scholarly journals Bayesian metamodeling of complex biological systems across varying representations

2021 ◽  
Author(s):  
Barak Raveh ◽  
Liping Sun ◽  
Kate L White ◽  
Tanmoy Sanyal ◽  
Jeremy Tempkin ◽  
...  

Comprehensive modeling of a whole cell requires an integration of vast amounts of information on various aspects of the cell and its parts. To divide-and-conquer this task, we introduce Bayesian metamodeling, a general approach to modeling complex systems by integrating a collection of heterogeneous input models. Each input model can in principle be based on any type of data and can describe a different aspect of the modeled system using any mathematical representation, scale, and level of granularity. These input models are (i) converted to a standardized statistical representation relying on Probabilistic Graphical Models, (ii) coupled by modeling their mutual relations with the physical world, and (iii) finally harmonized with respect to each other. To illustrate Bayesian metamodeling, we provide a proof-of-principle metamodel of glucose-stimulated insulin secretion by human pancreatic β-cells. The input models include a coarse-grained spatiotemporal simulation of insulin vesicle trafficking, docking, and exocytosis; a molecular network model of glucose-stimulated insulin secretion signaling; a network model of insulin metabolism; a structural model of glucagon-like peptide-1 receptor activation; a linear model of a pancreatic cell population; and ordinary differential equations for systemic postprandial insulin response. Metamodeling benefits from decentralized computing, while often producing a more accurate, precise, and complete model that contextualizes input models as well as resolves conflicting information. We anticipate Bayesian metamodeling will facilitate collaborative science by providing a framework for sharing expertise, resources, data, and models, as exemplified by the Pancreatic β-Cell Consortium.

2021 ◽  
Vol 118 (35) ◽  
pp. e2104559118 ◽  
Author(s):  
Barak Raveh ◽  
Liping Sun ◽  
Kate L. White ◽  
Tanmoy Sanyal ◽  
Jeremy Tempkin ◽  
...  

Comprehensive modeling of a whole cell requires an integration of vast amounts of information on various aspects of the cell and its parts. To divide and conquer this task, we introduce Bayesian metamodeling, a general approach to modeling complex systems by integrating a collection of heterogeneous input models. Each input model can in principle be based on any type of data and can describe a different aspect of the modeled system using any mathematical representation, scale, and level of granularity. These input models are 1) converted to a standardized statistical representation relying on probabilistic graphical models, 2) coupled by modeling their mutual relations with the physical world, and 3) finally harmonized with respect to each other. To illustrate Bayesian metamodeling, we provide a proof-of-principle metamodel of glucose-stimulated insulin secretion by human pancreatic β-cells. The input models include a coarse-grained spatiotemporal simulation of insulin vesicle trafficking, docking, and exocytosis; a molecular network model of glucose-stimulated insulin secretion signaling; a network model of insulin metabolism; a structural model of glucagon-like peptide-1 receptor activation; a linear model of a pancreatic cell population; and ordinary differential equations for systemic postprandial insulin response. Metamodeling benefits from decentralized computing, while often producing a more accurate, precise, and complete model that contextualizes input models as well as resolves conflicting information. We anticipate Bayesian metamodeling will facilitate collaborative science by providing a framework for sharing expertise, resources, data, and models, as exemplified by the Pancreatic β-Cell Consortium.


2015 ◽  
Vol 227 (3) ◽  
pp. 153-165 ◽  
Author(s):  
Saeed Alshahrani ◽  
Mohammed Mashari Almutairi ◽  
Shams Kursan ◽  
Eduardo Dias-Junior ◽  
Mohamed Mahmoud Almiahuob ◽  
...  

The products of theSlc12a1andSlc12a2genes, commonly known as Na+-dependent K+2Cl−co-transporters NKCC2 and NKCC1, respectively, are the targets for the diuretic bumetanide. NKCCs are implicated in the regulation of intracellular chloride concentration ([Cl−]i) in pancreatic β-cells, and as such, they may play a role in glucose-stimulated plasma membrane depolarization and insulin secretion. Unexpectedly, permanent elimination of NKCC1 does not preclude insulin secretion, an event potentially linked to the homeostatic regulation of additional Cl−transporters expressed in β-cells. In this report we provide evidence for such a mechanism. Mice lacking a single allele ofSlc12a2exhibit lower fasting glycemia, increased acute insulin response (AIR) and lower blood glucose levels 15–30 min after a glucose load when compared to mice harboring both alleles of the gene. Furthermore, heterozygous expression or complete absence ofSlc12a2associates with increased NKCC2 protein expression in rodent pancreatic β-cells. This has been confirmed by using chronic pharmacological down-regulation of NKCC1 with bumetanide in the mouse MIN6 β-cell line or permanent molecular silencing of NKCC1 in COS7 cells, which results in increased NKCC2 expression. Furthermore, MIN6 cells chronically pretreated with bumetanide exhibit increased initial rates of Cl−uptake while preserving glucose-stimulated insulin secretion. Together, our results suggest that NKCCs are involved in insulin secretion and that a singleSlc12a2allele may protect β-cells from failure due to increased homeostatic expression ofSlc12a1.


1990 ◽  
Vol 69 (1) ◽  
pp. 261-267 ◽  
Author(s):  
F. Dela ◽  
K. J. Mikines ◽  
B. Tronier ◽  
H. Galbo

Glucose-stimulated insulin secretion is depressed by training. To further elucidate the beta-cell adaptation to training, a nonglucose secretagogue was applied. Arginine was infused for 90 min to seven trained and seven untrained young men. Arginine and glucose concentrations increased identically in the groups. The insulin response was biphasic and waned despite increasing arginine concentrations. Both these phases as well as C-peptide responses were reduced in trained subjects, whereas proinsulin responses were similar in the groups. Identical increases were found in glucagon, growth hormone, catecholamines, and production and disappearance of glucose; identical decreases were found in free fatty acids, glycerol, and beta-hydroxybutyrate. In conclusion, in men training diminishes both arginine- and glucose-stimulated insulin secretion, indicating a profound beta-cell adaptation. Being enhanced, the effects of insulin on both production and disposal of glucose are changed in the opposite direction to beta-cell secretion by training. The responses of glucagon- and growth hormone-secreting cells to arginine do not change with training.


1996 ◽  
Vol 134 (3) ◽  
pp. 362-365 ◽  
Author(s):  
Bo Ahrén ◽  
Hillevi Larsson

Ahrén B, Larsson H. Peptide YY does not inhibit glucose-stimulated insulin secretion in humans. Eur J Endocrinol 1996;134:362–5. ISSN 0804–4643 Peptide YY (PYY) is localized to gut and pancreatic endocrine cells. It may therefore be involved in the regulation of insulin secretion as a gut hormone as well as an islet local regulator. In laboratory animals, the peptide inhibits stimulated insulin secretion, but its effects in humans are not known. We therefore infused PYY intravenously at a low (1 pmol·kg−1 · min−1; N = 4) or a high rate (5 pmol · kg−1 · min−1; N= 5) for 120 min in healthy women aged 52 years. After 30 min of infusion, glucose (0.5 g/kg) was injected rapidly. In separate control experiments, saline was infused instead of PYY. We found that PYY did not inhibit the acute insulin response to glucose or affect the glucose elimination rate, and PYY was also without influence on the basal plasma glucose and serum insulin levels during the 30-min infusion before the challenge with glucose. We therefore conclude that intravenous infusion of PYY does not affect glucose-stimulated insulin secretion in man. Bo Ahrén, Department of Medicine, Malmö University Hospital, S-205 02 Malmö, Sweden


2003 ◽  
Vol 284 (1) ◽  
pp. E129-E137 ◽  
Author(s):  
Elisabeth Qvigstad ◽  
Ingrid L. Mostad ◽  
Kristian S. Bjerve ◽  
Valdemar E. Grill

We tested the effects of acute perturbations of elevated fatty acids (FA) on insulin secretion in type 2 diabetes. Twenty-one type 2 diabetes subjects with hypertriglyceridemia (triacylglycerol >2.2 mmol/l) and 10 age-matched nondiabetic subjects participated. Glucose-stimulated insulin secretion was monitored during hyperglycemic clamps for 120 min. An infusion of Intralipid and heparin was added during minutes 60–120. In one of two tests, the subjects ingested 250 mg of Acipimox 60 min before the hyperglycemic clamp. A third test (also with Acipimox) was performed in 17 of the diabetic subjects after 3 days of a low-fat diet. Acipimox lowered FA levels and enhanced insulin sensitivity in nondiabetic and diabetic subjects alike. Acipimox administration failed to affect insulin secretion rates in nondiabetic subjects and in the group of diabetic subjects as a whole. However, in the diabetic subjects, Acipimox increased integrated insulin secretion rates during minutes 60–120 in the 50% having the lowest levels of hemoglobin A1c (379 ± 34 vs. 326 ± 30 pmol · kg−1 · min−1without Acipimox, P < 0.05). A 3-day dietary intervention diminished energy from fat from 39 to 23% without affecting FA levels and without improving the insulin response during clamps. Elevated FA levels may tonically inhibit stimulated insulin secretion in a subset of type 2 diabetic subjects.


2001 ◽  
Vol 281 (1) ◽  
pp. E171-E179 ◽  
Author(s):  
Albert Salehi ◽  
Bo-Guang Fan ◽  
Mats Ekelund ◽  
Gunnar Nordin ◽  
Ingmar Lundquist

We examined the relation between nutrient-stimulated insulin secretion and the islet lysosome acid glucan-1,4-α-glucosidase system in rats undergoing total parenteral nutrition (TPN). During TPN treatment, serum glucose was normal, but free fatty acids, triglycerides, and cholesterol were elevated. Islets from TPN-infused rats showed increased basal insulin release, a normal insulin response to cholinergic stimulation but a greatly impaired response when stimulated by glucose or α-ketoisocaproic acid. This impairment of glucose-stimulated insulin release was only slightly ameliorated by the carnitine palmitoyltransferase 1 inhibitor etomoxir. However, in parallel with the impaired insulin response to glucose, islets from TPN-infused animals displayed reduced activities of islet lysosomal enzymes including the acid glucan-1,4-α-glucosidase, a putative key enzyme in nutrient-stimulated insulin release. By comparison, the same lysosomal enzymes were increased in liver tissue. Furthermore, in intact control islets, the pseudotetrasaccharide acarbose, a selective inhibitor of acid α-glucosidehydrolases, dose dependently suppressed islet acid glucan-1,4-α-glucosidase and acid α-glucosidase activities in parallel with an inhibitory action on glucose-stimulated insulin secretion. By contrast, when incubated with intact TPN islets, acarbose had no effect on either enzyme activity or glucose-induced insulin release. Moreover, when acarbose was added directly to TPN islet homogenates, the dose-response effect on the catalytic activity of the acid α-glucosidehydrolases was shifted to the right compared with control homogenates. We suggest that a general dysfunction of the islet lysosomal/vacuolar system and reduced catalytic activities of acid glucan-1,4-α-glucosidase and acid α-glucosidase may be important defects behind the impairment of the transduction mechanisms for nutrient-stimulated insulin release in islets from TPN-infused rats.


1989 ◽  
Vol 257 (1) ◽  
pp. E43-E48 ◽  
Author(s):  
K. J. Mikines ◽  
F. Dela ◽  
B. Tronier ◽  
H. Galbo

Physical training decreases glucose-stimulated insulin secretion. To further explore the influence of the level of daily physical activity on beta-cell secretion, the effect of 7 days of bed rest was studied in six young, healthy men by sequential hyperglycemic clamp technique (7, 11, and 20 mM glucose, each step lasting 90 min). At 11 and 20 mM glucose, insulin concentrations in plasma were higher after (87 +/- 11 and 303 +/- 63 microU/ml) than before (63 +/- 5 and 251 +/- 50 microU/ml, P less than 0.05) bed rest. Also C-peptide levels were higher after bed rest than before during glucose stimulation. The responses of other hormones, metabolites, or electrolytes influencing beta-cell secretion were not influenced by bed rest. In spite of increased insulin levels after bed rest, glucose disposal at 20 mM of glucose was significantly lower after bed rest than before. It is concluded that bed rest for 7 days increases the glucose-stimulated insulin response, at least partly due to a beta-cell adaptation increasing glucose-stimulated insulin secretion. However, the insulin secretion does not increase adequately compared with the peripheral insulin resistance induced by bed rest.


2021 ◽  
Author(s):  
Stephanie A. Campbell ◽  
Jocelyn Bégin ◽  
Cassandra L. McDonald ◽  
Ben Vanderkruk ◽  
Tabea L. Stephan ◽  
...  

During pancreas development, endocrine progenitors differentiate into the islet-cell subtypes, which undergo further functional maturation in postnatal islet development. In islet b-cells, genes involved in glucose-stimulated insulin secretion are activated and glucose exposure increases the insulin response as b-cells mature. Here, we investigated the role of H3K4 trimethylation in endocrine cell differentiation and functional maturation by disrupting TrxG complex histone methyltransferase activity in mouse endocrine progenitors. In the embryo, genetic inactivation of TrxG component <i>Dpy30</i> in NEUROG3+ cells did not affect the number of endocrine progenitors or endocrine cell differentiation. H3K4 trimethylation was progressively lost in postnatal islets and the mice displayed elevated non-fasting and fasting glycemia, as well as impaired glucose tolerance by postnatal day 24. Although postnatal endocrine cell proportions were equivalent to controls, islet RNA-sequencing revealed a downregulation of genes involved in glucose-stimulated insulin secretion and an upregulation of immature b-cell genes. Comparison of histone modification enrichment profiles in NEUROG3+ endocrine progenitors and mature islets suggested that genes downregulated by loss of H3K4 trimethylation more frequently acquire active histone modifications during maturation. Taken together, these findings suggest that H3K4 trimethylation is required for the activation of genes involved in the functional maturation of pancreatic islet endocrine cells.


1999 ◽  
Vol 277 (1) ◽  
pp. E93-E102 ◽  
Author(s):  
Bo Ahrén ◽  
Per Sauerberg ◽  
Christian Thomsen

Increased insulinotropic activity by the cholinergic agonist carbachol exists in insulin-resistant high fat-fed C57BL/6J mice. We examined the efficiency and potency of carbachol to potentiate glucose-stimulated insulin secretion and to improve glucose tolerance in these animals. Intravenous administration of carbachol (at 15 and 50 nmol/kg) markedly potentiated glucose (1 g/kg)-stimulated insulin secretion in mice fed both a control and a high-fat diet (for 12 wk), with a higher relative potentiation in high fat-fed mice measured as increased (1–5 min) acute insulin response and area under the 50-min insulin curve. Concomitantly, glucose tolerance was improved by carbachol. In fact, carbachol normalized glucose-stimulated insulin secretion and glucose tolerance in mice subjected to a high-fat diet. Carbachol (>100 nmol/l) also potentiated glucose-stimulated insulin secretion from isolated islets with higher efficiency in high fat-fed mice. In contrast, binding of the muscarinic receptor antagonist [ N- methyl-3H]scopolamine to islet muscarinic receptors and the contractile action of carbachol on ileum muscle strips were not different between the two groups. We conclude that carbachol normalizes glucose tolerance in insulin resistance.


Sign in / Sign up

Export Citation Format

Share Document