Few studies have comprehensively examined amphibian granular gland secretions for novel insulinotropic peptides. This study involved isolation and characterisation of biologically active peptides from the skin secretions of Rana palustris frogs. Crude secretions obtained by mild electrical stimulation from the dorsal skin surface were purified by reversed-phase HPLC on a semipreparative Vydac C18 column, yielding 80 fractions. These fractions were assayed for insulin-releasing activity using glucose-responsive BRIN-BD11 cells. Acute 20 min incubations were performed in Krebs Ringer bicarbonate buffer supplemented with 5.6 mmol/l glucose in the absence (control) and presence of various fractions. Fractions 29-54 and fractions 68-75 showed significant 2.0-6.5-fold increases in insulin-releasing activity (P<0.001). The fractions showing most prominent insulinotropic activity were further purified to single homogeneous peaks, which, on testing, evoked 1.5-2.8-fold increases in insulin release (P<0.001). The structures of the purified peptides were determined by mass spectrometry and N-terminal amino acid sequencing. Electrospray ionisation ion-trap mass spectrometry analysis revealed molecular masses of 2873.5-8560.4 Da. Sufficient material was isolated to determine the primary amino acid sequence of the 2873.5 Da peptide, revealing a 27 amino acid sequence, ALSILRGLEKLAKMGIALTNCKATKKC, repressing palustrin-1c. The database search for this peptide showed a 48% homology with brevinin-1, an antimicrobial peptide isolated from various Rana species, which itself stimulated insulin release from BRIN-BD11 cells in a concentration-dependent manner. In conclusion, the skin secretions of R. palustris frogs contain a novel class of peptides with insulin-releasing activity that merit further investigation.

There are conflicting reports on the effect of serum from patients with insulin-dependent diabetes mellitus (IDDM) or normal human serum on beta-cell function and insulin secretion. Here, we report that the sera of newly diagnosed IDDM patients potently suppresses insulin secretion from a clonal rat pancreatic beta-cell line (BRIN-BD11), but do not alter cell viability. Indeed, the viability of the beta-cells was not significantly different between cells cultured in 10% (v/v) IDDM sera, normal human sera, or fetal calf serum after 24, 48 and 72 h. Alanine-stimulated insulin secretion from cells cultured for 24 h in (10% v/v) IDDM patient sera was reduced to 48% of that secreted from cells cultured in (10% v/v) normal human sera. After depletion of the complement components C1q and C3, the inhibition of insulin secretion induced by IDDM patient sera was significantly reversed (no significant difference was observed between cells cultured in complement-depleted IDDM patient sera and cells cultured in normal human sera or complement-depleted normal human sera). The concentration of glutamic acid decarboxylase (GAD) autoantibodies was markedly increased in the sera of six out of nine newly diagnosed IDDM patients in this study, whereas insulin auto-antibodies (IAA) were detected in the sera of three of the nine patients and islet-cell antibodies (ICA) in the sera of five of them. In addition, the concentration of soluble terminal complement complexes (SC5-9) was greater in some of the beta-cell culture media samples after 24 h incubation when the incubation medium was supplemented with IDDM patient sera than when supplementation was with normal human sera. We propose that the mechanism of sera-induced inhibition of insulin secretion from clonal beta-cells may involve complement- and cytokine-stimulated intracellular events that attenuate the metabolite-induced secretory process.

We have previously reported that newly diagnosed Type-1 diabetic patient sera potently suppressed insulin secretion from a clonal rat pancreatic beta-cell line (BRIN BD11) but did not alter cell viability. Here, we report that apoptosis in BRIN BD11 cells incubated in various sera types (fetal calf serum (FCS), normal human serum and Type-1 diabetic patient) was virtually undetectable. Although low levels of necrosis were detected, these were not significantly different between cells incubated in sera from different sources. ATP levels were reduced by approximately 30% while nitrite production increased twofold from BRIN BD11 cells incubated for 24 h in the presence of Type-1 diabetic patient sera compared with normal human sera. Additionally, ATP levels were reduced by approximately 40% and DNA fragmentation increased by more than 20-fold in BRIN BD11 cells incubated in FCS in the presence of a pro-inflammatory cytokine cocktail (interleukin-1beta, tumour necrosis factor-alpha and interferon-gamma), compared with cells incubated in the absence of cytokines. Nitric oxide production from BRIN BD11 cells was markedly increased (up to 10-fold) irrespective of sera type when the cytokine cocktail was included in the incubation medium. Type-1 diabetic patient sera significantly (P<0.001) raised basal levels of intracellular free Ca(2+ )concentration ([Ca(2+)](i)) in BRIN BD11 cells after a 24-h incubation. The alteration in [Ca(2+)](i) concentration was complement dependent, as removal of the early complement components C1q and C3 resulted in a significant reduction (P<0.01) of sera-induced [Ca(2+)](i )changes. We propose that the mechanism of Type-1 diabetic patient sera-induced inhibition of insulin secretion from clonal beta-cells may involve complement-stimulated elevation of [Ca(2+)](i) which attenuates the nutrient-induced insulin secretory process possibly by desensitizing the cell to further changes in Ca(2+).