Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 11 (1), 635

The in Vivo Genetic Program of Murine Primordial Lung Epithelial Progenitors


The in Vivo Genetic Program of Murine Primordial Lung Epithelial Progenitors

Laertis Ikonomou et al. Nat Commun.


Multipotent Nkx2-1-positive lung epithelial primordial progenitors of the foregut endoderm are thought to be the developmental precursors to all adult lung epithelial lineages. However, little is known about the global transcriptomic programs or gene networks that regulate these gateway progenitors in vivo. Here we use bulk RNA-sequencing to describe the unique genetic program of in vivo murine lung primordial progenitors and computationally identify signaling pathways, such as Wnt and Tgf-β superfamily pathways, that are involved in their cell-fate determination from pre-specified embryonic foregut. We integrate this information in computational models to generate in vitro engineered lung primordial progenitors from mouse pluripotent stem cells, improving the fidelity of the resulting cells through unbiased, easy-to-interpret similarity scores and modulation of cell culture conditions, including substratum elastic modulus and extracellular matrix composition. The methodology proposed here can have wide applicability to the in vitro derivation of bona fide tissue progenitors of all germ layers.

Conflict of interest statement

The authors declare no competing interests.


Fig. 1
Fig. 1. Identification and lineage tracing of Nkx2-1+ lung epithelial cells.
a Schematic of Nkx2-1CreERT2; R26RnT/nG mice. b Lineage-tracing experimental design using the mice described in (a) that express either nuclear tdTomato (nT; turquoise pseudocolor) prior to Cre recombination vs. nuclear GFP (nG; green pseudocolor) after Cre recombination. c Confocal micrographs of lung paraffin tissue sections showing lineage-labeled cells co-stained with markers of lung epithelial cell types. Arrows indicate lineage-traced (ntdTomato-/nGFP+; nT-/nG+) cell co-localization with markers of lung epithelial cells. For acetyl-α-tubulin and PDPN stains, yellow arrows indicate region shown in inset. Representative images from three lineage-traced embryos, Scale bars: 50 μm. d Schematic of the Nkx2-1GFP knock-in reporter mouse (upper panel) with Nkx2-1 ISH at E9.5 (lower left panel) and Nkx2-1GFP reporter expression in forebrain, thyroid and lung domains at E10.0 (lower right panel). Notice absence of Nkx2-1GFP expression in wild-type littermate. NB: the nuclear GFP lineage tracer in panels ac (nG) is a different GFP than the knock-in cytoplasmic Nkx2-1GFP reporter shown in dg. e Epifluorescence stereomicrographs of Nkx2-1GFP expression time course during lung development in the Nkx2-1GFP knock-in mouse demonstrate that the reporter is faithful and specific. Nkx2-1GFP+ thyroid is situated in front of the trachea at E13.5 (arrowhead). DF dark field, BF bright field. Representative images from embryos derived from three to ten independent litters per time point. f Confocal micrographs of adult Nkx2-1GFP mouse lung cryosections. NKX2-1GFP expression is evident in club (SCGB1A1), Type II alveolar epithelial (SFTPC), and basal cells (P63) but low or undetectable in ciliated (acetylated α-tubulin) and Type I alveolar epithelial cells (PDPN). The PDPN micrograph is a maximum intensity projection of six 0.82 μm optical slices. Representative images from three adult mice. Scale bars: 20 μm. g Bivariate flow cytometry dot plot indicating populations with various levels of NKX2-1GFP and EPCAM (color gates) and h RT-qPCR analysis of sorted populations showing enrichment of proximal and distal lung marker expression in the EPCAM+ NKX2-1GFP+ fraction, N = 3 independent sorts, error bars represent standard deviation.
Fig. 2
Fig. 2. RNA-Seq analysis of purified mouse embryonic Nkx2-1+ populations.
a Schematic of embryo dissection and NKX2-1GFP+ cell sorting at the lung primordium stage (E9.0, 18–23 somites) and at E13.5. The Nkx2-1GFP+ lung, thyroid, and forebrain domains were micro-dissected using an epifluorescence stereomicroscope. At E13.5, thyroid is separated from the trachea prior to enzyme digestion and sorting (left panels). Bivariate flow cytometry dot plots showing sorted NKX2-1GFP+ cell populations (middle panel) and pre-specified foregut endoderm (ENDM1+EPCAM+) and ectoderm (ENDM1EPCAM+) (right panel). b FACS-purified cell populations used in RNA-Seq analysis. The same colors are consistently used in subsequent figures to identify the respective populations. c The number of cells recovered by flow cytometry and normalized per embryo for the NKX2-1GFP+ populations (lung, thyroid, and forebrain) and foregut endoderm. The number of sorts: N = 7 for the lung, thyroid, and forebrain; N = 5 for foregut endoderm, error bars represent standard deviation. d Nkx2-1 expression (RNA-Seq normalized counts) in sorted Nkx2-1GFP+ and Nkx2-1GFP-negative populations. e Nkx2-1 counts for all Nkx2-1GFP+ samples (N = 3 lung, thyroid, and forebrain samples) mapped on the Nkx2-1 locus. Normalized Nkx2-1 counts for each sample are indicated on the right. f Principal component analysis (PCA) plot of the eight populations depicted in (b). Arrows connect specified endodermal or ectodermal populations and their respective precursor stages. The partial overlap of the Nkx2-1 lung and forebrain field populations is most probably due to the fact that both populations are heterogeneous and contain mesenchymal, endothelial, neuronal, non-lung foregut, and other lineages.
Fig. 3
Fig. 3. Gene expression profiles of Nkx2-1+ endodermal and ectodermal populations.
a Heatmaps containing row-normalized z-scores and representing unsupervised hierarchical clustering of samples analyzed in population RNA-Seq. Each heatmap depicts the pairwise comparison of specified endodermal or ectodermal Nkx2-1+ populations and their precursor stage (foregut endoderm and ectoderm, respectively). The number of genes used to create each heatmap appears underneath the heatmap. The cutoff criteria used define the number of genes for each heatmap are: false discovery rate (FDR) < 0.05, average expression > 0 and |log2(fold change)| > 2. b Heatmap of differentially expressed transcripts between the two Nkx2-1+ endodermal progenitor cell populations (lung and thyroid). The same cutoff criteria as in (a) were used. c Heatmap of differentially expressed transcripts across the three Nkx2-1+ cell populations (tissue-specific gene “signatures”). The same cutoff criteria as in (a) were used with the only difference of |log2(fold change)| > 3.
Fig. 4
Fig. 4. Differential regulation of the Hippo, Wnt and Tgf-β superfamily pathways in lung specification.
a Heatmap of normalized enrichment scores (NES) for pathways that are differentially regulated in specified Nkx2-1+ embryonic populations by gene set enrichment analysis (GSEA). The KEGG pathways database was used as basis for GSEA and an FDR cutoff of 0.25 was used for inclusion of pathways for each pairwise comparison. Pathways that were investigated are indicated with an asterisk (*). b Heatmaps of selected transcripts for the Hippo, Wnt, and Tgf-β superfamily pathways showing differential expression between pre-specified foregut endoderm and Nkx2-1+ lung primordium. c Whole-mount confocal micrographs of uncultured, freshly isolated embryos at 25–28 somite stage (upper panel), or mouse foregut explants (lower panel). Foreguts were explanted at the eight-somite stage and cultured for 2 days. Nkx2-1 expression demarcates the lung and thyroid domains. Co-staining with Sox2 (upper panel) and E-cad (lower panel) is used to visualize foregut endoderm; Arrow = Thyroid, Arrowhead = Lung. d Whole-mount confocal micrographs of mouse foregut explants. Foreguts were explanted at the 6–8 somite stage and cultured for 3 days. E-cad co-staining marks foregut endoderm; Arrow = Thyroid, Arrowhead = Lung.
Fig. 5
Fig. 5. ECM modulation during lung specification.
a Projection strength scores for the PSC-derived lung progenitors described in Serra et al.. b Schematic illustrating directed differentiation of mouse PSCs into lung epithelial lineages using 2D-gelatin-coated plates as well as 3D-Matrigel-coated plates. For these experiments, we used a mouse ESC line with an Nkx2-1mCherry reporter. The 2D condition leads to a low yield of Nkx2-1mCherry+/Epcam+ cells and a dominant population of Nkx2-1mCherry+/Epcam cells. c, d 3D Matrigel leads to increased numbers of Nkx2-1mCherry+ cells compared with 2D gelatin at both D8 and D14, Scale bar: 100 µm. e Bivariate flow cytometry plots for D14 lung progenitors showing expression of Nkx2-1mCherry and BV421-Epcam. f Quantification of Nkx2-1mCherry+/Epcam+ populations shown in e as a percent of cells (left panel) and cell yield per starting (D0) undifferentiated ESC (right panel), P-value < 0.01 by two-tailed t tests, N = 3 independent experiments, error bars represent standard deviation. g RT-qPCR data from downstream differentiation of 3D-Matrigel Nkx2-1mCherry+Epcam+ sorted populations shown in d. Fold changes relative to undifferentiated cells, statistics from two-tailed, unpaired t tests, N = 3 independent experiments, error bars represent standard deviation. h D14 EPCAM+ cells from both 2D-gelatin-coated and 3D-Matrigel-coated plates were assessed by bulk RNA-seq. The resulting samples were then hierarchically clustered using Euclidean distances. i Principal Component Analysis (PCA) of bulk RNA-Seq samples. The top 500 most variable genes were used in this analysis and they are listed in Supplementary Data 2.
Fig. 6
Fig. 6. Similarity analysis of in vitro PSC-derived and in vivo lung progenitors.
a Schematic depicting the four types of in vitro PSC-derived progenitors using a lung specification cocktail (in vitro-derived lineages) and derivation of similarity scores to in vivo populations (reference lineages) via two methods, namely linear projection analysis (LAP) and gene set analysis (GSA) (bd). Projections strength scores from LAP analysis for the four populations shown in a, including projections to lung endodermal populations (b), non-lung endoderm (c), and mesenchymal-like populations (d). Scores shown in Fig. 5a are re-plotted here for comparison purposes. e Projection scores from gene set analysis using the reference cell populations from bd and Supplementary Fig. 7A with bulk RNA-Seq data.

Similar articles

See all similar articles


    1. Murry CE, Keller G. Differentiation of embryonic stem cells to clinically relevant populations: lessons from embryonic development. Cell. 2008;132:661–680. doi: 10.1016/j.cell.2008.02.008. - DOI - PubMed
    1. Ikonomou L, Kotton DN. Derivation of endodermal progenitors from pluripotent stem cells. J. Cell Physiol. 2015;230:246–258. doi: 10.1002/jcp.24771. - DOI - PMC - PubMed
    1. Domyan ET, et al. Signaling through BMP receptors promotes respiratory identity in the foregut via repression of Sox2. Development. 2011;138:971–981. doi: 10.1242/dev.053694. - DOI - PMC - PubMed
    1. Goss AM, et al. Wnt2/2b and beta-catenin signaling are necessary and sufficient to specify lung progenitors in the foregut. Dev. Cell. 2009;17:290–298. doi: 10.1016/j.devcel.2009.06.005. - DOI - PMC - PubMed
    1. Harris-Johnson KS, Domyan ET, Vezina CM, Sun X. beta-Catenin promotes respiratory progenitor identity in mouse foregut. Proc. Natl Acad. Sci. USA. 2009;106:16287–16292. doi: 10.1073/pnas.0902274106. - DOI - PMC - PubMed

LinkOut - more resources