This chapter summarizes the findings of metabolic programming in the "pup-in-cup" rat model and small litter model and explores the possible reversal of the metabolic effects by nutritional interventions that vary the degree of calorie reduction in the postweaning period. Since the development of many organs and biological systems in mammalian newborns continues in the immediate postnatal period, these systems could be subject to plasticity due to any nutritional interventions imposed during infancy only. The contribution of developmental programming due to nutritional experience during the early phases of life to long-term predisposition for obesity and associated health complications in adulthood is now supported by compelling evidence from epidemiological observations, as well as some human studies and extensive animal investigations. The findings from numerous animal models involving different types of nutritional modifications during the suckling period may have relevance to human infant feeding practices resulting in overnutrition and other dietary modifications.

The pancreas is an exocrine and an endocrine organ. The acinar cells secrete proteolytic, lipolytic and amylytic enzymes in the pancreatic juice that flows through a ductal network to the duodenum. These enzymes are important for digestion. The pancreatic islets of Langerhans in the pancreas secrete hormones into the blood that are essential for maintenance of normal blood glucose levels or euglycemia. Diabetes mellitus is a disease of the endocrine pancreas. Type 1 diabetes is characterized as an absolute lack of insulin in the B-cells of the pancreatic islet. Insulin must be replaced for optimal control of blood glucose levels. Type 2 diabetes is characterized by a relative insufficiency of insulin secretion and peripheral insulin resistance. Oral hypoglycemic agents, injectable incretins and insulin are used to control blood glucose levels in type 2 diabetes. Oral hypoglycemic agents can target insulin secretion, hepatic glucose release, kidney glucose transporters, and glucose uptake and metabolism in insulin target tissues. Carbohydrate metabolism in the GI tract is also a target for reducing prandial glucose levels in blood. Maintaining euglycemia often requires combination pharmacologic therapy.

To evaluate the role of sphingosine kinase 1 (SphK1) in insulin secretion, we used stable transfection to knock down the expression of the Sphk1 gene in the rat insulinoma INS-1 832/13 cell line. Cell lines with lowered Sphk1 mRNA expression and SphK1 enzyme activity (SK11 and SK14) exhibited lowered glucose- and 2-aminobicyclo[2,2,1]heptane-2-carboxylic acid (BCH) plus glutamine-stimulated insulin release and low insulin content associated with decreases in the mRNA of the insulin 1 gene. Overexpression of the rat or human Sphk1 cDNA restored insulin secretion and total insulin content in the SK11 cell line, but not in the SK14 cell line. The Sphk1 cDNA-transfected SK14 cell line expressed significantly less SphK1 activity than the Sphk1 cDNA-transfected SK11 cells suggesting that the shRNA targeting SK14 was more effective in silencing the exogenous rat Sphk1 mRNA. The results indicate that SphK1 activity is important for insulin synthesis and secretion.

The metabolic programming effects of nutritional modifications in the immediate postnatal life are increasingly recognized to independently contribute to the development of metabolic syndrome in later life. Adjustment of litter size in rodents has been used to induce either under- or overnourishment in the immediate postnatal life of the offspring. While undernourishment led to growth retardation in the offspring, overnourishment produced increased body weight gains, hyperinsulinemia and hyperleptinemia. Overnourishment during the suckling period induced several adaptations in the energy circuitry in the hypothalamus of the offspring predisposing them for the onset of obesity later in life. Another approach for a nutritional modification in the immediate postnatal period is the artificial rearing of newborn rat pups on a high-carbohydrate (HC) milk formula without changes in the total calorie availability. Hyperinsulinemia, immediately evident in the HC pups, persisted in the post-weaning period even after withdrawal of the HC milk. Significant alterations in pancreatic islets supported chronic hyperinsulinemia in the HC rats. Alterations in the gene expression of hypothalamic neuropeptides predisposing to hyperphagia were evident during the period of the HC dietary modification. The persistence of these hypothalamic adaptations supported the obese phenotype in adult HC rats. A transgenerational effect gave rise to the development of chronic hyperinsulinemia and adult-onset obesity in the offspring of the HC female rats. Other studies have shown that lactation by a diabetic, obese or malnourished mother resulted in predisposition for the onset of metabolic disorders in the offspring. These observations from animal studies on the metabolic programming effects due to altered nutritional experiences in the immediate postnatal life strongly suggest that altered feeding practices for infants (formula feeding and early introduction of infant foods) could contribute to the rising incidence of overweight/obesity in children and adults.

Atrial natriuretic peptide (ANP) levels correlate with hyperglycemia in diabetes mellitus, but ANP effects on pancreatic islet β-cell insulin secretion are controversial. ANP was investigated for short- and long-term effects on insulin secretion and mechanisms regulating secretion in isolated rat pancreatic islets. A 3-h incubation with ANP did not affect basal or glucose-stimulated islet insulin secretion. However, 7-day culture of islets with 5.5 mM glucose and ANP (1 nM - 1 μM) markedly inhibited subsequent glucose (11 mM)-stimulated insulin secretion; total islet insulin content was not affected. Following ANP removal for 24 h, the islet insulin-secretory response to glucose was restored. The insulin-secretory response to other insulin secretagogues, including α-ketoisocaproic acid, forskolin, potassium chloride, and ionomycin were also markedly inhibited by chronic exposure to ANP. However, the combination of potassium chloride and α-ketoisocaproic acid was sufficient to overcome the inhibitory effects of ANP on insulin secretion. The glucose-stimulated increases in islet ATP levels and the ATP/ADP ratio were completely inhibited in ANP 7-day-treated islets vs. control; removal of ANP for 24 h partially restored the glucose response. ANP did not affect islet glycolysis. ANP significantly increased levels of islet activated hormone-sensitive lipase and the expression of uncoupling protein-2 and peroxisome proliferator-activated receptor-δ and -α. Although islet ANP-binding natriuretic peptide receptor-A levels were reduced to 60% of control after 7-day culture with ANP, the ANP-stimulated cGMP levels remained similar to control islet levels. Thus, long-term exposure to ANP inhibits glucose-stimulated insulin secretion and ATP generation in isolated islets.

A link has been made between ANP and diabetes mellitus since hyperglycemia and diabetes are associated with elevated ANP levels. Natriuretic peptide receptor (NPR)‐A and NPR‐B mRNAs were detected in insulin secreting INS‐1E insulinoma cells and isolated rat pancreatic islets, but only NPR‐A protein was expressed on islet α‐and β‐cells. Culture of INS‐1E cells and islets with ANP (0.01–1 μM) increased cell proliferation, DNA biosynthesis and cyclin D2 mRNA levels. An analogue of cGMP, the second messenger of NPR‐A, also elevated DNA biosynthesis and cyclin D2 mRNA levels. Enhanced DNA biosynthesis and cyclin D2 mRNA induced by ANP were inhibited by LY294002, an inhibitor of PI3K. ANP and 8‐bromo‐cGMP also stimulated the phosphorylation of Akt (387±10% of control (P<0.05)) and its downstream target forkhead box O1a (FoxO1a) (175±11% of control (P<0.05)), a transcriptional repressor of cyclin D2; LY294002 inhibited these responses. The transcription of pancreas duodenum homeobox‐1 (PDX‐1) and glucokinase regulated by FoxO1a was increased 2‐ to 3‐fold with ANP present. The evidence suggests that NPR‐A stimulation results in activation of a growth promoting signaling pathway in the pancreatic β‐cell that includes PI3K/Akt/FoxO1a/cyclin D2. The activation of PI3K/Akt by ANP or 8‐bromo‐cGMP promotes cyclin D2, PDX‐1 and glucokinase transcription by phosphorylating and restricting FoxO1a activity.

Cells from primary islets and beta-cell lines form pseudoislets (PIs) in static cultures. Interestingly, MIN6 beta-cells with aberrant regulation of proliferation form PIs which cease to grow after a week in culture. This growth arrest is attributed to a pro-apoptotic and anti-proliferative PI environment. We hypothesized that cell necrosis due to poor nutrient transport in dishes rather than apoptosis effects the observed PI size restriction. Formation of beta-cell PIs was explored in stirred-suspension bioreactors with enhanced mass transfer. Cells in stirred-suspension proliferated continuously and the PI size increased for two weeks. Bioreactor PIs displayed regulated basal insulin secretion and enhanced responsivity to glucose and incretins. Compared to dishes, cell viability in the bioreactor was higher with lower released lactate dehydrogenase activity. Similar expression of p21 and p27 in monolayers and PIs did not suggest an anti-proliferative PI milieu. Caspase-2, -8 and -9 activities were comparable in dish and bioreactor PIs, and the latter continued to grow after one week of culture. Thus, apoptosis is not sufficient to explain the differences in PI size between dishes and bioreactor. Moreover, the bioreactor method described here may be used to generate PIs with increased cell viability and function for research and clinical applications.

Sphingosine kinase (SPHK) catalyzes sphingosine 1-phosphate production, promoting cell survival and reducing apoptosis in isolated rat pancreatic islets. Glucose, the primary islet β-cell growth factor and insulin secretagogue, increased islet SPHK activity by 3- to 5-fold following acute (1 h) or prolonged (7 days) stimulation. Prolonged stimulation of islets with glucose induced SPHK1a and SPHK2 mRNA levels; there were no changes in SPHK protein expression. To isolate the metabolic effects of glucose on SPHK activation, islets were stimulated with glucose analogs or metabolites. 2-deoxy-D-glucose (2-DG), an analog phosphorylated by glucokinase but not an effective energy source, activated SPHK similarly to glucose. In contrast, 3-o-methylglucose (3-oMeG), which is transported but neither phosphorylated nor metabolized, did not increase islet SPHK activity. Glyceraldehyde and α-ketoisocaproic acid (KIC), metabolites that stimulate glycolysis and the citric acid cycle, respectively, did not activate islet SPHK. Moreover, inorganic phosphate blocked glucose-induced SPHK activation. A role for SPHK activity in β-cell growth was confirmed when small interfering (si)SPHK2 RNA transfection reduced rat insulinoma INS-1e cell SPHK levels and activity and cell growth. Glucose induced an early and sustained increase in islet SPHK activity that was dependent on glucose phosphorylation, but independent of ATP generation or new protein biosynthesis. Glucose-supported β-cell growth appears to be in part mediated by SPHK activity.

The adult differentiated insulin-secreting pancreatic islet beta-cell experiences slow growth. This study shows that atrial natriuretic peptide (ANP) stimulates cell proliferation and [(3)H]thymidine incorporation in INS-1E glucose-sensitive rat beta-cell line cells and isolated rat islet DNA. In addition, cGMP, the second messenger of natriuretic peptide receptors (NPR) A and B, stimulated islet DNA biosynthesis. The NPR-A receptor was expressed in INS-1E cells and islets. ANP-stimulated INS-1E cell DNA biosynthesis was blocked by preincubation with LY294002 (50 microM), an inhibitor of phosphatidylinositol 3'-kinase (PI3K). An indicator of cell cycle progression, cyclin D2 mRNA was increased by 2- to 3-fold in ANP- or 8-Br-cGMP-treated INS-1E cells and islets, and these responses were inhibited by LY294002. ANP and 8-Br-cGMP stimulated the phosphorylation of Akt and Foxo1a in INS-1E cells and islets, and LY294002 inhibited these responses. In contrast, ANP reduced the levels of phospho-ERK in INS-1E cells. Pancreas duodenum homeobox-1 (PDX-1) is essential for pancreas development, insulin production, and glucose homeostasis, and ANP increased PDX-1 mRNA levels by 2- to 3-fold in INS-1E cells and islets. The levels of glucokinase mRNA in islets and INS-1E cells were also increased in response to ANP. The evidence suggests that pancreatic beta-cell NPR-A stimulation results in activation of a growth-promoting signaling pathway that includes PI3K/Akt/Foxo1a/cyclin D2. These data support the conclusion that the activation of Akt by ANP or 8-Br-cGMP promotes cyclin D2, PDX-1, and glucokinase transcription by phosphorylating and restricting Foxo1a activity.

Natriuretic peptide receptor‐A (NPR‐A) is a guanylyl cyclase‐linked receptor that generates cGMP in response to atrial natriuretic peptide (ANP). Since elevated ANP levels are associated with diabetes mellitus and hyperglycemia, ANP effects on cultured isolated rat pancreatic islets and insulinoma cells (INS‐1e) were determined. Islets and INS‐1e cells expressed NPR‐A mRNA and protein. To determine the chronic effects of ANP, islets were cultured at 5.5 mM glucose with ANP (1 μM) for 7 days, which increased [3H]thymidine incorporation to 200±36% of control (p<0.05); similarly, ANP increased INS‐1e cell [3H]thymidine incorporation to 163±18% of control (p<0.05). 8‐Bromo‐cGMP (0.5 mM) also stimulated islet [3H]thymidine incorporation to 207±15% of control (p<0.001). The mTOR inhibitor rapamycin failed to inhibit ANP‐stimulated [3H]thymidine incorporation. In ANP‐treated INS‐1e cells, the phosphatidylinositol 3‐kinase (PI3K)/Akt/Foxo1a pathway was activated. Phospho‐Akt and phospho‐Foxo1a were increased by 387±77% and 170 ±12% of control (p<0.05), respectively, and this response was blocked by the PI3K inhibitor LY294002. Cyclin D2 mRNA was upregulated in ANP treated islets and INS‐1e cells by 336±71% and 332±36% of control (p<0.05), respectively. The evidence suggests that the mitogenic response stimulated by ANP treatment involves activation of the PI3K/Akt/Foxo1a/cyclin D2 pathway.

Alternative insulin therapies are being sought that will provide euglycemic control for people with diabetes mellitus. The epidermis is a self-renewing tissue that is easily accessible and can provide large numbers of autologous cells that can be used for generating insulin-secreting skin substitutes. Lentiviral vectors have been engineered to produce a fusion protein between the furin-cleavable proinsulin and the self-dimerization mutant of FK506-binding protein to yield bioactive insulin in keratinocytes; this insulin is released as a response to exogenous administration of a small organic molecule, rapamycin. The engineered keratinocytes retained normal morphology and grew in a manner similar to lentiviral-treated control cells. Epidermal keratinocytes in culture and in stratified bioengineered epidermis released insulin within 30 minutes after addition of rapamycin, and secretion slowed or stopped within 2-3 hours after removal of the inducing agent. When the cells were implanted into athymic mice that had been rendered diabetic with streptozotocin (STZ), insulin was detected in the plasma within 1 hour after addition of rapamycin. Concomitantly, serum glucose decreased to normal levels even in diabetic animals with severe hyperglycemia. Repeated rapamycin administration yielded similar results. These experiments provide proof-of-concept that insulin released from the skin in a regulatable manner can reverse hyperglycemia.

Atrial natriuretic peptide (ANP) is a stimulus for particulate guanylyl cyclase and cyclic GMP (cGMP) formation through natriuretic peptide receptors A and B. Natriuretic peptide receptor subtypes A, B, and C mRNAs were identified by RT‐PCR in rat pancreatic islets of Langerhans. Culture of isolated rat islets at 5.5 mM glucose with ANP (1 M) for 7 days resulted in increased [3H]thymidine incorporation that was 200±36% of control (P<0.05). Rat insulinoma INS‐1e cell [3H]thymidine incorporation were similarly stimulated by ANP to 163±18% (P<0.05) of control. 8‐Bromo‐cGMP (0.5 mM) also stimulated islet [3H]thymidine incorporation to 207±15% of control (P<0.001). The expression of cyclin D2 mRNA was also increased by ANP. The total protein content of INS‐1e cells did not change during ANP stimulation. The total insulin content of pancreatic islets did not change during 7 day culture with ANP. However, the islet insulin secretory response to 11 mM glucose was markedly inhibited (74±7%) (P<0.01) by ANP after 7 day culture (but not a 3 hour pretreatment). Thus, ANP stimulates a mitogenic response in islets and INS‐1 cells that can be mimicked by cGMP. Long‐term ANP treatment also blunts the β‐cell insulin secretory response to glucose, but does not significantly alter total insulin content.

Despite many years of research, daily insulin injections remain the gold standard for diabetes treatment. Gene therapy may provide an alternative strategy by imparting the ability to secrete insulin from an ectopic site. The epidermis is a self-renewing tissue that is easily accessible and can provide large numbers of autologous cells to generate insulin-secreting skin substitutes. Here we used a recombinant retrovirus to modify human epidermal keratinocytes with a gene encoding for human proinsulin containing the furin recognition sequences at the A-C and B-C junctions. Keratinocytes were able to process proinsulin and secrete active insulin that promoted glucose uptake. Primary epidermal cells produced higher amounts of insulin than cell lines, suggesting that insulin secretion may depend on the physiological state of the producer cells. Modified cells maintained the ability to stratify into 3-dimensional skin equivalents that expressed insulin at the basal and suprabasal layers. Modifications at the furin recognition sites did not improve proinsulin processing, but a single amino acid substitution in the proinsulin B chain enhanced C-peptide secretion from cultured cells and bioengineered skin substitutes 10- and 28-fold, respectively. These results suggest that gene-modified bioengineered skin may provide an alternative means of insulin delivery for treatment of diabetes.

Pancreatic islet β-cells undergo apoptosis (APO) following culture with cytokines (CTK) in an in vitro model for type 1 diabetes mellitus. The aim was to determine if sphingosine 1-phosphate (S1P) modulates the biochemical pathways that mediate the β-cell APO response. Isolated rat islets or cells were pretreated with S1P (400 nM) for 5 h prior to and during CTK stimulation for 24-48 h. APO was determined by TUNEL staining. CTK (1 ng/ml interleukin-1β and 5 ng/ml interferon-γ ) induced 28±2% APO in islet cells after 48 h. S1P alone did not affect islet cell APO, however, S1P together with CTK reduced APO to 13±3% (P<0.001 vs. CTK) of total cells. Caspase 3 activity in INS-1e cells was also a measure of APO. CTK increased caspase 3 activity after 24 h to 193±33% of control (C) cells without CTK treatment (P<0.001). The presence of S1P or dihydro-S1P (400 nM) with CTK reduced INS-1e cell caspase 3 activity to 127±15% C (P>0.05) and 105±14% C (P>0.05), respectively; the occurrence of APO cells with S1P alone was 94±7% C (P>0.05). Forskolin (50 nM) reduced CTK-induced caspase 3 activity to 121±5% C (P>0.05). Analysis by real-time PCR showed that the mRNA expression for the anti-APO gene Bcl-xL increased by 337±34% C (P<0.001) in CTK-treated islets, and that S1P augmented the expression to 570±63% C (P<0.01); with S1P alone Bcl-xL mRNA levels were 108±15% C. In conclusion, S1P protects the islet β-cell from cytokine-induced APO to a similar extent as forskolin, and increased expression of Bcl-xL and reduced activity of caspase 3 play a role in the antiapoptotic effects. (Supported by Juvenile Diabetes Research Foundation grant 1-2002-613)

Cytokines mediate pancreatic islet beta-cell apoptosis and necrosis, leading to loss of insulin secretory capacity and type 1 diabetes mellitus. The cytokines, IL-1 beta and interferon-gamma, induced terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate nick end labeling ( TUNEL) staining of rat islet cells within 48 h by about 25 - 30%, indicative of apoptosis and/or necrosis. Sphingosine 1-phosphate (S1P) at nanomolar concentrations significantly reduced islet cell cytokine-induced TUNEL staining. Similar effects were observed in INS-1 cells. The dihydro analog of S1P also reduced the percentage of TUNEL stained islet and INS-1 cells, whereas the S1P receptor antagonist BML-241 blocked the protective effects. Pertussis toxin did not affect the S1P protective response. In the presence of a phospholipase C antagonist, U73122, there was significant inhibition of the S1P protective effects against apoptosis/necrosis. S1P stimulated INS-1 cell protein kinase C activity. Carbamylcholine chloride acting through muscarinic receptors also inhibited cytokine-induced TUNEL staining in pancreatic islet cells. S1P and/or dihydro-S1P also antagonized cytokine-induced increases in cytochrome c release from mitochondria and caspase-3 activity in INS-1 cells, which are indicative of cell apoptosis vs. necrosis. S1P failed to affect nitric oxide synthase activity after 48 h. Thus, the evidence suggests that S1P acting on S1P receptors coupled to G(q) mediates protective effects on islet beta-cells against cytokine-induced apoptosis.

Sphingosine-1 phosphate (S1P) is a bioactive sphingolipid with the potential to mobilize Ca2+, to inhibit apoptosis, and to promote mitogenesis. Sphingosine kinase (SPHK) and S1P were characterized in INS-1 insulinoma cells and isolated rat islets of Langerhans. SPHK activity increased in INS-1 cell homogenates treated with interleukin-1beta (IL-1beta) or tumor necrosis factor-alpha (TNF-alpha), and responses were additive. IL-1beta or TNF-alpha increased islet SPHK activity within 15 min to 1 h; activity remained elevated after 8 h. SPHK2 was the predominant active isoform in INS-1 cells; little or no SPHK1 activity was detected. Cytokines increased endogenous S1P biosynthesis in 32P(i)-prelabeled INS-1 cells, and cycloheximide inhibited the response after 8 h, suggesting that protein synthesis mediated the response. There was no [32P]S1P release from cells. Compared with basal values, IL-1beta and TNF-alpha induced increases in SPHK1a mRNA levels relative to 18S ribosomal RNA in INS-1 cells within 1 h; relative SPHK2 mRNA levels were unchanged after cytokine treatment. IL-1beta, but not TNF-alpha, induced relative SPHK1a mRNA expression levels within 1 h in islets, whereas SPHK2 mRNA levels were unchanged. Thus, IL-1beta and TNF-alpha induced an early and sustained increase in SPHK activity in INS-1 cells and isolated islets, suggesting that S1P plays a role in the pathological response of pancreatic beta-cells to cytokines.