Activation of protein kinase C is not required for glyceraldehyde-stimulated insulin secretion from rat islets

Proteomics-Based Identification of Interaction Partners of the Xenobiotic Detoxification Enzyme FMO3 Reveals Involvement in Urea Cycle

The hepatic xenobiotic metabolizing enzyme flavin-containing monooxygenase 3 (FMO3) has been implicated in the development of cardiometabolic disease primarily due to its enzymatic product trimethylamine-N oxide (TMAO), which has recently been shown to be associated with multiple chronic diseases, including kidney and coronary artery diseases. Although TMAO may have causative roles as a pro-inflammatory mediator, the possibility for roles in metabolic disease for FMO3, irrespective of TMAO formation, does exist. We hypothesized that FMO3 may interact with other proteins known to be involved in cardiometabolic diseases and that modulating the expression of FMO3 may impact on these interaction partners. Here, we combine a co-immunoprecipitation strategy coupled to unbiased proteomic workflow to report a novel protein:protein interaction network for FMO3. We identified 51 FMO3 protein interaction partners, and through gene ontology analysis, have identified urea cycle as an enriched pathway. Using mice deficient in FMO3 on two separate backgrounds, we validated and further investigated expressional and functional associations between FMO3 and the identified urea cycle genes. FMO3-deficient mice showed hepatic overexpression of carbamoylphosphate synthetase (CPS1), the rate-limiting gene of urea cycle, and increased hepatic urea levels, especially in mice of FVB (Friend leukemia virus B strain) background. Finally, overexpression of FMO3 in murine AML12 hepatocytes led to downregulation of CPS1. Although there is past literature linking TMAO to urea cycle, this is the first published work showing that FMO3 and CPS1 may directly interact, implicating a role for FMO3 in chronic kidney disease irrespective of TMAO formation.

Effector systems involved in the insulin secretory responses to efaroxan and RX871024 in rat islets of Langerhans

One component of the mechanism by which imidazoline compounds promote insulin secretion involves closure of ATP-sensitive K+ channels in the beta-cell plasma membrane. Recently, however, it has also been proposed that these compounds may exert important effects on more distal effector systems. In the present work, we have investigated the contribution played by protein kinases A and C to the insulin secretory responses of isolated rat islets of Langerhans treated with efaroxan and RX871024 (1-phenyl-2-(imidazolin-2-yl) benzimidazole). Removal of extracellular Ca2+ or blockade of voltage-sensitive Ca2+ channels prevented stimulation of insulin secretion by efaroxan, confirming a critical role for increased Ca2+ influx in the secretory response. By contrast, inhibition of protein kinases A or C failed to alter efaroxan-induced insulin secretion. RX871024 dose-dependently increased insulin secretion from cultured islets incubated with 20 mM glucose. This effect was unaffected by modulation of protein kinase C, but was significantly attenuated by a selective inhibitor of protein kinase A (Rp-cAMPs). Measurements of cAMP revealed that RX871024 increased the islet cAMP content by more than 3-fold; reaching values similar in magnitude to those elicited by 50 microM 3-isobutyl-1-methyl xanthine. The results reveal that neither protein kinase A nor protein kinase C is obligatory for stimulation of insulin secretion by imidazolines. However, they suggest that a rise in cAMP may contribute to the amplified secretory response observed when cultured islets are incubated with RX871024 in the presence of a stimulatory glucose concentration.

Arachidonic acid-induced down-regulation of protein kinase C delta in beta-cells

We have previously identified expression of multiple protein kinase C (PKC) isoforms in insulinoma-derived beta-cells and whole islets. Both PKC delta and PKC alpha appear to be the more abundantly expressed isoforms. In this report we studied the effects of arachidonic acid (AA) on the subcellular distribution of PKC alpha and PKC delta. AA has been reported to activate both PKC alpha and PKC delta and it is thought to be an important second messenger in beta-cells. Here we report that AA interacted with and altered beta-cell pools of PKC delta preferentially over PKC alpha. AA (100 microM) over the course of 45 min reduced cytosolic levels of PKC delta (to 40 +/- 15%, compared to time zero control) leaving membrane- and cytoskeleton-associated levels near control levels. Analysis of whole cell homogenates showed a slight down-regulation of PKC delta indicating proteolysis. The down-regulation of cytosolic PKC delta appeared to be isoform specific since cytosolic PKC alpha remained at control levels over the time course. The response was dose-dependent and negligible at concentrations below 30 microM and occurred, at least partially, in the cytosolic compartment of the cell. Indomethacin also down-regulated cytosolic PKC delta preferentially over PKC alpha possibly through accumulation of AA. These findings suggest that cytosolic PKC delta may be a downstream target of this beta-cell second messenger.

A myristoylated pseudosubstrate peptide inhibitor of protein kinase C: effects on glucose- and carbachol-induced insulin secretion

We have used synthetic pseudosubstrate peptide inhibitors of protein kinase C (PKC) to re-examine the role of conventional isoforms of PKC in the insulin secretory response of intact rat islets of Langerhans to glucose and to the cholinergic agonist carbachol (CCh). One peptide was modified by N-terminal myristoylation (PKC-myr20-28) to allow its use in intact beta-cells. Maximal inhibition of PKC activity in vitro required 10-fold less of this peptide (PKC-myr20-28) than of its non-myristoylated analogue. The maximum inhibitory concentration of PKC-myr20-28 had little effect on islet protein kinase A or Ca2+/calmodulin kinase activities. PKC-myr20-28 (25-100 microM) caused a dose-dependent inhibition of phorbol myristate acetate (PMA)-induced insulin secretion from intact rat islets but non-myristoylated peptides had little effect on the secretory response to PMA. A concentration of PKC-myr20-28 (100 microM) which maximally inhibited PMA-induced insulin secretion, also inhibited the secretory response to CCh, but did not affect glucose-stimulated insulin secretion from intact islets. These results indicate that myristoylation of pseudosubstrate peptides increases their potency as inhibitors and that PKC-myr20-28 is a selective and cell-permeant inhibitor of PMA-sensitive isoforms of PKC. They also suggest that the activation of PMA-sensitive PKC isoforms mediates the stimulatory effects of CCh, but is not obligatory for glucose-induced insulin secretion from pancreatic beta-cells.

Inhibition of glucose-stimulated insulin secretion by Ro 31-8220, a protein kinase C inhibitor

The involvement of the family of protein kinase C (PKC) isoenzymes in the secretory response of rat islets of Langerhans to glucose, the major insulin secretagogue, was investigated using the PKC inhibitor Ro 31-8220, a derivative of staurosporine. Ro 31-8220 was a more selective PKC inhibitor than staurosporine in islets, having minimal effects on protein kinases activated by cyclic AMP or by Ca(2+) and calmodulin. The secretory response to 4βPMA, an activator of phorbol ester-sensitive isoforms of PKC, was abolished by Ro 31-8220. Basal insulin secretion (2MM: glucose) was not affected by Ro 31-8220, but 20MM: glucose-induced insulin release was inhibited in a dose-dependent manner, maximally by ∼50% at 10 µM: Ro 31-8220. Higher concentrations of Ro 31-8220 (507gmM: ) did not further inhibit the secretory response to glucose and also caused ∼50% inhibition of insulin secretion stimulated by 10MM: glyceraldehyde. Ca(2+)-stimulated insulin secretion from electrically permeabilised islets was not inhibited by Ro 31-8220. Calphostin C, which inhibits some isoforms of PKC by interacting with the diacylglycerol binding site, unexpectedly caused a large (∼10-fold) increase in secretion at 2MM: glucose, so could not be used in islets to further investigate the involvement of phorbol ester-sensitive PKC isoforms in the insulin secretory process. One possible explanation for our results using Ro 31-8220 is that phorbol ester-insensitive isoforms of PKC (ζ and/orι) are involved in glucose-stimulated insulin secretion from rat islets.

Protein phosphorylation in the regulation of insulin secretion: the use of site-directed inhibitory peptides in electrically permeabilised islets of Langerhans

We have used electrically permeabilised rat islets of Langerhans to investigate the role of protein phosphorylation in the regulation of insulin secretion using pseudosubstrate inhibitory peptides for cyclic AMP-dependent protein kinase (PKA) and for protein kinase C (PKC). The protein kinase inhibitor (PKI) peptide, PKI(6-22), completely inhibited the effects of cyclic AMP on islet PKA activity in vitro, on endogenous protein phosphorylation and on insulin secretion. This peptide had no significant effect on islet PKC activity in vitro, on Ca(2+)-induced protein phosphorylation and on secretory responses to Ca2+ or to the PKC activator, 4 beta-phorbol myristate acetate (PMA). The PKC pseudosubstrate inhibitory peptide, PKC(19-36), caused a marked inhibition of islet PKC activity in vitro and inhibite PMA-induced insulin secretion without affecting secretory responses to cyclic AMP and Ca2+. These results demonstrate that PKA- and PKC-induced protein phosphorylation is obligatory for cyclic AMP- and PMA-stimulated insulin secretion, respectively, and suggest that there is little "crosstalk" between the response elements of the secretory pathways to the different second messengers, at least after the generation of the messengers within the beta-cells.

Staurosporine inhibits protein kinases activated by Ca2+ and cyclic AMP in addition to inhibiting protein kinase C in rat islets of Langerhans

Staurosporine has been used in several studies to investigate the role of protein kinase C (PKC) in secretory responses of islets of Langerhans to insulin secretagogues. We have assessed the effect of staurosporine on: [i] islet PKC activity in vitro; [ii] the stimulation of insulin secretion by nutrient secretagogues and [iii] the stimulation of protein phosphorylation and insulin secretion in electrically permeabilised islets. All experiments were carried out on rat isolated islets of Langerhans, either intact or permeabilised by high voltage discharge (3.4 kV/cm). The activity of PKC partially purified from rat islets was inhibited by staurosporine (1.6-400 nM) in a concentration-dependent manner. Staurosporine also inhibited insulin secretion stimulated by both glucose and glyceraldehyde, with maximal effects at 50 nM. After prolonged exposure of islets to the tumour-promoting phorbol ester, 4 beta phorbol myristate acetate (4 beta PMA), a procedure which depletes islet PKC activity, staurosporine still inhibited both glucose- and glyceraldehyde-stimulated insulin release. In electrically permeabilised islets, staurosporine inhibited both Ca(2+)- and cyclic AMP-stimulated protein phosphorylation and insulin secretion. These results suggest that staurosporine should not be used as a selective inhibitor of PKC in rat islets.

Mastoparan stimulates insulin secretion from pancreatic beta-cells by effects at a late stage in the secretory pathway

Mastoparan (MP) is a component of wasp venom which stimulates secretion from a number of cell types. We have used intact and electrically permeabilised islets of Langerhans to investigate the mechanisms through which MP stimulates insulin secretion from pancreatic beta-cells. MP caused a temperature-dependent and dose-related stimulation of insulin secretion from intact islets at a substimulatory concentration (2 mM) of glucose, which was not dependent upon the presence of extracellular Ca2+. MP also stimulated ATP-independent insulin secretion from electrically permeabilised islets in which intracellular Ca2+ was clamped at a substimulatory concentration (50 nM). MP-induced insulin secretion was not inhibited by down-regulation of islet protein kinase C, nor by the protein kinase inhibitor staurosporine, nor by the cyclic AMP antagonist Rp-adenosine 3',5'-cyclic phosphorothioate. However, MP-induced secretion from permeabilised islets was inhibited by the presence of guanosine 5'-O-2-thiodiphosphate. These results suggest that MP stimulates insulin secretion by a mechanism that is independent of changes in cytosolic Ca2+ or protein kinase activation, but which is dependent, at least in part, upon activation of a GTP-binding protein at a late stage in the secretory process.

Stimulation of HIT-T15 insulinoma cells by glyceraldehyde does not require its metabolism

The addition of the triose D-glyceraldehyde (5-20 mM) to HIT-T15 hamster insulinoma cells caused a rapid, marked depolarisation of the plasma membrane accompanied by a pronounced intracellular acidification, an increase in the cytosolic free calcium concentration [Ca2+]i and enhanced secretion of insulin. D-glyceraldehyde did not reduce the rate of efflux of 86Rb+ from loaded perifused cells. All of the above effects of D-glyceraldehyde were also observed in response to L-glyceraldehyde. The changes in membrane potential and intracellular pH (pHi) caused by D-glyceraldehyde were unaffected by the glycolytic inhibitor iodoacetate, by K(+)-channel blockers (tolbutamide and tetraethylammonium), or by inhibitors of the transport of lactate (alpha-fluorocinnamate), alanine (methylaminoisobutyrate) or glucose (phloretin, phlorrizin). The glyceraldehyde-induced depolarisation and acidification were also observed in the absence of extracellular Ca2+ or Na+. The increase in [Ca2+]i evoked by D-glyceraldehyde was reversed by removal of Ca2+ from the medium. The formation of lactate by HIT-T15 cells was not significantly increased by addition of 10 mM D-glyceraldehyde or L-glyceraldehyde. In contrast, 10 mM glucose caused an approximately fourfold rise in lactate production. The oxidation of D-glyceraldehyde by HIT-T15 cells was also extremely modest compared to glucose oxidation by these cells. These results suggest that the stimulation of HIT-T15 cells by either D-glyceraldehyde of L-glyceraldehyde does not require metabolism of the triose within the cell and may not involve closure of nucleotide-sensitive K+ channels. We propose that the electrogenic transport of glyceraldehyde across the plasma membrane, possibly via H+ cotransport, might lead to depolarisation and hence to Ca2+ entry into the cell.