Glucagon-like-peptide-1 secretion from canine L-cells is increased by glucose-dependent-insulinotropic peptide but unaffected by glucose.

Glucagon-like peptide-1(7-36)amide (GLP-1) is a potent insulinotropic peptide released from the small intestine. To investigate the regulation of GLP-1 secretion, we established a GLP-1 release assay based on primary canine intestinal L-cells. The ileal mucosa was digested with collagenase/EDTA to a single cell suspension and enriched for L-cells by counterstream centrifugal elutriation. We performed release assays on the cultured cells after 36 h, and GLP-1 in the supernatant was determined by enzyme-linked immunoabsorbent assay (ELISA). Glucose-dependent insulinotropic peptide (GIP) dose dependently stimulated the release of GLP-1 and resulted in a 2-fold increase at 100 nM GIP. This effect was fully inhibited by 10 nM somatostatin. However, neither basal or GIP stimulated GLP-1 secretion were affected by ambient glucose concentrations from 5-25 mM. The receptor-independent secretagogues beta phorbol myristate acetate and forskolin dose dependently increased the secretion of GLP-1; effects inhibited by staurosporine and H8 respectively. Costimulation with GIP and phorbol ester, but not forskolin, resulted in an additive response. Furthermore, the effect of GIP could be inhibited by H8 but not by staurosporine. These results indicate that glucose does not directly stimulate canine L-cells. It is more probable that glucose releases GIP from the upper intestine that in turn stimulates GLP-1 secretion. The ability of GIP to stimulate GLP-1 secretion is probably mediated through activation of protein kinase A.

Proglucagon processing profile in canine L cells expressing endogenous prohormone convertase 1/3 and prohormone convertase 2.

The tissue-specific differential processing of proglucagon (Pg) yields glucagon in pancreatic A cells and glucagon-like peptide-1 (GLP-1), GLP-2, and glicentin in intestinal L cells. It has been suggested that the difference in Pg cleavage in A and L cells is due to the presence of distinct prohormone convertases (PC) in the two cell types, PC1/3 in the L cell and PC2 in the A cell. PC2 has been shown to cleave the N-terminal part of Pg, being essential for glucagon formation and PC1/3 to cleave the C-terminal part of Pg, leading to the formation of GLP-1. However, some of the cleavage sites in Pg have not proven to be substrates exclusively for either PC2 or PC1/3, and the cleavage profile of Pg in a primary cultured L cell has not yet been correlated with the actual presence of PC2 and PC1/3 in the L cell. We demonstrate here the presence of PC1/3, PC2, and the PC2 chaperone 7b2, in L cells using light immunohistochemistry on sections from canine ileum and on a canine intestinal cell culture enriched for L cells. Analysis of the cultured L cells, using gel chromatography and RIA, confirms the classical intestinal cleavage profile of Pg, resulting in mainly glicentin, oxyntomodulin, GLP-1-(7-37), and GLP-2. Despite the presence of 7b2 and mature PC2, as demonstrated by Western blot, absolute minimal amounts of glucagon were detected. These data show that the presence of intracellular PC2 and 7b2 in a primary cell possessing Pg does not have to lead to the formation of glucagon. This formation must then require an additional element to occur, or alternatively, the results could be explained by a canine specific organization of PC2 and Pg into separate compartments, which would prevent interaction.

Receptor dimerization determines the effects of growth hormone in primary rat adipocytes and cultured human IM-9 lymphocytes.

Binding of GH to its cell surface receptors is thought to result in the formation of a complex comprised of one molecule of hormone per two molecules of receptor. It has been proposed that this hormone-induced receptor dimerization is important for the mechanism of signal transduction. We have developed a mathematical model for quantitative evaluation of the biological responses associated with sequential receptor dimerization. Based on these predictions, we have investigated whether GH-induced receptor dimerization plays a role in two classical effects of GH, i.e. stimulation of lipogenesis in primary rat adipocytes and GH receptor down-regulation in cultured human IM-9 lymphocytes. Model predictions of biological responses linked to dimer formation yielded a bell-shaped pattern, with self-antagonism at high GH concentrations when monomeric GH-receptor complexes become predominant. The GH lipogenic bioactivity curve was indeed biphasic and first increased in a concentration-dependent manner between 10(-10)-10(-8) M GH (ED50, 0.5 nM), up to a maximum of 1.7-fold stimulation above basal. Then, the response decreased continuously above 5 x 10(-8) M GH, returning to basal levels around 10(-5) M GH. Incubation of IM-9 cells with wild-type human GH resulted in a dose-dependent loss of their surface receptors. In contrast, a human GH analog (G120R), mutated in the second binding surface of the hormone and, therefore, unable to induce GH receptor dimerization, failed to induce receptor down-regulation in the IM-9 cells. Furthermore, when added together with wild-type human GH, human GH(G120R) inhibited, in a concentration-dependent manner, the down-regulation induced by wild-type human GH. Taken together, these data support the hypothesis that receptor dimerization is critical for the stimulation by GH of both lipogenesis in primary rat adipocytes and receptor down-regulation in cultured human IM-9 lymphocytes.

Developmental expression of proprotein convertase 1/3 in the rat.

We have isolated a clone that has 3' end sequence identity with prohormone convertase 1/3 (PC1/3) from a rat islet cDNA library. Northern blot analysis and immunocytochemical studies have confirmed its presence in the endocrine pancreas. Analysis of poly A mRNA from various adult tissues demonstrated that it was relatively abundant in whole brain, lung and spleen, but not detectable in kidney, testis and heart. Using probes consisting of either the coding region or the 3' end sequences, the mRNA transcripts identified were 5.0, 3.0 and 8.5 kb. The 8.5 kb transcript detected has not been described previously. RT-PCR of RNA isolated from rat embryonic tissues using a primer set corresponding to the 3' end of the PC1/3 sequence showed a steady increase of expression in fetal pancreas and intestine during the course of development. In contrast, comparatively high and constant levels of PC1/3 expression were detected in fetal lung, whereas low and constant expression was detected in fetal liver. Double immuno-staining showed that PC1/3 was co-localised with insulin throughout development, and at mid-gestation, PC1/3 immunoreactivity could also be detected within glucagon-producing cells in the developing pancreas. Thus, we have identified a novel PC1/3 mRNA transcript in the rat by using sequence-specific probes and have demonstrated that the developmental expression of prohormone convertase PC1/3 is confined primarily to pancreas and intestine, suggesting that it may play a possible role in regulating growth and differentiation of these tissues.

Immunocytochemical evidence for a paracrine interaction between GIP and GLP-1-producing cells in canine small intestine.

Glucagon-like-peptide 1 (GLP-1) released from the intestine is considered to be an important incretin. We have recently demonstrated that glucose-dependent insulinotropic peptide (GIP) stimulated GLP-1 secretion from canine ileal L cells in culture. To investigate further the interplay between GLP-1- and GIP-secreting cells, we set out to determine the exact location and abundance of both cell types throughout the canine intestine. Canine small intestine was subdivided into 15-20 segments and investigated by immunocytochemistry with computer-assisted imaging. The abundance of GIP-, GLP-1- and somatostatin-immunoreactive cells was determined. GIP-secreting K cells were equally distributed in duodenum and jejunum, with the GLP-1-secreting L cells concentrated in the jejunum (5% duodenum, 73% jejunum and 22% ileum). These results indicated that the middle section of the small intestine containing 69% of the K cells also contained 51% of the L cells. Double immunostaining confirmed this overlap and furthermore over 30% of the L cells in this region were found adjacent to K cells. These results suggest the existence of a paracrine interaction between the K and L cells and indicate the importance of the jejunum in the regulation of insulin release by enteric-derived incretins.