JOURNAL OF BIOLUMINESCENCE AND CHEMILUMINESCENCE VOL 5 165-170
(1990)
Chemiluminescence Properties of Human, Canine and Rat Polymorphonuclear Cells Joachim Lindena” and Hannelore Burkhardt Institute of Clinical Biochemistry, Medical School Hannover, Konstanty-Gustschow-Str. 8, 3000 Hannover 61, FRG
Human as well as canine and rat polymorphonuclear cells (PMN) were separated from whole blood by centrifugation. Two-step discontinuous Percoll gradients with distinct different densities were used. The chemiluminescence properties of the isolated P M N and of phagocytes in small quantities of whole blood were compared in luminol-enhanced assays after stimulation with various agents: non-opsonized zymosan (3.5g/l), phorbol myristate mol/l) and N-formylacetate (PMA, 2.8 X lop6 mol/l), calcium ionophore A 23187 methionyl-leucyl-phenylalanine (FMLP, 3.5 X mol/l). The isolated cells of the three species responded t o all of the various stimuli. Species-related sensitivity could be ordered: human > canine > rat. Response t o the various agents in the human cells can be ranked: PMA 3 A 23187 > zymosan > FMLP; for the dog: A 23187 > PMA > zymosan > FMLP; and for the rat: zymosan 5 PMA > FMLP 3 A 23187. Time course and peak maximum response were different upon stimulation in the absence and presence of autologous plasma. Distinct soluble stimuli resulted in maximum responses below the baseline in the whole blood assays with canine (FMLP) and rat (FMLP, A 23187) phagocytes. Keywords: Granulocyte; man; dog; rat; chemiluminescence
INTRODUCTION
It has been proposed, that a variety of pathological conditions result from the discharge of reactive oxygen species, lysosomal enzymes and cationic proteins. These are released from polymorphonuclear cells (PMN). Resulting disease syndromes include: adult respiratory distress syndrome (ARDS) in the lung, myocardial injury after infarction as well as glomerular basement
destruction in the kidney and during extracorporeal circulation procedures (Henson and Johnston, 1987; Knudsen et af., 1985). Investigation of pathogenic mechanisms underlying these events is best conducted using various animal models. This study compares luminolenhanced chemiluminescence induced by a particulate and several soluble stimuli of highly purified PMN from dog and rat to those for humans.
Abbreviations: PMN, polymorphonuclear cells; PMA, phorbol myristate acetate; FMLP, N-formyl-methionyl-leucylphenylalanine; ARDS, adult respiratory distress syndrome; PBS, phosphate-buffered saline Dulbecco without Ca” and Mg2+; MEM, Minimal Essential Medium Dulbecco for chemiluminescence with N-2-hydroxyethylpiperazine-N-2-ethanesulphonic acid (Hepes), without phenol red, without glutamine; DMSO, dimethyl sulphoxide; CL, chemiluminescence. *Author for correspondence. 0884-399619010301hS-06$05 .OO 01990 by John Wiley & Sons. Ltd
Received Sepremher I988
J. LINDENA AND H. BURKHARDT
166
MATERIAL AND METHODS
Haemoglobin
Haemoglobin was determined in whole blood as cyanmethaemoglobin (van Kampen and Zijlstra, 1961). Sigma Chemicals Co., St. Louis, USA: Zymosan A; N-formyl-l-L-methionyl-L-leucyl-L-phenylalanine (FMLP); ionophore A 23187; isolation of polymorphonuclearcells phorbol myristate acetate (PMA). - Boehringer Mannheim,' FRG: luminol (5- 9 vol. Percoll were mixed with 1 vol. NaCl amino1 - 2.3 - dihydro - 1.4 - phthalazinedione); 1.5 mol/l. The stock isoosmotic Percoll was phophate-buffered saline Dulbecco without Ca2+ further diluted to the desired concentration with and Mg2+ (PBS); Minimal Essential Medium NaCl 0.15 mol/l. Dulbecco for chemiluminescence with N-2Densities of the respective Percoll solutions for hydroxyethylpiperazine-N-2-ethane sulphonic the different species, blood volume added to the acid (Hepes), without phenol red, without gluta- gradient, and the gravity forces applied to the mine (MEM). gradient were as follows: E. Merck, Darmstadt, FRG: sodium chloride; Turk's solution; May Grunwald stain. (1) Man: p = 1.0945kg/l, p = 1.0779kg/l, 4ml Pharmacia Fine Chemicals, Sweden: Percoll blood, 350g for 15min. (Polyvinylpyrolidone-coated silica gel) for density (2) Dog: p = 1.0945kg/l, p = 1.0756kg/l, 3ml gradient centrifugatin. blood mixed with 1ml NaCl 0.15 mol/l, 350g Fresenius, Bad Homburg, FRG: sodium citrate for 25 min. solution (31.1 g/l). (3) Rat: p = 1.1005kg/l, p = 1.0815kg/l, 2ml Hoffmann-La Roche, Grenzacli-Whylen, blood mixed with 2 ml NaCl 0.15 mol/l, 350 g FRG: Heparin (Liquemin 25 000). for 20 min. Chemicals, media and reagents
Blood
Human venous blood, anticoagulated with sodium citrate (9 vol. blood 1 vol. sodium citrate; final concentration: 3.1 g/l) was obtained from healthy blood donors from the local blood bank. Canine and rat blood, anticoagulated with heparin (9 vol. blood 1 vol. heparin; final heparin concentration 50 U/ml), was obtained from beagle dogs by cephalic vein puncture and from Lewis/Ztm rats by catheter technique (Lindena et af., 1984, 1986). Blood was taken from all species between 08.00 and 09.00 to minimize daytime dependence on respiratory burst activity of phagocytic blood cells (Heberer et al., 1982).
+
+
Plasma
An aliquot of blood was removed and centrifuged at 12,OOOg for 2min at room temperature (Eppendorf centrifuge 3200, Netheler and Hinz, Hamburg).
A two-step discontinuous gradient was established using 4ml each of the aforementioned density concentrations. Blood was layered on the gradient and centrifuged (20 "C, Minifuge GL with a swing-out rotor, Heraeus Christ). Thrombocytes and mononuclear cells banded on the surface of the upper layer; PMN banded at the interface of the two density concentrations and erythrocytes banded at the bottom. PMN were separated. Independent of visible contamination, erythrocytes were removed from this fraction with hypotonic lysis. Erythrocytes, even in minimal concentrations, act as quencher and active scavenger of reactive oxygen species (Allen et al., 1985; Lindena et al., 1987). PMN were washed twice in PBS. Their concentration was adjusted to 2.5 X lo6 celldm1 in MEM according to staining with Turk's solution. Differential counts were performed with May-Grunwald Giemsa stained cytocentrifuged preparations (isolated PMN) or smears (whole blood). Chemiluminescence
Stimuli. FMLP and non-opsonized zymosan, washed twice with PBS, were suspended in MEM
HUMAN, CANINE AND RAT PMN CHEMILUMINESCENCE
to mol/l and lOOg/l, respectively, and stored frozen at -70°C in small aliquots. FMLP and zymosan final assay concentrations were 3.5 x 10-6mol/l and 3.Sg/l, respectively. PMA and calcium ionophore A 23187 were sus ended in DMSO to stock solutions of 1.6 x 10- mol/l and lop2 mol/l, respectively, and stored frozen at -70°C in small aliquots. PMA and A 23187 final assa concentrations were 2.8 X low6mol/l and 10- Y molll, respectively. Final DMSO assay concentrations below 0.06% (vlv) did not affect the chemiluminescence reactions.
167
Resulting recoveries were 80.1%, 66.3% and 69% for man, dog and rat, respectively. Viability, checked with the trypan blue exclusion method, was greater than 98%.
a
Assay. Chemiluminescence (CL) assays of whole blood phagocytes and isolated PMN were performed in polystyrene chemiluminescence vials at 37°C using a Biolumat 9505 equipped with a six-channel device (Berthold, Wildbad, FRG). The vials were not stirred during the assay. CL reaction mixtures for isolated PMN and whole blood were composed as follows (Lindena et al., 1987): Isolated PMN: 500 pl MEM and 10 pl luminol (22.6mmoUl) as well as either 20pl autologous plasma (CL 1) or no plasma (CL 2) plus 20p1 PMN suspension (0.5 x lo5 cells). After 5min incubation, the reaction was started with 20 p1 of the respective stimulus. Controls without stimulus were run. Whole blood: 500p1 MEM and lop1 luminol (22.6mmol/l) plus 40p1 whole blood (CL 1). After 5min of incubation, the reaction was started with 20pl of the respective stimulus. A control without stimulus was run.
CL was continuously recorded. Two CL parameters were calculated from the measurements: peak maximum countslmin (cpmlmin x 0.5 x lo5 phagocytes). The bimodal response pattern of the FMLP stimulation (Dahlgren, 1987) is quantitatively better reflected as integral counts with respect to time (integral counts/h X 0.5 X lo5 phagocytes); peak time values (time in min after starting the Biolumat required to reach the peak maximum). RESULTS Separation of PMN
PMN constituted 9&99% of the recovered cells in the PMN layer; less than 1 % were monocytes.
Chemiluminescence
Peak CL responses and peak times for whole blood and isolated PMN are shown in Table 1. Note that for FMLP, the results are given as integral counts. In nearly every case, the magnitude of CL response was greatest in the human PMN. This was followed by the canine and rat. In human and canine cells the largest peak maximum values with simultaneously shortest peak time were obtained after stimulation with calcium ionophore A23187. FMLP was the weakest stimulus for all species. Response above baseline could not be detected in the canine and rat whole blood assays. With isolated PMN, a weak response was detectable. In contrast to both other soluble stimuli, zymosan and A 23187 responses with isolated PMN were generally higher in CL1 (addition of autologous plasma) than in CL2 (without autologous plasma). This does not hold true for neither rat whole blood, where A 23187 stimulation is too weak to be detected, nor for isolated rat PMN, where the extremely weak response was even lower in the presence of plasma (CL 1). DISCUSSION
The most common stimulants of oxidative metabolism of phagocytes are the particulate stimulus zymosan and the soluble stimuli PMA, FMLP and calcium ionophore A 23187. Their modes of cellular action leading to the emission of light and chemiluminescence assessment are distinctly different. Zymosan and FMLP act via complement receptor type 3 and specific FMLP receptor, respectively.These generate several intracellular messengers (Cheung et al., 1983; Lindena et al, 1987; Rossi et al., 1985; Wilson et al., 1987). PMA is known to bind directly to and mediate a redistribution of protein kinase C. A similar role has been discussed for A 23187 (Onozaki et al., 1983; Wilson et al., 1987; Wolfson et al. ,1985). Highly sensitive chemiluminogenic probing techniques allows measurement of phagocyte oxygenation activity using diluted native whole
168
J. LINDENA AND H. BURKHARDT
Table 1. Chemiluminescence in whole blood and isolated granulocytes of man and animals Stimuli Zymosan 3.5 g/l Whole blood CL" CL1 Man Dog Rat Timeg Man Dog Rat Granulocyte Clb CLI" Man Dog Rat Time Man Dog Rat CL2d Man Dog Rat Time Man Dog Rat
211 151 88 24.7 29.0 19.2
f rt
f rt
f f
72.9" 27.5 34.5 2.9 2.5 2.5
6412 f 1511 2479 f 647 1338 k 479 29.4 rt 2.5 5.1 37.7 f 25.0 rt 2.4 4920 f 865 1861 t 624 678 f 281 32.0 rt 3.4 50.7 f 8.1 81.0 f 16.4
PMA 2.8 x 10 6mol/l
668 56.7 51.0 17.4 32.1 19.3
t 229 rt
f rt
f f
10.5 21.8 2.6 3.0 3.5
22946 f 9395 1520 f 246 996 f 312 12.7 f 1.7 4.4 31.4 f 13.9 t 1.8 24982 f 8932 2908 f 696 1142 f 262 6.0 f 0.68 1.6 f 0.46 12.4 f 1.3
A 23187 10 5mol/l
784 2005
f 5
0 5.2 f 4.8 f 0
FMLP 3.5 x 10 6mol/l
104 611 0.80 0.63
1342 f
438'
0 0
-
18587 f 4259 18764 f 5053 54931 rt 13430 8450 2 2405 87.1 f 25.3 3261 ? 1067 5.7 rt 0.99 3.2 f 0.50 1.5 7.5 rt 25872 ? 11527 10786 ? 2364 41575 f 8500 10766 f 2609 352 f 115 4 8 6 8 f 1223 2.5 f 0.79 1.5 f 0.24 1.1 i 0.20 -
= 4 0 ~anticoagulated l whole blood at luminol concentration of 22.6mmol/l as mean f standard deviation with n = 30 (zymosan:man) and n = 20 (all other stimuli and species). b20pI PMN suspension (0.5 x l o 5 cells) at luminol concentration of 22.6mmol/l as mean f standard deviation with n = 30 (zymosan:man), n = (zymosan: dog, rat) and n = 10 (all other stimuli and species). Reaction mixture with 20 pl autologous plasma. dReaction mixture without autologous plasma. Response recorded for zymosan, PMA and A 23187 as peak maximum counts x 103/min x 0.5 x lo5'phagocytes. 'Response recorded for FMLP as integral counts x 103/h x 0.5 x l o 5 phagocytes. gTime, in minutes, to reach the peak maximum.
blood. The peak maximum of whole blood phagocytes is approximately 5% of that for an equivalent number of isolated PMN (Table I). This is a physical consequence of photon absorption by haemoglobin and scavenging of reactive oxygen species by erythrocytes (Allen et ul., 1985; Lindena et al., 1987). FMLP, resembling naturally occurring bacterial products, is the weakest soluble stimulus for oxygen metabolite generation (Test and Weiss, 1984). It did not produce a response in canine and rat whole blood. Observed species-related differences are not due to variations in haemoglobin concentrations. These are nearly identical: 138g/l, 149g/l and 131g/l for man ( n = 30), dog ( n = 20) and rat ( n = 20),
respectively. Nor is the erythrocyte number responsible for these differences: 5.0 x 10'2/1,6.4 x 10l2/1and 6.8 x 10'2/1, respectively. Sodium citrate (final concentration: 3.1 g/l) and heparin (final concentrations: 5 Uiml) are the most usual in vitro anticoagulants in clinical practice. However, prevention of blood clotting in laboratory animals, including dog and rat, require\ up to ten times higher anticoagulant concentrations (Friedel and Mattenheimer, 1970). A heparin concentration of SO Uiml in blood of laboratory animals has no cell-toxic effect as judged from plasma enzyme activity determinations (Lindena et ul., 1984. 1986) and from enzyme release experiments (Friedel and Mattenheimer,
169
HUMAN, CANINE AND RAT PMN CHEMILUMINESCENCE
1970). Observed species-related differences are not due to different anticoagulants used. Human heparinized whole blood (5 U/ml) compared to citrated (3.1 g/l) whole blood (set as 100%) responds to zymosan, PMA, A 23187 and FMLP in the whole blood CL (n = 3, mean k SD) with 98.9 ? 4.0%, 94.6 f 5.2%, 108 k 6.9% and 102 k 8,5%, respectively, and in the granulocyte CL (n = 3, mean k SD) for CLl with 105 f 3.2%, 95.4 k 3.696, 103 k 4.6% and 99.5 k 4.8%, respectively, and for CL2 with 95.8 k 4.5%, 99.2 k 2.1%, 101 k 7.2% and 104 k 6.0%, respectively. Zymosan particles, in the presence of plasma, are coated with the complement fragment iC3b. This is recognized by complement receptor type 3. In the absence of plasma, this receptor mediates the interaction of the zymosan polysaccharide component glucan (Lindena et al., 1987). The higher photon yield for zymosan in the presence of plasma (CL1) during stimulation of isolated PMN can, therefore, be explained by the additional coating of complement fragment iC3b. The soluble stimuli PMA, a croton oil derivate, and FMLP, an analogue of naturally occurring bacterial products, gave a smaller response in the presence of plasma. This is a result of their mode of action, which is independent of plasma (Rossi et al., 1986; Wilson et al., 1987; Wolfson et al., 1985). In this case the scavenging properties of plasma come to the fore, mediated for instance by ceruloplasmin (Goldstein et al., 1979) and albumin (Holt et al., 1984). The response pattern induced by calcium ionophore was unexpected-underlying the proposed plasma-independent mechanism of activation (Onozaki et al., 1983)-with significantly higher peak maximum values in the presence of plasma (CL1). This is indicative of plasma factor requirement for amplified response, at least in man and dog. This phenomenon, however, was not observed in rat, where A 23187 is the weakest stimulus tested. In contrast to man and most animal species previously tested (Hornebeck et al., 1987; Roberts et al., 1987; Suzuki et al., 1985), canine granulocytes are unresponsive to FMLP as a chemotaxin, aggregant or stimulator of chemiluminescence (Redl et al., 1983). The weak response in isolated canine PMN remains unclear. PMN of various species respond differently to various stimuli. Only selective and substantiated conclusions could be drawn from comparative observations made on animal and human PMN.
Acknowledgement This work was supported by the Deutsche Forschungsgemeinschaft.
REFERENCES Allen, R. C., Mead, M. E. and Kelly, J. L. (1985). Phagocyte oxygenation activity measured by chemiluminesccncc and chemiluminescence probing. in CRC Handbook of Methods for Oxygen Radical Research, Greenwald, R. A. (Ed.), CRC Press, Boca Raton, pp. 343-351. Cheung, K., Archibald, A . C. and Robinson, M. F. (1983). The origin of chemiluminescence produced by neutrophils stimulated by opsonized zymosan, J . Immunol., 130, 2324-2329. Dahlgren, C. (1987). Polymorphonuclear leukocyte chemiluminescence induced by formylmethionyl-leucylphenylalanine and phorbol myristate acetate: Effects of catalase and superoxide dismutase. Agents Actions, 21, 104-112. Friedel, R. and Mattenheimer, H. (1970). Release of metabolic enzymes from platelets during blood clotting of man, dog, rabbit and rat. Clin. Chim. Acta, 30, 37-46. Goldstein, J . M., Kaplan, H . B., Edelson, H . S. and Weissman, G. (1979). Ceruloplasmin a scavenger of superoxide anion radicals. J. Biol. Chem., 254, 4040-4045 Heberer, M., Emst, M., Durig, M., Allgower, M. and Fischer, H. (1982). Measurement of chemiluminescence in freshly drawn human blood.Klin. Wschr., 60, 1443-1448. Henson, P. M. and Johnston, Jr. B. B. (1987). Tissue injury in inflammation. Oxidants, proteinases and cationic proteins. J . Clin. Invest., 79, 669-674. Holt, M. E., Ryall, M. E. T. and Campbell, A. K. (1984). Albumin inhibits human polymorphonuclear leucocyte luminol-dependent chemiluminescence: evidence for oxygen radical scavenging. Br. J Exp. Pathol, 65, 231-241. Hornebeck, W., Soleihac, J. M., Tixier, J. M., Mocar, E. and Robert, L. (1987). Inhibition by elastase inhibitors of the Formyl Met Leu Phe-induced chemotaxis of rat polymorphonuclear leukocytes. Cell. Biochem. Funci., 5, 113-122. van Kampen, E. J. and Zijlstra, W. G. (1961). Standardization of hemoglobinometry. 11. The hemiglobincyanide method. Clin. Chim. Acta, 6, 538-544. Knudsen, F., Nielsen, A. H., Pedersen, J. 0. and Jersild, C. (1985). On the kinetics of complement activation, leucopenia and granuloyte-elastase release induced by haemodialysis. Scand. J . Clin. Lab. Invest., 45, 759--766. Lindena, J . Biittner, D. and Trautschold, I. (1984). Biological, analytical and experimental components of variance in a long-term study of plasma constituents in rat. J . Clin. Chem. Clin. Biochem., 22, 97-104. Lindena, J . and Trautschold, I. (1986). Catalytic enzyme activity concentration in plasma of man, sheep, dog, cat, rabbit, guinea pig, rat and mouse. Approach to a quantitative diagnostic enzymology, I. communication. J . Clin. Chem. Clin. Biochem., 24, 11-18. Lindena, J., Burkhardt, H . and Dwenger, A. (1987). Mechanisms of non-opsonized zymosan-induced and luminol-enhanced chemiluminescence in whole blood and isolated phagocytes. J . Clin. Chem. Clin. Biochem., 25, 765-778.
170 Onozaki, K . , Takenawa, T., Homma, Y. and Hashimato, T. (1983). The mechanism of macrophage activation induced by Ca” ionophore. Cell. Immunol., 75, 242-254. Redl, H., Flynn, P. J., Lamche, H., Schiesser, A . , Schlag, G. and Hammerschmidt, D.E. (1983). Aggregation, chernotaxis and chemiluminesccnce of canine granulocytes. Studies utilizing improved cell preparation techniques. Inflammation, I , 61-80. Roberts, R. L., Hatori, N., Drury, J . K . and Stiehm, E. R. (1987). Purification and properties of porcine polymorphonuclear cells. J . Immunol. Methods, 103, 21-32. Rossi, F., Della Bianca, V . . Grzeskowiak, M. and Zeni, L. (1986). Mechanism of oxygen free radicals production in granulocytes. In Oxygen Free Radicals in Shock, Novelli, G . P. and Ursini, F. (Eds), Karger, Basel, pp. 15-28. Suzuki, K., Asaoka, K . , Takahashi, K . and Fujikura, T.
J. LINDENA AND H. BURKHARDT (1985). Differences among primates in defence against infection: sensitivity of polymorphonuclear leukocytes to f Met-Leu-Phe. Cell. Biochem. Funct., 3, 297-303. Test, S. T. and Weiss, S. J . (1984). Quantitative and temporal characterization of the extracellular H,O, pool generated by human neutrophils. J . Biol.Chem., 259, 399-405. Wilson, E., Rice, W. G . , Kinkade, Jr., J . M., Merrill, J r . , A. H . Arnold, R. R . and Lamheth, J . D . (1987). Protein kinase C inhibition by sphingoid long-chain bases: effects on secretion in human neutrophils. Arch. Biochem. Biophys., 259, 204-214. Wolfson, M., McPhail, L.C., Nasrallah, V. N. and Snyderman, R. (1985). Phobol myristate acetate mediates redistribution of protein kinase C i n human neutrophils: potential role in t h e activation of the respiratory burst enzyme. J . Immunol., 135, 2051-2062.
(1990)
Chemiluminescence Properties of Human, Canine and Rat Polymorphonuclear Cells Joachim Lindena” and Hannelore Burkhardt Institute of Clinical Biochemistry, Medical School Hannover, Konstanty-Gustschow-Str. 8, 3000 Hannover 61, FRG
Human as well as canine and rat polymorphonuclear cells (PMN) were separated from whole blood by centrifugation. Two-step discontinuous Percoll gradients with distinct different densities were used. The chemiluminescence properties of the isolated P M N and of phagocytes in small quantities of whole blood were compared in luminol-enhanced assays after stimulation with various agents: non-opsonized zymosan (3.5g/l), phorbol myristate mol/l) and N-formylacetate (PMA, 2.8 X lop6 mol/l), calcium ionophore A 23187 methionyl-leucyl-phenylalanine (FMLP, 3.5 X mol/l). The isolated cells of the three species responded t o all of the various stimuli. Species-related sensitivity could be ordered: human > canine > rat. Response t o the various agents in the human cells can be ranked: PMA 3 A 23187 > zymosan > FMLP; for the dog: A 23187 > PMA > zymosan > FMLP; and for the rat: zymosan 5 PMA > FMLP 3 A 23187. Time course and peak maximum response were different upon stimulation in the absence and presence of autologous plasma. Distinct soluble stimuli resulted in maximum responses below the baseline in the whole blood assays with canine (FMLP) and rat (FMLP, A 23187) phagocytes. Keywords: Granulocyte; man; dog; rat; chemiluminescence
INTRODUCTION
It has been proposed, that a variety of pathological conditions result from the discharge of reactive oxygen species, lysosomal enzymes and cationic proteins. These are released from polymorphonuclear cells (PMN). Resulting disease syndromes include: adult respiratory distress syndrome (ARDS) in the lung, myocardial injury after infarction as well as glomerular basement
destruction in the kidney and during extracorporeal circulation procedures (Henson and Johnston, 1987; Knudsen et af., 1985). Investigation of pathogenic mechanisms underlying these events is best conducted using various animal models. This study compares luminolenhanced chemiluminescence induced by a particulate and several soluble stimuli of highly purified PMN from dog and rat to those for humans.
Abbreviations: PMN, polymorphonuclear cells; PMA, phorbol myristate acetate; FMLP, N-formyl-methionyl-leucylphenylalanine; ARDS, adult respiratory distress syndrome; PBS, phosphate-buffered saline Dulbecco without Ca” and Mg2+; MEM, Minimal Essential Medium Dulbecco for chemiluminescence with N-2-hydroxyethylpiperazine-N-2-ethanesulphonic acid (Hepes), without phenol red, without glutamine; DMSO, dimethyl sulphoxide; CL, chemiluminescence. *Author for correspondence. 0884-399619010301hS-06$05 .OO 01990 by John Wiley & Sons. Ltd
Received Sepremher I988
J. LINDENA AND H. BURKHARDT
166
MATERIAL AND METHODS
Haemoglobin
Haemoglobin was determined in whole blood as cyanmethaemoglobin (van Kampen and Zijlstra, 1961). Sigma Chemicals Co., St. Louis, USA: Zymosan A; N-formyl-l-L-methionyl-L-leucyl-L-phenylalanine (FMLP); ionophore A 23187; isolation of polymorphonuclearcells phorbol myristate acetate (PMA). - Boehringer Mannheim,' FRG: luminol (5- 9 vol. Percoll were mixed with 1 vol. NaCl amino1 - 2.3 - dihydro - 1.4 - phthalazinedione); 1.5 mol/l. The stock isoosmotic Percoll was phophate-buffered saline Dulbecco without Ca2+ further diluted to the desired concentration with and Mg2+ (PBS); Minimal Essential Medium NaCl 0.15 mol/l. Dulbecco for chemiluminescence with N-2Densities of the respective Percoll solutions for hydroxyethylpiperazine-N-2-ethane sulphonic the different species, blood volume added to the acid (Hepes), without phenol red, without gluta- gradient, and the gravity forces applied to the mine (MEM). gradient were as follows: E. Merck, Darmstadt, FRG: sodium chloride; Turk's solution; May Grunwald stain. (1) Man: p = 1.0945kg/l, p = 1.0779kg/l, 4ml Pharmacia Fine Chemicals, Sweden: Percoll blood, 350g for 15min. (Polyvinylpyrolidone-coated silica gel) for density (2) Dog: p = 1.0945kg/l, p = 1.0756kg/l, 3ml gradient centrifugatin. blood mixed with 1ml NaCl 0.15 mol/l, 350g Fresenius, Bad Homburg, FRG: sodium citrate for 25 min. solution (31.1 g/l). (3) Rat: p = 1.1005kg/l, p = 1.0815kg/l, 2ml Hoffmann-La Roche, Grenzacli-Whylen, blood mixed with 2 ml NaCl 0.15 mol/l, 350 g FRG: Heparin (Liquemin 25 000). for 20 min. Chemicals, media and reagents
Blood
Human venous blood, anticoagulated with sodium citrate (9 vol. blood 1 vol. sodium citrate; final concentration: 3.1 g/l) was obtained from healthy blood donors from the local blood bank. Canine and rat blood, anticoagulated with heparin (9 vol. blood 1 vol. heparin; final heparin concentration 50 U/ml), was obtained from beagle dogs by cephalic vein puncture and from Lewis/Ztm rats by catheter technique (Lindena et af., 1984, 1986). Blood was taken from all species between 08.00 and 09.00 to minimize daytime dependence on respiratory burst activity of phagocytic blood cells (Heberer et al., 1982).
+
+
Plasma
An aliquot of blood was removed and centrifuged at 12,OOOg for 2min at room temperature (Eppendorf centrifuge 3200, Netheler and Hinz, Hamburg).
A two-step discontinuous gradient was established using 4ml each of the aforementioned density concentrations. Blood was layered on the gradient and centrifuged (20 "C, Minifuge GL with a swing-out rotor, Heraeus Christ). Thrombocytes and mononuclear cells banded on the surface of the upper layer; PMN banded at the interface of the two density concentrations and erythrocytes banded at the bottom. PMN were separated. Independent of visible contamination, erythrocytes were removed from this fraction with hypotonic lysis. Erythrocytes, even in minimal concentrations, act as quencher and active scavenger of reactive oxygen species (Allen et al., 1985; Lindena et al., 1987). PMN were washed twice in PBS. Their concentration was adjusted to 2.5 X lo6 celldm1 in MEM according to staining with Turk's solution. Differential counts were performed with May-Grunwald Giemsa stained cytocentrifuged preparations (isolated PMN) or smears (whole blood). Chemiluminescence
Stimuli. FMLP and non-opsonized zymosan, washed twice with PBS, were suspended in MEM
HUMAN, CANINE AND RAT PMN CHEMILUMINESCENCE
to mol/l and lOOg/l, respectively, and stored frozen at -70°C in small aliquots. FMLP and zymosan final assay concentrations were 3.5 x 10-6mol/l and 3.Sg/l, respectively. PMA and calcium ionophore A 23187 were sus ended in DMSO to stock solutions of 1.6 x 10- mol/l and lop2 mol/l, respectively, and stored frozen at -70°C in small aliquots. PMA and A 23187 final assa concentrations were 2.8 X low6mol/l and 10- Y molll, respectively. Final DMSO assay concentrations below 0.06% (vlv) did not affect the chemiluminescence reactions.
167
Resulting recoveries were 80.1%, 66.3% and 69% for man, dog and rat, respectively. Viability, checked with the trypan blue exclusion method, was greater than 98%.
a
Assay. Chemiluminescence (CL) assays of whole blood phagocytes and isolated PMN were performed in polystyrene chemiluminescence vials at 37°C using a Biolumat 9505 equipped with a six-channel device (Berthold, Wildbad, FRG). The vials were not stirred during the assay. CL reaction mixtures for isolated PMN and whole blood were composed as follows (Lindena et al., 1987): Isolated PMN: 500 pl MEM and 10 pl luminol (22.6mmoUl) as well as either 20pl autologous plasma (CL 1) or no plasma (CL 2) plus 20p1 PMN suspension (0.5 x lo5 cells). After 5min incubation, the reaction was started with 20 p1 of the respective stimulus. Controls without stimulus were run. Whole blood: 500p1 MEM and lop1 luminol (22.6mmol/l) plus 40p1 whole blood (CL 1). After 5min of incubation, the reaction was started with 20pl of the respective stimulus. A control without stimulus was run.
CL was continuously recorded. Two CL parameters were calculated from the measurements: peak maximum countslmin (cpmlmin x 0.5 x lo5 phagocytes). The bimodal response pattern of the FMLP stimulation (Dahlgren, 1987) is quantitatively better reflected as integral counts with respect to time (integral counts/h X 0.5 X lo5 phagocytes); peak time values (time in min after starting the Biolumat required to reach the peak maximum). RESULTS Separation of PMN
PMN constituted 9&99% of the recovered cells in the PMN layer; less than 1 % were monocytes.
Chemiluminescence
Peak CL responses and peak times for whole blood and isolated PMN are shown in Table 1. Note that for FMLP, the results are given as integral counts. In nearly every case, the magnitude of CL response was greatest in the human PMN. This was followed by the canine and rat. In human and canine cells the largest peak maximum values with simultaneously shortest peak time were obtained after stimulation with calcium ionophore A23187. FMLP was the weakest stimulus for all species. Response above baseline could not be detected in the canine and rat whole blood assays. With isolated PMN, a weak response was detectable. In contrast to both other soluble stimuli, zymosan and A 23187 responses with isolated PMN were generally higher in CL1 (addition of autologous plasma) than in CL2 (without autologous plasma). This does not hold true for neither rat whole blood, where A 23187 stimulation is too weak to be detected, nor for isolated rat PMN, where the extremely weak response was even lower in the presence of plasma (CL 1). DISCUSSION
The most common stimulants of oxidative metabolism of phagocytes are the particulate stimulus zymosan and the soluble stimuli PMA, FMLP and calcium ionophore A 23187. Their modes of cellular action leading to the emission of light and chemiluminescence assessment are distinctly different. Zymosan and FMLP act via complement receptor type 3 and specific FMLP receptor, respectively.These generate several intracellular messengers (Cheung et al., 1983; Lindena et al, 1987; Rossi et al., 1985; Wilson et al., 1987). PMA is known to bind directly to and mediate a redistribution of protein kinase C. A similar role has been discussed for A 23187 (Onozaki et al., 1983; Wilson et al., 1987; Wolfson et al. ,1985). Highly sensitive chemiluminogenic probing techniques allows measurement of phagocyte oxygenation activity using diluted native whole
168
J. LINDENA AND H. BURKHARDT
Table 1. Chemiluminescence in whole blood and isolated granulocytes of man and animals Stimuli Zymosan 3.5 g/l Whole blood CL" CL1 Man Dog Rat Timeg Man Dog Rat Granulocyte Clb CLI" Man Dog Rat Time Man Dog Rat CL2d Man Dog Rat Time Man Dog Rat
211 151 88 24.7 29.0 19.2
f rt
f rt
f f
72.9" 27.5 34.5 2.9 2.5 2.5
6412 f 1511 2479 f 647 1338 k 479 29.4 rt 2.5 5.1 37.7 f 25.0 rt 2.4 4920 f 865 1861 t 624 678 f 281 32.0 rt 3.4 50.7 f 8.1 81.0 f 16.4
PMA 2.8 x 10 6mol/l
668 56.7 51.0 17.4 32.1 19.3
t 229 rt
f rt
f f
10.5 21.8 2.6 3.0 3.5
22946 f 9395 1520 f 246 996 f 312 12.7 f 1.7 4.4 31.4 f 13.9 t 1.8 24982 f 8932 2908 f 696 1142 f 262 6.0 f 0.68 1.6 f 0.46 12.4 f 1.3
A 23187 10 5mol/l
784 2005
f 5
0 5.2 f 4.8 f 0
FMLP 3.5 x 10 6mol/l
104 611 0.80 0.63
1342 f
438'
0 0
-
18587 f 4259 18764 f 5053 54931 rt 13430 8450 2 2405 87.1 f 25.3 3261 ? 1067 5.7 rt 0.99 3.2 f 0.50 1.5 7.5 rt 25872 ? 11527 10786 ? 2364 41575 f 8500 10766 f 2609 352 f 115 4 8 6 8 f 1223 2.5 f 0.79 1.5 f 0.24 1.1 i 0.20 -
= 4 0 ~anticoagulated l whole blood at luminol concentration of 22.6mmol/l as mean f standard deviation with n = 30 (zymosan:man) and n = 20 (all other stimuli and species). b20pI PMN suspension (0.5 x l o 5 cells) at luminol concentration of 22.6mmol/l as mean f standard deviation with n = 30 (zymosan:man), n = (zymosan: dog, rat) and n = 10 (all other stimuli and species). Reaction mixture with 20 pl autologous plasma. dReaction mixture without autologous plasma. Response recorded for zymosan, PMA and A 23187 as peak maximum counts x 103/min x 0.5 x lo5'phagocytes. 'Response recorded for FMLP as integral counts x 103/h x 0.5 x l o 5 phagocytes. gTime, in minutes, to reach the peak maximum.
blood. The peak maximum of whole blood phagocytes is approximately 5% of that for an equivalent number of isolated PMN (Table I). This is a physical consequence of photon absorption by haemoglobin and scavenging of reactive oxygen species by erythrocytes (Allen et ul., 1985; Lindena et al., 1987). FMLP, resembling naturally occurring bacterial products, is the weakest soluble stimulus for oxygen metabolite generation (Test and Weiss, 1984). It did not produce a response in canine and rat whole blood. Observed species-related differences are not due to variations in haemoglobin concentrations. These are nearly identical: 138g/l, 149g/l and 131g/l for man ( n = 30), dog ( n = 20) and rat ( n = 20),
respectively. Nor is the erythrocyte number responsible for these differences: 5.0 x 10'2/1,6.4 x 10l2/1and 6.8 x 10'2/1, respectively. Sodium citrate (final concentration: 3.1 g/l) and heparin (final concentrations: 5 Uiml) are the most usual in vitro anticoagulants in clinical practice. However, prevention of blood clotting in laboratory animals, including dog and rat, require\ up to ten times higher anticoagulant concentrations (Friedel and Mattenheimer, 1970). A heparin concentration of SO Uiml in blood of laboratory animals has no cell-toxic effect as judged from plasma enzyme activity determinations (Lindena et ul., 1984. 1986) and from enzyme release experiments (Friedel and Mattenheimer,
169
HUMAN, CANINE AND RAT PMN CHEMILUMINESCENCE
1970). Observed species-related differences are not due to different anticoagulants used. Human heparinized whole blood (5 U/ml) compared to citrated (3.1 g/l) whole blood (set as 100%) responds to zymosan, PMA, A 23187 and FMLP in the whole blood CL (n = 3, mean k SD) with 98.9 ? 4.0%, 94.6 f 5.2%, 108 k 6.9% and 102 k 8,5%, respectively, and in the granulocyte CL (n = 3, mean k SD) for CLl with 105 f 3.2%, 95.4 k 3.696, 103 k 4.6% and 99.5 k 4.8%, respectively, and for CL2 with 95.8 k 4.5%, 99.2 k 2.1%, 101 k 7.2% and 104 k 6.0%, respectively. Zymosan particles, in the presence of plasma, are coated with the complement fragment iC3b. This is recognized by complement receptor type 3. In the absence of plasma, this receptor mediates the interaction of the zymosan polysaccharide component glucan (Lindena et al., 1987). The higher photon yield for zymosan in the presence of plasma (CL1) during stimulation of isolated PMN can, therefore, be explained by the additional coating of complement fragment iC3b. The soluble stimuli PMA, a croton oil derivate, and FMLP, an analogue of naturally occurring bacterial products, gave a smaller response in the presence of plasma. This is a result of their mode of action, which is independent of plasma (Rossi et al., 1986; Wilson et al., 1987; Wolfson et al., 1985). In this case the scavenging properties of plasma come to the fore, mediated for instance by ceruloplasmin (Goldstein et al., 1979) and albumin (Holt et al., 1984). The response pattern induced by calcium ionophore was unexpected-underlying the proposed plasma-independent mechanism of activation (Onozaki et al., 1983)-with significantly higher peak maximum values in the presence of plasma (CL1). This is indicative of plasma factor requirement for amplified response, at least in man and dog. This phenomenon, however, was not observed in rat, where A 23187 is the weakest stimulus tested. In contrast to man and most animal species previously tested (Hornebeck et al., 1987; Roberts et al., 1987; Suzuki et al., 1985), canine granulocytes are unresponsive to FMLP as a chemotaxin, aggregant or stimulator of chemiluminescence (Redl et al., 1983). The weak response in isolated canine PMN remains unclear. PMN of various species respond differently to various stimuli. Only selective and substantiated conclusions could be drawn from comparative observations made on animal and human PMN.
Acknowledgement This work was supported by the Deutsche Forschungsgemeinschaft.
REFERENCES Allen, R. C., Mead, M. E. and Kelly, J. L. (1985). Phagocyte oxygenation activity measured by chemiluminesccncc and chemiluminescence probing. in CRC Handbook of Methods for Oxygen Radical Research, Greenwald, R. A. (Ed.), CRC Press, Boca Raton, pp. 343-351. Cheung, K., Archibald, A . C. and Robinson, M. F. (1983). The origin of chemiluminescence produced by neutrophils stimulated by opsonized zymosan, J . Immunol., 130, 2324-2329. Dahlgren, C. (1987). Polymorphonuclear leukocyte chemiluminescence induced by formylmethionyl-leucylphenylalanine and phorbol myristate acetate: Effects of catalase and superoxide dismutase. Agents Actions, 21, 104-112. Friedel, R. and Mattenheimer, H. (1970). Release of metabolic enzymes from platelets during blood clotting of man, dog, rabbit and rat. Clin. Chim. Acta, 30, 37-46. Goldstein, J . M., Kaplan, H . B., Edelson, H . S. and Weissman, G. (1979). Ceruloplasmin a scavenger of superoxide anion radicals. J. Biol. Chem., 254, 4040-4045 Heberer, M., Emst, M., Durig, M., Allgower, M. and Fischer, H. (1982). Measurement of chemiluminescence in freshly drawn human blood.Klin. Wschr., 60, 1443-1448. Henson, P. M. and Johnston, Jr. B. B. (1987). Tissue injury in inflammation. Oxidants, proteinases and cationic proteins. J . Clin. Invest., 79, 669-674. Holt, M. E., Ryall, M. E. T. and Campbell, A. K. (1984). Albumin inhibits human polymorphonuclear leucocyte luminol-dependent chemiluminescence: evidence for oxygen radical scavenging. Br. J Exp. Pathol, 65, 231-241. Hornebeck, W., Soleihac, J. M., Tixier, J. M., Mocar, E. and Robert, L. (1987). Inhibition by elastase inhibitors of the Formyl Met Leu Phe-induced chemotaxis of rat polymorphonuclear leukocytes. Cell. Biochem. Funci., 5, 113-122. van Kampen, E. J. and Zijlstra, W. G. (1961). Standardization of hemoglobinometry. 11. The hemiglobincyanide method. Clin. Chim. Acta, 6, 538-544. Knudsen, F., Nielsen, A. H., Pedersen, J. 0. and Jersild, C. (1985). On the kinetics of complement activation, leucopenia and granuloyte-elastase release induced by haemodialysis. Scand. J . Clin. Lab. Invest., 45, 759--766. Lindena, J . Biittner, D. and Trautschold, I. (1984). Biological, analytical and experimental components of variance in a long-term study of plasma constituents in rat. J . Clin. Chem. Clin. Biochem., 22, 97-104. Lindena, J . and Trautschold, I. (1986). Catalytic enzyme activity concentration in plasma of man, sheep, dog, cat, rabbit, guinea pig, rat and mouse. Approach to a quantitative diagnostic enzymology, I. communication. J . Clin. Chem. Clin. Biochem., 24, 11-18. Lindena, J., Burkhardt, H . and Dwenger, A. (1987). Mechanisms of non-opsonized zymosan-induced and luminol-enhanced chemiluminescence in whole blood and isolated phagocytes. J . Clin. Chem. Clin. Biochem., 25, 765-778.
170 Onozaki, K . , Takenawa, T., Homma, Y. and Hashimato, T. (1983). The mechanism of macrophage activation induced by Ca” ionophore. Cell. Immunol., 75, 242-254. Redl, H., Flynn, P. J., Lamche, H., Schiesser, A . , Schlag, G. and Hammerschmidt, D.E. (1983). Aggregation, chernotaxis and chemiluminesccnce of canine granulocytes. Studies utilizing improved cell preparation techniques. Inflammation, I , 61-80. Roberts, R. L., Hatori, N., Drury, J . K . and Stiehm, E. R. (1987). Purification and properties of porcine polymorphonuclear cells. J . Immunol. Methods, 103, 21-32. Rossi, F., Della Bianca, V . . Grzeskowiak, M. and Zeni, L. (1986). Mechanism of oxygen free radicals production in granulocytes. In Oxygen Free Radicals in Shock, Novelli, G . P. and Ursini, F. (Eds), Karger, Basel, pp. 15-28. Suzuki, K., Asaoka, K . , Takahashi, K . and Fujikura, T.
J. LINDENA AND H. BURKHARDT (1985). Differences among primates in defence against infection: sensitivity of polymorphonuclear leukocytes to f Met-Leu-Phe. Cell. Biochem. Funct., 3, 297-303. Test, S. T. and Weiss, S. J . (1984). Quantitative and temporal characterization of the extracellular H,O, pool generated by human neutrophils. J . Biol.Chem., 259, 399-405. Wilson, E., Rice, W. G . , Kinkade, Jr., J . M., Merrill, J r . , A. H . Arnold, R. R . and Lamheth, J . D . (1987). Protein kinase C inhibition by sphingoid long-chain bases: effects on secretion in human neutrophils. Arch. Biochem. Biophys., 259, 204-214. Wolfson, M., McPhail, L.C., Nasrallah, V. N. and Snyderman, R. (1985). Phobol myristate acetate mediates redistribution of protein kinase C i n human neutrophils: potential role in t h e activation of the respiratory burst enzyme. J . Immunol., 135, 2051-2062.
