A P M I S 99: 965-971, 1991
Granulocyte functions and Neisseria rnenirzgitiaids: influence of properdindeficient serum CLAES SODERSTROM', JEAN HENRIK BRACONIER', ANDERS G. SJOHOLM', and BRITT THURESSON' Department of Infectious Diseases' and Department of Medical Microbiology', University of Lund, Lund, Sweden
Soderstrom, C., Braconier, J. H., Sjoholm, A. G. & Thuresson, B. Granulocyte functions and Neisseria meningitidis: influence of properdin-deficient serum. APMIS 99: 965-971, 1991. Granulocyte-mediated reactions such as opsonization, chemotaxis, and release of granulocyte myeloperoxidase and lactoferrin were studied in properdin-deficient and normal human serum incubated with serogroup A and W-135 meningococci. There were no differences between the sera when serogroup A meningococci were studied. Opsonic and chemotactic activity were impaired against serogroup W-135 meningococci in properdin-deficient serum. Restitution with properdin restored both activities. We found similar release of myeloperoxidase and lactoferrin from granulocytes challenged with serogroup A or W-135 meningococci in either sera. These findings are in accordance with the clinical observations of meningococcal infections caused by serogroup W- 135 in properdindeficient patients as well as the absence of infections caused by serogroup A meningococci. Key words: Properdin deficiency; Neisseria meningitidis; opsonophagocytic killing; chemotactic activity. Claes Soderstrom, Department of Infectious Diseases, University Hospital of Lund, S-221 85 Lund, Sweden.
T h e complement component properdin stabilizes the alternative pathway C 3 convertase C3bBb (5). Inherited deficiency or dysfunction of properdin predisposes t o meningococcal disease (2, 15, 18-20). Several C3-dependent functions a r e impaired in properdin-deficient sera (1, 2, 18, 22-24). The meningococcal serogroups B, C, Y a n d W-I35 have been identified in patients with deficiency o r dysfunction of properdin (2, 18, 19, 24). Serogroup W-135 accounts for about 50% of meningococcal infections among properdindeficient patients (24). Infection due to serogroup A has so far not been reported in these pa tien ts. Serum bactericidal reactions require antibody
Received July 26, 1989. Accepted March 1 1 , 1991.
and a n intact complement system and are probably of predominant importance in defense against Neisseria meningitidis (8). There is also evidence for a role of granulocytes in defense against meningococcal disease (3, 16). In the present investigation, the effect of properdindeficient serum o n granulocyte-mediated reactions against serogroup A a n d W-135 meningococci was studied.
MATERIALS A N D METHODS Bacteria N. meningitidis serogroup A (strain A-3269), serogroup W-135 (strain W-7911) and Staphylococcus aureus (strain 502 A) were used (14, 23). The serogroup W-135 strain was isolated from a patient with properdin deficiency (23). The bacteria were stored and cultured as previously described (14, 23). For radiolabelling, the bacteria were grown overnight in
965
SUDERSTRUM et a!.
Muller-Hinton broth (23). One ml from the overnight culture was added to 8 ml of fresh Miiller-Hinton broth supplemented with 16 pCi D-(U'4C)glucose (Amersham International, England). The meningococci were harvested after five h growth, and were then washed three times in isotonic veronal buffered saline with 0.1% gelatin, 0.15 mM Ca2+, 0.5 mM Mg2+ and 0.2% glucose (wt/vol) (GVB). The meningococci were adjusted to about 8 x lo7viable organisms per ml in GVB. Sera Pooled normal human serum from 20 healthy blood donors was used as a control. The properdindeficient serum was from a healthy adult and was obtained before and one month after immunization with a meningococcal (polysaccharide A and C) vaccine, respectively (23). The sera were stored in aliquots at - 80°C. The properdin-deficient and the control serum showed no antibodies against serogroup W-135 as determined by ELISA (24). The control and preimmune properdin-deficient serum had similar levels of serogroup A antibodies (23). Alternative pathway-mediated reactions were studied in sera chelated with 10 mM EGTA (ethylene glycol-tetraacetic acid) (6) and reconstituted with 4 mM Mgz+. Granulocytes Human granulocytes were prepared from peripheral blood by using Isopaque-Ficoll (Pharmacia Fine Chemicals, Uppsala, Sweden). The bottom layer was mixed with 0.87% NH,Cl for lysis of red cells, washed twice, and suspended in GVB. Purified properdin Properdin was purified as described earlier (18) and added at a physiological concentration (19 mg/l) in restitution experiments. Phagocytic killing About 4 x lo6 granulocytes and 6 x lo6 bacteria were incubated during agitation at 37°C with serum at various concentrations (total volume 0.25 ml). Aliquots of 20 pl were removed at timed intervals, diluted in sterile distilled water with 5% saponin (wt/ vol) at+4"C for five min in order to rupture the granulocytes (25) and then further diluted in phosphate-buffered saline (PBS) (0.03 M sodium phosphate, 0.12 M NaCI, pH 7.2). The total number of viable meningococci was determined by colony counting using the pour-plate method. All assays were performed in duplicate. Commercial IgG, 50 g/l, (Gammonativo Kabi, Sweden) was used as a source of anti-meningococcal antibody (23). Bacterial suspensions (6 x lo7 bacterialiml) were preincubated for 15 min at 37°C with equal volumes (0.25 ml) of IgG at 50 and 12.5 g/l, or with buffer alone. After washing, the bacteria were used in the opsonophagocytic
966
assay together with Mg2+EGTA-chelatedserum (concentration 25%). Uptake of radiolabelled bacteria About 2 x lo7 radiolabelled bacteria were agitated with 4 x lo6 human granulocytes in GVB-diluted serum at 37°C. The sera were used at final concentrations of 5-25"?' in a total volume of 1.0 ml. At different intervals, aliquots of 0.3 ml were transferred to 1.O ml ice-cold GVB. After centrifugation (850 x g , seven min, 4°C) the supernatants were transferred to 10 ml Insta-gel (Packard, Downers Grove, Ill., USA). The granulocytes were separated from extracellular meningococci by differential centrifugation (three times, 80 x g, seven min, + 4°C). The final pellet was dissolved in 0.2 ml Soluene-350 (Packard, Downers Grove, Ill., USA), and transferred to 10 ml Insta-gel. After 24 h at+4"C, radioactivity was counted in a scintillation counter (Packard Tri-Carb 460 C). The uptake of radiolabelled bacteria was expressed as a percentage of the total radioactivity in each incubation mixture.
+
Intracellular killing S. aureus 502 A (about 2 x 10') were preincubated with 50% control serum for three min in a total volume of 2.0 ml. After washing, the opsonized staphylococci (about 8 x lo7) were incubated with granulocytes (about 5 x lo6) at 37°C for 15 rnin in a total volume of 1.8 ml. Lysostaphin (48 IU) in ice-cold buffer was added (final volume 3 ml) to eliminate extracellular S. aureus (14). After one min, lysostaphin was removed by dilution and repeated washings at +4"C. The washed granulocytes (about 6 x lo5) were incubated in buffer or serum at a concentration of 36% in a total volume of 0.25 ml. Surviving bacteria were quantified after lysis of the granulocytes by viable counts using the pour-plate method. Granulocyte chemotaxis Chemotaxis was studied by the under-agarose technique (12) as modified by Forsgren & Schmeling (7). Sera at a concentration of 50% were incubated with meningococci (1 x 109/ml)at 37°C for 30 rnin (unchelated serum) and 60 rnin (Mg2+EGTA-chelated serum). Bacteria were removed by centrifugation. The supernatant sera were heat treated (56°C 30 min) and used as chemoattractants. The culture dishes were incubated for three h at 37°C in a humidified atmosphere with 5% C02. The chemotaxis was quantified by measuring the leading front of a greatly enlarged projection of the migration patterns. Results were expressed as a chemotactic index, i.e. directed migration / random migration. Release of myeloperoxidase and lactoferrin during phagocytosis About 4 x lo7 meningococci were incubated with 4 x lo6human granulocytes and serum at 37°C during
GRANULOCYTE FUNCTIONS AND NEISSERIA MENINGITIDIS
agitation (total volume 0.5 ml). Aliquots of 0.2 ml were removed after 30 min and granulocytes were separated by centrifugation (850 x g, seven min, 4"C), and the content of myeloperoxidase and lactoferrin in the supernatants was determined by radioimmunoassay (1 3).
+
TABLE 1. Opsonophagocytic killing of serogroup A meningococci" in properdin-deficient serum (PDs) obtained before and four weeks after vaccination with serogroup A and C polysaccharides Bacterial killing. (YO) Before vaccination After vaccination Serum Serum+ Serum Serum+
PMNb
RESULTS
Opsonophagocy t ic killing Moderately reduced phagocytic killing of serogroup W-135 was found in the presence of properdin-deficient serum as compared with control serum at concentrations of 10%. Due to the bactericidal activity of serum, it was difficult to assess the contribution of phagocytic killing (Fig. 1). Opsonic activity mediated through the alternative pathway was studied with Mg2+ EGTA at serum concentrations of 25%, where no bactericidal activity was observed in the properdin-deficient serum. The opsonic activity was moderate (bacterial killing 15% after 30 min and 71% after 60 min). Serogroup A meningococci were efficiently opsonized by pre- and postimmune properdindeficient serum (Table 1). Addition of granulocytes did not increase the bactericidal effect
I
107
PMNb
99.2 PDs' 0 0 98.3 PDsd 0 99.3 99.7 99.4 "6 x lo6 bacteria were incubated with 4 x lo6 human granulocytes bfor 60 min at serum concentrations of '5% and d25%.
against serogroup A meningococci of chelated properdin-deficient or control serum (concentration 25%) (data not shown).
Uptake of radiolabelled meningococci by granulocytes The uptake of radiolabelled serogroup W-135 meningococci was reduced in unchelated properdin-deficient serum (concentration 25%) (Fig. 2). Alternative pathway-mediated uptake of radiolabelled bacteria was markedly reduced in properdin-deficient serum (concentration 25%). Addition of purified properdin restored the uptake in unchelated and MgEGTA-chelated, properdin-deficient serum, respectively. The uptake of radiolabelled serogroup A menin1000 7
PDs+P
I
3
5
NHs 50.
5[J)
\PNH, +PMN
5
1
15
-
1
30
Minutes
Fig. I . Opsonophagocytic killing of N. meningitidis serogroup W-135 in control serum (NHs) and properdin-deficient serum (PDs). About 6 x lo6 bacteria were incubated with 4 x lo6 granulocytes (PMN) at a serum concentration of 10%. Means and ranges for surviving bacteria are shown.
967
SODERSTROM et a[.
gococci was efficient in both properdin-deficient serum and pooled control serum whether sera were unchelated or MgEGTA-chelated. Intracellular killing In experiments with meningococci we found that the number of surviving granulocyte-associated meningococci was low, and there was no difference whether the granulocytes were ruptured in saponin or kept in PBS before plating. This indicated efficient intracellular killing of meningococci both in normal and properdindeficient serum. S. aureus was studied to provide more precise information on the influence of properdin-deficient serum on intracellular bacterial killing. Staphylococci are, in contrast to meningococci, rather resistant to phagocytic killing. Furthermore, separation of viable intracellular staphylococci is reliable with the use of lysostaphin to eliminate extracellular organisms (14). Intracellular killing of staphylococci was stimulated equally in properdin-deficient serum and control serum.
meningococci presensitized in :
P)
a
-n Q)
.-m>
*
1-
30
60
Minutes
Fig. 3. Influence of IgG on opsonophagocytic killing of serogroup W-135 meningococci. About 2.5 x lo7 bacteria were incubated with IgG, 3.1 mg and 12.5 mg, or buffer at 37°C for 15 min. After washings the bacteria were further incubated with Mg2+EGTAchelated control serum o and properdin-deficient serum A (concentration 25%) together with granulocytes. Viable bacteria were expressed as a percentage, using the number surviving in buffer alone as reference.
968
Influence of IgG on granulocyte-mediated killing Earlier studies did not indicate influence of IgG on bactericidal activity of chelated properdin-deficient serum on serogroup W- 135 meningococci unless properdin was added (23). However, granulocyte phagocytic killing was stimulated in a dose-dependent manner by addition of IgG (Fig. 3). Granulocyte chemotaxis A low chemotactic activity was generated by serogroup W-135 in properdin-deficient serum. With the addition of properdin, the deficient serum supported the generation of chemotactic activity equal to that of the control serum (Table 2). The chemotactic activity generated by serogroup A in properdin-deficient serum was efficient and similar to that of the control serum. Release of myeloperoxidase and lactoferrin The release of granulocyte myeloperoxidase and lactoferrin during phagocytosis did not differ between control serum and properdin-deficient serum. About 25% of myeloperoxidase and 15% of lactoferrin (of the total granulocyte content) TABLE 2. Chemotactic activity generated by N . meninnitidis seronroup A and serogroup W-13.5 Chemotactic index" SerogrouD A Unchelated serum MgEGTA-chelated serum NHSb 1.8 (1.6-1.9)' 2.0 (1.8-2.1) NHS+P" 1.9 (1.8-1.9) 1.9 (1.8-2.0) PDS' 1.8 (1.7-1.9) 1.7 (1.5-1.9) PDS + P 1.9 (1.9-2.0) 1.9 (1.7-2.0) Chemotactic index" serogroup W-135 Unchelated serum MgEGTA-chelated serum NHsb 1.5 (1.4-1.6)' 1.6 (1.5-1.8) NHs+P" 1.5 (1.4-1.6) 1.7 (1.6-1.8) PDs' 1.1 (0.9-1.2) 1.1 (1.1-1.2) PDs+P 1.6 (1.4-1.8) 1.7 (1.61.8) aChemotactic index (directed migrationirandom migration) of sera (serum concentration 50%) incubated with 1 x lo9 serogroup W-135 meningococci for 30 min in unchelated serum and 60 min in MgEGTAchelated serum bNormal human serum (pooled control) 'Mean values and ranges for triplicate determinations dProperdin (19 mg added) 'Properdin-deficient serum.
GRANULOCYTE FUNCTIONS AND N E I S S E R I A M E N I N G I T I D I S
were released during phagocytosis of serogroup A. When serogroup W-135 was studied, 35% of myeloperoxidase and 15% of lactoferrin were released. The spontaneous release of myeloperoxidase and lactoferrin (in buffer control) did not exceed 5%. DISCUSSION The association between complement deficiency and meningococcal disease has shown the importance of an intact complement system. The clinical manifestations of meningococcal disease differ between patients with defects in the terminal components compared to patients with properdin-defiency (1 5). The total lack of serum bactericidal activity, as seen in patients with C5C8 deficiency, clearly demonstrates the importance of serum bactericidal reactions. The clinical course of meningococcal disease in these patients is, however, usually mild, especially when compared to that in properdin-deficient patients (1 5). These findings suggest that the elimination of meningococci through granulocytes may be important. In the present study, we found that properdindeficient serum failed to support efficient granulocyte-mediated elimination (phagocytic killing, uptake of radiolabelled bacteria and generation of chemotactic activity) of N . meningitidis serogroup W-135. In contrast to the observations for serogroup W- 135 meningococci, the properdin-deficient serum was as efficient as the control serum against serogroup A meningococci. This efficient elimination was not affected by the presence of anticapsular antibodies as judged from the opsonophagocytic killing, that was similar in a preimmune properdin-deficient serum and in serum obtained after vaccination with serogroup A and C polysaccharide vaccine. It is possible that other antibodies, such as noncapsular antibodies, could contribute to the efficient opsonization or that the need for opsonic antibodies against serogroup A is less pronounced in the opsonophagocytic assay system used. It is also known that the surface structure of a microorganism is decisive for the activation of the alternative pathway (9). Serogroup A meningococci lack sialic acid as a repeating residue of
the capsule, in contrast to serogroup W-135 (4). In the absence of sialic acid, serogroup A act as a stronger activator of the alternative pathway (4).It is possible that the need for properdin under these circumstances is less pronounced. In a restrospective study of the relationship between meningococcal serogroups and the course of the disease in the Netherlands, the overall case fatality rate was 5.1% (21). The highest fatality rate was found for serogroup W-135 disease (18%) and the lowest for serogroup A (2.1%). As mentioned earlier, W-135 is the most frequent serogroup found in disease among properdin-deficient patients. Our study has shown that properdin is needed for efficient elimination of serogroup W-135. This may partly explain the relationship between properdin-deficiency and serogroup W- 135 disease. In a previous study we have shown that vaccine-induced anti-capsular antibodies against serogroup W- 135 enhance both bactericidal and granulocyte-mediated elimination (24). IgG antibodies also enhance alternative pathwaymediated bactericidal activity but only in the presence of properdin (23). In this study we found that IgG enhanced the phagocytic killing by a properdin-independent mechanism. The increase in phagocytic killing could be due to Fcmediated phagocytosis (17) as well as to enhanced deposition of opsonic C3 fragments (17) on the bacterial surface. Our study indicated rapid killing of meningococci following ingestion by phagocytes, as shown by others (3, 16). It has been reported that intracellular killing of some bacteria is enhanced by immunoglobulins and complement in the surrounding medium (1 1). We found no evidence that properdin-deficient serum differed from normal serum when intracellular killing of S. aureus was studied. Phagocytosis is associated with release of granulocyte components mostly from secondary (lactoferrin) but also from primary granules (myeloperoxidase) (10). Both increased and decreased release may influence the inflammatory reaction and the course of disease. In previous experiments we have observed reduced release of lactoferrin and myeloperoxidase during phagocytosis of Streptococcus pneumoniae in properdin-deficient serum (unpublished data). However, we found that properdin-deficient serum was as efficient as control serum in in969
SODERSTROMe l a[.
ducing release by the granule components during phagocytosis of meningococci. Previous studies of defense mechanisms against N. meningitidis in properdin deficiency have focused on serum bactericidal reactions, with results that cannot fully explain the association of the deficiency state with meningococcal disease (2, 23). The present findings suggest that impaired granulocyte functions related to opsonization and chemotaxis of serogroup W-135 meningococci might contribute to the clinical findings.
complement pathway activation. Infect. Immun. 55: 174-180, 1987. 10. Leffell, M . S. & Spitznagel, J. K.: Fate of hu-
11.
2. This work was supported by grants from the Crafoord Foundation in Lund, the Royal Physiographical Society in Lund, the Swedish Society of Medicine and the Swedish Medical Research Council (B89-16x47921). We thank Dr Tor Olofsson of the Department of Internal Medicine, Research Laboratory for Clinical Hematology at the University of Lund for valuable help with determinations of lactoferrin and myeloperoxidase.
3.
14.
REFERENCES 1. Braconier, J. H . , Odeberg, H. & Sjoholm, A. G.: Granulocyte phagocytosis of Streptococcus pneumoniae in properdin deficient serum. Infect. Immun. 40: 219-224, 1983. 2. Densen, I?, Weiler, J. M., G r i f f s , J. M . & Hoffmann, L. G. : Familial properdin deficiency and fatal meningococcemia. Correction of the bactericidal effect by vaccination. N. Engl. J. Med. 316: 922-926, 1987. 3. DeVoe, I. W : Egestion of degraded meningococci by polymorphonuclear leukocytes. J. Bacteriol. 125: 258-266, 1976. 4. Di Ninno, J. L. & Chenier, V K.: Activation of complement by Neisseria meningitidis. FEMS Microbiol. Lett. 12: 55-60, 1981. 5. Fearon, D. T & Austen, K. E : Properdin: Binding to C3b and stabilization of the C3b-dependent C3 convertase. J. Exp. Med. 142: 856-863, 1975. 6. Forsgren, A. & Quie, I? G.: Opsonic activity in human serum chelated with ethylene glycoltetraacetic acid. Immunol. 26: 1251-1256, 1974. 7. Forsgren, A. & Schmeling, D.: Effect of antibiotics on chemotaxis of human leukocytes. J. Antimicrob. Agents. Chemother. 11: 580-584, 1977. 8. Goldschneider, I., Gotschlich, E. C. & Artenstein, M . S.: Human immunity to the meningococcus. I. The role of humoral antibodies. J. Exp. Med. 129: 1307-1326, 1969. 9. Jarvis, G. A . & Vedros, N. A.: Sialic acid of group B Neisseria meningitidis regulates alternative
970
15.
16.
17.
18.
19.
20.
21.
man lactoferrin and myeloperoxidase in phagocytizing human neutrophils: Effects of immunoglobulin G subclasses and immune complexes coated on latex beads. Infect. Immun. 12: 813-820, 1975. Leijh, P C. J., van den Barselaar, M . T , van Zwet, T L., Daha, R. & van Furth, R.: Requirement of extracellular complement and immunoglobulin for intracellular killing of microorganisms by human monocytes. J. Clin. Invest. 63: 772-784, 1979. Nelson, R., Quie, P & Simmons, R.: Chemotaxis under agarose: a new and simple method for measuring chemotaxis and spontaneous migration of human polymorphonuclear leukocytes and monocytes. J. Immunol. 115: 1650-1656, 1975. Olsson, I., Olofsson, T , Ohlsson, K . & Gustavsson, A,: Serum and plasma myeloperoxidase, elastase and lactoferrin content in acute myeloid leukaemia. Scand. J. Haematol. 22: 397406, 1979. Rollof: J., Braconier, J. H . , Soderstrom, C. & Nilsson-Ehle, P: Interference of Staphylococcus aureus lipase with human granulocyte function. Eur. J. Clin. Microbiol. Infect. Dis. 7: 505-510, 1988. Ross, S. C. & Densen, P: Complement deficiency states and infection: epidemiology, pathogenesis and consequences of Neisserial and other infections in an immune deficiency. Medicine 63: 243-273, 1984. Ross, S. C., Rosenthal, I? J., Berberich, H. M. & Densen, P: Killing of Neisseria meningitidis by human neutrophils: Implications for normal and complement-deficient individuals. J. Infect. Dis. 155: 1266-1275, 1987. Schribner, D. J. & Fahrney, D.: Neutrophil receptors for IgG and complement: their roles in the attachment and ingestion phases of phagocytosis. J. Immunol. 116: 892-897, 1976. Sjoholm, A. G., Braconier, J. H. & Soderstrom, C.: Properdin deficiency in a family with fulminant meningococcal infections. Clin. Exp. Immunol. 50: 291-297, 1982. Sjoholm, A. G., Kuijper, E. J., Tijssen, C. C., Jansz, A , , Bol, I?, Spanjaard, L. & Zanen, H. C.: Dysfunctional properdin in a Dutch family with meningococcal disease. N. Engl. J. Med. 319: 33-37, 1988. Sjoholm, A . G., Soderstrom, C. & Nilsson, L-A.: A second variant of properdin deficiency: the detection of properdin at low concentrations in affected males. Complement 5: 130-140, 1988. Spanjaard, L., Bol, P,DeMarie, S. & Zanen, H. C.: Association of meningococcal serogroups and types with the course of disease, the Netherlands
GRANULOCYTE FUNCTIONS AND NEISSERIA MENINGITIDIS
1959-1983 (n= 1221). In: Poolman, J. T , Zanen, H. C., Meyer, 7:F , Heckels, J. E., Makela, P H., Smith, H. & Beuvery, E. C. (Eds.): Gonococci and meningococci. Martinus Nijhoff. Dordrecht 1988, pp. 175-180. 22. Soderstrom, C., Braconier, J. H., Christensen, K. K . , Christensen, 19 & Sjoholm, A . G.: Opsonization of group B streptococci in properdin deficient serum. Acta path. microbiol. immunol. scand. Sect. C. 93: 251-256, 1985. 23. Soderstrom, C., Braconier, J. H., Danielsson, D. & Sjoholm, A . G.: Bactericidal activity for Neisseria
meningitidis in properdin-deficient sera. J. Infect. Dis. 156: 107-112, 1987. 24. Soderstrom, C., Braconier, J. H., Kayhty, H., Sjoholm, A . G. & Thuresson, B.: Immune response to tetravalent meningococcal vaccine: opsonic and bactericidal functions of normal and properdin deficient sera. Eur. J. Clin. Microbiol. Infect. Dis. 8: 220-224, 1989. 25. Watt, 19 J.: The fate of gonococci in polymorphonuclear leucocytes. J. Med. Microbiol. 3: 501-509, 1980.
97 1
Granulocyte functions and Neisseria rnenirzgitiaids: influence of properdindeficient serum CLAES SODERSTROM', JEAN HENRIK BRACONIER', ANDERS G. SJOHOLM', and BRITT THURESSON' Department of Infectious Diseases' and Department of Medical Microbiology', University of Lund, Lund, Sweden
Soderstrom, C., Braconier, J. H., Sjoholm, A. G. & Thuresson, B. Granulocyte functions and Neisseria meningitidis: influence of properdin-deficient serum. APMIS 99: 965-971, 1991. Granulocyte-mediated reactions such as opsonization, chemotaxis, and release of granulocyte myeloperoxidase and lactoferrin were studied in properdin-deficient and normal human serum incubated with serogroup A and W-135 meningococci. There were no differences between the sera when serogroup A meningococci were studied. Opsonic and chemotactic activity were impaired against serogroup W-135 meningococci in properdin-deficient serum. Restitution with properdin restored both activities. We found similar release of myeloperoxidase and lactoferrin from granulocytes challenged with serogroup A or W-135 meningococci in either sera. These findings are in accordance with the clinical observations of meningococcal infections caused by serogroup W- 135 in properdindeficient patients as well as the absence of infections caused by serogroup A meningococci. Key words: Properdin deficiency; Neisseria meningitidis; opsonophagocytic killing; chemotactic activity. Claes Soderstrom, Department of Infectious Diseases, University Hospital of Lund, S-221 85 Lund, Sweden.
T h e complement component properdin stabilizes the alternative pathway C 3 convertase C3bBb (5). Inherited deficiency or dysfunction of properdin predisposes t o meningococcal disease (2, 15, 18-20). Several C3-dependent functions a r e impaired in properdin-deficient sera (1, 2, 18, 22-24). The meningococcal serogroups B, C, Y a n d W-I35 have been identified in patients with deficiency o r dysfunction of properdin (2, 18, 19, 24). Serogroup W-135 accounts for about 50% of meningococcal infections among properdindeficient patients (24). Infection due to serogroup A has so far not been reported in these pa tien ts. Serum bactericidal reactions require antibody
Received July 26, 1989. Accepted March 1 1 , 1991.
and a n intact complement system and are probably of predominant importance in defense against Neisseria meningitidis (8). There is also evidence for a role of granulocytes in defense against meningococcal disease (3, 16). In the present investigation, the effect of properdindeficient serum o n granulocyte-mediated reactions against serogroup A a n d W-135 meningococci was studied.
MATERIALS A N D METHODS Bacteria N. meningitidis serogroup A (strain A-3269), serogroup W-135 (strain W-7911) and Staphylococcus aureus (strain 502 A) were used (14, 23). The serogroup W-135 strain was isolated from a patient with properdin deficiency (23). The bacteria were stored and cultured as previously described (14, 23). For radiolabelling, the bacteria were grown overnight in
965
SUDERSTRUM et a!.
Muller-Hinton broth (23). One ml from the overnight culture was added to 8 ml of fresh Miiller-Hinton broth supplemented with 16 pCi D-(U'4C)glucose (Amersham International, England). The meningococci were harvested after five h growth, and were then washed three times in isotonic veronal buffered saline with 0.1% gelatin, 0.15 mM Ca2+, 0.5 mM Mg2+ and 0.2% glucose (wt/vol) (GVB). The meningococci were adjusted to about 8 x lo7viable organisms per ml in GVB. Sera Pooled normal human serum from 20 healthy blood donors was used as a control. The properdindeficient serum was from a healthy adult and was obtained before and one month after immunization with a meningococcal (polysaccharide A and C) vaccine, respectively (23). The sera were stored in aliquots at - 80°C. The properdin-deficient and the control serum showed no antibodies against serogroup W-135 as determined by ELISA (24). The control and preimmune properdin-deficient serum had similar levels of serogroup A antibodies (23). Alternative pathway-mediated reactions were studied in sera chelated with 10 mM EGTA (ethylene glycol-tetraacetic acid) (6) and reconstituted with 4 mM Mgz+. Granulocytes Human granulocytes were prepared from peripheral blood by using Isopaque-Ficoll (Pharmacia Fine Chemicals, Uppsala, Sweden). The bottom layer was mixed with 0.87% NH,Cl for lysis of red cells, washed twice, and suspended in GVB. Purified properdin Properdin was purified as described earlier (18) and added at a physiological concentration (19 mg/l) in restitution experiments. Phagocytic killing About 4 x lo6 granulocytes and 6 x lo6 bacteria were incubated during agitation at 37°C with serum at various concentrations (total volume 0.25 ml). Aliquots of 20 pl were removed at timed intervals, diluted in sterile distilled water with 5% saponin (wt/ vol) at+4"C for five min in order to rupture the granulocytes (25) and then further diluted in phosphate-buffered saline (PBS) (0.03 M sodium phosphate, 0.12 M NaCI, pH 7.2). The total number of viable meningococci was determined by colony counting using the pour-plate method. All assays were performed in duplicate. Commercial IgG, 50 g/l, (Gammonativo Kabi, Sweden) was used as a source of anti-meningococcal antibody (23). Bacterial suspensions (6 x lo7 bacterialiml) were preincubated for 15 min at 37°C with equal volumes (0.25 ml) of IgG at 50 and 12.5 g/l, or with buffer alone. After washing, the bacteria were used in the opsonophagocytic
966
assay together with Mg2+EGTA-chelatedserum (concentration 25%). Uptake of radiolabelled bacteria About 2 x lo7 radiolabelled bacteria were agitated with 4 x lo6 human granulocytes in GVB-diluted serum at 37°C. The sera were used at final concentrations of 5-25"?' in a total volume of 1.0 ml. At different intervals, aliquots of 0.3 ml were transferred to 1.O ml ice-cold GVB. After centrifugation (850 x g , seven min, 4°C) the supernatants were transferred to 10 ml Insta-gel (Packard, Downers Grove, Ill., USA). The granulocytes were separated from extracellular meningococci by differential centrifugation (three times, 80 x g, seven min, + 4°C). The final pellet was dissolved in 0.2 ml Soluene-350 (Packard, Downers Grove, Ill., USA), and transferred to 10 ml Insta-gel. After 24 h at+4"C, radioactivity was counted in a scintillation counter (Packard Tri-Carb 460 C). The uptake of radiolabelled bacteria was expressed as a percentage of the total radioactivity in each incubation mixture.
+
Intracellular killing S. aureus 502 A (about 2 x 10') were preincubated with 50% control serum for three min in a total volume of 2.0 ml. After washing, the opsonized staphylococci (about 8 x lo7) were incubated with granulocytes (about 5 x lo6) at 37°C for 15 rnin in a total volume of 1.8 ml. Lysostaphin (48 IU) in ice-cold buffer was added (final volume 3 ml) to eliminate extracellular S. aureus (14). After one min, lysostaphin was removed by dilution and repeated washings at +4"C. The washed granulocytes (about 6 x lo5) were incubated in buffer or serum at a concentration of 36% in a total volume of 0.25 ml. Surviving bacteria were quantified after lysis of the granulocytes by viable counts using the pour-plate method. Granulocyte chemotaxis Chemotaxis was studied by the under-agarose technique (12) as modified by Forsgren & Schmeling (7). Sera at a concentration of 50% were incubated with meningococci (1 x 109/ml)at 37°C for 30 rnin (unchelated serum) and 60 rnin (Mg2+EGTA-chelated serum). Bacteria were removed by centrifugation. The supernatant sera were heat treated (56°C 30 min) and used as chemoattractants. The culture dishes were incubated for three h at 37°C in a humidified atmosphere with 5% C02. The chemotaxis was quantified by measuring the leading front of a greatly enlarged projection of the migration patterns. Results were expressed as a chemotactic index, i.e. directed migration / random migration. Release of myeloperoxidase and lactoferrin during phagocytosis About 4 x lo7 meningococci were incubated with 4 x lo6human granulocytes and serum at 37°C during
GRANULOCYTE FUNCTIONS AND NEISSERIA MENINGITIDIS
agitation (total volume 0.5 ml). Aliquots of 0.2 ml were removed after 30 min and granulocytes were separated by centrifugation (850 x g, seven min, 4"C), and the content of myeloperoxidase and lactoferrin in the supernatants was determined by radioimmunoassay (1 3).
+
TABLE 1. Opsonophagocytic killing of serogroup A meningococci" in properdin-deficient serum (PDs) obtained before and four weeks after vaccination with serogroup A and C polysaccharides Bacterial killing. (YO) Before vaccination After vaccination Serum Serum+ Serum Serum+
PMNb
RESULTS
Opsonophagocy t ic killing Moderately reduced phagocytic killing of serogroup W-135 was found in the presence of properdin-deficient serum as compared with control serum at concentrations of 10%. Due to the bactericidal activity of serum, it was difficult to assess the contribution of phagocytic killing (Fig. 1). Opsonic activity mediated through the alternative pathway was studied with Mg2+ EGTA at serum concentrations of 25%, where no bactericidal activity was observed in the properdin-deficient serum. The opsonic activity was moderate (bacterial killing 15% after 30 min and 71% after 60 min). Serogroup A meningococci were efficiently opsonized by pre- and postimmune properdindeficient serum (Table 1). Addition of granulocytes did not increase the bactericidal effect
I
107
PMNb
99.2 PDs' 0 0 98.3 PDsd 0 99.3 99.7 99.4 "6 x lo6 bacteria were incubated with 4 x lo6 human granulocytes bfor 60 min at serum concentrations of '5% and d25%.
against serogroup A meningococci of chelated properdin-deficient or control serum (concentration 25%) (data not shown).
Uptake of radiolabelled meningococci by granulocytes The uptake of radiolabelled serogroup W-135 meningococci was reduced in unchelated properdin-deficient serum (concentration 25%) (Fig. 2). Alternative pathway-mediated uptake of radiolabelled bacteria was markedly reduced in properdin-deficient serum (concentration 25%). Addition of purified properdin restored the uptake in unchelated and MgEGTA-chelated, properdin-deficient serum, respectively. The uptake of radiolabelled serogroup A menin1000 7
PDs+P
I
3
5
NHs 50.
5[J)
\PNH, +PMN
5
1
15
-
1
30
Minutes
Fig. I . Opsonophagocytic killing of N. meningitidis serogroup W-135 in control serum (NHs) and properdin-deficient serum (PDs). About 6 x lo6 bacteria were incubated with 4 x lo6 granulocytes (PMN) at a serum concentration of 10%. Means and ranges for surviving bacteria are shown.
967
SODERSTROM et a[.
gococci was efficient in both properdin-deficient serum and pooled control serum whether sera were unchelated or MgEGTA-chelated. Intracellular killing In experiments with meningococci we found that the number of surviving granulocyte-associated meningococci was low, and there was no difference whether the granulocytes were ruptured in saponin or kept in PBS before plating. This indicated efficient intracellular killing of meningococci both in normal and properdindeficient serum. S. aureus was studied to provide more precise information on the influence of properdin-deficient serum on intracellular bacterial killing. Staphylococci are, in contrast to meningococci, rather resistant to phagocytic killing. Furthermore, separation of viable intracellular staphylococci is reliable with the use of lysostaphin to eliminate extracellular organisms (14). Intracellular killing of staphylococci was stimulated equally in properdin-deficient serum and control serum.
meningococci presensitized in :
P)
a
-n Q)
.-m>
*
1-
30
60
Minutes
Fig. 3. Influence of IgG on opsonophagocytic killing of serogroup W-135 meningococci. About 2.5 x lo7 bacteria were incubated with IgG, 3.1 mg and 12.5 mg, or buffer at 37°C for 15 min. After washings the bacteria were further incubated with Mg2+EGTAchelated control serum o and properdin-deficient serum A (concentration 25%) together with granulocytes. Viable bacteria were expressed as a percentage, using the number surviving in buffer alone as reference.
968
Influence of IgG on granulocyte-mediated killing Earlier studies did not indicate influence of IgG on bactericidal activity of chelated properdin-deficient serum on serogroup W- 135 meningococci unless properdin was added (23). However, granulocyte phagocytic killing was stimulated in a dose-dependent manner by addition of IgG (Fig. 3). Granulocyte chemotaxis A low chemotactic activity was generated by serogroup W-135 in properdin-deficient serum. With the addition of properdin, the deficient serum supported the generation of chemotactic activity equal to that of the control serum (Table 2). The chemotactic activity generated by serogroup A in properdin-deficient serum was efficient and similar to that of the control serum. Release of myeloperoxidase and lactoferrin The release of granulocyte myeloperoxidase and lactoferrin during phagocytosis did not differ between control serum and properdin-deficient serum. About 25% of myeloperoxidase and 15% of lactoferrin (of the total granulocyte content) TABLE 2. Chemotactic activity generated by N . meninnitidis seronroup A and serogroup W-13.5 Chemotactic index" SerogrouD A Unchelated serum MgEGTA-chelated serum NHSb 1.8 (1.6-1.9)' 2.0 (1.8-2.1) NHS+P" 1.9 (1.8-1.9) 1.9 (1.8-2.0) PDS' 1.8 (1.7-1.9) 1.7 (1.5-1.9) PDS + P 1.9 (1.9-2.0) 1.9 (1.7-2.0) Chemotactic index" serogroup W-135 Unchelated serum MgEGTA-chelated serum NHsb 1.5 (1.4-1.6)' 1.6 (1.5-1.8) NHs+P" 1.5 (1.4-1.6) 1.7 (1.6-1.8) PDs' 1.1 (0.9-1.2) 1.1 (1.1-1.2) PDs+P 1.6 (1.4-1.8) 1.7 (1.61.8) aChemotactic index (directed migrationirandom migration) of sera (serum concentration 50%) incubated with 1 x lo9 serogroup W-135 meningococci for 30 min in unchelated serum and 60 min in MgEGTAchelated serum bNormal human serum (pooled control) 'Mean values and ranges for triplicate determinations dProperdin (19 mg added) 'Properdin-deficient serum.
GRANULOCYTE FUNCTIONS AND N E I S S E R I A M E N I N G I T I D I S
were released during phagocytosis of serogroup A. When serogroup W-135 was studied, 35% of myeloperoxidase and 15% of lactoferrin were released. The spontaneous release of myeloperoxidase and lactoferrin (in buffer control) did not exceed 5%. DISCUSSION The association between complement deficiency and meningococcal disease has shown the importance of an intact complement system. The clinical manifestations of meningococcal disease differ between patients with defects in the terminal components compared to patients with properdin-defiency (1 5). The total lack of serum bactericidal activity, as seen in patients with C5C8 deficiency, clearly demonstrates the importance of serum bactericidal reactions. The clinical course of meningococcal disease in these patients is, however, usually mild, especially when compared to that in properdin-deficient patients (1 5). These findings suggest that the elimination of meningococci through granulocytes may be important. In the present study, we found that properdindeficient serum failed to support efficient granulocyte-mediated elimination (phagocytic killing, uptake of radiolabelled bacteria and generation of chemotactic activity) of N . meningitidis serogroup W-135. In contrast to the observations for serogroup W- 135 meningococci, the properdin-deficient serum was as efficient as the control serum against serogroup A meningococci. This efficient elimination was not affected by the presence of anticapsular antibodies as judged from the opsonophagocytic killing, that was similar in a preimmune properdin-deficient serum and in serum obtained after vaccination with serogroup A and C polysaccharide vaccine. It is possible that other antibodies, such as noncapsular antibodies, could contribute to the efficient opsonization or that the need for opsonic antibodies against serogroup A is less pronounced in the opsonophagocytic assay system used. It is also known that the surface structure of a microorganism is decisive for the activation of the alternative pathway (9). Serogroup A meningococci lack sialic acid as a repeating residue of
the capsule, in contrast to serogroup W-135 (4). In the absence of sialic acid, serogroup A act as a stronger activator of the alternative pathway (4).It is possible that the need for properdin under these circumstances is less pronounced. In a restrospective study of the relationship between meningococcal serogroups and the course of the disease in the Netherlands, the overall case fatality rate was 5.1% (21). The highest fatality rate was found for serogroup W-135 disease (18%) and the lowest for serogroup A (2.1%). As mentioned earlier, W-135 is the most frequent serogroup found in disease among properdin-deficient patients. Our study has shown that properdin is needed for efficient elimination of serogroup W-135. This may partly explain the relationship between properdin-deficiency and serogroup W- 135 disease. In a previous study we have shown that vaccine-induced anti-capsular antibodies against serogroup W- 135 enhance both bactericidal and granulocyte-mediated elimination (24). IgG antibodies also enhance alternative pathwaymediated bactericidal activity but only in the presence of properdin (23). In this study we found that IgG enhanced the phagocytic killing by a properdin-independent mechanism. The increase in phagocytic killing could be due to Fcmediated phagocytosis (17) as well as to enhanced deposition of opsonic C3 fragments (17) on the bacterial surface. Our study indicated rapid killing of meningococci following ingestion by phagocytes, as shown by others (3, 16). It has been reported that intracellular killing of some bacteria is enhanced by immunoglobulins and complement in the surrounding medium (1 1). We found no evidence that properdin-deficient serum differed from normal serum when intracellular killing of S. aureus was studied. Phagocytosis is associated with release of granulocyte components mostly from secondary (lactoferrin) but also from primary granules (myeloperoxidase) (10). Both increased and decreased release may influence the inflammatory reaction and the course of disease. In previous experiments we have observed reduced release of lactoferrin and myeloperoxidase during phagocytosis of Streptococcus pneumoniae in properdin-deficient serum (unpublished data). However, we found that properdin-deficient serum was as efficient as control serum in in969
SODERSTROMe l a[.
ducing release by the granule components during phagocytosis of meningococci. Previous studies of defense mechanisms against N. meningitidis in properdin deficiency have focused on serum bactericidal reactions, with results that cannot fully explain the association of the deficiency state with meningococcal disease (2, 23). The present findings suggest that impaired granulocyte functions related to opsonization and chemotaxis of serogroup W-135 meningococci might contribute to the clinical findings.
complement pathway activation. Infect. Immun. 55: 174-180, 1987. 10. Leffell, M . S. & Spitznagel, J. K.: Fate of hu-
11.
2. This work was supported by grants from the Crafoord Foundation in Lund, the Royal Physiographical Society in Lund, the Swedish Society of Medicine and the Swedish Medical Research Council (B89-16x47921). We thank Dr Tor Olofsson of the Department of Internal Medicine, Research Laboratory for Clinical Hematology at the University of Lund for valuable help with determinations of lactoferrin and myeloperoxidase.
3.
14.
REFERENCES 1. Braconier, J. H . , Odeberg, H. & Sjoholm, A. G.: Granulocyte phagocytosis of Streptococcus pneumoniae in properdin deficient serum. Infect. Immun. 40: 219-224, 1983. 2. Densen, I?, Weiler, J. M., G r i f f s , J. M . & Hoffmann, L. G. : Familial properdin deficiency and fatal meningococcemia. Correction of the bactericidal effect by vaccination. N. Engl. J. Med. 316: 922-926, 1987. 3. DeVoe, I. W : Egestion of degraded meningococci by polymorphonuclear leukocytes. J. Bacteriol. 125: 258-266, 1976. 4. Di Ninno, J. L. & Chenier, V K.: Activation of complement by Neisseria meningitidis. FEMS Microbiol. Lett. 12: 55-60, 1981. 5. Fearon, D. T & Austen, K. E : Properdin: Binding to C3b and stabilization of the C3b-dependent C3 convertase. J. Exp. Med. 142: 856-863, 1975. 6. Forsgren, A. & Quie, I? G.: Opsonic activity in human serum chelated with ethylene glycoltetraacetic acid. Immunol. 26: 1251-1256, 1974. 7. Forsgren, A. & Schmeling, D.: Effect of antibiotics on chemotaxis of human leukocytes. J. Antimicrob. Agents. Chemother. 11: 580-584, 1977. 8. Goldschneider, I., Gotschlich, E. C. & Artenstein, M . S.: Human immunity to the meningococcus. I. The role of humoral antibodies. J. Exp. Med. 129: 1307-1326, 1969. 9. Jarvis, G. A . & Vedros, N. A.: Sialic acid of group B Neisseria meningitidis regulates alternative
970
15.
16.
17.
18.
19.
20.
21.
man lactoferrin and myeloperoxidase in phagocytizing human neutrophils: Effects of immunoglobulin G subclasses and immune complexes coated on latex beads. Infect. Immun. 12: 813-820, 1975. Leijh, P C. J., van den Barselaar, M . T , van Zwet, T L., Daha, R. & van Furth, R.: Requirement of extracellular complement and immunoglobulin for intracellular killing of microorganisms by human monocytes. J. Clin. Invest. 63: 772-784, 1979. Nelson, R., Quie, P & Simmons, R.: Chemotaxis under agarose: a new and simple method for measuring chemotaxis and spontaneous migration of human polymorphonuclear leukocytes and monocytes. J. Immunol. 115: 1650-1656, 1975. Olsson, I., Olofsson, T , Ohlsson, K . & Gustavsson, A,: Serum and plasma myeloperoxidase, elastase and lactoferrin content in acute myeloid leukaemia. Scand. J. Haematol. 22: 397406, 1979. Rollof: J., Braconier, J. H . , Soderstrom, C. & Nilsson-Ehle, P: Interference of Staphylococcus aureus lipase with human granulocyte function. Eur. J. Clin. Microbiol. Infect. Dis. 7: 505-510, 1988. Ross, S. C. & Densen, P: Complement deficiency states and infection: epidemiology, pathogenesis and consequences of Neisserial and other infections in an immune deficiency. Medicine 63: 243-273, 1984. Ross, S. C., Rosenthal, I? J., Berberich, H. M. & Densen, P: Killing of Neisseria meningitidis by human neutrophils: Implications for normal and complement-deficient individuals. J. Infect. Dis. 155: 1266-1275, 1987. Schribner, D. J. & Fahrney, D.: Neutrophil receptors for IgG and complement: their roles in the attachment and ingestion phases of phagocytosis. J. Immunol. 116: 892-897, 1976. Sjoholm, A. G., Braconier, J. H. & Soderstrom, C.: Properdin deficiency in a family with fulminant meningococcal infections. Clin. Exp. Immunol. 50: 291-297, 1982. Sjoholm, A. G., Kuijper, E. J., Tijssen, C. C., Jansz, A , , Bol, I?, Spanjaard, L. & Zanen, H. C.: Dysfunctional properdin in a Dutch family with meningococcal disease. N. Engl. J. Med. 319: 33-37, 1988. Sjoholm, A . G., Soderstrom, C. & Nilsson, L-A.: A second variant of properdin deficiency: the detection of properdin at low concentrations in affected males. Complement 5: 130-140, 1988. Spanjaard, L., Bol, P,DeMarie, S. & Zanen, H. C.: Association of meningococcal serogroups and types with the course of disease, the Netherlands
GRANULOCYTE FUNCTIONS AND NEISSERIA MENINGITIDIS
1959-1983 (n= 1221). In: Poolman, J. T , Zanen, H. C., Meyer, 7:F , Heckels, J. E., Makela, P H., Smith, H. & Beuvery, E. C. (Eds.): Gonococci and meningococci. Martinus Nijhoff. Dordrecht 1988, pp. 175-180. 22. Soderstrom, C., Braconier, J. H., Christensen, K. K . , Christensen, 19 & Sjoholm, A . G.: Opsonization of group B streptococci in properdin deficient serum. Acta path. microbiol. immunol. scand. Sect. C. 93: 251-256, 1985. 23. Soderstrom, C., Braconier, J. H., Danielsson, D. & Sjoholm, A . G.: Bactericidal activity for Neisseria
meningitidis in properdin-deficient sera. J. Infect. Dis. 156: 107-112, 1987. 24. Soderstrom, C., Braconier, J. H., Kayhty, H., Sjoholm, A . G. & Thuresson, B.: Immune response to tetravalent meningococcal vaccine: opsonic and bactericidal functions of normal and properdin deficient sera. Eur. J. Clin. Microbiol. Infect. Dis. 8: 220-224, 1989. 25. Watt, 19 J.: The fate of gonococci in polymorphonuclear leucocytes. J. Med. Microbiol. 3: 501-509, 1980.
97 1
