THE JOURNAL OF INFECTIOUS DISEASES. VOL. 138, NO.3. SEPTEMBER 1978 © 1978 by The University of Chicago. 0022-1899/78/3803-0010$00.81
Participation of Immunoglobulin and the Alternative Complement Pathway in Opsonization of Bacteroides fragilis and Bacteroides thetaiotaomicron From the Departments of Medicine and Microbiology, University of Cincinnati College of Medicine, Cincinnati, Ohio
Ann B. Bjornson and H. Stephen Bjornson
Studies during the past decade have established that erythrocytes and pneumococci can be opsonized by the sequential reactions of antibody and classical complement components Cl, C4, C2, and C3 [1, 2]. The participation of an alternative complement pathway in opsonization of Streptococcus pneumoniae, Staphylococcus aureus, Proteus mirabilis, Serratia marcescens, Escherichia coli, and Pseudomonas aeruginosa has also been suggested by several studies [3-8]. Received for publication December 27, 1977, and in revised form March 8,1978. This work was supported by grant no. AI-13688 from the U.S. Public Health Service and by contract no. DAMD-17-76-C-6023 from the U.S. Army Medical Research and Development Command. The authors thank Dr. Michael Frank, National Institutes of Health, Bethesda, Maryland, Dr. George F. Brooks, Veterans Administration Hospital, Martinez, California, and Dr. Beatrice Lampkin, Children's Hospital Research Foundation, Cincinnati, Ohio, for the C4-deficient guinea pigs, C8-deficient human serum, and hypogammaglobulinemic human sera, respectively. We also thank Barbara Kitko and Constance Zellner for their technical assistance. Please address requests for reprints to Dr. Ann B. Bjornson, Division of Infectious Diseases, Department of Medicine, University of Cincinnati College of Medicine, 231 Bethesda Avenue, Cincinnati, Ohio 45267.
However, minimal information is available regarding specific humoral and cellular host defense mechanisms against gram-negative, anaerobic microorganisms. Casciato et al. demonstrated that, under aerobic conditions, clinical isolates of Bacteroides fragilis and Bacteroides thetaiotaomicron were not susceptible to the bacteriolytic activity of human serum [9]. Bjornson et al. subsequently showed that, under anaerobic conditions, strains of B. fragilis and B. thetaiotaomicron were phagocytosed and killed intracellularly by human polymorphonuclear leukocytes in the presence of pooled normal human serum (PNHS) but not by either leukocytes or serum alone [10]. Similar results were obtained when the experiments were carried out in an aerobic environment, with the exception that B. fragilis was phagocytosed and killed intracellularly by leukocytes to some extent in the absence of serum. The present investigation was undertaken to determine the requirements for immunoglobulin and complement in phagocytosis and intracellular killing of strains of Bacteroides by leukocytes. All of the experiments to be described were carried out under anaerobic conditions [10].
351
Downloaded from http://jid.oxfordjournals.org/ at UQ Library on July 28, 2015
Studies were conducted to determine the requirements for immunoglobulin and complement for opsonization of Bacteroides fragilis and Bacteroides thetaiotaomicron. The ability of human sera depleted of immunoglobulin or complement components to promote phagocytosis and intracellular killing of the strains of Bacteroides by human leukocytes was measured in vitro under anaerobic conditions. Neither hypogammaglobulinemic sera nor pooled normal human serum (PNHS) heated at 56 C for 30 min supported phagocytosis and killing of the strains of Bacteroides. Sera depleted of terminal complement components by treatment with inulin or cobra venom factor and C8-deficient human serum did not support phagocytosis of the test strains. PNHS depleted of C3, factor B, or factor D also did not support phagocytosis of either strain. Dose-dependent restoration of the opsonic activity of factor B-depleted serum was accomplished by purified human factor B but not by human C2. The results indicated that immunoglobulin and components of the alternative complement pathway participate in opsonization of the strains of Bacteroides tested in this study.
352
Materials and Methods
C and then transferred to a tube of deoxygenated thioglycolate broth that was supplemented with 0.5% yeast extract and 1% glucose and incubated at 37 C overnight. The bacteria were washed once and diluted in deoxygenated diluent immediately before the experiments. Leukocytes were prepared from the plasma of dextransedimented, heparinized blood obtained from healthy adult volunteers. The leukocyte-rich plasma was divided and centrifuged at 200 g for 10 min, and the supernatants were discarded. The leukocytes were washed twice and resuspended in deoxygenated diluent. Various combinations of serum (10% concentration), leukocytes (5 X 106 ) , bacteria (1.0 X 106) , and diluent in a final volume of 1 ml were added together in plastic-capped tubes. The reagents were added to the tubes in an anaerobic glove box (Coy Manufacturing, Ann Arbor, Mich.). In addition, the tubes to be used were equilibrated overnight in the glove box, and the sera were equilibrated in the glove box on ice for 1 hr before their addition to the respective reaction mixtures. The internal atmosphere of the glove box contained 85% nitrogen, 10% hydrogen, 5% carbon dioxide, and not >25 ppm of oxygen. The pH and p02 (oxygen pressure) in each of the different reaction mixtures were measured before incubation at 37 C. The mean pH of the mixtures (±SE) was 6.95 ± 0.05, and the p02 ranged from 19 to 25 mm Hg. The next steps in the procedure were carried out entirely in the glove box. The tubes were incubated for 3 hr at 37 C on a rotating platform, and samples were removed at zero-time and at 30,90, and 180 min. The samples were diluted in deoxygenated distilled water, and the dilutions were plated in thioglycolate agar by the pourplate method. The plates from all experiments were incubated anaerobically in the glove box for four days at 37 C, and the colonies were enumerated to determine the total number of bacteria surviving in each reaction mixture. In preliminary experiments, surviving extracellular and total bacteria were measured in reaction mixtures containing serum, leukocytes, and bacteria, or leukocytes and bacteria alone. Extracellular counts were found to be similar if not identical to total counts, a finding which indicated that phagocytosis was rapidly followed
Downloaded from http://jid.oxfordjournals.org/ at UQ Library on July 28, 2015
Sera. Sera obtained from C6-deficient rabbits (Rancho de Conejo, Vista, Calif.), C4-deficient guinea pigs, and a C8-deficient human were divided into small aliquots and frozen at -70 C. The C4-deficient guinea pigs were provided by Dr. Michael Frank, National Institutes of Health, Bethesda, Md., and the C8-deficient serum by Dr. George F. Brooks, Veterans Administration Hospital, Martinez, Calif. Hypogammaglobulinemic human sera were provided by Dr. Beatrice Lampkin, Children's Hospital Research Foundation, Cincinnati, Ohio. For PNHS, sera were obtained from 25 healthy adult donors, pooled, and stored as described above. PNHS was depleted of factor B by heating at 50 C for 30 min [11]; this serum will be referred to as RB. Factor D-depleted PNHS (RD) was prepared by a minor modification of the method of Martin et al. [12]. PNHS treated with 0.002 M EDTA was applied to a 350-ml Sephadex G-75 column equilibrated with saline containing 0.002 M EDTA. The excluded peak was concentrated and dialyzed against 0.01 M phosphate-buffered saline, pH 7.4. Properdin-depleted serum (RP) was prepared by adsorption of PNHS with zymosan at 17 C in the presence of magnesium ions [13]. PNHS was depleted of C3 (R3) by treatment with 0.15 M hydrazine [14] and of terminal complement components by treatment with inulin (10 mg/ml) or cobra venom factor (50 units/ ml; Cordis Laboratories, Miami, Fla.) at 37 C for 1 hr. Measurement of serum opsonic activity. A minor modification of the method of Hirsch and Strauss was used for measurement of the bactericidal activity of the leukocytes [15]. The strains of Bacteroides were isolated from clinical specimens in the Surgical Bacteriology Laboratory of the University of Cincinnati Medical Center, Cincinnati, Ohio. Diluent used in the experiments was Hanks' balanced salt solution (Microbiological Associates, Bethesda, Md.) containing 0.1% gelatin that was deoxygenated by boiling immediately before the experiments. Bacteria were maintained in thioglycolate medium at -70 C; before each experiment, a tube of thioglycolate medium was inoculated from a frozen culture. The tube was incubated overnight at 37
Bjornson and Bjornson
353
Opsonization of Bacteroides
Figure 1. Alkaline polyacrylamide discontinuous gel electrophoresis of 20 p.g of factor B, purified from the pseudoglobulin fraction of pooled normal human serum by the method of Gotze and Muller-Eberhard [11]. Top of band is at left.
Results
In our preliminary experiments, a standard anticomplementary procedure (heating at 56 C for 30 min) was used to inactivate complement in PNHS. In addition, hypogammaglobulinemic seFigure 2. Opsonic activity of human sera depleted of complement or immunoglobulin for Bacteroides fragilis (strain no. 1365) and Bacteroides thetaiotaomicron (strain no. 1343). The following reaction mixtures were tested: (1) pooled normal human serum (PNHS) and bacteria; (2) leukocytes and bacteria; (3) PNHS, leukocytes, and bacteria; (4) heated PNHS (at 56 C for 30 min), leukocytes, and bacteria; (5) hypogammaglobulinemic serum (HS) no. 1, leukocytes, and bacteria; (6) HS no. 2, leukocytes, and bacteria; (7) HS no. 3, leukocytes, and bacteria. HS no. 1, 2, and 3 contained 240-330 mg of IgG 1100 ml, 16- 29 mg of IgM/100 ml, and 1 log in bacterial counts of both strains of Bacteroides. These results indicated that both complement and immunoglobulin in normal human serum were required for opsonization of B. fragilis and B. thetaiotaomicron. To determine the participation of late-acting complement components C6-C9, we used C8-deficient human serum and C6-deficient rabbit serum. The opsonic activity of the C8-deficient human serum was as efficient as the opsonic activity of PNHS (figure 3). (The opsonic activity of the C6-deficient rabbit serum could not be evaluated since normal rabbit serum was also found to be unable to support phagocytosis and
a thetaiotaomicron
I x 101
IxIO' ' - - - - - - ' - - - - - ' - - - - - - - - ' lx 10·L--='-~--c'c---------' o 30 90 180 0 30 90 180 INCUBATION (min)
Downloaded from http://jid.oxfordjournals.org/ at UQ Library on July 28, 2015
by intracellular killing and that killing could therefore be used as a method for measurement of opsonization. Purified proteins. Factor B was purified from the pseudoglobulin fraction of PNHS by the method of Cotze and Muller-Eberhard [11]. By alkaline polyacrylamide discontinuous gel electrophoresis [16], the preparation contained a single band (figure I). By immunoelectrophoretic analysis using antiserum to factor B, the protein migrated in the f3 position and did not contain activated factor B (B). Purified human C2 was purchased from Cordis Laboratories.
354
Bjornson and Bjornson
"'l
B.thetaiataomicron
o
30
_ _ _-"I
90
~
IxIO'LI_-"I
0
~ I
~
90
Figure 3. Opsonic activity of C6and C8-deficient sera for Bacteroides fragilis (strain no. 1365) and Bacteroides thetaiotaomicron (strain no. 1343). The following reaction mixtures were tested: (1) leukocytes and bacteria; (2) pooled normal human serum, leukocytes, and bacteria; (3) C8-deficient human serum, leukocytes, and bacteria; (4) normal rabbit serum, leukocytes, and bacteria; (5) C6-deficient rabbit serum, leukocytes, and bacteria. The points represent mean values of two to five experiments, and the vertical bars represent the SE of the mean.
---!I
100
INCUBATION (min)
killing of the test strains.) These results indicated that C8 and C9 were not required for opsonization of the strains of Bacteroides. The participation of the alternative complement pathway in opsonization of the bacteria was next investigated. Human sera depleted of terminal complement components by treatment with inulin or cobra venom factor, recognized activating substances of the alternative pathway, were unable to support phagocytosis of the test strains (figure 4). (The opsonic activity of C4deficient guinea pig serum could not be evaluated since the maximal reduction in bacterial ~ thetaiotaomicron
Ix 10'
counts in the presence of normal or C4-deficient sera was only 0.5 log.) R3 was also unable to opsonize the test strains effectively (figure 5). In addition, the opsonic activity of PNHS was markedly reduced by heating at 50 C for 30 min, which removes factor B (RB), or by depleting the serum of factor D by molecular sieve chromatography on Sephadex G-75 (RD). The only differences between the opsonic requirements of the two strains were observed when the opsonic activity of RP was tested. The opsonic activity of RP for B. fragilis was equivalent to the opsonic activity of PNHS. The kinetics of opsonization
IxlO'
I
30
90
IBO
Figure 4. Participation of the alternative complement pathway in opsonization of Bacteroides fragilis (strain no. 1365) and Bacteroides thetaiotaomicron (strain no. 1343). The following reaction mixtures were tested: (1) normal guinea pig serum, leukocytes, and bacteria; (2) C4-deficient guinea pig serum, leukocytes, and bacteria; (3) pooled normal human serum (PNHS). leukocytes, and bacteria; (4) PNHS treated with 10 mg of inulin/ml (at 37 C for 60 min), leukocytes, and bacteria; (5) PNHS treated with 50 units of cobra venom factor/ml (at 37 C for 60 min), leukocytes, and bacteria. The poin ts represent mean values of two to five experiments, and the vertical bars represent the SE of the mean.
Downloaded from http://jid.oxfordjournals.org/ at UQ Library on July 28, 2015
I X lo'L--,"~_ _~_
IxlO'
355
Opsonization of Bacteroides
lx 10'
of B. thetaiotaomicron by RP was markedly decreased, and a significant reduction in bacterial counts was only observed after incubation for 3 hr. The results provided indirect evidence to indicate an absolute requirement for C3 and factors Band D for opsonization of both strains of Bacteroides and a partial requirement for properdin for opsonization of B. thetaiotaomicron but not for opsonization of B. fragilis. Since heating at 50 C for 30 min is known to inactivate C2 as well as factor B, heated serum was supplemented with purified human C2 or factor B, and opsonic activity was determined. Addition of two different preparations of purified human C2 to heated PNHS at a concentration that restored hemolytic activity to normal did not substantially increase the opsonic activity of the serum for the bacterial strains (figure 6). Opsonic activity of the heated serum was fully restored to normal by physiologic concentrations of two different preparations of purified human factor B but not by purified C2. In addition, restoration of opsonic activity of heated serum by factor B was dose-dependent (figure 7). These results indicated a requirement for factor B for opsonization of the bacteroides strains. Discussion In the present investigation, evidence was pro-
~
thetoiotaomicron
I xlO'
~
'3 I xlO' 4
-~
90
------Teo
~
'
Participation of Immunoglobulin and the Alternative Complement Pathway in Opsonization of Bacteroides fragilis and Bacteroides thetaiotaomicron From the Departments of Medicine and Microbiology, University of Cincinnati College of Medicine, Cincinnati, Ohio
Ann B. Bjornson and H. Stephen Bjornson
Studies during the past decade have established that erythrocytes and pneumococci can be opsonized by the sequential reactions of antibody and classical complement components Cl, C4, C2, and C3 [1, 2]. The participation of an alternative complement pathway in opsonization of Streptococcus pneumoniae, Staphylococcus aureus, Proteus mirabilis, Serratia marcescens, Escherichia coli, and Pseudomonas aeruginosa has also been suggested by several studies [3-8]. Received for publication December 27, 1977, and in revised form March 8,1978. This work was supported by grant no. AI-13688 from the U.S. Public Health Service and by contract no. DAMD-17-76-C-6023 from the U.S. Army Medical Research and Development Command. The authors thank Dr. Michael Frank, National Institutes of Health, Bethesda, Maryland, Dr. George F. Brooks, Veterans Administration Hospital, Martinez, California, and Dr. Beatrice Lampkin, Children's Hospital Research Foundation, Cincinnati, Ohio, for the C4-deficient guinea pigs, C8-deficient human serum, and hypogammaglobulinemic human sera, respectively. We also thank Barbara Kitko and Constance Zellner for their technical assistance. Please address requests for reprints to Dr. Ann B. Bjornson, Division of Infectious Diseases, Department of Medicine, University of Cincinnati College of Medicine, 231 Bethesda Avenue, Cincinnati, Ohio 45267.
However, minimal information is available regarding specific humoral and cellular host defense mechanisms against gram-negative, anaerobic microorganisms. Casciato et al. demonstrated that, under aerobic conditions, clinical isolates of Bacteroides fragilis and Bacteroides thetaiotaomicron were not susceptible to the bacteriolytic activity of human serum [9]. Bjornson et al. subsequently showed that, under anaerobic conditions, strains of B. fragilis and B. thetaiotaomicron were phagocytosed and killed intracellularly by human polymorphonuclear leukocytes in the presence of pooled normal human serum (PNHS) but not by either leukocytes or serum alone [10]. Similar results were obtained when the experiments were carried out in an aerobic environment, with the exception that B. fragilis was phagocytosed and killed intracellularly by leukocytes to some extent in the absence of serum. The present investigation was undertaken to determine the requirements for immunoglobulin and complement in phagocytosis and intracellular killing of strains of Bacteroides by leukocytes. All of the experiments to be described were carried out under anaerobic conditions [10].
351
Downloaded from http://jid.oxfordjournals.org/ at UQ Library on July 28, 2015
Studies were conducted to determine the requirements for immunoglobulin and complement for opsonization of Bacteroides fragilis and Bacteroides thetaiotaomicron. The ability of human sera depleted of immunoglobulin or complement components to promote phagocytosis and intracellular killing of the strains of Bacteroides by human leukocytes was measured in vitro under anaerobic conditions. Neither hypogammaglobulinemic sera nor pooled normal human serum (PNHS) heated at 56 C for 30 min supported phagocytosis and killing of the strains of Bacteroides. Sera depleted of terminal complement components by treatment with inulin or cobra venom factor and C8-deficient human serum did not support phagocytosis of the test strains. PNHS depleted of C3, factor B, or factor D also did not support phagocytosis of either strain. Dose-dependent restoration of the opsonic activity of factor B-depleted serum was accomplished by purified human factor B but not by human C2. The results indicated that immunoglobulin and components of the alternative complement pathway participate in opsonization of the strains of Bacteroides tested in this study.
352
Materials and Methods
C and then transferred to a tube of deoxygenated thioglycolate broth that was supplemented with 0.5% yeast extract and 1% glucose and incubated at 37 C overnight. The bacteria were washed once and diluted in deoxygenated diluent immediately before the experiments. Leukocytes were prepared from the plasma of dextransedimented, heparinized blood obtained from healthy adult volunteers. The leukocyte-rich plasma was divided and centrifuged at 200 g for 10 min, and the supernatants were discarded. The leukocytes were washed twice and resuspended in deoxygenated diluent. Various combinations of serum (10% concentration), leukocytes (5 X 106 ) , bacteria (1.0 X 106) , and diluent in a final volume of 1 ml were added together in plastic-capped tubes. The reagents were added to the tubes in an anaerobic glove box (Coy Manufacturing, Ann Arbor, Mich.). In addition, the tubes to be used were equilibrated overnight in the glove box, and the sera were equilibrated in the glove box on ice for 1 hr before their addition to the respective reaction mixtures. The internal atmosphere of the glove box contained 85% nitrogen, 10% hydrogen, 5% carbon dioxide, and not >25 ppm of oxygen. The pH and p02 (oxygen pressure) in each of the different reaction mixtures were measured before incubation at 37 C. The mean pH of the mixtures (±SE) was 6.95 ± 0.05, and the p02 ranged from 19 to 25 mm Hg. The next steps in the procedure were carried out entirely in the glove box. The tubes were incubated for 3 hr at 37 C on a rotating platform, and samples were removed at zero-time and at 30,90, and 180 min. The samples were diluted in deoxygenated distilled water, and the dilutions were plated in thioglycolate agar by the pourplate method. The plates from all experiments were incubated anaerobically in the glove box for four days at 37 C, and the colonies were enumerated to determine the total number of bacteria surviving in each reaction mixture. In preliminary experiments, surviving extracellular and total bacteria were measured in reaction mixtures containing serum, leukocytes, and bacteria, or leukocytes and bacteria alone. Extracellular counts were found to be similar if not identical to total counts, a finding which indicated that phagocytosis was rapidly followed
Downloaded from http://jid.oxfordjournals.org/ at UQ Library on July 28, 2015
Sera. Sera obtained from C6-deficient rabbits (Rancho de Conejo, Vista, Calif.), C4-deficient guinea pigs, and a C8-deficient human were divided into small aliquots and frozen at -70 C. The C4-deficient guinea pigs were provided by Dr. Michael Frank, National Institutes of Health, Bethesda, Md., and the C8-deficient serum by Dr. George F. Brooks, Veterans Administration Hospital, Martinez, Calif. Hypogammaglobulinemic human sera were provided by Dr. Beatrice Lampkin, Children's Hospital Research Foundation, Cincinnati, Ohio. For PNHS, sera were obtained from 25 healthy adult donors, pooled, and stored as described above. PNHS was depleted of factor B by heating at 50 C for 30 min [11]; this serum will be referred to as RB. Factor D-depleted PNHS (RD) was prepared by a minor modification of the method of Martin et al. [12]. PNHS treated with 0.002 M EDTA was applied to a 350-ml Sephadex G-75 column equilibrated with saline containing 0.002 M EDTA. The excluded peak was concentrated and dialyzed against 0.01 M phosphate-buffered saline, pH 7.4. Properdin-depleted serum (RP) was prepared by adsorption of PNHS with zymosan at 17 C in the presence of magnesium ions [13]. PNHS was depleted of C3 (R3) by treatment with 0.15 M hydrazine [14] and of terminal complement components by treatment with inulin (10 mg/ml) or cobra venom factor (50 units/ ml; Cordis Laboratories, Miami, Fla.) at 37 C for 1 hr. Measurement of serum opsonic activity. A minor modification of the method of Hirsch and Strauss was used for measurement of the bactericidal activity of the leukocytes [15]. The strains of Bacteroides were isolated from clinical specimens in the Surgical Bacteriology Laboratory of the University of Cincinnati Medical Center, Cincinnati, Ohio. Diluent used in the experiments was Hanks' balanced salt solution (Microbiological Associates, Bethesda, Md.) containing 0.1% gelatin that was deoxygenated by boiling immediately before the experiments. Bacteria were maintained in thioglycolate medium at -70 C; before each experiment, a tube of thioglycolate medium was inoculated from a frozen culture. The tube was incubated overnight at 37
Bjornson and Bjornson
353
Opsonization of Bacteroides
Figure 1. Alkaline polyacrylamide discontinuous gel electrophoresis of 20 p.g of factor B, purified from the pseudoglobulin fraction of pooled normal human serum by the method of Gotze and Muller-Eberhard [11]. Top of band is at left.
Results
In our preliminary experiments, a standard anticomplementary procedure (heating at 56 C for 30 min) was used to inactivate complement in PNHS. In addition, hypogammaglobulinemic seFigure 2. Opsonic activity of human sera depleted of complement or immunoglobulin for Bacteroides fragilis (strain no. 1365) and Bacteroides thetaiotaomicron (strain no. 1343). The following reaction mixtures were tested: (1) pooled normal human serum (PNHS) and bacteria; (2) leukocytes and bacteria; (3) PNHS, leukocytes, and bacteria; (4) heated PNHS (at 56 C for 30 min), leukocytes, and bacteria; (5) hypogammaglobulinemic serum (HS) no. 1, leukocytes, and bacteria; (6) HS no. 2, leukocytes, and bacteria; (7) HS no. 3, leukocytes, and bacteria. HS no. 1, 2, and 3 contained 240-330 mg of IgG 1100 ml, 16- 29 mg of IgM/100 ml, and 1 log in bacterial counts of both strains of Bacteroides. These results indicated that both complement and immunoglobulin in normal human serum were required for opsonization of B. fragilis and B. thetaiotaomicron. To determine the participation of late-acting complement components C6-C9, we used C8-deficient human serum and C6-deficient rabbit serum. The opsonic activity of the C8-deficient human serum was as efficient as the opsonic activity of PNHS (figure 3). (The opsonic activity of the C6-deficient rabbit serum could not be evaluated since normal rabbit serum was also found to be unable to support phagocytosis and
a thetaiotaomicron
I x 101
IxIO' ' - - - - - - ' - - - - - ' - - - - - - - - ' lx 10·L--='-~--c'c---------' o 30 90 180 0 30 90 180 INCUBATION (min)
Downloaded from http://jid.oxfordjournals.org/ at UQ Library on July 28, 2015
by intracellular killing and that killing could therefore be used as a method for measurement of opsonization. Purified proteins. Factor B was purified from the pseudoglobulin fraction of PNHS by the method of Cotze and Muller-Eberhard [11]. By alkaline polyacrylamide discontinuous gel electrophoresis [16], the preparation contained a single band (figure I). By immunoelectrophoretic analysis using antiserum to factor B, the protein migrated in the f3 position and did not contain activated factor B (B). Purified human C2 was purchased from Cordis Laboratories.
354
Bjornson and Bjornson
"'l
B.thetaiataomicron
o
30
_ _ _-"I
90
~
IxIO'LI_-"I
0
~ I
~
90
Figure 3. Opsonic activity of C6and C8-deficient sera for Bacteroides fragilis (strain no. 1365) and Bacteroides thetaiotaomicron (strain no. 1343). The following reaction mixtures were tested: (1) leukocytes and bacteria; (2) pooled normal human serum, leukocytes, and bacteria; (3) C8-deficient human serum, leukocytes, and bacteria; (4) normal rabbit serum, leukocytes, and bacteria; (5) C6-deficient rabbit serum, leukocytes, and bacteria. The points represent mean values of two to five experiments, and the vertical bars represent the SE of the mean.
---!I
100
INCUBATION (min)
killing of the test strains.) These results indicated that C8 and C9 were not required for opsonization of the strains of Bacteroides. The participation of the alternative complement pathway in opsonization of the bacteria was next investigated. Human sera depleted of terminal complement components by treatment with inulin or cobra venom factor, recognized activating substances of the alternative pathway, were unable to support phagocytosis of the test strains (figure 4). (The opsonic activity of C4deficient guinea pig serum could not be evaluated since the maximal reduction in bacterial ~ thetaiotaomicron
Ix 10'
counts in the presence of normal or C4-deficient sera was only 0.5 log.) R3 was also unable to opsonize the test strains effectively (figure 5). In addition, the opsonic activity of PNHS was markedly reduced by heating at 50 C for 30 min, which removes factor B (RB), or by depleting the serum of factor D by molecular sieve chromatography on Sephadex G-75 (RD). The only differences between the opsonic requirements of the two strains were observed when the opsonic activity of RP was tested. The opsonic activity of RP for B. fragilis was equivalent to the opsonic activity of PNHS. The kinetics of opsonization
IxlO'
I
30
90
IBO
Figure 4. Participation of the alternative complement pathway in opsonization of Bacteroides fragilis (strain no. 1365) and Bacteroides thetaiotaomicron (strain no. 1343). The following reaction mixtures were tested: (1) normal guinea pig serum, leukocytes, and bacteria; (2) C4-deficient guinea pig serum, leukocytes, and bacteria; (3) pooled normal human serum (PNHS). leukocytes, and bacteria; (4) PNHS treated with 10 mg of inulin/ml (at 37 C for 60 min), leukocytes, and bacteria; (5) PNHS treated with 50 units of cobra venom factor/ml (at 37 C for 60 min), leukocytes, and bacteria. The poin ts represent mean values of two to five experiments, and the vertical bars represent the SE of the mean.
Downloaded from http://jid.oxfordjournals.org/ at UQ Library on July 28, 2015
I X lo'L--,"~_ _~_
IxlO'
355
Opsonization of Bacteroides
lx 10'
of B. thetaiotaomicron by RP was markedly decreased, and a significant reduction in bacterial counts was only observed after incubation for 3 hr. The results provided indirect evidence to indicate an absolute requirement for C3 and factors Band D for opsonization of both strains of Bacteroides and a partial requirement for properdin for opsonization of B. thetaiotaomicron but not for opsonization of B. fragilis. Since heating at 50 C for 30 min is known to inactivate C2 as well as factor B, heated serum was supplemented with purified human C2 or factor B, and opsonic activity was determined. Addition of two different preparations of purified human C2 to heated PNHS at a concentration that restored hemolytic activity to normal did not substantially increase the opsonic activity of the serum for the bacterial strains (figure 6). Opsonic activity of the heated serum was fully restored to normal by physiologic concentrations of two different preparations of purified human factor B but not by purified C2. In addition, restoration of opsonic activity of heated serum by factor B was dose-dependent (figure 7). These results indicated a requirement for factor B for opsonization of the bacteroides strains. Discussion In the present investigation, evidence was pro-
~
thetoiotaomicron
I xlO'
~
'3 I xlO' 4
-~
90
------Teo
~
'
