Hydrogen Peroxide-Reducing Factor Released by PC12D Cells Increases Cell Tolerance against Oxidative Stress.

PC12D cells, a subline of rat adrenal pheochromocytoma PC12 cells, extend neurites rapidly in response to differentiation stimuli and are used to investigate the molecular mechanisms of neurite extension. In the present study, we found significant tolerance of PC12D cells against Parkinson's disease-related stimuli such as dopamine and 6-hydroxydopamine; this tolerance was significantly decreased by a change in the medium. Conditioned medium from PC12D cells induced tolerance against oxidative stress, which suggests that cytoprotective factor may be released by PC12D cells into the culture medium. Conditioned medium-induced tolerance was not found for PC12 cells or human neuroblastoma SH-SY5Y cells. A cytoprotective factor generated by PC12D cells exhibited hydrogen peroxide-reducing activity. Chemical characterization showed that this cytoprotective factor is water soluble and has a molecular weight about 1000 Da, and that its activity is inhibited by sodium cyanide. Release of this cytoprotective factor was increased by differentiation stimuli and oxidative stress. Taken together, these results suggest that release of a hydrogen peroxide-reducing factor by PC12D cells increases cell tolerance against oxidative stress. This study provides new insights into the antioxidative properties of factors in extracellular fluid.

Giant rhyolite lava dome formation after 7.3 ka supereruption at Kikai caldera, SW Japan.

Kikai submarine caldera to the south of the Kyushu Island, SW Japan, collapsed at 7.3 ka during the latest supereruption (>500 km3 of magma) in the Japanese Archipelago. Multi functional research surveys of the T/S Fukae Maru in this caldera, including multi-beam echosounder mapping, remotely operated vehicle observation, multi-channel seismic reflection survey, and rock sampling by dredging and diving, provided lines of evidence for creation of a giant rhyolite lava dome (~32 km3) after the caldera collapse. This dome is still active as water column anomalies accompanied by bubbling from its surface are observed. Chemical characteristics of dome-forming rhyolites akin to those of presently active small volcanic cones are different from those of supereruption. The voluminous post-caldera activity is thus not caused simply by squeezing the remnant of syn-caldera magma but may tap a magma system that has evolved both chemically and physically since the 7.3-ka supereruption.

Neurite outgrowth of NG108-15 cells induced by heat shock protein 90 inhibitors.

We previously reported that radicicol (Rad) and geldanamycin (Geld), heat shock protein 90 (Hsp90) inhibitors, potentiate neurite growth of cultured sensory neurons from chick embryo. We now show that the antibiotics induce neurite growth in NG108-15 cells. Treatment of the cells with these drugs caused transient decrease in protein levels of Raf1, ERK1/2, phosphorylated ERK1/2, Akt1, and CDK4. The neurite growth of NG108-15 induced by the inhibitors was blocked by actynomycin D, but the neurite growth stimulated by dbcAMP in the cells was not affected. The neurite growth could be due to a change in the synthesis of some specific protein(s) and is speculated to be due to the transient downregulation of particular-signaling molecules stabilized by Hsp90.

NGF-dependent formation of ruffles in PC12D cells required a different pathway from that for neurite outgrowth.

Two signaling pathways, phosphoinositide 3-kinase (PI-3k)/Akt and Ras/MAPK, are major effectors triggered by nerve growth factor (NGF). Rac1, Cdc42 and GSK-3beta are reported to be targets of PI-3k in the signal transduction for neurite outgrowth. Immediately after NGF was added, broad ruffles were observed temporarily around the periphery of PC12 cells prior to neurite growth. As PC12D cells are characterized by a very rapid extension of neurites in response to various agents, the signaling pathways described above were studied in relation to the NGF-induced formation of ruffles and outgrowth of neurites. Wortmannin, an Akt inhibitor (V), and GSK-3beta inhibitor (SB425286) suppressed the neurite growth in NGF-treated cells, but not in dbcAMP-treated cells. The outgrowth of neurites induced by NGF but not by dbcAMP was inhibited with the expression of mutant Ras. But upon the expression of dominant-negative Rac1, cells often extended protrusions, incomplete neurites, lacking F-actin. Intact neurites were observed in cells with dominant-negative Cdc42. These results suggest that NGF-dependent neurite outgrowth occurs via a mechanism involving activation of the Ras/PI-3K/Akt/GSK-3beta pathway, while dbcAMP-dependent neurite growth might be induced in a distinct manner. However, inhibitors for GSK-3beta and PI-3k (wortmannin) did not suppress the NGF-dependent formation of ruffles. In addition, the formation of ruffles was not inhibited by the expression of mutant Ras. On the other hand, it was suppressed by the expression of dominant-negative Rac1 or Cdc42. These results suggest that the NGF-induced ruffling requires activation of Rac1 and Cdc42, but does not require Ras, PI-3k, Akt and GSK-3beta. Taken together, the NGF-dependent formation of ruffles might not require Ras/PI-3k/Akt/GSK-3beta, but these pathways might contribute to the formation of intact neurites due to combined actions including Rac1.

Fates of Cdh23/CDH23 with mutations affecting the cytoplasmic region.

BUS/Idr mice carrying a mutant waltzer allele (vbus) are characterized by splayed hair bundles in inner ear sensory cells, providing a mouse homolog of USH1D/DFNB12. RT-PCR-based screening for the presence of mutations in mouse Cdh23, the gene responsible for the waltzer phenotype, has identified a G>A mutation in the donor splice site of intron 67 (Cdh23:c.9633+1G>A: GenBank AF308939.1), indicating that two altered Cdh23 molecules having intron-derived COOH-terminal structures could be generated in BUS mouse tissues. Immunochemical analyses with anti-Cdh23 antibodies showed, however, no clear Cdh23-related proteins in vbus/vbus tissues, while the antibodies immunoreacted with approximately 350 kDa proteins in control mice. Immunofluorescent experiments revealed considerable weakening of Cdh23 signals in sensory hair cell stereocilia and Reissner's membrane in the vbus/vbus inner ear, and transmission electron microscopy demonstrated abundant autophagosome/autolysosome vesicles, suggesting aberrant Cdh23:c.9633+1G>A-derived protein-induced acceleration of lysosomal bulk degradation of proteins. In transfection experiments, signal sequence-preceded FLAG-tagged transmembrane plus cytoplasmic regions (TMCy) of tissue-specific Cdh23(+/-68) isoforms were localized to filamentous actin-rich protrusions and the plasma membrane of cultured cells, whereas FLAG-TMCy:c.9633+1G>A proteins were highly insoluble and retained in the cytoplasm. In contrast, FLAG-tagged TMCy:p.Arg3175His and human TMCy:c.9625_9626insC forms were both localized to the plasma membrane in cultured cells, allowing prediction that USH1D-associated CDH23:p.Arg3175His and CDH23:c.9625_9626insC proteins could be transported to the plasma membrane in vivo. The present results thus suggest different fates of CDH23/Cdh23 with mutations affecting the cytoplasmic region.

Genetic regulation mediated by thiamin pyrophosphate-binding motif in Saccharomyces cerevisiae.

The expression of genes of Saccharomyces cerevisiae encoding the enzymes involved in the metabolism of thiamin (THI genes) is co-ordinately repressed by exogenous thiamin and induced in the absence of thiamin. In this yeast THI regulatory system acts mainly at the transcriptional level, thiamin pyrophosphate (TDP) seems to serve as a corepressor, and genetic studies have identified three positive regulatory factors (Thi2p, Thi3p and Pdc2p). We found in a DNA microarray analysis that the expression of THI genes increased 10- to 90-fold in response to thiamin deprivation, and likewise, the expression of THI2 and THI3 increased 17-fold and threefold, respectively. After transfer from repressing to inducing medium, the promoter activity of both THI2 and THI3 increased in parallel with that of PHO3, one of THI genes. The stimulation of THI3 promoter activity was diminished by deletion of THI3, indicative of the autoregulation of THI3. The THI genes were not induced when THI2 was expressed from the yeast GAL1 promoter in a thi3Delta strain or when THI3 was expressed in a thi2Delta strain, suggesting that Thi2p and Thi3p participate simultaneously in the induction. When mutant Thi3p proteins lacking TDP-binding activity were produced in the thi3Delta strain, THI genes were expressed even under thiamin-replete conditions. This result supports the hypothesis that Thi3p senses the intracellular signal of the THI regulatory system to exert transcriptional control. Furthermore, Thi2p and Thi3p were demonstrated to bind each other and this interaction was partially diminished by exogenous thiamin, suggesting that Thi2p and Thi3p stimulate the expression as a complex whose function is disturbed by TDP bound to Thi3p. We discuss the possibility that the induction of THI genes is triggered by the activation of the complex attributed to decrease in intracellular TDP and the elevated complex in the autoregulatory fashion further upregulates THI genes. This is the first report of the involvement of the TDP-binding motif in genetic regulation.

Cyclosporin A treatment upregulates Id1 and Smad3 expression and delays skeletal muscle regeneration.

The molecular signaling pathway linked to muscle regeneration has not yet been identified. Previously, we demonstrated that mice treated with cyclosporin A (CsA), a calcineurin inhibitor, failed to regenerate normally after muscle damage. Using reverse transcription (RT)-PCR, Western blot and immunohistochemical analysis, we investigated whether the amounts of nuclear factor of activated T cells (NFAT), myocyte-enhancer factor 2 (MEF2), the MyoD family, Id-1, and Smad3 change in the regenerating muscle after CsA treatment. Adult male ICR mice were subjected to a bupivacaine injection into the tibialis anterior muscle, and were treated with either CsA (25 mg/kg) or vehicle once daily. They were killed at 1, 2, 4, 6, 9 and 14 days post injury. RT-PCR analysis did not show a significant difference in MEF2s, MyoD and myogenin mRNA levels in the regenerating muscle in either placebo- and CsA-administered mice. In contrast, a significant increase in MRF4 mRNA was seen in CsA-administered mice compared to the placebo-treated mice at 4 and 9 days post surgery. In CsA-treated mice, the level of Id1 mRNA was elevated at day 9 relative to the placebo-treated mice. After 6 days, the CsA-treated mice possessed more abundant proliferating cell nuclear antigen (PCNA) and cyclin D1 protein in many satellite cells and/or myoblast-like cells in the regenerating muscle. The amount of myostatin, TGF-beta2 and Smad3 mRNA and proteins was increased more markedly in the mice treated with CsA. After 9 days, many satellite cells and/or myoblasts showed apparent co-localization of both MyoD and Smad3 in CsA-, but not in placebo-, treated mice. Our results demonstrated that CsA treatment upregulates Id1 and Smad3 expression and delays skeletal muscle regeneration in vivo.

Glycosylation site for chondroitin sulfate on the neural part-time proteoglycan, neuroglycan C.

Neuroglycan C (NGC) is a membrane-spanning chondroitin sulfate (CS) proteoglycan that is expressed predominantly in the central nervous system (CNS). NGC dramatically changed its structure from a proteoglycan to a nonproteoglycan form with cerebellar development, whereas a small portion of NGC molecules existed in a nonproteoglycan form in the other areas of the mature CNS, suggesting that the CS glycosylation of NGC is developmentally regulated in the whole CNS. As primary cultured neurons and astrocytes from cerebral cortices expressed NGC in a proteoglycan form and in a nonproteoglycan form, respectively, CS glycosylation seems to be regulated differently depending on cell type. To investigate the glycosylation process, cell lines expressing a proteoglycan form of NGC would be favorable experimental models. When a mouse NGC cDNA was transfected into COS 1, PC12D, and Neuro 2a cells, only Neuro 2a cells, a mouse neuroblastoma cell line, expressed NGC bearing CS chains. In PC12D cells, although three intrinsic CS proteoglycans were detected, exogenously expressed NGC did not bear any short CS chains just like NGC in the mature cerebellum. This suggests that the addition of CS chains to the NGC core protein is regulated in a manner different from that of other CS proteoglycans. As the first step in investigating the CS glycosylation mechanism using Neuro 2a cells, we determined the CS attachment site as Ser-123 on the NGC core protein by site-directed mutagenesis. The CS glycosylation was not necessary for intracellular trafficking of NGC to the cell surface at least in Neuro 2a cells.

Possible involvement of myosin-X in intercellular adhesion: importance of serial pleckstrin homology regions for intracellular localization.

Subcellular fractionation experiments with mouse hepatocytes, combined with sodium dodecylsulfate (SDS)-polyacrylamide gel electrophoresis (PAGE)-immunoblot analysis using antibodies against two different tail regions of mouse myosin-X demonstrated a 240 kDa molecular mass to be associated with the plasma membrane-rich P2 fraction. The basolateral plasma membrane fraction, but not the brush border fraction, isolated from renal cortices also contained the 240 kDa form of myosin-X. In an attempt to assess relative contributions of possible functional domains in the tail of myosin-X to localization and function, cDNA corresponding to all three pleckstrin homology (PH) domains and different regions (PH1, 2 and 3, and the two subdomains of PH1: PHS1 and PHS2), as well as the myosin tail homology 4 domain (MyTH4) and the band4.1/ezrin/radixin/moesin-like domain (FERM) were separately inserted into the pEGFP vector and expressed in cultured COS-1 cells. As a result, two distinct regions responsible for localization were identified with regard to PH: one covers all three forms that tends to localize to regions of dynamic actin, such as membrane ruffles, lamellipodia and thick cortical actin bundles at the sites of cell-cell adhesion in a Rac- and Cdc42-dependent manner. The other covers PHS1 and PH2 that localizes to filopodia, filopodial puncta and the sites of intercellular adhesion in a Cdc42-dependent manner. Expression of green fluorescent protein (GFP)-MyTH4 fusion protein resulted in formation of phalloidin-positive granules, while GFP-FERM affected the actin cytoskeletal system in a distinctly different way. Taken altogether, the results lend support to the view that myosin-X is involved in cell-cell adhesion-associated signaling-linked membrane and/or cytoskeleton reorganization.

Serum response factor plays an important role in the mechanically overloaded plantaris muscle of rats.

Molecular signaling pathways linking the hypertrophy after mechanical overloading in vivo have not been identified. Using western blot analysis, immunoprecipitation, and immunohistochemistry, we investigated the effect of the mechanical overloading state on RhoA, serum response factor (SRF), and MyoD in the rat plantaris muscle. Adult male rats (10 weeks of age) were used in this experiment. Compensatory enlargement of the plantaris muscle was induced in one leg of each rat by surgical removal of the ipsilateral soleus and gastrocnemius muscles. In the normal plantaris muscle of rats, slight expression of RhoA and SRF was observed in the quiescent satellite cells possessing CD34 and c-Met. Western blotting using the homogenate of whole muscle clearly showed that mechanical overloading of the plantaris muscle significantly increased the amount of RhoA during 3-6 days postsurgery. Threonine phosphorylation of SRF occurred at 2-4 h after mechanical overloading. The most marked increase in SRF protein was observed in the hypertrophied muscle at 6 days postsurgery. At 2 days postoperation, SRF immunoreactivity was not detected in the proliferating satellite cells possessing bromodeoxyuridine and in the infiltrating macrophages expressing ED1 in the overloaded muscle by surgical removal. The SRF protein was colocalized with RhoA, FAK, and myogenin but not Myf-5 in many mononuclear cells at 6 days of functional overload. At this time, MyoD immunoreactivity was detected in the cytoplasm of mononuclear cells (possibly satellite cell-derived myoblasts) possessing SRF protein at the nucleus. These results suggest that the signaling pathway through RhoA-FAK-SRF is important to the differentiation of satellite cells by interacting MyoD and myogenin in the hypertrophied muscle of rats.

Calcineurin is a potent regulator for skeletal muscle regeneration by association with NFATc1 and GATA-2.

The molecular signaling pathways involved in regeneration after muscle damage have not been identified. In the present study, we tested the hypothesis that calcineurin, a calcium-regulated phosphatase recently implicated in the signaling of fiber-type conversion and muscle hypertrophy, is required to induce skeletal muscle remodeling. The amount of calcineurin and dephosphorylated nuclear factor of activated T cells c1 (NFATc1) proteins was markedly increased in the regenerating muscle of rats. The amount of calcineurin co-precipitating with NFATc1 and GATA-2, and NFATc1 co-precipitating with GATA-2 gradually increased in the tibialis anterior muscle after bupivacaine injection. Calcineurin protein was present in the proliferating satellite cells labeled with BrdU in the damaged muscle after 4 days. In contrast, calcineurin was not detected in the quiescent nonactivating satellite cells expressing Myf-5. At 4 days post injection, many macrophages detected in the damaged and regenerating area did not possess calcineurin protein. Calcineurin protein was abundant in many myoblasts and myotubes that expressed MyoD and myogenin at 4 and 6 days post injection. In the intact muscle, no immunoreactivity of calcineurin or BrdU was detected in the cell membrane, cytosol or the extracellular connective tissue. In mice, intraperitoneal injection of cyclosporin A, a potent inhibitor of calcineurin, induced extensive inflammation, marked fiber atrophy, the appearance of immature myotubes, and calcification in the regenerating muscle compared with phosphate-buffered saline-administered mice. Thus, calcineurin may have an important role in muscle regeneration in association with NFATc1 and GATA-2.

The reciprocal change of neurotrophin-4 and glial cell line-derived neurotrophic factor protein in the muscles, spinal cord and cerebellum of the dy mouse.

Laminin alpha2 (merosin)-deficient congenital muscular dystrophy (CMD) patients show progressive muscle fiber necrosis and ineffective muscle regeneration, probably due to a lower formation of multinucleated myotubes due to an adhesion defect of myoblasts to each other. Some recent studies found that CMD patients have a white matter disorder and cerebellum atrophy. In the spinal cord of dy mice, a model of CMD, inducible nitric oxide synthase (iNOS) was markedly expressed. Using Western blotting and immunohistochemical analyses, we investigated the levels of neurotrophin-4 (NT-4), brain-derived neurotrophic factor, glial cell line-derived neurotrophic factor (GDNF) and ciliary neurotrophic factor (CNTF) in the central nervous system and skeletal muscles of dy mice. In the dy mice, the microtubule-associated protein-2 (MAP-2) protein level was markedly decreased in the Purkinje and granule cells of the cerebellum, and in lumbar motoneurons of the spinal cord. The motoneurons and axons of dy mice possessed lower expressions of phosphorylated tau. The amount of NT-4 was markedly lower in the cerebellum, spinal cord and hindlimb muscles of dy mice. In dy mice, GDNF was markedly enhanced in the Purkinje and granule cells of the cerebellum, in many lumbar motoneurons, and in the regenerating atrophied fibers. The CNTF protein level did not differ in the hindlimb muscles between the normal and dy mice. Therefore, GDNF could act to inhibit the death of Purkinje and granular neurons, and motoneurons, and to promote the remodeling of the neuromuscular junction of atrophied muscle fibers of dy mice. Furthermore, dy mice include neurogenic abnormalities in the cerebellum and spinal cord along with myogenic disorder of muscle fibers.