Protein inhibitor of neuronal nitric oxide synthase (PIN) is a new regulator of glucose-induced insulin secretion.

We previously showed that pancreatic beta-cells express neuronal nitric oxide synthase (nNOS) that controls insulin secretion through two catalytic activities: nitric oxide (NO) production and cytochrome c reductase activity. We now provide evidence that the endogenous protein inhibitor of nNOS (PIN) is expressed in rat pancreatic islets and INS-1 cells. Double-immunofluorescence studies showed a colocalization of PIN with both nNOS and myosin Va in insulin-secreting beta-cells. Electron microscopy studies confirmed that PIN is mainly associated with insulin secretory granules and colocated with nNOS in the latter. In addition, PIN overexpression in INS-1 cells enhanced glucose-induced insulin secretion, which is only partly reversed by addition of an NO donor, sodium nitroprusside (SNP), and unaffected by the inhibitor of cytochrome c reductase activity, miconazole. In contrast, the pharmacological inhibitor of nNOS, Nomega-nitro-l-arginine methyl ester, amplified glucose-induced insulin secretion, an effect insensitive to SNP but completely normalized by the addition of miconazole. Thus, PIN insulinotropic effect could be related to its colocalization with the actin-based molecular motor myosin Va and as such be implicated in the physiological regulation of glucose-induced insulin secretion at the level of the exocytotic machinery.

Cellulose membrane supported peptide arrays for deciphering protein-protein interaction sites: the case of PIN, a protein with multiple natural partners.

Cellulose membrane supported peptide arrays, prepared according to the Spot method, allow the rapid identification and characterization of protein-protein interaction sites. Here, the method was used to screen reactive peptides from different proteins that bind to a single molecule, the PIN protein. PIN possesses two binding grooves, that have been shown to interact with several targets, including neuronal NO synthase, dynein intermediate chain, myosin V, the proapoptotic protein Bim, the scaffolding proteins DAP1alpha and gephyrin, and the transcription factor NRF-1. Arrays of peptides representing sequences of these targets were probed for reactivity with GST-tagged PIN, enabling the precise identification of binding motifs. Binding motifs were then minimized to seven or eight amino acid long peptides: YSKETQT for dynein IC, CDKSTQT for Bim, KDTGIQVD for nNOS, QSVGVQV for DAP1alpha and EDKNTMTD for myosin V. Alascan and substitution analysis provided proof that the Gln residue is critical for the interaction and cannot be easily replaced. Positions -1 and +1, just flanking the pivotal Gln, are also important; they consist of hydrophobic residues (Thr, Val) that could only be replaced by hydrophobic or aromatic amino acids. Position -4 is also critical for binding, with its Asp or Ser being replaceable to some extent. Alignment of sequences of proteins known to bind PIN shows that the most frequent amino acids in the motif are DKGTQT, consistent with the Spot results. We postulate that the degenerate character of binding to PIN is based on the propensity of several sequences to adopt a beta-strand conformation that allows the Gln residue to position itself in the PIN channel and on the conformational breathing of the PIN binding groove.

Changes in the dimeric state of neuronal nitric oxide synthase affect the kinetics of secretagogue-induced insulin response.

We previously showed that pancreatic beta-cells express a neuronal isoform of nitric oxide synthase (nNOS) that controls insulin secretion by exerting two enzymatic activities: nitric oxide (NO) production and cytochrome c reductase activity. We now bring evidence that two inhibitors of nNOS, N-omega-nitro-l-arginine methyl ester (l-NAME) and 7-nitroindazole (7-NI), increase glucose-induced insulin secretion but affect beta-cell function differently. In the presence of l-NAME, insulin response is monophasic, whereas 7-NI preserves the normal biphasic secretory pattern. In addition, the alterations of beta-cell functional response induced by the inhibitors also differ by their sensitivity to a substitutive treatment with sodium nitroprusside, a chemical NO donor. These differences are probably related to the nature of the two inhibitors. Indeed, using low-temperature SDS-PAGE and real-time analysis of nNOS dimerization by surface plasmon resonance, we could show that 7-NI, which competes with arginine and tetrahydrobiopterin (BH(4)), an essential cofactor for nNOS dimer formation, inhibits dimerization of the enzyme, whereas the substrate-based inhibitor l-NAME stabilizes the homodimeric state of nNOS. The latter effect could be reproduced by the two endogenous inhibitors of NOS, N-omega-methyl-l-arginine and asymmetric dimethylarginine, and resulted interestingly in a reduced ability of the protein inhibitor of nNOS (PIN) to dissociate nNOS dimers. We conclude that intracellular factors able to induce abnormalities in the nNOS monomer/dimer equilibrium could lead to pancreatic beta-cell dysfunction.