Extracts of human term amniotic, placental, and chorion/decidua tissue contained, respectively, 4.36 +/- 2.79 (pmol/g wet wt; mean +/- S.E.M.: n = 5). 2.78 +/- 0.5 (n = 5) and 0.68 +/- 0.68 (n = 5) peptide YY (PYY)-like immunoreactivity. Using a specific PYY antiserum, gel filtration chromatography and reverse-phase high performance liquid chromatography (HLPC), amniotic, placental and fetal intestinal tissue extracts were demonstrated to contain PYY-like immunoreactivity consisting of equal amounts of PYY1-36 and PYY3-36. The presence of pancreatic polypeptide was not detected in any of the extracts. Positive immunohistochemical staining for PYY was seen in extravillous trophoblasts in the decidual septa and fetal membranes, the syncytiotrophoblast and cytotrophoblasts, amniotic epithelial cells and in maternal decidual stromal cells. Positive staining for PYY was found at the earliest date examined (9.5 weeks) and remained present throughout pregnancy to term. PYY1-36 and PYY3-36 may play important roles in human pregnancy, acting via endocrine or paracrine mechanisms.

To determine the role of insulin-like growth factors (IGFs) and their binding proteins (IGFBPs) in the development of the pancreas, and specifically of the islets of Langerhans, we have examined the cellular distribution and developmental changes in the expression of IGFs and IGFBPs in the pancreas of the fetal and neonatal rat between 19.5 days of gestation and postnatal day 28. This represents a period of substantial growth and restructuring of the beta cell component in islets of this species. IGF-I, IGF-II, and IGFBPs-1 to -6 mRNAs were localized by in situ hybridization, and peptides by immunohistochemistry, in histological sections. IGF-II mRNA was highly expressed in islet cells and some ductal epithelial cells in late fetal and early neonatal life, but was barely detectable by postnatal day 28. IGF-II peptide showed a similar distribution. IGF-I mRNA was barely detected in the fetus or neonate and was localized predominantly in the ductal and acinar tissues after postnatal day 7. IGF-I immunoreactivity was associated with some islet cells in the fetus and neonate, suggesting an endocrine rather than a paracrine source. We performed co-localization studies to assess whether the distribution of IGFs within the pancreas might be due to a sequestration by locally produced IGFBPs. The presence of mRNAs for both IGFBPs-1 and -2 was minimal in the pancreas prior to postnatal day 7, although subsequently IGFBP-1 mRNA was seen in islet cells, while IGFBP-2 mRNA was localized in both islets and acinar tissues. In contrast, both IGFBPs-1 and -2 immunoreactivities were identified in islets from late fetal life, suggesting a circulatory source for these IGFBPs during early pancreatic development. IGFBPs-3 to -5 mRNAs and immunoreactivities were identified within islet cells throughout fetal and neonatal life, with IGFBPs-3 and -5 being mainly associated with the alpha cell-rich islet mantle. The results show a compartmentalization of IGFs within pancreatic tissue, reflecting both paracrine and endocrine sources. The localization and action of IGFs in pancreas likely involves sequestration and distribution by endogenous as well as circulating IGFBPs.

Interleukin-1beta (IL-1beta), tumour necrosis factor-alpha (TNF-alpha) and interferon-gamma (IFN-gamma) contribute to the initial stages of the autoimmune destruction of pancreatic beta cells. IL-1beta is released by activated macrophages resident within islets, and its cytotoxic actions include a stimulation of nitric oxide (NO) production and the initiation of apoptosis. Insulin-like growth factors (IGFs)-I and -II prevent apoptosis in non-islet tissues. This study investigated whether IGFs are cytoprotective for isolated islets of Langerhans from non-obese diabetic mice (NOD) mice exposed to cytokines. Pancreatic islets isolated from 5-6-week-old, pre-diabetic female NOD mice were cultured for 48 h before exposure to IL-1beta (1 ng/ml), TNF-alpha (5 ng/ml), IFN-gamma (5 ng/ml) or IGF-I or -II (100 ng/ml) for a further 48 h. The incidence of islet cell apoptosis was increased in the presence of each cytokine, but this was significantly reversed in the presence of IGF-I or -II (IL-1beta control 3.5+/-1.6%, IL-1beta 1 ng/ml 27.1+/-5.8%, IL-1beta+IGF-I 100 ng/ml 4.4+/-2.3%, P<0.05). The majority of apoptotic cells demonstrated immunoreactive glucose transporter 2 (GLUT-2), suggesting that they were beta cells. Islet cell viability was also assessed by trypan blue exclusion. Results suggested that apoptosis was the predominant cause of cell death following exposure to each of the cytokines. Co-incubation with either IGF-I or -II was protective against the cytotoxic effects of IL-1beta and TNF-alpha, but less so against the effect of IFN-gamma. Exposure to cytokines also reduced insulin release, and this was not reversed by incubation with IGFs. Immunohistochemistry showed that IGF-I was present in vivo in islets from pre-diabetic NOD mice which did not demonstrate insulitis, but not in islets with extensive immune infiltration. Similar results were seen for IGF-binding proteins (IGFBPs). These results suggest that IGFs protect pre-diabetic NOD mouse islets from the cytotoxic actions of IL-1beta, TNF-alpha and IFN-gamma by mechanisms which include a reduction in apoptosis.

ABSTRACT

The presence of insulin-like growth factors (IGFs) in blood is regulated by their association with specific IGF-binding proteins (IGFBPs). In turn, the level of IGFBPs in the blood is likely to depend on a dynamic equilibrium between peptide production and clearance to extravascular tissues or organ-specific degradation. Since circulating IGFBPs may largely derive from liver we have employed partial hepatectomy in the rat to study the clearance rate of endogenous IGFBPs from blood once a major site of production is removed. Adult male rats were partially hepatectomized and serum and the remaining liver removed between 30 min and 7 days after surgery. Ligand blot analysis revealed two major species of IGFBP, of 28–30 kDa and 40–44 kDa in sera from control rats or sham-operated rats respectively. The larger species corresponded in size to rat IGFBP-3, but the smaller form was not recognized by antisera against rat IGFBP-1, bovine IGFBP-2 or human IGFBP-5 following Western immunoblot. Following hepatectomy, the levels of both IGFBP forms in the serum declined within 30 min and were barely detectable after 3 h or 6 h. They began to increase again in serum 24 h following surgery. The reduction in IGFBPs following hepatectomy was not primarily due to degradation by specific proteases in serum. Circulating levels of insulin were increased fivefold 3 h after hepatectomy but subsequently returned to control values. The rise in insulin was accompanied by a significant (P < 0·05) reduction in circulating IGF-I after 3 h which persisted at 24 h. Glucose levels in serum showed a transient but non-significant reduction between 90 min and 6 h after hepatectomy. Total RNA was extracted from remnant liver and subjected to Northern blot hybridization with ³²P-labelled cDNAs encoding rat IGFBP-1, -2 or -3. Messenger RNA encoding IGFBP-1 was barely detectable in liver from control or sham-operated animals, but increased within 30 min of partial hepatectomy and peaked at 3 h. It subsequently declined and was again barely detectable after 24 h. No expression of IGFBP-2 or -3 mRNAs was found by Northern blot analysis in the liver of control animals or following partial hepatectomy. These results suggest that both IGF-I and IGFBPs in rat serum decreased rapidly following partial hepatectomy, and that this was due largely to the rapid clearance of the peptide and its binding proteins once the major source of production was removed. A rapid induction of IGFBP-1 in the remaining liver may be unrelated to the circulating IGFBPs since immunoreactive IGFBP-1 was not detected in rat serum.

Journal of Endocrinology (1993) 137, 271–280