Complement C3 deficiency attenuates chronic hypoxia-induced pulmonary hypertension in mice

Abstract

Background: Evidence suggests a role of both innate and adaptive immunity in the development of pulmonary arterial hypertension. The complement system is a key sentry of the innate immune system and bridges innate and adaptive immunity. To date there are no studies addressing a role for the complement system in pulmonary arterial hypertension.

Methodology/principal findings: Immunofluorescent staining revealed significant C3d deposition in lung sections from IPAH patients and C57Bl6/J wild-type mice exposed to three weeks of chronic hypoxia to induce pulmonary hypertension. Right ventricular systolic pressure and right ventricular hypertrophy were increased in hypoxic vs. normoxic wild-type mice, which were attenuated in C3-/- hypoxic mice. Likewise, pulmonary vascular remodeling was attenuated in the C3-/- mice compared to wild-type mice as determined by the number of muscularized peripheral arterioles and morphometric analysis of vessel wall thickness. The loss of C3 attenuated the increase in interleukin-6 and intracellular adhesion molecule-1 expression in response to chronic hypoxia, but not endothelin-1 levels. In wild-type mice, but not C3-/- mice, chronic hypoxia led to platelet activation as assessed by bleeding time, and flow cytometry of platelets to determine cell surface P-selectin expression. In addition, tissue factor expression and fibrin deposition were increased in the lungs of WT mice in response to chronic hypoxia. These pro-thrombotic effects of hypoxia were abrogated in C3-/- mice.

Conclusions: Herein, we provide compelling genetic evidence that the complement system plays a pathophysiologic role in the development of PAH in mice, promoting pulmonary vascular remodeling and a pro-thrombotic phenotype. In addition we demonstrate C3d deposition in IPAH patients suggesting that complement activation plays a role in the development of PAH in humans.

Figure 1. C3d deposition in human PAH.

(A–D) Lung sections from non-PAH or IPAH patients (n = 3) were stained with α-C3d and α-SMA antibody and counterstained with DAPI to detect nuclei. Images shown are representative (E) Quantification of C3d staining in non-PAH and IPAH patients. Bars represent the mean ± SD (n = 4). *P<0.05.

Figure 2. C3d deposition in chronic hypoxia-induced PH in mice.

(A–D) Lungs from normoxic and hypoxic C57Bl/6J mice were stained with α-C3d and α-SMA antibody and counterstained with DAPI to detect nuclei(n = 4). (E) Quantification of C3d staining in WT vs. C3−/− mice in normoxia and hypoxia. Bars represent the mean ± SD (n = 4). *P<0.05. (F) Representative Western blot for C3d in normoxic vs. hypoxic C57Bl/6J mice. (G) Quantification of Western blots for C3d in normoxic vs. hypoxic mice. Bars represent the mean ± SD (n = 4). *P<0.05. (H) C5a was immunoprecipitated from lung or plasma of normoxic and hypoxic C57Bl/6J mice and analyzed by Western blot (rC5a = recombinant mouse C5a).

Figure 3. Genetic deletion of C3 attenuates CH-induced PAH in mice.

WT or C3−/− mice were exposed to CH to induce PAH. After 3 weeks (A) RVSP and (B) RVH (Fulton index), were determined. Bars represent the mean ± SD (n = 8–12). *P<0.05.

Figure 4. Loss of C3 attenuates muscularization of pulmonary arterioles in response to CH.

(A–B) Representative photomicrographs of lungs stained against SMA from (A) hypoxic WT and (B) C3−/− mice. Arrows indicate muscularized arterioles. (C) Quantification of the number of partially and fully muscularized arterioles (<100 µm) per high power field (200× total magnification). Bars represent the mean ± SD (n = 4). *P<0.05.

Figure 5. Loss of C3 attenuates vascular wall thickening in response to CH.

(A–B) Representative photomicrographs of muscularized arterioles from (A) hypoxic WT and (B) hypoxic C3−/− mice. (C) Quantification of the % wall thickness of peripheral arterioles (<100 µm). Bars represent the mean ± SD (n = 4). *P<0.05.

Figure 6. Complement C3a and C5a do not promote hPASMC proliferation.

(A) hASMC and (B) hPASMC were treated with C3a or C5a with or without concurrent PDGF (10 ng/ml) stimulation and assessed for proliferation. Bars represent mean ± SD of four individual experiments. *P<0.05 vs. PDGF. n.s. = not significant.

Figure 7. Markers of Inflammation and endothelial dysfunction in WT vs. C3 −/− mice.

(A) IL-6 mRNA was measured by quantitative rtPCR in RNA prepared from normoxic or hypoxic WT and C3−/− lungs. IL-6 levels were normalized to the house keeping gene β2-microglobulin. (B) ICAM-1 was quantified by ELISA in lung homogenates from normoxic or hypoxic WT or C3−/− mice. (C) ET-1 was quantified by ELISA in plasma from normoxic or hypoxic WT or C3−/− mice. Bars represent mean ± SD (n = 4) for A–C. *P<0.05.

Figure 8. Loss of C3 prevents platelet activation caused by CH.

(A) Bleeding time of WT and C3−/− mice exposed to normoxia or CH (n = 5–7). (B) Scatter plot showing platelet population in platelet rich plasma. All experiments were similarly gated to the area encircled. (C) Flow cytometry histogram demonstrating that the gated cell population is positive for the platelet marker CD41. (D–E) Representative flow cytometry histograms of platelets from (D) hypoxic WT or (E) hypoxic C3−/− mice stained with P-selectin antibody or isotype control. (F) Percent P-selectin positive platelets in PRP isolated from normoxic and hypoxic WT or C3−/− mice (n = 6). Bars represent mean ± SD. *P<0.05; n.s. = not significant.

Figure 9. Loss of C3 prevents hypoxia-induced TF upregulation.

(A) Representative Western blot analysis of normoxic or hypoxic WT and C3−/− lungs for TF expression. (B) Densitometric anlysis of Western blots from (A). Bars represent the mean ± SD (n = 4 animals per group). *P<0.05.

Figure 10. Decreased fibrin deposition in hypoxic C3−/− mice.

(A) Representative images of lung sections from normoxic or hypoxic WT or C3 −/− mice stained for fibrin(ogen) and nuclei (DAPI). Scale bar represents 150 µm. (B) % Area fibrin staining per total lung area (n = 4) in hypoxic or normoxic WT or C3−/− mice. (C) Representative photomicrograph of WT hypoxic mouse lung showing colocalization (yellow to orange) of Fibrin (green) and vWF (red). Scale bar represents 50 µm. (A, B, E). Bars represent the mean ± SD. *P<0.05; n.s. = not significant.