The ability of granulocytes to generate superoxide anions and hypochlorite during phagocytosis: comparison of neonatal granulocytes with adult granulocytes.

Biol Neonate 1990;58:145-151

© 1990 S. Karger AG, Basel 0006-3126/90/0583-0145S2.75/0

The Ability of Granulocytes to Generatè Superoxide Anions and Hypochlorite during Phagocytosis: Comparison of Neonatal Granulocytes with Adult Granulocytes Akira Nishidaa, Hirokazu Kimuraa, Katsuaki Sugiokab, Minoru Nakanob a Department o f Neonatology and Laboratory, Gunma Children’s Medical Center, Hokkitsu; bCollege o f Medical Care and Technology, Gunma University, Maebashi, Gunma, Japan

Key Words. Granulocytes • Superoxide anions • Hypochlorite • Cypiridina luciferin analog • Luminol • Chemiluminescence

Introduction Many investigators have employed lu minol (5-amino-2,3-dihydro- 1,4-phtalazinedione) (L) as an agent to amplify the lumi nescence response of activated macrophages and leukocytes [1-4]. L-dependent chemilu minescence (L-CL) by phagocytizing leuko cytes had been reported to be highly depen dent upon a myeloperoxidase (MPO)-mediated reaction that probably involves the generation of hypochlorite (OCL) [3]. The

luminescence is believed to be in good agree ment with bactericidal activity of leukocytes [4], Individuals with MPO deficiency, how ever, are usually healthy, whereas those with chronic granulomatous disease, who are characterized by failure to generate Superox ide anions (O 2), suffer from recurrent severe infections. Accordingly, it is of clinical im portance to check leukocytes for their ability to generate O 2 during phagocytosis. Recently, it was found that a chemilumi nescence probe with cypiridina luciferin

Downloaded by: East Carolina University - Laupus Library 150.216.68.200 - 1/16/2019 12:51:47 PM

Abstract. The ability of granulocytes to generate superoxide anions (O 2 ) and hypochlorite (OCL) during phagocytosis was investigated using peripheral blood samples from adults and cord blood samples from neonates, using the chemiluminescence probe cypiridina luciferin analog (CLA) for O 2 generation and luminol (L) for OCL generation. OCL generation by granulocytges was also monitored by taurine chloramine formation. The chemiluminescence probe based upon CLA was highly specific for and sensitive to O 2 and could be adopted to determine O 2 generation in terms of xanthine oxidase units. The CLA-dependent chemilu minescence by cord blood granulocytes was significantly higher than that by normal adult granulocytes. Taurine chloramine formation was significantly correlated with the L-dependent chemiluminescence (L-CL). Thus, the L-CL is considered to be mainly involved in OCL generated by phagocytizing granulocytes. L-CLs by cord blood granulocytes and normal adult granulocytes were essentially the same during phagocytosis.

Nishida/Kimura/Sugioka/Nakano

146

Materials and Methods Preparation o f Leukocyte. Blood was collected from 17 full-term infants and from 15 healthy adults. The gestational age o f the newborns ranged from 3841 weeks. All were products o f normal pregnancies and vaginal deliveries. Apgar scores were ë 8 at 5 min. The immediate neonatal period was uncompli cated. Neonates with risk factors for infection (rup ture o f membranes more than 24 h before delivery, maternal péripartum fever or amnionitis) and those who had received antimicrobial therapy were ex cluded from the study. Blood was drawn in heparin (20 unit/ml), and leukocytes were isolated by the method described by Rosen [7], Granulocytes were suspended at 1 X 107/ml in Hank’s balanced salt solu tion at pH 7.4 (HBSS) and stored on ice until needed (up to 3 h). Opsonized Zymosan. Zymosan (Sigma Chemicals, St. Louis, Mo., USA) in 20 mM veronal buffer at pH 7.4 (4 mg/ml) was boiled for 100 min and cooled to room temperatue. The zymosan was collected by cen trifugation for 10 min at 1,000 rpm, suspended to 30 mg/ml in pooled serum, incubated at 37 °C for 30 min and collected by centrifugation. The opsonized zymosan (OZ) was suspended in HBSS and washed twice with the same solution. Washed OZ was then suspended to 20 mg/ml in HBSS and was stored at -2 0 ° C until needed. CLA Solution. CLA, which had been synthesized according to the literature [8], was dissolved to 56 pg/ml in doubly distilled water and stored at -2 0 ° C until needed. CLA concentrations were based upon e 410 nm = 8,900 A/“1 cm -1.

L Solution. Luminol (Sigma) was dissolved to 1 mM in mildly alkaline water and was adjusted to pH 7.4 with 0.1 N HCi. This solution was stored at -2 0 °C .

Enzymes. Bovine red cell superoxide dismutase and bovine liver catalase were purchased from Sigma Chemicals. An aliquot o f the commercial catalase, suspended in 0.5 ml o f 0.1% thymol, was dialyzed twice against 3 liters o f 0.01 M Na-phosphate buffer at pH 7.4 before use. Luminescence Measurement. CLA-CL was mea sured with a Luminescence Reader (Aloka Co., To kyo, Japan), as described previously [9]. The standard assay mixture for phagocytizing system contained 4 X 10s granulocytes, 2 mg of OZ, and 1 \iM CLA in HBSS in a total volume o f 2 ml. The experiment with the same mixture from which OZ was omitted was used as control. All components, except for OZ or both OZ and CLA, were preincubated for 3 min, and the reaction was initiated by the addition o f OZ or both OZ and CLA. In some cases, superoxide dismu tase, catalase, or N aN i was added prior to the incuba tion. During the luminescence measurement, the in cubation mixture was agitated by rotation at 37 °C in the Luminescence Reader. L-CL was detected by the same procedure as for CLA-CL, except that 1 \iM L was used instead o f 1 \iM CLA. Taurine Chloramine Assay. OCL production by granulocytes was measured in terms o f taurine chlor amine formation by a slight modification o f the pro cedure o f Weiss et al. [10, II]. Reaction mixture con tained 1 X 106 granulocytes, 2 mg o f OZ, and 15 mM taurine (Sigma) in 2 ml o f Dulbecco’s phosphate-buff ered saline at pH 7.4. OZ was the last component added. The reaction was maintained for 60 min at 37 °C and was stopped by adding o f 50 pg o f catalase. Two milliliters o f the catalase-treated sample was then incubated with 0.12 m M 5-thio-2-nitro-benzoic acid (TNB) for 5 min at room temperature and cen trifuged for 5 min at 500 g. The resulting clear super natant was used for determination of the concentra tions o f 5-5'-dithiobis (2-nitrobenzoic acid) (DTNB), an oxidation product o f TNB by hypochlorite, using s 412 nm = 13,600 M~' c m -1. TNB was prepared by reducing DTNB (Wako Chemicals. Tokyo, Japan) with sodium borohydride [12]. Statistical Analysis. Statistical analysis was per formed using unpaired t test for variance and linear regression.


analog (2-methyl-6-phenyl-3,7-dihydroimidazo [1,2-a] pyrazin-3-one) (CLA) was highly specific for and sensitive to 05 generated by phagocitizing monocytes [5] and granulo cytes [6]. The present work was undertaken to com pare the abilities of granulocytes of cord blood from neonates with those of peripheral blood from adults to generate Oj and OCl~, using CLA and L as luminescence probes, respectively, during phagocytosis.

Assay of O ï and OCT by Chemiluminescence

147

Results

Fig. 1. CLA-CL o f granulocytes during phagocyto sis. The complete incubation mixture contained 1 gAf CLA, 4 X 105 granulocytes (cord blood) and 2 mg of OZ in HBSS (total volume, 2ml). (1) Both granulo cytes and OZ or only OZ were omitted from the com plete incubation mixture; (2) complete incubation mixture; (3) complete mixture + superoxide dismutase (SOD; 0.5 pA/); (4) complete mixture + NaN 1 (0.5 mA/); (5) complete mixture + catalase (20 pg/ml).

quantitatively expressed in terms of xan thine oxidase units by comparing the maxi mum light intensity in the xanthine oxidase system and that in the phagocytizing granu locyte system. With this chemiluminescence probe, granulocytes of peripheral blood from 10 adults and those of cord blood from 12 neo nates were checked for CLA-CL response to OZ. As shown in figure 2, the maximum light intensities were 167,000 ± 29,000 counts/min for granulocytes of cord blood from neonates and 127,000 ± 25,000 counts/min for granulocytes of peripheral blood from adults, which correspond to 576 ± 99 and 439 ± 86 xanthine oxidase units, respectively. Clearly, the ability of neonatal


CLA-CL. Little or no detectable emission was observed in granulocyte- or granulocyteOZ-suspension in the Luminescence Reader under the usual conditions used for lumines cence measurements. When CLA in water was added to HBSS with or without granulo cytes from neonates, luminescence occurred immediately, and its intensity did not change with time during 15 min of observa tion (non specific luminescence) (fig. 1, curve 1). Superoxide dismutase (0.5 pAf) or catalase (20 pg/ml) did not affect the nonspe cific luminescence (data not shown). On the other hand, granulocytes suspended in HBSS containing CLA emitted a strong light fol lowing the addition of OZ. Light intensity (count/min) increased as a function of time, reached maximum, and then decreased ex ponentially (fig. 1, curve 2). Superoxide dis mutase (0.5 \lM) reduced the light intensity to the non specific luminescence level (fig. 1, curve 3), indicating the participation of O 2 in CLA-CL by OZ-stimulated granulocytes. NaN3, which is known to be a scavenger of singlet molecular oxygen OO2) [13] and an inhibitor of hemoproteins [14], at the con centration of 0.5 mM slightly reduced the CLA-dependent luminescence by phagocytizing granulocytes, but catalase at the con centration of 20 pg/ml did not alter it. Under our experimental conditions, max imal light intensity obtained with granulo cytes of cord blood from neonates (corrected for non specific luminescence) was propor tional to the number of cells up to 1 X 106/ 2 ml. These results were essentially the same as those obtained with granulocytes of pe ripheral blood from adults. As reported previously [5, 6], the ability of granulocytes to generate O 2 could be


148

granulocytes is greater than that of normal adult granulocytes with respect to generating O 2 during phagocytosis. The statistical sig nificance for the difference was p < 0.01. L-CL and OCF Formation. When granu locytes of peripheral blood from 10 adults and those of cord blood from 10 neonates were examined for L-CL response to OZ, the maximum light intensities were 33,500 ± 12,500 counts/min for neonatal samples and

Discussion The luminescence measurement using CLA was first adopted for the quantitative determination of both xanthine and xan thine oxidase [15]. In the present study, this


Fig. 2. CLA-CL and L-CL. The incubation sys tems o f the CLA-CL were essentially the same as those in the legend for figure 1. Maximum light inten sity was corrected by subtraction o f the blank value. Xanthine oxidase unit was obtained from maximum light intensity, using the equation described in a pre vious report [6]. L-CL was detected by the same pro cedure as for CLA-CL, except that 1 \iM L was used instead of 1 \iM CLA. Horizontal bars indicate mean value and vertical bars indicate standard deviation.

35,700 ± 8,100 counts/min for adult sam ples (fig. 2). With granulocytes of peripheral blood from 5 adults and those of cord blood from 5 neonates (both 1 X 106 cells/2 ml), 63.6 ± 20.0 nmol of the chloramine/h for the sam ples from adults and 60.8 ± 14.2 nmol of the chloramine/h for the samples from neonates were found to be generated under OZ-stimulated conditions. The assays clearly show that, under OZ-stimulated conditions, OCL production by granulocytes from adults is essentially the same as that by granulocytes from neonates. Others [1] have reported that L-CL by phagocytizing granulocytes was slightly sup pressed by adding the catalytic amount of superoxide dismutase, indicating a partial involvement of Oï in the L-CL. However, the maximum light intensities of L-CL were significantly correlated with taurine chloramine formation (OCL produc tion) during phagocytosis of granulocytes from cord blood of neonates and from pe ripheral blood of adults (correlation coeffi cient 0.800, p < 0.01; fig. 3). CLA-CL and L-CL. The maximum light intensities of L-CL are highly correlated with the maximum light intensities of CLA-CL in phagocytizing granulocytes (fig. 4). The slopes of each regression line are essentially the same. The regression line of adult granu locytes is shifted to right as compared to the cord blood granulocytes.

149

Assay o f O î and OCL by Chemiluminescence

o cord PMN

probe was applied to the assay for O 2 gener ated by phagocytizing granulocytes in terms of xanthine oxidase units. It was found that a non specific weak luminescence which was not influenced by superoxide dismutase or catalase, appeared when CLA in water was mixed with HBSS. However, this weak lumi nescence was found to be distinguished from CLA-CL by phagocytizing granulocytes, with respect to their intensity and time course as well as to the inhibition by super oxide dismutase. It was also observed that catalase had no effect on CLA-CL of phago cytizing granulocytes as was the case for the hypoxanthine-xanthine oxidase system. This indicates that the hydroxyl radical, which could be generated by the metal-catalyzed Haber-Weiss reaction, is not involved in the above luminescence process. Judging from the structure of CLA, '02 would also partic ipate in CLA-CL. However, participation of '02 and the MPO + H 2O 2 reaction in the above CLA-CL appears to be negligible, since there was little or no effect of azide, a scavenger of '02 [13] and an inhibitor of

• a d u lt PMN

Fig. 4. CLA-CL and L-CL. CLA-CL and L-CL were detected by the same procedure as in figure 2.

MPO [14]. Thus, the chemiluminescence probe based upon CLA is considered to be highly specific for O 2 generated by phagocy tizing monocytes [5] and granulocytes [6]. Leukocytes used for the chemilumines cence measurements in the present study


Fig. 3. L-dependent chemilumi nescence and taurine chloramine formation. The reaction mixture of taurine chloramine assay contained 1 X 106 granulocytes, 2 mg o f OZ, and 15 mAf taurine in 2 ml o f Dulbecco’s phosphate-buffered saline. The reaction was stopped by addi tion o f catalase and was incubated with 0.12 mM TNB. L-CL was de tected by the same procedure as in figure 2.

contained 65-90% of granulocytes and 05% of monocytes. Under comparable condi tions, L-CL [16] and CLA-CL [5] in acti vated monocyte suspension have been re ported to be about 50 and 30% of that in activated granulocyte suspension, respec tively. Furthermore, lymphocyte suspen sions did not produce L-CL and CLA-CL when incubated with OZ (data not shown). Thus, L-CL and CLA-CL in our activated leukocyte system are mainly evoked by acti vated granulocytes. The method to measure O 2 production by reduction of cytochrome C has been widely accepted. When O 2 production and lumines cence were both measured, as a function of time following activation of the granulo cytes, a direct correspondance between inte grated light yield and net O 2 production was reported [6]. Thus, the rate of O 2 generation should be proportional to CLA-dependent light intensity during phagocytosis. Results obtained with the CLA method clearly indicate that granulocytes of cord blood are more potent in the generation of O 2 during phagocytosis than granulocytes of peripheral blood from adults, identical to those reported by others who used the cyto chrome C-reduction method for the estima tion of Oj-generated [17, 18]. It has been reported, however, that with luminol as a luminescence, probe cord blood granulocytes are significantly [19] or slightly [20] lower than normal adult granulocytes in their chemiluminescence response during phagocytosis. In contrast to this, our data indicate that L-CL during phagocytosis does not differ in cord blood and adult granulo cytes. The hydrogen peroxide-MPO-chloride system is capable of generating a powerful oxidant OCL [21], which can be trapped by


taurine, yielding taurine chloramine [22, 23]. This chloramine has been known to ox idize TNB to DTNB. Similarly, intact hu man neutrophils stimulated with OZ gener ated a stable species, in the presence of exogeneously added taurine, which could oxid ize TNB to DTNB, reflecting the production of OCL. The present data on quantitation of taurine chloramine formation by phagocytizing granulocytes showed no significant difference between cord and adult samples. De Chatalet et al. [3] reported that L-CL was highly dependent upon a MPO-mediated reaction that probably involves the genera tion of OCL. The results obtained here clearly indicate that taurine chloramine gen eration was highly correlated with L-CL. Thus, the L-CL is considered to be mainly involved in OCL, generated by phagocytizing granulocytes. Cord blood granulocytes required a large quantity of O 2 as compared to the adult granulocytes for obtaining the same intensity of the L-CL (fig. 4). This may be due to the lower MPO activity in cord blood granulo cytes. Cord blood granulocytes may not be rep resentative of circulating peripheral blood granulocytes from newborns. However, based on the fact that Oj-and OCL-generating activities of cord blood granulocytes measured with OZ prepared with adult sera are higher than and essentially the same as those of adult granulocytes, the ability to generate O 2 is probably correlated with the phagocytizing activity against infection. In general, neonates are more prone to infection than adults. This may be due to the rapid depletion of granulocyte activities ob served in stressed neonates and the lower opsonin activity in neonate sera, compared with that in adult sera [24],


150

Assay o f O j and OCl~ by Chemiluminescence

151

1 Allen RC, Roose LD: Phagocytic activation of luminol-dependent chemiluminescence in rabbit alveolar and peritoneal macrophages. Biochem Biophys Res Commun 1976;69:245-252. 2 Willson ME, Trush MA, Van Dyke K, Kyle JM, Mullet MD, Neal WA: Luminol-dependent chem iluminescence analysis o f opsono-phagocytic dys function. J Immunol Methods 1978;23:315-326. 3 De Châtelet LR, Long GD, Shirley PS, Bass DA, Thomas MJ, Henderson FW, Cohen MS: Mecha nism o f luminol-dependent chemiluminescence of human neutrophils. J Immunol 1982; 129:1589— 1593. 4 Welch WD: Correlation between measurements o f the luminol-dependent chemiluminescence re sponse and bacterial susceptibility to phagocyto sis. Infect Immun 1980;30:370-374. 5 Sugioka K, Nakano M, Kurashige S, Akuzawa Y, Goto T: The chemiluminescent probe with a Cypiridina luciferin analog, 2-mcthyl-3,7-dihydroimidazo [ 1,2-a] pyrazin-3-one, specific and sensitive for O 2 production by phagocytizing mac rophages. FEBS Lett 1986;197:27-30. 6 Nakano M, Sugioka K, Ushijima Y: Chemilumi nescence probe with Cypridina luciferin analog, 2-methyl-6-phenyl-3,7-dihydroimidazo [ 1,2-a] pyrazin-3-one, for estimating the ability o f human granulocytes to generate Os. Anal Biochem 1986; 159:363-367. 7 Rosen H, KlebanoffSJ: Chemiluminescence and superoxide production by myeloperoxidase-defi cient leukocytes. J Clin Invest 1976;58:50-60. 8 Inoue S, Sugiura S, Kakoi H, Goto T: Cypridina bioluminescence VI, a new route for the synthesis o f Cypridina luciferin and its analogs. Tetrahed Lett 1976:1609-1610. 9 Ushijima Y, Nakano M, Takyu C, Inaba H: Chemiluminescence in L-tyrosine-HjOj -horsera dish peroxidase system: Possible formation o f ty rosine cation radical. Biochem Biophys Res Com mun 1985;128:963-941. 10 Weiss SJ, Klein R, Slivka A, Wei M: Chlorination o f taurine by human neutrophils. Evidence o f hypochlorous acid generation. J Clin Invest 1982; 70:598-607. 11 Lampert MB, Weiss SJ: The chlorinating potential of the human monocyte. Blood 1983;62:645-651. 12 Aune TM, Thomas EL: Accumulation o f hypothiocyanite ion during peroxidase-catalyzed oxi

13

14

15

16

17

18

19

20

21

22

23

24

dation o f thiocyanate ion. Eur J Biochem 1977;80: 209-214. Hasty N, Merkel PB, Radlick P, Keans DR: Role o f azide with singlet oxygen. Tetrahed Lett 1972; 49-52. Nakagawa A, Nathan CF, Cohn ZA: Hydrogen peroxide metabolism in human monocytes during differentiation in vitro. J Clin Invest 1982;68: 1234-1250. Goto T, Takagi T: Chemiluminescence o f a Cypr idina luciferin analogue, 2-methyl-6-phenyl-3,7dihydroimidazo [ 1,2-a] pyrazin-3-one, in the pres ence of the xanthine-xanthine oxidase system. Bull Chem Soc Jpn 1980;53:833-834. Tono-Oka T, Ueno N, Matsumoto T, Ohkawa M, Matsumoto S: Chemiluminescence o f whole blood. 1. A simple and rapid method for the esti mation o f phagocytic function o f granulocytes and opsonic activity in whole blood. Clin Immunol Immunopathol 1983;26:66-75. Strauss RG, Snyder EL: Activation and activity of the superoxide-generating system o f neutrophils from human infants. Pediatr Res 1983; 17:662-664. Ambruso DR, Bentwood B, Henson PM. John ston RB: Oxidative metabolism of cord blood neutrophils: Relationship to content and degranu lation o f cytoplasmic granules. Pediatr Res 1984; 18:1148-1153. Van Epps DE, Goodwin JS, Murphy S: Agc-dcpcndent variation in polymorphonuclear leukocytes chemiluminescence. Infect Immun 1978;22:57-61. Strauss RG, Seifert MJ: Oxidative metabolism in cord-blood polymorphonuclear leukocytes. Arch Dis Child 1978;53:78-80. Harrison JE, Schultz J: Studies on the chlorinating activity o f myeloperoxidase. J Biol Chem 1976; 251:1371-1374. Zgliczynski JM, Stelmaszyriska T. Domariski J, Ostrowski W: Chloramines as intermediates of oxidation reaction o f amino acid by myeloperoxi dase. Biochim Biophys Acta 1971;235:419-424. Stelmaszyriska T, Zgliczynski JM: Myeloperoxi dase of human neutrophilic granulocytes as chlori nating enzyme. Eur J Biochem 1974;45:305-312. Stossel TP, Alper CA, Rosen FS: Opsonic activity in the newborn: role o f properdin. Pediatrics 1973;52:134-137. Akira Nishida, MD, The Neonatal Unit Children’s Hospital Hachiouji 4-33-13 Daimachi, Hachiouji, Tokyo 193 (Japan)


References

A comparison of the metabolic response to phagocytosis in human granulocytes and monocytes.

Superoxide dismutase activity of human granulocytes and lymphocytes.

Trajectories of human granulocytes.

Granulocytes and airway reactivity.

Editorial: Transfusions of filtered granulocytes.

H2O2 release from human granulocytes during phagocytosis. I. Documentation, quantitation, and some regulating factors.

[Basophilic granulocytes and autoimmune diseases. Can basophilic granulocytes modulate B-cell functions in systemic lupus erythematosus?].

CD64 expression by neutrophil granulocytes.

Endometrial granulocytes in proliferative endometrium.

Application of monoclonal antibody against granulocytes of scallop Chlamys farreri on granulocytes occurrence at different developmental stages and antigenic cross-reactivity of granulocytes in five other bivalve species.

Mepacrine uptake by granulocytes [proceedings].

H2O2 release from human granulocytes during phagocytosis. Relationship to superoxide anion formation and cellular catabolism of H2O2: studies with normal and cytochalasin B-treated cells.

Interferon-gamma-activated human granulocytes kill ingested Mycobacterium fortuitum more efficiently than normal granulocytes.

Viability of granulocytes obtained by filtration leukapheresis.

Kinetics of phagocytosis and intracellular killing of Candida albicans by human granulocytes and monocytes.

[Eosinophilic granulocytes in the cerebrospinal fluid].

Impaired phagocytosis of Staphylococcus aureus by granulocytes and monocytes of AIDS patients.

Cryopreservation of human granulocytes in liquid nitrogen.

Raman microspectroscopic approach to the study of human granulocytes.

In vitro effects of bisphosphonates on chemotaxis, phagocytosis, and oxidative burst of neutrophil granulocytes.

Kinetics of phagocytosis of Staphylococcus aureus and Escherichia coli by human granulocytes.

New evidence for the role of NADH oxidase in phagocytosis by human granulocytes.

Reactive Oxygen Species, Granulocytes, and NETosis.

Staining of eosinophil granulocytes in chemotactic studies.