Faculty Bibliography

Brain-derived neurotrophic factor (BDNF) activates a variety of signaling molecules to exert various functions in the nervous system, including neuronal differentiation, survival, and regulation of synaptic plasticity. Previously, we have suggested that BIT/SHPS-1 (brain immunoglobulin-like molecule with tyrosine-based activation motifs/SHP substrate 1) is a substrate of Shp-2 and is involved in BDNF signaling in cultured cerebral cortical neurons. To elucidate the biological function of BIT/SHPS-1 in cultured cerebral cortical neurons in connection with its role in BDNF signaling, we generated recombinant adenovirus vectors expressing the wild type of rat BIT/SHPS-1 and its 4F mutant in which all tyrosine residues in the cytoplasmic domain of BIT/SHPS-1 were replaced with phenylalanine. Overexpression of wild-type BIT/SHPS-1, but not the 4F mutant, in cultured cerebral cortical neurons induced tyrosine phosphorylation of BIT/SHPS-1 itself and an association of Shp-2 with BIT/SHPS-1 even without addition of BDNF. We found that BDNF-promoted survival of cultured cerebral cortical neurons was enhanced by expression of the wild type and also 4F mutant, indicating that this enhancement by BIT/SHPS-1 does not depend on its tyrosine phosphorylation. BDNF-induced activation of mitogen-activated protein kinase was not altered by the expression of these proteins. In contrast, BDNF-induced activation of Akt was enhanced in neurons expressing wild-type or 4F mutant BIT/SHPS-1. In addition, LY294002, a specific inhibitor of phosphatidylinositol 3-kinase, blocked the enhancement of BDNF-promoted neuronal survival in both neurons expressing wild-type and 4F mutant BIT/SHPS-1. These results indicate that BIT/SHPS-1 contributes to BDNF-promoted survival of cultured cerebral cortical neurons, and that its effect depends on the phosphatidylinositol 3-kinase-Akt pathway. Our results suggest that a novel action of BIT/SHPS-1 does not occur through tyrosine phosphorylation of BIT/SHPS-1 in cultured cerebral cortical neurons.

Apoptosis occurs not only in mitotic cells but also in postmitotic neuronal cells. We previously suggested that the tumor suppressor gene p53 is required for DNA strand break-induced apoptosis in dissociated culture of cerebellar granule neurons. In this study, we examined the role of p53 in apoptosis using organotypic slice culture of cerebellum from p53 null and wild-type mice. Exposure to bleomycin significantly increased the numbers of TUNEL-, p53-, and c-Jun-positive neurons in the wild-type mouse cerebellar internal granular layer (IGL) and Purkinje cell layer (PL). However, in p53-deficient mice, these responses were not observed. These results are consistent with our previous observations in dissociated neuronal culture showing that the amount of c-Jun protein increases significantly after addition of bleomycin in p53 wild-type cerebellar granule cells. The results presented here also indicate that p53 is involved in DNA strand break-induced apoptosis of fully postmitotic central nervous system neurons and suggest that c-Jun expression occurs downstream of p53 expression.

Brain-derived neurotrophic factor (BDNF), a member of the neurotrophins, promotes differentiation and survival and regulates plasticity of various types of neurons. BDNF binds to TrkB, a receptor tyrosine kinase, which results in the activation of a variety of signaling molecules to exert the various functions of BDNF. Shp-2, a Src homology 2 domain-containing cytoplasmic tyrosine phosphatase, is involved in neurotrophin signaling in PC12 cells and cultured cerebral cortical neurons. To examine the roles of Shp-2 in BDNF signaling in cultured rat cerebral cortical neurons, the wild-type and phosphatase-inactive mutant (C/S mutant) forms of Shp-2 were ectopically expressed in cultured neurons using recombinant adenovirus vectors. We found that several proteins tyrosine-phosphorylated in response to BDNF showed enhanced levels of tyrosine phosphorylation in cultured neurons infected with C/S mutant adenovirus in comparison with those infected with the wild-type Shp-2 adenovirus. In addition, in immunoprecipitates with anti-Shp-2 antibody, we also observed at least four proteins that displayed enhanced phosphorylation in response to BDNF in cultured neurons infected with the C/S mutant adenovirus. We found that the Shp-2-binding protein, brain immunoglobulin-like molecule with tyrosine-based activation motifs (BIT), was strongly tyrosine-phosphorylated in response to BDNF in cultured neurons expressing the C/S mutant of Shp-2. In contrast, the level of BDNF-induced phosphorylation of mitogen-activated protein kinase and coprecipitated proteins with anti-Trk and Grb2 antibodies did not show any difference between neurons infected with these two types of Shp-2. Furthermore, the survival effect of BDNF was enhanced by the wild type of Shp-2, although it was not influenced by the C/S mutant of Shp-2. These results indicated that in cultured cerebral cortical neurons Shp-2 is specifically involved in the regulation of several tyrosine-phosphorylated proteins, including BIT, in the BDNF signaling pathway. In addition, the phosphatase Shp-2 may not influence the level of BDNF-induced activation of mitogen-activated protein kinase in cultured cortical neurons. Further, Shp-2 may have potential to positively regulate BDNF-promoting neuronal survival.

Shp2, a protein tyrosine phosphatase possessing SH2 domains, is utilized in the intracellular signaling of various growth factors. Shp2 is highly expressed in the CNS. Brain-derived neurotrophic factor (BDNF), a member of the neurotrophin family, which also shows high levels of expression in the CNS, exerts neurotrophic and neuromodulatory effects in CNS neurons. We examined how BDNF utilizes Shp2 in its signaling pathway in cultured cerebral cortical neurons. We found that BDNF stimulated coprecipitation of several tyrosine-phosphorylated proteins with anti-Shp2 antibody and that Grb2 and phosphatidylinositol 3-kinase (PI3-K) were coprecipitated with anti-Shp2 antibody in response to BDNF. In addition, both anti-Grb2 and anti-PI3-K antibodies coprecipitated Shp2 in response to BDNF. The BDNF-stimulated coprecipitation of the tyrosine-phosphorylated proteins, Grb2, and PI3-K with anti-Shp2 antibody was completely inhibited by K252a, an inhibitor of TrkB receptor tyrosine kinase. This BDNF-stimulated Shp2 signaling was markedly sustained as well as BDNF-induced phosphorylation of TrkB and mitogen-activated protein kinases. In PC12 cells stably expressing TrkB, both BDNF and nerve growth factor stimulated Shp2 signaling similarly to that by BDNF in cultured cortical neurons. These results indicated that Shp2 shows cross-talk with various signaling molecules including Grb2 and PI3-K in BDNF-induced signaling and that Shp2 may be involved in the regulation of various actions of BDNF in CNS neurons.

Tumor suppressor gene p53 is a critical regulator of the cellular response to DNA damage. To examine the function of p53 in postmitotic CNS neurons, we cultured cerebellar granule cells from 15-day-old wild type and p53-deficient mice, and analyzed changes of protein expression in apoptosis elicited by DNA damage. When cerebellar granule cells from wild type mice were treated with bleomycin, a DNA strand-break inducing agent, neuronal death occurred. In contrast, cells from p53-deficient mice were resistant to bleomycin-induced neuronal death. Furthermore, cells from p53 heterozygous mice showed an intermediate resistance between wild type and p53-deficient mice. These results show that p53 is required for the bleomycin-induced cerebellar granule cell death. To examine which proteins are involved in this apoptosis, we examined changes in protein levels of the Bcl-2 family, including Bcl-2, Bcl-X and Bax. The relative amounts of these proteins did not change after bleomycin treatment, suggesting that the changes in the levels of these Bcl-2 family proteins are not necessary for apoptosis in this system. In contrast, the levels of c-Jun protein significantly increased 6 h after treatment with bleomycin in wild type but not in p53-deficient cerebellar granule cells. These results raise the possibility that c-Jun is required for p53-dependent neuronal apoptosis induced by bleomycin.

To study the involvement of the xeroderma pigmentosum group A gene (XPA) in neuronal apoptosis, we cultured cerebellar neurons from mice lacking XPA gene (XPA-/-) and induced apoptosis by exposure to UV irradiation or medium containing a low concentration of potassium (low-K+ medium). When cerebellar neurons from postnatal days 15-16 wild-type mice were treated with UV irradiation, apoptotic neuronal death was observed after 24-48 h. About 60% of neurons survived 48 h after UV irradiation at a dose of 5 J/m2. On the other hand, neurons from XPA-/- mice showed a significantly increased vulnerability to UV irradiation, and >90% of neurons died 48 h after UV irradiation at a dose of 5 J/m2. In contrast, low-K+ medium induced apoptosis of neurons from mice of each genotype with the same kinetics. These results suggest that the XPA gene is involved in neuronal DNA repair and that it thereby influences apoptosis induced by DNA damage in cultured cerebellar neurons.

The tumour suppressor p53 gene serves as a critical regulator of the cell cycle and of apoptosis following the exposure of normal cells to DNA damage. To examine the role of p53 in postmitotic CNS neurons, we cultured cerebellar neurons from normal wild-type mice and mutant p53-null mice under various conditions inducing neuronal death. When cerebellar neurons from 15- to 16-day postnatal wild-type mice were treated with ionizing radiation or DNA-damaging agents, massive neuron death occurred after 24-72 h. In contrast, neurons from p53-/- mice evidently resisted gamma-irradiation and some DNA-damaging agents, such as etoposide and bleomycin. On the other hand, low-K+ medium-induced apoptosis of cerebellar neurons was not affected by p53 status. Neither cell cycle progression nor DNA synthesis occurred during cell death induced by gamma-irradiation and low-K+ medium, as well as in normal cultures of p53+/+ and p53-/- neurons. These results suggest that p53 is required for the apoptotic death of postmitotic cerebellar neurons induced by DNA strand breaks.

The tumor suppressor p53 gene plays a key role in controlling the cell cycle checkpoint and in apoptosis following the exposure of normal cells to DNA damage. To investigate the role of p53 in cytosine arabinoside (Ara C)-induced cell death of CNS neurons, we examined the effect of Ara C on the survival of cultured cerebellar granule neurons from normal wild-type and p53 null mutant mice. When the neurons from wild-type mice were cultured with Ara C, they gradually died after 24 h in culture. In contrast, the neurons from p53 null mutant mice showed a resistance to the Ara C neurotoxicity. These results indicate that p53 involves Ara C-induced apoptosis in cultured cerebellar granule neurons, in which DNA damage may initiate the apoptotic death program of the neurons.