点击显示 收起
Department of Pediatrics, Divisions of Infectious Diseases and Neonatology and Department of Pathology, School of Medicine, University of California at San Diego, La Jolla
We examined the virulence role of group B Streptococcus (GBS) -hemolysin/cytolysin (h/c) in a neonatal-rabbit model of GBS pulmonary infection. Rabbits infected intratracheally with wild-type (wt) GBS developed focal pneumonia and, by 18 h after infection, had 100-fold more bacteria in lung tissue than did rabbits infected with a h/c mutant. Mortality (40% vs. 0%), development of bacteremia, and mean bacterial blood counts were all significantly higher in the rabbits challenged with wt GBS than in those challenged with the h/c mutant. Lung compliance during mechanical ventilation was impaired after injection of wt GBS but not after injection of the h/c mutant strain. This work, to our knowledge, provides the first in vivo evidence for a critical role of the h/c toxin in GBS neonatal pneumonia and in the breakdown of the pulmonary barrier to systemic infection.
Group B Streptococcus (GBS) is the leading cause of serious bacterial infections in human newborns. Between 15% and 30% of healthy women are colonized with GBS in the genitourinary or lower gastrointestinal tract [1]. Vertical transmission of GBS from mother to infant can occur after ascending infection of the placental membranes or aspiration of contaminated vaginal fluids during delivery. Early-onset GBS disease is typically heralded by the development of respiratory symptoms in the first several hours of life, reflecting an initial pulmonary focus of infection. Pathologic changes in neonatal GBS pneumonia include bacterial and neutrophilic infiltrates, intraalveolar edema, hyaline membranes, focal atelectasis, and evidence of pulmonary epithelial and endothelial cell infiltrates [2]. In nearly all cases, the bacterium penetrates lung barriers to reach the bloodstream, which frequently results in complications such as septicemia and meningitis.
The specific molecular determinants that allow GBS to breach the lung barrier and produce systemic infection are unknown. A candidate virulence factor that may play a role in this process is the GBS -hemolysin/cytolysin (h/c). This surface-associated toxin is produced by the vast majority of GBS clinical isolates and generates the zone of hemolysis surrounding colonies plated on blood agar media. The cylE gene in the GBS chromosome is required for h/c production and is sufficient to confer hemolysis when cloned in Escherichia coli [3]. The association between cylE and h/c production has been confirmed in multiple GBS strains of varying serotypes and by single-gene complementation to restore wild-type (wt) toxin production levels [3]. In tissue culture studies, GBS h/c has been shown to produce direct cytolytic injury to human pulmonary epithelium and endothelium by forming pores in the target cell membrane [4]. Subcytolytic concentrations of h/c can also promote GBS invasion of cultured lung epithelial cells and trigger the release of the neutrophil chemokine interleukin-8 (IL-8) [5]. The phospholipid dipalmotyl phosphatidylcholine (DPPC), a major component of human surfactant, inhibits GBS h/cmediated cytotoxicity, invasion, and IL-8 activation, providing a potential link to the increased susceptibility of premature, surfactant-deficient neonates to lung damage and sepsis from GBS infection [4, 5]. GBS h/c production has been clearly associated with increased virulence in intravenous challenge of mice and rabbits [6, 7], but these experiments have bypassed the lung and, thereby, the natural route of neonatal infection.
In this present study, we have developed a novel neonatal-rabbit model of early-onset GBS pneumonia and sepsis through direct transcutaneous instillation into the trachea. Using a wt strain and its isogenic cylE knockout h/c-deficient (h/c) mutant, we specifically examine the contribution of the h/c toxin to GBS pneumonia, lung function, and disruption of the pulmonary barrier to bloodstream infection.
Materials and methods.
The wt GBS strain used in these studies was NCTC10/84 (1169-NT1), a serotype V isolate from the blood of a septic neonate [8]. The corresponding h/c mutant NCTC:cylEcat was generated by precise in-frame allelic exchange mutagenesis [3]. GBS strains were grown to logarithmic phase (OD600 = 0.4, or 1 × 108 cfu/mL) in Todd-Hewitt broth (THB), washed in PBS, then resuspended in PBS at the appropriate concentration for use in the infectious challenges.
Pregnant New Zealand white rabbit does were obtained from Irish Farms and were acclimated for 57 days before term (day 31). Does were allowed to deliver litters normally and to nurse pups for up to 24 h. Pups were then removed, kept under a radiant warmer bed, and sedated with ketamine/xylazine (25/8 mg/kg) subcutaneously. The mean ± SD weight of rabbit pups used for these studies was 0.059 ± 0.017 kg, with a range of 0.0250.106 kg. Skin was sterilized with isopropyl alcohol, and the GBS suspension (5 mL/kg) was injected percutaneously into the trachea. The standard dose was 3 × 107 cfu/kg (range, 2.5 × 1075.0 × 107 cfu/kg); a higher inoculum of 3 × 109 cfu/kg was used for a small number of additional histopathologic studies. Treated rabbits were immediately provided with supplemental oxygen, allowed to recover on the warmer bed, and then returned to the doe for 18 h. Rabbit pups were observed for respiratory function, color, and signs of distress, and any that appeared to be in extremis before the end of the study were euthanized. At the 18-h end point, surviving rabbit pups were sedated with ketamine/xylazine and anesthetized with pentobarbital (10 mg/kg) intraperitoneally before placement of a blunt 18-gauge needle as a tracheostomy tube. Rabbits were then transferred to a specially built Plexiglas plethysmograph (Ridout Plastics) warmed to 39°C and were ventilated in parallel (InfantStar 950; Infrasonics) under 100% O2. Ventilatory settings were as follows: rate, 30 breaths/min; maximum inspiratory pressure, 15 cm H2O; and end expiratory pressure, 2 cm H2O. Flow into and out of the plethysmograph was measured with a pneumotachograph and pressure transducer (model LCVR ± 2 cm H2O; Celesco). Airway pressure was measured with a pressure transducer (model LCVR ± 50 cm H2O; Celesco). Respiratory system compliance (Crs) was calculated by integrating the flow data and dividing the integrated flow by the change in airway pressure. After 25 min of ventilation, rabbits were euthanized with an overdose of pentobarbital. The chest cavity was opened, the right bronchus was ligated with a suture, and the right lung was excised and homogenized using a Tissue Tearor (BioSpec Products). A cardiac puncture was performed for blood collection. Bacterial counts were determined by plating serial dilutions of lung homogenate or blood on THB agar for overnight incubation and enumeration of colony-forming units. The left lung was fixed at 20 cm H2O of pressure in 3.7% formalin and embedded in paraffin, and the sections were stained with hematoxylin-eosin for histologic analysis. All aspects of this protocol were reviewed and approved by the Animal Subjects Committee at the University of California at San Diego. Statistical analysis was performed by nonparametric Mann-Whitney U test, with P < .05 considered to be significant.
Results.
Rabbit pups in their first day of life were infected intratracheally with an equivalent inoculum (3 × 107 cfu/kg) of wt GBS (n = 23), the isogenic h/c mutant (n = 15), or saline alone (control, n = 4). Within 6 h of injection, several rabbits infected with wt GBS showed tachypnea and labored breathing, clinical findings that were absent in the h/c mutantinfected rabbits and controls. By 18 h after injection, 9/23 (39%) of the wt GBSinfected rabbits had died, and the remainder of the treatment group exhibited obvious cyanosis, lethargy, and respiratory distress (figure 1A). In contrast, no rabbits challenged with the h/c mutant or saline alone died, and all were normal in appearance, with minimal or no respiratory symptoms at 18 h after injection (figure 1A). Quantitative cultures for GBS were performed on lung tissue; all surviving rabbits were euthanized, and blood was collected. Compared with wt GBSinfected rabbits, rabbits challenged with the GBS h/c mutant had mean GBS counts in the lungs that were 100-fold lower (>1.0 × 106 vs. 1.0 × 104 cfu/g lung tissue; P < .01) and mean levels of bacteremia that were 100-fold lower (>1.0 × 104 vs. <1.0 × 102 cfu/mL blood; P < .05) (figure 1B). Forty percent of rabbits challenged with the h/c mutant did not have detectable (<20 cfu/mL) GBS in the bloodstream. These results demonstrate that the h/c toxin contributes to GBS proliferation in lung tissues, promotes penetration of the pulmonary and epithelial barriers, and produces high lethality in the neonatal infection model.
Sections of lung tissues from wt GBSinfected, h/c mutantinfected, and control rabbits were examined at 18 h after injection for histopathologic changes. Scattered foci of neutrophil infiltration were observed throughout the alveolar epithelium of wt GBSinfected rabbits (figure 2A, lower inoculum), whereas such changes were extremely rare in the rabbits challenged with the h/c mutant (figure 2A, lower inoculum) and in saline-treated controls (data not shown). On histologic examination, alveolar expansion was notably diminished in the wt GBSinfected rabbit lungs, compared with those from rabbits infected with the h/c mutant (figure 2A, lower inoculum). However, diffuse or severe bronchopneumonia was not observed in any group, and the alveolar architecture appeared relatively intact. Because wt GBSinfected rabbits consistently developed high-grade bacteremia, our results demonstrate that severe epithelial and endothelial barrier injury is not a prerequisite for GBS systemic dissemination. Severe early-onset GBS pneumonia can be associated with very high bacterial densities (1 × 1091 × 1011 cfu/g lung tissue) [9]. Therefore, for comparison, we infected an additional group of rabbit pups with a 100-fold higher GBS inoculum ( 3 × 109 cfu/kg body weight). In these experiments, wt GBSinfected rabbits developed severe respiratory distress and cyanosis by 6 h after injection, at which point rabbits were euthanized for lung histopathologic analysis. Numerous foci of bacterial infiltration were present (figure 2A, higher inoculum), with clear bronchopneumonia and dense neutrophilic infiltration of the alveoli. Evidence of neutrophilic infiltration was also present in rabbits infected with the higher dose of the h/c mutant (figure 2A), although at a markedly reduced severity compared with that in the wt GBSinfected rabbits.
To assess the contribution of the h/c toxin to lung dysfunction in early-onset GBS infection, rabbit pups surviving to 18 h after injection were sedated, tracheostomized, and ventilated in a warmed plethysmograph. Crs was calculated from volume and pressure changes during ventilation and was analyzed by 2 methods. We first calculated the change in Crs over time (slope) through the initial ventilation phase (010 min), during which a rapid increase in ventilator-induced alveolar expansion should occur [10, 11]. We found that rabbits infected with wt GBS had a significantly reduced initial Crs slope, compared with that measured for rabbits infected with the h/c mutant (figure 2B). An additional compliance measurement was taken for several rabbits at the end of the 25-min ventilation period. Whereas rabbits challenged with the h/c mutant were able to increase Crs to a level comparable to that in saline-treated controls, rabbits infected with wt GBS failed to do so (figure 2C). These data indicate that, even though only limited foci of focal pneumonia were observed at the 18-h time point, the expected alveolar expansion with mechanical ventilation was distinctly impaired in the wt GBSinfected rabbits. Our findings may reflect a direct effect of the h/c toxin on pulmonary function or, alternatively, an indirect effect related to the increased numbers of wt versus mutant GBS resulting from h/c-mediated resistance to immunological clearance.
Discussion.
The pathogenesis of GBS pneumonia and the consequent breakdown of the lung barrier to systemic infection are not well understood. The histopathologic changes observed in autopsy studies and in a primate model of intraamniotic infection [9] reveal widespread injury to lung cells, which could be the direct result of bacterial toxins or the indirect result of the induced inflammatory response. Moreover, GBS can invade alveolar epithelial and pulmonary endothelial cells, a process that could facilitate bloodstream entry even through intact cellular barriers [12, 13]. In premature infants, immature host defenses and the lack of pulmonary surfactant represent additional predisposing factors leading to a greatly increased risk of severe GBS pneumonia and systemic disease.
We believe that our neonatal model of GBS disease can add to our understanding of the common clinical scenario in which the neonate aspirates GBS-contaminated fluids in utero or during delivery and then gradually develops increasing respiratory symptomatology within the first several hours of life. Our results indicate that the h/c toxin plays a critical role in GBS-induced pulmonary dysfunction and in the ability of the pathogen to penetrate the lung barrier, establish bacteremia, and produce high mortality. The deleterious effects of h/c in the initial pulmonary stages of neonatal GBS infection likely reflect a combination of the toxin's cytolytic, proinvasive, and proinflammatory properties, with the ultimate balance of the individual virulence phenotypes dependent on the magnitude and tempo of bacterial exposure.
Our discovery of a key role for the h/c mutant in the pathogenesis of early-onset pneumonia may explain the observed benefits of surfactant preparations in reducing GBS infection in ventilated animal models [14] and in retrospective studies of human infants receiving empiric replacement therapy [15]. Our model will facilitate future studies examining the pathophysiology of neonatal pneumonia and will help to determine whether DPPC or other anti-h/c therapies could benefit infants experiencing, or at high risk of developing, early-onset GBS infection.
References
1. Moore MR, Schrag SJ, Schuchat A. Effects of intrapartum antimicrobial prophylaxis for prevention of group-B-streptococcal disease on the incidence and ecology of early-onset neonatal sepsis. Lancet Infect Dis 2003; 3:20113. First citation in article
2. Vollman JH, Smith WL, Ballard ET, Light IJ. Early onset group B streptococcal disease: clinical, roentgenographic, and pathologic features. J Pediatr 1976; 89:199203. First citation in article
3. Pritzlaff CA, Chang JC, Kuo SP, Tamura GS, Rubens CE, Nizet V. Genetic basis for the -haemolytic/cytolytic activity of group B Streptococcus. Mol Microbiol 2001; 39:23647. First citation in article
4. Nizet V, Gibson RL, Chi EY, Framson PE, Hulse M, Rubens CE. Group B streptococcal beta-hemolysin expression is associated with injury of lung epithelial cells. Infect Immun 1996; 64:381826. First citation in article
5. Doran KS, Chang JC, Benoit VM, Eckmann L, Nizet V. Group B streptococcal -hemolysin/cytolysin promotes invasion of human lung epithelial cells and the release of interleukin-8. J Infect Dis 2002; 185:196203. First citation in article
6. Liu GY, Doran KS, Lawrence T, et al. Sword and shield: linked group B streptococcal -hemolysin/cytolysin and carotenoid pigment function to subvert host phagocyte defense. Proc Natl Acad Sci USA 2004; 101:144916. First citation in article
7. Ring A, Depnering C, Pohl J, Nizet V, Shenep JL, Stremmel W. Synergistic action of nitric oxide release from murine macrophages caused by group B streptococcal cell wall and -hemolysin/cytolysin. J Infect Dis 2002; 186:151821. First citation in article
8. Wilkinson HW. Nontypable group B streptococci isolated from human sources. J Clin Microbiol 1977; 6:1834. First citation in article
9. Rubens CE, Raff HV, Jackson JC, Chi EY, Bielitzki JT, Hillier SL. Pathophysiology and histopathology of group B streptococcal sepsis in Macaca nemestrina primates induced after intraamniotic inoculation: evidence for bacterial cellular invasion. J Infect Dis 1991; 164:32030. First citation in article
10. Davis JM, Veness-Meehan K, Notter RH, Bhutani VK, Kendig JW, Shapiro DL. Changes in pulmonary mechanics after the administration of surfactant to infants with respiratory distress syndrome. N Engl J Med 1988; 319:4769. First citation in article
11. Yang QH, Lai-Fook SJ. Effect of lung inflation on regional lung expansion in supine and prone rabbits. J Appl Physiol 1991; 71:7682. First citation in article
12. Rubens CE, Smith S, Hulse M, Chi EY, van Belle G. Respiratory epithelial cell invasion by group B streptococci. Infect Immun 1992; 60:515763. First citation in article
13. Gibson RL, Lee MK, Soderland C, Chi EY, Rubens CE. Group B streptococci invade endothelial cells: type III capsular polysaccharide attenuates invasion. Infect Immun 1993; 61:47885. First citation in article
14. Herting E, Sun B, Jarstrand C, Curstedt T, Robertson B. Surfactant improves lung function and mitigates bacterial growth in immature ventilated rabbits with experimentally induced neonatal group B streptococcal pneumonia. Arch Dis Child Fetal Neonatal Ed 1997; 76:F38. First citation in article
15. Herting E, Gefeller O, Land M, et al. Surfactant treatment of neonates with respiratory failure and group B streptococcal infection. Pediatrics 2000; 106:95764; discussion 1135. First citation in article