We have previously shown that isolated islets embedded in type 1 collagen gel in the presence of a defined medium undergo transdifferentiation within 96 h to duct epithelial structures. The aim of this study was to identify the factors implicated in this process. Freshly isolated canine islets were embedded in type 1 collagen gel, Matrigel or agarose for up to 120 h and cultured in (i) Dulbecco's modified Eagle's medium (DMEM)/F12 plus cholera toxin (CT), (ii) medium CMRL1066 plus CT, (iii) CMRL1066 plus forskolin and (iv) CMRL1066 alone. At 16 h, intracellular levels of cAMP (fmol/10(3) islets) were increased in groups i-iii (642+/-17, 338+/-48, 1128+/-221) compared with group iv (106+/-19, P<0.01). Epithelial differentiation correlated with the total amount of intracellular cAMP measured over 120 h. Islet-epithelial transformation during the initial 36 h was associated with a wave of apoptosis which was followed by a wave of cell proliferation. During epithelial differentiation there was a progressive loss of all islet hormones and the concomitant expression of cytoskeletal proteins characteristic of duct epithelial cells. Islets in collagen and Matrigel demonstrated high rates of epithelial differentiation (63+/-2% and 71+/-4% respectively) compared with those in agarose gel (0+/-0%, P<0.001). Islets suspended in DMEM/F12 plus CT supplemented with soluble laminin or fibronectin did not undergo transformation. Prior incubation of freshly isolated islets with an integrin-binding arginine-glycine-aspartate motif-presenting synthetic peptide also reduced islet transformation. These studies confirm the biological potential of islets of Langerhans to differentiate to duct epithelial structures. cAMP-mediated signal transduction and an appropriate integrin-matrix interaction are necessary for this process to proceed.
R Wang, J Li and L Rosenberg
R Wang, J Li and N Yashpal
The limitation of available islets for transplantation is a major obstacle for the treatment of diabetes through islet therapy. However, islet monolayers expanded ex vivo may provide a source of progenitor cells and a model to help understand islet development from precursor cell types. The existence of progenitor cells within the islets is highly likely, yet, to date, no fully defined or characterized postnatal stem cell has been isolated, expanded or marked. Our study evaluates the expression of progenitor markers, including the haematopoietic stem cell marker c-Kit, in epithelial monolayers derived from postnatal rat islets through immunofluorescence and RT-PCR, and the ability of precursor-rich monolayers to reform islet-like structures. Islets formed confluent monolayers when cultured on a type I collagen gel which lacked endocrine phenotypes but were positive for cytokeratin 20 and contained an increased proportion of proliferating c-Kit-expressing cells, with the proportion reaching a maximum of 45+/-6% at 8 weeks of culture. Evaluation of transcription factors at the mRNA level revealed constant PDX-1, ngn3 and Pax4 expression, while undifferentiated cell markers, such as Oct4 and alpha-fetoprotein, were also detected frequently after 4 weeks of culture. Changing the extracellular matrix protein to laminin-rich Matrigel, the monolayers re-formed islet-like clusters that secreted insulin in a glucose-responsive fashion. Our data show that islets can be expanded ex vivo to form epithelial monolayers with rich undifferentiating cell populations that are characterized by cells expressing the progenitor markers. These monolayers are capable of extensive proliferation and retain plasticity to form new islet cells, and c-Kit-expressing cells may play an important role in new islet cluster formation.
R Wang, N Yashpal, F Bacchus and J Li
Hepatocyte growth factor (HGF) has been suggested to be a potent regulator of β-cell function and proliferation. The purpose of this study was to investigate whether HGF could regulate the proliferation and differentiation of islet-derived epithelial monolayers into insulin-producing cells. We have generated islet-derived epithelial monolayers that are enriched with cells expressing c-Kit, a tyrosine kinase receptor and putative marker, from isolated postnatal rat islets. Monolayers were cultured on type I collagen gel and treated in defined differentiation medium with or without HGF (50 ng/ml) for 7 days. Subsequently, the expression of transcription factors and pancreatic endocrine cell markers as well as c-Kit expression were compared between the HGF (HGF+), no HGF treatment (HGF−) and monolayers without differentiation medium (control) groups, using immunocytochemical and RT-PCR approaches. We observed that the number of c-Kit-, glucose transport type 2 (Glut2)- and the transcription factor pancreatic duodenal homeobox-1 (PDX-1)-expressing cells were significantly increased in the HGF+ group. The expression of insulin at the mRNA and protein level was also increased in this treatment group with a 1.7-fold increase in basal insulin release and a 2.3-fold increase in insulin content in comparison with the HGF− group. A high proliferative capacity was also found in the HGF+ group. Co-localization of insulin and PDX-1 or Glut2 was revealed frequently in cells treated with HGF+ with occasional co-staining of c-Kit and insulin observed. This study showed that HGF can activate the proliferation and differentiation of islet-derived epithelial monolayer into insulin-producing cells. However, no formation of islet-like clusters was observed. Taken together, this study implies that HGF mediates differentiation of immature cell types into insulin-expressing cells; however, HGF supplementation alone is insuffcient in restoring full β-cell function.
R Wang, J Li, N Yashpal and N Gao
There has over the last several years been renewed interest in developing a system for generating new islets and a search for a self-renewing population in the pancreas. In particular, the neural stem cell marker nestin has been implicated as an islet precursor marker and its immunoreactivity has been localized in the islets of Langerhans. This study examines islet-derived epithelial monolayers expanded ex vivo to provide a source of nestin-expressing progenitor cells – a model that will help us understand the role of nestin-expressing cells in islet cell development. When cultured on a type I collagen gel, islets formed confluent monolayers which lacked endocrine phenotypes but were positive for cytokeratin 20 and contained an increased proportion of proliferating nestin-expressing cells, reaching a maximum of 54±10%. Co-expression studies demonstrated that the nestin-positive cells are heterogeneous, with some nestin-expressing cells co-localizing with the transcription factor PDX-1 and glucose transporter type 2 or lack of co-expression with vimentin. When clonal populations of nestin-positive cells were expanded and subjected to a differentiation protocol, only a population that expressed the transcription factor PDX-1 at the mRNA level was capable of re-expressing insulin at the mRNA and protein level. In conclusion, these studies demonstrate that expanded nestin-expressing cells in vitro from islet-derived epithelial monolayers are heterogeneous; clonal analysis of nestin-positive cells reveals that a distinct subpopulation of nestin/PDX-1-expressing cells is capable of forming insulin-producing cells.
Cun Li, Thomas J McDonald, Guoyao Wu, Mark J Nijland and Peter W Nathanielsz
Neurons controlling appetite are located in the hypothalamic arcuate nuclei (ARH). Offspring appetite regulation has been shown to be modified by dysregulation of ARH nuclear development. Most ARH developmental studies have been in altricial rodents whose hypothalamic development is predominantly postnatal. In primates including humans, much development of hypothalamic appetite regulatory centers occurs before birth. We hypothesized that i) appetitive peptides are abundantly expressed by 90 percent gestation (0.9G), ready for postnatal function; ii) by 0.9G, intrauterine growth restriction (IUGR) increases the orexigenic:anorexigenic peptide ratio; iii) IUGR increases fetal glucocorticoid receptor (GR) expression; and iv) IUGR decreases STAT3, which signals inhibition of appetite. We developed a fetal baboon IUGR model resulting from reduced maternal nutrition. Pregnant baboons were fed ad libitum, controls (CTR; n=24), or 70% CTR diet to produce IUGR (n=14). C-section was performed at 0.9G. In CTR (n=7) and IUGR (n=6) fetal brains, ARH appetite regulatory peptides (neuropeptide Y (NPY) and proopiomelanocortin (POMC)) were quantified immunohistochemically. Fetal plasma cortisol was raised in IUGR fetuses. We observed that NPY and POMC were well expressed by 0.9G. IUGR increased NPY, GR, and active phosphorylated GR and decreased POMC and phosphorylated form of STAT3. We conclude that IUGR dysregulates ARH development in ways that will reset the appetitive neuropeptide balance in favor of increased appetite drive in postnatal life. We postulate that changes in peptide abundance are in part due to increased fetal cortisol and ARH GR. These changes may contribute to predisposition to obesity in IUGR offspring.
J L Környei, X Li, Z M Lei and Ch V Rao
The present study investigated the mechanisms involved in the mitogenic action of epidermal growth factor (EGF) in cultured human myometrial smooth muscle cells. The cells contained EGF/transforming growth factor-α (TGF-α) receptors as well as EGF and TGF-α mRNA transcripts and the corresponding proteins. Culturing with human EGF resulted in concentration- and time-dependent increases in cell density. The maximal increase was seen at 1 nm followed by a decrease to control levels at 100 nm EGF. The EGF increased cell density from 4 to 8 days followed by a plateau coinciding with the cells reaching confluence. EGF treatment concomitantly decreased the average size of cells. TGF-α mimicked EGF and there was no synergism between the two, suggesting a common mechanism of action. Although the presence of 10% fetal bovine serum enhanced overall cell growth, it was not required for EGF and TGF-α action. The receptor antibody, which is directed against the extracellular domain and can inhibit ligand binding to the receptors, dramatically inhibited the basal cell growth and exogenous EGF reversed the antibody effect. While TGF-α antibody was only marginally effective, EGF antibody had no effect on basal cell growth. Lavendustin (a tyrosine kinase inhibitor), calphostin (a protein kinase C inhibitor), but not H-89 (a protein kinase A inhibitor), inhibited EGF action. Indomethacin, a cyclo-oxygenase inhibitor, completely inhibited, whereas nordihydroguaiaretic acid, a lipoxygenase inhibitor, slightly inhibited EGF action. While estradiol-17β modestly inhibited basal as well as EGF-stimulated myometrial smooth muscle cell density, progesterone had no effect.
In summary, mitogenic action of EGF in human myometrial smooth muscle cells does not require serum components and it involves tyrosine kinase and protein kinase C signaling and eicosanoids from the cyclo-oxygenase pathway of arachidonic acid metabolism.
Journal of Endocrinology (1995) 146, 261–270
Qinglei Li, Fermin Jimenez-Krassel, Anilkumar Bettegowda, James J Ireland and George W Smith
Despite ample evidence pointing to an obligatory involvement of progesterone in ovulation, the mechanisms responsible for the ovulation promoting effects of intrafollicular progesterone are unclear. The objectives of this study were to determine if ovulation, luteinization and the gonadotropin surge-induced regulation of select extracellular matrix-degrading enzymes and their inhibitors, and mRNAs for prostaglandin (PG) biosynthesis and metabolizing enzymes are blocked following suppression of the intrafollicular increase in progesterone. Bovine preovulatory follicles were injected with the 3 β-hydroxysteroid dehydrogenase inhibitor trilostane or diluent and collected at 0, 12, and 24 h after GnRH induction of the preovulatory LH surge. Intrafollicular trilostane administration blocked the preovulatory increase in follicular fluid progesterone resulting in concentrations similar to those observed at time 0 post-GnRH injection. The preovulatory increase in follicular fluid PGE2 and PGF2α was reduced in trilostane-treated follicles and accompanied by upregulation of prostaglandin dehydrogenase mRNA in the granulosal and thecal cells. However, follicle rupture was not blocked by inhibition of the preovulatory rise in intrafollicular progesterone, and normal serum progesterone concentrations were observed during subsequent luteal development. Effects of trilostane administration on preovulatory changes in mRNA abundance and protein/activity in preovulatory follicles for most regulators of extracellular matrix remodeling examined were distinct from changes previously observed following the inhibition of intrafollicular prostaglandin synthesis. Results suggest that the preovulatory increase in intrafollicular progesterone may not be obligatory for bovine follicle rupture, luteinization, or regulation of prominent matrix-degrading proteinases and their inhibitors associated with ovulation.
Zengmin Wang, Gail J Mick, Rongrong Xie, Xudong Wang, Xuemei Xie, Guimei Li and Kenneth L McCormick
Both increased adrenal and peripheral cortisol production, the latter governed by 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1), contribute to the maintenance of fasting blood glucose. In the endoplasmic reticulum (ER), the pyridine nucleotide redox state (NADP/NADPH) is dictated by the concentration of glucose-6-phosphate (G6P) and the coordinated activities of two enzymes, hexose-6-phosphate dehydrogenase (H6PDH) and 11β-HSD1. However, luminal G6P may similarly serve as a substrate for hepatic glucose-6-phophatase (G6Pase). A tacit belief is that the G6P pool in the ER is equally accessible to both H6PDH and G6Pase. Based on our inhibition studies and kinetic analysis in isolated rat liver microsomes, these two aforesaid luminal enzymes do share the G6P pool in the ER, but not equally. Based on the kinetic modeling of G6P flux, the ER transporter for G6P (T1) preferentially delivers this substrate to G6Pase; hence, the luminal enzymes do not share G6P equally. Moreover, cortisol, acting through 11β-HSD1, begets a more reduced pyridine redox ratio. By altering this luminal redox ratio, G6P flux through H6PDH is restrained, allowing more G6P for the competing enzyme G6Pase. And, at low G6P concentrations in the ER lumen, which occur during fasting, this acute cortisol-induced redox adjustment promotes glucose production. This reproducible cortisol-driven mechanism has been heretofore unrecognized.
X Li, H Cui, B Sandstedt, H Nordlinder, E Larsson and T J Ekström
We have studied the insulin-like growth factor-II gene (IGF2) promoter usage in normal human liver from fetal to late adult life by quantifying the specific transcripts by RNase protection assays using exon-specific probes. While the fetal liver uses only three promoters (P2, P3, P4) for the transcription of IGF2, all four promoters can be used from the age of 2 months after birth.
The levels of the individual promoter transcripts vary substantially during development and the P3 promoter, which is a highly active fetal promoter, was not used by all the investigated adult patients but was detected in 30% of the adult group as a whole. The PI promoter, which has previously been considered as the only one responsible for IGF2 transcription in the postnatal/adult liver, displayed a trend of increasing relative and absolute activity throughout life, but in some adult cases it was found to be less active than the P4 promoter. The P4 promoter displayed an age-related trend of decreasing activity from a very high fetal level, but individual exceptions were apparent. The P2 promoter transcript, peaking at the age of 2 months, showed a relatively even absolute amount from 18 months onwards. Thus, while P2 and P3 were both found to reach their highest activity after birth, the P4 promoter displayed its highest transcription at the fetal stage.
The total IGF2 transcription, primarily from P2, P3 and P4, was found to peak shortly after birth. After this age, the P3 promoter transcript declined most rapidly and a low or zero amount was detected in adulthood. From the age of 18 months to old adulthood the total IGF2 mRNA, derived primarily from P1, P2 and P4, displayed a relatively even amount (approximately one tenth) of that seen at the peak at 2 months. This data may be important in relation to translatability of the various IGF2 transcripts.
Journal of Endocrinology (1996) 149, 117–124
J Patel, K Landers, H Li, R H Mortimer and K Richard
The development of fetal thyroid function is dependent on the embryogenesis, differentiation, and maturation of the thyroid gland. This is coupled with evolution of the hypothalamic–pituitary–thyroid axis and thyroid hormone metabolism, resulting in the regulation of thyroid hormone action, production, and secretion. Throughout gestation there is a steady supply of maternal thyroxine (T4) which has been observed in embryonic circulation as early as 4 weeks post-implantation. This is essential for normal early fetal neurogenesis. Triiodothyronine concentrations remain very low during gestation due to metabolism via placental and fetal deiodinase type 3. T4 concentrations are highly regulated to maintain low concentrations, essential for protecting the fetus and reaching key neurological sites such as the cerebral cortex at specific developmental stages. There are many known cell membrane thyroid hormone transporters in fetal brain that play an essential role in regulating thyroid hormone concentrations in key structures. They also provide the route for intracellular thyroid hormone interaction with associated thyroid hormone receptors, which activate their action. There is a growing body of experimental evidence from rats and humans to suggest that even mild maternal hypothyroxinemia may lead to abnormalities in fetal neurological development. Our review will focus on the ontogeny of thyroid hormone in fetal development, with a focus on cell membrane transporters and TR action in the brain.