Leveraging Temporal Wnt Signal for Efficient Differentiation of Intestinal Stem Cells in an Organoid Model

Abstract

The homeostasis of the intestinal epithelium heavily relies on the self-renewal and differentiation of intestinal stem cells (ISCs). Although the orchestration of these processes by signaling pathways such as the Wnt, BMP, Notch, and MAPK signals has been extensively studied, the dynamics of their regulation remains unclear. Our study explores how the Wnt signaling pathway temporally regulates the differentiation of ISCs into various cell types in an intestinal organoid system. We report that the duration of Wnt exposure following Notch pathway inactivation significantly influences the differentiation direction of intestinal epithelial cells toward multiple secretory cell types, including goblet cells, enteroendocrine cells (EECs), and Paneth cells. This temporal regulation of Wnt signaling adds another layer of complexity to the combination of niche signals that govern cell fate. By manipulating this temporal signal, we have developed optimized protocols for the efficient in vitro differentiation of ISCs into EECs and goblet cells. These findings provide critical insights into the dynamic regulation of ISC differentiation and offer a robust platform for future investigations into intestinal biology and potential therapeutic applications.

Introduction

The homeostasis of the intestinal epithelium is driven by actively renewing intestinal stem cells (ISCs) located at the crypt base [1,2]. These stem cells undergo self-renewal and differentiation processes, continuously producing various differentiated cell types essential for maintaining the integrity and function of the intestinal epithelium [3,4]. This includes absorptive enterocytes, secretory goblet cell, enteroendocrine cell (EEC), Paneth cell, and tuft cell, each with unique functional roles [5]. Goblet cells, for example, secrete mucins to create a protective layer over the intestinal epithelium [6,7], while EECs sense luminal content and release numerous hormones that regulate digestive processes [8 –10]. Paneth cells also play a crucial role in supporting the self-renewal of ISCs [11,12] and provide an additional protective layer for the intestinal epithelium by secreting antibacterial peptides [13,14].

The self-renewal and differentiation of ISCs are tightly regulated by a combination of signaling pathways that form distinct gradients along the crypt–villus axis [15 –20]. The dynamic equilibrium of these pathways, notably the Wnt, BMP, Notch, and MAPK signals, orchestrates the intricate self-renewal and differentiation processes of ISCs, ultimately guiding the cell fate specification toward multiple functional mature cell types [19,21 –24]. Specifically, Notch signaling regulates the specification of absorptive versus secretory cell types, where Notch pathway inactivation induces secretory cell differentiation [22,25].

These cellular processes of the intestinal epithelium can be reproduced in vitro using an intestinal organoid system [26 –28]. This system, generated from isolated crypts or single ISCs in a 3D environment, recapitulates the cellular constitution, structural organization, and function of the intestinal epithelium [29,30]. Moreover, the intestinal organoid model captures the dynamic self-renewal and differentiation processes in the intestinal epithelium, offering a physiologically relevant model to explore the regulation of cellular behavior of intestinal epithelial cells [29,31]. Past studies have leveraged this system to dissect the signals and their combinations in differentiating intestinal epithelial cells [22,32]. These include the interplay between Wnt and Notch signals [22], the role of the MAPK pathway in EEC differentiation [24], and the influence of the BMP gradient in EEC subtype specification [33].

While the core signals governing intestinal epithelial differentiation have been defined, how these pathways dynamically regulate cell fate is yet to be fully understood. In this study, we further explored the signals that control the differentiation of intestinal epithelial cell types using the intestinal organoid system. We uncovered that the Wnt signal regulates the specification of secretory cell types in a temporally regulated manner, where the duration of Wnt exposure following Notch inactivation dramatically affects the differentiation direction of intestinal epithelial cell types toward multiple secretory cell types, including goblet cells, EECs, and Paneth cells. This adds another layer of regulation over the combination of niche signals. By integrating this temporal Wnt signal with other niche signals, we developed an optimized and efficient differentiation protocol to generate EECs and goblet cells. This work enhances our understanding of intestinal biology and provides a robust platform for future research.

Materials and Methods

All experiments involving animal tissues in this study were performed in accordance with guidelines approved by the Laboratory Animal Ethics Committee of Tongji University (approval no.: TJAB04720101).

Mouse intestinal crypt isolation and culture

Crypts were isolated as previously described [26]. Briefly, the proximal half of the small intestine from 6- to 12-week-old C57BL/6J mice was harvested, opened longitudinally, and washed three times with cold Dulbecco's phosphate-buffered saline (DPBS) to remove luminal content. The intestine tissue was then cut into pieces with scissors and washed with cold DPBS 5–10 times by pipetting up and down using a 10 mL pipette. Tissue fragments were incubated with 2 mM ethylenediaminetetraacetic acid (EDTA) in DPBS on ice for 30 min. After the removal of EDTA, tissue fragments were washed with DPBS to release crypts. The released crypts were collected and passed through a 70-μm cell strainer. Isolated crypts were embedded in Matrigel (Corning) and plated at the center of wells in 48-well plates.

The basal medium contained advanced Dulbecco's modified Eagle's medium (DMEM)/F12 (Gibco) supplemented with B-27 (Gibco), 10 mM HEPES (Corning), 2 mM GlutaMAX (Gibco), penicillin–streptomycin (Gibco), and 1 mM N-acetyl-l-cysteine (Sigma). Crypt culture medium was made by supplementing the basal medium with 50 ng/mL recombinant murine epidermal growth factor (EGF) (PeproTech), 2 μM DMH1 (ApexBio), 10% R-spondin1 conditioned medium, 4 μM CHIR99021 (CHIR, ApexBio), and 1 mM valproic acid (VPA or V, Sigma). Cells were passaged every 4 days and typically used 2 days after passage.

Direct differentiation of small ISCs

For cell differentiation experiments, the cell culture medium was removed and replaced with a medium containing growth factors and small molecules. The medium was changed every 1–2 days depending on the differentiation condition used.

EEC differentiation was as follows: recombinant murine EGF (50 ng/mL), DMH1 (2 μM), DAPT (10 μM; TargetMol), CHIR (4 μM) for 2 days, and then replace CHIR with Wnt-C59 (5 μM; ApexBio) and add PD0325901 (1 μM; ApexBio) for 2 days.

Goblet cell differentiation was as follows: recombinant human fibroblast growth factor (FGF)-basic (FGF-2, 50 ng/mL; PeproTech), DMH1 (2 μM), DAPT (10 μM; TargetMol), CHIR (4 μM) for 1 day, and then replace CHIR with Wnt-C59 (5 μM) for 3 days.

RNA extraction and reverse transcription-quantitative polymerase chain reaction

Organoids or differentiated cells were harvested and washed with ice-cold DPBS before lysis in the TRIzol reagent (TransGen Biotech). Total RNA was isolated according to the manufacturer's instructions. Up to 1 μg of total RNA was used for cDNA synthesis using the NovoScriptPlus Reverse Transcriptase kit (NovoProtein). Quantitative polymerase chain reaction (qPCR) was performed using the ChamQ Universal SYBR qPCR Master Mix (Vazyme) on a CFX384 Real-Time System (BioRad). Relative gene expression levels were calculated using the ΔΔCt method. A list of the primer sequences for qPCR is provided in Table 1.

Table 1.

Primers Used for Real-Time Polymerase Chain Reaction Analysis

Gene	Sequence
Hprt	F-TCAGTCAACGGGGGACATAAA
Hprt	R-GGGGCTGTACTGCTTAACCAG
Clca1	F-ACTAAGGTGGCCTACCTCCAA
Clca1	R-GGAGGTGACAGTCAAGGTGAGA
Chga	F-CAGCTCGTCCACTCTTTCCG
Chga	R-CCTCTCGTCTCCTTGGAGGG
Ngn3	F-AGTGCTCAGTTCCAATTCCAC
Ngn3	R-CGGCTTCTTCGCTTTTTGCTG
Neurod1	F-ATGACCAAATCATACAGCGAGAG
Neurod1	R-TCTGCCTCGTGTTCCTCGT
Lyz1	F-GGAATGGATGGCTACCGTGG
Lyz1	R-CATGCCACCCATGCTCGAAT
Muc2	F-AGGGCTCGGAACTCCAGAAA
Muc2	R-CCAGGGAATCGGTAGACATCG
Atoh1	F-GCTGTGCAAGCTGAAGGG
Atoh1	R-TCTTGTCGTTGTTGAAGG
Sst	F-GACCTGCGACTAGACTGACC
Sst	R-CCAGTTCCTGTTTCCCGGTG
Cck	F-GAAGAGCGGCGTATGTCTGT
Cck	R-CCAGAAGGAGCTTTGCGGA
Gip	F-AACTGTTGGCTAGGGGACAC
Gip	R-TGATGAAAGTCCCCTCTGCG
Gcg	F-CTTCCCAGAAGAAGTCGCCA
Gcg	R-GTGACTGGCACGAGATGTTG
Pyy	F-ACGGTCGCAATGCTGCTAAT
Pyy	R-GACATCTCTTTTTCCATACCGCT
Sct	F-GACCCCAAGACACTCAGACG
Sct	R-TTTTCTGTGTCCTGCTCGCT
Tph1	F-AACAAAGACCATTCCTCCGAAAG
Tph1	R-TGTAACAGGCTCACATGATTCTC
Sox9	F-GAGCCGGATCTGAAGAGGGA
Sox9	R-GCTTGACGTGTGGCTTGTTC
Alpi	F-AGGATCCATCTGTCCTTTGG
Alpi	R-ACGTTGTATGTCTTGGACAG
Gfi1	F-AGAAGGCGCACAGCTATCAC
Gfi1	R-GGCTCCATTTTCGACTCGC
Spdef	F-AAGGCAGCATCAGGAGCAATG
Spdef	R-CTGTCAATGACGGGACACTG
Klf4	F-GTGCCCCGACTAACCGTTG
Klf4	R-GTCGTTGAACTCCTCGGTCT
Tff3	F-TTGCTGGGTCCTCTGGGATAG
Tff3	R-TACACTGCTCCGATGTGACAG
Lgr5	F-ACCCGCCAGTCTCCTACATC
Lgr5	R-GCATCTAGGCGCAGGGATTG
Myc	F-ATGCCCCTCAACGTGAACTTC
Myc	R-GTCGCAGATGAAATAGGGCTG
Axin2	F-TGACTCTCCTTCCAGATCCCA
Axin2	R-TGCCCACACTAGGCTGACA
Jun	F-CCTTCTACGACGATGCCCTC
Jun	R-GGTTCAAGGTCATGCTCTGTTT
Lef1	F-TGTTTATCCCATCACGGGTGG
Lef1	R-CATGGAAGTGTCGCCTGACAG

Flow cytometry

Differentiated cell colonies were dissociated from the Matrigel with TrypLE (Gibco) and dissociated into single cells by repeated pipetting using a P1000 pipette tip. The digestive process was stopped by dilute samples in advanced DMEM/F12 containing 20% fetal bovine serum (FBS; Vistech) and then centrifuged at 800 relative centrifugal force (RCF) for 5 min at 4°C. To mark dead cells, samples were incubated in 4′,6-diamidino-2-phenylindole (DAPI, 1:1,000; ApexBio) diluted in 2% FBS/2 mM EDTA/DMEM for 20 min at room temperature, then centrifuged at 800 RCF for 5 min at 4°C and washed in 2% FBS/2 mM EDTA/DMEM for three times. Cells were then incubated in 4% paraformaldehyde (Sangon Biotech) for 10 min at room temperature, washed with 2% bovine serum albumin (BSA; Sigma)/DPBS, and then permeabilized in DPBS containing 0.5% Triton X-100 (Sigma), 0.5% Tween 20 (Sigma), and 2% BSA for 1 h at room temperature.

Cells were washed with ice-cold DPBS and filtered into fluorescence-activated cell sorting (FACS) tubes through a 40 μm cell strainer (Falcon). FACS was performed on a BD FACSVerse (BD Biosciences), and data were analyzed in FlowJo. The gating strategy was consistent throughout this study (Supplementary Fig. S1A).

Immunostaining

Immunofluorescence was performed as previously described. Briefly, organoids were released from Matrigel with DPBS, fixed with 4% paraformaldehyde (Sangon Biotech) for 10 min at room temperature, and permeabilized in DPBS containing 0.5% Triton X-100 (Sigma), 0.5% Tween 20 (Sigma), and 2% BSA(Sigma) for 1 h at room temperature. Samples were blocked and incubated in a blocking buffer (0.1% BSA, 0.5% Tween 20 in DPBS) containing primary antibodies at 4°C overnight. Primary antibodies used in this study were as follows: rabbit anti-lysozyme (1:500; Invitrogen), rabbit anti-mucin 2 (1:200; ProteinTech), mouse anti-Chr-A (1:100; Santa Cruz), rat anti-somatostatin (1:200; Millipore), mouse anti-glucagon (1:200; Santa Cruz), rabbit anti-Neurod1 (1:200; Abcam), and mouse anti-Klf4 (1:100; Santa Cruz).

Samples were washed and incubated with secondary antibodies (1:1,000; Invitrogen) for 1 h at room temperature, counterstained with DAPI (1 μg/mL; ApexBio), imaged on a Zeiss LSM880 or Olympus IX73, and analyzed in ImageJ. Alkaline phosphatase activity was detected using a commercial kit (Beyotime) according to the manufacturer's instructions. Specific markers for each cell type are summarized in Table 2.

Table 2.

Annotation of Genes

Gene	Gene name	Description
Ngn3	Neurogenin 3	Marker of intestinal secretory progenitor cell
Neurod1	Neuronal differentiation 1	Marker of late secretory progenitor cell and EEC
Sox9	SRY-box transcription factor 9	Marker of ISC and progenitor cell
Atoh1	Atonal bHLH transcription factor 1	Marker of intestinal secretory progenitor cell
Chga	Chromogranin A	Typical marker of EEC
Mc4r	Melanocortin 4 receptor	Typical marker of EEC
Nts	Neurotensin	Typical marker of EEC
Nkx2-2	NK2 homeobox 2	Typical marker of EEC
Isl1	ISL LIM homeobox 1	Typical marker of EEC
Gcg	Glucagon	Typical marker of EEC subtype (L cell)
Sst	Somatostatin	Typical marker of EEC subtype (D cell)
Cck	Cholecystokinin	Typical marker of EEC subtype (I cell)
Gip	Gastric inhibitory polypeptide	Typical marker of EEC subtype (K cell)
Pyy	Peptide YY	Typical marker of EEC subtype (L cell)
Sct	Secretin	Typical marker of EEC subtype (S cell)
Tph1	Tryptophan hydroxylase 1	Typical marker of EEC
Muc2	Mucin2	Typical marker of goblet cell
Clca1	Chloride channel accessory 1	Typical marker of goblet cell
Tff3	Trefoil factor 3	Typical marker of goblet cell
Gfi1	Growth factor independent 1 transcriptional repressor	Typical transcription factor regulating goblet cell differentiation
Spdef	SAM pointed domain containing ETS transcription factor	Typical transcription factor regulating goblet cell differentiation
Klf4	KLF transcription factor 4	Typical transcription factor regulating goblet cell differentiation
Alpi	Alkaline phosphatase	Typical marker of enterocyte
Lyz1	Lysozyme 1	Typical marker of Paneth cell
Tert	Telomerase reverse transcriptase	Typical marker of ISC
Ascl2	Achaete-scute family bHLH transcription factor 2	Typical marker of ISC
Olfm4	Olfactomedin 4	Typical marker of ISC
Cdca7	Cell division cycle associated 7	Typical marker of ISC
Lgr5	Leucine-rich repeat containing G protein-coupled receptor 5	Typical marker of ISC

EECs, enteroendocrine cells; ISCs, intestinal stem cells.

GLP-1 secretion assay

GLP-1 secretion assay was performed as previously described [34]. Clones from 48-well plates were collected in 1.5-mL tubes and incubated in basal medium [Hanks' balanced salt solution (Life Technologies) supplemented with 10 mM HEPES, 0.1% fatty acid-free BSA, and no glucose, pH 7.4] for 2 h in a thermomixer at 300 rotations per min. Clones were then washed and incubated in 100 μL basal solution containing 1 mg/mL Diprotin-A (Sigma Aldrich) for 1 h. The supernatant was collected and the clones were further incubated in 100 μL basal solution containing 1 mg/mL Diprotin A and 10 mM glucose for 1 h. The supernatant was collected and clones were then lysed in CelLytic M buffer (Sigma-Aldrich). GLP-1 concentration was determined by Multi-Species GLP-1 total ELISA (Millipore) and DNA was quantified using the PicoGreen Kit (Invitrogen).

Library preparation for RNAseq

Total RNA of each sample was extracted using the TRIzol reagent (TransGen Biotech), quantified and qualified by the Agilent 2100 Bioanalyzer (Agilent Technologies), NanoDrop (Thermo Fisher Scientific), and 1% agarose gel. One microgram of total RNA with an RNA integrity number value above 6.5 was used for the following library preparation. Next-generation sequencing library preparations were constructed according to the manufacturer's protocol of VAHTS mRNA-seq V3 Library Prep Kit for Illumina (NR611). Sequencing was carried out on a NovaSeq6000 System (Illumina).

RNAseq data analysis

Quality control of the raw sequencing counts was initially performed using FastQC (v0.12.1) to assess metrics such as base quality, sequence length, and adapter contamination. After the QC analysis, Illumina-specific adapters and low-quality bases were trimmed using Trim Galore (v0.6.10) using default parameters. The cleaned reads were then aligned to the Mus musculus reference genome (mm10) using HISAT2 (v2.2.1). Postalignment, the sequence alignment/map (SAM) files were converted into binary alignment/map format and sorted using SAMtools (v1.13). Gene quantification was executed using FeatureCounts (v2.0.1) to generate a count table based on the aligned reads. The count data underwent normalization using variance-stabilizing transformation methods implemented in DESeq2. Differentially expressed genes (DEGs) between experimental conditions were identified based on adjusted P values and log₂-fold change, following false discovery rate correction. The results were visualized using volcano plots constructed with ggplot2.

To interpret the biological relevance of the DEGs, Gene Ontology (GO) enrichment analysis was conducted using the clusterProfiler package. Significance of the enriched terms was determined based on adjusted P values following the Benjamini–Hochberg correction.

Statistical analysis

Data are presented as mean ± standard deviation. Statistical analysis was performed in GraphPad Prism software version 7.0. Two-tailed Student's t-test, one-way analysis of variance (ANOVA) with Dunnett's multiple comparisons test, and one-way ANOVA with Tukey's multiple comparisons test were used to determine the statistical significance. The significance levels are ****P < 0.0001; ***P < 0.001; **P < 0.01; and *P < 0.05.

Data and code availability

The raw sequence data reported in this article have been deposited in the Genome Sequence Archive [35] in the National Genomics Data Center [36], China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA: CRA013070) that are publicly accessible at (https://ngdc.cncb.ac.cn/gsa). All the custom codes written in R are available from the corresponding authors upon reasonable request.

Results

Secretory cell differentiation induced by a combination of niche signals

Mouse ISCs cultured with EGF, Noggin, and R-Spondin1 (ENR condition) form organoids containing multiple spontaneously differentiated cell types, including enterocytes, goblet cells, EECs, and Paneth cells [26]. Adding CHIR99021 and VPA (CV) maintains ISCs, producing cultures rich in ISCs but lacking differentiated cells [22]. We assessed combinations of niche signals on ISC differentiation, including EGF (E), BMP inhibitor Noggin (N), Wnt activator CHIR99021 (CHIR) and R-Spondin1 (R), Wnt inhibitor Wnt-C59 (C59), HDAC inhibitor VPA (V), and Notch inhibitor DAPT (D), for their ability to induce secretory cell differentiation.

As shown in previous studies, DAPT markedly increased differentiation into secretory cells [22,25], including EEC and goblet cell, as indicated by the elevated expression of markers such as Chga, Gcg, and Clca1 (Fig. 1A). VPA inhibited their differentiation (Fig. 1A). Prior studies suggest that Wnt activation induces Paneth differentiation, while inhibition promotes goblet differentiation, although both yield suboptimal EEC differentiation (Fig. 1B) [22]. Remarkably, the omission of Wnt modulators (EN.D condition) induces higher EEC differentiation than EN.D.C59 or EN.D.CHIR (Fig. 1A–D), as demonstrated by immunostaining of the EEC marker Chga (Fig. 1B) and flow cytometry analysis (Fig. 1C, D, and Supplementary Fig. S1A). Removing EGF decreased viability, and so, it was retained. Under EN.D, cells differentiated to goblet cells, EECs, Paneth cells, and enterocytes (Supplementary Fig. S1B).

FIG. 1.

Combinations of small molecules promote secretory lineage differentiation. (A) RT-qPCR quantification of Chga, Gcg, and Clca1 expression in multiple differentiation conditions. E, EGF; N, DMH1; R, R-spondin CM; C59, Wnt-C59; V, VPA; D, DAPT. One-way ANOVA with Dunnett's multiple comparisons test between each condition with ENR condition; Data are represented as the mean ± SD (n = 3); the significance levels are ****P < 0.0001; ***P < 0.001; **P < 0.01; and *P < 0.05. (B) Confocal images showing immunostaining of Chga expression in cells differentiated for 4 days in multiple conditions as indicated. Scale bars, 50 μm. (C) Representative flow cytometry plots of Chga⁺ cells from cells differentiated in multiple conditions as indicated. (D) Quantification of percentage of Chga⁺ cells in conditions as indicated, n = 3 samples for each condition. One-way ANOVA with Tukey's multiple comparisons test; Data are represented as the mean ± SD; ****P < 0.0001; **P < 0.01; and *P < 0.05. (E) Representative confocal images of Gcg and Sst in cells differentiated in EN.D condition for 4 days. Scale bars, 50 μm. (F) Time course analysis of Ngn3, Neurod1, and Chga expression for cells in differentiation conditions as indicated. ANOVA, analysis of variance; RT-qPCR, reverse transcription-quantitative polymerase chain reaction; SD, standard deviation.

We also identified Sst- or Gcg-positive EEC subtypes, although at a lower frequency (Fig. 1E). Tracking the expression of EEC transcription factors Ngn3 and Neurod1 over time showed that the condition without Wnt modulator (EN.D) exhibited elevated Chga after day 4 (Fig. 1F) but lower Ngn3 than the condition with Wnt activator (EN.D.CHIR, Fig. 1F), while C59 rapidly downregulated Ngn3 after day 1 (Fig. 1F). Conversely, the EN.D condition induced a higher expression of transcription factors important for goblet cell differentiation, such as Atoh1, Gfi1, and Spdef compared with the EN.D.C59 condition. However, the EN.D condition did not induce the higher expression of goblet cell differentiation markers (Clca1 and Muc2) than the condition with Wnt inhibitor (EN.D.C59) (Supplementary Fig. S1C). This discrepancy suggests potential temporal regulation by the Wnt signal.

Temporal Wnt signal influences secretory cell fate determination

In the intestinal crypt, cells originating from ISCs continuously migrate along the crypt–villus axis, simultaneously altering cell fate. We propose that cell fate is not only influenced by combinations of static niche signals but also by dynamic niche signals that vary spatiotemporally within the crypt. Among these niche signals, the Wnt signal is crucial in modulating secretory cell fate specification. To assess this, we further explored the temporal influence of Wnt on secretory cell specification following Notch inhibition. This was done by modulating the duration of Wnt activation after DAPT addition (Fig. 2A).

FIG. 2.

The effect of temporal Wnt signal on secretory cell fate determination. (A) ISCs were initially cultured in EN.D.CHIR condition. At various time points as indicated, they were then transitioned to the EN.D.C59 condition, with differentiation persisting for a total of 96 h. (B) Normalized mRNA expression levels of Muc2, Chga, Lyz1, Gcg, and Ngn3 at 96 h. Data are represented as the mean ± SD (n = 3). Experiments were performed three times with similar results. (C) mRNA expression of key transcription factors (Sox9, Atoh1, Ngn3, Neourod1) and markers for secretory cell types (Chga, Gcg for EEC, Clca1 for goblet cells, and Lyz1 for Paneth cells) during ISC differentiation under multiple conditions. For EN.D.CHIR-EN.D.C59 condition, cells were first cultured in EN.D.CHIR condition for 2 days and then shifted to the EN.D.C59 condition for another 2 days. For EN.D.CHIR and EN.D.C59 conditions, differentiation processed in the unchanged condition for 4 days, with a medium refreshment at day 2. (D) Representative confocal images of Neurod1 in colonies cultured for 3 days in multiple conditions as indicated. Scale bars, 50 μm. (E) Quantification of the proportion of Neurod1⁺ cells in colonies as in (D). One-way ANOVA with Dunnett's multiple comparisons test between each condition with the optimized EEC differentiation condition; Data are represented as the mean ± SD (n = 3); ****P < 0.0001; ***P < 0.001; and *P < 0.05. (F) Representative confocal images of EEC (Chga), goblet cells (Muc2), and Paneth cells (Lyz) for cells cultured in conditions as indicated. Scale bars, 50 μm. EEC, enteroendocrine cells; ISCs, intestinal stem cells.

We found that prolonged high Wnt (96 h condition) increased the expression of Paneth cell marker Lyz1 but diminished goblet and EEC markers Muc2 and Chga postdifferentiation (Fig. 2B). In contrast, continuous Wnt inhibition (0 h condition) showed higher Muc2 expression, moderate Chga, and negligible Lyz1 (Fig. 2B). Meanwhile, transitional Wnt (48 h, for instance) increased Chga but reduced Muc2 and Lyz1 expression (Fig. 2B). Of note, Muc2 expression peaked at 12–18-h Wnt exposure rather than 0 h, suggesting that brief Wnt activation following Notch inhibition enhances goblet cell differentiation (Fig. 2B).

Concurrently, Chga expression peaked at 42–48 h, which showed slightly different kinetics from Gcg, which peaked at 18–30 h. The Paneth cell marker Lyz1 reached a maximum at the 96-h condition (Fig. 2B). Interestingly, Ngn3 showed a continuous increase in expression with longer Wnt exposure, suggesting Wnt plays a role in regulating Ngn3 expression (Fig. 2B).

We also monitored the expression of key transcription factors during intestinal cell differentiation, including Sox9, Atoh1, Ngn3, and Neurod1. Continuously subjected to high Wnt signals (EN.D.CHIR), cells showed elevated Ngn3 and eventually expressed high Lyz1 by day 4. In persistent Wnt inactivation (END.C59 condition), cells expressed reduced Sox9, Ngn3, and Neurod1 and eventually expressed high levels of goblet cell marker Clca1. Under conditions with transitional Wnt state (EN.D.CHIR-EN.D.C59 condition), cells initially upregulated Sox9 and Ngn3 and then downregulated their expression after Wnt inhibition, while upregulating EEC markers Neurod1, Chga, and Gcg (Fig. 2C). This suggests that the downregulation of Ngn3 after switching to Wnt inhibition (EN.D.CHIR-EN.D.C59 condition) may be essential for the maturation of differentiated EECs (Fig. 2C). Furthermore, the quantification of Neurod1-positive cells on day 3 across various conditions revealed a pattern consistent with their gene expression levels.

Specifically, the two-step differentiation protocol yielded the highest number of Neurod1-positive cells (Fig. 2D, E). Immunostaining of goblet cell, EEC, and Paneth cell markers confirmed these findings (Fig. 2F). We also observed consistent morphology changes (Supplementary Fig. S2A) and the upregulation of most EEC subtype markers under these conditions (Supplementary Fig. S2B).

These results suggest that the timing and duration of Wnt activation following Notch inhibition influence the specific cell types that differentiate from ISCs. In particular, a short duration of Wnt activation promotes goblet cell differentiation, an intermediate duration promotes EEC differentiation, and a long duration promotes Paneth cell differentiation.

Combination of temporal Wnt and MAPK inhibition for EEC differentiation

Given that a transitional Wnt signal promoted EEC differentiation, we sought to further optimize the EEC differentiation condition (Fig. 3A). We started with the EN.D condition as a basal condition, with the highest EEC differentiation, as in Fig. 1A. The addition of MAPK inhibitor PD0325901 (Pd) further increased EEC differentiation (the EN.D.Pd condition in Fig. 3B, C) [24]. Moreover, adding an initial Wnt activation step further enhanced EEC differentiation (the EN.D.CHIR-EN.D.Pd condition in Fig. 3B, C), which showed a high percentage of EECs in the organoids (Fig. 3D).

FIG. 3.

Optimize EEC differentiation by Wnt regulation. (A) Scheme of introducing PD0325901 and two-step Wnt modulation in EEC differentiation protocol. (B) Time course analysis of Atoh1, Ngn3, Neurod1, and Chga expression during differentiation in multiple conditions. A medium change was performed on day 1. (C) The expression of Neurod1 and Chga following a 5-day differentiation protocol in multiple conditions. (D) Representative confocal images of EEC (Chga) in colonies following a 5-day differentiation protocol in EN.D.CHIR (1 day) – EN.D.Pd (4 days) condition. Scale bars, 50 μM. (E) Time course analysis of Ngn3, Neurod1, Chga, and Gcg expression in multiple conditions during 4-day differentiation as indicated. (F) Representative flow cytometry plots of Chga⁺ cells from cells differentiated in conditions as indicated. (G) Quantification of the percentage of Chga⁺ cells in conditions as indicated, n = 3 samples from each condition. Data are represented as the mean ± SD; one-way ANOVA with Tukey's multiple comparisons test; ****P < 0.0001; ***P < 0.001; **P < 0.01; and *P < 0.05. (H) Representative confocal images of EECs (Chga), Paneth cells (Lyz), and goblet cells (Muc2), EEC subtypes (Sst and Gcg) in EN.D, EN.D.CHIR (2 days)-EN.D.C59 (2 days), and EN.D.CHIR (2 days)-EN.D.C59.Pd (2 days) conditions. Scale bars, 50 μm. (I) Time course analysis of multiple EEC subtype marker expression in multiple differentiation conditions as indicated.

In addition, adding Wnt-C59 (C59) in the second step to inhibit Wnt signaling further increased EEC differentiation (Fig. 3E, I). We also found that a 2-day Wnt activation step induced comparable Chga but higher Gcg compared with a 1-day Wnt activation step and was thus maintained in the optimized EEC differentiation protocol (Fig. 3E). Under this protocol, Chga⁺ EECs constituted 12.4% of the organoid cells (Fig. 3F, G), compared with 0.21% in the ENR condition (Fig. 1C). The optimized protocol generated enriched EECs, but low levels of goblet cells, Paneth cells, or enterocytes (Fig. 3H and Supplementary Fig. S3B). In addition, the EEC subtype markers also significantly increased (Fig. 3I and Supplementary Fig. S3A). Cells expressing the EEC subtype marker Gcg or Sst were also detected in the organoids (Fig. 3H).

The combination of EGFR, Wnt, and Notch inhibitors has been previously reported to efficiently induce EEC differentiation [24]. We conducted a comparative analysis through immunostaining and quantification of EEC marker Chga and also performed qPCR quantification for EEC-specific markers Chga and Neurod1. Our findings validate the superiority of the optimized protocol (EN.D.CHIR-EN.D.C59.Pd), demonstrating the highest EEC proportion and marker gene expression (Fig. 4A–C). Furthermore, it has been reported that the BMP signaling gradient along the crypt–villus axis modulates hormone expression [33]. We tested the effect of BMP modulation in our optimized protocol and observed a notable decrease in Gcg-expressing L cells and a modest reduction in SSt-expressing D cells upon BMP addition (Fig. 4D). Moreover, our optimized protocol led to a significant augmentation in GLP-1 secretion; a fivefold increase of GLP-1 secretion following glucose stimulation, suggesting these EECs were mature and functional (Fig. 4E).

FIG. 4.

Optimized condition yields high-purity and mature EECs. (A) Representative confocal images show EEC (Chga) under ENR and multiple differentiation conditions. Scale bar, 50 μm. (B) Percentage of Chga⁺ cells quantified from conditions in (A). One-way ANOVA with Dunnett's multiple comparisons test between each condition with the optimized condition; mean ± SD (n = 3); significance: ****P < 0.0001; ***P < 0.001; and **P < 0.01. (C) RT-qPCR quantification of Chga and Neurod1 expression in conditions as in (A). One-way ANOVA with Dunnett's multiple comparisons test between each condition with the optimized condition; mean ± SD (n = 4); significance: ****P < 0.0001. (D) EEC colonies differentiated for 4 days; BMP4 was added for either the last 24 or 96 h. Scale bar, 50 μm. (E) Glucose-stimulated GLP-1 secretion assessed in differentiated EECs.

To elucidate the gene expression profile, we performed RNAseq analysis on EECs generated using our optimized protocol. EEC lineage-specific markers such as Mc4r, Nts, Isl1, Nkx2–2, and Chga, along with subtype-specific markers such as Cck, Sct, and Gcg, were significantly upregulated (Fig. 5A, B). Conversely, markers associated with ISCs or progenitor cells, including Olfm4, Lgr5, Ascl2, Tert, and Cdca7, were downregulated (Fig. 5A, B). GO enrichment analyses aligned with these findings, showing upregulation of processes such as peptide hormone secretion, hormone and fatty acid metabolism, and stimulus response (Fig. 5C), thereby corroborating the efficiency and specificity of EEC differentiation protocol.

FIG. 5.

RNA sequencing uncovers EEC gene profile. (A) Heatmap of DEGs between cells cultured in ENR and optimized EEC differentiation condition. (B) Volcano plot of gene expression comparison for cells as in (A). x-axis: log₂ fold change; y-axis: -log10 (P value). Genes with P < 0.01 and fold change >2 are marked red; specific genes are labeled. (C) GO enrichment analysis of DEGs. Top 10 enriched GO terms displayed. DEGs, differentially expressed genes; GO, Gene Ontology.

Combination of temporal Wnt and fibroblast growth factor receptor activation for goblet cell differentiation

We further optimized the goblet cell differentiation protocol from ISCs. Based on Notch and Wnt inhibition (D.C59 condition), we performed a combination screening of multiple conditions (Fig. 6A). Instead of directly transitioning to Wnt inhibition (D.C59) from CV, we tested conditions by treating the cells with a Wnt activator (D.CHIR) for 1 day before transitioning to Wnt inhibition (D.C59). We found that (1) replacing EGF with FGF2 in the presence of the BMP inhibitor (N) (F.N.D.C59 vs. E.N.D.C59) markedly increased goblet cell marker Muc2 and Clca1 expression. (2) Further switching to a two-step protocol consisting of a first Wnt activation step (D.CHIR) and a second Wnt inhibition step (D.C59) increased goblet cell differentiation [F.N.D.C59 vs. F.N.D.CHIR (1 day)–F.N.D.C59 (3 days) in Fig. 6A]. This optimized condition induced apparent morphological change in the organoid cells, resembling goblet cells (Fig. 6B), corroborated by Muc2 staining (Fig. 6C).

FIG. 6.

Optimized differentiation condition of goblet cell by regulating Wnt signaling. (A) mRNA expression of goblet cell markers Clca1 and Muc2 in colonies cultured in multiple conditions as indicated, n = 3 samples from each condition. Data are represented as the mean ± SD; one-way ANOVA with Tukey's multiple comparisons test; ****P < 0.0001. (B) Morphology of differentiated cell colonies following differentiation for 4 days in multiple conditions as indicated. Scale bars, 50 μm. (C) Representative confocal images of goblet cells (Muc2) in colonies cultured in multiple conditions as indicated. Scale bars, 50 μm. (D) Representative flow cytometry plots of Muc2⁺ cells from colonies differentiated in multiple conditions as indicated. (E) Quantification of percentages of Muc2⁺ cells in differentiation conditions as indicated. Data are represented as the mean ± SD (n = 3); one-way ANOVA with Tukey's multiple comparisons test; ****P < 0.0001; ***P < 0.001. (F) Representative confocal images of EECs (Chga), goblet cells (Muc2), and Paneth cells (Lyz), and Alp staining images in colonies differentiated for 4 days in FN.D.CHIR (1 day)-FN. D.C59 (3 days) condition. Scale bars, 50 μm.

The refined goblet cell differentiation protocol yielded 34.5% Muc2+ cells compared with 5.11% in the ENR condition (Fig. 6D, E). Furthermore, this condition led to low EEC, Paneth cell, or enterocyte differentiation (Fig. 6F), demonstrating its specificity. Mechanistically, combining FGF2 with a two-step Wnt induced a temporary upregulation of transcription factors that are important for the differentiation of goblet cells, such as Sox9, Gfi1, Atoh1, Klf4, and Spdef, compared with other conditions (Fig. 7A). Confirming this, our day 2 quantification showed that this two-step Wnt differentiation approach resulted in the most substantial elevation in the number of Klf4-positive cells (Fig. 7B, C). This increase in Klf4-positive cells was consonant with the gene expression data and strongly correlated with increased goblet cell differentiation (Fig. 7D).

FIG. 7.

Wnt signaling regulates the expression of key transcript factors during goblet cell differentiation. (A) Time course analysis of key transcript factor expression during 4-day differentiation in multiple conditions as indicated. (B) Representative confocal images of Klf4 in colonies cultured in multiple conditions as indicated. Scale bars, 50 μm. (C) Quantification of Klf4⁺ cells in colonies as in (B). One-way ANOVA with Dunnett's multiple comparisons test between each condition with the optimized condition; mean ± SD (n = 3); significance: **P < 0.01 and *P < 0.05. (D) Time course analysis of goblet-specific marker expression during 4-day differentiation in multiple conditions as indicated.

Further comparisons were made between our optimized goblet cell differentiation protocol and previously established methods [22]. Immunofluorescence images, cell quantification, and qPCR analyses collectively revealed a significant enhancement in goblet cell differentiation efficiency using our proposed protocol (Fig. 8A–C). RNAseq analysis on differentiated goblet cells demonstrated an upregulation of goblet cell-associated markers such as Spdef, Muc2, Muc3, Clca1, and Tff3, and a downregulation of stem cell-related genes, including Olfm4, Lgr5, Aqp4, and Cdk6 (Fig. 8D, E). GO enrichment analysis showed an upregulation of processes such as peptide secretion and membrane potential (Fig. 8F), affirming the functional maturity of the goblet cells generated. Lastly, we demonstrated that the Wnt3a protein could achieve comparable results with CHIR99021, not only in inducing differentiation but also in upregulating Wnt-relative genes (Supplementary Fig. S4A–E).

FIG. 8.

Optimized conditions enhance goblet cell differentiation and gene expression. (A) Representative confocal images depict goblet cells (Muc2) in ENR and multiple differentiation conditions. Scale bar, 50 μm. (B) Proportion of Muc2⁺ cells quantified from conditions in (A). One-way ANOVA with Dunnett's multiple comparisons test between each condition with the optimized condition; mean ± SD (n = 3); significance: ****P < 0.0001; ***P < 0.001; and *P < 0.05. (C) RT-qPCR reveals Muc2 and Tff3 expression levels from conditions in (A). One-way ANOVA with Dunnett's multiple comparisons test between each condition with the optimized condition; mean ± SD (n = 4); significance: ****P < 0.0001. (D) Heatmap displays DEGs between cells cultured in ENR and the optimized condition. (E) Volcano plot compares gene expression between conditions as in (A). x-axis: log₂ fold change; y-axis: -log10 (P value). Genes with P < 0.01 and fold change >2 are marked red; specific genes are labeled. (F) GO enrichment analysis of DEGs; top 10 enriched GO terms listed.

Discussion

In the present study, we have elucidated a new facet of Wnt signaling within the scope of ISC differentiation. Our results underscore the temporal dynamics of Wnt signaling, specifically post-Notch inactivation, as a crucial determinant in shaping the cell fate of intestinal secretory cell types. This introduces an additional layer of complexity to the regulatory signals governing the cell fate and behavior of ISCs and their progeny. It implies that cell fate determination extends beyond simply amalgamating static niche signals, incorporating a dynamic and temporal component instead.

Indeed, the differentiation process of ISCs along the crypt–villus axis in vivo is highly dynamic [4,5,21]. We propose a model in which cell fate specification transpires concurrently with the migration of differentiating cells, governed by the duration and intensity of Wnt signal exposure. In this model, cells migrating downward and immersed in a high Wnt environment differentiate into Paneth cells, whereas those migrating upward evolve into goblet cells or EECs, with their fate contingent upon the length of high Wnt exposure near the crypt base.

Insights into temporal Wnt signaling dynamics have capacitated us to optimize protocols for the efficient generation of goblet cells and EECs in vitro. Given the significant roles these cells fulfill in maintaining intestinal integrity and physiologic processes such as mucus secretion, nutrient sensing, and hormonal regulation [7,9,10], this carries substantial therapeutic interest.

We used the intestinal organoid model to probe signaling requirements for intestinal epithelial cell fate determination. This model faithfully replicates the cellular constitution and structural organization of the epithelium, thereby facilitating the investigation of dynamic self-renewal and differentiation processes in vitro [30]. Utilizing this model, we were able to identify the combinations of signals that are essential and sufficient for generating various intestinal cell types. Notwithstanding its robustness, we acknowledge that the organoid model may not fully emulate the complexity of the in vivo environment, potentially neglecting pivotal systemic and local influences such as immune cells, vascular cells, and microbiota that could affect the differentiation and maturation of functional intestinal epithelial cells [37].

Future research could benefit from integrating small-molecule screening or more physiologically relevant in vitro models incorporating additional intestinal constituents. Alternatively, in vivo models could provide a more comprehensive understanding of the extrinsic signals involved in modulating cell fate determination of intestinal epithelial cells. Importantly, recent studies have shown that cells along the crypt–villus axis are zonated and undergo transcriptomic changes, and EECs switch hormone expression profiles [19,33,38]. Incorporating this perspective into our protocols could enhance their utility by enabling the identification of distinct EEC and goblet cell subtypes present in the cultures.

Moreover, our differentiation platform paves the way for further exploration of transcription factor dynamics during the differentiation of intestinal epithelial cells. This promising avenue for future research holds potential implications for therapeutic development.

Footnotes

Acknowledgments

The authors thank Chunying Li, Yongfei Hou, and Tianzeng Sun for R-Spondin1-conditioned medium production.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by the National Key R&D Program of China (2019YFA0110003, 2020YFA0112500), the National Natural Science Foundation of China (31970820), the Key Project of the Science and Technology of Shanghai Municipality (19JC1415300), and the Fundamental Research Funds for the Central Universities (22120230292).

Supplementary Data

Supplementary Figure S1

Supplementary Figure S2

Supplementary Figure S3

Supplementary Figure S4

References

Barker

, van Es

, Kuipers

, Kujala

, van den Born

, Cozijnsen

, Haegebarth

, Korving

, Begthel

, Peters

and Clevers

. (2007). Identification of stem cells in small intestine and colon by marker gene Lgr5. Nature, 449:1003–1007.

Ritsma

, Ellenbroek

SIJ

, Zomer

, Snippert

, de Sauvage

, Simons

, Clevers

and van Rheenen

. (2014). Intestinal crypt homeostasis revealed at single-stem-cell level by in vivo live imaging. Nature, 507:362–365.

Beumer

and Clevers

. (2021). Cell fate specification and differentiation in the adult mammalian intestine. Nat Rev Mol Cell Biol, 22:39–53.

Van Der Flier

and Clevers

. (2009). Stem cells, self-renewal, and differentiation in the intestinal epithelium. Annu Rev Physiol, 71:241–260.

Gehart

and Clevers

. (2019). Tales from the crypt: new insights into intestinal stem cells. Nat Rev Gastroenterol Hepatol, 16:19–34.

Birchenough

, Johansson

, Gustafsson

, Bergström

and Hansson

. (2015). New developments in goblet cell mucus secretion and function. Mucosal Immunol, 8:712–719.

Gustafsson

and Johansson

MEV

. (2022). The role of goblet cells and mucus in intestinal homeostasis. Nat Rev Gastroenterol Hepatol, 19:785–803.

Furness

, Rivera

, Cho

, Bravo

and Callaghan

. (2013). The gut as a sensory organ. Nat Rev Gastroenterol Hepatol, 10:729–740.

Gribble

and Reimann

. (2019). Function and mechanisms of enteroendocrine cells and gut hormones in metabolism. Nat Rev Endocrinol, 15:226–237.

10.

Gribble

and Reimann

. (2016). Enteroendocrine cells: chemosensors in the intestinal epithelium. Ann Rev Physiol, 78:277–299.

11.

Sato

, van Es

, Snippert

, Stange

, Vries

, van den Born

, Barker

, Shroyer

, van de Wetering

and Clevers

. (2011). Paneth cells constitute the niche for Lgr5 stem cells in intestinal crypts. Nature, 469:415–418.

12.

Dayton

and Clevers

. (2017). Beyond growth signaling: Paneth cells metabolically support ISCs. Cell Res, 27:851–852.

13.

Bevins

and Salzman

. (2011). Paneth cells, antimicrobial peptides and maintenance of intestinal homeostasis. Nat Rev Microbiol, 9:356–368.

14.

Pierre

, Peters

, La Torre

, Sidebottom

, Tao

, Zhu

, Cham

, Wang

, Kambal

, Harris

, Silva

, et al. (2023). Peptide YY: a Paneth cell antimicrobial peptide that maintains Candida gut commensalism. Science, 381:502–508.

15.

and Chen

Y-G

. (2015). Regulation of intestinal stem cell fate specification. Sci China Life Sci, 58:570–578.

16.

Farin

, Van Es

and Clevers

. (2012). Redundant sources of Wnt regulate intestinal stem cells and promote formation of Paneth cells. Gastroenterology, 143:1518–1529 e1517.

17.

Yan

, Janda

, Chang

, Zheng

, Larkin

, Luca

, Chia

, Mah

, Han

and Terry

. (2017). Non-equivalence of Wnt and R-spondin ligands during Lgr5+ intestinal stem-cell self-renewal. Nature, 545:238–242.

18.

Farin

, Jordens

, Mosa

, Basak

, Korving

, Tauriello

, De Punder

, Angers

, Peters

and Maurice

. (2016). Visualization of a short-range Wnt gradient in the intestinal stem-cell niche. Nature, 530:340–343.

19.

Beumer

, Puschhof

, Yengej

, Zhao

, Martinez-Silgado

, Blotenburg

, Begthel

, Boot

, van Oudenaarden

, Chen

and Clevers

. (2022). BMP gradient along the intestinal villus axis controls zonated enterocyte and goblet cell states. Cell Rep, 38:110438.

20.

, Zhang

, Tong

W-G

, Tawfik

, Ross

, Scoville

, Tian

, Zeng

, He

and Wiedemann

. (2004). BMP signaling inhibits intestinal stem cell self-renewal through suppression of Wnt-β-catenin signaling. Nat Genet, 36:1117–1121.

21.

Zhu

, Hu

and Xi

. (2021). The cellular niche for intestinal stem cells: a team effort. Cell Regen, 10:1–16.

22.

Yin

, Farin

, van Es

, Clevers

, Langer

and Karp

. (2014). Niche-independent high-purity cultures of Lgr5+ intestinal stem cells and their progeny. Nat Methods, 11:106–112.

23.

Kraiczy

, McCarthy

, Malagola

, Tie

, Madha

, Boffelli

, Wagner

, Wang

and Shivdasani

. (2023). Graded BMP signaling within intestinal crypt architecture directs self-organization of the Wnt-secreting stem cell niche. Cell Stem Cell, 30:433–449 e438.

24.

Basak

, Beumer

, Wiebrands

, Seno

, van Oudenaarden

and Clevers

. (2017). Induced quiescence of Lgr5+ stem cells in intestinal organoids enables differentiation of hormone-producing enteroendocrine cells. Cell Stem Cell, 20:177–190 e174.

25.

VanDussen

, Carulli

, Keeley

, Patel

, Puthoff

, Magness

, Tran

, Maillard

, Siebel

, Kolterud

, Grosse

, et al. (2012). Notch signaling modulates proliferation and differentiation of intestinal crypt base columnar stem cells. Development, 139:488–497.

26.

Sato

, Vries

, Snippert

, van de Wetering

, Barker

, Stange

, van Es

, Abo

, Kujala

, Peters

and Clevers

. (2009). Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature, 459:262–265.

27.

Rahmani

, Breyner

, Su

, Verdu

and Didar

. (2019). Intestinal organoids: a new paradigm for engineering intestinal epithelium in vitro. Biomaterials, 194:195–214.

28.

Sprangers

, Zaalberg

and Maurice

. (2021). Organoid-based modeling of intestinal development, regeneration, and repair. Cell Death Differ, 28:95–107.

29.

Clevers

. (2016). Modeling development and disease with organoids. Cell, 165:1586–1597.

30.

Takahashi

, Fujishima

and Kengaku

. (2021). Modeling intestinal stem cell function with organoids. Int J Mol Sci, 22:10912.

31.

Serra

, Mayr

, Boni

, Lukonin

, Rempfler

, Challet Meylan

, Stadler

, Strnad

, Papasaikas

, Vischi

, Waldt

, et al. (2019). Self-organization and symmetry breaking in intestinal organoid development. Nature, 569:66–72.

32.

Mead

, Hattori

, Levy

, Imada

, Goto

, Vukovic

, Sze

, Kummerlowe

, Matute

, Duan

, Langer

, et al. (2022). Screening for modulators of the cellular composition of gut epithelia via organoid models of intestinal stem cell differentiation. Nat Biomed Eng, 6:476–494.

33.

Beumer

, Artegiani

, Post

, Reimann

, Gribble

, Nguyen

, Zeng

, Van den Born

, Van Es

and Clevers

. (2018). Enteroendocrine cells switch hormone expression along the crypt-to-villus BMP signalling gradient. Nat Cell Biol, 20:909–916.

34.

Petersen

, Reimann

, Bartfeld

, Farin

, Ringnalda

, Vries

, van den Brink

, Clevers

, Gribble

and de Koning

. (2014). Generation of L cells in mouse and human small intestine organoids. Diabetes, 63:410–420.

35.

Chen

, Chen

, Zhang

, Zhu

, Tang

, Wang

, Dong

, Zhang

, Yu

, Sun

, Chi

, et al. (2021). The genome sequence archive family: toward explosive data growth and diverse data types. Genomics Proteomics Bioinformatics, 19:578–583.

36.

Members

C-N

and Partners. (2022). Database resources of the national genomics data center, China national center for bioinformation in 2022. Nucleic Acids Res, 50:D27–D38.

37.

Yin

, Mead

, Safaee

, Langer

, Karp

and Levy

. (2016). Engineering stem cell organoids. Cell Stem Cell, 18:25–38.

38.

Moor

, Harnik

, Ben-Moshe

, Massasa

, Rozenberg

, Eilam

, Bahar Halpern

and Itzkovitz

. (2018). Spatial reconstruction of single enterocytes uncovers broad zonation along the intestinal villus axis. Cell, 175:1156.e15–1167.e15.