Edward L. McEwen

Translational control is required for the unfolded protein response and in vivo glucose homeostasis.

The accumulation of unfolded protein in the endoplasmic reticulum (ER) attenuates protein synthesis initiation through phosphorylation of the alpha subunit of eukaryotic translation initiation factor 2 (eIF2alpha) at Ser51. Subsequently, transcription of genes encoding adaptive functions including the glucose-regulated proteins is induced. We show that eIF2alpha phosphorylation is required for translation attenuation, transcriptional induction, and survival in response to ER stress. Mice with a homozygous mutation at the eIF2alpha phosphorylation site (Ser51Ala) died within 18 hr after birth due to hypoglycemia associated with defective gluconeogenesis. In addition, homozygous mutant embryos and neonates displayed a deficiency in pancreatic beta cells. The results demonstrate that regulation of translation through eIF2alpha phosphorylation is essential for the ER stress response and in vivo glucose homeostasis.

Muscarinic receptor sequestration in SH-SY5Y neuroblastoma cells is inhibited when clathrin distribution is perturbed

Abstract: The possibility that clathrin plays a role in the agonist‐mediated sequestration of muscarinic cholinergic receptors in human SH‐SY5Y neuroblastoma cells has been investigated by the application of experimental paradigms previously established to perturb clathrin distribution and receptor cycling events. Preincubation of SH‐SY5Y cells under hypertonic conditions resulted in a pronounced inhibition of agonist‐induced muscarinic receptor sequestration (70–80% at 550 mOsm), which was reversed when cells were returned to isotonic medium. Depletion of intracellular K+ or acidification of the cytosol also resulted in >80% inhibition of muscarinic receptor sequestration. Under conditions of hypertonicity, depletion of intracellular K+, or acidification of cytosol, muscarinic receptor‐stimulated phosphoinositide hydrolysis and Ca2+ signaling events were either unaffected or markedly less inhibited than receptor sequestration. That these same experimental conditions did perturb clathrin distribution was verified by immunofluorescence studies. Hypertonicity and depletion of intracellular K+ resulted in a pronounced accumulation of clathrin in the perinuclear region, whereas acidification of the cytosol resulted in the appearance of microaggregates of clathrin throughout the cytoplasm and at the plasma membrane. The results are consistent with the possibility that muscarinic receptors in SH‐SY5Y cells are endocytosed via a clathrin‐dependent mechanism.

Cytoskeletal and phosphoinositide requirements for muscarinic receptor signaling to focal adhesion kinase and paxillin

Abstract: The mechanism whereby agonist occupancy of muscarinic cholinergic receptors elicits an increased tyrosine phosphorylation of focal adhesion kinase (FAK) and paxillin has been examined. Addition of oxotremorine‐M to SH‐SY5Y neuroblastoma cells resulted in rapid increases in the phosphorylation of FAK (t1/2 = 2 min) and paxillin that were independent of integrin‐extracellular matrix interactions, cell attachment, and the production of phosphoinositide‐derived second messengers. In contrast, the increased tyrosine phosphorylations of FAK and paxillin were inhibited by inclusion of either cytochalasin D or mevastatin, agents that disrupt the cytoskeleton. Furthermore, phosphorylation of FAK and paxillin could be prevented by addition of either wortmannin or LY‐294002, under conditions in which the synthesis of phosphatidylinositol 4‐phosphate was markedly attenuated. These results indicate that muscarinic receptor‐mediated increases in the tyrosine phosphorylation of FAK and paxillin in SH‐SY5Y neuroblastoma cells depend on both the maintenance of an actin cytoskeleton and the ability of these cells to synthesize phosphoinositides.

A Rapid Attenuation of Muscarinic Agonist Stimulated Phosphoinositide Hydrolysis Precedes Receptor Sequestration in Human SH-SY-5Y Neuroblastoma Cells*

Agonist occupancy of muscarinic cholinergic receptors in human SH-SY-5Y neuroblastoma cells elicited two kinetically distinct phases of phosphoinositide hydrolysis when monitored by either an increased mass of inositol 1,4,5-trisphosphate, or the accumulation of a total inositol phosphate fraction. Within 5s of the addition of the muscarinic agonist, oxotremorine-M, the phosphoinositide pool was hydrolyzed at a maximal rate of 9.5%/min. This initial phase of phosphoinositide hydrolysis was short-lived (t1/2=14s) and after 60s of agonist exposure, the rate of inositol lipid breakdown had declined to a steady state level of 3.4%/min which was then maintained for at least 5–10 min. This rapid, but partial, attenuation of muscarinic receptor stimulated phosphoinositide hydrolysis occurred prior to the agonist-induced internalization of muscarinic receptors.

Agonist‐Induced Endocytosis of Muscarinic Cholinergic Receptors: Relationship to Stimulated Phosphoinositide Turnover

Abstract: The ability of muscarinic cholinergic receptors to activate phosphoinositide turnover following agonist‐induced internalization has been investigated. Incubation of SH‐SY5Y neuroblastoma cells with oxotremorine‐M resulted in a time‐dependent endocytosis of both muscarinic receptors and α subunits of Gq and G11, but not of isoforms of phosphoinositide‐specific phospholipase C, into a subfraction of smooth endoplasmic reticulum (V1). Agonist‐induced increases in diacylglycerol mass and in 32P‐phosphatidate labeling, much of which was of the tetraenoic species, were also observed in the V1 fraction, but these increases persisted when the agonist‐induced translocation of receptors into the V1 fraction was blocked. All enzymes of the phosphoinositide cycle were detectable in the V1 fraction. However, with the exception of phosphatidylinositol 4‐kinase, none was enriched when compared with cell lysates. Both 32P‐labeling studies and enzyme assays point to a very limited capacity of this fraction to synthesize phosphatidylinositol 4,5‐bisphosphate, whereas the synthesis of phosphatidylinositol 4‐phosphate is robust. These results indicate that endocytosed receptors do not appear to retain their ability to activate phosphoinositide turnover. The availability of the substrate for phospholipase C, phosphatidylinositol 4,5‐bisphosphate, may be one factor that limits the activity of muscarinic receptors in this subcellular compartment.

Contribution of G protein activation to fluoride stimulation of phosphoinositide hydrolysis in human neuroblastoma cells.

Abstract: To examine the possibility that NaF enhances phosphoinositide‐specific phospholipase C (PIC) activity in neural tissues by a mechanism independent of a guanine nucleotide binding protein (Gp), we have evaluated the contribution of Gp activation to NaF‐stimulated phosphoinositide hydrolysis in human SK‐N‐SH neuroblastoma cells. Addition of NaF to intact cells resulted in an increase in the release of inositol phosphates (450% of control values; EC50 of ∼ 8 mM). Inclusion of U‐73122, an aminosteroid inhibitor of guanine nucleotide‐regulated PIC activity in these cells, resulted in a dose‐dependent inhibition of NaF‐stimulated inositol lipid hydrolysis (IC50 of ∼ 3.5 μM). When added to digitonin‐permeabilized cells, NaF or guanosine‐5′‐O‐thiotriphosphate (GTPγS) resulted in a three‐ and sevenfold enhancement, respectively, of inositol phosphate release. In the combined presence of optimal concentrations of NaF and GTPγS, inositol phosphate release was less than additive, indicative of a common site of action. Inclusion of 2–5 mM concentrations of guanosine‐5′‐O‐(2‐thiodiphosphate) (GDPβS) fully blocked phosphoinositide hydrolysis elicited by GTPγS, whereas that induced by NaF was partially inhibited (65%). However, preincubation of the cells with GDPβS resulted in a greater reduction in the ability of NaF to stimulate inositol phosphate release (87% inhibition). Both GTPγS and NaF‐stimulated inositol phosphate release were inhibited by inclusion of 10 μM U‐73122 (54–71%). The presence of either NaF or GTPγS also resulted in a marked lowering of the Ca2+ requirement for activation of PIC in permeabilized cells. These results indicate that in SK‐N‐SH cells, little evidence exists for direct stimulation of PIC by NaF and that the majority of inositol phosphate release that occurs in the presence of NaF can be attributed to activation of Gp.

Agonist-Specific Calcium Signaling and Phosphoinositide Hydrolysis in Human SK-N-MCIXC Neuroepithelioma Cells

Abstract: Fura‐2 digital imaging microfluorimetry was used to evaluate the Ca2+ signals generated in single clonal human neuroepithelioma cells (SK‐N‐MCIXC) in response to agonists that stimulate phosphoinositide hydrolysis. Addition of optimal concentrations of either endothelin‐1 (ET‐1), ATP, oxotremorine‐M (Oxo‐M), or norepinephrine (NE) all resulted in an increase in the concentration of cytosolic calcium (Ca2+i) but of different magnitudes (ET‐1 = ATP> NE). The Ca2+ signals elicited by the individual agonists also differed from each other in terms of their latency of onset, rate of rise and decay, and prevalence of a sustained phase of Ca2+ influx. The Ca2+ signals that occurred in response to ATP had a shorter latency and more rapid rates of rise and decay than those observed for the other three agonists. Furthermore, a sustained plateau phase of the Ca2+ signal, which was characteristic of the response to Oxo‐M, was observed in <40% of cells stimulated with ET‐1 and absent from Ca2+ signals elicited after NE addition. Removal of extracellular Ca2+ enhanced the rate of decay of Ca2+ signals generated by ATP, ET‐1, or Oxo‐M and, when evident, abolished the sustained phase of Ca2+ influx. In the absence of extracellular Ca2+, NE elicited asynchronous multiple Ca2+ transients. In either the absence or presence of extracellular Ca2+,>94% of cells responded to ET‐1 or ATP, whereas corresponding values for Oxo‐M and NE were ∼74 and ∼48%. Sequential addition of agonists to cells maintained in a Ca2+‐free buffer indicated that each ligand mobilized Ca2+ from a common intracellular pool. When monitored as a release of a total inositol phosphate fraction, all four agonists elicited similar (four‐ to sixfold) increases in phosphoinositide hydrolysis. However, the addition of ET‐1 or ATP resulted in larger increases in the net formation of inositol 1,4,5‐trisphosphate than did either Oxo‐M or NE. These results indicate that, in SK‐N‐MCIXC cells, the characteristics of both Ca2+ signaling and inositol phosphate production are agonist specific.

The Cellular Response to Unfolded Proteins in the Endoplasmic Reticulum

The endoplasmic reticulum (ER) is the organelle where approximately one-third of all proteins are folded and assembled into higher molecular weight structures before transit to their final destinations. In addition, the ER is the major calcium storage organelle in the cell and it responds to stimuli to release calcium into the cytosol in a regulated manner. The ER is exquisitely sensitive to alterations in homeostasis. Upon disruption of ER function, signals are initiated that propagate both adaptive and cell death responses. A number of stresses such as altered or inhibition of asparagine-linked glycosylation, reduction of disulfide bonds, expression of mutant or even some wild-type proteins, or glucose deprivation activate signaling pathways that are collectively termed the unfolded protein response (UPR). The activation of this pathway likely is important in health and disease as many genetic diseases result from defective folding of mutant proteins. In addition, protein aggregation is an initiating event for many neurodegenerative diseases, such as Alzheimer’s disease. Our recent studies also suggest that the UPR is important as a central regulator of glucose metabolism as disruptions in this pathway may lead to pancreas dysfunction and hyperglycemia. Understanding the molecular basis for this cellular response should provide avenues to intervene in disease states.