Recurrent correlation associative memories: a feature space perspective

In this paper, we analyze a model of recurrent kernel associative memory (RKAM) recently proposed by Garcia and Moreno. We show that this model consists in a kernelization of the recurrent correlation associative memory (RCAM) of Chiueh and Goodman. In particular, using an exponential kernel, we obtain a generalization of the well-known exponential correlation associative memory (ECAM), while using a polynomial kernel, we obtain a generalization of higher order Hopfield networks with Hebbian weights. We show that the RKAM can outperform the aforementioned associative memory models, becoming equivalent to them when a dominance condition is fulfilled by the kernel matrix. To ascertain the dominance condition, we propose a statistical measure which can be easily computed from the probability distribution of the interpattern Hamming distance or directly estimated from the memory vectors. The RKAM can be used below saturation to realize associative memories with reduced dynamic range with respect to the ECAM and with reduced number of synaptic coefficients with respect to higher order Hopfield networks.

Neural associative memory storing gray-coded gray-scale images

We present a neural associative memory storing gray-scale images. The proposed approach is based on a suitable decomposition of the gray-scale image into gray-coded binary images, stored in brain-state-in-a-box-type binary neural networks. Both learning and recall can be implemented by parallel computation, with time saving. The learning algorithm, used to store the binary images, guarantees asymptotic stability of the stored patterns, low computational cost, and control of the weights precision. Some design examples and computer simulations are presented to show the effectiveness of the proposed method.

The acquisition of an insulin-secreting phenotype by HGF-treated rat pancreatic ductal cells (ARIP) is associated with the development of susceptibility to cytokine-induced apoptosis

The elucidation of mechanisms regulating the regeneration and survival of pancreatic beta cells has fundamental implications in the cell therapy of type 1 diabetes. The present study had the following three aims: 1. to investigate whether pancreatic ductal epithelial cells can be induced to differentiate into insulin-producing cells by exposing them to hepatocyte growth factor (HGF); 2. to characterize some of the molecular events leading to their differentiation toward a beta-cell-like phenotype; 3. to evaluate the susceptibility of newly differentiated insulin-secreting cells to cytokine-induced apoptosis, a mechanism of beta-cell destruction occurring in type 1 diabetes. We demonstrated that HGF-treated rat pancreatic ductal cell line (ARIP) cells acquired the capability to transcribe the insulin gene and translate its counterpart protein. HGF-treated cells also exhibited a glucose-dependent capability to secrete insulin into the cultured medium. Expression analysis of some of the genes regulating pancreatic beta-cell differentiation revealed a time-dependent transcription of neurogenin-3 and Neuro-D in response to HGF. Finally, we determined the susceptibility to proinflammatory cytokine (PTh1)-induced apoptosis by incubating HGF-treated and untreated ARIP cells with a cocktail of interleukin-1 beta (IL-1beta), tumor necrosis factor-alpha (TNF-alpha) and interferon-gamma (IFN-gamma). Such treatment induced apoptotic death, as determined by the TUNEL technique, in about 40% of HGF-treated, insulin-secreting ARIP cells, while untreated ARIP cells were resistant to PTh1-induced apoptosis. In conclusion, we showed that HGF promotes the differentiation of ARIP cells into pancreatic beta-cell-like cells, and that the differentiation toward an insulin-secreting phenotype is associated with the appearance of susceptibility to cytokine-induced apoptosis.

The role of GLP-1 in the life and death of pancreatic beta cells

Glucagon-like peptide-1 (GLP-1), a peptide hormone produce by intestinal cells, has recently been shown to be capable of modulating islet cell mass. Administration of GLP-1 to rodent models of type 2 diabetes ameliorates insulin secretion, induces the replication of islet cells, and promotes islet-cell neogenesis from pancreatic ductal cells susceptible to transdifferentiate in insulin-producing cells. In addition, an anti-apoptotic effect of GLP-1 has been described in hyperglycemic animal models, using freshly isolated human islets or cultured beta cell lines exposed to various pro-apoptotic stimuli. The aim of this article is to review those reports that have emphasized the role of GLP-1 as a regulator of islet cell mass.

Cultured pancreatic ductal cells undergo cell cycle re-distribution and beta-cell-like differentiation in response to glucagon-like peptide-1

The intestinal hormone glucagon-like peptide-1 (GLP-1) has been shown to promote an increase in pancreatic beta-cell mass via proliferation of islet cells and differentiation of non-insulin-secreting cells. In this study, we have characterized some of the events that lead to the differentiation of pancreatic ductal cells in response to treatment with human GLP-1. Rat pancreatic ductal (ARIP) cells were cultured in the presence of GLP-1 and analyzed for cell counting, cell cycle distribution, expression of cyclin-dependent-kinase (Cdk) inhibitors, transcription of beta-cell-specific genes, loss of ductal-like phenotype and acquisition of beta-cell-like gene expression profile. Exposure of ARIP cells to 10 nM GLP-1 induced a significant reduction in the cell replication rate and a significant decrease in the percentage of cells in S phase of the cell cycle. This was associated with an increase in the number of cells in G0-G1 phase and a reduction of cells in G2-M phase. Western blot analysis for the Cdk inhibitors, kinase inhibitor protein 1 (p27(Kip1)) and Cdk-interacting protein 1 (p21(Cip1)), demonstrated a significant increase in p27(Kip1) and p21(Cip1) levels within the first 24 h from the beginning of GLP-1 treatment. As cells slowed down their proliferation rate, GLP-1 also induced a time-dependent expression of various beta-cell-specific mRNAs. The glucose transporter GLUT-2 was the first of those factors to be expressed (24 h treatment), followed by insulin (44 h) and finally by the enzyme glucokinase (56 h). In addition, immunocytochemistry analysis showed that GLP-1 induced a time-dependent down-regulation of the ductal marker cytokeratin-20 (CK-20) and a time-dependent induction of insulin expression. Finally, GLP-1 promoted a glucose-dependent secretion of insulin, as demonstrated by HPLC and RIA analyses of the cell culture medium. The present study has demonstrated that GLP-1 induces a cell cycle re-distribution with a decrease in cell proliferation rate prior to promoting the differentiation of cells towards an endocrine-like phenotype.

Pancreatic beta-cells expressing the Arg64 variant of the beta(3)-adrenergic receptor exhibit abnormal insulin secretory activity

The Arg64 beta(3)-adrenergic receptor (beta(3)AR) variant is associated with an earlier age of onset of diabetes and lower levels of insulin secretion in humans. The aims of this study were to investigate whether beta(3)AR is expressed by islet cells, if receptor binding affects insulin secretion and, finally, if the beta(3)AR Arg64 variant induces abnormal insulin secretory activity. Human pancreas extracts were subjected to RT-PCR, Western blotting and immunostaining analyses. DNA sequencing and Western blotting demonstrated that the beta(3)AR gene is transcribed and translated in the human pancreas; immunostaining showed that it is expressed by the islets of Langerhans. Cultured rat beta-cells responded to human beta(3)AR agonists in a dose- and time-dependent manner. Transfection of cultured rat beta-cells with the wild-type human beta(3)AR produced an increased baseline and ligand-dependent insulin secretion compared with parental cells. On the other hand, cells transfected with the Arg64 variant of the beta(3)AR secreted less insulin, both spontaneously and after exposure to human beta(3)AR agonists. Furthermore, while transfection with the wild-type beta(3)AR preserved the glucose-dependent secretion of insulin, expression of the variant receptor rendered the host cells significantly less responsive to glucose. In summary, cells express the beta(3)AR, and its activation contributes to the regulation of insulin secretion. These findings may help explain the low levels of insulin secretion in response to an i.v. glucose tolerance test observed in humans carrying the Arg64 polymorphism.

Enhancing insulin action: from chemical elements to thiazolidinediones

Type 2 diabetes is characterized by two fundamental biological defects: a reduced glucose-dependent insulin secretion and an increased resistance to the action of insulin at the level of various target tissues. While the use of agents to improve the insulin secretory activity of the islets of Langerhans has witnessed the flourishing of several new drugs over the years, a much greater difficulty has been experienced in the search for insulin-sensitizing drugs. The aim of this article is to critically review this topic, and to emphasize the importance of providing alternative strategies for the management of Type 2 diabetes.

Glucagon-like peptide 1 induces differentiation of islet duodenal homeobox-1-positive pancreatic ductal cells into insulin-secreting cells

Glucagon-like peptide-1 (GLP-1) is an incretin hormone capable of restoring normal glucose tolerance in aging glucose-intolerant Wistar rats. Whether the antidiabetic properties of GLP-1 are exclusively due to its insulin secretory activity remains to be determined. A GLP-1-dependent differentiation of pancreatic precursor cells into mature beta-cells has recently been proposed. The aim of this study was to investigate whether pancreatic ductal epithelial cells could be differentiated into insulin-secreting cells by exposing them to GLP-1. Rat (ARIP) and human (PANC-1) cell lines, both derived from the pancreatic ductal epithelium, were used to test this hypothesis. A major difference distinguishes these two cell lines: whereas ARIP cells spontaneously express the beta-cell differentiation factor islet duodenal homeobox-1 (IDX-1), PANC-1 cells are characteristically IDX-1 negative. GLP-1 induced the differentiation of ARIP cells into insulin-synthesizing cells, although it did not affect the phenotype of PANC-1 cells, as determined by fluorescence-activated cell sorting (FACS) analysis. Differentiation of ARIP cells by exposure to human GLP-1 occurs in a time- and dose-dependent manner, and this is associated with an increase in IDX-1 and insulin mRNA levels. Secretion of insulin was also induced in a parallel manner, and it was regulated by the concentration of glucose in the culture medium. Interestingly, PANC-1 cells, when stably transfected with human IDX-1, gained responsiveness to GLP-1 and were able to differentiate into beta-cells, as determined by FACS analysis, insulin gene expression, intracellular insulin content, and insulin accumulation in the culture medium. Finally, we demonstrated that the receptor for GLP-1 is constitutively expressed by ARIP and PANC-1 cells and that the mRNA level for this transcript was increased by cellular transfection with human IDX-1. In summary, our study provides evidence that GLP-1 is a differentiation factor for pancreatic ductal cells and that its effect requires the expression of IDX-1.

Glucagon-like peptide-1: a major regulator of pancreatic beta-cell function

Glucagon-like peptide-1 (GLP-1) is a gut hormone synthesized by post-translational processing in intestinal L-cells, and it is released in response to food ingestion. GLP-1 stimulates insulin secretion during hyperglycemia, suppresses glucagon secretion, stimulates (pro)-insulin biosynthesis and decreases the rate of gastric emptying and acid secretion. GLP-1 has also been shown to have a pro-satiety effect. In addition, it has been demonstrated that a long-term infusion with GLP-1, or exendin-4, a long-acting analog of human GLP-1, increases beta-cell mass in rats. In conclusion, GLP-1 appears to regulate plasma glucose levels via various and independent mechanisms. GLP-1 is an excellent candidate option for the treatment of patients with type 2 diabetes mellitus.

Glucagon-like peptide-1 induces cell proliferation and pancreatic-duodenum homeobox-1 expression and increases endocrine cell mass in the pancreas of old, glucose-intolerant rats

Glucose homeostasis in mammals is maintained by insulin secretion from the beta-cells of the islets of Langerhans. Type 2 diabetes results either from primary beta-cell failure alone and/or a failure to secrete enough insulin to overcome insulin resistance. Here, we show that continuous infusion of glucagon-like peptide-1 (7-36) (GLP-1; an insulinotropic agent), to young and old animals, had effects on the beta-cell of the pancreas other than simply on the insulin secretory apparatus. Our previous studies on a rodent model of glucose intolerance, the aging Wistar rat, show that a plateau in islet size, insulin content, and beta-cell mass is reached at 13 months, despite a continuing increase in body weight. Continuous sc infusion of GLP-1 (1.5 pM/kg x min), over 5 days, resulted in normal glucose tolerance. Our current results in both young and old rats demonstrate that treatment caused an up-regulation of pancreatic-duodenum homeobox-1 (PDX-1) expression in islets and total pancreas, induced pancreatic cell proliferation, and beta-cell neogenesis. The effects on levels of PDX-1 messenger RNA were abrogated by simultaneous infusion of Exendin (9-39), a specific antagonist of GLP-1. PDX-1 protein levels increased 4-fold in whole pancreata and 6-fold in islets in response to treatment. Beta-cell mass increased to 7.2 +/- 0.58 from 4.88 +/- 0.38 mg, treated vs. control, respectively, P < 0.02. Total pancreatic insulin content also increased from 0.55 +/- 0.02 to 1.32 +/- 0.11 microg/mg total pancreatic protein. Therefore, GLP-1 would seem to be a unique therapy that can stimulate pancreatic cell proliferation and beta-cell differentiation in the pancreas of rodents.

Novel sulfonylurea and non-sulfonylurea drugs to promote the secretion of insulin

The onset of type 2 diabetes is characterized by two determining factors: the insufficient ability to secrete insulin and/or the resistance to its biological action. Although in a very small proportion of individuals, one of those two metabolic abnormalities is the leading cause of diabetes, in most subjects, the coexistence of both appears to be necessary for the clinical manifestation of diabetes. Current biomedical research continues to clarify the relative contributions of these defects to the pathogenesis of type 2 diabetes, and novel pharmacological agents are specifically designed to correct either the impaired insulin secretory activity or the resistance to the action of insulin. The aim of this article is to provide a critical review of new sulfonylurea and non-sulfonylurea drugs that have been recently introduced for the treatment of diabetes, as well as drugs that are still under investigation and are likely to be made available in the near future.

Regulation of pancreatic PC1 and PC2 associated with increased glucagon-like peptide 1 in diabetic rats

The pancreatic processing enzymes, PC1 and PC2, convert proinsulin to insulin and convert proglucagon to glucagon and glucagon-like peptide 1 (GLP-1). We examined the effect of streptozotocin (STZ) treatment on the regulation of these enzymes and the production of insulin, glucagon, and GLP-1 in the rat. Pancreatic PC1 and PC2 mRNA increased >2-fold and >4-fold, respectively, in rats receiving intraperitoneal STZ (50 mg/kg) daily for 5 days. Immunocytochemistry revealed that, although pancreatic islet cells in the STZ-treated rats were sparse and atrophic PC1, PC2, glucagon, and GLP-1 immunoreactivity increased dramatically in the remaining islet cells. Heightened PC1 and PC2 expression was seen in cells expressing glucagon but not in insulin-expressing cells. Furthermore, in STZ-treated rats, bioactive GLP-1(7-36 amide) accumulated in pancreatic extracts and serum 3- and 2.5-fold, respectively, over control animals. This treatment also caused a 2-fold increase in the ratio of amidated forms of GLP-1 immunoreactivity to total glucagon immunoreactivity in the pancreas but did not affect the ratio of proinsulin to insulin. We conclude that hyperglycemic rats have an increased expression of prohormone converting enzymes in islet alpha cells, leading to an increase in amidated GLP-1, which can then exert an insulinotropic effect on the remaining beta cells.

Signalling via receptor tyrosine kinase modulates the expression of the DNA repair enzyme XPD in cultured cells

Oxidative damage to DNA has been documented in cells isolated from subjects with diabetes. Herein, we evaluate the mechanism(s) that regulate the expression of the DNA repair enzyme XPD. CHO cells transfected with the human insulin receptor (CHO/HIRc) showed a threefold increase in the level of XPD mRNA when compared to control CHO/neo cells (P < 0.01). The addition of insulin to serum-starved cells led to an increase in XPD mRNA levels in both CHO/neo and CHO/HIRc cells, in a time and dose dependent fashion. Insulin acted primarily by inducing XPD transcription. Moreover, inhibition of protein synthesis by cyclohexamide induced a marked degradation of XPD mRNA levels in insulin treated cells. Site-directed mutagenesis of the tyrosine-kinase domain of the insulin receptor abolished the increase in XPD mRNA resulting from the transfection with wild type insulin receptors (P < 0.001). Western blot analysis of cell extracts from CHO/neo and CHO/HIRc cells revealed an increase in XPD counterpart protein was also induced by transfecting cells with the human insulin receptor. Evaluation of DNA damage by means of internucleosomal fragmentation showed a dramatic decrease in DNA fragmentation in CHO cells transfected with wild-type insulin receptor compared to control CHO/neo cells. DNA fragmentation was further decreased by the addition of insulin in the culture medium. In summary, our data indicates that activation of the insulin receptor plays an important role in the cellular response leading to repair of damaged DNA.

Control of glucose homeostasis by incretin hormones

In humans as well as in other animal species, the ingestion of food provides the fundamental source of energy for various cellular activities. The intake of food and the ability of controlling the plasma levels of substrates for energy production involve complex mechanisms that ensure a constantly adequate supply of metabolites both in the fasting and in the fed state. A number of hormonal peptides released from the gastrointestinal (GI) tract in response to the ingestion of food have been shown to play a critical role in the postprandial control of glucose homeostasis. They are known to act through three main mechanisms of action. These include; (1) stimulation of insulin secretion of pancreatic islet (beta) cells; (2) inhibition of hepatic gluconeogenesis by suppression of glucagon secretion; and (3) inhibition of GI motility. While for some of these hormones all three mechanisms of action are utilized under physiological conditions, others preferentially use one or a combination of two mechanisms for lowering postprandial hyperglycemia. Although the term glucoincretins (or incretins, or insulinotropic hormones) etymologically only describes factors capable of inducing insulin secretion, it is more frequently used to identify a larger class of peptides that, rather than manifesting a specific mechanism of action (i.e., insulin secretion), share the ability of controlling glucose excursion in the fed state (with or without a direct insulinotropic effect). The latter more inclusive meaning, incretins, is used in this article. This review summarizes recent advances on synthesis, secretion, blood plasma patterns, and metabolism of some of the major GI regulatory peptides acting in the postprandial state.

Novel therapeutic strategies for the treatment of type 2 diabetes

Diabetes mellitus is the most common endocrine disease, accounting for over 200 million people affected worldwide. It is characterized by a lack of insulin secretion and/or increased cellular resistance to insulin, resulting in hyperglycemia and other metabolic disturbances. People with diabetes suffer from increased morbidity and premature mortality related to cardiovascular, microvascular and neuropathic complications. The Diabetes Control and Complication Trial (DCCT) has convincingly demonstrated the relationship of hyperglycemia to the development and progression of complications and showed that improved glycemic control reduced these complications. Although the DCCT exclusively studied patients with Type 1 diabetes, there is ample evidence to support the belief that the same relationship between metabolic control and clinical outcome exists in patients with Type 2 diabetes. Therefore, a major effort should be made to develop and implement more effective treatment regimes. This article reviews those novel drugs that have been recently introduced for the management of Type 2 diabetes, or that have reached an advanced level of study and will soon be proposed for preliminary clinical trials. They include: (i) compounds that promote the synthesis/secretion of insulin by the beta-cell; (ii) inhibitors of the alpha-glucosidase activity of the small intestine; (iii) substances that enhance the action of insulin at the level of the target tissues; and (iv) inhibitors of free fatty acid oxidation.

Regulation of glucose transporters and hexose uptake in 3T3-L1 adipocytes: glucagon-like peptide-1 and insulin interactions

Glucagon-like peptide-1 (7-36 amide) (GLP-1) is known to increase insulin release when given as a bolus in the fasted and fed state. GLP-1 also increases glucose uptake and lipid synthesis in cultured adipocytes. In this study we investigated the effects of GLP-1 on glucose uptake and on the levels of expression of the facilitative glucose transporters, GLUT1 and GLUT4, in fully differentiated 3T3-L1 adipocytes. Cells were incubated with GLP-1 (10 nM) with or without insulin (10 and 100 nM) for 24 h. Under these conditions, GLP-1 alone caused an increase in basal and acute insulin-stimulated glucose uptake along with an increase in GLUT1 and GLUT4 protein levels. However, there was no change in the expression of GLUT1 and GLUT4 mRNAs. In the absence of GLP-1, prolonged exposure to insulin caused a marked reduction in the levels of GLUT4 mRNA and protein, and an inhibition of glucose uptake after an acute exposure to insulin. This insulin-induced down-regulation of GLUT4 was prevented when GLP-1 was present during the 24-h treatment. In contrast, the acute insulin-stimulated glucose uptake could not be restored by GLP-1. GLP-1 is therefore the first gut hormone shown to be capable of modulating glucose transporter levels in cultured adipocytes.

The effects of GLP-1 on insulin release in young and old rats in the fasting state and during an intravenous glucose tolerance test

Glucose intolerance is a common feature of the aging process, and aging per se is an etiologic factor for Type II diabetes mellitus. To characterize the beta cell abnormalities that occur with aging, we looked at the serum glucose and insulin levels of six young (3-month) and six old (22-month) Wistar rats at 0, 2, 4, 7, 10, 15, 20, and 30 minutes after an intravenous glucose load (IVGTT, 0.5 g/kg glucose). We found that the fasting glucose and insulin levels were not significantly different between young and old rats. However, peak glucose levels were significantly higher in the old (349 +/- 10 mg/dl) compared to the young (250 +/- 7 mg/dl) animals (p < .0001). Insulin levels in the young animals peaked at 2 minutes (859 +/- 171 pmol/l) with a quick return toward fasting levels by 7 minutes. The old animals had a delayed and blunted insulin response to glucose, achieving lower peak insulin levels (656 +/- 164 pmol/l) 7 minutes after the glucose load. As insulin levels are also positively modulated by incretin hormones, we quantitated the fasting insulin responses of young and old animals to .05, 0.1, 0.2, 0.4, and 0.5 nmol/kg intravenous glucagon-like peptide-1 (GLP-1), the most potent incretin known. Insulin responses were similar in both age groups, with maximum insulin responses seen at 0.4 nmol/kg. GLP-1, in conjunction with the IVGTT, restored the acute insulin response to glucose and increased the clearance of glucose in the old animals. It therefore appears that old animals have an impaired glucose-mediated insulin release but maintain their insulin responsivity to GLP-1. This makes it a likely candidate in the treatment of Type II diabetes.

Glucagon-like peptide-1 can reverse the age-related decline in glucose tolerance in rats

Wistar rats develop glucose intolerance and have a diminished insulin response to glucose with age. The aim of this study was to investigate if these changes were reversible with glucagon-like peptide-1 (GLP-1), a peptide that we have previously shown could increase insulin mRNA and total insulin content in insulinoma cells. We infused 1.5 pmol/ kg-1.min-1 GLP-1 subcutaneously using ALZET microosmotic pumps into 22-mo-old Wistar rats for 48 h. Rat infused with either GLP-1 or saline were then subjected to an intraperitoneal glucose (1 g/kg body weight) tolerance test, 2 h after removing the pump. 15 min after the intraperitoneal glucose, GLP-1-treated animals had lower plasma glucose levels (9.04+/-0.92 mmol/liter, P < 0.01) than saline-treated animals (11.61+/-0.23 mmol/liter). At 30 min the plasma glucose was still lower in the GLP-1-treated animals (8.61+/-0.39 mmol/liter, P < 0.05) than saline-treated animals (10.36+/-0.43 mmol/liter). This decrease in glucose levels was reflected in the higher insulin levels attained in the GLP-1-treated animals (936+/-163 pmol/liter vs. 395+/-51 pmol/liter, GLP-1 vs. saline, respectively, P < 0.01), detected 15 min after glucose injection. GLP-1 treatment also increased pancreatic insulin, GLUT2, and glucokinase mRNA in the old rats. The effects of GLP-1 were abolished by simultaneous infusion of exendin [9-39], a specific antagonist of GLP-1. GLP-1 is therefore able to reverse some of the known defects that arise in the beta cell of the pancreas of Wistar rats, not only by increasing insulin secretion but also by inducing significant changes at the molecular level.

Overexpression and activation of the insulin receptor enhances expression of ERCC-1 messenger ribonucleic acid in cultured cells

In this study, a partial hamster complementary DNA encoding ERCC-1, a member of the DNA excision repair gene family, has been cloned. The nucleic acid and amino acid sequences were highly homologous to those of human and mouse ERCC-1. The hamster ERCC-1 gene was expressed as a 1.2-kilobase message in cultured Chinese hamster ovary cells. Northern (RNA) blot analysis revealed that overexpression of the insulin receptor or various growth factor receptor tyrosine kinases in Chinese hamster ovary cells increased ERCC-1 messenger RNA (mRNA) levels. This effect did not occur in cells overexpressing mutated insulin receptors that are known to have impaired kinase-related signaling. Increased ERCC-1 expression correlated with resistance to UV exposure. Fluorescent-activated cell sorter analysis of confluent cell populations indicated no differences in cell cycle distribution. Furthermore, no significant relationship was demonstrated between the relative expression of ERCC-1 mRNA and the rate of glucose utilization. Insulin enhanced the accumulation of ERCC-1 mRNA in serum-deprived cells expressing wild-type insulin receptors. The potential role for activation of the insulin receptor and related growth factor receptors in ERCC-1 gene expression and function remains to be defined.

Pancreatic reg gene expression is inhibited during cellular differentiation

BACKGROUND AND OBJECTIVE

Factors that control pancreatic regenerating (reg I) gene expression are unknown, but it is believed that its expression may correspond with cellular differentiation. The authors recently demonstrated that reg I is expressed in AR42J, a rat acinar cell line whose state of differentiation can be modulated by dexamethasone. They used this line to study reg I expression during cellular proliferation and differentiation.

METHODS

After treatment of cells with 10 nmol/L dexamethasone, proliferation was assayed by thymidine incorporation; differentiation by expression of elastase I mRNA. Reg I mRNA levels were measured using a rat reg I cDNA probe, and reg I protein levels assayed by enzyme-linked immunosorbent assay of cellular lysates with a polyclonal antibody. The effect of gastrin, cholecystokinin and glucagon on reg I expression was also studied.

RESULTS

When compared with controls, treatment with dexamethasone caused thymidine incorporation to decrease and elastase mRNA levels to increase. Reg I mRNA decreased from controls of 100 +/- 16% to 40 +/- 18% (p < 0.05), and reg I protein levels decreased as well. Gastrointestinal hormones had no significant effect on either elastase or reg I gene expression.

CONCLUSIONS

Expression of reg I inversely correlates with the level of cellular differentiation, can be modulated via the glucocorticoid receptor, and is a potential marker of gastrointestinal epithelial differentiation. Despite its presence within a pancreatic acinar cell line, reg I gene expression is not modulated by gastrointestinal hormones.

Differential expression of reg-I and reg-II genes during aging in the normal mouse

A cDNA termed reg (for regenerating gene) has been isolated from a rat pancreatic DNA library. Reg expression has been shown to correlate with changes in beta cell mass and function. This finding has been recently challenged by studies showing a non-beta-cell-dependent regulation of reg expression. All studies to date, however, have neglected the fact that two nonallelic reg genes (reg-I and reg-II) exist in several species. In studying the regulation of each individual copy gene, we investigated reg-I and -II gene expression in a naturally occurring modification of beta-cell physiology: normal aging. RNA was isolated from individual pancreata of 1-, 3-, 9-, 20-, and 30-month-old C57BL/6J mice (n > or = 3 per group) and subjected to slot-blot analysis using homologous probes for reg-I, reg-II, insulin, and elastase-I. A progressive age-dependent decrease in total reg mRNA levels (reg-I and -II) was detected. At 30 months of age, total reg mRNA levels were approximately 45% of the level detected in 1-month-old mice (p = .01). This paralleled the decrease in insulin mRNA levels (p = .01), which fell below 50%; by contrast, mRNA levels for elastase-I increased with age (p = .05). Analysis of RNA isolated from purified islets did not reveal any mRNA for reg, suggesting that in the normal mouse, reg is primarily a product of the exocrine pancreas. Reg mRNA were detectable in RNA extracts from stomach, duodenum, and small intestine. By hybridization of total pancreatic RNA with oligonucleotide probes which specifically recognize reg-I or reg-II sequences, we show that reg-I mRNA levels declined with age (p = .001) while reg-II mRNA levels remained unchanged. These data demonstrate that in mouse pancreas the two nonallelic reg genes are differentially expressed during aging and that the decrease in reg-I mRNA levels parallels the decrease in insulin gene expression. Differential regulation of reg-I and reg-II genes may explain the presence of conflicting data in the current literature.

Molecular investigation of age-related changes in mouse endocrine pancreas

Aging is an etiologic factor in non-insulin-dependent diabetes mellitus. While the effect of aging on insulin secretion has been described by several classic studies, the characterization of the molecular basis of beta-cell abnormalities is still under way. We recently demonstrated in rats that aging is associated not only with a reduction in insulin secretion but also with diminished levels of intracellular insulin content and the mRNA for insulin. In this study, we investigated whether the molecular abnormalities previously described in the rat beta cell were also present in the mouse (C57BL/6J). Total cellular RNA was isolated from individual pancreata of 3-, 9-, and 30-month-old mice (n = 6 per age group). Samples were subjected to slot-blot analysis by using homologous probes for insulin, glucagon, somatostatin, glucose transporter-2 (glut-2), glucokinase, elastase-I, and beta-actin. We observed a progressive age-dependent decrease in insulin mRNA levels: insulin mRNA levels decreased by 40% with age (p = .007). This paralleled decreases in glut-2 (p = .001) mRNA levels, but it was in contrast with glucokinase mRNA levels which increased markedly (p = .0003). Somatostatin mRNA levels were unchanged, glucagon mRNA levels decreased modesty (p = .01), and mRNA levels for elastase-I and beta-actin increased with age (p = .0001 for either one). In summary, it appears that in the mouse a progressive decline in the activity of the endocrine pancreas occurs with aging. This phenomenon seems to affect only the beta cells and not the alpha or delta cells of the islet of Langerhans or the exocrine pancreas. This progressive decline may represent the biological features of the age-dependent risk for the development of diabetes.

Regenerating (reg) and insulin genes are expressed in prepancreatic mouse embryos

The pancreatic regenerating (reg) gene is proposed to be involved in pancreatic beta-cell growth. Up- or down-regulation of reg gene expression has been shown to parallel variations in beta-cell mass and function in the adult pancreas. In several species at least two nonallelic reg genes have been identified. In this study we investigated the expression of each individual reg gene (reg-I and reg-II) during embryogenesis in the mouse. Single mouse embryos were harvested at 8.5, 9, 10, and 12 days of development, homogenized and subjected individually to reverse transcription (RT)-PCR, with a single primer pair to amplify both reg-I and -II mRNAs. Southern blot analysis of the RT-PCR products revealed the presence of reg mRNA at day 9 of embryogenesis, just before the beginning of pancreatic organogenesis. Slot-blot analysis with internal oligonucleotide probes that specifically recognize reg-I or -II sequences demonstrated that only reg-I mRNA was present in day 9 and day 10 prepancreatic embryos. Reg-II mRNA was not detected until day 12, a stage corresponding to late organogenesis. RT-PCR for insulin mRNA from the same samples used for the amplification of reg mRNA showed that the earliest insulin expression occurred at day 8.5, and coincided with the onset of reg-I expression. Hybridization with gene-specific oligonucleotide probes revealed that only insulin-II mRNA was detectable at this time. Insulin-I mRNA was not detectable until day 12 and coincided with early reg-II expression. These results suggest that the two nonallelic reg genes and the two insulin genes are expressed differentially during early embryogenesis. Differential expression of reg-I and -II suggests that they may be induced by different and independent stimuli and have distinct functions.

Insulin release and insulin mRNA levels in rat islets of Langerhans cultured on extracellular matrix

Primary culture of rat islets of Langerhans lose glucose responsiveness and eventually die when cultured for a long period of time. In this study we evaluated the effect of matrigel, a basement membrane extract, on (i) islet cell survival, (ii) cell responsiveness following a glucose challenge, and (iii) mRNA levels for insulin, glucagon, and somatostatin. Pancreatic islets were isolated by collagenase digestion and plated in culture dishes either coated or not with a matrigel layer. Using the reverse hemolytic plaque assay, we determined the total number of insulin-secreting cells and the amount of insulin secreted by individual beta cells. After 1 h of exposure to 5 mM glucose, beta cells from 6-month-old rat islets cultured for 6 weeks on matrigel showed an equal number of insulin-secreting cells compared to freshly isolated islets cultured for only 3 days in the absence of matrigel (39.5 +/- 2.5 vs. 37.1 +/- 2.6%). Furthermore, the release of insulin by cells cultured on matrigel for 6 weeks increased in a glucose-dependent manner (p < 0.001) and showed an ED50 of 7 mM. However, the amount of insulin released per single beta cell was reduced by 40-60% (p < 0.02) compared to that released from isolated beta cells derived from a 3-day culture of islets. Finally, there was a 35-55% increase (p < 0.05) in the levels of insulin, glucagon, and somatostatin mRNAs in cells cultured for 6 weeks on matrigel. These data suggest a trophic effect of matrigel on the maintenance of normal beta-cell activity and function and may lead the way to the development of a new model for the study of pancreatic islets in long-term culture.

Pancreatic regeneration (reg) gene expression in a rat model of islet hyperplasia

BACKGROUND

After pancreatectomy, regeneration of acinar and islet elements occurs. Recent data from a model of islet hyperplasia in the hamster suggested that induction of a local pancreatic factor stimulates islet formation. We postulate that the reg gene may be this factor.

METHODS

We studied reg expression during induction of islet growth by using a similar model in the rat. Rats underwent surgical wrapping of the splenic lobe of the pancreas or sham operation.

RESULTS

At 2 days ductular proliferation and immunohistochemical evidence of insulin within ductular epithelia were evident in the wrapped lobe. By 14 and 56 days islet number per square millimeter increased 63% and 43%, respectively. Reg mRNA levels, measured by Northern blot analysis with a rat reg cDNA probe (n = 5), increased 300% at 2 days in the wrapped lobe and decreased to that of unwrapped controls by 14 days. In situ hybridization showed localization of reg to the acinar cells. In unwrapped gastric lobes of animals who underwent surgical wrapping of the splenic lobe, no change in islet number per square millimeter or reg gene expression was noted.

CONCLUSIONS

Surgical wrapping of the pancreatic splenic lobe induces local reg gene expression that is temporally associated with duct cell hyperplasia. This is followed by islet formation within the wrapped lobe. Reg may play a role in the induction of new islets from ductular precursors and in maintenance of normal islet function.

Pancreatic thread protein is mitogenic to pancreatic-derived cells in culture

BACKGROUND & AIMS

Pancreatic thread proteins (PTPs) are acinar cell products and members of the regenerating gene (reg) family. reg expression increases during islet regeneration, is depressed during aging-related islet dysfunction, and may be important in beta-cell growth and maintenance. The aim of this study was to examine the genetic expression of reg in pancreatic-derived cells in vitro and the mitogenic effect of PTP/reg protein on these cells.

METHODS

reg gene expression was measured by Northern analysis in three rat pancreatic cell lines: ARIP (ductal), AR42J (acinar), and RIN (beta-cell). PTP/reg protein was isolated from bovine and human pancreas. Cells were cultured with PTP/reg for 72 hours, and thymidine incorporation was measured.

RESULTS

reg messenger RNA was detected in AR42J but not in ARIP or RIN. PTP/reg protein was mitogenic to RIN and ARIP in a dose-related fashion but not to AR42J. It was not mitogenic to cultured mature rat islets.

CONCLUSIONS

reg messenger RNA is expressed in acinar but not in beta-cell or ductal pancreatic cell lines. PTP/reg protein was mitogenic to both beta-cell and ductal cell lines but not to mature, nondividing islets. This supports the hypothesis that PTP/reg protein is an acinar cell-derived mediator of beta-cell growth and may be involved in modulating the duct-to-islet axis.

Age-dependent reduction in insulin secretion and insulin mRNA in isolated islets from rats

Aging is an etiologic factor in non-insulin-dependent diabetes mellitus. To characterize the beta-cell abnormalities that occur with age, we investigated glucose-stimulated insulin release, pancreatic insulin content, and mRNA levels for islet-specific genes in aging Wistar rats. Ten minutes after glucose stimulation, 6-mo-old islets had approximately 40% more cells secreting insulin than 24-mo-old islets (P < 0.0001); after 1 h, 67 +/- 1.0% islets from 6-mo-old rats secreted insulin, compared with 51 +/- 3.5% from 24-mo-old rats (P < 0.0001). The amount of insulin secreted by each beta-cell was also less in the older animals (P < 0.0001). Despite increases in islet size (P < 0.0001) and beta-cell number (P < 0.0001) with age, whole pancreas insulin content showed that 24-mo-old pancreas had less insulin than 6-mo-old pancreas (0.61 +/- 0.06 vs. 0.84 +/- 0.08 microgram/mg pancreatic protein; P < 0.05). Finally, insulin mRNA levels declined to 50% of the newborn value in 24-mo-old islets (P < 0.0001), whereas glucagon mRNA levels showed a very modest decline with age. Somatostatin mRNA levels did not vary significantly. In summary, it appears that in Wistar rats there is a progressive decline in beta-cell activity with age. This decline may represent the biological features of the age-dependent risk of developing diabetes.

Optimization neural network for solving flow problems

This paper describes a neural network for solving flow problems, which are of interest in many areas of application as in fuel, hydro, and electric power scheduling. The neural network consist of two layers: a hidden layer and an output layer. The hidden units correspond to the nodes of the flow graph. The output units represent the branch variables. The network has a linear order of complexity, it is easily programmable, and it is suited for analog very large scale integration (VLSI) realization. The functionality of the proposed network is illustrated by a simulation example concerning the maximal flow problem.

A synthesis procedure for brain-state-in-a-box neural networks

In this paper, some new qualitative properties of discrete-time neural networks based on the "brain-state-in-a-box" model are presented. These properties concern both the characterization of equilibrium points and the global dynamical behavior. Next, the analysis results are used as guidelines in developing an efficient synthesis procedure for networks that function as associative memories. A constrained design algorithm is presented that gives completely stable dynamical neural networks sharing some interesting features. It is guaranteed the absence of nonbinary stable equilibria, that is stable states with nonsaturated components. It is guaranteed that in close proximity (Hamming distance one) of the stored patterns there is no other binary equilibrium point. Moreover, the presented method allows one to optimize a design parameter that controls the size of the attraction basins of the stored vectors and the accuracy needed in a digital realization of the network.

Sensitivity of equilibrium points in continuous-time Hopfield's network

The sensitivity of continuous-time Hopfield neural network is investigated. The relative sensitivity of hyperbolic equilibrium points, with respect to changes in the interconnections, is computed. Then it is shown that the minimum sensitivity of equilibria corresponds to the minimum scattering of weights about their mean value. This permits a selection among the many available synthesis methods for associative memories.

The two nonallelic insulin-like growth factor-I genes in Xenopus laevis are differentially regulated during development

To study the potential role of insulin-like growth factor-I (IGF-I) during early embryogenesis, we have used the amphibian Xenopus laevis, a versatile model of vertebrate development. Using polymerase chain reaction (PCR)-based cloning strategies, we have previously identified two different nonallelic Xenopus IGF-I genes (IGF-I' and IGF-I"). Both are expressed in similar quantities in adult liver. We now report the use of a modification of the reverse transcription-PCR method, designated RNA template-specific PCR (RS-PCR), to detect IGF-I messenger RNA (mRNA) from single oocytes and embryos. The same primer pair was used to amplify both IGF-I' and IGF-I" mRNAs. Slot blot analysis of the RS-PCR products with internal oligonucleotide probes that specifically recognize IGF-I' or IGF-I" sequences revealed that only IGF-I' mRNA was present in follicles surrounding mature (stage VI) oocytes; neither IGF-I mRNA was present in follicles surrounding less mature oocytes (stages I and IV) or within oocytes or unfertilized eggs. After fertilization, IGF-I' mRNA was first detected during early organogenesis (stages 21-23), and increased during subsequent stages of development. To localize early IGF-I' expression, a stage 27 embryo was sliced into 24-microns cross-sections; RS-PCR and slot blot analysis were performed on RNA extracts from consecutive sections. IGF-I' mRNA was expressed in all sections, but was most abundant in the body region from which the visceral organs were developing. In contrast, to the early expression of IGF-I', IGF-I" mRNA was not detected until stage 41, a period corresponding to premetamorphic growth. Reverse PCR for Xenopus GH mRNA demonstrated that the onset of GH gene expression was coincident with the onset of IGF-I" gene expression (stage 41). These data suggest that the early expression of IGF-I' may be GH independent, whereas later expression of IGF-I" is GH dependent. We conclude that the two nonallelic IGF-I genes are expressed differentially in a stage-specific manner and suggest that IGF-I' may play an important role during early organogenesis, whereas IGF-I" may be important for premetamorphic growth.

Assignment of the human pancreatic regenerating (REG) gene to chromosome 2p12

A cDNA termed reg (for regenerating gene) has been isolated and characterized from a rat pancreatic library. Expression of reg is markedly increased in regenerating islets and decreased when insulin gene expression is inhibited. These findings have led to the hypothesis that reg may be involved in the expansion of beta-cell mass during regeneration as well as in the maintenance of normal beta-cell function. The human reg gene has a high degree of similarity to the rat reg gene. To determine the chromosomal location of the human reg gene, we analyzed two panels of mouse- or hamster-human hybrid cell lines containing a single human chromosome or several different human chromosomes. DNA extracts from these cell lines were analyzed for the presence of the human reg gene by polymerase chain reaction. In addition, human metaphase chromosomes were used for fluorescence in situ hybridization to further confirm the chromosomal assignment and to determine the subchromosomal localization. With these approaches, we show that the human reg gene is located on the short arm of chromosome 2 near the centromere at band 2p12.

RNA Template-Specific Polymerase Chain Reaction (RS-PCR) : A Modification of RNA-PCR that Dramatically Reduces the Frequency of False Positives

Reverse transcription of RNA followed by the polymerase chain reaction (RT-PCR or RNA-PCR) is an extraordinarily sensitive method to detect as few as 1-100 copies of a specific RNA (1-3). However, we and others have found that false positives caused by contamination with minute quantities of DNA (i.e., cDNAs, plasmid DNAs, genomic DNA, or PCR carryover) is a major shortcoming of the method even when meticulous laboratory technique is employed (4-6). RNA template-specific PCR (RS-PCR) is a modification of conventional RNA-PCR in which RNA is reverse-transcribed with a primer that contains at its 5' end a unique nucleotide sequence that may then be exploited in the PCR to amplify preferentially the RNA-derived sequence. RS-PCR retains full sensitivity, but reduces dramatically the frequency of false positives (7-9).

Existence of binary invariant sets in feedback neural networks with application to synthesis

The design of fully connected discrete-time neural networks, with threshold units, can be performed by solving a set of linear inequalities. Using this formulation, the design problem can be handled by several powerful algorithms. This approach is extended to the design of continuous-time neural networks with sigmoidal units, which represent a more realistic model of analog circuit implementations. The proposed method is based on some theoretical results proved in this paper.

Diabetes-related autoantibodies do appear in children with coeliac disease

Humoral immune factors related to type 1 diabetes have been investigated in children with coeliac disease. Anti-insulin (IAAb), immunoglobulin (alpha IgAb), islet cell (ICA) and glucagon autoantibodies were examined in 15 children with coeliac disease at diagnosis (group 1), in 15 children with coeliac disease following a gluten-free diet (group 2) and in 30 control patients (groups 3 and 4). IAAb were present in 27% of group 1 and in 20% of group 2 patients and alpha IgAb were significantly increased in group 1 and 2 patients; two patients in group 2 were positive for ICA; none of the coeliac disease patients were positive for anti-glucagon antibodies. The levels of anti-gliadin antibodies in group 1 were positively correlated with those of alpha IgAb. Coeliac disease-related HLA antigens were not correlated with antibody presence. The presence of diabetes-related humoral immune factors in coeliac disease raises the question as to whether or not they are predictive of subclinical pancreatic damage or whether they are simply indicators of a more general autoimmune diathesis.

Autoantibodies to insulin do appear in non-diabetic patients with autoimmune disorders: comparison with anti-immunoglobulin antibodies and other autoimmune phenomena

Insulin- and anti-immunoglobulin-antibodies have been recently reported in pre-diabetic subjects: the former has been proposed as a predictive marker of Type I diabetes in non-diabetic-subjects. To evaluate the diabetes-related specificity of these antibodies, the presence of insulin autoantibodies, using a recently developed and highly sensitive competitive radioimmune assay, and of anti-immunoglobulin antibodies together with that of immune complexes and of other autoantibodies has been investigated in patients with organ- or non-organ-specific autoimmune diseases. One hundred and eleven serum samples were assayed from patients with Graves' disease, primary hypothyroidism, chronic autoimmune thyroiditis, Addison's disease, chronic autoimmune hepatitis, pernicious anemia, lupus erythematosus, and rheumatoid arthritis, together with 45 serum samples from normal subjects. From patients with autoimmune diseases, 32.4% of all sera revealed values of insulin autoantibodies above the limit of positivity (p less than 0.001); anti-immunoglobulin antibodies were present in 4.1% of patients (NS); immune complexes were found in 19.5% (NS) of all patients, but in 38% of patients with Graves' disease and chronic hepatitis (p less than 0.02). There was a trend for multiple autoantibody positivity to be associated with high levels of insulin autoantibodies (p less than 0.05). Thus, whereas contrary to expectation anti-immunoglobulin antibodies are not associated with non-diabetes-related autoimmune diseases, increased humoral immunoresponsiveness to endogenous insulin appears to be related to autoimmunity in general rather than restricted to Type I diabetes.

Humoral and cellular immunological factors as possible markers of clinical relapse in HLA-typed Graves' patients followed with time

Humoral and cellular immune factors were studied in 33 newly diagnosed Graves' patients at diagnosis and in 12 of these patients at regular intervals thereafter. All the patients were treated with carbimazole for 15 months (initially 60 mg and then 20 mg supplemented with L-Thyroxine). The incidence of relapse after treatment was 42%. Thyrotropin receptor antibodies (TRAb), T-cell subsets, K and NK cells and mononuclear cells expressing surface antigen markers of different activation were evaluated respectively by the use of a radioimmunoassay and a panel of monoclonal antibodies. Patients in the follow-up study were HLA-A, B, C and D typed. TRAb levels (91%) and levels of 4F2-positive cells (73%) and class II-positive lymphocytes (69.6%) were significantly increased in newly diagnosed Graves' patients in comparison with normal controls, whereas CD8 cells were significantly decreased. There was a significant inverse correlation between the increase in 4F2-positive cells and TRAb values. In the follow-up study both humoral and cellular immunological parameters showed a wide variation in levels, but TRAb, 4F2 and L243 values declined on average with respect to diagnosis. After 15 months some patients still showed abnormal values of activated T cells and TRAb values. All patients who relapsed (42%) after medical treatment showed a significant increase of 4F2-positive cells, and some of TRAb, some time before the appearance of clinical symptoms. Finally, no correlation was found between HLA type and relapse of the disease.(ABSTRACT TRUNCATED AT 250 WORDS)

Quantitative immunological differences between newly diagnosed Graves' disease patients and relapsed patients

Class 2-positive T cells, T-cell and mononuclear cell subsets, thyrotropin receptor antibodies (TRAb) and immune complexes were evaluated in 34 newly diagnosed Graves' patients and in 13 relapsed patients before a cycle of specific medical treatment. Class II-positive T lymphocytes were detected by monoclonal antibodies against different epitopes of class II antigens, whereas 4F2-positive cells were detected by 4F2 monoclonal antibody. 4F2-positive cells were statistically increased in newly diagnosed Graves' patients compared to relapsed patients (p less than 0.05). An increased percentage of class II activated T cells, detected by monoclonal antibodies L243, was found in newly diagnosed patients in comparison with relapsed subjects (p less than 0.025). Newly diagnosed Graves' patients showed a significant decrease in the ratio suppressor/cytotoxic T cells in comparison with normal control subjects but not with relapsed patients. Ninety-one % of newly diagnosed Graves' patients showed a high TRAb value, whereas only 69% of relapsed patients showed increased values (p less than 0.025). No difference was observed in the immune complex positivity between newly diagnosed Graves' patients and relapsed subjects. In conclusion, both humoral and cellular immune differences were found in relapsed patients vs newly diagnosed Graves' patients. The immunological abnormalities are quantitatively more pronounced in the latter group.