Angiotensin-converting enzyme defines matrikine-regulated inflammation and fibrosis

Abstract

The neutrophil chemoattractant proline-glycine-proline (PGP) is generated from collagen by matrix metalloproteinase-8/9 (MMP-8/9) and prolyl endopeptidase (PE), and it is concomitantly degraded by extracellular leukotriene A4 hydrolase (LTA4H) to limit neutrophilia. Components of cigarette smoke can acetylate PGP, yielding a species (AcPGP) that is resistant to LTA4H-mediated degradation and can, thus, support a sustained neutrophilia. In this study, we sought to elucidate if an antiinflammatory system existed to degrade AcPGP that is analogous to the PGP-LTA4H axis. We demonstrate that AcPGP is degraded through a previously unidentified action of the enzyme angiotensin-converting enzyme (ACE). Pulmonary ACE is elevated during episodes of acute inflammation, as a consequence of enhanced vascular permeability, to ensure the efficient degradation of AcPGP. Conversely, we suggest that this pathway is aberrant in chronic obstructive pulmonary disease (COPD) enabling the accumulation of AcPGP. Consequently, we identify a potentially novel protective role for AcPGP in limiting pulmonary fibrosis and suggest the pathogenic function attributed to ACE in idiopathic pulmonary fibrosis (IPF) to be a consequence of overzealous AcPGP degradation. Thus, AcPGP seemingly has very divergent roles: it is pathogenic in its capacity to drive neutrophilic inflammation and matrix degradation in the context of COPD, but it is protective in its capacity to limit fibrosis in IPF.

Keywords: Fibrosis; Pulmonology.

Figure 1. An extracellular enzymatic activity that degrades AcPGP.

Tissue homogenates (A) or BALF and serum (B) from naive Balb/c mice were incubated with AcPGP, and degradation was assessed after 24 hours by mass spectrometry. (C and D) Serum from naive lta4h^–/– mice and littermate controls was incubated with AcPGP and degradation assessed at multiple time points by mass spectrometry (C) or release of free proline (D). Serum from naive Balb/c mice was heat inactivated (HI) for 10 min at 100°C (E) or preincubated with selective inhibitors (all at 1 μM) for 30 min at 37°C (F) prior to incubation with AcPGP. Degradation of AcPGP was subsequently assessed by mass spectrometry (E and F). A and B are from 4 mice/group and representative of 3 experiments. C–F are triplicates and representative of 2 independent studies. Results depicted as mean ± SEM. *P < 0.05 or **P < 0.01 using Mann–Whitney statistical test (A, B, and E) or Kruskal-Wallis with Dunn’s post test (F).

Figure 2. Angiotensin-converting enzyme (ACE) degrades AcPGP.

Serum from naive Balb/c mice was preincubated with multiple concentrations (ranging from 1 nM to 100 μM) of captopril (A) or enalapril (B) for 30 min at 37°C prior to incubation with AcPGP and degradation of peptide was assessed after 24 hours by mass spectrometry. (C) Plasma from ACE-KO mice (ACE 1 = null for ACE; ACE 14/14 = null for somatic ACE; n = 14) and littermate controls (n = 9) were incubated at 37°C with AcPGP and degradation of peptide assessed after 24 hours by mass spectrometry. Recombinant ACE (0.001–0.1 U/ml) and prolidase (1 U/ml) were incubated with AcPGP and degradation assessed at multiple time points by mass spectrometry. ***P < 0.001 (D) or release of free proline (E). (F) Recombinant human ACE was incubated with AcPGP or Mca-RPPGFSAFK(Dnp)-OH and various concentrations of captopril. Degradation of AcPGP was assessed after 24 hours by mass spectrometry and of Mca- RPPGFSAFK(Dnp)-OH fluorometrically. Levels of ACE protein in tissue homogenates (G) or BALF and serum (H) from naive Balb/c mice were determined by ELISA. (I) Schematic of the proteolytic cascade that defines the availability of collagen-derived matrikines PGP and AcPGP. A, B, D, E, and F are triplicates and representative of 2 independent studies. G and H are from 4 mice/group and representative of 2 experiments. Results depicted as mean ± SEM. *P < 0.05 or **P < 0.01 using Mann–Whitney statistical test (C and H) or Kruskal-Wallis with Dunn’s post test (G).

Figure 3. ACE is elevated in BALF following pulmonary challenge to promote AcPGP degradation and is inhibited by captopril.

(A and B) Balb/c mice were administered 10 μg LPS i.n. and culled after 6, 12, or 24 hours. BALF ACE protein levels were determined by ELISA (A). BALF from these mice was incubated with AcPGP and degradation assessed after 24 hours by mass spectrometry (B). (C–E) Naive Balb/c mice were administered histamine i.v. and culled after 30 min. BALF AcPGP–degrading activity was assessed by mass spectrometry (C) and ACE protein levels by ELISA (D). (E) Correlation between BALF ACE protein levels and BALF albumin levels (determined by ELISA). (F–H) Balb/c mice were administered 1 μg MIP-2 i.n. and culled after 6 hours. BALF AcPGP–degrading activity was assessed by mass spectrometry (F) and ACE protein levels by ELISA (G). (H) Correlation between BALF ACE protein levels and BALF albumin levels (determined by ELISA). (I and J) Lta4h^–/– mice and littermate controls were administered 10 μg LPS i.n. and 2 mg captopril/PBS control i.p. Mice were culled after 24 hours, and levels of PGP (I) and AcPGP (J) in BALF were assessed by mass spectrometry. Experiments are from 4 (F and G) or 5 (A–D, I and J) mice/group and representative of 2 experiments. Spearman correlation (E and H) is representative of 2 experiments, each with at least 4 mice per group. Results depicted as mean ± SEM. *P < 0.05 or **P < 0.01 using Mann–Whitney statistical test (C and H) or Kruskal-Wallis with Dunn’s post test (G). Spearman correlation from 2 experiments with at least 5 mice per group.

Figure 4. ACE is elevated in BALF of IPF patients.

BALF ACE levels in human subjects with IPF (n = 10) or COPD (n = 17) and healthy controls (n = 17) were determined by ELISA. Results depicted as mean ± SEM. *P < 0.05 using unpaired t test.

Figure 5. ACE is elevated in BALF following bleomycin exposure to promote AcPGP degradation.

(A) C57BL/6 mice were administered bleomycin i.t. and culled after 1, 2, and 3 weeks. BALF was incubated with AcPGP and degradation assessed after 24 hours by mass spectrometry. (B–E) C57BL/6 mice were administered bleomycin i.t. and culled after 3 and 7 days. BALF AcPGP–degrading activity was assessed by mass spectrometry (B) and ACE protein levels by ELISA (C). BALF AcPGP–degrading activity was assessed with and without preincubation with 1 μM captopril (D). (E) Macrophages and PMNs were enumerated in BALF 3 and 7 days after bleomycin exposure. Experiments are from 4 mice/group. Results depicted as mean ± SEM. *P < 0.05 and **P < 0.01 using Mann–Whitney statistical test.

Figure 6. AcPGP inhibits bleomycin-induced pulmonary fibrosis.

C57BL/6 mice were administered bleomycin i.t. followed after a week by daily i.t. administrations of 125 μg or 250 μg AcPGP per mouse or PBS for 2 weeks (A). Representative H&E stained lung sections (B–G) and quantification of the fibrosis using the Ashcroft scoring system (H). Images are 20× and 200×, as indicated; scale bars: 500 μM. (I) Total lung hydroxyproline measurements. Experiments are from at least 5 mice/group and representative of 2 independent experiments. Results depicted as mean ± SEM. *P < 0.05 or **P < 0.01 using Kruskal-Wallis with Dunn’s post test.

Figure 7. Efficacy of peptide analogs of AcPGP against fibrosis.

C57BL/6 mice were administered bleomycin i.t. followed after a week by daily i.t. administrations of 250 μg AcPGP, PGP, PGG, or PBS for 2 weeks (A). Representative H&E-stained lung sections (B–I). Images are 20× and 200×; scale bars: 500 μM. (J) Total lung hydroxyproline measurements. Experiments are from at least 5 mice/group and representative of 2 independent experiments. Results depicted as mean ± SEM. **P < 0.01 using Kruskal-Wallis with Dunn’s post test.

Figure 8. A PGP antagonist abrogates the antifibrotic effects of captopril.

C57BL/6 mice were provided with water or water supplemented with captopril and administered bleomycin i.t., followed a week later by i.t. administrations of 50 μg RTR or PBS 3 times a week for 2 weeks (A). Representative H&E-stained lung sections (B–I) and quantification of the fibrosis using the Ashcroft scoring system (J). Images are 20× and 200×. (K) Total lung hydroxyproline measurements. Experiments are from 3–5 mice/group and representative of 2 independent experiments. Results depicted as mean ± SEM. *P < 0.05 using Kruskal-Wallis with Dunn’s post test.