Natalia Petersen,1 Frank Reimann,2 Sina Bartfeld,1 Henner F. Farin,1 Femke C. Ringnalda,1 Robert G.J. Vries,1

Stieneke van den Brink,1 Hans Clevers,1 Fiona M. Gribble,2 and Eelco J.P. de Koning1,3,4

Generation of L Cells in Mouseand Human Small IntestineOrganoids

Upon a nutrient challenge, L cells produceglucagon-like peptide 1 (GLP-1), a powerfulstimulant of insulin release. Strategies to augmentendogenous GLP-1 production include promotingL-cell differentiation and increasing L-cell number.Here we present a novel in vitro platform to generatefunctional L cells from three-dimensional cultures ofmouse and human intestinal crypts. We show thatshort-chain fatty acids selectively increase thenumber of L cells, resulting in an elevation of GLP-1release. This is accompanied by the upregulation oftranscription factors associated with the endocrinelineage of intestinal stem cell development. Thus,our platform allows us to study and modulate thedevelopment of L cells in mouse and human cryptsas a potential basis for novel therapeutic strategiesin patients with type 2 diabetes.Diabetes 2014;63:410–420 | DOI: 10.2337/db13-0991

Impairment of insulin secretion is a hallmark of type 2diabetes. New strategies in the treatment of type 2 di-abetes are based on the glucose-lowering effects of theintestinally produced hormone glucagon-like peptide 1(GLP-1), which augments glucose-dependent insulinrelease, improves b-cell survival, and promotes satiety(1–3). GLP-1–producing L cells are scattered in the in-testinal epithelium among enterocytes and other

secretory cells. They also produce GLP-2 and peptide YY.GLP-1 is released in response to ingested nutrients andis rapidly degraded by the enzyme dipeptidyl peptidase4. Current antihyperglycemic agents include inhibitorsof dipeptidyl peptidase 4, which enhance the bio-availability of endogenously secreted GLP-1 and GLP-1receptor agonists.

Alternatively, increasing the L-cell number to augmentGLP-1 secretion can be a useful therapeutic strategy.L cells are generated from stem cells at the base of in-testinal crypts. The intestinal stem cells proliferate andgive rise to transit-amplifying progenitor cells that sub-sequently differentiate (4). Enteroendocrine cells andcells from other secretory cell lineages, such as goblet andPaneth cells, originate from a common progenitor cell (5–7). Later in differentiation, endocrine cell progenitorsexpress neurogenin 3 (Ngn3) (8). Insight into the de-velopment of L cells, and the determination of factorsand downstream signaling pathways that drive L-celldifferentiation are hampered by the lack of an in vitrosystem that allows the study of L cells in their regular cellenvironment. Therefore, we applied a three-dimensionalintestinal crypt culture system that was developed re-cently at the Hubrecht Institute (9). In this system, in-testinal crypts are grown as self-renewing organoids thatcontinuously produce differentiated epithelial cells, in-cluding chromogranin (Chg)A–positive cells, similar tointestinal crypts in vivo (4,9,10). So far, it has not been

1Hubrecht Institute/KNAW and University Medical Centre Utrecht, Utrecht, theNetherlands2Cambridge Institute for Medical Research, Department of Clinical Biochemistry,Addenbrooke’s Hospital, Cambridge, U.K.3Department of Nephrology, Leiden University Medical Center, Leiden, theNetherlands4Department of Endocrinology, Leiden University Medical Center, Leiden, theNetherlands

Corresponding author: Natalia Petersen, n.petersen@hubrecht.eu, or EelcoJ.P. de Koning, e.koning@hubrecht.eu.

Received 27 June 2013 and accepted 7 October 2013.

This article contains Supplementary Data online at http://diabetes.diabetesjournals.org/lookup/suppl/doi:10.2337/db13-0991/-/DC1.

© 2014 by the American Diabetes Association. See http://creativecommons.org/licenses/by-nc-nd/3.0/ for details.

410 Diabetes Volume 63, February 2014

TECHNOLOGIC

ALADVANCES

established whether these ChgA-positive cells inorganoids are representative of L cells in vivo.

Here we show a continuous generation of mouse andhuman L cells in vitro, with a possibility to observe L-celldevelopment in real time and selectively modulate thegeneration of L cells by nutrient stimulation with short-chain fatty acids (SCFAs).

RESEARCH DESIGN AND METHODS

Animals

Animal experiments were conducted under the approvalof the animal care committee of the Royal NetherlandsAcademy of Arts and Sciences (permit HI 11.2503).Twelve-week-old C57BL/6 mice were purchased fromCharles River (L’Arbresle, France). Mice expressing yellowfluorescent protein (YFP) under transcriptional controlof the proglucagon promotor (GLU-Venus mice) (11)were bred on a C57BL/6 background in our animal fa-cility. Mice were fed regular chow ad libitum.

Human Tissue

Surgically resected intestinal tissues or endoscopic biopsysamples were obtained from the University MedicalCentre Utrecht, which was approved by the ethics com-mittee of the University Medical Centre Utrecht. In-formed consent was provided by all subjects.

Crypt Isolation

Mouse and human intestinal crypts were isolated frommouse and human jejunum, as described previously(9,12,13), and were seeded into Matrigel (BD Bio-sciences), in which they grew into organoids (9,13).Crypts were cultured in advanced Dulbecco’s modifiedEagle’s medium/F12 containing 100 units/mL penicillin/streptomycin, 10 mmol/L HEPES, 2 mmol/L Glutamax,supplements N2 (13) and B27 (13), and 50 ng/mLmurine or human epidermal growth factor (for mouseand human crypt culture, respectively) (all from LifeTechnologies); and 1 mmol/L N-acetylcysteine (Sigma-Aldrich) and murine Noggin and human R-spondin-1,both as 10% conditioned medium (13). The culture me-dium for human crypts was additionally supplementedwith 50% Wnt-3A conditioned medium (13), 10 nmol/Lgastrin (Sigma-Aldrich), 10 mmol/L nicotinamide (Sigma-Aldrich), 500 nmol/L inhibitor of transforming growthfactor-b type I receptor ALK5 kinase (A83-01) (TocrisBioscience), and 10 mmol/L p38 mitogen–activated pro-tein kinase inhibitor (SB202190) (Sigma-Aldrich) (13).The medium was refreshed every 2–3 days. Mouse andhuman colon crypts were cultured using the same me-dium composition as human jejunum. Analysis of humanand mouse organoids was performed on 1.5- or 2-month-old cultures. For passage, organoids were removed fromthe Matrigel, mechanically dissociated using a glass Pas-teur pipette, pelleted, and replated in fresh Matrigel in24-well plates. Mouse organoids were passaged every 5days with a 1:4 splitting ratio. Human organoids werepassaged every 10–14 days with a 1:8 split ratio. To test

the effect of SCFAs on L-cell differentiation, a combinationof acetate, propionate, and butyrate (5, 1, and 1 mmol/L,respectively) was added to the culture medium. These SCFAconcentrations were chosen based on earlier in vitro studies(14) and the ratios of these SCFAs in plasma and intestinallumen (15). For control mouse organoids, regular mediumwithout SCFAs was used. For dose testing in Supplemen-tary Fig. 2F, different concentrations of SCFA combinationwere used with a constant ratio of 5:1:1 for acetate/buty-rate/propionate, respectively.

To improve differentiation of human organoids duringSCFA testing, Wnt-3A, nicotinamide, A-83-01, and SB202190inhibitors were omitted (13). Human and mouse organo-ids were collected for analysis 48 h after SCFA addition.

Immunostaining and 5-Ethynyl-29-DeoxyuridineLabeling

For immunostaining, organoids were fixed in 4% para-formaldehyde, permeabilized with 0.3% Triton X-100,and blocked with 3% donkey serum. Organoids were in-cubated overnight with primary antibodies against GLP-1(Phoenix Pharmaceuticals), mucin (sc-15334; Santa CruzBiotechnology), lysozyme (Lyz1) (A0099; Dako), ChgA(sc-1488; Santa Cruz Biotechnology), or ChgC (sc-1491;Santa Cruz) at 4°C. Alexa Fluor 568 donkey anti-goat andAlexa Fluor 488 donkey anti-rabbit (Invitrogen) were usedas secondary antibodies. Images were acquired on a confo-cal laser-scanning microscope (SP5; Leica) using LAS soft-ware. The percentage of L cells in organoids was determinedbased on the number of L cells and DAPI-positive cells in3-mm optical slices from z-stacks with a distance of 3 mmbetween the slices. For 5-ethynyl-29-deoxyuridine (EdU)labeling, mouse organoids were incubated in 10 mmol/LEdU (Click-it; Invitrogen) for 30 min and human organoids2 h before fixation. The detection was performed accordingto the manufacturer’s protocol.

Quantitative PCR Analysis

Total RNA was extracted from organoids using Trizol(Invitrogen) and reverse-transcribed with a Fermentaskit. Quantitative real-time PCR was performed on a real-time PCR System (Bio-Rad) using SYBR green assays. Wetested GAPDH, HPRT, and beta-2 microglobulin (B2M) asendogenous control gene and found B2M to be moststable during organoid culture and passaging (data notshown). Gene expression of L-cell–specific functionalmarkers (11) was analyzed in sorted L cells fromorganoids, fresh villi, and crypts. Markers of intestinalcell types were analyzed in whole organoids (Table 1).Transcription factors associated with L-cell developmentwere tested in whole organoids and sorted L cells.

Villus and Crypt L-Cell Isolation and Fluorescence-Activated Cell Sorting

For comparison of primary and organoid-derivedL cells, freshly isolated small intestine crypts fromGLU-Venus mouse and small intestines of GLU-Venusmouse organoids after five to eight passages were

diabetes.diabetesjournals.org Petersen and Associates 411

Table 1—Quantitative RT-PCR primers

Gene Species Alias Forward Reverse

Beta-2 microglobulin Mouse B2m CTGGTGCTTGTCTCACTGAC GTTCAGTATGTTCGGCTTCC

Preproglucagon Mouse Gcg GCTTATAATGCTGGTGCAAG TTCATCTCATCAGGGTCCTC

Sodium-glucose linkedtransporter 1 Mouse Sglt1 CAAAGTGACCACTTCCAATG GTACCGTTGGAGGCTTCTTC

Glucokinase Mouse Gck TGAAGACGAAACACCAGATG GTCCAGGAAGTCAGAGATGC

Glut5, fructose transporter Mouse Slc2a5 CCAATATGGGTACAACGTAGC TTCTGTCGTAGTAGGTGTCATTG

Sodium channel, voltage-gated,type III, a-subunit Mouse Scn3a TCGATTCAGTGCCACCTC ATTCTTCGTCCAGTCAGGAG

K+ voltage-gated channel,subfamily S, 2 Mouse Kcns2 GCACACAGACAAGCACTCAC GGTCATGCTGGAGATGC

Calcium channel,voltage-dependent,P/Q type, a-1A subunit Mouse Cacna1a ACAACATCCTGTTTGCTGTG CAACCAGTTCCAAGTGTTCC

Free fatty acid receptor 2 Mouse Ffar2 ACCAGAGGAGAACCAGGTAG CGTGAGGATCAAGGAACTG

Secretin Mouse Sct CTGCTGCTCTCCAGTTCC CTCGCTGGTGAACATTCC

Tryptophan hydroxylase Mouse Tph TGGACAGCTGAGAGTCTTTG CAAAGGGCTTGACTTTGG

Gastric inhibitory polypeptide(glucose-dependentinsulinotropic peptide) Mouse Gip TGAAGACCTGCTCTCTGTTG CTGCGTACCTTGGACCTC

Chromogranin A Mouse ChgA TTCCATGCAGGCTACAAAG GTCTTTCCATCTCCATCCAC

Lysozyme 1 Mouse Lyz1 GGAATGGATGGCTACCGTGG CATGCCACCCATGCTCGAAT

Intestinal trefoil factor Mouse Itf CCTCTGGCTAATGCTGTTG CAGTCCACTCTGACATTTGC

Leucine-rich repeat containingG protein-coupled receptor 5 Mouse Lgr5 CTTTGACACACATTCCCAAG AAATTCTGTAGCGCTTCCTC

Prominin 1 Mouse CD133 (Prom1) ATGCAGGAGGAAGTGCTTG AGTCCTGGTCTGCTGGTTAG

Intestinal fatty acid binding protein Mouse I-Fabp CGGCACGTGGAAAGTAGACC AATGGTCCAGGCCCCAGTGA

Neurogenin 3 Mouse Ngn3 GCATGCACAACCTCAACTC TTTGTAAGTTTGGCGTCATC

Neurogenic differentiation factor 1 Mouse Neurod1 AGGTGGTACCTTGCTACTCC TGAAAGAGAAGTTGCCATTG

Aristaless related homeobox factor Mouse Arx GTTACCAGCTGGAGGAACTG GGCCTCTGTCAGGTCCAG

Forkhead box A2, transcriptvariant 1 Mouse Foxa1/2 GAGCCATCCGACTGGAG ATGTGTTCATGCCATTCATC

Beta-2 microglobulin Human B2M GCGCTACTCTCTCTTTCTGG GCTGGATGACGTGAGTAAAC

Preproglucagon Human GCG CAAGGCAGCTGGCAACGT CAAGGCAGCTGGCAACGT

Secretin Human SCT CACTCAGACGGGACGTTCAC ATGCTGTTCTCTGCGTCCTG

Tryptophan hydroxylase Human TPH TGCGGACTTGGCTATGAACT GGAATACGGTTCCCCAGGTC

Gastric inhibitory polypeptide(glucose-dependentinsulinotropic peptide) Human GIP GGCAGTGGGACTAGGAGAGA GGCTGCTCACCTTAGCATGA

Chromogranin A Human CHGA ACTCCGAGGAGATGAACGGA CTTGGAGAGCGAGGTCTTGG

Lysozyme 1 Human LYZ1 CGCTACTGGTGTAATGATGG TTTGCACAAGCTACAGCATC

Intestinal trefoil factor Human ITF, TFF3 CTTGCTGTCCTCCAGCTC CAGTCCACCCTGTCCTTG

Leucine-rich repeat containingG protein-coupled receptor 5 Human LGR5 GAGAAAGCATTTGTAGGCAAC ATCTCCCAACAAACTGGATG

Prominin 1 Human CD133 GTCCTGGGGCTGCTGTTTAT TCCACCACATTTGTTACAGCA

Intestinal fatty acid binding protein 2 Human I-FABP, FABP2 CACAGTCAAAGAATCAAGCAC TCGTTTCCATTGTCTGTCC

Neurogenin 3 Human NGN3 AGTTGGCACTGAGCAAGC AGTGCCGAGTTGAGGTTG

Neurogenic differentiation factor 1 Human NEUROD1 TGAGACTATCACTGCTCAGG CACTCTCGCTGTACGATTTG

Aristaless related homeobox factor Human ARX GTGTGGGCTGTCTCAGG CGACACCCAGCTTTCATC

Forkhead box A2, transcript variant 1 Human FOXA1/2 GGGAGCGGTGAAGATGGA TCATGTTGCTCACGGAGGAGTA

Primers were generated using Primer 3 software. Relative expression values were derived using the DDCT method with B2M as theendogenous reference.

412 In Vitro Platform for L-Cell Studies Diabetes Volume 63, February 2014

dissociated with 0.05% trypsin-EDTA (Life Technolo-gies) at 37°C into single cells, centrifuged in 4% FBS inPBS at 300g, and immediately sorted by flow cytom-etry. Villi were collected from intestinal fragmentsprior to crypt isolation, washed in PBS containing 10%FBS, and dissociated into single cells, as described forcrypt cell isolation. YFP-positive cells were separatedby flow cytometry as described previously (11).

Time-Lapse Imaging

YFP fluorescence and bright-field images of GLU-Venusmouse organoids in a glass-bottom 24-well plate weremade every 3 h during 3 days using a wide-field LeicaAF7000 microscope equipped with a thermostaticchamber with humidity control and 5% CO2.

GLP-1 Secretion Assay

Basic medium for static incubations contained Hanks’balanced salt solution (Life Technologies) supplementedwith 10 mmol/L HEPES, 0.1% fatty acid–free BSA, andno glucose, pH 7.4 (NaOH). Organoids from 24-wellplates were collected in 1.5-mL Eppendorf tubes (1 wellper tube) and incubated in the basic medium for 2 h ina thermomixer at 300 rotations per minute. Then orga-noids were washed and incubated in 50 mL of basicmedium containing 1 mg/mL diprotin A (Sigma-Aldrich)for 1 h. The supernatant was collected, and the organoids

were incubated in 50 mL of 10 mmol/L glucose in basicmedium with diprotin A for 1 h, followed by collection ofthe supernatant. Organoids were then lysed in CelLytic Mbuffer (Sigma-Aldrich). GLP-1 concentrations in theorganoid cell lysates and supernatants were determined byMulti-Species GLP-1 total ELISA (Millipore) and HumanTotal GLP-1 multi-array (Meso Scale Diagnostics). DNAwas extracted from the organoid cell lysate (16)and quantified using the PicoGreen kit (Invitrogen). GLP-1content was normalized to the DNA content of organoids.

Statistical Analysis

All data are expressed as the mean 6 SEM. One-wayANOVA with post hoc Tukey tests was used for thecomparison of dose testing for the SCFA combinationand the expression of genes associated with GLP-1 se-cretion in sorted L cells. For comparison of basal andstimulated GLP-1 secretion data, the Kolmogorov-Smirnov test was applied. A Student nonpaired t test wasused for all other comparisons. Significance level was setat P , 0.05.

RESULTS

L Cells Are Generated in Organoids Derived FromMurine and Human Intestinal Crypts

Small intestinal crypts formed organoids that consistedof several outwardly protruding crypt domains and

Figure 1—A: Mouse organoids embedded in Matrigel and cultured for 4 days. B: Human organoids in Matrigel, cultured for 10 days.L cells in villus (C) and crypt region (D) of mouse organoids and in human organoids (E ) are identified by GLP-1 immunostaining (green).Scale bars: 20 mm.

diabetes.diabetesjournals.org Petersen and Associates 413

a central villus domain (Fig. 1A and B and SupplementaryFig. 1A and B). In accordance with the cell composition ofcrypts in vivo, the crypt domain in organoids consisted ofa crypt base, containing stem cells and Paneth cells;a zone of dividing multipotent transit-amplifying cells;and differentiating cells, such as goblet cells and enter-oendocrine cells (16) (Supplementary Figs. 1A–F and 2B).The villus domain mostly consisted of absorptive andsecretory cells (17,18) and showed a spheroid or oval-shaped structure in mouse organoids (Fig. 1A) con-taining mostly nondividing cells (Supplementary Fig.1A). Human organoids formed fold-like extensions (Fig.1B), and the central part of the organoid often hadmany single dividing cells (Supplementary Fig. 1B).Human organoids showed mucin- and Lyz1-positivecells but no clearly defined Lyz1 granules or mucindroplets, characteristic for Paneth and goblet cells(Supplementary Fig. 1D and F). Mouse organoids had5.2 6 0.3 crypt domains per organoid 4 days after pas-sage and human organoids had 11.1 6 0.4 crypt domainsper organoid 10 days after passage. L cells, identified asGLP-1–immunoreactive cells, were mostly found in cryptregions as single cells that were polarized toward theexterior basal side of the organoids (Fig. 1C–E). In mouseorganoids, L cells constituted 0.5 6 0.05% of all orga-noid cells, which corresponds to 2.1 6 0.1 cells peraverage-sized organoid (ranging from 0 to 15 cells,depending on organoid size). The organoids generatedL cells at a relatively constant rate for 48–96 h aftersplitting (Supplementary Fig. 2F). In human organoids,GLP-1–positive cells constituted 0.1 6 0.001% of allorganoid cells, or 1.5 6 0.03 cells per organoid. Aftera 30-min EdU pulse, no double-positive cells for GLP-1and EdU were found in a total of 300 human or mouseorganoids (data not shown), indicating that mature L cellsdo not divide. Free fatty acid receptor 2 (FFAR2) appearedexclusively on GLP-1–positive cells (Supplementary Fig.2A). GLP-1–immunoreactive cells also expressed vesiclemarkers ChgA and ChgC (Supplementary Fig. 2B and C,respectively). These characteristics were still present inL cells from 6-month-old cultures of mouse organoids(data not shown).

GLU-Venus mice express YFP in intestinal L cellsunder the control of the proglucagon promoter (11).Intestinal organoid cultures from these mice provideda unique opportunity to monitor real-time L-cell de-velopment (Fig. 2A). The majority of L cells appearednear the crypt base and moved away from the crypt baseconcomitantly with a proliferative expansion of thetransit-amplifying compartment. The fluorescence dis-appeared 3–4 days (Fig. 2A) after first becoming detect-able, indicating the loss of glucagon promoter activity orL-cell death. The disappearance of the fluorescent signalis unlikely to reflect photobleaching, as the cells wereonly excited for 70 ms every 3 h. As.90% of labeled cellswere double-positive when Venus-expressing organoidswere immunostained for GLP-1 (data not shown), the

fluorescent time course is likely to reflect the turnover ofGLP-1–expressing cells.

To test whether in vitro–generated L cells are func-tionally mature, we used GLU-Venus mice to comparefluorescence-activated cell (FAC)-sorted primary L cellsfrom the small intestine and L cells from organoids after 6passages. Estimated by FAC sorting, the percentage ofL cells in the organoids was similar to that observed infresh small intestine crypts (Supplementary Fig. 2H) andwas in line with our calculations based on microscopy. Wecompared gene expression of specific functional markersin L cells isolated from organoids and from freshly pre-pared villi and crypts (Fig. 2B). Proglucagon gene expres-sion was higher in L cells from villi compared with L cellsfrom crypts and organoids (Fig. 2B). We found that Sglt1,Glut5, Gck, Ffar2, Kcnq1, Scn3a, Kir6.2, and Cacna1a ex-pression was maintained in organoid-derived L cells, in-dicating that L cells produced in our cultures retain themolecular profile of their counterparts in vivo and expressstimulus-secretion–coupling components for GLP-1release.

We further analyzed the expression of transcriptionfactors associated with L-cell development (8,19–23) inL cells from organoids and from freshly prepared crypts(Supplementary Fig. 3A–D). Organoid L cells had a simi-lar expression pattern to that of crypt-derived L cells andshowed reduced expression of Ngn3 compared withwhole organoids, consistent with its role as early endo-crine marker (8).

SCFAs Increase the Number of L Cells and GLP-1Secretion in Mouse and Human Small IntestineOrganoids

SCFAs are known to stimulate GLP-1 secretion, and wetested their effect on L-cell development. The addition ofSCFAs as a combination of 5 mmol/L acetate, 1 mmol/Lpropionate, and 1 mmol/L butyrate almost doubledthe number of L cells in mouse (Fig. 3A and B) andhuman organoids (Fig. 3C) within 48 h. We testedseveral concentrations of the combination of the threeSCFAs (Supplementary Fig. 2F) and found that lowerconcentrations of SCFAs did not increase L-cell num-ber. Concentrations of SCFAs higher than those usedabove did not result in a further increase in L-cellnumber (not significant, by one-way ANOVA). Pro-pionate and butyrate (but not acetate) in concen-trations .1 mmol/L had a toxic effect, resulting inchanges in organoid morphology and reduction ofgrowth (data not shown). Increasing the acetate levelsalone did not have an additional effect on L-cellnumbers (data not shown).

The increase in L-cell number was accompanied bya higher GLP-1 content in mouse and human organoids(Fig. 3D and E). To test whether SCFA treatment gen-erates nutrient-responsive L cells, we measured glucose-induced GLP-1 secretion. In SCFA-treated mouseorganoids, both basal and stimulated GLP-1 secretion

414 In Vitro Platform for L-Cell Studies Diabetes Volume 63, February 2014

were higher than in the control (Fig. 3F). Both controland SCFA-treated mouse organoids showed a twofoldincrease in GLP-1 release during stimulation with 10mmol/L glucose (Fig. 3F), indicating that this effectwas due to the increased number of L cells but not tothe ability to increase GLP-1 release during the nu-trient challenge. Human organoids showed very lowbasal secretion (Fig. 3G), consistent with fewer L cells.There was a twofold increase in GLP-1 secretion fromcontrol organoids upon glucose stimulation (Fig. 3G). Af-ter the SCFA treatment, both basal and stimulated GLP-1secretion from human organoids were increased (Fig. 3G).

To test the effect of SCFA on differentiation of theother cell lineages, we performed marker gene expressionanalysis (Fig. 4A and C and Table 1). In mouse and hu-man organoids, SCFA treatment had no effect on ex-pression of Lgr5 and CD133 (markers for stem cells and

early progenitors respectively) (18), ITF (goblet cells),Lyz1 (Paneth cells), or I-FABP (enterocytes), but the pan-endocrine cell marker ChgA was increased during SCFAtreatment. In accordance with our data on GLP-1 con-tent, Gcg expression was elevated in mouse and humanorganoids compared with the control. Gene expression ofthe enterochromaffin cell marker Tph or the K-cellmarker Gip did not change in mouse organoids afterSCFA treatment, indicating a specific induction of L-celldifferentiation (Fig. 4A). In human organoids, the ex-pression of SCT and TPH was increased (Fig. 4C). Next, toevaluate the potential of organoids to generate L cells, weanalyzed gene expression of the transcription factorsNgn3 (19,20), Neurod1 (21), Foxa1/2 (22), and Arx (23)associated with early and late L-cell development inorganoids. In mouse and human organoids, gene ex-pression of Neurod1 and Foxa1/2 was upregulated, but

Figure 2—A: Development of an L cell in an organoid crypt. A picture of the same organoid was taken at different time points duringculture. White arrows indicate the position of a group of Paneth cells at the crypt base. Scale bars: 20 mm. B: Gene expression in FAC-sorted primary L cells from crypts, villi, and organoid-derived L cells: Gcg, Sglt1, Ffar2, Glut5, Gck, Kir6.2, Scn3a, and Cacna1a. Data aregene expression determined by the DDCT method with B2m as the endogenous control presented as the mean6 SEM; n is the number ofexperiments from four to six samples from organoid-derived L cells and four samples from primary L cells from different mice. **P < 0.01,by one-way ANOVA with post hoc Tukey test.

diabetes.diabetesjournals.org Petersen and Associates 415

Figure 3—Effect of SCFA on L-cell numbers in small intestine organoids. A: Mouse SCFA-treated organoids (c and d) produce more GLP-1–positive cells than control organoids (a and b). Z-stack projections of 20 optical slices covering the entire organoid. Scale bar: 20 mm.B and C: The number of L cells increased after 48 h of SCFA treatment (5 mmol/L acetate, 1 mmol/L propionate, and 1 mmol/L butyrate).D and E: SCFAs increased the total GLP-1 content of organoids. F and G: SCFA-mediated increase in GLP-1 secretion. Data are pre-sented as the mean 6 SEM; n is the number of experiments from three independent platings for each series. *P < 0.05, **P < 0.005.

416 In Vitro Platform for L-Cell Studies Diabetes Volume 63, February 2014

the expression of Ngn3 or Arx was not (Fig. 4B and D). Sometranscription factors that regulate cell development are alsoexpressed after cell maturation and may be involved in themaintenance of the mature state of the cell. Therefore, wealso compared the expression of transcription factors in thepure L-cell fraction from control and SCFA-treated mouseorganoids and found no differences (Fig. 4E).

DISCUSSION

The main findings of our study are that L cells can becontinuously generated from murine and human in-testine in a three-dimensional culture system that allows

real-time studies of L-cell development. Using organoidculture of GLU-Venus mouse crypts, we were able toobserve the “birth” of L cells within a crypt environmentin vitro and study gene expression from in vitro–derivedL cells. The number of L cells can be selectively modu-lated by nutrient factors (SCFAs). This illustrates the useof this novel platform to investigate factors that modu-late the L-cell number and function in order to increaseendogenous GLP-1 production for the treatment ofpatients with type 2 diabetes.

The number of L cells in organoids was close to thatobserved in freshly isolated mouse small intestinal crypts

Figure 4—A and B: Expression of main intestinal cell-type markers in organoids after SCFA treatment: Gcg, Secr, Tph, Gip, ChgA, Lyz1,Itf, Lgr5, and CD133, and the enterocyte marker I-Fabp. C and D: Effect of SCFAs on transcription factors associated with L-cell de-velopment in whole organoids: NGN3; NEUROD1 (ND1), ARX, and FOXA1/2. Data are gene expression determined by the DDCT method,with B2m as an endogenous control presented as the mean 6 SEM; n is the number of experiments from four platings for mouseorganoids and three independent platings for human organoids. E: Expression of transcription factors associated with L-cell developmentin sorted L cells from control and SCFA-treated organoids. Data are relative gene expression determined by the DDCT method andpresented as the mean 6 SEM; n = 5 for control organoids from five platings and n = 3 for SCFA-treated organoids from two platings.*P < 0.05.

diabetes.diabetesjournals.org Petersen and Associates 417

and villi. The majority of L cells were in crypt domains ofthe organoids, similar to reports in freshly isolated tis-sue, where L cells are also found mainly in crypts andlower villi (24). It is assumed that in vivo the cells fromthe crypt region are being “pushed” toward the villusby dividing transit-amplifying cells and then shed fromthe surface of the villus. In the organoids, the “villus”space is smaller and is “shared” among several crypts (9),so the migration patterns of cells in organoids may notmimic the situation in intact intestine. While we did notobserve major expression differences for most L-cellmarkers in cells derived from murine organoids or freshtissue, proglucagon was expressed at significantly higherlevels in villus L cells. A possible explanation is an in-crease of proglucagon expression as L cells mature duringthe migration from the crypts to the villus top. Otherfactors predominantly present in the crypt environment,such as Wnt and Notch signaling, may suppress celldifferentiation. This might be of special importance inthe human organoid cultures, where differences in cul-ture medium may interfere with the establishment ofa distinct crypt-villus axis and may also directly affectenteroendocrine cell maturation consistent with the ob-served lower L-cell density.

L cells from our mouse and human organoid culturescoexpress neuroendocrine granule markers ChgA andChgC and the FFAR2, and are able to secrete GLP-1 inresponse to nutrient stimulation. The increase in GLP-1secretion during glucose stimulation was comparable tothat reported in previous studies on mouse primary L cells(11). Similar expression of genes associated with stimulussecretion coupling for GLP-1 release in primary L cells andL cells from organoids further indicates that the organoidsystem is an excellent model for the generation of func-tioning L cells. It has some advantages over the estab-lished systems because L cells in organoids are retained ina polarized environment and can be monitored over theirentire life span. In addition, organoids can be establishedfrom human tissue. Organoids thus have substantial po-tential for translational studies and drug testing, particu-larly for drugs targeting L-cell differentiation.

Earlier in vivo studies on animals reported that di-etary fiber increases GLP-1 levels (25–28) and colonicL-cell numbers (29,30). This effect was attributed to theproduction of SCFAs by intestinal microflora during thefermentation of fiber (30). SCFAs have been shown toelevate the proglucagon gene in rat intestine (28,31). Inour in vitro system, SCFAs enhanced L-cell differentia-tion in mouse and human small intestinal crypts. In-creased proglucagon gene expression, elevated GLP-1content, and glucose-stimulated increase in GLP-1 se-cretion in mouse and human organoids indicate thatL cells generated during SCFA treatment are functional.Importantly, SCFAs did not change the organoid growthpattern or expression of stem cell, goblet cell, and Panethcell markers. The effect of SCFAs was selective to L cellsin mouse organoids because pan-endocrine marker

ChgA gene levels increased, but markers of other endo-crine cell types (Sct, Tph, and Gip) did not change. Hu-man SCFA-treated organoids showed an upregulation ofTPH and SCT, indicating an additional effect on en-terochromaffin cells and S cells, respectively. Whilesome overlap in secretin and GLP-1 expression has beenobserved in mice (32,33), enterochromaffin cells arethought to derive through a lineage distinctive fromL cells, and we did not observe any colocalization ofserotonin and GLP-1 in human organoids (data notshown). It is thus possible that, at least in humanorganoids, SCFAs increase maturation to differententeroendocrine cell types. It should be rememberedthat L cells coexpress a number of different hormones,including cholecystokinin, GIP, and peptide YY, withGIP being more prevalent in the proximal small in-testine and peptide YY increasing toward the distal in-testine (32,34). Therefore, although an increase in L-cellsecretion may be beneficial for glucose control in type 2diabetes, the concomitant stimulation of a range of gutpeptides could lead to additive anorexia but also tounwanted side effects, for example, in overstimulationof the exocrine pancreas and the gallbladder. Dissectionof the pathway by which SCFAs enhance L-cell de-velopment could provide valuable knowledge on how toenhance naturally regulated GLP-1 production from Lcells in duodenum, jejunum, ileum, and colon. After EdUstaining, we found no dividing L cells, regardless ofwhether organoids were treated with SCFAs (data notshown). This indicates that existing L cells do notcontribute to increased L-cell numbers and that SCFAslikely affect endocrine progenitors in the organoids.Ngn3 defines the endocrine commitment of a secretorycell progenitor (8,19,20), and, consistent with that, wefound that the expression of Ngn3 is lower in matureL cells than in whole organoids that also contain apopulation of early endocrine-committed cells. SCFAsdid not change the gene expression of Ngn3 in wholeorganoids, but increased expression of transcriptionfactors Neurod1 and Foxa1/2, which are downstream ofNgn3 and have been specifically associated with L-celldevelopment (21,22). Based on this, we suggest thatSCFAs act on late enteroendocrine precursors of L cells,in which the expression of Ngn3 is already fading (20).SCFA treatment did not have additional effects on theexpression of transcription factors in sorted L cells.FFAR2 mediates the stimulatory effect of SCFAs onGLP-1 secretion (14); however, it remains to be estab-lished whether it is involved in L-cell differentiation.Because SCFAs are GLP-1 secretagogues, it is possiblethat persistent activation of existing L cells may stim-ulate the development of L cells in surrounding areas.

In conclusion, the self-renewing three-dimensionalintestinal crypt culture system allows the productionof functional L cells, and can be used to study thedevelopment of L cells and their changes in (patho)physiological conditions that can be modeled in vitro.

418 In Vitro Platform for L-Cell Studies Diabetes Volume 63, February 2014

It is a promising platform for compound screeningand studies on L-cell function. We demonstrate thatL-cell differentiation can be selectively increased bySCFAs. Further identification of the signaling mech-anisms induced by SCFAs may be a useful tool in ourpursuit for better treatments for type 2 diabetes andobesity.

Acknowledgments. The authors thank the following colleagues fromthe Hubrecht Institute: Stefan van der Elst for performing fluorescence-activated cell sorting; Jori Tip-Leenders for technical assistance; Benaissa ElHaddouti for animal care; Marc de Wetering for a sample of human colonorganoids; Anko de Graaff and the Hubrecht Imaging Center for supporting theimaging; and Harry Begthel for assistance with immunostaining. The authorsalso thank Chris van der Bent of the Department of Endocrinology, LeidenUniversity Medical Center, for performing the Mesoscale GLP-1 assay.

Funding. This study was partly funded by the Bontius Foundation.

Duality of Interest. No potential conflicts of interest relevant to thisarticle were reported.

Author Contributions. N.P. designed the study, researched the data,and wrote the manuscript. F.R. contributed to the study design and discussion.S.B. contributed a method for human organoid maintenance. H.F.F., R.G.J.V.,H.C., and F.M.G. contributed to the discussion. F.C.R. and S.v.d.B. contributedto cell culture optimization. E.J.P.d.K. designed the study and wrote themanuscript. All authors reviewed/edited the manuscript. N.P. is the guarantorof this work and, as such, had full access to all the data in the study and takesresponsibility for the integrity of the data and the accuracy of the data analysis.

References1. Elrick H, Stimmler L, Hlad CJ Jr, Arai Y. Plasma insulin responses to oral

and intravenous glucose administration. J Clin Endocrinol Metab 1964;24:1076–1082

2. Nauck MA, Homberger E, Siegel EG, et al. Incretin effects of increasingglucose loads in man calculated from venous insulin and C-peptide re-sponses. J Clin Endocrinol Metab 1986;63:492–498

3. Kieffer TJ, Habener JF. The glucagon-like peptides. Endocr Rev 1999;20:876–913

4. van der Flier LG, Clevers H. Stem cells, self-renewal, and differentiation inthe intestinal epithelium. Annu Rev Physiol 2009;71:241–260

5. Yang Q, Bermingham NA, Finegold MJ, Zoghbi HY. Requirement of Math1for secretory cell lineage commitment in the mouse intestine. Science2001;294:2155–2158

6. Jensen J, Pedersen EE, Galante P, et al. Control of endodermal endocrinedevelopment by Hes-1. Nat Genet 2000;24:36–44

7. Stanger BZ, Datar R, Murtaugh LC, Melton DA. Direct regulation of in-testinal fate by Notch. Proc Natl Acad Sci USA 2005;102:12443–12448

8. Lee CS, Perreault N, Brestelli JE, Kaestner KH. Neurogenin 3 is essential forthe proper specification of gastric enteroendocrine cells and the mainte-nance of gastric epithelial cell identity. Genes Dev 2002;16:1488–1497

9. Sato T, Vries RG, Snippert HJ, et al. Single Lgr5 stem cells build crypt-villusstructures in vitro without a mesenchymal niche. Nature 2009;459:262–265

10. Sato T, Clevers H. Growing self-organizing mini-guts from a single intestinalstem cell: mechanism and applications. Science 2013;340:1190–1194

11. Reimann F, Habib AM, Tolhurst G, Parker HE, Rogers GJ, Gribble FM.Glucose sensing in L cells: a primary cell study. Cell Metab 2008;8:532–539

12. Snippert HJ, van der Flier LG, Sato T, et al. Intestinal crypt homeostasisresults from neutral competition between symmetrically dividing Lgr5 stemcells. Cell 2010;143:134–144

13. Sato T, Stange DE, Ferrante M, et al. Long-term expansion of epithelialorganoids from human colon, adenoma, adenocarcinoma, and Barrett’sepithelium. Gastroenterology 2011;141:1762–1772

14. Tolhurst G, Heffron H, Lam YS, et al. Short-chain fatty acids stimulateglucagon-like peptide-1 secretion via the G-protein-coupled receptorFFAR2. Diabetes 2012;61:364–371

15. Cummings JH. Short chain fatty acids in the human colon. Gut 1981;22:763–779

16. Laird PW, Zijderveld A, Linders K, Rudnicki MA, Jaenisch R, Berns A.Simplified mammalian DNA isolation procedure. Nucleic Acids Res 1991;19:4293

17. Sato T, van Es JH, Snippert HJ, et al. Paneth cells constitute the niche forLgr5 stem cells in intestinal crypts. Nature 2011;469:415–418

18. Clevers H. Searching for adult stem cells in the intestine. EMBO Mol Med2009;1:255–259

19. Schonhoff SE, Giel-Moloney M, Leiter AB. Minireview: development anddifferentiation of gut endocrine cells. Endocrinology 2004;145:2639–2644

20. Jenny M, Uhl C, Roche C, et al. Neurogenin3 is differentially required forendocrine cell fate specification in the intestinal and gastric epithelium.EMBO J 2002;21:6338–6347

21. Naya FJ, Huang HP, Qiu Y, et al. Diabetes, defective pancreatic morpho-genesis, and abnormal enteroendocrine differentiation in BETA2/neuroD-deficient mice. Genes Dev 1997;11:2323–2334

22. Ye DZ, Kaestner KH. Foxa1 and Foxa2 control the differentiation of gobletand enteroendocrine L- and D-cells in mice. Gastroenterology 2009;137:2052–2062

23. Du A, McCracken KW, Walp ER, et al. Arx is required for normal enteroen-docrine cell development in mice and humans. Dev Biol 2012;365:175–188

24. Aiken KD, Kisslinger JA, Roth KA. Immunohistochemical studies indicatemultiple enteroendocrine cell differentiation pathways in the mouse prox-imal small intestine. Dev Dyn 1994;201:63–70

25. Kok NN, Morgan LM, Williams CM, Roberfroid MB, Thissen JP, DelzenneNM. Insulin, glucagon-like peptide 1, glucose-dependent insulinotropicpolypeptide and insulin-like growth factor I as putative mediators of thehypolipidemic effect of oligofructose in rats. J Nutr 1998;128:1099–1103

26. Zhou J, Martin RJ, Tulley RT, et al. Dietary resistant starch upregulatestotal GLP-1 and PYY in a sustained day-long manner through fermentationin rodents. Am J Physiol Endocrinol Metab 2008;295:E1160–E1166

27. Massimino SP, McBurney MI, Field CJ, et al. Fermentable dietary fiberincreases GLP-1 secretion and improves glucose homeostasis despiteincreased intestinal glucose transport capacity in healthy dogs. J Nutr1998;128:1786–1793

28. Reimer RA, McBurney MI. Dietary fiber modulates intestinal proglucagonmessenger ribonucleic acid and postprandial secretion of glucagon-likepeptide-1 and insulin in rats. Endocrinology 1996;137:3948–3956

29. Everard A, Lazarevic V, Derrien M, et al. Responses of gut microbiota andglucose and lipid metabolism to prebiotics in genetic obese and diet-induced leptin-resistant mice. Diabetes 2011;60:2775–2786

30. Kaji I, Karaki S, Tanaka R, Kuwahara A. Density distribution of free fattyacid receptor 2 (FFA2)-expressing and GLP-1-producing enteroendo-crine L cells in human and rat lower intestine, and increased cellnumbers after ingestion of fructo-oligosaccharide. J Mol Histol 2011;42:27–38

diabetes.diabetesjournals.org Petersen and Associates 419

31. Tappenden KA, Drozdowski LA, Thomson AB, McBurney MI. Short-chainfatty acid-supplemented total parenteral nutrition alters intestinal structure,glucose transporter 2 (GLUT2) mRNA and protein, and proglucagon mRNAabundance in normal rats. Am J Clin Nutr 1998;68:118–125

32. Habib AM, Richards P, Cairns LS, et al. Overlap of endocrine hormoneexpression in the mouse intestine revealed by transcriptional profiling andflow cytometry. Endocrinology 2012;153:3054–3065

33. Rindi G, Ratineau C, Ronco A, Candusso ME, Tsai M, Leiter AB. Targetedablation of secretin-producing cells in transgenic mice reveals a commondifferentiation pathway with multiple enteroendocrine cell lineages in thesmall intestine. Development 1999;126:4149–4156

34. Habib AM, Richards P, Rogers GJ, Reimann F, Gribble FM. Co-localisationand secretion of glucagon-like peptide 1 and peptide YY from primarycultured human L cells. Diabetologia 2013;56:1413–1416

420 In Vitro Platform for L-Cell Studies Diabetes Volume 63, February 2014