Yoshihiko Kaisho

Cloning and expression of a cDNA encoding a novel human neurotrophic factor

A cDNA encoding a novel human neurotrophic factor (designated nerve growth factor‐2; NGF‐2) was cloned from a human glioma cDNA library using a synthetic DNA corresponding to human nerve growth factor (NGF). The cloned cDNA encodes a polypeptide composed of 257 amino acid residues including a prepro‐sequence of 138 residues and a mature region of 119 residues. The amino acid sequence of human NGF‐2 exhibits 58% similarity with that of human NGF. Conditioned medium of COS‐7 cells transfected with an expression plasmid for human NGF‐2 cDNA supported the survival of sensory neurons isolated from dorsal root ganglia of embryonic chicks. A 1.5 kb of NGF‐2 mRNA can be detected from an early development stage in rat brain, by Northern blotting analysis.

Overexpression of GPR40 in Pancreatic β-Cells Augments Glucose-Stimulated Insulin Secretion and Improves Glucose Tolerance in Normal and Diabetic Mice

OBJECTIVE GPR40 is a G protein–coupled receptor regulating free fatty acid–induced insulin secretion. We generated transgenic mice overexpressing the hGPR40 gene under control of the mouse insulin II promoter and used them to examine the role of GPR40 in the regulation of insulin secretion and glucose homeostasis. RESEARCH DESIGN AND METHODS Normal (C57BL/6J) and diabetic (KK) mice overexpressing the hGPR40 gene under control of the insulin II promoter were generated, and their glucose metabolism and islet function were analyzed. RESULTS In comparison with nontransgenic littermates, hGPR40 transgenic mice exhibited improved oral glucose tolerance with an increase in insulin secretion. Although islet morphologic analysis showed no obvious differences between hGPR40 transgenic and nontransgenic mice, isolated islets from hGPR40 transgenic mice had enhanced insulin secretion in response to high glucose (16 mmol/l) compared with those from nontransgenic mice, and they both had similar low glucose (3 mmol/l)-stimulated insulin secretion. In addition, hGPR40 transgenic islets significantly increased insulin secretion against a naturally occurring agonist palmitate in the presence of 11 mmol/l glucose. hGPR40 transgenic mice were also found to be resistant to high-fat diet–induced glucose intolerance, and hGPR40 transgenic mice harboring KK background showed augmented insulin secretion and improved oral glucose tolerance compared with nontransgenic littermates. CONCLUSIONS Our results suggest that GPR40 may have a role in regulating glucose-stimulated insulin secretion and plasma glucose levels in vivo and that pharmacological activation of GPR40 may provide a novel insulin secretagogue beneficial for the treatment of type 2 diabetes.

Tissue Distribution and Immunocytochemical Localization of Neurotrophin‐3 in the Brain and Peripheral Tissues of Rats

Abstract: The tissue distribution of neurotrophin‐3 (NT‐3) was investigated in rats at 1 month of age using a newly established, sensitive two‐site enzyme immunoassay system for NT‐3, as well as the immunocytochemical localization of this protein. The immunoassay for NT‐3 enabled us to quantify NT‐3 at levels > 3 pg per assay. In the rat brain, NT‐3 was detectable only in the olfactory bulb (0.54 ng/g wet weight), cerebellum (0.71 ng/g), septum (0.91 ng/g), and hippocampus (6.3 ng/g). By contrast, NT‐3 was widely distributed in peripheral tissues. Appreciable levels of NT‐3 were also found in the thymus (31 ng/g), heart (38 ng/g), diaphragm (21 ng/g), liver (45 ng/g), pancreas (892 ng/g), spleen (133 ng/g), kidney (40 ng/g), and adrenal gland (46 ng/g). An antibody specific for NT‐3 bound to pyramidal cells in the CA2‐CA4 regions of the hippocampus, to A cells in the islets of Langerhans in the pancreas, to unidentified cells in the red pulp of the spleen, to liver cells, and to muscle fibers in the diaphragm from rats at 1 month of age. Molecular masses of NT‐3‐immunoreactive proteins in the hippocampus and pancreas were 14 and 12 kDa, respectively. Thus, in rats, NT‐3 was detected in restricted regions of the brain and in the visceral targets of the nodose ganglia at high concentrations. Our present results suggest that NT‐3 not only functions as a classical target‐derived neurotrophic factor but also can play other roles.

Targeted disruption of the neurotrophin-3 gene with lacZ induces loss of trkC-positive neurons in sensory ganglia but not in spinal cords

We have replaced the NT-3 gene with Escherichia coli-derived lacZ gene by means of homologous recombination in embryonic stem cells and thus produced null mutant mice. Mice homozygous for this mutation developed to birth, but most of them could not suck well and died within 2 days after birth. The surviving homozygous mutant mice displayed movement disorder similar to ataxia. The expression of lacZ was widely distributed in the target tissues of peripheral nerves, spinal motor neurons, lumbar dorsal root ganglia and trigeminal ganglia during the prenatal periods. A neuroanatomical examination revealed that there was marked cell reduction present in trigeminal and lumbar dorsal root ganglia in the developing homozygous mutant mice. In these tissues, the expression of trkC, a high-affinity receptor for NT-3, was markedly reduced. In contrast, we did not find any morphological abnormalities, significant cell loss or decreased levels of trkC expression in the motor neurons present in the ventral horn of the spinal cord. These results indicate that the absence of the NT-3 gene leads to a defect in the sensory nervous system, but it may be complemented by other neurotrophins in the motor nervous system during the development.

Regional expression of the nerve growth factor gene family in rat brain during development

The developmental expression patterns of three members (NGF, NGF-2/NT-3, and BDNF) of the NGF family in rat brain are different. NGF-2/NT-3 mRNA was the first detected during development followed by NGF and BDNF mRNAs. A substantial amount of NGF mRNA was found to be synthesized in the hippocampus and the cortex, and this regional expression pattern did not change during development. In contrast, NGF-2/NT-3 mRNA was detected in almost all the brain regions examined in the early developmental stage. In the late stage, the transcript was found in high concentration only in the hippocampus and the cerebellum. BDNF mRNA was widely distributed, and its level was augmented in the late developmental stage.

Production, purification and characterization of biologically active recombinant human nerve growth factor

The human NGF gene was isolated and inserted downstream from murine leukemia virus LTR in a plasmid having dihydrofolate reductase cDNA. The expression plasmid was introduced into CHO cells. Selection of the transformants for the resistance to methotrexate gave a CHO cell line which produced human NGF at a level of 4 mg/L in the culture medium. The recombinant human NGF was purified to near homogeneity from the culture supernatant. The NH2-terminal amino acid sequence, the COOH-terminal amino acid (Ala), and the amino acid composition of the human NGF were identical to those deduced from the nucleotide sequence of the human NGF gene. The recombinant human NGF was composed of 120 amino acid residues. Three disulfide linkages were determined to be Cys15-Cys80, Cys-58-Cys108, and Cys68-Cys110; the locations were identical to those in the mouse 2.5S NGF molecule. The specific biological activity of the recombinant human NGF was comparable with that of authentic mouse 2.5S NGF as determined by stimulation of neurite outgrowth from PC12 cells.

Expression of neurotrophin genes in the brain of senescence-accelerated mouse (SAM) during postnatal development

We compared the expression patterns of neurotrophin genes in the brain of senescence-accelerated mouse (SAMP8) which shows age-related impairment of learning behavior, with SAMR1 control which shows normal aging. By Northern blot analysis, NT-3 mRNA levels in the cortex were higher in SAMP8 than in SAMR1 mice during development, whereas in the midbrain, hippocampus and forebrain, NT-3 expression levels in SAMP8 were lower than those in SAMR1. At early stages, although NGF mRNA levels in SAMP8 were lower than those in SAMR1, BDNF mRNA levels were almost equivalent in both strains. By in situ hybridization analysis, NT-3 mRNA signals in the CA1 and CA2 regions in SAMP8 were shown to be reduced at early stages. However, BDNF mRNA signals were almost equivalent in both SAMR1 and SAMP8.

Analysis of neurotrophin-3 expression using the lacZ reporter gene suggests its local mode of neurotrophic activity

We replaced the mouse neurotrophin-3 gene with the Escherichia coli-derived lacZ gene by means of homologous recombination. The mice with this mutation were useful models for studying the distribution of neurotrophin-3 expression in vivo, because visualization by 5-bromo-4-chloro-3-indoyl-beta-D-galactopyranoside (X-Gal) staining was simple and rapid compared with in situ hybridization or immunohistochemistry. Whole-mount staining of mutant embryos at embryonic day 10 revealed that lacZ, a reporter for the neurotrophin-3 gene, was expressed in the mesencephalon, mandibular arch and somites. In the embryos at days 13-17, lacZ was markedly expressed in the peripheral target tissues of sensory and sympathetic neurons. We also found that spinal motor neurons and sensory neurons in trigeminal and dorsal root ganglia express lacZ. Some of these X-Gal staining regions overlapped with the sites expressing trkC, a high-affinity receptor for neurotrophin-3. The distribution of X-Gal staining in heterozygotes and homozygotes was similar to that of neurotrophin-3 messenger RNA detected by in situ hybridization. However, there was less lacZ expression in the dorsal root ganglia of homozygotes than neurotrophin-3 expression in wild-type mice. These results suggest that the neurotrophin-3 produced in the dorsal root ganglia also plays a role in the survival of some of the neurotrophin-3-positive neurons and that the local mode of neurotrophic activity is widely distributed.

Developmental changes of neurotrophin-3 level in the mouse brain detected by a highly sensitive enzyme immunoassay

Levels of neurotrophin-3 (NT-3) in the mouse brain were measured by a highly sensitive enzyme immunoassay (EIA). The monoclonal antibody, 3W3, was labeled with beta-galactosidase, followed by measurement of galactosidase activity. The detection limit of the EIA system was 0.4 pg/well (4 pg/ml). At 1 and 8 weeks of age, the highest level of NT-3 was detected in the hippocampus, a relatively high level also observed in the cerebellum. In contrast, in the cortex, the striatum, the diencephalon, the midbrain, and the brainstem, NT-3 levels were low. Furthermore, we examined the developmental changes of NT-3 level in the hippocampus and the cerebellum. In the hippocampus, the NT-3 levels were more than 20 ng/g tissue from 1 week to 14 weeks of age, but at 20 weeks of age the level decreased to about half. In the cerebellum, although the NT-3 level was high at 1 week of age, the levels were gradually decreased to one-fourth by 20 weeks of age. In peripheral tissues, a large amount of NT-3 protein was observed in the heart.

Neurotrophin-3 is expressed in the posterior lobe of mouse cerebellum, but does not affect the cerebellar development

We replaced the neurotrophin-3 (NT-3) gene with Escherichia coli-derived lacZ via homologous recombination in embryonic stem (ES) cells and generated the mutant mice. Here we show the in vivo expression of NT-3 in the cerebellum during the postnatal period. A high level of lacZ expression was found in the granular layer of posterior lobe (lobules VII to X) in the postnatal NT-3(+/-) cerebellum. The expression in these regions was reduced with age. Although the Purkinje cells are considered to be a target of NT-3 and the NT-3(-/-) mice displayed severe moving disorders like ataxia, no histological abnormality was observed in their cerebellum. These findings suggest that the NT-3 expressed in the cerebellum gives some trophic effects primarily to the posterior lobe, however, the deficiency does not affect its development.