A Dynamic Variation of Pulmonary ACE2 Is Required to Modulate Neutrophilic Inflammation in Response to Pseudomonas aeruginosa Lung Infection in Mice

Abstract

Angiotensin-converting enzyme 2 (ACE2) is a potent negative regulator capable of restraining overactivation of the renin-angiotensin system, which contributes to exuberant inflammation after bacterial infection. However, the mechanism through which ACE2 modulates this inflammatory response is not well understood. Accumulating evidence indicates that infectious insults perturb ACE2 activity, allowing for uncontrolled inflammation. In the current study, we demonstrate that pulmonary ACE2 levels are dynamically varied during bacterial lung infection, and the fluctuation is critical in determining the severity of bacterial pneumonia. Specifically, we found that a pre-existing and persistent deficiency of active ACE2 led to excessive neutrophil accumulation in mouse lungs subjected to bacterial infection, resulting in a hyperinflammatory response and lung damage. In contrast, pre-existing and persistent increased ACE2 activity reduces neutrophil infiltration and compromises host defense, leading to overwhelming bacterial infection. Further, we found that the interruption of pulmonary ACE2 restitution in the model of bacterial lung infection delays the recovery process from neutrophilic lung inflammation. We observed the beneficial effects of recombinant ACE2 when administered to bacterially infected mouse lungs following an initial inflammatory response. In seeking to elucidate the mechanisms involved, we discovered that ACE2 inhibits neutrophil infiltration and lung inflammation by limiting IL-17 signaling by reducing the activity of the STAT3 pathway. The results suggest that the alteration of active ACE2 is not only a consequence of bacterial lung infection but also a critical component of host defense through modulation of the innate immune response to bacterial lung infection by regulating neutrophil influx.

Figure 1.. Pulmonary ACE2 expression and activity are dynamically variable In response to Pseudomonas aeruginosa infection.

A bacterial pneumonia model was established by instilling 30μl saline containing 1X10⁶ Pseudomonas aeruginosa (Pao1 strain, P.a) through intubation with an otoscope. At time post bacterial inoculation, mice were sacrificed, and bronchiole alveoli lavage fluid (BALF) or lung tissues were collected. (A). ACE2 activity in BALF was measured by Fluro-substrate based assay. The RLU was normalized by recovery volume of respective BALF. The ACE2 gene expression level was measured by quantitative RT- PCR and displayed as ACE2 mRNA expression level relative to a housekeeping gene RPLO mRNA expression level (B), and ACE2 protein level in mouse lung homogenates was determined by ELISA, and the concentration was normalized by protein concentration (C). (D), Neutrophil numbers in BALF were determined by flow cytometry, gated as Ly6G⁺CD11b⁺ in live cells (7AAD⁻). In all samples n≥5. All comparisons were done between the non-treated group (0) and treated group at the time (day) of sacrifice post bacterial inoculation. Data were analyzed for statistical significance by two-tailed student’s T-test or analysis of variance (ordinary one way ANOVA multiple comparisons) using Prism software (GraphPad). * p<0.05, ** p<0.01 *** p<0.001.

Figure 2.. Pre-existing and persistent lack of the active ACE2 exacerbates the severity and worsens the prognosis of bacterial pneumonia.

(A) Schematic depicting the strategy of ACE2 inhibitor (MLN4760, 2.5mg/kg) treatment. (B) Bodyweight change at the time of sacrifice. (C) Survival curve of mice in respective experimental groups as indicated. (D). The degree of lung permeability from mice of respective experimental groups was determined by wet/dry ratio. (E) Levels of total protein in BALF from mice of the indicated experimental group. (F) Neutrophil accumulation in mouse lungs with pre-existing and persistent deficiency of the active ACE2 was assayed by flow cytometry gated as Ly6G⁺CD11b⁺ living cells in BALF. (G) Bacterial load in mouse BALF was determined by colony-forming unit (CFU). The volume of individual BALF normalized the CFU (indicated as CFU/ml). (H-I) Pro-inflammatory cytokine and chemokine levels of CXCL5 and KC were determined by ELISA. (J) Representative micrograph of pathohistology of lungs from indicated groups at the time post bacterial inoculation. dpi: day post, inoculation. (K-L) Representative immunofluorescence images from an indicated experimental group of mice for inducible nitric oxidase (iNOS, K), and neutrophil marker myeloperoxidase (MPO, L). (M) Pathological scores of mouse lungs from wild type mice with or without MLN4760 treatment and ACE2^ko mice were calculated. Data were analyzed for statistical significance by two-tailed student’s T-test or analysis of variance (ordinary one way ANOVA multiple comparisons) using Prism software (GraphPad).In all experiments, n≥4 *p<0.05, ** p<0.01, ***p<0.001. Scale bar = 50μm.

Figure 3.. Deficiency of active ACE2 induced lung injury is due to heightened neutrophilic inflammation.

(A). Schematic elaborating the strategy of ACE2 inhibitor (MLN4760, 2.5mg/kg) and anti-Ly6G (5mg/kg) rat antibody treatment. (B). Neutrophil counts in BALF from experimental groups were determined by flow cytometry gated as live Ly6G⁺CD11b⁺ cells. (C). Bodyweight changes 48-hour post bacterial lung infection in comparison to initial body weight. (D-E). Levels of pro-inflammatory cytokine and inflammatory cell recruiting chemokine in BALF from experimental groups as indicated were detected by ELISA. Final concentrations were normalized by recovery volume. (F-G). Representative micrograph demonstrating immunofluorescent staining for inflammation marker inducible Nitric oxide synthase (iNOS, green) and neutrophil marker myeloperoxidase (MPO, red) in lung sections from mice as indicated. Data were analyzed for statistical significance by two-tailed student’s T-test or analysis of variance (ordinary one way ANOVA multiple comparisons) using Prism software (GraphPad).For each experimental group n≥4. *p<0.05 **p<0.01, ***p<0.001. N.S. No statistic significance. Scale bar = 50μm.

Figure 4.. Interrupting active ACE2 recovery attenuates the resolution of lung inflammation in bacterial pneumonia.

(A) Timelines outline the experimental protocol. (B) Bodyweight changes relative to initial body weight at six dpi with or without interrupting active ACE2 recovery by ACE2 inhibitor MLN4760. (C) Neutrophil accumulation in mouse lung infected by bacteria as manifested by live Ly6G⁺CD11b⁺ cells in BALF using flow cytometry. (D-E). Interrupting active ACE2 recovery prolonged elevated pro-inflammatory cytokine (Mip2 and KC) levels in BALF determined by ELISA. (F). H & E staining micrograph showing histology of mouse lung with perturbed active ACE2 recovery at six dpi. G-H: Representative micrograph demonstrating immunofluorescence for neutrophil marker myeloperoxidase (G, MPO, red), inflammation marker inducible Nitric oxide synthase (H, iNOS, green in lung sections from mice as indicated. Data were analyzed for statistical significance by two-tailed student’s T-test using Prism software (GraphPad). For each experimental group n≥4. *p<0.05 **p<0.01. Scale bar = 50μm.

Figure 5.. Precluding the active ACE2 reduction leads to impaired neutrophil recruitment, exuberant bacterial burden, and enhanced mortality in bacterial pneumonia.

(A). Schematic depicting the strategy of recombinant ACE2 treatment. (B). The percentage of survived mice with or without recombinant human ACE2 at two days post bacterial lung infection. (C). Weight loss in mice that underwent bacterial pneumonia with or without rhACE2. (D). Bacterial load recovered from BALF. Final bacterial counts were normalized by recovery volume. (E). Neutrophil numbers in BALF were determined by flow cytometry gated as Ly6G⁺CD11b⁺ in live cells (7AAD⁻). F-I: Pro-inflammatory cytokine and chemokine levels of CXCL5, G-CSF, Mip-2, and KC were determined by ELISA. (J) Representative micrograph of pathohistology of lungs from indicated groups at the time post bacterial inoculation. dpi: day post, inoculation. (K-L) Representative immunofluorescence images from an indicated experimental group of mice for inducible nitric oxidase (iNOS, K), and neutrophil marker myeloperoxidase (MPO, L). Data were analyzed for statistical significance by two-tailed student’s T-test using Prism software (GraphPad). In all experimental groups, n≥4. *p<0.05, ** p<0.01, ***p<0.001. Scale bar = 50μm.

Figure 6.. Recombinant human ACE2 treatment post initial inflammatory response mitigates lung inflammation and injury in bacterial pneumonia.

(A) Timeline depicting the experimental protocol. (B) Bodyweight changes relative to that before bacterial pneumonia model initiation with or without ACE2 intervention 48-hours post bacterial inoculation. (C) The severity of lung edema in wild type mice with or without recombinant ACE2 was determined by lung tissue wet/dry ratio measurement. (D) Protein levels in BALF from mice with or without recombinant human ACE2 was determined. (E) Bacterial load in BALF from mice of the individual experimental group was determined as CFU/ml. (F) The effect of recombinant ACE2 that was given post initial inflammatory process on the neutrophil accumulation in bacterially infected mouse lung was measured by flow cytometry as indicated as live Ly6G⁺CD11b⁺ cells in BALF. (G-H). Active ACE2 enhancement led to reduced pro-inflammatory cytokine (Mip2 and KC) levels in BALF determined by ELISA. (I). Representative H & E staining micrographs are revealing histopathology in mouse lungs from respective experiment groups as indicated. (J-K) Representative immunofluorescence images from an indicated experimental group of mice for inducible nitric oxidase (iNOS, J), a marker of inflammation and neutrophil marker myeloperoxidase (MPO, K). (L). Pathological scores of mouse lungs with or without recombinant ACE2 treatment post initial inflammation. Data were analyzed for statistical significance by two-tailed student’s T-test using Prism software (GraphPad).For each experimental group n≥4. * p<0.05, ** p<0.01 and *** p<0.001. Scale bar = 50μm.

Figure 7.. Pulmonary epithelial ACE2 is required to buffer exuberant neutrophilic lung inflammation in response to bacterial lung infection.

(A). Representative images immunofluorescent staining for foxj1 (red) and ACE2 (green), demonstrating ACE2 is predominantly expressed in foxj1⁺ cells (upper panel), and the expression is reduced 48-hour post bacterial infection(lower panel). (B). ACE2^Δfoxj11 mice demonstrated more severe mortality in bacterial pneumonia (C). ACE2^Δfoxj1 mice exhibited more weight loss in bacterial pneumonia in comparison to wild type counterpart. (D-E). Bacterially infected ACE2^Δfoxj1 mouse demonstrated an increased permeability as evidenced by elevated wet/dry ratio (D) and protein levels in BALF (E). (F) Neutrophil accumulation in ACE2^Δfoxj1 mouse lung infected by pseudomonas bacteria was exacerbated as manifested by increased Ly6G⁺CD11b⁺ live cells in BALF using flow cytometry. (G) Bacterial load in BALF from ACE2^Δfoxj1 mouse lung infected by pseudomonas bacteria was not significant in comparison to that in the wild type counterparts. (H-I). Excising ACE2 gene from foxj1⁺ cells led to enhanced pro-inflammatory cytokine production (KC and Mip2) as measured by ELISA. (J). Representative micrograph showing histology of lungs of ACE2^Δfoxj1 mice with pseudomonas bacterial infection. K-L Immunofluorescent images of inflammation marker inducible Nitric oxide synthase (iNOS, green, K) and neutrophil marker myeloperoxidase (MPO, red L) in ACE2^Δfoxj1 mice with a bacterial lung infection. (M). Pathological scores of mouse lungs from wild type and were calculated. Data were analyzed for statistical significance by two-tailed student’s T-test using Prism software (GraphPad). In all experimental groups, n≥4. * p<0.05, ** p<0.01 and *** p<0.001. Scale bar = 50μm.

Figure 8.. ACE2 regulates IL-17-mediated neutrophil infiltration by modulating STAT3 signaling.

(A). IL-17A concentration in BALF from mice with or without P.a lung infection was determined by ELISA. (B). Neutrophil counts in BALF from wild-type mice with P.a lung infection with or without pre-existing reduced ACE2 and in the presence or absence of A IL-17A neutralizing antibody. The neutrophils were detected by flow cytometry gated as Ly6G⁺CD11b⁺ live cells. The counts were normalized with the volume of BALF in the individual experimental group of mice. (C). Recombinant IL-17A (100 ng/ml, 30 μl intra-nasal instillation) induces neutrophil infiltration in wild type mice lungs, and the induction is more potent in ACE2 deficient mice. Recombinant human ACE2 inhibits the IL-17A induced neutrophil infiltration in mice (D-E). rhACE2 alleviates rIL-17 mediated cytokine and chemokine production in BALF. (F). IL-17A production in BALF from mice that ACE2 activity was manipulated either by pharmacological reagents or genetically modified. IL-17A concentration was determined by ELISA and normalized with respective BALF volume. (G-H). rhACE2 inhibited IL-17A induced STAT3 phosphorylation in vitro (G, mouse lung organoids) and in vivo (H, mouse lung). (I). rhACE2 represses bacterial lung infection induced STAT3 activation in both wild-type and ACE2 deficient mice. (J). Blocking STAT3 signaling alleviates the lack of pulmonary ACE2 induced exuberant neutrophil infiltration in bacterially infected mouse lung as manifested by western blot for phosphorylated STAT3. WP1066, a STAT3 antagonist. Data were analyzed for statistical significance by two-tailed student’s T-test or analysis of variance (ordinary one way ANOVA multiple comparisons) using Prism software (GraphPad).In all experimental groups, n≥3. * p<0.05, ** p<0.01 and *** p<0.001.

Figure 9.. Schematic of the working hypothesis delineating the role of ACE2 to modulate neutrophilic inflammation in a bacterial pneumonia mouse model.

Bacterial lung infection leads to pulmonary dynamic variation, which is required to modulate neutrophil influx. Pre-existing and persistent enhanced or reduced levels of active ACE2 that disrupts ACE2 dynamic during bacterial pneumonia results in a worsened outcome. Also, interrupted ACE2 recovery delays the resolution process of neutrophilia and lung inflammation in bacterial pneumonia. However, a therapeutic window for the improved outcome using active ACE2 is tangible.