Nox2 NADPH oxidase is dispensable for platelet activation or arterial thrombosis in mice

Abstract

Deficiency of the Nox2 (gp91phox) catalytic subunit of nicotinamide adenine dinucleotide phosphate (NADPH) oxidase is a genetic cause of X-linked chronic granulomatous disease, a condition in which patients are prone to infection resulting from the loss of oxidant production by neutrophils. Some studies have suggested a role for superoxide derived from Nox2 NADPH oxidase in platelet activation and thrombosis, but data are conflicting. Using a rigorous and comprehensive approach, we tested the hypothesis that genetic deficiency of Nox2 attenuates platelet activation and arterial thrombosis. Our study was designed to test the genotype differences within male and female mice. Using chloromethyl-dichlorodihydrofluorescein diacetate, a fluorescent dye, as well as high-performance liquid chromatography analysis with dihydroethidium as a probe to detect intracellular reactive oxygen species (ROS), we observed no genotype differences in ROS levels in platelets. Similarly, there were no genotype-dependent differences in levels of mitochondrial ROS. In addition, we did not observe any genotype-associated differences in platelet activation, adhesion, secretion, or aggregation in male or female mice. Platelets from chronic granulomatous disease patients exhibited similar adhesion and aggregation responses as platelets from healthy subjects. Susceptibility to carotid artery thrombosis in a photochemical injury model was similar in wild-type and Nox2-deficient male or female mice. Our findings indicate that Nox2 NADPH oxidase is not an essential source of platelet ROS or a mediator of platelet activation or arterial thrombosis in large vessels, such as the carotid artery.

Graphical abstract

Figure 1.

Nox2 deficiency does not alter platelet ROS production in male or female mice. Levels of intracellular ROS, detected by oxidation of CM-H₂DCF, were measured in the presence or absence of stimulation with convulxin or thrombin in washed platelets from male (A-B) and female (C-D) WT (+/y or +/+) or Nox2-deficient (−/y or −/−) mice. The fluorescent signal generated as the result of oxidation of CM-H₂DCF to CM-DCF is presented as fold change over the signal observed with platelets from sex-matched WT mice without thrombin or convulxin. Data are presented as mean ± standard error (n = 6 or 7 mice in each group). *P < .05 vs. WT (+/y or +/+) without convulxin/thrombin, ^$P < .05 vs. Nox2-deficient (−/y or −/−) without convulxin/thrombin, ^@P < .05 vs. WT (+/y or +/+) with thrombin (0.02 U/mL), ^#P < .05 vs. Nox2-deficient (−/y or −/−) with thrombin (0.02 U/mL), 2-way ANOVA with the Tukey test for multiple comparisons.

Figure 2.

The amounts of superoxide-specific 2-OH-E⁺generated from the oxidation of DHE are similar in Nox2-deficient and WT mice. Bead-purified platelets (pooled) from male Nox2-deficient (−/y) or WT (+/y) mice were incubated with DHE (25 µM), followed by activation (no stirring) or aggregation (with stirring) with thrombin (0.05 U/mL) and convulxin (50 ng/mL). The superoxide-specific oxidation product 2-OH-E⁺ was quantitated in platelet pellets (intracellular signal) (A) and supernatant fractions (extracellular signal) (B) using HPLC coupled with electrochemical detection and authentic standards. Data are presented as mean ± standard error (n = 3 or 4 mice in each group). *P < .05 vs. RP. 2-OH-E⁺ generation within different experimental conditions (ie, RP, Act, or Agg) and genotypes was analyzed using 2-way ANOVA with the Tukey test. Act, activated platelets; Agg, aggregated platelets; RP, resting platelets.

Figure 3.

Mitochondrial ROS generation is not altered in Nox2-deficient mice. Washed platelets from male and female mice deficient in Nox2 (−/y or −/−) or WT littermates (+/y or +/+) were incubated with 10 µM of DHR1,2,3 (A-D) or MitoSox (E-H) for 25 minutes at 37°C, in the presence or absence of stimulation with convulxin or thrombin. Samples were analyzed by flow cytometry. Data are presented as mean ± standard error (n = 4 or 5 mice in each group). *P < .05 vs. WT (+/y or +/+) without convulxin/thrombin, ^$P < .05 vs. Nox2-deficient (−/y or −/−) without convulxin/thrombin, ^@P < .05 vs. WT (+/y or +/+) with convulxin (25 ng/mL)/thrombin (0.02 U/mL), ^#P < .05 vs. Nox2-deficient (−/y or −/−) with convulxin (25 ng/mL)/thrombin (0.02 U/mL), 2-way ANOVA with the Tukey test for multiple comparisons.

Figure 4.

Activation of α_IIbβ₃and P-selectin surface expression are not influenced by Nox2 deficiency in male or female mice. Activation of α_IIbβ₃ (detected by JON/A binding) (A-D) and surface expression of P-selectin (E-H) were measured by flow cytometry in washed platelets from male and female WT (+/y or +/+) or Nox2-deficient (−/y or −/−) mice following activation with convulxin or thrombin. Data are presented as mean ± standard error (n = 5 or 6 mice in each group). *P < .05 vs. WT (+/y or +/+) without convulxin/thrombin, ^$P < .05 vs. Nox2-deficient (−/y or −/−) without convulxin/thrombin, ^@P < .05 vs. WT (+/y or +/+) with convulxin (25 ng/mL)/thrombin (0.02 U/mL), ^#P < .05 vs. Nox2-deficient (−/y or −/−) with convulxin (25 ng/mL)/thrombin (0.02 U/mL), 2-way ANOVA with Tukey test for multiple comparisons.

Figure 5.

Platelet aggregation and accumulation/thrombi formation ex vivo are similar in WT and Nox2-deficient male or female mice. Washed platelets from male and female WT (+/y or +/+) or Nox2-deficient (−/y or −/−) mice were activated with 25 ng/mL convulxin (A-C) or 0.02 U/mL thrombin (D-F). Representative aggregation tracings for convulxin (A) and thrombin (D). Quantitative data for the percentage light transmission at different time points for convulxin in male (B) and female (C) mice and for thrombin in male (E) and female (F) mice. Washed platelets from male and female WT and Nox2-deficient mice were perfused over a collagen surface for 5 minutes in a microfluidic flow chamber at a shear rate of 2000/s. (G) Representative images of platelet accumulation after 5 minutes of perfusion for male and female mice. (H) The time course of accumulation of platelets/thrombi development was calculated as the surface area covered by platelets in a fixed field. (I) Total thrombi area after 5 minutes of perfusion was calculated as the average surface area covered by platelets in 5 representative fields. Data are presented as mean ± standard error (n = 5-7 mice in each group). Platelet aggregation data were analyzed using multiple Student t tests with Holm-Sidak test for comparison between the groups. Platelet accumulation over time was compared using 2-way ANOVA repeated measures, and total thrombi areas on collagen surface were analyzed by 1-way ANOVA, followed by the Tukey test for multiple comparisons. *P < .05 vs. +/y, ^$P< .01 vs. −/y.

Figure 6.

Platelets from CGD patients have similar aggregation and adhesion responses as platelets from healthy control subjects. Platelet aggregation was examined using washed platelets (A-D) or PRP (E-H). Representative aggregation tracings for washed platelets stimulated with thrombin (0.05 U/mL; A) or collagen (1.0 μg/mL; B). Summary data for aggregation kinetics with thrombin (C) or collagen (D). Representative tracings for dose-dependent platelet aggregation to collagen with PRP from control subjects (E) and CGD patients (F). Summary data for maximal aggregation (G) and aggregation kinetics (H) with 1.0 µg/mL collagen of platelets from control subjects vs. CGD patients. Accumulation of platelet thrombi over collagen was examined in a microfluidic flow chamber at a shear rate of 2000/s. (I) Representative images of platelet accumulation/thrombi formation with platelets from control subjects vs. CGD patients. (J) Time course of accumulation of platelets/thrombi development, calculated as the surface area covered by platelets in a fixed field. (K) Total thrombi area after 5 minutes of perfusion, calculated as the average surface area covered by platelets in 5 representative fields. Composite data are presented from 3 separate runs (2 CGD patients and 3 age-matched control subjects; 1 CGD patient donated a blood sample 2 times).

Figure 7.

Deficiency in Nox2 does not influence susceptibility to carotid artery thrombosis in male or female mice. The time to stable occlusion of the carotid artery following photochemical injury was measured in male and female WT (+/y or +/+) or Nox2-deficient (−/y or −/−) mice. (A) Time to stable occlusion. (B) Percentage of mice with a patent carotid artery (free of stable occlusion) as a function of time after injury. Data are presented as mean ± standard error (n = 7-10 mice in each group). Data were analyzed using 2-way ANOVA with the Tukey test for multiple comparisons.