Acta path, microbiol. scand. Sect. C, 84: 112-118, 1976
INFLUENCE O F ANTIBODIES AND THERMOLABILE SERUM FACTORS ON THE BACTERICIDAL ACTIVITY O F HUMAN NEUTROPHIL GRANULOCYTES CLAUS0. SOLBERG, KARENELINACHRISTIE, BODILLARSEN and OLAVT 0 N D E R The University of Bergen, School of Medicine, Medical Department B and Broegelmann Research Laboratory for Microbiology, The Gade Institute, Bergen, Norway
Solberg, C. O., Christie, K. E., Larsen, B. & Tmder, 0. Influence of antibodies and thermolabile serum factors on the bactericidal activity of human neutrophil granulocytes. Acta path. microbiol. scand. Sect. C , 84: 112-118, 1976. The influence of serum antibodies and thermolabile serum factors on the intracellular killing of Staphylococcus aureus by neutrophil granulocytes has been examined using a method which facilitates a precise in vitro evaluation of the phagocytic and bactericidal activities of human polymorphonuclear leucocytes. The bactericidal activity of the granulocytes was significantly less pronounced in the presence of serum absorbed with Staph. aureus or inactivated at 56" C for 30 minutes than in the presence of untreated serum. Specific antibodies seemed to stimulate the intracellular killing of bacteria more than thermolabile serum factors. Key words: Granulocytes; phagocytosis; intracellular killing; influence of serum factors.
C. 0. Solberg, Medical Department B, 5016 Haukeland sykehus, Bergen, Norway.
Received 25.viii.75
Accepted 23.x.75
Phagocytosis and intracellular killing of bacteria are important physiological functions of neutrophil granulocytes. I t is well known that antibodies and heat labile serum factors markedly enhance the phagocytic activity of the granulocytes (4, 5 , 6, 15). Whether these factors also influence the intracellular killing of bacteria is poorly understood. If this problem is to be studied, analyses of the phagocytic as well as the bactericidal activities of the granulocytes are required. However, studies of the interactions of bacteria and granulocytes have usually been confined to the ingestion phase or phagocytosis and little attention has been paid to the dy112
namics of the intracellular phase (for review see 1, 3, 5 ) . I n previous investigations, a method has been developed which facilitates a precise in vitro evaluation of the phagocytic and the bactericidal activities of the granulocytes ( 7 , 9, 11). Using this method, it was demonstrated that normal human serum by way of the granulocytes markedly enhanced not only the phagocytosis but also the intracellular killing of bacteria ( 10). The present study is concerned with the influence of serum antibodies and heat labile serum factors on the intracellular killing of bacteria by neutrophil granulocytes.
MATERIALS AND METHODS
Leucocytes Leucocytes obtained from normal individuals by “Isopaque”/dextran sedimentation of heparinized venous blood ( 10 units heparin per ml blood) were twice washed in heparinized saline ( 1 unit heparin per ml saline) by centrifugation at 500 g for 5 minutes ( 7 ) . A differential count was made and the cells were resuspended to 107 neutrophils per ml in Hanks balanced salt solution containing 0.1 per cent gelatin. The functional integrety of the isolated neutrophils remained intact as measured by latex particle phagocytosis and 95 to 99 per cent resisted staining with trypan blue. Eosinophil granulocyte contamination in 50 consecutive specimens varied from 0 to 7 per cent (mean 3 per cent), basophil granulocyte contamination from 0 to 2 per cent (mean 1 per cent), and lymphocyte-monocyte contamination from 12 to 23 per cent (mean 17 per cent). Small amounts of autologous serum, though less than 0.02 per cent, remained. Bacteria Staphylococcus aureus “Oxford” (Heatley strain, obtained from the National Collection of Type Cultures, CoIindaIe, London, 1958) was used as the test organism ( 7 ) . The bacteria were cultured overnight in Penassay broth (Difco), twice washed in 0.45 per cent saline and suspended in Hanks balanced salt solution to an optical density of 0.6 at 620 nm in a Beckman spectrophotometer ( 7 ) . This suspension was diluted in Hanks balanced salt solution containing 0.1 per cent gelatin to a concentration of 8-12 >: 107 colony forming units per ml. Serum Fresh, pooled normal serum from six adults was stored at -30” C in 1 ml aliquots. Immediately before each experiment, serum was thawed and added to Hanks balanced salt solution containing 0.1 per cent gelatin to make appropriate serum dilutions. All experiments in the present study were performed with the same pool of serum. Heat labile serum factors were inactivated by incubation of freshly thawed normal serum for 30 minutes at 56” C. With a view to absorption of antibodies in the serum to Staph. aureus, 4 ml serum diluted 1:5 in Hanks balanced salt solution were mixed with approximately 2 ml of packed bacteria (harvest from I0 Penassay agar plates). T h e bacteria had been twice washed in 0.45 per cent saline by centrifugation at 5000 g for 20 minutes at 4” C. T h e mixture was incubated for 45 minutes at 4” C, centrifuged at 5000 g for 20 minutes and the super-
natant (absorbed serum) stored in 1 ml aliquots. I n slide and tube agglutination tests, the absorbed serum gave only trace agglutination in the 1:5 dilution in contrast to unabsorbed serum which agglutinated in dilutions up to 1:512.
Test Procedure 0.5 ml leucocyte suspension, 0.1 ml bacteria suspension, and 0.4 ml diluted serum were added to 12 x 75 mm disposable plastic tubes. This provided about 2 bacteria per granulocyte and final serum concentrations as those indicated in the figures (vide infra). The tubes were incubated at 37” C using end over end rotation to promote contact between bacteria and leucocytes. Samples were removed periodically for determinations of the total number of viable bacteria and the number of viable intracellular bacteria. The total number of viable bacteria was determined after osmotic disruption of the leucocytes by adding 0.01 ml leucocyte-bacteria suspension to 1 ml distilled water. Quantitation of viable bacteria was made from appropriate dilutions of this suspension using a standard pour-plate technique and Penassay agar (Difco) . The number of viable intracellular bacteria was determined as described earlier ( 7 ) : 0.01 ml of the leucocyte-bacteria suspension and 1 mI Hanks balanced salt solution containing 0.1 per cent gelatin, 500 pg streptomycin, 500 units penicillin G and 2 mg phenylbutazone were incubated at 37” C for 15 minutes and centrifuged for 10 minutes at 500 g. The cellular pellet was twice washed in 5 ml Hanks balanced salt solution by centrifugation at 500 g for 10 minutes and resuspended in 1 ml distilled water to allow osmotic disruption of the leucocytes to occur. Quantitation of viable bacteria was made by the standard pour-plate technique. The bactericidal capacity of the granulocytes is proportional to the total number of bacteria killed and inversely proportional to the total number of viable bacteria or number of viable intracellular bacteria ( 7 ) . The number of bacteria phagocytized equals the number of viable intracellular bacteria plus the number of bacteria killed ( 7 ) . Controls Controls consisting of mixtures of bacteria and serum without leucocytes served to detect any direct bactericidal effect of the serum or serum factors, and mixtures of leucocytes and bacteria incubated without rotation served to detect any extracellular bactericidal activity caused, for example, by enzymes liberated from damaged granulocytes. No reduction in the number of viable bacteria was observed in these control tests during 2 hours’ incubation.
113
Statistical Method The Wilcoxon-test for two samples was used.
RESULTS
Influence of Inactivated Serum
Leucocyte-bacteria suspensions were incubated with inactivated and normal serum and the total number of viable bacteria and the number of viable intracellular bacteria were determined. In the tests containing normal serum a t a final concentration of 1 per cent or more, phagocytosis and killing of bacteria occurred
INACTIVATED SERUM
rapidly (Figs. 1 and 2 ) . In the tests with inactivated serum, the reduction in the total number of viable bacteria was significantly less pronounced and the number of viable intracellular bacteria increased slowly during the early phase of incubation; later it remained higher, indicating that the phagocytosis and, to a minor degree, the bactericidal activity of the granulocytes were reduced. Influence of Absorbed Serum
A test similar to that described above was performed using absorbed and normal serum.
NORMAL SERUM SERUM CONCENTRATION (PER CENT)
I0'
2w
4
lo6
Ym 4
>
LL 0 fY
w
m
f=
lo5
j
,
5 15
1
45
I
90
1
120
-
I
1
5 15
I
45
I
90
I
120
MINUTES MINUTES Fig. 1 . Total number of viable bacteria during incubation of granulocyte-bacteria suspensions with inactivated and normal serum (mean of four experiments). The differences between 0.25, 1.0 and 10 per cent inactivated and normal serum at 45, 90 and 120 minutes were significant ( ~ 5 0 . 0 2 5 ) .
114
)RMAL SERUM
ACTMATED SERUM SERUM CONCENTRAlION (PER CENT I
5 I5
SERM CONCENTRATION
9
45
90
li0
MINUTES
5 15
45
90
120
MINUTES
Fig. 2. Number of viable intracellular bacteria during incubation of granulocyte-bacteria suspensions with inactivated and normal serum (mean of four experiments). The differences between 0.25, 1.0 and 10 per cent inactivated and normal serum at 45, 90 and 120 minutes were significant ( ~ 5 0 . 0 2 5 ) .
The phagocytic and bactericidal activities of the granulocytes were significantly less pronounced in the presence of absorbed serum than in the presence of normal serum (Fig. 3). If inactivated absorbed serum was added to leucocyte-bacteria suspensions, the phagocytic and bactericidal activities of the granulocytes would be as poor as that to be seen if Hanks balanced salt solution had been used (10). Conversely, if absorbed serum was added to leucocyte-bacteria suspensions containing inactivated serum, the phagocytic and bactericidal activities of the granulocytes would increase to approximately the same level as that to be seen if normal serum had been used. DISCUSSION
Many of the studies of interactions of bacteria and granulocytes have been confined to
phagocytosis and little attention has been paid to the dynamics of the intracellular phase (1, 3, 5 ) . This has partly been due to technical difficulties. A major problem has remained namely the separation of extracellular and intracellular bacteria in an in vitro phagocytic system in order to evaluate the intracellular bactericidal processes (2, 7, 8, 9, 12). By our method, extracellular bacteria are effectively controlled by antibiotics which do not inactivate intracellular bacteria (9). Killing of extracellular bacteria by antibiotics takes, however, 10-15 minutes. During this period, inactivation of intracellular bacteria by granulocyte enzymes may significantly obscure the results (9). Thus, inhibition of the bactericidal activity of the granulocytes by phenylbutazone is a prerequisite for the determination of the number of viable intracellular bacteria while extracellular killing by antibiotics takes place. Accordingly, 115
-
ABSORBED SERUM
NORMAL SERUM
TOTAL BACTERIA
o-
1o7
- - - - -o INTRACELLULAR BACTERIA
4
Ly.
lo6
a
: 4
>
B
?\ I \
[L
w m
5
2
I I
lo5
\ '
9
\ \
loL
5 15
90
45
I20
MINUTES
5
i5
9'0
45
l2C
MINUTES
Fig. 3. Viable bacterial counts during incubation of granulocyte-bacteria suspensions with absorbed and normal serum at concentrations of 10 per cent (mean of four experiments). T h e differences between absorbed and normal serum at 45, 90 and 120 minutes were significant ( ~ 5 0 . 0 2 5 ) .
the essence of our method is the combined use of phenylbutazone for the inhibition of intracellular killing of bacteria and antibiotics for the control of extracellular bacteria. In the tests using absorbed serum, large numbers of bacteria remained in extracellular positions. However, significant numbers of bacteria were phagocytized and killed, indicating either that phagocytosis and intracelluIar killing of bacteria may take place in the absence of specific antibodies or that small amounts of antibodies were still present in 116
the absorbed serum or leucocyte suspension. Agglutination tests demonstrated that the amounts of specific antibodies in absorbed serum were too small to promote the phagocytosis and intracellular killing of Staph. aureus ( 10). Furthermore, the small amount of serum in the leucocyte suspension can also be considered to be negligible, since the phagocytic and bactericidal activities of the granulocytes were significantly less pronounced in the presence of Hanks balanced salt solution than in the presence of absorbed se-
rum (10). However, when the absorbed serum and the leucocyte suspension are added together, as in the present test system, the amount of antibodies may increase to a concentration of such magnitude that complement components will be activated and significant phagocytosis and intracellular killing of bacteria will be promoted. O n the other hand, it has been demonstrated that Fc-binding of nonspecific antibodies to Staph. aureus may stimulate the phagocytosis of the bacteria by human neutrophils, probably by activating complement components ( 13, 14). It remains to be determined whether the phagocytosis and intracellular killing of bacteria by the granulocytes in the presence of absorbed serum is due to small amounts of specific antibodies still present in the test suspension or to Fc-binding of nonspecific antibodies to the bacteria. The phagocytic and bactericidal activities of the granulocytes were significantly less pronounced in the presence of inactivated serum than in the presence of normal serum. Even after incubation for two hours, the total number of viable bacteria and the number of viable intracellular bacteria were markedly higher in the tests with inactivated serum. O n the other hand, both phagocytosis and intracellular killing of bacteria significantly increased when inactivated serum was added to the leucocyte-bacteria suspensions instead of absorbed serum, indicating that the stimulating effect of specific antibodies on the bactericidal activity of the granulocytes was superior to that of heat-labile serum factors. The reason why the bactericidal activity of the granulocytes was reduced in the absence of specific antibodies or heat labile serum factors remains unknown. However, if significant phagocytosis is to occur, attachment of sensitized bacteria to the phagocytic cell membrane by the Fc part of specific antibody molecules is required (Fc-receptor) (4, 5 , 6, 15). Heat labile serum factors, mainly complement components, probably stimulate phagocytosis by facilitating the attachment of sensitized bacteria to the phagocyte membrane (C-receptor). As soon as attachment
occurs, the bacterium will be surrounded by the leucocyte cell membrane which is forming a phagocytic vacuole into which bactericidal enzymes are released. The attachment of sensitized bacteria by specific antibodies and heat labile serum factors probably acts as a stimulus for the release of these enzymes. This would explain the markedly reduced bactericidal activity of the granulocytes in the absence of specific antibodies or heat labile serum factors.
REFERENCES
1. Brandt, L.: Studies on the phagocytic activity of neutrophilic leucocytes. Scand. J. Haematol., Suppl. 2, 1967. 2. Craig, C. P. & Suter, E.: Extracellular factors influencing staphylocidal capacity of human polymorphonuclear leucocytes. J. Immunol. 97: 287-296, 1966. 3. Hirsch, J . G.: Neutrophil and eosinophil leucocytes. In: Zweifach, B. W., Grant, L. & McCluskey, R. T. (Eds.): The inflammatory process. Academic Press, Inc., New York, 1965, p. 245-280. 4. Messner, R. P. & Jelinek, J . : Receptors for human yG globulin on human neutrophils. J. Clin. Invest. 49: 2165-2171, 1970. 5. Quie, P . G.: Disorders of phagocyte function. Year Book Medical Publishers, Inc., Chicago, 1972. 6. Quie, P . G., Messner, R. P. & Williams, R. C., Jr.: Phagocytosis in subacute bacterial endocarditis. Localization of the primary opsonic site to Fc fragment. J. Exper. Med. 128: 553570, 1968. 7. Solberg, C. 0.: Enhanced susceptibility to infection. A new method for the evaluation of neutrophil granulocyte functions. Acta path. microbiol. scand. Sect. B, 80: 10-18, 1972. 8. Solberg, C. 0.: Evaluation of neutrophil granulocyte functions. Acta path. microbiol. scand. Sect. B, 80: 559-563, 1972. 9. Solberg, C. 0.: Protection of phagocytized bacteria against antibiotics. A new method for the evaluation of neutrophil granulocyte functions. Acta med. scand. 191: 383-387, 1972. 10. Solberg, C. 0. & Hellum, K. B.: Influence of serum on the bactericidal activity of neutrophi1 granulocytes. Acta path. microbiol. scand. Sect. B, 81: 621-626, 1973. 11. Solberg, C. 0. & Hellum, K. B.: Neutrophil granulocyte function in bacterial infections. Lancet ZZ: 727-730, 1972. 12. Suzuki, J . B., Booth, R . R . & Grecr, N . : Eva-
117
luation of phagocytic activity by ingestion of labeled bacteria. J. Infect. Dis. 123: 93-96, 1971. 13. van Oss, C. I. & Stinson, M . W.: Immunoglobulins as specific opsonins. I. The influence of polyclonal and monoclonal immunoglobulins on in vitro phagocytosis of latex particles and staphylococci by human neutrophils. J. Reticuloendothel. SOC. 8: 397-406, 1970.
118
14. van Oss, C . I., Woeppel, M . S. & Marquart, S . E.: Immunoglobulins as specific opsonins. 111. The opsonizing power of fragments of polyclonal and monoclonal immunoglobulin G. J. Reticuloendothel. SOC.13: 221-230, 1973. 15. Williams, R . C., Jr.: Opsonins in phagocytosis. I n Williams, R. C., Jr. & Fudenberg, H. H. (Eds.) : Phagocytic mechanism in health and disease. Georg Thieme Publishers, Stuttgart, 1972, p. 167-178.
INFLUENCE O F ANTIBODIES AND THERMOLABILE SERUM FACTORS ON THE BACTERICIDAL ACTIVITY O F HUMAN NEUTROPHIL GRANULOCYTES CLAUS0. SOLBERG, KARENELINACHRISTIE, BODILLARSEN and OLAVT 0 N D E R The University of Bergen, School of Medicine, Medical Department B and Broegelmann Research Laboratory for Microbiology, The Gade Institute, Bergen, Norway
Solberg, C. O., Christie, K. E., Larsen, B. & Tmder, 0. Influence of antibodies and thermolabile serum factors on the bactericidal activity of human neutrophil granulocytes. Acta path. microbiol. scand. Sect. C , 84: 112-118, 1976. The influence of serum antibodies and thermolabile serum factors on the intracellular killing of Staphylococcus aureus by neutrophil granulocytes has been examined using a method which facilitates a precise in vitro evaluation of the phagocytic and bactericidal activities of human polymorphonuclear leucocytes. The bactericidal activity of the granulocytes was significantly less pronounced in the presence of serum absorbed with Staph. aureus or inactivated at 56" C for 30 minutes than in the presence of untreated serum. Specific antibodies seemed to stimulate the intracellular killing of bacteria more than thermolabile serum factors. Key words: Granulocytes; phagocytosis; intracellular killing; influence of serum factors.
C. 0. Solberg, Medical Department B, 5016 Haukeland sykehus, Bergen, Norway.
Received 25.viii.75
Accepted 23.x.75
Phagocytosis and intracellular killing of bacteria are important physiological functions of neutrophil granulocytes. I t is well known that antibodies and heat labile serum factors markedly enhance the phagocytic activity of the granulocytes (4, 5 , 6, 15). Whether these factors also influence the intracellular killing of bacteria is poorly understood. If this problem is to be studied, analyses of the phagocytic as well as the bactericidal activities of the granulocytes are required. However, studies of the interactions of bacteria and granulocytes have usually been confined to the ingestion phase or phagocytosis and little attention has been paid to the dy112
namics of the intracellular phase (for review see 1, 3, 5 ) . I n previous investigations, a method has been developed which facilitates a precise in vitro evaluation of the phagocytic and the bactericidal activities of the granulocytes ( 7 , 9, 11). Using this method, it was demonstrated that normal human serum by way of the granulocytes markedly enhanced not only the phagocytosis but also the intracellular killing of bacteria ( 10). The present study is concerned with the influence of serum antibodies and heat labile serum factors on the intracellular killing of bacteria by neutrophil granulocytes.
MATERIALS AND METHODS
Leucocytes Leucocytes obtained from normal individuals by “Isopaque”/dextran sedimentation of heparinized venous blood ( 10 units heparin per ml blood) were twice washed in heparinized saline ( 1 unit heparin per ml saline) by centrifugation at 500 g for 5 minutes ( 7 ) . A differential count was made and the cells were resuspended to 107 neutrophils per ml in Hanks balanced salt solution containing 0.1 per cent gelatin. The functional integrety of the isolated neutrophils remained intact as measured by latex particle phagocytosis and 95 to 99 per cent resisted staining with trypan blue. Eosinophil granulocyte contamination in 50 consecutive specimens varied from 0 to 7 per cent (mean 3 per cent), basophil granulocyte contamination from 0 to 2 per cent (mean 1 per cent), and lymphocyte-monocyte contamination from 12 to 23 per cent (mean 17 per cent). Small amounts of autologous serum, though less than 0.02 per cent, remained. Bacteria Staphylococcus aureus “Oxford” (Heatley strain, obtained from the National Collection of Type Cultures, CoIindaIe, London, 1958) was used as the test organism ( 7 ) . The bacteria were cultured overnight in Penassay broth (Difco), twice washed in 0.45 per cent saline and suspended in Hanks balanced salt solution to an optical density of 0.6 at 620 nm in a Beckman spectrophotometer ( 7 ) . This suspension was diluted in Hanks balanced salt solution containing 0.1 per cent gelatin to a concentration of 8-12 >: 107 colony forming units per ml. Serum Fresh, pooled normal serum from six adults was stored at -30” C in 1 ml aliquots. Immediately before each experiment, serum was thawed and added to Hanks balanced salt solution containing 0.1 per cent gelatin to make appropriate serum dilutions. All experiments in the present study were performed with the same pool of serum. Heat labile serum factors were inactivated by incubation of freshly thawed normal serum for 30 minutes at 56” C. With a view to absorption of antibodies in the serum to Staph. aureus, 4 ml serum diluted 1:5 in Hanks balanced salt solution were mixed with approximately 2 ml of packed bacteria (harvest from I0 Penassay agar plates). T h e bacteria had been twice washed in 0.45 per cent saline by centrifugation at 5000 g for 20 minutes at 4” C. T h e mixture was incubated for 45 minutes at 4” C, centrifuged at 5000 g for 20 minutes and the super-
natant (absorbed serum) stored in 1 ml aliquots. I n slide and tube agglutination tests, the absorbed serum gave only trace agglutination in the 1:5 dilution in contrast to unabsorbed serum which agglutinated in dilutions up to 1:512.
Test Procedure 0.5 ml leucocyte suspension, 0.1 ml bacteria suspension, and 0.4 ml diluted serum were added to 12 x 75 mm disposable plastic tubes. This provided about 2 bacteria per granulocyte and final serum concentrations as those indicated in the figures (vide infra). The tubes were incubated at 37” C using end over end rotation to promote contact between bacteria and leucocytes. Samples were removed periodically for determinations of the total number of viable bacteria and the number of viable intracellular bacteria. The total number of viable bacteria was determined after osmotic disruption of the leucocytes by adding 0.01 ml leucocyte-bacteria suspension to 1 ml distilled water. Quantitation of viable bacteria was made from appropriate dilutions of this suspension using a standard pour-plate technique and Penassay agar (Difco) . The number of viable intracellular bacteria was determined as described earlier ( 7 ) : 0.01 ml of the leucocyte-bacteria suspension and 1 mI Hanks balanced salt solution containing 0.1 per cent gelatin, 500 pg streptomycin, 500 units penicillin G and 2 mg phenylbutazone were incubated at 37” C for 15 minutes and centrifuged for 10 minutes at 500 g. The cellular pellet was twice washed in 5 ml Hanks balanced salt solution by centrifugation at 500 g for 10 minutes and resuspended in 1 ml distilled water to allow osmotic disruption of the leucocytes to occur. Quantitation of viable bacteria was made by the standard pour-plate technique. The bactericidal capacity of the granulocytes is proportional to the total number of bacteria killed and inversely proportional to the total number of viable bacteria or number of viable intracellular bacteria ( 7 ) . The number of bacteria phagocytized equals the number of viable intracellular bacteria plus the number of bacteria killed ( 7 ) . Controls Controls consisting of mixtures of bacteria and serum without leucocytes served to detect any direct bactericidal effect of the serum or serum factors, and mixtures of leucocytes and bacteria incubated without rotation served to detect any extracellular bactericidal activity caused, for example, by enzymes liberated from damaged granulocytes. No reduction in the number of viable bacteria was observed in these control tests during 2 hours’ incubation.
113
Statistical Method The Wilcoxon-test for two samples was used.
RESULTS
Influence of Inactivated Serum
Leucocyte-bacteria suspensions were incubated with inactivated and normal serum and the total number of viable bacteria and the number of viable intracellular bacteria were determined. In the tests containing normal serum a t a final concentration of 1 per cent or more, phagocytosis and killing of bacteria occurred
INACTIVATED SERUM
rapidly (Figs. 1 and 2 ) . In the tests with inactivated serum, the reduction in the total number of viable bacteria was significantly less pronounced and the number of viable intracellular bacteria increased slowly during the early phase of incubation; later it remained higher, indicating that the phagocytosis and, to a minor degree, the bactericidal activity of the granulocytes were reduced. Influence of Absorbed Serum
A test similar to that described above was performed using absorbed and normal serum.
NORMAL SERUM SERUM CONCENTRATION (PER CENT)
I0'
2w
4
lo6
Ym 4
>
LL 0 fY
w
m
f=
lo5
j
,
5 15
1
45
I
90
1
120
-
I
1
5 15
I
45
I
90
I
120
MINUTES MINUTES Fig. 1 . Total number of viable bacteria during incubation of granulocyte-bacteria suspensions with inactivated and normal serum (mean of four experiments). The differences between 0.25, 1.0 and 10 per cent inactivated and normal serum at 45, 90 and 120 minutes were significant ( ~ 5 0 . 0 2 5 ) .
114
)RMAL SERUM
ACTMATED SERUM SERUM CONCENTRAlION (PER CENT I
5 I5
SERM CONCENTRATION
9
45
90
li0
MINUTES
5 15
45
90
120
MINUTES
Fig. 2. Number of viable intracellular bacteria during incubation of granulocyte-bacteria suspensions with inactivated and normal serum (mean of four experiments). The differences between 0.25, 1.0 and 10 per cent inactivated and normal serum at 45, 90 and 120 minutes were significant ( ~ 5 0 . 0 2 5 ) .
The phagocytic and bactericidal activities of the granulocytes were significantly less pronounced in the presence of absorbed serum than in the presence of normal serum (Fig. 3). If inactivated absorbed serum was added to leucocyte-bacteria suspensions, the phagocytic and bactericidal activities of the granulocytes would be as poor as that to be seen if Hanks balanced salt solution had been used (10). Conversely, if absorbed serum was added to leucocyte-bacteria suspensions containing inactivated serum, the phagocytic and bactericidal activities of the granulocytes would increase to approximately the same level as that to be seen if normal serum had been used. DISCUSSION
Many of the studies of interactions of bacteria and granulocytes have been confined to
phagocytosis and little attention has been paid to the dynamics of the intracellular phase (1, 3, 5 ) . This has partly been due to technical difficulties. A major problem has remained namely the separation of extracellular and intracellular bacteria in an in vitro phagocytic system in order to evaluate the intracellular bactericidal processes (2, 7, 8, 9, 12). By our method, extracellular bacteria are effectively controlled by antibiotics which do not inactivate intracellular bacteria (9). Killing of extracellular bacteria by antibiotics takes, however, 10-15 minutes. During this period, inactivation of intracellular bacteria by granulocyte enzymes may significantly obscure the results (9). Thus, inhibition of the bactericidal activity of the granulocytes by phenylbutazone is a prerequisite for the determination of the number of viable intracellular bacteria while extracellular killing by antibiotics takes place. Accordingly, 115
-
ABSORBED SERUM
NORMAL SERUM
TOTAL BACTERIA
o-
1o7
- - - - -o INTRACELLULAR BACTERIA
4
Ly.
lo6
a
: 4
>
B
?\ I \
[L
w m
5
2
I I
lo5
\ '
9
\ \
loL
5 15
90
45
I20
MINUTES
5
i5
9'0
45
l2C
MINUTES
Fig. 3. Viable bacterial counts during incubation of granulocyte-bacteria suspensions with absorbed and normal serum at concentrations of 10 per cent (mean of four experiments). T h e differences between absorbed and normal serum at 45, 90 and 120 minutes were significant ( ~ 5 0 . 0 2 5 ) .
the essence of our method is the combined use of phenylbutazone for the inhibition of intracellular killing of bacteria and antibiotics for the control of extracellular bacteria. In the tests using absorbed serum, large numbers of bacteria remained in extracellular positions. However, significant numbers of bacteria were phagocytized and killed, indicating either that phagocytosis and intracelluIar killing of bacteria may take place in the absence of specific antibodies or that small amounts of antibodies were still present in 116
the absorbed serum or leucocyte suspension. Agglutination tests demonstrated that the amounts of specific antibodies in absorbed serum were too small to promote the phagocytosis and intracellular killing of Staph. aureus ( 10). Furthermore, the small amount of serum in the leucocyte suspension can also be considered to be negligible, since the phagocytic and bactericidal activities of the granulocytes were significantly less pronounced in the presence of Hanks balanced salt solution than in the presence of absorbed se-
rum (10). However, when the absorbed serum and the leucocyte suspension are added together, as in the present test system, the amount of antibodies may increase to a concentration of such magnitude that complement components will be activated and significant phagocytosis and intracellular killing of bacteria will be promoted. O n the other hand, it has been demonstrated that Fc-binding of nonspecific antibodies to Staph. aureus may stimulate the phagocytosis of the bacteria by human neutrophils, probably by activating complement components ( 13, 14). It remains to be determined whether the phagocytosis and intracellular killing of bacteria by the granulocytes in the presence of absorbed serum is due to small amounts of specific antibodies still present in the test suspension or to Fc-binding of nonspecific antibodies to the bacteria. The phagocytic and bactericidal activities of the granulocytes were significantly less pronounced in the presence of inactivated serum than in the presence of normal serum. Even after incubation for two hours, the total number of viable bacteria and the number of viable intracellular bacteria were markedly higher in the tests with inactivated serum. O n the other hand, both phagocytosis and intracellular killing of bacteria significantly increased when inactivated serum was added to the leucocyte-bacteria suspensions instead of absorbed serum, indicating that the stimulating effect of specific antibodies on the bactericidal activity of the granulocytes was superior to that of heat-labile serum factors. The reason why the bactericidal activity of the granulocytes was reduced in the absence of specific antibodies or heat labile serum factors remains unknown. However, if significant phagocytosis is to occur, attachment of sensitized bacteria to the phagocytic cell membrane by the Fc part of specific antibody molecules is required (Fc-receptor) (4, 5 , 6, 15). Heat labile serum factors, mainly complement components, probably stimulate phagocytosis by facilitating the attachment of sensitized bacteria to the phagocyte membrane (C-receptor). As soon as attachment
occurs, the bacterium will be surrounded by the leucocyte cell membrane which is forming a phagocytic vacuole into which bactericidal enzymes are released. The attachment of sensitized bacteria by specific antibodies and heat labile serum factors probably acts as a stimulus for the release of these enzymes. This would explain the markedly reduced bactericidal activity of the granulocytes in the absence of specific antibodies or heat labile serum factors.
REFERENCES
1. Brandt, L.: Studies on the phagocytic activity of neutrophilic leucocytes. Scand. J. Haematol., Suppl. 2, 1967. 2. Craig, C. P. & Suter, E.: Extracellular factors influencing staphylocidal capacity of human polymorphonuclear leucocytes. J. Immunol. 97: 287-296, 1966. 3. Hirsch, J . G.: Neutrophil and eosinophil leucocytes. In: Zweifach, B. W., Grant, L. & McCluskey, R. T. (Eds.): The inflammatory process. Academic Press, Inc., New York, 1965, p. 245-280. 4. Messner, R. P. & Jelinek, J . : Receptors for human yG globulin on human neutrophils. J. Clin. Invest. 49: 2165-2171, 1970. 5. Quie, P . G.: Disorders of phagocyte function. Year Book Medical Publishers, Inc., Chicago, 1972. 6. Quie, P . G., Messner, R. P. & Williams, R. C., Jr.: Phagocytosis in subacute bacterial endocarditis. Localization of the primary opsonic site to Fc fragment. J. Exper. Med. 128: 553570, 1968. 7. Solberg, C. 0.: Enhanced susceptibility to infection. A new method for the evaluation of neutrophil granulocyte functions. Acta path. microbiol. scand. Sect. B, 80: 10-18, 1972. 8. Solberg, C. 0.: Evaluation of neutrophil granulocyte functions. Acta path. microbiol. scand. Sect. B, 80: 559-563, 1972. 9. Solberg, C. 0.: Protection of phagocytized bacteria against antibiotics. A new method for the evaluation of neutrophil granulocyte functions. Acta med. scand. 191: 383-387, 1972. 10. Solberg, C. 0. & Hellum, K. B.: Influence of serum on the bactericidal activity of neutrophi1 granulocytes. Acta path. microbiol. scand. Sect. B, 81: 621-626, 1973. 11. Solberg, C. 0. & Hellum, K. B.: Neutrophil granulocyte function in bacterial infections. Lancet ZZ: 727-730, 1972. 12. Suzuki, J . B., Booth, R . R . & Grecr, N . : Eva-
117
luation of phagocytic activity by ingestion of labeled bacteria. J. Infect. Dis. 123: 93-96, 1971. 13. van Oss, C. I. & Stinson, M . W.: Immunoglobulins as specific opsonins. I. The influence of polyclonal and monoclonal immunoglobulins on in vitro phagocytosis of latex particles and staphylococci by human neutrophils. J. Reticuloendothel. SOC. 8: 397-406, 1970.
118
14. van Oss, C . I., Woeppel, M . S. & Marquart, S . E.: Immunoglobulins as specific opsonins. 111. The opsonizing power of fragments of polyclonal and monoclonal immunoglobulin G. J. Reticuloendothel. SOC.13: 221-230, 1973. 15. Williams, R . C., Jr.: Opsonins in phagocytosis. I n Williams, R. C., Jr. & Fudenberg, H. H. (Eds.) : Phagocytic mechanism in health and disease. Georg Thieme Publishers, Stuttgart, 1972, p. 167-178.
