancreatic acinar, ductal and centroacinar cells</article-title><source>Semin Cell Dev Biol</source><volume>23</volume><fpage>711</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1016/j.semcdb.2012.06.008</pub-id></element-citation></ref><ref id="bib12"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>D’Amour</surname><given-names>KA</given-names></name><name><surname>Bang</surname><given-names>AG</given-names></name><name><surname>Eliazer</surname><given-names>S</given-names></name><name><surname>Kelly</surname><given-names>OG</given-names></name><name><surname>Agulnick</surname><given-names>AD</given-names></name><name><surname>Smart</surname><given-names>NG</given-names></name><etal/></person-group><year>2006</year><article-title>Production of pancreatic hormone-expressing endocrine cells from human embryonic stem cells</article-title><source>Nat Biotechnol</source><volume>24</volume><fpage>1392</fpage><lpage>401</lpage><pub-id pub-id-type="doi">10.1038/nbt1259</pub-id></element-citation></ref><ref id="bib13"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Fujikawa</surname><given-names>T</given-names></name><name><surname>Oh</surname><given-names>SH</given-names></name><name><surname>Pi</surname><given-names>L</given-names></name><name><surname>Hatch</surname><given-names>HM</given-names></name><name><surname>Shupe</surname><given-names>T</given-names></name><name><surname>Petersen</surname><given-names>BE</given-names></name></person-group><year>2005</year><article-title>Teratoma formation leads to failure of treatment for type I diabetes using embryonic stem cell-derived insulin-producing cells</article-title><source>Am J Pathol</source><volume>166</volume><fpage>1781</fpage><lpage>91</lpage><pub-id pub-id-type="doi">10.1016/S0002-9440(10)62488-1</pub-id></element-citation></ref><ref id="bib14"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Furuyama</surname><given-names>K</given-names></name><name><surname>Kawaguchi</surname><given-names>Y</given-names></name><name><surname>Akiyama</surname><given-names>H</given-names></name><name><surname>Horiguchi</surname><given-names>M</given-names></name><name><surname>Kodama</surname><given-names>S</given-names></name><name><surname>Kuhara</surname><given-names>T</given-names></name><etal/></person-group><year>2011</year><article-title>Continuous cell supply from a Sox9-expressing progenitor zone in adult liver, exocrine pancreas and intestine</article-title><source>Nat Genet</source><volume>43</volume><fpage>34</fpage><lpage>41</lpage><pub-id pub-id-type="doi">10.1038/ng.722</pub-id></element-citation></ref><ref id="bib14a"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gao</surname><given-names>R</given-names></name><name><surname>Ustinov</surname><given-names>J</given-names></name><name><surname>Korsgren</surname><given-names>O</given-names></name><name><surname>Otonkoski</surname><given-names>T</given-names></name></person-group><year>2005</year><article-title>In vitro neogenesis of human islets reflects the plasticity of differentiated human pancreatic cells</article-title><source>Diabetologia</source><volume>48</volume><fpage>2296</fpage><lpage>2304</lpage><pub-id pub-id-type="doi">10.1007/s00125-005-1935-8</pub-id></element-citation></ref><ref id="bib15"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gradwohl</surname><given-names>G</given-names></name><name><surname>Dierich</surname><given-names>A</given-names></name><name><surname>LeMeur</surname><given-names>M</given-names></name><name><surname>Guillemot</surname><given-names>F</given-names></name></person-group><year>2000</year><article-title>neurogenin3 is required for the development of the four endocrine cell lineages of the pancreas</article-title><source>Proc Natl Acad Sci USA</source><volume>97</volume><fpage>1607</fpage><lpage>11</lpage><pub-id pub-id-type="doi">10.1073/pnas.97.4.1607</pub-id></element-citation></ref><ref id="bib16"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Gu</surname><given-names>G</given-names></name><name><surname>Dubauskaite</surname><given-names>J</given-names></name><name><surname>Melton</surname><given-names>DA</given-names></name></person-group><year>2002</year><article-title>Direct evidence for the pancreatic lineage: NGN3+ cells are islet progenitors and are distinct from duct progenitors</article-title><source>Development</source><volume>129</volume><fpage>2447</fpage><lpage>57</lpage></element-citation></ref><ref id="bib17"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hanna</surname><given-names>J</given-names></name><name><surname>Saha</surname><given-names>K</given-names></name><name><surname>Pando</surname><given-names>B</given-names></name><name><surname>van Zon</surname><given-names>J</given-names></name><name><surname>Lengner</surname><given-names>CJ</given-names></name><name><surname>Creyghton</surname><given-names>MP</given-names></name><etal/></person-group><year>2009</year><article-title>Direct cell reprogramming is a stochastic process amenable to acceleration</article-title><source>Nature</source><volume>462</volume><fpage>595</fpage><lpage>601</lpage><pub-id pub-id-type="doi">10.1038/nature08592</pub-id></element-citation></ref><ref id="bib18"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hao</surname><given-names>E</given-names></name><name><surname>Tyrberg</surname><given-names>B</given-names></name><name><surname>Itkin-Ansari</surname><given-names>P</given-names></name><name><surname>Lakey</surname><given-names>JR</given-names></name><name><surname>Geron</surname><given-names>I</given-names></name><name><surname>Monosov</surname><given-names>EZ</given-names></name><etal/></person-group><year>2006</year><article-title>Beta-cell differentiation from nonendocrine epithelial cells of the adult human pancreas</article-title><source>Nat Med</source><volume>12</volume><fpage>310</fpage><lpage>6</lpage><pub-id pub-id-type="doi">10.1038/nm1367</pub-id></element-citation></ref><ref id="bib19"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hebrok</surname><given-names>M</given-names></name></person-group><year>2012</year><article-title>Generating beta cells from stem cells-the story so far</article-title><source>Cold Spring Harbor Perspect Med</source><volume>2</volume><fpage>a007674</fpage><pub-id pub-id-type="doi">10.1101/cshperspect.a007674</pub-id></element-citation></ref><ref id="bib20"><element-citation publication-type="book"><person-group person-group-type="author"><name><surname>Henquin</surname><given-names>J-C</given-names></name></person-group><year>2005</year><article-title>Cell biology of insulin secretion</article-title><person-group person-group-type="editor"><name><surname>Kahn</surname><given-names>CR</given-names></name><name><surname>Weir</surname><given-names>GC</given-names></name><name><surname>King</surname><given-names>GL</given-names></name><name><surname>Jacobsen</surname><given-names>AM</given-names></name><name><surname>Moses</surname><given-names>AC</given-names></name><etal/></person-group><source>Joslin’s diabetes mellitus</source><publisher-loc>Boston, MA, USA</publisher-loc><publisher-name>Lippincott Williams & Wilkins</publisher-name><fpage>83</fpage><lpage>107</lpage></element-citation></ref><ref id="bib21"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Heremans</surname><given-names>Y</given-names></name><name><surname>Van De Casteele</surname><given-names>M</given-names></name><name><surname>in’t Veld</surname><given-names>P</given-names></name><name><surname>Gradwohl</surname><given-names>G</given-names></name><name><surname>Serup</surname><given-names>P</given-names></name><name><surname>Madsen</surname><given-names>O</given-names></name><etal/></person-group><year>2002</year><article-title>Recapitulation of embryonic neuroendocrine differentiation in adult human pancreatic duct cells expressing neurogenin 3</article-title><source>J Cell Biol</source><volume>159</volume><fpage>303</fpage><lpage>12</lpage><pub-id pub-id-type="doi">10.1083/jcb.200203074</pub-id></element-citation></ref><ref id="bib22"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Hoesli</surname><given-names>CA</given-names></name><name><surname>Johnson</surname><given-names>JD</given-names></name><name><surname>Piret</surname><given-names>JM</given-names></name></person-group><year>2012</year><article-title>Purified human pancreatic duct cell culture conditions defined by serum-free high-content growth factor screening</article-title><source>PLOS ONE</source><volume>7</volume><fpage>e33999</fpage><pub-id pub-id-type="doi">10.1371/journal.pone.0033999</pub-id></element-citation></ref><ref id="bib23"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Immervoll</surname><given-names>H</given-names></name><name><surname>Hoem</surname><given-names>D</given-names></name><name><surname>Sakariassen</surname><given-names>PO</given-names></name><name><surname>Steffensen</surname><given-names>OJ</given-names></name><name><surname>Molven</surname><given-names>A</given-names></name></person-group><year>2008</year><article-title>Expression of the “stem cell marker” CD133 in pancreas and pancreatic ductal adenocarcinomas</article-title><source>BMC Cancer</source><volume>8</volume><fpage>48</fpage><pub-id pub-id-type="doi">10.1186/1471-2407-8-48</pub-id></element-citation></ref><ref id="bib24"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Inada</surname><given-names>A</given-names></name><name><surname>Nienaber</surname><given-names>C</given-names></name><name><surname>Katsuta</surname><given-names>H</given-names></name><name><surname>Fujitani</surname><given-names>Y</given-names></name><name><surname>Levine</surname><given-names>J</given-names></name><name><surname>Morita</surname><given-names>R</given-names></name><etal/></person-group><year>2008</year><article-title>Carbonic anhydrase II-positive pancreatic cells are progenitors for both endocrine and exocrine pancreas after birth</article-title><source>Proc Natl Acad Sci USA</source><volume>105</volume><fpage>19915</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1073/pnas.0805803105</pub-id></element-citation></ref><ref id="bib25"><element-citation publication-type="book"><person-group person-group-type="author"><name><surname>Klimstra</surname><given-names>DS</given-names></name><name><surname>Hruban</surname><given-names>RH</given-names></name><name><surname>Pitman</surname><given-names>MB</given-names></name></person-group><year>2007</year><article-title>VII - Alimentary Tract, 30 - Pancreas</article-title><person-group person-group-type="editor"><name><surname>Mills</surname><given-names>SE</given-names></name></person-group><source>Histology for Pathologists</source><publisher-loc>Philadelphia</publisher-loc><publisher-name>Lippincott Williams & Wilkins</publisher-name><fpage>723</fpage><lpage>60</lpage></element-citation></ref><ref id="bib26"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Koblas</surname><given-names>T</given-names></name><name><surname>Pektorova</surname><given-names>L</given-names></name><name><surname>Zacharovova</surname><given-names>K</given-names></name><name><surname>Berkova</surname><given-names>Z</given-names></name><name><surname>Girman</surname><given-names>P</given-names></name><name><surname>Dovolilova</surname><given-names>E</given-names></name><etal/></person-group><year>2008</year><article-title>Differentiation of CD133-positive pancreatic cells into insulin-producing islet-like cell clusters</article-title><source>Transplant Proc</source><volume>40</volume><fpage>415</fpage><lpage>8</lpage><pub-id pub-id-type="doi">10.1016/j.transproceed.2008.02.017</pub-id></element-citation></ref><ref id="bib27"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kopp</surname><given-names>JL</given-names></name><name><surname>Dubois</surname><given-names>CL</given-names></name><name><surname>Schaffer</surname><given-names>AE</given-names></name><name><surname>Hao</surname><given-names>E</given-names></name><name><surname>Shih</surname><given-names>HP</given-names></name><name><surname>Seymour</surname><given-names>PA</given-names></name><etal/></person-group><year>2011</year><article-title>Sox9+ ductal cells are multipotent progenitors throughout development but do not produce new endocrine cells in the normal or injured adult pancreas</article-title><source>Development</source><volume>138</volume><fpage>653</fpage><lpage>65</lpage><pub-id pub-id-type="doi">10.1242/dev.056499</pub-id></element-citation></ref><ref id="bib28"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Kroon</surname><given-names>E</given-names></name><name><surname>Martinson</surname><given-names>LA</given-names></name><name><surname>Kadoya</surname><given-names>K</given-names></name><name><surname>Bang</surname><given-names>AG</given-names></name><name><surname>Kelly</surname><given-names>OG</given-names></name><name><surname>Eliazer</surname><given-names>S</given-names></name><etal/></person-group><year>2008</year><article-title>Pancreatic endoderm derived from human embryonic stem cells generates glucose-responsive insulin-secreting cells in vivo</article-title><source>Nat Biotechnol</source><volume>26</volume><fpage>443</fpage><lpage>52</lpage><pub-id pub-id-type="doi">10.1038/nbt1393</pub-id></element-citation></ref><ref id="bib29"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lardon</surname><given-names>J</given-names></name><name><surname>Corbeil</surname><given-names>D</given-names></name><name><surname>Huttner</surname><given-names>WB</given-names></name><name><surname>Ling</surname><given-names>Z</given-names></name><name><surname>Bouwens</surname><given-names>L</given-names></name></person-group><year>2008</year><article-title>Stem cell marker prominin-1/AC133 is expressed in duct cells of the adult human pancreas</article-title><source>Pancreas</source><volume>36</volume><fpage>e1</fpage><lpage>6</lpage><pub-id pub-id-type="doi">10.1097/mpa.0b013e318149f2dc</pub-id></element-citation></ref><ref id="bib30"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lawson</surname><given-names>DA</given-names></name><name><surname>Xin</surname><given-names>L</given-names></name><name><surname>Lukacs</surname><given-names>RU</given-names></name><name><surname>Cheng</surname><given-names>D</given-names></name><name><surname>Witte</surname><given-names>ON</given-names></name></person-group><year>2007</year><article-title>Isolation and functional characterization of murine prostate stem cells</article-title><source>Proc Natl Acad Sci USA</source><volume>104</volume><fpage>181</fpage><lpage>6</lpage><pub-id pub-id-type="doi">10.1073/pnas.0609684104</pub-id></element-citation></ref><ref id="bib31"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Lupi</surname><given-names>R</given-names></name><name><surname>Del Guerra</surname><given-names>S</given-names></name><name><surname>Tellini</surname><given-names>C</given-names></name><name><surname>Giannarelli</surname><given-names>R</given-names></name><name><surname>Coppelli</surname><given-names>A</given-names></name><name><surname>Lorenzetti</surname><given-names>M</given-names></name><etal/></person-group><year>1999</year><article-title>The biguanide compound metformin prevents desensitization of human pancreatic islets induced by high glucose</article-title><source>Eur J Pharmacol</source><volume>364</volume><fpage>205</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1016/S0014-2999(98)00807-3</pub-id></element-citation></ref><ref id="bib32"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>McKnight</surname><given-names>KD</given-names></name><name><surname>Wang</surname><given-names>P</given-names></name><name><surname>Kim</surname><given-names>SK</given-names></name></person-group><year>2010</year><article-title>Deconstructing pancreas development to reconstruct human islets from pluripotent stem cells</article-title><source>Cell Stem Cell</source><volume>6</volume><fpage>300</fpage><lpage>8</lpage><pub-id pub-id-type="doi">10.1016/j.stem.2010.03.003</pub-id></element-citation></ref><ref id="bib33"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Mellitzer</surname><given-names>G</given-names></name><name><surname>Bonne</surname><given-names>S</given-names></name><name><surname>Luco</surname><given-names>RF</given-names></name><name><surname>Van De Casteele</surname><given-names>M</given-names></name><name><surname>Lenne-Samuel</surname><given-names>N</given-names></name><name><surname>Collombat</surname><given-names>P</given-names></name><etal/></person-group><year>2006</year><article-title>IA1 is NGN3-dependent and essential for differentiation of the endocrine pancreas</article-title><source>EMBO J</source><volume>25</volume><fpage>1344</fpage><lpage>52</lpage><pub-id pub-id-type="doi">10.1038/sj.emboj.7601011</pub-id></element-citation></ref><ref id="bib34"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Minami</surname><given-names>K</given-names></name><name><surname>Okuno</surname><given-names>M</given-names></name><name><surname>Miyawaki</surname><given-names>K</given-names></name><name><surname>Okumachi</surname><given-names>A</given-names></name><name><surname>Ishizaki</surname><given-names>K</given-names></name><name><surname>Oyama</surname><given-names>K</given-names></name><etal/></person-group><year>2005</year><article-title>Lineage tracing and characterization of insulin-secreting cells generated from adult pancreatic acinar cells</article-title><source>Proc Natl Acad Sci USA</source><volume>102</volume><fpage>15116</fpage><lpage>21</lpage><pub-id pub-id-type="doi">10.1073/pnas.0507567102</pub-id></element-citation></ref><ref id="bib35"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Miyatsuka</surname><given-names>T</given-names></name><name><surname>Kosaka</surname><given-names>Y</given-names></name><name><surname>Kim</surname><given-names>H</given-names></name><name><surname>German</surname><given-names>MS</given-names></name></person-group><year>2011</year><article-title>Neurogenin3 inhibits proliferation in endocrine progenitors by inducing Cdkn1a</article-title><source>Proc Natl Acad Sci USA</source><volume>108</volume><fpage>185</fpage><lpage>90</lpage><pub-id pub-id-type="doi">10.1073/pnas.1004842108</pub-id></element-citation></ref><ref id="bib36"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Miyatsuka</surname><given-names>T</given-names></name><name><surname>Li</surname><given-names>Z</given-names></name><name><surname>German</surname><given-names>MS</given-names></name></person-group><year>2009</year><article-title>Chronology of islet differentiation revealed by temporal cell labeling</article-title><source>Diabetes</source><volume>58</volume><fpage>1863</fpage><lpage>8</lpage><pub-id pub-id-type="doi">10.2337/db09-0390</pub-id></element-citation></ref><ref id="bib37"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Noguchi</surname><given-names>H</given-names></name><name><surname>Xu</surname><given-names>G</given-names></name><name><surname>Matsumoto</surname><given-names>S</given-names></name><name><surname>Kaneto</surname><given-names>H</given-names></name><name><surname>Kobayashi</surname><given-names>N</given-names></name><name><surname>Bonner-Weir</surname><given-names>S</given-names></name><etal/></person-group><year>2006</year><article-title>Induction of pancreatic stem/progenitor cells into insulin-producing cells by adenoviral-mediated gene transfer technology</article-title><source>Cell Transplant</source><volume>15</volume><fpage>929</fpage><lpage>38</lpage><pub-id pub-id-type="doi">10.3727/000000006783981431</pub-id></element-citation></ref><ref id="bib38"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Nostro</surname><given-names>MC</given-names></name><name><surname>Keller</surname><given-names>G</given-names></name></person-group><year>2012</year><article-title>Generation of beta cells from human pluripotent stem cells: potential for regenerative medicine</article-title><source>Semin Cell Dev Biol</source><volume>23</volume><fpage>701</fpage><lpage>10</lpage><pub-id pub-id-type="doi">10.1016/j.semcdb.2012.06.010</pub-id></element-citation></ref><ref id="bib39"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Pan</surname><given-names>FC</given-names></name><name><surname>Wright</surname><given-names>C</given-names></name></person-group><year>2011</year><article-title>Pancreas organogenesis: from bud to plexus to gland</article-title><source>Dev Dyn</source><volume>240</volume><fpage>530</fpage><lpage>65</lpage><pub-id pub-id-type="doi">10.1002/dvdy.22584</pub-id></element-citation></ref><ref id="bib40"><element-citation publication-type="book"><person-group person-group-type="author"><name><surname>Pictet</surname><given-names>R</given-names></name><name><surname>Rutter</surname><given-names>WJ</given-names></name></person-group><year>1972</year><article-title>Development of the embryonic endocrine pancreas</article-title><person-group person-group-type="editor"><name><surname>Steiner</surname><given-names>DF</given-names></name><name><surname>Freinkel</surname><given-names>N</given-names></name></person-group><source>Handbook of physiology: a critical, comprehensive presentation of physiological knowledge and concepts</source><comment>section 7: Endocrinology</comment><publisher-name>American Physiological Society</publisher-name><fpage>25</fpage><lpage>55</lpage></element-citation></ref><ref id="bib41"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rescan</surname><given-names>C</given-names></name><name><surname>Le Bras</surname><given-names>S</given-names></name><name><surname>Lefebvre</surname><given-names>VH</given-names></name><name><surname>Frandsen</surname><given-names>U</given-names></name><name><surname>Klein</surname><given-names>T</given-names></name><name><surname>Foschi</surname><given-names>M</given-names></name><etal/></person-group><year>2005</year><article-title>EGF-induced proliferation of adult human pancreatic duct cells is mediated by the MEK/ERK cascade</article-title><source>Lab Invest</source><volume>85</volume><fpage>65</fpage><lpage>74</lpage><pub-id pub-id-type="doi">10.1038/labinvest.3700204</pub-id></element-citation></ref><ref id="bib42"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rovira</surname><given-names>M</given-names></name><name><surname>Scott</surname><given-names>SG</given-names></name><name><surname>Liss</surname><given-names>AS</given-names></name><name><surname>Jensen</surname><given-names>J</given-names></name><name><surname>Thayer</surname><given-names>SP</given-names></name><name><surname>Leach</surname><given-names>SD</given-names></name></person-group><year>2010</year><article-title>Isolation and characterization of centroacinar/terminal ductal progenitor cells in adult mouse pancreas</article-title><source>Proc Natl Acad Sci USA</source><volume>107</volume><fpage>75</fpage><lpage>80</lpage><pub-id pub-id-type="doi">10.1073/pnas.0912589107</pub-id></element-citation></ref><ref id="bib43"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Rozzo</surname><given-names>A</given-names></name><name><surname>Meneghel-Rozzo</surname><given-names>T</given-names></name><name><surname>Delakorda</surname><given-names>SL</given-names></name><name><surname>Yang</surname><given-names>SB</given-names></name><name><surname>Rupnik</surname><given-names>M</given-names></name></person-group><year>2009</year><article-title>Exocytosis of insulin: in vivo maturation of mouse endocrine pancreas</article-title><source>Ann N Y Acad Sci</source><volume>1152</volume><fpage>53</fpage><lpage>62</lpage><pub-id pub-id-type="doi">10.1111/j.1749-6632.2008.04003.x</pub-id></element-citation></ref><ref id="bib44"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sander</surname><given-names>M</given-names></name><name><surname>Neubuser</surname><given-names>A</given-names></name><name><surname>Kalamaras</surname><given-names>J</given-names></name><name><surname>Ee</surname><given-names>HC</given-names></name><name><surname>Martin</surname><given-names>GR</given-names></name><name><surname>German</surname><given-names>MS</given-names></name></person-group><year>1997</year><article-title>Genetic analysis reveals that PAX6 is required for normal transcription of pancreatic hormone genes and islet development</article-title><source>Genes Dev</source><volume>11</volume><fpage>1662</fpage><lpage>73</lpage><pub-id pub-id-type="doi">10.1101/gad.11.13.1662</pub-id></element-citation></ref><ref id="bib45"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sapir</surname><given-names>T</given-names></name><name><surname>Shternhall</surname><given-names>K</given-names></name><name><surname>Meivar-Levy</surname><given-names>I</given-names></name><name><surname>Blumenfeld</surname><given-names>T</given-names></name><name><surname>Cohen</surname><given-names>H</given-names></name><name><surname>Skutelsky</surname><given-names>E</given-names></name><etal/></person-group><year>2005</year><article-title>Cell-replacement therapy for diabetes: generating functional insulin-producing tissue from adult human liver cells</article-title><source>Proc Natl Acad Sci USA</source><volume>102</volume><fpage>7964</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.1073/pnas.0405277102</pub-id></element-citation></ref><ref id="bib46"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sato</surname><given-names>T</given-names></name><name><surname>Vries</surname><given-names>RG</given-names></name><name><surname>Snippert</surname><given-names>HJ</given-names></name><name><surname>van de Wetering</surname><given-names>M</given-names></name><name><surname>Barker</surname><given-names>N</given-names></name><name><surname>Stange</surname><given-names>DE</given-names></name><etal/></person-group><year>2009</year><article-title>Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche</article-title><source>Nature</source><volume>459</volume><fpage>262</fpage><lpage>5</lpage><pub-id pub-id-type="doi">10.1038/nature07935</pub-id></element-citation></ref><ref id="bib47"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Smith</surname><given-names>SB</given-names></name><name><surname>Qu</surname><given-names>HQ</given-names></name><name><surname>Taleb</surname><given-names>N</given-names></name><name><surname>Kishimoto</surname><given-names>NY</given-names></name><name><surname>Scheel</surname><given-names>DW</given-names></name><name><surname>Lu</surname><given-names>Y</given-names></name><etal/></person-group><year>2010</year><article-title>Rfx6 directs islet formation and insulin production in mice and humans</article-title><source>Nature</source><volume>463</volume><fpage>775</fpage><lpage>80</lpage><pub-id pub-id-type="doi">10.1038/nature08748</pub-id></element-citation></ref><ref id="bib48"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Solar</surname><given-names>M</given-names></name><name><surname>Cardalda</surname><given-names>C</given-names></name><name><surname>Houbracken</surname><given-names>I</given-names></name><name><surname>Martin</surname><given-names>M</given-names></name><name><surname>Maestro</surname><given-names>MA</given-names></name><name><surname>De Medts</surname><given-names>N</given-names></name><etal/></person-group><year>2009</year><article-title>Pancreatic exocrine duct cells give rise to insulin-producing beta cells during embryogenesis but not after birth</article-title><source>Dev Cell</source><volume>17</volume><fpage>849</fpage><lpage>60</lpage><pub-id pub-id-type="doi">10.1016/j.devcel.2009.11.003</pub-id></element-citation></ref><ref id="bib49"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sosa-Pineda</surname><given-names>B</given-names></name><name><surname>Chowdhury</surname><given-names>K</given-names></name><name><surname>Torres</surname><given-names>M</given-names></name><name><surname>Oliver</surname><given-names>G</given-names></name><name><surname>Gruss</surname><given-names>P</given-names></name></person-group><year>1997</year><article-title>The Pax4 gene is essential for differentiation of insulin-producing beta cells in the mammalian pancreas</article-title><source>Nature</source><volume>386</volume><fpage>399</fpage><lpage>402</lpage><pub-id pub-id-type="doi">10.1038/386399a0</pub-id></element-citation></ref><ref id="bib50"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Suckale</surname><given-names>J</given-names></name><name><surname>Solimena</surname><given-names>M</given-names></name></person-group><year>2010</year><article-title>The insulin secretory granule as a signaling hub</article-title><source>Trends Endocrinol Metab</source><volume>21</volume><fpage>599</fpage><lpage>609</lpage><pub-id pub-id-type="doi">10.1016/j.tem.2010.06.003</pub-id></element-citation></ref><ref id="bib51"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sugiyama</surname><given-names>T</given-names></name><name><surname>Rodriguez</surname><given-names>RT</given-names></name><name><surname>McLean</surname><given-names>GW</given-names></name><name><surname>Kim</surname><given-names>SK</given-names></name></person-group><year>2007</year><article-title>Conserved markers of fetal pancreatic epithelium permit prospective isolation of islet progenitor cells by FACS</article-title><source>Proc Natl Acad Sci USA</source><volume>104</volume><fpage>175</fpage><lpage>80</lpage><pub-id pub-id-type="doi">10.1073/pnas.0609490104</pub-id></element-citation></ref><ref id="bib52"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Sussel</surname><given-names>L</given-names></name><name><surname>Kalamaras</surname><given-names>J</given-names></name><name><surname>Hartigan-O’Connor</surname><given-names>DJ</given-names></name><name><surname>Meneses</surname><given-names>JJ</given-names></name><name><surname>Pedersen</surname><given-names>RA</given-names></name><name><surname>Rubenstein</surname><given-names>JL</given-names></name><etal/></person-group><year>1998</year><article-title>Mice lacking the homeodomain transcription factor Nkx2.2 have diabetes due to arrested differentiation of pancreatic beta cells</article-title><source>Development</source><volume>125</volume><fpage>2213</fpage><lpage>21</lpage></element-citation></ref><ref id="bib53"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Swales</surname><given-names>N</given-names></name><name><surname>Martens</surname><given-names>GA</given-names></name><name><surname>Bonne</surname><given-names>S</given-names></name><name><surname>Heremans</surname><given-names>Y</given-names></name><name><surname>Borup</surname><given-names>R</given-names></name><name><surname>Van de Casteele</surname><given-names>M</given-names></name><etal/></person-group><year>2012</year><article-title>Plasticity of adult human pancreatic duct cells by neurogenin3-mediated reprogramming</article-title><source>PLOS ONE</source><volume>7</volume><fpage>e37055</fpage><pub-id pub-id-type="doi">10.1371/journal.pone.0037055</pub-id></element-citation></ref><ref id="bib54"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Takahashi</surname><given-names>K</given-names></name><name><surname>Tanabe</surname><given-names>K</given-names></name><name><surname>Ohnuki</surname><given-names>M</given-names></name><name><surname>Narita</surname><given-names>M</given-names></name><name><surname>Ichisaka</surname><given-names>T</given-names></name><name><surname>Tomoda</surname><given-names>K</given-names></name><etal/></person-group><year>2007</year><article-title>Induction of pluripotent stem cells from adult human fibroblasts by defined factors</article-title><source>Cell</source><volume>131</volume><fpage>861</fpage><lpage>72</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2007.11.019</pub-id></element-citation></ref><ref id="bib55"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Tashiro</surname><given-names>F</given-names></name><name><surname>Niwa</surname><given-names>H</given-names></name><name><surname>Miyazaki</surname><given-names>J</given-names></name></person-group><year>1999</year><article-title>Constructing adenoviral vectors by using the circular form of the adenoviral genome cloned in a cosmid and the Cre-loxP recombination system</article-title><source>Hum Gene Ther</source><volume>10</volume><fpage>1845</fpage><lpage>52</lpage><pub-id pub-id-type="doi">10.1089/10430349950017527</pub-id></element-citation></ref><ref id="bib55a"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Thatava</surname><given-names>T</given-names></name><name><surname>Nelson</surname><given-names>TJ</given-names></name><name><surname>Edukulla</surname><given-names>R</given-names></name><name><surname>Sakuma</surname><given-names>T</given-names></name><name><surname>Ohmine</surname><given-names>S</given-names></name><name><surname>Tonne</surname><given-names>JM</given-names></name></person-group><year>2011</year><article-title>Indolactam V/GLP-1-mediated differentiation of human iPS cells into glucose-responsive insulin-secreting progeny</article-title><source>Gene Ther</source><volume>18</volume><fpage>283</fpage><lpage>93</lpage><pub-id pub-id-type="doi">10.1038/gt.2010.145</pub-id></element-citation></ref><ref id="bib56"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Trautmann</surname><given-names>B</given-names></name><name><surname>Schlitt</surname><given-names>HJ</given-names></name><name><surname>Hahn</surname><given-names>EG</given-names></name><name><surname>Lohr</surname><given-names>M</given-names></name></person-group><year>1993</year><article-title>Isolation, culture, and characterization of human pancreatic duct cells</article-title><source>Pancreas</source><volume>8</volume><fpage>248</fpage><lpage>54</lpage><pub-id pub-id-type="doi">10.1097/00006676-199303000-00017</pub-id></element-citation></ref><ref id="bib57"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Vardanyan</surname><given-names>M</given-names></name><name><surname>Parkin</surname><given-names>E</given-names></name><name><surname>Gruessner</surname><given-names>C</given-names></name><name><surname>Rodriguez Rilo</surname><given-names>HL</given-names></name></person-group><year>2010</year><article-title>Pancreas vs. islet transplantation: a call on the future</article-title><source>Curr Opin Organ Transplant</source><volume>15</volume><fpage>124</fpage><lpage>30</lpage><pub-id pub-id-type="doi">10.1097/MOT.0b013e32833553f8</pub-id></element-citation></ref><ref id="bib58"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Vierbuchen</surname><given-names>T</given-names></name><name><surname>Wernig</surname><given-names>M</given-names></name></person-group><year>2011</year><article-title>Direct lineage conversions: unnatural but useful?</article-title><source>Nat Biotechnol</source><volume>29</volume><fpage>892</fpage><lpage>907</lpage><pub-id pub-id-type="doi">10.1038/nbt.1946</pub-id></element-citation></ref><ref id="bib59"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>P</given-names></name><name><surname>Rodriguez</surname><given-names>RT</given-names></name><name><surname>Wang</surname><given-names>J</given-names></name><name><surname>Ghodasara</surname><given-names>A</given-names></name><name><surname>Kim</surname><given-names>SK</given-names></name></person-group><year>2011</year><article-title>Targeting SOX17 in human embryonic stem cells creates unique strategies for isolating and analyzing developing endoderm</article-title><source>Cell Stem Cell</source><volume>8</volume><fpage>335</fpage><lpage>46</lpage><pub-id pub-id-type="doi">10.1016/j.stem.2011.01.017</pub-id></element-citation></ref><ref id="bib60"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>S</given-names></name><name><surname>Jensen</surname><given-names>JN</given-names></name><name><surname>Seymour</surname><given-names>PA</given-names></name><name><surname>Hsu</surname><given-names>W</given-names></name><name><surname>Dor</surname><given-names>Y</given-names></name><name><surname>Sander</surname><given-names>M</given-names></name><etal/></person-group><year>2009</year><article-title>Sustained Neurog3 expression in hormone-expressing islet cells is required for endocrine maturation and function</article-title><source>Proc Natl Acad Sci USA</source><volume>106</volume><fpage>9715</fpage><lpage>20</lpage><pub-id pub-id-type="doi">10.1073/pnas.0904247106</pub-id></element-citation></ref><ref id="bib61"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Wang</surname><given-names>S</given-names></name><name><surname>Yan</surname><given-names>J</given-names></name><name><surname>Anderson</surname><given-names>DA</given-names></name><name><surname>Xu</surname><given-names>Y</given-names></name><name><surname>Kanal</surname><given-names>MC</given-names></name><name><surname>Cao</surname><given-names>Z</given-names></name><etal/></person-group><year>2010</year><article-title>Neurog3 gene dosage regulates allocation of endocrine and exocrine cell fates in the developing mouse pancreas</article-title><source>Dev Biol</source><volume>339</volume><fpage>26</fpage><lpage>37</lpage><pub-id pub-id-type="doi">10.1016/j.ydbio.2009.12.009</pub-id></element-citation></ref><ref id="bib62"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Warren</surname><given-names>L</given-names></name><name><surname>Manos</surname><given-names>PD</given-names></name><name><surname>Ahfeldt</surname><given-names>T</given-names></name><name><surname>Loh</surname><given-names>YH</given-names></name><name><surname>Li</surname><given-names>H</given-names></name><name><surname>Lau</surname><given-names>F</given-names></name><etal/></person-group><year>2010</year><article-title>Highly efficient reprogramming to pluripotency and directed differentiation of human cells with synthetic modified mRNA</article-title><source>Cell Stem Cell</source><volume>7</volume><fpage>618</fpage><lpage>30</lpage><pub-id pub-id-type="doi">10.1016/j.stem.2010.08.012</pub-id></element-citation></ref><ref id="bib63"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Xu</surname><given-names>X</given-names></name><name><surname>D’Hoker</surname><given-names>J</given-names></name><name><surname>Stange</surname><given-names>G</given-names></name><name><surname>Bonne</surname><given-names>S</given-names></name><name><surname>De Leu</surname><given-names>N</given-names></name><name><surname>Xiao</surname><given-names>X</given-names></name><etal/></person-group><year>2008</year><article-title>Beta cells can be generated from endogenous progenitors in injured adult mouse pancreas</article-title><source>Cell</source><volume>132</volume><fpage>197</fpage><lpage>207</lpage><pub-id pub-id-type="doi">10.1016/j.cell.2007.12.015</pub-id></element-citation></ref><ref id="bib64"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Yatoh</surname><given-names>S</given-names></name><name><surname>Dodge</surname><given-names>R</given-names></name><name><surname>Akashi</surname><given-names>T</given-names></name><name><surname>Omer</surname><given-names>A</given-names></name><name><surname>Sharma</surname><given-names>A</given-names></name><name><surname>Weir</surname><given-names>GC</given-names></name><etal/></person-group><year>2007</year><article-title>Differentiation of affinity-purified human pancreatic duct cells to beta-cells</article-title><source>Diabetes</source><volume>56</volume><fpage>1802</fpage><lpage>9</lpage><pub-id pub-id-type="doi">10.2337/db06-1670</pub-id></element-citation></ref><ref id="bib65"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Yoo</surname><given-names>AS</given-names></name><name><surname>Sun</surname><given-names>AX</given-names></name><name><surname>Li</surname><given-names>L</given-names></name><name><surname>Shcheglovitov</surname><given-names>A</given-names></name><name><surname>Portmann</surname><given-names>T</given-names></name><name><surname>Li</surname><given-names>Y</given-names></name><etal/></person-group><year>2011</year><article-title>MicroRNA-mediated conversion of human fibroblasts to neurons</article-title><source>Nature</source><volume>476</volume><fpage>228</fpage><lpage>31</lpage><pub-id pub-id-type="doi">10.1038/nature10323</pub-id></element-citation></ref><ref id="bib66"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhang</surname><given-names>D</given-names></name><name><surname>Jiang</surname><given-names>W</given-names></name><name><surname>Liu</surname><given-names>M</given-names></name><name><surname>Sui</surname><given-names>X</given-names></name><name><surname>Yin</surname><given-names>X</given-names></name><name><surname>Chen</surname><given-names>S</given-names></name><etal/></person-group><year>2009</year><article-title>Highly efficient differentiation of human ES cells and iPS cells into mature pancreatic insulin-producing cells</article-title><source>Cell Res</source><volume>19</volume><fpage>429</fpage><lpage>38</lpage><pub-id pub-id-type="doi">10.1038/cr.2009.28</pub-id></element-citation></ref><ref id="bib67"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhou</surname><given-names>Q</given-names></name><name><surname>Brown</surname><given-names>J</given-names></name><name><surname>Kanarek</surname><given-names>A</given-names></name><name><surname>Rajagopal</surname><given-names>J</given-names></name><name><surname>Melton</surname><given-names>DA</given-names></name></person-group><year>2008</year><article-title>In vivo reprogramming of adult pancreatic exocrine cells to beta-cells</article-title><source>Nature</source><volume>455</volume><fpage>627</fpage><lpage>32</lpage><pub-id pub-id-type="doi">10.1038/nature07314</pub-id></element-citation></ref><ref id="bib68"><element-citation publication-type="journal"><person-group person-group-type="author"><name><surname>Zhou</surname><given-names>Q</given-names></name><name><surname>Law</surname><given-names>AC</given-names></name><name><surname>Rajagopal</surname><given-names>J</given-names></name><name><surname>Anderson</surname><given-names>WJ</given-names></name><name><surname>Gray</surname><given-names>PA</given-names></name><name><surname>Melton</surname><given-names>DA</given-names></name></person-group><year>2007</year><article-title>A multipotent progenitor domain guides pancreatic organogenesis</article-title><source>Dev Cell</source><volume>13</volume><fpage>103</fpage><lpage>14</lpage><pub-id pub-id-type="doi">10.1016/j.devcel.2007.06.001</pub-id></element-citation></ref></ref-list></back><sub-article article-type="article-commentary" id="SA1"><front-stub><article-id pub-id-type="doi">10.7554/eLife.00940.019</article-id><title-group><article-title>Decision letter</article-title></title-group><contrib-group content-type="section"><contrib contrib-type="editor"><name><surname>Rossant</surname><given-names>Janet</given-names></name><role>Reviewing editor</role><aff><institution>University of Toronto</institution>, <country>Canada</country></aff></contrib></contrib-group></front-stub><body><boxed-text><p>eLife posts the editorial decision letter and author response on a selection of the published articles (subject to the approval of the authors). An edited version of the letter sent to the authors after peer review is shown, indicating the substantive concerns or comments; minor concerns are not usually shown. Reviewers have the opportunity to discuss the decision before the letter is sent (see <ext-link ext-link-type="uri" xlink:href="http://elife.elifesciences.org/review-process">review process</ext-link>). Similarly, the author response typically shows only responses to the major concerns raised by the reviewers.</p></boxed-text><p>[Editors’ note: this article was originally rejected after discussions between the reviewers, but the authors were invited to resubmit after an appeal against the decision.]</p><p>Thank you for choosing to send your work entitled “Expansion and Conversion of Human Pancreatic Ductal Cells into Insulin-Secreting Endocrine Cells” for consideration at <italic>eLife</italic>. Your full submission has been peer reviewed by one of our Senior editors, Janet Rossant, and two other reviewers, and the decision was reached after discussions between the reviewers. We regret to inform you that your work will not be considered further for publication at this point.</p><p>The reviewers and the Senior editor have had an extensive online discussion about your paper, after exchanging the reviews. While they all feel that the experiments are carefully carried out, and the data presented are robust, in the end they were not convinced that the study as a whole provided a major step forward in the drive towards generating functional beta cells from other cell types. It was noted that you have not demonstrated whether the use of ductal cells (CD133<sup>+</sup> cells) is advantageous over the use of other cell types for transdifferentiation. What would happen if the same factors were used to reprogram other cell types, even non-pancreatic cells, such as fibroblasts? The fact that ductal cells can respond to exogenous transcription factors does not directly demonstrate that these cells have latent potential to form beta cells, as claimed in the Abstract. It was also noted that the beta cells produced are not apparently fully mature and, therefore, the long-term significance of this approach for human therapy is unclear. The ability to grow human ductal cells in vitro is interesting and a further analysis of the non-transduced cells to respond to external signals and undergo differentiation into different cell types would be interesting.</p><p>Given these major concerns and the amount of extra work that would be needed to address them, the decision is to reject the manuscript at this time. A majorly enhanced experimental study including assessing whether ductal cells are uniquely responsive to these inducing factors, better characterization of the cells produced, and a further analysis of the unmanipulated ductal cells could form the basis of a new submission at a later date. The major points from the full reviews are provided below.</p><p><italic>Reviewer #1</italic>:</p><p>In this manuscript the authors show that they can sort human cadaveric pancreas tissue with antibody to CD133 and that this enriches for pancreatic ductal epithelium. They then show that they can generate clonal spheres from these cells that can be passaged several generations in culture. They then infect these cultures with adenovirus expressing first neurogenin and then additional sets of beta cell inducing transcription factors and show that they can induce expression of endocrine phenotypes in the spheres and that a combination of 4 factors generates insulin-producing cells that show some degree of glucose regulation. The final most successful converted cells express 7% of the levels of insulin found in adult islets, but when transplanted to the kidney capsule in mice, they were able to detect some human insulin in serum. However, in these grafts cell survival was poor, so they were unable to test the ability of the cells to rescue diabetic mice.</p><p>This study is well performed and does indicate that ductal cells may be responsive to exogenous transcription factors that can drive towards an islet cell fate.</p><p>Major comments:</p><p>1) It is not clear whether conversion to islet cells is a unique property of ductal cells or whether other pancreatic cells or other cell types could respond in the same way. Other groups have suggested that hepatocytes can be transdifferentiated to insulin-producing cells- is this system more or less effective?</p><p>2) The final cocktail of transcription factors is stated to produce monohormonal insulin-producing cells, but this is not directly shown in the figures. This is an important point because many other differentiation assays generate fetal-like polyhormonal cells that cannot respond to glucose in the manner of adult beta cells. The cells produced here are not fully functional adult-type cells.</p><p>3) How long does expression of the exogenous factors persist? Is it required for ongoing maintenance of the cells or can you demonstrate independence of the exogenous factors?</p><p>4) How sure are they that the starting population is pure ductal cells, given that CD133 is not exclusively expressed in ductal epithelium in the pancreas? Can they double sort with a general epithelial marker to further purify the population, given that CD133 only enriches 4-fold for sphere-forming cells?</p><p><italic>Reviewer #2</italic>:</p><p>In this study the authors describe a method to isolate and expand human ductal cells using an antibody against CD133. Using culture conditions similar to those described by Dr. Hans Clevers (Sato et al.), they were able to culture CD133<sup>+</sup> cells as epithelial spheres that maintain a ductal phenotype and lack acinar and endocrine markers. Furthermore, the authors were able to reprogram the CD133-enriched population to endocrine cells by infecting isolated CD133<sup>+</sup> cells and/or CD133<sup>+</sup> -derived spheres with adenoviruses expressing ngn3, MAFA, PAX6 and PDX1. On average the authors are able to generate 10% insulin+ cells that resemble fetal beta cells, as they secrete insulin in response to low level of glucose (2mM), but fail to respond to higher glucose concentration (11mM). Following transplantation of the reprogrammed spheres into NSG mice, they observed that survival of the graft after transplantation was poor. However, they were able to detect human insulin in the serum of the host mice and showed that insulin levels increased after glucose challenge, suggesting in vivo maturation, albeit few mice were analysed. In general this work is very well done, with convincing images and clear data. There are only some minor points that need to be clarified.</p><p>Minor comments:</p><p>1) As CD133 is detected in centroacinar cells as well as in ductal cells (Immervoll et al JHC 2011), the authors should include additional acinar markers in their expression profile (<xref ref-type="fig" rid="fig1">Fig 1 D</xref>) to exclude acinar contamination.</p><p>2) Please include the average percentage of CD133<sup>+</sup> cells detected in human pancreas and the purity of the sorted populations.</p><p>3) <xref ref-type="fig" rid="fig2">Figure 2A</xref> please include co-staining of KRT19 and CD133.</p><p>4) The authors state “Time-lapse imaging revealed that spheres arose from single isolated CD133<sup>+</sup> ductal cells”. However, this statement is not accurate unless the purity of the sort was 100%.</p><p>5) Does the percentage of CD133<sup>+</sup> cells decrease in culture? What is the percentage of CD133 after 3 months in culture?</p><p>6) The authors state that Insulin<sup>+</sup> IPCs did not express other islet hormones, the authors should include co-staining of c-peptide with GCG, PP and Ghrelin in <xref ref-type="fig" rid="fig5">Figure 5D</xref>.</p><p><italic>Reviewer #3</italic>:</p><p>The claim that adult human pancreatic duct cells have a latent capacity for endocrine differentiation is correct, but only in this context of extremely strong transcription-factor-based enforced differentiation. I always wonder what MNP6 would do to a non-pancreatic cell type, and therefore if this effect is a specific latency of pancreatic duct cells or not (the paragraph starting ‘The transcription factors <italic>MafA</italic>, <italic>Neurog3</italic> and <italic>Pdx1</italic>…’ is more important as a result if there is something these cells can do that is not ever seen with the 4-factor combination MNP6 (4TF) or 4TFM).</p><p>Essential is the claim that the insulin-producing cells are mono-hormone-positive, but this is not shown in the paper. The authors refer to the <xref ref-type="fig" rid="fig5">Figure 5D</xref> as the one showing no double-positives, but no Gcg, PP, or Ghrl are shown here.</p><p>Some more clarity on this aspect seems critical. Assays for Gcg and other hormones are either referred to as data not shown, or this aspect is stated but the figure does not have the supporting data. Gcg immunodetection was done on cells from N alone? Gcg, PP were tested on the 4-factor combinations?</p><p>Some explanation of why spontaneous (that is, non-TF-enforced) differentiation to endocrine fates was ruled out.</p><p>Is Pax6 already expressed in the MNP-adenovirus cultures? If so, why is more needed?</p><p>The part on “We sought new methods to mature the cells” (my words) reads strangely. It's the same method, just extending the time frame, I believe, and I would simplify this text. Another comment here is that we revert to the MNP6 mixture (4TF), having just been told that Pdx1 can be removed without impact - why?</p><p><xref ref-type="fig" rid="fig2s1">Figure 2C–figure supplement 1</xref> has an incorrect y-axis. 10,000 percent to the log(base10) is 4, correct. This graph needs altering – why not just “fold” for cell number? Related to this, Discussion asserts 3,000-fold, but this is just once? Up to 3,000-fold, and more typically xx-fold? Why do the cultures suddenly go bad at G7 (see text) – what happens – sudden apoptosis? #48 seems to be continuing even at G7 – please amend this text.</p><p>Does the 3-factor system (MNP) work in the authors' hands on mouse acini? And/or duct?</p><p>Does the CD133 separation method include centroacinar cells (CAC) or not? What is the photograph in <xref ref-type="fig" rid="fig1">Figure 1</xref> – ‘tip’ of duct dives out of plane of section before it is reached, and therefore the CAC cannot be seen in this panel? CAC could have a specific latency not seen in duct cells.</p><p>The idea that lineage-tracing methods are hard to apply in human cells should be stated, as everything else hinges on numerical arguments on cleanliness of cell separations, etc.</p><p>chgA is an endocrine differentiation marker. This statement is meant to indicate that full-blown differentiation to the hormone-expressing state requires a substantial pulse of Ngn3. What about other hormones, even indicating non-pancreatic cell types?</p><p>A major point is the longevity of the pulse of Ngn3 and the other factors achieved with these methods, and the detection of a program that runs from the endogenous loci with or without the continuous presence of N, MNP, MNP6 (4TF, 4TFM). Would this method pulse the cells or not, and is continuous presence of some of the factors preventing their full differentiation to the most mature state?</p><p>Title of the section starting ‘Although ELISA studies readily detected Proinsulin production by IPCs…’ reads odd to me: ‘genetic conversion’ reads as if the genome of the adult duct cells is being altered in some manner.</p><p>Systematic removal of factors from MNP6 mixtures: Why can Pdx1 be removed without any impact?</p><p>Various ‘obvious’ markers were not tested, or the manuscript is incorrect in not showing such ‘easily pointed out’ questions. Pdx1 is produced, by immunofluorescence assay, within these induced beta cells? To normal levels? MafA/B, etc? Nkx6.1 was assessed, but the ‘dogmatic’ mature β-cell TF list should be addressed, at least.</p><p>It does seem difficult to follow 4V, 4TF, 4TFM, MNP. Seems as if there is a mixed descriptor used for the same manipulation, at least sometimes; I suggest simply checking for a way of making the text fully consistent throughout.</p></body></sub-article><sub-article article-type="reply" id="SA2"><front-stub><article-id pub-id-type="doi">10.7554/eLife.00940.020</article-id><title-group><article-title>Author response</article-title></title-group></front-stub><body><p>We are grateful that the initial review has provided such useful suggestions for improving our study, and that the overall view of the experimental strategy, concepts, and data quality were so positive. We have provided experimental data, most of it new, to address all the remaining major concerns summarized in the decision letter, including (1) new data with human fibroblasts to assess whether primary human pancreatic ductal cells are uniquely responsive to conversion conditions identified here, (2) better characterization of the cells produced, especially studies to establish their mono-hormonal development, and (3) further analysis of unmanipulated ductal cells. We trust our responses to these requests for additional data meet or exceed the reviewers' expectations.</p><p>In addition to the detailed responses to specific points below, we would like to address the general point about the importance of the findings presented here. We thank the reviewers for the opportunity to clarify why our work “represents a major step forward in the drive toward generating functional beta cells from other cell types”.</p><p>An important, and oft-cited ‘benchmark’, in this effort is production of insulin-producing cells from stem cell lines like human ES cells (D’Amour et al., 2006; <xref ref-type="bibr" rid="bib28">Kroon et al., 2008</xref> and subsequent work). However, these cells develop largely as poly-hormonal cells (most frequently expressing both glucagon and insulin) and fail to secrete insulin in response to glucose or other secretagogues unless transplanted for >2 months in mice (reviewed in <xref ref-type="bibr" rid="bib38">Nostro and Keller, 2012</xref>). Thus, it has remained unknown whether human cells can develop solely in vitro to generate glucose-responsive insulin-secreting progeny and without tumorigenicity. Our data indicate that in principle this can be achieved. Moreover, the depth and quality of phenotyping we perform to characterize achievement of beta-cell fate (in our view) is unmatched by any other prior study, with inclusion of ultrastructural, in vitro, and in vivo secretion data.</p><p>A second as yet unanswered question in the field is whether human pancreatic cells can be converted using transgenic methods toward a ?-cell fate. The Melton group (<xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>) established that this was possible in mice using viral methods in exocrine acinar cells, but since that seminal study the field has yearned to know how relevant this strategy might be for human cells, especially human pancreatic cells. We feel that our work unequivocally establishes that human pancreatic ductal exocrine cells have latent potential to produce functional endocrine islet-like cells.</p><p>Third, there is a longstanding debate about the ability of pancreatic ductal cells to generate endocrine progeny, with the majority of experimental assessment of this important question previously performed in transgenic mice. Thus, we feel our work provides important unique evidence that pancreatic ductal (or centroacinar) cells have such endocrine potential. Moreover, our work addresses a gap in knowledge in our field about (and a new system for evaluating) the potential of human ductal cells for alternate fates.</p><p>The identification of a new cellular source and genetic strategy for generating progeny with features of functional ?-cells should advance the long-term development of cell replacement options in type 1 diabetes. Thus, we feel quite strongly that our work provides a major contribution by addressing and answering multiple outstanding questions in diabetes research and pancreas biology. We hope that our revision clarifies this contribution and is now deemed worthy of publication in <italic>eLife.</italic></p><p>Reviewer #1:</p><p><italic>1) It is not clear whether conversion to islet cells is a unique property of ductal cells or whether other pancreatic cells or other cell types could respond in the same way. Other groups have suggested that hepatocytes can be transdifferentiated to insulin-producing cells – is this system more or less effective</italic>?</p><p>We fully agree with this reviewer and the similar comment from Reviewer #3 that additional assessment of other human cells would enhance the impact and conclusions from our study. Thus, we attempted to convert human adult dermal fibroblasts using our conversion methods; these cells have been previously used successfully in other reprogramming experiments (<xref ref-type="bibr" rid="bib65">Yoo et al., 2011</xref>). With the same 4TFM condition and Insulin or ChromograninA (ChgA) mRNA levels as readouts, we detected no clear evidence of conversion toward an endocrine or β-cell fate (n ≥ 6 from 3 independent experiments: <xref ref-type="fig" rid="fig5s4">Figure 5–figure supplement 4</xref>; see Materials and Methods for details). This indicates that human adult dermal fibroblasts have little to no ?-cell conversion potential. Despite repeated attempts, we were unable to assess the conversion property of human acinar cells due to their inability to grow in culture (<xref ref-type="fig" rid="fig1">Figure 1B</xref> CD133-negative population), consistent with other reports that primary acinar cells are difficult to culture. Thus, although we are unable to test all human cell types in this way, our work suggests that primary adult pancreatic ductal cells have a special latency for conversion toward an endocrine islet and beta cell-like fate.</p><p>To our knowledge, work by Sapir and colleagues (<xref ref-type="bibr" rid="bib45">Sapir et al., 2005</xref>) is the only report describing the use of human primary hepatocytes for β-cell transdifferentiation. In this report, however, the conversion toward an insulin-producing fate was comparatively poor; resulting cells produced about 10,000-fold lower insulin mRNA level than that of human islets, about 3–4 orders of magnitude lower than from conversion of pancreatic duct spheres. In addition, characteristic dense core vesicles in converted hepatocytes were not observed, indicating insufficient conversion towards ?-cells. In our report, we have presented evidence for acquisition of several characteristics of insulin-producing cells by converted duct spheres, including development of characteristic dense core vesicles, high Insulin content (estimated 49–77% total insulin in each insulin-producing cell compared to human islets), absence of glucagon or other islet hormone co-expression, modest glucose sensing and insulin secretion capacity, and other features detailed in our results.</p><p>These features also distinguish our work from studies reporting insulin-producing cells produced from in vitro differentiation of human embryonic stem cells (hESCs) and induced pluripotent stem cells (iPSCs). In our opinion, the qualitative and quantitative level of beta-cell phenotypes achieved by genetically-directed duct sphere conversion is higher than that achieved by prior hESC or iPSC studies in vitro. Moreover, our findings suggest that genetic methods may enhance beta-cell development from such cell lines</p><p>In sum, we believe conversion of human adult duct spheres is by far the most robust method yet reported for generating insulin-producing cells from human primary cells. We incorporated these points in the revised manuscript to clearly demonstrate the advantage of ductal cell use in human. We thank the reviewer for pointing this out.</p><p><italic>2) The final cocktail of transcription factors is stated to produce monohormonal insulin-producing cells, but this is not directly shown in the figures. This is an important point because many other differentiation assays generate fetal-like polyhormonal cells that cannot respond to glucose in the manner of adult beta cells. The cells produced here are not fully functional adult-type cells</italic>.</p><p>We used immunochemistry to detect co-expression of C-peptide with GCG, PPY, and GHRL, and included such GCG and GHRL co-staining images in a new <xref ref-type="fig" rid="fig5">Figure 5D</xref>. We found no PPY-positive cells from 7,487 GFP-positive cells we screened from two different samples. We found three GCG- positive cells out of 4,460 GFP-positive cells screened and none were co-labeled with C-peptide.</p><p><italic>3) How long does expression of the exogenous factors persist? Is it required for ongoing maintenance of the cells or can you demonstrate independence of the exogenous factors</italic>?</p><p>Expression of the exogenous factors persisted after two weeks of maturation period (<xref ref-type="fig" rid="fig5s3">Figure 5–figure supplement 3</xref>) and at least 5 months, evidenced by the GFP-positive insulin-producing cells in transplanted mice (<xref ref-type="fig" rid="fig5s2">Figure 5–figure supplement 2A</xref>). Transgenes delivered by adenovirus do not generally persist long term (<xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>), especially in dividing cells. Persistent transgene expression noted in our study likely results from the known ability of Neurog3 to induce cell cycle exit (<xref ref-type="bibr" rid="bib35">Miyatsuka et al., 2011</xref>), thereby preventing dilution of transgene-encoded factors by cell division. This precluded tests to demonstrate independence of the transgene-encoded factors for ongoing maintenance of the converted-cell phenotypes observed. We clarified this point in our revised manuscript.</p><p><italic>4) How sure are they that the starting population is pure ductal cells, given that CD133 is not exclusively expressed in ductal epithelium in the pancreas</italic>?</p><p>In human pancreas, CD133 is expressed exclusively in ductal cells and centroacinar cells, but not other cell types in pancreas, including acinar or islet endocrine cells. This has been shown by us (<xref ref-type="fig" rid="fig1">Figure 1C</xref>, <xref ref-type="fig" rid="fig1s1">Figure 1–figure supplement 1B</xref>) and by others (Immervoll et al., BMC cancer, 2008; Lardon et al., Pancreas, 2008). In addition, we assessed the purity of the ductal cell fraction collected by qRT-PCR and immunostaining of CD133<sup>+</sup> “sorted” cells (<xref ref-type="fig" rid="fig1">Figure 1D–F</xref>). Thus, we are confident that CD133-based sorting can efficiently eliminate contaminating native islet endocrine cells, permitting conclusions about conversion toward endocrine cell fate. We understand that centroacinar cells may still be included in our CD133<sup>+</sup> fraction. Therefore, we modified the phrase “purify ductal cells” to “fractionate ductal cells” in our revised manuscript.</p><p>Can they double sort with a general epithelial marker to further purify the population, given that CD133 only enriches 4-fold for sphere-forming cells?</p><p>Ductal cells already constitute large portion in unsorted cell population, ranging from 30% to 40% (<xref ref-type="bibr" rid="bib8">Bouwens and Pipeleers, 1998</xref>). Therefore, an average of 4-fold enrichment is expected even by achieving pure ductal cell isolations. Supporting this is our finding that the CD133-negative population is completely devoid of the sphere-forming cells (<xref ref-type="fig" rid="fig1">Figure 1B</xref>).</p><p>Reviewer #2:</p><p><italic>In general this work is very well done, with convincing images and clear data</italic>.</p><p>We thank the reviewer for this positive assessment.</p><p><italic>1) As CD133 is detected in centroacinar cells as well as in ductal cells (Immervoll et al JHC 2011), the authors should include additional acinar markers in their expression profile (</italic><xref ref-type="fig" rid="fig1"><italic>Fig 1 D</italic></xref><italic>) to exclude acinar contamination</italic>.</p><p>As requested, we performed qPCR with additional acinar marker carboxyl ester lipase (<italic>CEL</italic>) and included this data in <xref ref-type="fig" rid="fig1s1">Figure 1–figure supplement 1C</xref>.</p><p><italic>2) Please include the average percentage of CD133</italic><sup><italic>+</italic></sup> <italic>cells detected in human pancreas and the purity of the sorted populations</italic>.</p><p>As suggested, we calculated the average percentage of CD133<sup>+</sup> fraction as 32.73% (n=32). We included this in our revised manuscript. To assess the purity of the sorted populations, we included qRT-PCR and immunostaining data of the “sorted” CD133+ and CD133-negative population with various markers in <xref ref-type="fig" rid="fig1">Figure 1D–F</xref>.</p><p><italic>3)</italic> <xref ref-type="fig" rid="fig2"><italic>Figure 2A</italic></xref> <italic>please include co-staining of KRT19 and CD133</italic>.</p><p>As suggested, we performed co-staining of KRT19 and CD133 on spheres and included these new data in <xref ref-type="fig" rid="fig2s1">Figure 2–figure supplement 1D</xref>.</p><p><italic>4) The authors state “Time-lapse imaging revealed that spheres arose from single isolated CD133</italic><sup><italic>+</italic></sup> <italic>ductal cells”. However, this statement is not accurate unless the purity of the sort was 100%</italic>.</p><p>We agree with this point. To eliminate any possible confusion, we modified the sentence to “Time-lapse imaging revealed that spheres arose from single cells.”</p><p><italic>5) Does the percentage of CD133</italic><sup><italic>+</italic></sup> <italic>cells decrease in culture? What is the percentage of CD133 after 3 months in culture</italic>?</p><p>We used immunochemistry to quantify CD133 expression in cells comprising G1 and G7 spheres and quantified CD133<sup>+</sup> cells. 95.0 ± 3.0% and 98.1 ± 0.58% cells were positive for CD133 in G1 and G7 spheres, respectively (n=6 sections each, ≥ 1300 cells counted). Rare CD133-negative cells are likely due to the exclusion of cell apical regions during tissue section, where CD133 expression is localized. We included this data in <xref ref-type="fig" rid="fig2s1">Figure 2–figure supplement 1D</xref> and E, and evaluated these results in our discussion.</p><p><italic>6) The authors state that Insulin</italic><sup><italic>+</italic></sup> <italic>IPCs did not express other islet hormones, the authors should include co-staining of c-peptide with GCG, PP and Ghrelin in</italic> <xref ref-type="fig" rid="fig5"><italic>Figure 5D</italic></xref>.</p><p>We agree and now include appropriate images of the requested co-immunostaining results. We performed co-staining of C-peptide with GCG, PPY, and GHRL, and included GCG and GHRL co-staining images in a new <xref ref-type="fig" rid="fig5">Figure 5D</xref>. We found no PPY-positive cells from 7,487 GFP-positive cells we screened from two different samples. We found three GCG-positive cells out of 4,460 GFP-positive cells screened and none were co-labeled with C-peptide.</p><p>Reviewer #3:</p><p><italic>The claim that adult human pancreatic duct cells have a latent capacity for endocrine differentiation is correct, but only in this context of extremely strong transcription-factor-based enforced differentiation</italic>.</p><p>We did not know if this comment required a response but for completeness, we included the following paragraph in our manuscript discussion.</p><p>“During mouse pancreas development, Neurog3 levels surges in a subset of pancreatic progenitors located in primitive ducts...”</p><p><italic>I always wonder what MNP6 would do to a non-pancreatic cell type, and therefore if this effect is a specific latency of pancreatic duct cells or not (the paragraph starting ‘The transcription factors MafA, Neurog3 and Pdx1…’ is more important as a result if there is something these cells can do that is not ever seen with the 4-factor combination MNP6 (4TF) or 4TFM)</italic>.</p><p>Please see our response to the similar comment (#1) from the first reviewer.</p><p>In addition, in every experiment and analysis we performed, we included spheres (derived from CD133<sup>+</sup> pancreatic ducts) infected with control viruses (Ad-RFP or Ad-GFP) but otherwise identical conditions as with N, 4TF or 4TFM spheres. We did not detect evidence of endocrine cell conversion by qRT-PCR, immunohistochemistry, and Insulin C-peptide ELISA (data from these controls are all included and labelled appropriate a “N”, “4TF” or “4TFM”). Therefore, we concluded that spontaneous conversion does not occur at detectable frequency.</p><p><italic>Essential is the claim that the insulin-producing cells are mono-hormone-positive, but this is not shown in the paper. The authors refer to the</italic> <xref ref-type="fig" rid="fig5"><italic>Figure 5D</italic></xref> <italic>as the one showing no double-positivities, but no Gcg, PP, or Ghrl are shown here. Some more clarity on this aspect seems critical. Assays for Gcg and other hormones are either referred to as data not shown, or this aspect is stated but the figure does not have the supporting data. Gcg immunodetection was done on cells from N alone? Gcg, PP were tested on the 4-factor combinations</italic>?</p><p>We agree that the claim of mono-hormonal development is important to document. Please see also our response to a similar comment (#2) from the first reviewer. We were unable to detect mRNA or protein for <italic>GCG</italic> and <italic>PPY</italic> in Ad-RFP-Neurog3 infected spheres by qRT-PCR, now stated in the manuscript and in a revised <xref ref-type="fig" rid="fig4s1">Figure 4–figure supplement 1A</xref>. For 4TFM spheres, we performed co- staining of C-peptide with GCG, PPY, and Ghrelin, and included GCG and GHRL co-staining images in a new <xref ref-type="fig" rid="fig5">Figure 5D</xref>. We found no PPY-positive cells from 7,487 GFP-positive cells we screened from two different samples. We found three GCG-positive cells out of 4,460 GFP-positive cells screened and none were co-labeled with C-peptide. Thus we feel our claim of mono-hormonal insulin<sup>+</sup> cells is well documented.</p><p><italic>Some explanation of why spontaneous (that is, non-TF-enforced) differentiation to endocrine fates was ruled out</italic>.</p><p>We initially attempted to induce spontaneous differentiation of pancreatic ductal cells using systematic variations of culture conditions, but these efforts proved unsuccessful. This motivated us to induce Neurog3 to mimic the embryonic endocrine cell development explained above. We detailed this, as requested, in our revised manuscript.</p><p>In addition, as we described above, in every experiment and analysis we performed we included spheres (derived from pancreatic ducts) infected with control viruses (Ad-RFP or Ad-GFP) but otherwise identical conditions as with N, 4TF or 4TFM spheres. These controls served to monitor the possibility of spontaneous differentiation. Using the thorough analyses presented in our manuscript, we were unable to detect any sign of spontaneous endocrine cell differentiation in these controls. Therefore, we concluded that spontaneous conversion occurred at negligible frequency.</p><p><italic>Is Pax6 already expressed in the MNP-adnovirus cultures? If so, why is more needed</italic>?</p><p>We thank the reviewer for this good question: Pax6 mRNA level in MNP spheres was less than 0.03% of levels detected in primary human islets. This finding promoted us to include this critical factor in our cocktail. To document this result, we have now included Pax6 qRT-PCR data in <xref ref-type="fig" rid="fig4s1">Figure 4–figure supplement 1</xref> and in our revised manuscript.</p><p><italic>The part on “We sought new methods to mature the cells” (my words) reads strangely. It's the same method, just extending the time frame, I believe, and I would simplify this text</italic>.</p><p>In addition to the extended time frame, we constructed new adenovirus permitting simultaneous expression of three factors (Pdx1, MafA, and Pax6) in a single virus, to increase the chance of these 3 factors being expressed in single cell. Therefore, we used only 2 viruses for 4TF (as opposed to 4 individual viruses for 4V). We clarified this in our revised manuscript.</p><p><italic>Another comment here is that we revert to the MNP6 mixture (4TF), having just been told that Pdx1 can be removed without impact – why</italic>?</p><p>Even though exogenous PDX1 is not required for the conversion observed here, we found that exogenous PDX1 increased INS expression by two fold (<xref ref-type="fig" rid="fig4">Figure 4E</xref>). This motivated us to include PDX1 in our MAFA-PAX6-PDX1 virus (Ad-GFP-M6P) construction.</p><p><xref ref-type="fig" rid="fig2s1"><italic>Figure 2–figure supplement 1C</italic></xref> <italic>has an incorrect y-axis. 10,000 percent to the log(base10) is 4, correct. This graph needs altering – why not just “fold” for cell number? Related to this, Discussion asserts 3,000-fold, but this is just once? Up to 3,000-fold, and more typically xx-fold? Why do the cultures suddenly go bad at G7 (see text) – what happens – sudden apoptosis? #48 seems to be continuing even at G7 – please amend this text</italic>.</p><p>We thank the reviewer for finding this error and we have amended the data presentation appropriately.</p><p>As was shown in <xref ref-type="fig" rid="fig2s1">Figure 2–figure supplement 1C</xref>, we observed only one sample reached to G7. We agree with the reviewer and changed “>3,000-fold” to “up to 3,200-fold”.</p><p>We found the cells stopped growing and therefore could not be passaged in spheres after G7. We revised the text as suggested to “ductal cell expansion was not achieved, and spheres were not formed” in our manuscript.</p><p><italic>Does the 3-factor system (MNP) work in the authors' hands on mouse acini? And/or duct</italic>?</p><p>In preliminary studies, we have found that the MNP combination can induce insulin expression in cultured mouse ductal cells. If deemed important by the editors and reviewers, we are happy to include this unpublished data.</p><p><italic>Does the CD133 separation method include centroacinar cells (CAC) or not? What is the photograph in</italic> <xref ref-type="fig" rid="fig1"><italic>Figure 1</italic></xref> <italic>– ‘tip’ of duct dives out of plane of section before it is reached, and therefore the CAC cannot be seen in this panel? CAC could have a specific latency not seen in duct cells</italic>.</p><p>We also think this is an interesting point and included this in our discussion. Briefly, CD133 is expressed in CACs in human pancreas, so we cannot exclude the possibility that CACs are included in CD133 sorted cell fraction. Indeed this is highly likely. We were unable to test the presence of CACs in our sorted CD133<sup>+</sup> fraction due to a lack of specific “human” CAC marker. However, the sphere-forming efficiency approached 1 in 5 cells in our assay, and the frequency of CACs is thought to be much lower than this.</p><p>The idea that lineage-tracing methods are hard to apply in human cells should be stated, as everything else hinges on numerical arguments on cleanliness of cell separations, etc.</p><p>We have included a sentence in the Discussion to cover this point.</p><p><italic>chgA is an endocrine differentiation marker. This statement is meant to indicate that full-blown differentiation to the hormone-expressing state requires a substantial pulse of Ngn3. What about other hormones, even indicating non-pancreatic cell types</italic>?</p><p>We performed qRT-PCR with CCK (Cholecystokinin) and GAST (Gastrin) – hormones expressed in the intestine – in Neurog3-overexpressed spheres and were unable to measure any detectable mRNA. We included this in our manuscript.</p><p><italic>A major point is the longevity of the pulse of Ngn3 and the other factors achieved with these methods, and the detection of a program that runs from the endogenous loci with or without the continuous presence of N, MNP, MNP6 (4TF, 4TFM). Would this method pulse the cells or not, and is continuous presence of some of the factors preventing their full differentiation to the most mature state</italic>?</p><p>Expression of the exogenous factors persisted after two weeks of maturation period (<xref ref-type="fig" rid="fig5s3">Figure 5–figure supplement 3</xref>) and at least 5 months, evidenced by the GFP-positive insulin-producing cells in transplanted mice (<xref ref-type="fig" rid="fig5s2">Figure 5–figure supplement 2A</xref>). Thus, the current method does not permit 'pulse' expression of the factors. Transgenes delivered by adenovirus do not generally persist long term (<xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>), especially in dividing cells. Persistent transgene expression noted in our study likely results from the known ability of Neurog3 to induce cell cycle exit (<xref ref-type="bibr" rid="bib35">Miyatsuka et al., 2011</xref>), thereby preventing dilution of transgenes and transgene-encoded factors by cell division. This precluded tests to demonstrate independence of the transgene-encoded factors for ongoing maintenance of the converted-cell phenotypes observed. We also agree with the reviewer that it is possible the continuous presence of some of the factors may prevent full differentiation toward more physiologically-mature endocrine cells. However, we do not feel these limitations change the main conclusions of our study. We have included a discussion of these points in our revised manuscript.</p><p><italic>Title of the section starting ‘Although ELISA studies readily detected Proinsulin production by IPCs…’ reads odd to me: ‘genetic conversion’ reads as if the genome of the adult duct cells is being altered in some manner</italic>.</p><p>We agree with this reviewer’s point. We changed the phrase “genetic conversion” to “conversion”.</p><p><italic>Systematic removal of factors from MNP6 mixtures: Why can Pdx1 be removed without any impact</italic>?</p><p>It is unclear why exogenous PDX1 is not critical for ductal sphere conversion toward an endocrine fate. One possibility is that our findings indicate distinct transcription factor requirements of human ductal epithelium (compared to mouse acinar cells and human hepatocytes that require Pdx1 expression for conversion, like in <xref ref-type="bibr" rid="bib67">Zhou et al., 2008</xref>). We included a discussion of this point in our revised manuscript.</p><p><italic>Various ‘obvious’ markers were not tested, or the manuscript is incorrect in not showing such ‘easily pointed out’ questions. Pdx1 is produced, by immunofluorescence assay, within these induced beta cells? To normal levels? MafA/B, etc? Nkx6.1 was assessed, but the ‘dogmatic’ mature β-cell TF list should be addressed, at least</italic>.</p><p>We agree that MAFA and PDX1 are generally accepted adult β-cell markers in human. However, viral-expressed exogenous MAFA and PDX1 proteins from Ad-GFP-M6P adenovirus precluded us from detecting endogenous MAFA and PDX1 expression in IPCs. Based on this constraint, we used immunostaining to verify expression of other known ?-cell specific markers such as NKX6.1, IAPP, and PC1/3. We emphasized this point in our revised text.</p><p><italic>It does seem difficult to follow 4V, 4TF, 4TFM, MNP. Seems as if there is a mixed descriptor used for the same manipulation, at least sometimes; I suggest simply checking for a way of making the text fully consistent throughout</italic>.</p><p>We have striven to achieve consistent use of these acronyms throughout the text in our revised manuscript.</p></body></sub-article></article> |