Research in this laboratory focuses on the regulation of biosynthesis, processing, intracellular trafficking, and secretion of neuropeptides and peptide hormones in neuronal and endocrine cells with an emphasis on the cellular organization of these biological processes. Neuropeptides and peptide hormones are synthesized from larger multivalent precursors, and the biosynthesis may be regulated at the transcriptional, post-transcriptional (RNA splicing), translational, or post-translational levels. Through these regulatory mechanisms, expression of the same gene coding for a particular precursor can result in a different set of related processed peptides in different cells. The peptides secreted ultimately dictate the identity and function of that peptidergic cell. Such regulatory mechanisms and the cellular components that make up the biosynthetic and regulated secretory pathway are studied. A recent major focus of the laboratory has been on the mechanisms by which neuropeptides, peptide hormones, and their processing enzymes are sorted at the trans-Golgi network and targeted to the regulated secretory pathway and post-Golgi transport of hormone containing vesicles to the release site. Mechanisms involved and under study include novel conformation dependant sorting signals, sorting receptors and cholesterol-rich lipid microdomains (rafts). The aim of this research is to understand molecular mechanisms of biosynthesis and sorting of proteins to the regulated secretory pathway in peptidergic neurons and endocrine cells in normal and disease states. Such studies have provided an insight into the molecular basis of diseases such as familial hyperproinsulemia and obesity which result from intracellular missorting of proinsulin and cocaine amphetamine regulated transcript (CART) respectively to the constitutive pathway.
These studies are conducted using a multidisciplinary approach. Whole animals, "knock-out " mice, fresh tissues and organs, and dissociated cell cultures are analyzed using various biochemical, pharmacological, immunological, and recombinant-DNA techniques. These include protein and peptide separation, enzymology, radioimmunoassay, immunocytochemistry, confocal microscopy, recombinant-DNA methodology, and in situ hybridization.
The membrane-bound form of Carboxypeptidase E (CPE) was identified as a sorting receptor for the regulated secretory pathway through studies carried out in our lab. We have generated a molecular model of CPE, and structural alignment with other carboxypeptidases revealed a putative sorting signal binding domain that is independent of the enzymatic site. In order to study the sorting signal binding site, we generated a number of mutant forms of CPE and expressed them in insect Sf9 cells using a baculovirus vector. We developed a binding assay using the membranes from infected cells expressing CPE, and were able to map a sorting signal binding site in CPE. Two basic amino acid residues, Arg255 and Lys260, were found to be critical for binding to the acidic amino acids on the sorting signal of pro-opiomelanocortin (POMC), p roinsulin and brain derived neurotrophic factor (BDNF). In the anterior loop of pituitary and neurons derived from CPE knock out mice, the regulated secretion of BDNF was impaired.
The focus of research in our laboratory lies primarily with characterizing the mechanisms of protein trafficking to the regulated secretory pathway in endocrine cells. Specifically, we have identified the membrane form of carboxypeptidase E (CPE) as a sorting receptor for several prohormones, including pro-opiomelanocortin (POMC), proenkephalin and p roinsulin. We have developed a model for the trafficking of p roinsulin through the regulated secretory pathway (Figure 1). Hexamers of p roinsulin bind to the sorting receptor, CPE, at the trans-Golgi network, or TGN. Within the immature secretory granule, p roinsulin is converted to insulin through the sequential actions of the prohormone convertases, PC1 and PC2. Mature insulin is then retained in the immature secretory granule, and is eventually stored within the mature secretory granule. We are making use of this model in order to determine a cellular mechanism underlying a form of hyperp roinsulinemia.
Familial hyperp roinsulinemia (FH) is a genetic condition resulting in the presence of significantly elevated levels of plasma p roinsulin. Some patients develop diabetes as a result of the increased secretory demand. Associated with this disease are a number of point mutations in the p roinsulin gene: three that result in a biologically defective form of insulin, and four (His10Asp, Arg65Leu, Arg65Pro and Arg65His) that result in the defective sorting and processing of p roinsulin. Studies from animal and cell models have shown that the sorting of His10Asp (also known as B10) is inefficient, in that it is partially diverted to the constitutive secretory pathway. However, the sorting of the other p roinsulin mutants has not been studied. We hypothesize that the mutant p roinsulins found in FH are missorted to the constitutive secretory pathway, due to impaired binding to CPE either at the level of the TGN or within the immature secretory granule. The inefficient sorting and secretion of the mutant p roinsulins would then account for the high plasma p roinsulin levels observed in FH. In order to study the sorting behavior of the mutant p roinsulins found in FH, we are engineering an endocrine cell line to produce mutant forms of p roinsulin. We make use of biochemical and cell biological techniques, such as radioimmunoassay, high pressure liquid chromatography, fluorescence immunocytochemistry and receptor binding assays, in order to examine the trafficking of the mutant p roinsulins. The results from these experiments suggest that the mutant forms of p roinsulin found in FH show some impairment in sorting to the regulated secretory pathway and in binding to CPE.
The Leu34Phe ProCART Mutation causes CART Deficiency leading to Obesity in Humans
Cocaine and amphetamine regulated transcript (CART) is an anorexigenic neuropeptide, and its absence in CART / mice results in obesity. We show that obese humans bearing a Leu34Phe mutation in pro-CART have diminished serum levels of bioactive CART, and elevated amounts of partially processed pro-CART. Leu34Phe pro-CART expressed in AtT20 cells was missorted and secreted constitutively as unprocessed proCART; while wild type pro-CART was sorted to the regulated secretory pathway vesicles and processed to CART. The defective intracellular sorting of Leu34Phe proCART would account for the reduced levels of bioactive CART in affected humans. These results suggest that the obesity observed in humans bearing the Leu34Phe mutation could be due to a putative deficiency in hypothalamic bioactive CART.
Endocrine and neuroendocrine cells synthesize prohormones and proneuropeptides, which are sorted into, and proteolytically processed in the granules of the regulated secretory pathway. The processed bioactive peptides are stored in the granules until the cell receives an external stimulus that triggers their release. This is in contrast to the constitutive secretory pathway that secretes other proteins in an unregulated manner.
We have used pro-opiomelanocortin (POMC), as our model prohormone to investigate the mechanism by which endocrine and neuroendocrine cells sort cargo to the regulated secretory pathway (RSP). Our studies show that the amino-terminal domain of POMC contains a loop structure required for the sorting of POMC to the RSP. This structure is not involved in the aggregation of POMC to form multimers, or to another prohormone, proenkephalin, indicating that aggregation alone is insufficient to effect sorting of POMC to the RSP. We now have structural data from NMR that has identified a potential motif consisting of two acidic and two hydrophobic residues within the N-terminal domain of POMC. These residues are highly conserved across species. Site-directed mutagenesis studies have confirmed the importance of these specific residues in the sorting of POMC to the RSP in PC12 cells. This motif in POMC specifically binds to a sorting receptor, the membrane form of carboxypeptidase E (CPE), to effect sorting of POMC to the RSP. Our NMR data also provided information about the molecular nature of the putative binding site on CPE. Thus by molecular modeling we were able to identify a region encompassing Arg255 and Lys260 on CPE as the sorting signal binding site. The sorting signal motif in POMC potentially represents a general consensus motif that is present also in other prohormones such as proinsulin, brain derived neurotrophic factor (BDNF), and proenkephalin . The techniques we use include transfection of wild type and mutant forms of prohormones into endocrine or neuroendocrine cells and following the path through which they get secreted, either via the constitutive or the regulated secretory pathway. Our main assays are immunofluorescent microscopy, pulse-chase radiolabeling, radio-immuno assays and western blots.
RSP Sorting Signals and Gene Therapy
We have recently been able to use knowledge gained from the identification of sorting signals and applied it to a form of gene therapy necessitating the diversion of prohormones to the constitutive pathway in the salivary gland.
After being targeted to the immature secretory granule from the TGN, prohormones are processed through the actions of the prohormone convertase enzymes, PC1 (also known as PC3) or PC2. These convertase enzymes are expressed exclusively in neural and endocrine cells, and cleave prohormones at pairs of basic amino acids. Following this endoproteolytic cleavage, the C-terminal basic residues are trimmed by the enzymatically active form of CPE to yield mature, bioactive peptide hormones.
Since the processing enzymes (PC1, PC2 and CPE) are active only within the high Ca 2+, low pH environment of the secretory granule, they must be sorted along with their prohormone substrates to the regulated secretory pathway. These enzymes may share a common mechanism for sorting that involves lipid microdomains, or "rafts". Rafts are microdomains within cell membranes that are enriched in saturated lipids, such as sphingolipids, and cholesterol. Rafts may serve to cluster protein complexes, such as those involved in signal transduction, thereby facilitating signaling. Using a number of techniques such as pulse-chase, subcellular fractionation, lipid analysis by quantitative TLC and fluorescence immunocytochemistry, we have demonstrated that cholesterol is the most abundant lipid in the membranes of the TGN and secretory granules, and that the function of CPE as a sorting receptor is dependent on cholesterol. Raft association of CPE likely facilitates its interaction with its prohormone cargo. We also found that sorting of PC1 is mostly dependent on a transmembrane domain (aa617-638). This sequence is raft associated and sufficient to target PC1 (its native molecule) or a reporter protein to the regulated secretory pathways.
Molecular modeling of these enzymes has revealed an amphipathic alpha helix within the C-terminal domains of these proteins. These regions have recently been shown to be raft associated and to mediate the targeting of PC1, PC2 and CPE to the RSP. The amphipathic alpha helical region of the processing enzymes may therefore serve dual functions: as a membrane anchor within lipid microdomains of the TGN and secretory granules, and as a sorting domain.
Secretory granules (or large dense-core vesicles, LDCVs) are unique organelle in which neuropeptides and/or hormones are packaged and stored for secretion via the regulated secretory pathway (RSP) upon stimulation in neuroendocrine and endocrine cells. Secretory granule formation is therefore a prerequisite step to fulfill the precise physiological functions of neuropeptides and hormones.
Despite vigorous studies to understand the mechanism(s) for secretory granule biogenesis for the past several decades, not much information is available so far.
One of the goals in our laboratory is to uncover the molecular mechanism(s) responsible for the biogenesis of secretory granules in neuroendocrine and endocrine cells. In order to achieve this goal, we employ techniques such as electron microscopy, confocal microscopy as well as other biochemical and cell biological methods in these studies to analyze secretory granule biogenesis. We utilize neuroendocrine and endocrine model cell lines such as PC12, AtT-20 and 6T3 cells. In addition, we utilize technologies to knock-down gene expression (i.e. antisense RNA, siRNA) in neuroendocrine cells such as PC12 to study the role of proteins involved in dense-core secretory granule biogenesis. Using an antisense RNA technology, we have identified that chromogranin A (CgA) is a key molecule involved in secretory granule biogenesis in neuroendocrine PC12 cells, and re-expression of CgA rescued the RSP and hormone secretion in endocrine 6T3 cells (Kim et al., 2001). In vivo study using transgenic mice expressing antisense RNAs against CgA has confirmed the role of CgA in secretory granule biogenesis in adrenal chromaffin cells (Kim et al., 2005). Moreover, using DNA microarray technology, real-time PCR and proteomics, we identified another key protein, protease nexin-1 (PN-1), involved in secretory granule biogenesis (Kim and Loh, 2006). PN-1 was up-regulated in the presence of CgA in 6T3 cells and prevented the degradation of granule proteins in the Golgi complex (Kim and Loh, 2006). Currently we are interested in a signal transduction pathway for PN-1 induced by CgA in AtT-20 and 6T3 endocrine cells.
Trafficking, processing and induction of granule biogenesis
Cholesterol and secretory granules
Cholesterol is an abundant lipid in eukaryotic membranes, implicated in numerous structural and functional capacities. We have investigated the mechanism by which cholesterol affects secretory granule biogenesis in vivo using Dhcr7-/- and Sc5d-/- mouse models of the human diseases, Smith-Lemli-Opitz syndrome (SLOS) and lathosterolosis. These homozygous-recessive multiplemalformation disorders are characterized by the functional absence of one of the last two enzymes in the cholesterol biosynthetic pathway, resulting in the accumulation of precursors. Cholesterol-deficient mice exhibit a significant decrease in the numbers of secretory granules in the pancreas, pituitary and adrenal glands. Moreover, there was an increase in morphologically aberrant granules in the exocrine pancreas of Dhcr7-/- acinar cells. Regulated secretory pathway function was also severely diminished in these cells, but could be restored with exogenous cholesterol. Sterol precursors incorporated in artificial membranes resulted in decreased bending rigidity and intrinsic curvature compared with cholesterol, thus providing a cholesterol-mediated mechanism for normal granule budding, and an explanation for granule malformation in SLOS and lathosterolosis.
Park, Lou, Phillips, Cawley, Loh
Post-Golgi transport of hormone and BDNF vesicles for activity-dependent secretion is important in mediating endocrine function and synaptic plasticity. We examined the role of the cytoplasmic tail of vesicular transmembrane CPE in the transport of POMC/ACTH and BDNF vesicles in the endocrine corticotrophic cell line AtT-20 and hippocampal neurons, respectively. Overexpression of the CPE cytoplasmic tail diminished localization of endogenous POMC and BDNF in the processes of AtT-20 cells and hippocampal neurons. Furthermore, live-cell imaging showed that overexpression of the CPE tail decreased both the velocity and processivity of POMC and BDNF vesicle movement in live primary anterior pituitary cells and hippocampal neurons. Our findings demonstrated that the CPE tail is involved in the processive trafficking of POMC and BDNF vesicles to the plasma membrane for secretion. We then performed pulldown experiments using both AtT-20 cell and mouse brain cytosol in vitro and showed that the cytoplasmic tail of CPE interacted with the motor adaptor protein dynactin and microtubule motors kinesin-2, kinesin-3, and cytoplasmic dynein. Moreover, competition assays using a CPE tail peptide verified specific interaction between the CPE tail and dynactin. Thus, the mechanism for the transport of POMC and BDNF vesicles to the release site for activity-dependent secretion in endocrine cells and neurons requires the interaction of vesicular CPE cytoplasmic tail to anchor these organelles to the microtubule motors.
Elucidation of molecules that control cancer cell growth and invasion will greatly facilitate the identification of biomarkers that can predict impending tumor metastasis while providing targets for therapy. Despite the huge repertoire of biomarkers reported to be useful in identifying aggressive tumors and predicting prognosis, for most cancers there is a dearth of reliable markers for predicting future metastasis from biopsies or resected primary tumors. Recently, we identified a new molecule, an alternatively spliced isoform of carboxypeptidase E (CPE-ΔN) that lacks the N-terminus present in the WT CPE. Human tumor cell lines from liver, colon, breast, prostate, colon, head, and neck that are highly metastatic showed higher expression of CPE-ΔN than did matched tumor lines with low metastatic potential. We also demonstrated that CPE-ΔN plays a role in promoting growth and cell invasion in human cancer cells. Overexpression of CPE-ΔN in the low-metastatic human hepatocellular carcinoma (HCC) cells significantly increased cell proliferation and migration by upregulating expression of a metastasis gene. When si-RNA downregulated CPE-ΔN expression in highly metastatic cell lines from breast, prostate, head and neck, colon, and liver, the si-CPE treated cells showed inhibition. In all the tumor lines, suppression of CPE expression led to a 70 to 85 percent inhibition of invasion. In vivo animal studies, mice with the si-scr-HCC-derived tumors developed intrahepatic metastasis and extrahepatic metastasis to the lung while mice inoculated with si-CPE-ΔN-HCC derived tumors had smaller tumors and failed to demonstrate metastasis. Our in vitro and in vivo results demonstrated that CPE-ΔN is a newly discovered mediator of metastasis of several human tumor cells.
To determine if CPE-ΔN is a useful biomarker for predicting recurrence and metastasis, we analyzed two retrospective cohorts of patients in blinded studies. Using quantitative RT-PCR and Western blots, we measured CPE-ΔN mRNA and protein levels in primary tumors from HCC and colon cancer patients. RT-PCR verified that only CPE-ΔN mRNA but not WT CPE mRNA was expressed in primary HCC or colonic tumors. Using qRT-PCR, we compared CPE-ΔN mRNA in the primary tumor (T) with that in surrounding non-tumor (N) tissue and determined the ratio (T/N). In 89.8 percent of HCC patients who were disease-free one year after surgery, their CPE-ΔN mRNA ratio was less than or equal to 2, whereas 92 percent of patients with extra- or intrahepatic metastasis/recurrence exhibited a T/N ratio greater than 2. Thus, CPE-ΔN is potentially a powerful prognostic biomarker for predicting future metastasis.
All related topics
All related news