Mechanism of Antibody-Mediated Reduction of Nasopharyngeal Colonization by Haemophilus influenzae Type b Studied in an Infant Rat Model

0 The Journal of Infectious Diseases 1996; 174:1337-40 1996 by The University of Chicago. All rights reserved. 0022-1899/96/7406-0028$01.00 1 Received 4 December 1995; revised 26 June 1996. Presented in part: European Research Conference, Immunology of Infec tions, Castelvecchio Pascoli, Italy, 30 September to 5 October 1995. Experimental procedures for use with animals were reviewed by the National Public Health Institute's Laboratory Animal Committee and approved by the Provincial Board. Financial support: Finnish Cultural Foundation. Institute , Mannerheimintie 166, SF-00300 Helsinki , Finland 2 National Public Health Institute , Helsinki , Finland; Department of Medical Microbiology, University of Amsterdam , Netherlands; Lederle-Praxis Biologicals, Rochester, New York Mechanism of Antibody-Mediated Reduction o f Nasopharyngeal Colonization by Haemophilus infiuenzae Type b Studied in an Infant R a t Model The mechanism of antibody-mediated reduction of Haemophilus influenzae type b (Hib) carriage was studied in the infant rat colonization model. Monoclonal Hib polysaccharide (PS) antibody (MAb) given intranasally or intraperitoneally and human secretory anti-Hib PS IgA given intranasally inhibited colonization by Hib during the entire follow-up period (2-48 h after challenge) but did not affect colonization by Hi, a noncapsulated variant of Hib. Ftab'), fragments, prepared from the MAb or from human serum anti-Hib IgG reduced Hib colonization as efficiently as the unc1eaved molecules. Complement depletion by cobra venom treatment had no effect on the antibody-mediated reduction of Hib colonization. These results indicate that Fc-mediated activities of immunoglobulins are not essential in the reduction of Hib colonization. Instead, antibodies to Hib most likely reduce colonization by a direct effect on growth of the bacteria or their adherence to the nasopharyngeal mucosa. - Hib cell with the capsular antibodies may sterically hinder the association of Hib to the epithelium and prevent the growth of Hib [5]. However, complement and phagocytosis may also be involved in the clearance of Hib from the mucosal surface. In the present study, we used the infant rat model to study the mechanism of anti-Hib PS-mediated inhibition of Hib col onization. We compared the kinetics of colonization of Hib and its noncapsulated variant on the mucosa in the presence of Hib PS antibodies. We also investigated the role of activities mediated by the Fc part of the antibody molecule in the clear ance of Hib by using the F(ab')2 fraction of the anti-Hib PS antibody in the colonization experiments. Since antibody-medi ated killing of Hib and phagocytosis require complement, the contribution of complement was determined by using comple ment-depleted infant rats. Materials and Methods Outbred, specific pathogen - free albino Wistar rats were ob tained from the Animal Center, University of Helsinki. Hib strain 760705b+ (su-R), isolated from cerebrospinal fluid, and its noncap sulated variant Hi 760705bo (Str") [6] were stored in skim milk at - 70C and cultured for animal inoculations as described [3]. The 2 strains have identical reactivities with monoclonal antibody (MAb) specific for lipopolysaccharide and outer membrane proteins PI, P2, P5, and P6. The outer membrane protein and lipopolysaccha ride profiles on SDS-PAGE are identical. Two lots of anti-Hib PS murine MAb E117-5 [7], containing ~800 and 1200 ILg/mL, respectively, of anti-Hib PS IgG lK were used. A control meningococcal porA MAb was received from G. van den Dobbelsteen (RIVM, Netherlands). A human serum IgG fraction containing 36 ILg/mL anti-Hib PS IgG was made from a pooled sample of serum from 10 adults vaccinated with Hib vac cines (half with a polysaccharide and half with a conjugate vaccine) [3]. As a control, IgG fraction containing no detectable anti-Hib PS was made from a pooled sample of serum from children. After immunization, breast milk containing 4.1 p,g/mL of anti-Hib PS, mainly sIgA of the IgAl isotype, was obtained from a mother who had been vaccinated with a Hib conjugate [3]. Preimmunization milk contained <0.2 p,g/mL anti-Hib PS. Ftab"), fragments were prepared from MAbs and human IgG fractions by pepsin (Merck, Darmstadt, Germany) digestion (pep sin to IgG ratio, 1:2.5 by weight; pepsin to MAb ratio, 1:1 by weight) [8]. Ffab"), fragments were separated from Fe fragments by gel filtration on a Superdex 200 column (Pharmacia LKB Bio technology, Uppsala, Sweden). The purity of the F(ab')z fractions was confirmed by SDS-PAGE and, in the case of Frah'), from the Mab, by EIA using Fab- and Fe-specific conjugates to mouse IgG (A-3682 and A-9309; Sigma, St. Louis). Fe contamination in Ftab"), fraction from the MAb was 2%, as calculated by comparing the titers of the Mab and the Frab"), derived from the Mab by EIA with an Fe-specific conjugate. Cobra venom anticomplementary factor (CoVF) from Naja naja kaouthia (C-8406; Sigma) was administered to infant rats by ip injection (0.2 mL/pup, 20 U of CoVFIlOO g of body weight). The activity of complement in serum of CoVF-treated and control animals (given PBS instead ofCoVF) was determined by a comple ment assay [9]. Results were expressed as titers (reciprocals of dilutions hemolyzing 50% of the sensitized sheep erythrocytes, as estimated by eye). Serum complement titers declined from 16 to < 1 within 3 h after ip injection of CoVF. Bacteria (~2500 cfu) and antibody were administered inl, as described previously [3]. Antibody was administered 3 h before bacterial challenge, or bacteria were mixed with the antibody just before inoculation. Antibody or PBS (0.15-0.2 mL) were given ip 3 h before administration of bacteria. During follow-up after the injection of 40-55 p,g of MAb/pup, the level of serum anti Hib PS remained high (8-12 p,g/mL). Nasal wash samples were obtained from the infant rats at 6,24, and 48 h or only at 48 h after bacterial challenge. The time point of 48 h was chosen because >90% of pups have been shown to be colonized 48 h after an inl inoculation of 2500 cfu of Hib [3]. The samples were cultured as described [3], and Hib and Hi colo nies were counted. Samples with no colonies 100 cfu/mL) were assigned a value of 30 cfu/mL. Student's t test, assuming equal variances, was used to compare the geometric mean concentrations (GMC) of Hib or Hi in the groups of animals. P < .05 was considered statistically significant. Blood samples were taken from the tail vein of unanesthetized pups (pooled samples) or from hearts of CO 2-anesthetized pups (individual samples) to determine anti-Hib PS concentrations by RIA (samples stored at -20C) or complement depletion (samples stored at -70C). Results In the absence of anti-Hib PS antibodies, the GMC of Hib 760705b+ in nasal wash samples increased from .......,2000 cfu/mL at 2 h to 9800 cfu/mL at 6 h after inl bacterial challenge, at which point it started to decrease slowly, being .......,2000 cfu/ml. (...truncated)