Stem Cell

Reassessing endothelial-to-mesenchymal transition in mouse bone marrow: insights from lineage tracing models – Nature Communications

Time Stamp: December 19, 2023 7:00 PM
Source Node: 388887

Animal experiments and genetically modified mice

All experiments were performed according to the ethics approvals of the Institutional Animal Care and Use Committee of Central South University, Changsha, Hunan, China (Approval#: 2020sydw0800). Mice were maintained at a maximum of five mice per cage under a standard 12 h light/dark cycle, and had access to food and water ad libitum.

The following strains were used: C57BL/6J, B6.Cg-Tg(Prrx1-cre)1Cjt/J⁴⁷ (The Jackson Laboratory), B6.129(Cg)-Leprtm2(cre)Rck/J⁴⁸ (The Jackson Laboratory), B6.Cg-Gt(ROSA)26Sortm9(CAG-tdTomato)Hze/J⁴⁹ (The Jackson Laboratory), B6.129S-Cdh5tm1(rtTA-tetO-Cre)Smoc (Shanghai Model Organisms Center, NM-KI-18006), B6.Cg-Tg(Tek-cre/ERT2)1Arnd/ArndCnrm⁵⁰ (EMMA).

Pooled sample from both male and female mice were used in the FACS experiments and male mice were used in other experiments at the indicated postnatal ages. To generate Prrx1-Cre;R26T and Tek-CreERT2;R26T double transgenic mice, male Prrx1-Cre or Tek-creERT2 mice were bred with female Rosa26-LSL-tdTomato mice. In the tamoxifen treatment, mice were intraperitoneally injected with tamoxifen (Sigma‒Aldrich, T5648) at a dose of 75 mg/kg body weight for 5 consecutive days. For doxycycline treatment, juvenile mice and pregnant dams were orally administered with doxycycline (MCE, HY-N0565B) in their drinking water⁵¹ (0.2 g doxycycline and 2 g sucrose per 100 mL water) for 5 and 4 consecutive days, respectively. In the 5-FU/PBS treatment, mice received 5-FU (Sigma‒Aldrich, F6627) at a dose of 150 mg/kg body weight or equal volumes of PBS for one dose via the intraperitoneal route.

Isolation and culture of BMSCs

Protocols for isolating BMSCs from mouse bone marrow⁵² and compact bone⁵³ were combined with some modifications. Briefly, tibiae and femurs were collected from euthanized mice and cleaned of attached muscles, connective tissues, and the epiphysis in PBS supplemented with 2% FBS. The cleaned bones were then cut into small pieces with scissors in stromal cell growth medium composed of αMEM supplemented with 15% qualified FBS (Gibco, 12664-025). Subsequently, the mixtures of bone chips and bone marrow were plated in tissue culture vessels. After 16–24 h, the culture medium was aspirated, and the cells were gently washed with PBS before being resupplied with fresh growth medium. The adherent cells were cultured for at least 2 passages to eliminate ECs and hematopoietic cells⁵² before being subjected to further analyses.

Purification of Tomato⁺ BMSCs using FACS

To purify Tomato⁺ BMSCs from Cdh5-tetO-Cre;R26T or Tek-CreERT2;R26T mice, cultured BMSCs were detached from culture vessels using accutase (Gibco), passed through 40 μm cell strainers, and centrifuged at 300 × g for 5 min. The pellet was resuspended in PBS containing 2% FBS, and the Tomato⁺ cells were analyzed in the PE channel of the FACSAria II Cell Sorter (BD Biosciences). Single, viable Tomato⁺ cells were then sorted into PBS containing 10% FBS. The sorted cells were centrifuged at 300 × g for 5 min and plated onto tissue culture vessels in stromal cell growth medium.

Osteogenic, adipogenic, and chondrogenic differentiation of cultured BMSCs

BMSCs in culture were subjected to osteogenic, adipogenic, or chondrogenic differentiation using the respective differentiation medium (Cyagen Biosciences) for up to 3 weeks following the manufacturer’s instructions. The differentiated cells were stained with Alizarin Red, Oil Red O, and Toluidine Blue, respectively.

Isolation and culture of bone marrow ECs

Tibiae and femurs obtained from euthanized mice were cleaned of attached muscles, connective tissues, and the epiphysis. The cleaned bones were then cut into small pieces in digestion medium composed of αMEM supplemented with 2% FBS, 1 mg/mL type II collagenase (Gibco, 17101015), 1 mg/mL Dispase II (Sigma‒Aldrich, 04942078001), and 1 mg/mL DNase (Sigma‒Aldrich, 10104159001). The mixtures of bone chips and bone marrow were incubated in an orbital shaker at 120 r.p.m., 37 °C for 45 min. Subsequently, the samples were filtered through cell strainers and centrifuged, and the pellet was plated onto tissue culture vessels coated with collagen I (from rat tail, BD biosciences, 354236) in endothelial growth medium (EGM, Lonza, CC-3202). After 16–24 h, the culture medium was aspirated, and the cells were gently washed with PBS. The adherent cells containing ECs, hematopoietic cells, and BMSCs were subjected to flow cytometry or immunostaining analysis. Alternatively, the adherent cells were resupplied with fresh EGM to subsequently enrich for EMCN⁺ BMSCs.

Purification of EMCN⁺ BMSCs using MACS and FACS

To isolate EMCN⁺ BMSCs, bone marrow cells cultured in EGM were detached from culture vessels using accutase. The cells were then passed through cell strainers, centrifuged, and resuspended in PBS containing 2% FBS. Subsequently, an EMCN-PE antibody (Santa Cruz, sc-65495) and anti-PE microbeads (Miltenyi Biotec, 130-048-801) were used to purify EMCN⁺ cells following the manufacturer’s instructions. The MACS-purified EMCN⁺ cells were immortalized with the SV40T antigen⁵⁴ constructed in a recombinant lentivirus vector (Shanghai GeneChem Co., Ltd.). The immortalized cells were propagated for a few passages and then stained with the EMCN-PE antibody (eBioscience, 12-5851-82) to enrich for EMCN⁺ cells by FACS for 3 rounds.

Flow cytometry

Sample preparation for uncultured bone marrow cells

Tibiae and femurs were collected and cleaned as described above and cut into small pieces in the digestion medium. The samples were then incubated in an orbital shaker at 120 r.p.m., 37 °C for 45 min, filtered through cell strainers, and centrifuged. The pellet was resuspended and subjected to flow cytometry analysis. Alternatively, the resuspended cells were depleted for lineage-positive cells using a lineage cell depletion kit (MiltenyiBiotec, 130-090-858) as per the manufacturer’s instructions before flow cytometry analysis.

Sample preparation for cultured bone marrow cells

The culture medium was aspirated, and the cells were washed with PBS. Subsequently, accutase was added to the culture vessels to detach the cells while preserving their endothelial markers. The detached cells were then collected and used for subsequent analyses.

Flow cytometry analysis and FACS

Cell suspensions were blocked with CD16/32 (clone 93, Biolegend, 101320, 1:200) at room temperature for 10 min and stained with the following antibodies at room temperature for 30 min: CD31-BV711 (clone 390, Biolegend, 303136, 1:100), EMCN-AF488/eFluor660 (clone V.7C7, e-Bioscience, 50-5851-82/53-5851-82, 1:100), TIE2-APC (clone TEK4, e-Bioscience, 17-5987-82, 1:100), CDH5-BV421 (clone 11D4.1, BD OptiBuild, 747749, 1:100), PDGFRα-APC/BV421 (clone APA5, BD Pharmingen, 562777/562774, 1:100), CD51-APC (clone RMV-7, Elabscience, E-AB-F1235E, 1:100), Sca-1 (clone D7, Biolegend, 108111, 1:100), CD44-APC (clone IM7, Biolegend, 103011, 1:100), CD29-APC (clone HMβ1-1, Biolegend, 102215, 1:100), CD45-PercpCy5.5/APC-CY7 (clone 30-F11, BD Pharmingen, 550994/557659, 1:100), Lineage cocktail-PercpCy5.5 (including Ly76, Ly-6G/Ly-6C, B220, CD11b, CD3e, clones TER119, RB6-8C5, RA3-6B2, M1/70, and 145-2C11, respectively, BD Pharmingen, 561317, 1:50), CD71-PercpCy5.5 (clone C2, BD Pharmingen, 562858, 1:100), CD19-APC (clone 1D3, eBioscience, 17-0193-80, 1:200), CD3-APC (clone 17A2, Cell Signaling Technology, 24265s, 1:200), Ly-6G-APC (clone 1A8, BD Pharmingen, 560599, 1:100), CD11b-FITC (clone M1/70, BD Pharmingen, 557396, 1:100), BV711 rat IgG2a κ isotype control (clone RTK2758, Biolegend, 400551, 1:100), BV421 rat IgG2a κ isotype control (clone R35-95, BD Horizon, 562602, 1:100), eFluor660 rat IgG2a κ isotype control (clone eBR2a, e-Bioscience, 50-4321-82, 1:100), APC rat IgG1 κ isotype control (clone eBRG1, e-Bioscience, 17-4301-82, 1:100). Cells were then washed with PBS and centrifuged. The pellet was resuspended in PBS supplemented with 2% FBS and subjected to flow cytometry analysis. To identify singlet cells, a gate was drawn based on the forward scatter area (FSC-A) versus forward scatter height (FSC-H) or forward scatter width (FSC-W) features, and dead cells were distinguished using Zombie Aqua dye (Biolegend, 423101, 1:1000).

Equipment

Flow cytometry analyses were performed on FACSCanto II Cell Analyzer (BD Biosciences) and Aurora Analyzer (Cytek Biosciences). Cell sorting was carried out using the FACSAria II Cell Sorter (BD Biosciences). Data were collected using Flowjo CE (Tree Star) or FACS DIVA software (6.1.3, BD Biosciences). The raw data were analyzed by FlowJo software (V10, Tree Star).

Single-cell RNA sequencing and analysis

Sample preparation

Tibiae and femurs were collected from 5-week-old wild-type mice (n = 3 mice), cleaned of attached muscles, and cut into small pieces in the digestion medium. The released bone marrow (bm) and bone fragments (bo) were treated as separate samples and incubated in an orbital shaker at 120 r.p.m., 37 °C for 45 min. Subsequently, the samples were filtered through cell strainers and centrifuged. For the bm sample, lineage-positive cells were depleted using MACS (MiltenyiBiotec, 130-090-858), and then Lin⁻CD45⁻CD71⁻CD3⁻CD19⁻ cells were sorted using FACS. For the bo sample, erythrocytes were lysed (ThermoFisher Scientific, A1049201) and Lin⁻CD45⁻CD71⁻CD3⁻CD19⁻ cells were sorted using FACS.

Single-cell library preparation, sequencing, and analysis

Single-cell mRNA libraries were prepared for sequencing using the Chromium Next GEM Single Cell 3’ GEM, Library & Gel Bead Kit v3.1 (#1000121). The samples were sequenced using an Illumina NovaSeq 6000. Reads from scRNA-seq were aligned to mm10 and collapsed into UMI counts using 10x Genomics Cell Ranger software (version 4.0.0) with default parameters. Further analyses were performed using Seurat 3.0 in the R statistical language.

For quality control, the following cells were excluded from analysis: (1) Cells with more than 20% mitochondrial gene expression. (2) Cells in the top 2% quantile of nGene and nUMI. (3) Cells with the value of log10(GenesPerUMI) no more than 0.8. (4) Hematopoietic clusters and small clusters without clear characteristics of BMSCs or ECs. A total of 5554 cells were included for analysis. The re-clustering of ECs was performed using a resolution of 0.1. Doublets ratios in the EC subclusters were analyzed using the scDblFinder package (v.1.8.0) with default parameters²⁶.

When cells were fractioned into two subsets, one positive for a target gene and one negative for the same gene, the cells with gene reads greater than zero were defined as positive, otherwise as negative.

Pseudotime analysis were performed between the EC clusters and each BMSC subtypes using the Monocle package⁵⁵. We selected the top 1000 significantly differentially expressed genes as the ordering genes for the trajectory reconstruction.

All plots were generated using the ggplot2 and VennDiagram packages in R (4.0.2). Boxplots are displayed as follows: the median (middle line), the first and third quartiles (lower and upper edges of the “boxes”), the largest/smallest values no further than 1.5 times the distance between the first and third quartiles (upper/lower whiskers), data beyond the end of the whiskers (individually plotted dots), and the mean (small dots within “boxes”).

Reanalysis of scRNA-seq datasets from published literature

The original datasets were downloaded from the NCBI GEO database, and Cell Ranger software was used to conduct preliminary data analysis and generate the gene expression matrix. Following quality control, clusters were identified and named based on the original literature with some modifications. Hematopoietic clusters and small clusters without clear characteristics of BMSCs or ECs were excluded from analysis or visualization. The data were subsequently analyzed similarly to the scRNA-seq analysis of the bo/bm samples.

Immunofluorescence assay and image acquisition

Immunostaining of tibia/femur sections

Tibiae and femurs were collected and fixed in 4% PFA at 4 °C for 24 h. After fixation, the samples were decalcified in 0.5 M EDTA (pH = 7.4) at 4 °C with constant shaking for 1–3 days and immersed in a 30% sucrose solution for 24 h before being embedded in O.C.T. Cryostat sections were generated at a thickness of 10–20 μm. These sections were rehydrated in PBS, blocked with PBS containing 5% donkey serum or 4% BSA at room temperature for 30 min, and then probed with primary antibodies diluted in the blocking solution overnight at 4 °C. After removing the primary antibodies, the sections were washed in PBS and stained with secondary antibodies for 45 min at room temperature. The nuclei were counterstained with DAPI (Vector) before mounting the sections with coverslips.

Immunostaining of cultured cells

The cells were plated on coverslips and allowed to grow before being fixed with 4% PFA at room temperature for 30 min. Subsequently, the coverslips were blocked with PBS containing 5% donkey serum at room temperature for 30 min. The cells were then incubated with primary antibodies overnight at 4 °C. Following the removal of the primary antibodies, the sections were washed in PBS and stained with secondary antibodies for 45 min at room temperature. The nuclei were counterstained with DAPI before mounting the coverslips on glass slides.

The following primary antibodies were used for immunostaining: CD31 (Abcam, ab28364, 1:100 for coverslips or R&D Systems, FAB3628G, 1:100 for bone sections), Endomucin (Santa Cruz, sc-65495, 1:200), alpha smooth muscle actin (Abcam, ab124964, 1:400), CD45 (BD Pharmingen, 557659, 1:200), and RUNX2 (Cell Signaling Technology, 12556 S, 1:400).

The following secondary antibodies were used in immunostaining (all obtained from Jackson ImmunoResearch): donkey anti-rabbit Alexa Fluor 488 (711-545-152, 1:400) and Alexa Fluor 647 (711-605-152, 1:400); donkey anti-rat Alexa Fluor 488 (712-545-150, 1:400), Alexa Fluor 594 (712-585-150, 1:400), and Alexa Fluor 647 (712-605-150, 1:400); and donkey anti-goat Alexa Fluor 488 (705-545-147, 1:400).

Quantification of colocalization in immunostaining analysis

In wild-type mice, we quantified the colocalization of EMCN and αSMA in EMCN⁺αSMA⁺ cells that surrounded or aligned with arteries or arterioles. Similarly, in Cdh5-tetO-Cre;R26T mice, we evaluated the colocalization of Tomato fluorescence and αSMA in αSMA⁺ cells surrounding or aligning with arteries/arterioles. Single channel images were generated using ZEN software (2.3, Zeiss) and analyzed with the Colocalization Finder tool in ImageJ software (1.54f, NIH, Bethesda, MD, USA). The Pearson correlation coefficient (PCC) was calculated between EMCN and αSMA, as well as Tomato fluorescence and αSMA. The PCC ranges from −1 to 1, with a value of 1 indicating a perfect linear relationship between the distributions of the two probes, −1 indicating a perfect inverse relationship, and 0 indicating an uncorrelated distribution of the probes⁵⁶.

RNAscope ISH analysis

Tibiae and femurs were collected from euthanized mice and fixed with 4% PFA at 4 °C for 24 h. Decalcification was then performed using fast decalcification buffer (Beijing Pursuit Bio Co., Ltd.) at 4 °C with constant shaking for 20 h. After that, the samples were dehydrated in a series of ethanol solutions and cleared with xylene. Subsequently, the samples were embedded in paraffin wax to generate sections at a thickness of 10 μm. These sections were deparaffinized and rehydrated before undergoing ISH using the RNAscope® Multiplex Fluorescent Reagent Kit v2 following the manufacturer’s protocol. The target probes used in ISH were Prrx1 (ACD 485231-C2, hybridizing with NM_011127.2, nucleotides 254–1726) and Lepr (ACD 471171, hybridizing with NM_146146.2, nucleotides 3220–4109). The mRNA signals were detected with opal570 and opal690, respectively. A probe against the bacterial dapB gene was used as a negative control. The sections were then subjected to immunostaining with CD31 and EMCN antibodies as described above.

Image acquisition

Images of immunostaining for cultured cells or bone sections were acquired using a Zeiss Axio Imager M2 microscope equipped with an ApoTome.2 system. Images of Alizarin Red, Oil Red O, and Toluidine Blue staining were acquired using an Olympus CX31 optical microscope. ZEN (2.3) and ImageJ (1.54f) software were used for image processing.

Statistics & Reproducibility

The results are presented as the mean ± S.E.M. Differences between experimental groups were evaluated by two-tailed Student’s t-test (adjusted for unequal variances with Welch’s test where appropriate), one-way ANOVA, or two-tailed Wilcox rank-sum test, as indicated in the figure legends. GraphPad Prism software (v8.4) or R (4.0.2) was used for statistical analyses. p value less than 0.05 was considered statistically significant. No statistical method was used to predetermine the sample size. Required experimental sample sizes were estimated based on previous established protocols in the field. The experiments were not randomized, and the investigators were not blinded to the experimental allocation or outcome assessment.