RANK drives structured intestinal epithelial expansion during pregnancy

Mice

Rank conditional mice (Rank^flox) were generated in our laboratory and have been previously described¹¹. The following additional mouse strains were used: Rankl conditional mice (Rankl^flox)⁵¹, Traf6 conditional mice (Traf6^flox)³¹, constitutively active RANK mutant over-expressing mice (caRANK^LSL)²⁹, Rnf43 conditional mice (Rnf43^flox) and Znrf3 conditional mice (Znrf3^flox)²⁸. Vil^cre mice³³. Apc^min/+ mice³², Twist2^cre mice⁵², Cd4^cre mice⁴⁰, Rorgt^cre mice⁵³, tdTomato reporter mice⁵⁴ and Lgr5-eGFP-IRES-creERT2 mice⁵⁵ were purchased from the Jackson laboratories. All mouse lines were maintained on a C57BL/6J genetic background and housed under specific pathogen-free conditions. Mouse cages were individually ventilated and subjected to ambient temperature of 22 ± 1 °C under a 14 h–10 h light–dark cycle. Mouse genotypes were assessed by PCR. For all experiments, only littermate and sex-matched mice were used, unless otherwise specified. Control littermates of caRANK^vil-Tg mice were defined as Vil^cre mice, heterozygous caRANK^LSL mice or wild-type mice (negative for Vil^cre and negative for caRANK^LSL).

We did not observe any apparent differences among control littermates with different genotypes in any experiments. To exclude the potential effects of the Rank deletion in the intestine in timed pregnancy/lactation studies, Rank^WT and Rank^Δvil female littermates were crossed to wild-type syngeneic C57BL/6J male breeders, resulting in RANK-sufficient fetuses with a comparable genetic background. For the offspring analysis, we used the Rank^WT and Rank^Δvil female littermates who delivered more than five mice to avoid the effects of different offspring numbers. All mice were bred, maintained, examined and euthanized in accordance with institutional animal care guidelines and ethical animal license protocols approved by the legal authorities. All experimental animal projects performed at Institute of Molecular Biotechnology of the Austrian Academy of Sciences (IMBA), Vienna BioCenter were approved by the Federal Ministry of Education, Science and Research. Animal experiments using Germ-free mice at the University of British Columbia were approved by the University of British Columbia Animal Care Committee. Animal experiments using germ-free mice at Kiel University were approved prior to the study by the committee for animal welfare of the state of Schleswig-Holstein (V242-7224.121-33). Timed matings were performed in germ-free and specific-pathogen-free mice to achieve syngenic (both parents C57BL/6) and semiallogenic breedings (male BALB/C, female C57BL/6).

Mouse intestinal organoids

Mouse intestinal organoids were established as described previously²¹. In brief, freshly isolated intestinal crypts were mixed with 10 μl Matrigel (Corning, 356231) and placed on a warmed 24-well plate dish to let them polymerize. The crypts were then cultured with a ROCK inhibitor (Y-27632; Sigma-Aldrich, Y0503, 10 μM) and ENR (EGF/NOGGIN/R-spondin) medium composed of advanced Dulbecco’s modified Eagle’s medium/F12 (DMEM) supplemented with penicillin–streptomycin, 10 mM HEPES, GlutaMAX, N2 (Life Technologies), B27 (Life Technologies) and 1 mM N-acetylcysteine (Sigma-Aldrich), 50 ng ml⁻¹ mouse recombinant epidermal growth factor (EGF; Peprotech), R-spondin1 (conditioned medium from 293T-HA-RspoI-Fc cells, 10% final volume), and 100 ng ml⁻¹ NOGGIN (Peprotech). Passage was performed weekly at a 1:6 split ratio. To explore ERK/MAPK-dependent phenotypes, we cultured organoids using a low concentration of EGF (50 ng ml⁻¹ to 50 pg ml⁻¹) in ENR medium (E^lowNR medium).

To assess organoid survival (Fig. 1f), 25 irradiated organoids were dissociated into single cells by TrypLE (Thermo Fisher Scientific) and DNase I (Worthington Biochemical) treatments for 5 min at 37 °C and subsequent vigorous pipetting through a p200 pipette. The dissociated cells were mixed with 15 μl Matrigel and seeded into each well of a 48-well plate. The organoids were then further cultured in WENR medium. WENR medium was composed of WNT3A (conditioned medium (CM) from WNT3A L-cells, 50% final volume), 10 μM Y-27632 (ROCK inhibitor, StemCell Technologies) and 10 μM nicotinamide (Sigma-Aldrich), on the basic ENR medium. Bright-field images of organoids were taken using a Carl Zeiss Axiovert.A1 microscope. For the measurement of organoid size, organoid areas in horizontal cross sections were measured using Fiji software (ImageJ, v.2.3.0).

Human intestinal organoids

Patient recruitment and sample collection

Intestinal biopsy specimens were collected from the duodenum of children undergoing diagnostic endoscopy. This study was conducted with informed patient and/or caretaker consent as appropriate, and with full ethical approval by East of England – Cambridge South Research Ethics Committee (REC-12/EE/0482).

Human organoid cultures

Human intestinal organoids were generated from mucosal biopsy specimens by isolating intestinal crypts and culturing those in Matrigel (Corning) using medium described previously⁵⁶. The medium was replaced every 48–72 h. Once organoids were established, they were further cultured in an expansion medium composed of advanced DMEM/F12 supplemented with penicillin–streptomycin, 10 mM HEPES, GlutaMAX, N2 (Life Technologies), B27 (Life Technologies), 1 mM N-acetylcysteine (Sigma-Aldrich), R-spondin1 (conditioned medium from 293T-HA-RspoI-Fc cells, 10% final volume), 100 ng ml⁻¹ NOGGIN (Peprotech), 10 nM human gastrin I (Sigma-Aldrich), 500 nM A83-01 (Tocris), WNT3A (conditioned medium from WNT-producing L-cell line, 50% final volume), 50 ng ml⁻¹ mouse recombinant epidermal growth factor (EGF; Peprotech), 100 ng ml⁻¹ human insulin-like growth factor-1 (IGF-1; BioLegend), and 50 ng ml⁻¹ human recombinant fibroblast growth factor-basic (FGF-2; Peprotech)⁴³. To test the role of RANKL under suboptimal growth conditions, we used a growth-factor-reduced condition lacking EGF, IGF-1 and FGF-2 from the expansion medium. To assess organoid survival, organoids were irradiated and subsequently dissociated into single cells using TrypLE (Thermo Fisher Scientific) and DNase I (Worthington Biochemical) treatments for 5 min at 37 °C and subsequent vigorous pipetting using a p200 pipette. The dissociated cells were mixed with 15 μl Matrigel (Corning, 356231) and seeded into a 48-well plate. The organoids were then further cultured in expansion medium supplemented with the ROCK inhibitor Y-27632 (10 μM, StemCell Technologies). Human BMPR1A mutant organoids have been described previously⁴⁴.

Apc
^min/+ tumoroids

Small intestinal adenomas were collected from Apc^min/+ heterozygous mice. Tissues were incubated with Gentle Cell Dissociation Reagent (StemCell Technologies) for 15 min at room temperature and vortexed vigorously to remove non-transformed crypts surrounding the tumour. The remaining tissue was minced into 2–5 mm fragments, and further digested in TrypLE (Thermo Fisher Scientific) and DNase I (Worthington Biochemical) for 10 min at 37 °C. The supernatant was collected and centrifuged at 300g for 5 min, suspended in Matrigel and seeded into a 24-well plate. The seeded cells were cultured with advanced DMEM/F12 supplemented with penicillin–streptomycin, 10 mM HEPES, GlutaMAX, N2 (Life Technologies), B27 (Life Technologies) and 50 ng ml⁻¹ mouse recombinant epidermal growth factor (EGF; Peprotech). The medium was replaced every 2 days. Tumoroids were passaged every 5 days.

Ex vivo maintenance of mesenchymal cell of the lamina propria

The protocol was adapted from a previous study⁵⁷. In brief, half of the upper small intestinal tissue from nulliparous female mice was washed with cold PBS, Peyer’s patches were removed manually, and then the remaining specimens were incubated in 10 ml of gentle dissociation solution (HBSS with 10 mM EDTA and 1 mM DTT (Sigma-Aldrich)) on ice for 20 min. The tissues were shaken vigorously, and the supernatant was discarded. The remaining tissue fragments were cut into 2–5 mm fragments and seeded in DMEM/F12 supplemented with 10% fetal bovine serum (FBS) and 50 ng ml⁻¹ mouse recombinant epidermal growth factor (EGF; Peprotech). Once mesenchymal cells started outgrowth from the tissue fragment, attached cells were dissociated with trypsin, seeded in six-well dishes and subsequently grown to expand mesenchymal cells. Then, 3 days before stimulation with recombinant mouse prolactin (rmProlactin) (Peprotech), the culture medium was changed to DMEM with 10% charcoal-stripped FBS and 50 ng ml⁻¹ of EGF.

MTT assay

Organoid growth was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT) assay. Organoids were incubated with MTT (0.5 mg ml⁻¹, final concentration; Sigma-Aldrich) for 4 h at 37 °C and the cells containing formazan were subsequently solubilized with 10% SDS in 0.01 M HCl. The absorbance of the formazan product was measured at 550 nm using BioTek Synergy 2. Absorbance at 720 nm was subtracted from sample values measured at 550 nm. Furthermore, the absorbance values of wells containing Matrigel and medium, but not organoids, were subtracted as background controls.

Flow cytometry

Intestinal organoids

For flow cytometry, organoids were dissociated with TrypLE (Thermo Fisher Scientific) and DNase I (Worthington Biochemical) for 5 min at 37 °C and subsequent vigorous pipetting. The cell suspension was washed with DMEM/F12 medium containing 10% FBS. Dead cells were fluorescently labelled using a fixable viability dye (eBioscience, 1:1,200). Antibody labelling of cells was performed in FACS staining buffer (PBS supplemented with 2% FCS and 2 mM EDTA) on ice for 30 min after blocking Fc receptors. Fc receptors were blocked with anti-CD16/32 antibodies (BD Pharmingen, 1:100). The following antibody was used: anti-CD44 (IM7, eBioscience, 1:200). Lgr5-eGFP⁺CD44⁺ cells were assessed on an LSRII cytometer (BD Biosciences) using FACSDiva (BD Biosciences). The data were analysed using the FlowJo software (Treestar).

Mouse lamina propria cells

A total of 20 cm of upper small intestinal tissue was washed with cold PBS, Peyer’s patches were removed manually and the remaining specimens were then incubated in 10 ml of gentle dissociation solution (HBSS with 10 mM EDTA and 1 mM DTT (Sigma-Aldrich)) on ice for 20 min. The tissues were shaken vigorously, and the supernatant was discarded. The remaining tissue fragments were washed with 10 ml of HBSS buffer, cut into 2–5 mm fragments and further digested in dissociation solution (advanced DMEM with 0.15 mg ml⁻¹ of collagenase P (Roche), 0.8 mg ml⁻¹ of dispase (Gibco) and 400 IU ml⁻¹ of DNase I (Worthington)) using the GentleMACS dissociator (Miltenyi) at 37 °C for 1 h. The cell suspension was filtered through a 100 μm cell strainer into a 50 ml tube, then centrifuged at 300g for 5 min and the supernatant was discarded. Dead cells were fluorescently labelled using a fixable viability dye (eBioscience, 1:1,200). Antibody labelling of cells was performed in DMEM supplemented with 2%) on ice for 30 min after blocking Fc receptors. Fc receptors were blocked with anti-CD16/32 antibodies (BD Pharmingen, 1:100). The following antibodies was used: anti-CD31 (MEC13.3, BioLegend, 1:300) and anti-podoplanin (8.1.1, BioLegend, 1:300). TdTomato expression in podoplanin⁺CD31⁻ mesenchymal cells was assessed on the LSRII cytometer (BD Biosciences) using FACSDiva (BD Biosciences). The data were analysed using the FlowJo software (Treestar).

3D-imaging and quantifications of intestinal tissue

Tissue preparation and imaging

Intestinal tissues were fixed in 4% paraformaldehyde at 4 °C for 20 h. The fixed samples were then incubated in 1% Triton X-100 solution at 4 °C overnight for permeabilization. The tissues were subsequently incubated in DAPI (1:500; Invitrogen, D3571), phalloidin (1:400; Invitrogen, A30107) or 1,1′-dioctadecyl-3,3,3′,3′-tetramethylindodicarbocyanine, 4-chlorobenzenesulfonate salt (DiD, 1: 500: Invitrogen, D7757) at 4 °C for 48 h, followed by three washes with fresh PBS over 30 min periods. The labelled samples were transferred into RapiClear 1.49 (Sunjin Lab) overnight. The samples were mounted in a 0.50 mm i-spacer (Sunjin Lab) for confocal imaging. Images were acquired using the Zeiss LSM 700 confocal microscope. The z-step size was set to 2.15 μm. Arivis Vision 4D was used for 3D Image visualization as shown in Figs. 2a,c and 3a,c and Extended Data Figs. 8j and 10d,e. Imaris was used for the image visualization shown in Supplementary Videos 1–4.

Measurement of volume, surface area and length of villi

For measuring the volume, surface and length of the villi from three-dimensional images, a custom ImageJ macro was created. The MorpholibJ library (v.1.4.1) (https://imagej.net/plugins/morpholibj) and ImageScience library (v.3.1.0) (https://imagescience.org/meijering/software/imagescience/) were used. In the first step, the image was downsampled, the crypt region was annotated manually on several 2D-slices, then interpolated to cover the volume in 3D. The crypts were removed to focus on the villi only. To create seed objects and separate the individual villi, a combination of binary operations and Laplacian of Gaussian filtering was used iteratively. The resulting objects were then regrown to their original size by 3D watershed and villi were analysed separately. Length measurement was performed using a distance map starting at the base of the villi, then reading out the maximum intensity. Volume and surface measurements were also performed on the segmented objects using MorpholibJ library.

EdU incorporation assay

Cell cycle analysis in organoids

For cell cycle analysis, organoids were incubated with 10 μM 5-ethynyl-2′-deoxyuridine (EdU) for 1 h and subsequently dissociated with TrypLE and DNase I. After dead cells were fluorescently labelled with a viability dye (eBioscience, 1:1,200), dissociated cells were fixed, permeabilized using a fixation/permeabilization kit (eBioscience) and finally stained using the Click-iT EdU kit (Life Technologies). Cell cycle stages were analysed using flow cytometry. For whole-mount imaging of EdU labelling, organoids were incubated with 10 μM EdU for 2 h and subsequently fixed and counterstained with DAPI to visualize nuclei. Images were acquired using a Zeiss LSM 700 confocal microscope.

In vivo labelling of epithelial cells in the mouse intestine

EdU (Sigma-Aldrich) was administered at 100 mg per kg body weight intraperitoneally. The time of day for EdU delivery was consistent for all the animals used. Then, 24 h after administration, mouse intestinal tissues were collected and fixed in 4% paraformaldehyde at 4 °C for 20 h. Whole-mount tissues were incubated in 1% Triton X-100 solution at 4 °C overnight for permeabilization. The EdU-incorporated cells were labelled using the Click-iT Edu kit (Life Technologies). The tissues were subsequently incubated in DAPI (1:500; Invitrogen, D3571) at 4 °C for 48 h, followed by three washes with fresh PBS over 30 min periods. The labelled samples were transferred into RapiClear 1.49 (Sunjin Lab) overnight. The samples were mounted in a 0.50 mm i-spacer (Sunjin Lab) for confocal imaging. Images were acquired using the Zeiss LSM 700 confocal microscope. The z-step size was set to 2.15 μm. Intestinal epithelial cell migration distance was defined as the distance from the crypt base to the EdU-positive cells that had migrated the farthest and was measured using Imaris software.

Histology and immunohistochemistry

For histological analysis, the dissected mouse intestines or human intestinal organoids were fixed in 4% paraformaldehyde overnight at 4 °C, dehydrated and embedded in paraffin. 2 μm paraffin-sections were deparaffinized by xylene substitute (Thermo Fisher Scientific, Shandon) and rehydrated. Rehydrated sections were stained with H&E for morphological assessment. For immunohistochemistry, after rehydration of the sections, epitopes were retrieved using sodium citrate pH 6 with 0.05% Tween-20 for 30 min or using the BOND Enzyme Pretreatment Kit (Leica AR9551) for 5 min. The sections were blocked for 1 h in 5% BSA (VWR Life Science) and 10% goat serum (Sigma-Aldrich, 9023) and incubated with primary antibodies against phospho-histone H3 (1:100; CellPath, PBC-ACI3130C), mouse OLFM4 (1:800; Cell Signaling Technology, 39141), human OLFM4 (1:100; Cell Signaling Technology, 14369), cleaved caspase-3 (1:100; Cell Signaling Technology, 9661) or CRE (1:100, Cell Signaling Technology, 15036), all diluted in blocking solution. For detecting M cells, sections were blocked for 1 h in 2% BSA (ready to use; VWR Life Science) and 5% rabbit serum (Sigma-Aldrich, R9133) and incubated with a primary antibody against glycoprotein 2 (1:150, MBL Life Science, D278-3). The sections were subsequently incubated with a secondary antibody (HRP-polymer rabbit, DCS (PD000POL-K)) and DAB (Abcam, ab64238). Finally, the sections were counter-stained with non-acidified haematoxylin (Thermo Fisher Scientific, 6765002). For detecting CRE, TMB substrate (SZABO SCANDIC) was used as replacement for DAB, in combination with Nuclear fast red. Slides were then scanned using the Mirax Scanner (Zeiss) and representative images were acquired using the Panoramic Viewer Software v.2.4.0 (3DHistech). The sections were examined with blinding to the genotype of the mice.

Immunofluorescence staining of frozen intestinal sections and organoids

Intestinal tissues were isolated from mice following trans-cardiac perfusion with PBS containing heparin and snap-frozen in Optimal Cutting Temperature (OCT) compound (Sakura). 14 μm cryosections were prepared, air-dried at room temperature for 1 h and subsequently fixed in ice-cold acetone at −20 °C for 10 min. The slides were blocked in 0.3% H₂O₂ for 60 min, Avidin/Biotin blocking buffer (Vector Laboratories) for 15 min and 10% goat serum (Alexa Fluor Tyramide SuperBoost Kit) for 60 min at room temperature, and subsequently stained with biotinylated anti-RANK antibodies (BAF692; R&D systems; 1:50) or biotinylated anti-RANKL antibodies (13-5952-82; Invitrogen; 1:150) at 4 °C overnight. The Tyramide Signal Amplification (TSA) System (Alexa Fluor Tyramide SuperBoost Kit) was used according to the manufacturer’s protocol. For further multiplexing, additional stainings were performed after the TSA fluorescence protocol. In brief, slides were stained at 4 °C overnight with anti-PDGFRα antibodies (AF1062, R&D Systems, 1:150) in 2% BSA/PBST (0.1% Tween-20), followed by donkey anti-goat Alexa Fluor 555 (A21432, Invitrogen, 1:500) as the secondary antibody. For the detection of intestinal epithelial cells, anti-EPCAM Alexa Fluor 488 (118210, BioLegend, 1:100) was used. Phalloidin and DAPI were used for membrane staining and nuclear counterstaining, respectively. Confocal images were obtained using the Zeiss LSM 700 and Zeiss LSM 710 microscopes.

For the whole-mount staining of mouse and human organoids, organoids were fixed with 4% PFA at room temperature for 15 min, followed by incubation with blocking and permeabilization solution consisting of 0.2% Triton X-100, 0.1% Tween-20, 2% BSA and 2% normal goat serum in PBS at room temperature for 1 h. Mouse organoids were stained at 4 °C overnight with anti-mouse OLFM4 (1:400; Cell Signaling Technology, 39141) in blocking and permeabilization solution. Goat anti-rabbit Alexa Fluor 633 (1:500; Invitrogen, A21072) was used as a secondary antibody. Human organoids were stained with anti-human OLFM4 antibodies (1:100; Cell Signaling, 14369) at 4 °C overnight and the TSA Fluorescence System (Alexa Fluor Tyramide SuperBoost Kit) was used according to the manufacturer’s protocol. Phalloidin and DAPI were used for membrane staining and nuclear counterstaining, respectively. Confocal images were obtained using Zeiss LSM 700 and Zeiss.

Whole-mount imaging of the mammary gland

Mammary glands were dissected from mice, spread on glass slides and fixed in Carnoy’s fixative (60% ethanol, 30% chloroform and 10% glacial acetic acid) overnight. The slides were washed in 70% ethanol for 15 min, 30% for 15 min, rinsed in distilled water for 5 min, stained in carmine alum stain (2.5 g alum potassium sulfate and 1.0 g carmine in 500 ml of double-distilled H₂O) and then washed in 70% ethanol until fat was clear and glands still visible. Subsequently, the slides were dehydrated in 95% ethanol and 100% ethanol for 1 h, respectively, followed by 1 h in xylene. The dehydrated samples were mounted with EukitNeo mounting medium. Whole-mount images were obtained using Zeiss Axio Zoom.V16.

Western blotting

Western blotting was performed using standard protocols. Total protein was extracted from isolated intestinal epithelial cells. To isolate intestinal epithelial cells, intestinal tissues were minced into around 5 mm fragments and further incubated with Gentle Cell Dissociation Reagent (StemCell Technologies) for 15 min at room temperature. The tissue fragments were vigorously resuspended and isolated intestinal epithelial cells collected by passing through a 70 µm cell strainer (SZABO SCANDIC). Isolated intestinal epithelial cells were then lysed in RIPA buffer containing a cocktail of protease and phosphatase inhibitors (Thermo Fisher Scientific, 78440). Blots were blocked for 1 h with 5% bovine serum albumin (BSA) in TBST (1× Tris-buffered saline (TBS) and 0.1% Tween-20) and then incubated overnight at 4 °C with primary antibodies, diluted in 5% BSA in TBST (1:1,000 dilution). Blots were washed three times in TBST for 15 min, then incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (1:5,000 dilution; GE Healthcare, NA9340V) for 45 min at room temperature, washed three times in TBST for 15 min and visualized using enhanced chemiluminescence (ECL Plus, Pierce, 1896327). The following primary antibodies were used: anti-β-actin (1:1,000; Sigma-Aldrich, A5316); anti-IκBα (1:1,000; Cell Signaling, 9247) and anti-phospho-IκBα (Ser32/36) (1:1,000; Cell Signaling, 9246). Secondary antibodies were anti-rabbit IgG HRP (1:5,000; GE Healthcare, NA9340V) and anti-mouse IgG HRP (1:5,000; Promega, W4021).

qPCR

Total RNA was extracted from intestinal organoids or intestinal epithelial cells. To isolate intestinal epithelial cells, intestinal tissues were minced into around 5 mm tissue pieces and then incubated with Gentle Cell Dissociation Reagent (StemCell Technologies) for 15 min at room temperature. The tissue fragments were vigorously resuspended and the isolated intestinal epithelial cells collected by passing through a 70 µm cell strainer (SZABO SCANDIC). Total RNA isolation was performed using the RNA isolation kit (VBCF) which uses a lysis step based on guanidine thiocyanate (adapted from ref. ⁵⁸) and magnetic beads (GE Healthcare, 65152105050450) applied on a KingFisher instrument (Thermo Fisher Scientific). After 5 min incubation at room temperature, DNA was digested with DNase I (New England BioLabs) for 15 min at room temperature, followed by a series of washing steps. RNA was eluted from the beads in 50 μl RNase-free water for 2 min at room temperature. Equivalent quantities of total RNA were reverse transcribed to synthesize cDNA using a LunaScript RT SuperMix Kit (New England BioLabs). qPCR was performed using Luna Universal qPCR master Mix (New England BioLabs). Primer sequences were as follows: Gapdh forward, CATCACTGCCACCCAGAAGACTG; Gapdh reverse, ATGCCAGTGAGCTTCCCGTTCAG; Rank forward, CCCAGGAGAGGCATTATGAG; Rank reverse, CACACACTGTCGGAGGTAGG; Rankl forward, GTGAAGACACACTACCTGACTCC; Rankl reverse, GCCACATCCAACCATGAGCCTT; Birc2 forward, CCACTTCAGACACCCCAGGA; Birc2 reverse, TTCCGAACTTTCTCCAGGGC; Birc3 forward, GCGTTCAGAGCCTAGGAAGT; Birc3 reverse, GTGAGATGACAGGGAGGGGA; Tnfaip3 forward, AGCAAGTGCAGGAAAGCTGGCT; Tnfaip3 reverse, GCTTTCGCAGAGGCAGTAACAG; Bcl2 forward, CCTGTGGATGACTGAGTACCTG; Bcl2 reverse, AGCCAGGAGAAATCAAACAGAGG; Bcl2l1 forward, GCCACCTATCTGAATGACCACC; Bcl2l1 reverse, AGGAACCAGCGGTTGAAGCGC; Lgr5 forward, CGGGACCTTGAAGATTTCCT; Lgr5 reverse, GATTCGGATCAGCCAGCTAC; Bmp2 forward, TGCTTCTTAGACGGACTGCG; Bmp2 reverse, TGGGGAAGCAGCAACACTAG; Id2 forward, CCAGAGACCTGGACAGAACC; Id2 reverse, CGACATAAGCTCAGAAGGGAAT; Id3 forward, AGCTCACTCCGGAACTTGTG; Id3 reverse, AGAGTCCCAGGGTCCCAAG; GABDH forward, AATGAAGGGGTCATTGATGG; GABDH reverse, AAGGTGAAGGTCGGAGTCAA; BIRC2 forward, CAGACACATGCAGCTCGAATGAG; BIRC2 reverse, CACCTCAAGCCACCATCACAAC; BIRC3 forward, GCTTTTGCTGTGATGGTGGACTC; BIRC3 reverse, CTTGACGGATGAACTCCTGTCC; TNFAIP3 forward, CTCAACTGGTGTCGAGAAGTCC; TNFAIP3 reverse, TTCCTTGAGCGTGCTGAACAGC; BCL2L1 forward, GCCACTTACCTGAATGACCACC; BCL2L1 reverse, AACCAGCGGTTGAAGCGTTCCT; BMP2 forward, TGTATCGCAGGCACTCAGGTCA; BMP2 reverse, CCACTCGTTTCTGGTAGTTCTTC; ID2 forward, TTGTCAGCCTGCATCACCAGAG; ID2 reverse, AGCCACACAGTGCTTTGCTGTC; OLFM4 forward, GACCAAGCTGAAAGAGTGTGAGG; OLFM4 reverse, CCTCTCCAGTTGAGCTGAACCA.

QuantSeq 3′ mRNA-seq

Library preparation

The protocol for total RNA extraction was performed as described above in the ‘qPCR’ section. RNA quantification and quality control were performed using a DNF-471 Standard Sensitivity RNA Analysis kit (Agilent) with a fragment analyzer. Equivalent quantities of total RNA were used for library preparation using the Lexogen QuantSeq 3′ mRNA-Seq Library Prep Kit FWD from Illumina. The DNF-474 High Sensitivity NGS Fragment Analysis Kit (1–6,000 bp) (Agilent) was used to determine the quality of the library with a fragment analyzer. 3′ RNA-seq (QuantSeq) reads were prepared for analysis by removing adaptor contamination, poly(A) read through and low-quality tails using bbmap v.36.92. Libraries were pooled at an equimolar ratio and sequenced on an Illumina HiSeq 2500 instrument using the single-read 50-read mode.

Data analysis

RNA-seq reads were trimmed using BBDuk v38.06 (ref=polyA.fa.gz,truseq.fa.gz k=13 ktrim=r useshortkmers=t mink=5 qtrim=r trimq=10 minlength=20). Reads mapping to abundant sequences included in the iGenomes UCSC GRCm38 reference (mouse rDNA, mouse mitochondrial chromosome, phiX174 genome, adapter) were removed using bowtie2 v.2.3.4.1 alignment. The remaining reads were analysed using genome and gene annotation for the GRCm38/mm10 assembly obtained from Mus musculus Ensembl release 94. Reads were aligned to the genome using star v.2.6.0c and reads in genes were counted with featureCounts (subread v.1.6.2) using strand-specific read counting for QuantSeq experiments (-s 1). Differential gene expression analysis on raw counts was performed using DESeq2, over-representation analysis with clusterProfiler v.4.4.4 and gene set enrichment analysis with fgsea v.1.22.0. The relevant signalling processes and biological functions were evaluated using the commercial QIAGEN’s Ingenuity Pathway Analysis (IPA, Qiagen; www.qiagen.com/ingenuity) software. The z score was applied to predict a cellular process’ directional change, such as activating or inhibiting a cellular pathway. The Benjamini–Hochberg method was used to adjust canonical pathway P values.

Single-cell sorting of mouse intestinal lamina propria cells for sequencing

The protocol was modified from a previous study²⁶. In brief, 20 cm of upper small intestinal tissue was carefully washed with cold PBS, Peyer’s patches were removed manually, and then the remaining specimens were incubated in 10 ml of gentle dissociation solution (HBSS with 10 mM EDTA and 1 mM DTT (Sigma-Aldrich)) on ice for 20 min. The tissues were shaken vigorously, and the supernatant was collected in a new conical tube, washed with HBSS buffer and suspended in 10 ml of HBSS buffer, suspension ‘A’. The remaining tissue fragments were washed with 10 ml of HBSS buffer, cut into 2–5 mm fragments and were further digested in dissociation solution (advanced DMEM with 0,15 mg ml⁻¹ of collagenase P (Roche), 0.8 mg ml⁻¹ of dispase (Gibco), and 400 IU ml⁻¹ of DNase I (Worthington)) using a GentleMACS dissociator (Miltenyi) at 37 °C for 1 h. The cell suspension was filtered through a 100 μm cell strainer into a 50 ml tube, then centrifuged at 300g for 5 min, and the supernatant was discarded. The cell pellets were then combined with the cell suspension ‘A’. Dead cells were fluorescently labelled using a fixable viability dye (eBioscience, 1:1,200). Antibody labelling of cells was performed in DMEM supplemented with 2%) on ice for 30 min after blocking Fc receptors. Fc receptors were blocked with anti-CD16/32 antibodies (BD Pharmingen, 1:100). The following antibodies were used: anti-CD45 (IM7, eBioscience, 1:400) and EPCAM (G8.8, BioLegend, 1:800). Using FACSAria III Cell Sorter (BD), CD45-positive and EPCAM-negative cells (Immune cells) and CD45⁻ and EPCAM⁻ cells (mesenchymal cells) were enriched by cell sorting.

scRNA-seq

Library preparation from mouse intestinal organoids

Control mouse intestinal organoids and organoids cultured in the presence of recombinant mouse RANK ligand (rmRANKL, Oriental Yeast) for 12 h were dissociated with TrypLE (Thermo Fisher Scientific) and DNase I (Worthington Biochemical) for 5 min at 37 °C and subsequent vigorous pipetting through a p200 pipette. The cell suspension was washed with DMEM/F12 medium containing 10% FBS. Cell viability and efficiency of dissociation were determined using Nucleocounter NC-250 (Chemometec) before the single cells were loaded into one channel of a 10x Chromium microfluidics chip to package them into one library. scRNA-seq libraries were generated using 10x Genomics kits. The libraries were sequenced on an Illumina NovaSeq 6000.

Library preparation from mouse intestinal lamina propria cells

For each sample, 1 million cells were fixed for 22 h at 4 °C, quenched and stored at −80 °C according to 10x genomic Fixation of Cells & Nuclei for Chromium Fixed RNA profiling (CG000478, 10X Genomics, Pleasanton, CA) using the Chromium Next GEM Single Cell Fixed RNA Sample preparation kit (PN-1000414, 10X Genomics). In total, 250,000 cells per sample were used for probe hybridization using the Chromium Fixed RNA Kit, mouse WTA probes (PN-1000496, 10X Genomics), pooled at equal numbers and washed following the Pooled Wash Workflow following the Chromium Fixed RNA Profiling Reagent kit protocol (CG000527, 10X Genomics). GEMs were generated using Next GEM ChipQ (PN-1000422, 10X Genomics) on the Chromium X (10X Genomics) system with a target of 10,000 cells recovered and libraries prepared according to the manufacturer instructions (CG000527, 10x Genomics). Sequencing was performed using NovaSeq S4 lane PE150 (Illumina) with a target of 15,000 reads per cell. Alignment of the samples was performed using the 10x Genomics Cell Ranger 7.1.0 multi pipeline.

Data analysis (mouse intestinal organoids)

Reads were aligned to the reference mouse genome (mm10) downloaded from the 10x Genomics website (v.2020-A) using the Cell Ranger (v.5.0.1) count function with the default parameters. Genome annotation corresponded to Ensembl v98. The median number of unique molecular identifiers (UMIs) per cell was between 23,501 and 25,058, with a median of 3,808–4,300 genes detected per condition. The computational analysis of the 10x Genomics UMI count matrices was performed using the R package Seurat (v.4.0.5). Cells were subjected to a quality-control step, keeping those cells expressing more than 1,000 genes and with less than 20% of UMIs assigned to mitochondrial genes. Those thresholds were chosen after visual inspection of the distributions. Using this filtering, we retained between 844 and 2,299 cells, with a median of 3,877–4,512 genes per cell detected per condition. Genes expressed in less than three cells for a sample independently or in less than five cells when the samples were merged, were removed from the analyses. Each dataset was subjected separately to normalization, identification of highly variable genes and scaling using the SCTransform function. After obtaining principal components with RunPCA for each sample independently, we integrated them using reciprocal PCA (RPCA) to identify anchors with the FindIntegrationAnchors function (setting the reduction parameter to rpca), as we expect some cell type differences after rmRANKL treatment, therefore avoiding a possible overintegration.

To annotate cell populations, we performed an unsupervised clustering analysis using the Louvain algorithm with a resolution of 0.7 in a shared nearest neighbours graph constructed with the first 20 principal components, as implemented in the FindClusters and FindNeighbors Seurat functions. Nonlinear dimensional reduction for visualization was performed using the RunUMAP function with the same principal components. Cluster 6 was further subdivided in an unsupervised manner using the FindSubCluster function with a 0.6 resolution, enabling us to separate goblet and Paneth cells without splitting the rest of the clusters any further. Markers in each cluster were identified using the FindConservedMarkers and FindAllMarkers functions in the log-normalized counts by using the Wilcoxon rank-sum test. Genes with P value < 0.05 (adjusted by Bonferroni’s correction) and a log₂-transformed fold change of >0.25 were retained. Clusters were annotated in accordance with those makers, as well as considering small intestinal cell-type markers from previous studies^22,26,59. To further confirm our classifications, cell type annotations from the small intestine scRNA-seq dataset from a previous study²² were transferred using the TransferData function in Seurat after removing distal cells and simplifying the TA annotation in the reference. We used UCell to obtain scores for gene sets of interest in each cell. The plots were generated using the DimPlot and VlnPlot functions from Seurat as well as the ggplot2 and pheatmap R libraries.

Data analysis (human intestinal crypt cells)

For the computational analysis of scRNA-seq data from human intestinal crypt cells, the 10x Genomics scRNA-seq expression matrix of human intestinal crypt cells from a previous study⁴³ was downloaded from the Gene Expression Omnibus (GSM3389578). Cells were already filtered in the dataset. Clustering and UMAP dimensionality reduction were performed with Seurat using similar parameters as in their study, that is, considering the first 25 principal components and a k.param of 20 for FindNeighbors and a resolution of 0.6 in FindClusters. A small cluster corresponding to non-epithelial cells was detected and removed from the analyses, redoing the downstream analyses and unsupervised clustering with a 0.8 resolution. The clusters were annotated considering markers and labels from the original paper.

Data analysis (mouse lamina propria cells)

Sample demultiplexing and read alignment were performed using the Cell Ranger (v.7.2.0) multi-function with the default parameters, considering the reference mouse genome (mm10) downloaded from the 10x Genomics website (v.2020-A) and the Chromium_Mouse_Transcriptome_Probe_Set_v1.0.1_mm10-2020-A.csv probe set. The median number of UMIs per cell was between 4,063 and 5,902, with a median of 2,050–2,589 genes detected per condition. The computational analysis of the 10x Genomics UMI count matrices was performed using Seurat (v.4.2.0). Cells were subjected to a quality control step, keeping those cells expressing more than 500 genes, 1,000 UMIs and with less than 5% of UMIs assigned to mitochondrial genes and cells considered singlets by scDblFinder (v.1.12.0) with the default parameters. Using this filtering, we retained between 6,822 and 9,396 cells, with a median of 2,008–2,451 genes per cell detected per condition. Genes expressed in less than three cells for a sample independently were removed from the analyses. Each dataset was subjected separately to normalization, identification of highly variable genes and scaling using the SCTransform function with vst.flavor v.2. We integrated the data using canonical correlation analysis.

To annotate cell populations, we performed an unsupervised clustering analysis using the Louvain algorithm with a resolution of 0.5 in a shared nearest-neighbours graph constructed with the first 17 principal components. Cluster 5 and 12 were subset, reintegrated with canonical correlation analysis after normalization with SCTransform and reclustered in an unsupervised manner with a 0.8 resolution and 15 principal components, allowing to further separate CD4 T cells and ILCs. Markers in each cluster were identified using the FindConservedMarkers and FindAllMarkers functions in the log-normalized counts by using the Wilcoxon rank-sum test. Genes with P value < 0.05 (adjusted by Bonferroni’s correction) and a log₂-transformed fold change of >0.25 were retained. Clusters were annotated in accordance with those markers, as well as considering small intestinal cell type markers as follows; naive CD4 T cells: Ccr7, Klf2, Sell. Activated T cells: Cd40Ig, Cd4. Th1 cells: Il12rb2, Ccr5. T helper 17 cells: Il17a, Il17f, Rora. Regulatory T cells: Foxp3, Ctla4, Il10, Tnfrsf4. Memory T cells: Zbtb16, Zfp683. CD8 T cells: Cd8a, Itgae, Gzma. CD4⁻CD8⁻ T cells: Trdc, Cd163l1, Ly6g5b, Cd3e. ILC1: Tbx21, Tyrobp, Ccl3, Xcl1, Il13. ILC2: Gata3, IL17rb, Hs3st1. ILC3: Rorc, Il22, Slc6a20a. B cells: Cd79a, Cd19, Pax5. Plasma cells: Igha, Igkc, Jchain, Xbp1, Mzb1. Macrophage: Cd14, Unc93b1, Lyz2, Il1b. PDGFRA^lowCD81⁺ trophocytes: Cd81, Ackr4, Cd34, Grem1, Col14a1, Dcn. PDFGRA^lowGREM1^med stromal cells: Dkk2, Wnt2b. PDGFRA^lowGREM1⁻ stromal cells: Sfrp1, Frzb, Fgfr2. PDGFRA^high telocytes: Pdgfra, Bmp7, Bmp5, Wif1, Chrd, Dkk3. Myofibroblast: Myh11, Hhip, Npnt. Smooth muscle cells: Atp1b2, Des, Fhl5, Rgs4. Vascular endothelial cells: Pecam1, Plvap, Flt1. Lymphatic endothelial cells: Lyve1, Mmrn1, Rspo3. Glia cells: Gpr37l1, Sox10, Kcna1.

Collection of milk and serum

Lactating female mice were separated from their offspring at lactation day 8 and fasted for 5 h from 10:00 to 15:00. Subsequently, they were anaesthetized with isoflurane (2% induction and 1% maintenance) and injected with 2 IU of oxytocin (Sigma-Aldrich, O3251) intraperitoneally. Expressed milk was collected with a P20 pipette. Serum was collected from inferior vena cava from mice anaesthetized with ketamine–xylazine and pooled into Micro sample tube Lithium heparin (Sarstedt). Serum samples were centrifuged at 2,000g for 10 min at room temperature twice to separate from cells. All samples were then stored at −80 °C for further analysis. The concentrations of milk IgA and IgG were measured with an ELISA kit (Bethyl Laboratories).

MS analysis

Samples were prepared by adding 100 μl of a methanol/ethanol mixture (4:1, v/v) to 25 μl of the respective serum or milk samples in a 1.5 ml tube, followed by vortexing, incubation and centrifugation. The supernatants were transferred to HPLC vials and measured consecutively with reversed-phase (RP) and hydrophilic interaction chromatography (HILIC) on-line coupled to liquid chromatography–tandem mass spectrometry (LC–MS/MS). Then, 2.5 μl of each sample was pooled for quality control. Metabolite extracts were separated (HILIC) on a SeQuant ZIC-pHILIC HPLC column (Merck, 100 × 2.1 mm; 5 µm) or a RP-column (Waters, ACQUITY UPLC HSS T3 150 × 2.1; 1.8 μm) with a flow rate of 100 µl min⁻¹, using the Ultimate 3000 HPLC system coupled to a Q-Exactive Focus (both Thermo Fisher Scientific). In HILIC, the gradient was ramped up in 21 min from 90% A (100% acetonitrile) to 60% B (25% ammonium bicarbonate in water). In RP, the 20 min gradient started with 99% A (0.1% formic acid in water) and ramped up to 60% B (0.1% formic acid in acetonitrile). Eluting compounds were directly ionized by electrospray ionization in polarity switching mode. Spectra were acquired in data-dependent acquisition mode using high-resolution tandem mass spectrometry. The ionization potential was set to +3.5/−3.0 kV, the sheath gas flow was set to 20, and an auxiliary gas flow of 5 was used. Obtained datasets were processed by Compound Discoverer 3.0 (Thermo Fisher Scientific). Annotation was conducted by searching the metabolite databases (mzCloud, our in-house database, ChemSpider, BioCyc, Human Metabolome Database, KEGG, MassBank and MetaboLights) with a mass accuracy of 3 ppm for precursor masses and, if applicable, 10 ppm for fragment ion masses.

For measurement of triglycerides with LC–MS, lipids were extracted using chloroform–methanol extraction from each sample. The chloroform phase was removed and diluted 1:1 with methanol and 1 μl of each sample was directly injected on a Kinetex C8 column (100 Å, 150 × 2.1 mm) using a 20 min gradient of 80% A (60% acetonitrile, 10 mM ammonium acetate, 0.1% formic acid, 40% water) to 95% B (90% isopropanol, 10 mM ammonium acetate, 0.1% formic acid and 5% water) using a flow rate of 100 µl min⁻¹ and a 60 °C column temperature. Triglycerides were detected and quantified in the positive-ion mode as their ammonium adducts.

Metabolic studies

For analysis of offspring delivered from Rank^WT or Rank^Δvil female mice, the mice were fed normal chow from weaning age until 4 weeks of age, after which they were fed normal chow or HFD (60% kcal% fat, Research Diets, D12492i) for up to 25 weeks. The pups were weekly weighed starting from postnatal day 7 until 25 weeks. For oral glucose-tolerance tests, mice (aged 25 weeks) were fasted overnight and were then administrated an oral glucose bolus by gavage (2 g per kg for normal chow-fed mice and 1 g per kg for HFD-fed mice). Glucose concentrations were measured using glucometers from blood taken by tail nick at 0, 15, 30, 45, 60 and 120 min after glucose ingestion, using a handheld blood glucose meter (One Touch UltraEasy; Lifescan). The area under the glucose-tolerance test curve was calculated for each mouse using GraphPad Prism v.9.3.1c (GraphPad Software). For analysis of insulin levels, tail-vein blood samples were added to Micro sample tube Lithium heparin (Sarstedt) to avoid blood clotting. Plasma insulin levels were measured using the Alpco Mouse Ultrasensitive Insulin ELISA (80-INSMSU-E10). For each litter of offspring, blood samples were taken the same time of the day.

Statistics and reproducibility

All values are expressed as means ± s.e.m. GraphPad Prism 8 software was used to perform statistical analyses. All details of the statistical tests used are stated in the figure legends. Two-tailed Student’s t-tests, two-tailed Mann–Whitney U-tests and one-way ANOVA with two-tailed Tukey’s test were used as described in the figure legends. Two-way ANOVA was used to compare two groups over time. Survival curves were compared using the log-rank (Mantel–Cox) test. Unless otherwise specified in the main text or figure legends, all experiments reported in this study were repeated at least two independent times.

Reporting summary

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

• SEO Powered Content & PR Distribution. Get Amplified Today.
• PlatoData.Network Vertical Generative Ai. Empower Yourself. Access Here.
• PlatoAiStream. Web3 Intelligence. Knowledge Amplified. Access Here.
• PlatoESG. Carbon, CleanTech, Energy, Environment, Solar, Waste Management. Access Here.
• PlatoHealth. Biotech and Clinical Trials Intelligence. Access Here.
• Source: https://www.nature.com/articles/s41586-024-08284-1