Modeling Drug-Induced Neuropathy Using Human iPSCs for Predictive Toxicology.

Drugs under development can cause unpredicted toxicity in humans due to differential drug responsiveness between humans and other disease models, resulting in clinical trial failures. Human induced pluripotent stem cells (iPSCs) are expected to represent a useful tool for toxicity testing. However, among many assays, appropriate cellular assays for predicting neurotoxicity in an iPSC-based model are still uncertain. Here we generated neurons from iPSCs of Charcot-Marie-Tooth disease (CMT) patients. Some CMT patients are sensitive to anticancer drugs and present with an adverse reaction of neuropathy. We analyzed cellular phenotypes and found that mitochondria in neurites of CMT neurons were morphologically shorter and showed slower mobility compared to control. A neurosphere assay showed that treatment with drugs known to cause neuropathy caused mitochondrial aggregations in neurites with adenosine triphosphate shortage in both CMT and control neurons, although more severely in CMT. These findings suggest that the genetically susceptible model could provide a useful tool to predict drug-induced neurotoxicity.

Axotomy induces axonogenesis in hippocampal neurons through STAT3.

After axotomy of embryonic hippocampal neurons in vitro, some of the axotomized axons lose their identity, and new axons arise and grow. This axotomy-induced axonogenesis requires importin, suggesting that some injury-induced signals are transported via axons to elicit axonogenesis after axotomy. In this study, we show that STAT3 is activated in response to axotomy. Because STAT3 was co-immunoprecipitated with importin ß in the axotomized neurons, we suggest that STAT3 is retrogradely transported as molecular cargo of importin α/ß heterodimers. Indeed, inhibition of importin α binding with STAT3 resulted in the attenuation of axonogenesis. Silencing STAT3 blocked the axonogenesis, demonstrating that STAT3 is necessary for axotomy-induced axonogenesis. Furthermore, the overexpression of STAT3 enhanced axotomy-induced axonogenesis. Taken together, these results demonstrate that activation and retrograde transport of STAT3 in injured axons have key roles in the axotomy-induced axonogenesis of hippocampal neurons.

Localization of Caspr2 in myelinated nerves depends on axon-glia interactions and the generation of barriers along the axon.

Cell recognition proteins of the contactin-associated protein (Caspr) family demarcate distinct domains along myelinated axons. Caspr is present at the paranodal junction formed between the axon and myelinating glial cells, whereas Caspr2 is localized and associates with K(+) channels at the adjacent juxtaparanodal region. Here we investigated the distribution of Caspr2 during development of peripheral nerves of normal and galactolipids-deficient [ceramide galactosyl transferase (CGT)-/-] mice. This mutant exhibits paranodal abnormalities, lacking all putative adhesion components of this junction, including Caspr, contactin, and neurofascin 155. In sciatic nerves of this mutant, Caspr2 was not found at the juxtaparanodal region but was concentrated instead at the paranodes with Kv1.2. Similar distribution of Caspr2 was found in the PNS of contactin knock-out mice, which also lack Caspr in their paranodes. During development of wild-type peripheral nerves, Caspr2 and Kv1.2 were initially detected at the paranodes before relocating to the adjacent juxtaparanodal region. This transition was not observed in CGT mice, where Caspr2 and Kv1.2 remained paranodal. Double labeling for Caspr and Caspr2 demonstrated that these two related proteins occupied mutually excluding domains along the axon and revealed the presence of both paranodal and internodal barrier-like structures that are delineated by Caspr. Finally, we found that the disruption of axon-glia contact in CGT-/- nerves also affects the localization of the cytoskeleton-associated protein 4.1B along the axon. Altogether, our results reveal a sequential appearance of members of the Caspr family at different domains along myelinated axons and suggest that the localization of Caspr2 may be controlled by the generation of Caspr-containing barriers along the axon.

Type II brain 4.1 (4.1B/KIAA0987), a member of the protein 4.1 family, is localized to neuronal paranodes.

Histochemical analyses of type II brain 4.1/4.1B/KIAA0987, a member of the protein 4.1 family, were carried out in rat brain. In situ hybridization (ISH) showed that type II brain 4.1 mRNA is expressed in a variety of neuronal cells. In particular, type II brain 4.1 mRNA was actively transcribed in the cells of the mesencephalon and the brainstem, which have large myelinated nerve fibers. Expression of type II brain 4.1 mRNA was not observed at least in glial cells distributed in nerve fiber tracts. In immunohistochemical studies using anti-type II brain 4.1-specific antibody, the major immunosignals appeared as brilliant pairs of dots along nerve fibers. Such immunosignals were detected throughout the brain, but were highly concentrated in nerve fiber tracts. These data suggested that type II brain 4.1 is predominantly localized to neuronal paranodes. Detailed analysis concentrating on the nodal region indicated that type II brain 4.1 is present at the paranodal membrane but not in the axoplasm. Weaker type II brain 4.1-specific immunosignals were observed along the internodal membrane of myelinated axons and in the cytoplasm of some neuronal cells. Finally, comparative immunohistochemical studies using antibodies against the other three protein 4.1 family members, type I brain 4.1/4.1N/KIAA0338, erythroid type 4.1 (4.1R) and 4.1G, demonstrated that each of these proteins is distributed in a unique pattern in the cerebellum. Our results are the first to show that type II brain 4.1 is the only member of the protein 4.1 family localized to neuronal paranodes.

Cellular and subcellular localization of a newly identified member of the protein 4.1 family, brain 4.1, in the cerebellum of adult and postnatally developing rats.

For obtaining a deeper insight into the properties of a newly characterized member of the protein 4.1 family, brain 4.1, the cellular and subcellular localization was investigated in the cerebellar cortex of adult and postnatally developing rats. Fluorescent immunohistochemical observations showed that brain 4.1 localized predominantly to glomeruli in the granular layer and throughout the molecular layer in adult rat cerebellar cortex. Analysis of subcellular localization of brain 4.1 by immuno-electron microscopy further demonstrated that presynaptic terminals of mossy fibers and parallel fibers, cytoplasm of granule cells and cytoplasm and/or processes of glial cells contained brain 4.1 while postsynaptic regions of the dendrites of granule cells and Purkinje cells, axons and myelin sheaths did not. Thus, one of the major subcellular destination of brain 4.1 was presynaptic terminal in the cerebellum. This was further supported by the fact that the immunostaining pattern of brain 4.1 in the cerebellum changed in a similar way to that of a synaptic terminal marker, synaptophysin during the postnatal development. Immunoblot analysis also demonstrated that contents of brain 4.1 isoforms varied in parallel with the changes of the immunostaining pattern. Biochemical analysis confirmed the presence of brain 4.1 at synaptic terminals, but there was no obvious correlation between each isoform and its subcellular localization. These results suggested that brain 4.1 is involved in the formation and maintenance of synapse as a membrane skeletal component at presynaptic terminals in the cerebellum.

Molecular characterization of a new member of the protein 4.1 family (brain 4.1) in rat brain.

In addition to the well-known erythroid 4.1 gene, two human genes (KIAA0338 and 4.1G) have recently been identified as members of the protein 4.1 family of genes. We compared the expression levels of these three genes and found that the KIAA0338 gene was predominantly expressed in human brain. To further characterize this novel protein 4.1, called brain 4.1, we isolated rat brain 4.1 cDNA and analyzed its gene products in rat brain. The results indicated that the mRNA and protein products of the brain 4.1 gene were more abundant in brain compared to any other tissues examined. The brain 4.1 mRNA appeared as multiple bands with estimated sizes of 3.9 kb, 6.2 kb and 8.7 kb on RNA blotting analysis, and was found to consist of various alternative forms as reported previously for the erythroid 4. 1 gene. As for the brain 4.1 gene product, many isoforms discernible by immunoblotting analysis were also observed depending on the tissue type and the brain region. The existence of multiple forms of the brain 4.1 implies that it has multiple and diverse functions like the erythroid 4.1 gene product.

Characterization of a new beta-spectrin gene which is predominantly expressed in brain.

We recently identified a gene which shows high similarity to the beta-spectrin gene but with a different chromosomal location from either of the two known beta-spectrin genes [T. Nagase, K.-I. Ishikawa, D. Nakajima, M. Ohira, N. Seki, N. Miyajima, A. Tanaka, H. Kotani, N. Nomura, O. Ohara, Prediction of the coding sequences of unidentified human genes: VII. The complete sequences of 100 new cDNA clones from brain which can code for large proteins in vitro, DNA Res. 4 (1997) 141-150]. In order to further characterize this new spectrin gene and its product, we isolated the rat counterpart of this gene and analyzed it in terms of its protein coding sequence, the tissue distribution of its mRNA and the product, and the regional distribution of the mRNA and the product in the brain. The results indicated that this gene was most abundantly transcribed in the brain and neurons were the predominant cell-type to express this gene. In particular, Purkinje cells were the richest in this gene product, and this new form of beta-spectrin was found more prominently in the dendrites than in the cell bodies. Since the expression pattern and the subcellular localization of this gene product were quiet distinct from those of the two beta-spectrin isoforms already characterized, this beta-spectrin gene would play an important role in neuronal membrane skeleton although it has been overlooked to date.

[Serial changes of atrial natriuretic peptide and brain natriuretic peptide during cesarean section under spinal anesthesia].

Both atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) are considered to play a role in regulation of body fluid volume and blood pressure. Serial changes of their plasma concentrations are known to reflect blood volume. In this study, we investigated the serial changes in maternal blood concentrations of ANP and BNP in those who underwent cesarean section under spinal anesthesia. Blood samples were obtained before anesthesia, before surgery, at the end of surgery, and on the next morning. Samples were also compared with those from normal pregnant women. Throughout the cesarean section, ANP did not show any significant changes, while BNP remained within normal ranges during anesthesia and increased on the next morning. We concluded that ANP and BNP were regulated differentially, and that BNP reflected blood volume significantly during and after spinal anesthesia.

A new solid-phase chemical DNA sequencing method which uses streptavidin-coated magnetic beads.

To simplify the chemical DNA sequencing protocol, we developed a new solid-phase method which uses streptavidin-coated magnetic beads. This method is based on the finding that the biotinylated DNA-streptavidin complex was stable under the conditions for some chemical sequencing reactions. The 5'-biotinylated DNA generated by the polymerase chain reaction was first captured by streptavidin-coated magnetic beads and then subjected to a set of simplified chemical sequencing reactions on the beads at room temperature. Followed by the piperidine cleavage reaction, the products were resolved by gel electrophoresis, transferred onto a nylon membrane and visualized by chemiluminescent detection. As a consequence, high-quality sequencing ladders were obtained, due to complete removal of contaminating chemicals, without the time-consuming precipitation/centrifugation steps used in the conventional chemical sequencing protocol.

Identification of the cells expressing cot proto-oncogene mRNA.

The cell types expressing cot proto-oncogene mRNA were identified by in situ hybridization (ISH) histochemistry. Among a variety of adult mouse tissues examined, four types of glandular cells expressing cot gene were identified: (1) granular duct cells in the submandibular and sublingual glands; (2) serous cells in the parotid gland; (3) peptic (chief) cells in gastric glands; and (4) goblet cells in colonic glands. Investigation of the developmentally regulated expression of cot mRNA using tissues of 14-day and 18-day embryos, newborn and weanling mice showed that cot gene is expressed only in morphologically differentiated and functionally activated cells of these four types. No other types of cells showing ISH signals were observed. Based on these results, cot gene expressions in cultured cells of colonic adenocarcinomas and gastric adenocarcinomas were examined. SW 480 and WiDr cells showed high expression of this gene and so should be useful for functional analysis of Cot kinase. The expression patterns of cot gene in tumor tissues of the parotid gland, and gastric and colonic glands were investigated. Two of the tissues overexpressed this gene markedly, suggesting that overproduction of Cot kinase may be one cause of their transformation.

The murine cot proto-oncogene: genome structure and tissue-specific expression.

We cloned and analyzed the murine cot proto-oncogene and examined its tissue-specific expression in fetal, newborn and adult mice. Genomic cot DNA consists of eight exons, spanning more than 25 kb, and all intron-exon borders are well conserved as compared to the human homolog. Analysis of the full-length cot cDNA revealed that it contained an open reading frame of 1,401 nucleotides, like human cot proto-oncogene. The sequence identity between murine and human cot gene is 84.4% at the nucleotide level and 93.9% at the deduced amino acid level. On northern blot analysis of poly (A)+ RNA, the cot message was detected at 2.9 kb in size. Expression of the cot gene was observed in many tissues from fetal to adult mice, though the level of expression was low in all tissues examined.

Ontogeny of macrophage function to release superoxide anion in conventional and germfree mice.

To determine whether the presence of bacterial flora contributes to the ontogenic development of macrophage function, the ability of macrophages to release superoxide anion (O2-) in response to stimulation with phorbol myristate acetate was compared in conventional and germfree mice of various ages after birth. One-week-old conventional mice showed a very low level of O2- release by their macrophages, and gradual increases were observed in 2-, 3-, and 4-week-old mice in an age-dependent manner. Macrophages from germfree mice always showed a significantly lower level of O2- release compared with conventional mice of the same age; however, age-dependent functional development was seen also in the germfree group. The poor level of O2- release by macrophages from adult germfree mice could be restored to more than the level by conventional mice when the mice were conventionalized for 3 weeks. These results suggested that the ontogenic development of macrophage function is not controlled by the presence of bacterial flora but that the full-scale expression of function at each age is under the influence of microflora.

Characterization of mRNA responsible for induction of functional sodium channels in Xenopus oocytes.

When Xenopus laevis oocytes were microinjected with poly(A)+ mRNA isolated from adult rat brains or electric organs of Electrophorus electricus, the oocytes developed functional sodium channels. Upon application of veratrine, the microinjected oocytes exhibited transient depolarization, resulting in spontaneous repetitive spikes in some occasions, and action potentials. These responses were mediated mainly by external Na ions, prolonged by scorpion toxin, completely blocked by tetrodotoxin, and suppressed by local anesthetics. Thus the mRNA-induced sodium channels exhibited essentially all the functional properties expected for native sodium channels in nerve and muscle membranes. Rat brain mRNA was fractionated into 4 fractions by sucrose gradient centrifugation. Each fraction and various combinations of them were examined for the efficiency in inducing functional sodium channels in Xenopus oocytes. A fraction corresponding to mRNA of approximately 30S to 46S was found to contain all mRNA necessary for the expression of the channels, indicating that mRNA of smaller sizes expected to code for smaller polypeptides may not be required.

Cell-mediated lysis of heat-inactivated influenza virus-coated murine targets.

The involvement of inoculated virus antigens in the induction of target susceptibility to cytotoxic T lymphocyte (CTL)-mediated lysis was investigated using heat-inactivated influenza virus, PR8 strain, and various inhibitors in comparison to the cases for live or ultraviolet (u.v.)-irradiated influenza and Sendai viruses. Induction of target susceptibility with heated PR8 was not inhibited by cycloheximide and actinomycin D as in the case of u.v.-irradiated Sendai virus, whereas live virus and u.v.-irradiated PR8 were inhibited under conditions which suppress protein synthesis. Induction of target susceptibility with the live and inactivated PR8 tested was suppressed in the presence of chloroquine, contrary to the case of Sendai virus, and was dependent on the cleavage type of influenza virus haemagglutinin. These findings suggest that the viral target antigens recognized by CTL in heated PR8-coated targets came from inoculated virus proteins, whereas those in PR8-infected or u.v.-irradiated PR8-coated targets involved newly synthesized viral proteins. The former viral target antigens seem to be transferred or processed from the endosome, depending on low pH fusion in the endosomes into which they were engulfed. In this point, the induction of viral target antigens with heated PR8 was different from that induced by u.v.-inactivated Sendai virus. Targets made with heated PR8 were recognized by cross-reactive CTL over the HA subtype.

Ontogeny of macrophage-mediated protection against Listeria monocytogenes.

We investigated the ontogenic development of macrophage functions which are important in the expression of host defense against infection by Listeria monocytogenes. Macrophage functions, including accumulation in response to local stimuli, chemotaxis in vitro, and intracellular killing, as well as number of macrophages, were examined by using mice 1, 2, 3, 4, and 8 weeks old. The number of peritoneal macrophages was extremely low in younger mice even when their body weights were taken into consideration. Macrophage accumulation in response to infectious stimulus with viable listeria was poor in younger mice and showed an age-dependent development. In younger groups, chemotaxis in vitro was as immature as chemotaxis in vivo. In 1- and 2-week-old mice, macrophages did not show any intracellular killing activity against L. monocytogenes, but killing was observed in mice over 3 weeks of age. These functions developed in an age-dependent manner and reached the 8-week-old adult level after the mice were 4 weeks of age. In adult mice, these macrophage functions were shown to be enhanced after immunization with viable listeria; however, such an immunization-induced enhancement was very poor in the younger groups of mice. Protection judged by mortality and in vivo bacterial growth was weaker in the younger groups against both primary and secondary challenges. In vivo protection against L. monocytogenes seemed to develop in the same age-dependent manner as the development of macrophage functions. These results indicate that age-dependent immaturity of macrophage functions mainly comprises the age-dependent immaturity of protection against L. monocytogenes.

Sleep apnea with unusual EEG changes in Jakob-Creutzfeldt disease.

Disruption of the periodic pattern in the EEG was observed in a patient with Jakob-Creutzfeldt disease. Periods of apnea (during 'sleep') were associated with abrupt cessation of the periodic complexes and their replacement by 7-9 c/sec waves which persisted throughout the apneic episode. The periodic complexes reappeared on termination of the apnea. This close temporal association between disruption of the periodicity and apneic episodes suggests proximity of respiratory neurons and the deep subcortical pacemaker.