Detection and quantification of epithelial progenitor cell populations in human healthy and IPF lungs

Abstract

Background: In the human lung, epithelial progenitor cells in the airways give rise to the differentiated pseudostratified airway epithelium. In mice, emerging evidence confers a progenitor function to cytokeratin 5 (KRT5(+)) or cytokeratin 14 (KRT14(+))-positive basal cells of the airway epithelium. Little is known, however, about the distribution of progenitor subpopulations in the human lung, particularly about aberrant epithelial differentiation in lung disease, such as idiopathic pulmonary fibrosis (IPF).

Methods: Here, we used multi-color immunofluorescence analysis to detect and quantify the distribution of airway epithelial progenitor subpopulations in human lungs obtained from healthy donors or IPF patients.

Results: In lungs from both, healthy donors and IPF patients, we detected KRT5(+)KRT14(-), KRT5(-)KRT14(+) and KRT5(+)KRT14(+) populations in the proximal airways. KRT14(+) cells, however, were absent in the distal airways of healthy lungs. In IPF, we detected a dramatic increase in the amount of KRT5(+) cells and the emergence of a frequent KRT5(+)KRT14(+) epithelial population, in particular in distal airways and alveolar regions. While the KRT14(-) progenitor population exhibited signs of proper epithelial differentiation, as evidenced by co-staining with pro-SPC, aquaporin 5, CC10, or MUC5B, the KRT14(+) cell population did not co-stain with bronchial/alveolar differentiation markers in IPF.

Conclusions: We provide, for the first time, a quantitative profile of the distribution of epithelial progenitor populations in human lungs. We show compelling evidence for dysregulation and aberrant differentiation of these populations in IPF.

Keywords: Basal cells; Epithelium; Human explants; Progenitor cells; Subpopulations.

Fig. 1

Proximal-to-distal distribution of KRT5⁺ and KRT14⁺ basal cells subpopulations in the healthy human lung. a, b Sections from healthy human lungs (6 donors) were co-stained for Keratin (KRT) 5 (green), KRT14 (red) and counterstained with DAPI (blue). Two representative pictures, with corresponding high magnification panels, are shown for the conducting airways (a left and right panels) and the distal airways (b left and right panels). Arrow: KRT5⁺ cell with characteristic luminal epithelial cell morphology. Full bars = 500 μm. Scattered bars = 100 μm. c Isolated primary human bronchial epithelial cells (hBEC) were co-stained for KRT 5 (green), KRT14 (red) and counterstained with DAPI (blue). The panels show: Keratin5-DAPI (left), Keratin14-DAPI (middle) and merged (right) images. Full bars = 500 μm. Scattered bars = 100 μm

Fig. 2

Coexpression of p63 with KRT5 and KRT14 in the healthy human lung. Sections from healthy human lungs (6 donors) were co-stained either for KRT5 (green) and p63 (red) (a left; b left), or KRT14 (red) and p63 (green) (a right; b right) and counterstained with DAPI (blue). One representative picture, with the corresponding high magnification panel, is shown for each staining, for the conducting (a) and distal (b) healthy airways. Arrow: KRT5^-p63⁺ cell. Full bars = 500 μm. Scattered bars = 100 μm

Fig. 3

Quantification of total basal cells and KRT5⁺ and KRT14⁺ basal cell subpopulations in the healthy human lung. Cells were counted according to their positivity for p63, KRT5 or KRT14 immunostaining in conducting airways (1 donor, 2 samples, n = 7 assessed fields/sample) and in distal airways (5 donors, 1-3 samples/donor, n = 34 assessed bronchioles). Data are expressed as mean ± SEM. A 2-Way ANOVA with Bonferroni post-test revealed a significant interaction according to the type of airway (Conducting versus distal, p < 0.0001); all 6 quantified populations were significantly different between conducting and distal airways (***p < 0.001). Each dataset (conducting and distal) was then compared separately with a One-Way ANOVA test: ***p < 0.001 versus “Conducting KRT5⁺”, ^### p < 0.001 versus “Distal KRT5⁺”

Fig. 4

KRT5⁺ and KRT14⁺ basal cell subpopulations in the IPF lung. Sections from IPF human lungs (5 patients) were co-stained for KRT5 (green), KRT14 (red) and counterstained with DAPI (blue). a Two representative pictures, with corresponding high magnification panels, are shown for the conducting airways. b 4 representative pictures, with corresponding high magnification panels, illustrate the distribution patterns (metaplastic, simple and pod-like) of KRT5⁺ and KRT14⁺ basal cell subpopulations in the distal IPF lung. Full bars = 500 μm. Scattered bars = 100 μm

Fig. 5

Coexpression of p63 with KRT5 and KRT14 in the IPF lung. a, b Sections from IPF human lungs (5 patients) were co-stained either for KRT5 (green) and p63 (red) (a left; b left), or KRT14 (red) and p63 (green) (a right; b right) and counterstained with DAPI (blue). One representative picture, with the corresponding high magnification panel, is shown for each staining, for the conducting (a) and distal (b) healthy airways. c Characteristic KRT5⁺ pod, stained for KRT5 (green) and p63 (red). Arrow: KRT5⁺p63^- cell. Full bars = 500 μm. Scattered bars = 100 μm

Fig. 6

Quantification of total basal cells and KRT5⁺ and KRT14⁺ basal cell subpopulations in the IPF lung. a, b Cells were counted according to their positivity for p63, KRT5 or KRT14 immunostaining in (a) conducting airways (1 donor, 2 samples, n = 7 assessed fields/sample) and (b) in distal regions (5 donors, 1-3 samples/donor, n = 34 assessed fields in bronchiolized regions (striped bars), n = 24 assessed fields in fibrotic regions (black bars), n = 11 assessed fields in non fibrotic regions (white bars)). Data are expressed as mean ± SEM. a A 1-Way ANOVA revealed a significant interaction (p < 0.0001). The data were further analyzed for significance with a Dunn’s post-test (**p < versus “KRT5⁺”). b A 2-Way ANOVA with Bonferroni post-test revealed a significant interaction according to the type of remodeled region (p < 0.0001). Each dataset (bronchiolization, fibrotic, non fibrotic) was then compared separately with a 1-Way ANOVA test, with a Dunn’s post-test: ***p < 0.001 versus the neighboring “Bronchiolization” column, ^### p < 0.001 between “Bronchiolization KRT5⁺” and “Bronchiolization KRT14⁺”

Fig. 7

Costaining of KRT5 and KRT14 with alveolar and bronchial epithelial differentiation markers in the distal IPF lung. Human distal IPF lungs (5 patients) were costained for KRT5 or KRT14 and AQP5 (Alveolar type 1 cell marker), ProSPC (Alveolar type 2 cell marker), acTUB (ciliated cell marker), CC10 (Clara cell marker) and Mucin 5B (secretory cell marker). One picture with the corresponding high magnification is shown for each staining. a Left: KRT5 (green) and AQP5 (red). Right: KRT14 (red) and AQP5 (green). b Left: KRT5 (green) and ProSPC (red). Right: KRT14 (red) and ProSPC (green). c Left: KRT5 (green) and acTUB (red). Right: KRT14 (red) and AQP5 (green). d Left: KRT5 (green) and CC10 (red). Right: KRT14 (red) and CC10 (green). e Left: KRT5 (green) and MUC5B (red). Right: KRT14 (red) and MUC5B (green). Full bars = 100 μm. Scattered bars = 50 μm

Fig. 8

Costaining of KRT5 with developmental pathway markers in the distal IPF lung. a, b, c, d Sections from IPF human lungs (5 patients) were co-stained for KRT5 (green) and (a) SHH (red), (b) GLI1 (red), (c) SOX9 (red), and (d) HES1 (red), and counterstained with DAPI (blue). One representative picture, with the corresponding high magnification panel, is shown for each staining. Full bars = 100 μm. Scattered bars = 50 μm