Endoplasmic Reticulum Stress, Pancreatic -Cell ...perspectivesinmedicine.cshlp.org/content/2/9/a007666.full.pdf · Endoplasmic Reticulum Stress, Pancreatic b-Cell Degeneration, and

Endoplasmic Reticulum Stress, Pancreatic b-CellDegeneration, and Diabetes

Feroz R. Papa

Department of Medicine, The Diabetes Center, The Lung Biology Center, and The California Institute forQuantitative Biosciences, University of California, San Francisco, San Francisco, California 94143-2520

Correspondence: [email protected]

Overwhelming of protein folding in the endoplasmic reticulum (ER)—referred to as “ERstress”—activates a set of intracellular signaling pathways termed the unfolded protein re-sponse (UPR). Beneficial outputs of the UPR promote adaptation in cells experiencing man-ageably low levels of ER stress. However, if ER stress reaches critically high levels, the UPRuses destructive outputs to trigger programmed cell death. Genetic mutations in various UPRcomponents cause inherited syndromes of diabetes mellitus in both rodents and humans,implicating the UPR in the proper functioning and survival of pancreatic isletb cells. Markersof chronically elevated ER stress, terminal UPR signaling, and apoptosis are evident inb cellsin these rare disorders; these markers are similarly present in islets of human patients withcommon forms of diabetes. These findings promise to enhance our molecular understandingof human diabetes significantly and may lead to new and effective therapies.

About one-third of the proteome consists ofsoluble and transmembrane proteins that

use the secretory pathway to localize to intracel-lular organelles, the plasma membrane, or theextracellular space (Gething and Sambrook1992; Gaut and Hendershot 1993). During theirbiogenesis, these secretory proteins are first in-jected into the endoplasmic reticulum (ER),wherein they must fold to their native confor-mations. Within the ER, these proteins also un-dergo various posttranslational modifications,including glycosylation and disulfide bond for-mation; abundant ER-resident enzymes cata-lyze these reactions on the secretory protein cli-ents. Molecular chaperones in the ER cyclicallybind and release the client proteins as they foldto their native shapes, shielding them from ag-

gregation. Glycosylating enzymes add and trimglycan groups, and oxido-reductases catalyzedisulfide bond formation (Sevier and Kaiser2002; Tu and Weissman 2004). Together, theseenzymatic processes maximize the probabilitythat secretory proteins are properly folded, mod-ified, and assembled into multiprotein com-plexes in the ER before they traffic farther down-stream in the secretory pathway.

The effort of these protein-folding ma-chines notwithstanding, a substantial fractionof secretory proteins normally fails to fold prop-erly in the ER. Because secretory proteins oftenmediate crucial signaling roles (e.g., cell surfacereceptors or polypeptide hormones), incom-pletely folded forms are not tolerated, and in-stead are disposed of through discriminating

Editors: Jeffrey A. Bluestone, Mark A. Atkinson, and Peter Arvan

Additional Perspectives on Type I Diabetes available at www.perspectivesinmedicine.org

Copyright # 2012 Cold Spring Harbor Laboratory Press; all rights reserved; doi: 10.1101/cshperspect.a007666

Cite this article as Cold Spring Harb Perspect Med 2012;2:a007666

1

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org

on March 28, 2021 - Published by Cold Spring Harbor Laboratory Press http://perspectivesinmedicine.cshlp.org/Downloaded from

http://perspectivesinmedicine.cshlp.org/

quality-control systems. Through a processcalled ER-associated degradation (ERAD), un-folded proteins are removed to the cytosol forsubsequent ubiquitylation and degradation bythe 26S proteasome (McCracken and Brodsky2003; Meusser et al. 2005; Smith et al. 2011). ERunfolded secretory proteins are also disposed ofthrough autophagy (Kaniuk et al. 2007; Yori-mitsu and Klionsky 2007). These quality-con-trol processes are highly stringent, and whenthey operate properly, optimal protein productsare produced and secreted by the cell.

However,cellsfrequentlyencounterenviron-mental challenges during which protein-fold-ing demand in the ER exceeds capacity. Duringsuch imbalanced states of “ER stress,” secretoryproteins start to accumulate in incompletelymodified and unfolded forms at significantlevels within the ER. Diverse challenges, bothphysiological and pathological, can provokeER stress. For instance, ischemia causes nutrientand oxygen deprivation to deplete cellular en-ergy stores, which, in turn, compromises theenergy-intensive processes of ER protein mod-ification and folding (Kaufman 2002). ER stressappears to occur naturally in cells that undergodifferentiation into specialized professionalsecretory types (e.g., terminal differentiationof B lymphocytes into immunoglobulin-pro-ducing plasma cells) (Reimold et al. 2001).

Overproduction of secretory proteins may inand of itself generate ER stress; this is especiallytrue for mutant forms of secretory proteins thatare particularly difficult to fold. For instance,pancreatic islet b cells sustain a high rate ofinsulin production and secretion, which canrise even greater in stressed states that are eitheracquired, or caused by genetic mutations (de-scribed in detail below).

The presence of unfolded proteins in the ERduring stress triggers a set of intracellular sig-naling pathways called the unfolded proteinresponse (UPR) (Bernales et al. 2006). Cellsare alerted to the presence of unfolded proteinswithin the ER by three widely expressed ERtransmembrane signaling proteins called pro-tein kinase RNA (PKR)-like ER kinase (PERK),activating transcription factor-6 (ATF6), andinositol-requiring enzyme-1 (IRE1a) (Fig. 1)(Tirasophon et al. 1998; Harding et al. 1999;Yoshida et al. 2000). These three signaling pro-teins become activated through direct and/or indirect binding of unfolded proteins(Credle et al. 2005; Zhou et al. 2006; Gardnerand Walter 2011). Combinatorial signals fromIRE1a, PERK, and ATF6 initially trigger tran-scriptional programs that up-regulate genesencoding many of the aforementioned ER chap-erones, oxidoreductases, and ERAD compo-nents (Travers et al. 2000). By increasing the

ER stress

IRE1α PERK

Lumenaldomain

KinaseKinase

ER lumen

Cytosol

bZIPtranscriptionfactor

Cell fateoutcome

RNase

ATF6

Figure 1. Proximal sensors of endoplasmic reticulum (ER) stress—IRE1a, ATF6, PERK. Combinatorial outputsfrom these three ER transmembrane sensors are integrated over time to determine cell fate outcomes under ERstress.

F.R. Papa

2 Cite this article as Cold Spring Harb Perspect Med 2012;2:a007666

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



complement of ER protein-folding and quality-control enzymes, the UPR enhances the cell’scapacity to sustain protein secretion duringtimes of high demands. The UPR also imposesa transient translational block during ER stress,thereby concentrating available resources to al-low preexisting proteins to fold before new onesare made. If these adaptive UPR outputs aresuccessful, the decline in levels of unfolded pro-teins causes UPR signaling to wane as homeo-stasis is restored (Merksamer et al. 2008). Thus,the physiological outputs of the UPR in cellsthat experience manageable levels of ER stresspromote homeostasis.

However, if ER stress persists at irremediablyhigh levels, the UPR switches its physiologicaloutputs from promoting adaptation to insteadpromoting self-destruction; this culminates inprogrammed cell death, usually through apop-tosis (Fig. 2). The commitment to cell destruc-tion appears to occur as a consequence of high-level/chronic activation of the UPR operating inalternate modes—we refer to this condition as a“terminal UPR.” During terminal UPR signal-ing, cells may transition through intermediate

dysfunctional states, before finally undergoingprogrammed cell death. For instance, some celltypes become dedifferentiated or chronically in-flamed under irremediable ER stress (Zhanget al. 2006). Stress levels in the ER are reflectedthrough the activation levels of IRE1a, PERK,and ATF6; therefore, these upstream sensors arecentrally poised to receive and transmit the in-formation needed by the cell to commit in abinary manner either to adaptation or to self-destruction.

By relegating highly stressed cells to apopto-sis, multicellular organisms may be exercisingan extreme but definitive form of protein qual-ity control. At the cost of culling some irrevers-ibly stressed cells, multicellular organisms mayderive physiological benefits. Just as reasonably,if the apoptotic process becomes too vigorous,organisms may suffer organ failure caused by aninsufficient mass of functioning cells. This par-ticular point is vividly on display in many infor-mative genetic syndromes of diabetes mellituscaused by the degeneration of insulin-producingpancreatic isletb cells in the face of irremediableER stress—these are discussed next.

Apoptosis

High-level/persistentUPR signaling

Thresholds separating

cell states

Intermediatedysfunctional

states:(dedifferentiation,inflammation, etc.)

Time

Low-level/transientUPR signaling

Homeostasis Severity ofER stress

Figure 2. Divergent cell fates result from both the magnitude and duration of ER stress signaling. Adaptive UPRoutputs can contain low levels of ER stress by reducing the concentration of unfolded proteins in the ER.However, continued activation of the UPR sensors indicates the inability to reestablish homeostasis. Dependingon the time and severity of stress experienced by the cell, thresholds separating distinct cell states are crossed. IfER stress is unrelieved, UPR signaling morphs to terminal states, promoting cell dedifferentiation and inflam-mation and eventually triggering apoptosis.

ER Stress, Pancreatic b-Cell Degeneration

Cite this article as Cold Spring Harb Perspect Med 2012;2:a007666 3

ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



UPR MUTATIONS CAUSE RARE, INHERITEDDIABETIC SYNDROMES

Pancreatic islet b cells are specialized cells thatproduce and secrete the hormone insulin in re-sponse to increases in ambient blood glucoselevels. By setting off a signal transduction cas-cade when insulin binds its receptor on insulin-responsive cells in peripheral tissues, glucoseenters these target cells, causing energy produc-tion, and removing the stimulus for further in-sulin release as blood glucose levels normalize.This tight glucostatic cycle is dysregulated in thedisease diabetes mellitus, ultimately because ofan insufficient mass of functioningbcells to pro-duce the amounts of insulin needed in the fast-ed and postprandial states for normoglycemia.

b cells contain highly developed ERs be-cause they are charged with the task of produc-ing insulin continuously during their lives. Ithas been estimated that eachb cell produces ap-proximately 1 million molecules of insulin everyminute (Scheuner and Kaufman 2008). Insulinbiogenesis requires a complex series of molecu-lar biosynthetic events that are initiated in theER (Steiner 2000). The insulin precursor, pre-pro-insulin, is cotranslationally translocatedinto the ER lumen, whereupon its signal se-quence is removed, generating pro-insulin. ER-resident oxido-reductases catalyze formation ofthree intramolecular disulfides in pro-insulin tohelp it fold to its native shape. The importanceof oxidative folding for structural maturationand trafficking of pro-insulin to the Golgi andsecretory granules is shown vividly by the “Aki-ta” mouse mutant, which cannot perform thisprocess. The Akita mouse expresses a pro-insu-lin variant gene, Ins2 (C96Y)—“Akita” insulin.Because Ins2 (C96Y) lacks a cysteine needed toform one of the intramolecular disulfide bondsthat helps it fold the ER, it can progress no far-ther than this organelle in the secretory pathway(Wang et al. 1999; Ron 2002; Izumi et al. 2003).In contrast, wild-type pro-insulin folds oxida-tively to its native state in the ER and is properlytrafficked to downstream Golgi and secretorygranules, where it is further processed by endo-proteases that remove its C-peptide to generatemature insulin (Liu et al. 2005).

Akita is the first of several textbook exam-ples considered that link ER stress to the deathof b cells and diabetes (see Fig. 3). Diabetes inthe Akita mouse is not caused directly by reduc-tion of mature insulin levels. Despite retainingthree normal insulin gene copies (mice possesstwo distinct insulin-encoding genes), the miceinstead suffer from insufficient insulin produc-tion secondary tob-cell loss. Akita insulin causesa toxic gain-of-function diabetic syndrome(Wang et al. 1999). By accumulating in the ERas a conformationally altered immature species,Akita insulin may act as a “proteotoxin” thatexhausts homeostatic UPR outputs, and insteadtriggers a terminal UPR (Ron 2002). This causesb cells in the Akita mouse to deterministicallyenter the apoptotic pathway, leading diabetes todevelop predictably �4 wk after birth (Oyado-mari et al. 2002). Rare infantile diabetes-causingAkita-like insulin mutations have been recentlydescribed in humans (Stoy et al. 2007). Intrigu-ingly, genetic removal of a downstream pro-ap-optotic UPR transcription factor called C/EBPhomologous protein (CHOP) ameliorates b-cell loss and diabetes in the Akita background,emphasizing the central role the terminal UPRplays in b-cell degeneration (Oyadomari et al.2002).

Another striking example of ER stress andUPR dysregulation causing diabetes is seen inthe Perk knockout mice. Homozygous deletionin mice of the gene encoding the UPR sensorPERK causes massive and rapid b-cell apopto-sis, leading to infantile diabetes (Delepine et al.2000; Harding et al. 2000b). Perk knockout micealso develop pancreatic exocrine insufficiencyand show growth defects early in life. These de-fects are believed to be secondary to dysfunctionand death of several different important profes-sional secretory cell types; intriguingly, diabetesmellitus is one of the earliest and most severephenotypes in the mutant animals. A rare hu-man diabetic syndrome caused by PERK-nullgene mutations (called Wolcott-Rallison syn-drome) phenocopies many of the features ofthe Perk knockout animals.

Underlying mechanisms of cell degenera-tion due to PERK deficiency are understood inconsiderable detail, although many important

F.R. Papa


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



questions remain. PERK is an ER transmem-brane kinase that dimerizes under ER stress.Dimerization causes the kinase domains totrans-autophosphorylate, which consequentlyincreases phosphorylation activity against itsdownstream substrate, the eIF2a translationinitiation factor. Activated PERK phosphory-lates eIF2a on the Ser-51 residue, which causestranslation to cease globally because it depletesthe eIF2 . GTP. Met-tRNAi complex needed toinitiate cap-dependent mRNA translation. Ithas been proposed that because of the absenceof PERK, b cells cannot properly attenuatetranslation to match ER protein-folding capac-ity. As a consequence, they suffer deposition ofunfolded proteins in the ER (Harding et al.2000b). Consistent with this notion, the ERs

of b cells in Perk2/2 mice are distended withelectron-dense proteinaceous material, and theislets show a high rate of apoptosis. Additional-ly, there appears to be a marked decline in b-cellproliferation early in neonatal life in the knock-out animals. The inability to compensate forincreased rates of apoptosis through increasingb-cell proliferation is likely to be an importantcomponent in the endocrine pancreatic failurein the Perk2/2 animals (Gupta et al. 2009).

Translation attenuation through phosphor-ylation of eIF2a also occurs through other ki-nases that sense different upstream stresses. Forinstance, GCN2 and PKR are two widely ex-pressed eIF2a kinases that become active underamino acid deprivation and accumulation ofdouble-stranded RNAs, respectively. GCN2 is

ER stress

ER unfolded proteins:Akita insulin, islet

amyloid polypeptide(IAPP)

ATF6 txn factorcleaved

Chaperone and XBP1genes trancribed

Perkkinase

activated

IRE1αkinase/RNase

activated

XBP1 mRNAspliced

eIF2αtranslation

factorphosphorylated Co-chaperone

produced

IPKp58

ATF4 txn factortranslated

ERAD factors (e.g., WFS1)/ER membrane biosynthetic

enzymes

CHOP txn factorproduced

ProapoptoticBax/Bak activated

AntiapoptoticBcl-2 proteins

repressed

Geneticsyndromes:

prodiabetic

antidiabetic

Mitochondrialcytochrome “c ”

released;effector caspases

activated

Apoptosis

ASK1/JNK1phosphorylated

ER-localizedmRNA decay

XBP1s txnfactor

Figure 3. UPR players and effects of gene mutations on b-cell degeneration and diabetes. Early adaptive eventsare shown in green, and subsequent pro-apoptotic events are shown in red. A point of no return is crossed whenmitochondria are permeabilized by the terminal gatekeepers Bax and Bak to liberate cytochrome “c”—mito-chondrial outer membrane permeabilization (MOMP). Gene mutations promoting diabetes (red asterisks)remove adaptive players in the UPR; downstream removal of pro-apoptotic CHOP is protective (green asterisk).



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



the most conserved of eIF2a kinases because itis the only one present in unicellular eukaryotes.Indeed, PERK appears to have evolved in meta-zoans through duplication and rearrangementof the genes encoding GCN2 and the most an-cient UPR sensor, IRE1a. PERK signaling maynormally be used as a glucose-sensing mecha-nism to prime the secretory apparatus for theupcoming synthesis and structural maturationof pro-insulin in response to changes in glucoselevels in the blood (Trusina et al. 2008). However,the strategy of limiting translation is fraughtwith the danger that should upstream stress re-main unrelieved, cells may never resume trans-lation to levels needed to recover viability. Thus,there are safety valves that extinguish PERK sig-naling after a time window has elapsed.

The importance for glycemic control of anescape from this global translational block isseen in two mouse mutant models: (1) Aknock-in mouse mutant expressing an unphos-phorylatable eIF2a version (S51A) develops asevere wasting syndrome shortly after life dueto arrested hepatic gluconeogenesis in the ho-mozygote (Scheuner et al. 2001), and a milderinsulin-resistant hyperglycemic syndrome inthe heterozygote (Scheuner et al. 2005). (2)p58IPK is a co-chaperone produced severalhours after the UPR has been initiated. p58IPK

may help to close a timing loop necessary toturn off the UPR by inhibiting eIF2a kinases,including PERK. Homozygous loss of the geneencoding p58IPK causes diabetes (Ladiges et al.2005; Oyadomari et al. 2006; Laybutt et al.2007). By the time p58IPK is produced, if cellshave not yet returned to homeostasis, the UPRmay switch from promoting homeostasis to anapoptotic mode. In the absence of p58IPK, acontinued translational block through PERKmay signal a frustrated UPR cycle. Together,the p58IPK, eIF2a (S51A), and PERK geneticmodels offer fascinating insights into the im-portance of temporal control of UPR signalingfor cell fate.

Although its most proximal effect is in reg-ulating translation under ER stress, PERK alsopromotes transcriptional changes through thedownstream UPR transcription factor ATF4,which becomes preferentially translated even

as cap-dependent translation declines globally.It is known that many ATF4 gene targets haveanti-oxidant functions in the cell (Harding et al.2000a). Thus, because ATF4 responses are abro-gated in the Perk2/2 animals, it is unclear towhat degree the removal of these potentiallycytoprotective outputs contributes to the dia-betic phenotype in the mice. Indeed, mouseembryonic fibroblasts derived from eitherPerk2/2 or Atf42/2 mice appear to be basallyunder high oxidative stress—defined here as ac-cumulation of reactive oxygen species (ROS). Ithas been estimated generally that �25% of cel-lular ROS originates from the ER through theprocess of oxidative protein folding. Actingthrough protein disulfide isomerases (PDIs),the ERO1 oxido-reductase transfers electronsfrom cysteine sulfhydryl groups on ER-translo-cated proteins to molecular oxygen, stoichio-metrically generating H2O2 as an ROS by-product. For professional secretory cells, ROSproduction by the ER may be even higher. Be-cause the b-cell ER needs to produce three di-sulfides for each molecule of pro-insulin it syn-thesizes (i.e., 3 million disulfides/min per cell),a large load of ROS is necessarily produced thatneeds to be continually disposed of. Futiledisulfide bond formation, due to an unpairedcysteine as in Akita insulin, is another source ofelevated ROS from the ER. It is also known thatb cells normally express limiting levels of cellu-lar enzymes that detoxify ROS (glutathione per-oxidase and catalase) (Lenzen et al. 1996), per-haps making b cells especially susceptible tooxidative stress. An ER-resident ROS-degradingenzyme called PRDX4 that uses H2O2 generatedby ERO1 in a salvage pathway to promote oxi-dation of secretory clients was recently described(Zito et al. 2010); intriguingly, overexpression ofPRDX4 in the b cell protects against a form ofexperimental diabetes caused by theb-cell toxinstreptozotocin (STZ) (Ding et al. 2010).

Another genetic example of ER stress-in-duced diabetes comes from the Wfs12/2 mouse,and in humans with Wolfram syndrome. WFS1is an ER transmembrane protein whose losscauses early-onset diabetes, neurodegeneration,and optic and auditory defects (Riggs et al.2005). WFS1 protein is widely expressed in

F.R. Papa


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



diverse tissues and is thought to aid proteinassembly and ERAD (Fonseca et al. 2005), aswell as control processing of the UPR sensorATF6 (Fonseca et al. 2005, 2010). As withPERK and p58IPK deficiency, the earliest defectsmanifest in pancreatic b cells.

The aforementioned examples amply illus-trate the principle that ER unfolded protein ac-cumulation and removal of key UPR functionspromote apoptosis in b cells to cause diabetes.These experimental genetic syndromes and rarenaturally occurring diseases link b-cell degen-eration to diabetes through mechanisms of ERstress and dysregulated UPR activity (summa-rized in Fig. 3); we predict that more such ex-amples will be found as newer mouse UPR ge-netic models are generated and human genomesequencing efforts expand. These fascinatingexperimental and rare inherited disorders man-ifest with near-complete penetrance. The com-monality of phenotype begs the question of whypancreatic b cells should be so disproportion-ately affected in UPR mutants (as comparedwith other tissues). Some thoughts on this mat-ter are offered in the next section on more wide-spread forms of human diabetes. Indeed, theexcitement of these experimental findings pro-ceeds from logically extending many of the keyconcepts to ask whether lessons learned can ad-vance our understanding of common humandiabetic syndromes—that is, types 1 and 2 andgestational diabetes. It would be exciting if thefundamental knowledge gained regarding UPRsignaling can successfully be applied and trans-lated to these human diabetes syndromes. Im-portantly, it may also allow rational therapeuticassaults to be mounted, because durable treat-ments for these diseases have remained elusive.The application of these principles for under-standing common forms of human diabetes isconsidered next.

ER STRESS AND THE UPR IN COMMONFORMS OF HUMAN DIABETES

We predict that the rare ER stress disease modelsdiscussed above will be powerfully informativefor understanding underlying mechanisms ofb-cell functional shutdown and degeneration

incommonhumandiabetes.Asindividualunits,the functioning b cells in a pancreas may expe-rience increasing and unresolvable ER stress asthey compensate for neighboring b cells thathave become dysfunctional through disease andaging. Studies confirm that b cells of mice mayalready be functioning (even in healthy states) atlevels of UPR activation that are significantlygreater than in other professional secretory cells(Iwawaki et al. 2004). Therefore, without a widemargin for further homeostatic adjustment, bcells could quickly cross a threshold that putsthem at risk for dedifferentiation and apoptosisthrough a terminal UPR.

Thus, as per-cell ER stress levels rise andterminal UPR outputs stochastically cause thedeath of individual cells, vicious cycles leadinginexorably on to whole pancreatic organ failureshould set in. In this unifying scheme (Fig. 4),the upstream stresses differ for types 1 and 2diabetes, but the downstream outcomes willbe common: For type 1 diabetes (DM1), a dis-ease that results from autoimmune attack by T-lymphocytes against b cells, as b-cell functiondegenerates, remaining cells in islets would nec-essarily have to compensate by overworking andmay themselves experience critical thresholds ofER stress. Thus, UPR-mediated apoptosis fromwithin the b cell may synergize with pro-deathprocesses initiated by autoimmune attack fromwithout through cells of the innate and adaptiveimmune system.

For type 2 diabetes (DM2), a disease pro-voked by peripheral insulin resistance, b cellsare forced to compensate by increasing insulinproduction to abnormally high levels (Karamet al. 1963). Successfulb-cell compensation mayprevent progression to frank diabetes, but insome insulin-resistant patients, the dysfunctionand death of enough b cells (roughly half of theoriginal b-cell mass) occurring over many yearsmay lead to a tipping point and organ failurebeyond which normoglycemia cannot be main-tained. The inability to compensate for declin-ing numbers of b cells through cell proliferationin some populations may contribute to this pro-cess. Interestingly, although a pre-diabetic(compensated) state may exist stably for years,progression to frank type diabetes occurs on a



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



much shorter time scale (perhaps weeks) (Ste-fan et al. 1982; Weir and Bonner-Weir 2004),indicating that an “acute-on-chronic” organshutdown may be occurring; the often dramaticinitial presentation of type 1 diabetes is alsoconsistent with such dynamics. Intriguingly,both disorders may have a long metastable pe-riod of failing compensation, which perhapscould be exploited through intervention inthose at risk (as in the Diabetes PreventionProgram, DPP) (Knowler et al. 2002). Althoughtype 1 diabetes has traditionally been regardedprimarily as a discrete and homogeneous disor-der (when compared with the greater heteroge-neity evident in b-cell function of patients withtype 2 diabetes), recent findings of persistent C-peptide production and preserved b-cell func-tion decades after diagnosis in distinct groups ofpatients may indicate inherent differences inpopulations with regard to the ability to toleratestress in the b cell (Wang et al. 2012).

Experimental lessons gleaned from the rare,deterministic UPR diabetic disorders, describedin the last section, must be interpreted and ap-plied cautiously to common human diabetes(both types 1 and 2), which occur heteroge-neously and probabilistically, and after alsocarefully considering other disease factors thatare likely at play. Because it evolved through

the study of lower eukaryote models such asyeast, the study of signaling in the UPR and itsphysiological consequences is often confinedand ascribed to cell-autonomous effects (Walterand Ron 2011). However, it is very likely thatchanges in cell–cell signaling caused by ERstress strongly impact inter-organ functioningin mammals. For instance, ER stress signaling ininflamed insulin-responsive target tissues suchas liver and fat was shown to reduce insulinsignaling (Ozcan et al. 2004); this should havethe effect of increasing the burden for insulinproduction “at a distance” in b cells.

Additionally, dysregulated UPR signaling inimmune cells that neighbor b cells in the isletsmay also be pathogenic. The secretory pathwayis the gateway for the structural maturation ofabout one-third of the proteome; therefore,small changes in ER stress signaling and UPRoutputs could have widespread effects. As im-mune cells operate through cell-surface recep-tors and soluble secreted proteins, UPR defectsleading to immune cell dysfunction could con-ceivably promote loss of self-tolerance againstb cells. In addition, local inflammation in theislet from pro-inflammatory cytokines such asinterleukin 1-b (IL-1b) released locally in theislet by cells of the innate immune system canadversely affect b-cell function (also see below);

ER stressDiabetes

Apoptosis

Autoimmuneattack onβ cells

(type 1)

Peripheralinsulin

resistance(type 2) Unfolded

proteinresponse

(UPR)

β cellsoverworked Homeostasis

β celldysfunction/

death

Oxidativedamage (ROS)/inflammation(IL-1β)

Figure 4. Conjectural scheme for a central role of ER stress and divergent UPR signaling in human diabetes (bothtypes 1 and 2). See text for details.

F.R. Papa


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



interestingly, a link between upstream ER stressand IL-1b was recently noted (Menu et al.2012).

UPR studies are often performed with un-natural ER stress agents that deterministicallypush cells rapidly into apoptosis. More careshould be given to studying subtle disease-rele-vant in vivo defects that develop over longerperiods in murine models. Measuring ER stressin b cells of living humans is currently not fea-sible, because the pancreatic islet b-cell massand insulin secretory function during life can-not be studied simultaneously. However, pan-creatic autopsy series from type 2 diabetic pa-tients clearly show a reduction in b-cell mass(Yoon et al. 2003) concomitant with activationin UPR apoptotic markers. In addition, humantype 2 diabetic islets contain protein aggregatesin the form of amyloid (Westermark et al. 1992).Islet amyloid is composed of a 37-residue amy-loidogenic polypeptide called islet amyloidpolypeptide (IAPP). IAPP spontaneously formsER membrane-damaging sheets of amyloid (Sa-waya et al. 2007). Therefore, IAPP could be an-other ER stress link promoting b-cell deaththrough activation of the pro-apoptotic tran-scription factor CHOP (Huang et al. 2007).

Oxidative stress and inflammation engen-dered by ER stress may further drive a terminalUPR into a common terminal pathway leadingto b-cell demise in both types 1 and 2 diabetes.Indeed, one recent study linked IAPP to produc-tion of pro-inflammatory 1L-1b by the NLRP3inflammasome, a multiprotein complex that re-sponds to internal danger signals (Masters et al.2010). We propose a unified model that inte-grates oxidative and inflammatory damage asdriving links and amplifiers for ER stress inthe pathogenesis of common forms of humandiabetes—types 1 and 2 (Fig. 4). Similarly, theseprinciples could apply during gestation, duringwhich increasing insulin requirements necessi-tate an approximate doubling of the b-cell massin pregnant mothers (Karnik et al. 2007). Insome individuals (perhaps those who are al-ready at risk for DM2), this could increase riskfor gestational diabetes (GDM) (Buchanan andXiang 2005). Indeed, it is well appreciated thatGDM puts affected mothers at risk for DM2

later in life, perhaps by reducing the existingbaseline mass of b cells after parturition.

KEY COMPONENTS OF THE UPRAPOPTOTIC SWITCH AND THERAPEUTICSTRATEGIES

We turn now to a more rigorous and mechanis-tic review of how the UPR appears to switchfrom promoting homeostasis to instead causingcell destruction and apoptosis (see Fig. 5)(Merksamer and Papa 2010). We propose thatmany key components of this terminal UPRswitch may pose attractive targets for drugging.

As mentioned above, the UPR is triggeredby three ER-resident transmembrane proteins:PERK, ATF6, and IRE1a. The presence of un-folded ER proteins is thought to activate each ofthese three proximal detectors through directand/or indirect sensing mechanisms (Kohno2007). IRE1a is a type-I transmembrane pro-tein that contains three domains: an amino-ter-minal domain that resides in the ER lumen andsenses unfolded proteins, a cytosolic serine/threonine kinase domain, and a carboxy-termi-nal cytosolic endoribonuclease (RNase) domain(Tirasophon et al. 1998; Wang et al. 1998).IRE1a becomes activated when monomers oli-gomerize into either dimers or higher-orderoligomers, causing trans-autophosphorylationof the kinase domains, which, in turn, activatesthe RNase domains. Two opposing models havebeen proposed to explain how IRE1a changesits oligomerization state. It has been suggestedthat the ER-resident chaperone immunoglobu-lin-binding protein (BiP) functions as a key reg-ulator by binding to IRE1a and inhibiting itsoligomerization. When unfolded proteins accu-mulate, BiP dissociates from IRE1a to preferen-tially engage and refold them, freeing up IRE1ato oligomerize (Bertolotti et al. 2000). The sec-ond model proposes that unfolded proteinsbind directly to the lumenal domain of IRE1a,which, in turn, induces its oligomerization(Credle et al. 2005; Gardner and Walter 2011).Structural data are lacking to definitively rule in(or rule out) each of the two models. BecausePERK has an amino-terminal lumenal domainhomologous to IRE1a, it is very likely to



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



PPP

P

Unfolded proteins

Unfolded proteins

A

B

SH SH

SH SHSH

SH SH SH SH SH SH

SHSH

SH

HS

SH SH SH SH SH

PERKBiP

SustainedmRNA degradation

mRNA

AAAA

elF2α

ATF4

CHOP

Antiapoptotic Bcl2proteins(Bcl2, Bcl-XL, Mcl-1)

Sustainedtranslationattenuation

Caspase 2ASK1

JNK

Bim

NOXA

PUMA Bid

Bim

Bax and BakMitochondria

Cytochrome c release and activation ofapoptosome and effector caspases

SH

BiP SH SH PERK

ER lumen

Low ER stress

HS

Amino acid import

Glutathione biosynthesis

High ER stress

IREIα

IREIα

TransientmRNAdegradation

5′

5′5′

5′

AAAA

AAAA AAAA

IntronXBP1u mRNA

XBP1sXBP1s mRNA

mRNARibosome

Translocon Cytosol

elF2α

ATF4

ATF6 (N) XBP1s Chaperones

Transienttranslationattenuation

Modifiying enzymes

ERAD components

ATF4

Lipid biosynthesis

P

P

PP PP

P

P

PP

PP

PP PPP

P PPP

PPP

P PPP

SH SH

PPP P

PPP P

PPPP

Figure 5. (See legend on facing page.)

F.R. Papa


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



become activated by similar principles (Berto-lotti et al. 2000; Liu et al. 2000).

A different mechanism governs ATF6 activa-tion, which is an ER-resident type-II transmem-brane protein that exists basally as an oxidizedmonomer, dimer, and/or oligomer associatedwith BiP. Under ER stress, ATF6 dissociatesfrom BiP and conserved intramolecular and/or intermolecular disulfide bonds in the lu-menal domain of ATF6 become reduced. Theresulting reduced ATF6 monomers translocateto the Golgi and become cleaved by the Site-1and Site-2 proteases (Haze et al. 1999; Shenet al. 2005; Nadanaka et al. 2007). This liberatesthe amino-terminal cytosolic fragment of ATF6,producing a soluble variant called ATF6(N),which is a basic leucine zipper (bZiP) transcrip-tion factor. ATF6(N) transcriptional targets ap-pear to be largely adaptive because they includegenes encoding chaperones as well as the mRNAencoding the adaptive XBP1 transcription fac-tor (Yoshida et al. 2001, 2003; Rutkowski et al.2006; Wu et al. 2007) (see below). The conse-quences of PERK and IRE1a signaling, however,appear to be context and time dependent, andthey lead to divergent cell fate outcomes basedon these variables.

PERK and IRE1a may operate as “stress in-tegrators” that also use a time variable to deter-mine if cells have crossed a threshold beyondwhich a different cell fate outcome is triggered.We have proposed that if adaptive outputs fromPERK and IRE1a succeed in reducing ER stress,homeostasis will be restored and the cell will beaverted from downstream destructive out-comes. The adaptive responses through PERKand IRE1a involve several outputs and can beconceptualized as two negative-feedback loopsacting on two different time scales: a fast nega-tive-feedback loop that rapidly decreases the

influx of proteins into the ER; and a slow neg-ative-feedback loop that requires de novomRNA and protein synthesis to increase thefolding capacity of the ER, and is thereforeslower (Trusina et al. 2008). As shown in Figure2, the amplitude and/or strength of the pertur-bant stress is an additional variable that willdetermine if cells have crossed a cell fate-deter-mining threshold (i.e., a strong stress requiresless time than a weaker stress to push cells intoapoptosis; conversely, weak stresses may buildup over time to produce the same effect asa strong, short stress). Activated PERK phos-phorylates eIF2a, which impedes subsequentrounds of translation initiation (Harding et al.1999). In addition, IRE1a rapidly degradesseveral hundred ER-localized mRNAs uponactivation (Hollien and Weissman 2006; Hanet al. 2009). Transient translation attenuation(through PERK) and mRNA decay (throughIRE1a) should constitute a fast negative-feed-back loop because together they should rapidlyreduce the protein load on the ER. This providesthe ER’s folding machinery an extended oppor-tunity to fold existing unfolded proteins and theERAD machinery an extended time period todegrade them.

Additionally, both PERK and IRE1a havesynthetic outputs, which also work as slowernegative-feedback loops. For instance, althoughPERK activation inhibits cap-dependent trans-lation, some mRNAs having small upstreamopen reading frames (uORFs) become prefer-entially translated during this block. For exam-ple, as mentioned earlier, the pro-survival tran-scription factor ATF4 has a target set thatincludes activities that attenuate oxidative stress(Harding et al. 2000a). The synthetic output ofIRE1a occurs when it catalyzes the unconven-tional splicing of XBP1u mRNA into XBP1s

Figure 5. (Continued) Molecular details of the UPR homeostatic–apoptotic switch. (A) adaptive UPR eventsreduce protein load acutely through reversible translational attenuation and mRNA decay (mediated by PERKand IRE1a, respectively). Synthesis of gene products through de novo transcription/translation enhances ERprotein-folding functions and ER-associated degradation (ERAD). (B) The terminal UPR occurs when ERunfolded proteins cannot be sufficiently reduced, and therefore UPR sensor signaling is not quelled. In thatinstance, destructive UPR outputs occur through continued IRE1a mRNA endonucleolytic decay and contin-ued translational blocks through PERK. Downstream amplification of the terminal UPR occurs through JNKand CHOP impinging on Bax and Bak to promote MOMP.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



mRNA, which encodes the bZIP transcriptionfactor X-box-binding protein 1 (XBP1s) (Yoshi-da et al. 2001; Calfon et al. 2002). ATF6(N) andXBP1s work together to increase transcriptionrates of genes encoding ER-resident chaperones,protein-modification enzymes, ERAD compo-nents, and lipid biosynthetic enzymes to aug-ment ER size and to increase the ER’s foldingand degradation activities (Yamamoto et al.2007). Collectively, these negative-feedbackloops should reduce the concentration of un-folded proteins in the ER, and as the concentra-tion of unfolded proteins decreases, the UPRshuts off.

The conversion of adaptive UPR signalinginto destruction may occur as the negative-feed-back loops collapse and morph into feed-forwarddestructive loops under high ER stress (lowerpanel of Fig. 5). Events heralding this con-version include transcription of the late ATF4target, the pro-apoptotic transcription factorCHOP. That CHOP is clearly an important me-diator between the UPR and the apoptotic ma-chinery was shown by protection of Chop2/2

mice against ER-stress-induced b-cell apoptosisin several experimental models of diabetes (Songet al. 2008). Another definitive step markingentry into apoptosis occurs when the outermitochondrial membrane (OMM) is permea-bilized and cytochrome c is released to acti-vate executioner caspases—designated MOMP(Shore et al. 2011). This intrinsic (mitochondri-al) apoptotic pathway is typically triggered inresponse to intracellular stresses including DNAdamage and viral infections and is regulated bythe Bcl-2 protein family (Youle and Strasser2008).

The Bcl-2 family is divided into three groups:multidomain pro-apoptotic proteins (e.g., BAX,BAK), anti-apoptotic proteins (e.g., BCL-2,BCL-XL), and pro-apoptotic BH3-only proteins(e.g., BID, BAD, BIM, NOXA, PUMA) (Brunelleand Letai 2009). In response to ER stress, thepro-apoptotic BH3-only proteins are transcrip-tionallyorposttranslationallyactivatedtostimu-late pro-apoptotic BAX and BAK either directlyor indirectly through antagonizing anti-apop-totic members. Once activated, BAX and/orBAK form homo-oligomers in the OMM to

initiate MOMP (Wei et al. 2001). Of the BH3-only family, PUMA, NOXA, BID, and BIM havebeen implicated in apoptosis triggered by ERstress (Li et al. 2006; Puthalakath et al. 2007;Upton et al. 2008). Moreover, the adaptor pro-tein CRK has recently been shown to contain a“BH3-like” domain that undergoes proteolyticprocessing under ER stress and is required forefficient ER stress-induced apoptosis (Austgenet al. 2012). CHOP mRNA levels increase sharp-ly during a terminal UPR through up-regulationby the transcription factor ATF4 downstreamfrom PERK and eIF2a. Providing an importantlink between the terminal UPR and a BH3-onlyprotein, CHOP increases Bim transcription dur-ing ER stress (Puthalakath et al. 2007). In addi-tion to regulating Bim expression, CHOPantag-onizes the expression of anti-apoptotic Bcl-2.

Another terminal UPR component may bethe mitogen-activated protein kinase (MAPK)c-Jun amino-terminal kinase (JNK) operatingdownstream from IRE1a and in parallel withthe PERK–ATF4–CHOP arm. JNK is activatedby cytokines and several intracellular stressesand can promote apoptosis when hyperacti-vated (Weston and Davis 2007). JNK signalingincreases downstream from IRE1a, the tumornecrosis factor receptor-associated factor 2(TRAF2), and the MAP3K apoptosis signal-reg-ulating kinase 1 (ASK1) (Urano et al. 2000;Nishitoh et al. 2002). JNK may promote apop-tosis by interacting with Bcl-2 family members;specifically, JNK may phosphorylate and inhibitthe anti-apoptotic proteins BCL-2, BCL-XL,and MCL-1. Furthermore, JNK can also phos-phorylate and activate several BH-3 only pro-teins, including BID and BIM to promote apop-tosis (Weston and Davis 2007).

Because the UPR simultaneously transmitssurvival and apoptotic outputs, understandingthe interplay between these competing signalsis necessary to elucidate the mechanism bywhich cells decide whether to continue to at-tempt adaptation or to initiate cell death. Wehave proposed that tonic high-level activationof both PERK and IRE1awill signal an inabilityto adapt and initiate the aforementioned de-structive outputs. In support of this notion ex-perimentally, overexpression of either PERK or

F.R. Papa


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



IRE1a, which leads to their spontaneous oligo-merization and activation, is typically sufficientto cause apoptosis. Intertwining with the Bcl-2family, sustained PERK activity may be neces-sary to build CHOP levels to a required thresh-old to stimulate Bcl-2 proteins to commit toapoptosis. In addition, sustained PERK activityshould result in protracted translation attenua-tion, which would be incompatible with cell sur-vival. Similarly, sustained mRNA degradationmediated by hyperactivated IRE1a may depleteER cargo and protein-folding activities (Hanet al. 2009). This last point may have furthersignificance for b-cell degeneration in that deg-radation of insulin mRNA during high ER stressby IRE1a in any oneb cell, as we and others haveshown (Pirot et al. 2007; Lipson et al. 2008;Han et al. 2009), should simply shift the burdenof insulin production to another b cell, therebypromoting organ failure. This may occur aswhole-scale mRNA decay by IRE1a at the ERmembrane consumes the very mRNAs encodingthe very same activities needed to salvage ERfunction (chaperones, oxidoreductases, etc.).

We conclude this review by highlightingsome select UPR targets that could be modulat-ed to bias toward cell survival. We propose thatUPR components present rich and attractivetargets for pharmacological intervention. How-ever, the parallel and cross-wired networks mayrequire several nodes to be targeted simultane-ously to effect robust therapeutic effects. Nev-ertheless, for cell degenerative diseases such asdiabetes, the biasing of the UPR homeostatic–apoptotic switch to favor cell survival couldpotentially be disease-modifying, thus payoffsmay be large. One strategy is to prolong theadaptive phases of the UPR to maximize chanc-es of recovery (examples of such targets wouldinclude XBP1 and ATF6). Another route wouldbe to inhibit key mediators of apoptosis (CHOP,BAX, BAK). Related to this, the strategy of pre-conditioning by preemptive adaptive UPR acti-vation could be beneficial. Potential timerssuch as p58IPK may also pose attractive targets.Indeed, a small molecule called salubrinal wasshown to block phosphatases mediating eIF2adephosphorylation to enhance cell survival un-der ER stress (Boyce et al. 2005).

Similarly, preemptive activation of the ki-nase PERK was shown through a proof-of-con-cept strategy (using synthetic dimerizable mod-ules) to enhance survival in cell culture modelsof ER and oxidative stress (Lu et al. 2004). It isunclear, however, whether long-term PERK ac-tivation with its attendant consequences of in-hibiting translation could be a viable strategyfor cytoprotection in vivo. In addition, we(Han et al. 2008) and others (Lin et al. 2007)have shown that IRE1a activation can prolongsurvival under ER stress. The basis for this par-ticular strategy rests on a highly unusual rela-tionship between IRE1a’s two catalytic domainsthat we discovered (Papa et al. 2003). We showedthat the kinase of IRE1a can be engaged with adesigner kinase inhibitor to trigger RNase ac-tivity and force splicing of the XBP1 mRNA,leading to production of the pro-survival XBP1stranscription factor, while bypassing the kinaseactivity. Cells subjected to these maneuvers pre-emptively enjoy a small, but significant, mea-sure of cytoprotection when they are challengedby ER stress (Han et al. 2008). However, kinase-active versions of IRE1a that become hyperac-tivated in their RNase domain cause massiveendonucleolytic degradation of mRNAs local-izing to the ER membrane during cotransla-tional translocation of protein products (Hanet al. 2009). This event was shown to push cellsinto a terminal UPR. Inhibition of the RNasedomains by small-molecule direct inhibitorsmay be therapeutic in these contexts (Papan-dreou et al. 2010).

Finally, it has been shown that small chem-ical chaperones that could act as templates forprotein folding and that may have global effectson stabilizing protein conformations also affordsignificant protection, directly in animals, toforestall diabetes (Ozcan et al. 2006).

In conclusion, ER stress and its remedia-tion in pancreatic b cells, as a set of relatedtopics, will continue to attract attention fromexperts in fields ranging from cell biology topathology, signal transduction, epidemiology,pharmacology, and medicine. We predict thatthe next few years will bring many more chal-lenges, as well as opportunities for discoveryand invention.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



REFERENCES

Austgen K, Johnson ET, Park TJ, Curran T, Oakes SA. 2012.The adaptor protein CRK is a pro-apoptotic transducerof endoplasmic reticulum stress. Nat Cell Biol 14: 87–92.

Bernales S, Papa FR, Walter P. 2006. Intracellular signalingby the unfolded protein response. Annu Rev Cell Dev Biol22: 487–508.

Bertolotti A, Zhang Y, Hendershot LM, Harding HP, Ron D.2000. Dynamic interaction of BiP and ER stress transduc-ers in the unfolded-protein response. Nat Cell Biol 2:326–332.

Boyce M, Bryant KF, Jousse C, Long K, Harding HP, Scheu-ner D, Kaufman RJ, Ma D, Coen DM, Ron D, et al. 2005.A selective inhibitor of eIF2adephosphorylation protectscells from ER stress. Science 307: 935–939.

Brunelle JK, Letai A. 2009. Control of mitochondrial apo-ptosis by the Bcl-2 family. J Cell Sci 122: 437–441.

Buchanan TA, Xiang AH. 2005. Gestational diabetes melli-tus. J Clin Invest 115: 485–491.

Calfon M, Zeng H, Urano F, Till JH, Hubbard SR, HardingHP, Clark SG, Ron D. 2002. IRE1 couples endoplasmicreticulum load to secretory capacity by processing theXBP-1 mRNA. Nature 415: 92–96.

Credle JJ, Finer-Moore JS, Papa FR, Stroud RM, Walter P.2005. On the mechanism of sensing unfolded protein inthe endoplasmic reticulum. Proc Natl Acad Sci 102:18773–18784.

Delepine M, Nicolino M, Barrett T, Golamaully M, LathropGM, Julier C. 2000. EIF2AK3, encoding translation ini-tiation factor 2-a kinase 3, is mutated in patients withWolcott-Rallison syndrome. Nat Genet 25: 406–409.

Ding Y, Yamada S, Wang KY, Shimajiri S, Guo X, TanimotoA, Murata Y, Kitajima S, Watanabe T, Izumi H, et al. 2010.Overexpression of peroxiredoxin 4 protects against high-dose streptozotocin-induced diabetes by suppressing ox-idative stress and cytokines in transgenic mice. AntioxidRedox Signal 13: 1477–1490.

Fonseca SG, Fukuma M, Lipson KL, Nguyen LX, Allen JR,Oka Y, Urano F. 2005. WFS1 is a novel component of theunfolded protein response and maintains homeostasis ofthe endoplasmic reticulum in pancreatic b-cells. J BiolChem 280: 39609–39615.

Fonseca SG, Ishigaki S, Oslowski CM, Lu S, Lipson KL,Ghosh R, Hayashi E, Ishihara H, Oka Y, Permutt MA,et al. 2010. Wolfram syndrome 1 gene negatively regulatesER stress signaling in rodent and human cells. J ClinInvest 120: 744–755.

Gardner BM, Walter P. 2011. Unfolded proteins are Ire1-activating ligands that directly induce the unfolded pro-tein response. Science 333: 1891–1894.

Gaut JR, Hendershot LM. 1993. The modification and as-sembly of proteins in the endoplasmic reticulum. CurrOpin Cell Biol 5: 589–595.

Gething M-J, Sambrook J. 1992. Protein folding in the cell.Nature 355: 33–45.

Gupta S, McGrath B, Cavener DR. 2009. PERK regulates theproliferation and development of insulin-secreting b-celltumors in the endocrine pancreas of mice. PLoS ONE 4:e8008.

Han D, Upton JP, Hagen A, Callahan J, Oakes SA, Papa FR.2008. A kinase inhibitor activates the IRE1a RNase toconfer cytoprotection against ER stress. Biochem BiophysRes Commun 365: 777–783.

Han D, Lerner AG, Vande Walle L, Upton JP, Xu W, Hagen A,Backes BJ, Oakes SA, Papa FR. 2009. IRE1a kinase acti-vation modes control alternate endoribonuclease out-puts to determine divergent cell fates. Cell 138: 562–575.

Harding HP, Zhang Y, Ron D. 1999. Protein translation andfolding are coupled by an endoplasmic-reticulum-resi-dent kinase. Nature 397: 271–274.

Harding HP, Novoa I, Zhang Y, Zeng H, Wek R, Schapira M,Ron D. 2000a. Regulated translation initiation controlsstress-induced gene expression in mammalian cells. MolCell 6: 1099–1108.

Harding HP, Zhang Y, Bertolotti A, Zeng H, Ron D. 2000b.Perk is essential for translational regulation and cell sur-vival during the unfolded protein response. Mol Cell 5:897–904.

Haze K, Yoshida H, Yanagi H, Yura T, Mori K. 1999. Mam-malian transcription factor ATF6 is synthesized as a trans-membrane protein and activated by proteolysis in re-sponse to endoplasmic reticulum stress. Mol Biol Cell10: 3787–3799.

Hollien J, Weissman JS. 2006. Decay of endoplasmic reticu-lum-localized mRNAs during the unfolded protein re-sponse. Science 313: 104–107.

Huang CJ, Lin CY, Haataja L, Gurlo T, Butler AE, Rizza RA,Butler PC. 2007. High expression rates of human isletamyloid polypeptide induce endoplasmic reticulumstress-mediated b-cell apoptosis, a characteristic of hu-mans with type 2 but not type 1 diabetes. Diabetes 56:2016–2027.

Iwawaki T, Akai R, Kohno K, Miura M. 2004. A transgenicmouse model for monitoring endoplasmic reticulumstress. Nat Med 10: 98–102.

Izumi T, Yokota-Hashimoto H, Zhao S, Wang J, Halban PA,Takeuchi T. 2003. Dominant negative pathogenesis bymutant proinsulin in the Akita diabetic mouse. Diabetes52: 409–416.

Kaniuk NA, Kiraly M, Bates H, Vranic M, Volchuk A, Bru-mell JH. 2007. Ubiquitinated-protein aggregates form inpancreatic b-cells during diabetes-induced oxidativestress and are regulated by autophagy. Diabetes 56:930–939.

Karam JH, Grodsky GM, Forsham PH. 1963. Excessive in-sulin response to glucose in obese subjects as measured byimmunochemical assay. Diabetes 12: 197–204.

Karnik SK, Chen H, McLean GW, Heit JJ, Gu X, Zhang AY,Fontaine M, Yen MH, Kim SK. 2007. Menin controlsgrowth of pancreatic b-cells in pregnant mice and pro-motes gestational diabetes mellitus. Science 318: 806–809.

Kaufman RJ. 2002. Orchestrating the unfolded protein re-sponse in health and disease. J Clin Invest 110: 1389–1398.

Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF,Lachin JM, Walker EA, Nathan DM. 2002. Reduction inthe incidence of type 2 diabetes with lifestyle interventionor metformin. N Engl J Med 346: 393–403.

F.R. Papa


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



Kohno K. 2007. How transmembrane proteins sense endo-plasmic reticulum stress. Antioxid Redox Signal 9: 2295–2303.

Ladiges WC, Knoblaugh SE, Morton JF, Korth MJ, SopherBL, Baskin CR, MacAuley A, Goodman AG, LeBoeuf RC,Katze MG. 2005. Pancreatic b-cell failure and diabetes inmice with a deletion mutation of the endoplasmic retic-ulum molecular chaperone gene P58IPK. Diabetes 54:1074–1081.

Laybutt DR, Preston AM, Akerfeldt MC, Kench JG, BuschAK, Biankin AV, Biden TJ. 2007. Endoplasmic reticulumstress contributes to b cell apoptosis in type 2 diabetes.Diabetologia 50: 752–763.

Lenzen S, Drinkgern J, Tiedge M. 1996. Low antioxidantenzyme gene expression in pancreatic islets comparedwith various other mouse tissues. Free Radic Biol Med20: 463–466.

Li J, Lee B, Lee AS. 2006. Endoplasmic reticulum stress-induced apoptosis: Multiple pathways and activation ofp53—Up-regulated modulator of apoptosis (PUMA)and NOXA by p53. J Biol Chem 281: 7260–7270.

Lin JH, Li H, Yasumura D, Cohen HR, Zhang C, Panning B,Shokat KM, LaVail MM, Walter P. 2007. IRE1 signalingaffects cell fate during the unfolded protein response.Science 318: 944–949.

Lipson KL, Ghosh R, Urano F. 2008. The role of IRE1a in thedegradation of insulin mRNA in pancreatic b-cells. PLoSONE 3: e1648.

Liu CY, Schroder M, Kaufman RJ. 2000. Ligand-indepen-dent dimerization activates the stress response kinasesIRE1 and PERK in the lumen of the endoplasmic reticu-lum. J Biol Chem 275: 24881–24885.

Liu M, Li Y, Cavener D, Arvan P. 2005. Proinsulin disulfidematuration and misfolding in the endoplasmic reticu-lum. J Biol Chem 280: 13209–13212.

Lu PD, Jousse C, Marciniak SJ, Zhang Y, Novoa I, ScheunerD, Kaufman RJ, Ron D, Harding HP. 2004. Cytoprotec-tion by pre-emptive conditional phosphorylation oftranslation initiation factor 2. EMBO J 23: 169–179.

Masters SL, Dunne A, Subramanian SL, Hull RL, TannahillGM, Sharp FA, Becker C, Franchi L, Yoshihara E, Chen Z,et al. 2010. Activation of the NLRP3 inflammasome byislet amyloid polypeptide provides a mechanism for en-hanced IL-1b in type 2 diabetes. Nat Immunol 11:897–904.

McCracken AA, Brodsky JL. 2003. Evolving questions andparadigm shifts in endoplasmic-reticulum-associateddegradation (ERAD). Bioessays 25: 868–877.

Menu P, Mayor A, Zhou R, Tardivel A, Ichijo H, Mori K,Tschopp J. 2012. ER stress activates the NLRP3 inflam-masome via an UPR-independent pathway. Cell DeathDis 3: e261.

Merksamer PI, Papa FR. 2010. The UPR and cell fate at aglance. J Cell Sci 123: 1003–1006.

Merksamer PI, Trusina A, Papa FR. 2008. Real-time redoxmeasurements during endoplasmic reticulum stress re-veal interlinked protein folding functions. Cell 135:933–947.

Meusser B, Hirsch C, Jarosch E, Sommer T. 2005. ERAD:The long road to destruction. Nat Cell Biol 7: 766–772.

Nadanaka S, Okada T, Yoshida H, Mori K. 2007. Role ofdisulfide bridges formed in the lumenal domain ofATF6 in sensing endoplasmic reticulum stress. Mol CellBiol 27: 1027–1043.

Nishitoh H, Matsuzawa A, Tobiume K, Saegusa K, Takeda K,Inoue K, Hori S, Kakizuka A, Ichijo H. 2002. ASK1 isessential for endoplasmic reticulum stress-induced neu-ronal cell death triggered by expanded polyglutaminerepeats. Genes Dev 16: 1345–1355.

Oyadomari S, Koizumi A, Takeda K, Gotoh T, Akira S, ArakiE, Mori M. 2002. Targeted disruption of the Chop genedelays endoplasmic reticulum stress-mediated diabetes.J Clin Invest 109: 525–532.

Oyadomari S, Yun C, Fisher EA, Kreglinger N, Kreibich G,Oyadomari M, Harding HP, Goodman AG, Harant H,Garrison JL, et al. 2006. Cotranslocational degradationprotects the stressed endoplasmic reticulum from proteinoverload. Cell 126: 727–739.

Ozcan U, Cao Q, Yilmaz E, Lee AH, Iwakoshi NN, OzdelenE, Tuncman G, Gorgun C, Glimcher LH, HotamisligilGS. 2004. Endoplasmic reticulum stress links obesity,insulin action, and type 2 diabetes. Science 306: 457–461.

Ozcan U, Yilmaz E, Ozcan L, Furuhashi M, Vaillancourt E,Smith RO, Gorgun CZ, Hotamisligil GS. 2006. Chemicalchaperones reduce ER stress and restore glucose homeo-stasis in a mouse model of type 2 diabetes. Science 313:1137–1140.

Papa FR, Zhang C, Shokat K, Walter P. 2003. Bypassing akinase activity with an ATP-competitive drug. Science302: 1533–1537.

Papandreou I, Denko NC, Olson M, Van Melckebeke H, LustS, Tam A, Solow-Cordero DE, Bouley DM, Offner F, NiwaM, et al. 2010. Identification of an Ire1a endonucleasespecific inhibitor with cytotoxic activity against humanmultiple myeloma. Blood 117: 1311–1314.

Pirot P, Naamane N, Libert F, Magnusson NE, Orntoft TF,Cardozo AK, Eizirik DL. 2007. Global profiling of genesmodified by endoplasmic reticulum stress in pancreaticbcells reveals the early degradation of insulin mRNAs. Di-abetologia 50: 1006–1014.

Puthalakath H, O’Reilly LA, Gunn P, Lee L, Kelly PN, Hun-tington ND, Hughes PD, Michalak EM, McKimm-Breschkin J, Motoyama N, et al. 2007. ER stress triggersapoptosis by activating BH3-only protein Bim. Cell 129:1337–1349.

Reimold AM, Iwakoshi NN, Manis J, Vallabhajosyula P, Szo-molanyi-Tsuda E, Gravallese EM, Friend D, Grusby MJ,Alt F, Glimcher LH. 2001. Plasma cell differentiation re-quires the transcription factor XBP-1. Nature 412: 300–307.

Riggs AC, Bernal-Mizrachi E, Ohsugi M, Wasson J, Fatrai S,Welling C, Murray J, Schmidt RE, Herrera PL, PermuttMA. 2005. Mice conditionally lacking the Wolfram genein pancreatic islet b cells exhibit diabetes as a result ofenhanced endoplasmic reticulum stress and apoptosis.Diabetologia 48: 2313–2321.

Ron D. 2002. Proteotoxicity in the endoplasmic reticulum:Lessons from the Akita diabetic mouse. J Clin Invest 109:443–445.

Rutkowski DT, Arnold SM, Miller CN, Wu J, Li J, GunnisonKM, Mori K, Sadighi Akha AA, Raden D, Kaufman RJ.2006. Adaptation to ER stress is mediated by differential



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



stabilities of pro-survival and pro-apoptotic mRNAs andproteins. PLoS Biol 4: e374.

Sawaya MR, Sambashivan S, Nelson R, Ivanova MI, SieversSA, Apostol MI, Thompson MJ, Balbirnie M, Wiltzius JJ,McFarlane HT, et al. 2007. Atomic structures of amyloidcross-b spines reveal varied steric zippers. Nature 447:453–457.

Scheuner D, Kaufman RJ. 2008. The unfolded protein re-sponse: A pathway that links insulin demand with b-cellfailure and diabetes. Endocr Rev 29: 317–333.

Scheuner D, Song B, McEwen E, Liu C, Laybutt R, GillespieP, Saunders T, Bonner-Weir S, Kaufman RJ. 2001. Trans-lational control is required for the unfolded protein re-sponse and in vivo glucose homeostasis. Mol Cell 7:1165–1176.

Scheuner D, Vander Mierde D, Song B, Flamez D, CreemersJW, Tsukamoto K, Ribick M, Schuit FC, Kaufman RJ.2005. Control of mRNA translation preserves endoplas-mic reticulum function in b cells and maintains glucosehomeostasis. Nat Med 11: 757–764.

Sevier CS, Kaiser CA. 2002. Formation and transfer of di-sulphide bonds in living cells. Nat Rev Mol Cell Biol 3:836–847.

Shen J, Snapp EL, Lippincott-Schwartz J, Prywes R. 2005.Stable binding of ATF6 to BiP in the endoplasmic retic-ulum stress response. Mol Cell Biol 25: 921–932.

Shore GC, Papa FR, Oakes SA. 2011. Signaling cell deathfrom the endoplasmic reticulum stress response. CurrOpin Cell Biol 23: 143–149.

Smith MH, Ploegh HL, Weissman JS. 2011. Road to ruin:Targeting proteins for degradation in the endoplasmicreticulum. Science 334: 1086–1090.

Song B, Scheuner D, Ron D, Pennathur S, Kaufman RJ. 2008.Chop deletion reduces oxidative stress, improves b cellfunction, and promotes cell survival in multiple mousemodels of diabetes. J Clin Invest 118: 3378–3389.

Stefan Y, Orci L, Malaisse-Lagae F, Perrelet A, Patel Y, UngerRH. 1982. Quantitation of endocrine cell content in thepancreas of nondiabetic and diabetic humans. Diabetes31: 694–700.

Steiner DF. 2000. New aspects of proinsulin physiologyand pathophysiology. J Pediatr Endocrinol Metab 13:229–239.

Stoy J, Edghill EL, Flanagan SE, Ye H, Paz VP, Pluzhnikov A,Below JE, Hayes MG, Cox NJ, Lipkind GM, et al. 2007.Insulin gene mutations as a cause of permanent neonataldiabetes. Proc Natl Acad Sci 104: 15040–15044.

Tirasophon W, Welihinda AA, Kaufman RJ. 1998. A stressresponse pathway from the endoplasmic reticulum to thenucleus requires a novel bifunctional protein kinase/en-doribonuclease (Ire1p) in mammalian cells. Genes Dev12: 1812–1824.

Travers KJ, Patil CK, Wodicka L, Lockhart DJ, Weissman JS,Walter P. 2000. Functional and genomic analyses revealan essential coordination between the unfolded proteinresponse and ER-associated degradation. Cell 101: 249–258.

Trusina A, Papa FR, Tang C. 2008. Rationalizing translationattenuation in the network architecture of the unfoldedprotein response. Proc Natl Acad Sci 105: 20280–20285.

Tu BP, Weissman JS. 2004. Oxidative protein folding in eu-karyotes: Mechanisms and consequences. J Cell Biol 164:341–346.

Upton JP, Austgen K, Nishino M, Coakley KM, Hagen A,Han D, Papa FR, Oakes SA. 2008. Caspase-2 cleavage ofBID is a critical apoptotic signal downstream of endo-plasmic reticulum stress. Mol Cell Biol 28: 3943–3951.

Urano F, Wang X, Bertolotti A, Zhang Y, Chung P, HardingHP, Ron D. 2000. Coupling of stress in the ER to activa-tion of JNK protein kinases by transmembrane proteinkinase IRE1. Science 287: 664–666.

Walter P, Ron D. 2011. The unfolded protein response: Fromstress pathway to homeostatic regulation. Science 334:1081–1086.

Wang XZ, Harding HP, Zhang Y, Jolicoeur EM, Kuroda M,Ron D. 1998. Cloning of mammalian Ire1 reveals diver-sity in the ER stress responses. EMBO J 17: 5708–5717.

Wang J, Takeuchi T, Tanaka S, Kubo SK, Kayo T, Lu D, TakataK, Koizumi A, Izumi T. 1999. A mutation in the insulin 2gene induces diabetes with severe pancreatic b-cell dys-function in the Mody mouse. J Clin Invest 103: 27–37.

Wang L, Lovejoy NF, Faustman DL. 2012. Persistence ofprolonged C-peptide production in type 1 diabetes asmeasured with an ultrasensitive C-peptide assay. DiabetesCare 35: 465–470.

Wei MC, Zong WX, Cheng EH, Lindsten T, Panoutsakopou-lou V, Ross AJ, Roth KA, MacGregor GR, Thompson CB,Korsmeyer SJ. 2001. Proapoptotic BAX and BAK: A req-uisite gateway to mitochondrial dysfunction and death.Science 292: 727–730.

Weir GC, Bonner-Weir S. 2004. Five stages of evolving b-celldysfunction during progression to diabetes. Diabetes 53:S16–S21.

Westermark P, Johnson KH, O’Brien TD, Betsholtz C. 1992.Islet amyloid polypeptide—A novel controversy in dia-betes research. Diabetologia 35: 297–303.

Weston CR, Davis RJ. 2007. The JNK signal transductionpathway. Curr Opin Cell Biol 19: 142–149.

Wu J, Rutkowski DT, Dubois M, Swathirajan J, Saunders T,Wang J, Song B, Yau GD, Kaufman RJ. 2007. ATF6a op-timizes long-term endoplasmic reticulum function toprotect cells from chronic stress. Dev Cell 13: 351–364.

Yamamoto K, Sato T, Matsui T, Sato M, Okada T, Yoshida H,Harada A, Mori K. 2007. Transcriptional induction ofmammalian ER quality control proteins is mediated bysingle or combined action of ATF6a and XBP1. Dev Cell13: 365–376.

Yoon KH, Ko SH, Cho JH, Lee JM, Ahn YB, Song KH, Yoo SJ,Kang MI, Cha BY, Lee KW, et al. 2003. Selective b-cell lossanda-cell expansion in patients with type 2 diabetes mel-litus in Korea. J Clin Endocrinol Metab 88: 2300–2308.

Yorimitsu T, Klionsky DJ. 2007. Eating the endoplasmicreticulum: Quality control by autophagy. Trends CellBiol 17: 279–285.

Yoshida H, Okada T, Haze K, Yanagi H, Yura T, Negishi M,Mori K. 2000. ATF6 activated by proteolysis binds in thepresence of NF-Y (CBF) directly to the cis-acting elementresponsible for the mammalian unfolded protein re-sponse. Mol Cell Biol 20: 6755–6767.

Yoshida H, Matsui T, Yamamoto A, Okada T, Mori K. 2001.XBP1 mRNA is induced by ATF6 and spliced by IRE1 in

F.R. Papa


ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



response to ER stress to produce a highly active transcrip-tion factor. Cell 107: 881–891.

Yoshida H, Matsui T, Hosokawa N, Kaufman RJ, Nagata K,Mori K. 2003. A time-dependent phase shift in the mam-malian unfolded protein response. Dev Cell 4: 265–271.

Youle RJ, Strasser A. 2008. The BCL-2 protein family: Op-posing activities that mediate cell death. Nat Rev Mol CellBiol 9: 47–59.

Zhang K, Shen X, Wu J, Sakaki K, Saunders T, RutkowskiDT, Back SH, Kaufman RJ. 2006. Endoplasmic reticulum

stress activates cleavage of CREBH to induce a systemicinflammatory response. Cell 124: 587–599.

Zhou J, Liu CY, Back SH, Clark RL, Peisach D, Xu Z, Kauf-man RJ. 2006. The crystal structure of human IRE1 lu-minal domain reveals a conserved dimerization interfacerequired for activation of the unfolded protein response.Proc Natl Acad Sci 103: 14343–14348.

Zito E, Melo EP, Yang Y, Wahlander A, Neubert TA, Ron D.2010. Oxidative protein folding by an endoplasmicreticulum-localized peroxiredoxin. Mol Cell 40: 787–797.



ww

w.p

ersp

ecti

vesi

nm

edic

ine.

org



August 14, 20122012; doi: 10.1101/cshperspect.a007666 originally published onlineCold Spring Harb Perspect Med

Feroz R. Papa Diabetes

-Cell Degeneration, andβEndoplasmic Reticulum Stress, Pancreatic

Subject Collection Type I Diabetes

DiabetesThe Pathogenesis and Natural History of Type 1

Mark A. Atkinson Disease SubtypesPrediction, Significance, and Detection of Distinct Humoral Autoimmunity in Type 1 Diabetes:

S. EisenbarthMassimo Pietropaolo, Roberto Towns and George

Diabetes and Other Autoimmune Diseases?Do MHCII-Presented Neoantigens Drive Type 1

Philippa Marrack and John W. KapplerDegeneration, and Diabetes

-CellβEndoplasmic Reticulum Stress, Pancreatic

Feroz R. Papa

Treatment of Type 1 DiabetesClinical Immunologic Interventions for the

Anette-G. ZieglerLucienne Chatenoud, Katharina Warncke and

Localization, and RegulationIslet Autoantigens: Structure, Function,

al.Peter Arvan, Massimo Pietropaolo, David Ostrov, et

Update on Islet TransplantationMichael McCall and A.M. James Shapiro

Environmental Triggers of Type 1 DiabetesMikael Knip and Olli Simell

Destruction in the Diagnosis of Type 1 Diabetes-CellβImmunologic and Metabolic Biomarkers of

Jasmin Lebastchi and Kevan C. HeroldFar

The Story So−− Cells from Stem CellsβGenerating

Matthias Hebrok

Tissues in Immunodeficient MiceDiabetes by Engraftment of Functional Human Advancing Animal Models of Human Type 1

Shultz, et al.Michael A. Brehm, Alvin C. Powers, Leonard D.

Antigen Targets of Type 1 Diabetes AutoimmunityBart O. Roep and Mark Peakman

Diabetes in Mice and HumansBreakdown in Peripheral Tolerance in Type 1

BluestoneLukas T. Jeker, Hélène Bour-Jordan and Jeffrey A.

Innate ImmunityConnecting Type 1 and Type 2 Diabetes through

Justin I. Odegaard and Ajay Chawla

1 DiabetesAntigen-Specific Therapeutic Approaches in Type

et al.Xavier Clemente-Casares, Sue Tsai, Carol Huang, Diabetes

Increased Frequency of Insulin-Dependent The Hygiene Hypothesis: An Explanation for the

Jean-François Bach and Lucienne Chatenoud

http://perspectivesinmedicine.cshlp.org/cgi/collection/ For additional articles in this collection, see

Copyright © 2012 Cold Spring Harbor Laboratory Press; all rights reserved


http://perspectivesinmedicine.cshlp.org/cgi/collection/

http://perspectivesinmedicine.cshlp.org/cgi/collection/


Documents

Endoplasmic Reticulum Stress, Pancreatic -Cell ...perspectivesinmedicine.cshlp.org/content/2/9/a007666.full.pdf · Endoplasmic Reticulum Stress, Pancreatic b-Cell Degeneration, and