Deactivation of Norepinephrine by Peroxynitrite as a New Pathogenesis in the Hypotension of Septic Shock

The protocols of this investigation were reviewed and approved by the institutional animal care committee of Fukui Medical University (Fukui, Japan). Male Wistar rats weighing 250–300 g were used.

A stock solution (8 × 10⁻²m ONOO⁻¹in 3 × 10⁻¹N NaOH) was purchased from Dojindo Laboratories (Kumamoto, Japan). As alkaline conditions are required for ONOO⁻¹stability, the stock solution was diluted to appropriate concentrations with 3 × 10⁻¹N NaOH. The ONOO⁻¹was added to a 5 × 10⁻²m sodium phosphate buffer solution containing norepinephrine at a pH of 7.4 and a temperature of 37°C. The half-life of ONOO⁻¹at a pH of 7.4 is less than 1 s. 11To avoid decomposition of ONOO⁻¹before full mixing, reaction solutions in tubes were rapidly stirred during bolus ONOO⁻¹additions. The final pHs of all reaction solutions were checked to ensure adequate buffering. To exclude the influence of possible decomposition products of ONOO⁻¹, such as nitrate or nitrite, decomposed ONOO⁻¹was used as a control. Decomposed ONOO⁻¹was made by incubation for 10 min in a buffer with a pH of 7.4. 16,17 

Norepinephrine (3 × 10⁻⁵m) was incubated with 10⁻³m SIN-1 in a 5 × 10⁻²m sodium phosphate buffer solution at a pH of 7.4 and a temperature of 23°C for 0, 10, 30, 90, or 180 min. Under these conditions, SIN-1 produces submicromolar ONOO⁻¹per minute for several hours continuously. 18 

Norepinephrine oxidation was confirmed spectrophotometrically 12with a Shimazu UV-160A (accuracy, ± 0.005 absorbance; reproducibility, ± 0.002 absorbance; Shimazu, Tokyo, Japan). Norepinephrine has a maximal absorbance at 280 nm at a pH of 7.4. However, once it is oxidized, norepinephrine has dual peaks of absorbance at 295 nm (2) and 370 nm (1) when an appropriate concentration is used as a base line. The numbers in parentheses indicate relative amounts of absorbance.

The rats were anesthetized by isoflurane inhalation and were killed by decapitation; the thoracic aortas were isolated for the functional experiments. 19The thoracic aorta was placed in Krebs Henseleit solution (10⁻³m; NaCl, 118; KCl, 4.7; NaHCO³, 25; KH²PO⁴, 1.2; MgSO⁴, 1.2; CaCl², 2.5; and glucose, 10 m; pH, 7.4). Helical strips were carefully prepared under a dissecting microscope. To avoid the possible involvement of endothelium-derived relaxing factors in the mechanical response, the lack of relaxation to acetylcholine was confirmed after the endothelium was rubbed off with filter paper. 19Each strip was carefully suspended in an organ chamber containing 30 ml Krebs Henseleit solution bubbled with 95% O²–5% CO²at 37°C, and the tension changes were recorded isometrically. A resting tension of 0.5 g was applied over a 1-h equilibration period.

As 0.1 ml norepinephrine pretreated with ONOO⁻¹was added into the organ chamber containing 30 ml Krebs Henseleit solution, the final concentration of norepinephrine acting on the vascular strips was 1/300 of the norepinephrine pretreated with ONOO⁻¹. The final concentrations of norepinephrine in the Krebs Henseleit solution and the concentrations of ONOO⁻¹in the reaction with norepinephrine in the sodium phosphate buffer are shown in the text and in the figures.

Contractions were expressed in terms of milligrams of contractile force developed divided by milligrams of wet tissue weight. Relaxations were expressed as a percentage of norepinephrine-induced contraction.

Because ONOO⁻¹decomposes so rapidly, it is not possible to determine a deactivation rate constant by measuring norepinephrine activity over time. Therefore, bimolecular rate constants for reactions of ONOO⁻¹and norepinephrine were determined by a competition assay, as previously described in detail. 16,20Based on the concentration response for ONOO⁻¹-mediated deactivation, an ONOO⁻¹concentration of 5 × 10⁻⁵m was chosen, which would inhibit more than 90% but less than 100% of 10⁻⁷m norepinephrine–induced contraction so that protection by cysteine could be precisely measured. Norepinephrine solutions were prepared in a sodium phosphate buffer, and cysteine was added immediately before ONOO⁻¹. Remaining norepinephrine activity was determined by vasocontraction in a functional experiment. By comparing the ability of different cysteine concentrations to prevent ONOO⁻¹-mediated deactivation of the fixed concentration of norepinephrine, the rate constant for the ONOO⁻¹–norepinephrine reaction can be calculated on the basis of the following equation:

The terms F and k  describe the fraction of norepinephrine deactivated by ONOO⁻¹and the rate constant, respectively. Cysteine reacts with ONOO⁻¹at a known rate (k  ^cysteine) of 5 × 10³m/s at a pH of 7.4 and a temperature of 37°C. When F[norepinephrine]/(1 − F) k  ^cysteineis plotted as a function of cysteine concentration, the reciprocal of the slope gives k  ^{norepinephrine}in molar per second.

During anesthesia with an intramuscular injection of pentobarbitone (30 mg/kg), a patent airway was established by tracheotomy and subsequent tracheal cannulation during spontaneous ventilation. The internal carotid artery and vein were cannulated to create a route for continuous measurement of arterial pressure and a route for infusion, respectively. 21The infusion was maintained with physiologic saline (mixture of 3.4 mg/ml pentobarbitone for anesthetic maintenance) at a rate of 2.0 ml · kg⁻¹· h⁻¹. The rectal temperature was kept stable at 37°C by warm infusion and by laying pieces of aluminum foil over each rat. Drugs were administered intravenously by a microinjector, and each volume was 25–35 μl.

Peroxynitrite and SIN-1 were purchased from Dojindo Laboratories (Kumamoto, Japan), and the other drugs were purchased from Sigma (St. Louis, MO).

Results are expressed as mean ± SD. Group differences were analyzed by one-way analysis of variance and Scheffé F test as post hoc  comparisons for multiple comparison at a significance level of 0.05. These analyses were performed on a personal computer using Stat View II 4.0 software (Abacus Concepts, Berkeley, CA).

After treatment with ONOO⁻¹, the maximal absorbance of norepinephrine shifted from 280 (inset in fig. 1) to 295 nm (fig. 1). In addition, a new peak appeared at 370 nm. The rate of absorbance between 295 and 370 nm was approximately 2:1. The absorbances at 295 and 370 nm increased in a ONOO⁻¹concentration–dependent manner.

Norepinephrine induced concentration-dependent contraction in endothelium-denuded aorta, and 10⁻⁷m norepinephrine induced submaximal contraction. Norepinephrine pretreated with ONOO⁻¹decreased its contractile force in a ONOO⁻¹concentration–dependent manner (EC⁵⁰= 5.1 × 10⁻⁵m;fig. 2A, solid circles). Representative tracings on contraction by norepinephrine pretreated with ONOO⁻¹are shown in figure 2B.

Conversely, norepinephrine pretreated with decomposed ONOO⁻¹(up to 3 × 10⁻⁴m) did not decrease its contractile force (fig. 2A, open circles). An excessively high concentration of decomposed ONOO⁻¹decreased the contractile force of norepinephrine (EC⁵⁰= 2.3 × 10⁻³m).

The decrease in the ONOO⁻¹-pretreated norepinephrine–induced contractile force was pH dependent (i.e ., the decrease reduced at pretreatment in a lower pH buffer;fig. 3).

To test the competition between oxidized and nonoxidized norepinephrine, after norepinephrine (10⁻⁷m) pretreated with ONOO⁻¹(3 × 10⁻⁴m), which had no contractile force (as shown in fig. 2), was added to the Krebs Henseleit solution in an organ chamber, norepinephrine was cumulatively applied. The contraction induced by norepinephrine was as much as that induced without norepinephrine pretreated with ONOO⁻¹(fig. 4).

Norepinephrine (10⁻⁷m) incubated with SIN-1 (10⁻³m) decreased its vasocontractile force in an incubation time–dependent manner (fig. 5). Ten-minute incubation with SIN-1 decreased the vasocontractile force of norepinephrine significantly, and 180-min incubation deactivated norepinephrine completely. The vasorelaxant activities of SIN-1 were stable (approximately 20% of norepinephrine induced contraction) for 180 min (data not shown).

In a preliminary experiment, it was confirmed that cysteine pretreated with ONOO⁻¹had no effect on norepinephrine vasoactivities (n = 5). The rate reaction between ONOO⁻¹and norepinephrine was determined by measuring protection as a function of increasing free cysteine concentrations. By comparing the relative amounts of cysteine and norepinephrine and the known rate constant for the ONOO⁻¹–cysteine reaction, the rate constant for the ONOO⁻¹–norepinephrine reaction was determined to be 6 × 10²m/s (fig. 6).

Intravenous administration of 1 μg/kg norepinephrine increased systolic and diastolic arterial blood pressure significantly (table 1). The increases were prompt and transient. In contrast, administration of norepinephrine pretreated with ONOO⁻¹had no significant effect on arterial blood pressure for 1 h. Figure 7shows the representative continuously monitored arterial blood pressure results.

The maximal absorbance of norepinephrine shifted from 280 to 295 nm after treatment with ONOO⁻¹in a ONOO⁻¹concentration–dependent manner (fig. 1). In addition, a new, smaller peak appeared at 370 nm (the rate of absorbance between 295 and 370 nm was approximately 2:1). As these spectral characteristics corresponded with those of oxidized norepinephrine, 12ONOO⁻¹certainly oxidized norepinephrine in this study. The activities of norepinephrine pretreated with ONOO⁻¹on vascular smooth muscle contractions (fig. 2) and on consequent increases of blood pressure (table 1) were decreased. During the pretreatment of norepinephrine with ONOO⁻¹, ONOO⁻¹oxidizes norepinephrine, and the rest of ONOO⁻¹decomposes in a sodium phosphate buffer. 11Therefore, there are three possible mechanisms to explain the decrease in vasocontractile force of norepinephrine pretreated with ONOO⁻¹in our study: (1) oxidized norepinephrine does not induce vascular smooth muscle contraction, (2) decomposed ONOO⁻¹relaxes vascular smooth muscle and consequently inhibits norepinephrine-induced vasocontraction because ONOO⁻¹or decomposed ONOO⁻¹stimulates cyclic GMP synthesis 22and relaxes 23in vascular smooth muscle, and (3) both (1) and (2). Norepinephrine pretreated with 3 × 10⁻⁴m decomposed ONOO⁻¹did not decrease its vasocontractile activity (fig. 2), and the same concentration of ONOO⁻¹deactivated norepinephrine completely. Therefore, it was suggested that (1) is the most likely mechanism for the decrease in vasocontractile force of norepinephrine pretreated with ONOO⁻¹.

As norepinephrine (10⁻⁷m) pretreated with ONOO⁻¹(3 × 10⁻⁴m) had no contractile force, as shown in figure 2, it is believed that almost all of the norepinephrine was oxidized norepinephrine. During coexistence of the oxidized norepinephrine, norepinephrine-induced contractions were not inhibited (fig. 4). Therefore, oxidized norepinephrine does not have enough affinity to adrenoceptors to inhibit normal norepinephrine binding to adrenoceptors competitively.

It is well known that vascular hyporeactivity to catecholamines, such as dopamine, norepinephrine, and epinephrine, occurs in septic shock. Although the precise mechanisms of vascular hyporeactivity have not been elucidated completely, large amounts of NO produced by iNOS in response to bacterial endotoxins and/or inflammatory cytokines seem to participate in this hyporeactivity. 1–3Recently it was reported that the exposure of norepinephrine to NO or NO-related compounds leads to 6-nitronorepinephrine, 7which has poor vasoactivities. Norepinephrine was almost completely converted by NO or NO-related compounds to 6-nitronorepinephrine, not to oxidized norepinephrine. This nitration reaction requires extremely high NO concentrations (at least > 10⁻²m) and is very slow (rate constant, < 0.1 m/s). 7Therefore, the reaction hardly occurs under normal physiologic conditions, and it may not be a predominant reaction even in sepsis, in which the concentration of NO is micromolar. 24In fact, although 6-nitronorepinephrine was identified in rat brain, the concentration was very low (4 × 10⁻⁸m). 25Conversely, multiple experiments have demonstrated that NO becomes highly reactive when converted to secondary reactive nitrogen species such as ONOO⁻¹. 26Also in this study, ONOO⁻¹reacted with and deactivated norepinephrine at a much faster rate (6 × 10²m/s) and at a much lower concentration (3 × 10⁻⁵m) than NO. These results suggest that the oxidation of norepinephrine by ONOO⁻¹may be a more predominant reaction than the nitration of norepinephrine by NO in sepsis.

The rate constant for ONOO⁻¹and norepinephrine is not fast compared with that of other important antioxidants that react with ONOO⁻¹(table 2). 27However, the speed of the reaction is in proportion not only to the rate constant but also to the concentration of the reactant with ONOO⁻¹. Large amounts of other reactive oxygen species are produced, and the consequent loss of antioxidant concentrations occurs in septic shock. 28Conversely, the concentration of norepinephrine increases because the endogenous norepinephrine released increases in shock, and exogenous norepinephrine is administered in the treatment of hypotension. Therefore, it is possible that ONOO⁻¹might react with norepinephrine, as well as with antioxidants, in septic shock.

It has been suggested that the net exposure to ONOO⁻¹should be shown as the area under the curve of time versus  concentration. 29Therefore, even low concentrations of ONOO⁻¹may be able to deactivate norepinephrine during long-term incubation. Thus, norepinephrine was incubated with an ONOO⁻¹producer, SIN-1. This drug releases both a few micromolars of NO and O²⁻¹· continuously and, consequently, ONOO⁻¹. 30One millimolar of SIN-1 released submicromolar ONOO⁻¹continuously in sodium phosphate buffer solution at a pH of 7.4 and a temperature of 23°C 18and decreased the vasocontractile force of norepinephrine in an incubation time–dependent manner (fig. 5). SIN-1 also relaxes vascular smooth muscle, but the vasorelaxant activities were stable during this study. Therefore, it is believed that the time-dependent decrease in vasocontractile force of norepinephrine must be via  the deactivation of norepinephrine by ONOO⁻¹gradually released from SIN-1. Although ONOO⁻¹is produced from stimulated vascular smooth muscles 31and endothelial cells, 32it is not known how much ONOO⁻¹is produced in the vascular system environment during sepsis. However, it may be noteworthy that leukocytes accelerated to adhere to vascular endothelium by endotoxin can generate submicromolar ONOO⁻¹, 33and the ONOO⁻¹continuously generated during sepsis may be enough to deactivate norepinephrine.

Lower pH shortens the half-life of ONOO⁻¹. 33This may explain why the contraction by norepinephrine pretreated with ONOO⁻¹at lower pH was stronger than that at higher pH (fig. 3). It is well known that lactic acidosis is frequently present in septic shock patients due to decreased blood flow to visceral organs, as well as impaired oxygen extraction. 34The correction of pH by alkaline salts may put patients at risk for deactivation of norepinephrine in the presence of ONOO⁻¹.

Oxidized norepinephrine administered intravenously had no effect on blood pressure (fig. 7, table 1). This result shows that oxidized norepinephrine cannot be reduced to norepinephrine even by the various antioxidants existing in vivo , such as glutathione, cysteine, vitamin C, and vitamin E—i.e ., the reaction between ONOO⁻¹and norepinephrine was irreversible in vivo . If the reaction was reversible and the reduction of oxidized norepinephrine to norepinephrine occurred, the norepinephrine must have increased arterial blood pressure.

In conclusion, norepinephrine was oxidized and deactivated by ONOO⁻¹at a rate constant of 6 × 10²m/s irreversibly in a pH-dependent manner. These results may, at least in part, account for the loss of vasoconstrictor tone to norepinephrine in septic shock.

The authors thank Joseph S. Beckman, Ph.D. (Professor, Department of Biochemistry and Biophysics, Oregon State University, Corvallis, Oregon), for beneficial advice on ONOO⁻¹treatments.