Platelet-derived Wnt antagonist Dickkopf-1 is implicated in ICAM-1/VCAM-1-mediated neutrophilic acute lung inflammation

Abstract

Neutrophil infiltration represents the early acute inflammatory response in acute lung injury. The recruitment of neutrophils from the peripheral blood across the endothelial-epithelial barrier into the alveolar airspace is highly regulated by the adhesion molecules on alveolar epithelial cells (AECs). Wnt/β-catenin signaling is involved in the progression of inflammatory lung diseases including asthma, emphysema, and pulmonary fibrosis. However, the function of Wnt/β-catenin signaling in acute lung inflammation is unknown. Here, we identified platelet-derived Dickkopf-1 (Dkk1) as the major Wnt antagonist contributing to the suppression of Wnt/β-catenin signaling in AECs during acute lung inflammation. Intratracheal administration of Wnt3a or an antibody capable of neutralizing Dkk1 inhibited neutrophil influx into the alveolar airspace of injured lungs. Activation of Wnt/β-catenin signaling in AECs attenuated intercellular adhesion molecule 1 (ICAM-1)/vascular cell adhesion molecule 1 (VCAM-1)-mediated adhesion of both macrophages and neutrophils to AECs. Our results suggest a role for Wnt/β-catenin signaling in modulating the inflammatory response, and a functional communication between platelets and AECs during acute lung inflammation. Targeting Wnt/β-catenin signaling and the communication between platelets and AECs therefore represents potential therapeutic strategies to limit the damage of acute pulmonary inflammation.

Figure 1

Wnt/β-catenin signaling is inhibited during acute pulmonary inflammation. (A-B) Two-hit LPS and MV model: Mice were treated with LPS and MV (LM) or nonventilated (Control). The mRNA levels (A) of Wnt signaling components in the whole-lung tissues and freshly isolated AECs of LM-treated animals were analyzed by real-time PCR and plotted as a ratio against the control animals, with 18S rRNA as a reference gene. The protein levels (B) of ABC and total β-catenin in the whole-lung tissues of the control and LM-treated animals were determined by western blotting, with β-actin as an internal control. The density of the bands was quantified by ImageJ and normalized to β-actin. The results were expressed as a ratio of the control mice. Data are expressed as the mean ± SEM (n = 5 animals) and statistical significance was determined by Student t test. *P < .05 vs control. (C-D) Influenza viral pneumonia model: Mice were intranasally inoculated with H1N1 influenza A/PR/8/34 virus (250 pfu) and housed for 1 to 7 days (n = 3 per group). The mRNA levels (C) of Wnt signaling components in the whole-lung tissues were measured by real-time PCR and normalized to control animals (without infection) using 18S rRNA as a reference gene. The protein levels (D) of ABC, total β-catenin, and Dkk1 in whole-lung tissue of infected animals were determined by western blotting using β-actin as an internal control. The density of the ABC bands was quantified by ImageJ, normalized to control mice, and labeled on the top. PCR, polymerase chain reaction; rRNA, ribosomal RNA.

Figure 2

Dkk1 is accumulated in the lung during acute pulmonary inflammation. (A-C) Two-hit LPS and MV model. (A) The mRNA expression of Wnt antagonists (sFRP1 and Dkk1-4) in the whole-lung tissues of LM-challenged and control mice was determined by real-time PCR and plotted as the fold change over control animals with 18S rRNA as a reference gene. (B) The protein levels of Dkk1 in the whole-lung tissues of control and LM-challenged animals were measured by western blotting using β-actin as a loading control. The density of the bands was quantified by ImageJ and normalized to β-actin. The result was expressed as a ratio to control animals. (C) The Dkk1 concentrations in the BAL and plasma of LM-challenged mice were measured by ELISA. (D-G) Influenza viral pneumonia model. (D) The mRNA level of Dkk1 in the whole-lung tissues of H1N1 influenza virus PR/8-infected mice for 0 to 7 days was analyzed by real-time PCR and normalized to the control mice with 18S rRNA as a reference gene. (E) The Dkk1 concentrations in the BAL and plasma of H1N1 influenza virus PR/8-infected mice were measured by ELISA. (F) Immunostaining of Dkk1 in the lungs of control, LM-challenged, and H1N1 influenza virus PR/8-infected (7 dpi) mice. Scale bar = 75 µm. (G) The protein levels of T1α (AEC I marker), CD141 (endothelial cell marker), IgM (epithelial-endothelial barrier damage marker), and Dkk1 in the BALs of H1N1 influenza virus PR/8-infected mice (day 0 to day 7) were determined by western blotting. Data are expressed as the mean ± SEM (n = 3-8 animals) and statistical significance is determined by the Student t test. *P < .05 vs control.

Figure 3

Dkk1 is induced in activated platelets during acute pulmonary inflammation-associated thrombocytopenia. (A-B) The circulating platelet count in H1N1 influenza virus PR/8-infected, LM-challenged, and live bacteria E coli–infected mice. (C) Platelets were isolated from H1N1 influenza virus PR/8-infected mice (2 dpi) and labeled with PE-CD62P and FITC-CD41 (1 μg/mL of each). Surface expression of CD62P (activation marker) was detected by FACS. (D-F) Protein levels of Dkk1 and CD41 (platelet marker) in circulating platelets in LM-challenged, live bacteria E coli–infected, and PR/8-infected mice were measured by western blot with β-actin as a loading control. Data shown are the mean ± SEM (n = 3-8 animals) and statistical significance is tested by the Student t test. *P < .01 vs control mice.

Figure 4

Dkk1 is released from activated platelets and binds to AECs during acute pulmonary inflammation. PRP, isolated from LM-challenged mice, was incubated with 100 μM SFLLRN for 20 minutes (A) or 5 to 20 minutes (B) and then centrifuged. The protein levels of Dkk1 and CD41 in the pellet were determined by western blotting (A) and the Dkk1 level in the supernatant was determined by ELISA (B). (C) Primary mouse AEC II was cultured for 4 days and then stimulated with TNFα and IL-1β (20 ng/mL of each) for 24 hours. Recombinant mouse Dkk1 was added to the cells at the indicated concentrations for 2 hours. Cell-based ELISA was performed to quantify the binding of Dkk1 to the stimulated AECs. (D) The mRNA levels of the Dkk1 receptors (KRM1, KRM2, LRP5, and LRP6) were analyzed by real-time PCR in the TNFα/IL-1β–stimulated AECs and normalized to control. Data are expressed as the mean ± SEM from 3 independent preparations.

Figure 5

Intratracheal instillation of anti-Dkk1 antibody or Wnt3a attenuates neutrophil infiltration during LM-induced acute pulmonary inflammation. Mice were intratracheally instilled with LPS together with isotype-matched control IgG or function-blocking monoclonal antibody against Dkk1 (86 µg/mouse) and then mechanically ventilated (LM). Mice were also intratracheally coadministered with control (Con_CM) or Wnt3a-conditioned medium (Wnt3a_CM) along with LPS and ventilated as above. (A) Differential immune cell number in the BAL was counted. The mRNA levels of ICAM-1 (B), VCAM-1 (C), and IL-6 (E) in the whole-lung tissues were measured by real-time PCR and were normalized to blank control. (D) Concentrations of IL-6 in BAL were measured by ELISA. Data are expressed as the mean ± SEM (n = 5-8 animals). Statistical significance was determined by 1-way ANOVA analysis with post hoc Tukey test. *, **, and # represent P < .05.

Figure 6

The platelet-derived Wnt antagonist Dkk1 is implicated in the stimulation of TNFα-mediated ICAM-1/VCAM-1 expression in AECs. E10 cells were incubated with 50% Con_CM or Wnt3a_CM together with 200 μM Dkk1 for 12 hours and then stimulated with 20 ng/mL TNFα for 2 hours (A-B) or 24 hours (C). Platelets were isolated from LM-challenged mice, resuspended in Tyrode-HEPES buffer, and treated with 0.1 U/mL thrombin for 90 minutes. Platelet releasate (PR) was collected and added to cultured primary mouse AECs together with isotype-matched control IgG or a function-blocking monoclonal antibody against Dkk1 (10 μg/mL) for 24 hours. After that, mouse AECs were stimulated with TNFα (20 ng/mL) for 2 hours (D-E). The mRNA levels of ICAM-1 (A,D) and VCAM-1 (B,E) were measured by real-time PCR using mouse 18S rRNA as a reference gene and normalized to control untreated cells. (C) Protein levels of ICAM-1, VCAM-1, and ABC were determined by western blotting with β-actin as an internal control. Data are expressed as the mean ± SEM from 3 independent preparations and statistical significance was tested with ANOVA analysis followed by post hoc Tukey test. *, **, ***, and # represent P < .05.

Figure 7

Wnt/β-catenin signaling reduces ICAM-1/VCAM-1–mediated adhesion of alveolar macrophages and neutrophils to AECs. E10 cells were incubated with 50% Con_CM or Wnt3a_CM together with 200 μM Dkk1 for 12 hours and then stimulated with 20 ng/mL TNFα for 24 hours, or E10 cells were incubated with control IgG or blocking mAbs against ICAM-1 or VCAM-1 for 30 minutes and then stimulated with 20 ng/mL TNFα for 24 hours. Next, adhesion of freshly isolated AMϕ or neutrophils to E10 cells was assayed. The values are presented as a percentage of attached AMϕ or neutrophils from 3 independent experiments and are tested for statistical significance by ANOVA analysis followed by the post hoc Tukey test. Data are expressed as the mean ± SEM. *, **, ***, #, ##, and ### represent P < .05.