Localization of Secretion-Related Proteins in Pancreatic Islets of Langerhans

National Institute Of Biomedical Imaging And Bioengineering, Bethesda

Investigators

Linked publications, trials & patents

Abstract

Previous cell culture work showed that DCV numbers are increased 2- or 3-fold following over expression of mammalian IA-2 in mouse MIN6 cells, as assayed by vesicle counts from electron micrographs. We have demonstrated that IDA-1/IA-2 is up-regulated in pag-3 mutants, perhaps similarly resulting in increased DCV numbers and increased neurosecretion. Although technically challenging in C. elegans, we sought to determine by electron microscopy if the overexpression of ida-1 observed in pag-3 mutants was accompanied by increased numbers of DCVs per neuron. Wild type and mutant animals were fixed, sectioned and examined by electron microscopy. We focused on the ventral nerve cord region and identified presynaptic regions by serial sections;individual neurons in any section were not precisely identified. We found that the average density of presynaptic region DCVs in pag-3 mutants was about twice the value of that in wild type controls. We also quantitated DCV numbers in ida-1 mutants and found a 50% reduction compared to wild type. Of particular interest, we found that the pag-3;ida-1 double mutants eliminated the increased DCV numbers observed in the pag-3 single mutant. Our results demonstrate that presynaptic DCV numbers correlate with neurosecretion-based phenotypes in the mutants studied. All of these enhanced phenotypes are dependent on wild type IDA-1 activity. Our current study adds to a growing body of evidence linking IA-2-related protein levels to neuroendocrine secretion and highlights the usefulness of model organisms in dissecting basic biological mechanisms. For human diseases that might benefit from increased neurosecretion, such as diabetes mellitus, targeted knockdown of Gfi-1 function represents a potential therapeutic target. Although there is substantial evidence linking IA-2 protein levels to DCV numbers and secretion, the molecular mechanisms underlying DCV homeostasis and its link to transcriptional control remain unclear. The identification of PAG-3/Gfi-1 as a component of this regulatory system provides another clue for increasing our understanding of these molecular mechanisms. To investigate further the role of IA-2, which is a major autoantigen in type-1 diabetes and is found in dense-core vesicles (DCVs) of many neuroendocrine tissues. The luminal domain of IA-2 has a significant homology with the protein RESP18, which is also found in the lumen of the endoplasmic reticulum of neuroendocrine cells. Very little is known about the function of RESP18, but it has been hypothesized that the protein is implicated in the regulation of dense-core vesicle exocytosis in pancreatic islet cells. We have tested this hypothesis by localizing RESP18 using immunoelectron microscopy with anti-RESP18 polyclonal antibody, detected by colloidal gold conjugated to secondary antibody. We found that RESP18 is mainly expressed in the lumen of DCVs, and only to a lesser extent in the ER and Golgi network. Furthermore, double labeling experiments with 5-nm and 10-nm gold nanoparticles showed that RESP18 is co-localized with insulin in pancreatic beta cell vesicles, as well as with glucagon in pancreatic alpha cell vesicles. One of the limitations of immunolabeling of plastic sections is difficulty in penetration of the labels after embedding. Experiments are therefore being performed to immunolabel cells that have been frozen and cryosectioned after cryoprotection with gelatin and sucrose (Tokuyasu method). Frozen Islets are cryosectioned using a Leica Ultracut UCT ultramicrotome with an FCS cryosectioning accessory. After thawing, immuno-gold labeling and embedding the sections, it is planned to perform electron tomography to determine the penetration of gold nanoparticles of different diameters, in multiple labeling experiments.