Ghrelin is a newly identified gastric peptide hormone that has various important functions, including growth-hormone release and appetite stimulation. Ghrelin-immunoreactive cells (ghrelin cells) are characterized by X-type endocrine cells in the rat stomach. In the present study, we analysed ghrelin cells in fundi of stomach from ICR mice and Syrian hamsters immunohistochemically, immunoelectron microscopically and morphometrically, and compared the results with those from Wistar rats. Immunohistochemistry revealed that ghrelin cells were sparsely distributed in the proper gastric glands in all species. The number of ghrelin cells per unit area in hamsters was significantly lower than that in rats. Immunoelectron microscopy detected ghrelin immunolabelling in granules in the X-type endocrine cells. However, the diameter of granules in the hamsters was significantly smaller than that in the mice and rats. Gastric ghrelin contents were determined by radioimmunoassay, and levels in the hamsters were significantly lower than those in mice and rats. The results from mice were identical to those from rats. In conclusion, gastric ghrelin cells in mice and hamsters are characterized by X-type endocrine cells, as has been observed in rats. However, the data indicated that gastric ghrelin production was lower in hamster than in mouse or rat.

Hydrogen (H2) acts as a therapeutic antioxidant. However, there are few reports on H2 function in other capacities in diabetes mellitus (DM). Therefore, in this study, we investigated the role of H2 in glucose transport by studying cultured mouse C2C12 cells and human hepatoma Hep-G2 cells in vitro, in addition to three types of diabetic mice [Streptozotocin (STZ)-induced type 1 diabetic mice, high-fat diet-induced type 2 diabetic mice, and genetically diabetic db/db mice] in vivo. The results show that H2 promoted 2-[14C]-deoxy-d-glucose (2-DG) uptake into C2C12 cells via the translocation of glucose transporter Glut4 through activation of phosphatidylinositol-3-OH kinase (PI3K), protein kinase C (PKC), and AMP-activated protein kinase (AMPK), although it did not stimulate the translocation of Glut2 in Hep G2 cells. H2 significantly increased skeletal muscle membrane Glut4 expression and markedly improved glycemic control in STZ-induced type 1 diabetic mice after chronic intraperitoneal (i.p.) and oral (p.o.) administration. However, long-term p.o. administration of H2 had least effect on the obese and non-insulin-dependent type 2 diabetes mouse models. Our study demonstrates that H2 exerts metabolic effects similar to those of insulin and may be a novel therapeutic alternative to insulin in type 1 diabetes mellitus that can be administered orally.

No abstract

No abstract

Peripheral obstruction of intrahepatic portal vein branches was detected by dynamic sequential computed tomography during arterial portography and subsequently confirmed surgically in 9 patients with hepatic neoplasm (7 hepatocellular carcinomas, 1 cholangiocarcinoma, and 1 metastatic lymphadenopathy from gastric carcinoma). In 1 patient, 2 obstructed segments were seen. Eight of these 10 segments showed more dense staining than other regions of the liver during infusion hepatic angiography. Retrograde opacification of the peripheral venules of the obstructed portion was seen in 2 of these 8 segments. This pattern was attributed to trans-sinusoidal or peripheral arterioportal shunting. In 5 cases, the segmental staining obscured the tumor stain, making the tumor appear larger than it actually was or causing it to be missed altogether.

In this paper we propose a minimum α-information method to maximize and minimize information contained in hidden units. The method aims to interpret internal representations and to improve generalization performance. The α-information minimization forces hidden units to have maximum information or minimum information, depending on the importance of hidden units. To interpret internal representation, we have only to see a small number of hidden units with maximum information, ignoring hidden units with minimum information. In addition, by minimizing the α-information, unnecessary information can be eliminated, leading to better generalization. First, α-information was applied to a simple data compression problem in which four characters can be compressed into one hidden unit. Then we applied α-information minimization to one of the most challenging topics in neural networks, namely, the description and understanding of natural languages.As an example to demonstrate the acquisition of descriptive adequacy by neural networks, we dealt with the inference of well-formedness of an artificial language close to English. Experimental results confirmed that explicit internal representations can be obtained and generalization can be significantly improved. 1. Introduction. Many attempts have been made to describe learning by neural networks from the information theoretic point of view. Information has been maximized [12], [18]-[20] or minimized [1], [8], [13], depending on problems. Information maximization, applied to supervised learning, has been used to make hidden units respond clearly to input patterns [12]. Thus, it is easy to interpret interpretations. However, it is clear that information maximization cannot deal well with ambiguous input patterns, because hidden units must respond explicitly to any input pattern. By explicit responses to input patterns, improved generalization performance cannot be expected. To increase generalization performance, information has been minimized [1], [8], [13]. Information minimization prevents hidden units from responding too clearly to those input patterns.Thus, if information can be maximized and minimized at the same time, it is expected to have internal representations for better interpretation and improved generalization. The α-information is proposed to represent a situation in which information minimization and information maximization are both needed. When the α-information, represented in the difference between Shannon and Rényi entropy, is minimized, information can be maximized or minimized. Thus, α-information minimization performs an operation of information maximization and information minimization at the same time. Thus, the method can be used to interpret explicitly internal representations and to improve generalization.

To examine both of the target vessels and the optimal time of their endothelial denudation to study vascular restenosis after balloon injury in cholesterol-loaded rabbits, we made 36 atherosclerotic rabbits by feeding a hypercholesterol diet, and histologically examined the onset time and the development of atherosclerosis. Atheromatous changes were observed first after the 5th week in the thoracic aorta from the start of the diet, and then extended to the abdominal aorta, coronary artery with time. The atherosclerotic lesions in the thoracic aorta and the proximal portion of the coronary artery showed high-grade concentric intimal thickening with luminal stenosis. The abdominal aortic lesion mildly progressed. In the renal, carotid and femoral arteries, in contrast, slight atheroscleromatous changes developed during the diet period. These results suggest that the thoracic and abdominal aortas and the coronary artery would be suitable as target vessels to study vascular restenosis after balloon injury, and the endothelial denudation of these vessels should be performed between the 8th and 15th week in this diet protocol for an accurate analysis.

Ghrelin is a novel peptide hormone, originally identified in the rat and human stomach that plays various important roles. In the present study, we report the intra-renal localization of ghrelin in laboratory rodents. Kidneys from 3 month-old mice, rats and hamsters of both sexes were analyzed by immunohistochemistry. Positive signals were clearly observed in the epithelium of the distal tubules, whereas other segments of the nephron or interstitial cells, including juxtaglomerular cells, showed negative reactions. Pre-embedding immunoelectron microscopy revealed positive signals exclusively on the basolateral membrane in the distal tubular cells and in the collecting ducts. In addition, prepro-ghrelin gene expression was assessed by RT-PCR, and the expected 329-bp prepro-ghrelin mRNA was clearly detected in the kidney. On Western blot analysis, although a specific band for ghrelin (3 kDa) was not detected in the kidney, the expected band for prepro-ghrelin (13 kDa) was clearly detected in both the stomach and the kidney. This paper clarified the intra-renal localization of ghrelin.