Dual response of BDNF to sublethal concentrations of beta-amyloid peptides in cultured cortical neurons.

Beta-amyloid (Abeta) deposition is one important pathological hallmark in Alzheimer's disease (AD). However, low levels of Abeta may modify critical endogenous protection systems before neurodegeneration occurs. We examined the time-course effect of sublethal concentrations of Abeta on total BDNF (panBDNF), BDNF transcripts (I, II, IV and VI), trkB.FL, trkB.T1 and p75(NGFR) mRNA expression in cultured cortical neurons. We have shown that Abeta exhibited a dual response on BDNF mRNA, i.e. an increase at short times (3-5 h) and a dramatic decrease at longer times (24 or 48 h). The early increase in BDNF expression seems to be driven by increased expression of transcripts I and IV. The BDNF drop was specific since did not occur for other mRNAs examined. The BDNF protein content showed a similar profile but did not follow the dramatic reduction as its encoding mRNA. These observations may help to explain cognitive deficits observed at initial stages of AD.

Regulation of mitochondrial mRNA stability by RNase L is translation-dependent and controls IFNalpha-induced apoptosis.

Interferons (IFNs) inhibit the growth of many different cell types by altering the expression of specific genes. IFNs activities are partly mediated by the 2'-5' oligoadenylates-RNase L RNA decay pathway. RNase L is an endoribonuclease requiring activation by 2'-5' oligoadenylates to cleave single-stranded RNA. Here, we present evidence that degradation of mitochondrial mRNA by RNase L leads to cytochrome c release and caspase 3 activation during IFNalpha-induced apoptosis. We identify and characterize the mitochondrial translation initiation factor (IF2mt) as a new partner of RNase L. Moreover, we show that specific inhibition of mitochondrial translation with chloramphenicol inhibits the IFNalpha-induced degradation of mitochondrial mRNA by RNase L. Finally, we demonstrate that overexpression of IF2mt in human H9 cells stabilizes mitochondrial mRNA, inhibits apoptosis induced by IFNalpha and partially reverses IFNalpha-cell growth inhibition. On the basis of our results, we propose a model describing how RNase L regulates mitochondrial mRNA stability through its interaction with IF2mt.

Protective effect of BDNF against beta-amyloid induced neurotoxicity in vitro and in vivo in rats.

We examined the potential protective effect of BDNF against beta-amyloid-induced neurotoxicity in vitro and in vivo in rats. In neuronal cultures, BDNF had specific and dose-response protective effects on neuronal toxicity induced by Abeta(1-42) and Abeta(25-35). It completely reversed the toxic action induced by Abeta(1-42) and partially that induced by Abeta(25-35). These effects involved TrkB receptor activation since they were inhibited by K252a. Catalytic BDNF receptors (TrkB.FL) were localized in vitro in cortical neurons (mRNA and protein). In in vivo experiments, Abeta(25-35) was administered into the indusium griseum or the third ventricle and several parameters were measured 7 days later to evaluate potential Abeta(25-35)/BDNF interactions, i.e. local measurement of BDNF release, number of hippocampal hilar cells expressing SRIH mRNA and assessment of the corpus callosum damage (morphological examination, pyknotic nuclei counting and axon labeling with anti-MBP antibody). We conclude that BDNF possesses neuroprotective properties against toxic effects of Abeta peptides.

Effect of aging on brain-derived neurotrophic factor, proBDNF, and their receptors in the hippocampus of Lou/C rats.

The interest in understanding healthy aging has prompted scientist to look for animal models presenting this feature. Lou/C rats, an inbred strain of Wistar origin, is an animal model of successful aging with a longer lifespan and preserved memory capacities than most laboratory rat strains. In an attempt to shed light on this remarkable aging feature, we investigated the hippocampal patterns (mRNA and proteins) of some protective and plasticity-related molecules, i.e., brain-derived neurotrophic factor (BDNF), its precursor proBDNF, and its receptors (i.e., TrkB.FL, TrkB.T1, TrkB.T2, and p75). Using different experimental approaches, we compared these characteristics in young and aged Lou/C versus matched Wistar rats (the most appropriate controls). Data showed that young and aged Lou/C rats had higher amounts of BDNF and proBDNF content than Wistar rats. In contrast, proBDNF content was reduced in aged Lou/C rats and increased in aged Wistar rats. With aging, Lou/C rats showed a weaker decrease in TrkB.FL receptors than Wistar rats and no changes in TrkB.T1 receptors, which, contrarily, were increased in aged Wistar rats. Overall, these observations could account for the preserved cognitive performances and memory-dependent mechanisms, such as the unaltered long-term potentiation (LTP), throughout the lifespan recently reported in the Lou/C strain. Data suggest that boosting the expression or activity of these endogenous protective systems may be a promising alternative for combating some age-related cognitive declines. Therefore, Lou/C rats represent an interesting model of healthy aging for studying plasticity-related processes that evolve from youthfulness to aging.

Spatial memory training modifies the expression of brain-derived neurotrophic factor tyrosine kinase receptors in young and aged rats.

Aging leads to alterations in the function of the hippocampus, a brain structure largely involved in learning processes. This study aimed at examining the basal levels and the impact of a learning-associated task on brain-derived neurotrophic factor (BDNF), on BDNF full-length catalytic receptor (TrkB.FL) and on the truncated forms (TrkB.T1 and TrkB.T2) receptor expression (mRNA and protein) in the hippocampus of young (2-month-old) and aged (24-month-old) Wistar rats. Spatial memory was evaluated using a water-maze procedure involving visible and invisible platform location learning. Aged rats showed higher latencies during the first two training days but rapidly exhibited learning performances similar to patterns observed with young rats. Real-time PCR measurements showed that aged rats had significantly higher levels of trkB.FL mRNAs than young rats under basal conditions. In situ hybridization analysis indicated that the highest level of trkB.FL mRNA (mRNA encoding for TrkB.FL receptor) was noted in the dentate gyrus, and in the CA2 and CA3 hippocampal layers. In contrast, there was no marked difference in trkB.T1 signal in any hippocampal region. Training induced a significant reduction in trkB.FL mRNA levels solely in aged rats. In contrast, in young and aged rats, trkB.T2 mRNA levels were significantly increased after training. Measurements of proteins revealed that learning significantly increased TrkB.FL content in aged rats. Untrained aged rats presented higher levels of BDNF and brain-derived neurotrophic factor precursor (proBDNF) proteins than young rats. Training strongly increased precursor BDNF metabolism in young and aged rats, resulting in increased levels of proBDNF in the two groups but in old rats the mature BDNF level did not change. This study shows that Wistar rats present age-related differences in the levels of BDNF and TrkB isoforms and that spatial learning differentially modifies some of these parameters in the hippocampus.

In vivo brain-derived neurotrophic factor release and tyrosine kinase B receptor expression in the supraoptic nucleus after osmotic stress stimulus in rats.

Brain-derived neurotrophic factor (BDNF) is a member of the neurotrophin family involved in plasticity and neuroprotective processes. In recent years, we have reported the presence of BDNF mRNA in the supraoptic nucleus (SON) as well its sensitivity to osmotic stress. The rat SON is a relatively homogenous nucleus mainly consisting of magnocellular soma with their dendritic processes. BDNF may be released from dendrites to the extracellular space to stimulate tyrosine kinase (Trk) B receptors which are hypothetically present on these subcellular SON compartments. The main goal of this work was thus to study the presence and the in vivo BDNF-IR release from SON using the push-pull perfusion technique following systemic (i.p.) or local (within the SON) osmotic stimulation. BDNF was detected by immunocytochemistry and its release was measured by immunological assay (ELISA). Likewise, TrkB receptor localization in the SON-mRNA and their respective proteins-were studied by in situ hybridization and immunohistofluorescence techniques, respectively. Phosphorylation of CREB was detected by immunohistofluorescence. We present here direct evidence of in vivo dendritic BDNF release from SON which is highly sensitive to osmotic stress. The osmotic response latency period clearly depends on the mode of stimulus application (210 min for i.p. route vs. 15 min for intra-SON administration). The fact that BDNF is released as a very rapid peak when osmotic stimulation is locally applied is strong evidence in favor of an intra-SON origin of this secretion. Osmotic stress also increased phosphorylated cAMP response element binding protein immunoreactivity in the SON. In addition, we show in control rats that truncated forms of tyrosine kinase B receptor 2 mRNA represent the most abundant messenger in the SON as compared with brain-derived neurotrophic factor full-length catalytic receptor or truncated forms of tyrosine kinase B receptor 1 mRNA. In conclusion, it is likely that BDNF and their receptors are involved in neuronal plasticity changes induced by osmotic stress in the SON.

The 2'-5' oligoadenylate/RNase L/RNase L inhibitor pathway regulates both MyoD mRNA stability and muscle cell differentiation.

The 2'-5' oligoadenylate (2-5A)/RNase L pathway is one of the enzymatic pathways induced by interferon. RNase L is a latent endoribonuclease which is activated by 2-5A and inhibited by a specific protein known as RLI (RNase L inhibitor). This system has an important role in regulating viral infection. Additionally, variations in RNase L activity have been observed during cell growth and differentiation but the significance of the 2-5A/RNase L/RLI pathway in these latter processes is not known. To determine the roles of RNase L and RLI in muscle differentiation, C2 mouse myoblasts were transfected with sense and antisense RLI cDNA constructs. Importantly, the overexpression of RLI in C2 cells was associated with diminished RNase L activity, an increased level of MyoD mRNA, and accelerated kinetics of muscle differentiation. Inversely, transfection of the RLI antisense construct was associated with increased RNase L activity, a diminished level of MyoD mRNA, and delayed differentiation. In agreement with these data, MyoD mRNA levels were also decreased in C2 cells transfected with an inducible RNase L construct. The effect of RNase L activity on MyoD mRNA levels was relatively specific because expression of several other mRNAs was not altered in C2 transfectants. Therefore, RNase L is directly involved in myoblast differentiation, probably through its role in regulating MyoD stability. This is the first identification of a potential mRNA target for RNase L.

Age-related changes in brain-derived neurotrophic factor and tyrosine kinase receptor isoforms in the hippocampus and hypothalamus in male rats.

A large amount of aging individuals show diminished cognitive and endocrine capabilities. The main brain areas involved in these changes are the hippocampus and hypothalamus, two regions possessing high plasticity and implicated in cognitive and endocrine functions, respectively. Among neurotrophins (considered as genuine molecular mediators of synaptic plasticity), brain-derived neurotrophic factor (BDNF) exhibits in adult rats, the highest concentrations in the hippocampus and hypothalamus. Most of neuronal effects of BDNF are mediated through high-affinity cell surface BDNF tyrosine kinase receptors (TrkB). Different TrkB isoforms are issued by alternative splicing of mRNA encoding for TrkB (trkB mRNA) generating at least three different TrkB receptors with different signaling capabilities. The goal of this study was to examine simultaneously the expression (mRNAs and proteins) of BDNF and its three specific receptors, in the hippocampus and hypothalamus throughout lifespan in rats. We observed that BDNF essentially increased during the first 2 postnatal weeks in the hippocampus and hypothalamus, with no close correlation to its mRNA levels. In these regions, mRNA encoding for BDNF full-length catalytic receptor (trkB.FL mRNA) showed no important changes throughout life but of the mRNA truncated forms of TrkB receptors (trkB.T1 mRNA and trkB.T2 mRNA) trkB.T1 mRNA strongly increased after birth, then remaining stable during aging. trkB.T2 mRNA gradually decreased from 1 postnatal week becoming undetectable in the hippocampus in old-rats. Proteins issued from these mRNAs showed substantial quantitative modifications with aging. From 2 months old, the BDNF full-length catalytic receptor (TrkB.FL) gradually and significantly decreased in the hippocampus and the hypothalamus. Of the truncated forms of TrkB receptors (TrkB.T1 and TrkB.T2) TrkB.T1, which is essentially localized in glial cells, significantly increased from the first postnatal week in the hippocampus and in the hypothalamus, remaining stable during aging but reduced in old rats. TrkB.T2 which similarly to TrkB.FL has a neuronal localization also gradually decreased in the hippocampus and in the hypothalamus throughout lifespan. These reductions were significant at 21 and 30 days old, respectively. All the changes reported here could contribute to the reduced plasticity of these regions observed in old rats.

Characterization of RNABP, an RNA binding protein that associates with RNase L.

The 2',5'-oligoadenylate (2-5A)/RNase L pathway is one of several enzymatic pathways induced by interferons (IFN). RNase L is a latent endoribonuclease that is activated on its binding by 2-5A and inhibited by the ribonuclease L inhibitor (RLI). We have shown previously by coimmunoprecipitation that RNase L may be associated with a 90-kDa RNA binding protein (RNABP), identified with a monoclonal antibody (mAb) raised against an RNase L complex purified under native conditions on 2-5A-sepharose. Here we confirm, by gel-filtration and pull-down analysis, the association of RNase L and RNABP, and we demonstrate that this association is significantly increased in the presence of 2-5A. Moreover, we found that RNABP protein levels decrease during terminal differentiation in various cell lines but do not vary during vesicular stomatitis virus (VSV) or encephalomyocarditis virus (EMCV) infection or following IFN-alpha/beta treatment. In this latter case, although total cellular RNABP levels do not vary, the amount of RNABP found in the cytoplasm increases in comparison to that found in the nucleus, indicating a cytoplasmic localization of RNABP after IFN-alpha/beta treatment. Finally, we demonstrate the interaction between RNase L and RNABP in intact cells. Microinjection of an mAb against RNABP into HeLa cells inhibits RNase L antiviral activity and partially inhibits the IFN-alpha/beta-induced antiviral activity.

The RNase L inhibitor (RLI) is induced by double-stranded RNA.

The (2-5A)-RNase L pathway is an important component of interferon (IFN) action. Its central role in the antiviral effect of IFN against Picornaviridae has been clearly demonstrated. We have characterized and cloned a new component of this pathway, the RNase L inhibitor (RLI). RLI is a cellular protein whose mRNA is not regulated by IFN but is induced by viruses, such as encephalomyocarditis virus (EMCV). RLI inhibits RNase L during the time course of EMCV infection, and overexpression of RLI in HeLa cells partially reverses the antiviral action of IFN against EMCV. The replicative complexes of several viruses consist of double-stranded RNA structures. These dsRNAs could activate gene transcription as demonstrated for IFNs and could be responsible for RLI induction. We describe the increased expression of RLI mRNA and RLI protein induced by synthetic dsRNAs, such as poly(I):poly(C). This induction gives rise to an inhibition of the 2-5A-binding activity of RNase L. The inhibition of RNase L activity is transcient, probably due to the rapid turnover of RLI protein.

Placental BDNF/TrkB signaling system is modulated by fetal growth disturbances in rat and human.

The brain-derived neurotrophic factor (BDNF) has been shown to exert an important role during implantation, placental development, and fetal growth control in mice. Its expression is closely related to the nutritional status in several tissues such as in the nervous system. In a previous study, we demonstrated that maternal undernutrition (MU), during the perinatal life, modified both the BDNF and its functional receptor, the tyrosine kinase receptor B (TrkB) gene expression in the brain of growth-restricted rat offspring during sensitive developmental windows, suggesting that these early modifications may have long-lasting consequences. In the present study, we measured BDNF/TrkB mRNA and protein levels in rat placentas from mothers submitted to a 50% food restriction during gestation, and in human placentas from pregnancies with fetal growth restriction or fetal macrosomia. In the rat, two subtypes of placental TrkB receptors have been identified: the TrkB-FL and TrkB-T1 receptors. We found that MU induced intrauterine growth restriction (IUGR) of fetuses at term and decreased the placental BDNF mRNA and protein levels. Placentae from undernourished mothers exhibited an increased mRNA expression of TrkB-FL whereas both TrkB-FL and TrkB-T1 receptors proteins levels were not modified. In human IUGR placentas, both BDNF and TrkB receptor mRNA expressions were up-regulated. Finally, although neither BDNF nor TrkB mRNA levels were altered by fetal macrosomia alone, BDNF mRNA levels were decreased when macrosomia was associated with maternal type 1 diabetes. These results show that the placental BDNF/TrkB system is modulated in rats and humans during pregnancies with fetal growth perturbations and is affected by the maternal energetic status. These data suggest that this system may exert an important role for the feto-placental unit development and that it may also be implicated in the etiology of pathologies related to placental and fetal growth disturbances.

Perinatal undernutrition modifies cell proliferation and brain-derived neurotrophic factor levels during critical time-windows for hypothalamic and hippocampal development in the male rat.

Maternal perinatal undernutrition (MPU) modifies the activity of the hypothalamic-pituitary-adrenal axis and sensitises to the development of metabolic and cognitive adult diseases. Because the hypothalamus and hippocampus are involved in the regulation of neuroendocrine activity, energy metabolism and cognition, we hypothesised that a maternal 50% food restriction (FR50) from day 14 of pregnancy (E14) until postnatal day 21 (P21) would affect the development of these structures in male rat offspring. Protein and mRNA levels of brain-derived neurotrophic factor (BDNF) and cell proliferation [analysed by 5-bromodeoxyuridine (BrdU) incorporation] were compared in both control and FR50 rats from E21 to P22. Although the pattern of the evolution of BDNF concentration and cell proliferation throughout development was not strikingly different between groups, several disturbances at specific developmental stages were observed. FR50 rats exhibited a delayed increase of hippocampal BDNF content whereas, in the hypothalamus, BDNF level was augmented from E21 to P14 and associated, at this latter stage, with an increased mRNA expression of TRkB-T2. In both groups, a correlation between BDNF content and the number of BrdU positive cells was noted in the dentate gyrus, whereas opposite variations were observed in CA1, CA2 and CA3 layers, and in the arcuate and ventromedial nuclei. In the hippocampus, P15-FR50 rats showed an increased number of BrdU positive cells in all regions, whereas, at P22, a decrease was observed in the CA2. In the hypothalamus, between E21 and P8, MPU increases the number of BrdU positive cells in all regions analysed and, until P15, marked differences were noticed in the median eminence, the paraventricular nucleus and the arcuate nucleus. Taken together, the results obtained in the present study show that MPU changes the time course of production of BDNF and cell proliferation in specific hippocampal and hypothalamic areas during sensitive developmental windows, suggesting that these early perinatal modifications may have long-lasting consequences.

An antiviral state induced in HeLa cells by microinjected poly(rI).poly(rC).

The direct introduction with micropipettes of poly(rI).poly(rC) into the cytoplasm of several human cell lines inhibited the multiplication of vesicular stomatitis virus. This antiviral activity was at least partly due to interferon (IFN) production and secretion from the injected cells since it was species-specific, partly neutralized by IFN antibodies and was transmissible to non-adjacent cells. This suggests a mechanism of IFN induction involving the internalization of poly(rI).poly(rC) and its interaction with an intracellular target.

Microinjected interferon does not promote an antiviral response in Hela cells.

Human fibroblast interferon (Hu IFN beta) was directly introduced with glass micropipets into the cytoplasm of Hela cells. Such an injection of more than 10(4) molecules per cell failed to induce any antiviral state when challenged with vesicular stomatitis virus (VSV). These findings are discussed in relation to the possible role of internalization in the mechanism of antiviral action of interferon.

Regeneration of enzyme activity after western blot: activation of RNase L by 2-5A on filter--importance for its detection.

A rapid and convenient new procedure for detecting RNase L activity following Western blot by renaturation of the enzyme on the nitrocellulose sheets is described. This method allows the simultaneous analysis of enzymatic activity (e.g., cleavage of poly(uridylic acid)-3'-[32P]pCp) and RNase L binding to radioactivE probes (e.g., 2-5A-3'-[32P]pCp) in the same sample. Unlike previously published methods, this procedure eliminates interference from proteases or other RNases during the analysis of RNase L activity. The detection of RNase(s) L is also affected by the presence of endogenous 2-5A, 2-5A derivatives, or other possible "inhibitors" in cell extracts; this Western blot assay allows of RNase(s) L to be detected independently of intracellular 2-5A or analogs. Differences between the procedures used so far and this Western blot technique can indeed be demonstrated. It is shown with this Western blot assay that although RNase L has been described as a protein of 185-200 kDa under nondenaturating conditions, its 80-kDa (and 40-kDa) component is able to bind 2-5A and to cleave poly(uridylic acid) in a 2-5A-dependent way, independently of other subunit(s) or cofactor(s).

Polyclonal antibodies against RNase L. Subcellular localization of this enzyme in mouse cells.

RNase L activated by 2-5A (a series of 2'-5'-linked adenylic oligoribonucleotides) is a key enzyme of the interferon system. To study RNase L (endonuclease L) in intact cells independently of intracellular 2-5A and of its activity, we have developed polyclonal antibodies against RNase L. RNase L from mouse spleen was purified on a column of 2-5A-Sepharose and used to immunize rabbits in co-injection with polyadenylic-polyuridylic acid as adjuvant. Antibodies were purified by chromatography on Affi-Gel blue and 2-5A-Sepharose-immobilized RNase L. These polyclonal antibodies immunoprecipitate the 80- and 40-kDa forms of RNase L in mouse spleen. In Western blot, only the 80-kDa form of RNase L is recognized by these antibodies. These purified antibodies were used to localize RNase L in the cytoplasm of intact mouse NIH 3T3 cells by immunofluorescence. The cytoplasmic localization of RNase L was confirmed by its 2-5A binding activity after cellular fractionation.

Requirement for endocytosis of poly(rI).poly(rC) to generate toxicity on interferon-treated LM cells.

Poly(rI).poly(rC) induces a lytic reaction in interferon-treated mouse LM cells. We have attempted to determine whether the intracellular penetration of poly(rI).poly(rC) is a prerequisite for cell lysis and to gain some insight into the pathway followed. We found that poly(rI).poly(rC) coupled to Sepharose beads was unable to generate lysis of interferon-treated cells whereas the cells underwent lysis after microinjection of poly(rI).poly(rC). Some inhibitors of endocytosis were found to inhibit the development of the lytic reaction. Lysosomotropic amines or a low temperature (19 degrees C) blocked endocytosis of poly(rI).poly(rC) but did not prevent its uptake. The internalization of poly(rI).poly(rC) was energy-dependent and was blocked when sodium azide and 2-deoxyglucose were added simultaneously. We conclude that poly(rI).poly(rC) is internalized and reaches an acidic compartment before triggering the lytic reaction in the cell.

The 2-5A/RNase L/RNase L inhibitor (RLI) [correction of (RNI)] pathway regulates mitochondrial mRNAs stability in interferon alpha-treated H9 cells.

Interferon alpha (IFNalpha) belongs to a cytokine family that exhibits antiviral properties, immuno-modulating effects, and antiproliferative activity on normal and neoplasic cells in vitro and in vivo. IFNalpha exerts antitumor action by inducing direct cytotoxicity against tumor cells. This toxicity is at least partly due to induction of apoptosis. Although the molecular basis of the inhibition of cell growth by IFNalpha is only partially understood, there is a direct correlation between the sensitivity of cells to the antiproliferative action of IFNalpha and the down-regulation of their mitochondrial mRNAs. Here, we studied the role of the 2-5A/RNase L system and its inhibitor RLI in this regulation of the mitochondrial mRNAs by IFNalpha. We found that a fraction of cellular RNase L and RLI is localized in the mitochondria. Thus, we down-regulated RNase L activity in human H9 cells by stably transfecting (i) RNase L antisense cDNA or (ii) RLI sense cDNA constructions. In contrast to control cells, no post-transcriptional down-regulation of mitochondrial mRNAs and no cell growth inhibition were observed after IFNalpha treatment in these transfectants. These results demonstrate that IFNalpha exerts its antiproliferative effect on H9 cells at least in part via the degradation of mitochondrial mRNAs by RNase L.

Cloning and characterization of a RNAse L inhibitor. A new component of the interferon-regulated 2-5A pathway.

The 2-5A/RNase L system is considered as a central pathway of interferon (IFN) action and could possibly play a more general physiological role as for instance in the regulation of RNA stability in mammalian cells. We describe here the expression cloning and initial characterization of RLI (for RNase L inhibitor), a new type of endoribonuclease inhibitor. RLI cDNA codes for a 68-kDa polypeptide whose expression is not regulated by IFN. Its expression in reticulocyte extracts antagonizes the 2-5A binding ability and the nuclease activity of endogenous RNase L or the cloned 2DR polypeptide. The inhibition requires the association of RLI with the nuclease and is dependent on the ratio between the two proteins. Likewise RLI is coimmunoprecipitated with the RNase L complex by a nuclease-specific antibody. RLI does not lead to 2-5A degradation or to irreversible modification of RNase L. The overexpression of RLI in stably transfected HeLa cells inhibits the antiviral activity of IFN on encephalomyocarditis virus but not on vesicular stomatitis virus. RLI therefore appears as the first described and potentially important mediator of the 2-5A/RNase L pathway.