Female mice lacking Pald1 exhibit endothelial cell apoptosis and emphysema

Abstract

Paladin (Pald1, mKIAA1274 or x99384) was identified in screens for vascular-specific genes and is a putative phosphatase. Paladin has also been proposed to be involved in various biological processes such as insulin signaling, innate immunity and neural crest migration. To determine the role of paladin we have now characterized the Pald1 knock-out mouse in a broad array of behavioral, physiological and biochemical tests. Here, we show that female, but not male, Pald1 heterozygous and homozygous knock-out mice display an emphysema-like histology with increased alveolar air spaces and impaired lung function with an obstructive phenotype. In contrast to many other tissues where Pald1 is restricted to the vascular compartment, Pald1 is expressed in both the epithelial and mesenchymal compartments of the postnatal lung. However, in Pald1 knock-out females, there is a specific increase in apoptosis and proliferation of endothelial cells, but not in non-endothelial cells. This results in a transient reduction of endothelial cells in the maturing lung. Our data suggests that Pald1 is required during lung vascular development and for normal function of the developing and adult lung in a sex-specific manner. To our knowledge, this is the first report of a sex-specific effect on endothelial cell apoptosis.

Figure 1

Paladin has a broad expression in the postnatal lung. (a) Endothelial cells and pneumocytes type II were isolated from lungs of mT/mG mice expressing endothelial-specific Tie2-Cre (left) or pneumocyte type II-specific SPC-Cre (right). Q-PCR of sorted single cells indicates Pald1 mRNA expression in both endothelial cells and non-endothelial cells, including pneumocytes type II, both at 3 weeks and 3 months of age. Expression was normalized to 18 S RNA. (b) Pald1 LacZ reporter activity (blue) is detected in Pald1 ^LacZ/LacZ mice, 5 days, 4 weeks and 19 weeks after birth. LacZ is broadly expressed in the lung tissue, except for the bronchial epithelium, which shows no reporter activity (arrow). Scale bar = 200 µm.

Figure 2

Paladin is expressed both in the epithelial and mesenchymal compartment of the postnatal lung (4 weeks). (a–e) Combined X-gal (black) and immunofluorescence staining of Pald1 ^LacZ/LacZ mice show Pald1 LacZ expression in the vasculature, i.e. endothelial cells in capillaries (a) but to a lesser extent in endothelial cells of larger blood vessels (b) as indicated by Erg staining (endothelial cell nuclei, green). In large blood vessels LacZ expression can be detected in vascular smooth muscle cells (c, α-smooth muscle actin, red). Paladin LacZ reporter is also active in pneumocytes type II (d, SPC, green) and pneumocytes type I/II (e, cytokeratin, red). Scale bar = 20 µm.

Figure 3

Pald1 ^−/− mice show reduced complexity of the lung tissue. (a,b) Dark field images of distal airspace from 4 weeks old female Pald1 wild type (a) and knock-out animals (b), suggest a general airspace enlargement in the Pald1 knock-out lungs. Scale bars = 200 µm. (c) Quantification of interseptal alveolar distance using Mean Linear Intercept (MLI) of hematoxylin-eosin stained lung tissue at P5 (n = 3), 4 (n = 5) and 19 weeks (n = 4) shows increased air spaces in female knock-out animals at all stages, but was particularly increased at 4 weeks. ANOVA per age group: P5 p < 0.0001; 4 weeks p = 0.0002; 19 weeks p = 0.0004. Pald1 ^−/− was compared to Pald1 ^+/+ of the same sex within each age group. (d) MLI of Pald1 wild type, heterozygous and knock-out female mice at the indicated ages. ANOVA per age group: 4 weeks p = 0.0001, 19 weeks p = 0.0007. Pald1 ^−/− and Pald1 ^+/− were compared to Pald1 ^+/+ within each age group. Error bars: SD, *p ≤ 0.05, ***p ≤ 0.001, ****p ≤ 0.0001.

Figure 4

Pald1 ^−/− show a decrease in the endothelial cell population at 4 weeks. (a–e) Quantification of the relative proportion of endothelial cells (a,b, Erg, n = 3–5), pneumocytes type II (c, SPC, n = 3), pneumocytes type I/II (d, cytokeratin, n = 3), and macrophages (e, CD68, n = 3–7). The specific cell type contribution was unaltered in Pald1 ^−/− mice except for a 15% decrease of the endothelial cell population at 4 weeks in Pald1 ^−/− mice. Scale bar = 20 µm, Error bars: SD, t-test between genotypes of each age group. ***p ≤ 0.001.

Figure 5

Increased apoptosis and proliferation of endothelial cells in Pald1 knock-out female lungs. (a,b) Quantification of cleaved caspase-3 positive (CC3+) cells in 4 and 19-week old lungs revealed a significant increase in the number of cleaved caspase-3 positive endothelial cells (Erg positive) at 4 weeks of age and at 19 weeks in female (a, n = 3–4), but not in male Pald1 ^−/− mice (b, n = 3–4). There was no significant increase in non-endothelial cells (Erg negative) at both 4 and 19 weeks. (c,d) Quantification of Ki67 positive cells in 4 and 19-week old lungs revealed an increased number of Ki67-positive endothelial cells (Erg positive) in female (c, n = 3–4), but not in male Pald1 ^−/− mice (d, n = 3–4). There was no significant increase in non-endothelial cells (Erg negative) at both 4 and 19 weeks. Error bars: SD, t-test between genotypes within each age group. * = p ≤ 0.05, **** = p ≤ 0.0001.

Figure 6

Proteomics data overview and significantly differentially expressed proteins. (a) Proteomics reveals sex-dependent protein expression differences in lung tissue. The heatmap shows proteome data overview of 8635 proteins with overlapping quantification in all 4 weeks of age lung samples using hierarchical clustering. Columns and rows represent lung samples and proteins, respectively. The samples are labels as sex_genotype_ID. The major factor for clustering is sex, and not genotype. (b) Boxplot of the three significantly differentially expressed proteins. The y-axis shows the ratios to the pool. For each gene, the ratios to the pool in the wild type group (WT) and knock-out group (KO) are plotted in the boxplot, respectively.