This article is mentioned in:

1. Endoplasmic reticulum (ER) stress

The ER is a key organelle in cell survival and normal cellular function. In the ER, nascent proteins are folded with the assistance of molecular chaperones and folding enzymes. Correctly folded proteins are subsequently transported to the Golgi apparatus, whereas unfolded and misfolded protein are retained in the ER or returned to the cytoplasm by the process of ER-associated degradation (ERAD) and ultimately degraded by the proteasome. Various factors have been reported to contribute to ER stress, including: i) Disturbances in cellular redox regulation, as a consequence of oxidants, hypoxia or reducing agents, which affect the disulfide bonds of proteins in the ER lumen; ii) glucose deprivation, which disrupts N-linked protein glycosylation in the ER; iii) disruption of Ca2+, which impairs the functions of Ca2+-dependent chaperones, including glucose-regulated protein (GRP) 78, GRP94 and calreticulin; iv) viral infections which result in the presence of virus-encoded proteins; v) a high-fat diet and vi) protein mutations (1). ER stress activates a stress response signaling network known as the unfolded protein response (UPR), which is capable of inducing apoptosis and inflammatory reactions (2). Of note, the ER also has a role in lipid biosynthesis, including lipid membrane synthesis and controlling the synthesis of cholesterol and other membrane lipid components (3).

2. ER stress and the UPR

The UPR is mediated by at least three transmembrane proteins, including inositol-requiring enzyme 1 (IRE1), protein kinase R (PKR)-like ER kinase (PERK) and activating transcription factor 6 (ATF6) (2,4). Under unstressed conditions, ER stress transducers, predominantly IRE1, PERK and ATF6, are maintained in an inactive state by binding to the major ER chaperone GRP78 at the ER lumen periphery. Under conditions leading to ER stress, GRP78 is displaced to interact with misfolded luminal proteins, resulting in the release and activation of IRE1, PERK and ATF6. In ER stress, PERK phosphorylates the α subunit of eukaryotic translation initiation factor 2 (eIF2α), leading to a rapid reduction in the initiation of mRNA translation and a reduction in the level of newly synthesized proteins in the ER. However, eIF2α phosphorylation by PERK enables the translation of ATF4. ATF4 promotes apoptosis by inducing the transcription of genes involved in apoptosis, including tribbles homolog 3 and C/EBP homologous protein (CHOP).

ER stress-induced IRE1 activation initiates non-spliceosomal splicing of the mRNA of the transcription factor X-box binding protein 1 (XBP1). XBP1 controls protective responses to ER stress, for example by upregulating the transcription of genes encoding ER chaperones and ERAD components. Using its non-specific endoribonuclease activity, IRE1 induces the degradation of ER-localized mRNAs, a process known as regulated IRE1-dependent decay (5). This process further reduces protein accumulation in the ER. Of note, IRE1 also activates c-Jun N-terminal kinase (JNK) by recruiting the scaffold protein tumor necrosis factor receptor-associated factor 2 (TRAF2), as well as apoptosis signal-regulating kinase (ASK1) and caspase 12 (2,6,7).

Following translocation to the Golgi apparatus under ER stress, ATF6 is cleaved by site 1 and site 2 proteases in the process of regulated intramembrane proteolysis. The cytoplasmic region of ATF6 is an active transcription factor responsible for the transactivation of various genes encoding ER chaperones, ERAD components and protein foldases. ATF6 and PERK also activate nuclear factor κ-light-chain-enhancer of activated B cells (NFκB) during ER stress (8–11). Increased levels of JNK and NFκB promote the production of pro-inflammatory cytokines, which cause further inflammation and ER stress (12).

3. ER stress in adipose tissue (AT)

Adipose tissue is increasingly recognized as a tissue containing a molecular network that connects obesity, adipokine secretion, chronic inflammation and insulin resistance (IR). In AT, adipocytes have been found to constitute ≤50% of the total number of cells, with other cell types including preadipocytes, macrophages and vascular cells. Under conditions of ER stress, AT secretes a number of protein signals and factors. A previous study showed that UPR markers are overexpressed in the AT of obese rodents (13). In adult-derived human adipocyte stem cells (ADHASCs), ER stress has been reported to increase the levels of ER stress genes and eIF2α phosphorylation (14). Furthermore, AT inflammation, which is observed in obesity and diabetes, is associated with the infiltration of macrophages into the AT. This may be triggered by adipocyte death, the secretion of adipokines, including tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6), and increased levels of adipocyte-associated chemokines, including monocyte chemoattractant protein 1 (MCP-1). Similar to other cells that demonstrate a high secretory capacity, including mature B lymphocytes, liver cells and pancreatic β cells, adipocytes have also been reported to adapt their ER capabilities under conditions of stress, particularly in diabetes and obesity (15). ER stress has been observed in human AT in numerous studies, indicating that ER stress may play a crucial role in AT disorders, including inflammation and apoptosis (16–18). In obesity, the AT is poorly oxygenated (19,20), which leads to AT hypoxia. This interferes with disulfide bonding in the ER lumen, resulting in ER stress, initiation of the UPR and increased levels of ER stress markers in the AT, including CHOP and GRP78 (13,21–23). Hosogai et al (22) reported that, in 3T3-L1 adipocytes, hypoxia is associated with ER stress and increased levels of GRP78 and CHOP. Furthermore, using ATF4, GADD34 and ATF3 as markers of the apoptotic pathway, Sharma et al (17) found that ER stress induced apoptosis. Hypoxia does not only stimulate the inflammatory response of macrophages (24,25), but also induces apoptosis and G0/G1-phase cell cycle arrest through AKT and JNK (26). Furthermore, Yin et al (27) found that hypoxia induces cell death by promoting the release of free fatty acids (FFAs) and inhibiting glucose uptake in adipocytes via the inhibition of the insulin signaling pathway.

4. IR in brief

Prior to discussing the association between IR and AT ER stress, an understanding of IR is required. IR is caused by an impaired sensitivity of target organs, including AT, the liver and muscle, to insulin. Insulin regulates glucose uptake in the liver and muscle. Furthermore, insulin is a key regulator of circulating FFA concentrations. In AT, insulin decreases lipolysis and thereby reduces FFA efflux from adipocytes; in skeletal muscle, insulin predominantly induces glucose uptake by stimulating the translocation of glucose transporter type 4 (GLUT4) to the plasma membrane. Furthermore, in the liver, insulin inhibits gluconeogenesis by reducing the activity of key enzymes. Therefore IR has been suggested to increase circulating FFA concentrations, leading to ectopic fat deposition that impedes insulin-mediated glucose uptake in skeletal muscle and increases glucose generation in the liver (28). A combination of IR and abnormalities in insulin secretion can lead to Type 2 diabetes mellitus (T2DM).

5. IR at the molecular level

In normal conditions, insulin is associated with complex signaling cascades. Briefly, insulin receptor-mediated tyrosine phosphorylation of insulin receptor substrates (IRSs) induces the activation of at least two major pathways: The phosphatidylinositol 3-kinase (PI3K)-AKT and mitogen-activated protein kinase (MAPK) pathways. The PI3K-AKT pathway is primarily responsible for the effect of insulin on glucose uptake and the suppression of gluconeogenesis, whereas the MAPK pathway regulates gene expression and interacts with the PI3K-AKT pathway to control cell growth and differentiation.

While tyrosine phosphorylation activates, serine phosphorylation of IRSs at specific serine residues inhibits insulin signaling. IκB kinase-β (IKK-β), JNK 1 and MAPKs are examples of serine kinases that phosphorylate IRS1 and consequently inactivate its insulin signaling activity. Of note, these serine kinases are also mediators of inflammatory signaling pathways, demonstrating that an inhibitory crosstalk may exist between inflammatory and insulin signaling at a molecular level. The association between cytokine signaling and the inhibition of insulin signaling has also been indicated by the presence of molecular mediators, including suppressor of cytokine signaling (SOCS) 1 and 3 and nitric oxide (NO). IL-6 has been shown to induce the activation of SOCS proteins during inflammation, causing the ubiquitinylation and degradation of IRS proteins. Endogenous NO production by inducible NO synthase (iNOS), under the action of numerous inflammatory cytokines, can limit IRS1 and iNOS activity, resulting in reduced AKT activity; AKT is a key mediator of IRS signaling (29).

6. Adipocyte ER stress and IR: Mechanisms and significance

ER stress was first observed by Ozcan et al (13) in the AT of obese mice and proposed as a risk factor for IR. It has since been demonstrated that stress-inducing conditions, such as obesity, are not only associated with ER stress and the apoptosis of β-cells, hepatocytes and adipocytes, but also with metabolic disorders, particularly IR. In patients with IR, treatment with taurine-conjugated ursodeoxycholic acid (TUDCA), a conjugated bile acid derivative that inhibits ER stress-induced apoptosis, was observed to increase insulin sensitivity (30). Furthermore, the administration of chaperones, including 4-phenyl butyric acid (PBA), trimethylamine N-oxide dihydrate, dimethyl sulfoxide and 150 kDa oxygen-regulated protein, which protect cells from ER stress by stabilizing protein conformation and improving ER folding capacity, was found to increase insulin sensitivity in obese diabetic mice (31,32). These data indicate that ER stress in adipocytes, hepatocytes and β-cells may, at least in part, initiate IR, as well as aggravate pre-existing IR, particularly in the context of obesity. The mechanism by which ER stress interferes with insulin receptor signaling is multifactorial, and the role of adipose cells/AT in IR is discussed later in this review.

7. Role of the IRE1-JNK-IRS-1 signaling pathway

It has been suggested that ER stress in adipocytes may disrupt insulin signaling through the activation of IRE1. ER stress induces insulin receptor signaling through increasing the serine phosphorylation and decreasing the tyrosine phosphorylation of IRS-1, leading to IR (33). As mentioned previously, IRS-1 becomes inactivated upon phosphorylation of specific serine residues (34). Several intracellular serine kinases may mediate this IRS-1 phosphorylation, including IKK, JNK, mammalian target of rapamycin (mTOR) and protein kinase C-θ (PKC-θ). Insulin-resistant states, including obesity and T2DM, are associated with the activation of JNK and/or IKK, which leads to serine phosphorylation of IRS1 and thus the induction of IR (35–39). In obesity, JNK is activated by IRE1 kinase activity as a consequence of AT ER. Furthermore, PKR, which is activated by saturated FAs, ceramides and lipopolysaccharide, is capable of inhibiting insulin signaling directly by phosphorylating IRS1 on serine residues or indirectly by stimulating JNK activity (40). The IRE1-JNK signaling pathway has also been reported to directly inhibit cytoplasmic insulin signaling in ob/ob mice due to JNK-mediated phosphorylation of IRS-1 at serine-307 (13). Additionally, the inflammatory cytokines secreted by adipocytes and macrophages in AT in obese patients can activate JNK pathways (41). Furthermore, inflammatory cytokines can impair insulin signaling by interfering with IRS-1-insulin receptor interactions and promoting IRS-1 degradation (42). However, the effect of ER stress on insulin signaling requires further investigation in order to elucidate the specific mechanisms underlying the development of IR in AT in obese patients. In a study on AT from obese volunteers, approximately a two-fold upregulation of phosphorylated JNK-1 and an upregulation of the spliced form of XBP1, which is part of the IRE-1/XBP1 proximal ER stress sensor, was observed in the AT (43) (Fig. 1).

Figure 1 Association between ER stress in adipocytes and IR. The process is multifactorial and highly interlinked. ER, endoplasmic reticulum; IR, insulin resistance; ROS, reactive oxygen species; FFA, free fatty acid; NO, nitric oxide.

8. Role of adipokines

Several investigations in rodents and humans have found an association between IR and ER stress in AT, as well as increased lipolysis and altered adipokine production (13,16–18,44). In ER stress-associated conditions, including obesity, a decrease in leptin and adiponectin secretion and an increase in IL-6 secretion may be observed (44). Furthermore, studies have shown that adiponectin folding and multimerization are impaired in obese states due to a decreased expression of ER disulfide-bond A oxidoreductase-like protein (DsbA-L), which leads to ER stress (45,46). However, CHOP is upregulated under ER stress, which impairs resistin transcription in adipocytes and alters its secretion (47). Of note, a number of studies have shown that a decreased expression of the cytokine adiponectin may promote IR in obesity (48–50). ER stress has been reported to decrease levels of high-molecular-weight fractions as well as total adiponectin in human adipocytes, which may have role in insulin sensitivity and metabolic syndrome (51,52). Decreased levels of adiponectin are observed in conditions associated with excessive nutrient intake, including obesity. Adiponectin signaling targets adenosine monophosphate-activated protein kinase, which is a negative regulator of mTOR. High levels of mTOR in turn induce serine phosphorylation of IRS1. Therefore, when adiponectin levels are low, the inhibition of IRS1 causes IR through the activation of the mTOR signaling pathway. An investigation using tunicamycin and thapsigargin as ER stress inducers in ADHASCs found that ER stress decreased adiponectin and increased TNF-α mRNA expression, in addition to decreasing levels of inhibitor of NFκB-α (IκB) protein (14).

Resistin is an adipokine secreted by preadipocytes of human AT (53) and has been associated with IR. ER stress has been found to reduce resistin mRNA expression in 3T3-L1 adipocytes in a time- and dose-dependent manner, indicating that resistin is regulated by ER stress (47). Furthermore, in adipocytes, ER stress affects IR through resistin. However, further investigations are required as resistin was observed to be positively correlated with IR in a study involving individuals who were obese and diabetic (54), while no correlation was observed in other studies involving individuals with a normal weight and children who were obese (55–57).

9. Role of inflammatory cytokines during ER stress with IR

In a study by Hotamisligil (58), inflammation was observed to induce IR through the inhibition of IRS in the insulin signaling pathway. Furthermore, inflammation and ER stress in AT are known to exhibit a dual relationship. Inflammation can cause ER stress in AT, while ER stress, either due to obesity or obesity-associated consequences, such as IR and increased FFA levels, can increase the activity of anti- or pro-inflammatory proteins. Studies have shown that obese patients exhibit increased plasma levels of C-reactive protein, inflammatory cytokines, including TNF-α, IL-6, MCP-1 and IL-8, and the multifunctional proteins leptin (59) and osteopontin (60,61), indicating a continuous low-grade inflammation in the obese state.

All three subdivisions of ER stress associated with the UPR contribute to the low-grade inflammation associated with obesity. Of note, IRE1 activates JNK and NFκB through TRAF2 and ASK1. In addition, PERK inhibits IκB through eIF2α phosphorylation. PKR also phosphorylates eIF2α and activates JNK and IKK, while the ATF6 branch of the pathway activates NFκB. Finally, ER stress activates cyclic adenosine monophosphate (cAMP) responsive element-binding protein, hepatocyte-specific, which, together with NFκB, augments the transcription of genes involved in inflammation (38,62). The role of IκB and IKK in ER-induced inflammation has been investigated in ADHASCs. The IκB family of proteins controls the activation of NFκB, whose entry into the nucleus can initiate inflammation. Therefore, IκB regulates NFκB by sequestering it in the cytoplasm. In turn, IκB can be degraded by IKK, which results in entry of the NFκB dimers into the nucleus and inflammation. Therefore, ER stress may induce inflammation through a decrease in IκB levels as a consequence of increased IKK activity (58).

IL-6 and TNF-α, levels of which are increased in obesity and IR, have been proposed to induce ER stress and consequently promote IR in a positive feedback manner (63). In ATs, TNF-α inhibits lipogenesis and adiponectin expression via the inhibition of peroxisome proliferator-activated receptor-γ (PPAR-γ)-mediated mechanisms (64–66). It has been suggested that ER stress may activate TNF-α, which is then involved in inflammatory processes and the PPAR-γ-mediated effects on adipocytes (67). Investigations using PPAR activators such as thiazolidinediones have verified that the transcriptional activity of PPAR-γ is required for the maintenance of insulin sensitivity and lipid metabolism (26). A previous study investigated TNF-α in the AT of lean and obese subjects using reverse transcription-polymerase chain reaction methodology (43). Neutralizing TNF-α in rats and knocking out the TNF-α or TNF-α receptor 1 genes in mice have also shown protective effects against IR induced by diet and genetic obesity (68,69). Furthermore, TNF-α activates several IR-related pathways, including IKK-β and SOCS3, in cultured murine adipocytes (70,71). TNF-α has been suggested to affect insulin sensitivity by modifying the expression of IRS1, GLUT4, adiponectin and PPAR-α (66,72). Of note, TNF-α also induces reactive oxygen species (ROS) generation through the activation of the nicotinamide adenine dinucleotide phosphate-oxidase (NADPH oxidase) (73,74). In addition, another inflammatory cytokine, IL-1β, plays an important role in inducing ER stress in adipocytes in obese individuals by increasing the levels of iNOS, which generates NO. In turn, NO inhibits the activity of the ER Ca2+ pump, resulting in a depletion of ER Ca2+ stores, and thus inducing ER stress (75,76).

Obesity-induced ER stress was recently investigated in AT using the 12/15-lipoxygenase (12/15-LO) enzyme, which is involved in a unique inflammatory pathway that regulates the ER stress response in key cells, tissues and organs, including adipocytes, pancreatic islets and the liver (77). The addition of the 12/15-LO cleavage products, 12-hydroxyeicosatetraenoic acid and 12-hydroperoxyeicosatetraenoic acid, to differentiated 3T3-L1 adipocytes was observed to induce the expression and activation of ER stress markers, including GRP78, XBP1, phosphorylated-PERK and phosphorylated-IRE1α. The study also found that 12/15-LO upregulated IL-12 expression. These findings may represent a novel therapeutic strategy for alleviating ER-stress associated inflammation, β-cell dysfunction and IR, thereby reducing metabolic complications associated with visceral adiposity by inhibiting 12/15-LO activation or downstream IL-12 signaling (77).

10. Role of DsbA-L

DsbA-L is a protein that has been proposed to have a role in the downregulation of adiponectin during ER stress-induced autophagy in adipocytes (78). It was revealed that ER stress-induced autophagy plays an important role in obesity-induced adiponectin downregulation in adipocytes, and that increasing the expression of DsbA-L may increase adiponectin levels and lead to enhanced insulin sensitivity in vitro and in vivo. Therefore, increasing DsbA-L expression levels could be a novel approach to protect cells from obesity-induced ER stress and improve insulin sensitivity. Of note, cellular DsbA-L levels were stimulated by the PPAR-γ agonist rosiglitazone, an insulin-sensitizing drug (78).

11. A direct role for ER stress in insulin signaling in AT

Xu et al (44) demonstrated a direct role for ER stress in insulin signaling in adipose cells. ER stress inducers were observed to decrease insulin signaling in 3T3-L1 adipocytes without affecting insulin-stimulated glucose uptake (79), in a manner independent of the IRE1/JNK pathway (44). Notably, ER stress has also been revealed to directly increase lipolysis by downregulating the expression of the lipid droplet-associated protein perilipin A (80,81).

12. Role of FFAs

The roles of hypoxia and inflammatory mediators in AT ER stress have been discussed previously in this review. A number of studies have demonstrated that elevated FFA levels may also have an important role in the induction of ER stress in various cells, including adipocytes (62,82); however, further investigation is required. This theory is supported by the fact that numerous obese individuals exhibit elevated FFA plasma levels (83,84). Hotamisligil (58) showed that inflammation may lead to IR by inhibiting IRSs in the insulin signaling pathway, and it is well established that inflammation inhibits the action of insulin by increasing the levels of FFAs and decreasing those of adiponectin in the blood. Therefore, it is possible that both inflammatory cytokines and FFAs may target IRSs, leading to IR (85,86). High FFA levels have also been shown to downregulate PPAR-γ protein and mRNA expression, further enhancing IR (87). In AT, ER stress promotes FFA efflux from adipocytes, and high levels of circulating FFA have been suggested to be the cellular basis of lipotoxicity, dyslipidemia and IR (13,17,88–90). Furthermore, accumulating evidence suggests that saturated long-chain FFAs, primarily palmitate, and, to a lesser extent, unsaturated long chain FFAs may induce ER stress and mediate β-cell apoptosis, ultimately leading to IR and T2DM (91–96). In a recent study, a unique lipolysis pathway was reported that occurred in response to ER stress in adipocytes. This pathway occurred independently of hormone-sensitive lipolysis, but was associated with elevated cAMP production and PKA activity (81). Chemically induced ER stress was revealed to activate cAMP/PKA and extracellular signal-regulated kinase 1/2 (ERK1/2) and regulate lipolysis in ER-stressed adipocytes, with PKA being an acute regulator and ERK1/2 a chronic regulator (81). Notably, it has been suggested that JNK (97) and PKC (98) may also modulate lipolysis and that ERK1/2 and JNK are activated during ER stress.

13. Role of ROS and ER stress in adipocytes

ROS have been reported to play an important role in the ER stress response in adipocytes, which then directly or indirectly contributes to IR (21). Increased ROS generation has also been observed in response to high levels of FFAs in the AT of obese mice. Furthermore, TNF-α has been shown to induce ROS generation through the activation of NADPH oxidase (74,75,99). As a consequence of their oxidizing effects on nascent proteins and their action on Ca2+ channels, ROS lower Ca2+ availability and increase the number of misfolded and unfolded proteins in the ER, which further increases ER stress (39,100). This FFA-mediated ROS generation model provides another mechanism by which ER stress may be induced in AT and subsequently lead to IR.

14. Conclusion and future directions

In conclusion, this review demonstrates the existence of a strong association between adipocyte ER stress and IR that is complex and multifactorial. Therefore, the inhibition of ER stress may lead to the discovery of novel therapeutics for the treatment of metabolic diseases, including T2DM; investigations are currently underway in this area. To date, at least two chaperones, PBA and TUDCA, have been approved by the Food and Drug Administration and have been shown to relieve ER stress-mediated pathologies, including IR, in hepatocytes, adipocytes and β-cells (31,32,94). However, further investigations are required before these strategies can be applied in patients for the treatment of metabolic and nutritional disorders, including obesity and IR.

Abbreviations:

References