Abstract: To elucidate the role of innate immunity in the pathogenesis of meningitis, the effects of cerebrospinal fluid (CSF) and of sera from patients with bacterial and abacterial meningitis were studied on monocytes because of their resemblance in function with brain microglial cells. Monocytes were isolated from the mononuclear fraction of blood of healthy donors. They were incubated in growth medium single and in the presence of purified lipopolysaccharide or lipoteichoic acid of intact cells of Streptococcus pneumoniae and of samples of CSF and sera of patients. Samples were drawn from 10 patients with bacterial and 26 with abacterial meningitis. Concentrations of tumor necrosis factor α, interleukin (IL) 1β, IL-6, IL-10, IL-12, and interferon γ were estimated in cell supernatants by enzyme-linked immunosorbent assay and of procalcitonin by an immunochemiluminometric assay. It was found that the addition of CSF from patients with bacterial meningitis induced higher concentrations of IL-1β, IL-6, IL-10, IL-12, and procalcitonin than controls and than the addition of CSF from patients with abacterial meningitis. Intact cells of S. pneumoniae failed to elicit similar responses to CSF of patients. The presence of CSF from patients with abacterial meningitis induced higher levels of interferon γ and IL-6 than controls. It is concluded that cells of the innate immunity play a considerable role in the pathogenesis of meningitis. CSF triggering elicits cytokine responses that are higher in bacterial than abacterial meningitis; IL-1β, IL-6, IL-10, and procalcitonin are important mediators in bacterial meningitis and IL-6 and interferon γ in abacterial meningitis.
*Department of Neurology, †Fourth Department of Internal Medicine and ‡First Department of Paediatrics, University of Athens, Medical School, Greece.
Address correspondence and reprint requests to E.J. Giamarellos-Bourboulis, MD, PhD, Fourth Department of Internal Medicine, ATTIKON University Hospital, 1 Rimini Street, 124 64 Athens, Greece. E-mail: firstname.lastname@example.org.
The role of cells of the innate immune system in the pathogenesis of meningitis is not fully elucidated. Although cells of innate immunity are considered to play a considerable role in pathogenesis, their participation is based on evidence derived from studies of other types of infection.1 Innate immunity in the central nervous system is represented by the function of microglial cells and of resident macrophages in the perivascular spaces and the choroid plexus.2 Triggering of these cells by microbial pathogens seems to follow the same cascade as that of blood monocytes.3-5 It is presumed that these cells elicit proinflammatory and anti-inflammatory cytokines often estimated in the cerebrospinal fluid (CSF) of patients with meningitis of bacterial or abacterial origin.6,7
The present study aimed to define the significance of cytokines elicited after triggering of cells of the innate immune system in the pathogenesis of meningitis and to study probable differences between bacterial and abacterial pathogens. Based on the presumption that glial cells represent the function of tissue macrophage in the brain so that they may behave in vitro as monocytes, purified monocytes collected from healthy volunteers were cultured in the presence of samples of CSF and of serum of patients with bacterial and abacterial meningitis.
PATIENTS AND METHODS
A total of 36 patients were enrolled in the study. They comprised consecutive admissions for meningitis over the period January 2002 to December 2002 from the Fourth Department of Internal Medicine of the University of Athens. Exclusion criteria comprised the following: (a) HIV infection, (b) any underlying malignancy, and (c) intake of any antimicrobial agent before admission. The study was approved by the ethics committee of the hospital, and written informed consent was collected from the patients' relatives.
Meningitis was diagnosed in any admitted patient with the following criteria: (a) typical history with fever and headache, (b) at least 1 of 3 physical findings comprising nuchal rigidity, Kernig sign, and Bruzinski sign, and (c) pleocytosis of the CSF. Pleocytosis was defined as any more than 10 white blood cells in the CSF.3
Patients were then divided as those who have abacterial or bacterial meningitis. Diagnosis of abacterial meningitis was based on the following criteria8: (a) less than 500 white blood cells per microliter of CSF, predominant lymphocytes, (b) normal values of glucose of the CSF, and (c) negative search for any bacterial causative pathogen and for tuberculosis. Diagnosis of bacterial meningitis was based on the following criteria9: (a) more than 500 white blood cells per microliter of CSF, predominant neutrophils, (b) decreased glucose and increased total protein of the CSF, and (c) positive microbiological findings for any bacterial causative pathogen.
CSF was collected in all patients upon admission to the hospital after lumbar puncture under aseptic conditions. A thorough microbiological search was then performed for the presence of any bacterial pathogen. That search comprised latex agglutination of CSF and urine for Streptococcus pneumoniae, Neisseria meningitidis, and Haemophilus influenzae (BioMérieux, Paris, France); Gram smear of the CSF, Ziehl-Neelsen staining of the CSF; blood cultures and cultures of the CSF. A total of 7 mL of CSF was collected from each patient. One milliliter was placed into EDTA-coated tubes (Vacutainer, Becton Dickinson, Cockeysville, Md) and applied for white blood cell counting on a standard Neubauer counter. Another milliliter was collected into sterile tubes and applied for measuring of glucose and total protein levels by standard tests (Abbott Diagnostics, Chicago, Ill). Three milliliters of CSF was equally divided and plated onto MacConkey agar, chocolate agar enriched with nicotinamide adenine dinucleotide and hemoglobin and blood agar (Becton Dickinson). MacConkey agar plates were incubated for 18 hours at 35°C; chocolate agar and blood agar plates were incubated for 48 hours at 37°C in an atmosphere of 5% CO2. The remaining 2 mL of CSF was centrifuged; half of the supernatant was collected into sterile tubes for estimation of procalcitonin (PCT) by the assay described below. The remaining half was kept refrigerated at −20°C. It was removed from the fridge before being added into cell cultures.
Blood samples were also taken upon admission after venipuncture of 1 forearm vein concomitantly to lumbar puncture. A volume of 10 mL was added in tubes with thioglycolate medium and incubated for 7 days at 35°C. Another volume of 5 mL was centrifuged; collected serum was applied for estimation of PCT and of C-reactive protein. PCT was measured by the assay described below and C-reactive protein by nephelometry (Boehring, Mannheim, Germany; lower detection limit 3.2 mg/L). A serum sample was kept refrigerated at −70°C and removed from the fridge before being added into cell cultures.
In Vitro Stimulation of Human Monocytes
Human monocytes were isolated from 2 healthy adult volunteers, as already described.10 Briefly, heparinized venous blood was layered over Ficoll Hypaque (Biochrom, Berlin, Germany) and centrifuged. The separated mononuclear cells were washed 3 times with phosphate-buffered saline (pH 7.2; Merck, Darmstadt, Germany) and resuspended in RPMI 1640 (Biochrom) supplemented with 10% fetal bovine serum (Biochrom) and 2 mmol/L of glutamine (Biochrom) in the presence of 100 U/mL penicillin G and 0.1 mg/mL of streptomycin (Sigma, St Louis, Mo). After incubation for 1 hour at 37°C in 5% CO2, nonadherent cells were removed, whereas adherent monocytes were washed 3 times with Hanks solution (Biochrom). Cells were then harvested by 0.25% trypsin/0.02% EDTA (Biochrom) and resuspended in RPMI 1640 supplemented with 10% fetal bovine serum and 2 mmol/L glutamine in a 12-well plate at a density of 1 × 105 cells/well. The final volume of fluid per well was 2.4 mL, and the purity of monocytes was greater than 95%, as assessed by Giemsa staining. Cells were left to incubate for 30 minutes at 37°C in 5% CO2, when 100 μL of sampled CSF and 100 μL of sampled serum of each patient were added into different wells. Cells were subsequently incubated for 18 hours at 37°C in 5% CO2 when culture supernatants were collected and stored at −70°C until assayed. Each separate samples of CSF and of serum were added to wells containing monocytes of each 1 of the 2 volunteers; the wells were run in duplicate.
A total of 17 wells with monocytes of each individual were also prepared. Five of these wells were left unstimulated and served as controls; in the remaining wells with monocytes, the following stimulants were added: purified endotoxin (lipopolysaccharide [LPS]) of Escherichia coli 0144:H4 (Sigma) at a final concentration of 10 ng/mL in 5; purified lipoteichoic acid (LTA) of Staphylococcus aureus (Sigma) at a final concentration of 10 ng/mL in 5; and viable colonies of 2 isolates of S. pneumoniae were added at a final inoculum of 1 × 106 colony-forming units/mL in 4. The applied isolates were derived from 2 patients with meningitis other than those enrolled in the present study. They represented stock isolates kept refrigerated in skimmed milk (Becton Dickinson) at −70°C. After regrowth onto blood agar plates, single colonies were incubated for 24 hours in thioglycolate medium at 37°C in a shaking water bath. They were then washed 3 times with phosphate-buffered saline to remove any free LTA and added in the wells.
Measurements of Cytokines and PCT
Tumor necrosis factor α (TNF-α), interleukin (IL) 1β, IL-6, IL-10, IL-12, and interferon γ (IFN-γ) were measured in culture supernatants with an enzyme immunoassay (Diaclone, Paris, France). Lowest limits of detection were 15.1 pg/mL for TNF-α; 10 pg/mL for IL-1β, IL-6, and IFN-γ; and 30 pg/mL for IL-10 and IL-12. All measurements were performed in duplicate, and cytokine concentrations were expressed as pg/104 monocytes.
Estimation of PCT in samples of sera and CSF and in culture supernatants was performed in duplicate with an assay based on immunochemoluminescence (BRAHMS Diagnostica GmbH, Berlin, Germany; lower limit of detection 0.08 ng/mL). Concentrations were expressed as ng/104 monocytes.
Characteristics of CSF were expressed by their means (± SD), and concentrations of cytokines, by their medians and 95% confidence intervals (CIs). Characteristics of CSF between patients with abacterial and bacterial meningitis were compared by the Mann-Whitney U test. Comparisons of cytokines released in cell culture supernatants of controls and monocytes stimulated by LPS, LTA, S. pneumoniae isolates, CSF, and sera sampled from patients with abacterial and bacterial meningitis were performed by Mann-Whitney U test with a correction according to Bonferroni.
The ratios of concentrations of PCT and cytokines of cell supernatants after stimulation with samples from patients with abacterial and bacterial meningitis were counted. Ratios were compared by Mann-Whitney U test.
Correlations between the absolute counts of neutrophils and lymphocytes of the CSF and the level of released cytokines after stimulation or the concentrations of PCT in CSF and serum were performed according to the Spearman rank of order (rs).
Any P value less than 0.05 was considered as significant.
Demographic characteristics of patients enrolled in the study are given in Table 1. None of the patients had a positive blood culture. Concentrations of IL-1β, IL-6, and IL-12 in cell supernatants after triggering with LTA, LPS, S. pneumoniae, and samples of CSF and sera sampled from patients with bacterial and abacterial meningitis are shown in Figure 1. Concentrations were found elevated compared with controls after stimulation with CSF from patients with bacterial meningitis. Similar increases were achieved by LPS and LTA; S. pneumoniae did not induce any increase of IL-1β, IL-6, or IL-12. Median ratios of IL-1β released by monocytes after stimulation with samples of CSF to IL-1β released after stimulation with respective samples of serum were 0.87 for abacterial and 4.21 for bacterial meningitis (P = 0.001). Respective values for IL-6 were 1.20 and 1.07 (not significant), and for IL-12, 0.98 and 1.62 (P = 0.042).
Concentrations of TNF-α and IFN-γ in cell supernatants after triggering with LTA, LPS, S. pneumoniae, and samples of CSF and sera sampled from patients with bacterial and abacterial meningitis are shown in Figure 2. Concentrations of TNFα were found elevated compared with controls after stimulation with LTA, LPS, and S. pneumoniae but not with samples drawn from patients with meningitis. IFN-γ was considerably elevated after stimulation with CSF from patients with abacterial meningitis but not after stimulation with LPS, LTA, or S. pneumoniae. Median ratios of TNF-α released by monocytes after stimulation with samples of CSF to TNF-α released after stimulation with respective samples of serum were 0.89 for abacterial and 1.64 for bacterial meningitis (not significant). Respective values for IFN-γ were 3.21 and 1.89 (not significant).
Concentrations of IL-10 and PCT in cell supernatants after triggering with LTA, LPS, S. pneumoniae, and samples of CSF and sera sampled from patients with bacterial and abacterial meningitis are shown in Figure 3. Concentrations of IL-10 were found elevated compared with controls after stimulation with CSF and sera drawn from patients with meningitis. PCT was considerably elevated only after stimulation with CSF from patients with bacterial meningitis. Both IL-10 and PCT failed to elevate after stimulation with LPS, LTA, or S. pneumoniae. Median ratios of IL-10 released by monocytes after stimulation with samples of CSF to IL-10 released after stimulation with respective samples of serum were 1.28 for abacterial and 1.26 for bacterial meningitis (not significant). Respective values for PCT were 0.87 and 0.39 (P = 0.007).
Positive correlation was only found between the absolute neutrophil counts of the CSF and IL-1β of cell supernatants (rs = +0.610, P = 0.027).
Despite the existence of various hypotheses based on other types of infection, the exact mechanism of the pathogenesis of meningitis is not fully elucidated. Many authors have measured elevated levels of proinflammatory and anti-inflammatory mediators in the CSF of patients with meningitis, but the exact mechanism of their production is based on the hypotheses of the action of bacterial products on cells of the innate immunity.1,2 Because resident meningeal and perivascular macrophages are nowadays considered to play a pivotal role in pathogenesis,11 the present study aimed to estimate differences in the release of cytokines in the culture supernatants of monocytes collected from healthy volunteers after triggering with samples of CSF and serum from patients with bacterial and abacterial meningitis. This is the first study, to our knowledge, where a similar attempt is made; furthermore, the number of patients enrolled (Table 1) is greater than those applied in former studies6-8,11 comprising the estimation of single cytokines in the CSF and serum of patients with meningitis.
It was of striking importance to find that CSF derived from patients with bacterial meningitis elicited greater levels of IL-1β, IL-6, IL-10, and IL-12 than serum from patients with bacterial meningitis (Figs. 1 and 3). Similar levels were produced after triggering by LPS and LTA but not by intact cells of S. pneumoniae, despite the fact that most cases of enrolled patients with bacterial meningitis were caused by S. pneumoniae (Table 1). This is consistent with the finding that intact bacteria of S. pneumoniae may not elicit cytokine responses from innate immune cells; S. pneumoniae requires prior autolysis of its bacterial cell.13 S. pneumoniae was capable to induce TNF-α that was not increased in the presence of CSF of patients with bacterial meningitis (Fig. 2). The panel of produced cytokines after triggering with CSF of patients with bacterial meningitis is consistent with enormous signaling for brain damage, brain edema by IL-1β, intense inflammatory reaction by IL-6, and induction of TH1 response by IL-12 which heralds further brain damage.1,3,8 As a consequence, the higher level of cytokines induced by the CSF of patients with bacterial compared with abacterial meningitis might explain the differences of the clinical picture of both situations. It should also be mentioned that the higher cytokine responses after triggering with CSF compared with sera of patients with bacterial meningitis might signify that meningitis, at its initial stages, is a confined inflammation in the brain without similar bloodstream spreading.
Triggering of monocytes by CSF from patients with abacterial meningitis was not accompanied by an inflammatory response of similar intensity as after triggering with samples from patients with bacterial meningitis. IL-6 and IFN-γ were the predominant cytokines (Figs. 1 and 2), as also found in the CSF of patients with viral meningitis.6
The presented model applied monocytes from healthy donors to explain differences in the pathogenesis of bacterial and abacterial meningitis. Its adequacy is based on results that microglia constitute representatives of the innate immune system in the brain,1 thus eliciting cytokine responses after triggering with LPS or constituents of pneumococcal cell wall.14 Similar increases in the concentrations of IL-10 in the CSF have also been described after inoculation of H. influenzae in a rat model of meningitis.15 The design of the present study did not comprise any measurements of the concentrations of IL-8. This was based on the assertion that IL-8 is a nonspecific chemokine also secreted by neutrophils.10
PCT was highly elevated in the sera of patients with bacterial compared with abacterial meningitis (Table 1), as expected according to other authors.9,12 However, its production by monocytes after triggering with CSF from patients with bacterial meningitis was higher than with sera from the same patients or than with CSF from patients with abacterial meningitis. The latter observations and the concomitant increase of the anti-inflammatory cytokine IL-10 (Fig. 3) underline the hypothesis that PCT might behave as an anti-inflammatory mediator.16 These findings also demonstrate that, for the production of PCT in the bloodstream, other cells than the monocytes are also effectuated. The latter hypothesis is further intensified by the lack of production of PCT after triggering with LPS, LTA, or S. pneumoniae.
The presented findings revealed the importance of the cells of the innate immunity for the pathogenesis of meningitis. Direct triggering elicits cytokine responses that are higher in bacterial than abacterial meningitis; IL-1β, IL-6, IL-10, and PCT are important mediators in bacterial meningitis and IL-6 and IFN-γ in abacterial meningitis. Further research is mandatory to fully elucidate the clinical relevance of these findings.
1. Scheld WM, Koedel U, Nathan B, et al. Pathophysiology of bacterial meningitis: mechanisms of neuronal injury. J Infect Dis
2. Aloisi F. Immune function of microglia. Glia
3. Dong Y, Benveniste EN. Immune function of astrocytes. Glia
4. Hauwel M, Furon E, Canova C, et al. Innate (inherent) control of brain infection, brain inflammation and brain repair: the role of microglia, astrocytes, "protective" glial stem cells and stroma; ependymal cells. Brain Res Brain Res Rev
5. Carpentier PA, Begolka WS, Olson JK, et al. Differential activation of astrocytes by innate and adaptive immune stimuli. Glia
6. Dalai I, Tzhori S, Somekh E, et al. Cytokine profiles in cerebrospinal fluid of children with echovirus type 4 meningitis. Pediatr Neurol
7. Sprenger H, Rosler A, Tonn A, et al. Chemokines in the cerebrospinal fluid of patients with meningitis. Clin Immunol Immunopathol
8. Ichiyama T, Isumi H, Yoshitomi T, et al. NF-κB activation in cerebrospinal fluid cells from patients with meningitis. Neurol Res
9. Gendrel D, Raymond J, Assicot M, et al. Measurement of procalcitonin levels in children with bacterial or viral meningitis. Clin Infect Dis
10. Giamarellos-Bourboulis EJ, Plachouras D, Tzivra A, et al. Stimulation of innate immunity by susceptible and multidrug-resistant Pseudomonas aeruginosa
: an in vitro and in vivo study. Clin Exp Immunol
11. Polfiet MMJ, Zwijnenburg PJG, van Furth AM, et al. Meningeal and perivascular macrophages of the central nervous system play a protective role during bacterial meningitis. J Immunol
12. Shimetani N, Shimetani K, Mori M. Levels of three inflammatory markers, C-reactive protein, serum amyloid A protein and procalcitonin in the serum and cerebrospinal fluid of patients with meningitis. Scand J Clin Lab Invest
13. Koedel U, Scheld WM, Pfister HW. Pathogenesis and pathophysiology of pneumococcal meningitis. Lancet Infect Dis
14. Häusler KG, Prinz M, Nplte C, et al. Interferon-γ differentially modulates the release of cytokines and chemokines in lipopolysaccharide- and pneumococcal cell wall-stimulated mouse microglia and macrophages. Eur J Neurosci
15. Bakhiet M, Mustafa M, Zhu J, et al. Induction of cytokines and autoantibodies in cerebrospinal fluid (CSF) during experimental meningitis. Clin Exp Immunol
© 2006 Lippincott Williams & Wilkins, Inc.
16. Giamarellos-Bourboulis EJ, Mega A, Grecka P, et al. Procalcitonin: a marker to clearly differentiate systemic inflammatory response syndrome and sepsis in the critically ill patient. Intensive Care Med