Figures
Abstract
Background
Inflammation is associated with perinatal brain injury but the underlying mechanisms are not completely characterized. Stimulation of Toll-like receptors (TLRs) through specific agonists induces inflammatory responses that trigger both innate and adaptive immune responses. The impact of engagement of TLR2 signaling pathways on the neonatal brain is still unclear. The aim of this study was to investigate the potential effect of a TLR2 agonist on neonatal brain development.
Methodology/Principal Findings
Mice were injected intraperitoneally (i.p.) once a day from postnatal day (PND) 3 to PND11 with endotoxin-free saline, a TLR2 agonist Pam3CSK4 (5 mg/kg) or Lipopolysaccharide (LPS, 0.3 mg/kg). Pups were sacrificed at PND12 or PND53 and brain, spleen and liver were collected and weighed. Brain sections were stained for brain injury markers. Long-term effects on memory function were assessed using the Trace Fear Conditioning test at PND50. After 9 days of Pam3CSK4 administration, we found a decreased volume of cerebral gray matter, white matter in the forebrain and cerebellar molecular layer that was accompanied by an increase in spleen and liver weight at PND12. Such effects were not observed in Pam3CSK4-treated TLR 2-deficient mice. Pam3CSK4-treated mice also displayed decreased hippocampus neuronal density, and increased cerebral microglia density, while there was no effect on caspase-3 or general cell proliferation at PND12. Significantly elevated levels of IL-1β, IL-6, KC, and MCP-1 were detected after the first Pam3CSK4 injection in brain homogenates of PND3 mice. Pam3CSK4 administration did not affect long-term memory function nor the volume of gray or white matter.
Citation: Du X, Fleiss B, Li H, D'angelo B, Sun Y, Zhu C, et al. (2011) Systemic Stimulation of TLR2 Impairs Neonatal Mouse Brain Development. PLoS ONE 6(5): e19583. https://doi.org/10.1371/journal.pone.0019583
Editor: Adam J. Ratner, Columbia University, United States of America
Received: September 26, 2010; Accepted: April 12, 2011; Published: May 6, 2011
Copyright: © 2011 Du et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by grants from the Swedish Medical Research Council (VR K2009-54X-21119-01-4 to XW, VR K2007-54x-14185-06-3 to CM, VR 2006-3396 to HH), Wilhelm and Martina Lundgren (vet2-41/2010 to XW), Signhild Engkvists Stiftelse (to XW), Linnéa och Josef Carlssons stiftelse (HH), National Natural Science Foundation of China (30973240 to XW), Swedish governmental grants to researchers in the public health service (ALFGBG2863 to HH) and Medical Research Council in United Kingdom (P19381 to HH). Åhlén stiftelse (CM), National Institutes of Health, USA (RO1 NIAID-AI067353-01A1 to OL). EU FP6, Neobrain (036534 to HH and CM) and the European Union (HEALTH-F2-2009-241778 to CM). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: OL served as a consultant to MedImmune and Biosynexus. OL's laboratory receives material and/or sponsored research support from 3M Drug Delivery Systems and VentiRx. None of the other authors declare any conflicts.
Introduction
Improved neonatal and intensive care has enabled the survival of preterm infants with very low birth weights. These infants are at increased risk for nosocomial infection, and Staphylococcus epidermidis is the predominant pathogen isolated from blood cultures obtained in the neonatal intensive care unit [1], [2], [3], [4], [5]. Increasing evidence suggests that neonatal brain injury is associated with infection/inflammation, but the underlying mechanisms are incompletely characterized [6], [7], [8]. Preterm infants in particular have an increased risk of brain injury, which is predominantly located in the cerebral white matter, although recently a high frequency of grey matter injury has also been reported [9]. Moreover, very low birth weight premature infants manifest cerebellar abnormalities [6].
Infection/inflammation stimulates innate and subsequent adaptive immune responses via the Toll-like Receptor (TLR) family of pattern-recognition receptors that can be stimulated with specific agonists. TLRs exist in a wide range of tissues outside the immune system, including the central nervous system (CNS). TLR2 forms heterodimers with TLR1 and TLR6, and these receptor complexes recognize molecules expressed on Gram-positive bacteria, such as peptidoglycan, lipopeptides, and lipoproteins, and they also mediate recognition of whole bacteria such as Staphylococcus epidermidis [10], [11], [12]. Of note, TLR2 is selectively up-regulated in the peripheral blood mononuclear cells of human newborns infected with Gram-positive bacteria [13]. With respect to the CNS, a role for TLR2 signaling in adult mouse brain injury has been suggested, as summarized by a most recent review [14] but there are few reports that define the role of TLR2 signaling in neonatal brain injury. However, there are studies that suggest that TLR2 and TLR4 are the principal TLRs present on microglia which are involved in the innate immune response to infection/hypoxia-ischemia; for a most recent review, please see [15]. Of note, neonates demonstrate a distinct functional expression of the TLR system [16], [17], and therefore studies of outcome in adult models cannot be directly extrapolated to newborns. In the present study, we hypothesized that stimulation of TLR2 during a critical period of neonatal brain development would have a detrimental effect on the immature brain, which may be measurable as changes in adult behavior. We used a synthetic lipopeptide, Pam3CysSerLys4 (Pam3CSK4), as a specific TLR2 agonist [18], that was administrated systemically to newborn wild-type and TLR2 deficient mice from postnatal day (PND) 3 to PND11 to evaluate short and long-term effects on the developing mouse brain.
Methods
Ethics statement
The animal experiments were approved by the local Animal Ethics Committee at the University of Gothenburg (Ethical approval 350-2009).
Animals
Time-mated pregnant C57BL/6 wild-type mice were purchased from Charles River Laboratories (Sulzfeld, Germany) and gave birth in the animal facility (Experimental Biomedicine, University of Gothenburg, Gothenburg, Sweden). B6.129-Tlr2tm1Kir/J (TLR2 –deficient) mice were purchased from the Jackson Laboratory (US) and bred in the animal facility. The day of birth was defined as postnatal day (PND) 0. Mice were housed with a 12-hour light/dark cycle with ad libidium access to a standard laboratory chow diet (B&K, Solna, Sweden) and drinking water was provided.
Drug administration
Offspring of both genders of C57BL/6 wild type mice were randomly divided into three groups. i) Negative control mice treated with endotoxin-free saline (10 µl/g, sb776, Sigma, USA, n = 12); ii) Pam3CSK4 (5 mg/kg, Invitrogen, n = 11) treated mice; and iii) Lipopolysaccharide, (LPS, 0.3 mg/kg, Escherichia coli 055:B5; Sigma, Stockholm, Sweden) injected mice (n = 13). LPS animals were used for comparison, as we have previously shown that repeated administration of this dose of LPS from PND3 to PND11 induces neonatal brain white/gray matter injury [19]. Offspring of both genders of TLR2-deficient mice were randomly divided into two treatment groups: i) Pam3CSK4, treated (5 mg/kg, n = 8) mice; ii) endotoxin-free saline treated (10 µl/g, sb776, Sigma, USA, n = 10). Mice were injected intraperitoneally (i.p.) once a day from PND3 to PND11. Pups were sacrificed at PND12 and PND53 and brain (including cerebrum and cerebellum), spleen, and liver were collected and weighed.
Immunohistochemical staining
Mice at PND12 and PND53 were deeply anesthetized and perfused intracardially with saline followed by 5% buffered formaldehyde (Histofix; Histolab, Gothenburg, Sweden). Brains were removed and fixed in 5% buffered formaldehyde for 18–24 hours and processed to paraffin. The cerebrum was cut into 10-µm coronal sections and collected at 50-section intervals. Serial sections were used for histologic stains, as previously described [20]. Briefly, nonspecific binding was blocked for 30 minutes with 4% horse serum or 4% goat serum in phosphate-buffered saline. The following primary antibodies were used: microtubule associated protein-2 (MAP-2; clone HM-2, Sigma), rabbit anti-myelin basic protein (MBP, Sternberger Monoclonal Incorporated, SMI 94, Lutherville, Massachusetts), active form of caspase-3 (557038, BD Bioscience Pharmingen); anti-neuronal nuclear antigen (NeuN) (MAB377B, Chemicon), anti-Ki67 (NCL-KI-67-MMI, Novocasta), and anti-Iba-1 (019-19741 Wako). Primary antibodies were incubated for 60 minutes at room temperature followed by the corresponding biotinylated secondary antibodies (all from Vector, Burlingame, California) also for 60 minutes at room temperature. Visualization was performed using Vectastain ABC Elite with 0.5 mg/mL 30-diaminobenzidine enhanced with 15 mg/mL ammonium nickel sulfate, 2 mg/mL β-D-glucose, 0.4 mg/mL ammonium chloride, and 0.01 mg/mL β-glucose oxidase (all from Sigma).
The cerebellum from PND12 mice was cut into 10 µm sagittal sections and collected at 50-section intervals. Serial sections were used for thionin/fuchsin acid staining as described previously [21].
Gray and white matter volume measurement
The forebrain gray matter area was determined by measuring the MAP-2 immunoreactive area from 6 serial sections per animal. The cerebral subcortical white matter area was determined by measuring MBP immunoreactive area in 6 serial sections per animal. The area of the molecular cell layer and granule cell layer of the cerebellum were measured in thionin/fuchsin acid stained sections in 8 serial sections. Micro Image, version 4.0 (Micro-Macro AB, Gothenburg, Sweden) was used for all the above measurements. The volume was calculated from area measurements according to the Cavalieri's Principal as described previously [20], using the following formula: V = SA · p · T, where V is the total volume, SA is the sum of the areas measured, p is the inverse of the section sampling fraction, and T is the section thickness.
Cell counting
NeuN-positive cells were counted in all cornu ammonis (CA) fields and dentate gyrus (DG) in 2 sections through the anterior hippocampus and Iba-1 positive cells in the right hemisphere in 4 sections, using stereological principles (Stereo investigator 7, System Inc, Magdeburg, Germany), with a counting frame of 40×40 µm per section for NeuN, and 150×150 µm per section for Iba-1. Ki67-positive cells were counted within the area of the granule cell layer (GCL), including the subgranular zone (SGZ) in the DG and CA of the hippocampus. Caspase-3 positive cells were counted in the right hemisphere (4 levels) and subcortical white matter (4 levels) and Purkinje cells in the posterior lobes in all 8 serial sections of the cerebellum. The average number of positive cells/mm2 was calculated.
Cytokine/chemokine assay
Cytokine/chemokines were measured in whole brain homogenate supernatants from PND3 wild type mice sacrificed 6 hours after i.p. treatment with endotoxin-free saline (10 µl/g, n = 6), LPS (0.3 mg/kg; n = 7) or Pam3CSK4 (5 mg/kg; n = 7). Mice were deeply anesthetized and perfused intracardially with saline and brains were removed and snap frozen. Brains were homogenized by sonication in ice-cold homogenization buffer containing 1% protease inhibitor cocktail (P8340, Sigma-Aldrich) and 3% EDTA in 0.1 M phosphate buffered saline, centrifuged at 4°C once at 900×g for 10 minutes and then at 10,000×g for 15 minutes, supernatants were collected and stored at −80°C until use. Concentrations of IL-1β, IL-6, KC, MCP1, IL-10, IL-17 and TNFα were measured using Bio-plex Multiplex Cytokine Assay (Bio-Rad laboratories, Hercules, CA). Results were normalized to the amount of protein per well, as determined using a Bio-Rad DC protein assay.
Trace Fear Conditioning test
Long term memory function was measured via Trace Fear Conditioning test at PND50, in an Automatic Reflex Conditioner 7531 (inside dimensions 390×95×165 mm; Cat No: 7530, Ugo Basile, Italy) as previously described [22], [23] with some modifications. Animals were timed for freezing within a 2 min time period recorded by digital video cameras. Freezing was defined as absence of movement except for respiration. The procedure was conducted over 2 days. On day 1 freezing was scored prior to mice receiving a pairing of a tone (20 seconds, 80 dB, 670 kHz) and a shock (2 seconds, 0.5 mA). The time interval between the tone and the shock was 2 seconds. On day 2, freezing was scored pre-tone and the tone was then presented once for 30 seconds, 80 dB, 670 kHz. No shock was administered and freezing was scored for 2 minutes after the tone presentation (tone-elicited freezing, post-tone freezing).
Statistics
Statistical Package for the Social Sciences (SPSS 17.0) and StatView (5.0.1) were used for all analyses. One-way ANOVA followed by LSD post hoc test was used for comparison of data from more than two groups. For all other analysis, Student's unpaired t-test was used for comparison. Results are presented as mean ± standard error of the mean (SEM). P<0.05 was considered statistically significant.
Results
Brain, liver, and spleen weight changes at PND12
After repeated administration of 5 mg/kg Pam3CSK4 once a day from PND3 to PND11, brain weight was decreased compared with endotoxin-free saline-treated animals at PND12. In contrast, there was no difference between endotoxin-free saline-treated animals and LPS-treated animals (Figure 1A). We found no infarctions, dilatation of the cerebral ventricles, or morphological signs of cell death in any of the brain regions examined after administration of Pam3CSK4 or LPS.
Quantitative analysis of the cerebral weight (A), spleen weight (B), liver weight (C), body weight (D) at PND12 and PND53 after Pam3CSK4 administration from PND 3 to 11 in while type mice. *p<0.05; ***p<0.001.
There was a significant increase in both the absolute spleen and liver weights in animals treated with Pam3CSK4 and LPS compared with those treated with endotoxin-free saline at PND12 (Fig. 1B, 1C) as well as the relative spleen and liver weight to body weight ratio (data not shown). The whole body weight was not different between groups at PND12 (Fig. 1D). No mortality or other signs of morbidity were found during the entire study period.
Gray and white matter changes in the cerebrum at PND12
To examine the gray matter and white matter changes after Pam3CSK4 treatment, the cerebral gray matter volume was measured using immunohistochemical staining for the neuronal marker MAP-2, and subcortical white matter volume was measured using the myelin marker MBP (Fig. 2A). At PND12, significantly decreased cerebral white matter (Fig. 2B) and gray matter volume (Fig. 2C) were found both in the Pam3CSK4 and LPS-treated mice compared with saline-treated mice. In contrast, Pam3CSK4 administration to TLR 2 -deficient mice from PND3 to PND11 once a day did not result in any significant differences between the Pam3CSK4 treatment group and saline controls, with respect to both the white (Fig. 3A) and gray (Fig. 3B) matter volume. Similarly, brain weight (Fig. 3C), body weight (Fig. 3D), spleen weight (Fig. 3E) and liver weight (Fig. 3F) also did not change in the Pam3CSK4-injected TLR2-deficient animals compared to endotoxin-free saline-treated animals. These findings further confirm that the observed white/gray matter changes following Pam3CSK4 administration in wild type mice are TLR2- dependent.
Representative microphotographs of MBP staining in the subcortical area (A), Quantitative analysis of subcortical white matter volume (B) cerebral gray matter volume (C) and the number of NeuN positive cells in both DG and CA of hippocampus(D), at PND 12 after Pam3CSK4 administration from PND3 to PND11 in while type mice. *p<0.05; ** p<0.01.
Quantitative analysis of subcortical white matter volume (A) cerebral gray matter volume (B) cerebral weight (C), body weight (D) spleen weight (E), and liver weight (F) at PND12 after Pam3CSK4 administration from PND 3 to 11 in TLR2 deficient mice.
LPS-induced inflammation reduces hippocampal neurogenesis in adult rats [24]. To investigate the specific impact of Pam3CSK4-exposure on post-mitotic neurons in the hippocampus, NeuN positive cells were counted in the dentate gyrus (DG) and the CA fields. The density of NeuN-positive cells in the CA fields was significantly decreased both in Pam3CSK4 and LPS-treated mice compared with endotoxin-free saline-treated animals at PND12 (Fig. 2D). There was no difference in the density of NeuN positive cells in the DG between groups (Fig. 2D). Cell proliferation and apoptosis at PND12 were examined by staining brain sections with cell proliferation marker Ki67, and the apoptosis marker active caspase-3. There were no significant differences in number of Ki67 positive cells (Fig. 4A) or number of active caspase-3 positive cells in either the cerebral gray matter or the subcortical white matter among the three groups (Fig. 4B).
Representative photomicrographs and quantitative analysis of Ki67 (A), and active Caspase-3(B) at PND 12 after Pam3CSK4 administration from PND 3 to PND11 in while type mice. ** p<0.01. The pictures show representative positive staining of Ki67 (A) and caspase-3 (B).
Inflammation after Pam3CSK4 administration
To characterize the inflammatory response after Pam3CSK4 treatment, we first analyzed the cytokine/chemokine production by multiplex ELISA in brain homogenate samples at 6 hours after the first Pam3CSK4 treatment at PND3, in comparison with saline and LPS treated mice. It was found that 5 mg/kg Pam3CSK4 treatment induced elevated levels of IL-1ß, IL-6, KC, MCP-1, similar to those cytokines and chemokines induced by 0.3 mg/kg LPS (Fig. 5). IL-1ß was an exception in that a significant increase was noted in Pam3CSK4-treated pups compared with LPS-treated pups. Of note, IL-6 was significantly increased by Pam3CSK4 but not by LPS. TNF-α levels did not change in either of the two treatment groups. IL-10 and IL-17 levels were below the limits of detection in all brain homogenate samples tested.
Cytokine/chemokine changes in brain homogenates at 6 hours after the first Pam3CSK4 administration at PND3 in while type mice. *, p<0.05; **, p<0.01; ***, p<0.001.
To further examine the inflammatory response, we stained brain sections for the microglia marker Iba-1 (Fig. 6A). There was a significant increase of Iba-1 positive cells in the Pam3CSK4-treated group compared with endotoxin-free saline treated animals, while there was no difference between the LPS-treated group and endotoxin-free saline group (Fig. 6B).
Representative microphotograph of Iba-1 in the hippocampus area (A) and Iba-1 positive cells counts (B) at PND 12 after Pam3CSK4 administration from PND 3 to PND11 in while type mice. **p<0.01.
Cerebellar changes at PND12
To investigate the effect of Pam3CSK4 administration on the neonatal cerebellum, molecular cell layer and granule cell layer volumes were measured and the density of Purkinje cells was counted (Fig. 7A). There was a significant decrease in the molecular cell layer volume in Pam3CSK4-treated mice but not LPS treated mice compared with saline-treated pups (Fig. 7B), while there were no differences in the granule cell layer volume (Fig. 7C) or the number of Purkinje cells between the three groups (Fig. 7D).
Representative microphotograph of the cerebellar lobe (A) with arrow indicating Purkinje cell under higher magnification. Quantitative analysis of the total volume of molecular cell layer (B), granule cell layer (C), and Purkinje cell counts (D) in cerebellum at PND 12 after Pam3CSK4 administration from PND 3 to PND 11 in while type mice. Arrow indicates Purkinje cell with higher magnification. *p<0.05.
Long-term effects of neonatal Pam3CSK4 administration
Since we found a decrease in gray and white matter volumes and a decrease in the number of neurons in hippocampus at PND 12 after Pam3CSK4 administration, we examined whether these early brain alterations persisted to young adulthood and related to hippocampus-dependent learning and memory deficits. To examine any long term effect of Pam3CSK4 administration and associated neonatal brain injury on learning and memory function, the Trace Fear Conditioning test was conducted at PND50 (Fig. 8A). At PND53, mice were killed and the weight of the cerebrum, liver, spleen and the whole body were measured. In addition, signs of brain injury for both gray and white matter injury were examined.
Schematic diagram of Trace Fear Conditioning test (A), US: unconditioned stimulus, the foot shock. CS: conditioned stimulus, a particular context and/or a cue. Histograms displays post-tone freezing in the Trace Fear Conditioning test at PND50 (B) in while type mice.
For the Trace Fear Conditioning test, no significant differences were found between the three groups (Fig. 8B). At PND53, the weight of the cerebrum, liver, spleen, and body were not different between groups, except for a decreased liver weight in the LPS-treated group (P = 0.025, Fig. 1C). Neither gray matter nor white matter volumes were different among all three groups (data not shown).
Discussion
In the present study, we found that repeated systemic administration of a TLR2 agonist induced elevated cytokine/chemokine levels in brain homogenates, reduced neonatal gray and white matter volume and hippocampal neuron density, and increased number of microglia cells. By adulthood, brain injury had recovered and there was no detectable long-term change in memory function. To our knowledge, this is the first report of the role of TLR2 agonists on short and long term neonatal brain development. The present study provides important direct evidence that systemic inflammation via TLR2 may exert negative effects on neonatal brain development.
In the rodent, there is a major growth spurt of the brain in the first postnatal week [25], which equates to the second-third trimester in human pregnancy, a developmental window when white matter damage or deficiency of white matter growth is presumed to occur in the human. We used a repeated Pam3CSK4 exposure model from PND3 up to PND11, therefore, covering the period of rapid brain growth in rodents.
To ensure a biologic effect, we used a relatively high dose of Pam3CSK4 (5 mg/kg) compared with other in vivo studies in the adult, that range from 5 µg/kg to 2 mg/kg [26], [27], [28], [29]. However, 5 mg/kg Pam3CSK4 and 0.3 mg/kg LPS treatment produced almost identical levels of KC and MCP-1 in brain homogenates, and despite this relatively high dose, we found no mortality or other signs of morbidity. Similarly, in previous studies we found no adverse effects using the same dose of the TLR2 agonist Lipoteichoic acid (LTA) [30]. These observations suggest that TLR2 agonists have relatively lower potency in neonatal mice compared with the TLR4 agonist LPS.
TLR2 mRNA and protein is expressed in the cortex in embryonic and early postnatal stages of development [31], with relatively low expression before birth that increases during the first 2 weeks of life [32]. Loss of TLR2 does not appear to result in direct defects in cerebral development [31]. However, TLR2 mRNA is expressed constitutively in the adult mouse brain [33] and TLR2 deficiency results in impaired neurogenesis in the hippocampus by adulthood [34]. TLR2 mRNA and genes related to the TLR2 signaling pathway was shown to be induced in the ipsilateral mouse brain hemispheres after transient middle cerebral artery occlusion (MCAO) [35], [36]. Moreover, adult TLR2-deficient mice demonstrated reduced brain damage and improved functional outcome after MCAO [35], [36], [37], though contradictory results have demonstrated a TLR-2 dependent increased brain infarct size after cerebral ischemia/reperfusion injury [38]. Further, in adult mice, hyaluronan blocks oligodendrocyte progenitor maturation and remyelination through TLR2 pathway [39]; Intrathecal administration of Pam3CSK4 induces the pathophysiological hallmarks of bacterial meningitis and neuronal damage in a TLR2-dependent fashion in adult rats [40]. Overall, these observations suggest that activation of TLR2 may be detrimental in acute CNS injury, and together with our observations that a TLR2 agonist causes neonatal brain injury, is suggestive that TLR2 antagonists may have potential as novel neuroprotective agents.
Pam3CSK4 is a synthetic tripalmitoylated lipopeptide that mimics the acylated amino terminus of bacterial lipoproteins, such as is found on the nosocomial pathogen Staphylococcus epidermidis. Pam3CSK4 is specifically recognized by a heterodimer of TLR2 and TLR1, stimulation then resulting in the activation of intracellular signaling events. Although there are few studies on the role of TLR2 in immature brain injury, the TLR2 pathway is thought to play a role in Group B streptococcus-induced neurodegeneration [41]. Moreover, in postnatal day 11 rats, intracisternal injection with 0.5 µg of Pam3CSK4 in the infant rat model of experimental pneumococcal meningitis is capable of inducing a neuroinflammatory response but does not induce hippocampal apoptosis [42]. However, the TLR2 agonist LTA, does not affect vulnerability to hypoxia-ischemia in immature rats [30] and deletion of the gene encoding the adaptor protein MyD88, important for signaling downstream of TLR2 and other TLRs, did not protect the immature brain from hypoxic-ischemic brain injury [14]. Together with our present findings, showing that forebrain and cerebellar volume were recovered by PND53, this suggests that the role of TLR2 in brain injury may be context dependent, with a role in neurodegeneration by whole bacteria that may require engagement of multiple pattern recognition receptors, including several TLRs, NOD-like receptors (NLRs), and complement systems, but a more limited, reversible effect in the context of a pure TLR2 agonist. Similarly, following prenatal stimulation of TLR4 by LPS, there is a transient decrease in myelination and functional outcome which is reversed later in development [43].
Demonstrating the transient effects of Pam3CSK4 on brain injury, the Trace Fear Conditioning test did not detect any learning and memory deficit in young adulthood. Trace Fear Conditioning to either a cue or a context represents a form of associative learning and memory test that has been well characterized in many species [44], and used as a sensitive method to detect hippocampus-dependent learning and memory including in mice [45]. We have previously shown that this is a sensitive test to detect learning and memory function recovery after neonatal hypoxia-ischemia induced brain injury [46]. However, we cannot exclude the possibility of long-term subtle changes in brain structure and functions in the present studies that were not detectable with the present methods, and this will need to be further investigated in the future.
Injury to the cerebellum is becoming increasingly recognized in preterm infants [47], [48]. Also in animal models, reduced number of neurons in cerebellum has been reported in the postnatal guinea-pig [49] and fetal sheep [50] following intrauterine growth-restriction. Moreover, a recent study found a diffuse pattern of cerebellar white matter damage in animals exposed to LPS while there was no obvious injury to the cerebellar cortex or of Purkinje cells [51]. In the present study, we found a significant decrease in the volume of the molecular layer after Pam3CSK4 treatment while there were no differences in the granule cell layer or number of Purkinje cells between groups. These observations suggest that TLR effects on the cerebellum may be region specific.
We found a significant increase in the number of microglia in Pam3CSK4 treated mice, but we saw no such increase in LPS treated animals. It is generally accepted that microglia are responsible for the innate immune response and that microglia express all TLRs, including TLR2 and TLR4, at readily detectable levels [52]. Thus, direct TLR2 stimulation could lead to the activation of microglia and release of pro-inflammatory cytokines, chemokines and free radicals, which could cause toxicity to neurons or oligodendrocytes [53], [54]. Indeed, levels of IL-1ß, IL-6, KC and MCP-1 significantly increased at 6 hours after the first Pam3CSK4 injection at PND3, indicating that the observed gray/white matter changes in the neonatal brain might be at least partly due to cytokine/chemokine toxicity to neurons/oligodendrocytes. Similar levels of most cytokines were seen after both Pam3CSK4 and LPS treatment, except for IL-1ß and IL-6, which was only significantly elevated following TLR2 agonist stimulation but not LPS stimulation. Of note, such observations may be consistent with the polarization of neonatal mononuclear cells towards relatively high TLR2-mediated IL-6 production [55]. Whether such differences in cytokine responses between Pam3CSK4 and LPS treatment contributed to the differences in microglia activation between these two treatments will be the subject of future investigation. Interestingly, IL-1ß is known to sensitize excitotoxic neonatal brain injury [56] and blocking of the IL-1ß receptor protects the immature brain from hypoxic-ischemic brain damage [57]. We did not observe differences in markers of proliferation or apoptosis at least not at PND 12 and 53, but decreased mature neuronal number suggests that effects of Pam3CSK4 on cell survival may have occurred at a time point prior to that examined.
The liver and spleen play a central role in immune responses and the liver is crucial in metabolizing microbial constituents such as Pam3CSK4. Thus the transient enlargement of the spleen and liver in Pam3CSK4 treated mice may indicate an acute reaction of the adaptive immune system and attempts to remove Pam3CSK4 in the blood.
Although our study demonstrates that repeated, high-dose, systemic administration of a TLR2 agonist can lead to CNS injury, it is important to note that these effects are likely context-dependent. Indeed, vaccines containing TLR2 agonists, including intradermal bacille Calmette-Guerin (BCG; Mycobacterium bovis) [58] and certain formulations of the intramuscular Haemophilus influenzae type b vaccine [59], have been safely and effectively administered to millions of infants. This underscores the importance of context, including route, frequency, and dose of administration when considering the impact of TLR agonists in injury models.
In conclusion, we found that systemic administration of a TLR2 agonist to neonatal mice caused transient gray and white matter injury in both the cerebrum and cerebellum. This suggests that engagement of the TLR2 pathway can have detrimental effects on the developing brain, and may play a role in neonatal brain injury associated with bacterial sepsis. However, neonatal brain injury is often multifactorial, and TLR2 agonist effects may interact with other exposures such as hypoxia/ischemia and/or be involved in a broader inflammatory response following Gram-positive bacterial exposure. Accordingly, it is possible that during Gram-positive bacterial infection, combined insults, including those driven via TLR2, may cause long-lasting functional or structural deficits.
Author Contributions
Conceived and designed the experiments: XW CM HH OL. Performed the experiments: XD BF HL BD YS. Analyzed the data: XD BF HL BD YS CZ HH XW OL CM. Contributed reagents/materials/analysis tools: XW XD BF HL BD YS CZ. Wrote the paper: XD XW CM BF CZ HH OL.
References
- 1. Stoll BJ, Hansen N, Fanaroff AA, Wright LL, Carlo WA, et al. (2002) Late-onset sepsis in very low birth weight neonates: the experience of the NICHD Neonatal Research Network. Pediatrics 110: 285–291.
- 2. Stoll BJ, Hansen N, Fanaroff AA, Wright LL, Carlo WA, et al. (2002) Changes in pathogens causing early-onset sepsis in very-low-birth-weight infants. N Engl J Med 347: 240–247.
- 3. Strunk T, Richmond P, Simmer K, Currie A, Levy O, et al. (2007) Neonatal immune responses to coagulase-negative staphylococci. Curr Opin Infect Dis 20: 370–375.
- 4. Bizzarro MJ, Raskind C, Baltimore RS, Gallagher PG (2005) Seventy-five years of neonatal sepsis at Yale: 1928–2003. Pediatrics 116: 595–602.
- 5. Khashu M, Osiovich H, Henry D, Al Khotani A, Solimano A, et al. (2006) Persistent bacteremia and severe thrombocytopenia caused by coagulase-negative Staphylococcus in a neonatal intensive care unit. Pediatrics 117: 340–348.
- 6. Volpe JJ (2009) Brain injury in premature infants: a complex amalgam of destructive and developmental disturbances. Lancet Neurol 8: 110–124.
- 7. Dammann O, Leviton A (1997) Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res 42: 1–8.
- 8. Wolfberg AJ, Dammann O, Gressens P (2007) Anti-inflammatory and immunomodulatory strategies to protect the perinatal brain. Semin Fetal Neonatal Med 12: 296–302.
- 9. Inder TE, Huppi PS, Warfield S, Kikinis R, Zientara GP, et al. (1999) Periventricular white matter injury in the premature infant is followed by reduced cerebral cortical gray matter volume at term. Ann Neurol 46: 755–760.
- 10. Strunk T, Power Coombs MR, Currie AJ, Richmond P, Golenbock DT, et al. (2010) TLR2 mediates recognition of live Staphylococcus epidermidis and clearance of bacteremia. PLoS One 5: e10111.
- 11. Strunk T, Currie A, Richmond P, Simmer K, Burgner D (2010) Innate immunity in human newborn infants: prematurity means more than immaturity. J Matern Fetal Neonatal Med.
- 12. Chau TA, McCully ML, Brintnell W, An G, Kasper KJ, et al. (2009) Toll-like receptor 2 ligands on the staphylococcal cell wall downregulate superantigen-induced T cell activation and prevent toxic shock syndrome. Nat Med 15: 641–648.
- 13. Zhang JP, Yang Y, Levy O, Chen C (2010) Human Neonatal Peripheral Blood Leukocytes Demonstrate Pathogen-Specific Coordinate Expression of TLR2, TLR4/MD2 and MyD88 During Bacterial Infection In Vivo. Pediatr Res Aug 27:
- 14. Mallard C, Wang X, Hagberg H (2009) The role of Toll-like receptors in perinatal brain injury. Clin Perinatol 36: 763–772, v–vi.
- 15. Volpe JJ, Kinney HC, Frances EJ, Rosenberg PA (2011) The developing oligodendrocyte: key cellular target in brain injury in the premature infant. Int J Dev Neurosci Mar 5:
- 16. Levy O (2007) Innate immunity of the newborn: basic mechanisms and clinical correlates. Nat Rev Immunol 7: 379–390.
- 17. Prabhudas M, Adkins B, Gans H, King C, Levy O, et al. (2011) Challenges in infant immunity: implications for responses to infection and vaccines. Nat Immunol 12: 189–194.
- 18. Aliprantis AO, Yang RB, Mark MR, Suggett S, Devaux B, et al. (1999) Cell activation and apoptosis by bacterial lipoproteins through toll-like receptor-2. Science 285: 736–739.
- 19. Wang X, Hellgren G, Lofqvist C, Li W, Hellstrom A, et al. (2009) White matter damage after chronic subclinical inflammation in newborn mice. J Child Neurol 24: 1171–1178.
- 20. Wang X, Hagberg H, Zhu C, Jacobsson B, Mallard C (2007) Effects of intrauterine inflammation on the developing mouse brain. Brain Res 1144: 180–185.
- 21. Svedin P, Hagberg H, Mallard C (2009) Expression of MMP-12 after Neonatal Hypoxic-Ischemic Brain Injury in Mice. Dev Neurosci 31: 427–436.
- 22. Atkins AL, Mashhoon Y, Kantak KM (2008) Hippocampal regulation of contextual cue-induced reinstatement of cocaine-seeking behavior. Pharmacol Biochem Behav 90: 481–491.
- 23. Saxe MD, Battaglia F, Wang JW, Malleret G, David DJ, et al. (2006) Ablation of hippocampal neurogenesis impairs contextual fear conditioning and synaptic plasticity in the dentate gyrus. Proc Natl Acad Sci U S A 103: 17501–17506.
- 24. Monje ML, Toda H, Palmer TD (2003) Inflammatory blockade restores adult hippocampal neurogenesis. Science 302: 1760–1765.
- 25. Dobbing J, Sands J (1979) Comparative aspects of the brain growth spurt. Early Hum Dev 3: 79–83.
- 26. Ha T, Hu Y, Liu L, Lu C, McMullen JR, et al. (2010) TLR2 ligands induce cardioprotection against ischaemia/reperfusion injury through a PI3K/Akt-dependent mechanism. Cardiovasc Res 87: 694–703.
- 27. Hua F, Ma J, Ha T, Kelley J, Williams DL, et al. (2008) Preconditioning with a TLR2 specific ligand increases resistance to cerebral ischemia/reperfusion injury. J Neuroimmunol 199: 75–82.
- 28. Mersmann J, Berkels R, Zacharowski P, Tran N, Koch A, et al. (2010) Preconditioning by toll-like receptor 2 agonist Pam3CSK4 reduces CXCL1-dependent leukocyte recruitment in murine myocardial ischemia/reperfusion injury. Crit Care Med 38: 903–909.
- 29. Long EM, Millen B, Kubes P, Robbins SM (2009) Lipoteichoic acid induces unique inflammatory responses when compared to other toll-like receptor 2 ligands. PLoS One 4: e5601.
- 30. Eklind S, Arvidsson P, Hagberg H, Mallard C (2004) The role of glucose in brain injury following the combination of lipopolysaccharide or lipoteichoic acid and hypoxia-ischemia in neonatal rats. Dev Neurosci 26: 61–67.
- 31. Okun E, Griffioen KJ, Son TG, Lee JH, Roberts NJ, et al. (2010) TLR2 activation inhibits embryonic neural progenitor cell proliferation. J Neurochem 114: 462–474.
- 32. Lathia JD, Okun E, Tang SC, Griffioen K, Cheng A, et al. (2008) Toll-like receptor 3 is a negative regulator of embryonic neural progenitor cell proliferation. J Neurosci 28: 13978–13984.
- 33. Laflamme N, Soucy G, Rivest S (2001) Circulating cell wall components derived from gram-negative, not gram-positive, bacteria cause a profound induction of the gene-encoding Toll-like receptor 2 in the CNS. J Neurochem 79: 648–657.
- 34. Rolls A, Shechter R, London A, Ziv Y, Ronen A, et al. (2007) Toll-like receptors modulate adult hippocampal neurogenesis. Nat Cell Biol 9: 1081–1088.
- 35. Ziegler G, Harhausen D, Schepers C, Hoffmann O, Rohr C, et al. (2007) TLR2 has a detrimental role in mouse transient focal cerebral ischemia. Biochem Biophys Res Commun 359: 574–579.
- 36. Lehnardt S, Lehmann S, Kaul D, Tschimmel K, Hoffmann O, et al. (2007) Toll-like receptor 2 mediates CNS injury in focal cerebral ischemia. J Neuroimmunol 190: 28–33.
- 37. Tang SC, Arumugam TV, Xu X, Cheng A, Mughal MR, et al. (2007) Pivotal role for neuronal Toll-like receptors in ischemic brain injury and functional deficits. Proc Natl Acad Sci U S A 104: 13798–13803.
- 38. Hua F, Ma J, Ha T, Kelley JL, Kao RL, et al. (2009) Differential roles of TLR2 and TLR4 in acute focal cerebral ischemia/reperfusion injury in mice. Brain Res 1262: 100–108.
- 39. Sloane JA, Batt C, Ma Y, Harris ZM, Trapp B, et al. (2010) Hyaluronan blocks oligodendrocyte progenitor maturation and remyelination through TLR2. Proc Natl Acad Sci U S A 107: 11555–11560.
- 40. Hoffmann O, Braun JS, Becker D, Halle A, Freyer D, et al. (2007) TLR2 mediates neuroinflammation and neuronal damage. J Immunol 178: 6476–6481.
- 41. Lehnardt S, Henneke P, Lien E, Kasper DL, Volpe JJ, et al. (2006) A mechanism for neurodegeneration induced by group B streptococci through activation of the TLR2/MyD88 pathway in microglia. J Immunol 177: 583–592.
- 42. Sellner J, Grandgirard D, Gianinazzi C, Landmann RM, Leib SL (2009) Effects of Toll-like receptor 2 agonist Pam(3)CysSK(4) on inflammation and brain damage in experimental pneumococcal meningitis. J Neuroimmunol 206: 28–31.
- 43. Rousset CI, Chalon S, Cantagrel S, Bodard S, Andres C, et al. (2006) Maternal exposure to LPS induces hypomyelination in the internal capsule and programmed cell death in the deep gray matter in newborn rats. Pediatr Res 59: 428–433.
- 44. Kim JJ, Jung MW (2006) Neural circuits and mechanisms involved in Pavlovian fear conditioning: a critical review. Neurosci Biobehav Rev 30: 188–202.
- 45. Gerlai R (1998) Contextual learning and cue association in fear conditioning in mice: a strain comparison and a lesion study. Behav Brain Res 95: 191–203.
- 46. Wang X, Han W, Du X, Zhu C, Carlsson Y, et al. (2010) Neuroprotective Effect of Bax-Inhibiting Peptide on Neonatal Brain Injury. Stroke Sep;41(9): 2050–5.
- 47. Messerschmidt A, Brugger PC, Boltshauser E, Zoder G, Sterniste W, et al. (2005) Disruption of cerebellar development: potential complication of extreme prematurity. AJNR Am J Neuroradiol 26: 1659–1667.
- 48. Volpe JJ (2009) Cerebellum of the premature infant: rapidly developing, vulnerable, clinically important. J Child Neurol 24: 1085–1104.
- 49. Mallard C, Loeliger M, Copolov D, Rees S (2000) Reduced number of neurons in the hippocampus and the cerebellum in the postnatal guinea-pig following intrauterine growth-restriction. Neuroscience 100: 327–333.
- 50. Reddy K, Mallard C, Guan J, Marks K, Bennet L, et al. (1998) Maturational change in the cortical response to hypoperfusion injury in the fetal sheep. Pediatr Res 43: 674–682.
- 51. Dean JM, Farrag D, Zahkouk SA, El Zawahry EY, Hagberg H, et al. (2009) Cerebellar white matter injury following systemic endotoxemia in preterm fetal sheep. Neuroscience 160: 606–615.
- 52. Bsibsi M, Ravid R, Gveric D, van Noort JM (2002) Broad expression of Toll-like receptors in the human central nervous system. J Neuropathol Exp Neurol 61: 1013–1021.
- 53. Nadeau S, Rivest S (2003) Glucocorticoids play a fundamental role in protecting the brain during innate immune response. J Neurosci 23: 5536–5544.
- 54. Lehnardt S, Massillon L, Follett P, Jensen FE, Ratan R, et al. (2003) Activation of innate immunity in the CNS triggers neurodegeneration through a Toll-like receptor 4-dependent pathway. Proc Natl Acad Sci U S A 100: 8514–8519.
- 55. Angelone DF, Wessels MR, Coughlin M, Suter EE, Valentini P, et al. (2006) Innate immunity of the human newborn is polarized toward a high ratio of IL-6/TNF-alpha production in vitro and in vivo. Pediatr Res 60: 205–209.
- 56. Favrais G, Schwendimann L, Gressens P, Lelievre V (2007) Cyclooxygenase-2 mediates the sensitizing effects of systemic IL-1-beta on excitotoxic brain lesions in newborn mice. Neurobiol Dis 25: 496–505.
- 57. Hagberg H, Gilland E, Bona E, Hanson LA, Hahin-Zoric M, et al. (1996) Enhanced expression of interleukin (IL)-1 and IL-6 messenger RNA and bioactive protein after hypoxia-ischemia in neonatal rats. Pediatr Res 40: 603–609.
- 58. Heldwein KA, Liang MD, Andresen TK, Thomas KE, Marty AM, et al. (2003) TLR2 and TLR4 serve distinct roles in the host immune response against Mycobacterium bovis BCG. J Leukoc Biol 74: 277–286.
- 59. Latz E, Franko J, Golenbock DT, Schreiber JR (2004) Haemophilus influenzae type b-outer membrane protein complex glycoconjugate vaccine induces cytokine production by engaging human toll-like receptor 2 (TLR2) and requires the presence of TLR2 for optimal immunogenicity. J Immunol 172: 2431–2438.