Single-Cell RNA Sequencing-Based Characterization of Resident Lung Mesenchymal Stromal Cells in Bronchopulmonary Dysplasia

Ivana Mižíková; Flore Lesage; Chanele Cyr-Depauw; David P Cook; Maria Hurskainen; Satu M Hänninen; Arul Vadivel; Pauline Bardin; Shumei Zhong; Olli Carpén; Barbara C Vanderhyden; Bernard Thébaud

doi:10.1093/stmcls/sxab023

. 2022 Jan 24;40(5):479–492. doi: 10.1093/stmcls/sxab023

Single-Cell RNA Sequencing-Based Characterization of Resident Lung Mesenchymal Stromal Cells in Bronchopulmonary Dysplasia

Ivana Mižíková ^1,², Flore Lesage ^1,³, Chanele Cyr-Depauw ^1,³, David P Cook ^4,⁵, Maria Hurskainen ^1,^3,^6,⁷, Satu M Hänninen ⁸, Arul Vadivel ¹, Pauline Bardin ^1,³, Shumei Zhong ¹, Olli Carpén ⁸, Barbara C Vanderhyden ^3,^4,⁹, Bernard Thébaud ^1,^3,^10,^✉

PMCID: PMC9199848 PMID: 35445270

Abstract

Late lung development is a period of alveolar and microvascular formation, which is pivotal in ensuring sufficient and effective gas exchange. Defects in late lung development manifest in premature infants as a chronic lung disease named bronchopulmonary dysplasia (BPD). Numerous studies demonstrated the therapeutic properties of exogenous bone marrow and umbilical cord-derived mesenchymal stromal cells (MSCs) in experimental BPD. However, very little is known regarding the regenerative capacity of resident lung MSCs (L-MSCs) during normal development and in BPD. In this study we aimed to characterize the L-MSC population in homeostasis and upon injury. We used single-cell RNA sequencing (scRNA-seq) to profile in situ Ly6a⁺ L-MSCs in the lungs of normal and O₂-exposed neonatal mice (a well-established model to mimic BPD) at 3 developmental timepoints (postnatal days 3, 7, and 14). Hyperoxia exposure increased the number and altered the expression profile of L-MSCs, particularly by increasing the expression of multiple pro-inflammatory, pro-fibrotic, and anti-angiogenic genes. In order to identify potential changes induced in the L-MSCs transcriptome by storage and culture, we profiled 15 000 Ly6a⁺ L-MSCs after in vitro culture. We observed great differences in expression profiles of in situ and cultured L-MSCs, particularly those derived from healthy lungs. Additionally, we have identified the location of Ly6a⁺/Col14a1⁺ L-MSCs in the developing lung and propose Serpinf1 as a novel, culture-stable marker of L-MSCs. Finally, cell communication analysis suggests inflammatory signals from immune and endothelial cells as main drivers of hyperoxia-induced changes in L-MSCs transcriptome.

Keywords: lung MSC, bronchopulmonary dysplasia, lung development, scRNA-seq

Graphical Abstract

Significance Statement.

Characterizing the L-MSC population in normally developing lung and upon injury is pivotal in understanding if resident L-MSCs fail to prevent neonatal lung injury and whether they are feasible candidates for cell therapy. Using scRNA-seq, we provide a first detailed report of the characteristics and behavior of L-MSC in situ and in vitro, during both, normal development and oxygen-induced injury on single-cell level and propose novel L-MSC markers. Additionally, we provide proof that the L-MSC transcriptome is broadly affected by culture, which needs to be considered prior to potential therapeutic use of these cells.

Introduction

Late lung development represents an important period in lung maturation marked by an exponential increase in the gas exchange surface area by forming the most distal respiratory units, the alveoli. Within these units, respiration takes place across a thin (0.2-2 μm) alveolo-capillary barrier. Formation of alveolar structures, a process known as alveolarization, is facilitated by spatially and temporarily coordinated interactions between diverse cell types and the pulmonary microenvironment.¹ Defects in late lung development in humans manifest as bronchopulmonary dysplasia (BPD), a multifactorial disease occurring as a consequence of premature birth, respiratory distress, and associated treatments in neonatal intensive care. BPD is the most common chronic disease in children and a leading cause of death in children under the age of 5.^1,2 BPD is also associated with neurodevelopmental delay, increased incidence of asthma, re-hospitalizations and early-onset emphysema.^3,4

To date, multiple studies have demonstrated the lung protective effects of exogenous, bone marrow (BM)- or umbilical cord (UC)-derived, mesenchymal stromal cells (MSCs) in experimental BPD models.^5-10 The discovery of lung resident (L-)MSCs prompted questions regarding the apparent insufficient regenerative capacity of L-MSCs in lung injury.¹¹ Characterizing the L-MSC population in homeostasis and upon injury is pivotal in understanding the apparent contradiction between the therapeutic effects of exogenous MSCs, while the resident population fails to prevent neonatal lung injury from occurring. However, very little is currently known about the role of L-MSCs in postnatal lung development and in BPD. Lung stromal cells, including lipofibroblasts, myofibroblasts, and matrix fibroblasts are a potent source of inter-cellular signaling and are known to play an important role in BPD pathogenesis.¹² However, how L-MSCs communicate with other cell populations and contribute to the development of BPD remains unknown.

While most authors report that L-MSCs can differentiate, to some extent, into chondroblasts, osteoblasts, and adipocytes,¹³ form colonies in vitro,^13,14 and express classical MSC markers THY1 (CD90), NT5E (CD73), and ENG (CD105),^13,15 no L-MSC-specific marker has yet been established. Due to the lack of standardization for L-MSC identification, as well as differences in expression profiles between species, no single marker has been broadly accepted. Lung mesenchymal progenitor cell markers have been proposed,^13,15-18 including LY6A, often referred to as SCA-1 (stem cell antigen 1).^16-19 LY6A was proposed as a defining progenitor marker for mesenchymal cell lineages in the lung¹⁹ and LY6A⁺ mesenchymal lung cells were shown to promote colony formation, proliferation and differentiation of epithelial progenitor cells.²⁰

In the study presented here we identify, for the first time, the transcriptome of Ly6a⁺ L-MSCs in healthy and diseased developing mouse lungs. We hypothesized, that O₂-exposure (a well-established model to mimic BPD) significantly impacts the phenotype and function of L-MSCs, as well as cellular communication between L-MSCs and other cell populations in the developing lung. We identify perturbations to the phenotype and functional properties of L-MSCs in this model. Furthermore, we report extensive single-cell RNA sequencing (scRNA-seq) profiling of L-MSCs in the lungs of 36 healthy and O₂-exposed mice at 3 developmental timepoints (P3, P7, and P14). Finally, we investigate cultured Ly6a⁺ L-MSCs and Ly6a^– mouse lung stromal cells by scRNA-seq. We identify changes in L-MSCs transcription profile induced by storage and culture and present novel, culture-stable marker for this rare progenitor population.

Materials and Methods

Experimental Animals

Pregnant C57BL/6N mice were purchased from Charles Rivers Laboratories, Saint Constant, QC, Canada at embryonic day (E)15. Mice were housed by the Animal Care and Veterinary Service of the University of Ottawa in accordance with institutional guidelines. All study protocols were approved by the animal ethics and research committee of the University of Ottawa (protocol OHRI-1696) and conducted according to guidelines from the Canadian Council on Animal Care (CCAC). Mouse pups born on the same day were randomized at the day of birth [postnatal day (P)0] and divided into equal-sized litters of 6-8 pups/cage. Cages were then maintained in either room air (normoxia, 21% O₂) or normobaric hyperoxia (85% O₂) until the day of harvest. The hyperoxic environment was maintained in sealed plexiglass chambers with continuous oxygen monitoring (BioSpherix, Redfield, NY). Mice were maintained in 12/12 hours light/dark cycle and received food ad libidum. In order to avoid confounding factors associated with oxygen toxicity, nursing dams were rotated between normoxic and hyperoxic group every 48 hours. Euthanasia was performed by an intraperitoneal (i.p.) injection of 10 μL/g Pentobarbital Sodium (CDMV, Saint-Hyacinthe, QC, Canada).

Lung Isolation

The detailed procedure is provided in Supplementary Method S1. A detailed flowchart illustrating the allocation of each mice to respective experimental groups is depicted in Supplementary Fig. S1.

Mean Linear Intercept Measurement

Paraffin-embedded tissue blocks were sectioned at 4 μm, stained with hematoxylin and eosin (H&E) stain, and scanned using the Axio Scan.Z1 (Zeiss, Oberkochen, Germany). The mean linear intercept (MLI) was estimated with Fiji/ImageJ software using a 64-point grid as described previously.²¹ A total of 20 randomly selected 500 µm × 500 µm fields of view were assessed in each lung.

Fluorescent Activated Cell Sorting

The number of cells in the single-cell suspension was estimated using the EVE NanoEnTek automatic cell counter and a total of 1 × 10⁶ cells/sample were resuspended in 550 μL of FACS buffer (5%, v/v, FBS and 1 mM EDTA in 1 × DPBS). Cells were then incubated at RT in the dark with 2 μL/1 × 10⁶ cells of CD16/32 antibody for 15 minutes. Following blocking, cells were centrifuged and pellets were resuspended in 1:100 mixture of panel of antibodies: FITC-CD31, AF647-CD45, Pe/Cy7-CD326, and BV421-LY-6A/E (Supplementary Supplementary Table S1). Cells were incubated with antibodies for 20 minutes in dark at RT, pelleted and washed 3× with FACS buffer. FACS was performed immediately using a MoFlo XDP (XDP, Beckman Coulter, Fullerton, CA, USA) and compensation and analysis was done using Summit v.5.4 at the Ottawa Hospital Research Institute (OHRI) StemCore facility.

Cell Culture and Storage

The detailed procedure is provided in Supplementary Method S2.

Colony Formation Assay

The detailed procedure is provided in Supplementary Method S3.

MSCs Surface Marker Profiling

Cultured, passage 3 CD31^–/CD45^–/EpCAM^–/LY6A⁺ L-MSCs were profiled for MSC surface markers by flow cytometry. Briefly, 3 × 10⁵ cells/sample were resuspended in 200 μL of FACS buffer in 96-well plate and incubated at RT in the dark with 2 μL/1 × 10⁶ cells of CD16/32 antibody for 15 minutes. Cells were then divided to 3 equal fractions, centrifuged and resuspended in one of the following 1:100 mixture of antibodies: (1) BV421-CD31, Pe/Cy7-conjugated CD326, PE-CD73, and AF488- D105; (2) AF647- CD45, BV421-LY-6A/E, PE-conjugated CD34, and AF488-CD146; (3 PB-CD90.2 (Supplementary Table S1). Cells were incubated with antibodies for 20 minutes in dark at RT, pelleted and washed 3× with FACS buffer. Flow cytometry was performed immediately using a MoFlo XDP (XDP, Beckman Coulter, Fullerton, CA, USA) and compensation and analysis was done using Summit v.5.4 at the OHRI core facility.

Osteogenic Differentiation

The detailed procedure is provided in Supplementary Method S4.

Adipogenic differentiation

The detailed procedure is provided in Supplementary Method S5.

Chondrogenic differentiation

The detailed procedure is provided in Supplementary Method S6.

Fluorescent in situ hybridization

The detailed procedure and a list of used probes are provided in Supplementary Method S7.

Multiplexing Samples for scRNA-seq

Multiplexing was performed according to the MULTI-seq protocol.²² The detailed procedure is provided in Supplementary Method S8.

scRNA-seq Library Preparation and Sequencing

Single-cell suspensions were processed using the 10× Genomics Single Cell 3ʹ v3 RNA-seq kit by Ottawa Hospital Research Institute Stem Core Laboratories. Gene expression libraries were prepared according to the manufacturer’s protocol. MULTI-seq barcode libraries were retrieved from the samples and libraries were prepared independently, as described previously.²² Final libraries were sequenced on the NextSeq500 platform (Illumina) to reach an approximate depth of 20 000-25 000 reads/cell.

scRNA-seq Data Processing, Analyses, and Quantification

The detailed procedure is provided in Supplementary Method S9.

Statistical Analysis

All statistical analyses were performed with GraphPad Prism 8.0. The presence of potential statistical outliers was determined by Grubbs’ test. Data are presented as means ± SD. Differences in case of 2-member groups were evaluated either by unpaired Student’s t-test or multiple unpaired Student’s t-test with correction for multiple comparisons using the Holm-Šidák method. P-values of ˂.05 were considered as  significant and depicted as following: *P ˂ .05'; **P ˂ .01; ***P ˂ .001; ****P ˂ .0001.

Results

The Developing Murine Lung Contains a Population of L-MSC Marked by the Expression of Ly6a

In order to understand the expression patterns unique to LY6A⁺ L-MSCs in the developing lung, we took advantage of a publicly available scRNA-seq dataset from newborn mice.²³ Within this dataset, we analyzed 7994 stromal cells from normoxia or hyperoxia-exposed developing mouse pups on postnatal days (P)3, 7, and 14, clustered into 6 distinct populations (Fig. 1A). Based on the expression pattern of commonly used MSC markers (Fig. 1B, Supplementary Fig. S2A), we selected Ly6a as most suitable marker to identify L-MSC in lung stroma. We then subsetted the Ly6a⁺ cells, belonging mostly to the Col14a1⁺ fibroblasts, as a separate, seventh cluster (Fig. 1B). Differential gene expression analysis revealed that Ly6a⁺ L-MSCs could be characterized by the expression of additional markers, including Lum, Serpinf1, or Dcn, with Lum being the single most unique identifier of the population (Fig. 1C, Supplementary Table S2). It was previously shown to inhibit migration, invasion, and tube-formation in BM-MSCs,²⁴ and was implicated in epithelial-mesenchymal transition and fibrocyte differentiation.²⁵ While Ly6a⁺ L-MSCs expressed additional MSC markers Mcam, Alcam, and Eng, their expression did not serve as a reliable indicator of Ly6a⁺ L-MSCs (Fig. 1D).

Figure 1. — Gene expression profile of *Ly6a*⁺ L-MSCs during late lung development. **(A)** Six clusters of stromal cells were previously identified in developing lungs. In the dataset re-analyzed here mice were exposed to room air (21% O₂) or hyperoxia (85%O₂) from P1 onward and lungs were harvested at P3, P7, and P14. UMAP plot depicting the distribution of lung stromal cells based on the developmental age and oxygen exposure. **(B)** UMAP plots showing the expression of *Ly6a* mRNA (left panel) within the lung stroma and new cluster identities, including the *Ly6a*⁺L-MSCs. **(C)** Heatmap of top 10 most differentially expressed genes across stromal clusters depicted in panel (B). **(D)** Dotplot depicting expression of routine MSC markers in lung stromal populations. **(E)** Relative contribution of *Ly6a*⁺ and *Ly6a*^– cells in developing lung stroma at P3, P7, and P14. n = 6 animals/group. Data are presented as means ± SD. Statistical analyses were performed with GraphPad Prism 8.0 and the presence of potential statistical outliers was determined by Grubbs’ test. Significance was evaluated by multiple unpaired Student’s t-test with Holm-Šidák correction for *Ly6a*⁺ and *Ly6a*^– cells separately. P-value of <.05 was considered significant and is depicted. **(F)** Dotplot depicting the expression of most differentially expressed genes in *Ly6a*⁺ L-MSCs during normal lung development. **(G)** Dotplot depicting the expression of genes that are differentially expressed specifically in *Ly6a*⁺ L-MSCs and not in other lung stromal clusters during normal lung development. **(H)** Selected developmental age-associated signaling pathways in the *Ly6a*⁺ L-MSC cluster identified by gene set enrichment analysis (GSEA). All terms are significantly enriched (adjusted P-value < .05). Normalized enrichment scores (NES) values were computed by gene set enrichment analysis on fold change-ranked genes. Expression values in Heatmap represent Z-score-transformed log(TP10k+1) values. Expression levels in Dotplots and UMAP plots are presented as log(TP10k+1) values. Log(TP10k+1) corresponds to log-transformed UMIs per 10k. Data depicted in the figure are derived from the same data set as data included in previously published study licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/): Hurskainen et al²³ . While all figures depict unique information, due to their origin, cells and clusters distribution depicted in panels A and B resembles data presented in the abovementioned study.

The Transcription Profile and Signaling Activity of Ly6a⁺ L-MSCs Change Significantly During Postnatal Lung Development

We first aimed to understand how the L-MSC population changes in the postnatal developing lung. While the size of the population remained unchanged between P3 and P7, the second week of lung development in healthy mice was associated with an increase in the size of the Ly6a⁺ stromal population (Fig. 1E, Supplementary Table S3). Similarly, differential state analysis (DSA) in normally developing lungs revealed that the gene expression patterns in L-MSCs changed more predominantly between P7 and P14, as evidenced by both, changes in relative expression and proportion of expressing cells (Fig. 1F, Supplementary Table S4). Although the expression of genes such as Apoe, Inmt, Klf9, and Abca1 was drastically increased in L-MSCs, these genes were also considerably upregulated in Ly6a^– stromal cells (Supplementary Table S4). The largest L-MSC-specific expression changes were observed for Mmp3, C1s1, Podn, Dlk1, and Agtr2 (Fig. 1G). Gene set enrichment analysis (GSEA) identified extracellular matrix (ECM) formation, vascular development, and wound healing among the activated pathways (Fig. 1H, Supplementary Table S5).

Next, to further understand how L-MSCs send and receive signals during postnatal development, we performed a cell communication analysis. We inferred developmental age-induced cellular communications between Ly6a⁺ L-MSCs and other lung populations using the NicheNet tool^23,26 (Fig. 2A, Supplementary Figs. S3-S5 and Table S6). During development L-MSCs received signals from several cell populations, including endothelial cells, interstitial macrophages (Int Mf), alveolar epithelial type 2 (AT2) cells, and stromal cells  (Fig. 2A). Col4a1, Fat1, Hmgb2, Vcam1, and Hc were identified as most potent ligands, targeting numerous downstream genes in the developing L-MSCs, including Klf9, Top2a, and other strongly de-regulated genes (Figs. 1F and 2A, B, Supplementary Table S4). Furthermore, L-MSCs produced numerous ligands, targeting most lung cell populations, including itself (Fig. 2A). Among the most broadly acting ligands were Agt, App, and Apoe (Fig. 2C). Functional enrichment analysis (FEA) revealed that the expression of the L-MSC-produced ligands was associated with pathways related to angiogenesis, cell migration, adhesion and chemotaxis, and ECM organization (Supplementary Fig. S2B and Table S7).

Figure 2. — Age-associated gene expression and signaling in the developing *Ly6a*⁺ L-MSCs. **(A)** Circos plot showing inferred cell communications between *Ly6a*⁺ L-MSCs and other populations in the developing mouse lung. Cell communications associated with increasing developmental age are depicted. Cell types in the top right correspond to receiver populations with the largest expression changes in response to increasing age. These cell types are connected to the sender cell types expressing ligands predicted to promote this response. Ligands expressed by the same cell population are colored the same. **(B)** Heatmap depicting predicted target genes for ligands most likely to be received by normally developing *Ly6a*⁺ L-MSC population as indicated in (A). The intensity of expression is indicated as specified by the color legend. **(C)** Heatmap depicting predicted target genes for ligands sent by *Ly6a*⁺ L-MSC population in normally developing lungs as indicated in (A). The intensity of expression is indicated as specified by the color legend.

The Transcription Profile and Signaling Activity of Ly6a⁺ L-MSC Change Significantly During Postnatal Lung Development in Response to Hyperoxia

Hyperoxia induced an increase in proportion of Ly6a⁺ stromal cells as determined by scRNA-seq analysis at P14 (Fig. 1E, Supplementary Table S3). This was consistent with increased proportion of LY6A⁺ stromal cells in hyperoxia-exposed lungs at P7 as measured by flow cytometry (Supplementary Fig. S2C. D). In order to identify hyperoxia-induced changes in gene expression specific to Ly6a⁺ L-MSCs, we performed a DSA for both, Ly6a⁺ L-MSC population and non-progenitor Ly6a^– stromal cells (Supplementary Table S8). Hyperoxia-induced expression changes most distinctive of Ly6a⁺ L-MSCs are illustrated in Fig. 3A. Exposure to hyperoxia was associated with Ly6a⁺ L-MSCs-specific increase in expression of multiple pro-inflammatory (Cxcl1, Ccl2), as well as pro-fibrotic and anti-angiogenic (Timp1, Serpina3n) genes (Fig. 3A,  Supplementary Table S8). Gene set enrichment analysis of hyperoxia-induced changes in gene expression revealed an activation in inflammatory pathways, as well as decrease in pathways associated with arterial development and morphogenesis (Fig. 3B, Supplementary Table S9). When inspecting pathways altered by hyperoxia exclusively in Ly6a⁺, but not Ly6a^– stromal cells, activation of cytokine and chemokine signaling, cell cycle regulation, and senescence were most noticeable (Supplementary Fig. S2G and Table S9).

Figure 3. — Hyperoxia-induced gene expression and signaling in the developing *Ly6a*⁺ L-MSCs. **(A)** Dotplot depicting the expression of markers specifically altered by hyperoxia exposure in *Ly6a*⁺ and *Ly6a*^– cells in the developing mouse lung. **(B)** Selected hyperoxia-regulated signaling pathways in the *Ly6a*⁺ L-MSC cluster identified by gene set enrichment analysis (GSEA). All terms are significantly enriched (adjusted P-value <.05). Normalized enrichment scores (NES) values were computed by gene set enrichment analysis on fold change-ranked genes. **(C)** Circos plot showing inferred cell communications between *Ly6a*⁺ L-MSCs and other populations in the developing mouse lung. Cell communications induced by exposure to hyperoxia are depicted. Cell types in the top right correspond to receiver populations with the largest expression changes in response to hyperoxia. These cell types are connected to the sender cell types expressing ligands predicted to promote this response. Ligands expressed by the same cell population are colored the same. **(D)** Heatmap depicting predicted target genes for ligands most likely to be received by *Ly6a*⁺ L-MSC population in hyperoxia as indicated in **(C)**. The intensity of expression is indicated as specified by the color legend. **(E)** Heatmap depicting predicted target genes for ligands sent by *Ly6a*⁺ L-MSC population in hyperoxia as indicated in **(C)**. The intensity of expression is indicated as specified by the color legend.

To further understand the fate of Ly6a⁺ L-MSCs in hyperoxia-induced injury, we performed a cell communication analysis using the NicheNet tool, inferring hyperoxia-induced cellular communications^23,26 (Fig. 3C, Supplementary Figs. S6 and S7, Table S10). Ly6a⁺ L-MSCs in hyperoxia-exposed lungs received signals from several cell populations, including immune cells, capillary and arterial endothelial cells, mesothelial cells and Col13a1⁺ fibroblasts (Fig. 3C). Further, we inferred genes in Ly6a⁺ L-MSCs most likely to be targeted by the received signals (Fig. 3D). Multiple ligands, such as Apoe, Il1a, Ifng, and Mmp9 were predicted to target the expression of pro-inflammatory, pro-fibrotic and anti-angiogenic genes discussed above, including Timp1, Cxcl1, and Icam1 (Fig. 3D). Expression of these target genes was elevated in Ly6a⁺ L-MSCs by hyperoxia exposure (Fig. 3A). Finally, ligands produced by Ly6a⁺ L-MSCs affected multiple cell populations, including alveolar macrophages, ciliated and AT2 cells, capillary and vein endothelium and other stromal populations. Among the most broadly acting ligands produced by Ly6a⁺ L-MSCs were Bmp4, Bmp5, Col4a1, and Tnc (Fig. 3A). Inferred target genes in receiving cells targeted by majority of these ligands included Ccnd1, Cdkn1a, Icam1, and Hmox1 (Fig. 3E). According to FEA, expression of the L-MSC-produced ligands was associated with pathways related to vessel morphogenesis, epithelial cell proliferation, cell chemotaxis, and immune homeostasis and response (Supplementary Fig. S2H and Table S11).

Murine Ly6a⁺/Col14a1⁺ L-MSCs Localize to Perivascular Regions of the Developing Lung

Next, we aimed to localize the Ly6a⁺ L-MSCs in the developing lung using FISH. L-MSCs were identified as Ly6a⁺/Col14a1⁺ cells. Double-positive L-MSCs in both, normally and aberrantly developing lungs localized predominantly to perivascular regions of large vessels with more double-positive cells observed in hyperoxia-exposed lungs (Fig. 4A).

Figure 4. — Identification of *Ly6a*⁺ L-MSCs in the developing lung. **(A)** Fluorescent RNA in situ hybridization showing localization of L-MSCs identified by the co-expression of *Ly6a* (white) and *Col14a1* (green) in lungs of room air (21% O₂) or hyperoxia (85%O₂)-exposed developing mice. Scale bar = 200 µm for low-magnification (×5, top panels) windows, 50 µm for higher-magnification (×40, bottom left panels) windows, and 20 µm for high-magnification (×63, bottom right panels) windows. Four 14-day-old animals/group were analyzed. Expression levels in Dotplot are presented as log(TP10k+1) values. Log(TP10k+1) corresponds to log-transformed UMIs per 10k. **(B)** Fluorescent RNA in situ hybridization showing co-expression of *Ly6a* (white), *Col14a1* (green), and *Timp1* (pink) in lungs of room air (21% O₂) or hyperoxia (85%O₂)-exposed developing mice. Scale bar = 200 µm for low-magnification (×5, top left) windows, 50 µm for higher-magnification (×40, top right) windows, and 20 µm for high-magnification (×63, bottom panels) windows. Four 14-day-old animals/group were analyzed. **(C)** Fluorescent RNA in situ hybridization showing co-expression of *Ly6a* (white), *Col14a1* (green), and *Serpina3n* (pink) in lungs of room air (21% O₂) or hyperoxia (85%O₂)-exposed developing mice. Scale bar = 200 µm for low-magnification (×5, top left) windows, 50 µm for higher-magnification (×40, top right) windows, and 20 µm for high-magnification (×63, bottom panels) windows. Four 14-day-old animals/group were analyzed.

Additionally, we aimed to validate some of the novel normoxic and hyperoxic L-MSC markers as suggested by scRNA-seq analysis (Figs. 1C and 3A). Ly6a⁺ L-MSCs were co-stained for the hyperoxia-associated markers Timp1 and Serpina3n (Fig. 4B and C, respectively). In both instances triple-positive cells were observed in the regions adjacent to large vessels (highlighted by white squares in low-magnification panels). These cells were not only more abundant in the lungs from BPD mice, but the expression levels of both Timp1 and Seprina3n were increased in the diseased lungs (see higher-magnification panels Fig. 4B, C).

Hyperoxia Exposure Does Not Impact Clonal or Differentiation Potential of LY6A⁺ L-MSCs

In order to verify their progenitor cell-like properties, we isolated and studied LY6A⁺ L-MSCs from healthy and hyperoxia-exposed developing mouse pups. An arrest in lung development was induced by exposing newborn mouse pups to normobaric hyperoxia (85% O₂; Fig. 5A). CD31^–/CD45^–/EpCAM^–/LY6A⁺ L-MSCs were isolated from 7-day-old healthy (21% O₂-exposed) or diseased (85% O₂-exposed) mouse pups (Fig. 5B) and examined for the hallmarks of the MSC phenotype in vitro. While lungs of hyperoxia-exposed pups consistently yielded higher numbers of LY6A⁺ L-MSCs (Fig. 5B), no differences in the appearance (Fig. 5C), differentiation capacity (Fig. 5C), expression of surface markers (Fig. 5D), or clonal abilities (Fig. 5E) were observed between the cells isolated from healthy and diseased animals. LY6A⁺ L-MSCs isolated from both healthy and hyperoxia-exposed mice had a fibroblast-like appearance and expressed classical markers of MSCs in vitro (Fig. 5C, D). In order to investigate their differentiation capacity, LY6A⁺ L-MSCs were induced to differentiate along the osteogenic, chondrogenic, and adipogenic lineages. Both normoxia and hyperoxia-derived LY6A⁺ L-MSCs produced osteogenic and chondrogenic matrix  (Fig. 5C). However, only a single sample of normoxia-derived LY6A⁺ L-MSCs produced a small number of adipocytes, and no lipogenic differentiation was observed in hyperoxia-derived LY6A⁺ L-MSCs (Supplementary Fig. S2E). Postnatal hyperoxia exposure had no effect on colony-forming capacity of LY6A⁺ L-MSCs as assessed by single-cell plating colony-forming assay. Both normoxia and hyperoxia-derived LY6A⁺ L-MSCs produced colonies of various sizes. While larger colonies consisted of fibroblast-like spindle-shaped cells, smaller colonies were formed by cells with a large cytoplasm (Fig. 5E). Inconsistent differentiation capacity and colony formation might suggest a heterogeneous nature of cultured LY6A⁺ L-MSCs.

Figure 5. — Characterization of LY6A⁺ L-MSCs in normal and impaired mouse lung development. **(A)** Mouse pups were exposed to room air (21% O₂, gray) or hyperoxia (85%O₂, blue) from P1 onward. Mice were harvested on postnatal day (P)7. Representative histological sections from lungs developing in 21% O₂ or 85% O₂. Lung morphometry was quantified by the mean linear intercept (MLI) measurement. n = 7 animals/group. Scale bar = 100 µm. **(B)** LY6A⁺ L-MSCs were identified by flow cytometry as CD45-AF647^–/CD31-FITC^–/CD326(EPCAM)-PeCy7^–/LY6A(SCA1)-BV421⁺ cells and their proportion in lung homogenates was quantified. n = 8-9 animals/group. **(C)** Representative images of undifferentiated LY6A⁺ L-MSCs and LY6A⁺ L-MSCs differentiated toward osteogenic and chondrogenic lineages and stained with Alizarin Red S or Alcian blue, respectively. Experiments were performed in quadruplets. Scale bar = 250 µm or 1000 µm. **(D)** Expression of routine MSCs surface markers in cultured LY6A⁺ L-MSCs isolated from room air (21% O₂, gray bars) or hyperoxia-exposed (85%O₂, purple bars) developing pups as determined by flow cytometry. n = 3-4 animals/group. **(E)** Quantification and representative images of colony formation of cultured LY6A⁺ L-MSCs isolated from room air (21% O₂, gray bars) or hyperoxia-exposed (85%O₂, purple bars). n = 4 animals/group. Scale bar =100 µm. All data are presented as means ± SD. Statistical analyses were performed with GraphPad Prism 8.0. The presence of potential statistical outliers was determined by Grubbs’ test. Significance was evaluated by unpaired Student’s t test for analysis in panels (A) and (B), and by multiple unpaired Student’s t-test with Holm-Šidák correction in panels (D) and (E). P-values of <.05 were considered significant and are depicted.

Cell Culture Alters the Gene Expression Profile of LY6A⁺ L-MSCs

For therapeutic applications, MSCs are typically culture expanded, then frozen, over the short, or long term and thawed prior to administration. These various steps may alter the properties of the cell product. In order to understand changes in the L-MSCs expression profile induced by storage and culture, we performed a scRNA-seq analysis of cultured LY6A⁺ and LY6A^– lung stromal cells isolated from 7-day-old healthy (21% O₂-exposed) or diseased (85% O₂-exposed) mouse pups (Fig. 6A, Supplementary Fig. S2C-F). We sequenced over 15 000 cultured CD31^–/CD45^–/EpCAM^–/LY6A^– and CD31^–/CD45^–/EpCAM^–/LY6A⁺ cells and identified 4 distinct clusters (Fig. 6A-C, Supplementary Tables S12 and S13). While normoxia and hyperoxia-derived LY6A^– stromal cells contributed to all 4 clusters, very few LY6A⁺ cells could be found in clusters 2 and 3 (Fig. 6B). The presence of distinct clusters within the L-MSCs population is consistent with the heterogeneous phenotype of cultured L-MSCs described above (Fig. 5E). In line with this finding, the highest levels of routine MSC markers, such as Thy1, Eng, Alcam, or Mcam, were found in the largest cluster 0, while very little expression was seen in the 2 smallest clusters (Fig. 6D). While still expressing routine MSC markers to some level, cluster 1 was characterized by its distinct expression of Cck, previously found to attenuate p53-mediated apoptosis in lung cancer²⁷ (Fig. 3A-C). Cluster 2 was distinguished by the expression of pro-adipogenic markers, such as Igfbp2 and Col4a1, as well as markers of myofibroblasts (Des) and alveolar epithelium (Krt8 and Prnp2) (Fig. 6C, Supplementary Table S13). Cluster 3 was characterized by the expression of multiple osteogenic markers, including Cryab, Postn, and Ngfr. Interestingly, the expression of both Postn, as well as another cluster 3 marker Col18a1, was previously reported in BPD patients and hyperoxia-exposed developing mice.^28,29

Figure 6. — Gene expression profile of cultured normoxia and hyperoxia-derived LY6A⁺ L-MSCs. **(A)** LY6A⁺ and LY6A^– stromal cells isolated from lungs of room air (21% O₂) or hyperoxia (85%O₂)-exposed developing mice were frozen, cultured, and sequenced at passage 3. n = 3 animals/group. scRNA-seq identified 4 clusters of cultured LY6A⁺ and LY6A^– stromal cells. **(B)** Relative distribution of room air (21% O₂) or hyperoxia (85%O₂)-derived LY6A⁺ and LY6A^– cells to the 4 different clusters. n = 3 animals/group. **(C)** Heatmap of top 10 most differentially expressed genes across clusters depicted in panel (A). **(D)** Violin plots depicting expression of routine MSC markers in cultured stromal populations. **(E)** Dotplot depicting expression of oxygen-specific markers in LY6A⁺ and LY6A^– cultured lung stromal cells. Expression values in Heatmap and violin plots represent Z-score-transformed log(TP10k+1) values. Expression levels in Dotplot and UMAP plot are presented as log(TP10k+1) values. Log(TP10k+1) corresponds to log-transformed UMIs per 10k.

Next, we aimed to identify the best markers for cultured L-MSCs (Supplementary Tables S14-S18). We compared the gene expression profiles of LY6A⁺ and LY6A^– stromal cells (Supplementary Tables S15-S18) and identified differentially expressed genes between normoxia- and hyperoxia-derived subsets of these populations (Supplementary Tables S17 and S18). In comparison to LY6A^– cells, LY6A⁺ cells were characterized by high expression of Actg2, Col1a2, Serpinf1, Prrx1, and Lxn, and by low expression of smooth muscle cell (SMC) marker Tagln2,³⁰ alveolar progenitor marker Tm4sf1,³¹ and Prdx6 (Fig. 6E, Supplementary Tables S15 and S16). From these markers hyperoxia exposure further specifically increased the expression of Actg2, and decreased the expression of Tagln2, Tm4sf1, and Prdx6 in LY6A⁺ cells. Hyperoxic LY6A⁺ cells were additionally distinguished by expression of Ptn, Adamts5, Rbp1, and Col3a1 (Fig. 6E, Supplementary Table S17). Expression of Prrx1 and Serpinf1 is known to favor an osteogenic phenotype, and Serpinf1 is known to inhibit adipogenesis.^32,33

In order to identify L-MSCs expression patterns maintained after cell culture and storage, we next compared expression of the most promising markers of Ly6a⁺ L-MSCs in both, in situ and in vitro datasets from cells isolated at P7 (Supplementary Fig. S2I, J). This analysis revealed that a large portion of the expression profile characteristic for Ly6a⁺ L-MSCs in situ (Supplementary Fig. S2I,J) is lost when cells are frozen and cultured, including the expression of promising markers, such as Lum, Ptn, Dcn, or Pi16 (Supplementary Fig. S2H). Furthermore, while the expression pattern of some markers, such as Serpina3n or C3 persisted in cultured cells, the portion of the cells expressing the gene was diminished (Fig. 7A, B). The most suitable in situ or in vitro-specific identifying markers of Ly6a⁺ L-MSCs are depicted in Fig. 4A, B. Among the most stable markers of Ly6a⁺ L-MSCs, resistant to changes induced by culture, were Serpinf1 and Postn (Fig. 7A,B, Supplementary Fig. S2I, J). In order to confirm the viability of Serpinf1 as potential novel marker for L-MSCs, we performed FISH in developing lungs at P14. Triple-positive cells could be found in lungs of both, normoxic and hyperoxic mice (Fig. 7C). No differences were apparent in Serpinf1 expression intensity between the 2 groups.

Figure 7. — Identification of novel markers for in situ and cultured *Ly6a*⁺ L-MSCs. **(A)** Identifying markers were first established in the in situ, or cultured *Ly6a*⁺ and *Ly6a*^– lung stromal cells based on Supplementary Fig. 1H, I. Dotplot depicts the expression levels of those markers, most suitable for identification of *Ly6a*⁺ and *Ly6a*^– lung stromal cells in situ in normoxic or hyperoxic animals at P7. (B) Identifying markers were first established in the in situ, or cultured *Ly6a*⁺ and *Ly6a*^– lung stromal cells based on Supplementary Fig. 1H, I. Dotplot depicts the expression levels of those markers, most suitable for identification of normoxia-derived and hyperoxia-derived LY6A⁺ and LY6A^– lung stromal cells in culture. **(C)** Fluorescent RNA in situ hybridization showing co-expression of *Ly6a* (white), *Col14a1* (green), and *Serpinf1* (pink) in lungs of room air (21% O₂) or hyperoxia (85%O₂)-exposed developing mice. Scale bar = 200 µm for low-magnification (×5, top left) windows, 50 µm for higher-magnification (×40, top right) windows, and 20 µm for high-magnification (×63, bottom panels) windows. Four 14-day-old animals/group were analyzed.

Finally, new expression patterns arose particularly in hyperoxia-derived Ly6a⁺ L-MSCs after cell culture. While a high Ptn, Lum, Dcn, Col3a2, and Col14a1 expression was initially characteristic of both, hyperoxia and normoxia-derived Ly6a⁺ L-MSCs, in cultured L-MSCs this was true only for the hyperoxia-derived Ly6a⁺ L-MSCs (Supplementary Fig. S2I, J). This expression pattern denotes that not only does the L-MSC transcriptome change in culture but that the cells isolated from lungs of diseased mice tend to retain their expression profile and, potentially, progenitor-like nature longer.

Discussion

Our current knowledge regarding the identity and properties of tissue resident MSCs remains limited. Most studies analyse L-MSCs in culture after isolation with one, or several MSC markers. However, no explicit rules regarding which markers represent the L-MSC population the best exist to date. The progenitor-like characteristics of these cells have been established in culture,^13,14 but it is not yet known why L-MSCs fail to prevent the lung injury or restore damage in the lung. While L-MSCs were previously found in bronchoalveolar lavage of BPD patients,³⁴ it is not known whether this is due to increased apoptosis and subsequent shedding from the lung or is a sign of activation and proliferation of L-MSC and hence of increased numbers in BPD patients. Here, we provide an extensive scRNA-seq--based analysis of L-MSCs in healthy and O₂-exposed developing mouse lung, as well as in culture.

The use of omics approaches to study tissue-specific MSCs in vivo has been previously proposed.³⁵ In the study presented here, we utilize scRNA-seq to study L-MSCs immediately after isolation (in situ) without confounding procedures, such as FACS, cell culture and storage, and hence preserve the in vivo activation status of the different lung populations as much as possible. We used a publicly available scRNA-seq dataset from newborn mice,²³ which contains 7994 stromal cells from normoxia or hyperoxia-exposed developing mice clustered into 6 populations. More in-depth description of the 6 stromal clusters was published by the authors previously.²³ Additionally, distinction between Col13a1⁺ and Col14a1⁺ fibroblasts has been previously reported in multiple species.³⁶Col14a1⁺ fibroblasts co-expressing high levels of Dcn and Col1a1 also correspond to interstitial fibroblasts, while Col13a1⁺ fibroblasts co-expressing Inmt can also be labeled as lipofibroblasts, as identified in scRNA-seq studies by others.³⁷Col14a1/Dcn/Col1a1 expressing cells are also referred to as matrix-producing fibroblasts by some authors,³⁸ with Mfap4³⁹ and Mfap5^23,40 as additional proposed markers of ECM-producing cells. Distinct clusters of SMC, fibromyocytes, and myofibroblasts with similar gene expression profile had also been previously reported, with Acta2 and Cnn1 serving as markers of SMC and fibromyocytes and Acta2/Tgfbi expression pattern defining pulmonary myofibroblasts.^{23,36,37,40-42}

We selected Ly6a to identify L-MSCs for 2 reasons: (1) Ly6a is one of the most commonly used L-MSC markers and its expression has been shown in specific progenitor-like populations, (2) Ly6a was the only known MSC marker forming a visible subcluster within the lung mesenchyme of early postnatal mouse pups. We identified novel markers of L-MSCs, including Lum, Serpinf1, and Dcn. Next, we showed how the L-MSC’s transcriptome changes during the course of normal lung development and in hyperoxia, and explored the communication between L-MSCs and other lung cell populations.

Hyperoxia-exposure, used as a model for BPD, was associated in Ly6a⁺ L-MSCs with increased expression of pro-inflammatory, pro-fibrotic, and anti-angiogenic genes, including Timp1, Cxcl1, and Ccl2. Increased expression of both Timp1 and Ccl2 was previously reported in hyperoxia-exposed rodents,^43,44 and in plasma⁴⁵ or tracheal aspirates (TA)⁴⁶ of BPD patients. Timp1 expression was further increased in fibrotic foci in chronic BPD⁴⁷ and in the lungs of ventilated newborns.⁴⁸ Cell communication inference analysis revealed, that upon hyperoxia exposure, L-MSCs received ligands primarily from immune and endothelial cells, including Il1a, Mmp9, Ifng, and Fasl. Interestingly, many of received ligands were predicted to target the expression of pro-inflammatory, pro-fibrotic, and anti-angiogenic genes increased in hyperoxia-exposed L-MSCs. IFNγ and MMP9, which target the expression of both Timp1 and Cxcl1, were previously implicated in development of alveolar hypoplasia⁴⁹ and an increased expression of IFNγ was reported in TA of BPD patients.^34,50 Development of BPD was also associated with increased TA and plasma protein levels of ICAM1.^51,52 Furthermore, IL1A was shown to induce an inflammatory phenotype in lung fibroblasts.⁵³ Additionally, Fasl⁺ immune cells were shown to induce fibroblast cell death,^54,55 and its overexpression was associated with alveolar apoptosis and disturbed alveolar and vascular development.⁵⁶

Next, we investigated how the L-MSCs’ transcriptome changed due to culture and storage, both necessary steps for the preparation of a cell therapeutic product. ScRNA-seq analysis revealed, that following isolation, storage, and culture, most L-MSCs retain the expression of MSC markers, including Ly6a⁺. Cultured L-MSCs showed moderate ability to differentiate into chondrocytes and osteoblasts. However, we observed only one instance of successful differentiation along the adipogenic lineage, consistent with previous studies of L-MSCs in developing rats.¹³ Inconsistent differentiation capacity could be attributed to native or acquired heterogeneity within the cultured L-MSC population as indicated by the variable size and morphology of L-MSC-derived colonies. Importantly, such heterogeneity could indicate the existence of L-MSCs with varying progenitor-like capabilities, most likely impacting their therapeutic efficacy. Further, more detailed characterization of different L-MSCs subpopulations might be necessary in order to prepare a superior therapeutic product. ScRNA-seq revealed considerable changes in the transcriptome of L-MSCs in culture, implying that the cells studied and maintained in vitro for the purposes of therapeutic interventions are appreciably altered compared to L-MSCs in situ. Interestingly, we observed that the culture-induced transcription changes are less pronounced in L-MSCs derived from hyperoxia-exposed animals. This might suggest that hyperoxia primes L-MSCs to maintain certain characteristics, potentially in an attempt to trigger a repair mechanism. While the organism’s own resident L-MSCs fail to prevent the hyperoxia-induced lung damage, a therapeutic use of injury-primed L-MSCs might be more beneficial than L-MSCs from healthy individuals. Interestingly, tissue origin and microenvironment were shown to significantly impact the behavior and therapeutic efficacy of MSCs.^57,58 Moreover, conditioned media from BM-MSCs exposed ex vivo to hyperoxia exhibited superior therapeutic effects in the hyperoxia-induced rat BPD model when compared with media from BM-MSCs which were not pre-conditioned.⁵⁹

The localization of L-MSCs in the developing lungs has not yet been described. Here, we localized the Ly6a⁺/Col14a1⁺ L-MSC cells in the perivascular regions of both, healthy and diseased developing lungs by FISH. The double-positive L-MSCs in the hyperoxia-exposed lungs co-expressed Timp1 and Serpina3e, confirming the results of scRNA-seq analysis. Finally, as Ly6a is not expressed in human tissues, we aimed to identify additional markers to label L-MSCs, both in situ and in vitro. Lum, identified as marker of L-MSCs in situ, is known to be produced by MSCs. Within the lung, its expression was localized to peripheral lung and vessel walls.⁶⁰ While scRNA-seq revealed Lum as a promising L-MSCs marker in situ, its expression in culture was not well preserved. In comparison, the expression of Serpinf1 was well preserved in vitro, with the expression slightly increased in hyperoxic cells. Interestingly, Serpinf1 expression was previously reported to be increased in hyperoxia-exposed newborn mice and Serpinf1^–/– animals were protected from hyperoxia-induced lung injury.⁶¹Serpinf1 is also known as an anti-angiogenic and anti-migratory marker associated with aging MSCs.^61,62 In situ, Serpinf1 colocalized well with Ly6a⁺/Col14a1⁺ cells in both healthy and diseased lungs, suggesting Serpinf1 as promising new marker for L-MSCs.

To our knowledge, this is the first report studying the characteristics and behavior of L-MSC in situ and in vitro, during both health and disease. We described in detail the transcriptome and cellular communication of this resident cell population in homeostasis and in oxygen-induced injury. In addition, we established several markers that can be used to identify L-MSC in vitro and in vivo, both in healthy and diseased lungs. Additional studies will be needed to further unravel the heterogeneity of this population, as well as their therapeutic capabilities.

Conclusion

In conclusion, we provide, for the first time, a detailed scRNA-seq profile of L-MSCs in both, normally developing lung, and in the oxygen-induced lung injury. We further unraveled the cellular communication of this rare cell population in homeostasis and disease in order to mechanistically understand its endogenous repair capabilities, as well as its potential use as an exogenous cell therapeutic product. Additionally, we identified broad changes in L-MSCs transcriptional profile induced by storage and culture, which was more pronounced in the cells derived from diseased animals and propose a novel, culture-stable marker Serpinf1. The knowledge of how cultured L-MSCs differ from their counterparts in situ has great importance in designing L-MSC-based cell therapies.

Supplementary Material

sxab023_suppl_Supplementary_Data_S1

Click here for additional data file.^{(119.9KB, pdf)}

sxab023_suppl_Supplementary_Data_S2

Click here for additional data file.^{(145.4KB, pdf)}

sxab023_suppl_Supplementary_Data_S3

Click here for additional data file.^{(57.1KB, pdf)}

sxab023_suppl_Supplementary_Figure_S1

Click here for additional data file.^{(398.9KB, pdf)}

sxab023_suppl_Supplementary_Figure_S2

Click here for additional data file.^{(23.6MB, pdf)}

sxab023_suppl_Supplementary_Figure_S3

Click here for additional data file.^{(2MB, pdf)}

sxab023_suppl_Supplementary_Figure_S4

Click here for additional data file.^{(2.3MB, pdf)}

sxab023_suppl_Supplementary_Figure_S5

Click here for additional data file.^{(1.3MB, pdf)}

sxab023_suppl_Supplementary_Figure_S6

Click here for additional data file.^{(2.1MB, pdf)}

sxab023_suppl_Supplementary_Figure_S7

Click here for additional data file.^{(1.2MB, pdf)}

sxab023_suppl_Supplementary_Tables

Click here for additional data file.^{(26.1MB, xlsx)}

Acknowledgments

The authors acknowledge Prof. Zev Gartner (University of California, San Francisco) for kindly providing barcodes for multiplex labeling.

Funding

This study was supported by the Canadian Institutes of Health Research (CIHR), the German Research Foundation (Deutsche Forschungsgemeinschaft, #MI 2505_1-1), the Molly Towel Perinatal Research Foundation Postdoctoral Fellowship, the Canadian Lung Association—Breathing as One, the Frederick Banting and Charles Best Doctoral Scholarship, the Finnish Foundation for Pediatric Research, and the Finnish Sigrid Juselius Foundation.

Conflict of Interest

The authors declare no potential conflicts of interest.

Author Contributions

I.M.: conception and design, collection and assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript. F.L., C.C.-D., D.P.C.: collection and assembly of data, data analysis and interpretation, manuscript writing, final approval of manuscript. M.H., S.M.H., A.V., S.Z.: collection of data, final approval of manuscript. P.B.: data analysis, final approval of manuscript. O.C., B.C.V.: financial support, administrative support, final approval of manuscript. B.T.: conception and design, financial support, administrative support, final approval of manuscript.

Data Availability

All scRNA sequencing data, including raw fastq sequencing files, gene expression matrices, and cell metadata generated in this study are deposited in the NCBI’s Gene Expression Omnibus (GEO) database and can be accessed via accession code GSE198532. Code used for the analysis of scRNA-seq data is available at the public GitHub repository at: https://github.com/imizikova/scRNAseq-LMSCs-in-BPD.

References

1. Warburton D. Overview of lung development in the newborn human Neonatology. 2017;111:398-401. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Thébaud B, Goss KN, Laughon M, et al. Bronchopulmonary dysplasia Nat Rev Dis Primers. 2019;5:78. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Wong PM, Lees AN, Louw J, et al. Emphysema in young adult survivors of moderate-to-severe bronchopulmonary dysplasia Eur Respir J. 2008;32:321-328. [DOI] [PubMed] [Google Scholar]
4. Fawke J, S L, Kirkby J, et al. Lung function and respiratory symptoms at 11 years in children born extremely preterm: The EPICure study Am J Respir Crit Care Med. 2010;182:237-245. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. O’Reilly M, Möbius MA, Vadivel A, et al. Late rescue therapy with cord-derived mesenchymal stromal cells for established lung injury  in experimental bronchopulmonary dysplasia Stem Cells Dev. 2020;29:364-371. [DOI] [PubMed] [Google Scholar]
6. van Haaften T, Byrne R, Bonnet S, et al. Airway delivery of mesenchymal stem cells prevents arrested alveolar growth in neonatal lung injury in rats Am J Respir Crit Care Med. 2009;180:1131-1142. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Sutsko RP, Young KC, Ribeiro A, et al. Long-term reparative effects of mesenchymal stem cell therapy following neonatal hyperoxia-induced lung injury Pediatr Res. 2013;73:46-53. [DOI] [PubMed] [Google Scholar]
8. Pierro M, Ionescu L, Montemurro T, et al. Short-term, long-term and paracrine effect of human umbilical cord-derived stem cells in lung injury prevention and repair in experimental bronchopulmonary dysplasia Thorax. 2013;68:475-484. [DOI] [PubMed] [Google Scholar]
9. Augustine S, Cheng W, Avey MT, et al. Are all stem cells equal? Systematic review, evidence map, and meta-analyses of preclinical stem cell-based therapies for bronchopulmonary dysplasia Stem Cells Transl Med. 2020;9:158-168. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Aslam M, Baveja R, Liang OD, et al. Bone marrow stromal cells attenuate lung injury in a murine model of neonatal chronic lung disease Am J Respir Crit Care Med. 2009;180:1122-1130. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Collins JJP, Thébaud B.. Progenitor cells of the distal lung and their potential role in neonatal lung disease: progenitor cells of the distal lung Birth Defects Res A Clin Mol Teratol. 2014;100:217-226. [DOI] [PubMed] [Google Scholar]
12. Ushakumary MG, Riccetti M, Perl AT.. Resident interstitial lung fibroblasts and their role in alveolar stem cell niche development, homeostasis, injury, and regeneration. Stem Cells Transl Med. 2021;10:1021-1032. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Collins JJP, Lithopoulos MA, dos Santos CC, et al. Impaired angiogenic supportive capacity and altered gene expression profile of resident CD146 ⁺ mesenchymal stromal cells isolated from hyperoxia-injured neonatal rat lungs Stem Cells Dev. 2018;27:1109-1124. [DOI] [PubMed] [Google Scholar]
14. Lama VN, Smith L, Badri L, et al. Evidence for tissue-resident mesenchymal stem cells in human adult lung from studies of transplanted allografts J Clin Invest. 2007;117:989-996. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Rolandsson Enes S, Andersson Sjöland A, Skog I, et al. MSC from fetal and adult lungs possess lung-specific properties compared to bone marrow-derived MSC Sci Rep. 2016;6:29160. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Deng M, Li J, Gan Y, et al. Changes in the number of CD31^−;CD45^−;Sca-1⁺ cells and Shh signaling pathway involvement in the lungs of mice with emphysema and relevant effects of acute adenovirus infection. COPD. 2017;12:861-872. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Gong X, Sun Z, Cui D, et al. Isolation and characterization of lung resident mesenchymal stem cells capable of differentiating into alveolar epithelial type II cells: Isolation and characterization of LR-MSCs Cell Biol Int. 2014;38:405-411. [DOI] [PubMed] [Google Scholar]
18. Shi C, Cao X, Chen X, et al. Intracellular surface-enhanced Raman scattering probes based on TAT peptide-conjugated Au nanostars for distinguishing the differentiation of lung resident mesenchymal stem cells Biomaterials. 2015;58:10-25. [DOI] [PubMed] [Google Scholar]
19. McQualter JL, Brouard N, Williams B, et al. Endogenous fibroblastic progenitor cells in the adult mouse lung are highly enriched in the Sca-1 positive cell fraction Stem Cells. 2009;27:623-633. [DOI] [PubMed] [Google Scholar]
20. McQualter JL, Yuen K, Williams B, et al. Evidence of an epithelial stem/progenitor cell hierarchy in the adult mouse lung Proc Natl Acad Sci USA. 2010;107:1414-1419. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Salaets T, Gie A, Jimenez J, et al. Local pulmonary drug delivery in the preterm rabbit: feasibility and efficacy of daily intratracheal injections. Am J Physiol Lung Cell Mol Physiol. 2019;316:L589-L597. [DOI] [PubMed] [Google Scholar]
22. McGinnis CS, Patterson DM, Winkler J, et al. MULTI-seq: sample multiplexing for single-cell RNA sequencing using lipid-tagged indices. Nat Methods. 2019;16:619-626. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Hurskainen M, Mižíková I, Cook DP, et al. Single cell transcriptomic analysis of murine lung development on hyperoxia-induced damage. Nat Commun. 2021;12:1565. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. Malinowski M, Pietraszek K, Perreau C, et al. Effect of lumican on the migration of human mesenchymal stem cells and endothelial progenitor cells: involvement of matrix metalloproteinase-14. PLoS One. 2012;7:e50709. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Li L-F, Chu P-H, Hung C-Y, et al. Lumican regulates ventilation-induced epithelial-mesenchymal transition through extracelluar signal-regulated kinase pathway. Chest. 2013;143:1252-1260. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Browaeys R, Saelens W, Saeys Y.. NicheNet: modeling intercellular communication by linking ligands to target genes. Nat Methods. 2020;17:159-162. [DOI] [PubMed] [Google Scholar]
27. Han Y, Su C, Yu D, et al. Cholecystokinin attenuates radiation-induced lung cancer cell apoptosis by modulating p53 gene transcription. Am J Transl Res. 2017;9:638-646. [PMC free article] [PubMed] [Google Scholar]
28. Bozyk PD, Bentley JK, Popova AP, et al. Neonatal periostin knockout mice are protected from hyperoxia-induced alveolar simplication. PLoS One. 2012;7:e31336. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Mohamed WAW, Niyazy WH, Mahfouz AA.. Angiopoietin-1 and endostatin levels in cord plasma predict the development of bronchopulmonary dysplasia in preterm infants. J Trop Pediatr. 2011;57:385-388. [DOI] [PubMed] [Google Scholar]
30. Shudo Y, Cohen JE, Goldstone AB, et al. Isolation and trans-differentiation of mesenchymal stromal cells into smooth muscle cells: utility and applicability for cell-sheet engineering. Cytotherapy. 2016;18:510-517. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Zacharias WJ, Frank DB, Zepp JA, et al. Regeneration of the lung alveolus by an evolutionarily conserved epithelial progenitor. Nature. 2018;555:251-255. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Yu S, Li P, Li B, et al. RelA promotes proliferation but inhibits osteogenic and chondrogenic differentiation of mesenchymal stem cells. FEBS Lett. 2020;594:1368-1378. [DOI] [PubMed] [Google Scholar]
33. Gattu AK, Swenson ES, Iwakiri Y, et al. Determination of mesenchymal stem cell fate by pigment epithelium-derived factor (PEDF) results in increased adiposity and reduced bone mineral content. FASEB J. 2013;27:4384-4394. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Hikino S, Ohga S, Kinjo T, et al. Tracheal aspirate gene expression in preterm newborns and development of bronchopulmonary dysplasia: TAF cell gene expression predicting BPD. Pediatr Int. 2012;54:208-214. [DOI] [PubMed] [Google Scholar]
35. Sipp D, Robey PG, Turner L.. Clear up this stem-cell mess. Nature. 2018;561:455-457. [DOI] [PubMed] [Google Scholar]
36. Raredon MSB, Adams TS, Suhail Y, et al. Single-cell connectomic analysis of adult mammalian lungs. Sci Adv. 2019;5:eaaw3851. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Angelidis I, Simon LM, Fernandez IE, et al. An atlas of the aging lung mapped by single cell transcriptomics and deep tissue proteomics. Nat Commun. 2019;10:963. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Li R, Bernau K, Sandbo N, et al. Pdgfra marks a cellular lineage with distinct contributions to myofibroblasts in lung maturation and injury response. ELife. 2018;7:e36865. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Cohen M, Giladi A, Gorki A-D, et al. Lung single-cell signaling interaction map reveals basophil role in macrophage imprinting. Cell. 2018;175:1031-1044.e18. [DOI] [PubMed] [Google Scholar]
40. Guo M, Du Y, Gokey JJ, et al. Single cell RNA analysis identifies cellular heterogeneity and adaptive responses of the lung at birth. Nat Commun. 2019;10:37. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Travaglini KJ, Nabhan AN, Penland L, et al. A molecular cell atlas of the human lung from single-cell RNA sequencing. Nature. 2020;587:619-625. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Vila Ellis L, Cain MP, Hutchison V, et al. Epithelial vegfa specifies a distinct endothelial population in the mouse lung. Dev Cell. 2020;52:617-630.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
43. Hosford GE, Fang X, Olson DM.. Hyperoxia decreases matrix metalloproteinase-9 and increases tissue inhibitor of matrix metalloproteinase-1 protein in the newborn rat lung: association with arrested alveolarization. Pediatr Res. 2004;56:26-34. [DOI] [PubMed] [Google Scholar]
44. Kumar VHS, Lakshminrusimha S, Kishkurno S, et al. Neonatal hyperoxia increases airway reactivity and inflammation in adult mice: neonatal hyperoxia and AHR in mice. Pediatr Pulmonol. 2016;51:1131-1141. [DOI] [PubMed] [Google Scholar]
45. Schulz CG, Sawicki G, Lemke RP, et al. MMP-2 and MMP-9 and their tissue inhibitors in the plasma of preterm and term neonates. Pediatr Res. 2004;55:794-801. [DOI] [PubMed] [Google Scholar]
46. Fulton CT, Cui TX, Goldsmith AM, et al. Gene expression signatures point to a male sex-specific lung mesenchymal cell PDGF receptor signaling defect in infants developing bronchopulmonary dysplasia. Sci Rep. 2018;8:17070. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Dik WA, de Krijger RR, Bonekamp L, et al. Localization and potential role of matrix metalloproteinase-1 and tissue inhibitors of metalloproteinase-1 and -2 in different phases of bronchopulmonary dysplasia. Pediatr Res. 2001;50:761-766. [DOI] [PubMed] [Google Scholar]
48. De Paepe ME, Greco D, Mao Q.. Angiogenesis-related gene expression profiling in ventilated preterm human lungs. Exp Lung Res. 2010;36:399-410. [DOI] [PubMed] [Google Scholar]
49. Harijith A, Choo-Wing R, Cataltepe S, et al. A role for matrix metalloproteinase 9 in IFNγ-mediated injury in developing lungs: relevance to bronchopulmonary dysplasia. Am J Respir Cell Mol Biol. 2011;44:621-630. [DOI] [PMC free article] [PubMed] [Google Scholar]
50. Aghai ZH, Saslow JG, Mody K, et al. IFN-γ and IP-10 in tracheal aspirates from premature infants: relationship with bronchopulmonary dysplasia. Pediatr Pulmonol. 2013;48:8-13. [DOI] [PubMed] [Google Scholar]
51. Ballabh P, Kumari J, Krauss AN, et al. Soluble E-selectin, soluble L-selectin and soluble ICAM-1 in bronchopulmonary dysplasia, and changes with dexamethasone. Pediatrics. 2003;111: 461-468. [DOI] [PubMed] [Google Scholar]
52. Kojima T, Sasai M, Kobayashi Y.. Increased soluble ICAM-1 in tracheal aspirates of infants with bronchopulmonary dysplasia. Lancet. 1993;342:1023-1024. [DOI] [PubMed] [Google Scholar]
53. Suwara MI, Green NJ, Borthwick LA, et al. IL-1α released from damaged epithelial cells is sufficient and essential to trigger inflammatory responses in human lung fibroblasts. Mucosal Immunol. 2014;7:684-693. [DOI] [PMC free article] [PubMed] [Google Scholar]
54. Nareznoi D, Konikov-Rozenman J, Petukhov D, et al. Matrix metalloproteinases retain soluble FasL-mediated resistance to cell death in fibrotic-lung myofibroblasts. Cells. 2020;9:411. [DOI] [PMC free article] [PubMed] [Google Scholar]
55. Predescu SA, Zhang J, Bardita C, et al. Mouse lung fibroblast resistance to fas-mediated apoptosis is dependent on the baculoviral inhibitor of apoptosis protein 4 and the cellular FLICE-inhibitory protein. Front Physiol. 2017;8:128. [DOI] [PMC free article] [PubMed] [Google Scholar]
56. De Paepe M, E, Gundavarapu S, Tantravahi U, et al. Fas-ligand-induced apoptosis of respiratory epithelial cells causes disruption of postcanalicular alveolar development. Am J Pathol. 2008;173: 42-56. [DOI] [PMC free article] [PubMed] [Google Scholar]
57. Abreu SC, Rolandsson Enes S, Dearborn J, et al. Lung inflammatory environments differentially alter mesenchymal stromal cell behavior. Am J Physiol Lung Cell Mol Physiol. 2019;317: L823-L831. [DOI] [PMC free article] [PubMed] [Google Scholar]
58. Islam D, Huang Y, Fanelli V, et al. Identification and modulation of microenvironment is crucial for effective mesenchymal stromal cell therapy in acute lung injury. Am J Respir Crit Care Med. 2019;199:1214-1224. [DOI] [PubMed] [Google Scholar]
59. Waszak P, Alphonse R, Vadivel A, et al. Preconditioning enhances the paracrine effect of mesenchymal stem cells in preventing oxygen-induced neonatal lung injury in rats. Stem Cells Dev. 2012;21:2789-2797. [DOI] [PubMed] [Google Scholar]
60. Dolhnikoff M, Morin J, Roughley PJ, et al. Expression of lumican in human lungs. Am J Respir Cell Mol Biol. 1998;19:582-587. [DOI] [PubMed] [Google Scholar]
61. Chetty A, Bennett M, Dang L, et al. Pigment epithelium–derived factor mediates impaired lung vascular development in neonatal hyperoxia. Am J Respir Cell Mol Biol. 2015;52:295-303. [DOI] [PMC free article] [PubMed] [Google Scholar]
62. Liang H, Hou H, Yi W, et al. Increased expression of pigment epithelium-derived factor in aged mesenchymal stem cells impairs their therapeutic efficacy for attenuating myocardial infarction injury. Eur Heart J. 2013;34:1681-1690. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

sxab023_suppl_Supplementary_Data_S1

Click here for additional data file.^{(119.9KB, pdf)}

sxab023_suppl_Supplementary_Data_S2

Click here for additional data file.^{(145.4KB, pdf)}

sxab023_suppl_Supplementary_Data_S3

Click here for additional data file.^{(57.1KB, pdf)}

sxab023_suppl_Supplementary_Figure_S1

Click here for additional data file.^{(398.9KB, pdf)}

sxab023_suppl_Supplementary_Figure_S2

Click here for additional data file.^{(23.6MB, pdf)}

sxab023_suppl_Supplementary_Figure_S3

Click here for additional data file.^{(2MB, pdf)}

sxab023_suppl_Supplementary_Figure_S4

Click here for additional data file.^{(2.3MB, pdf)}

sxab023_suppl_Supplementary_Figure_S5

Click here for additional data file.^{(1.3MB, pdf)}

sxab023_suppl_Supplementary_Figure_S6

Click here for additional data file.^{(2.1MB, pdf)}

sxab023_suppl_Supplementary_Figure_S7

Click here for additional data file.^{(1.2MB, pdf)}

sxab023_suppl_Supplementary_Tables

Click here for additional data file.^{(26.1MB, xlsx)}

Data Availability Statement

[CIT0001] 1. Warburton D. Overview of lung development in the newborn human Neonatology. 2017;111:398-401. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0002] 2. Thébaud B, Goss KN, Laughon M, et al. Bronchopulmonary dysplasia Nat Rev Dis Primers. 2019;5:78. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0003] 3. Wong PM, Lees AN, Louw J, et al. Emphysema in young adult survivors of moderate-to-severe bronchopulmonary dysplasia Eur Respir J. 2008;32:321-328. [DOI] [PubMed] [Google Scholar]

[CIT0004] 4. Fawke J, S L, Kirkby J, et al. Lung function and respiratory symptoms at 11 years in children born extremely preterm: The EPICure study Am J Respir Crit Care Med. 2010;182:237-245. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0005] 5. O’Reilly M, Möbius MA, Vadivel A, et al. Late rescue therapy with cord-derived mesenchymal stromal cells for established lung injury  in experimental bronchopulmonary dysplasia Stem Cells Dev. 2020;29:364-371. [DOI] [PubMed] [Google Scholar]

[CIT0006] 6. van Haaften T, Byrne R, Bonnet S, et al. Airway delivery of mesenchymal stem cells prevents arrested alveolar growth in neonatal lung injury in rats Am J Respir Crit Care Med. 2009;180:1131-1142. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0007] 7. Sutsko RP, Young KC, Ribeiro A, et al. Long-term reparative effects of mesenchymal stem cell therapy following neonatal hyperoxia-induced lung injury Pediatr Res. 2013;73:46-53. [DOI] [PubMed] [Google Scholar]

[CIT0008] 8. Pierro M, Ionescu L, Montemurro T, et al. Short-term, long-term and paracrine effect of human umbilical cord-derived stem cells in lung injury prevention and repair in experimental bronchopulmonary dysplasia Thorax. 2013;68:475-484. [DOI] [PubMed] [Google Scholar]

[CIT0009] 9. Augustine S, Cheng W, Avey MT, et al. Are all stem cells equal? Systematic review, evidence map, and meta-analyses of preclinical stem cell-based therapies for bronchopulmonary dysplasia Stem Cells Transl Med. 2020;9:158-168. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0010] 10. Aslam M, Baveja R, Liang OD, et al. Bone marrow stromal cells attenuate lung injury in a murine model of neonatal chronic lung disease Am J Respir Crit Care Med. 2009;180:1122-1130. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0011] 11. Collins JJP, Thébaud B.. Progenitor cells of the distal lung and their potential role in neonatal lung disease: progenitor cells of the distal lung Birth Defects Res A Clin Mol Teratol. 2014;100:217-226. [DOI] [PubMed] [Google Scholar]

[CIT0012] 12. Ushakumary MG, Riccetti M, Perl AT.. Resident interstitial lung fibroblasts and their role in alveolar stem cell niche development, homeostasis, injury, and regeneration. Stem Cells Transl Med. 2021;10:1021-1032. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0013] 13. Collins JJP, Lithopoulos MA, dos Santos CC, et al. Impaired angiogenic supportive capacity and altered gene expression profile of resident CD146 ⁺ mesenchymal stromal cells isolated from hyperoxia-injured neonatal rat lungs Stem Cells Dev. 2018;27:1109-1124. [DOI] [PubMed] [Google Scholar]

[CIT0014] 14. Lama VN, Smith L, Badri L, et al. Evidence for tissue-resident mesenchymal stem cells in human adult lung from studies of transplanted allografts J Clin Invest. 2007;117:989-996. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0015] 15. Rolandsson Enes S, Andersson Sjöland A, Skog I, et al. MSC from fetal and adult lungs possess lung-specific properties compared to bone marrow-derived MSC Sci Rep. 2016;6:29160. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0016] 16. Deng M, Li J, Gan Y, et al. Changes in the number of CD31^−;CD45^−;Sca-1⁺ cells and Shh signaling pathway involvement in the lungs of mice with emphysema and relevant effects of acute adenovirus infection. COPD. 2017;12:861-872. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0017] 17. Gong X, Sun Z, Cui D, et al. Isolation and characterization of lung resident mesenchymal stem cells capable of differentiating into alveolar epithelial type II cells: Isolation and characterization of LR-MSCs Cell Biol Int. 2014;38:405-411. [DOI] [PubMed] [Google Scholar]

[CIT0018] 18. Shi C, Cao X, Chen X, et al. Intracellular surface-enhanced Raman scattering probes based on TAT peptide-conjugated Au nanostars for distinguishing the differentiation of lung resident mesenchymal stem cells Biomaterials. 2015;58:10-25. [DOI] [PubMed] [Google Scholar]

[CIT0019] 19. McQualter JL, Brouard N, Williams B, et al. Endogenous fibroblastic progenitor cells in the adult mouse lung are highly enriched in the Sca-1 positive cell fraction Stem Cells. 2009;27:623-633. [DOI] [PubMed] [Google Scholar]

[CIT0020] 20. McQualter JL, Yuen K, Williams B, et al. Evidence of an epithelial stem/progenitor cell hierarchy in the adult mouse lung Proc Natl Acad Sci USA. 2010;107:1414-1419. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0021] 21. Salaets T, Gie A, Jimenez J, et al. Local pulmonary drug delivery in the preterm rabbit: feasibility and efficacy of daily intratracheal injections. Am J Physiol Lung Cell Mol Physiol. 2019;316:L589-L597. [DOI] [PubMed] [Google Scholar]

[CIT0022] 22. McGinnis CS, Patterson DM, Winkler J, et al. MULTI-seq: sample multiplexing for single-cell RNA sequencing using lipid-tagged indices. Nat Methods. 2019;16:619-626. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0023] 23. Hurskainen M, Mižíková I, Cook DP, et al. Single cell transcriptomic analysis of murine lung development on hyperoxia-induced damage. Nat Commun. 2021;12:1565. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0024] 24. Malinowski M, Pietraszek K, Perreau C, et al. Effect of lumican on the migration of human mesenchymal stem cells and endothelial progenitor cells: involvement of matrix metalloproteinase-14. PLoS One. 2012;7:e50709. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0025] 25. Li L-F, Chu P-H, Hung C-Y, et al. Lumican regulates ventilation-induced epithelial-mesenchymal transition through extracelluar signal-regulated kinase pathway. Chest. 2013;143:1252-1260. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0026] 26. Browaeys R, Saelens W, Saeys Y.. NicheNet: modeling intercellular communication by linking ligands to target genes. Nat Methods. 2020;17:159-162. [DOI] [PubMed] [Google Scholar]

[CIT0027] 27. Han Y, Su C, Yu D, et al. Cholecystokinin attenuates radiation-induced lung cancer cell apoptosis by modulating p53 gene transcription. Am J Transl Res. 2017;9:638-646. [PMC free article] [PubMed] [Google Scholar]

[CIT0028] 28. Bozyk PD, Bentley JK, Popova AP, et al. Neonatal periostin knockout mice are protected from hyperoxia-induced alveolar simplication. PLoS One. 2012;7:e31336. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0029] 29. Mohamed WAW, Niyazy WH, Mahfouz AA.. Angiopoietin-1 and endostatin levels in cord plasma predict the development of bronchopulmonary dysplasia in preterm infants. J Trop Pediatr. 2011;57:385-388. [DOI] [PubMed] [Google Scholar]

[CIT0030] 30. Shudo Y, Cohen JE, Goldstone AB, et al. Isolation and trans-differentiation of mesenchymal stromal cells into smooth muscle cells: utility and applicability for cell-sheet engineering. Cytotherapy. 2016;18:510-517. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0031] 31. Zacharias WJ, Frank DB, Zepp JA, et al. Regeneration of the lung alveolus by an evolutionarily conserved epithelial progenitor. Nature. 2018;555:251-255. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0032] 32. Yu S, Li P, Li B, et al. RelA promotes proliferation but inhibits osteogenic and chondrogenic differentiation of mesenchymal stem cells. FEBS Lett. 2020;594:1368-1378. [DOI] [PubMed] [Google Scholar]

[CIT0033] 33. Gattu AK, Swenson ES, Iwakiri Y, et al. Determination of mesenchymal stem cell fate by pigment epithelium-derived factor (PEDF) results in increased adiposity and reduced bone mineral content. FASEB J. 2013;27:4384-4394. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0034] 34. Hikino S, Ohga S, Kinjo T, et al. Tracheal aspirate gene expression in preterm newborns and development of bronchopulmonary dysplasia: TAF cell gene expression predicting BPD. Pediatr Int. 2012;54:208-214. [DOI] [PubMed] [Google Scholar]

[CIT0035] 35. Sipp D, Robey PG, Turner L.. Clear up this stem-cell mess. Nature. 2018;561:455-457. [DOI] [PubMed] [Google Scholar]

[CIT0036] 36. Raredon MSB, Adams TS, Suhail Y, et al. Single-cell connectomic analysis of adult mammalian lungs. Sci Adv. 2019;5:eaaw3851. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0037] 37. Angelidis I, Simon LM, Fernandez IE, et al. An atlas of the aging lung mapped by single cell transcriptomics and deep tissue proteomics. Nat Commun. 2019;10:963. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0038] 38. Li R, Bernau K, Sandbo N, et al. Pdgfra marks a cellular lineage with distinct contributions to myofibroblasts in lung maturation and injury response. ELife. 2018;7:e36865. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0039] 39. Cohen M, Giladi A, Gorki A-D, et al. Lung single-cell signaling interaction map reveals basophil role in macrophage imprinting. Cell. 2018;175:1031-1044.e18. [DOI] [PubMed] [Google Scholar]

[CIT0040] 40. Guo M, Du Y, Gokey JJ, et al. Single cell RNA analysis identifies cellular heterogeneity and adaptive responses of the lung at birth. Nat Commun. 2019;10:37. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0041] 41. Travaglini KJ, Nabhan AN, Penland L, et al. A molecular cell atlas of the human lung from single-cell RNA sequencing. Nature. 2020;587:619-625. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0042] 42. Vila Ellis L, Cain MP, Hutchison V, et al. Epithelial vegfa specifies a distinct endothelial population in the mouse lung. Dev Cell. 2020;52:617-630.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0043] 43. Hosford GE, Fang X, Olson DM.. Hyperoxia decreases matrix metalloproteinase-9 and increases tissue inhibitor of matrix metalloproteinase-1 protein in the newborn rat lung: association with arrested alveolarization. Pediatr Res. 2004;56:26-34. [DOI] [PubMed] [Google Scholar]

[CIT0044] 44. Kumar VHS, Lakshminrusimha S, Kishkurno S, et al. Neonatal hyperoxia increases airway reactivity and inflammation in adult mice: neonatal hyperoxia and AHR in mice. Pediatr Pulmonol. 2016;51:1131-1141. [DOI] [PubMed] [Google Scholar]

[CIT0045] 45. Schulz CG, Sawicki G, Lemke RP, et al. MMP-2 and MMP-9 and their tissue inhibitors in the plasma of preterm and term neonates. Pediatr Res. 2004;55:794-801. [DOI] [PubMed] [Google Scholar]

[CIT0046] 46. Fulton CT, Cui TX, Goldsmith AM, et al. Gene expression signatures point to a male sex-specific lung mesenchymal cell PDGF receptor signaling defect in infants developing bronchopulmonary dysplasia. Sci Rep. 2018;8:17070. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0047] 47. Dik WA, de Krijger RR, Bonekamp L, et al. Localization and potential role of matrix metalloproteinase-1 and tissue inhibitors of metalloproteinase-1 and -2 in different phases of bronchopulmonary dysplasia. Pediatr Res. 2001;50:761-766. [DOI] [PubMed] [Google Scholar]

[CIT0048] 48. De Paepe ME, Greco D, Mao Q.. Angiogenesis-related gene expression profiling in ventilated preterm human lungs. Exp Lung Res. 2010;36:399-410. [DOI] [PubMed] [Google Scholar]

[CIT0049] 49. Harijith A, Choo-Wing R, Cataltepe S, et al. A role for matrix metalloproteinase 9 in IFNγ-mediated injury in developing lungs: relevance to bronchopulmonary dysplasia. Am J Respir Cell Mol Biol. 2011;44:621-630. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0050] 50. Aghai ZH, Saslow JG, Mody K, et al. IFN-γ and IP-10 in tracheal aspirates from premature infants: relationship with bronchopulmonary dysplasia. Pediatr Pulmonol. 2013;48:8-13. [DOI] [PubMed] [Google Scholar]

[CIT0051] 51. Ballabh P, Kumari J, Krauss AN, et al. Soluble E-selectin, soluble L-selectin and soluble ICAM-1 in bronchopulmonary dysplasia, and changes with dexamethasone. Pediatrics. 2003;111: 461-468. [DOI] [PubMed] [Google Scholar]

[CIT0052] 52. Kojima T, Sasai M, Kobayashi Y.. Increased soluble ICAM-1 in tracheal aspirates of infants with bronchopulmonary dysplasia. Lancet. 1993;342:1023-1024. [DOI] [PubMed] [Google Scholar]

[CIT0053] 53. Suwara MI, Green NJ, Borthwick LA, et al. IL-1α released from damaged epithelial cells is sufficient and essential to trigger inflammatory responses in human lung fibroblasts. Mucosal Immunol. 2014;7:684-693. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0054] 54. Nareznoi D, Konikov-Rozenman J, Petukhov D, et al. Matrix metalloproteinases retain soluble FasL-mediated resistance to cell death in fibrotic-lung myofibroblasts. Cells. 2020;9:411. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0055] 55. Predescu SA, Zhang J, Bardita C, et al. Mouse lung fibroblast resistance to fas-mediated apoptosis is dependent on the baculoviral inhibitor of apoptosis protein 4 and the cellular FLICE-inhibitory protein. Front Physiol. 2017;8:128. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0056] 56. De Paepe M, E, Gundavarapu S, Tantravahi U, et al. Fas-ligand-induced apoptosis of respiratory epithelial cells causes disruption of postcanalicular alveolar development. Am J Pathol. 2008;173: 42-56. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0057] 57. Abreu SC, Rolandsson Enes S, Dearborn J, et al. Lung inflammatory environments differentially alter mesenchymal stromal cell behavior. Am J Physiol Lung Cell Mol Physiol. 2019;317: L823-L831. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0058] 58. Islam D, Huang Y, Fanelli V, et al. Identification and modulation of microenvironment is crucial for effective mesenchymal stromal cell therapy in acute lung injury. Am J Respir Crit Care Med. 2019;199:1214-1224. [DOI] [PubMed] [Google Scholar]

[CIT0059] 59. Waszak P, Alphonse R, Vadivel A, et al. Preconditioning enhances the paracrine effect of mesenchymal stem cells in preventing oxygen-induced neonatal lung injury in rats. Stem Cells Dev. 2012;21:2789-2797. [DOI] [PubMed] [Google Scholar]

[CIT0060] 60. Dolhnikoff M, Morin J, Roughley PJ, et al. Expression of lumican in human lungs. Am J Respir Cell Mol Biol. 1998;19:582-587. [DOI] [PubMed] [Google Scholar]

[CIT0061] 61. Chetty A, Bennett M, Dang L, et al. Pigment epithelium–derived factor mediates impaired lung vascular development in neonatal hyperoxia. Am J Respir Cell Mol Biol. 2015;52:295-303. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0062] 62. Liang H, Hou H, Yi W, et al. Increased expression of pigment epithelium-derived factor in aged mesenchymal stem cells impairs their therapeutic efficacy for attenuating myocardial infarction injury. Eur Heart J. 2013;34:1681-1690. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Single-Cell RNA Sequencing-Based Characterization of Resident Lung Mesenchymal Stromal Cells in Bronchopulmonary Dysplasia

Ivana Mižíková

Flore Lesage

Chanele Cyr-Depauw

David P Cook

Maria Hurskainen

Satu M Hänninen

Arul Vadivel

Pauline Bardin

Shumei Zhong

Olli Carpén

Barbara C Vanderhyden

Bernard Thébaud

Abstract

Graphical Abstract

Graphical Abstract.

Significance Statement.

Introduction

Materials and Methods

Experimental Animals

Lung Isolation

Mean Linear Intercept Measurement

Fluorescent Activated Cell Sorting

Cell Culture and Storage

Colony Formation Assay

MSCs Surface Marker Profiling

Osteogenic Differentiation

Adipogenic differentiation

Chondrogenic differentiation

Fluorescent in situ hybridization

Multiplexing Samples for scRNA-seq

scRNA-seq Library Preparation and Sequencing

scRNA-seq Data Processing, Analyses, and Quantification

Statistical Analysis

Results

The Developing Murine Lung Contains a Population of L-MSC Marked by the Expression of Ly6a

Figure 1.

The Transcription Profile and Signaling Activity of Ly6a+ L-MSCs Change Significantly During Postnatal Lung Development

Figure 2.

The Transcription Profile and Signaling Activity of Ly6a+ L-MSC Change Significantly During Postnatal Lung Development in Response to Hyperoxia

Figure 3.

Murine Ly6a+/Col14a1+ L-MSCs Localize to Perivascular Regions of the Developing Lung

Figure 4.

Hyperoxia Exposure Does Not Impact Clonal or Differentiation Potential of LY6A+ L-MSCs

Figure 5.

Cell Culture Alters the Gene Expression Profile of LY6A+ L-MSCs

Figure 6.

Figure 7.

Discussion

Conclusion

Supplementary Material

Acknowledgments

Funding

Conflict of Interest

Author Contributions

Data Availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

The Transcription Profile and Signaling Activity of Ly6a⁺ L-MSCs Change Significantly During Postnatal Lung Development

The Transcription Profile and Signaling Activity of Ly6a⁺ L-MSC Change Significantly During Postnatal Lung Development in Response to Hyperoxia

Murine Ly6a⁺/Col14a1⁺ L-MSCs Localize to Perivascular Regions of the Developing Lung

Hyperoxia Exposure Does Not Impact Clonal or Differentiation Potential of LY6A⁺ L-MSCs

Cell Culture Alters the Gene Expression Profile of LY6A⁺ L-MSCs