| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
,¶
* Neuronal Stem Cells Research Group,
Molecular Tumor Genetics Research Group, and
Molecular Immunology and Genetherapy Research Group, Max Delbrück Centre for Molecular Medicine, Berlin-Buch, Germany;
Institute of Biological Sciences, Institute of Neurosurgery, Stanford University, Stanford, CA, USA;
|| Volkswagenstiftung Research Group, Department of Experimental Neurology, and
¶ Institutes of Immunology, Charité, Berlin, Germany; and
# Department of Neurology, Charité University Medicine Berlin, Campus Virchow Klinikum, Berlin, Germany
2 Correspondence: CRTD-DFG-Center for Regenerative Therapies Dresden, Tatzberg 47-49, 01307 Dresden, Germany. E-mail: gerd.kempermann{at}crt-dresden.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: CD4 • glucocorticoids • rheumatoid arthritis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4
5)
. One of the prime examples of brain plasticity is adult neurogenesis: the exceptional case in which new neurons are generated throughout adulthood (6
7
8)
. One of the two regions of adult neurogenesis is the hippocampus, a brain region centrally involved in learning and memory (8
, 9)
. It was noted early that strong stressors robustly down-regulated adult hippocampal neurogenesis and that elevated glucocorticoid levels are involved in this process (3
, 10
, 11)
. On the other hand, exposure to an enriched environment or voluntary physical activity, which both increase adult neurogenesis (12
, 13)
, is also associated with elevated glucocorticoid levels, suggesting a dose dependency of the effect.
Among ethologically relevant stressors, immune responses rank highly (14)
. Peripheral lipopolysaccaride (LPS) application that initially activates an innate immune response robustly decreases adult neurogenesis (15
, 16)
. On the other hand, the infiltration of the brain by potentially autoreactive central nervous system (CNS)-specific T cells led to an increase in adult neurogenesis (17)
. However, so far, no information exists about the effect of an adaptive T-cell-dependent immune response during peripheral inflammation on adult neurogenesis or the mechanism of this phenomenon. We therefore studied examples of both adaptive and innate immune responses for their effect on adult hippocampal neurogenesis and the potential involvement of corticosterone in mediating such effects.
Adult hippocampal neurogenesis allows lifelong adaptation of the hippocampal neuronal network to levels of complexity and novelty encountered by an individual (7
, 18
, 19)
. During this process, systemic stimuli, such as locomotion, activate the proliferative precursor cells (12
, 19
20
21)
, whereas cognitive stimuli, as exemplified by the situation of environmental enrichment, recruit new neurons from the pool of newborn immature neurons (22
, 23)
. In this sense, physical activity would indicate to the brain that there is a high chance of encountering cognitively challenging situations. The hippocampal formation reacts by generating new neuronal progenitor cells on demand (24)
. We propose that adaptive immune responses might serve a similar purpose as physical activity, signaling a possible need for new progenitor cells into the neurogenic regions of the brain.
As a model for peripheral T-cell activation, we used experimental antigen-induced arthritis (AIA) in the knee joint of female C57Bl/6 mice. We also injected mice either with Staphylococcus enterotoxin (SEB) as specific model for peripheral T-cell activation or with LPS as model for a primary innate immune response. Precursor cell proliferation and neurogenesis in the dentate gyrus were assessed at different time points after the onset of the immune response. Levels of corticosterone in the hippocampus, the major glucocorticoid in rodents, were measured in all treatment groups. The direct effect of high and low corticosterone on precursor cell proliferation and differentiation was tested in primary hippocampal precursor cell cultures.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
For AIA induction, mice were immunized by subcutaneous injection of 100 µg methylated bovine serum albumin (mBSA; Sigma-Aldrich, Steinheim, Germany) in 50 µl PBS, emulsified in 50 µl complete Freund’s adjuvant (CFA; Sigma-Aldrich, Steinheim, Germany), additionally supplemented with 4 mg/ml of the heat-inactivated Mycobacterium tuberculosis strain H37RA (Difco, Detroit, MI, USA), or incomplete Freund’s adjuvant (IFA; Pierce, Rockford, IL, USA) on d –21 (CFA) and –14 (IFA). In parallel to each antigen-specific immunization, 200 ng of Bordetella PTX (Calbiochem, La Jolla, CA, USA) was injected intraperitoneally. Arthritis was induced on d 0 by injection of 100 µg mBSA dissolved in 20 µl PBS into the left knee joint, while 20 µl PBS was injected into the right knee joint as a control. As a standard, mBSA-induced AIA in mice was observed over 3 wk. Animals were thus analyzed at d 3, 7, and 21 after arthritis induction. Days 1 to 3 after arthritis induction are considered the acute phase and are associated with knee joint swelling, exudates, and massive granulocytic infiltration (Fig. 1A1, A2
). The consecutive days up to d 21 are defined as the chronic phase and are characterized by synovial hyperplasia, mononuclear cell infiltration, fibrosis, and cartilage and bone erosion (Fig. 1A3, A4
). The knee joint diameter was measured using a caliper (Mattoon, Tokyo, Japan) at d 0 and 1 after arthritis induction. Joint swelling was expressed as the difference (in mm) between d 0 and 1 of the same knee joint.
|
SEB and LPS model
To systemically activate Vβ-positive T cells, SEB S4881 (Sigma-Aldrich) was applied. Female adult C57BL6 mice (
5/group) received a single intraperitoneal injection of SEB in the concentration of 0.05, 0.5, or 2 µg/ml dissolved in 0.9% NaCl. LPS (from Escherichia coli 0111:B4; L3012; Sigma-Aldrich) injections were done in parallel with the SEB injections in 5 female C57Bl/6 animals at 0.5 µg/ml in 0.9% NaCl. Pure 0.9% NaCl-treated mice served as controls. A single intraperitoneal injection of BrdU (Sigma-Aldrich) was applied 3 d later. Mice were killed 24 h after the BrdU injection, and their brains were investigated as described.
Immunohistochemistry
Animals were deeply anesthetized with ketamine and perfused transcardially with 0.9% NaCl. The brains were dissected from the skulls, and from 3 animals/group, one hemisphere was homogenized for corticosterone RIA into dilution reagent (see below); the other hemisphere or the whole brains were fixed overnight in 4% paraformaldehyde. Before being sectioned from a dry-ice-cooled copper block on a sliding microtome (Leica, Bensheim, Germany), the brains were kept in 30% sucrose in 0.1 M phosphate buffer, pH 7.4, overnight. Brains were cut in the coronal plane in 40-µm-thick sections and cryoprotected. For BrdU detection, DNA was denatured in 2 N HCL for 30 min at 37°C. Free floating sections were then rinsed in 0.1 M borate buffer, pH 8.5, and thoroughly washed in TBS, pH 7.4. To block endogenous peroxidase reactions, sections were pretreated with 0.6% H2O2. Sections were incubated with primary antibodies in TBS supplemented with 0.1% TritonX-100 and 3% donkey serum (TBS-plus) overnight at 4°C. After the sections were rinsed in TBS and a blocking step in TBS-plus was performed, an incubation step followed, with the biotinylated secondary antibody diluted 1:500 in TBS-plus. ABC reagent (Vectastain Elite; Vector Laboratories, Burlingame, CA, USA) was applied for 1 h at a concentration of 9 µl/ml for each reagent. Diaminobenzidine (DAB; Sigma-Aldrich) was used as a chromogen at a concentration of 0.25 mg/ml in TBS with 0.01% H2O2 and 0.04% nickel chloride, followed by rinsing with tap water and TBS. The sections were mounted on slides and coverslipped with Neomount.
Cell cultures were fixed with cooled 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 20–30 min. After being washed with TBS, cells were blocked with 3% donkey serum (Millipore/Chemicon, Elze, Germany) containing 0.2% Triton X-100. Primary antibodies were diluted in blocking buffer, and the cells were incubated overnight. After being washed with TBS, secondary antibody was diluted in TBS and the cells were incubated for 2 h at room temperature.
The primary antibodies were applied in the following concentrations: goat anti-doublecortin (DCX; 1:200; Santa Cruz Biotechnologies, Heidelberg, Germany), mouse anti-NeuN (1:100; Millipore/Chemicon, Elze, Germany), rabbit anti-Iba-1 (1:1000, Millipore/Chemicon), and rat anti-BrdU (1:500; Biozol, Echingen, Germany). As secondary antibodies, we used biotinylated donkey-anti-goat, donkey-anti-IgG, and donkey-anti-rabbit, conjugated to different fluorophores (all 1:250; Dianova, Hamburg, Germany).
RIA for corticosterone
Hippocampal samples were taken from 5 animals treated either with SEB, LPS, or vehicle only. Also, one hippocampus of 3 animals from each of the AIA groups [AIA, pertussis toxin with Freund’s adjuvant (FA-PTX), and control] was subjected to corticosterone measurements using an RIA (Diagnostic System Laboratories, Inc., Webster, TX, USA) according to the manufacturer’s protocol. The brain samples were homogenized in the dilution reagent supplied by the manufacturer, and 1:100 dilutions were used in the assay.
Neuronal precursor cell cultures
Neuronal precursor cells were derived from the dentate gyrus of postnatal day 20 mice, as described previously (25)
. The pooled tissue was dissociated to obtain a single-cell suspension. To assess precursor cell enrichment, the cell mixture underwent centrifugation at 20,000 g for 30 min in a continuous Percoll gradient. Cell culture dishes were coated with 10 µg/ml polylysine overnight at room temperature and with 5 µg/ml laminin (Sigma-Aldrich) at 37°C overnight.
After excess coating solution was removed, cells were plated directly onto the surface. The cultures were maintained in proliferation medium consisting of Neurobasal with B27 or N2 supplement, 2 mM Glutamax (Invitrogen, Carlsbad, CA, USA), 1 µg/ml Pen-Strep (Sigma-Aldrich), 20 ng/ml human fibroblast growth factor-2, and 20 ng/ml human epidermal growth factor (Pepro Tech, London, UK). The medium was replaced with fresh medium the next day. Subsequently, the cultures were fed with new medium every 2 or 3 d by replacing 75% of the medium. We maintained a cell density of 104 cells/cm2. For stress hormone experiments, cultures were treated for 24 h with corticosterone (Sigma-Aldrich; 1.5 or 0.25 µM in 0.001% ethanol) or an ethanol vehicle (26). In the blocking experiments, viral vector infection was performed 16 h before hormone treatment. BrdU (20 µM; Sigma-Aldrich) was added to cultures 20 min before the incubation with corticosterone for 24 h, and cultures were processed for BrdU immunostaining as described. BrdU-positive cells in culture were counted in 5 fields/well (center and at 3, 6, 9, and 12 o’clock) using a light field microscope (Leica). Trypan blue staining (0.8 mM, 2 min; Sigma-Aldrich) was performed in 200 µl cell suspension, and cells were counted in 10 µl in a hemicytometer to obtain the percentage of dead cells after 24 h corticosterone treatment.
Virus preparation
The viral vector containing a transdomain form of the glucocorticoid receptor (GR) and thus inhibiting its translocation into the nucleus was kindly provided by Prof. Robert Sapolsky (Biological Sciences, Stanford University, Stanford, CA, USA). The construction is covered in detail by Kaufer et al. (26)
. Viral production was done according to the protocol provided by the Sapolsky laboratory (27)
. Briefly, the virus is generated by transfection of plasmids into embryonic day 5 cells using lipofectamine and superinfecting 24 h later with the helper virus d120 at a multiplicity of infection of 0.3. Amplicons are harvested by freeze-thaw lysis followed by sucrose cushion centrifugation at 20,000 g overnight. Pelleted viruses are resuspended in ddH2O and frozen at –80°C. Expression of the genes encoding the enhanced green fluorescent protein fusion constructs was directly assayed by examining enhanced green fluorescent protein levels under fluorescent microscopy (Leica).
Statistics
Numerical data (shown as means±SE) were analyzed by 1-way ANOVA and Fisher’s test, where appropriate, using Stat view 5.01 for Macintosh (SAS Institute, Cary, NC, USA). Key ANOVA parameters (F values, degrees of freedom) are given in the description of results. Where appropriate, Student’s t test was performed. The level of significance was preset at P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 29)
with no known phenotype in the CNS.
At either 3 d or 3 wk after induction of AIA, we found no significant difference in the number of precursor cells of the dentate gyrus between AIA animals and the two control conditions: mice treated with BSA and FA-PTX (3 d: P=0.1881) and naive controls (P=0.7282, ANOVA P=0.404, F3,16=0.964; 3 wk: P=0.1305, ANOVA P=0.426, F3,16=3.986; Fig. 1C
). At 7 d after AIA induction, the degree of knee swelling as a measure of the local inflammatory response in AIA was correlated with an increased number of BrdU-positive cells in the gyrus dentatus (r2=0.75; Fig. 1D
). At 7 d after AIA induction, we detected a significant increase in the number of BrdU-labeled cells in the gyrus dentatus compared with FA-PTX-treated (P=0.0188) and untreated controls (P=0.0122, ANOVA P=0.0029, F3,16=8.883; Fig. 1B, C
). The majority of BrdU-positive cells expressed doublecortin (DCX), indicating that they were in the neuronal lineage. The number of BrdU-labeled DCX-positive cells was significantly increased in AIA animals compared with FA-PTX-treated mice (P=0.0001) and untreated controls (P=0.0006, ANOVA P<0.0001, F3,16=17.072). FA-PTX alone had no effect on the generation of DCX-positive cells (Fig. 1B, E
). There was no effect on the generation of new microglial cells (as shown by Iba1 staining) between the groups, showing that no activation of resident inflammatory cells in the CNS occurred in neither of the treatments (Fig. 1E
). We did see an effect on BrdU-positive cells of undetermined phenotype (Fig. 1E
, BrdU other) in the FA-PTX and the AIA groups, possibly indicating an effect on astrocytes.
Systemic activation of CD4-positive T cells in an SEB model increases the numbers of neuronal progenitor cells in the dentate gyrus
As a second model for specific T-cell activation without using any adjuvant, we used SEB (30)
. Similar to the AIA experiment, application of SEB and thus activating T cells in the periphery resulted in a significant increase in BrdU-positive cells in the dentate gyrus (P=0.0014, F3,15=6.808; Fig. 2A
) and BrdU-positive/DCX-positive cells in the dentate gyrus compared with controls (P=0.0109, ANOVA P=0.019, F2,10=11.797; Fig. 2B
). Thus, peripheral activation of T lymphocytes led to an increase in the number of proliferating progenitor cells in the neuronal lineage. When we stained SEB-treated vs. nontreated C57Bl/6 brain sections for CD3 to detect T cells, we found <10 CD3-positive T cells per whole brain in either group (Fig. 2C
).
|
AIA, LPS, and SEB differently increase corticosterone levels in the brain
First, we measured corticosterone levels in the hippocampus 12 and 24 h after SEB or LPS application. As expected, peripheral LPS injection led to a significant long-term 4- to 5-fold increase in corticosterone levels in the hippocampus at both time points [12 h: control (CTR) 126±23 vs. LPS 520±53 pg/dcl, P=0.001; 24 h: CTR 120±19 vs. LPS 525±37 pg/dcl, P=0.0012; Fig. 3A
].
|
A transient 2-fold increase was detected 12 h after SEB injection (CTR 126±23 vs. SEB 268±40 pg/dcl, P=0.014; Fig. 3A
) followed by a decrease toward baseline at 24 h after injection (CTR 120±19 vs. SEB 78±11 pg/dcl; Fig. 3A
).
Due to the different time course of AIA, we measured corticosterone levels at 3, 7, and 21 d after AIA induction. The peak of corticosterone levels in the hippocampus at d 7 coincided with the peak of progenitor cell proliferation (CTR 7 d 130±16 vs. AIA 7 d 286±22 pg/dcl, P=0.003; Fig. 3B
).
Dose-dependent effects of corticosterone levels on survival and proliferation of neuronal precursor cells in vitro
To investigate the direct effects of corticosterone on hippocampal precursor cells, we treated primary neuronal progenitor cultures with low (0.25 nM) and high (1.5 nM) corticosterone (high/lowCORT) concentrations. Proliferation was measured by BrdU incorporation similar to the in vivo assay. Cell death was evaluated by trypan blue. In 3 independent experiments, we detected an increase in the average number of BrdU cells per field in the lowCORT group and a decrease in the highCORT group vs. the noCORT group (GFP: noCORT 19.6±2.0 vs. lowCORT 46.7±1.9 vs. highCORT 10.2±1.05, ANOVA P=0.0024; F3,15=9.445; Fig. 3C
). The decrease of BrdU-labeled cells in the highCORT group vs. the noCORT group was paralleled by a 40% increase in the numbers of trypan blue-labeled cells (noCORT 0.21±0.12 vs. highCORT 44.29±12.57%, P=0.0001; Fig. 3D
).
To show that the observed effect was mediated by the GR, we preincubated the neuronal precursor cells with a Herpes simplex virus-1 amplicon vector construct expressing a transdomain form of GR to block GR translocation and thus activation of downstream targets. With this treatment, we could restore BrdU cell numbers and survival in the highCORT group (Fig. 3C, D
). There were no differences between GFP and no-vector control treatment (data not shown).
Blocking of the GR also inhibited the proliferative effect of low corticosterone application. BrdU levels were similar across the groups [transdomain GR vector (TD): noCORT 18.1±1.1 vs. lowCORT 22.6±3.5; highCORT 17.6±0.6; Fig. 3C
].
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In line with this key property of the model, we did not detect changes in microglial proliferative activity in all experimental groups. However, we found significantly increased numbers of proliferating DCX-positive precursor cells in AIA animals compared with controls. The increased cell proliferation correlated with the progress and severity of the disease as measured by knee joint swelling and was paralleled by an increased corticosterone level in the hippocampus. In line with our findings, it has been reported that in AIA, chronic activation of the hypothalamic-pituitary-adrenal axis is seen at 7–21 d after adjuvant injection (31)
, causing an increase in serum corticosterone (32)
. Similarly, physical activity, known to be a positive regulator of adult neurogenesis, also moderately up-regulates serum and hippocampal corticosterone levels (33)
.
Sloviter et al. (34)
have shown that removing the adrenal gland and blocking glucocorticoid production resulted in a near-complete loss of granule cells in the dentate gyrus, suggesting a major survival effect of systemic glucocorticoids under physiological conditions. After Gould and colleagues (35
36
37)
reported an increase in adult hippocampal neurogenesis in adrenalectomized rats, adrenal hormones were believed to quite generally suppress adult neurogenesis with some reports, including those on exercise effects on adult neurogenesis, challenging that paradigm (38
, 39)
.
We could here replicate the antiproliferative effect of corticosterone with either a high dose of corticosterone in vitro or a 5-fold increase of corticosterone in the hippocampus after systemic LPS injection. In similar experiments by two other groups using systemic LPS injections to trigger an innate immune response, microglia activation in general (16)
and IL-6 in particular (15)
have been implied as mediators for decreased neurogenesis. In addition to the cytokine-mediated effects, our data point toward a possible pivotal role of chronically elevated hippocampal corticosterone levels in blunting neurogenesis during systemic inflammation. To distinguish between cytokine and corticosterone effects, additional research is needed. The response triggered by LPS is also to a certain degree dependent on the exact composition of the concrete preparation used. Given the effect size in the present experiment and the general validity of the model, this issue is unlikely to confound our interpretation.
When we applied a low dose of corticosterone on neuronal precursor cells in vitro or triggered a small or transient increase of corticosterone in the hippocampus by systemic SEB injection or AIA induction, we detected an increase in neuronal precursor cell proliferation.
In mature neurons, mineralocorticoid receptor (MR) and GR are coexpressed in abundance (40)
. The paradoxical actions of glucocorticoids are most evident through their involvement in central functions essential for neuronal survival and remodeling as well as neuronal death. The beneficial or homeostatic effects of glucocorticoids are predominantly mediated by the MR, with high affinity to glucocorticoids, while neurotoxic effects are mainly dependent on GR, with lower glucocorticoid affinity than MR (41
, 42)
. Recent studies (43)
highlight the neuroprotective potential of MR overexpression on granule cells in the dentate gyrus. Although mineralocorticoid activity has been implied in the process of neurogenesis (44)
, in our previous immunohistochemical study (45)
, we could not detect any MR expression during the proliferative stages of neuronal progenitor maturation in the adult hippocampus. MR expression was only present at advanced postmitotic stages (young mature neurons), identified by the expression of calbindin. GR expression, on the other hand, was seen in 50% of the mitotic radial-glia-like precursor cell population and at a later neuronal precursor maturation stage corresponding with a DCX-positive/nestin-negative expression pattern (45)
. While the dual actions of glucocorticoids on mature neurons can be explained by the balanced action of the bipolar MR/GR system (1
, 41)
, direct regulation of neuronal progenitor cell proliferation by glucocorticoids is most likely facilitated through the GR. We could strengthen this hypothesis by showing that the high- and low-corticosterone effects could both be blocked by a viral vector construct specifically blocking GR binding only (26)
. Future in vivo studies including titration of GR-blocking-vector effects, as well as receptor-binding studies, which were not in the focus of the present study, will elucidate the kinetics of this phenomenon. From our observations here in this first study, we suggest that differential levels of corticosterone in the hippocampus caused by either a peripheral innate (LPS) or an adaptive (SEB, AIA) immune response could account for bidirectional changes in adult hippocampal neurogenesis.
In addition, the observed effects might be indirect and involve secondary signaling from other cells, most notably interneurons that are known to modulate neuronal precursor cell fate (46)
. Recent experimental evidence (47
, 48)
indicates that T cells can instruct microglia to adopt a neuroprotective phenotype, which in turn could mediate the neurogenic effect observed in our models of peripheral T-cell activation. However, our in vitro data indicate that a similar effect of corticosterone is also found in isolated precursor cells.
We also cannot exclude that apart from regulations by glucocorticoids, differences in the cytokine profile or concentration over the time course of AIA might have an effect on adult hippocampal neurogenesis. In the acute phase, IL12/p40 is the dominant serum cytokine; in later stages of AIA, serum IL-1 and IL-6 concentrations increase (49)
. However, the expected effect of these cytokines would be a decrease rather than an increase of neurogenesis, as has been shown for microglia-derived IL-6 after systemic LPS injection (15)
. Therefore, the cytokines produced in AIA are unlikely candidates as direct mediators of precursor cell proliferation. Persistent pain in AIA would result in prolonged stress and consecutive down-regulation of adult neurogenesis (50)
, as numerous other studies of stress effects on adult neurogenesis have shown (reviewed, e.g., in ref. 12
). LPS, in contrast, might mimic such stress-like effects that are paralleled by high glucocorticoid levels in the hippocampus.
Application of SEB was used as a model of a more T-cell-specific peripheral immune response. SEB engages the immune system by stimulating T cells bearing Vβ chains of the T-cell receptor (51)
. SEB also stimulates the HPA axis by inducing corticosteroid release (52)
. In concordance to the AIA experiment, we report here a significant increase in hippocampal neurogenesis after systemic SEB application, pointing toward a neurogenic effect by peripheral T-cell activation. Michal Schwartz’s group (17)
has first reported neurogenic effects of CNS-specific T cells. The mode of action of these potential autoreactive CNS-specific T cells with regard to neurogenesis is still under investigation. As one possible mechanism, CNS-specific T cells might enter the CNS and stimulate neurogenesis by secreting brain-derived neurotrophic factor in proximity to the progenitor cells (17
, 53)
. In partial contrast to those findings, we show here that peripheral activation of potential nonencephalotogenic, non-CNS-specific T cells either by a local inflammation of the knee joint or by subjection to T-cell-stimulating agents triggers hippocampal neurogenesis without obvious T-cell infiltration into the brain parenchyma. In concordance with findings by Ekdahl et al. (48)
, our data do not support the idea that cell-cell contact-dependent T-lymphocyte interaction in the neurogenic niche is of significant importance for neurogenesis.
We obviously cannot rule out that cells other than lymphocytes infiltrating the brain from the periphery contributed to the observed effects. We could not detect changes in the number of CD3+ cells between vehicle and SEB-treated mice, but other bone marrow-derived cells could be involved in mediating the transient increase of precursor cell proliferation. For the cerebellum, it has been reported that chronic inflammation dramatically increases heterokaryon formation (54)
. Although fusion of bone marrow-derived cells has been detected in Purkinje cells (55)
, fusion has been extremely rare and shows a high tropism. So far, no fusion events in the dentate gyrus have been reported (56)
. This possible novel type of interaction between bone marrow and brain cells has enriched the search for a mediator between immune response and neurogenesis and eventually cognitive or motor functions.
The well-known positive effects of exercise (20
, 33)
and enriched environment (57
, 58)
on adult neurogenesis are paralleled by an increase in corticosterone levels in the hippocampus. We here propose that a mild transient activation of the adaptive immune system could act on neurogenesis in the same way as voluntary exercise and mild acute stress by elevating corticosterone levels just enough to trigger changes in cell-based synaptic plasticity, with GR occupation as the primary sensor. Stressors that trigger cell proliferation in the dentate gyrus might increase the pool of precursor cells in situations in which a network modulation is demanded by a situation of novelty and complexity (7)
. Among other factors, an adaptive immune response could modulate neurogenesis levels via a transient and small increase of hippocampal glucocorticoids, and thereby increase the potential for network adaptation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication September 11, 2008. Accepted for publication April 16, 2009.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
0.05 vs. control.
10/hippocampus) were detected in hippocampi of SEB-treated animals. Scale bar = 50 µm.