Single-cell guided prenatal derivation of primary fetal epithelial organoids from human amniotic and tracheal fluids

Gerli, Mattia Francesco Maria; Calà, Giuseppe; Beesley, Max Arran; Sina, Beatrice; Tullie, Lucinda; Sun, Kylin Yunyan; Panariello, Francesco; Michielin, Federica; Davidson, Joseph R.; Russo, Francesca Maria; Jones, Brendan C.; Lee, Dani Do Hyang; Savvidis, Savvas; Xenakis, Theodoros; Simcock, Ian C.; Straatman-Iwanowska, Anna A.; Hirst, Robert A.; David, Anna L.; O’Callaghan, Christopher; Olivo, Alessandro; Eaton, Simon; Loukogeorgakis, Stavros P.; Cacchiarelli, Davide; Deprest, Jan; Li, Vivian S. W.; Giobbe, Giovanni Giuseppe; De Coppi, Paolo

doi:10.1038/s41591-024-02807-z

Download PDF

Article
Open access
Published: 04 March 2024

Single-cell guided prenatal derivation of primary fetal epithelial organoids from human amniotic and tracheal fluids

Nature Medicine volume 30, pages 875–887 (2024)Cite this article

24k Accesses
1 Citations
2146 Altmetric
Metrics details

Subjects

Abstract

Isolation of tissue-specific fetal stem cells and derivation of primary organoids is limited to samples obtained from termination of pregnancies, hampering prenatal investigation of fetal development and congenital diseases. Therefore, new patient-specific in vitro models are needed. To this aim, isolation and expansion of fetal stem cells during pregnancy, without the need for tissue samples or reprogramming, would be advantageous. Amniotic fluid (AF) is a source of cells from multiple developing organs. Using single-cell analysis, we characterized the cellular identities present in human AF. We identified and isolated viable epithelial stem/progenitor cells of fetal gastrointestinal, renal and pulmonary origin. Upon culture, these cells formed clonal epithelial organoids, manifesting small intestine, kidney tubule and lung identity. AF organoids exhibit transcriptomic, protein expression and functional features of their tissue of origin. With relevance for prenatal disease modeling, we derived lung organoids from AF and tracheal fluid cells of congenital diaphragmatic hernia fetuses, recapitulating some features of the disease. AF organoids are derived in a timeline compatible with prenatal intervention, potentially allowing investigation of therapeutic tools and regenerative medicine strategies personalized to the fetus at clinically relevant developmental stages.

Derivation of self-renewing lung alveolar epithelial type II cells from human pluripotent stem cells

Article 15 November 2019

Controlled modelling of human epiblast and amnion development using stem cells

Article 11 September 2019

Human mid-trimester amniotic fluid (stem) cells lack expression of the pluripotency marker OCT4A

Article Open access 31 May 2019

Main

Although prenatal diagnosis of congenital anomalies adopts sophisticated genetic and imaging analyses^1,2, prediction of severity remains challenging, limiting patient-specific parental counseling. Patient stratification for prenatal therapy has shown level 1 evidence for improved outcomes in congenital diaphragmatic hernia (CDH)^3,4, twin-to-twin transfusion syndrome (TTTS)⁵ and myelomeningocele (MMC)⁶. For other conditions, such as lower urinary tract obstruction (LUTO)⁷, where vesico-amniotic shunting is technically possible, appropriate patient selection remains a hurdle. The lack of autologous models of developing human tissues is a bottleneck to these advancements.

Organoids are three-dimensional (3D) models recapitulating some biological and pathophysiological features of patient’s tissues in vitro. Autologous organoids can be derived from human embryonic stem cells⁸ or through induced pluripotent stem (iPS) cells. iPS cell-derived organoids have been generated from reprogrammed AF-derived fetal cells^9,10. These organoids resemble fetal-like tissues, but the extensive manipulation and lengthy differentiation protocols reduce patient fidelity and hinder applicability for prenatal disease modeling and targeted therapy.

In contrast, primary organoids, requiring minimal in vitro manipulation, have been derived from human discarded postnatal biological samples (for example, urine, menstrual flow, PAP smear and bronchoalveolar lavage^11,12,13,14). In prenatal medicine, primary organoids were generated from fetal tissues collected postmortem through biobanks¹⁵; however, accessing fetal tissues is associated with ethico-legal restrictions hampering their research^{16,17,18,19,20}. Current methods for primary fetal organoid derivation are destructive, restricting the use for prenatal modeling, diagnostics and regenerative medicine^21,22. Furthermore, fetal material can be procured mostly up to 20–22 weeks post-conception, making investigation of later developmental stages unfeasible²³. Here we present derivation of primary human fetal epithelial organoids, of multiple tissue identities, obtained from fetal fluids collected at prenatal diagnostic and therapeutic interventions during the second and third trimesters. This allows generation of organoids alongside continuation of pregnancy.

Around 30,000 amniocenteses are performed annually in the UK²⁴ for confirmatory diagnosis and advanced diagnostics². Moreover, amniodrainage is routine treatment for polyhydramnios^25,26 and TTTS²⁷. Finally, prenatal MMC repair and fetoscopic endoluminal tracheal occlusion (FETO) for CDH provide additional access to fetal fluids during pregnancy^3,4,6. The AF surrounds, supports and protects the fetus during development. Its origin and recirculation follow complex dynamics, progressing together with fetal and extra-embryonic tissue development. AF is highly heterogeneous in origin and composition, containing secretions and cells from various fetal tissues such as the gastrointestinal tract, kidney and lung^28,29; however, a detailed map of the AF epithelial populations and their potential is lacking. AF harbors stem cells from mesenchymal and hematopoietic niches^30,31,32 but most AF cells are epithelial and have only been partially characterized. Using single-cell sequencing, we investigated human AF epithelial cells (AFEpCs), highlighting that these originate from multiple developing tissues. We then explored whether this cell population contains lineage-committed progenitors capable of forming tissue-specific primary fetal organoids. Additionally, we expanded our findings to tracheal fluid (TF) epithelial cells obtained from CDH cases during FETO³³.

This work demonstrates that diverse epithelial stem cell populations shed into the AF and TF can form epithelial organoids resembling their tissue of origin. Autologous derivation of primary fetal organoids during continuing pregnancies could enable the development of advanced prenatal models and improve counseling and designing personalized therapies. Finally, AF organoids (AFOs) offer the possibility of researching later gestational stages currently inaccessible, being beyond the limits of termination of pregnancy.

Results

Single-cell mapping the human AF to investigate presence of tissue-specific fetal epithelial progenitors

We collected AF from 12 pregnancies (Supplementary Tables 1 and 2) and isolated, using fluorescence-activated cell sorting (FACS), the viable nucleated cells with an intact cell membrane (Fig. 1a and Extended Data Fig. 1a). We then performed 3′ single-cell RNA sequencing (scRNA-seq) and generated an unsupervised Uniform Manifold Approximation and Projection (UMAP) using Seurat v.4 (Fig. 1a and Extended Data Fig. 1b). The SingleR package was applied to automatically annotate the epithelial cluster based on primary human cell atlas data³⁴, then confirmed by expression of pan-epithelial marker genes (Fig. 1a,b and Extended Data Fig. 1c,d). Protein validation conducted using flow cytometry, confirmed broad presence of epithelial cell adhesion molecule (EpCAM) and ECAD in the majority of viable AF cells (Fig. 1c). We then probed the AFEpC cluster for the presence of specific gastrointestinal, kidney and lung signatures using single-cell gene set enrichment analysis (scGSEA; Fig. 1d). Finally, we scored these cells for canonical tissue-specific progenitor markers: LGR5, OLFM4, LRIG1, CDX2, CD44, LYZ, SMOC2 and PROCR (gastrointestinal); PAX2, PAX8, LHX1, JAG1, SIX2, RET, HNF4A, GATA3, POU3F3 and WT1 (kidney); and NKX2-1, SOX9, ETV4, ETV5, GATA6 and ID2 (lung). This indicated the presence of gastrointestinal, renal and pulmonary epithelial progenitor cells in AF (Fig. 1e).

**Fig. 1: Single-cell analysis of the AF content.**

Generation of primary fetal epithelial human AFOs

To investigate formation of AFOs we seeded viable AF cells in Matrigel droplets and cultured them in an ad hoc-defined generic epithelial medium without tissue-specific signals (Supplementary Table 3). Individual AF cells began proliferating and self-organizing to form 3D organoids, visible within 2 weeks. To establish clonal lines, individual AFOs were picked, dissociated into single cells and replated (Fig. 2a,b and Extended Data Fig. 2a). Using this method, we derived 423 AFO lines from 42 AF samples (16–34 weeks gestational age (GA); Supplementary Tables 1 and 2). The clonal origin of AFOs was further confirmed by single-cell AFEpC culture (Extended Data Fig. 2b). AFOs showed multiple morphologies, expanded up to passage 20 and successfully cryopreserved, providing evidence of self-renewal and long-term culture (Fig. 2c,d and Extended Data Fig. 2b–e). Organoid formation was observed in 89.7% of AF samples, with a median formation efficiency of 0.011% (one organoid formed per 7.4 × 10³ cells; Fig. 2b and Extended Data Fig. 2c). We found no meaningful association between GA and AFO formation efficiency. We imaged two distinct organoid morphologies through X-ray phase-contrast computed tomography (PC-CT), confirming the applicability of the method for organoid characterization (Fig. 2e). Immunostaining confirmed cell proliferation (Ki67) and lack of apoptosis (cleaved caspase 3) within AFOs (Fig. 2f). We confirmed the AFO’s epithelial identity by staining for pan-epithelial markers (EpCAM, ECAD and pan-cytokeratin) and showing absence of the mesenchymal marker platelet-derived growth factor receptor α (PDGFRα). Notably, AFO’s epithelium is polarized as demonstrated by basolateral integrin β4 (ITGβ4), apical F-actin and zonula occludens-1 (ZO-1)-positive luminal tight junctions (Fig. 2g).

**Fig. 2: Generation of primary fetal epithelial AFOs.**

We then conducted bulk RNA sequencing to investigate the tissue identity of each clonal AFO line. As a control, we isolated tissue-derived human organoids from fetal small intestine (n = 5), stomach (n = 3), lung (n = 6), kidney (n = 4), bladder (n = 1) and placenta (n = 1; Fig. 1e). Unsupervised principal-component analysis (PCA) conducted on bulk RNA-seq data (n = 121 AFOs, 23 AF samples; Supplementary Table 2) showed three AFO clusters colocalizing with small intestinal, pulmonary and renal fetal tissue-derived control organoids (Fig. 2h and Extended Data Fig. 2f). No colocalization was observed with bladder or stomach organoids, whereas one AFO clustered with the placental control. Cluster-based Gene Ontology analysis corroborated the single tissue identity of each AFO clone, showing upregulation of pathways specific to the assigned tissues (Extended Data Fig. 2g). Annotation was further confirmed by Euclidean hierarchical clustering (Extended Data Fig. 2h). Finally, scRNA-seq conducted on AFOs from each cluster (small intestine n = 3, kidney n = 6 and lung n = 4; Supplementary Table 2) highlighted multiple epithelial cell clusters (Fig. 2i). Overall, this provides evidence that intestinal, renal and pulmonary AFEpCs can give rise to clonal AFO lines reflecting their respective tissue of origin.

Characterization and maturation of SiAFOs

Small intestinal AFOs (SiAFOs) expanded consistently for more than ten passages, forming crypt-like structures (Fig. 3a and Extended Data Fig. 3a). Moreover, 5-ethynyl-2'-deoxyuridine (EdU) incorporation indicated cell proliferation at the SiAFO crypt base (Fig. 3a). Bulk RNA-seq of 23 SiAFOs showed expression of typical intestinal stem/progenitor (LGR5, OLMF4, LRIG1 and SMOC2), Paneth (LYZ), goblet (MUC2 and CLCA1), endocrine (CHGA) and enterocyte (ALPI, FABP1, VIL1, EZR, KRT20 and ATP1A1) markers (Fig. 3b). Immunostaining for the stem cell marker olfactomedin 4 (OLFM4) and intestinal epithelial cytokeratin 20 (KRT20) confirmed the presence of a crypt–villus axis. SiAFO immunostaining showed markers of numerous intestinal cell types (Fig. 3c,d) such as Paneth cells (lysozyme (LYZ)) and enterocytes (fatty acid-binding protein 1 (FABP1)). Notably, SiAFOs lack lung (NKX2-1) and kidney (PAX8)-specific markers (Extended Data Fig. 3b). scRNA-seq confirmed the presence of tissue-specific cellular identities such as intestinal stem cells, proliferating transit amplifying, enterocytes, goblet and enteroendocrine cells (Fig. 3e).

**Fig. 3: Characterization and maturation of small intestine AFO.**

We then performed a maturation assay culturing SiAFOs in intestinal-specific medium for 14 d. Upon maturation, SiAFOs displayed more prominent budding structures, resembling small intestinal crypt-like organization. Immunofluorescent (IF) staining showed chromogranin A (CHGA)-positive enteroendocrine cells and stronger presence of goblet cell marker mucin 2 (MUC2) compared to pre-maturation (Fig. 3f and Extended Data Fig. 3c). In addition, quantitative reverse transcription PCR (RT–qPCR) following DAPT treatment demonstrated downregulation of Notch target genes (HES1 and OLFM4), stem/progenitor cells, Paneth cells and Wnt target genes (LGR5, LYZ and AXIN2). While enterocyte and goblet cell markers (FABP1, ALPI and MUC2) were upregulated, the enteroendocrine marker CHGA was not. Notably, Notch inhibition did not drive increased expression of ATOH1 nor its downstream target DLL1 (Extended Data Fig. 3d).

To evaluate SiAFO functional capacity, we assessed the digestive activity of two small intestinal brush border enzymes, dipeptidyl peptidase IV (peptide hydrolysis; Fig. 3g) and disaccharidase (sucrose hydrolysis; Fig. 3h). Finally, to investigate the potential of SiAFOs for tissue engineering, we performed an intestinal ring formation assay (Fig. 3i). Three hours after seeding, SiAFOs started fusing; by day 2 a more complex tubular budding structure had formed with full ring maturation at day 10. MicroCT revealed SiAFO ring self-organization, forming a tube-like structure with lumen and buddings resembling intestinal architecture (Fig. 3j, Extended Data Fig. 3e and Supplementary Video 1). SiAFO rings manifested correct cell polarity (KRT20 and ITGβ4) and tight junctions (ZO-1), in addition to enterocytes (FABP1 and KRT20), enteroendocrine (CHGA), Paneth (LYZ) and goblet secretory cells (MUC2). Moreover, SiAFO rings maintained proliferation (Ki67) in the crypt-like portion (Fig. 3k, Extended Data Fig. 3f and Supplementary Video 2). The overall marker profile of SiAFO rings (bulk RNA-seq in Fig. 3b), revealed strong upregulation of genes typical of functionally differentiated intestinal cells, particularly of mucin secretory lineages, brush border enzymes and enterocytes.

Characterization and differentiation of KAFOs

Kidney tubule AFOs (KAFOs) expanded long-term (up to passage 10), while maintaining proliferation (Ki67; Fig. 4a and Extended Data Fig. 4a) and mostly manifested a compact morphology with several lines showing a cystic structure. Bulk RNA-seq of 54 KAFOs (19 AF samples; 18–34 weeks GA) showed expression of canonical developmental renal epithelial and nephron progenitor genes (PAX2, PAX8, LHX1 and JAG1), while lacking cap mesenchyme markers (SIX2, CITED1 and GDNF). Moreover, we detected distal (PCBD1, SLC41A3 and POU3F3) and proximal (ABCC1, ABCC3, ABCC4 and CUBN) tubule gene expression. Collecting duct marker GATA3 was expressed, while the loop of Henle marker UMOD was not. Podocyte markers (WT1, NPHS1 and NPHS2) were not expressed in KAFOs except PODXL and MAFB (Fig. 4b and Extended Data Fig. 4b). Based on this, we concluded that KAFOs have a tubuloid-like phenotype and express markers belonging to different renal tubule segments, confirmed by protein expression of PAX8 and LHX1. KAFOs also displayed the segment-specific kidney tubule proteins GATA3 and ECAD (distal tubule/collecting duct). Conversely, we detected LTL, representing proximal tubule identity. Of note, KAFOs exhibited a mixed tubular phenotype, coexpressing GATA3 and LTL or presenting only GATA3. The presence of polarized tubular microvilli was validated by immunofluorescence for acetylated tubulin (Ac-αTUB) (Fig. 4c,d and Extended Data Fig. 4c). Additionally, KAFOs showed increased intracellular thallium fluorescence compared to fetal lung organoids (FLOs), comparable to control fetal kidney organoids (FKOs), indicating the presence of functional voltage-gated potassium channels (Fig. 4e). Moreover, KAFOs display functional epithelial tight junctions, with apical ZO-1 (Extended Data Fig. 4d) and intact barrier integrity with 67.9% of KAFOs impermeable to inulin-FITC diffusion, decreased to 24.4% upon ethylenediaminetetraacetic acid (EDTA) treatment (Fig. 4f and Extended Data Fig. 4e). Further scRNA-seq characterization conducted on KAFOs confirmed the presence of multiple renal tubule cells such as ureteric tip/stalk and distal tubule or early nephron and nephron progenitor cells (Fig. 4g). Notably, some KAFO lines (21 of 54) expressed ureteric bud marker RET (Fig. 4b and Extended Data Fig. 4b). RET protein staining on the luminal side presented comparably to fetal kidney sections and FKOs. (Fig. 4h and Extended Data Fig. 4f). In addition, 85.7% of RET-expressing KAFO lines exhibited compact morphology, this decreased to 54.5% in RET-negative lines, exhibiting more cystic or mixed morphology (Fig. 4i). Finally, we adapted a reported protocol to differentiate KAFOs toward a distal/collecting duct phenotype. After 14 d of vasopressin and arginine-aldosterone stimulation, KAFOs manifested markers of the collecting duct (AQP2) and distal tubules (SLC12A1 and CALB1), which were lower in expansion medium (Fig. 4j and Extended Data Fig. 4g). Additionally, differentiated KAFOs displayed more CALB1-positive cells upon immunostaining (21 ± 6.6%) and increased CALB1 gene expression (Fig. 4k).

**Fig. 4: Characterization and differentiation of kidney tubule AFOs.**

Characterization and differentiation of LAFOs

Lungs are major cellular contributors to AF, continuously releasing fluid into the amniotic cavity. Consequently, lung AFOs (LAFOs) formed from the majority of our samples. We isolated, clonally expanded and sequenced 43 LAFO lines from 12 AF samples spanning 16–34 weeks GA. LAFOs were propagated up to 21 passages maintaining high proliferation, confirmed by Ki67 staining (Fig. 5a and Extended Data Fig. 5a). LAFOs can have cystic or compact morphology (Extended Data Fig. 5b,c). Bulk RNA-seq indicated expression of multiple pulmonary markers in 43 LAFOs, manifesting stem/progenitor cell markers (NKX2-1, FOXA2, SOX2, SOX9, TP63 and GATA6), alveolar type 1 (HOPX, PDPN, AGER and AQP5) and alveolar type 2 cell-related genes (SFTPA1, SFTPA2, SFTPB, SFTPC, SFTPD, ABCA3 and LAMP3). Mature basal cell markers KRT5, TROP2 and NGFR were rarely detected in LAFOs as well as the ciliated cell transcription factor FOXJ1, which showed sporadic low expression in the expansion medium. Furthermore, the early secretory cell marker SCGB3A2 was highly expressed compared to the absent mature club cell marker SCGB1A1. Secretory goblet cell marker MUC5AC was expressed in tissue-derived control FLOs, but several LAFO lines showed low/negligible expression. Last, while the neuroendocrine cell marker ASCL1 was expressed in almost all LAFOs, only a small number showed expression of CHGA (Fig. 5b and Extended Data Fig. 5d). Immunostaining demonstrated homogenous protein expression of the stem cell markers NKX2-1 and SOX2 and basal cell marker P63 (Fig. 5c,d). Pro-surfactant protein C (proSFTPC) was absent (Extended Data Fig. 5e).

**Fig. 5: Characterization and differentiation of lung AFOs.**

We then investigated the proximal and distal differentiation of LAFOs. scRNA-seq indicated successful proximal airway differentiation with newly formed specialized cell clusters such as ciliated and deuterosomal cells, associated with an increase in secretory cells (Fig. 5e; top UMAP). Notably, we also observed polarized epithelium with motile cilia on LAFO luminal surfaces (Supplementary Video 3). Immunofluorescent staining confirmed the presence of luminal Ac-αTUB cilia on proximally differentiated LAFOs, which was further corroborated by ciliated epithelia marker FOXJ1 nuclear expression (Fig. 5f). In addition, the appearance of keratin 5 (KRT5; mature basal cells) and secretory marker mucin (MUC5AC) concomitantly with the maintenance of SOX2, corroborated proximal lung differentiation (Fig. 5f and Extended Data Fig. 5f). FOXJ1-positive cells (62.8 ± 8.2%) increased in the differentiated LAFOs compared to controls in expansion, whereas the number of KRT5-positive cells (3.8 ± 2.2%) did not (Fig. 5g and Extended Data Fig. 5f). Moreover, proximal LAFOs showed increased gene expression of airway markers such FOXJ1, TUBA1A and SCGB1A1, when compared to undifferentiated controls (Fig. 5h). Cilia were analyzed in detail by transmission electron microscopy (TEM), displaying normal rootlets with associated mitochondria. The ciliary axonemes also displayed normal structures with radial spokes and a normal central microtubule pair, outer and inner dynein arms (Fig. 5i and Extended Data Fig. 5g). When directed toward a distal phenotype, LAFOs increased protein expression of the AT2 marker SFTPB, displaying different cellular localization in independent LAFO lines. Some presented within intracellular granules, others as luminal surfactant accumulation, possibly indicating a more mature state (Fig. 5j and Extended Data Fig. 5h), alongside increased expression of SFTPA1, SFTPA2, SFTPB and SFTPD (Fig. 5k). Notably, scRNA-seq revealed that distally differentiated LAFOs showed increased SCGB3A2⁺/SFTPB⁺ lower airway progenitor cells, emergence of a Hillock-like cluster and reduction of club cells (Fig. 5e; bottom UMAP). Finally, ultrastructural analysis revealed that distal LAFOs contain lamellar bodies with a normal structure, typical of surfactant-secreting cells (Fig. 5l). Overall, this indicates progression of LAFOs toward more mature lung phenotypes.

Characterization of AF and TF organoids from CDH fetuses

CDH is a rare congenital malformation where the diaphragm fails to close, with herniation of the abdominal organs into the chest (OMIM 142340). Consequently, fetal lungs are mechanically compressed, limiting growth of both respiratory and vascular compartments³⁵. To test our platform for disease modeling, we derived lung organoids from both AF and TF of fetuses with severe/moderate CDH-related lung hypoplasia (Fig. 6a and Supplementary Table 1). Fluids were obtained at FETOs, from patients not receiving steroids^3,4. AFOs were derived as above; however, due to the low TF sample volume (1–3 ml) and high cell viability (60–75%) we omitted sorting to preserve cell numbers. Similar to control (CT) LAFOs (Fig. 5), we successfully generated CDH AFOs from 16 AF samples (16 of 20; 80%) and CDH TF organoids (TFOs) from 7 TF samples (7 of 17; 41.2%; Fig. 6a and Extended Data Figs. 2a and 6a,b). CDH LAFOs and lung TFOs (LTFOs) expanded up to passage 10, manifesting morphology consistent with CT LAFOs (Fig. 6b–d). PCA confirmed that all TF-derived organoids had lung identity and demonstrated clustering of CDH LAFOs/LTFOs with an associated shift from CT LAFOs (Extended Data Fig. 6c). CDH organoids expressed NKX2-1, SOX2 and P63 lung markers, along with Ki67 (Fig. 6b,c,e). Notably, CDH organoids showed higher SOX9 immunofluorescence positivity than CT LAFOs (Fig. 6e,f and Extended Data Fig. 6d), suggesting a more prominent stem/progenitor identity³⁶. CDH LAFOs/LTFOs were generated from samples taken at the two FETO-related interventions: (1) occlusive endotracheal balloon insertion (28–31 weeks GA); and (2) balloon removal (32–34 weeks GA; Extended Data Fig. 6e). Paired analysis was not possible due to limited sample availability and therefore pooled analysis was performed. Our RNA-seq data (n = 30 CDH LAFOs from eight patients, n = 23 CDH LTFOs from four patients) highlighted expression of lung epithelial stem/progenitor markers at levels similar to GA-matched CT LAFOs (Fig. 6g and Extended Data Fig. 7). SOX9 expression was remarkably downregulated in CDH organoids generated post-FETO, consistent with increased tissue maturation (Fig. 6h). Comparative analysis between CDH and GA-matched control LAFOs showed a reduction in differentially expressed genes (DEGs) (P < 0.01, logfold change (FC) > 2) between organoids generated from samples before (380) and after (102) FETOs (Fig. 6i, with further comparisons shown in Extended Data Fig. 8). Gene Ontology analysis identified upregulated pathways related to surfactant production and metabolism in CDH organoids, upregulation of phosphatidylcholine metabolism, and downregulation of pathways related to laminin interaction, integrin/ECM interaction and ECM proteoglycans (P < 0.05, logFC > 1), more evident in post-FETO organoids (Extended Data Fig. 6f,g). SFTPC expression was also validated through immunofluorescent staining (Fig. 6j).

**Fig. 6: Generation, differentiation and characterization of LAFOs and LTFOs from CDH pregnancies.**

We then differentiated CDH LAFOs/LTFOs. Similar to controls, proximal CDH LAFOs showed cilia via TEM, acetylated α-tubulin (Ac-αTUB), FOXJ1 and SOX2 expression (Fig. 6k and Extended Data Fig. 6h). The ciliary beating frequency (CBF) for proximal CT and CDH LAFOs was shown to be within the normal physiological range using a high-speed camera (Fig. 6l and Supplementary Video 4). Finally, when subjected to distal differentiation, CDH organoids showed lamellar bodies and expressed SFTPB (Fig. 6m). scRNA-seq found substantial differences in cellular composition between differentiated and undifferentiated CDH LAFOs, compared to control LAFOs (Fig. 6n). CDH LAFOs/LTFOs showed decreased basal and club cell populations, consistent with CDH models³³. We detected an increased percentage of pulmonary neuroendocrine cells in CDH LAFOs in expansion and differentiation, consistent with human CDH tissue specimens and animal models^37,38. Notably, we revealed increased representation of AT2 cells in CDH LAFOs/LTFOs in conjunction with further differences in proximally and distally differentiated CDH organoids. CDH LAFOs did not display the expected increase in club cells following proximal differentiation but showed an increased deuterosomal cell fraction. Last, distalized CT LAFOs displayed an appearance of Hillock cells and lower airway progenitor cells, absent in distal CDH LAFOs. Of note, and in contrast, distalized CDH LTFOs displayed fewer basal cells (Fig. 6n).

Discussion

This work demonstrates that the AF contains tissue-specific fetal epithelial progenitor cells originating from various developing organs. We show that, under defined culture conditions, these cells form epithelial organoids resembling their tissues of origin (small intestine, kidney and lung). Finally, we provide evidence that lung organoids derived from AF and TF of fetuses affected by CDH exhibit features of the disease.

Personalized therapeutic modeling of congenital conditions must be implemented prenatally. Here, we demonstrate derivation of autologous primary fetal organoids from AF and TF, sampled for clinical purposes while allowing continuation of pregnancy. The AFO technology uses widely available samples, requiring minimal manipulation and applying established culture techniques. Fluid sampling to AFO characterization and expansion is implemented within 4–6 weeks, a timeline relevant to prenatal intervention, counseling and therapy, with great advantages over iPS cell-dependent methods, requiring 5–9 months to produce organoids^9,10,39.

For this work we compiled a single-cell atlas of unperturbed human AF. AF cells are known to be mostly epithelial, with their origin often ascribed to skin, kidneys and fetal membranes⁴⁰. Initial AFEpC automatic annotation and reference integration attempts were unsuccessful. The AF has a nonspecific ambient RNA signature and contains cells that lack adhesion and are exposed to nutrients/gas imbalances. Moreover, currently available fetal atlases only cover earlier gestational stages and the cells analyzed are captured within their native tissue environment^{41,42,43,44,45} (Extended Data Fig. 1e). Our map expands on the heterogeneity of the human AFEpC, showing these to exhibit multiple tissue origins being therefore distinct from previously reported placental-derived amniotic epithelial cells^46,47. Within AFEpCs we identified gastrointestinal, renal and pulmonary epithelial stem/progenitor cells. In culture, these lineage-committed progenitors generated clonal organoids capable of long-term expansion. Remarkably, upon maturation, SiAFOs, KAFOs and LAFOs acquired further tissue-specific differentiation hallmarks.

Derivation of SiAFOs was rare, with success in only two AF samples (16–17 weeks GA, one obtained from a termination of pregnancy). The occurrence of AFEpCs with intestinal progenitor features is interesting, as after the breakdown of the anal membrane (12 weeks GA), sphincters are expected to retain the intestinal content⁴⁸; however, colonocytes are reported in second trimester AF⁴⁹ suggesting possible release of gastrointestinal cells later in pregnancy. Thus, SiAFOs could find therapeutic use, for example where prenatal diagnosis of short-bowel syndrome is made. Notably, matured SiAFOs acquired further tissue-derived small intestinal organoid hallmarks⁵⁰.

KAFOs were easily derived across all gestational ages, likely due to excretion from developing kidneys²⁹. KAFOs recapitulated tubuloid identity¹⁴ through expression of nephron progenitor and tubule markers. Upon differentiation, KAFOs upregulated functional renal proteins such as AQP2, SLC12A1 and CALB1.

LAFO derivation was frequent, due to continuous provision of lung progenitors to AF via regular circulation of fluid through fetal lungs⁵¹, interrupted during FETO, possibly explaining the differences between CDH LAFOs and CDH LTFOs⁵². Current CDH patient stratification relies on simple and reliable prenatal imaging parameters⁵³. FETO is of great survival benefit to the most severely affected patients, but better predictive functional biomarkers are needed as 60% of treated fetuses do not survive to hospital discharge⁵⁴. For instance, previous work on TF showed a different miRNA signature (miR-200) in patients who were responsive to the intervention; however, we could not assess this in our dataset due to transforming growth factor β and BMP inhibition interference⁵⁵. We showed that CDH LAFOs/LTFOs manifest altered expression of surfactant protein genes. Consistently, AT2/AT1 increase was previously suggested as a hallmark of hypoplastic and CDH lungs^56,57. Our RNA-seq analyses demonstrated increased AT2 gene expression in CDH organoid cells compared to GA-matched controls, suggesting that these manifest some disease features, supporting their potential use in prenatal regenerative medicine. Future investigations will determine whether there is correlation between alterations in CDH LAFOs/LTFOs and clinical outcomes of the fluid donors. If successful, AFOs could become a complementary prenatal prognostic tool. It should be noted that AF sampling after FETO is conducted before balloon removal, therefore CDH LAFOs may not reliably reflect changes in the fetal lung in response to FETO. CDH LTFOs that are sampled from the occluded trachea will do so⁵⁸. Of note, although TF sampling at the time of FETO is less widely applicable, this gives access to pure fetal lung/airway cells that produce only lung organoids.

Among 239 organoids sequenced, we found only one to have an unknown phenotype (Extended Data Fig. 2e), ascribed to a non-clonality at the picking stage. For future endeavors, optimizing AFO derivation could involve cost-effective methods such as PCR and image-based characterization, rather than RNA sequencing. Disease modeling requires a larger CDH patient cohort for platform validation to finely model the disease and test treatments. Finally, the present system is limited to the epithelial compartment and cannot model complex conditions involving, for example, mesenchymal and vascular compartments. This might be overcome by culturing organoids with mesenchymal and endothelial cells. Notably, mesenchymal and hematopoietic cells can be isolated from AF^30,59, and direct reprogramming of AF cells to the endothelial lineage has been reported⁶⁰. Hence, AF can provide different cell lineages to generate more-complex prenatal models. Future work will also focus on deriving organoids for each AF-exposed tissue. As an example, we identified one AFO with placental identity, suggesting the possible presence of additional organoid-forming AFEpCs.

In conclusion, we report derivation of fetal epithelial organoids of different tissues from fetal fluids, through minimally invasive sampling. This is achieved from continuing pregnancies, within a GA window beyond the one currently accessible from fetal tissue obtained from termination of pregnancies. Small intestine, kidney tubule and lung AFOs are expandable and can be functionally matured with great potential for regenerative medicine and personalized disease modeling.

Methods

Ethics and informed consent

Fetal fluid samples were collected from all participants after written informed consent in compliance with the Declaration of Helsinki. Ethical approval was given by the NHS Health Research Authority, in accordance with the Governance Arrangements for Research Ethics Committees and complied fully with the standard operating procedures for research ethics committees in the UK. Fetal fluid samples were collected from the University College London Hospital (UCLH) Fetal Maternal Unit (FMU) under the London Bloomsbury Research Ethics Committee (REC 14/LO/0863 IRAS ID 133888) in compliance with UK national guidelines (Review of the Guidance on the Research Use of Fetuses and Fetal Material, 1989, Cm762). Fetal fluid samples were also collected and used at UZ Leuven (ethics committee no. S53548). Control fetal tissue samples were sourced via the Joint Medical Research Council (MRC)/Wellcome Trust Human Developmental Biology Resource under informed ethical consent from all donors, undertaken with Research Tissue Bank ethical approval (project 200478: UCL REC 18/LO/0822, IRAS ID 244325; Newcastle REC 18/NE/0290, IRAS ID 250012). Pediatric intestinal tissue sample was obtained upon informed consent and ethical approval for the use of human tissue obtained from the East of England, Cambridge Central Research Ethics Committee (REC 18/EE/1050). Details on recruitment and ethics oversight are also provided in the Reporting Summary.

AF collection and isolation of the viable cell fraction

Euploid AF samples (amniocenteses and amniodrainages) were collected from UCLH FMU and UZ Leuven as part of standard patient clinical care. After collection, fluids were stored at 4 °C until processing. AF samples were passed through 70-μm and 40-μm cell strainers and transferred to 50-ml tubes before being centrifuged at 300g for 10 min at 4 °C. Supernatant was discarded, pellet resuspended in 5–10 ml FACS blocking buffer containing 1% FBS and 0.5 mM EDTA in PBS and transferred to FACS tubes. Cells were incubated with 5 μg ml⁻¹ Hoechst (Sigma-Aldrich, 33342) for 40 min at 37 °C and then counterstained with 2 μg ml⁻¹ PI (Sigma-Aldrich, P4170) for 5 min at room temperature. Viable cells were sorted using a FACSAria III (BD), unselected for side and forward scatter, but gated for Hoechst⁺ and PI⁻. Viability was confirmed through a Live/Dead Acridine Orange/PI fluorescent Luna cell counter.

Derivation and culture of human AFOs

Viable AF cells were resuspended in cold Matrigel (Corning, 354230) and plated at a density of 6 × 10⁴ live cells per 30 μl droplet onto a pre-warmed 24-well plate. Cells were cultured in an ad hoc chemically defined generic medium (AFO expansion medium; Supplementary Table 3) supplemented with Rho-kinase inhibitor (ROCKi; Tocris) and Primocin (Invivogen, NC9141851) for the first 3 d of culture. To establish clonal organoid lines, single organoids formed at P0 (day 14–20) were manually picked under the microscope to be clonally expanded. Each individual organoid was assigned an ID line and transferred to a 0.5-ml tube pre-coated with 1% BSA (Sigma-Aldrich). Organoids were resuspended in TrypLE (Thermo, 12605010) and incubated for 5 min at 37 °C. After digestion, organoids were disaggregated by pipetting and an additional 400 μl ice-cold Advanced DMEM/F12 supplemented with Glutamax, P/S and HEPES (ADMEM+++) were added. Organoids were precipitated with a minicentrifuge for 2 min and a second washing passage with ADMEM+++ was repeated. After centrifugation, the pellet was resuspended in 20 μl cold Matrigel (Corning, 354230) and plated in a pre-warmed 48-well plate. The plate was incubated for 20 min at 37 °C and AFO expansion medium was added with ROCKi and Primocin (Invivogen, NC9141851) for the first 3 d (Supplementary Table 3). Medium was replaced every 3–4 d. After approximately 10–14 d, grown organoids were passaged as described below.

Derivation and culture of human fetal LTFOs

Euploid fetal TFs were collected during procedures of FETO carried out at the UCLH or UZ Leuven (ethical approval REC 14/LO/0863 IRAS 133888 and ethics committee number S53548, respectively), kept refrigerated and processed within 24–48 h. We collected TF samples before the insertion of the balloon (pre-FETO) and after its removal (post-FETO). Due to the nature of the TF samples, mostly small (1–3 ml) and containing a majority of living cells, FACS sorting was not performed as it was not deemed necessary. TFs were transferred into 15-ml tubes on ice, washed with ice-cold ADMEM+++ and centrifuged at 300g for 5 min at 4 °C. Supernatant was discarded and cells were resuspended in 1 ml ADMEM+++. Cells were counted with Trypan blue to normalize and exclude dead cells and then plated in Matrigel (Corning, 354230) droplets as described for the AF cells above. Plates were incubated for 20 min at 37 °C and human fetal lung organoid medium (Supplementary Table 3) supplemented with ROCKi and Primocin (Invivogen, NC9141851) was added. The medium was changed every 3 d. LTFOs were clonally expanded with the same methodology described above for the AFOs and passaged as described below.

Passaging of organoids

Depending on number and size, organoids were passaged for expansion into a 24 or 12-well plate after clonal picking. Afterwards organoids were usually split 1:2 to 1:3 every 10–14 d of culture. For passaging, the medium was aspirated and ice-cold ADMEM+++ was added to each well. Matrigel droplets were disrupted and collected into a 15-ml tube on ice. Organoids were washed with 10 ml cold ADMEM+++ and centrifuged at 300g for 5 min at 4 °C. Large and cystic organoids were resuspended in 1 ml ADMEM+++ and mechanically disaggregated using a P1000 pipette. If small, organoids were instead disrupted enzymatically as follows. The medium was aspirated and organoid pellets were resuspended in 300 μl TrypLE. After incubation for 3–7 min at 37 °C, organoids were pipetted with P200 to break them down into single cells. Cold ADMEM+++ was added up to 10 ml and the sample was centrifuged at 300g for 5 min at 4 °C. The supernatant was discarded and the cell pellet was resuspended in Matrigel (Corning, 354230) and plated. The plate was incubated for 20 min at 37 °C to allow the Matrigel to solidify, upon which AFO expansion culture medium was added with ROCKi. The medium was changed every 3 d.

Fetal tissue collection and generation of a control primary fetal organoids library

Control fetal tissue organoid derivation was conducted as follows:

Human fetal tissue-derived small intestinal organoids

Fetal small intestines were processed as previously described⁶¹. The tissue was washed with PBS, cleared of any mesenteric tissue and fat, then cut longitudinally. The villi were scratched away using a glass coverslip. The remaining tissue was cut into 2–3-mm pieces, washed vigorously and incubated in 2 mM EDTA in PBS for 30 min for 5 min on an orbital shaker. The supernatant containing the intestinal crypts was centrifuged at 800g for 5 min at 4 °C. After being washed in ADMEM+++ and centrifuged, the pellet was resuspended in Matrigel (Corning, 354230) and plated in presence of Primocin (Invivogen, NC9141851) and ROCKi. The recipe for the medium is in Supplementary Table 3.

Human fetal tissue-derived kidney tubule organoids

The process was adapted following a previously published protocol for deriving adult tubuloids¹⁴. Briefly, fetal kidneys were collected, washed in ice-cold HBSS and minced to isolate the cortical tissue. The tissue was washed in 10 ml basal medium and the supernatant was removed when the tissue pieces settled at the bottom of the tube. After being washed several times in ADMEM+++, the tubule fragments were isolated by 1 mg ml⁻¹ collagenase digestion (C9407, Sigma) on an orbital shaker for 30–45 min at 37 °C. Fragments were further washed in basal medium with 2% FBS and centrifuged at 300g for 5 min at 4 °C. Pellets were resuspended in Matrigel (Corning, 354230) and cultured in kidney organoid medium (Supplementary Table 3) supplemented with ROCKi and Primocin (Invivogen, NC9141851).

Human fetal tissue-derived lung organoids

Fetal lung tissue was processed by adapting a previously published protocol¹¹. Briefly, fetal lungs were minced and washed in ADMEM+++. Tissue fragments were digested in ADMEM+++ containing 1 mg ml⁻¹ collagenase (C9407, Sigma) on an orbital shaker at 37 °C for 30–60 min. The digested tissue was shaken vigorously and strained over a 100-μm filter. Tissue fragments were washed in ice-cold basal medium with 2% FBS and centrifuged at 300g for 5 min at 4 °C. Supernatant was discarded and pellet was resuspended in Matrigel (Corning, 354230) and cultured in lung medium (Supplementary Table 3) supplemented with ROCKi and Primocin (Invivogen, NC9141851).

Human fetal tissue-derived stomach organoids

Fetal stomach organoids were isolated from specimens following an established dissociation protocol¹⁵. Briefly, stomachs were cut open and mucus was removed with a glass coverslip and mucosa was stripped from muscle layer. Mucosa samples were cut into pieces of 3–5 mm and washed in HBSS until the supernatant was clear. The tissue was incubated in chelating buffer supplemented with 2 mM EDTA for 30 min at room temperature. Tissue fragments were squeezed with a glass slide to isolate the gastric glands, which were transferred in ADMEM+++, strained through at 40 μm and centrifuged at 300g for 5 min at 4 °C. The pellet was resuspended in Matrigel (Corning, 354230) and plated. Gastric medium (Supplementary Table 3) was added with ROCKi and Primocin (Invivogen, NC9141851).

Human tissue-derived placental and fetal bladder organoids

Fetal bladder or placental biopsies were isolated and washed with ice-cold HBSS. Briefly, the samples were minced and washed in ADMEM+++. Tissue fragments were digested in ADMEM+++ containing 1 mg ml⁻¹ collagenase (C9407, Sigma), 2.4 U ml⁻¹ dispase (Thermo Fisher, 17105041) and 0.1 mg ml⁻¹ DNase (Merck, 260913) on an orbital shaker at 37 °C for 20–60 min. The digested tissue was then strained through at 70 μm followed by 40 μm. Tissue fragments were washed in ice-cold basal medium with 10% FBS in 50-ml Falcon tubes and centrifuged at 300g for 10 min at 4 °C. The supernatant was discarded and the pellet was resuspended in Matrigel (Corning, 354230) and cultured in expansion medium (Supplementary Table 3) supplemented with ROCKi and Primocin (Invivogen, NC9141851).

Organoid cryopreservation and thawing

After 7–10 d of culture, organoids were dissociated enzymatically as described above. The final cell pellet was resuspended in 1:1 ADMEM+++ and freezing medium (80% FBS and 20% dimethylsulfoxide). Cryovials were stored at −80 °C overnight and then transferred to LN2 for long-term storage. For organoid thawing, cryovials were equilibrated on dry ice and then placed at 37 °C. Vial content was rapidly transferred to 15-ml Falcon tubes containing 9 ml ice-cold ADMEM+++, then centrifuged at 300g for 5 min at 4 °C. The supernatant was discarded and the pellet was resuspended in cold Matrigel (Corning, 354230). After 20 min of incubation at 37 °C, the medium was supplemented with ROCKi and replaced after 3 d.

Evaluation of organoid formation efficiency and area

Organoid formation efficiency was determined by counting the number of organoids at P0. The total number of formed organoids per well was manually counted approximately 14 d after seeding of the AF cells in Matrigel. The efficiency was determined by calculating the total number of grown organoids divided by the number of viable single cells initially plated. The organoid area was determined by measuring the perimeter of each organoid in different ×5 fields acquired at the Zeiss Axio Observer A1 and using ImageJ software. The number of formed organoids over passages was calculated by counting the organoids in each ×5 field and normalized by field size.

Organoid maturation/differentiation

For SiAFOs, after manual passaging, organoids were seeded in triplicate in Matrigel (Corning, 354230) and cultured in AFO expansion medium. After approximately 7 d, human small intestine medium was used (Supplementary Table 3) for 14 d. Basal culture medium was the same but without the addition of CHIR99021. DAPT (Notch inhibitor) 10 μm was added to basal culture medium for 48 h to stimulate differentiation. For KAFOs, after either manual or enzymatic passaging, organoids were seeded in triplicate in Matrigel and cultured in AFO expansion medium. After approximately 7–10 d, distal/collecting duct kidney differentiation medium (Supplementary Table 3) was used for 14 d⁶². For LAFOs, after manual passaging, organoids were seeded in triplicate in Matrigel and cultured in AFO expansion medium for approximately 10 d. For lung proximal differentiation, PneumaCult ALI Medium (Stem Cell Technologies, 05001) was used for 14 d. For distalization, organoids were exposed for 14 d to a previously reported medium⁶³ (Supplementary Table 3).

Intestinal ring formation in collagen hydrogel

The assay was adapted from a previously published protocol⁶⁴. SiAFOs were expanded for 7 d in AFO expansion medium. After the medium was removed and the wells were washed once with PBS, Matrigel droplets were transferred to 1% BSA pre-coated Eppendorf tubes and dissolved in Cell Recovery Solution (Corning, 354253) for 45 min on ice. Organoids were centrifuged at 300g for 5 min at 4 °C and the supernatant was removed. The organoids were collected and resuspended in 120 μl collagen hydrogel (collagen type I 0.75 mg ml⁻¹, DMEM-F12 1×, HEPES 1 M, MilliQ to volume, pH 7) and plated in an ultra-low adherent 24-well plate making a circular shape around the edges of the well. The plate was incubated for 30 min at 37 °C and medium was added in the middle of the well to allow homogeneous detachment of the collagen ring. Intestinal rings were cultured in suspension in intestinal medium for 10 d.

Whole-mount immunofluorescence

Before fixation, organoids were retrieved from Matrigel using Cell Recovery Solution for 45 min on ice. Organoids were collected in a 15-ml tube precoated with 1% BSA in PBS and fixed with 4% PFA for 20 min at room temperature. Samples were washed three times with PBS for 5 min and spun down at 300g for 5 min at 4 °C. Whole-mount immunostaining was performed by blocking and permeabilizing the organoids with PBS-Triton X-100 0.5% with 1% BSA for 1 h at room temperature. Primary antibodies were incubated in blocking/permeabilization buffer for 24 h at 4 °C in agitation. After being extensively washed with PBS-Triton 0.2%, organoids were incubated with secondary antibodies and Hoechst overnight at 4 °C in agitation. After incubation, organoids were further washed and resuspended in PBS in preparation for confocal imaging. For tissue clearing of the SiAFO ring, a previously published protocol was adapted⁶⁵. EdU staining was performed with the Click-iT EdU Alexa Fluor 568 Imaging kit (Life Technologies) following the manufacturer’s protocol. A full list of antibodies is available in Supplementary Table 4.

Image acquisition

Phase-contrast images were acquired using Zeiss Axio Observer A1 and Zeiss ZEN (v.3.1) software. IF images of whole-mount staining and sections were acquired on a Leica SP5 or a Zeiss LSM 710 confocal microscope using ×20, ×25, ×40 and ×63 immersion objectives. Image analysis and z-stack projections were generated using ImageJ (https://imagej.nih.gov/ij/). Videos of SiAFO rings were processed using Imaris (v.2).

X-ray PC-CT

The imaging of the organoids was performed using PC-CT at beamline I13-1 (coherence branch) of the Diamond Light Source (www.diamond.ac.uk). The X-ray energy was 9.7 keV and the system resolution was 1.6 μm. The organoids were imaged embedded in HistoGel (Epredia HistoGel). The PC-CT scan entailed the acquisition of 2,000 equally spaced projections through a 180º rotation of the specimen. The total scan time was approximately 1 h. The ‘single image’ phase retrieval operation⁶⁶ was applied to the acquired projections, with the estimated phase to attenuation ratio (referred to as δ:β ratio) set at 250. Both phase retrieval and tomographic slice reconstructions were performed using Savu⁶⁷ and 3D images were generated using Drishti^68,69.

microCT

After 10 d of culture, SiAFO rings were collected and processed for microCT scanning. Rings were fixed for 1 h in 4% PFA and extensively washed in PBS. The specimen was iodinated overnight by immersion in 1.25% potassium triiodide in 10% formalin solution. After being rinsed in deionized water, the sample was wrapped in laboratory wrapping film and mounted in HistoGel (Epredia, HG-4000-012) in a 1.5-ml tube. MicroCT scanning was performed using a Nikon Med-X microCT scanner (Nikon Metrology). The specimen was mounted and held in pace using a drill chuck to ensure centralized rotational positioning. Whole specimen scans were acquired using an X-ray energy of 120 kV, current of 50 μA, exposure time of 1,000 ms, four frames per projection, a detector gain of 24 dB and an optimized number of projections of 2,258. A tungsten target was used and an isotropic voxel size of 3.57 μm was achieved. Reconstructions were carried out using modified Feldkamp filtered back projection algorithms with CTPro3D (Nikon, Metrology, v.XT 5.1.43) and post-processed using VGStudio MAX (Volume Graphics, v.3.4).

SiAFO dipeptidyl peptidase IV and disaccharidase assays

For both assays organoids were plated in 48-well plates, 15 μl basement membrane extract per well, in triplicate. For the dipeptidyl protease assay, organoids were washed in PBS and then incubated at 37 °C with 200 μl per well in Gly-Pro p-nitroanilide hydrochloride (Sigma, G0513) dissolved in PBS at a concentration of 1.5 mM (or PBS alone in control wells). During incubation, samples were agitated on an orbital shaker (60 r.p.m.) and supernatants were sampled at 20, 40 and 60 min. Absorbance (415 nm) was measured with a plate reader (Bio-Rad) and the concentration was determined by comparison to a 4-nitroaniline (Sigma, 185310) standard curve (0–200 µg ml⁻¹) and normalized per mg organoid lysate protein (Pierce BCA Protein Assay kit, Thermo Scientific). For the disaccharidase assay, basement membrane extract was removed by adding Cell Recovery Solution for 40 min at 4 °C and organoids were placed in 1.5-ml Eppendorf tubes, one well per Eppendorf. Organoids were washed once in PBS and then incubated at 37 °C with 200 µl per Eppendorf of 56 mM sucrose in PBS (or PBS alone for controls). During incubation, samples were agitated on an orbital shaker (60 r.p.m.). Supernatants were collected at 0, 30, 60, 90 and 120 min and sampled for glucose detection using the Amplex Red glucose/glucose oxidase assay kit (Thermo Fisher, A22189) according to the manufacturer’s protocol. Briefly, 50 µl of the reaction working solution was added to 50 µl of the test samples (diluted 1:4 with 1× reaction buffer) in a 96-well black flat-bottom microtiter plate in duplicate and incubated in the dark for 30 min at room temperature. Fluorescence (excitation 535 nm and emission 590 nm) was measured using a Tecan microplate reader (Infinite M1000 PRO). Glucose concentration was determined by comparison to a glucose standard curve and normalized per mg organoid lysate protein (Pierce BCA Protein Assay kit, Thermo Scientific, 23225).

KAFO potassium ion channel assay

KAFOs were expanded at least in triplicate in 96-well plates (10 µl Matrigel per well). FluxOR II Green Potassium Ion Channel Assay was performed according to the manufacturers’ instructions (F20017, Thermo). Briefly, medium was removed and 80 µl 1× loading buffer was added to each well and incubated for 30 min at room temperature and 30 min at 37 ºC to facilitate dye entry. After removing the loading buffer, 80 µl assay buffer was added to each well. A microplate reader (SoftMax Pro v.7.1.2) was set at an excitation wavelength of 480 nm and an emission wavelength of 545 nm and recorded every 5 s for 5 min. After 5 min of plate recording, voltage-gated channels were stimulated with 20 µl High Potassium Stimulus Buffer containing 2 mM thallium sulfate (Tl₂SO₄) and 10 mM potassium sulfate (K₂SO₄). The plate was read once again every 5 s for 5 min. The average of replicates was normalized to the control (assay buffer) and the number of cells.

KAFO epithelial barrier integrity assay

Expanded KAFOs were collected in Cell Recovery Solution for 45 min on ice in 15-ml tubes precoated with 1% BSA. Afterwards, organoids were equally divided into two tubes for the two conditions (EDTA⁺ and EDTA⁻), washed once with ADMEM+++ and centrifuged at 300g for 5 min at 4 °C. The supernatant was removed and, to disrupt the epithelial barrier integrity, organoids were incubated with or without 200 µl 4 mM EDTA in PBS on ice for 15 min. Organoids were centrifuged at 60g for 5 min and the EDTA solution was removed. Subsequently, 200 µl 500 µg ml⁻¹ 2–5 kDa inulin-FITC⁷⁰ (Sigma, F3272) resuspended in ADMEM+++ was added to both conditions. Organoids were then incubated for 60 min at 37 °C and directly imaged using an LSM 710 Zeiss confocal microscope. The proportion of organoids with intact barrier integrity (organoids without inulin signals inside the lumen) of each group was calculated for quantification analysis.

Ciliary beat frequency analysis

Organoids were seeded into eight-well glass-bottom slides (ibidi, 80806) and differentiated toward the lung proximal lineage for 14 d as described above. For CBF analysis, motile cilia grown inside organoids were observed using an inverted microscope system (Nikon Ti-U; Nikon NIS-Elements v.5.41.02 software) with a digital high-speed video camera (Prime BSI Express, Teledyne Photometrics). Videos were recorded at a rate of approximately 178 frames per second using a ×20 objective with ×1.5 magnifier. For each subject a minimum of five organoids were studied. The video of each organoid was divided into 16 small regions of interest (256 × 256) for analysis. The time taken for five full ciliary beats was recorded. Clearly visible cilia from every small region of interest were counted for five full beats and the number of high-speed video frames for five full beats was noted. CBF (Hz) was calculated as (178/(number of frames for five beats)) × 5.

TEM

Organoids in the matrix were fixed in a mix of 2.5% glutaraldehyde and 4% paraformaldehyde in 0.1 M Sorensen’s buffer, pH 7.3, and washed with 0.1 M Sorensen’s buffer. They were postfixed in 1% aqueous solution of osmium tetroxide, washed and dehydrated through an increasing series of ethanol solutions, followed by propylene oxide (Merck). Organoids were embedded in TAAB812 resin (TAAB Laboratory Equipment) and cut to approximately 70-nm thick sections using a Leica UC7 Ultramicrotome (Leica Microsystems). Sections were collected onto copper mesh grids and contrasted for 2 min with 4% uranyl acetate solution in methanol (VWR), followed by 2 min in lead citrate (Reynolds’ solution). Samples were viewed on a JEOL JEM-1400 TEM (JEOL) with an accelerating voltage of 120 kV. Digital images were collected with a Xarosa digital camera using Radius software (both from EMSIS).

Flow cytometry

Cold-stored AF was processed as described above for viable cell sorting. Viable cells were resuspended in FACS blocking buffer (FBB) and incubated for 30 min at 4 °C with the following fluorochrome-conjugated antibodies: APC/Fire 750 anti-human CD324 (E-cadherin) (BioLegend, 324122, 5 µl per tube), APC/Fire 750 anti-human CD326 (EpCAM) (BioLegend, 324233, 5 µl per tube). Cells were washed in 10 ml FBB and analyzed using a BD FACSymphony A5 (BD FACSDiva v.8.0.1 software). Data were processed on FlowJo (v.10.15).

RNA isolation and RT–qPCR

Organoids were collected from Matrigel with Cell Recovery Solution for 45 min on ice. Cells were then washed in ice-cold PBS to remove leftover Matrigel. Organoids were centrifuged at 300g for 5 min at 4 °C and the supernatant was discarded. Pellet was resuspended and lysed with RLT buffer (QIAGEN). Total RNA was isolated with an RNeasy Micro or Mini kit (QIAGEN) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop (Thermo). Complementary DNA was prepared using High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, 4368813). Quantitative real-time PCR detection was performed using PowerUp SYBR Green Master Mix (Applied Biosystems, A25742) and StepOnePlus Real-Time PCR System (Applied Biosystems). Assays for each sample were run in triplicate and were normalized to the housekeeping gene β-actin. Primer sequences are listed in Supplementary Table 5.

Bulk RNA sequencing

RNA was extracted as described above and stored at −80 °C until processing. NEBNext Low-Input RNA library preparation and sequencing were performed by the UCL genomics facility. Single-end bulk RNA sequencing was conducted on an Illumina NextSeq 2000. Then, 100 cycles were run to achieve an average of 5 million reads per sample.

Transcriptome bioinformatics analysis

Quality control was conducted on FASTQ raw sequences using v.0.11.9 FastQC (https://github.com/s-andrews/FastQC). Then, TrimGalore! v.0.6.6 (https://github.com/FelixKrueger/TrimGalore) was used to trim low-quality reads (quality 20 and length 70). STAR v.2.7.1a (https://github.com/alexdobin/STAR) was applied to align FASTQ sequences to the National Center for Biotechnology Information (NCBI) human reference genome GRCh38.p13 (www.ncbi.nlm.nih.gov/assembly/GCF_000001405.39/). featureCounts v.1.6.3 (https://doi.org/10.1093/bioinformatics/btt656) quantified the expression of individual genes to generate the raw count matrix, using the GRCh38.104 gene annotation (www.ensembl.org/Homo_sapiens/Info/Index). Default parameters were used for both alignment and quantification. The generated count matrix was further processed with a custom R script. Genes with fewer than ten reads across three samples were removed. Gene IDs were included using the added BioMart package (www.ensembl.org/info/data/biomart/biomart_r_package.html). CPM normalization was completed with the edgeR package (www.bioconductor.org/packages/release/bioc/html/edgeR.html).

ComBat_seq batch correction (rdrr.io/bioc/sva/man/ComBat_seq.html) was applied between the five batches. ggplot2 was used for graph generation, including PCA and dot plots generated from the normalized CPM matrix. Clustered heat maps were generated with pheatmap (cran.r-project.org/web/packages/pheatmap/) to enable hierarchical comparisons.

For the comparison analyses, the pheatmap hierarchical clustering DESeq2 (bioconductor.org/packages/release/bioc/html/DESeq2.html) with standard parameters was used to determine DEGs. A volcano plot was generated with ggplot2 to highlight statistically significant DEGs. Metascape Gene Annotation and Analysis Resource v.3.5 (metascape.org) was used to determine Gene Ontology pathway activation using the DEGs identified previously. A log fold change cutoff of 1 or 2 was applied as stated. An adjusted P value of 0.05 was used as the significance threshold.

scRNA-seq of AF cells

Viable AF cells were isolated using FACS as described above. To preserve AF cell heterogeneity, we did not gate by forward or side scatter. Hoechst was used to identify nucleated cells and exclude debris; PI was used to identify dead and damaged cells. Viable cells were then immediately processed for cDNA library preparation. Library generation was conducted following the 10x Genomics Chromium Next GEM Single Cell 3ʹ Reagent kits v.3.1 (Dual Index). Libraries were sequenced using NovaSeq 6000. Data processing was conducted with v.6.0.1 CellRanger. Ambient RNA correction was carried out with CellBender v.0.2.2 on standard parameters (Broad Institute; github.com/broadinstitute/CellBender). Analysis was performed with Seurat v.4.1.1 within a custom R script that was used for further downstream processing. Cells with fewer than 150 features were removed to prevent doublets or cells of low quality. There were 38,880 total cells and 33,934 cells after filtering. A single small cluster with high mitochondrial percentage genes was removed. No cell cycle correction was carried out. Normalization was then carried out using the NormalizeData function, with a logNormalize method and a scale factor of 10,000. Batch correction was completed through Seurat’s IntegrateData function after assessing for integration anchors based on the 2,000 most variable features. The object was scaled using ScaleData, RunPCA and FindNeighbors determined for 20 principal components. UMAPs and violin plots were generated using ggplot2 and normalized gene expression was always shown, with violin plots showing averaged normalized gene expression within the identified epithelial cluster. SingleR³⁴ v.1.6.1 was used to label the epithelial cluster. A single-cell experiment (github.com/drisso/SingleCellExperiment) object was analyzed using Human Primary Cell Atlas Data (www.humancellatlas.org), accessed via celldex (github.com/LTLA/celldex). Determining the tissue of origin for the AFEpCs was conducted through scGSEA, using escape package v.1.12.0 (bioconductor.org/packages/escape) with the enrichIt function on standard parameters and the C8 cell type signature gene set from the Broad Institute as a reference (gsea-msigdb.org/gsea/msigdb). The Seurat function ModuleScore was then used to score these tissue cells for a range of progenitor markers (‘Results’).

Single-cell RNA sequencing of organoids

Organoids were expanded or differentiated according to the above protocols. After washing with PBS, organoids were collected in Cell Recovery Solution for 45 min at 4 ºC for Matrigel removal and disaggregated to single cells by incubation with TrypLE for 7–10 min at 37 ºC. Cells were washed with FBB containing 1% FBS and 0.5 mM EDTA in PBS and centrifuged at 300g for 5 min at 4 ºC. Cells were resuspended in FBB and cell viability was confirmed using a Live/Dead Acridine Orange/PI fluorescent Luna cell counter. Cells were prepared at 1,000 cells per µl and cDNA library generation was completed following the 10x Genomics Chromium Next GEM Single Cell 3ʹ kit v.3.1 (Dual Index). Sequencing was completed by the UCL genomics facility. Data were pre-processed using v.6.0.1 CellRanger scRNA-seq. Upon formation of the count matrices, analysis was continued within a custom R script using Seurat v.4.1.1. A total of 41 organoids were sequenced across six lanes, meaning that deconvolution was required. No batch correction or cell-cycle correction was carried out. Deconvolution of the data was achieved in a two-step process. Initially, single-nucleotide polymorphism profiles were generated using CellSNP-lite v.1.2.0 from the same AFO bulk RNA-seq data for organoids from the same samples as sequenced through scRNA-seq. The NCBI Human single-nucleotide polymorphism dataset was used as a reference (https://ftp.ncbi.nlm.nih.gov/snp/organisms/human_9606/VCF/00-common_all.vcf.gz). The scRNA-seq data were then deconvoluted using Vireo v.0.5.6 on a cell-by-cell basis. The total number of cells was 36,463. All cells not confidently assigned (as determined by Vireo) were excluded (n = 3,805, 9.8%). All barcodes identified as doublets by Vireo were removed (n = 5,356, 13.8%). Differentiated/mature organoids were sequenced in different lanes from the undifferentiated to enable separation. Cell barcode patient separations are provided as supplementary data on the Gene Expression Omnibus. A 30% mitochondrial gene cutoff was applied to the SiAFOs but this was deemed unneeded for KAFOs or LAFOs. SingleR v.1.61 was used to label epithelial cells as discussed previously. Labeling of the different epithelial cell types present was conducted differently for each tissue. For the SiAFO, CellTypist (www.celltypist.org) was used with Cells_Intestinal_Tract as input⁷¹. This was performed in conjunction with manual marker analysis to label cluster by cluster. For LAFOs, CellTypist was also used with three input references: Cells_Fetal_Lung⁴¹, Cells_Lung_Airway⁷² and Human_Lung_Atlas⁷³, in conjunction with marker analysis. KAFOs were labeled with the automatic kidney-labeling tool DevKidCC (github.com/KidneyRegeneration/DevKidCC)⁴⁴, combined with marker analysis.

Statistics and reproducibility

Statistical analysis was conducted on data from at least three independent experimental or biological replicates wherever possible, as stated in the figure legends. Results are expressed as mean ± s.d. or s.e.m., as the median and quartiles (25% and 75% percentiles) or 95% CI range. Statistical significance was analyzed using unpaired or paired t-tests for comparisons between two different experimental groups. Statistical significance was assessed using one-way or two-way ANOVA with Dunnett’s, Holm–Šídák or Tukey’s multiple-comparisons test for analysis among more than two groups. *P < 0.03, **P < 0.002, ***P < 0.0002, ****P < 0.0001 were considered significant. Exact P values are stated in each figure legend where appropriate. Statistical analysis was carried out using R and GraphPad Prism v.10 software.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Raw individual-level data and combined processed data of the bulk RNA sequencing (AFOs, TFOs and fetal tissue-derived organoids) and scRNA-seq (AF and AFOs) have been uploaded to the NCBI Gene Expression Omnibus (GSE220994). These data are openly available with no restriction or time limit. Questions or additional requests can be directed to the corresponding authors. Source data are provided with this paper.

Code availability

Code is available on Zenodo at 10.5281/zenodo.8124205. This includes R scripts used for annotation, downstream analysis and visualization. This code is openly available with no restriction or time limit. Direct questions and additional requests can be directed to the corresponding authors.

References

Davidson, J. R. et al. Fetal body MRI and its application to fetal and neonatal treatment: an illustrative review. Lancet Child Adolesc. Health 5, 447–458 (2021).
Article PubMed PubMed Central Google Scholar
Lord, J. et al. Prenatal exome sequencing analysis in fetal structural anomalies detected by ultrasonography (PAGE): a cohort study. Lancet 393, 747–757 (2019).
Article CAS PubMed PubMed Central Google Scholar
Deprest, J. A. et al. Randomized trial of fetal surgery for moderate left diaphragmatic hernia. N. Engl. J. Med. 385, 119–129 (2021).
Deprest, J. A. et al. Randomized trial of fetal surgery for severe left diaphragmatic hernia. N. Engl. J. Med. 385, 107–118 (2021).
Senat, M.-V. et al. Endoscopic laser surgery versus serial amnioreduction for severe twin-to-twin transfusion syndrome. N. Engl. J. Med. 351, 136–144 (2004).
Adzick, N. S. et al. A randomized trial of prenatal versus postnatal repair of myelomeningocele. N. Engl. J. Med. 364, 993–1004 (2011).
Article CAS PubMed PubMed Central Google Scholar
Morris, R. K. et al. Percutaneous vesicoamniotic shunting versus conservative management for fetal lower urinary tract obstruction (PLUTO): a randomised trial. Lancet 382, 1496–1506 (2013).
Article PubMed PubMed Central Google Scholar
Takasato, M. et al. Directing human embryonic stem cell differentiation towards a renal lineage generates a self-organizing kidney. Nat. Cell Biol. 16, 118–126 (2013).
Article PubMed Google Scholar
Michielin, F. et al. The microfluidic environment reveals a hidden role of self-organizing extracellular matrix in hepatic commitment and organoid formation of hiPSCs. Cell Rep. 33, 108453 (2020).
Article CAS PubMed PubMed Central Google Scholar
Kunisaki, S. M. et al. Human induced pluripotent stem cell-derived lung organoids in an ex vivo model of the congenital diaphragmatic hernia fetal lung. Stem Cells Transl. Med. 10, 98–114 (2021).
Article CAS PubMed Google Scholar
Sachs, N. et al. Long-term expanding human airway organoids for disease modeling. EMBO J. 38, e100300 (2019).
Article PubMed PubMed Central Google Scholar
Cindrova-Davies, T. et al. Menstrual flow as a non-invasive source of endometrial organoids. Commun. Biol. 4, 651 (2021).
Article CAS PubMed PubMed Central Google Scholar
Lõhmussaar, K. et al. Patient-derived organoids model cervical tissue dynamics and viral oncogenesis in cervical cancer. Cell Stem Cell 28, 1380–1396 (2021).
Article PubMed Google Scholar
Schutgens, F. et al. Tubuloids derived from human adult kidney and urine for personalized disease modeling. Nat. Biotechnol. 37, 303–313 (2019).
Article CAS PubMed Google Scholar
Giobbe, G. G. et al. SARS-CoV-2 infection and replication in human gastric organoids. Nat. Commun. 12, 1–14 (2021).
Article Google Scholar
Cao, K. X., Booth, A., Ourselin, S., David, A. L. & Ashcroft, R. The legal frameworks that govern fetal surgery in the United Kingdom, European Union, and the United States. Prenat. Diagn. 38, 475–481 (2018).
Article PubMed PubMed Central Google Scholar
Vergallo, G. M. Freedom of scientific research and embryo protection under Italian and European Court of Human Rights’ jurisprudence. Brief European legislation overview. Eur. J. Health Law 28, 3–25 (2021).
Article Google Scholar
Regulation of stem cell research in Europe. EuroStemCell https://www.eurostemcell.org/regulation-stem-cell-research-europe
Servick, K. Biden administration scraps human fetal tissue research restrictions. Science https://doi.org/10.1126/SCIENCE.ABJ0479 (2021).
McCune, J. M. & Weissman, I. L. The ban on US government funding research using human fetal tissues: how does this fit with the NIH mission to advance medical science for the benefit of the citizenry? Stem Cell Rep. 13, 777–786 (2019).
Article Google Scholar
de Coppi, P. et al. Regenerative medicine: prenatal approaches. Lancet Child Adolesc. Health 6, 643–653 (2022).
Article PubMed PubMed Central Google Scholar
Calà, G., Sina, B., De Coppi, P., Giobbe, G. G. & Gerli, M. F. M. Primary human organoids models: current progress and key milestones. Front. Bioeng. Biotechnol. 11, 320 (2023).
Article Google Scholar
Gerrelli, D., Lisgo, S., Copp, A. J. & Lindsay, S. Enabling research with human embryonic and fetal tissue resources. Development 142, 3073–3076 (2015).
Article CAS PubMed Google Scholar
Shulman, L. P. & Elias, S. Amniocentesis and chorionic villus sampling: green-top guideline No. 8 July 2021. BJOG 129, 260–268 (2022).
Google Scholar
Dickinson, J. E. et al. Amnioreduction in the management of polyhydramnios complicating singleton pregnancies. Am. J. Obstet. Gynecol. 211, 434 (2014).
Article Google Scholar
Erfani, H. et al. Amnioreduction in cases of polyhydramnios: indications and outcomes in singleton pregnancies without fetal interventions. Eur. J. Obstet. Gynecol. Reprod. Biol. 241, 126–128 (2019).
Article PubMed Google Scholar
Senat, M.-V. et al. Endoscopic laser surgery versus serial amnioreduction for severe twin-to-twin transfusion syndrome. N. Engl. J. Med. 351, 136–144 (2004).
Article CAS PubMed Google Scholar
Underwood, M. A., Gilbert, W. M. & Sherman, M. P. Amniotic fluid: not just fetal urine anymore. J. Perinatol. 25, 341–348 (2005).
Article PubMed Google Scholar
Beall, M. H., van den Wijngaard, J. P. H. M., van Gemert, M. J. C. & Ross, M. G. Amniotic fluid water dynamics. Placenta 28, 816–823 (2007).
Article CAS PubMed Google Scholar
de Coppi, P. et al. Isolation of amniotic stem cell lines with potential for therapy. Nat. Biotechnol. 25, 100–106 (2007).
Article PubMed Google Scholar
Steigman, S. A. & Fauza, D. O. Isolation of mesenchymal stem cells from amniotic fluid and placenta. Curr. Protoc. Stem Cell Biol. Chapter 1, 1E2 (2007).
Ditadi, A. et al. Human and murine amniotic fluid c-Kit⁺Lin⁻ cells display hematopoietic activity. Blood 113, 3953–3960 (2009).
Article CAS PubMed Google Scholar
Wagner, R. et al. A tracheal aspirate-derived airway basal cell model reveals a proinflammatory epithelial defect in congenital diaphragmatic hernia. Am. J. Respir. Crit. Care Med. 207, 1214–1226 (2023).
Article CAS PubMed PubMed Central Google Scholar
Aran, D. et al. Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage. Nat. Immunol. 20, 163–172 (2019).
Article CAS PubMed PubMed Central Google Scholar
Zani, A. et al. Congenital diaphragmatic hernia. Nat. Rev. Dis. Prim. 8, 1–20 (2022).
Google Scholar
Nikolić, M. Z. et al. Human embryonic lung epithelial tips are multipotent progenitors that can be expanded in vitro as long-term self-renewing organoids. eLife 6, e26575 (2017).
Article PubMed PubMed Central Google Scholar
Ijsselstijn, H., Hung, N., De Jongste, J. C., Tibboel, D. & Cutz, E. Calcitonin gene-related peptide expression is altered in pulmonary neuroendocrine cells in developing lungs of rats with congenital diaphragmatic hernia. Am. J. Respir. Cell Mol. Biol. 19, 278–285 (1998).
Ijsselstijn, H., Gaillard, J. L. J., De Jongste, J. C., Tibboel, D. & Cutz, E. Abnormal expression of pulmonary bombesin-like peptide immunostaining cells in infants with congenital diaphragmatic hernia. Pediatr. Res. 42, 715–720 (1997).
Article CAS PubMed Google Scholar
Jensen, K. B. & Little, M. H. Organoids are not organs: sources of variation and misinformation in organoid biology. Stem Cell Rep. 18, 1255–1270 (2023).
Article Google Scholar
Hoehn, H. & Salk, D. Morphological and biochemical heterogeneity of amniotic fluid cells in culture. Methods Cell Biol. 26, 11–34 (1982).
Article CAS PubMed Google Scholar
He, P. et al. A human fetal lung cell atlas uncovers proximal-distal gradients of differentiation and key regulators of epithelial fates. Cell 185, 4841–4860 (2022).
Article CAS PubMed Google Scholar
Stewart, B. J. et al. Spatiotemporal immune zonation of the human kidney. Science 365, 1461–1466 (2019).
Article CAS PubMed PubMed Central Google Scholar
Hochane, M. et al. Single-cell transcriptomics reveals gene expression dynamics of human fetal kidney development. PLoS Biol. 17, e3000152 (2019).
Article PubMed PubMed Central Google Scholar
Wilson, S. B. et al. DevKidCC allows for robust classification and direct comparisons of kidney organoid datasets. Genome Med. 14, 1–25 (2022).
Article Google Scholar
Cao, J. et al. A human cell atlas of fetal gene expression. Science 370, eaba7721 (2020).
Article CAS PubMed PubMed Central Google Scholar
Miki, T., Lehmann, T., Cai, H., Stolz, D. B. & Strom, S. C. Stem cell characteristics of amniotic epithelial cells. Stem Cells 23, 1549–1559 (2005).
Article CAS PubMed Google Scholar
Parolini, O. et al. Concise review: isolation and characterization of cells from human term placenta: outcome of the first international workshop on placenta derived stem cells. Stem Cells 26, 300–311 (2008).
Article PubMed Google Scholar
Nievelstein, R. A. J., van der Werff, J. F. A., Verbeek, F. J., Valk, J. & Vermeij-Keers, C. Normal and abnormal embryonic development of the anorectum in human embryos. Teratology 57, 70–78 (1998).
Article CAS PubMed Google Scholar
Chitayat, D., Marion, R. W., Squillante, L., Kalousek, D. K. & Das, K. M. Detection and enumeration of colonic mucosal cells in amniotic fluid using a colon epithelial-specific monoclonal antibody. Prenat. Diagn. 10, 725–732 (1990).
Article CAS PubMed Google Scholar
Meran, L. et al. Engineering transplantable jejunal mucosal grafts using patient-derived organoids from children with intestinal failure. Nat. Med. 26, 1593–1601 (2020).
Article CAS PubMed Google Scholar
Muniraman, H. Fetal lung fluid: not the same as amniotic fluid. Pediatr. Pulmonol. 56, 1270 (2021).
Article PubMed Google Scholar
Glibert, W. M., Moore, T. R. & Brace, R. A. Amniotic fluid volume dynamics. Fetal Matern. Med. Rev. 3, 89–104 (1991).
Article Google Scholar
Davidson, J. et al. Motion corrected fetal body magnetic resonance imaging provides reliable 3D lung volumes in normal and abnormal fetuses. Prenat. Diagn. 42, 628–635 (2022).
Article PubMed PubMed Central Google Scholar
Wagner, R. et al. Proteomic profiling of hypoplastic lungs suggests an underlying inflammatory response in the pathogenesis of abnormal lung development in congenital diaphragmatic hernia. Ann. Surg. 278, e411–e421 (2023).
Pereira-Terra, P. et al. Unique tracheal fluid microRNA signature predicts response to FETO in patients with congenital diaphragmatic hernia. Ann. Surg. 262, 1130–1140 (2015).
Article PubMed Google Scholar
Davey, M. Surfactant levels in congenital diaphragmatic hernia. PLoS Med. 4, e243 (2007).
Article PubMed PubMed Central Google Scholar
Donahoe, P. K., Longoni, M. & High, F. A. Polygenic causes of congenital diaphragmatic hernia produce common lung pathologies. Am. J. Pathol. 186, 2532–2543 (2016).
Article PubMed PubMed Central Google Scholar
Evrard, V. A. et al. Intrauterine tracheal obstruction, a new treatment for congenital diaphragmatic hernia, decreases amniotic fluid sodium and chloride concentrations in the fetal lamb. Ann. Surg. 226, 753–758 (1997).
Article CAS PubMed PubMed Central Google Scholar
Loukogeorgakis, S. P. & de Coppi, P. Concise review: amniotic fluid stem cells: the known, the unknown, and potential regenerative medicine applications. Stem Cells 35, 1663–1673 (2017).
Article PubMed Google Scholar
Ginsberg, M. et al. Efficient direct reprogramming of mature amniotic cells into endothelial cells by ETS factors and TGFβ suppression. Cell 151, 559–575 (2012).
Article CAS PubMed PubMed Central Google Scholar
Giobbe, G. G. et al. Extracellular matrix hydrogel derived from decellularized tissues enables endodermal organoid culture. Nat. Commun. 10, 1–14 (2019).
Article Google Scholar
Uchimura, K., Wu, H., Yoshimura, Y. & Humphreys, B. D. Human pluripotent stem cell-derived kidney organoids with improved collecting duct maturation and injury modeling. Cell Rep. 33, 108514 (2020).
Article CAS PubMed PubMed Central Google Scholar
Jacob, A. et al. Differentiation of human pluripotent stem cells into functional lung alveolar epithelial cells. Cell Stem Cell 21, 472–4880 (2017).
Article CAS PubMed PubMed Central Google Scholar
Sachs, N., Tsukamoto, Y., Kujala, P., Peters, P. J. & Clevers, H. Intestinal epithelial organoids fuse to form self-organizing tubes in floating collagen gels. Development 144, 1107–1112 (2017).
Article CAS PubMed Google Scholar
Jafree, D. J., Long, D. A., Scambler, P. J. & Moulding, D. Tissue clearing and deep imaging of the kidney using confocal and two-photon microscopy. Methods Mol. Biol. 2067, 103–126 (2020).
Article CAS PubMed Google Scholar
Paganin, D., Mayo, S. C., Gureyev, T. E., Miller, P. R. & Wilkins, S. W. Simultaneous phase and amplitude extraction from a single defocused image of a homogeneous object. J. Microsc. 206, 33–40 (2002).
Article CAS PubMed Google Scholar
Atwood, R. C., Bodey, A. J., Price, S. W. T., Basham, M. & Drakopoulos, M. A high-throughput system for high-quality tomographic reconstruction of large datasets at Diamond Light Source. Philos. Trans. R. Soc. A: Math. Phys. Eng. Sci. 373, 20140398 (2015).
Article Google Scholar
Savvidis, S. et al. Monitoring tissue engineered constructs and protocols with laboratory-based X-ray phase contrast tomography. Acta Biomater. 141, 290–299 (2022).
Article CAS PubMed Google Scholar
Hu, Y., Limaye, A. & Lu, J. Three-dimensional segmentation of computed tomography data using Drishti Paint: new tools and developments. R. Soc. Open Sci. 7, 201033 (2020).
Article PubMed PubMed Central Google Scholar
Yousef Yengej, F. A. et al. Tubuloid culture enables long-term expansion of functional human kidney tubule epithelium from iPSC-derived organoids. Proc. Natl Acad. Sci. USA 120, e2216836120 (2023).
Article PubMed PubMed Central Google Scholar
Elmentaite, R. et al. Cells of the human intestinal tract mapped across space and time. Nature 597, 250–255 (2021).
Article CAS PubMed PubMed Central Google Scholar
Madissoon, E. et al. A spatially resolved atlas of the human lung characterizes a gland-associated immune niche. Nat. Genet. 55, 66–77 (2022).
Article PubMed PubMed Central Google Scholar
Sikkema, L. et al. An integrated cell atlas of the lung in health and disease. Nat. Med. 29, 1563–1577 (2023).

Download references

Acknowledgements

Research at Great Ormond Street Hospital NHS Foundation Trust and UCL Great Ormond Street Institute of Child Health is made possible by the NIHR Great Ormond Street Hospital (GOSH) Biomedical Research Centre (BRC). M.F.M.G. held an H2020 Marie Skłodowska-Curie Fellowship (843265, AmnioticID) and received support from the Wellcome Institutional Strategic Support Fund, UCL Global Engagement Fund, UCL Therapeutic Acceleration Support (TAS Call 10) and the NIHR GOSH BRC. G.C. holds a UCL Division of Surgery and Interventional Science PhD scholarship. M.A.B. holds a GOSH Child Health Research Charitable Incorporated Organization (GOSH-CC) PhD Studentship. L.T. holds an NIHR GOSH BRC Crick Clinical Research Training Fellowship. K.Y.S. holds a PhD Scholarship from Kidney Research UK (Paed_ST_005_220221129). J.R.D. holds an MRC Clinical Research Training Fellowship (MR/V006797/1). S.S. is a UKRI, EPSRC Doctoral Prize Fellow (EP/T517793/1). A.O. is supported by the Royal Academy of Engineering under the Chairs in Emerging Technologies scheme (CiET1819/2/78). The work of D.C. is supported by the Fondazione Telethon Core Grant, Armenise-Harvard Foundation Career Development Award, European Research Council (759154, CellKarma) and the Italian Ministry of Health (Piano Operativo Salute Traiettoria 3, ‘Genomed’). J.D. is funded by the GOSH-CC. F.M.R. is supported by a grant from My FetUZ at KU Leuven. B.C.J. is supported by the General Sir John Monash Foundation as a Chairman’s Circle John Monash Scholar. A.L.D. is supported by the UCLH BRC. V.S.W.L. is funded by the Francis Crick Institute, which receives its core funding from Cancer Research UK (CC2141), the UK MRC (CC2141) and the Wellcome Trust (CC2141). G.G.G. is supported by the UCL Therapeutic Acceleration Support LifeArc Fund Rare Diseases Call (TRO award no. 184646). G.G.G. and S.P.L. are supported by the NIHR GOSH BRC. F.M. and P.D.C. are supported by CDH-UK (16ICH03) and the BREATH Consortium (Longfonds Voorhen Astma Fonds 552269). P.D.C. is also supported by the NIHR (NIHR-RP-2014-04-046), H2020 (668294, INTENS), OAK Foundation (W1095/OCAY-14-191), GOSH-CC (V5201), the Wellcome/EPSRC Centre for Interventional and Surgical Sciences (WEISS), University College London and NIHR GOSH BRC. We acknowledge the provision of beamtime MG31437 on ID13-1 at the Diamond Light Source. Human fetal material used for generating our control organoids was provided by the Joint MRC/Wellcome (MR/R006237/1) Human Developmental Biology Resource (www.hdbr.org). We are grateful to A. Pellegata, M. Pellegrini, V. Karaluka, B. Crespo, A. Baulies, O. Xintarakou, D. Baidya, G. Cenedese, R. Hynds, D. Jafree, D. Long, S. Mantero, O.J. Arthurs, N. Elvassore, C. Pou Casellas, H. Clevers and M. Haniffa for useful discussion and advice. We are grateful to A. Eddaoudi and staff at the GOSICH Flow Cytometry facility, D. Moulding and staff at the GOSICH Imaging facility, T. Brooks, P. Niola, F. Crossby and UCL Genomics. We acknowledge the UCL Myriad High Performance Computing Facility (Myriad@UCL) and associated support services. We acknowledge Servier Medical Art by Servier, from which we used some images licensed under a Creative Commons Attribution 3.0 Unsupported License. We thank staff at the University of Leicester Core Biotechnology Services Electron Microscopy Facility for help with material processing. We thank the UCLH FMU research midwife team, R. Devlieger, L. De Catte, L. Lewi, S. Mastrodima-Polychroniou and G. Attilakos who helped to collect the fluid samples. We thank the UCLH Charity and GOSH-CC for supporting the CDH FETO service. Finally, we are grateful to the Human Developmental Cell Atlas Network, whose members provided valuable advice for the development of the project. This publication is part of the Human Cell Atlas (www.humancellatlas.org/publications).

Author information

These authors contributed equally: Mattia Francesco Maria Gerli, Giuseppe Calà.

Authors and Affiliations

Department of Surgical Biotechnology, Division of Surgery and Interventional Science, University College London, London, UK
Mattia Francesco Maria Gerli, Giuseppe Calà & Kylin Yunyan Sun
Great Ormond Street Institute of Child Health, University College London, London, UK
Mattia Francesco Maria Gerli, Giuseppe Calà, Max Arran Beesley, Beatrice Sina, Lucinda Tullie, Kylin Yunyan Sun, Federica Michielin, Joseph R. Davidson, Brendan C. Jones, Dani Do Hyang Lee, Theodoros Xenakis, Ian C. Simcock, Christopher O’Callaghan, Simon Eaton, Stavros P. Loukogeorgakis, Giovanni Giuseppe Giobbe & Paolo De Coppi
Politecnico di Milano, Milan, Italy
Beatrice Sina
Stem Cell and Cancer Biology Laboratory, The Francis Crick Institute, London, UK
Lucinda Tullie & Vivian S. W. Li
Armenise/Harvard Laboratory of Integrative Genomics, Telethon Institute of Genetics and Medicine, Pozzuoli, Italy
Francesco Panariello & Davide Cacchiarelli
Elizabeth Garrett Anderson Institute for Women’s Health, University College London, London, UK
Joseph R. Davidson, Anna L. David & Jan Deprest
Department of Development and Regeneration, Woman and Child and UZ Leuven Clinical Department of Obstetrics and Gynaecology, KU Leuven, Leuven, Belgium
Francesca Maria Russo, Anna L. David, Jan Deprest & Paolo De Coppi
Department of Medical Physics and Biomedical Engineering, University College London, London, UK
Savvas Savvidis & Alessandro Olivo
Department of Radiology, Great Ormond Street Hospital, London, UK
Ian C. Simcock
Department of Respiratory Sciences, University of Leicester, Leicester, UK
Anna A. Straatman-Iwanowska & Robert A. Hirst
Specialist Neonatal and Paediatric Surgery, Great Ormond Street Hospital for Children NHS Foundation Trust, London, UK
Stavros P. Loukogeorgakis & Paolo De Coppi
Department of Translational Medicine, University of Naples Federico II, Naples, Italy
Davide Cacchiarelli
Genomics and Experimental Medicine Program, Scuola Superiore Meridionale, Naples, Italy
Davide Cacchiarelli
Medical and Surgical Department of the Fetus, Newborn and Infant, Ospedale Pediatrico Bambino Gesù, IRCCS, Rome, Italy
Paolo De Coppi
NIHR Great Ormond Street Hospital Biomedical Research Centre, London, UK
Paolo De Coppi

Authors

Mattia Francesco Maria Gerli
View author publications
You can also search for this author in PubMed Google Scholar
Giuseppe Calà
View author publications
You can also search for this author in PubMed Google Scholar
Max Arran Beesley
View author publications
You can also search for this author in PubMed Google Scholar
Beatrice Sina
View author publications
You can also search for this author in PubMed Google Scholar
Lucinda Tullie
View author publications
You can also search for this author in PubMed Google Scholar
Kylin Yunyan Sun
View author publications
You can also search for this author in PubMed Google Scholar
Francesco Panariello
View author publications
You can also search for this author in PubMed Google Scholar
Federica Michielin
View author publications
You can also search for this author in PubMed Google Scholar
Joseph R. Davidson
View author publications
You can also search for this author in PubMed Google Scholar
Francesca Maria Russo
View author publications
You can also search for this author in PubMed Google Scholar
Brendan C. Jones
View author publications
You can also search for this author in PubMed Google Scholar
Dani Do Hyang Lee
View author publications
You can also search for this author in PubMed Google Scholar
Savvas Savvidis
View author publications
You can also search for this author in PubMed Google Scholar
Theodoros Xenakis
View author publications
You can also search for this author in PubMed Google Scholar
Ian C. Simcock
View author publications
You can also search for this author in PubMed Google Scholar
Anna A. Straatman-Iwanowska
View author publications
You can also search for this author in PubMed Google Scholar
Robert A. Hirst
View author publications
You can also search for this author in PubMed Google Scholar
Anna L. David
View author publications
You can also search for this author in PubMed Google Scholar
Christopher O’Callaghan
View author publications
You can also search for this author in PubMed Google Scholar
Alessandro Olivo
View author publications
You can also search for this author in PubMed Google Scholar
Simon Eaton
View author publications
You can also search for this author in PubMed Google Scholar
Stavros P. Loukogeorgakis
View author publications
You can also search for this author in PubMed Google Scholar
Davide Cacchiarelli
View author publications
You can also search for this author in PubMed Google Scholar
Jan Deprest
View author publications
You can also search for this author in PubMed Google Scholar
Vivian S. W. Li
View author publications
You can also search for this author in PubMed Google Scholar
Giovanni Giuseppe Giobbe
View author publications
You can also search for this author in PubMed Google Scholar
Paolo De Coppi
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

M.F.M.G., G.G.G. and P.D.C. conceived the study, outlined the experiments and isolated the first AFO lines. M.F.M.G. established the cell-sorting methodology. G.C. supported the experimental design and conducted the main organoid culture, characterization and functional experiments with the help of B.S., K.Y.S., B.C.J., M.F.M.G. and G.G.G. M.A.B. performed the transcriptomic analyses with the help and support from T.X., F.P. and D.C. L.T., S.E. and V.S.W.L. contributed to SiAFO characterization. F.M. contributed to LAFO data interpretation. J.R.D., S.E. and S.P.L. contributed to AF analyses, data interpretation and writing of the paper. S.S. and A.O. performed and interpreted X-ray PC-CT experiments. I.C.S. conducted microCT imaging analysis. A.A.S.-I. and R.A.H. generated and interpreted the TEM data. D.D.H.L. and C.O.C. conducted CBF analyses. A.L.D., F.M.R. and J.D. contributed to fetal fluid collection and data interpretation. M.F.M.G., G.C., S.E., G.G.G. and P.D.C. wrote the paper. All authors contributed to paper discussions and data interpretation.

Corresponding authors

Correspondence to Mattia Francesco Maria Gerli or Paolo De Coppi.

Ethics declarations

Competing interests

The views expressed are those of the authors and not necessarily those of the NHS, the NIHR or the Department of Health. The authors declare the following competing interests: a patent detailing the AFO technology was filed by UCL on 18 April 2023 entitled ‘Derivation of Primary Organoids from the Fetal Fluids’, under application no. GB2305703.7. M.F.M.G., G.G.G., P.D.C. and G.C. are listed as inventors on the patent. The other authors declare no competing interests.

Peer review

Peer review information

Nature Medicine thanks the anonymous reviewers for their contribution to the peer review of this work. Primary Handling Editor: Sonia Muliyil, in collaboration with the Nature Medicine team.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 AF single-cell RNA Sequencing.

(a) Schematic and phase-contrast image of fresh AF sample at collection (left). FACS plot showing heterogeneity of AF cells by forward and side scatter after sorting (right). (b) scRNA-seq UMAP analysis of n=12 AF samples; middle UMAP shows cell distribution across second and third trimester; right panel shows cells labeled by post-conception weeks (PCW). (c) UMAPs depicting expression of epithelial keratins within the AF epithelial cell cluster. (d) Upregulated GO pathways in the epithelial-labeled cluster of the scRNA-seq AF from DEGs calculated when compared to all other clusters (one-sided Fisher test, adjusted using Benjamini-Hochberg for multiple hypotheses). (e) Schematic showing the comparison between the gestational age weeks covered by reference fetal scRNA-seq atlases, highlighting the lack of late-stage development data.

Source data

Extended Data Fig. 2 Derivation and characterization of AFO.

(a) Schematic depicting the workflow for clonal AFO derivation. Image created and adapted in full licensed Biorender. (b) Phase-contrast images showing growth of clonal AFO from sorted AF epithelial cells cultured at a ratio of 5 cells in 3μL Matrigel droplet. (c) Percentage of samples that generated organoids at passage 0 (n=26/29 AF, n=16/20 CDH AF, n=7/17 TF samples, mean and 95% confidence interval [AF, 72.65, 97.81; CDH AF, 56.34, 94.27; TF, 18.44, 67.08]). (d) Phase-contrast images showing the recovery of AFO at passage 4 after cryopreservation and their expansion until passage 8 (Scale bar: 200 μm). (e) Phase-contrast images depicting two additional AFO lines from independent patients (scale bar: 200 μm). (f) PCA shows all sequenced AFO with the excluded unknown sample (top); PCA showing each organoid as its gestational age (bottom). (g) Top10 pathways from gene ontology analysis comparing each AFO identity to the other 2 AFO identities, (one-sided Fisher test, adjusted using Benjamini-Hochberg for multiple hypotheses). (h) Euclidean-clustered heatmap confirming the tissue-typing labeling of the AFO.

Source data

Extended Data Fig. 3 Characterization of SiAFO.

(a) Phase-contrast images showing SiAFOs at passage 6 and their expansion after cryopreservation until passage 10 (scale bar: 200 μm). (b) Whole-mount immunofluorescent staining showing the absence of NKX2-1 (lung) and PAX8 (kidney) in SiAFO (scale bar: 50 μm). (c) Whole-mount immunofluorescent staining for Chromogranin A (CHGA) and Mucin 2 (MUC2) on SiAFOs in expansion medium (scale bar: 50 μm). (d) RT–qPCR analysis of SiAFOs cultured in maturation medium without (Basal), with CHIR99021 (CHIR) or with DAPT (n=5 SiAFO clonal lines from n=2 patients; mean±SEM). (e) Stereoscopic image of an intestinal ring (scale bar: 1 mm) and microCT images showing the presence of budding (left) as well as luminal-like structures (L) (scale bars: 100 μm). (f) 3D z-stacks of SiAFO rings showing tight junctions (ZO-1), Paneth cells (LYZ), enterocytes (FABP1 and KRT20), intestinal secretory cells (MUC2) and enteroendocrine cells (CHGA); nuclei counterstained with Hoechst (scale bars:50μm).

Source data

Extended Data Fig. 4 Characterization of KAFO.

(a) Phase-contrast images showing the thawing of KAFOs after cryopreservation at passage 7 and their expansion until passage 10 (scale bar: 200 μm). (b) Full set of KAFOs subjected to bulk RNA-seq. (c) Double positive GATA3 and LTL cells confirm a mixed tubular phenotype; in the left panel a different phenotype of KAFOs was observed with GATA3 positive and LTL negative cells (scale bar: 50 μm). (d) Immunofluorescence showing presence of ZO-1-positive tight junctions in KAFOs (scale bar: 50 μm). (e) Representative example of inulin-permeability assay (scale bar: 50 μm). (f) Localization of RET protein signal in fetal kidney tissue slice and tissue-derived fetal kidney organoids (FKO; scale bar: 50 μm). (g) Distal and collecting duct markers are absent or poorly present in KAFOs in expansion medium; differentiation induces expression of CALB1, SLC12A1 and AQP2 (n=4 independent biological samples); nuclei counterstained with Hoechst (scale bar: 50 μm).

Extended Data Fig. 5 Characterization of LAFO.

(a) Phase-contrast images showing LAFOs after cryopreservation expanded to passage 21 (Scale bar: 200 μm). (b) Different LAFO lines’ morphological phenotypes (scale bars: 200 μm), quantified in (c) (n=10 AF samples; mean with 95% CI, ***P<0.001, one-way ANOVA with multiple comparisons). (d) Full set of sequenced LAFOs. (e) LAFOs show absence of surfactant protein C in expansion (Scale bar: 50 μm). (f) Airway markers (FOXJ1, ACATUB, MUC5AC, KRT5) are absent or poorly expressed at protein level in LAFOs cultured in expansion medium, but present in proximally differentiated LAFOs (n=5 independent biological samples, mean±SEM, ns=non-significant, two-tailed paired t-test) (Scale bar: 50 μm). (g) TEM showing normal cilia rootlets with associated mitochondria (green arrow); longitudinal section of the axonemes confirms a normal central microtubule pair (yellow arrow) and radial spokes (red arrow) (scale bars: 100 nm). (h) Distalised LAFOs produce surfactant protein B vs. control in expansion; nuclei counterstained with Hoechst (scale bars: 50 μm).

Source data

Extended Data Fig. 6 Derivation and characterization of CDH organoids.

(a) Comparison of organoid formation efficiency between non-CDH and CDH fluids at P0 (n=26 non-CDH AFs; n=16 CDH AFs; n=7 CDH TFs; median and quartiles). (b) Organoid formation efficiencies at various gestational age for CDH AF and TF. (c) PCA showing CDH organoids alongside all the AFO clusters. (d) Representative CDH organoid lines stained for SOX9 (scale bars: 50 μm). (e) Derivation of CDH AFO and TFO from different patients before (left) and after (right) FETO surgery (scale bar: 200 μm). (f) Heatmap showing comparison of lung surfactant genes expression between CDH organoids before and after FETO. (g) GO analysis of CDH LAFO and LTFO before and after FETO, when compared to GA-matched control LAFOs; red arrows highlight surfactant and matrix remodeling pathways. (h) Phase-contrast image of Proximal CDH LAFOs showing cilia and mucus/debris within the lumen. Immunofluorescence image showing Proximal CDH LAFOs expressing SOX2 (scale bars: 50 μm). (i) Ciliary beating frequency analysis of CDH organoids (n=49 organoids from 5 lines of 3 independent patients) vs CT LAFOs (n=40 organoids from 4 lines of 4 independent patients; **P=0.0014 unpaired t-test).

Source data

Extended Data Fig. 7 Dot plot showing expression of lung genes in CDH LAFO/LTFO.

Expression of a selection of lung genes associated with different lung cell types (log CPM). Relative expression within each gene is shown by color. All CDH organoids generated, LAFO and LTFO are shown, separated by before and after FETO. Proximal and Distal differentiated CDH organoids also shown, with colored bars used to pair differentiated organoid with the matching undifferentiated organoid line. They are shown alongside GA-matched LAFO controls and tissue-derived fetal organoid controls (not age-matched).

Extended Data Fig. 8 Additional CDH organoid gene expression analyses.

(a) Volcano plots showing significant DEGs (p<0.05, |LFC| > 2). Lung-relevant genes are highlighted. Comparison of organoids generated from AF or TF sampled after FETO with those before FETO. (b) Comparison of LAFO and LTFO from before and after FETO with GA-matched control LAFOs. (c) Comparison of CT LAFO between samples GA-matched to before and after FETO (d) Comparison of LAFO with LTFO, before and after FETO. (e) DEG gene list generated from comparing CDH LAFO to controls, before and after FETO.

Supplementary information

Supplementary Information

Supplementary Tables 1–5.

Reporting Summary

Supplementary Video 1

The 3D reconstruction of microCT imaging performed on a whole SiAFO ring.

Supplementary Video 2

Whole-mount 3D immunofluorescent staining and tissue clearing of SiAFO ring (nuclei in blue, KRT20 in red, ITGβ4 in green and CHGA in yellow).

Supplementary Video 3

High-speed video microscopy of beating cilia and whirling mucus and debris within the lumen of proximally differentiated LAFOs.

Supplementary Video 4

Phase-contrast video of beating cilia and whirling mucus and debris within the lumen of proximal CDH LAFOs.

Source data

Source Data Fig. 1

Flow cytometry data analysis.

Source Data Fig. 2

Statistical source data.

Source Data Fig. 3

Statistical source data.

Source Data Fig. 4

Statistical source data.

Source Data Fig. 5

Statistical source data.

Source Data Fig. 6

Statistical source data.

Source Data Extended Data Fig. 1

FACS data analysis.

Source Data Extended Data Fig. 2

Statistical source data.

Source Data Extended Data Fig. 3

Statistical source data.

Source Data Extended Data Fig. 5

Statistical source data.

Source Data Extended Data Fig. 6

Statistical source data.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Gerli, M.F.M., Calà, G., Beesley, M.A. et al. Single-cell guided prenatal derivation of primary fetal epithelial organoids from human amniotic and tracheal fluids. Nat Med 30, 875–887 (2024). https://doi.org/10.1038/s41591-024-02807-z

Download citation

Received: 02 November 2023
Accepted: 05 January 2024
Published: 04 March 2024
Issue Date: March 2024
DOI: https://doi.org/10.1038/s41591-024-02807-z

This article is cited by

Organoids grown from amniotic fluid could shed light on rare diseases
- Lilly Tozer
Nature (2024)

Subjects

Abstract

Similar content being viewed by others

Main

Results

Single-cell mapping the human AF to investigate presence of tissue-specific fetal epithelial progenitors

Generation of primary fetal epithelial human AFOs

Characterization and maturation of SiAFOs

Characterization and differentiation of KAFOs

Characterization and differentiation of LAFOs

Characterization of AF and TF organoids from CDH fetuses

Discussion

Methods

Ethics and informed consent

AF collection and isolation of the viable cell fraction

Derivation and culture of human AFOs

Derivation and culture of human fetal LTFOs

Passaging of organoids

Fetal tissue collection and generation of a control primary fetal organoids library

Human fetal tissue-derived small intestinal organoids

Human fetal tissue-derived kidney tubule organoids

Human fetal tissue-derived lung organoids

Human fetal tissue-derived stomach organoids

Human tissue-derived placental and fetal bladder organoids

Organoid cryopreservation and thawing

Evaluation of organoid formation efficiency and area

Organoid maturation/differentiation

Intestinal ring formation in collagen hydrogel

Whole-mount immunofluorescence

Image acquisition

X-ray PC-CT

microCT

SiAFO dipeptidyl peptidase IV and disaccharidase assays

KAFO potassium ion channel assay

KAFO epithelial barrier integrity assay

Ciliary beat frequency analysis

TEM

Flow cytometry

RNA isolation and RT–qPCR

Bulk RNA sequencing

Transcriptome bioinformatics analysis

scRNA-seq of AF cells

Single-cell RNA sequencing of organoids

Statistics and reproducibility

Reporting summary

Data availability

Code availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding authors

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Extended data

Supplementary information

Source data

Rights and permissions

About this article

Cite this article

Share this article

This article is cited by

Search

Quick links