Furthermore, dextran did not influence the TFA of endocardial vegetations, which was ... vegetations isolated from rabbits infected either with dextran-positive S.
INFECTION AND IMMUNITY, Dec. 1996, p. 5166–5170 0019-9567/96/$04.0010 Copyright q 1996, American Society for Microbiology
Vol. 64, No. 12
Role of Phagocytosis in Activation of the Coagulation System in Streptococcus sanguis Endocarditis MAURICE J. L. M. F. BANCSI,1* MARCEL H. A. M. VELTROP,1 ROGIER M. BERTINA,2 AND JAN THOMPSON1 Department of Infectious Diseases1 and Hemostasis and Thrombosis Research Center,2 Leiden University Hospital, Leiden, The Netherlands Received 18 July 1996/Returned for modification 27 August 1996/Accepted 27 September 1996
The formation of vegetations consisting of fibrin, cellular elements, humoral factors, and bacteria is the central event in the pathogenesis of bacterial endocarditis. Fibrin formation occurs on the vegetation, the coagulation system being activated locally via the expression of tissue factor (TF) on fibrin-adherent monocytes. This study was performed to assess the importance of phagocytosis of fibrin-adherent Streptococcus sanguis in the stimulation of TF expression on fibrin-adherent monocytes, as well as a role for “frustrated” phagocytosis. With the latter process, these cells are unable to remove bacteria from the fibrin surface but nonetheless might be activated to generate TF. We found that serum was not required for the stimulation of TF expression by fibrin-adherent monocytes in the presence of S. sanguis in an in vitro model for bacterial endocarditis. The bacterial adhesin dextran did not influence the TF activity (TFA) of fibrin-adherent monocytes: TFA was the same after stimulation with a dextran-positive streptococcus as with its dextran-negative mutant. Furthermore, dextran did not influence the TFA of endocardial vegetations, which was the same for vegetations isolated from rabbits infected either with dextran-positive S. sanguis or its dextran-negative mutant. These results do not support the hypothesis that in bacterial endocarditis (frustrated) phagocytosis significantly contributes to TF expression on vegetation-adherent monocytes. Fibronectin, however, although not influencing the fibrin binding of the streptococci, did enhance the TFA of monocytes in a concentrationdependent manner. We conclude that although streptococci do enhance expression of TFA on monocytes, phagocytosis and bacterial adhesins do not play a major role in this process. Stimulation of monocyte TFA may be more dependent on interactions between monocytes and the vegetational surface via fibronectin receptors, such as VLA 4 and VLA 5 (very late antigens 4 and 5). vegetational surface (14) but also its presence on the outer membrane of S. sanguis accounts for a lower 50% infective dose in the rabbit model of BE (i.e., the number of bacteria required to ensure infected vegetations in 50% of the rabbits that were injected with live bacteria) compared with an isogenic mutant strain of S. sanguis that lacks the ability to produce dextran (12). The tight adherence of streptococci may lead to “frustrated” phagocytosis by the monocytes, phagocytosing cells being unable to remove the bacteria from the vegetational surface, which also could result in de novo-synthesized TFA on the monocyte surface (18). In this study we investigated the contribution of phagocytosis and frustrated phagocytosis in the stimulation of TF expression on adherent monocytes by adherent bacteria. The role of phagocytosis was assessed by studying the effect of serum on the stimulation of TF expression (serum is thought to be an obligatory cofactor of cell-mediated phagocytosis). The eventual role of frustrated phagocytosis was addressed in both in vitro and in vivo experiments. Finally, the role of fibronectin in mediating frustrated phagocytosis was investigated. The influence of fibronectin on the adherence of streptococci to the fibrin plates was assessed, as well as the influence of fibronectin on the TFA of fibrin-adherent monocytes, both in the absence and presence of S. sanguis.
The activation of the coagulation system, which leads to the formation of endocardial vegetations, is an important feature in the course of bacterial endocarditis (BE). Monocytes play a central role in this process by the expression on their surface of tissue factor (TF), the activator of the extrinsic pathway of blood coagulation (1, 15). Phagocytosis of bacteria by monocytes on the vegetational surface might account for the expression of the TF activity (TFA), since these cells, when examined in suspension, were found to generate TF after serum-dependent phagocytosis of Streptococcus sanguis, whereas nonphagocytosing cells lack TF expression (17). TFA in that study was detectable after 4 to 5 h of incubation, indicating that this activity represents de novo-synthesized TF (17). In BE, however, monocytes and bacteria are adherent to the vegetational surface. Factors that mediate the initial adhesion of monocytes and/or bacteria may therefore affect the phagocytosis of the bacteria by the monocytes and thus may interfere with the expression of TF on the monocyte surface. The bacterial adhesion factor dextran, an outer membrane constituent of S. sanguis, and fibronectin, an extracellular matrix component, are thought to be involved in the initial adherence of the streptococci to the surface of the endocardial vegetations. In the rat model of BE, it was shown that inactivation of fibronectin binding of S. sanguis reduced the virulence of the bacterium in BE (10), while not only is dextran reported to be involved in the binding of the bacterium to the
MATERIALS AND METHODS Microorganism. The dextran-producing S. sanguis NCTC 7864 (D1) and its dextran-negative mutant (D2) were used. The dextran-negative streptococcus is a spontaneous mutant isolated from a culture of the parent dextran-positive strain. Both strains are the same as those used in previous studies (5, 12, 13, 15). Streptococci from an overnight culture in Todd-Hewitt broth (Oxoid, London, England) were washed three times in pyrogen-free saline. For in vivo experi-
* Corresponding author. Mailing address: Leiden University Hospital, Department of Infectious Diseases, C5P, P.O. Box 9600, 2300 RC Leiden, The Netherlands. Phone: 31-71-5262620. Fax: 31-71-5266758. 5166
VOL. 64, 1996
COAGULATION ASPECTS OF BACTERIAL ENDOCARDITIS
ments, bacteria were diluted in saline to a concentration of approximately 108 CFU per ml. For in vitro experiments, streptococci were diluted in RPMI 1640 culture medium (ICN Flow Laboratories, Irvine, Scotland) to the concentration indicated. In vivo model. Bacterial endocarditis was induced in male New Zealand White rabbits as previously described (7, 12). Briefly, rabbits, weighing about 2 kg, were anaesthetized by intramuscular injection of 1.5 ml of Hypnorm (Janssen Pharmaceutica, Tilburg, The Netherlands), and a polyethylene catheter (Portex, Hythe, England) was introduced into the left ventricle of the heart via the left carotid artery. The catheter was left in situ for the duration of the experiment. After 48 h, 108 CFU of live streptococci in 1 ml of pyrogen-free saline was injected intravenously in a marginal ear vein. Two days later, rabbits were sacrificed by intravenous injection of Euthesate (sodium pentobarbital; Apharma, Arnhem, The Netherlands). The heart was removed, and the endocardial vegetations were isolated aseptically. Care was taken to ensure that only vegetations were removed (6). Control rabbits were catheterized but received 1 ml of saline instead of streptococci after 48 h. Processing of vegetations. Isolated vegetations were weighed and homogenized (5%, wt/vol) in a buffer (pH 7.45) containing 10 mM HEPES (N-2hydroxyethylpiperazine-N9-2-ethanesulfonic acid) (Sigma, St. Louis, Mo.), 137 mM NaCl (Merck, Darmstadt, Germany), 11 mM a-D-glucose (BDH Chemicals Ltd., Poole England), 4 mM KCl (Merck), and 5 mg of bovine serum albumin (Sigma) per ml. Part of the homogenate was used for the determination of bacterial numbers and part was used for the assessment of the TFA. The degree of infection of the vegetations was determined by counting six 0.01-ml aliquots of a series of 10-fold dilutions on sheep blood agar after an overnight incubation at 378C (5) and is expressed as log CFU per gram of vegetation. The part of the homogenate to be used for assessment of TFA was frozen and thawed three times to lyse intact cells and then stored at 2708C. Blood cultures. Immediately before rabbits were sacrificed, 1 ml of blood was drawn from a marginal ear vein and collected in a vial containing 10 mg of EDTA, and 100 ml was plated on sheep blood agar plates. CFU were counted after overnight incubation at 378C. Monocytes. Heparinized buffy coat from 500 ml of peripheral venous blood of a healthy donor was diluted five times in phosphate-buffered saline (PBS) containing 0.5 U of heparin per ml and was layered on Ficoll-amidotrizoate (r 5 1,077 g/ml) (4). After differential centrifugation, the mononuclear-cell-rich interface, containing 30% monocytes, was carefully removed, washed three times in PBS-heparin, and resuspended in RPMI 1640 tissue culture medium at a concentration of 106 monocytes per ml. Ten milliliters of the cell suspension was injected into Teflon culture bags (16) and cultured overnight at 378C, with 5% CO2 and a relative humidity of 95%. The next day, the cells were recovered and used in the experiments. Fibrin plates. In a 24-well tissue culture plate (Costar, Cambridge, England), 200 ml of either soluble fibronectin containing fibrinogen (Sigma) or fibronectinfree fibrinogen (Kordia, Leiden, The Netherlands) (5 mg/ml) dissolved in a buffer containing 50 mM triethanolamine (TEA; Fluka BioChemika, Buchs, Switzerland)–100 mM NaCl (Merck) (pH 7.45) was mixed with 50 ml of 100 mM CaCl2 (Merck) before 10 ml of 0.5 U of thrombin (Central Laboratory of the Blood Transfusion Service, Amsterdam, The Netherlands) per ml was added to induce fibrin formation. The mixture was allowed to polymerize overnight at 48C. Fibronectin-free fibrinogen was supplemented with soluble exogenous fibronectin (Sigma) when indicated. The next day, the plates were used in experiments. Adherence of streptococci to fibrin plates. The adherence of the bacteria to the fibrin plates was determined as described before (1). In short, bacteria were layered on fibrin plates and incubated at 378C and 5% CO2. After 1 h, the plates were washed, the fibrin was removed from the 24-well tissue culture plate and homogenized, and the numbers of bacteria were determined in 0.01-ml samples of serial 10-fold dilutions, which were plated on sheep blood agar plates and incubated overnight at 378C (see above). TFA assay. TFA of the vegetations was measured by following the factor VII (FVII)-dependent factor X (FX) activation with an amidolytic assay for activated FX (FXa) (2). In short, homogenized vegetations (5%, wt/vol) were incubated with FVII and CaCl2 at 378C. After 20 min FX was added, and the reaction was allowed to proceed for 20 min. Next, the chromogenic substrate pefachrome FXa (15 mM in 50 mM EDTA; Kordia) was added. The colorimetric reaction was stopped after 20 min with 50% acetic acid (Merck), and the optical density at 405 nm was measured. The absorption value was converted to picomoles of FXa by using a calibration curve ranging from 0 to 30 nmoles of FXa. The FXa calibrator was prepared by complete activation of isolated FX with the FX activator from Russell’s viper venom (Chromogenix). The TFA was expressed as picomoles of FXa per gram of vegetation per minute. The experimental procedure for measuring TFA on the surface of fibrinadherent monocytes was adapted from Bom et al. (3) and modified as previously described (1). In short, approximately 1.5 3 106 adherent monocytes were stimulated for 4 h at 378C. After being washed, cells were incubated with FVII and CaCl2 for 20 min. Next, FX was added, and after 5 min of incubation a sample was taken for FXa measurement. TFA was expressed as picomoles of FXa. Factors VII and X were purified as described earlier (1). Statistical analysis. For determination of significance of differences between the TFAs of vegetations of noninfected control rabbits and of rabbits infected with S. sanguis D1 or S. sanguis D2, analysis of variance was used with Newman-
5167
TABLE 1. Effects of dextran on TFA, infection, and weight of endocardial vegetations of rabbits infected with dextran-positive S. sanguis D1 or its dextran-negative mutant D2a Exptl group
TFAb
Infectionc
Sterile (control) S. sanguis D1 S. sanguis D2
378 6 123 728 6 131d 747 6 122d
8.64 6 0.83 9.00 6 0.94
Vegetational weight (mg)
12.3 6 8.5 40.1 6 20.1e 39.2 6 18.7e
Results are means 6 standard deviations of at least four vegetations. Determined as picomoles of FXa per gram of vegetation per minute. Determined as log CFU per gram of vegetation. d P , 0.0015 compared with control. e P , 0.001 compared with control. a b c
Keuls post hoc correction for significance levels. The degree of infection was analyzed by the unpaired Student t test. Significance of difference for in vitro experiments was calculated by using the paired Student t test. The significance level (a) was 5%.
RESULTS Effects of dextran on infection, weight, and TFA of endocardial vegetations. To test in vivo the hypothesis that frustrated phagocytosis is involved in the stimulation of the monocytes to express TF in the course of BE, the effects of dextran, involved in the initial adherence to the vegetational surface, on infection, weight, and TFA of endocardial vegetations were assessed. All vegetations of rabbits injected with S. sanguis D1 or D2 were colonized with the respective strain (n 5 4 rabbits per group). After 48 h of infection no significant differences were found in bacterial titers for both strains, which ranged from 7.5 to 9.6 log CFU/g of vegetation (Table 1). The mean weight of vegetations of rabbits infected with S. sanguis D1 did not differ from those infected with S. sanguis D2, but the weights of both groups of vegetations were significantly higher than the weights of the vegetations of noninfected control rabbits (39.2 6 18.7 mg for D2 and 40.1 6 20.1 mg for D1 versus 12.3 6 8.5 mg for control vegetations, P , 0.001) (Table 1). Comparison of the TFA of vegetations of S. sanguis D1infected rabbits and S. sanguis D2-infected rabbits did not reveal a significant difference (Table 1). Effects of dextran and serum on the adherence of S. sanguis to fibrin plates. The effect of dextran was also studied in vitro, because the presence of dextran on the outer membrane of S. sanguis apparently had no effect on TFA, weight, and infection of the vegetations in the in vivo model. The adherence of streptococci to fibrin was determined as described in Materials and Methods. The attachment, calculated as the ratio of the number of adherent bacteria to the inoculum size, was found to be 5% for inocula ranging from 106 to 109 CFU both for S. sanguis D1 and for the mutant D2. Adding 10% serum or 1 mg of exogenous dextran (Sigma) per ml to the bacterial suspension during the incubation on the fibrin plates did not affect the attachment of the bacteria. Effects of serum and dextran on TFA of fibrin-adherent monocytes. After 4 h of coincubation of fibrin-adherent bacteria and monocytes in the presence of 10% serum or 1 mg of dextran per ml, the attachment of both S. sanguis D1 and D2 was not altered. Also, the number of adherent monocytes did not change during these experiments. These results ensured stable experimental conditions for the measurements of TF expression. Next, it was established whether 10% serum or 1 mg of exogenous dextran per ml influenced the magnitude of mono-
5168
BANCSI ET AL.
FIG. 1. Effect of 10% human serum on TFA of 1.5 3 106 fibrin-adherent monocytes incubated for 4 h in the presence or absence of 2 3 107 CFU of S. sanguis D1 or D2. Results are means 6 standard deviations of three duplicate experiments. o, serum; s, no serum.
cyte TFA in the absence or presence of bacteria. In Fig. 1, it is shown that serum did not enhance the TFA of the monocytes incubated with either S. sanguis D1 or D2. Addition of exogenous dextran to the D2 streptococci did not influence their stimulatory effect on the monocytes. Similarly, addition of exogenous dextran to the D1 streptococci had no stimulatory effect (Fig. 2). Figure 3 shows the ratio-dependent stimulation of the monocytes by both S. sanguis D1 and D2. In the presence of the dextran-producing streptococci, the monocytes were stimulated to the same degree as in the presence of the dextran-negative S. sanguis. A minimal ratio of 5 bacteria per monocyte was necessary to significantly enhance the TFA, while a ratio of 10 bacteria per monocyte induced maximal TFA on the monocytes. Effect of fibronectin on TFA of fibrin-adherent monocytes. Also, fibronectin is known to increase the attachment of S. sanguis to a vegetation (5). Thus its presence in the endocardial vegetation might facilitate tight attachment of the bacteria to
INFECT. IMMUN.
FIG. 3. Influence of S. sanguis-to-monocyte ratio (s.sa/mo) on stimulation of TFA, expressed as fold increase of TFA of 1.5 3 106 fibrin-adherent monocytes incubated for 4 h in the presence or absence of streptococci. Results are means 6 standard deviations of three duplicate experiments. z, S. sanguis D1; o, S. sanguis D2.
the vegetational surface. This, in turn, could lead to frustrated phagocytosis. Therefore, the role of fibronectin was also studied in the in vitro model of BE. There was no difference in the degree of direct binding to fibronectin, fibrinogen, or gelatin of the dextran-producing and the dextran-negative streptococci used in this study. Neither did the presence of fibronectin enhance the binding of the streptococci to fibrin (data not shown). The effect of fibronectin on the TFA of fibrin-adherent monocytes was assessed because these cells are known to have fibronectin receptors (15). The TFA of monocytes adherent on fibrin depleted of fibronectin (FN2) was compared with that of monocytes adherent on fibronectin containing fibrin (FN1). As shown in Fig. 4A, in both the presence and absence of streptococci the TFA of the monocytes was significantly enhanced when adherent to FN1 compared with FN2. Next, the TFA of monocytes adherent to fibrin and consisting of a mixture of fibronectin-depleted and fibronectin-containing fibrin (FN6) was measured and compared with FN1- and FN2-adherent monocytes. TFA of these FN6 monocytes was significantly higher than the TFA of FN2 monocytes and significantly lower than the TFA of FN1 monocytes (Fig. 4A). Thus, we hypothesized that the TFA of the fibrin-adherent monocytes was enhanced by the presence of fibronectin in a dose-dependent manner. Indeed, we found that increasing the amount of exogenous fibronectin added to the fibronectin-depleted fibrinogen before polymerization (Fig. 4B) also resulted in significant increases of monocytic TFA, in both the presence and absence of streptococci. DISCUSSION
FIG. 2. Effect of exogenous dextran on TFA of 1.5 3 106 fibrin-adherent monocytes incubated for 4 h in the presence or absence of 2 3 107 CFU of S. sanguis D1 or D2. Results are means 6 standard deviations of three duplicate experiments. _ ^, exogenous dextran; s, no exogenous dextran.
The aim of the present study was to investigate whether (frustrated) phagocytosis of adherent bacteria by adherent monocytes is a significant stimulus for TF expression in the course of BE. Also, a possible role of the adhesion factors dextran and fibronectin in such a frustrated phagocytosis was assessed. In a previous study, we found that monocytes in suspension express TFA upon serum-dependent phagocytosis of bacteria (17). However, although monocytes adherent on fibrin plates express a higher TFA in the presence of bacteria, they do so also in the absence of serum (1) (Fig. 1). Thus, normal phagocytosis
VOL. 64, 1996
COAGULATION ASPECTS OF BACTERIAL ENDOCARDITIS
5169
dextran-producing bacteria adhere more tightly to the fibrin surface through an additional interaction of dextran with fibrin. Subsequently, for monocytes, it would be difficult to remove these bacteria from the fibrin surface, resulting in frustrated phagocytosis and more TFA expressed on the monocytes than on monocytes after normal phagocytosis. However, the present results do not support this hypothesis. We found that dextran did not enhance the binding of S. sanguis to the fibrin surface, nor did it enhance the TFA of the fibrin-adherent monocytes (Fig. 2 and 3). The findings that both the dextran-producing strain and its dextran-negative mutant induce vegetations with an equal weight, similar bacterial titers, and comparable TFA are in agreement with the results from the in vitro experiments (Table 1). Thus, although dextran is a virulence factor in the initial stage of BE by facilitating the initial infection of an endocardial vegetation (11), it does not seem to play a significant role in the progress of the disease. Strains of S. sanguis are reported to show a fibronectinmediated binding to the vegetational surface (11). For the strain of S. sanguis used in our study, however, adherence was not enhanced by the presence of fibronectin in the fibrin plates. However, monocytes are also known to have fibronectin receptors on their cell surface (8, 9). We found that fibronectin enhanced TFA expression in a concentration-dependent manner (Fig. 4A and B). This stimulation of the monocytes was higher in the presence of streptococci. Therefore at least two different pathways lead to expression of TFA on fibrin-adherent monocytes. Since stimulation of TF expression by S. sanguis is not mediated by serum-dependent phagocytosis, as described above, other interactions must be more important. What these are, however, remain to be elucidated. With regard to the fibronectin-mediated expression of TF on fibrin-adherent monocytes, there are multiple receptors on these cells that can bind fibronectin or fibrinogen (8, 9). Engagement of the b1-a4 integrin VLA 4 (very late antigen 4) is a potent stimulus to induce TFA on monocytes in suspension (9), while engagement of the b1-a5 integrin VLA 5 may upregulate TFA via the complement receptor 3, which is a receptor for fibrinogen (8). Future studies will focus on blocking the interaction of these receptors with their ligands by monoclonal antibodies. ACKNOWLEDGMENT FIG. 4. Effect of fibronectin on TFA of 1.5 3 106 fibrin-adherent monocytes incubated for 4 h in the presence or absence of 2 3 107 CFU of S. sanguis. Results are means 6 standard deviations of three duplicate experiments. (A) Fibronectin-free fibrin plates compared with fibronectin-containing fibrin plates and with fibrin plates containing a mixture of both forms of fibrin. s, fibronectinfree fibrin; o, mixture of fibronectin-free and fibronectin-containing fibrin; ^, _ fibronectin-containing fibrin. (B) Fibrin plates containing fibronectin-free fibrin supplemented with various amounts of exogenous fibronectin. s, 0 mg of fibronectin per ml of fibrin; o, 50 mg of fibronectin per ml of fibrin; ^, _ 100 mg of fibronectin per ml of fibrin.
cannot account for this enhanced TFA expression and frustrated phagocytosis might be more important. Monocyte TFA plays a cardinal role in the pathogenesis of BE. Endocardial vegetations have been shown to have procoagulant activity (6), which was shown to be monocyte dependent (9). Bacterial adhesion factors such as dextran, which is present on the outer membrane of S. sanguis, and fibronectin receptors have been reported to facilitate the induction of BE in vivo (8, 9, 11, 14). We hypothesized that monocytes are stimulated to express TFA after frustrated phagocytosis of bacteria on the vegetational surface. In this hypothesis the
This study was supported by The Netherlands Heart Foundation grant number 91.058. REFERENCES 1. Bancsi, M. J. L. M. F., J. Thompson, and R. M. Bertina. 1994. Stimulation of monocyte tissue factor expression in an in vitro model of bacterial endocarditis. Infect. Immun. 62:5669–5672. 2. Bancsi, M. J. L. M. F., M. H. A. M. Veltrop, R. M. Bertina, and J. Thompson. 1996. Influence of monocytes and antibiotic treatment on tissue factor activity of endocardial vegetations in rabbits infected with Streptococcus sanguis. Infect. Immun. 64:448–451. 3. Bom, V. J. J., V. W. M. van Hinsbergh, H. H. Reinalda-Poot, R. W. Mohanlal, and R. M. Bertina. 1991. Extrinsic activation of human coagulation factors IX and X on the endothelial cell surface. Thromb. Haemostasis 66:283–291. 4. Bo¨yum, A. 1968. Isolation of mononuclear cells and granulocytes from human blood. Scand. J. Clin. Invest. 21(Suppl.):77. 5. Buiting, A. G. M., J. Thompson, J. J. Emeis, H. Mattie, E. J. P. Brommer, and R. Van Furth. 1988. Effects of tissue-type plasminogen activator on Streptococcus sanguis-infected endocardial vegetations in vitro. J. Antimicrob. Chemother. 21:609–621. 6. Buiting, A. G. M., J. Thompson, D. van der Keur, W. C. Schmall-Bauer, and R. M. Bertina. 1989. Procoagulant activity of endocardial vegetations in rabbits with Streptococcus sanguis endocarditis. Thromb. Haemostasis 62: 1029–1033. 7. Durack, D. T., and P. B. Beeson. 1972. Experimental bacterial endocarditis. I. Colonisation of a sterile vegetation. Br. J. Pathol. 53:44–49. 8. Fan, S. T., and T. S. Edgington. 1994. Coupling of the adhesive receptor
5170
9.
10. 11. 12.
13.
BANCSI ET AL.
CD11b/CD18 to functional enhancement of effector macrophage tissue factor response. J. Clin. Invest. 87:50–57. Fan, S. T., N. Mackman, M. Z. Cui, and T. S. Edgington. 1995. Integrin regulation of an inflammatory effector gene. Direct induction of tissue factor promoter by engagement of b1 or a4 integrin chains. J. Immunol. 154:3266– 3274. Lowrance, J. H., L. M. Baddour, and W. A. Simpson. 1990. The role of fibronectin binding in the rat model of experimental endocarditis caused by Streptococcus sanguis. J. Clin. Invest. 86:7–13. Manning, J. E., E. B. H. Hume, and K. W. Knox. 1994. An appraisal of the virulence factors associated with streptococcal endocarditis. J. Med. Microbiol. 40:110–114. Meddens, M. J. M., J. Thompson, P. C. H. Leijh, and R. van Furth. 1984. Role of granulocytes in the induction of an experimental endocarditis with a dextran-producing Streptococcus sanguis and its dextran-negative mutant. Br. J. Exp. Pathol. 65:257–265. Meddens, M. J. M., J. Thompson, H. Mattie, and R. van Furth. 1984. Role
Editor: V. A. Fischetti
INFECT. IMMUN.
14. 15. 16. 17.
18.
of granulocytes in the prevention and therapy of experimental Streptococcus sanguis endocarditis in rabbits. Antimicrob. Agents Chemother. 25:263–267. Scheld, W. M., J. A. Valone, and M. A. Sande. 1978. Bacterial adherence in the pathogenesis of endocarditis. J. Clin. Invest 61:1394–1404. Tho ¨rig, L., J. Thompson, F. Eulderink, J. J. Emeis, and R. van Furth. 1980. Effects of monocytopenia and anticoagulation in experimental Streptococcus sanguis endocarditis. Br. J. Exp. Pathol. 61:108–116. van der Meer, J. W. M., J. S. van de Gevel, I. Elzenga-Claesen, and R. van Furth. 1979. Suspension cultures of mononuclear phagocytes in the Teflon culture bag. Cell. Immunol. 42:208. van Ginkel, C. J. W., L. Tho ¨rig, J. Thompson, J. I. H. Oh, and W. van Aken. 1979. Enhancement of generation of monocyte tissue thromboplastin by bacterial phagocytosis: possible pathway for fibrin formation on infected vegetations in bacterial endocarditis. Infect. Immun. 25:388–395. van Ginkel, C. J. W., W. G. van Aken, J. I. H. Oh, and J. Vreeken. 1977. Stimulation of monocyte procoagulant activity by adherence to different surfaces. Br. J. Haematol. 37:25–45.
