| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online June 21, 2002 as doi:10.1096/fj.02-0143fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute for Developmental Research, Aichi Human Service Center, Kasugai, Aichi 480-0392, Japan
2Correspondence: Department of Perinatology, Institute for Developmental Research, Aichi Human Service Center, Kasugai, Aichi 480-0392, Japan. E-mail: katohsemba{at}inst-hsc.pref.aichi.jp
SPECIFIC AIMS
Generating new cells throughout life, the dentate gyrus of the hippocampus is essential for normal recognition memory performance. Reduction of brain-derived neurotrophic factor (BDNF) in this structure impairs its functions. We have conducted a search for compounds that stimulate endogenous BDNF production in the hippocampal granule neurons and examined whether a high level of BDNF enhances hippocampal neurogenesis.
PRINCIPAL FINDINGS
1. Riluzole enhances BDNF expression in the rat hippocampus
BDNF synthesis can be induced in an activity-dependent manner and is markedly stimulated by glutamate agonists and GABA antagonists. However, the majority of these compounds cause convulsions and/or cell death. We therefore investigated the effects of ion channel modulators on BDNF levels in hippocampal granule neurons without inducing cell death. Among the compounds tested, riluzole, a voltage-dependent sodium channel blocker, resulted in a marked increase in the concentration of BDNF protein after 6 h, when it was measured using a two-site enzyme immunoassay system. Bay-K8644, an activator of L-type voltage-dependent calcium channels, also enhanced BDNF levels slightly but not significantly. No effects were found with other ion channel modulators. Levels of BDNF mRNA and its protein reached respective peaks 4 and 8 h after riluzole injection. Western blotting analysis demonstrated increases not only in mature BDNF (molecular mass 14 kDa), but also precursor BDNF (30 kDa).
2. Distribution of riluzole-enhanced BDNF in the brain
Riluzole significantly increased BDNF levels in the rat hippocampus, entorhinal cortex, and hypothalamus 
5-, 10-, and 2-fold, respectively, compared with respective controls (Fig. 1
A). Results of cresyl-violet staining of hippocampal sections of riluzole-treated rats showed neither bulk loss of any neurons nor necrotic and/or atrophic cells after 48 h (Fig. 1B-a
). Analysis of apoptotic cells using semi-serial sections to those used for Fig. 2
C showed no alteration in number between the two groups even though riluzole was repeatedly injected. Immunohistochemical localization of BDNF showed that in control sections, the hilar region (single arrow) and the stratum radiatum of the CA3 region (double arrows) were stained with BDNF-specific antibodies (Fig. 1B-b
). In riluzole-injected rats, immunoreactivity was markedly enhanced after 8 h in the granule cell layer (GCL; arrowhead), the hilar region, and the stratum radiatum (Fig. 1B-c
), but not in the CA1 region (Fig. 1B-d
). Antibodies stained intensely the cytoplasm of granule cells and weakly cells in the subgranular layer (Fig. 1B-f
). Staining was enhanced in the pathways of mossy fibers of the hilar region and the stratum radiatum but not in neurons (Fig. 1B, e, f
). Speckled and/or homogeneous staining patterns of enhanced BDNF in hilar region and the stratum radiatum (Fig. 1B, e and f
, respectively) suggest localization in the extracellular space similar to the case with kainic acid-injected rats.
|
|
3. Repeated injections of riluzole cause persistent increase in hippocampal BDNF
BDNF protein was still relatively high 24 h after a single injection of riluzole. We examined whether repeated injections resulted in a persistent increase in endogenous BDNF. Riluzole at various doses was injected every 24 h for 5 days (see Fig. 2A
), and amounts of hippocampal neurotrophins were measured 24 h after each riluzole injection. BDNF was maintained at high levels at least for 5 days. In contrast, neurotrophin 3 levels were clearly reduced and nerve growth factor did not alter.
4. Repeated riluzole injections promote precursor cell proliferation but do not support cell survival
We first investigated the effects of repeated-riluzole injections on cell survival. As depicted in Fig. 2A
, riluzole administration began just after the second bromodeoxyuridine (BrdU) injection and was repeated according to the schedule. When BrdU+ cells were determined at 5 days, there was no significant difference in mean numbers in the GCL between controls and riluzole-injected rats (Fig. 2B
). Even though a high level of BDNF was maintained for 2 wk by riluzole injections, the number of BrdU+ cells was again similar in both groups. In the next step, we determined whether riluzole promoted proliferation of precursor cells. Riluzole was injected five times before BrdU labeling to generate a high level of BDNF for 5 days (Fig. 2A
). At 16 h after the last BrdU injection (at the 5 day point), more labeled cells were found in the GCL of riluzole-injected rats than in controls (Fig. 2C
, left). The mean increase for the entire GCL was threefold (Fig. 2C
, right). Numbers of BrdU+ cells 16 h later in control rats were lower than in the survival experiment (at the 5 day point) due to increased numbers of labeled cells at 48 h and no alteration during the next 3 days. To determine phenotypes of the BrdU+ cells, rats were maintained for another 4 wk after the last riluzole injection without additional treatment (see Fig. 2A
). In control rats, the mean number of BrdU+ cells at 32 days was reduced to about a half the peak value (cell numbers 48 h after BrdU injection). However, the number for the riluzole-injected group was still threefold that for the control group, suggesting no difference in the percentage of degenerating cells. Analyses of confocal microscopic images using sections from riluzole-injected rats showed that NeuN, a neuronal nuclear marker, was expressed in most granule neurons (Fig. 2D
; green cells). BrdU+ cells (red cells) found in the dentate gyrus coexpressed NeuN (yellow cells). In both groups, 90% of BrdU+ cells coexpressed NeuN and only 5% coexpressed S-100ß, a glial marker.
5. Intraventricular administration of BDNF-specific antibodies blocks riluzole-promoted proliferation of precursor cells
To determine whether riluzole-induced enhancement of BDNF is necessary to promote proliferation of granule precursors, we intraventricularly administrated BDNF-specific antibodies in addition to giving intraperitoneal injections of riluzole and BrdU. When riluzole injections were given > 3 days before BrdU labeling, the mean number of BrdU+ cells was twofold that in controls. In contrast, a single injection of riluzole had little effect on numbers of BrdU+ cells. Injections of BDNF-specific antibodies caused a significant decrease in numbers of BrdU+ cells in the GCL.
CONCLUSIONS
We here demonstrated that riluzole injections enhance not only the BDNF concentration, but also the numbers of newly generated cells in the rat hippocampus. Furthermore, the present investigation revealed repeated injections of riluzole to lack any influence on survival of precursor cells although they induced a persistent increase in BDNF. The indication is that riluzole promotes precursor proliferation and that this effect occurs through enhancement of BDNF. It has been observed that BDNF is elevated in association with an increase in numbers of BrdU+ cells in the hippocampus. The present study directly demonstrated that elevation of BDNF, particularly its persistent increase, is necessary for promoted proliferation of hippocampal precursor cells. Some steroid hormones and neurotransmitter receptor antagonists promote G1/S transition in the cell cycle in the short term, apparently due to direct effects on the cell cycle. However, this was not the case in the present study, considering that persistent, but not transient, enhancement of BDNF levels is necessary for increase in numbers of BrdU+ cells.
An enriched environment, dietary restriction, and apoptotic degeneration of neurons increase the number of BrdU+ cells in the hippocampal dentate gyrus. Although mechanisms underlying the increase in numbers observed here remain unclear, evidence is accumulating that induction of long-term potentiation (LTP) in the perforant path may promote precursor proliferation in this structure. On the other hand, BDNF plays a role in a form of LTP by enhancing glutamatergic synaptic transmission in presynaptic loci. BDNF enhancement is observed specifically in the hippocampus and entorhinal cortex.
BDNF synthesis is stimulated in an activity-dependent manner. In contrast, riluzole blocks voltage-dependent sodium channels and suppresses the activity of neurons. The question therefore arises as to why riluzole can stimulate BDNF synthesis. This may be answered by the fact that induction of LTP in the hippocampus is regulated by the concentration of extracellular adenosine, produced partly from ecto-ATP by ecto-nucleotidase. Blockade of voltage-dependent sodium channels can reduce ATP release, resulting in decreased ecto-adenosine levels, followed by induction of LTP. This is supported by the observation that agonists of adenosine receptors and enhancement of extracellular adenosine levels by intraventricular administration of apyrase (ectonucleases) both decrease riluzole-induced BDNF production (unpublished observations).
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0143fje; to cite this article, use FASEB J. (June 21, 2002) 10.1096/fj.02-0143fje 
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
