Experimental Pneumococcal Meningitis in Mice: A Model of Intranasal
Petra J. G. Zwijnenburg,1,2Tom van der Poll,2
Sandrine Florquin,3Sander J. H. van Deventer,2
John J. Roord,1and A. Marceline van Furth1,2
1Department of Pediatrics, University Hospital Vrije Universiteit,
and Departments of
Academic Medical Center, University of Amsterdam,
2Experimental Internal Medicine and
Amsterdam, The Netherlands
Effective laboratory animal models of bacterial meningitis are needed to unravel the patho-
physiology of this disease. Previous models have failed to simulate human meningitis by using
a directly intracerebral route of infection. Hyaluronidase is a virulence factor of Streptococcus
pneumoniae. In this study, a novel model of murine meningitis is described. Intranasal admin-
istration of S. pneumoniae with hyaluronidase induced meningitis in 50% of inoculated mice, as
defined by a positive cerebrospinal fluid (CSF) culture and an inflammatory infiltrate in the
meninges. None of the mice inoculated without hyaluronidase developed meningitis. Hyal-
uronidase was found to facilitate pneumococcal invasion of the bloodstream after colonization
of the upper respiratory tract. Meningitis was characterized by pleocytosis of CSF and the
induction of proinflammatory cytokines and CXC chemokines in brain tissue. These results
indicate that this murine model mimics important features of human disease and allow for the
Bacterial meningitis due to Streptococcus pneumoniae is as-
sociated with significant mortality and morbidity, despite anti-
biotic treatment . Knowledge of the pathogenesisofpneumo-
coccal meningitis is important not only because of the severity
of the disease but also because resistance to penicillin and other
antimicrobial agents has increased dramatically over the past
decade . An animal model simulating human disease is es-
sential to study the pathophysiology of meningitis and to in-
vestigate the effectiveness of vaccines and potential new ad-
junctive therapeutic strategies. Various animal models have
been established [3, 4]. Mice models generally involve direct
intracerebral injection of bacteria and therefore fail to resemble
many aspects of bacterial invasion in human disease.
Hyaluronidase is a virulence factor of many bacteria, in-
cluding S. pneumoniae. Hyaluronidase is an endoglycosidase
that cleaves hexosaminidic linkages, thereby degrading hyal-
uronic acid, a component of the extracellular matrix. S. pneu-
Received 7 July 2000; revised 10 January 2001; electronically published
1 March 2001.
Presented in part: 39th Interscience Conference on Antimicrobial Agentsand
Chemotherapy, San Francisco, September 1999 (abstract 2040).
All animal experimentation guidelines were followed and were approved
by the Institutional Animal Care and Use Committee of the Academic
Medical Center, Amsterdam.
Reprints or correspondence: P. J. G. Zwijnenburg, Dept. of Experimental
Internal Medicine, Academic Medical Center, Meibergdreef 9, 1105 AZ,
Amsterdam, The Netherlands (firstname.lastname@example.org).
The Journal of Infectious Diseases
? 2001 by the Infectious Diseases Society of America. All rights reserved.
moniae strains demonstrate a strong correlation between hyal-
uronidase activity and the capacity to induce meningitis [5–7].
In this study, we sought to establish a mouse model of
pneumococcal meningitis that mimics the common route of
infection in human disease (i.e., colonization of the upper res-
piratory tract mucosa, which is followed by invasion ofbacteria
into the bloodstream and subsequent invasion of the central
nervous system [CNS] via the blood-brain barrier [BBB]) [8, 9].
For that reason, we added increasing doses of purified hyal-
uronidase to intranasally inoculatedS.pneumoniae.Inaddition,
to assess the similarity of this murine model of pneumococcal
meningitis to human disease, we describe the local inflamma-
tory response by analyzing leukocyte influx and by measuring
concentrations of inflammatory mediators.
Materials and Methods
ingitis patient, was cultured overnight at 37?C in 80 mL of brain
heart infusion broth (BHI). At midlogarithmic growth phase, 5 mL
of this suspension was transferred to 25 mL of BHI. The bacterial
suspensions were grown at 37?C until an optical density of 1.0 at
a wavelength of 620 nm was achieved. Subsequently,thesuspension
was washed twice in sterile isotonic saline and was resuspended in
sterile isotonic saline, which corresponded to
exact number of colony-forming units was determined retrospec-
tively by the growth of serial dilutions on blood agar plates. Syn-
thetic hyaluronidase (0, 180, 360, or 560 U; Sigma) was added to
Female Swiss mice, weighing 12–14 g, were used.
S. pneumoniae (serotype 6A), isolated from a men-
by guest on February 2, 2016
1144 Zwijnenburg et al. JID 2001;183 (1 April)
tococcus pneumoniae (in cfu) and hyaluronidase (HL).
No. of mice that developed meningitis 48, 72, and 96 h after inoculation with Strep-
1 ? 104cfu5 ? 104cfu1 ? 105cfu5 ? 105cfu
48 h 72 h96 h48 h72 h 96 h 48 h 72 h 96 h48 h 72 h 96 h
none of the mice had developed meningitis. —, Not determined.
Data are no. of mice that developed meningitis/totalno.ofinoculatedmice.At24hafterinoculation,
Induction of meningitis.
halation of isoflurane (Abbott), and 50 mL of bacterial suspension
was inoculated intranasally or intravenously. Control mice were
inoculated with isotonic saline (
( ), or S. pneumoniae only (i.e., without hyaluronidase;
n p 8
). At 48, 72, or 96 h after inoculation, mice were anesthetized8
with Hypnorm (Janssen Pharmaceutica) and midazolam (Roche),
after which blood samples were obtained throughcardiacpuncture,
and cerebrospinal fluid (CSF) was collected by puncture from the
cisterna magna. White blood cells (WBCs) in CSF were counted
immediately in a hematometer (Emergo). Cytospins were made of
CSF and were stained using DiffQuick (Dade), according to the
Brains were removed, half of which were fixed in 10% buffered
formalin for histopathologic study. The other half were homoge-
nized in sterile saline. Homogenized brain tissue was incubated
with lysis buffer (300 mM NaCl, 15 mM Tris, 2 mM MgCl, 2 mM
Triton [X-100], Pepstatin A [20 ng/mL; pH 7.4], Leupeptin [20 ng/
mL; pH 7.4], and Aprotinine [20 ng/mL; pH 7.4]) for 30 min and
were centrifuged for 15 min at 1500 g. After plating on blood agar,
blood samples were transferred to EDTA tubes (BectonDickinson)
and were centrifuged at 1500 g for 10 min. Supernatants of brain
homogenates were stored at ?20?C for cytokine measurements. In
2 experiments, blood samples were obtained twice daily through
the experimentation period (at 7, 24, 31, 48, 55, and 72 h after
inoculation) from the tail vein and were plated on blood agar, to
provide insight into the development of bacteremia.
Quantification of bacterial outgrowth.
in sterile isotonic saline were made of CSF and blood and were
plated onto blood agar plates. Plates were incubated for 18 h at
37?C, after which colonies were counted.
Brains were fixed in 10% buffered for-
malin, and, after paraffin embedding, 4-mm sections were stained
with hematoxylin-eosin. All slides were coded and were scored
semiquantitatively by a pathologist who had no knowledge of the
inocula or CSF culture.
Levels of cytokines were measured by using com-
mercially available ELISAs, according to the manufacturer’s rec-
ommendations (KC, macrophage inflammatory protein [MIP]–2,
interleukin [IL]–1a, IL-1b, IL-6, interferon-g, and tumor necrosis
factor (TNF)–a; all obtained from R&D Systems).
Data are expressed as
tical analysis was performed by using Mann-Whitney U test. P !
was considered to be significant..05
Mice were lightly anesthetized by in-
), sterile hyaluronidase
n p 2
Serial 10-fold dilutions
Induction of meningitis.
ingitis when the CSF culture was positive, and histopathologic
Mice were considered to havemen-
evidence for meningitis was present. To identify the optimal
inoculum, mice were inoculated intranasally with
,, or cfu of S. pneumoniae and 180 U of
hyaluronidase, and were killed after 24, 48, 72, or 96 h. At 24
h after inoculation, none of the mice showed positive CSF
cultures or histopathologic signs of meningitis. At all other
timepoints (48, 72, and 96 h), 50% of all mice inoculated with
orcfu of S. pneumoniae and 180 U of hyal-
uronidase had developed meningitis (table 1). No further in-
crease in disease rate was found at 72 h after inoculation with
the addition of 360 or 540 U of hyaluronidase to the inoculum.
None of the mice inoculated intranasally with S. pneumoniae
only (i.e., without hyaluronidase) developed meningitis.
At 72 h after inoculation with
and 180 U of hyaluronidase, the mean WBC count of CSF of
mice with positive CSF cultures was
of CSF, whereas the mean WBC count of mice with negative
cultures was cells/mL of CSF (
entiation of this pleocytosis showed 70%–75% neutrophils,
20%–25% monocytes, and 2%–5% lymphocytes.
All mice that demonstrated growth of S. pneumoniae in CSF
a significant inflammatory infiltrate predominantly composed of
polymorphonuclear leukocytes in the meninges with a prefer-
ential concentration around leptomeningeal blood vessels, at 48
h after inoculation (figure 1). The inflammation was more pro-
nounced at the base of the brain than at the hemispheric con-
vexities. After 72 and 96 h, the inflammatory infiltrates in the
meninges were more pronounced and diffuse, with the devel-
opment of foci of necrotizing cerebritis and formation of small
brain abscesses in 70% of mice with meningitis. Control mice, as
well as mice with negative CSF culture, did not show any signs
of meningitis or other histopathologic lesions.
Mice did not show any signs of illness for the first 40–48 h after
signs of systemic toxicity. No neurological signs were observed.
Route of infection.
Blood samples obtained from all micethat
had been killed were cultured, to evaluate the route of infection.
Of mice inoculated with
moniae and 180 U of hyaluronidase, all mice (
developed meningitis were bacteremic at time of death, whereas
6 mice with positive blood cultures did not show any signs of
meningitis at time of death. Therefore, 80% of bacteremic mice
developed meningitis. Of 32 mice from which blood samples
cfu of S. pneumoniae
P ! .05
tocfu of S. pneu-
n p 24
by guest on February 2, 2016
JID 2001;183 (1 April) Pneumococcal Meningitis in Mice1145
stained brain tissue of a mouse that developed meningitis after inocu-
lation with Streptococcus pneumoniae and hyaluronidase (A) and of a
mouse that was inoculated with hyaluronidase alone (B); original mag-
nification, . Cerebral cortex is depicted by a broad arrow; me-10 ? 3.3
ninges by a thin arrow. Slides are from brain tissue obtained 96 h after
Representative histopathology of hematoxylin-eosin–
veloped and did not develop meningitis.
Brain cytokine concentrations in mice that de-
time in hMeningitisNo meningitis
5.65 ? 0.59
5.92 ? 2.21
5.28 ? 0.30
4.93 ? 0.33
2.86 ? 0.40
4.45 ? 0.16
1.68 ? 0.20
2.43 ? 0.27
4.37 ? 0.67
1.33 ? 0.13
0.83 ? 0.09
2.21 ? 0.17
9.00 ? 3.66
25.56 ? 13.3
9.40 ? 1.29
1.70 ? 0.65
3.84 ? 0.83
0.86 ? 0.11
0.77 ? 0.13
1.15 ? 0.36
1.08 ? 0.20
0.45 ? 0.06
0.32 ? 0.07
0.22 ? 0.08
51.3 ? 32.4
21.2 ? 7.5
11.1 ? 1.0
4.2 ? 0.3
5.4 ? 0.2
3.7 ? 0.2
8.6 ? 5.2
7.2 ? 1.3
9.2 ? 2.8
2.3 ? 0.2
2.8 ? 0.2
2.0 ? 0.1
cated. Results are from 4–10 mice/group. Time point after intranasal
inoculation with– cfu of Streptococcus pneumoniae with
or without hyaluronidase. IL, interleukin; MIP, macrophage inflamma-
tory protein; TNF-a, tumor necrosis factor.
aP values indicate differences between groups, usingMann-Whit-
ney U test.
Data are mean levels, unless otherwise indi- (ng/g)?SE
were collected repeatedly throughout the experimentation pe-
riod, blood cultures showed that, if bacteremia occurred, bac-
terial growth in blood could be demonstrated 7–31 h after in-
To assess the role of hyaluronidase in bacterial migration
across the BBB, we intravenously injected mice with either
cfu of S. pneumoniae (
n p 12
S. pneumoniae and 180 U of hyaluronidase (
after intravenous inoculation, 9 (75%) of 12 mice in each group
had positive CSF cultures, regardless of the addition of hyal-
uronidase. All mice with positive CSF culture also presented
mild to severe meningitis with formation of brain abscesses, as
Induction of cytokines and chemokines.
) alone orcfu of
). Two days
n p 12
of meningitis was associated with an increase in concentrations
of inflammatory cytokines such as TNF-a, IL-1a, and IL-1b
and IL-6, KC, and MIP-2 in brain tissue (table 2). TNF-a, IL-
1a, IL-6, and MIP-2 concentrations correlate with bacterial
counts in CSF (all), and IL-1b levels correlate with
P ! .01
severity of histological changes ().
P ! .05
In this study, we describe a new murine model of meningitis, in
which pneumococci with hyaluronidase are introduced intranasally.
The pleocytosis of the CSF and the differentiation of thecells
that we found are comparable with human meningitis. Fur-
thermore, the histopathologic findings during the course of the
disease in our model closely mimictheprogressionofmeningitis
in untreated patients.
We were able to induce meningitis in 50% of mice inoculated
intranasally with S. pneumoniaeand hyaluronidase, whereas
none ofthemice inoculatedwithS.pneumoniaealonedeveloped
meningitis. After intranasal administration, substances can be
delivered to the CNS either directly, via the olfactory nerve or
olfactory epithelium, or indirectly, via the systemic circulation
. In our experiments, all mice with meningitis were bacte-
remic at time of death, and sequential blood cultures showed
by guest on February 2, 2016
1146 Zwijnenburg et al.JID 2001;183 (1 April) Download full-text
bacterial growth at time points when CSF cultures were still
negative. Therefore, in our model, S. pneumoniae seems to gain
access to the subarachnoid space via the circulation.
Invasion of the CNS by pneumococci is initiated by coloni-
zation of the nasopharyngeal mucosa. Pneumococcal surfacead-
hesin A, pneumococcal surface protein A, and pneumolysin are
surface proteins of pneumococci, which are believed to play an
important role in adherence to the mucosa . By a proteolytic
step, pneumococci are capable of penetrating the basement
membrane and extracellular matrix, after which they may enter
the circulation. Plasminogen binding by bacteria has been sug-
gested to contribute to this phenomenon . In our model, the
addition of hyaluronidase probably facilitates invasion by loos-
ening the barrier function of the nasopharyngeal mucosa. Hyal-
uronidase does not seem to influence the passage of the bacteria
across the BBB, in light of the fact that we couldnotdemonstrate
any difference in bacterial migration across the BBB in the pres-
ence of hyaluronidase after intravenous administration of pneu-
mococci. Therefore, we assume that the function of hyaluroni-
dase in this model is predominantly facilitation of systemic
invasion after nasopharyngeal colonization.
The elevation of proinflammatory cytokines in brain tissue
during meningitis highlights the similarity of the model with
human disease. TNF-a, IL-1a, IL-1b, and IL-6 were demon-
strated elsewhere to be markedly elevatedinhumanCSFduring
bacterial meningitis [12, 13].
An important hallmark of bacterial meningitis is the develop-
ment of pleocytosis of CSF. Chemokines are considered to be
sets and play a critical role in meningeal inflammation .CXC
chemokines have been implicated in the development of neutro-
philic pleocytosis in CSF during bacterial meningitis [14, 15]. In
mice, the prototypic CXC chemokines are KC and MIP-2. We
found elevated MIP-2 and KC concentrations in homogenates
of brain tissue of mice with meningitis. No reports are available
on the production of KC and MIP-2 in mice during pneumo-
Pneumococcal meningitis is a life-threatening community-
acquired infection. The model for this disease described in this
study is unique, because it mimics the route of infection in
humans. The model is efficient and reproducible and deals with
an experimentally convenient latency, because mice develop
meningitis 2 days after inoculation. As such, it should be an
excellent model for addressing issues related to this clinically
1. Quagliarello V, Scheld WM. Bacterial meningitis: pathogenesis, pathophy-
siology, and progress. N Engl J Med 1992;327:864–72.
2. Novak R, Henriques B, Charpentier E, Normark S, Tuomanen E. Emergence
of vancomycin tolerance in Streptococcus pneumoniae. Nature 1999;399:
3. Tunkel AR, Scheld WM. Alterations of the blood-brain barrier in bacterial
meningitis: in vivo and in vitro models. Pediatr Infect Dis J 1989;8:911–3.
4. Tauber MG, Zwahlen A. Animal models for meningitis. Methods Enzymol
5. Kostyukova NN, Volkova MO, Ivanova VV, Kvetnaya AS.Astudyofpatho-
genic factors of Streptococcus pneumoniae strains causing meningitis.
FEMS Immunol Med Microbiol 1995;10:133–7.
6. Berry AM, Lock RA, Thomas SM, Rajan DP, Hansman D, Paton JC. Clon-
ing and nucleotide sequence of the Streptococcus pneumoniae hyaluroni-
dase gene and purification of the enzyme from recombinant Escherichia
coli. Infect Immun 1994;62:1101–8.
7. Berry AM, Paton JC. Additive attenuation of virulence of Streptococcus
pneumoniae by mutation of the genes encoding pneumolysin and other
putative pneumococcal virulence proteins. Infect Immun 2000;68:133–40.
8. Zhang JR, Tuomanen E. Molecular and cellular mechanisms for microbial
entry into the CNS. J Neurovirol 1999;5:591–603.
9. Tuomanen EI. The biology of pneumococcal infection. Pediatr Res 1997;42:
10. Mathison S, Nagilla R, Kompella UB. Nasal route for direct delivery of
solutes to the central nervous system: fact or fiction? J Drug Target1998;5:
11. Eberhard T, Kronvall G, Ullberg M. Surface bound plasmin promotes mi-
gration of Streptococcus pneumoniae through reconstituted basement
membranes. Microb Pathog 1999;26:175–81.
12. van Furth AM, Roord JJ, van Furth R. Roles of proinflammatory and anti-
inflammatory cytokines in pathophysiology of bacterial meningitis and
effect of adjunctive therapy. Infect Immun 1996;64:4883–90.
13. Diab A, Zhu J, Lindquist L, Wretlind B, Bakhiet M, Link H. Haemophilus
influenzae and Streptococcus pneumoniae induce different intracerebral
mRNA cytokine patterns during the courseofexperimentalbacterialmen-
ingitis. Clin Exp Immunol 1997;109:233–41.
14. Tauber MG, MoserB.Cytokinesandchemokinesinmeningealinflammation:
biology and clinical implications. Quiz 12. Clin Infect Dis 1999;28:1–11.
15. Spanaus KS, Nadal D, Pfister HW, et al. C-X-C and C-C chemokines are
expressed in the cerebrospinal fluid in bacterial meningitis and mediate
chemotactic activity on peripheral blood-derived polymorphonuclearand
mononuclear cells in vitro. J Immunol 1997;158:1956–64.
by guest on February 2, 2016