Lgr5⁺ cells regulate small intestinal morphogenesis before villification

Abstract

Lgr5 marks both adult intestinal stem cells and embryonic intestinal stem/progenitor cells. However, the stemness properties and physiological roles of embryonic intestinal Lgr5⁺ cells prior to villification (PVLCs) remain largely unknown. In this study, we show that PVLCs in the embryonic small intestine exhibit region-specific stemness, with progressively enhanced stemness potential from the proximal to distal region. Through inducible cell ablation and gene knockout experiments, we demonstrate that PVLCs regulate small intestinal morphogenesis via Hedgehog signaling in a region-dependent manner, with distal morphogenesis being more dependent on this mechanism. This study reveals the stemness and functional roles of PVLCs in the embryonic small intestine prior to villification, highlighting regionalized cellular heterogeneity as a critical determinant of intestinal morphogenesis.

Graphical Abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s13619-026-00284-y.

Background

Serving as an integral part of the digestive system, the intestine is responsible for physiological functions including food digestion, nutrient absorption, waste excretion, and pathogen defense (Zhu et al. 2021; Yang et al. 2025). To adapt to its dynamic environment, the intestinal epithelium undergoes continuous and rapid self-renewal, a process driven by intestinal stem cells residing at the base of crypts. A well-established marker of these stem cells is Lgr5 (Leucine-rich repeat-containing G protein-coupled receptor 5) (Barker et al. 2007). The regulatory mechanisms governing the maintenance and differentiation of Lgr5⁺ intestinal stem cells under both homeostatic and pathological conditions in adults have been extensively elucidated (McCarthy et al. 2020; Beumer and Clevers 2021; Li et al. 2024).

The establishment of future physiological functions in embryos requires proper intestinal development, which involves key processes including intestinal specification, regionalization, and the formation of villus (villification) and crypt structures (Guiu and Jensen 2015; Wang et al. 2019). We previously reported that Lgr5 expression can be detected in the intestinal epithelium as early as embryonic day (E) 9.5 in mice, and displays region-specific distribution before villification (Zhao et al. 2022). The villification initiates at E14.0-E14.5, followed by restricted Lgr5 expression in the inter-villus (future crypt) regions, which also harbor proliferative epithelial cells (Shyer et al. 2015). It is widely assumed that embryonic intestinal Lgr5⁺ cells function as stem or progenitor cells, based on their expression of stemness-associated genes and ability to generate lineage-tracing events (Dzama et al. 2017; Nigmatullina et al. 2017). However, due to the low frequency of lineage tracing events and the region-specific distribution of Lgr5⁺ cells (Dzama et al. 2017; Nigmatullina et al. 2017), the stemness potential and function of embryonic Lgr5⁺ cells remain controversial.

Epithelial-mesenchymal interactions play a critical role in intestinal development, with Hedgehog (Hh) signaling serving as a key mediator (Le Guen et al. 2015). During intestinal morphogenesis, epithelial cells secrete the functionally redundant ligands Sonic hedgehog (Shh) and Indian hedgehog (Ihh), which act on mesenchymal cells in a paracrine manner and coordinate with other signaling pathways to regulate morphogenetic processes such as villification (Ramalho-Santos et al. 2000; Madison et al. 2005; Mao et al. 2010; Walton et al. 2012; Shyer et al. 2015). However, the importance of Hh signaling derived from different epithelial cell subtypes remains unknown.

Previous studies have predominantly relied on gene knockout strategies to investigate gene functions during embryogenesis, overlooking the contributions of specific cell populations. Although cell ablation models are widely employed in adult tissues to study cell-type-specific functions (Metcalfe et al. 2014; Tan et al. 2021), such models have rarely been applied in embryonic contexts. In this study, we utilized Lgr5-DTR^+/- mice to ablate Lgr5⁺ cells in the embryonic small intestine and demonstrated that Lgr5⁺ cells regulate regional intestinal morphogenesis through Hh signaling. In addition, we demonstrated that the stemness potential of Lgr5⁺ cells gradually increased from the proximal to the distal region before villification. Our findings provide critical insights into the regulatory mechanisms governing small intestinal organogenesis and establish a foundational framework integrating morphogenesis with region-specific regulatory mechanisms.

Results

Lgr5⁺ cells exhibit dynamic spatiotemporal distributions in embryonic small intestine

Our previous single-cell RNA sequencing (scRNA-seq) analysis revealed that Lgr5 expression exhibits a regional difference with progressively increasing levels from the proximal (PSI) to the distal (DSI) small intestine during development (Zhao et al. 2022), consistent with a previous study (Maimets et al. 2022). This distribution pattern is distinct from the uniform location of Lgr5⁺ stem cells in the crypts of adult mouse intestines (Barker et al. 2007). Based on RNA counts from the dataset, we calculated the ratios of Lgr5⁺ cells in the PSI and DSI at different developmental stages. At E10.5, the ratios of Lgr5⁺ cells began to exhibit a clear proximal–distal difference. Prior to villification at E13.5, the ratio of Lgr5⁺ cells in the DSI was significantly higher than that in the PSI (Fig. 1A). RNAscope staining indicated that the expression of Lgr5 became concentrated in the inter-villus regions after villification (Fig. 1B, e.g., PSI at E15.5 and DSI at E17.5), gradually resembling the distribution pattern of adult Lgr5⁺ stem cells. Notably, the expression level of Lgr5 in the DSI remained higher than that in the PSI after villification at E17.5, which was not observed in Lgr5⁺ stem cells in adult intestines. We named these Lgr5⁺ cells prior to villification as PVLCs (Pre-villification Lgr5⁺ cells).

Fig. 1.

Spatiotemporal distributions of Lgr5⁺ cells in the embryonic mouse small intestine. A UMAP visualization of published scRNA-seq dataset of the small intestinal epithelium (SIE), along with the ratio of Lgr5⁺ cells in the PSI and DSI from E9.5 to E15.5. B RNAscope staining of Lgr5 in the PSI and DSI from E11.5 to E17.5 (n = 3). The dashed lines indicate the epithelial-mesenchymal interface. SIE: small intestinal epithelium; epi: epithelium; PSI: proximal small intestine; DSI: distal small intestine; E: embryonic day. Scale bar = 200 μm

Stemness potential of PVLCs progressively increases along the PSI-DSI axis

In the adult intestine, one of the characteristics of stem cells is the expression of stemness-related genes (Barker et al. 2007; Haber et al. 2017). To determine whether PVLCs share similar transcriptomic features in the embryonic small intestine, we performed differential gene expression analysis between PVLCs and other epithelial subpopulations (non-PVLCs) based on the scRNA-seq dataset. From E9.5 to E11.5, PVLCs exhibited minimal transcriptional divergence from non-PVLCs in both the PSI and DSI. However, the number of genes enriched in PVLCs was significantly increased in both regions at E13.5 (Fig. S1A), consistent with the elevated number of PVLCs at E13.5 (Fig. 1). Gene Ontology (GO)/Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses showed that the majority of the genes were associated with cell cycle process (Fig. 2A), including well-known genes such as Mki67, Top2a and Ube2c (Fig. S1B; Table S1). Moreover, PVLCs exhibited significantly higher cell cycle scores than non-PVLCs at E13.5 (Fig. 2B and S1C). These data suggest that PVLCs began to enter a more rapid cycling state than non-PVLCs at E13.5.

Fig. 2.

Stemness transition of PVLCs along the PSI-DSI axis. A GO/KEGG analysis of differentially expressed genes between Lgr5⁺ epithelial cells and Lgr5⁻ epithelial cells, analyzed from scRNA-seq. B Cell cycle scores of non-PVLCs and PVLCs at E13.5, analyzed from scRNA-seq. C Strategy for short-term and long-term lineage tracing of PVLCs in embryos. D Immunofluorescence staining of short-term lineage tracing events (ZsGreen) in the PSI and DSI at E15.5. E Quantification of number of lineage tracing events comprising 2–5 cells in (D). F Immunofluorescence co-staining of long-term lineage tracing events (ZsGreen) with MUC2 or CHGA along the PSI-DSI axis at P0. G Quantification of cell proportions of ZsGreen co-localized with MUC2 or CHGA relative to the total number of MUC2⁺ or CHGA⁺ cells in (F). TAM: tamoxifen; PSI: proximal small intestine; DSI: distal small intestine; E: embryonic day; P: postnatal day; PVLCs: pre-villification Lgr5⁺ cells. Scale bar = 200 μm (D), 100 μm (F). Data are shown as mean ± SD, n = 3 (D-G), ***p < 0.001, Wilcoxon test (B), Two-way ANOVA (E and G)

To validate the cycling state of PVLCs in vivo, the short-term lineage tracing experiments were conducted using Rosa26-ZsGreen^flox/flox;Lgr5-CreERT2^+/- mice during different time windows. The tracing process was induced by one dosage of tamoxifen (TAM) at E9.5, E11.5 or E13.5, respectively, before villification and terminated at E15.5 (Fig. 2C). When induction was initiated at E9.5, no tracing events were observed in the PSI or DSI at E15.5. Induction at E11.5 resulted in a small number of tracing events appearing in the DSI. In contrast, there was a significantly increasing number of tracing events in both the PSI and DSI when tracing was induced at E13.5 (Fig. 2D and E). To explore whether these cells possess stemness or progenitor characteristics similar to those of adult Lgr5⁺ intestinal stem cells, we initiated long-term lineage tracing of PVLCs starting from E13.5 and assessed differentiated lineages at birth (Postnatal day 0, P0) (Fig. 2C). In the PSI, MUC2⁺ goblet cells and CHGA⁺ enteroendocrine cells were widely distributed, but showed no or minimal overlaps with the tracing events generated by PVLCs. As the observation area shifted to the distal region, an increasing number of MUC2 and CHGA signals overlapped with the tracing events (Fig. 2F and G). Furthermore, PVLCs in the DSI displayed significantly higher CytoTRACE2 scores than those in the PSI (Fig. S1D). Taken together, these results suggest that the stemness potential of PVLCs gradually enhances from the proximal to the distal region before villification, which is different from earlier reports suggesting uniform stem cell feature of embryonic intestinal Lgr5⁺ cells (Dzama et al. 2017; Nigmatullina et al. 2017).

Ablation of PVLCs impairs small intestinal morphogenesis

To investigate the physiological role of PVLCs, we performed in vivo ablation of these cells using Lgr5-DTR^+/- (Lgr5-DTR-EGFP) mice (Fig. S2A). The diphtheria toxin (DT) administration into pregnant mice at E13.5 resulted in complete lethality of mutant embryos by E15.5 (Fig. S2B), accompanied by severe disruption of PSI and DSI structures (Fig. S2C). This outcome is likely attributed to the presence of Lgr5⁺ cells in other embryonic organs (Liu et al. 2014; Sayols et al. 2020; Watanabe et al. 2023), leading to global developmental defects after ablation. Since the ex vivo culture could maintain the proximal–distal heterogeneity of Lgr5 expression (Fig. S2D), the embryonic small intestines were isolated, cultured ex vivo and then treated with DT to induce PVLC ablation, thereby eliminating the in vivo side effects (Fig. 3A). To exclude the possibility of cytotoxic effects of DT, the control group (Lgr5-DTR^−/−) with DT treatment was included. After 3 or 5 days of culture, DT treatment did not cause obvious morphologic abnormalities (Fig. 3B and S2E). However, PVLC ablation in the mutant intestines induced region-specific defects. The PSI developed abnormal cyst-like structures, while the DSI showed more severe damage, including structural collapse and diametric reduction (Fig. 3B and C; Fig. S2E). This regional difference is consistent with the higher abundance of PVLCs in the distal region. The mutant PSI exhibited abnormal structures, characterized by villus absence and flattened epithelium (Fig. 3D). In the mutant DSI, epithelial structures were nearly depleted and mesenchymal proliferation was markedly reduced (Fig. 3D and E). Moreover, the ablation led to upregulated apoptosis in mesenchymal cells in both regions (Fig. 3D and E). The undermined mesenchymal proliferation and enhanced apoptosis may contribute to the collapse and narrowing of the DSI. Intriguingly, when ablation was induced at E12.5, cyst-like structures did not form in the PSI, but mesenchymal apoptosis was still increased (Fig. S2F). Since PVLCs are a population of epithelial cells, their ablation and the resulting impaired mesenchymal development highlight the essential role of epithelial-mesenchymal interactions during intestinal morphogenesis.

Fig. 3.

PVLCs regulate small intestinal morphogenesis. A Strategy for ablation of PVLCs in the embryonic PSI and DSI ex vivo. B Morphologies of embryonic PSI and DSI after PVLC ablation. C Quantification of SI diameters in (B). D Immunofluorescence staining of MKI67 and Cleaved Caspase 3 (CC3) in the embryonic PSI and DSI after PVLC ablation. E Quantification of MKI67⁺ cell ratio among mesenchymal cells and CC3⁺ cell number in (D). DT: diphtheria toxin; PSI: proximal small intestine; DSI: distal small intestine; mes: mesenchyme; E: embryonic day; PVLCs: pre-villification Lgr5⁺ cells. Scale bar = 100 μm (B), 200 μm (D). Data are shown as mean ± SD, n = 3 (B-E), *p < 0.05, ***p < 0.001, n.s.: not significant, Two-way ANOVA (C and E)

PVLCs regulate morphogenesis of the distal small intestine via Shh signaling

To elucidate the molecular mechanisms by which PVLCs regulate embryonic small intestinal development, we performed bulk RNA sequencing on PSI explants before and after PVLC ablation (Fig. 4A), as collapsed DSI explants were not suitable for sequencing. Compared to the control group, elimination of PVLCs induced significant transcriptomic alterations (Table S2). GO/KEGG analysis revealed that the downregulated genes were significantly enriched in the Hh signaling pathway, embryonic morphogenesis, and extracellular matrix (ECM)-receptor interaction (Fig. 4B). These included canonical mesenchymal markers (Pdgfra, Fn1, etc.), villification-associated genes (Fat4, Vangl1) (Rao-Bhatia et al. 2020), and components of the Hh signaling (Gli1, Ptch1, etc.) (Fig. 4C). These transcriptomic alterations were consistent with the structural defects following ablation.

Fig. 4.

PVLCs regulate small intestinal morphogenesis via Hh signaling ex vivo. A Strategy for ablation of PVLCs in the embryonic PSI ex vivo and subsequent bulk RNA-seq. B GO/KEGG analysis of downregulated genes after PVLC ablation in the PSI, analyzed from bulk RNA-seq. C Gene heatmap of indicated genes, analyzed from bulk RNA-seq. D Strategy for PVLC ablation in combination with SAG stimulation in the embryonic PSI and DSI ex vivo. E Morphologies of embryonic PSI and DSI after PVLC ablation in combination with SAG stimulation. F Immunofluorescence staining of Cleaved Caspase 3 (CC3) in the embryonic PSI and DSI after PVLC ablation in combination with SAG stimulation. G Quantification of SI diameters and CC3⁺ cell number in (F). DT: diphtheria toxin; SAG: Smoothened agonist; PSI: proximal small intestine; DSI: distal small intestine; E: embryonic day; PVLCs: pre-villification Lgr5⁺ cells. Scale bar = 100 μm (E), 200 μm (F). Data are shown as mean ± SD, n = 2 (B and C), n = 3 (E–G), *p < 0.05, ***p < 0.001, Two-way ANOVA (G)

Hh ligands are secreted by epithelial cells and act on mesenchymal cells via paracrine signaling during intestinal development (Walton et al. 2012). To explore whether PVLC-derived Hh signaling is indispensable for intestinal development, we first confirmed that embryonic small intestinal organoids could maintain the expression of Lgr5 (Fig. S3A and B), and then established non-contact co-culture experiments using a Transwell system with embryonic Lgr5-DTR^+/- epithelial organoids in the upper chamber and embryonic wild-type mesenchymal cells in the lower chamber (Fig. S3C), and found that ablation of PVLCs within organoids led to moderate downregulation of the mesenchymal marker Pdgfra (Fig. S3D), confirming their paracrine regulatory roles. SAG is a small molecule that activates Hh signaling as a Smoothened agonist (Chen et al. 2002). As shown in Fig. 4D and E, SAG rescued the developmental defects induced by PVLC depletion, including the disappearance of cystic structures in the PSI and restoration of collapsed architecture in the DSI. Immunofluorescence staining further demonstrated the recovery of the columnar morphology of epithelial cells and villification in the PSI, and normalization of intestinal diameter in the DSI (Fig. 4F and G). Importantly, mesenchymal apoptosis was significantly alleviated after SAG stimulation in both regions (Fig. 4F and G). These findings collectively indicate that PVLCs regulate embryonic small intestinal development via Hh signaling.

We next investigated whether PVLCs exert similar regulatory effects in vivo. Our scRNA-seq data and a previous study (Maimets et al. 2022) indicated that Shh exhibited a regional expression divergence with higher expression in DSI prior to villification at E13.5 (Fig. 5A), mirroring the distribution pattern of PVLCs. In contrast, Ihh and Dhh (Desert hedgehog) did not exhibit such regional specificity. Based on this regional specificity, we employed the Lgr5-CreERT2^+/-;Shh^flox/flox embryos to delete Shh in PVLCs (Fig. 5B and S4A). In mutant embryos, the DSI developed cystic structures (Fig. 5C), failed to undergo villification, and exhibited increased mesenchymal apoptosis, whereas the PSI remained unaffected (Fig. 5D and E). The DSI phenotype in vivo closely resembled the PSI phenotype ex vivo, which may be ascribed to the more profound effects induced by cell ablation. Intriguingly, in the Shh-deficient DSI, the failure of villification was accompanied by a high ratio of proliferative epithelial cells and their abnormal distribution (Fig. 5D and E). Furthermore, there was an increase in the ratio of LYZ1⁺ epithelial cells, while the ratio of MUC2⁺ goblet cells remained unchanged; however, MUC2 expression was markedly reduced (Fig. S4B and C). Taken together, these findings demonstrate that PVLCs regulate DSI morphogenesis and epithelial differentiation through Shh signaling.

Fig. 5.

PVLCs regulate DSI morphogenesis via Shh signaling in vivo. A Expression levels of Shh, Ihh and Dhh in the PSI and DSI epithelial cells at E13.5 and E15.5, analyzed from scRNA-seq. B Strategy for knockout of Shh in embryonic PVLCs in vivo. C Morphologies of embryonic stomach and intestine after Shh knockout. D Immunofluorescence staining of TUNEL and MKI67 in the embryonic PSI and DSI after Shh knockout. E Quantification of TUNEL⁺ cell number and MKI67⁺ cell ratio among epithelial cells in (D). TAM: tamoxifen; PSI: proximal small intestine; DSI: distal small intestine; E: embryonic day; PVLCs: pre-villification Lgr5⁺ cells. Scale bar = 1 mm (C), 200 μm (D). Data are shown as mean ± SD, n = 3 (C-E), *p < 0.05, ***p < 0.001, n.s.: not significant, Two-way ANOVA (E)

Discussion

The regional distribution of Lgr5⁺ cells in embryonic small intestinal development has been neglected. In this study, we found that Lgr5⁺ cells are more abundant and exhibit enhanced stemness potential in the DSI compared to the PSI prior to villification (Fig. 6). Given that Lgr5 is a canonical target of Wnt signaling (de Lau et al. 2014), this regional disparity may reflect a gradient of Wnt activity ascending from PSI to DSI (Spence et al. 2011; Maimets et al. 2022). Following villification, Lgr5⁺ cells become restricted to the inter-villus regions (future crypts), where they acquire stem/progenitor cell identity (Shyer et al. 2015), accompanied by a gradual loss of regional differences. Of note, our lineage-tracing experiments revealed that, in contrast to the DSI, the most inter-villus cells (including Lgr5⁺ stem cells) in the PSI do not originate from the PVLCs. Recent studies have identified alternative progenitor populations independent of Lgr5⁺ cells during small intestinal development, such as Trop2⁺, Cnx43⁺, and Bmi1⁺ cells (Mustata et al. 2013; Smith et al. 2025). However, it remains unclear whether these cells exhibit regional differences in distribution or developmental trajectories. Therefore, the progenitors of Lgr5⁺ stem cells in the PSI are still undefined. It should be noted that the lineage-tracing strategy used in this study does not distinguish between the initially labeled PVLCs and the newly generated PVLCs derived from them over time. As a result, the final clone size reflects the combined outcome of initial labeling efficiency, proliferative expansion, and ongoing PVLC production, rather than a direct measure of intrinsic stem cell capacity. This intrinsic feature of the experimental design complicates quantitative interpretation of clonal dynamics and limits the ability to infer absolute stemness from lineage tracing alone. Accordingly, our observations are interpreted as reflecting relative stemness potential rather than definitive intrinsic stemness.

Fig. 6.

Features and distributions of Lgr5⁺ cells (including PVLCs) during small intestinal morphogenesis. E: embryonic day; PVLCs: pre-villification Lgr5⁺ cells

Upon PVLC ablation or Shh signaling blockade, we observed elevated apoptosis of mesenchymal cells. Although Hh signaling has been reported to regulate apoptosis in tumor cells (Noguchi et al. 2015), its role in modulating cell survival during embryonic development remains poorly understood (Briscoe and Therond 2013). In contrast to adult tissues, embryonic mesenchymal cells are actively proliferating and more susceptible to niche perturbations. Given the essential role of epithelial-mesenchymal interactions and signaling crosstalk during intestinal development (Le Guen et al. 2015; Zhao et al. 2022), it is possible that Hh signaling could indirectly modulate cell survival through other signals. Furthermore, ablation of PVLCs led to a shift in epithelial cell morphology from columnar to flattened and disrupted epithelial integrity. These changes may also compromise the epithelial-mesenchymal interaction and subsequently influence mesenchymal behaviors, which warrants further investigation.

In this study, we observed phenotypic differences between the in vivo gene knockout model and the ex vivo cell ablation model, which may arise from several reasons. First, gene knockout affects the function of a single gene, whereas cell ablation eliminates all genetic activities within targeted cells, resulting in more severe consequences. Second, the DSI harbors a higher proportion of PVLCs and exhibits stronger Shh expression, making the DSI more susceptible to this genetic perturbation. Third, during small intestinal development, functional redundancy exists between Shh and Ihh (Ramalho-Santos et al. 2000; Madison et al. 2005; Mao et al. 2010), and a compensatory mechanism may mitigate the overall Hh signaling reduction followed by Shh deletion. Overall, the cell ablation model provides a more comprehensive assessment of the roles of specific cell populations, though it still has limitations, such as the difficulty in tissue-specific ablation.

Methods

Mice

Lgr5-DTR-EGFP strain was acquired from Genentech (#OM-214999). Rosa26-ZsGreen^flox/flox (#007906) strain was acquired from the Jackson Laboratory. Lgr5-CreERT2 (T003768) strain was obtained from GemPharmatech Co., Ltd. Shh^flox/flox strain was a gift from Dr. Shiwen Luo (Nanchang University). To obtain embryos, 8–12-week-old female mice were mated with adult males, and the day a vaginal plug was observed at noon was designated as E0.5. Both male and female embryos were included. All mice were maintained under specific pathogen-free (SPF) conditions with a 12-h light/dark cycle.

Antibodies

Primary antibodies: anti-Ki67 (ab15580, Abcam), anti-Cleaved Caspase 3 (9664S, CST), anti-E-cadherin (610182, BD Biosciences), anti-Muc2 (ab272692, Abcam), anti-Chromogranin A (ab15160, Abcam) and anti-Lysozyme (ab10850, Abcam). Secondary antibodies: Alexa Fluor 488 anti-mouse (715–545–150, Jackson ImmunoResearch) and Alexa Fluor TRITC anti-rabbit (711–025–152, Jackson ImmunoResearch).

Cell and gene manipulation in vivo

For in vivo lineage tracing and gene knockout experiments, pregnant mice were intraperitoneally injected with TAM (Targetmol, dissolved in corn oil) at 0.1 mg/g at the designated embryonic stages. For in vitro transient labeling experiments, organoids were treated with 4-OHT (4-Hydroxytamoxifen) at 0.5 μM (Sigma). For in vivo cell ablation experiments, pregnant mice were intraperitoneally injected with DT (Sigma, dissolved in PBS) at 50 ng/g.

Ex vivo culture

Ex vivo culture was conducted following a previous study (Walton and Kolterud 2014). In brief, segments of small intestine measuring 0.5–1.0 cm in length were carefully dissected from indicated embryos. After removal of surrounding connective tissues, intestinal fragments were placed onto Transwell inserts (8 μm, Corning) and cultured in BGJb medium (12591–038, Thermo Fisher) supplemented with penicillin–streptomycin (Thermo Fisher) and ascorbic acid (0.1 mg/mL, Selleck). Cultures were maintained at 37 °C in a humidified incubator with 5% CO₂, with medium refreshed every 24 h. DT (0.1 μg/mL) and SAG (1 μM, Selleck) were added to the culture medium.

In situ hybridization (RNAscope)

For in situ hybridization using the RNAscope platform, freshly harvested tissues were rapidly embedded in optimum cutting temperature (OCT) compound on dry ice and stored at -80 °C for no longer than one week. Cryosections were prepared at a thickness of 10 μm (Leica) and subsequently stored at -80 °C overnight. Prior to hybridization, sections were fixed in 10% neutral buffered formalin (Solarbio) for 15 min at 4 °C, followed by sequential dehydration in graded ethanol. Endogenous peroxidase activity was quenched by treatment with hydrogen peroxide (322381, ACD) for 10 min at room temperature. Protease digestion was then carried out using Protease III (322381, ACD) for 20 min at room temperature. Hybridization was performed at 40 °C for 2 h using the Lgr5 probe (312171, ACD). Signal amplification and chromogenic detection were conducted in accordance with the manufacturer’s protocol (323110, ACD). Images were acquired using a confocal microscope (FV3000, Olympus).

Immunofluorescence staining

Embryonic tissues were rinsed in cold PBS, fixed in 4% paraformaldehyde for 30 min, and subsequently dehydrated in 30% sucrose at room temperature overnight. Tissues were then embedded in OCT compound and frozen at -80 °C prior to cryosectioning at a thickness of 10 μm (Leica). Before staining, sections were air-dried at room temperature for 30 min, followed by immersion in PBS for 5 min to remove residual OCT. Antigen retrieval was performed by incubating the sections in citrate buffer in a 95 °C water bath for 25 min. Samples were then permeabilized with 0.1% Triton X-100 for 15 min at 4 °C, and then blocked with 3% bovine serum albumin (BSA) containing 0.01% Triton X-100 for 1 h at room temperature. Sections were incubated with primary antibodies overnight at 4 °C, followed by incubation with secondary antibodies and counterstaining with DAPI (Thermo Fisher) for 1 h at room temperature. The Terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) assay was conducted using the kit (C1089, Beyotime) based on the manufacturer’s protocol.

Hematoxylin and eosin (H&E) staining

Tissue sections were deparaffinized in xylene and rehydrated through a descending ethanol series into distilled water. Nuclear staining was performed with hematoxylin for 1–2 min, followed by rinsing under running tap water to remove excess dye. Differentiation was achieved using 1% acid alcohol, and nuclei were subsequently blued in 0.2% ammonia water. Sections were then counterstained with eosin for 20–30 s to visualize cytoplasmic and extracellular matrix components. Following staining, the sections were dehydrated through a graded ascending ethanol series, cleared in xylene, and mounted using a resin-based mounting medium. Bright-field images were acquired using a light microscope equipped with a digital imaging system.

Co-culture of embryonic intestinal organoids and mesenchymal cells

Embryonic intestinal organoids were cultured from wild-type, Lgr5-DTR^+/- or Rosa26-ZsGreen^flox/flox;Lgr5-CreERT2^+/- (for transient labeling of PVLCs) mouse embryos at E13.5 following previously described protocols (Zhao et al. 2022). PSI and DSI segments were mixed for the organoid culture and subsequent co-culture. The organoid culture medium (Advanced DMEM/F12, Thermo Fisher) consists of penicillin/streptomycin (Thermo Fisher), GlutaMAX (Thermo Fisher), N2 (Thermo Fisher), B27 (Thermo Fisher), N-acetylcysteine (Sigma-Aldrich), EGF (50 ng/mL, Peprotech), Noggin (100 ng/mL, Novoprotein), R-spondin1 (500 ng/mL, Novoprotein), CHIR-99021 (5 mM, Selleck), A83-01 (5 nM, MCE), and Y-27632 (100 nM, Selleck). To obtain total mesenchymal cells, wild-type embryonic intestines were enzymatically dissociated, and the resulting cell pellet was plated in culture dishes using DMEM (Corning) supplemented with 10% fetal bovine serum (FBS, Excell) for two-dimensional adherent culture. Both organoids and mesenchymal cells were passaged 2–3 times to achieve stable growth. Prior to co-culture, mesenchymal cells were enzymatically dissociated and counted, and organoids were resuspended and counted. Approximately 50 organoids were resuspended in 10 μL of Matrigel (Corning) and seeded into the center of the Transwell inserts (0.4 μm, Corning). Mesenchymal cells (4 × 10⁴) were resuspended in organoid culture medium and seeded into the lower chamber. The same medium was also added to the inserts to ensure adequate nutrition. After 24 h of co-culture, DT (0.1 μg/mL) was added to the medium and incubated for an additional 24 h. Mesenchymal cells were then harvested for downstream analyses.

Quantitative PCR (qPCR)

Total RNA was extracted from samples using TRIzol reagent (Thermo Fisher) following the manufacturer’s protocol. First-strand cDNA synthesis was carried out using the NovoScript Plus SuperMix (Novoprotein). Quantitative PCR was performed in technical triplicate using a LightCycler 480 System (Roche), with Gapdh serving as the reference gene. Relative gene expression levels were calculated using the ΔCT method. Primer sequences were summarized in Table S3.

Single-cell RNA sequencing (scRNA-seq) analysis

The E9.5-E15.5 scRNA-seq dataset was obtained from the published study (Zhao et al. 2022). The ratio of PVLCs was calculated based on RNA counts. To identify signature genes for PVLCs, the Seurat (v5.1.0) functions FindAllMarkers and FindMarkers were employed using default parameters. For a given cluster, positively expressed marker genes identified by FindAllMarkers were determined by comparing cells within the cluster to all other cells using the ‘wilcox’ method. GO/KEGG analyses were performed using Metascape.

Cell cycle scoring analysis

ScRNA-seq data were analyzed using Seurat (v5.1.0). Cell cycle-related genes were obtained from Seurat’s updated reference gene sets (cc.genes.updated.2019), including canonical S phase and G2/M phase markers. Only genes detected in the dataset were retained for downstream analyses. To quantitatively assess differences in cell cycle activity between Lgr5⁺ and Lgr5⁻ populations within each developmental stage, S phase (S.Score) and G2/M phase (G2M.Score) scores were calculated using Seurat’s CellCycleScoring function. Statistical comparisons between Lgr5⁺ and Lgr5⁻ cells were performed using two-sided Wilcoxon rank-sum tests, as the score distributions were non-Gaussian. For analyses involving multiple developmental stages, p-values were adjusted using the Benjamini–Hochberg false discovery rate (FDR) method. Statistical significance was reported using FDR-adjusted p-values.

CytoTRACE2 analysis

ScRNA-seq data were analyzed using Seurat (v5.1.0). To assess the differentiation potential of Lgr5⁺ cells, CytoTRACE2 (v1.1.0) was applied. Briefly, the processed Seurat object containing raw counts was used as input, and CytoTRACE2 scores were computed using the counts slot. Cells were classified as Lgr5⁺ or Lgr5^– based on expression of the Lgr5 gene (expression > 0 considered positive). CytoTRACE2 scores were subsequently added to the Seurat metadata, and the distributions of scores between PSI Lgr5⁺ and DSI Lgr5⁺ cells were compared using the Wilcoxon rank-sum test. Violin plots and boxplots were generated to visualize differences in differentiation potential.

Bulk RNA sequencing (RNA-seq)

Total RNA was extracted from samples using the RNeasy Mini Kit (Qiagen) according to the manufacturer’s instructions. cDNA library preparation was performed using the Ovation RNA-Seq System V2 kit (NuGEN). Libraries were subsequently sequenced on the Illumina NovaSeq platform using paired-end 150 bp reads (PE150). RNA-seq was conducted with two independent biological replicates. Genes with an absolute log₂ fold change > 1 and a p-value < 0.01 were considered significantly differentially expressed. GO/KEGG analyses were performed using Metascape.

Statistics

All statistical analyses were performed using GraphPad Prism 9 for Mac. Comparisons between two groups were evaluated using Student’s t-test, while multiple group comparisons were conducted using ordinary Two-way ANOVA followed by Tukey’s multiple comparisons. A P-value less than 0.05 was considered statistically significant. For certain statistical analyses, quantification was performed on a per-segment basis, with each segment representing a defined length of intestine as illustrated in the related images.

Abbreviations

Lgr5

Leucine-rich repeat-containing G protein-coupled receptor 5

PVLCs

Pre-villification Lgr5⁺ cells

scRNA-seq

Single-cell RNA sequencing

Shh

Sonic hedgehog

PSI/DSI

Proximal/Distal small intestine

DT

Diphtheria toxin

Authors’ contributions

L.Z., Y.X. and Y.S. performed experiments and analyzed results. W.S. and L.Z. analyzed sequencing data. S.L. provided the mouse strain. L.Z. and Y.-G.C. designed the project and wrote the manuscript, which was revised by Y.-G.C., L.Z. and H.L..

Funding

This work was supported by grants from the NSFC Excellence Research Group Program (32588201), the National Key Research and Development Program of China (2024YFA1307401, 2024YFA1107701, 2023YFA1800603), Shenzhen Medical Research Fund (B2302022), the Natural Science Foundation of Jiangxi Province (20224ACB209001), and the National Natural Science Foundation of China (32400681).

Data availability

The raw RNA-seq data have been deposited in the Genome Sequence Archive (GSA) (CRA035210).

Declarations

Ethics approval and consent to participate

All procedures involving animals were carried out in compliance with protocols approved by the Institutional Animal Care and Use Committee (IACUC) at Tsinghua University (approval number: 19-CYG1.G23-1).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests. Ye‐Guang Chen is the Editor‐in‐Chief of Cell Regeneration. He was not involved in the review or the decision related to this manuscript.