The critical roles of platelet activation and reduced NO bioavailability in fatal pulmonary arterial hypertension in a murine hemolysis model

Abstract

Pulmonary arterial hypertension (PAH) is suspected to be a strong mortality determinant of hemolytic disorders. However, direct contribution of acute intravascular hemolysis to fatal PAH has not been investigated. The roles of nitric oxide (NO) insufficiency and platelet activation in hemolysis-associated fatal PAH have been suspected but not been experimentally studied. We recently generated a unique intravascular hemolysis mouse model in which the membrane toxin, intermedilysin (ILY), exclusively lyses the erythrocytes of transgenically expressing human CD59 mice (ThCD59(RBC)), thereby inducing ILY-dose-dependent massive hemolysis. Using this murine hemolysis model, we found that the acute increase in pulmonary arterial pressure leading to right ventricle failure caused sudden death. Reduced NO bioavailability and massive platelet activation/aggregation leading to the formation of massive thrombosis specifically in the pulmonary microvasculature played the critical roles in pathogenesis of acute hemolysis-associated fatal PAH. Therapeutic interventions enhancing NO bioactivity or inhibiting platelet activation prevented sudden death or prolonged survival time via the suppression of the acute increase in pulmonary arterial pressure and improvement of right ventricle function. These findings further highlight the importance of the inhibition of platelet activation and the enhancement of NO bioavailability for the treatment and prevention of hemolysis-associated (fatal) PAH.

Figure 1

Further characterization of ILY-mediated hemolysis mouse model. (A-C) ThCD59^RBC erythrocyte susceptibility to lysis by different concentrations of NaCl (A), mellitin (B), or PLY (C) was similar to that of WT erythrocytes. Erythrocytes were incubated in serial concentrations of NaCl, mellitin, or PLY at 37°C for 30 minutes. Results represent mean ± SEM (n = 3, P > .05). (D) ILY-mediated hemolysis-associated sudden death. ILY₁₄₀FI induced 100% (12/12) mortality, but ILY₁₄₀SI (0/12) and ILY₇₀FI (0/8) did not cause death in any mouse. *P < .001 vs the mice with ILY₁₄₀FI. (E-F) Hematocrit- and plasma-free hemoglobin levels in the ThCD59^RBC and WT mice administered ILY injection. *P < .001 vs WT mice and #P < .01 vs the ThCD59^RBC mice administered ILY₇₀FI. Results represent mean ± SEM.

Figure 2

ILY₁₄₀FI-induced death in ThCD59^RBC mice associated with a significantly lower RRRVSP compared with ILY₁₄₀SI or ILY₇₀FI. (A-D) Representative tracings show the dynamic changes in RVSP in the WT (A) and ThCD59^RBC mice with ILY₁₄₀FI (B), and the ThCD59^RBC mice with ILY₁₄₀SI (C) and ILY₇₀FI (D). (E) RRRVSP comparison among these mice. RRRVSP (mm Hg/s) = increased systolic pulmonary arterial pressure (in mm Hg)/time (in seconds; black line in panel B). RRRVSP t test. *P < .05 vs ThCD59^RBC mice with ILY₁₄₀FI. Results represent mean ± SEM.

Figure 3

Systemic blood pressures and echocardiogram in ILY-treated mice. (A) Hemolysis in the ThCD59^RBC induced by ILY injection with 3 different dosing schedules induced an immediate, rapid decrease in MABP. Five minutes later, the mice treated with either ILY₁₄₀SI or ILY₇₀FI had a gradual increase in the MABP, but all mice treated with ILY₁₄₀FI died (and had no measurable MABP). *P < .05 (n = 4) compared with the respective baseline level of MABP. (B) Representative echocardiographic frames in the parasternal long-axis view show the dynamic changes in right ventricular volume in the ThCD59^RBC mice treated with different ILY dosing schedules (the full echocardiogram videos are showed in supplemental Videos 1-4). *P < .01 vs WT with ILY₁₄₀FI. (C) Increased percentage of maximal right ventricular area at end-diastole vs baseline right ventricular area in the mice treated with the ILY. *P < .05 vs ILY₁₄₀FI. (D) Elapsed time until maximal right ventricular dilation was achieved in the mice treated with ILY₁₄₀FI and ILY₁₄₀SI. *P < .05 vs ILY₁₄₀FI. Results represent mean ± SEM.

Figure 4

NO bioavailability and platelet activation in ILY-mediated hemolysis mice. (A) NO (measured as nitrite and nitrate) was reduced in ILY-injected ThCD59^RBC but not WT mice. *P < .05 vs WT. Values shown are mean ± SEM. (B-C) sP-selectin levels or platelet counts. *P < .05 vs WT with ILY₁₄₀FI and **P < .05 vs ThCD59^RBC with ILY₁₄₀FI. (D) ILY resulted in increased TAT in ThCD59^RBC but not in WT mice. Blood samples were collected 1 minute after ILY₁₄₀FI and 15 minutes after ILY₁₄₀SI and ILY₇₀FI from ThCD59^RBC and at 15 minutes after ILY₁₄₀FI from WT mice. Results represent mean values ± SEM. *P < .01 vs WT with ILY₁₄₀FI.

Figure 5

Representative images of integrin αIIb and P-selectin staining in lung. There is more extensive integrin αIIb and P-selectin staining (arrows) in small perialveolar blood vessels in the lungs of ThCD59^RBC mice treated with ILY₁₄₀FI than in those treated with ILY₁₄₀SI. Platelet aggregates were adherent to the vessel walls in the ThCD59^RBC mice treated with ILY₁₄₀FI; this is evident in the 60× magnified images. WT mouse lungs had no or very little staining. Lower magnification (4× and 10×) images for P-selectin staining of the different mice are shown in supplemental Figure 3.

Figure 6

Sildenfil prevented death or prolonged the survival time in the ThCD59^RBC mice treated with ILY₁₄₀FI. (A) cGMP levels in the lung of the ThCD59^RBC mice with or without sildenafil pretreatment. *P < .01 vs without sildenafil pretreatment. (B-C) Pretreatment with sildenafil reduced mortality (B) or prolonged survival time (C) in the ThCD59^RBC mice treated with ILY₁₄₀FI. *P < .05 vs ILY (or without sildenafil pretreatment). (D) Representative image shows the dynamic change of RVSP in the sildenafil-pretreated ThCD59^RBC mice given ILY₁₄₀FI. (E) Pretreatment with sildenafil suppressed the RRRVSP in the ThCD59^RBC mice given ILY₁₄₀FI. *P < .05 vs the mice without sildenafil pretreatment. (F) RRRVSP inversely correlated with the time until the death among ThCD59^RBC mice with the different treatments. (G) There were low levels of RRRVSP in the mice that survived with the different treatments.

Figure 7

Platelet depletion, platelet inhibition, or antithrombotic therapy prevented death or prolonged survival times in the ThCD59^RBC mice treated with ILY₁₄₀FI. The mortality rates (A), sP-selectin levels (B), platelet counts (C), lung P-selectin (D), and integrin αIIb staining (E) in the ThCD59^RBC mice pretreated with the different antithrombotic agents before ILY₁₄₀FI treatment. The WT mice treated with ILY₁₄₀FI were used as the control for sP-selectin and integrin αIIb staining and platelet counts. *P < .05 vs ThCD59^RBC without drug pretreatment.