Effect of extracellular adenosine triphosphate hydrolysis by apyrase on bleomycin-induced circulating and alveolar mononuclear phagocyte activation and lung inflammation.

Abstract

To the Editor: Lung fibrosis is characterized by extracellular matrix accumulation and remodeling of the lung interstitium. Although lung fibrosis has been extensively studied, there is still a lack of effective anti-pulmonary fibrosis drugs at present. The pathogenesis of lung fibrosis involves many aspects in which the lung macrophages play a particularly important role.[1] It has been reported that the manipulation of the monocyte/macrophage phenotype switch might be a potential target for many macrophage-mediated inflammatory disorders. In our previous study, mice that were exposed to bleomycin (BLM) showed a dynamic change of mononuclear phagocytes in the circulating system, lung alveoli, and interstitial compartments. The rapid increase of the number of circulating Ly6Chi monocytes after BLM stimulation, followed by the expansion of M2-like alveolar macrophages (AMϕ) numbers, are closely associated with lung inflammatory response and fibrosis.[2,3] Moreover, the level of adenosine triphosphate (ATP), which is a well-known dangerous signal molecule with pro-inflammatory properties, increased in bronchoalveolar lavage fluid (BALF) of patients with idiopathic pulmonary fibrosis as well as BLM-treated mice. The effects of ATP are mediated through P2 purinergic receptors (P2X7R), which are widely expressed on the surface of immune cells and other tissues. Experimental studies have shown that the elevating pulmonary ATP levels after exposure to lung injury can result in the inflammation and vascular leakage increase.[4] To our knowledge, the effects of ATP on the monocyte/macrophage phenotype during pulmonary inflammation, have not been adequately investigated. Our study hypothesizes that apyrase, an enzyme that hydrolyzes both ATP and adenosine diphosphate to adenosine monophosphate, might influence the polarization of lung macrophage and circulating monocytes to alleviate BLM-induced acute lung injury and fibrosis. The study was approved by local Ethics Committees (Characteristic Medical Center of Chinese Peoples Armed Police Force), and all animal care and experiments were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee of Characteristic Medical Center of Chinese Peoples Armed Police Force. A total of 200 male C57BL/6 mice of 8 to 10 weeks and weighing between 18 to 20 g, were randomly divided into four groups: normal saline (NS) only, BLM only, BLM plus NS (BLM+NS), and BLM plus apyrase (BLM+APY) (n = 50 for each group). Five animals were housed per polycarbonate cage under a controlled temperature of 22°C to 25°C, humidity of 45% to 50%, and a 12:12 hour light to dark cycle. All animals were adaptively housed for one week prior to the start of the experiment. In order to induce acute pulmonary fibrosis, BLM (Taihe Pharmaceutical Company, Tianjin, China) or saline was administered by oropharyngeal aspiration under light anesthesia with 2% isoflurane (Hebei Yipin Pharmaceutical Company, Shijiazhuang, Hebei, China). The BLM was dissolved in 40 μL saline for administration at a dosage of 2.5 mg/kg bodyweight. Apyrase (Sigma–Aldrich Company, St. Louis, Missouri, USA) was dissolved in 20 μL saline and delivered by oropharyngeal aspiration to the mice with the dosage of 40 mg/kg at 4 hours after BLM challenge. On days 1, 3, 7, 14, and 21, ten mice from each group were took for the following test [Figure 1A]. Half of the mice from each group received bronchoalveolar lavage. BALF was collected for ATP concentration detection using an ATP Bioluminescent Assay Kit (Sigma–Aldrich Company, St. Louis, MO, USA) by spectrophotometer (PerkinElmer Company, Hopkinton, MA, USA), alveolar macrophage phenotypic analysis (7-aminoactinomycin D [7-AAD], peridinin-chlorophyll-protein complex [Per CP]/Cy5.5 anti-CD64, phycoerythrin [PE] anti-CD206, Biolegend Company, San Diego, CA, USA) using flow cytometry (Beckman Company, Breya, CA, USA), and total cell count and macrophages, lymphocytes, neutrophils count by hematoxylin and eosin (HE) staining of cell slice. For the rest mice, peripheral blood was collected for monocytes phenotype analysis using flow cytometry (PE/Cy5.5 anti-mouse lymphocyte antigen 6complex, locus G [Ly6G], fluorescein isothiocyanate [FITC] anti-mouse lymphocyte antigen 6 complex, locus C [Ly6C], PE anti-mouse CD11b Biolegend Company, San Diego, CA, USA); and the right lungs were stained with HE and Masson's trichrome for assessing the inflammation score and fibrosis index according to clinically accepted protocol in a blinded fashion. In order to verify the expression of P2X7R in mouse mononuclear phagocytes, we sorted Ly6G-CD11b+ circulating monocytes from mice peripheral blood using flow cytometry (BD Company, Peabody, MA, USA) and AMϕ from BALF using F4/80+ (rat anti-F4/80 monoclonal antibody, Abcam Company, Cambridge Campus, Cambs, UK). Then, the expression of P2X7R in those mononuclear phagocytes was tested by immunofluorescent staining methods.Figure 1: (A) Study protocol. (B) The concertation of ATP in BALF on day 1. (C) The hemotoxylin and eosin (HE) staining of lung tissue on day 7 after BLM challenge (original magnification ×200). (D) Masson's trichrome staining of lung tissue on day 21 (original magnification ×400). (E) Apyrase partially represses alveolar macrophage M2 polarization. The left panel shows the representative pseudo-color plots of flow cytometry analysis of lung alveolar macrophages on day 14. The right panels show the comparisons of lung alveolar M1-like (F4/80+CD11c+CD206-) and M2-like (F4/80+CD11c+CD206+) macrophages across all groups on day 14, respectively. (F) Apyrase decreases circulating Ly6Chi monocytes in BLM challenged mice. Left panel shows the representative flow cytometry analysis (pseudo-color plots) of monocyte subsets. Right panels show the comparisons of Ly6Chi and Ly6Clo monocytes across all groups on day 3, respectively. ∗ P < 0.05; † P < 0.01 (n = 5 for each group). AMϕ: Alveolar macrophage; APY: Apyrase; ATP: Adenosine triphosphate; BALF: Bronchoalveolar lavage fluid; BLM: Bleomycin; CD: cluster of differentiation; Ly6C: Lymphocyte antigen 6 complex, locus C; NS: Normal saline; P2X7R: P2 type 7 purinergic receptors; SSC: Side scatter.Statistical analysis was performed using STATA version 14.1 (STATA Corp, College Station, TX, USA). Data conforming to normal distribution are presented as mean ± standard deviation. One-way analysis of variance followed by Tukey's test was used for comparisons. Data that do not conform to normal distribution are presented as median, the 25th percentile and the 75th percentile. Statistical analysis was performed using Kruskal–Walllis H test, followed by Nemenyi test for multiple comparisons. P < 0.05 indicates a statistically significant difference. According to the Masson's trichrome staining of BLM group on day 21, a mass of blue-dyed portion revealed the excess accumulation of collagen, which proved the successful establishment of lung fibrosis mice model. More than 95% of the sorted cells harvested from mice peripheral blood were Ly6G-CD11b+ monocytes [Supplementary Figure 1A, https://links.lww.com/CM9/B196]. And we got F4/80+ macrophages from BALF. P2X7R expressed in Ly6G-CD11b+ monocytes and F4/80+ macrophages were validated in our study, which was the foundation of our experimental intervention [Supplementary Figure 1, https://links.lww.com/CM9/B196]. Compared with NS (94.49 ± 5.67 nmol/L), the ATP concentration in BALF from the BLM (124.50 ± 15.33 nmol/L), and BLM+NS groups (125.00 ± 16.13 nmol/L) was significantly increased (P = 0.0486, P = 0.0446), whereas the ATP concentration in the BLM+APY group (92.72 ± 6.25 nmol/L) significantly decreased on day 1 compared with BLM+NS group (P = 0.0334). These results confirmed that BLM challenge can generate superfluous ATP in the lung alveoli, and apyrase can effectively decrease the excess ATP [Figure 1B]. According to the HE staining on day 7, the BLM group mice showed extensive alveolar thickening and clear inflammatory infiltrates, and treatment with apyrase markedly reduced the inflammatory response [Figure 1C]. Meanwhile, compared to the BLM+NS group (2.50 ± 0.24), the lung inflammation scores of the BLM+APY group (1.81 ± 0.37) on day 7 were significantly decreased (P = 0.0363). In the meantime, compared with BLM+NS control, the administration of apyrase reduced the number of total cells ([31.20 ± 2.66] × 105 counts/mL vs. [53.60 ± 11.24] × 105 counts/mL, P = 0.0021), macrophages ([24.37 ± 2.52] × 105 counts/mL vs. [42.73 ± 6.10] × 105 counts/mL, P = 0.0001), lymphocytes ([0.42 ± 0.25] × 105 counts/mL [1.20 ± 0.30] × 105 counts/mL, P = 0.0055), and neutrophils ([8.39 ± 1.93] × 105 counts/mL [13.38 ± 4.76] × 105 counts/mL, P = 0.0086) in BALF on day 7 [Supplementary Figure 2, https://links.lww.com/CM9/B196]. Masson's trichrome staining on day 21 showed that treatment with apyrase significantly reduced collagen deposition [Figure 1D]. The fibrosis index of the BLM+APY group (10.98 ± 3.15) decreased significantly when compared with BLM+NS control (15.60 ± 4.20, P = 0.0119) [Supplementary Figure 3, https://links.lww.com/CM9/B196]. We investigated the effect of apyrase treatment on monocyte/macrophage polarization. Results showed that BLM can restrain the AMϕ polarization to M1 and induce AMϕ polarization to M2. As shown in Figure 1E, on day 14, the percentage of M1 phenotype of AMϕ (F4/80+CD11c+CD206-) was about (85.00 ± 3.31)% in the NS group, while in BLM group it decreased to (37.64 ± 2.74)% (P < 0.0001). On the contrary, the percentage of M2 phenotype of AMϕ in NS group was significantly lower than BLM group ((15.14 ± 5.28)% vs. 56.06 ± 4.83%, P < 0.0001). Apyrase appears to partially inhibit BLM-induced AMϕ M2 polarization on day 14. When compared with BLM+NS group, the M1 phenotype of AMϕ in BLM+APY group increased ([47.94 ± 3.02]% vs. (36.37 ± 8.47)%, P = 0.0108) [Figure 1E]. Our previous results showed that Ly6Chi monocytes underwent a dynamic change following exposure to BLM, and reached the maximal level on day 3.[2] We then assessed the effect of apyrase treatment on circulating monocyte subsets on day 3. Compared with the BLM+NS group, apyrase treatment could reduce the percentage of Ly6Chi monocytes that was induced by BLM treatment ((69.77 ± 4.21)% vs. (76.70 ± 2.87)%, P = 0.0205). On the contrary, Ly6Clo monocyte percentage was higher in BLM+APY group when compared with BLM+NS group [Figure 1F]. Our data confirmed the expression of ATP receptor P2X7 in mouse peripheral blood monocytes and AMϕ. We demonstrated that apyrase inhibits the inflammatory response and the expansion of circulating Ly6Chi inflammatory monocytes during the acute phase of BLM induced-lung injury. In addition, apyrase reduces collagen deposition and M2 AMϕ polarization during the remodeling phase. The work of Li et al[5] demonstrated that ATP/P2X7R axis mediates the pathological process of allergic asthma by inducing M2 polarization of AMϕ. Our present work evidenced that apyrase can exhaust superfluous ATP and reverse M2 polarization to alleviate lung fibrosis. Furthermore, our work extended to monocyte and indicated that ATP mediated the monocyte phenotype differentiation which was similar to previous study.[5] The more elaborate mechanism of ATP/P2X7R axis in lung inflammation and lung fibrosis needed to be further studied by using P2X7R gene knockout animals. All in all, our findings highlight that consuming excess airway ATP by inhibiting ATP mediated monocyte/macrophage polarization may be a potential strategy for treating lung injury and fibrosis during BLM treatment. Funding This study was supported by a grant from Tianjin Municipal Science and Technology Committee (No. 18JCZDJC12000). Conflicts of interest None.