Ann. Medit. Burns Club - vol. V1 - n. 3 - September 1993

SUMMARY. Canine models inflicted with smoke inhalation injury were used in this experiment. The NAD(P)H oxidase activity and superoxide anion (02) production of neutrophil were studied in vitro with zymozan as stimulant. The concomitant changes of blood gas, lung water volume, chest X-ray, and pulmonary pathornorphology were also observed, in order to demonstrate the relationship between toxic oxygen metabolites and acute lung injury. The results showed that: 1) CO poisoning, hypoxaemia, metabolic acidosis, respiratory alkalosis and lung injury developed rapidly in the early stage after smoke inhalation; 2) the WBC count decreased rapidly at an early stage after smoke inhalation; 3) the activity of neutrophil NAD(P)H oxidase increased gradually from 30 min to 6 h and then decreased, reaching its pre-injury level at 12 h post-injury. In contrast, the 0-2 production of polymorphonuclear neutrophils (PMN) in vitro decreased progressively from 30 min to 6 h and returned to its pre-injury level at 12 h post-injury. We explain these follows: 1) neutrophils produced a large amount of superoxide anion in vivo within 6 h post-injury, accompanied by a huge consumption of NAD(P)H; 2) six hours postinjury, newborn granulocytes with less activity were released from the bone marrow into the blood stream, and the activity of inflammatory factors such as C5a etc. gradually decreased. It is thus proposed that neutrophils accumulate in pulmonary tissue after smoke inhalation and experience a "respiratory burst" characterized by the activation of cytoplasmic NAD(P)H oxidase and the production of a large amount Of 0-2 as well as other toxic oxygen radicals, which participate in the pathogenesis of acute lung injury in the early stage after smoke inhalation injury.

Neutrophils and their toxic oxygen metabolites have been implicated recently as the principle mediators of the increased alveolar capillary membrane (ACM) permeability in Adult Respiratory Distress Syndrome (Joseph and Ward, 1982; 1985). In order to clarify the role of these radicals in the pathogenesis of inhalation injury, we employed the model of canine smoke inhalation injury to observe the post-injury changes of superoxide anion (02) production and NAD(P)H oxidase activity of neutrophils in vitro.

Twenty male adult dogs, weighing 8.5-14 kg, were intravenously anaesthetized with 3% pentobarbital sodium at a dose of 20 to 30 mg/kg. The femoral artery was intubated with a silicone rubber tube, the pulmonary artery with a catheter and the trachea with a tracheal tube. The canine smoke inhalation injury model established in our Burn Institute was adopted (Zhu et al., 1988). The dogs inhaled smoke twice, 5 min each time with a 5 min interval.

In particular:

1. Gas analysis and the white cell count of femoral arterial blood were performed before and 5 min, 30 min, 2 h, 6 h and 12 h after inhalation injury.
2. Neutrophil 0-2 production and NAD(P)H oxidase activity were measured before and 30 min, 2 h and 12 h after inhalation injury.
3. Lung water volume (LWV, dry-wet weighing method) and the pulmonary pathomorphological study were carried out immediately after sacrifice of the dogs 12 h post-injury.
4. Chest roentgenograms were made before and 6 h and 9 h post injury.

Neutrophil 0-2 production was examined in vitro according to Michele Markert's method (Markert et al., 1984), as follows.

A 10 ral centrifuge tube was filled first with Percoll den-iixing solution and second with 5-7 m] of femoral arterial blood mixed with 0.85% sodium chloride containing heparin (10g/ml) and EDTA-Na2 (Img/ml). The primed tube was centrifuged at 2000 rpm for 20 min and then the supernatant and lymphocytes were discarded, while the granulocyte layer was collected and suspended; in this the contaminated erythrocytes were dissolved with distilled water. The suspension was then washed twice and regulated to a concentration of 1.25-5 x 10'/nil. The neutrophil viability was more than 95%, as identified by the method of trypan-blue dye exclusion.

Zymozan (Sigma) was mixed in HBSS (Hank's balanced salt solution, pH 7.2) to make a 10 mg/ml suspension, which was boiled for 30 min, washed, added with self-pre-injury serum (0.1 ml serum/mg zymozan), incubated at 37 'C for 30min and washed twice thereafter. It was finally regulated at 8 mg/ml, ready for use.

The 02 generation by polymorphonuelear neutrophils (PMN) was assayed indirectly by the superoxide-dismutase(SOD)-inhibitable reduction of ferricytochrome C. Briefly, PMN (0.7 ml/tube) were incubated in 1.5 ral HBSS containing 50 g/1 ferricytochrome C (30 mglml) and the stimulus (opsonized zymozan 6 ing) with and without SOD (30 pig) for 12 min at 37 'C in an agitator. The suspensions were centrifuged (150 g, 10 min at 4 'C) and 02 production of PMN was deterniined by the difference in the absorbance of the cultured PMN supernatant with and without SOD, which was obtained by spectrophotometry at a wave length of 550 nin.

The method used was David's (David et aL, 1977):

Neutrophil suspension was performed by the aforementioned method. The PMN suspension was mixed with the opsonized zymozan (10 mg/iffi) and incubated at 37 'C for 30 min. The mixture was then centrifuged. The pellet was mixed with sucrose and then homogenized. The homogenate supematant was aspirated after 500g and 27000g centrifugation successively. 0.2 nil of the supematant was put into one tube as sample and 0.2 m] of sucrose solution as control. 0.2 nil NAD(P)1-1 solution (freshly prepared, 0.2 mM) and 0.6 ml potassium phosphate buffer were added to each tube. The two cuvettes were then incubated at 25 'C for I hr and I ml of 0.4 N HC104 solution was added to each. The mixtures were then centrifuged at 3000 rpm for 10 min. 0.2 in] supernatant was then taken from each tube and added to another two matched tubes. 0.3 ml ION NaOH solution was then poured into each tube. The two tubes were agitated and incubated at 38 'C for 30 min and then 2.5 nil distilled water was poured into each tube. Spectrofluorometry was used to determine the fluorescence of the supernatants instantly with quinine sulphate as standard control.

2. Making of the NADP + standard curve

NADP + was dissolved in 0. 1 M Tris buffer and a series of gradient solutions, in which the final NADP + masses were 14.868 ng, 74.34 ng, 148.68 ng, 743.4 ng and 1486.8 ng respectively, was made. 0.2 in] was taken from each gradient and added with 0.3 ml 10 N NaOH respectively. Fluorometry was carried out as before. The fluorescent values were plotted as the horizontal coordinate and the substrate gradient weights as the vertical ordinate. Simple linear regression was applied to evaluate the "ear regression equation.

All data were expressed by :~ ±s. Stable preinjury values were used as control. The WBC count and 02 production presented by the ratio of the postinjury to pre-injury value, which was considered to be 1, were analysed by Student's t test. The data of NAD(P)H oxidase activity and blood gas analysis were treated by single factor analysis of variance. LAtV values were compared with those obtained in normal dogs (Wu et al., 1982) and Student's t test was employed (* 0.01 < p < 0.05, **.p < 0.01).

1. Changes of respiratory function

The respiratory rate decreased immediately postinjury and then gradually increased. Pa02 and PaC02 both decreased progressively. A-aD02 and Qs/Qt began to increase early, at 5 min post-injury, and maintained a high level thereafter. These findings indicated that there were severe hypoxia and hypoxaemia in the early postinjury stage.

2. Change of blood COHb (carboxyhaemoglobin %) (Table 1)

This increased markedly post-injury at an early stage, especially between 5 min and 2 h, which denoted obvious CO poisoning in the injured dogs.

3. Blood acid-base imbalance

HC03-, SBC, BEb and TC02 all began to decline 5 min post-injury, indicating the presence of metabolic acidosis. The decrease of PaC02 also indicated respiratory alkalosis.

4. Change of neutrophil 0~2 production (Table 2)

This decreased clearly post-injury, especially from 2 h to 6 h.

5. NADP + standard curve and regression equation

Y = 79.9077X - 52.1519
where X = fluorescent valute, Y = substrate weight, r = 0.993429, p < 0.01.

6. Change of neutrophil NAD(P)11 oxidase activit (Table 3)

This increased post-injury, markedly from 2 h to 6 h.

7. LWV

Normal group: 79.594 ± 0.974%; injured group: 84.34 ± 1.851% (p < 0.01).

8. Chest roentgenogram

There was tracheal constriction and a faint, opaque or uneven shadow in one or both sides of the lung fields.

9. WBC count (Table 4)

This decreased immediately at 5 min to 2 hr postinjury and increased sharply from 6 h to 12 h postinjury.

10. Pathomorphological changes

a) Macroscopically: there was a large amount of white and pink secretion flooding in the tracheobronchial lumen. The mucosa were congested and oedematous. It could be seen frequently that yellowish-white pseudomembranes coated up the tracheal inner wall. The lung was remarkably oedematous with patches of bleeding and atelectasis. When sectioned, there was a large amount of reddish and frothy secretion leaking from the cut surface of the lung. The lesions were found mostly in the middle and lower lung lobes.
b) Microscopically: there was aggregation of granulocytes, fibrins and erythrocytes in the bronchioles, as well as diffuse congestion, pulmonary oedema and granulocyte infiltrations. In some alveoli there was heavy bleeding or compensatory dilatation of the alveoli, and partial consolidation and collapse of the interstitial frame. There were also interstitial congestion, oedema (even breaking) and structural disruption.

After smoke inhalation in dogs, we observed carbon monoxide poisoning, hypoxaemia, metabolic and respiratory alkalosis, obvious congestion, oedema and haemorrhage of the tracheal mucosa, and in addition interstitial and alveolar pulmonary oedema, atelectasis and haemorrhage, indicating that the inflicted animals suffered from severe inhalation injury. All the indices proved that the model adopted in this experiment was stable and appropriate for the present study.

The experimental results suggest that the neutrophil oxygen metabolism is increased. We therefore suppose that neutrophil and its oxygen metabolites played a role in the initiation and development of inhalation pulmonary injury. This conclusion is based on the following reasoning:

1. The pathomorphological study found that there was granulocyte infiltration and/or aggregation within the alveoli, capillaries and intrapulmonary blood vessels, while the peripheral WBC count decreased sharply 5 min post-injury. It seemed possible that the peripheral granulocytes were trapped in the pulmonary circulation. The causes inducing the granulocyte aggregation were multiple. Besides the direct damage resulting from inhaled smoke, the secondary injury to the alveolar macrophage, leading to the release of chernotactic factors, was an important cause. Another important cause was the activated complements (Huang et al., 1984), which led to increasing C5a production (Gerd et aL, 1983), neutrophil chemotaxis and aggregation in the pulmonary capillaries.
2. The enhancement of neutrophil NAD(P)H oxidase activity post-injury suggests that aggregated neutrophils experienced a "respiratory burst" (Sbarra et al., 1959) under the action of chernotactic factors. The "respiratory burst" exhibited increased oxygen consumption and hexose monophosphate shunt metabolism. Simultaneously, the NAD(P)H oxidase located at the cellular membrane was synthesized and activated. It worked as one of the members of the electron transferring system and transferred one electron from NAD(P)H to the oxygen molecule to form a superoxide anion (02)Thus increased NAD(P)H oxidase activity implies increased production of 02- 02 can be further transferred by SOD to H202. Accompanying neutrophil degranulation, a large amount of lysosome enzymes were released to cause tissue injury in coordination with the oxygen radicals (Charles et al., 1983). MPO (myeloperoxidase) was also released by neutrophil degranulation and could act on H202 to form a large amount of 02 and HOCI with the participation of halogen. Furthermore, 02 and H202 could interact on each other to form OH- in the presence of ferrous ion (Fel+). These secondary oxygen radicals have stronger biological activities and can cause an increase of membranous lipid peroxidation of the pulmonary histocytes (alveolar epithelium, interstitial cells and endothelial capillary cells, inactivation of intracellular enzymes, DNA injury and destruction of pulmonary interstitial structural proteins and glucoproteins (Joseph and Ward, 1982, 1985; Huang et al., 1984). Also, OH- can inactivate a Pantitrypsin by oxidizing its residual methionine, so as to enhance the tissue injury actions of neutrophil lysosome enzymes, acidic hydrolase and elastase. As a result, pulmonary capillary endothelia, interstitial tissue and alveolar epithelia are injured and alveolar capillary membrane permeability is increased, which finally leads to pulmonary oedema, haemorrhage, etc.
3. The 0-2 production tended to decline from 30 min to 6 h post-injury and bounced back to approach pre-injury level at 12 h post-injury. This might be due to the fact that we carried out the experiment in vitro and the neutrophils had been activated in vivo' having experienced the "respiratory burst" accompanied by the production of a large amount of 0-2. The latter was further transformed by SOD into H202- 0-2 can pass through the cellular membrane into the cytoplasm either by anion passage or by direct spreading across the membrane. The intracellular glutathione synthetase was therefore activated to diminish H202, as shown in the following figure: The figure shows that NAD(P)H was therefore consumed to regenerate the reduced form of glutathione. When we examined PMN 02- production in vitro, there was therefore not enough NAD(P)H to be used and 0-2 production thus decreased. Some researchers have reported that the neutrophil "respiratory burst" lasts from about 30 to 60 min (Jardl et al., 1978). This may be also related to the PMN NAD(P)H and/or energy consumption postinjury. The time of test procedure in this experiment was more than 60 min from the harvesting of the neutrophils to the determination of 02 production. The value therefore declined.
4. The reason for neutrophil NAD(P)H oxidase activity decreasing and 02 production returning to the near pre-injury level at 12 h post-injury might be the decreased stimulation from inhalation injury. These results coincided with other results we obtained in past experiments showing that the activation peaks of inflammatory mediators, such as complements, also occurred 4 to 6 h post-injury. In addition, there might be newborn neutrophils released from the bone marrow into the blood stream, a possibility supported by the increase of the WBC count 6 hr after inhalation injury. As the newborn neutrophils had not yet been stimulated directly by inhalation injury and their chemotactic activity had not increased, they would not experience the "respiratory burst" in vivo and NAD(P)H consumption would be less.

To sum up, our results suggest that the neutrophils and their toxic oxygen metabolites play an important role in the pathogenesis of smoke inhalation injury in the early post-injury stage. However, as no direct method exists today for the examination of neutrophil function in vivo, investigation is needed to resolve this problem.

RESUME. Dans cet expériment nous avons utilisé des modèles canins atteints de lésion par inhalation de fumée. Nous avons étudié in vitro l'activité de NAD(P)H oxydase et la production par l'anion superoxyde (0-2) de neutrophiles avec zymozan comme stimulant. Nous avons aussi observé les variations concomitantes des gas sanguins, du volume de la liquide pulmonaire, des radiographies thoraciques et de la pathomorphologie pulmonaire afin de démontrer la rélation entre les métabolites de l'oxygène toxiques et la lésion pulmonaire aiguë. Les résultats ont montré que: 1) l'intoxication de CO, l'hypoxémie, l'acidose métabolique, l'alcalose respiratoire et les lésions pulmonaires se manifestaient rapidement dans la première phase après l'inhalation de fumée; 2) la numération des cellules sanguines blanches diminuait rapidement dans la première phase après l'inhalation de fumée; 3) l'activité de l'oxydase des neutrophiles NAD(P)H augmentait depuis 30 minutes jusqu'à 6 heures et ensuite diminuait pour retourner au niveau pré-lésionnel 12 heures après la lésion. Au contraire, la production de PMN in vitro diminuait progressivement depuis 30 minutes jusqu'à 6 heures et retournait au niveau pré-lésionnel 12 heures après la lésion. Nous en donnons les explications suivantes: 1) les neutrophiles ont produit une grande quantité d'anion superoxyde (02) in vivo entre 6 heures après la lésion, et dans le même temps une énorme consommation de NAD(P)H; 2) à 6 heures après la lésion, des granulocytes nouveaux-nés avec une activité mineure ont été libérés de la moelle dans la circulation sanguine, et l'activité des facteurs inflammatoires comme C5a etc. diminuait progressivement. Les auteurs en concluent que les neutrophiles s'accumulent dans les tissus pulmonaires après l'inhalation de fumée et subissent une "explosion respiratoire" caracterisée par l'activation de l'oxydase cytoplasmatique NAD(P)H et par la production d'une grande quantité de 0-2 et d'autres radicaux de l'oxygène qui participent à la pathogenèse de la lésion pulmonaire aiguë dans la première phase après la lésion par inhalation de fumée.

BIBLIOGRAPHY

1. Charles C.G., Spragg R., Revak S.D.: Pathogenesis of the Adult Respiratory Distress Syndrome: evidence of oxidant activity in bronchoalveolar lavage fluid. J. Clin. Invest., 71: 754-61, 1983.
2. David I., Dechatelet L.R., Spitanagel J.K. et al.: Comparison of NADH and NADPH oxidase activities in granules isolated from human polymorphonuclear leukocytes with a fluorometric assay. J. Clin. Invest., 59: 282-90, 1977.
3. Gerd T.O., Beauchamp C., Menapace D.: Oxygen radical dependent lung damage following thermal injury of rat skin. J. Trauma, 23: 369-77, 1983.
4. Huang Wen-hua et al.: The kinetic changes of serum complement hemolytic activities of total and alternative pathway as well as the serum C3 level following thermal injury. Acta Acad. Med. Mil. Tert. (Chinese), 6: 61, 1984.
5. Jardl R.C. et al.: Termination of the respiratory burst in human neutrophils. J. Clin. Invest., 61: 1176-85, 1978.
6. Joseph F.C., Ward P.A.: Role of oxygen-derived free radicals and metabolites in leukocytes dependent inflammatory reactions. Am. J. Pathol., 107: 379-418, 1982.
7. Joseph F.C., Ward P.A.: Polymorphonuclear leukocytemediated cell and tissue injury: oxygen metabolites and their relations to human disease. Hum. Path., 16: 973, 1985.
8. Markert M. et al.: Measurement of 0, production by human neutrophils. The preparation and assay of NAD(P)H oxidase containing particles from human neutrophils. Methods in Enzymology, 105: 358, 1984.
9. Sbarra A.J. Karnovsky M.L.: The biological basis of phagocytosis. 1: Metabolic changes during the ingestion of particles by polymorphonuclear leukocytes. J. Biol. Chem., 234:1355,1959.
10. Wu Zhenzhong et al.: The changes of canine lung water volume after severe steam inhalation injury. J. Third Mil. Med. Coll. (Chinese.), 4: 110-13, 1982.
11. Zhu Pei-fang, Li Ao (Ngao) et al.: Comparisons of early changes between severe steam and smoke inhalation injury in dogs. Chin. J. Plas. Surg. and Bums, 4: 136-8, 1988.