Single Cell Analysis of Lung Lymphatic Endothelial Cells and Lymphatic Responses during Influenza Infection

Abstract

Tissue lymphatic vessels network plays critical roles in immune surveillance and tissue homeostasis in response to pathogen invasion, but how lymphatic system per se is remolded during infection is less understood. Here, we observed that influenza infection induces a significant increase of lymphatic vessel numbers in the lung, accompanied with extensive proliferation of lymphatic endothelial cells (LECs). Single-cell RNA sequencing illustrated the heterogeneity of LECs, identifying a novel PD-L1⁺ subpopulation that is present during viral infection but not at steady state. Specific deletion of Pd-l1 in LECs elevated the expansion of lymphatic vessel numbers during viral infection. Together these findings elucidate a dramatic expansion of lung lymphatic network in response to viral infection, and reveal a PD-L1⁺ LEC subpopulation that potentially modulates lymphatic vessel remolding.

Keywords: Influenza infection; Lymphatic endothelial cells (LECs); PD-L1; Regeneration; Single cell; lung injury; scRNA-seq.

Figure 1.

Influenza infection induces lymphatic vessel dilation followed with an increase in number. (a) Experimental timeline indicating the administration of tamoxifen (TMX), influenza infection and the days post-infection (dpi). (b) Representative hematoxylin and eosin (H&E) staining of lung sections from control and infected Prox1-CreERT2; R26R^tdT mice at 6 and 28 dpi. (c) Immunofluorescence staining for VEGFR3 (green) and tdTomato (red) of lung sections from control and virus-infected mice. Note nuclei counterstained with DAPI (blue). Quantification of lymphatic vessels per field is shown on the right (mean± SEM. n = 4. n.s., not significant). (d) Light-sheet microscopy images of tdTomato expression in control and virus-infected lung tissues at 6 dpi. Quantification of the lymphatic vessel diameter is shown in the adjacent graph (mean± SEM, n = 3 *** p < 0.001). (e) Immunofluorescence of tdTomato in the virus-infected lung sections at 28 dpi. Quantification of lymphatic vessels is shown on the right (mean± SEM, n = 4 for uninfected, n = 11 for 28 dpi, **** p < 0.0001). Scale bars: 200 μm for b (60 μm for insets), 100 μm for c–e (10 μm for insets in c).

Figure 2.

Influenza infection induces transient but intensive proliferation of LECs accompanied with a change in lymphatic function. (a) Immunofluorescence staining for Ki67 (green) in tdTomato-labeled lymphatic endothelial cells (LECs) (red) in the lung at different time points post viral infection. White arrowheads indicate Ki67⁺tdTomato⁺ proliferating lymphatic endothelial cells. (b) Quantification of the percentage of Ki67⁺ cells in tdTomato⁺ LECs. Note proliferation peaks at 14 dpi (mean ± SEM, n = 4 for uninfected/6 dpi/14 dpi, n = 3 for 21 dpi/28 dpi, ***p < 0.001, **p < 0.01; n.s., not significant). (c & e) Representative images of dextran-FITC drainage in lungs and lung drainage lymph nodes. (d & f) Quantitation of FITC-dextran mean intensity (mean ± SEM, n = 3, ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05). (g) Tidal volume (TV) at different time points post infection (*p < 0.05). Scale bars: 100 μm for a and c.

Figure 3.

scRNA-Seq analysis identifies a PD-L1⁺ LEC subpopulation during influenza infection. (a) scRNA-seq experimental diagram. (b) UMAP clustering of single-cell transcriptomes of LECs purified from control (Prox1-CreERT2; R26R^tdT uninfected) and virus-infected lungs. Note lung tissues were collected from Prox1-CreERT2; R26R^tdT mice at 10 dpi. (c) Identification of 6 distinct LEC clusters (C1–C6) based on gene expression profiles. (d, e and f) Expression of tdTomato, Prox1 and Vegfr3 in distinct LEC clusters. (g) Gene Ontology (GO) analysis of signaling pathways enriched in cluster C1. −log10 (Q value) indicates the significance of enrichment. (h) The distribution of Pd-l1 expression across LEC populations.

Figure 4.

Pd-l1 conditional deletion leads to increased lymphatic vessels following viral infection. (a) H&E staining of lung sections from control (Prox1-Cre-ER; Pd-l1^loxp/+) and Pd-l1^ΔLEC mice at 28 dpi. (b) Quantification injury areas across the whole lungs between control and Pd-l1^ΔLEC mice (mean ± SEM, n = 5 for control, n = 3 for Pd-l1^ΔLEC, n.s.). (c) Body weight change of the mice following viral infection (mean ± SEM, n = 3 for control, n = 5 for Pd-l1^ΔLEC, p=0.322). (d) Immunofluorescence staining for VEGFR3 (red) and SPC (green) in the lungs at 28 dpi. SPC was used to determine alveolar regeneration. (e) Quantification of lymphatic vessels per field (0.339 mm²) (mean ± SEM, n = 28 for control, n = 25 for Pd-l1^ΔLEC, **p < 0.01). (f) Tidal volumes at 14 dpi (p = 0.7). Scale bars: 3 mm for a (60 μm for insets) and 20 μm for d.