Encapsulated Haemophilus influenzae Types a, c, and d

Transcription

1 INFECTION AND IMMUNITY, Mar. 1982, p Vol. 35, No /82/3759-5$2./ Participation of Complement in Host Defense Against Encapsulated Haemophilus influenzae Types a, c, and d C. JAMES CORRALL,' JERRY A. WINKELSTEIN,2 AND E. RICHARD MOXONl* Eudowood Division ofinfectious Diseases' and Eudowood Division ofimmunology,2 Departments ofpediatrics and Microbiology, The Johns Hopkins University School of Medicine, Baltimore, Maryland 2125 Received 3 July 1981/Accepted 27 October 1981 Each of the serotypes of Haemophilus influenzae (types a to f) may colonize the respiratory tract of humans, but only type b strains commonly cause invasive systemic infections. To investigate the role of complement in immunity to encapsulated non-type b strains, rats were depleted of C3 with cobra venom factor and challenged with representative serotypes of H. influenzae (type a, b, c, or d) by different routes. After intravenous challenge, rats depleted of C3 had a greater incidence and magnitude of bacteremia with each of the serotypes when compared with normal controls. Intraperitoneal inoculation of type b organisms resulted in meningitis in normal and C3-depleted rats, but only C3-depleted, and not normal, rats developed meningitis after inoculation of serotype a, c, or d. In contrast to systemic inoculation, intranasal challenge with the different serotypes resulted in bloodstream invasion and meningitis only after challenge with type b organisms. These data suggest that complement plays a significant role in immunity to encapsulated, non-type b H. influenzae through its effect on bloodstream clearance. Haemophilus influenzae (Hi) is a common cause of life-threatening infections in children. At least six different capsular serotypes exist (a to f), and although each of the serotypes may colonize respiratory mucosa, only type b strains commonly cause systemic infections (3, 4, 9). The basis for the unique proclivity of type b organisms to cause invasive disease is not known. The complement system plays an important role in host defense against a variety of bacterial, fungal, and viral infections. Previous studies have shown that infant rats depleted of C3 by cobra venom factor (CoVF) are more susceptible to Hi type b bacteremia and meningitis than are control animals (2). Infant rats offer a biologically relevant model of disease in humans since after intranasal (i.n.) challenge with various serotypes of Hi, induction of bacteremia and meningitis occurs only with strains of serotype b, but not with other encapsulated serotypes (7a). One possible basis for the relative pathogenicity of different Hi serotypes may be their differing interaction with the complement system. The following experiments were performed to determine whether C3 depletion in rats altered their susceptibility to infection with encapsulated Hi other than type b. MATERIALS AND METHODS Animals. The 9-day-old (average weight, 13 g) and 5-day-old newborn COBS/CD albino Sprague Dawley rats were purchased from Charles River Breeding Laboratories, Inc., Wilmington, Mass., and maintained as previously described (7). CoVF. Purified CoVF obtained from Naja naja venom was purchased from Cordis Laboratories, Miami, Fla. The CoVF was reconstituted in sterile saline to a concentration of 1 U/ml and frozen at -7 C until use. C3 depletion. The 9-day-old rats were injected intraperitoneally (i.p.) with either 125 U of CoVF per kg of body weight or an equal volume of sterile saline. The 5-day-old infant rats were injected i.p. with either 3 U of CoVF per kg of body weight or an equal volume of sterile saline. C3 determinations. Serum was obtained by serial tail bleeding from five rats at each time point, pooled, and stored at -7 C. C3 titers were deteminined simultaneously on all samples by the functional hemolytic assay of Shin and Mayer (1). Bacteria. The strains of encapsulated Hi types a 759 (strain Morgan), b (strain Simpkin), c (ATCC strain 97), and d (ATCC strain 98) and unencapsulated Hi (strains S and U-1) have been previously described (7a), as have the methods of growth (7), quantitation (7), and i.p. (11) or i.n. (7a) inoculation. All animals were inoculated with bacteria 3 h after treatment with either CoVF or sterile saline. Intravenous (i.v.) inoculation was performed in the 9-day-old rats via the lateral tail vein (7a). The 5-day-old newborn rats were inoculated by the i.p. or i.n. route. Sampling of blood and CSF. Cerebrospinal fluid (CSF) was collected by a previously described method (7a). Blood cultures were performed in 9-day-old rats by obtaining blood by intracardiac puncture; both.1 ml of undiluted blood and.1 ml of a 1:1 dilution

2 76 CORRALL, WINKELSTEIN, AND MOXON I- 'n _O C 7 6 a 5 c u i ) -_/ 1 2t Days After CoVF Administration FIG. 1. Effect of CoVF treatment on serum C3 titer of 3-day-old CoVF-treated rats (A) and control rats (A) and among 5-day-old CoVF-treated rats () and control rats (). The arrows signify the injection of CoVF. were plated. Blood cultures were performed in 5-dayold rats by tail vein puncture with.1 ml of undiluted blood plated as previously described (7a). Statistical analysis. Comparison of the incidence of bacteremia and meningitis was performed by the X2 test with Yates continuity correction, and the magnitude of bacteremia and the concentration of bacteria in CSF samples were compared by Student two-tailed t test (1). RESULTS Effect of CoVF on C3 titers. The serum C3 titer fell rapidly to less than 1%o of the control value in the 9-day-old weanling rats within 3 h after the CoVF administration and remained at this level for 5 days (Fig. 1). In 5-day-old rats, the serum C3 titers fell to less than 2% of control values 3 h after CoVF treatment and then dropped to approximately 1% of control values 1 day after CoVF treatment and remained markedly decreased through the fifth day after CoVF treatment before beginning to rise. Effect of complement depletion on the incidence and magnitude of bacteremia and meningitis. At 3 h after injection with either CoVF or saline, rats aged 9 days were challenged i.v. with encapsulated type a (Hia), b (Hib), c (Hic), or d (Hid) or non-encapsulated strain U-1 or S. The incidence and magnitude of bacteremia were significantly greater in C3-deficient animals challenged with Hic or Hid than in control rats (Table 1). Although the C3-depleted rats challenged with Hia had a higher incidence of bacteremia at 3 h than did control animals, the difference was not significant. Non-encapsulated strains (U-1 and S) did not cause bacteremia in either C3-depleted or control rats. The C3-depleted animals challenged with Hic had a higher magnitude of bacteremia at 6 h than they did at 3 h, a result similar to that obtained in C3-depleted animals challenged with Hib. Thus, after i.v. challenge with each of the four serotypes, complementdepleted rats, compared with controls, were more susceptible to bacteremia. To study the effect of complement depletion on the development of meningitis, 5-day-old newborn rats were inoculated i.p. Figure 2 shows the incidence of bacteremia and meningitis and the magnitude of bacteremia in C3- depleted and control rats 24 h after inoculation. Depletion of C3 enhanced the occurrence of bacteremia due to type a, c, or d. Meningitis was not observed in control rats inoculated with encapsulated strains other than type b. In contrast, more than 5% of C3-depleted rats challenged with types a, c, and d had meningitis as evidenced by inflammatory cells, visible organisms upon a Gram stain, and positive cultures of 6 o- 2 E4.Q CD g 3 QM ol Type a Type b Type c Tpa T Ty -1 I& Ie INFECT. IMMUN. S Type d P Value N.S. S.5 S.1 S1 FIG. 2. Effect of C3 depletion on the occurrence of bacteremia and meningitis in groups of six CoVFtreated () or control rats (), aged 5 days, inoculated i.p. with 13 encapsulated Hi type a, b, c, or d and cultured 24 h later. Animals with meningitis are designated -e or *. Solid horizontal bars indicate the mean of each group. The broken horizontal line indicates the lower limit for detection of blood stream bacteria (1 organisms per ml). The P value indicates the significance level for differences in the magnitude of bacteremia between CoVF-treated and control rats. Only surviving animals are depicted. N.S., Not significant. v.

3 VOL. 35, 1982 COMPLEMENT AND NON-TYPE b H. INFLUENZAE 761 TABLE 1. Effect of C3 depletion on the incidence and magnitude of bacteremia after i.v. inoculation of 3- month-old rats with 1' encapsulated (types a, b, c, d) or non-encapsulated Hi Bacteremia at 3 h Bacteremia at 6 h Incidencea (%) bacteremia Incidence' (%) bacteremia Type (strain) Complement status. Geometric mean Geometric mean (bacteria/ml) (bacteria/ml) a Normal () Undetectableb () Undetectable Depleted 3 (21) 1.5 x 11 () Undetectable b Normal 9 (64) 3.4 x (1) 3.1 x 12 Depleted 11 (78) 9. X (1) 3.4 x 13c c Normal () Undetectable () Undetectable Depleted 7 (5)c 9.3 x 1kc 14 (1)c 1.73 x 13c d Normal () Undetectable () Undetectable Depleted 5 (36)c 5.6 x 11 () Undetectable Untypable (S) Normal () Undetectable () Undetectable Depleted () Undetectable () Undetectable Untypable (U-1) Normal () Undetectable () Undetectable Depleted () Undetectable () Undetectable a Number of rats with bacteremia out of a total of 14 rats tested. b Limit of sensitivity is 11 bacteria per ml. c Difference between CoVF-treated and control rats is significant at P c.5. CSF. Deaths occurred in C3-depleted, but not control, rats (two with type c and one each with types b and d). To study the effect of complement depletion on respiratory tract invasion, a route analogous to infection in man, 5-day-old newborn rats were inoculated i.n. Cultures of blood were obtained 6 h after challenge since previous work indicated that, with this inoculum, the majority of rats inoculated with Hib become bacteremic at that time (7a). Bacteremia and meningitis occurred only in those animals challenged with Hi type b; bacteremia or meningitis did not occur in C3- depleted or control rats challenged with Hi type a, c, or d (Table 2). Thus, after i.n. challenge, there was no evidence that complement deple- TABLE 2. Effect of C3 depletion on the incidence and magnitude of bacteremia and meningitis after i.n. challenge of 5-day-old rats with 1' encapsulated (types a, b, c, d) or non-encapsulated Hi Incidence of Geometric mean GnidncoomeFatericmal Type (strain) Complement status bacteremia bacteremia mencidengt (teoncnw (% (bacteria/ml) (bacteria/mi)n a Normal /14 () Undetectablec /14 () Undetectable Depleted /8 () Undetectable /8 () Undetectable b Normal 8/12 (67) 9.8 x 12 3/12 (25) 1.36 x 12 Depleted 8/11 (73) 3. x 13 5/11 (45) 3.15 x 12 c Normal /15 () Undetectable /15 () Undetectable Depleted /1 () Undetectable /1 () Undetectable d Normal /16 () Undetectable /16 () Undetectable Depleted /17 () Undetectable /17 () Undetectable Untypable (S) Normal on (o) Undetectable on (o) Undetectable Depleted /17 () Undetectable /17 () Undetectable Untypable (U-1) Normal /13 () Undetectable /13 () Undetectable Depleted /1 () Undetectable /1 () Undetectable Number of rats with bacteremia/total number of rats tested. " Number of rats with meningitis/total number of rats tested. c Limit of sensitivity is 12 bacteria per ml.

4 762 CORRALL, WINKELSTEIN, AND MOXON tion enhanced the development of bacteremia or meningitis with serotype a, c, or d. DISCUSSION Previous studies showed that complement is important in host defense against Hib in an experimental infection of rats (2). The present studies indicate that complement also plays a significant role in the host's defense against encapsulated Hi types a, c, and d. Complement may be activated by Hi through the alternative or classical pathways (5, 8). The interaction of complement components with Hi may promote bacterial killing through either serum bactericidal activity, involving participation of the terminal complement components C5 to C9, or through opsonization leading to phagocytic ingestion. We inoculated strains of Hi representative of several different serotypes into rats in which C3 levels were substantially reduced by treatment with cobra venom. This allowed evaluation of the in vivo contribution to host defense of C3 to C9-mediated activities. The rat model of invasive Hi infection has been extensively studied and shown to be a biologically relevant model of the human infection. Previous studies established that the type b strain used in the present studies was substantially more virulent for the normal rat than the heterologous type a, c, or d strains of Hi (7a). Using different routes of inoculation (i.v., i.p., or i.n.), we sought to determine both the effect of C3 depletion on the virulence of non-type b encapsulated serotypes and the stage in the evolution of the invasive process in which complement manifested its effect. After i.v. inoculation, complement depletion enhanced the occurrence of bacteremia for each of the serotypes. Based on our previous studies, this result was anticipated for the type b strain. The data for the type a, c, and d organisms, however, indicated an important role for complement in bloodstream clearance. The effect of complement depletion on the pathogenicity of the type c strain was particularly striking. By 6 h, the frequency and magnitude of bacteremia were similar to those observed in rats inoculated with the type b strain. As anticipated from the above studies that used the i.v. challenge route, bacteremia was also substantially increased in complement-depleted rats inoculated i.p. However, these studies were principally devised to determine whether meningitis resulted. The occurrence of meningitis in complement-depleted rats challenged with the type a, c, and d serotypes was in marked contrast to the control animals challenged with these same serotypes. The fact that those complement-depleted animals who had a sustained high level bacteremia with the type a, INFECT. IMMUN. c, or d organisms also had meningitis is consistent with previous data with type b strains where bacteremia of equal to, or greater than, 13 bacteria per ml was found to be necessary, but not sufficient, to result in meningeal invasion (6). These data do not indicate whether complement depletion directly affected central nervous system invasion or whether the occurrence of meningitis was a function of decreased clearance of bacteria in the bloodstream. Our previous studies with another type b strain (2) strongly favored the latter possibility but do not necessarily pertain to other type b strains or to heterologous serotypes. In contrast to the effect of complement depletion in rats inoculated systemically, no detectable effect was observed in cobra venom-treated rats after i.n. challenge. Thus, C3 depletion did not seem to be a determinant of bloodstream invasion from the nasopharynx. These findings were of particular interest vis-a-vis a previous study with a different type b strain (Eagan) where cobra venom treatment significantly enhanced invasiveness (2). Thus, quantitative differences in the amount of capsule, as well as determinants of pathogenicity other than serotype, may be critical. Studies with both wild and isogenic or transformed strains are clearly indicated. In summary, these studies extend our previous investigations to indicate an important role for complement in host defense against encapsulated Hi other than type b. The effect of complement depletion seemed to result from impaired intravascular clearance of bacteria rather than enhanced bloodstream invasion from the nasopharynx. Type b strains possess unique virulence characteristics both as regards naturally occurring invasive infection of humans and experimental infections of rats. Differences in the pathogenicity of encapsulated (type b and nontype b) Hi may be determined at least in part by differences in their interaction with the complement system which plays an important role in the host's defense against bacteremia and meningitis. ACKNOWLEGMENTS We thank Diana M. Corral for assistance with animal inoculation and fluid sampling and Gail Nadolny for manuscript preparation. This research was supported in part by Public Health Service grants AI-7195, NS-12554, and AI from the National Institutes of Health and in part by the Board of the Hospital for Consumptives of Maryland (Eudowood). E.R.M. is a recipient of Research Career Development award 5 K4 AI-3 from the National Institute of Allergy and Infectious Diseases. LITERATURE CITED 1. Armitage, P Statistical methods in medical research, p Blackwell Scientific Publications, London.

5 VOL. 35, 1982 COMPLEMENT AND NON-TYPE b H. INFLUENZAE Crosson, F. J., J. A. Winkelstein, and E. R. Moxon Participation of complement in the nonimmune host defense against experimental Haemophilus influenzae type b septicemia and meningitis. Infect. Immun. 14: Dawson, B., and K. mn Incidence and type distribution of capsulated H. irnfluenzae strains. Br. Med. J. vol.: Greene, G. R Meningitis due to Haemophilus influenzae other than type b: case report and review of the literature. Pediatrics 62: Johnston, R. B., P. Anderson, and S. Newman Opsonization and phagocytosis of Haemophilus influenzae, p In S. H. Sell and D. T. Karzon (ed.), Haemophilus influenzae. Vanderbilt University Press, Nashville, Tenn. 6. Moxon, E. R., and P. T. Ostrow Haemophilus influenzae meningitis in infant rats. Role of bacteremia in pathogenesis of age-dependent inflammtory responses in cerebrospinal fluid. J. Infect. Dis. 135: Moxon, E. R., A. L. Smith, D. R. Averlll, and D. H. Smith Haemophilus influenzae meningitis in infant rats after intranasal inoculation. J. Infect. Dis. 129: a.Mozon, E. R., and K. A. Vaughn The type b capsular polysaccharide as a virulence determinant of Haemophilus injuenzae: studies using clinical isolates and laboratory transformants. J. Infect. Dis. 143: Quim, P. H., F. J. Croon, J. A. Winkelstein, ad E. R. Moxon Activation of the alternative complement pathway by Haemophilus influenzae type b. Infect. Immun. 16: Sell, S. H., D. J. Turner, and C. F. Federspck Natural infections with Haemophilus influenzae in children. I. Types identified, p In S. H. Sell and D. T. Karzon (ed.), Haemophilus influenzae. Vanderbilt University Press, Nashville, Tenn. 1. Shin, H. S., and M. M. Mayer The third component of the guinea pig complement system. Biochemistry 7: Smith, A. L., D. H. Smith, D. R. AverIl, J. Marino, ande. R. Moxon Production of Haemophilus influenzae b meningitis in infant rats by intraperitoneal inoculation. Infect. Inmun. 8: Downloaded from on October 31, 218 by guest