Secreted Extracellular Cyclophilin A Is a Novel Mediator of Ventilator-induced Lung Injury

Abstract

Rationale: Mechanical ventilation is a mainstay of intensive care but contributes to the mortality of patients through ventilator-induced lung injury. eCypA (extracellular CypA [cyclophilin A]) is an emerging inflammatory mediator and metalloproteinase inducer, and the gene responsible for its expression has recently been linked to coronavirus disease (COVID-19). Objectives: To explore the involvement of eCypA in the pathophysiology of ventilator-induced lung injury. Methods: Mice were ventilated with a low or high Vt for up to 3 hours, with or without blockade of eCypA signaling, and lung injury and inflammation were evaluated. Human primary alveolar epithelial cells were exposed to in vitro stretching to explore the cellular source of eCypA, and CypA concentrations were measured in BAL fluid from patients with acute respiratory distress syndrome to evaluate the clinical relevance. Measurements and Main Results: High-Vt ventilation in mice provoked a rapid increase in soluble CypA concentration in the alveolar space but not in plasma. In vivo ventilation and in vitro stretching experiments indicated the alveolar epithelium as the likely major source. In vivo blockade of eCypA signaling substantially attenuated physiological dysfunction, macrophage activation, and MMPs (matrix metalloproteinases). Finally, we found that patients with acute respiratory distress syndrome showed markedly elevated concentrations of eCypA within BAL fluid. Conclusions: CypA is upregulated within the lungs of injuriously ventilated mice (and critically ill patients), where it plays a significant role in lung injury. eCypA represents an exciting novel target for pharmacological intervention.

Keywords: acute respiratory distress syndrome; animal model; cyclosporin; matrix metalloproteinase; mechanical ventilation.

Figure 1.

eCypA (extracellular CypA [cyclophilin A]) concentration and correlations. (A) eCypA concentrations were determined in lung lavage fluid at the end of experiments lasting 1, 2, or 3 hours (or when mortality surrogates were reached if at a time point earlier than 3 h). Concentrations were shown to be correlated with markers of injury; that is, (B) Pa_O2, (C) lung lavage total protein, and (D) soluble RAGE. (E) eCypA concentrations in plasma were also determined at the end of experiments lasting 1, 2, or (up to) 3 hours. (F) AM cell-surface expression of the principal eCypA receptor CD147 was determined by flow cytometry of lung-cell suspensions at either 1 or 2 hours of ventilation. All data were determined to be nonnormally distributed and are displayed as individual data points, with solid lines indicating median values. Data in A, E, and F were evaluated by using a Kruskal-Wallis test followed by a Dunn multiple comparisons test. N = 5–6 for A and E, and N = 4–5 for F. *P < 0.05 and **P < 0.01. In B, C, and D, associations were evaluated by using the nonparametric Spearman correlation (r and P values given in figure). AM = alveolar macrophage; MFI = mean fluorescence intensity.

Figure 2.

Source of eCypA (extracellular CypA [cyclophilin A]). Mice were ventilated with a high Vt for 3 hours (or until mortality surrogates were met), and intracellular expression of CypA were determined by using flow cytometry on (A) type 1 alveolar epithelial cells (AECs) and (B) AMs and compared with expression in nonventilated (control) mice. Data are expressed as the MFI. Human AECs in culture were exposed to 6 hours of 20% stretching or were held static, and (C) concentrations of secreted IL8 were determined by using an ELISA. (D) Amounts of secreted eCypA were determined in cell supernatants by using Western blotting. The upper portion of D shows a representative blot with lanes from left to right showing 1) 20-kD molecular-weight marker, 2) stretch, 3) static, 4) stretch, 5) static, and 6) stretch. The lower portion of D shows the relative band intensity normalized to a “standard” sample loaded onto every gel. (E) Positive cell staining for 7-AAD and (F) supernatant concentration of LDH (lactate dehydrogenase) were determined to evaluate cell viability. Data in all panels except E were determined to be normally distributed and were evaluated by using a Student’s t test. Data in E were evaluated by using a Mann-Whitney test. Data are displayed as individual points, with solid lines indicating either the mean or median value for normal or nonnormal data, respectively. N = 6 for each data set. *P < 0.05 and **P < 0.01. 7-AAD = 7-aminoactinomycin D; AM = alveolar macrophage; MFI = mean fluorescence intensity.

Figure 3.

Biological effect of CypA (cyclophilin A) on macrophages in vitro. Human monocyte–derived macrophages were stimulated with 1 μg/ml of recombinant human CypA for 6 hours in vitro. Cell-surface expression of (A) ICAM-1, (B) CD86, (C) MHC-II, and (D) CD206 were determined by using flow cytometry. (E) Concentrations of secreted IL6 within culture supernatants were determined by using an ELISA. (F) Cell viability was evaluated via 7-AAD staining by using flow cytometry. Data in A, D, E, and F were determined to be normally distributed and were evaluated by using a Student’s t test, whereas data in B and C were nonnormal and were evaluated by using a Mann-Whitney test. Data are displayed as individual data points, with solid lines indicating the mean or median value for normal or nonnormal data, respectively. N = 6–7 for each data set. *P < 0.05, ***P < 0.001, and ****P < 0.0001. 7-AAD = 7-aminoactinomycin D; MFI = mean fluorescence intensity.

Figure 4.

Biological effect of CypA (cyclophilin A) on alveolar epithelial cells (AECs) in vitro. Human primary AECs were stimulated with 1 μg/ml of recombinant human CypA for 6 hours in vitro. (A) Cell-surface expression of ICAM-1 was determined by using flow cytometry. Concentrations of secreted (B) IL8 and (C) IL6 within culture supernatants were determined by using an ELISA. (D) Cell viability was evaluated via 7-AAD staining by using flow cytometry. Data in A and D were determined to be normally distributed and were evaluated by using a Student’s t test, whereas data in B and C were nonnormal and were evaluated by using a Mann-Whitney test. Data are displayed as individual points, with solid lines indicating either the mean or median value for normal or nonnormal data, respectively. The dashed lines represent positive control data from cells stimulated with 1 ng/ml of TNF (tumor necrosis factor) to demonstrate that cells were functionally responsive. N = 6 for each data set. 7-AAD = 7-aminoactinomycin D; MFI = mean fluorescence intensity.

Figure 5.

Physiological effects of eCypA (extracellular CypA [cyclophilin A]) inhibition during ventilator-induced lung injury in vivo. Mice were ventilated for up to 3 hours with a high Vt after intratracheal delivery of either the eCypA inhibitor MM-284 or vehicle (10% DMSO in saline). Changes in the PIP were determined (A) over time and (B) at the endpoint. Note that the “end” reflects the final measurement taken, regardless of the length of ventilation, as some animals in the vehicle group met the mortality surrogate before 3 hours. (C) Survival was evaluated over the time of ventilation. Changes in (D) Ers and (E) Rrs were calculated as the percentage increase at the end versus the start of ventilation. (F) Pa_O2 was evaluated over time, whereas the (G) lung wet/dry weight ratio, (H) lung lavage-fluid protein, and (I) lung lavage-fluid albumin were measured as endpoint markers of edema and permeability. Data in A, B, D, F, G, and H were determined to be normally distributed, whereas data in E and I were nonnormally distributed. A and F show time courses as the mean ± SD, and data were evaluated by using two-way ANOVA, with significant differences shown as the time by treatment group interaction. Survival in C was evaluated by using a log rank test. Data in other panels are displayed as individual points, with solid lines indicating the mean or median value for normal or nonnormal data, respectively. Data in B, D, G, and H were evaluated by using a Student’s t test, whereas data in E and I were evaluated by using a Mann-Whitney test. N = 7–8 for each data set. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Ers = respiratory system elastance; PIP = peak inspiratory pressure; Rrs = respiratory system resistance.

Figure 6.

Antiinflammatory effects of eCypA (extracellular CypA [cyclophilin A]) inhibition during ventilator-induced lung injury in vivo. Mice were ventilated for up to 3 hours with a high Vt after intratracheal delivery of either the eCypA inhibitor MM-284 or the vehicle (10% DMSO in saline). Lavage-fluid concentrations of (A) CXCL1, (B) IL6, (C) IL1β, and (E) soluble RAGE were determined by using an ELISA. (D) Cell-surface expression of ICAM-1 on AMs and type 1 AECs was determined by using flow cytometry of lung-cell suspensions. The numbers of neutrophils and inflammatory Ly6C⁺ monocytes were quantified in (F) lavage fluid and (G) lung tissue by using flow cytometry. (H) The amount of active MMP2 (matrix metalloproteinase 2) in lavage fluid was determined by using gelatin zymography. The upper portion of H shows representative gel with lanes from left to right showing the 1) 57-kD molecular-weight marker, 2–5) high Vt + vehicle, 6–8) high Vt + MM-284, and 9) MMP2 standard. The lower portion of H shows the relative band intensity normalized to a “standard” sample loaded onto every gel. Data in all panels except H were determined to be nonnormally distributed and were evaluated by using a Mann-Whitney test or Kruskal-Wallis test followed by a Dunn multiple comparisons test (D, F, and G). Data in H were evaluated by using a Student’s t test. Data are displayed as individual points, with solid lines indicating the mean or median value for normal or nonnormal data, respectively. N = 6–8 for each data set. *P < 0.05, **P < 0.01, and ***P < 0.001. AEC = alveolar epithelial cell; AM = alveolar macrophage.

Figure 7.

Human data. eCypA (extracellular CypA [cyclophilin A]) concentrations were determined by using an ELISA in randomly selected baseline BAL-fluid samples from patients with acute respiratory distress syndrome within the control arms of the previously published HARP (Hydroxymethylglutaryl–Coenzyme A Reductase Inhibition for Acute Lung Injury) and KARE (Keratinocyte Growth Factor for the Treatment of the Acute Respiratory Distress Syndrome) studies and were compared with samples from healthy volunteers. Data were assumed to be nonnormally distributed and are displayed as individual data points, with lines indicating the median value. Data were evaluated by using a Kruskal-Wallis test followed by a Dunn multiple comparisons test. **P < 0.01.