Clinical Science and Molecular Medicine (1975) 48, 201-212.
Factors influencing the entry of dye into neutrophil leucocytes in the Nitroblue Tetrazolium test
A. W. SEGAL
AND
A. J. LEVI
Northwich Park Hospital and Clinical Research Centre, Harrow, Middlesex
(Received 18 July 1974)
Summary
sample of the patient's blood with a solution of NBT. A sample of the mixture is then smeared on glass slides and the percentage of neutrophil leucocytes ('NBT score') containing discrete deposits of the dark-blue, water-insoluble diformazan, the product of reduction of NBT, is determined. A high NBT score (> 10%) was observed in patients with acute pyogenic infection, distinguishing them from normal subjects and patients with non-pyogenic disease, in whom a low score (~10%) was obtained (park et al., 1968). The addition of NBT to blood causes it to form a complex with and precipitate heparin and/or fibrinogen, to enter some neutrophilleucocytes as a result of phagocytosis and there be reduced (Segal & Levi, 1973). In the absence of defective reduction of dye, which occurs in the rare chronic granulomatous disease (Nathan, Baehner & Weaver, 1969), the NBT score is a measure of the phagocytosis of dye complexes by neutrophil leucocytes. In this study, an examination has been made of factors that could be responsible for the enhanced phagocytosis of dye complexes characteristic of a positive test. The possibility that low scores might indicate neutrophil dysfunction resulting from immaturity or exposure to immune complexes was also investigated.
1. The Nitroblue Tetrazolium (NBT) test is a measure of the phagocytosis of particulate complexes ofNBT and heparin and/or fibrinogen by neutrophil leucocytes. 2. Humoral factors in the plasma of ill patients stimulate neutrophil leucocytes of normal subjects to ingest these complexes. 3. The acute phase protein, orosomucoid and endotoxin stimulate reduction of NBT in vitro and could be responsible for some positive NBT tests in vivo. 4. The effect of factors promoting phagocytosis may be masked by an inability of neutrophil leucocytes to respond to stimulation. 5. This defective neutrophil function could result from the replacement of circulating neutrophil leucocytes by less mature, less actively phagocytic cells or by the presence of circulating immune complexes.
Key words: Nitroblue Tetrazolium test, infection, granulocytes, phagocytosis, acute phase proteins, immune complexes. Introduction
The Nitroblue Tetrazolium (NBT) test was introduced in 1968 by B. H. Park and colleagues (Park, Fikrig & Smithwick, 1968) as an empirical test to distinguish pyogenic bacterial infection from other diseases. The test is performed by incubating a
Subjects and methods
Subjects Serum samples were obtained in the course of routine investigations from patients on whom NBT tests had been performed by both the heparin and the ethylenediaminetetra-acetic acid (EDTA)/ficoll methods (Segal, Trustey & Levi, 1973) and stored at
Co~respondence: Dr A. W. Segal, The Hammersmith Hospital and Postgraduate Medical School, Du Cane Road. London WI2 OHS.
201
202
A. W. Segal and A. J: Levi
- 20°C before use. Pyogenic infection of less than 4 days' duration was classified as 'acute' and of greater than 10 days' duration as 'continuing'. Blood samples containing normal neutrophils were obtained from normal healthy adult volunteers and samples from different individuals were used in each experiment. Peripheral blood and bone marrow samples were obtained simultaneously from five patients with macrocytosis without megaloblastosis, in the course of routine investigations. Materials
Orosomucoid was isolated from 1 litre of human serum by ammonium sulphate precipitation, ionexchange chromatography and polyacrylamide gel electrophoresis (D. Ramsden & L. N. Louis, unpublished work). The precipitate produced by ammonium sulphate at a concentration of between 2·27 and 3·79 mol/l was redissolved in water and dialysed against running tap water and then against sodium acetate buffer (0·05 mol/I, pH 5·0). The solution in buffer was then applied to a carboxymethylcellulose column (50 cm x 7·5 ern) and the proteins were continuously eluted with sodium acetate buffer of increasing concentration from 0·05 to 0·4 mol/l. Orosomucoid-rich fractions were identified by radial immunodiffusion against rabbit anti-(human acid a'-glycoprotein) serum (Behringwerke; Mancini, Carbonara & Heremans, 1965), dialysed against distilled water and then concentrated to 5·0 ml in an ultracentrifugation cell with a IlM-I0 membrane (Amicon). Preparative polyacrylamide gel electrophoresis (Brownstone, 1969),with the tank and Tris-glycine buffer, pH 8·6, described by Charlwood (1969), was performed with a gel containing 8·0% (w/v) acrylamide and 5·0% (w/v) bisacrylamide. Fractions of the eluate containing pure orosomucoid were identified by two-dimensional crossed electrophoresis (Laurell, 1966), pooled, dialysed against water and freeze-dried. Rabbit anti-ovalbumin serum was a gift from Dr David Brown, Department of Immunology, The Hammersmith Hospital, London. Himalayan rabbits had been immunized by a deep intramuscular injection of ovalbumin (20 mg) in Freund's complete adjuvant, followed by several monthly booster injections. Equivalence was determined at 1·25 mgjrnl of ovalbumin by quantitative precipitation (Kabat, 1961). Anti-ovalbumin antibodies consisted almost entirely of the IgG class of immunoglobulin.
Prostaglandins were kindly donated by John E. Pike, Upjohn Pharmaceuticals, Michigan, U.S.A. The source of other materials is detailed in the text. Methods Effect of heterologous sera on reduction of NBT by normal neutrophilleucocytes. The effect of serum
from normal subjects and sick patients upon the reduction of NBT was investigated by isolating the neutrophil leucocytes and suspending them in the test sera before the addition of NBT. This was done in the presence of EDTA and ficoll (Gordon, Rowan, Brown & Carson, 1973), and heparin (Park et al., 1968) to simulate the conditions of the two main methods of performing NBT tests. Plasma, where EDTA was used as anticoagulant, and fibrinogen, where heparin was used, were added to the test system as a source of fibrinogen with which the dye forms a complex and precipitates to form the particulate object of phagocytosis (Segal & Levi, 1973). Preparations of normal neutrophilleucocytes were made from normal ABO blood group, 0 positive, blood as described previously (Segal & Levi, 1973). To examine the effect of heterologous serum, 100 III of serum from the normal subject or patient was gently mixed with 100 pI aliquots of the washed neutrophil preparation from normal subjects in disposable plastic test tubes (Luckham LIP 3S). After incubation at 37°C for 15 min, 100 III of plasma from the neutrophil donor, containing EDTA as anticoagulant (7·1 mmol/l), and 100 pI of 0·15 mol/I sodium chloride (saline) containing NBT (Sigma, 2·8 mmol/l) and ficoll (Pharmacia 400 mgjrnl) were added, the tubes were capped, mixed by gentle shaking, and incubated at 37°C for 30 min and then at room temperature for 15 min. Slide preparations were made, counterstained with Leishman's stain and the NBT score was determined (Segal et al., 1973). In some tests, a solution of fibrinogen and heparin was added in place of plasma from the neutrophil donor. Serum (100 Ill) was gently mixed with 100 III aliquots of the washed neutrophil suspension as described above and incubated at 37°C for 30 min. To each tube was then added 100 ttl of phosphate (0·09 moljl)-saline buffer (pH 7·4) containing fibrinogen (Kabi, 10 mg/m!) and
Neutrophil phagocytosis of dye complexes heparin (Weddel Pharmaceuticals, 50 i.u.jml) in solution, and 100 pI of a solution of NBT (2'8 mmol/I) in saline. The suspension was mixed by gentle shaking, incubated at 37°C for 15 min, and then at room temperature for a further 15 min and the NBT score determined. Concentrations of C-reactive protein and orosomucoid and NBT scores. Concentrations of these proteins were measured by the Mancini radial immunodiffusion technique (Mancini, Carbonara & Heremans, 1965). C-reactive protein was determined with commercially available Partagen plates (Behringwerke), Orosomucoid concentrations were measured by allowing 3 pI serum samples and standard sera (Behringwerke), diluted 1 in 11 in saline, to diffuse into plates containing 0·4 ml of specific rabbit anti-human serum (titre 0·4 mg/ml, Behringwerke) in 36·0 ml of 1'0% (w/v) agarose. Concentrations of orosomucoid and C-reactive protein were not determined in sera in which normal neutrophilleucocytes were resuspended. Effect of various compounds 011 NBT scores ill a model system. The model system was designed to result in the same final concentrations of component mixtures as those used in the standard NBT test. One volume (usually 0·25 ml) of the agent or dilutions thereof in phosphate-sodium chloride buffer was mixed with 2 vol. (usually 0·5 ml) of blood. This mixture was incubated at 3rC for 10 min, after which 1 vol. (usually 0·25 ml) of a solution of NBT (2'8 mmol/l) in saline was added. The mixture was gently shaken, incubated at 3rC for a further 20 min and NBT scores were determined in the usual way. Control experiments with buffer alone or buffer containing endotoxin (Escherichia coli, 0.127: B8; Difco, 30 pg/ml) as a stimulus to NBT reduction, were also performed. Initial studies were performed on blood containing EDTA as anticoagulant but other studies were performed with heparin as anticoagulant because of reports that EDTA decreased the sensitivity of the ofNBT test (Park & Good, 1970).The concentration of agents is expressed as the final concentration present after the addition of the dye. (a) Orosomucoid. The effect of orosomucoid was studied by incubating 10 pI of buffer containing orosomucoid (10 mg/ml) with 20 pI of blood containing heparin (25 i.u.jml) as anticoagulant for 10 min before the addition of 10 pI ofNBT (2'8 mmol/l) in saline (sixteen studies). Further studies were performed with orosomucoid (10 mg/ml) in buffer
203
which had been heated to 56°C for 30 min and with serial doubling dilutions of orosomucoid (0'6410·0 mg/ml) in phosphate-saline buffer. The effect of a combination of protamine sulphate and orosomucoid was studied by mixing 5 pI of protamine sulphate (Boots, 1 mg/ml) with 20 pI of blood and: 10pl of buffer; a solution of orosomucoid (lOmg/ml) in buffer; or endotoxin (20 pg/ml) in buffer. Control studies were performed by comparing the effect of orosomucoid with that of equal concentrations (10 mg/ml) of fetuin (Grand Island Biologic Co.) and of immunoglobulins (Lister Institute). (b) Endotoxin. Blood taken into EDTA (7'1 mmol/l) had endotoxin added to give concentrations from 1 pg/ml to 100 pg/ml. (c) Heparin. Blood was taken from healthy adults and 0'5 ml immediately added to each of seven plastic test tubes containing 0·25 ml of a mixture of buffer and sodium heparin (Evans Medical, preservative-free; 162 i.u.jrng) to give final heparin concentrations of 0,5, 5,0, 25, 50, 100, 135 and 250 i.u.jml, (d) Hormones and other compounds. These were added to heparinized blood (25 i.u.Iml) to give the following final concentrations: adrenaline (Macarthy's Pharmaceuticals), cortisone acetate (Roussel), hydrocortisone hemisuccinate (Organon) and noradrenaline (Winthrop), 10- 1 to 10- 9 mol/I; adenosine diphosphate (ADP, Sigma), 2'3 mmol-z-J nmol/l; adenosine cyclic monophosphate (cyclic AMP, Sigma), 5·0 pmol/I and 5·0 pmol/l; prostaglandins Et, s, and F 2 " , 2·8 and 0·28 pmol/l. (e) Autologous peripheral blood and bone marrow neutrophil leucocytes exposed to endotoxin and to heparin. Peripheral blood and bone marrow samples taken into EDTA (7'1 mmol/l) or heparin (50 i.u./ ml) as anticoagulant were exposed to endotoxin (10 pg/ml) in the model system; 0·5 nil of other samples were added to and mixed with 0·25 ml of buffer containing heparin (200 i.u./ml) to which was then added 0·25 ml of NBT (2'8 mmol/I). (f) Immune complexes. Ovalbumin-antiovalbumin immune complexes were made by quantitative precipitation. In a limited number of experiments, blood containing heparin (20 i.u.jml) as anticoagulant was exposed to immune complexes, at various antigen/antibody ratios, and in various dilutions, in the model system. The effect of the addition of immune complexes, NBT and endotoxin in varying combinations and sequence was also investigated.
A. W. Segal and A. J. Levi
204 Expression of results
Results
The NBT score is the percentage of neutrophil leucocytes showing reduction of NBT. The standardized NBT score is the NBT score as a percentage of the score obtained on maximal stimulation with endotoxin. Where appropriate, results are expressed as mean values ± SEM.
Effect of heterologous sera on reduction of NBT by normal neutrophil leucocytes
80
{;
( b) Heparin
70 60
•
o {;
•
50
• 40
{;
•
30
~ '"
IID
z
"0 c:
10
•
0 60
•
80
••
20
Ql
(; o
.
.,," '----"J.:.:..."~_L_• ·.....L.'_ _ .....L.I_ _
.L.._
:§> 0 50 "
___J
Ql
0
'"
ag'
•
20
o
{;
20
6.
40
6.
30
30 6.
l:i
0
6. 6.
20
.
.~, " ".,. ~, 10
l:i
6.6.
•
Z
"0 .S
."
50
u
IID
{;
10
l:i l:i
(;
o
30
60 ~ e;
•
•
6.
70
(a) EDTA/ficoll
40
Different sera had various effects on reduction of NBT by the same normal neutrophil leucocytes. There was significant correlation between an individual's original NBT score and the NBT score obtained after normal neutrophil leucocytes had been exposed to serum from that person (resuspended NBT score) in the presence of EDTA as anticoagulant (Fig. 1; r = 0-49, P < 0'002). The poor correlation (r = 0'29) between initial and resuspended NBT scores in the presence of heparin as anticoagulant is compatible with the variable reduction by normal neutrophil leucocytes in whole blood containing heparin (Segal et al., 1973). There was a close correlation between the initial and resuspended scores of patients with acute (r = 0'52, P
Factors influencing the entry of dye into neutrophil leucocytes in the Nitroblue Tetrazolium test
A. W. SEGAL
AND
A. J. LEVI
Northwich Park Hospital and Clinical Research Centre, Harrow, Middlesex
(Received 18 July 1974)
Summary
sample of the patient's blood with a solution of NBT. A sample of the mixture is then smeared on glass slides and the percentage of neutrophil leucocytes ('NBT score') containing discrete deposits of the dark-blue, water-insoluble diformazan, the product of reduction of NBT, is determined. A high NBT score (> 10%) was observed in patients with acute pyogenic infection, distinguishing them from normal subjects and patients with non-pyogenic disease, in whom a low score (~10%) was obtained (park et al., 1968). The addition of NBT to blood causes it to form a complex with and precipitate heparin and/or fibrinogen, to enter some neutrophilleucocytes as a result of phagocytosis and there be reduced (Segal & Levi, 1973). In the absence of defective reduction of dye, which occurs in the rare chronic granulomatous disease (Nathan, Baehner & Weaver, 1969), the NBT score is a measure of the phagocytosis of dye complexes by neutrophil leucocytes. In this study, an examination has been made of factors that could be responsible for the enhanced phagocytosis of dye complexes characteristic of a positive test. The possibility that low scores might indicate neutrophil dysfunction resulting from immaturity or exposure to immune complexes was also investigated.
1. The Nitroblue Tetrazolium (NBT) test is a measure of the phagocytosis of particulate complexes ofNBT and heparin and/or fibrinogen by neutrophil leucocytes. 2. Humoral factors in the plasma of ill patients stimulate neutrophil leucocytes of normal subjects to ingest these complexes. 3. The acute phase protein, orosomucoid and endotoxin stimulate reduction of NBT in vitro and could be responsible for some positive NBT tests in vivo. 4. The effect of factors promoting phagocytosis may be masked by an inability of neutrophil leucocytes to respond to stimulation. 5. This defective neutrophil function could result from the replacement of circulating neutrophil leucocytes by less mature, less actively phagocytic cells or by the presence of circulating immune complexes.
Key words: Nitroblue Tetrazolium test, infection, granulocytes, phagocytosis, acute phase proteins, immune complexes. Introduction
The Nitroblue Tetrazolium (NBT) test was introduced in 1968 by B. H. Park and colleagues (Park, Fikrig & Smithwick, 1968) as an empirical test to distinguish pyogenic bacterial infection from other diseases. The test is performed by incubating a
Subjects and methods
Subjects Serum samples were obtained in the course of routine investigations from patients on whom NBT tests had been performed by both the heparin and the ethylenediaminetetra-acetic acid (EDTA)/ficoll methods (Segal, Trustey & Levi, 1973) and stored at
Co~respondence: Dr A. W. Segal, The Hammersmith Hospital and Postgraduate Medical School, Du Cane Road. London WI2 OHS.
201
202
A. W. Segal and A. J: Levi
- 20°C before use. Pyogenic infection of less than 4 days' duration was classified as 'acute' and of greater than 10 days' duration as 'continuing'. Blood samples containing normal neutrophils were obtained from normal healthy adult volunteers and samples from different individuals were used in each experiment. Peripheral blood and bone marrow samples were obtained simultaneously from five patients with macrocytosis without megaloblastosis, in the course of routine investigations. Materials
Orosomucoid was isolated from 1 litre of human serum by ammonium sulphate precipitation, ionexchange chromatography and polyacrylamide gel electrophoresis (D. Ramsden & L. N. Louis, unpublished work). The precipitate produced by ammonium sulphate at a concentration of between 2·27 and 3·79 mol/l was redissolved in water and dialysed against running tap water and then against sodium acetate buffer (0·05 mol/I, pH 5·0). The solution in buffer was then applied to a carboxymethylcellulose column (50 cm x 7·5 ern) and the proteins were continuously eluted with sodium acetate buffer of increasing concentration from 0·05 to 0·4 mol/l. Orosomucoid-rich fractions were identified by radial immunodiffusion against rabbit anti-(human acid a'-glycoprotein) serum (Behringwerke; Mancini, Carbonara & Heremans, 1965), dialysed against distilled water and then concentrated to 5·0 ml in an ultracentrifugation cell with a IlM-I0 membrane (Amicon). Preparative polyacrylamide gel electrophoresis (Brownstone, 1969),with the tank and Tris-glycine buffer, pH 8·6, described by Charlwood (1969), was performed with a gel containing 8·0% (w/v) acrylamide and 5·0% (w/v) bisacrylamide. Fractions of the eluate containing pure orosomucoid were identified by two-dimensional crossed electrophoresis (Laurell, 1966), pooled, dialysed against water and freeze-dried. Rabbit anti-ovalbumin serum was a gift from Dr David Brown, Department of Immunology, The Hammersmith Hospital, London. Himalayan rabbits had been immunized by a deep intramuscular injection of ovalbumin (20 mg) in Freund's complete adjuvant, followed by several monthly booster injections. Equivalence was determined at 1·25 mgjrnl of ovalbumin by quantitative precipitation (Kabat, 1961). Anti-ovalbumin antibodies consisted almost entirely of the IgG class of immunoglobulin.
Prostaglandins were kindly donated by John E. Pike, Upjohn Pharmaceuticals, Michigan, U.S.A. The source of other materials is detailed in the text. Methods Effect of heterologous sera on reduction of NBT by normal neutrophilleucocytes. The effect of serum
from normal subjects and sick patients upon the reduction of NBT was investigated by isolating the neutrophil leucocytes and suspending them in the test sera before the addition of NBT. This was done in the presence of EDTA and ficoll (Gordon, Rowan, Brown & Carson, 1973), and heparin (Park et al., 1968) to simulate the conditions of the two main methods of performing NBT tests. Plasma, where EDTA was used as anticoagulant, and fibrinogen, where heparin was used, were added to the test system as a source of fibrinogen with which the dye forms a complex and precipitates to form the particulate object of phagocytosis (Segal & Levi, 1973). Preparations of normal neutrophilleucocytes were made from normal ABO blood group, 0 positive, blood as described previously (Segal & Levi, 1973). To examine the effect of heterologous serum, 100 III of serum from the normal subject or patient was gently mixed with 100 pI aliquots of the washed neutrophil preparation from normal subjects in disposable plastic test tubes (Luckham LIP 3S). After incubation at 37°C for 15 min, 100 III of plasma from the neutrophil donor, containing EDTA as anticoagulant (7·1 mmol/l), and 100 pI of 0·15 mol/I sodium chloride (saline) containing NBT (Sigma, 2·8 mmol/l) and ficoll (Pharmacia 400 mgjrnl) were added, the tubes were capped, mixed by gentle shaking, and incubated at 37°C for 30 min and then at room temperature for 15 min. Slide preparations were made, counterstained with Leishman's stain and the NBT score was determined (Segal et al., 1973). In some tests, a solution of fibrinogen and heparin was added in place of plasma from the neutrophil donor. Serum (100 Ill) was gently mixed with 100 III aliquots of the washed neutrophil suspension as described above and incubated at 37°C for 30 min. To each tube was then added 100 ttl of phosphate (0·09 moljl)-saline buffer (pH 7·4) containing fibrinogen (Kabi, 10 mg/m!) and
Neutrophil phagocytosis of dye complexes heparin (Weddel Pharmaceuticals, 50 i.u.jml) in solution, and 100 pI of a solution of NBT (2'8 mmol/I) in saline. The suspension was mixed by gentle shaking, incubated at 37°C for 15 min, and then at room temperature for a further 15 min and the NBT score determined. Concentrations of C-reactive protein and orosomucoid and NBT scores. Concentrations of these proteins were measured by the Mancini radial immunodiffusion technique (Mancini, Carbonara & Heremans, 1965). C-reactive protein was determined with commercially available Partagen plates (Behringwerke), Orosomucoid concentrations were measured by allowing 3 pI serum samples and standard sera (Behringwerke), diluted 1 in 11 in saline, to diffuse into plates containing 0·4 ml of specific rabbit anti-human serum (titre 0·4 mg/ml, Behringwerke) in 36·0 ml of 1'0% (w/v) agarose. Concentrations of orosomucoid and C-reactive protein were not determined in sera in which normal neutrophilleucocytes were resuspended. Effect of various compounds 011 NBT scores ill a model system. The model system was designed to result in the same final concentrations of component mixtures as those used in the standard NBT test. One volume (usually 0·25 ml) of the agent or dilutions thereof in phosphate-sodium chloride buffer was mixed with 2 vol. (usually 0·5 ml) of blood. This mixture was incubated at 3rC for 10 min, after which 1 vol. (usually 0·25 ml) of a solution of NBT (2'8 mmol/l) in saline was added. The mixture was gently shaken, incubated at 3rC for a further 20 min and NBT scores were determined in the usual way. Control experiments with buffer alone or buffer containing endotoxin (Escherichia coli, 0.127: B8; Difco, 30 pg/ml) as a stimulus to NBT reduction, were also performed. Initial studies were performed on blood containing EDTA as anticoagulant but other studies were performed with heparin as anticoagulant because of reports that EDTA decreased the sensitivity of the ofNBT test (Park & Good, 1970).The concentration of agents is expressed as the final concentration present after the addition of the dye. (a) Orosomucoid. The effect of orosomucoid was studied by incubating 10 pI of buffer containing orosomucoid (10 mg/ml) with 20 pI of blood containing heparin (25 i.u.jml) as anticoagulant for 10 min before the addition of 10 pI ofNBT (2'8 mmol/l) in saline (sixteen studies). Further studies were performed with orosomucoid (10 mg/ml) in buffer
203
which had been heated to 56°C for 30 min and with serial doubling dilutions of orosomucoid (0'6410·0 mg/ml) in phosphate-saline buffer. The effect of a combination of protamine sulphate and orosomucoid was studied by mixing 5 pI of protamine sulphate (Boots, 1 mg/ml) with 20 pI of blood and: 10pl of buffer; a solution of orosomucoid (lOmg/ml) in buffer; or endotoxin (20 pg/ml) in buffer. Control studies were performed by comparing the effect of orosomucoid with that of equal concentrations (10 mg/ml) of fetuin (Grand Island Biologic Co.) and of immunoglobulins (Lister Institute). (b) Endotoxin. Blood taken into EDTA (7'1 mmol/l) had endotoxin added to give concentrations from 1 pg/ml to 100 pg/ml. (c) Heparin. Blood was taken from healthy adults and 0'5 ml immediately added to each of seven plastic test tubes containing 0·25 ml of a mixture of buffer and sodium heparin (Evans Medical, preservative-free; 162 i.u.jrng) to give final heparin concentrations of 0,5, 5,0, 25, 50, 100, 135 and 250 i.u.jml, (d) Hormones and other compounds. These were added to heparinized blood (25 i.u.Iml) to give the following final concentrations: adrenaline (Macarthy's Pharmaceuticals), cortisone acetate (Roussel), hydrocortisone hemisuccinate (Organon) and noradrenaline (Winthrop), 10- 1 to 10- 9 mol/I; adenosine diphosphate (ADP, Sigma), 2'3 mmol-z-J nmol/l; adenosine cyclic monophosphate (cyclic AMP, Sigma), 5·0 pmol/I and 5·0 pmol/l; prostaglandins Et, s, and F 2 " , 2·8 and 0·28 pmol/l. (e) Autologous peripheral blood and bone marrow neutrophil leucocytes exposed to endotoxin and to heparin. Peripheral blood and bone marrow samples taken into EDTA (7'1 mmol/l) or heparin (50 i.u./ ml) as anticoagulant were exposed to endotoxin (10 pg/ml) in the model system; 0·5 nil of other samples were added to and mixed with 0·25 ml of buffer containing heparin (200 i.u./ml) to which was then added 0·25 ml of NBT (2'8 mmol/I). (f) Immune complexes. Ovalbumin-antiovalbumin immune complexes were made by quantitative precipitation. In a limited number of experiments, blood containing heparin (20 i.u.jml) as anticoagulant was exposed to immune complexes, at various antigen/antibody ratios, and in various dilutions, in the model system. The effect of the addition of immune complexes, NBT and endotoxin in varying combinations and sequence was also investigated.
A. W. Segal and A. J. Levi
204 Expression of results
Results
The NBT score is the percentage of neutrophil leucocytes showing reduction of NBT. The standardized NBT score is the NBT score as a percentage of the score obtained on maximal stimulation with endotoxin. Where appropriate, results are expressed as mean values ± SEM.
Effect of heterologous sera on reduction of NBT by normal neutrophil leucocytes
80
{;
( b) Heparin
70 60
•
o {;
•
50
• 40
{;
•
30
~ '"
IID
z
"0 c:
10
•
0 60
•
80
••
20
Ql
(; o
.
.,," '----"J.:.:..."~_L_• ·.....L.'_ _ .....L.I_ _
.L.._
:§> 0 50 "
___J
Ql
0
'"
ag'
•
20
o
{;
20
6.
40
6.
30
30 6.
l:i
0
6. 6.
20
.
.~, " ".,. ~, 10
l:i
6.6.
•
Z
"0 .S
."
50
u
IID
{;
10
l:i l:i
(;
o
30
60 ~ e;
•
•
6.
70
(a) EDTA/ficoll
40
Different sera had various effects on reduction of NBT by the same normal neutrophil leucocytes. There was significant correlation between an individual's original NBT score and the NBT score obtained after normal neutrophil leucocytes had been exposed to serum from that person (resuspended NBT score) in the presence of EDTA as anticoagulant (Fig. 1; r = 0-49, P < 0'002). The poor correlation (r = 0'29) between initial and resuspended NBT scores in the presence of heparin as anticoagulant is compatible with the variable reduction by normal neutrophil leucocytes in whole blood containing heparin (Segal et al., 1973). There was a close correlation between the initial and resuspended scores of patients with acute (r = 0'52, P
