Mouse lung lobes already differ molecularly before birth, study finds

The five lobes of the mouse lung are not molecularly interchangeable, even before the organ has finished taking shape. A new study published in Cell Regeneration reports that individual lung lobes of mouse embryos at embryonic day 15.5, a stage within the pseudoglandular period of lung development, already carry measurably different gene expression programs and different predicted proportions of at least one specialized endothelial cell type. The work, led by Yun Yang, Yonggao Fu and Shiting Li with senior author Xianfa Yang, suggests that the regional identity of lung lobes is established during embryogenesis rather than acquired later in life.

The research team dissected the five lobes of the E15.5 mouse lung, one left lobe and four right lobes designated cranial, middle, caudal and accessory, and subjected each to total transcriptome sequencing. Rather than sequencing only polyadenylated messenger RNA, the researchers used ribosomal RNA depletion to capture both coding and non-coding expressed genes across the whole transcriptome. For each lobe they generated three high-quality biological replicates, detecting on average more than 16,000 messenger RNAs and roughly 4,600 long non-coding RNAs per lobe. Replicates from the same lobe correlated strongly with one another, with pairwise correlation coefficients ranging from approximately 0.85 to 0.96, which the authors cite as evidence that the dataset is reproducible enough to support lobe-level comparisons.

When the researchers applied principal component analysis to the top 1,000 highly variable protein-coding genes, the biological replicates of each lobe clustered together and separated cleanly from the other lobes. The same analysis performed on the top 1,000 highly variable long non-coding RNAs produced a strikingly different picture: the samples became largely intermingled. This contrast led the team to conclude that the consistent inter-lobar divergence they observed occurs primarily at the level of protein-coding genes, a finding that shapes how future studies might approach lobar comparisons in the developing lung.

Differential expression analysis, carried out with criteria of an adjusted p-value below 0.05 and an absolute log2 fold change greater than 0.5, revealed that each lobe is enriched for its own gene module. The left lobe showed higher expression of a set of 55 genes associated with the positive regulation of immune system processes. The right cranial lobe, by contrast, preferentially expressed genes linked to cardiac muscle contraction and ventricular cardiac muscle tissue morphogenesis. Because the right cranial lobe sits close to the developing heart, the authors specifically checked for cardiomyocyte marker expression in the dissected lobes to rule out cardiac tissue contamination, and they report that this potential contamination was precluded.

The immune signature in the left lobe connects to an intriguing observation from other recent work cited in the paper: administering lipopolysaccharide to mice induces a stronger inflammatory response in the left lung lobe, with increased recruitment of IBA1-positive cells. The authors suggest that their embryonic data may indicate that lobe-specific immune predisposition is established during development, with the left lobe acquiring a distinct transcriptional program that could influence its sensitivity to inflammatory stimuli later in life. They are careful to frame this as a possibility that warrants further investigation, noting that the immune microenvironment of the embryonic lung remains largely unexplored.

To move from gene expression to cellular composition, the team built a framework they call CCC-Lobe, short for Cell type Composition Characterization resolved at Lobe resolution. This approach integrates their bulk lobe transcriptomes with a published stage-matched single-cell transcriptomic reference of the embryonic lung, using marker genes upregulated in each major cell type to sharpen the deconvolution signal. The analysis covered major epithelial subtypes, endothelial subtypes including general capillary cells, and mesenchymal populations such as Wnt2-positive fibroblasts and myofibroblasts. Marker gene expression patterns, such as Nkx2-1 in epithelial cells, Wnt2 in fibroblasts, and Aplnr and Cd93 in general capillary cells, were used to validate the inferred cell identities.

The deconvolution results pointed to lobe-specific differences in cell type abundance, with the most striking pattern involving general capillary endothelial cells, abbreviated gCap. The left lobe showed the highest predicted abundance of gCap cells, while the right cranial lobe showed the lowest. General capillary cells have been reported to serve as a progenitor reservoir for repairing damaged capillaries, so the authors argue that lobe identity should be considered when evaluating the embryonic origin and developmental regulation of these capillary progenitors. They also note that their finding contrasts with previous work that reported a uniform distribution of gCap cells between right and left lung lobes, and they attribute the difference to the lobe-resolved resolution of their analysis at this specific embryonic stage.

Because the cell type proportions were computationally inferred, the team sought independent validation. They reanalyzed a publicly available spatial transcriptomic dataset of stage-matched E15.5 embryonic lung and found that the expression of Aplnr, a marker of general capillary cells, followed the same pattern predicted by CCC-Lobe: highest enrichment in the left lobe and lowest in the right cranial lobe. They then performed RNAscope staining for Aplnr transcripts on E15.5 lung cryosections, quantifying Aplnr-positive cell proportions across six cryosections with QuPath-based cell classification, and this staining corroborated the inter-lobar differences. The authors additionally observed that the molecular divergence of gCap cells may have an anterior-posterior positional component within the left lobe, with relatively higher expression of Cd93 and Aplnr in the middle part of that lobe.

The study also validated lobe-specific molecular signatures experimentally at the level of individual genes. Quantitative PCR confirmed the relative expression patterns of Aldh1a2, Rgs5, Gata4 and Hoxb9 across the lobes, and immunofluorescence staining showed that the mesothelial marker GATA4 was obviously enriched in the accessory lobe, consistent with the transcriptomic data. Morphological tracking using an Nkx2-1-eGFP lineage reporter mouse documented the developmental trajectory leading up to the stage analyzed: a GFP-positive endodermal diverticulum emerges from the ventral foregut wall, bilateral primary bronchi form separately, a rudimentary lobar organization with a unilobed left and quadrilobed right lung is apparent by E12.0, and all five lobes are clearly discernible by E15.5 with well-organized epithelial tubules surrounded by dense mesenchyme.

The authors place their findings in a broader clinical context. Regional variation is a recognized feature of the adult lung, and several lung diseases show lobe preference: bronchial atresia most commonly affects the left upper lobe in humans, while congenital pulmonary airway malformations are more frequently observed in the lower lung lobes. The team proposes that such lobe-biased disease susceptibilities may have developmental origins, with lobe-specific signatures established at least in part during embryogenesis. They also suggest that the cardiac gene enrichment in the right cranial lobe may reflect shared morphogenetic cues or coordinated cardiopulmonary patterning, noting prior work on multipotent cardiopulmonary progenitors and Wnt2-positive progenitors that contribute to nearly all lung mesoderm lineages, and they propose that future lineage-tracing studies could clarify whether the right cranial lobe is a special developmental hotspot for cardiopulmonary progenitor differentiation. As the authors emphasize, their results are observational and stage-specific, based on E15.5 mouse embryos with three replicates per lobe, and they call for future studies incorporating more embryonic timepoints, spatial multi-omics, functional perturbation and clinical samples to define how lobe-specific programs are established and how they might influence lifelong pulmonary health and disease.

Subject of Research: Lobe-specific molecular and cellular heterogeneity in the embryonic mouse lung

Article Title: Early emergence of inter-lobar heterogeneity in mouse embryonic lung

Article References: Yang, Y., Fu, Y., Li, S., Liu, Z., Wen, M., Jing, N., & Yang, X. (2026). Early emergence of inter-lobar heterogeneity in mouse embryonic lung. Cell Regeneration, 15(1), Article 31. https://doi.org/10.1186/s13619-026-00303-y

Image Credits: AI Generated

DOI: 10.1186/s13619-026-00303-y

Keywords: mouse embryonic lung, lung lobes, pseudoglandular stage, transcriptomics, general capillary endothelial cells, gCap, E15.5, inter-lobar heterogeneity, CCC-Lobe, RNAscope, cardiopulmonary development, congenital lung disease