Kinetics of increased generation of (cid:1) NO in endotoxaemic rats as measured by EPR

Ferrous-diethyldithiocarbamate (Fe(DETC) 2 ) chelate is a lipophilic spin trap devel-oped for (cid:1) NO detection by electron paramagnetic resonance (EPR) spectroscopy. Using this spin trap we investigated the kinetics of (cid:1) NO production in endotoxaemia in rats induced by lipopolysaccharide ( Escherichia coli , 10 mg/kg). The NO-Fe(DE-TC) 2 complex was found to give a characteristic EPR signal, and the amplitude of the 3rd (high-field) component of its hyperfine splitting was used to monitor the level of (cid:1) NO. We found that in blood, kindey, liver, heart and lung (cid:1) NO production starts to increase as early as 2 h after LPS injection, reaches the maximum 6 h after LPS injection and then returns to basal level within further 12–18 h. Interestingly, in the eye bulb the maximum of (cid:1) NO production was detected 12 h after LPS, and the signal was still pronounced 24 h after LPS. In brief, the highly lipophilic exogenous spin trap, Fe(DETC) 2 is well suited for assessment of (cid:1) NO production in endotoxaemia. We demonstrated that the kinetics of increased production of (cid:1) NO in endotoxaemic organs, with the notable exception of

It is well known that expression of inducible · NO synthase (NOS-2) increases in endotoxic shock, and · NO produced by NOS-2 contributes to hypotension, vasoplegia, and also, indirectly, via peroxynitrite formation, to organ injury in endotoxaemia. Indeed, inhibition of NOS-2 in endotoxaemia reversed hypotension, and restored vascular responsiveness to vasoconstrictor agents (Thiemermann, 1997;Kirkeboen & Strand, 1999). Moreover, numerous reports have demonstrated organ-protective effects of NOS inhibitors, superoxide dismutase (SOD) mimetics, or peroxynitrite (ONOO -) decomposition catalysts (Salvemini et al., 1999). On the other hand, · NO produced by a constitutive isoenzyme, NOS-3, has been shown to play a protective role in endotoxaemia by its vasodilatory (Spain et al., 1994), antiplatelet (Shultz & Raij, 1992) or antileukocyte (Laszlo et al., 1995) activities. Interestingly, it was not long ago demonstrated that overexpression of NOS-3 in mice conferred increased resistance to LPS-induced mortality, further supporting the importance of the cytoprotective role of · NO generated by NOS-3 in endotoxaemia (Yamashita et al., 2000). The cytoprotective action of · NO is of particular importance in the endotoxaemic lung, as one of us has recently proposed (Chlopicki et al., 2001). Most surprisingly, quite recent papers suggest that even · NO synthesized by NOS-2 may be cytoprotective (Kubes, 2000). It might well be that the toxic effect of · NO results in the major part from rapid reaction with superoxide anions and peroxynitrite formation while · NO itself protects the tissue against injury, mostly via cGMP-mediated effects. This hypothesis could be validated by direct measurement of · NO and superoxide anions and peroxynitrite formation in endotoxaemia. If the production of · NO in endotoxaemia is accompanied by superoxide release, the amount of · NO that diffuses to target sites may be much lower than the expected on the basis of high NOS-2 expression. Here we attempted to measure directly · NO forma-tion in various organs by electron paramagnetic resonance (EPR) spectroscopy using exogenous spin trap -ferrous diethyldithiocarbamate, Fe(DETC) 2 . Preparation of NO-Fe(DETC) 2 solution. Solutions were prepared in PBS bubbled for at least 30 min with nitrogen to remove oxygen. Concentrations are expressed as the final concentrations of reagents in the measured samples. Ferrous sulphate (50 mg/l) dissolved in sodium citrate solution (250 mg/l) was added to a solution of sodium DETC (about 3 mM, i.e. 500 mg/l), and reduced with sodium dithionite (6 g/l). This mixture was supplemented with a solution of · NO (0.067 mM) obtained by reduction of a stoichiometric equivalent of sodium nitrite with an excess of sodium dithionite, according to Plonka et al. (1999). The solutions were prepared shortly (2-3 min) before experiments, and were mixed just before EPR measurement.

Experiments in vivo.
Endotoxaemia was induced by a single i.p. injection of LPS (10 mg/kg body mass, in 0.5 ml PBS). Animals were killed with nitrogen asphyxia 2.5, 4, 6, 8, 12, 18, and 24 h after LPS injection, whereupon blood (about 1 ml, left ventricle) and organs (lungs, liver, kidney, heart and eye bulbs) were isolated. Control animals were injected with 0.5 ml of sterile PBS. Thirty minutes prior to euthanasia, animals were given spin trap (DETC -500 mg/kg body mass, in 0.2 ml PBS, i.p.), and chelated iron(II) complexes (FeSO 4´7 H 2 O, 50 mg/kg plus sodium citrate -250 mg/kg body mass, s.c.) in two separate injections. Notably, DETC creates insoluble complexes with iron(II). Accordingly, to prevent their precipitation from injected solution, they must be given separately (Vanin et al., 1989). All animals of the test and control groups were treated identically, so that the time available for the spin trap and iron(II) ions to penetrate tissues and to create insoluble NO-binding complexes was exactly the same -30 min. The doses were chosen according to Kubrina et al. (1992). All the experiments were repeated at least 2-4 times, so as to collect sets of 2-4 independent organ samples per each timepoint. EPR measurements. Solution of NO-Fe-(DETC) 2 was frozen in standard glass pipes (inner diameter 4 mm, length 2 cm) in liquid nitrogen (77 K), immediately after preparation. Tissue and blood samples were prepared in similar pipes directly after isolation, and stored in liquid nitrogen until the EPR measurements. The spectra were recorded at 77 K, using an EPR X-band spectrometer Varian E-3 with a rectangular TE 102 resonant cavity, and a standard quartz Dewar. The EPR signal intensity was expressed as the amplitude of the third (high-field) hyperfine component of the signal (see Fig. 1), because the first and the second component often overlap with a strong line of Cu(DETC) 2 complexes (Andriambeloson et al., 1997;Kubrina et al., 1992).

RESULTS
A typical EPR signal of the spin adduct NO-Fe(DETC) 2 was obtained by mixing Fe(DETC) 2 complexes with a solution of · NO. The signal (g^= 2.035; g ïï = 2.02) with a well-resolved hyperfine splitting at g^is shown in Fig. 1. The third, high-field component of the signal could be distinguished in the spectra from all the investigated organs (see inserts in Fig. 2) and reflected the level of nitrosyl-iron complexes in the measured material.
Despite differences in the intensity of the signals between tissues (Fig. 2), the time course of changes in the signal intensity of NO-Fe(DETC) 2 complex in blood ( Fig. 2A), liver (Fig. 2B), lung (Fig. 2C), kidney (Fig. 2D) and heart (Fig. 2E) was similar. The maximum was reached around 6-8 h after administration of LPS, followed by a slow decrease in the signal amplitude, so that 24 h after LPS injection the signal was hardly detectable. In Vol. 50 Kinetics of · NO generation in endotoxaemic rats blood of endotoxaemic rats the signal had the lowest intensity among all the investigated materials ( Fig. 2A), whereas the signal from liver (Fig. 2B) was the strongest. Instead, blood revealed a strong triplet signal of haemoglobin-complexed nitric oxide -nitrosohaemoglobin (HbNO; Fig. 2A) which in other organs was poorly detectable.
In contrast with blood and other organs, in the eye bulbs the maximum could be observed 12 h after induction of endotoxaemia, and 24 h after LPS the signal was still pronounced (Fig. 2F).

DISCUSSION
We analyzed here the formation of · NO in rat endotoxaeamia by EPR using exogenous spin trap -Fe(DETC) 2 . Since DETC in vivo of-
ten creates EPR-detectable complexes with copper cations, the signal of which is superimposed on the signal of NO-Fe(DETC) 2 adducts, only the 3rd, high field component of the signal remains clearly measureable (Andriambeloson et al., 1997;Kubrina et al., 1992). This component remained detectable in the presence of HbNO signal in blood ( Fig. 2A, insert), and in a linear fashion corresponded with · NO concentration (now shown), therefore, it was used by us as the analytical line. Importantly, the weak signal of the spin adducts detectable in blood and strong in solid tissues shows that DETC, a hydrophobic NO-metric spin-trap may be considered a useful NO-metric agent for faithful determination of · NO contents in tissues, as suggested in numerous previous papers (Vanin & Kleschyov, 1998;Kubrina et al., 1992;Mülsch et al., 1992;Vanin et al., 1989). After chelating iron(II) cations, the complex precipitates, which makes it toxic, but which also prevents uncontrolled changes in the trap concentration or trap removal from the tissues (Vanin & Kleschyov, 1998). Our data show that in the majority of organs, · NO complexes are detectable as early as two hours after LPS injection. We also show that the intensity of · NO production reaches the maximum about 6 h after administration of LPS and then either stays at approximately the same level until 12 h and then decreases (kidney, lung, blood), or starts to diminish already 12 h after LPS (heart, liver). In all cases, 24-48 h after LPS the signal intensity returned to the basal level. Similar kinetics were observed for HbNO in endotoxaemic rat blood (Plonka et al., 1999;Westenberger et al., 1990), as well as lung (Plonka et al., 1999).
Notably, during endotoxaemia in rats in vivo the induction of NOS-2 as measured at the mRNA and protein level follows a different pattern. NOS-2 protein level reaches the maximum 12-24 h after LPS and is present at almost the same level 30 h after LPS (Brandes et al., 1999) while mRNA reaches the maximum after 12 h and is no longer detectable 30 h af-ter LPS administration. Also in isolated macrophages stimulated with LPS or cytokines both NOS-2 gene promotor activity, and NOS-2 protein reach the maximum 24 h after stimulation with LPS but not earlier (Cieslik et al., 2002). This is consistent with the hypothesis that production of · NO reaches its maximum early in endotoxaemia, while later, due to concomitant increase in superoxide generation, peroxynitrite is formed. This concept may be further supported by the fact that a single dose of 10 mg of LPS in vivo in rats leads to a marked increase in the expression and activity of NADPH oxidase and xanthine oxidase attaining its peak 12-30 h after LPS (Brandes et al., 1999). Accordingly, late endotoxaemia is associated with increased superoxide anion production, increased generation of ONOOand nitrotyrosine formation leading to dysfunction of various organs (Salvemini et al., 1999). Although without detailed measurements of superoxide and peroxynitrite production this hypothesis posesses only the value of speculation, we suggest that it is the most probable explanation of our results. A different time course of · NO production in endotoxaemia is observed in the eye bulbs, in which the maximum of the signal of NO-Fe(DETC) 2 complexes was seen 12 h after LPS and even 24 h after LPS it was still well detectable. The kinetics of the NO-Fe(DETC) 2 signal in the eye corresponds better than in other organs with the well known time course of the expression of NOS-2 protein in endotoxaemia. Accordingly, one could suggest that inflammatory formation of free radicals is limited in endotoxaemic eye (Rocha et al., 1997) and does not interfere with detection of · NO. This hypothesis may find its rationale in the fact that eye is an immunologically priviledged organ, with a unique pattern of inflammatory response (Streilein & Stein-Streilein, 2000;Taylor et al., 1998). On the other hand, the low intensity of the EPR signals detectable in the eye bulb, as compared to other organs and tissues, may be accounted for by poorer penetration of iron(II) ions or DETC through the non-vascularized parts of the organ and their slightly altered metabolism.
Spin trapping of · NO by a highly lipophilic exogenous spin trap -Fe(DETC) 2 , which traps · NO at the site of · NO synthesis, has turned out to deliver congruent information on the magnitude of · NO production in various tissues. However, additional studies are needed to clarify the basis for the incompatibility between the known kinetics of NOS-2 expression and the kinetics of · NO production, assessed here by EPR technique, in lung, liver, heart, kidney, as well as for the exceptional kinetics of · NO production in endotoxaemic eye. Progress towards a better picture of the time course of oxidative and "nitrosative" stress accompanying endotoxaemia in various parts of the organism will be achieved by supplementary ex vivo or in vivo kinetic measurements using DETC derivatives, such as N-(dithiocarboxy)sarcosine (Plonka et al., 2003). Direct and parallel measurements of superoxide and peroxynitrite production in endotoxaemic tissues are mandatory to explain the early drop in the level of · NO despite the persistence of NOS-2 expression. How to rescue · NO activity in the late endotoxaemia to take advantage of its cytoprotective properties without perpetuating its hypotensive effects presents a challenge for future investigations.