Differential Pneumococcal Growth Features in Severe Invasive Disease Manifestations

Abstract

The nasopharyngeal commensal Streptococcus pneumoniae can become invasive and cause metastatic infection. This requires the pneumococcus to have the ability to adapt, grow, and reside in diverse host environments. Therefore, we studied whether the likelihood of severe disease manifestations was related to pneumococcal growth kinetics. For 383 S. pneumoniae blood isolates and 25 experimental mutants, we observed highly reproducible growth curves in nutrient-rich medium. The derived growth features were lag time, maximum growth rate, maximum density, and stationary-phase time before lysis. First, the pathogenicity of each growth feature was probed by comparing isolates from patients with and without marked preexisting comorbidity. Then, growth features were related to the propensity of causing severe manifestations of invasive pneumococcal disease (IPD). A high maximum bacterial density was the most pronounced pathogenic growth feature, which was also an independent predictor of 30-day mortality (P = 0.03). Serotypes with an epidemiologically higher propensity for causing meningitis displayed a relatively high maximum density (P < 0.005) and a short stationary phase (P < 0.005). Correspondingly, isolates from patients diagnosed with meningitis showed an especially high maximum density and short stationary phase compared to isolates from the same serotype that had caused uncomplicated bacteremic pneumonia. In contrast, empyema-associated strains were characterized by a relatively long lag phase (P < 0.0005), and slower growth (P < 0.005). The course and dissemination of IPD may partly be attributable to the pneumococcal growth features involved. If confirmed, we should tailor the prevention and treatment strategies for the different infection sites that can complicate IPD. IMPORTANCE Streptococcus pneumoniae is a leading infectious cause of deaths worldwide. To understand the course and outcome of pneumococcal infection, most research has focused on the host and its response to contain bacterial growth. However, bacterial epidemiology suggest that certain pneumococcal serotypes are particularly prone to causing complicated infections. Therefore, we took the bacterial point of view, simply examining in vitro growth features for hundreds of pneumococcal blood isolates. Their growth curves were very reproducible. Certain poles of pneumococcal growth features were indeed associated with specific clinical manifestations like meningitis or pleural empyema. This indicates that bacterial growth style potentially affects the progression of infection. Further research on bacterial growth and adaptation to different host environments may therefore provide key insight into pathogenesis of complicated invasive disease. Such knowledge could lead to more tailored vaccine targets or therapeutic approaches to reduce the million deaths that are caused by pneumococcal disease every year.

Keywords: S. pneumoniae; bacterial growth; empyema; meningitis; pathogenesis; serotype.

FIG 1

Intrinsic growth features. For six blood culture isolates of S. pneumoniae (in different colors), six repeat measurements are displayed as growth curves with duplicates on three separate days (A) and their derived growth features as box plots (B). Visualizations are downloaded from the online interactive Growth Viewer database (https://fairdb.tenwiseservice.nl/GrowthViewer/). OD₆₀₀, optical density at 600 nm.

FIG 2

Pathogenic poles of growth features. Derived growth features for 367 S. pneumoniae blood culture isolates that had infected previously relatively healthy adults (n = 78) compared to relatively frail adults (n = 289) (Charlson comorbidity index score ≤ 2 versus >2). Each dot represents six repeat measurements for one isolate from a corresponding patient. The displayed growth features are lag phase (A), growth speed (B), maximum density (C), and stationary phase (D). OD₆₂₀, optical density at 620 nm.

FIG 3

Maximum density and invasive pneumococcal disease (IPD) mortality. Maximum density reached in vitro by S. pneumoniae blood culture isolates from patients who died and survived the episode of invasive pneumococcal disease. Each dot represents six repeat measurements for one isolate from a corresponding patient. OD₆₂₀, optical density at 620 nm; **, P < 0.005.

FIG 4

Serotypes’ epidemiological propensity for causing meningitis. Derived growth features for 383 S. pneumoniae blood culture isolates stratified according to the serotypes’ epidemiological propensity for causing meningitis (high, n = 108, pink; intermediate, n = 234, white; low, n = 41, green). Distributions are visualized as either Tukey box plots with mean (+) and outliers (·). The displayed growth features are lag phase (A), growth speed (B), maximum density (C), and stationary phase (D). The dashed horizontal line indicates a trend across the three categories. OD₆₂₀, optical density at 620 nm; *, P < 0.017 (Bonferroni corrected significance threshold); **, P < 0.005; ***, P < 0.0005.

FIG 5

Within-serotype differentials of meningitis cases. Derived growth features for S. pneumoniae blood culture isolates. The data are displayed for meningitis cases that show marked polarity for a growth feature within their serotype. Gray boxes represent isolates from patients with uncomplicated pneumonia caused by that serotype (from minimum to maximum, with horizontal bar at mean respiration median value), while pink dots represent isolates from patients with confirmed meningitis. The displayed growth features are lag phase (A), maximum density (B), and stationary phase (C). OD₆₂₀, optical density at 620 nm.

FIG 6

Serotypes epidemiologically associated with empyema. Derived growth features for 379 S. pneumoniae blood culture isolates stratified according to the serotypes’ epidemiological association with pleural empyema (yes, n = 204, blue; no, n = 175, yellow). Distributions are visualized in either Tukey box plots with mean (+) and outliers (·). The displayed growth features are lag phase (A), growth speed (B), maximum density (C), and stationary phase (D). OD₆₂₀, optical density at 620 nm; *, P < 0.017 (Bonferroni corrected significance threshold); **, P < 0.005; ***, P < 0.0005.