Soluble PD-L1 improved direct ARDS by reducing monocyte-derived macrophages

Abstract

Acute respiratory distress syndrome (ARDS) is common in intensive care units (ICUs), although it is associated with high mortality, no effective pharmacological treatments are currently available. Despite being poorly understood, the role of programmed cell death protein 1 (PD-1) and PD-ligand 1 (PD-L1) axis in ARDS may provide significant insights into the immunosuppressive mechanisms that occur after ARDS. In the present study, we observed that the level of soluble PD-L1 (sPD-L1), a potential activator of the PD-1 pathway, was upregulated in survivors of direct ARDS than in non-survivors. Administration of sPD-L1 in mice with direct ARDS relieved inflammatory lung injury and improved the survival rate, indicating the protective role of sPD-L1 in direct ARDS. Using high-throughput mass cytometry, we found a marked decrease in the number of lung monocyte-derived macrophages (MDMs) with proinflammatory markers, and the protective role of sPD-L1 diminished in ARDS mice with monocyte/macrophage depletion. Furthermore, PD-1 expression increased in the MDMs of patients and mice with direct ARDS. Finally, we showed that sPD-L1 induced MDM apoptosis in patients with direct ARDS. Taken together, our results demonstrated that the engagement of sPD-L1 on PD-1 expressing macrophages resulted in a decrease in pro-inflammatory macrophages and eventually improved direct ARDS. Our study identified a prognostic indicator for patients with direct ARDS and a potential target for therapeutic development in direct ARDS.

Fig. 1. Low levels of serum sPD-L1 indicated more severe disease and predicted worse prognosis in patients with direct ARDS.

A Serum sPD-L1 levels in healthy controls (n = 10), Non-ARDS patients with mechanical ventilation (n = 20) and ARDS patients (n = 44). B sPD-L1 levels of survivors (n = 20) and non-survivors (n = 10) of direct ARDS. C Correlations between APACHEII score and serum sPD-L1 levels in patients with direct ARDS (n = 30). D Correlations between oxygenation index and the levels of serum sPD-L1 in patients with direct ARDS (n = 30). *P < 0.05, **P < 0.01, ***P < 0.001, P value was determined by unpaired t test. ^####P < 0.01, P value was determined by Man–Whitney test.

Fig. 2. Administration of sPD-L1 attenuated inflammatory lung injury and improved survival rate in mice with direct ARDS.

A Time course of protein concentration in bronchoalveolar lavage fluid (BALF) of ARDS mice (n = 5/each group). B Time course of cell counts in BALF of ARDS mice (n = 5/each group). C Time course of PD-1 expression in CD45 + lung cells of ARDS mice (n = 5/each group). D PD-1 expression in CD45 + lung cells of ARDS mice at different time points post-infection (n = 5/each group). E Schematic depicting experiment workflow: The mice were intratracheal injected with PAO1(2 × 10⁶ CFU/mL, 50 µL/mouse) to establish the ARDS model, 6 h later the mice were randomly divided into two groups. The mice in sPD-L1 group were intravenous injected with 20 µg/100 µl sPD-L1 protein while those in PBS group were administrated with 100 μL PBS. The mortality was monitored for 7 days and. Every 24 h the mice were administrated with another dose of sPD-L1 or PBS. F Kaplan–Meier plot of survival rate of ARDS mice (n = 23/group) *P < 0.05 by log-rank test. G To detect the lung injury after administration of sPD-L1/PBS, mice were sacrificed at 12-h post infection (6-h post administration of sPD-L1/PBS). Protein concentration in BALF of Sham mice, IgG/PBS-treated ARDS mice, and sPD-L1-treated ARDS mice (n = 10–11/each group). H Cell counts of BALF of Sham mice, IgG/PBS-treated ARDS mice, and sPD-L1-treated ARDS mice (n = 10–11/each group). I Wet to dry ratio of the left lung in Sham mice, IgG/PBS-treated ARDS mice, and sPD-L1-treated ARDS mice (n = 10–11/each group). J HE staining of the left lung in control, IgG/PBS-treated ARDS mice, and sPD-L1-treated ARDS mice (n = 10–11/each group) The experiments have been repeated for at least three times. Each value represents the mean ± SEM of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, analyzed by t test.

Fig. 3. CyTOF revealed that administration of sPD-L1 reduced monocyte-derived macrophages.

A The frequency of common immune cells in CD45+live cells by manual gating. B Gate strategy for monocyte-derived macrophages in one of the samples. C The frequency of significant reduced cluster in CD45+live cells (cluster 44). D The viSNE plot layered with the significant reduced cluster analyzed by X shift (n = 4/each group) (cluster 44). E The viSNE plots of expression of the markers that were highly expressed in the significantly changed clusters (cluster 44). The color gradient (from low (blue) to high (red)) indicates the intensity of the markers. F The viSNE plots of the markers of M1 like and M2 like macrophages in cluster 44. P value was analyzed by Man–Whitney test.

Fig. 4. Flow cytometry verified that administration sPD-L1 reduced lung monocyte-derived macrophages.

A Gate strategy for dendritic cells, monocyte-derived macrophages, neutrophils and alveolar macrophages (MDMs were measured in lung tissue single-cell suspension while alveolar macrophages were assayed in BALF) (n = 9–12/each group). B Percentage of different cell populations in CD45-positive cells of lung or BALF (n = 9–12/each group). C Percentage of different cell populations in CD45-positive cells of blood. D PD-1 expression on MDMs of lung (n = 9–12/each group). E Representative histograms of PD-1 expression on MDM of lung. The cells were previously gated by CD45⁺CD11b⁺F4/80⁺ (n = 3–6/each group). The experiments have been repeated for at least three times. Each value represents the mean ± SEM of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, analyzed by t test. MDM monocyte-derived macrophages, Neu, neutrophil.

Fig. 5. The protective role of sPD-L1 was diminished in ARDS mice with monocytes/macrophages depletion.

A The flow cytometry of monocytes and macrophages in blood, lung and BALF of CL or L injected mice (n = 5/group). B The statistical graph of the frequency in CD45⁺cells of the cells in graph A (n = 5/group). C The protein concentration in BALF of mice with different treated methods (n = 5/group). D The cell counts in BALF of mice with different treated methods (n = 5–6/group). E) The wet to dry ratio of lung tissue of mice with different treated methods (n = 5–6/group). F The HE staining of lung tissue from different treated mice (n = 5–6/group). Each value represents the mean ± SEM of three independent experiments. The experiments have been repeated for at least three times. *P < 0.05, **P < 0.01, ***P < 0.001, analyzed by t test. Mo/Ma, monocyte/macrophage, MDM monocyte-derived macrophages, Neu, neutrophil.

Fig. 6. Administration of sPD-L1 induced apoptosis of monocyte-derived macrophages.

A Representative histograms of PD-1 expression on monocyte-derived macrophages of healthy controls and ARDS patients. B PD-1 expression on MDM of healthy controls, Non-ARDS patients and ARDS patients. (n = 5/group) C The apoptosis of macrophages by PI/Annexin-V staining. D Apoptosis of macrophages in healthy controls, Non-ARDS patients and ARDS patients. (n = 5/group), analyzed by annexin V-fluorescein isothiocyanate/PI double staining. Cells in the B2 and B4 quadrants(annexin V + /PI+ and annexin V + /PI−, respectively) were considered to be apoptotic. D Graphical representation of apoptosis (n = 5/group). Each value represents the mean ± SEM of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, analyzed by t test. MDM monocyte-derived macrophages; HC healthy controls.

Fig. 7. A schematic diagram explaining the role of sPD-L1 in direct ARDS.

In the setting of direct ARDS, soluble PD-L1 binded to the monocytederived dmacrophages with increased PD-1 expression and induced their apoptosis. The reduced MDMs alleviated the inflammatory lung injury and ultimately improved direct ARDS.