Activation and Activity of the Superoxide- Generating System of Neutrophils from Human Infants

Pediatr. Res. 17: 662-664 (1983)

RONALD G. STRAUSS(*~'AND ESTHER L. SNYDER Department of Pediatrics, University of Iowa College of Medicine, Iowa City, Iowa, USA

Summary stimuli triggers activation of a reduced pyridine nucleotide oxidase that transfers single electrons from NADPH to molecular oxygen, We investigated the superoxide-generating system of neutro- thus reducing oxygen molecules to form superoxide anion (4). As phils obtained from human neonates and their mothers. Using a a consequence of activation, neutrophils experience an oxidative standard assay to measure the cumulative production of superox- burst characterized by increased oxygen consumption, by the ide anion over 15 min, neonatal neutrophils generated equivalent production of superoxide anion, hydrogen peroxide, hydroxyl or increased quantities of superoxide when compared to mothers radical, singlet oxygen, and chemiluminescence, and by increased and controls. Nanamoles of superoxide produced by 2.5 X lo6 activity of the hexose monophosphate shunt (4). The importance unstimulated neutrophils was 7.95 +. 3.45 for neonates, 2.25 f 1.50 of these oxidative reactions has been established by studies of for mothers, and 2.70 + 0.90 for controls. Comparable values for patients with chronic granulomatous disease. Their neutrophils 2 neutrophils stimulated by phorbol myristate acetate were 97.95 cannot mount an oxidative burst and, consequently, are unable to 2.10, 84.75 7.50, and 88.50 2.55, respectively (neonates > + + efficiently kill bacteria. controls, P 0.02). When studied by a continuous assay to assess < To more completely investigate oxidative metabolism in neo- kinetics of superoxide generation, neonatal neutrophils were more natal neutrophils, we studied superoxide generation for two rea- rapidly activated (P 0.02) than were cells from mothers or < sons. (I) Superoxide production is the key reaction in the initiation controls. Seconds elapsed from the addition of 0.667 pg/ml phorbol of the oxidative burst (1 1); accordingly, its characteristics must be until maximal rates of superoxide production were 39.3 2 2.88, defined. (2) Techniques are available to precisely measure the 49.1 3.12, and 52.2 2 4.32 for neonates, mothers, and controls, + kinetics (both activation and activity) of the superoxide-generating respectively. After complete activation, the rates of superoxide system (8, 9). Kinetic abnornalities of the oxidative burst have production were similar for all subjects (about 8 nmole/min/2 x been detected in neonatal neutrophils by chemiluminescence (15, neutrophils under a variety of conditions). lo6 17, 21), and further studies are warranted because of the impre- The increased activity of some aspects of oxidative metabolism cision of the chemiluminescence assay (2, 14, 15, 21). that occurs in neutrophils from human neonates might be explained, at least in part, by a more easily activated superoxide- MATERIALS AND METHODS generating system that promptly initiates the oxidative burst. In contrast, those aspects reported to be decreased cannot be ex- Studies were approved by the local committee governing human plained by abnormalities of superoxide generation per se because investigations, and informed, written consent was obtained. Ve- maximal rates of superoxide production were comparable in neo- nous blood was collected from full-term infants (4-5 ml) and their natal and control neutrophils. Although the mechanisms respon- mothers within 24 h of birth. Adults served as controls. All subjects sible for decreased oxidative metabolism in neonatal neutrophils were healthy and were taking no medications on a regular basis. remain undefined, they apparently affect the oxidative burst at Cell suspensions containing greater than 90% neutrophils were steps subsequent to the early steps in superoxide generation. obtained by dextran sedimentation, Ficoll-Hypaque centrifugation and hypotonic lysis by an isolation technique previously Abbreviation shown to yield uniform suspensions of morphologically mature neutrophils from infants, mothers, and controls (18). Neutrophils PMA, phorbol myristate acetate were suspended in Krebs-Ringer .phosphate . buffer (pH 7.4). Standard assay to measure the cumulative generation of super- Newborn infants are more susceptible to infections and expe- oxide. The ability of neutrophils to produce and release superoxide rience greater morbidity and mortality from infections when com- anion is usually assessed by measuring the total quantity of pared to older individuals. Although several abnormalities of superoxide dismutase inhibited ferricytochrome C reduction oc- neonatal body defense mechanisms have been described, reports curring over a defined period of time (19). The total volume of all are contradictory and controversy exists regarding the significance reaction mixtures was 1.6 ml, and all reactions contained 0.1 ml of each defect. Neutrophils provide an important defense against of 1.2 mM ferricytochrome C (80 pM final concentration) and 0.3 bacterial infections. Although complete agreement does not exist ml of a solution containing either superoxide dismutase or bovine regarding the properties of neonatal neutrophils, abnormalities of serum albumin (30 pg of protein final concentration). Stimulation chemotaxis, phagocytosis, microbicidal activity, and biochemistry reactions received 1.0 ml of neutrophil suspension (2.5 x lo6 cells) have been reported (Reviewed in 16). Studies of oxidative metab- plus either 0.1 ml opsonized zymosan (50 zymosan particles per olism are particularly perplexing. Increased activity has been neutrophil) or 0.1 ml PMA. PMA (Consolidated Midland Corp., reported for some aspects including oxygen consumption (lo), Brewster, NY) was initially dissolved in dimethylsulfoxide at a superoxide anion generation (1, 12), hydrogen peroxide produc- concentration of 1 mg/ml and frozen in aliquots. It was thawed tion (20), hexose monophosphate shunt activity (3, 10, 17), and and prepared fresh daily by dilution with phosphate-buffered nitroblue tetrazolium dye reduction (5, 6, 10). In contrast, de- isotonic saline (pH 7.3) so that the final concentration was 0.667 creased activity has been reported for chemiluminescence (7, 15, pg/ml. Unstimulated reactions received an identical quantity of 17, 21) and hydroxyl radical formation (1). neutrophil suspension plus 0.1 ml of buffer. Components were Activation of neutrophils by phagocytosis or exposure to soluble incubated with constant agitation at 37OC for 15 min. Reactions SUPEROXIDE-GENERATING SYSTEM 663 were stopped by plunging the vials into an ice bath, and neutro- linear portion of the curve intercepted the stable preactivation phils were immediately removed by centrifugation. Supernatant baseline. Activity of the superoxide-generating system was deter- fluids were transferred to clean tubes and promptly read spectro- mined by the linear rate of absorbance change that followed photometrically at 550 nm. The quantity of superoxide anion complete activation. The initial rate was consistently linear, and produced was expressed as nmol ferricytochrome C reduced in it was computed using measurements that encompassed the first 15 rnin by 2.5 x 10%eutrophils using the extinction coefficient 2 min. The change in absorbance then slowed, and the later rate E550 nm = 2.1 x lo4 m-' cm-' . of ferricytochrome C reduction was measured during the fifth Continuous production of superoxide anion. Both activation and minute after addition of PMA. As in the standard assay, super- activity of thesuperoxide-generating system were measured by the oxide anion generation was expressed as nmole of ferricytochrorne continuous measurement of suveroxide dismutase inhibited ferri- C reduced. cytochrome C reduction. A keviously described spectrophoto- Statistical analysis was performed in consultation with Dr. Leon metric assay (8, 9) was modified. The reaction was monitored at F. Burmeister, Department of Preventive Medicine and Environ- 37°C by Gilford model 250 spectrophotometer attached to a mental Health. The significance of differences between means was model 6050 chart recorder (Gilford Instrument Laboratories, tested by the two group t, paired t and sign tests. Oberlin, OH) with the following settings: wave length 550 nm; slit width 0.03 mm; calibrated absorbance 0.6; chart speed 1 inch/ RESULTS min. Cuvettes were filled with 1.1 ml of buffer, and after prewarm- ing at 37"C, 0.2 ml of neutrophil suspension (2 x 10%ells), 0.1 The total quantity of superoxide anion released and accumu- ml ferricytochrorne C (8 x lop5M final concentration) and 0.1 of lated in the reaction mixture as reduced ferricytochrorne C during either superoxide dismutase or buffer was added. Factors known 15 min of incubation (standard assay) is recorded in Table 1. to influence results of the assay such as pH, concentration of Maternal and control values were similar under all experimental divalent cations, and quantities of neutrophils and stimulants were conditions. The mean value for superoxide generation by unstim- carefully controlled (8). Temperature was precisely maintained at ulated neonatal neutrophils was nearly 3-fold greater than that of 37"C, and the reaction was allowed to stabilize (no further change controls; however, the difference was not statistically significant. in absorbance) for 2-5 min. Neutrophils were stimulated by Superoxide generation was markedly increased in neutrophils adding 0.1 ml of PMA (final concentration either 0.067 or 0.667 from all subjects when they were stimulated either by phagocytosis yg/ml), and the reaction was recorded for 6 min. It was established of opsonized zymosan or by exposure to PMA. Superoxide gen- in pilot experiments that the quantity of ferricytochrome C used eration in response to PMA by neonatal neutrophils was signifi- would not limit the reaction, and that superoxide dismutase cantly greater (P < 0.02) than that of controls. completely inhibited the reduction of ferricytochrorne C. Results of the continuous assay are presented in Table 2. ~h example of the recorded reaction is presented in Figure 1. Maternal and control values did not differ under any of the Activation of the superoxide-generating system was defined as the conditions. Activation of the superoxide-generating system was seconds elapsed (lag time) between the addition of PMA and the significantly more rapid (P c: 0.02) in neonatal neutrophils than achievement of the maximal rate of ferricytochrorne C reduction. in controls at either concentration of PMA as indicated by shorter It was calculated by measuring the distance from the point at lag times. Rates of activation were similar for both doses of PMA. which PMA was added to the point at which the back-extrapolated Activity of the superoxide-generating system after complete activation was similar in neutrophils from neonates; their mothers and controls whether measured immediately or later (5 rnin after activation). The initial rate was dose-related (i.e., more ferricytochrome C was reduced per minute by neutrophils exposed to the lower dose of PMA) (P < 0.02). The difference was slight, and the mechanism remains undefined. It cannot be easily explained by a toxic effect of the higher dose of PMA because viability by dye exclusion was not decreased, and the later rate was similar regard- less of the dose of PMA. The mean later rate (5 rnin after complete activation) was significantly slower (P< 0.02) than the initial rate in all subjects, and presumably reflects a change (possibly oxidative auto-inactivation) in the superoxide-generating system of the neutrophils. The mechanism responsible cannot be defined by data from this study, but the decreased rate cannot be ascribed to substrate limitations imposed by ferricytochrome C consumption since it was present in great excess.

DISCUSSION Lag Time I Using the standard assay to measure the cumulative production of superoxide anion over time, we found that neonatal neutrophils -2-101 23456 generated equivalent or increased quantities of superoxide when compared to their mothers and to adult neutrophils. Although the Time in Minutes mean values of superoxide anion generation were larger for neo- Fig. 1. The line depicts the change in absorbance at 550 nm that occurs natal neutrophils under all experimental conditions, superoxide over time after the stimulation of neutrophils by phorbol myristate acetate production was significantly greater than that of mothers or (PMA). controls only when neutrophils were stimulated by PMA. These

Table 1. Superoxide anion production (nmole, mean f S.E.M.) by neutrophils from neonates, their mothers and controls (2.5 x 10" neutrophils incubated for 15 rnin at 37°C) Conditions Neonates (n = 7) Mothers (n = 7) Controls (n = 7) Unstimulated 7.95 + 3.45 2.25 + 1.50 2.70 +. 0.90 Zymosan stimulated 81.45 +- 5.70 70.80 + 6.30 77.85 + 5.25 PMA stimulated 97.95 + 2.10' 84.75 + 7.50 88.50 rt 2.55 ' P < 0.02 neonates versus controls. 664 STRAUSS AND SNYDER

Table 2. Rates of superoxide anion production by neutrophilsfrom neonates, their mothers and controls in response to phorbol myristate acetate 0.067 or 0.667 w/ml

Conditions Neonates (n = 10) Mothers (n = 10) Controls (n = 10) Initial lag (seconds f S.E.M.) 0.067 + pg/ml 42.1 + 2.46l 54.7 k 3.30 52.0 + 4.14 0.667 pg/ml 39.3 ? 2.8S1 49.1 f 3.12 52.2 k 4.32 Initial rate (nmole/min/2 X lo6 neutrophils + S.E.M.) 0.067 pg/ml 9.33 k 0.59 8.93 + 0.64 9.49 + 0.56 0.667 pg/ml 8.11 k 0.66 7.70 ? 0.45 8.08 + 0.6 1 Later rate (nmole/min/2 X lo6 neutrophils + S.E.M.) 0.067 pg/ml 6.30 f 0.38 7.22 + 0.43 7.31 f 0.30 0.667 pg/ml 6.50 r 0.42 6.37 ? 0.30 6.59 + 0.37 ' P < 0.02 neonates versus controls, P < 0.001 neonates versus mothers. findings are somewhat similar to those of Ambruso et al. (1) natal neutrophils and the mechanisms responsible remain unde- although they found significantly increased production in response fmed. to opsonized zymosan rather than to PMA. When studied by the continuous superoxide-generating assay REFERENCES AND NOTES 1. Ambmso, D. R., Altenburger, K. M., and Johnston, R. B. Jr.: Defective oxidative to assess the kinetics of the reaction, neonatal neutrophils were metabolism in newborn neutrophils: discrepancy between superoxide anion more rapidly activated than were cells from their mothers or and hydroxyl radical generation. Pediatrics, 64(S~pp[):722 (1979). controls. Once completely activated, however, the activitv of the 2. Andersen, B. V. and Brendzel, A. M.: Use of unique chemiluminescence spec- superoxide-genera&g sfstem was similar in all goups when trometer in a study of factors influencing granulocyte light emission. J.'I~- munol. Methods 19: L279 (1978). assessed by the rate of superoxide generation. After a 3. Anderson, D. C., Pickering, L. K., and Feigin, R. D.: Leukocyte function in few minutes of maximal activity, the rate of superoxide anion normal and infected neonates. J. Pediatr.., 85:-- 420 (1974).\- , production slowed both in neonatal and control neutrophils. This 4. Babior, B. M.: Oxygen-dependent microbial killing by phagocytes. N. Engl. J. decrease in activity was comparable in all groups during the first Med., 298: 659 (1978). 5. Chandler, B. D., Kapoor, N., Barker, B. E., Boyle, R. J., and Oh, W.: Nitroblue few moments, but then became quite variable with the passage of tetrazolium test in neonates. J. Pediatr., 92: 638 (1978). time. We are unaware of other reports in which neutrophils from 6. Humbert. J. R.. Kurth. M. L.. and Hathawav. W. E.: Increased reduction of nitroblue tetrazolium bv neutrophils of neGborn infants. Pediatrics, 45: 125 neonates were studied bv the continuous assav:,, however. mixed leukocyte suspensions from human fetuses have been evaluated (1970). 7. Mills, E. L., Thompson, T., Bjorksten, B., Filipovich, B. S., and Ouie, P. G.: The and were found to exhibit normal activation but a slightly slower chemiluminescence response and bactericidal activity of polymorphonuclear rate of superoxide generation than adult leukocytes (9). neutrophils from newborns and their mothers. Pediatrics, 63: 429 (1979). Results in Table 1 of the standard assay (cumulative production 8. Newburger, P. E., Chovaniec, M. E., and Cohen, H. J.: Activity and activation of over a futed period of time) canaot easily be directly compared to the granulocyte superoxide-generating system. Blood, 55: 85 (1980). 9. Newburger, P. E.: Superoxide generation by human fetal granulocytes. Pediatr. those gf the contirylous assay (Table 2). Under the conditions Res., 16: 373 (1982). employed, the maximal rate of superoxide generation was main- 10. Park, B. H., Holmes, B., and Good, R. A,: Metabolic activities in leukocytes of tained for only a few minutes in the continuous assay. As activity newborn infants. J. Pediatr., 76: 237 (1970). waned, an almost infmite number of rates were observed among 11. Root, R. K. and Metcalf, J. A,: H20Zrelease from human granulocytes during phagocytosis. J. Clin. Invest., 60: 1266 (1977). both individuals and groups of subjects. It was therefore impos- 12. Shigeoka, A. O., Charette, R. P., Wyman, M. L., and Hill, H. R.: Defective sible to accurately compute the total (cumulative) quantity of oxidative metabolic responses of neutrophils from stressed neonates. J. Pediatr., superoxide generated over time using data from the continuous 98: 392 (1981). assay unless one includes all of the rates of superoxide production 13. Shigeoka, A. 0. and Santos, J. I.: Functional analysis of neutrophil granulocytes from healthy, infected, and stressed neonates. J. Pediatr., 95: 454 (1979). that apply during the time interval studied. It is likely that the 14. Stevens, P., Winston, D. J., and VanDyke, K.: In vitro evaluation of opsonic and continuous assay more precisely assesses the status of the super- cellular granulocvte function by luminol-dependent chemiluminescence. Util- oxide-generating system. ity in with severe ne;ropenia and cellular deficiency states. Infect. Increased activity in neutrophils from healthy neonates has Immun., 22: 41 (1978). 15. Strauss, R. G., Rosenbergr, T. G., and Wallace, P. D.: Neutrophil chemilumi- been reported for some aspects of oxidative metabolism including nescence during the first month of life. Acta Haematol., 63: 326 (1980). oxygen consumption (lo), superoxide generation (I, 12), hydrogen 16. Strauss, R. G. and Mauer, A. M.: Formed elements of human blood; in Stave, peroxide production (20), hexose monophosphate shunt activity Perinatal physiology; 2nd ed., pp. 199-214 (Plenum Publishing, New York (3, 10, 17) and nitroblue tetrazolium dye reduction (5, 6, 10). --1978) -,. Findings in the present study suggest that these increases might 17. Strauss, R. G. and Seifert, M. J.: Oxidative metabolism in cord-blood polymorphonuclear leucocytes. Arch. Dis. Child., 53: 78 (1978). be explained, at least in part, by a more easily activated superox- 18. Strauss, R. G., Snyder, E. L., Wallace, P. D., and Rosenberger, T. G.: Oxygen- ide-generating system. On the other hand, chemiluminescence (7, detoxifying enzymes in neutrophils of infants and their mothers. J. Lab. Clin. 15, 17, 21) and hydroxyl radical formation (1) have been found to Med., 95: 897 (1980). be decreased in neonatal neutrophils. We cannot explain these 19. Strauss, R. G.: Neutrophil assays as modified for infants. J. Clin. Lab. Immunol., 3: 133 (1980). abnormalities by decreased activity of the superoxide-generating 20. Strauss, R. G. and Snyder, E. L.: Neutrophils from human infants exhibit system, at least during the early moments of activity, because the decreased viability. Pediatr. Res., 15: 794 (1981). maximal rates of superoxide production and the initial decreases 21. Van Epps, D. E., Goodwin, J. S., and Murphy, S.: Age-dependent variations in polymorphonuclear leukocyte chemiluminescence. Infect. Irnmun., 22: 57 as the response waned were similar in neonatal and control (1978). neutrophiis. The defect(s) must lie distal to the initial steps in 22. Requests for reprints should be addressed to: Dr. Ronald G. Strauss, Department superoxide anion production per se. It must be noted, however, of Pediatrics, University Hospitals and Clinics, Iowa City, IA 52242. that decreased oxidative metabolism has been observed most 23. Supported in part by grant R01 HD00964 and Research Career Development frequently in sick infants (3, 12, 13), and only healthy neonates Award KO4 HD00255 (Dr. Strauss) from the National Institute of Child Health and Human Development, National Institutes of Health. were evaluated in the present study; thus, the clinical relevance of 24. Received for publication July 2, 1982. those aspects of oxidative metabolism that are decreased in neo- 25. Accepted for publication December 21, 1982.