A mechanism of paraquat toxicity involving nitric oxide synthase

Abstract

Paraquat (PQ) is a well described pneumotoxicant that produces toxicity by redox cycling with cellular diaphorases, thereby elevating intracellular levels of superoxide (O-(2)). NO synthase (NOS) has been shown to participate in PQ-induced lung injury. Current theory holds that NO reacts with O-(2) generated by PQ to produce the toxin peroxynitrite. We asked whether NOS might alternatively function as a PQ diaphorase and reexamined the question of whether NO/O-(2) reactions were toxic or protective. Here, we show that: (i) neuronal NOS has PQ diaphorase activity that inversely correlates with NO formation; (ii) PQ-induced endothelial cell toxicity is attenuated by inhibitors of NOS that prevent NADPH oxidation, but is not attenuated by those that do not; (iii) PQ inhibits endothelium-derived, but not NO-induced, relaxations of aortic rings; and (iv) PQ-induced cytotoxicity is potentiated in cytokine-activated macrophages in a manner that correlates with its ability to block NO formation. These data indicate that NOS is a PQ diaphorase and that toxicity of such redox-active compounds involves a loss of NO-related activity.

Figure 1

NOS1 is a PQ diaphorase. (A) Stimulation of NOS1-dependent NADPH oxidation by increasing concentrations of PQ. NADPH oxidation was continuously monitored spectrophotometrically at 340 nm in a microtiter plate format for 1 min. (B) Inhibition of NO formation by increasing concentrations of PQ. The same microtiter plates used above were left at room temperature for 5 hr, and then an aliquot was assayed for total nitrite and nitrate. (C) Anaerobic reduction of PQ to its cation radical (PQ⨥) by NOS1. Formation of PQ⨥ was followed at 600 nm for 3 min. (D) Hanes plot (K_m and V_max values) for the PQ reductase activity of rat NOS1.

Figure 2

(A) NOS inhibitors l-NAME and l-NMMA have differential effects on NOS1-dependent NADPH oxidation. NADPH oxidation was continuously monitored spectrophotometrically at 340 nm in a microtiter plate format for 1 min in the presence (solid bars) or absence (open bars) of PQ (25 μM). NOS inhibitors were tested at a concentration of 1 mM. (B) Attenuation of PQ-induced endothelial cell injury by l-NAME, but not by l-NMMA. Endothelial cells were grown in 24-well plates and treated with increasing concentrations of either l-NAME or l-NMMA in the presence of 2 mM PQ. The percent release of LDH into the culture medium 24 hr after PQ treatment was used to assess cell injury. ∗ indicates a significant difference from control group, and ∗∗ indicates significant differences from the PQ group, P < 0.05.

Figure 3

(A) PQ inhibits EDRF activity. Rabbit aortic rings were constricted with 0.1 μM phenylephrine (initial contraction) and relaxed with increasing concentrations of acetylcholine in the presence or absence of increasing concentrations of PQ or l-NAME (2 mM). (B) PQ does not affect exogenous NO-induced aortic ring relaxation. Rabbit aortic rings were constricted with 1 μM phenylephyrine (initial contraction) and relaxed with increasing concentrations of NO in the presence or absence of 2 mM PQ. Relaxations are expressed as the percent of the initial phenylephrine-induced contraction.

Figure 4

(A) PQ-induced cell injury is potentiated in immunostimulated macrophages. Cell injury was assessed 24 hr after cytomix and PQ treatment by measuring LDH release. (B) Induction of NO formation by cytomix and its inhibition by PQ. NOS2 activity was assessed 24 hr after cytomix by measuring the accumulation of NO metabolites, nitrite and nitrate. Macrophages were grown in 24-well plates and immunostimulated by treatment with cytomix. Unstimulated (basal) and activated (cytomix) cells were treated with or without PQ (0.1 mM). (C) NO-mediated macrophage injury and its inhibition by PQ. Cell injury was assessed 24 hr after DETA NONOate and PQ treatment by measuring LDH release. (D) PQ does not alter the formation of NO metabolites from DETA NONOate. NO release was assessed 24 hr after DETA NONOate and PQ treatment by measuring the accumulation of nitrite and nitrate. Macrophages were grown in 24-well plates and treated with increasing concentrations of DETA NONOate in the presence or absence of 0.1 mM PQ.

Figure 5

(A) Decreased nitrotyrosine formation with PQ treatment. Protein nitration was assessed 24 hr after cytomix and PQ treatment by Western blot analysis. Macrophages were grown in 24-well plates and immunostimulated by treatment with cytomix. Unstimulated (basal) and activated (cytomix) cells were treated with or without PQ (0.1 mM). (B) Differential effect on PQ-mediated aconitase inactivation by cytomix and DETA NONOate treatments. Aconitase activity was measured in cell lysates 24 hr after cytomix, DETA NONOate, and PQ treatments. Macrophages were grown in 24-well plates and treated with cytomix or 250 μM DETA NONOate. Control, activated (cytomix), and DETA NONOate groups were treated with or without PQ (0.1 mM). ∗ indicates a significant difference from control group, P < 0.05.