Cell Metabolism

Perspective

The Role of Endoplasmic Reticulumin Hepatic Lipid Homeostasis and Stress Signaling

Suneng Fu,1 Steven M. Watkins,1,* and Gokhan S. Hotamisligil1,*1Department of Genetics and Complex Diseases, Harvard School of Public Health, Boston, MA 02115, USA*Correspondence: swatkins@hsph.harvard.edu (S.M.W.), ghotamis@hsph.harvard.edu (G.S.H.)DOI 10.1016/j.cmet.2012.03.007

The endoplasmic reticulum (ER) is a critical site of protein, lipid, and glucose metabolism, lipoprotein secre-tion, and calcium homeostasis. Many of the sensing, metabolizing, and signaling mechanisms for thesepathways exist within or on the ER membrane domain. Here, we review the cellular functions of ER, howperturbation of ER homeostasis contributes to metabolic dysregulation and potential causative mechanismsof ER stress in obesity, with a particular focus on lipids, metabolic adaptations of ER, and themaintenance ofits membrane homeostasis. We also suggest a conceptual framework of metabolic roundabout to integratekey mechanisms of insulin resistance and metabolic diseases.

Meeting Metabolic Challenges: The Function of theEndoplasmic ReticulumNutrient overload and chronic metabolic diseases are the hall-

marks of the 21st century’s public health challenges. Intake of

calories of all forms (lipids, proteins, and carbohydrates)

demands an adaptive response from the metabolic networks.

Chronic exposure to excessive or inappropriate nutrients places

a heavy burden on adaptive responses that ultimately fail and

cause a pattern of continuous, low-level inflammatory and stress

responses, which we refer to as metaflammation, leading ulti-

mately to the chronic metabolic diseases of modern life: obesity,

diabetes, cardiovascular disease, and other associatedmetabol-

ically driven pathologies (Gregor andHotamisligil, 2011). Interest-

ingly, thedetrimental effectsof acuteovernutritionare rarely toxic,

but instead present a transient challenge that is often adequately

met by the adaptive systems. Long-term exposure, however,

leads to a gradual loss of metabolic homeostasis, and diseases

emerge as consequences of chronic and incremental dysregula-

tion of metabolism. Chronic disease systems then are defined by

changes in both the quality and quantity of substrates utilized for

metabolism and by the inability of the metabolic control mecha-

nisms themselves to manage long-term exposures to nutrients.

Under both normal physiological fluctuations of nutrients and

conditions of excess, the endoplasmic reticulum (ER) has a vital

role in maintaining cellular and organismic metabolic homeo-

stasis. As we will discuss in this article, the ER provides signifi-

cant adaptive capacity to metabolism in managing the periodic

cycles associated with feeding, fasting and other metabolic

demands of limited duration. This organelle however, appears

to be less capable of providing the necessarymetabolic flexibility

to manage chronic and escalating metabolic challenges,

possibly due to a lack of selective pressures to develop such

counter measures during much of its evolution. Thus, the ER

represents both an evolutionary bottleneck leading to common

chronic diseases and a valuable target area for their prevention

and treatment. Several key biological functions of the ER with

relevance to metabolic aspects of ER stress and the unfolded

protein response pathways and their regulation upon exposure

to acute, cyclic, and chronic metabolic stresses will be dis-

cussed in the following sections with a heavier focus on liver.

Protein Synthesis and Secretion

Protein biosynthesis is themost enriched and best characterized

function of the rough ER (Song et al., 2010). It is estimated that up

to �75% proteins are synthesized on the ER and all membrane

or secreted proteins that enter the secretary pathway undergo

quality control for proper folding in the ER lumen (Andrews and

Tata, 1971). This is an extraordinary task, as millions of mole-

cules are processed through the ER in seconds, especially in

a highly synthetic organ like the liver, and close to 50% of newly

synthesized proteins fail to assume their properly folded forms.

This aspect of ER function has been extensively reviewed

recently and will not be covered in detail (Braakman and Bulleid,

2011; Hebert and Molinari, 2007). When ER is stressed with the

presence of misfolded proteins, the unfolded protein response

(UPR) is activated (Mori, 2000). The canonical UPR response

(discussed further below) aims to reduce protein synthesis, facil-

itate protein disposal, and enhance protein folding to mitigate

stress. If these efforts fail, UPR can also trigger death responses.

It is curious that despite having a rather robust and elegant

adaptive system, ER still fails to establish a healthy equilibrium

against chronic metabolic challenges. It is therefore possible

that metabolic stress introduces unusual constraints for the

adaptive capacity of the ER. For example, a proper ER response

is dependent on efficient nutrient management and accurate

metabolic control, and a dysfunctional ER, along with other

signals, intersects with a variety of inflammatory and stress

signaling networks that in the long run can also add insult to

the ER injury. Here, we will discuss how ER responses intersect

with metabolism with an emphasis on lipids, not to suggest

a superiority for this aspect of ER biology but since this aspect

of metabolism is often overlooked in the consideration of ER

adaptive biology and constitute the main focus of this article.

Intermediary Lipid Metabolism

The overabundance of nutrients such as lipids in obesity and

caloric surplus leads to aberrant lipid management and ectopic

fat accumulation. This lipotoxicity is a fundamental component

of metabolic disease and insulin resistance (Unger and Scherer,

2010; McGarry and Dobbins, 1999). Lipid metabolism occurs

primarily at the ER, where many of the enzymes involved in inter-

mediary and complex lipid metabolism reside, giving the ER a

Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc. 623

Figure 1. Lipid Metabolism at the Endoplasmic ReticulumThe depiction of lipid metabolism at the ER is not intended to be compre-hensive, but to provide a representation of the key functionalities of the ER insynthesizing, modifying, and exporting lipids.(A) Key metabolic processes in cholesterol metabolism occurring at the ERinclude control of cholesterol synthesis via SREBP release and upregulation ofcholesterol synthesis enzymes including HMG-CoA reductase (HMGR),oxidation of excess membrane cholesterol by Cyp7a, a key step in liberatingcholesterol for bile acid production and secretion, and the acylation ofcholesterol into cholesterol esters by acyl-CoA:acyltransferase (ACAT), which,like DGAT, creates a highly hydrophobic entity that seeds the formation of lipiddroplets within the membrane phase.(B) Key activities in fatty acid metabolism include synthesis of saturated fattyacids (SFA) via fatty acid synthase (FAS) and their elongation and desaturation

624 Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc.

Cell Metabolism

Perspective

central but under-investigated role in managing lipotoxicity. The

primary mechanisms for managing lipid substrates at the ER

are presented in Figure 1A. Among lipids, careful control of

cholesterol levels is particularly important as cholesterol exerts

a strong ordering effect on membranes through its hydrophobic

interactions with the fatty acids present in phospholipids.

Control of cellular cholesterol levels occurs at the ER through

pathways that sense the level of cholesterol or cholesterol

derivatives within the ER membrane itself and relay signals

controlling cholesterol synthesis and clearance. The primary

modes of cholesterol regulation at the ER include initiation of

cholesterol synthesis via the SCAP/SREBP2 pathway under

conditions of low ER cholesterol (Brown and Goldstein, 2009),

conversion of cholesterol to oxysterols and eventually bile acids

(Russell, 2003) for excretion in bile, and the production of choles-

terol esters, which can migrate out of the membrane into lipid

droplets by virtue of their hydrophobic properties (Martin and

Parton, 2006). Given that cholesterol is an evolutionarily

conserved building block and that it is critical to cell function, it

is likely that other pathways exist for sensing cholesterol and

triggering regulation of metabolism, but it is not yet clear how

quantitatively important those pathways are for ER cholesterol

homeostasis.

The composition of the intracellular fatty acid pool is also regu-

lated by the ER in order to meet the basic cellular demands for

synthesizing complex lipids with a broad variety of structures

(Figure 1B). Like cholesterol, the fatty acid composition of aggre-

gate lipid structures has a profound impact on the physical

properties of cellular membranes. The tools for modifying the

structure of fatty acids include desaturases, elongases, and

b oxidation cycles. The desaturases and elongases occur mainly

in the ER (Miyazaki and Ntambi, 2008); thus, the ER is a crucial

site for maintaining all aspects of fatty acid and lipid homeostasis

in cells. Additionally, the process of synthesizing fatty acids de

novo is controlled at least in part by fatty acid sensing mecha-

nisms within the ER via the feedback inhibition of the release of

the transcription factor SREBP-1 (Horton, 2002). In humans the

majority of de novo lipogenesis takes place in the liver while

adipose tissue de novo lipogenesis is suppressed (Shrago

et al., 1971). Consequently, human adipose tissue lipids are

mainly derived from the accumulation and modification of diet

and liver-produced fatty acids. This is a critical consideration in

making correlative evaluations related to the total amount of de

to longer-chainSFAandmonounsaturated fatty acids (MUFA)by the long-chainelongase (LCE) and stearoyl-CoA desaturase (SCD), respectively. Synthesis offatty acids is controlled by SREBP-1. Essential (polyunsaturated, PUFA) fattyacids are metabolized by desaturation and elongation of highly unsaturatedfatty acids (HUFA)bydesaturases (FADS) andelongases (ELOVL), respectively.(C) Key activities in the synthesis of glycerolipids including phospholipids (PL)and triglycerides (TG) include glycerol-3-phosphate acyltransferase (GPAT),which acylates the sn-1 position of glycerol-3-phosphate, acylglycer-olphosphate acyltransferase (AGPAT), which acylates the sn-2 position,phosphatidic acid phosphohydrolase (PAPH), which forms diacylglycerol, thediacylglycerol choline/ethanolaminephosphotransferases (CEPT) which formeither phosphatidylcholine or phosphatidylethanolamine from diacylglycerol,and monoacylglycerol acyltransferase (MGAT) or diacylcglycerol acyl-transferase (DGAT) which can acylate diacylglycerol to form triacylglycerol.The formation of triacylated TG seed the formation of a nonpolar phase withinthe membrane domain, and this can grow and mature into either a lipid dropletor a VLDL particle. Other important phospholipid synthesis pathways exist atthe ER but are not shown.

Cell Metabolism

Perspective

novo lipogenic products in blood or tissues, as these products

are strong indicators of hepatic lipogenesis.

Glycerolipid Synthesis, Lipid Droplet Formation,

and VLDL Assembly

There are two primary forms of complex lipids in the endo-

plasmic reticulum: amphiphilic phospholipids and neutral lipids

such as triglycerides and cholesterol esters (Figure 1C). Phos-

pholipids are essential for the assembly of membranes and the

vesicles that traffic proteins and other entities to the rest of the

cell. Triglycerides and cholesterol esters are sinks for excess

fatty acid and cholesterol, and by virtue of their hydrophobic

nature, seed the formation of lipid droplets within the ER

membrane (Martin and Parton, 2006; Farese and Walther,

2009). The formation of lipid droplets protects the cell from accu-

mulation of excess fatty acids and maintains the integrity of the

ER membrane. However, it may also consume ER membranes

during this process, although the significance of this biology to

ER function is not known. The primary pathway for the syntheses

of both phospholipids and triglycerides is the Kennedy pathway,

which sequentially acylates glycerol phosphate to phospholipids

or triglycerides and is largely present at the ER (Fagone and

Jackowski, 2009). Control of both the fatty acid composition

and the phospholipid type distribution has important effects on

membrane function (Sprong et al., 2001). Notably, the assembly

of lipid droplets for intracellular lipid storage requires a specific

increase in the synthesis of phosphatidylcholine at the ER,

potentially putting the balance of the ERmembrane composition

at risk (Krahmer et al., 2011). Thus, lipid metabolism plays an

important role in maintaining ER membrane function and its

disturbance, even in small ways, can pose major challenges

that lead to ER stress andmetabolic dysfunction (Fu et al., 2011).

The synthesis and compositional control of lipids in the ER also

needs to be tightly coordinated with lipid trafficking and secre-

tion. Hepatocytes and other cell types have the capacity to

assemble and secrete ApoB-containing lipoproteins comprising

ApoB, triglycerides, cholesterol esters, and phospholipids. In

hepatocytes, lipoprotein secretion serves not only to deliver lipid

to peripheral tissues but also to clear the hepatocyte of excess

cholesterol and fatty acid. Interestingly, the use of the VLDL

assembly mechanism to clear lipid comes at the cost of

increasing protein synthesis and lipid metabolism demands on

the ER, and the regulation of this process appears tightly linked

to fatty acid metabolism (Caviglia et al., 2011; Ota et al., 2008).

Consequently, signaling networks integral to the UPR program

are directly linked to lipogenic pathways and lipid secretion, to

coordinate the metabolic needs of a recovering or adapting

ER. It is also critical to mention that the biology of lipids is not

limited to their properties as building blocks and substrates but

also includes specific signaling molecules and hormones that

control local and systemic metabolic responses, which will be

discussed further in the following sections.

Calcium Homeostasis

Calcium and phosphate are two vital cellular signaling mole-

cules. However, to prevent the formation of calcium phosphate

precipitates, mechanisms have evolved to extrude the vast

majority of calcium out of the cell. As a result, the ratio of intracel-

lular calcium concentration to extracellular calcium concentra-

tion is maintained around 1:100,000 (�10 nM versus �1 mM)

(Clapham, 2007). Cellular processes that constantly require

high levels of calcium are compartmentalized into membrane-

enveloped organelles that include ER, mitochondria, Golgi

body, and others, and they require dedicated pumps to maintain

proper calcium concentrations. The calcium concentration in the

ER is roughly 10,000 times higher than that in the cytosol

(�0.1 mM versus �10 nM), and this gradient is maintained by

the active transport function of Sarco-/endoplasmic reticulum

calcium ATPase (SERCA) (Moore et al., 1975). Many ER chaper-

ones and post-translational modification enzymes depend on

calcium for proper function, and maintaining ER calcium stores

is critical for proper protein production (Michalak et al., 2002).

Additionally, the maintenance of extremely low cytosolic calcium

concentrations allows for a broad range of calcium-mediated

signaling events, most of which involve the controlled and tran-

sient release of calcium from the ER (Berridge et al., 2003). The

function of calcium transporters and channels is also modulated

by membrane lipid composition, protein-protein interaction, and

a range of post-translational modifications (Traaseth et al., 2008;

Li et al., 2009; Fu et al., 2011). Therefore, ER provides a platform

for the crosstalk of many cellular signaling pathways and meta-

bolic regulatory events through calcium fluxes and metabolism

(Verkhratsky, 2005).

ER Stress and Its Contribution to Insulin Resistanceand DiabetesIn the presence of misfolded proteins, ER initiates the unfolded

protein response through the actions of cannonical sensors

PERK, IRE1, and ATF6 in an attempt to restore ER homeostasis

(Walter and Ron, 2011; Malhotra and Kaufman, 2007; Mori,

2000). Collectively, these branches operate through suppression

of protein synthesis, facilitation of protein degradation, and

increase the availability of chaperones and, if all fails, can drive

cells to apoptosis (Figure 2A). An important link that led to the

concept of metabolic regulation by ER emerged from the recog-

nition of its ability to regulate insulin action and glucose metab-

olism (Ozcan et al., 2004; Hotamisligil, 2010b). Insulin is one of

the primary anabolic hormones in the body, stimulating glucose

and lipid uptake into the peripheral tissues and the synthesis of

proteins, glycogen, and lipids therein. As the major cellular

anabolic organelle, ER plays a critical role in insulin-induced

synthetic and growth pathways and the eventual development

of obesity under permissive environmental exposures as well

as the emergence of metabolic disorders during the course of

chronic over-nutrition and obesity (Hotamisligil, 2010b). There

are at least two mechanisms by which ER stress counters whole

body insulin-dependent biological processes: tissue insulin

resistance and b cell failure resulting in impaired insulin produc-

tion. While, these are the main aspects discussed here, it is

critical to recognize, however, that the ER’s impact on metabo-

lism goes beyond the boundaries of insulin to include direct

actions on many metabolic pathways independent of this

hormone, and beyond diabetes as metabolic disease.

ER Stress and Insulin Production

A robust ER function is paramount to b cell function and survival

with or without insulin resistance. This is evident in bothmice and

man, where mutations in PERK result in permanent neonatal

diabetes, such as the case in Wolcott-Rallison syndrome (Dele-

pine et al., 2000). Similar observations have also supported the

role of this branch of the UPR in b cell function in models where

Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc. 625

Figure 2. Cannonical Unfolded Protein Response(A) Canonical UPR. Unfolded or misfolded proteins activate the three ER stress sensors (IRE1, PERK, and ATF6) by releasing them from BiP binding. IRE1 isa ribonuclease that splices and religatesHAC1/Xbp1 transcripts for the synthesis of an active transcription factor to induce the expression of chaperones and lipidsynthesis enzymes. PERK is a kinase that phosphorylates eIF2a and suppresses protein synthesis. ATF6 is a transcription factor that once released from ER willtravel to the Golgi body, cleaved and translocated to the nucleus to activate the transcription of chaperone genes. Upon severe stress conditions, the synthesisof ATF4 is enhanced in an eIF2a phosphorylation independent manner that promotes apoptosis.(B) Contribution of UPR to both inflammatory and metabolic abnormalities in obesity. UPR activates JNK at least through three different mechanisms: PKR,IRE1/TRAF/ASK1, and PERK/CHOP/ERO1L/IP3R/CaMK. JNK promotes inflammation and apoptosis by phosphorylating and activating AP1 family of tran-scription factors. JNK also phosphorylates IRS molecules on serine residues and act as an important mediator of insulin resistance. UPR can also modulategluconeogenesis and lipogenesis in multiple ways. First, the insulin resistance resulting from UPR prevents the inactivation of Foxo1 and its gluconeogenicprogram by insulin. Second, the nuclear form of ATF6 inhibits gluconeogenesis and lipogenesis, respectively, by binding to TORC2 and SREBP2. Third, thespliced form of Xbp1 can directly or indirectly (through SREBP1) activate lipogenesis program while inhibiting gluconeogenesis. Lastly, eIF2a phosphorylationleads to the synthesis of CHOP and the activation of lipogenesis programs through CEBPa/b. The sometimes opposing effects of UPR in regulating gluco-neogenesis and lipogenesis are context dependent. The lipogenesis program in the obese liver also alters the membrane lipid composition of ER, inhibitionof SERCA and propagation of ER stress. Abbreviations: IRE1, inositol required 1; PERK, PKR-like endoplasmic reticulum localized kinase; ATF6, activatingtranscription factor 6.

Cell Metabolism

Perspective

eIF2a phosphorylation is altered (Harding et al., 2001; Zhang

et al., 2002). In the presence of insulin resistance, b cells prolif-

erate and secrete more insulin, which can account for up to

50% of total protein synthesis. When facing a constant demand

of insulin in the context of obesity and insulin resistance, transla-

tional suppression by eIF2a, while compromising the synthetic

capacity, may not be sufficient to alleviate ER stress if it is not

accompanied by proper coordination of the other mitigating

events, chief among them being the ER folding capacity. This

state creates yet another challenge since constitutively acti-

vated chaperoning programs may also not be compatible with

long-term cellular homeostasis (Trusina et al., 2008), In fact,

this is likely the case for most, if not all, chronic ER stress condi-

tions. For example, while critical for the UPR, when chronically

626 Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc.

produced ATF6 can also negatively regulate insulin production

by suppressing the transcription of the proinsulin gene (Seo

et al., 2008), and abnormal constitutive activation of ATF6 and

XBP1 contributes to b cell dysfunction and death in Ins2+/Akita

pancreatic b cells (Nozaki et al., 2004). Curiously, deletion of

Xbp1 also diminishes islet function and survival (Lee et al.,

2011). In other words, the chaperoning aspect of ER homeo-

stasis is a double-edged knife and needs to be precisely regu-

lated; too little or too much can both be dangerous. In addition

to the molecules in canonical UPR responses, there are other

mutations implicated in pancreatic dysfunction and diabetes in

humans, such as the Wolfram Syndrome 1 (Wfs1) gene, which

is induced by ER stress and regulates ER calcium homeostasis

and survival pathways (Ueda et al., 2005; Takei et al., 2006).

Cell Metabolism

Perspective

ER Stress and Insulin Resistance

Chronic ER stress is present in liver, b cells, brain, and adipose

tissue in animal models of obesity (Ozcan et al., 2004; Nakatani

et al., 2005), as well as obese humans (Boden et al., 2008;

Sharma et al., 2008; Gregor et al., 2009), but not uniformly in

skeletal muscle (Deldicque et al., 2011). In humans, weight loss

results in reversal of these stress indicators in adipose and liver

tissues (Gregor et al., 2009). Importantly, alleviation of ER stress

by chemical or molecular chaperones improves metabolic

control and insulin sensitivity in animal models (Ozcan et al.,

2006; Kammoun et al., 2009; Ozawa et al., 2005) and, most

importantly, in obese humans (Kars et al., 2010; Xiao et al.,

2011), indicating that ER dysfunction is a critical contributor to

chronic metabolic deterioration. The mechanism by which ER

stress impairs insulin action and metabolic control is complex

and can occur at several levels to impair signal transduction

and direct or indirect perturbation of specific metabolic

responses. At the signaling level, several mechanisms integrate

ER stress responses with inflammatory and stress signaling

cascades. These include activation of JNK through double-

stranded RNA-dependent protein kinase (PKR) (Nakamura

et al., 2010), via the IRE1/TRAF2/ASK1 (Urano et al., 2000) or

the PERK/CHOP/ERO1L/IP3R/CaMK (Timmins et al., 2009)

pathways, all of which result in impaired insulin receptor

signaling (Hirosumi et al., 2002; Ozcan et al., 2004; Sabio and

Davis, 2010) (Figure 2B). Genetic inhibition of certain aspects

of IRE-1 activity by Bax-inhibitor 1 in the liver tissue also protects

mice against insulin resistance and glucose intolerance (Bailly-

Maitre et al., 2010). Interestingly, PKR is also induced by ER

stress and required for JNK activation (Nakamura et al., 2010)

in cells and whole animals upon short- or long-term exposure

to lipids. PKR forms complexes with insulin receptor signaling

components, inflammatory signaling molecules, and protein

translation machinery, thereby integrating many metabolic and

stress signals (Nakamura et al., 2010). An intriguing possibility

remains that PKR may directly sense a metabolic signal and

serve to assemble a metaflammatory complex that is in play in

metabolic regulation. Other possibilities linking ER function to

metabolic control include alteration of the overall metabolic

context that indirectly contributes to impaired insulin action

through suppressing the secretion of metabolically favorable

hormones such as adiponectin or promote the release of

deleterious ones (Xu et al., 2010) or actions through the

networks between liver and other metabolic organs with the

central nervous system (Purkayastha et al., 2011). Finally, endo-

crine functions of gut and the gut microbiota, which have

profound effects on systemic metabolic homeostasis and

pathological outcomes, also are regulated by ER stress and

related inflammatory signaling pathways (Caricilli et al., 2011;

Kau et al., 2011).

Independent Metabolic Functions

As mentioned before, metabolism in general and glucose and

lipid metabolism in particular could also be regulated through

molecules involved in ER adaptive responses, independent of

insulin action or even ER stress, in a classic sense (Figure 2B).

ATF6 can inhibit the expression of gluconeogenic and lipogenic

genes by binding to TORC2 and SREBP2, respectively, and

suppress their function (Wang et al., 2009; Zeng et al., 2004).

XBP1 (and its yeast homolog, HAC1), as a transcription factor

in its own right, directly regulates the expression of a large set

of lipid metabolism genes and hyperactivation promotes hepatic

steatosis (Lee et al., 2008; Acosta-Alvear et al., 2007). XBP1has

also been suggested to suppress hepatic gluconeogenesis by

retaining Foxo1 in the cytosol independent of its own transcrip-

tion activation function (Zhou et al., 2011). Interestingly, deletion

of Xbp1 in the liver was also found to prevent steatosis and

hepatic insulin resistance (Jurczak et al., 2011). Lastly, the

UPR branch traditionally considered as the translational arm,

PERK/eIF2a phosphorylation, also controls lipogenesis through

CEBP/a, PPARg, and SREBP1 (Bobrovnikova-Marjon et al.,

2008). For example, PERK regulates the processing andmatura-

tion of SREBP and its target gene expression in the mammary

gland (Bobrovnikova-Marjon et al., 2008). Hence, there are

multiple direct paths to metabolic regulation from UPR and the

related inflammatory and stress signaling pathways independent

of insulin action (Figure 2B).

Mechanisms of ER Stress in a Metabolic Setting and theRole of LipotoxicityER stress can be envisioned to proceed in three modes: acute,

periodic, and chronic (Figure 3A). The acute form of ER stress

has been extensively studied and is typically induced by treat-

ment with chemicals (e.g., tunicamycin, DTT, calcium iono-

phores, thapsigargin, saturated fatty acids, and cholesterol)

that induce ER stress. The periodic mode of ER stress is largely

associated with rhythmic or transient physiological activities

(e.g., feeding-fasting cycles) (Pfaffenbach et al., 2010) or tran-

sient infusions of glucose or lipids (Boden et al., 2011), and ER

homeostasis is fully restored after each cycle. Just like acute

inflammation, UPR in this context is an essential part of homeo-

stasis. Chronic ER stress, however, is characterized by a state in

which ER homeostasis is not fully restored during periodic

cycles, either due to the continuous presence of a pre-existing

ER stress inducer or the engagement of UPR in a vicious cycle,

the byproducts of which also compromise ER function.

Chronic ER stress is the mode most relevant to metabolic

health in the condition of nutrient excess, obesity, and insulin

resistance. In recent years, a number of models have been

developed that may explain ER dysfunction and stress in ameta-

bolic context (Hotamisligil, 2010a; Tabas and Ron, 2011; Eizirik

and Cnop, 2010) (Figure 3B). Among mechanistic postulates

inducing ER stress in obesity are protein misfolding by inhibition

of protein N-glycosylation after fructose feeding (Qiu et al.,

2005), protein overload due to increased synthesis of insulin,

apoB, or acute phase proteins or antibodies in the pancreas or

liver of obese subjects (Su et al., 2009; Matveyenko et al.,

2009), lipid synthesis (Fu et al., 2011) or overload leading to the

production of toxic lipid species (Samuel et al., 2010; Summers,

2006; Turpin et al., 2006), or the disruption of membrane struc-

ture and inhibition of membrane protein function (Borradaile

et al., 2006; Li et al., 2004), inflammation reducing membrane,

and chaperone protein function (Uehara et al., 2006; Nakamura

et al., 2010), and defective autophagy (Yang et al., 2010). The

exact mechanism of chronic metabolic ER stress is likely to be

tissue specific and context dependent and may involve the

contribution of multiple pathways. For the purposes of this

manuscript, particular interest will be paid to lipid-induced stress

pathways and their connection to the induction of canonical ER

Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc. 627

Figure 3. Stress and Adaptation at the Endoplasmic Reticulum(A) Conceptual models for three modes of ER stress induction with relevanceto overnutrition and chronic disease. Periodic or adaptive UPR, typified bychanges in ER homeostasis during feeding/fasting cycles, or other transientrhythmic demands, and represented by the black line. In such cases, substrateoverflow and protein synthesis demands challenge the ER, but UPR andmetabolic adaptations such as insulin resistance serve as mechanisms forrestoring ER status. Acute nutrient-induced UPR and ER stress shown in thered line. This stress is typified by treatments that perturb the ER membranecomposition or functionality directly, such as excess palmitate, cholesterol orchemicals such as tunicamycin. The relevance of this mode of ER stress tophysiology is questionable, but many of the experimental protocols for theinduction of ER stress follow this model. Chronic ER stress and loss ofadaptive capacity, shown in blue and typified by long term overexposure tonutrients as in obesity. Under such conditions the ER does not have time,means, or capacity to fully restore functionality before facing another round ofsubstrate supply and demand for synthesis. Hence, the ER progressively loses

628 Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc.

Cell Metabolism

Perspective

stress. Again, this is not to suggest the relative importance over

others but to provide a thorough examination of this aspect.

Acute ER Stress Induction by Lipids

The role of the saturated fatty acid palmitate (16:0), on metabolic

disorders is widely acknowledged. For example, increased

consumption or elevated blood levels of saturated fatty acids

are associated with insulin resistance and the risk of diabetes

in humans (Xiao et al., 2006; Hu et al., 2001; Wang et al., 2003;

Vessby, 2000). This effect has been recapitulated in experi-

mental models, where treatment of cells or animals with palmi-

tate induces acute ER stress and cytotoxicity (Borradaile et al.,

2006; Wei et al., 2006; Erbay et al., 2009). This is likely because

palmitate does not effectively activate triglyceride synthesis

(Lee et al., 2010), and thus it can not be effectively sequestered

into triglycerides unless unsaturated fatty acids are simulta-

neously present or other countermeasures can be deployed

(Listenberger et al., 2003). Consequently, palmitate may be

metabolized to species that induce cytotoxicity: ceramides

(Holland et al., 2007; Unger, 2002), diacylglycerides (Samuel

et al., 2010), reactive oxygen species (Listenberger et al.,

2001), and highly saturated ER membrane phospholipids

(Ariyama et al., 2010).

Excess cholesterol can also induce acute ER stress. Choles-

terol concentrations in the ER are kept exceedingly low through

ER-resident mechanisms (Lange and Steck, 2008). The reason

for the maintenance of low concentrations of cholesterol in the

ER is not entirely clear, but conditions that elevate ER cholesterol

lead to ER stress and loss of SERCA-mediated calcium homeo-

stasis, presumably by promoting the formation of an ordered ER

membrane phase that perturbs SERCA structure (Feng et al.,

2003). Preventing externally loaded cholesterol from cycling to

the ER via inhibitors of cholesterol transport prevents choles-

terol-induced ER stress and cytotoxicity (Feng et al., 2003).

Cholesterol and palmitate can act additively to elicit cytoxicity,

but it is not known whether the effect is a result of a synergistic

membrane ordering effect, or by other additive biochemical

and molecular mechanisms (Pineau et al., 2009).

Unsaturated fatty acids including oleate and palmitoleate

effectively prevent acute ER stress induction under conditions

of either excess palmitate or cholesterol (Pineau et al., 2009; Lis-

tenberger et al., 2003; Peng et al., 2011). The reasons for this in

acute stress have to do with the ability of monounsaturated fatty

acids to support adaptive mechanisms including triglyceride

control over metabolism and its own membrane and folding homeostasis,leading to pathological states.(B) Factors influencing ER homeostasis and ER stress. Diet, excess caloricintake, and obesity are prime factors in taxing ER’s homeostatic mechanismsand capacity. In particular, saturated fat and cholesterol (1), when provided inexcess, can promote the formation of an ordered membrane domain in the ERthat inhibits VLDL formation and SERCA activity, causing a loss of both lipidclearance and calcium homeostatic capabilities. Additionally, saturated fattyacids increase ApoB production demands, but are not good substrates fortriglyceride synthesis, and inhibit the clearance of lipid from the ERmembranevia lipid droplet formation – and may increase the production of lipotoxicmediators such as ceramides and DAG. Fructose and hyperglycemia (2) canimpair normal protein glycosylation or activate lipogenesis and perturbmembrane properties, leading to UPR. Inflammatory signals (3) increasesdemand on hepatic ER for the synthesis and secretion of acute phase proteins,inhibits SERCA activity, and may compromise UPR. Oxidative stress (4)resulting from nutrient overload or mitochondrial dysfunction can impairSERCA and ER chaperone function.

Cell Metabolism

Perspective

synthesis (Lee et al., 2010; Listenberger et al., 2003). Genetic

inhibition of the endogenous ability to convert saturated fatty

acids to unsaturated fatty acids promotes ER stress in cells

and animals (Green and Olson, 2011; Flowers et al., 2008). In

fact, it is feasible to postulate that products of de novo lipogen-

esis such as palmitoleate act to mitigate the potential damaging

effects of saturated fatty acids that, when present in isolation,

disrupt metabolism (Cao et al., 2008; Riquelme et al., 2011).

Chronic ER Stress Induction by Lipids

While monounsaturated fatty acids are established attenuators

of acute palmitate-induced ER stress, excess fatty acid expo-

sure over long periods of time can exceed the capacity, espe-

cially of nonadipose cells, to maintain homeostasis, resulting in

chronic ER stress and lipotoxicity. For example, short-term treat-

ment of cells with oleic acid increases VLDL and triglyceride

secretion, whereas extended incubation of cells with oleic acid

can lead to ER stress and suppression of VLDL secretion (Ota

et al., 2008). Importantly, VLDL secretion and ER homeostasis

under conditions of chronic oleate exposure can be restored

in vivo and in vitro by PBA—a chemical chaperone, demon-

strating that the chronic effects of fatty acids on ER function

are related to compromised chaperone function and more

importantly could be remedied by interventions targeting ER

function (Ota et al., 2008). Likewise PBA alleviates lipid-induced

insulin resistance and b cell dysfunction in humans (Xiao et al.,

2011), although effects of PBA on systems unrelated to its chap-

erone role cannot be ruled out in either setting. Animal models

with increased fatty acid desaturation capacity in extrahepatic

sites are resistant to ER stress and the development ofmetabolic

abnormalities associated with obesity (Cao et al., 2008; Erbay

et al., 2009; Green andOlson, 2011). For example, macrophages

lacking the lipid binding protein aP2 strongly upregulate Scd1

expression, produce copious amounts of palmitoleate, and

become resistant to palmitate-induced ER stress (Erbay et al.,

2009). Adipose and liver tissues of mice with high levels of palmi-

toleate are also devoid of the hallmarks of metabolic disease

commonly found in dyslipidemic or obese mice, consistent

with a protective role for MUFA accumulation (Cao et al.,

2008). In these cases, the source of palmitoleate was extrahe-

patic, allowing the detection of the biology independent of

hepatic lipogenesis. In free-living humans, overall levels of palmi-

toleate in blood are dictated by rates of hepatic lipogenesis, and

therefore total blood palmitoleate may be associated with meta-

bolic risk and increased adiposity (Wang et al., 2003; Paillard

et al., 2008; Navina et al., 2011). Under conditions where palmi-

toleate is increased in away that is dissociated from hepatic lipo-

genesis, we and others have noted its beneficial effects (Cao

et al., 2008; Yang et al., 2011; Matsuzaka et al., 2007; Stefan

et al., 2010; Mozaffarian et al., 2010). In any case, the stress-miti-

gating and cytoprotective mechanisms of unsaturated fatty

acids may be of great importance in understanding ER function

and designing novel strategies for the prevention or treatment of

diabetes and atherosclerosis. While there are many seemingly

competing theories on which particular lipid species is directly

responsible for the acute effects of palmitate (Samuel et al.,

2010; Ussher et al., 2010; Holland et al., 2007), the chronic induc-

tion of ER stress may in fact result from many simultaneous

insults and respond tomany remedies such as reductions in lipid

exposure as a result of caloric restriction, weight loss, and

genetic inhibition of intracellular fatty acid trafficking and

signaling (Tsutsumi et al., 2011; Gregor et al., 2009; Erbay

et al., 2009).

Obesity induces a chronic state of ER stress in mice and hu-

mans (Hotamisligil, 2010b), and it is reasonable to presume

that the mechanism relates, at least in part, to a long-term expo-

sure to lipids. An emerging line of evidence indicates that chronic

ER stress in the liver may be related to subtle alterations in the

phospholipid composition of the ER (Fu et al., 2011). Even with

triglyceride and cholesterol ester synthesis capacity intact

(preventing acute induction of stress by fatty acids) long-term

exposure to lipids can induce the formation of lipid droplets.

These growing droplets require increased synthesis of phospha-

tidylcholine at the ER (Krahmer et al., 2011). In studies systemat-

ically examining hepatic lipid metabolism, we have recently

shown that an imbalance in the phosphatidylcholine/phosphati-

dylethanolamine concentrations in the ER from obese mice is

related to altered SERCA function (Fu et al., 2011). Inhibition of

PEMT, an ER enzyme that converts the endofacial phospholipid

phosphatidylethanolamine to the exofacial phosphatidylcholine,

reversed obesity-driven changes in ER membrane composition

and corrected SERCA dysfunction, ER stress, and abnormal

glucose metabolism (Fu et al., 2011). It is also worth noting

that lipids and fatty acids have a vast array of physical forms

that lead to a variety of functions in diverse lipid pools and their

influence may involve other pathways. Indeed, a recent study

highlighted the ability of the highly unsaturated fatty acid DHA

to induce autophagy (Caviglia et al., 2011), providing perhaps

another mechanism for the maintenance of ER homeostasis

(Yang et al., 2010). Lipids can also engage inflammatory path-

ways directly, although the molecular mediators and the lipid

species involved are not yet established (Wen et al., 2011; Naka-

mura et al., 2010). Considerably more research is necessary,

especially in the specific signaling pathways and functional

networks engaged by specific lipids in the future.

Integration of Stress, Insulin Resistance, and LipidMetabolic PathwaysThere are currently a number of models put forward to explain

themechanistic linkages among obesity, ER stress, insulin resis-

tance, and type 2 diabetes. These include chronic ‘‘metabolic’’

inflammation or metaflammation characterized by altered pro-

duction of adipokines/cytokines and the infiltration of immune

cells into tissues (Gregor and Hotamisligil, 2011; Olefsky and

Glass, 2010; Shoelson et al., 2006; Lumeng and Saltiel, 2011),

lipotoxicity and ectopic fat accumulation in the liver and perhaps

skeletal muscle (Larson-Meyer et al., 2011; Savage et al., 2007),

and decreased mitochondrial function (Bournat and Brown,

2010; Muoio and Newgard, 2008). These pathways have for

the most part been studied as independent mechanisms for

the induction of insulin resistance. Exploring these mechanisms

in isolation has been extremely useful in simplifying initial exper-

imental approaches; however, the results of these experiments

have led to competing reductionist and compartmentalized

proposals for the central mechanisms of insulin resistance. Inter-

preting these conflicting proposals is logically challenging, and

more integrative interpretations will likely be necessary. The

interconnectedness of metabolism as well as ER pathways

makes it possible to assign the causation of insulin resistance

Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc. 629

Figure 4. The Concept of MetabolicRoundabout; an Integrated View ofMetabolic DisruptionAn overview of the metabolic roundabout. Themetabolic roundabout regulates the flow of keymetabolic species (vehicles). The level of anyindividual metabolite is regulated by a network ofmetabolic pathways (denoted by orange andgreen arrows respectively). The ability of aneffector (such as insulin) to facilitate the flow of allmetabolites entering the roundabout is con-strained by the homeostatic balance (entrance/exit) of each individual metabolite. Blockages orconstrictions of any specific pathway linked to themetabolic circle will inevitably impair the flow ofother connected metabolic pathways. Nutrientoverabundance – induced traffic jams in theroundabout can be improved by (1) limiting theoverall inflow of traffic or increasing the number oflanes (dashed orange arrow), (2) rerouting of trafficthrough secondary pathways (dashed blackarrow), or, best of all, (3) repairing the initiallyderanged pathway (broken vehicle). The metabo-lites represented in the metabolic roundabout canbe lipids, inflammatory mediators, neuronalsignals and organelle components, and theirsources can be dietary, cellular and even gut mi-crobiota-derived. In the case of lipids, they couldbe recognized or uptaken by receptors, trans-porters and cellular organelles, and consequentlytrigger metaflammation, ER stress, and mito-chondrial dysfunction. The activation of chronicinflammation and impairment of ER and mito-chondria function can feed-forward contribute tofurther deterioration of the aberrant lipid profile by

reducing fatty acid oxidation, promoting lipolysis, and stimulating de novo lipogenesis. Inflammatory mediators can promote lipogenesis, impair mitochondrialrespiratory chain function (NO), and downregulate SERCA expression. Aberrant lipid profiles, impaired mitochondrial function and ER stress, in return, canpromote metaflammation. Lastly, ER stress activates inflammatory cascades (JNK/IKK/CREB pathways), generates ROS, leads to mitochondrial calciumoverload, and activates lipogenesis. Mitochondrial dysfunction, oxidative stress, metaflammation and lipogenesis further exacerbate ER stress by impairingprotein folding (ROS frommitochondria), protein overloading (acute phase proteins of inflammation) and lipid overloading (lipogenesis). Hence, lipid metabolism,chronic inflammation, and ER stress can all initiate and resolve the problems at the metabolic roundabout when appropriately regulated temporally and spatially.

Cell Metabolism

Perspective

and diabetes in vivo to more integrative mechanisms. A relevant

question is, therefore, what is the nature of insulin resistance

mechanisms and how are they organized to sustain a chronic

condition?

The observations listed above suggest that the pathways

leading to metabolic dysfunction are not structured in conven-

tional linear and hierarchical networks. Instead, we propose

that independent mechanisms for managing insulin signaling

andmetabolismmay resemble the flow of traffic in a roundabout,

so that individual pathways feed into and interact in a common

conduit (Figure 4). Such an arrangement of pathways for regu-

lating hormone action and metabolism provides a framework

for how individual or collective stresses may provide stimulus

to induce insulin resistance (or other metabolic outcomes), just

as evident in the quantized nature of traffic in a roundabout.

A small but concerted increase in traffic from all input pathways

is capable of stopping flow and disrupting function (i.e., insulin

resistance), but so is an acute level of traffic from any one

contributing pathway, or even a single blockage. As traffic

jams may be propagated if not resolved promptly, initial im-

pairment of singular metabolic functions may lead to chronic

activation of additional insulin resistance pathways, and the

deterioration of metabolic status, i.e., development of diabetes.

In this flow of carbon and energy the roads are all connected

and traffic can be rerouted. Impairments can be relieved without

630 Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc.

addressing all or even any of the original causes, either by iden-

tifying new exits (qualitative) or enhancing existing roadways

(quantitative) (Figure 4). As a qualitative example, the presence

of unsaturated fatty acids can reroute the metabolism of satu-

rated fatty acids and cholesterol toward storage or excretion

(in the forms of triglyceride or cholesterol ester-rich lipid droplets

and lipoproteins) and prevent insulin resistance, without the

need to lower the amount of saturated fatty acids or cholesterol

entering into the system. On the quantitative side, enhancement

of ER function by chemical or molecular chaperones can facili-

tate lipoprotein assembly and excretion and concurrently reduce

fatty acid loading into the roundabout by suppressing de novo

lipogenesis (Ota et al., 2008). Similar interconnectedness also

exists for inflammatory responses, and resolution of this state,

in the appropriate manner, can also liberate the congestion at

the roundabout and improve metabolism (Figure 4).

This roundabout concept is consistent with other proposed

models of insulin resistance including a prevalent hypothesis of

mitochondrial dysfunction involving the accumulation of incom-

pletely oxidized medium chain acylcarnitines (Muoio and New-

gard, 2008). A very important implication of the coordination

and rerouting property of the roundabout structure is that, it

permits the separation of cause and effect under certain condi-

tions and confounds simplistic inference of causality from

genetic studies, which are the foundation of many competing

Cell Metabolism

Perspective

hypotheses of lipotoxicity and insulin resistance. Thus, inte-

grated, systems-oriented approaches will be necessary in the

future, not only for modeling and examining the contribution of

individual pathways of insulin resistance and interactions

between mechanistic platforms, but also for evaluating the

efficacy of therapeutics in different tissues and pathological

conditions.

The lipid-centric presentation of this concept here is not to

implicate a diminished significance for glucose and/or amino

acid metabolism in the pathogenesis of diabetes and metabolic

syndrome or the interactive nature of pathways involving these

aspects of metabolism. The metabolism of glucose, lipids, and

amino acids are intimately coupled: both the carbon backbone

and the reducing equivalents are passed along among metabo-

lite classes, hence they too can be readily integrated into this

proposal (Figure 4). It is conceivable that the ‘‘traffic compo-

nents’’ in such a roundabout will not be limited to just metabo-

lites, but also include mediators of inflammation, functional

components of cellular organelles, and possibly other elements

such as the input from gut. The exact composition of this meta-

bolic roundabout may vary among different tissues at different

disease states due to their differences in metabolic wiring,

associated energy balance state, and duration of the stimulus.

For example, promoting de novo lipogenesis in adipose tissue

providing metabolic benefit (Cao et al., 2008) in contrast to the

established detrimental effects of lipogenesis in liver and periph-

eral tissues. Similarly, inflammation enhances short-term

mitochondrial function in the muscle and alters energy expendi-

ture, while causing insulin resistance in the liver and adipose

tissues and permanent tissue damage (Thaler et al., 2012).

Features and consequences of ER stress also differ between

metabolic tissues. For example, patterns of ER stress in skeletal

muscle do not resemble those observed in adipose tissue (Oz-

can et al., 2004; Deldicque et al., 2010, 2011). Therefore, the

roundabout organization of insulin resistance pathways

provides a versatile but simple framework for disruptive,

vicious-cycle-like propagation of chronic disease conditions

including diabetes.

Concluding RemarksMetabolism requires integration of many signals to manage

carbon sources and maintain homeostasis. The fundamental

nature of insulin resistance and metabolic diseases may there-

fore represent an entangled, metastable state of cellular and

whole-body metabolism that is resistant to the anabolic action

of insulin as well as other hormones, rather than the perturbation

of individual metabolic or signaling pathways. It is also clear that

not every aspect of metabolism is directly targeted by insulin or

disrupted by insulin resistance. Perturbations in the central

nervous system also play important roles in the pathogenesis

of metabolic diseases and inflammation (Gautron and Elmquist,

2011; Lumeng and Saltiel, 2011). Finally, the composition of the

gut flora can exert an enormous influence on systemic metabolic

outcomes and at times override the genetic and dietary impact

on disease manifestation (Kau et al., 2011). For this reason, the

mechanism of ER stress in the pathogenesis of abnormal insulin

action and diabetes shall be understood through the metabolic

action of ER and its role as the primary organelle of cellular

anabolic metabolism and a signal-integrating platform, so as

other mechanisms regulating systemic metabolic homeostasis

(Figure 4). Opportunities therefore exist to exploit mechanisms

relating to ER management of debilitating substrate flow for

disease prevention or treatment. If this flow is not interrupted

or diverted, overwhelming the ER structure and functional

capacity initiates a vicious cycle of stress, degradation in meta-

bolic control, modification of the ERmembrane itself and awors-

ening of ER stress and functionality. We suggest that lipids play

a unique and crucial role in this cycle because they are both

regulated by the ER and comprise the ER membrane itself. The

notion that increased capacity to cope with lipid overload spares

ER functionality is consistent with findings from genetic experi-

ments showing that by modifying the membrane phospholipid

composition (Fu et al., 2011), promoting the clearance of lipo-

toxic substrate (Listenberger et al., 2003; Flowers et al., 2008),

or by preventing the presentation of lipotoxic substrate to the

ER in the first place (Erbay et al., 2009), ER is protected from

stress and dysfunction with favorable organismic outcomes.

Thus, it may be prudent to consider mechanisms for controlling

ER homeostasis with a focus on the metabolic aspects of ER

function, particularly through the control of lipids, and to exercise

caution on enhancing insulin sensitivity or secretion to overcome

insulin resistance without addressing underlying mechanisms

impairing ER function and the whole organismal context,

including the microbiota, within which such manipulations are

conducted.

ACKNOWLEDGMENTS

We regret the omission of many valuable contributions of our colleagues dueto space limitations. We thank Drs. Ling Yang and Ana Paula Arruda forassistance in some of the illustrations. Research in the Hotamisligil lab is sup-ported by the National Institutes of Health, the Juvenile Diabetes ResearchFoundation, and the American Diabetes Association. S.F. is supported bythe Naomi Berry Fellowship Award.

REFERENCES

Acosta-Alvear, D., Zhou, Y., Blais, A., Tsikitis, M., Lents, N.H., Arias, C.,Lennon, C.J., Kluger, Y., and Dynlacht, B.D. (2007). XBP1 controls diversecell type- and condition-specific transcriptional regulatory networks. Mol.Cell 27, 53–66.

Andrews, T.M., and Tata, J.R. (1971). Protein synthesis by membrane-boundand free ribosomes of the developing rat cerebral cortex. Biochem. J. 124,883–889.

Ariyama, H., Kono, N., Matsuda, S., Inoue, T., and Arai, H. (2010). Decrease inmembrane phospholipid unsaturation induces unfolded protein response. J.Biol. Chem. 285, 22027–22035.

Bailly-Maitre, B., Belgardt, B.F., Jordan, S.D., Coornaert, B., von Freyend,M.J., Kleinridders, A., Mauer, J., Cuddy, M., Kress, C.L., Willmes, D., et al.(2010). Hepatic Bax inhibitor-1 inhibits IRE1alpha and protects from obesity-associated insulin resistance and glucose intolerance. J. Biol. Chem. 285,6198–6207.

Berridge, M.J., Bootman, M.D., and Roderick, H.L. (2003). Calcium signalling:dynamics, homeostasis and remodelling. Nat. Rev. Mol. Cell Biol. 4, 517–529.

Bobrovnikova-Marjon, E., Hatzivassiliou, G., Grigoriadou, C., Romero, M.,Cavener, D.R., Thompson, C.B., and Diehl, J.A. (2008). PERK-dependentregulation of lipogenesis during mouse mammary gland development andadipocyte differentiation. Proc. Natl. Acad. Sci. USA 105, 16314–16319.

Boden, G., Duan, X., Homko, C., Molina, E.J., Song, W., Perez, O., Cheung, P.,and Merali, S. (2008). Increase in endoplasmic reticulum stress-relatedproteins and genes in adipose tissue of obese, insulin-resistant individuals.Diabetes 57, 2438–2444.

Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc. 631

Cell Metabolism

Perspective

Boden, G., Song,W., Duan, X., Cheung, P., Kresge, K., Barrero, C., andMerali,S. (2011). Infusion of glucose and lipids at physiological rates causes acuteendoplasmic reticulum stress in rat liver. Obesity (Silver Spring) 19, 1366–1373.

Borradaile, N.M., Han, X., Harp, J.D., Gale, S.E., Ory, D.S., and Schaffer, J.E.(2006). Disruption of endoplasmic reticulum structure and integrity in lipotoxiccell death. J. Lipid Res. 47, 2726–2737.

Bournat, J.C., and Brown, C.W. (2010). Mitochondrial dysfunction in obesity.Curr. Opin. Endocrinol. Diabetes Obes. 17, 446–452.

Braakman, I., and Bulleid, N.J. (2011). Protein folding and modification in themammalian endoplasmic reticulum. Annu. Rev. Biochem. 80, 71–99.

Brown, M.S., and Goldstein, J.L. (2009). Cholesterol feedback: from Schoen-heimer’s bottle to Scap’s MELADL. J. Lipid Res. 50 (Suppl ), S15–S27.

Cao, H., Gerhold, K., Mayers, J.R., Wiest, M.M., Watkins, S.M., and Hotamisli-gil, G.S. (2008). Identification of a lipokine, a lipid hormone linking adiposetissue to systemic metabolism. Cell 134, 933–944.

Caricilli, A.M., Picardi, P.K., de Abreu, L.L., Ueno, M., Prada, P.O., Ropelle,E.R., Hirabara, S.M., Castoldi, A., Vieira, P., Camara, N.O., et al. (2011). Gutmicrobiota is a key modulator of insulin resistance in TLR 2 knockout mice.PLoS Biol. 9, e1001212.

Caviglia, J.M., Gayet, C., Ota, T., Hernandez-Ono, A., Conlon, D.M., Jiang, H.,Fisher, E.A., and Ginsberg, H.N. (2011). Different fatty acids inhibit apoB100secretion by different pathways: unique roles for ER stress, ceramide, and au-tophagy. J. Lipid Res. 52, 1636–1651.

Clapham, D.E. (2007). Calcium signaling. Cell 131, 1047–1058.

Deldicque, L., Cani, P.D., Philp, A., Raymackers, J.-M., Meakin, P.J., Ashford,M.L.J., Delzenne, N.M., Francaux, M., and Baar, K. (2010). The unfoldedprotein response is activated in skeletal muscle by high-fat feeding: potentialrole in the downregulation of protein synthesis. Am. J. Physiol. Endocrinol.Metab. 299, E695–E705.

Deldicque, L., Van Proeyen, K., Francaux, M., and Hespel, P. (2011).The unfolded protein response in human skeletal muscle is not involved inthe onset of glucose tolerance impairment induced by a fat-rich diet. Eur. J.Appl. Physiol. 111, 1553–1558.

Delepine, M., Nicolino, M., Barrett, T., Golamaully, M., Lathrop, G.M., andJulier, C. (2000). EIF2AK3, encoding translation initiation factor 2-alpha kinase3, is mutated in patients with Wolcott-Rallison syndrome. Nat. Genet. 25,406–409.

Eizirik, D.L., and Cnop, M. (2010). ER stress in pancreatic beta cells: the thinred line between adaptation and failure. Sci. Signal. 3, pe7.

Erbay, E., Babaev, V.R., Mayers, J.R., Makowski, L., Charles, K.N., Snitow,M.E., Fazio, S., Wiest, M.M., Watkins, S.M., Linton, M.F., and Hotamisligil,G.S. (2009). Reducing endoplasmic reticulum stress through a macrophagelipid chaperone alleviates atherosclerosis. Nat. Med. 15, 1383–1391.

Fagone, P., and Jackowski, S. (2009). Membrane phospholipid synthesis andendoplasmic reticulum function. J. Lipid Res. 50 (Suppl ), S311–S316.

Farese, R.V., Jr., and Walther, T.C. (2009). Lipid droplets finally get a littleR-E-S-P-E-C-T. Cell 139, 855–860.

Feng, B., Yao, P.M., Li, Y., Devlin, C.M., Zhang, D., Harding, H.P., Sweeney,M., Rong, J.X., Kuriakose, G., Fisher, E.A., et al. (2003). The endoplasmic retic-ulum is the site of cholesterol-induced cytotoxicity in macrophages. Nat. CellBiol. 5, 781–792.

Flowers, M.T., Keller, M.P., Choi, Y., Lan, H., Kendziorski, C., Ntambi, J.M.,and Attie, A.D. (2008). Liver gene expression analysis reveals endoplasmicreticulum stress and metabolic dysfunction in SCD1-deficient mice feda very low-fat diet. Physiol. Genomics 33, 361–372.

Fu, S., Yang, L., Li, P., Hofmann, O., Dicker, L., Hide,W., Lin, X., Watkins, S.M.,Ivanov, A.R., and Hotamisligil, G.S. (2011). Aberrant lipid metabolism disruptscalcium homeostasis causing liver endoplasmic reticulum stress in obesity.Nature 473, 528–531.

Gautron, L., and Elmquist, J.K. (2011). Sixteen years and counting: an updateon leptin in energy balance. J. Clin. Invest. 121, 2087–2093.

632 Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc.

Green, C.D., and Olson, L.K. (2011). Modulation of palmitate-inducedendoplasmic reticulum stress and apoptosis in pancreatic b-cells bystearoyl-CoA desaturase and Elovl6. Am. J. Physiol. Endocrinol. Metab. 300,E640–E649.

Gregor, M.F., and Hotamisligil, G.S. (2011). Inflammatory mechanisms inobesity. Annu. Rev. Immunol. 29, 415–445.

Gregor, M.F., Yang, L., Fabbrini, E., Mohammed, B.S., Eagon, J.C., Hotamisli-gil, G.S., and Klein, S. (2009). Endoplasmic reticulum stress is reduced intissues of obese subjects after weight loss. Diabetes 58, 693–700.

Harding, H.P., Zeng, H., Zhang, Y., Jungries, R., Chung, P., Plesken, H., Saba-tini, D.D., and Ron, D. (2001). Diabetes mellitus and exocrine pancreaticdysfunction in perk-/- mice reveals a role for translational control in secretorycell survival. Mol. Cell 7, 1153–1163.

Hebert, D.N., and Molinari, M. (2007). In and out of the ER: protein folding,quality control, degradation, and related human diseases. Physiol. Rev. 87,1377–1408.

Hirosumi, J., Tuncman, G., Chang, L., Gorgun, C.Z., Uysal, K.T., Maeda, K.,Karin, M., and Hotamisligil, G.S. (2002). A central role for JNK in obesity andinsulin resistance. Nature 420, 333–336.

Holland, W.L., Brozinick, J.T., Wang, L.-P., Hawkins, E.D., Sargent, K.M., Liu,Y., Narra, K., Hoehn, K.L., Knotts, T.A., Siesky, A., et al. (2007). Inhibition ofceramide synthesis ameliorates glucocorticoid-, saturated-fat-, and obesity-induced insulin resistance. Cell Metab. 5, 167–179.

Horton, J.D. (2002). Sterol regulatory element-binding proteins: transcriptionalactivators of lipid synthesis. Biochem. Soc. Trans. 30, 1091–1095.

Hotamisligil, G.S. (2010a). Endoplasmic reticulum stress and atherosclerosis.Nat. Med. 16, 396–399.

Hotamisligil, G.S. (2010b). Endoplasmic reticulum stress and the inflammatorybasis of metabolic disease. Cell 140, 900–917.

Hu, F.B., van Dam, R.M., and Liu, S. (2001). Diet and risk of Type II diabetes:the role of types of fat and carbohydrate. Diabetologia 44, 805–817.

Jurczak, M.J., Lee, A.H., Jornayvaz, F.R., Lee, H.Y., Birkenfeld, A.L., Guigni,B.A., Kahn, M., Samuel, V.T., Glimcher, L.H., and Shulman, G.I. (2011).Dissociation of inositol-requiring enzyme (IRE1a)-mediated c-Jun N-terminalkinase activation from hepatic insulin resistance in conditional X-box-bindingprotein-1 (XBP1) knock-out mice. J. Biol. Chem. 287, 2558–2567.

Kammoun, H.L., Chabanon, H., Hainault, I., Luquet, S., Magnan, C., Koike, T.,Ferre, P., and Foufelle, F. (2009). GRP78 expression inhibits insulin and ERstress-induced SREBP-1c activation and reduces hepatic steatosis in mice.J. Clin. Invest. 119, 1201–1215.

Kars, M., Yang, L., Gregor, M.F., Mohammed, B.S., Pietka, T.A., Finck, B.N.,Patterson, B.W., Horton, J.D., Mittendorfer, B., Hotamisligil, G.S., and Klein,S. (2010). Tauroursodeoxycholic Acid may improve liver and muscle but notadipose tissue insulin sensitivity in obese men and women. Diabetes 59,1899–1905.

Kau, A.L., Ahern, P.P., Griffin, N.W., Goodman, A.L., and Gordon, J.I. (2011).Human nutrition, the gut microbiome and the immune system. Nature 474,327–336.

Krahmer, N., Guo, Y., Wilfling, F., Hilger, M., Lingrell, S., Heger, K., Newman,H.W., Schmidt-Supprian, M., Vance, D.E., Mann, M., et al. (2011). Phosphati-dylcholine synthesis for lipid droplet expansion is mediated by localizedactivation of CTP:phosphocholine cytidylyltransferase. Cell Metab. 14,504–515.

Lange, Y., and Steck, T.L. (2008). Cholesterol homeostasis and the escapetendency (activity) of plasma membrane cholesterol. Prog. Lipid Res. 47,319–332.

Larson-Meyer, D.E., Newcomer, B.R., Ravussin, E., Volaufova, J., Bennett, B.,Chalew, S., Cefalu, W.T., and Sothern, M. (2011). Intrahepatic and intramyo-cellular lipids are determinants of insulin resistance in prepubertal children.Diabetologia 54, 869–875.

Lee, A.-H., Scapa, E.F., Cohen, D.E., and Glimcher, L.H. (2008). Regulation ofhepatic lipogenesis by the transcription factor XBP1. Science 320, 1492–1496.

Cell Metabolism

Perspective

Lee, J.N., Kim, H., Yao, H., Chen, Y., Weng, K., and Ye, J. (2010). Identificationof Ubxd8 protein as a sensor for unsaturated fatty acids and regulator oftriglyceride synthesis. Proc. Natl. Acad. Sci. USA 107, 21424–21429.

Lee, A.-H., Heidtman, K., Hotamisligil, G.S., and Glimcher, L.H. (2011). Dualand opposing roles of the unfolded protein response regulated by IRE1alphaand XBP1 in proinsulin processing and insulin secretion. Proc. Natl. Acad.Sci. USA 108, 8885–8890.

Li, Y., Ge, M., Ciani, L., Kuriakose, G., Westover, E.J., Dura, M., Covey, D.F.,Freed, J.H., Maxfield, F.R., Lytton, J., and Tabas, I. (2004). Enrichment ofendoplasmic reticulum with cholesterol inhibits sarcoplasmic-endoplasmicreticulum calcium ATPase-2b activity in parallel with increased order ofmembrane lipids: implications for depletion of endoplasmic reticulum calciumstores and apoptosis in cholesterol-loaded macrophages. J. Biol. Chem. 279,37030–37039.

Li, G., Mongillo, M., Chin, K.T., Harding, H., Ron, D., Marks, A.R., and Tabas, I.(2009). Role of ERO1-alpha-mediated stimulation of inositol 1,4,5-triphos-phate receptor activity in endoplasmic reticulum stress-induced apoptosis.J. Cell Biol. 186, 783–792.

Listenberger, L.L., Ory, D.S., and Schaffer, J.E. (2001). Palmitate-inducedapoptosis can occur through a ceramide-independent pathway. J. Biol.Chem. 276, 14890–14895.

Listenberger, L.L., Han, X., Lewis, S.E., Cases, S., Farese, R.V., Jr., Ory, D.S.,and Schaffer, J.E. (2003). Triglyceride accumulation protects against fattyacid-induced lipotoxicity. Proc. Natl. Acad. Sci. USA 100, 3077–3082.

Lumeng, C.N., and Saltiel, A.R. (2011). Inflammatory links between obesity andmetabolic disease. J. Clin. Invest. 121, 2111–2117.

Malhotra, J.D., and Kaufman, R.J. (2007). The endoplasmic reticulum and theunfolded protein response. Semin. Cell Dev. Biol. 18, 716–731.

Martin, S., and Parton, R.G. (2006). Lipid droplets: a unified view of a dynamicorganelle. Nat. Rev. Mol. Cell Biol. 7, 373–378.

Matsuzaka, T., Shimano, H., Yahagi, N., Kato, T., Atsumi, A., Yamamoto, T.,Inoue, N., Ishikawa, M., Okada, S., Ishigaki, N., et al. (2007). Crucial role ofa long-chain fatty acid elongase, Elovl6, in obesity-induced insulin resistance.Nat. Med. 13, 1193–1202.

Matveyenko, A.V., Gurlo, T., Daval, M., Butler, A.E., and Butler, P.C. (2009).Successful versus failed adaptation to high-fat diet-induced insulin resistance:the role of IAPP-induced beta-cell endoplasmic reticulum stress. Diabetes 58,906–916.

McGarry, J.D., and Dobbins, R.L. (1999). Fatty acids, lipotoxicity and insulinsecretion. Diabetologia 42, 128–138.

Michalak, M., Robert Parker, J.M., and Opas, M. (2002). Ca2+ signaling andcalcium binding chaperones of the endoplasmic reticulum. Cell Calcium 32,269–278.

Miyazaki, M., and Ntambi, J. (2008). Fatty acid desaturation and chain elonga-tion in mammals. In Biochemistry of lipids, lipoproteins and membranes, D.E.Vance and J.E. Vance, eds. (Amsterdam: Elsevier Science), pp. 191–211.

Moore, L., Chen, T., Knapp, H.R., Jr., and Landon, E.J. (1975). Energy-depen-dent calcium sequestration activity in rat liver microsomes. J. Biol. Chem. 250,4562–4568.

Mori, K. (2000). Tripartitemanagement of unfolded proteins in the endoplasmicreticulum. Cell 101, 451–454.

Mozaffarian, D., Cao, H., King, I.B., Lemaitre, R.N., Song, X., Siscovick, D.S.,and Hotamisligil, G.S. (2010). Trans-palmitoleic acid, metabolic risk factors,and new-onset diabetes in U.S. adults: a cohort study. Ann. Intern. Med.153, 790–799.

Muoio, D.M., and Newgard, C.B. (2008). Fatty acid oxidation and insulinaction: when less is more. Diabetes 57, 1455–1456.

Nakamura, T., Furuhashi, M., Li, P., Cao, H., Tuncman, G., Sonenberg, N., Gor-gun, C.Z., and Hotamisligil, G.S. (2010). Double-stranded RNA-dependentprotein kinase links pathogen sensing with stress and metabolic homeostasis.Cell 140, 338–348.

Nakatani, Y., Kaneto, H., Kawamori, D., Yoshiuchi, K., Hatazaki, M.,Matsuoka,T.-A., Ozawa, K., Ogawa, S., Hori, M., Yamasaki, Y., andMatsuhisa, M. (2005).

Involvement of endoplasmic reticulum stress in insulin resistance and dia-betes. J. Biol. Chem. 280, 847–851.

Navina, S., Acharya, C., DeLany, J.P., Orlichenko, L.S., Baty, C.J., Shiva, S.S.,Durgampudi, C., Karlsson, J.M., Lee, K., Bae, K.T., et al. (2011). Lipotoxicitycauses multisystem organ failure and exacerbates acute pancreatitis inobesity. Sci. Transl. Med. 3, 107ra110.

Nozaki, J.I., Kubota, H., Yoshida, H., Naitoh, M., Goji, J., Yoshinaga, T., Mori,K., Koizumi, A., and Nagata, K. (2004). The endoplasmic reticulum stressresponse is stimulated through the continuous activation of transcriptionfactors ATF6 and XBP1 in Ins2+/Akita pancreatic beta cells. Genes Cells 9,261–270.

Olefsky, J.M., and Glass, C.K. (2010). Macrophages, inflammation, and insulinresistance. Annu. Rev. Physiol. 72, 219–246.

Ota, T., Gayet, C., and Ginsberg, H.N. (2008). Inhibition of apolipoprotein B100secretion by lipid-induced hepatic endoplasmic reticulum stress in rodents. J.Clin. Invest. 118, 316–332.

Ozawa, K., Miyazaki, M., Matsuhisa, M., Takano, K., Nakatani, Y., Hatazaki,M., Tamatani, T., Yamagata, K., Miyagawa, J.-I., Kitao, Y., et al. (2005).The endoplasmic reticulum chaperone improves insulin resistance in type 2diabetes. Diabetes 54, 657–663.

Ozcan, U., Cao, Q., Yilmaz, E., Lee, A.-H., Iwakoshi, N.N., Ozdelen, E.,Tuncman, G., Gorgun, C., Glimcher, L.H., and Hotamisligil, G.S. (2004). Endo-plasmic 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, R.O.,Gorgun, C.Z., and Hotamisligil, G.S. (2006). Chemical chaperones reduceER stress and restore glucose homeostasis in a mouse model of type 2 dia-betes. Science 313, 1137–1140.

Paillard, F., Catheline, D., Duff, F.L., Bouriel, M., Deugnier, Y., Pouchard, M.,Daubert, J.-C., and Legrand, P. (2008). Plasma palmitoleic acid, a product ofstearoyl-coA desaturase activity, is an independent marker of triglyceridemiaand abdominal adiposity. Nutr. Metab. Cardiovasc. Dis. 18, 436–440.

Peng, G., Li, L., Liu, Y., Pu, J., Zhang, S., Yu, J., Zhao, J., and Liu, P. (2011).Oleate blocks palmitate-induced abnormal lipid distribution, endoplasmicreticulum expansion and stress, and insulin resistance in skeletal muscle.Endocrinology 152, 2206–2218.

Pfaffenbach, K.T., Nivala, A.M., Reese, L., Ellis, F., Wang, D., Wei, Y., andPagliassotti, M.J. (2010). Rapamycin inhibits postprandial-mediated X-box-binding protein-1 splicing in rat liver. J. Nutr. 140, 879–884.

Pineau, L., Colas, J., Dupont, S., Beney, L., Fleurat-Lessard, P., Berjeaud,J.-M., Berges, T., and Ferreira, T. (2009). Lipid-induced ER stress: synergisticeffects of sterols and saturated fatty acids. Traffic 10, 673–690.

Purkayastha, S., Zhang, H., Zhang, G., Ahmed, Z., Wang, Y., and Cai, D.(2011). Neural dysregulation of peripheral insulin action and blood pressureby brain endoplasmic reticulum stress. Proc. Natl. Acad. Sci. USA 108,2939–2944.

Qiu, W., Kohen-Avramoglu, R., Mhapsekar, S., Tsai, J., Austin, R.C., and Adeli,K. (2005). Glucosamine-induced endoplasmic reticulum stress promotesApoB100 degradation: evidence for Grp78-mediated targeting to proteasomaldegradation. Arterioscler. Thromb. Vasc. Biol. 25, 571–577.

Riquelme, C.A., Magida, J.A., Harrison, B.C., Wall, C.E., Marr, T.G., Secor,S.M., and Leinwand, L.A. (2011). Fatty acids identified in the Burmese pythonpromote beneficial cardiac growth. Science 334, 528–531.

Russell, D.W. (2003). The enzymes, regulation, and genetics of bile acidsynthesis. Annu. Rev. Biochem. 72, 137–174.

Sabio, G., and Davis, R.J. (2010). cJun NH2-terminal kinase 1 (JNK1): roles inmetabolic regulation of insulin resistance. Trends Biochem. Sci. 35, 490–496.

Samuel, V.T., Petersen, K.F., and Shulman, G.I. (2010). Lipid-induced insulinresistance: unravelling the mechanism. Lancet 375, 2267–2277.

Savage, D.B., Petersen, K.F., and Shulman, G.I. (2007). Disordered lipidmetabolism and the pathogenesis of insulin resistance. Physiol. Rev. 87,507–520.

Seo, H.-Y., Kim, Y.D., Lee, K.-M., Min, A.-K., Kim, M.-K., Kim, H.-S., Won,K.-C., Park, J.-Y., Lee, K.-U., Choi, H.-S., et al. (2008). Endoplasmic reticulum

Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc. 633

Cell Metabolism

Perspective

stress-induced activation of activating transcription factor 6 decreases insulingene expression via up-regulation of orphan nuclear receptor small hetero-dimer partner. Endocrinology 149, 3832–3841.

Sharma, N.K., Das, S.K., Mondal, A.K., Hackney, O.G., Chu, W.S., Kern, P.A.,Rasouli, N., Spencer, H.J., Yao-Borengasser, A., and Elbein, S.C. (2008).Endoplasmic reticulum stress markers are associated with obesity in nondia-betic subjects. J. Clin. Endocrinol. Metab. 93, 4532–4541.

Shoelson, S.E., Lee, J., and Goldfine, A.B. (2006). Inflammation and insulinresistance. J. Clin. Invest. 116, 1793–1801.

Shrago, E., Glennon, J.A., and Gordon, E.S. (1971). Comparative aspects oflipogenesis in mammalian tissues. Metabolism 20, 54–62.

Song, Y., Jiang, Y., Ying, W., Gong, Y., Yan, Y., Yang, D., Ma, J., Xue, X.,Zhong, F., Wu, S., et al. (2010). Quantitative proteomic survey of endoplasmicreticulum in mouse liver. J. Proteome Res. 9, 1195–1202.

Sprong, H., van der Sluijs, P., and van Meer, G. (2001). How proteins movelipids and lipids move proteins. Nat. Rev. Mol. Cell Biol. 2, 504–513.

Stefan, N., Kantartzis, K., Celebi, N., Staiger, H., Machann, J., Schick, F.,Cegan, A., Elcnerova, M., Schleicher, E., Fritsche, A., and Haring, H.U.(2010). Circulating palmitoleate strongly and independently predicts insulinsensitivity in humans. Diabetes Care 33, 405–407.

Su, Q., Tsai, J., Xu, E., Qiu, W., Bereczki, E., Santha, M., and Adeli, K. (2009).Apolipoprotein B100 acts as a molecular link between lipid-induced endo-plasmic reticulum stress and hepatic insulin resistance. Hepatology 50, 77–84.

Summers, S.A. (2006). Ceramides in insulin resistance and lipotoxicity. Prog.Lipid Res. 45, 42–72.

Tabas, I., and Ron, D. (2011). Integrating the mechanisms of apoptosisinduced by endoplasmic reticulum stress. Nat. Cell Biol. 13, 184–190.

Takei, D., Ishihara, H., Yamaguchi, S., Yamada, T., Tamura, A., Katagiri, H.,Maruyama, Y., and Oka, Y. (2006). WFS1 protein modulates the free Ca(2+)concentration in the endoplasmic reticulum. FEBS Lett. 580, 5635–5640.

Thaler, J.P., Yi, C.X., Schur, E.A., Guyenet, S.J., Hwang, B.H., Dietrich, M.O.,Zhao, X., Sarruf, D.A., Izgur, V., Maravilla, K.R., et al. (2012). Obesity is asso-ciated with hypothalamic injury in rodents and humans. J. Clin. Invest. 122,153–162.

Timmins, J.M., Ozcan, L., Seimon, T.A., Li, G., Malagelada, C., Backs, J.,Backs, T., Bassel-Duby, R., Olson, E.N., Anderson, M.E., and Tabas, I.(2009). Calcium/calmodulin-dependent protein kinase II links ER stress withFas and mitochondrial apoptosis pathways. J. Clin. Invest. 119, 2925–2941.

Traaseth, N.J., Ha, K.N., Verardi, R., Shi, L., Buffy, J.J., Masterson, L.R., andVeglia, G. (2008). Structural and dynamic basis of phospholamban and sarco-lipin inhibition of Ca(2+)-ATPase. Biochemistry 47, 3–13.

Trusina, A., Papa, F.R., and Tang, C. (2008). Rationalizing translation attenua-tion in the network architecture of the unfolded protein response. Proc. Natl.Acad. Sci. USA 105, 20280–20285.

Tsutsumi, A., Motoshima, H., Kondo, T., Kawasaki, S., Matsumura, T.,Hanatani, S., Igata, M., Ishii, N., Kinoshita, H., Kawashima, J., et al. (2011).Caloric restriction decreases ER stress in liver and adipose tissue in ob/obmice. Biochem. Biophys. Res. Commun. 404, 339–344.

Turpin, S.M., Lancaster, G.I., Darby, I., Febbraio, M.A., and Watt, M.J. (2006).Apoptosis in skeletal muscle myotubes is induced by ceramides and is posi-tively related to insulin resistance. Am. J. Physiol. Endocrinol. Metab. 291,E1341–E1350.

Ueda, K., Kawano, J., Takeda, K., Yujiri, T., Tanabe, K., Anno, T., Akiyama, M.,Nozaki, J., Yoshinaga, T., Koizumi, A., et al. (2005). Endoplasmic reticulumstress induces Wfs1 gene expression in pancreatic beta-cells via transcrip-tional activation. Eur. J. Endocrinol. 153, 167–176.

Uehara, T., Nakamura, T., Yao, D., Shi, Z.-Q., Gu, Z., Ma, Y., Masliah, E.,Nomura, Y., and Lipton, S.A. (2006). S-nitrosylated protein-disulphide isom-erase links protein misfolding to neurodegeneration. Nature 441, 513–517.

634 Cell Metabolism 15, May 2, 2012 ª2012 Elsevier Inc.

Unger, R.H. (2002). Lipotoxic diseases. Annu. Rev. Med. 53, 319–336.

Unger, R.H., and Scherer, P.E. (2010). Gluttony, sloth and the metabolicsyndrome: a roadmap to lipotoxicity. Trends Endocrinol. Metab. 21, 345–352.

Urano, F., Wang, X., Bertolotti, A., Zhang, Y., Chung, P., Harding, H.P., andRon, D. (2000). Coupling of stress in the ER to activation of JNK protein kinasesby transmembrane protein kinase IRE1. Science 287, 664–666.

Ussher, J.R., Koves, T.R., Cadete, V.J.J., Zhang, L., Jaswal, J.S., Swyrd, S.J.,Lopaschuk, D.G., Proctor, S.D., Keung,W.,Muoio, D.M., and Lopaschuk, G.D.(2010). Inhibition of de novo ceramide synthesis reverses diet-induced insulinresistance and enhances whole-body oxygen consumption. Diabetes 59,2453–2464.

Verkhratsky, A. (2005). Physiology and pathophysiology of the calcium store inthe endoplasmic reticulum of neurons. Physiol. Rev. 85, 201–279.

Vessby, B. (2000). Dietary fat and insulin action in humans. Br. J. Nutr. 83(Suppl 1 ), S91–S96.

Walter, P., and Ron, D. (2011). The unfolded protein response: from stresspathway to homeostatic regulation. Science 334, 1081–1086.

Wang, L., Folsom, A.R., Zheng, Z.J., Pankow, J.S., and Eckfeldt, J.H.; ARICStudy Investigators. (2003). Plasma fatty acid composition and incidence ofdiabetes in middle-aged adults: the Atherosclerosis Risk in Communities(ARIC) Study. Am. J. Clin. Nutr. 78, 91–98.

Wang, Y., Vera, L., Fischer, W.H., and Montminy, M. (2009). The CREB coac-tivator CRTC2 links hepatic ER stress and fasting gluconeogenesis. Nature460, 534–537.

Wei, Y., Wang, D., Topczewski, F., and Pagliassotti, M.J. (2006). Saturatedfatty acids induce endoplasmic reticulum stress and apoptosis independentlyof ceramide in liver cells. Am. J. Physiol. Endocrinol. Metab. 291, E275–E281.

Wen, H., Gris, D., Lei, Y., Jha, S., Zhang, L., Huang,M.T.-H., Brickey, W.J., andTing, J.P.-Y. (2011). Fatty acid-induced NLRP3-ASC inflammasome activationinterferes with insulin signaling. Nat. Immunol. 12, 408–415.

Xiao, C., Giacca, A., Carpentier, A., and Lewis, G.F. (2006). Differential effectsof monounsaturated, polyunsaturated and saturated fat ingestion on glucose-stimulated insulin secretion, sensitivity and clearance in overweight andobese, non-diabetic humans. Diabetologia 49, 1371–1379.

Xiao, C., Giacca, A., and Lewis, G.F. (2011). Sodium phenylbutyrate, a drugwith known capacity to reduce endoplasmic reticulum stress, partiallyalleviates lipid-induced insulin resistance and beta-cell dysfunction in humans.Diabetes 60, 918–924.

Xu, L., Spinas, G.A., and Niessen, M. (2010). ER stress in adipocytes inhibitsinsulin signaling, represses lipolysis, and alters the secretion of adipokineswithout inhibiting glucose transport. Horm. Metab. Res. 42, 643–651.

Yang, L., Li, P., Fu, S., Calay, E.S., and Hotamisligil, G.S. (2010). Defectivehepatic autophagy in obesity promotes ER stress and causes insulin resis-tance. Cell Metab. 11, 467–478.

Yang, Z.-H., Miyahara, H., and Hatanaka, A. (2011). Chronic administration ofpalmitoleic acid reduces insulin resistance and hepatic lipid accumulation inKK-Ay Mice with genetic type 2 diabetes. Lipids Health Dis. 10, 120.

Zeng, L., Lu, M., Mori, K., Luo, S., Lee, A.S., Zhu, Y., and Shyy, J.Y.-J. (2004).ATF6 modulates SREBP2-mediated lipogenesis. EMBO J. 23, 950–958.

Zhang, P., McGrath, B., Li, S., Frank, A., Zambito, F., Reinert, J., Gannon, M.,Ma, K., McNaughton, K., and Cavener, D.R. (2002). The PERK eukaryoticinitiation factor 2 alpha kinase is required for the development of the skeletalsystem, postnatal growth, and the function and viability of the pancreas.Mol. Cell. Biol. 22, 3864–3874.

Zhou, Y., Lee, J., Reno, C.M., Sun, C., Park, S.W., Chung, J., Lee, J., Fisher,S.J., White, M.F., Biddinger, S.B., and Ozcan, U. (2011). Regulation of glucosehomeostasis through a XBP-1-FoxO1 interaction. Nat. Med. 17, 356–365.