The endoplasmic reticulum (ER) is the principal site for folding and maturation of transmembrane, secretory and ER-resident proteins (Ref. Reference Lee1). Several post-translational modifications take place in the ER, including disulfide-bond formation, cleavage of the ER signal-recognition peptide, N-linked glycosylation and addition of glycophosphatidylinositol anchors. These post-translational modifications are required for the correct folding of transmembrane and secretory proteins (Ref. Reference Ellgaard and Helenius2). Importantly, ER-resident molecular chaperones are required for the successful completion of each folding step.

N-linked glycosylation of newly synthesised proteins entering the ER lumen is an essential part of protein maturation and folding in eukaryotic cells. During this process, the attached N-linked oligosaccharides become substrates for ER chaperones, including calnexin and calreticulin. These chaperones assist in protein folding by binding to the oligosaccharides, as well as unfolded regions in the glycoprotein (Refs Reference Hebert, Foellmer and Helenius3, Reference Sitia and Braakman4, Reference Ma and Hendershot5). As maturation continues, glucosidase enzymes remove glucose residues from the N-linked oligosaccharides. If the protein is properly folded, it is translocated to the Golgi complex for further processing (Refs Reference Hebert, Foellmer and Helenius3, Reference Ma and Hendershot5). If, however, protein folding is impaired, additional cycles of glucose addition (catalysed by glucosyltransferase), removal and chaperone binding continue until proper folding is achieved.

In addition to calnexin and calreticulin, the glucose-regulated proteins (GRPs), including GRP78 and GRP94, assist in proper protein folding by recognising exposed hydrophobic domains (Refs Reference Lee1, Reference Flynn6). GRP78 binding to ATP, and its subsequent conversion to ADP, causes a conformational change in GRP78 that enhances its affinity to unfolded proteins (Ref. Reference Ma and Hendershot5). The release of ADP from GRP78 by BiP-associated protein promotes another conformational change in GRP78 that releases the unfolded protein (Refs Reference Ma and Hendershot5, Reference Chung, Shen and Hendershot7). It is this repeated binding and release of GRP78 that promotes correct protein folding (Refs Reference Ma and Hendershot5, Reference Hendershot8).

ER stress and the unfolded protein response

Conditions that impair or change the folding capacity of the ER might lead to ER stress and induce the unfolded protein response (UPR), an integrated, intracellular signalling pathway that induces translational inhibition and ER chaperone upregulation (Refs Reference Kaufman9, Reference Ron10) (Fig. 1). Pharmacological disruption of ER homeostasis and protein folding can be achieved through the depletion of ER calcium (Ca²⁺) stores with Ca²⁺ ionophores or sarcoplasmic/ER Ca²⁺-ATPase pump inhibitors. Depletion of ER Ca²⁺ in vitro causes the accumulation of unfolded proteins by decreasing ER chaperone activity and their capacity to fold proteins. Physiologically, induction of ER stress can result from changes in the rate of protein synthesis and folding, ER Ca²⁺ homeostasis or glucose deprivation (Refs Reference Kaufman9, Reference Ron10).

Figure 1.

ER stress and the UPR. The UPR responds to ER stress through activation of the (a) IRE1, (b) ATF6 and (c) PERK pathways. Activation of these transmembrane proteins occurs following dissociation of GRP78 in response to ER stress. Once activated, the UPR functions as an intracellular signalling pathway to attenuate protein translation through eIF2α phosphorylation, increased ER chaperone expression and enhanced degradation of unfolded proteins by the proteasome. Abbreviations: ATF4, activating transcription factor 4; eIF2, eukaryotic translation initiation factor 2; ER, endoplasmic reticulum; ERAD, endoplasmic-reticulum-associated degradation; GRP78, glucose-related protein 78; IRE1, type-I ER transmembrane protein kinase; PERK, protein kinase receptor-like ER kinase; S1P and S2P, site-1 and site-2 proteases; sXBP1, spliced form of X-box-binding protein 1; UPR, unfolded protein response; uXBP1, unspliced form of X-box-binding protein 1.

Disruption in ER function (termed ER stress) that interferes with the folding and maturation of newly synthesised proteins induces the UPR. The UPR is mediated by three ER sensors: type-I ER transmembrane protein kinase (IRE1, encoded by ERN1), activating transcription factor 6 (ATF6) and the proline-rich, extensin-like receptor (PKR)-like ER kinase (PERK). Under normal homeostatic conditions, UPR activation is repressed by GRP78 binding to the ER lumenal domains of IRE1, ATF6 and PERK. However, following ER stress, dissociation of GRP78 causes activation of the IRE1, ATF6 and PERK pathways (Refs Reference Lee1, Reference Kaufman9, Reference Ron10, Reference Lawrence de Koning11, Reference Rutkowski and Kaufman12).

IRE1 is an ER transmembrane protein that binds to GRP78 under non-ER stress conditions and becomes active following dimerisation and autophosphorylation in the presence of unfolded proteins (Refs Reference Kaufman9, Reference Ron and Hubbard13, Reference Pincus14). Recent evidence suggests that activation of IRE1 occurs not only following GRP78 dissociation, but upon the subsequent binding of unfolded proteins to the IRE1 lumenal domain (Ref. Reference Pincus14). In this model, GRP78 reduces UPR sensitivity by inhibiting the binding of unfolded proteins to IRE1 under conditions of low stress and returning IRE1 to an inactive state when ER stress is alleviated (Ref. Reference Pincus14). Following autophosphorylation, the RNase property of IRE1 splices mRNA from the X-box-binding protein 1 gene (XBP1), removing an intron and enabling translation into an active transcription factor that induces ER chaperone gene expression and components of the ER-associated degradation (ERAD) pathway (Refs Reference Yoshida15, Reference Yoshida16, Reference Tirasophon17).

Similarly to IRE1, ATF6 is a transmembrane protein that binds GRP78 under non-ER stress conditions. Golgi localisation signals (GLSs) in the lumenal domain of ATF6 are exposed following GRP78 dissociation in the presence of unfolded proteins. The exposed GLSs promote the translocation of ATF6 from the ER to the Golgi (Refs Reference Chen, Shen and Prywes18, Reference Shen19). Proteolytic cleavage of ATF6 occurs in the Golgi by the site-1 and site-2 proteases (S1P and S2P, encoded by MBTPS1 and MBTPS2) (Refs Reference Haze20, Reference Ye21), which also cleave the sterol regulatory element binding proteins (SREBPs) (see below). Following protease cleavage, ATF6 translocates to the nucleus and induces XBP1 gene expression, thereby providing more substrate for IRE1 (Refs Reference Yoshida16, Reference Lee22).

Corresponding to the upregulation of UPR response genes, the accumulation of misfolded proteins leads to a decrease in protein translation via the PERK pathway. PERK is an ER transmembrane sensor that binds GRP78, and upon its dissociation, causes activation of its cytosolic kinase domain (Ref. Reference Ron10). ER-stress-induced activation of PERK leads to the phosphorylation of eukaryotic translation initiation factor 2α (eIF2α), causing a general inhibition of protein translation (Ref. Reference Harding, Zhang and Ron23). Although eIF2α phosphorylation inhibits general protein translation, certain UPR-associated mRNAs (including ATF4 and GRP78) overcome translation inhibition through an internal ribosome entry site (IRS) that allows their translation during eIF2α phosphorylation (Ref. Reference Ron10). Recently, studies have also demonstrated that sterol regulatory element binding protein-1 (SREBP-1) (discussed in detail below) can be translated by an IRS (Ref. Reference Damiano24).

Together, IRE1, ATF6 and PERK respond to ER stress by inhibiting general protein translation while specifically upregulating the expression of ER chaperones as well as cellular factors that comprise the ERAD pathway. Initially, the UPR provides a protective advantage for the cell, but prolonged ER stress activates UPR-dependent pathways inducing apoptosis (Refs Reference Feng25, Reference Hossain26, Reference Morishima27, Reference Nakagawa28).

Cellular cholesterol metabolism

In addition to being a precursor of steroid hormones, cholesterol is essential for the proper functioning of cellular membranes (Ref. Reference Maxfield and Wustner29). Although its distribution within membranes is essential for many biological functions, excess cholesterol can be toxic (Ref. Reference Maxfield and Tabas30). Therefore, levels of cholesterol are tightly regulated through cellular uptake, biosynthesis, trafficking, storage and excretion.

Exogenously derived cholesterol arrives at the cell membrane in low-density lipoprotein (LDL) particles that enter cells through an endocytic pathway regulated by the expression of the LDL receptor (LDLR). Through a series of steps, LDL-derived cholesterol exits the endocytic compartment and is transported to various cellular membranes, as required (Ref. Reference Chang31). Intracellular cholesterol trafficking involves the movement of cholesterol to or from various intracellular compartments such as the late endocytic compartment, the ER and cytosolic lipid droplets (Ref. Reference Soccio and Breslow32). Cholesterol is transported throughout the cell by incorporation into transport vesicles or by nonvesicular transport, including binding to cytosolic carrier proteins (Refs Reference Maxfield and Wustner29, Reference Soccio and Breslow32). In addition to obtaining cholesterol from external sources, cells have the capacity to endogenously synthesise cholesterol. The cholesterol biosynthetic pathway is made up of many enzymes that are regulated by the SREBPs. HMG-CoA reductase (HMGCR), the rate-limiting enzyme in the cholesterol biosynthesis pathway, is localised to the ER membrane where it converts HMG-CoA to mevalonate. Mevalonate production is the first step in a series of enzymatic reactions leading to cholesterol synthesis (Ref. Reference Espenshade and Hughes33).

Levels of intracellular cholesterol are highly regulated because excess unesterified cholesterol disrupts membranes, leading to cell toxicity (Ref. Reference Maxfield and Tabas30). Increased membrane cholesterol negatively regulates expression of LDLR and the enzymes responsible for cholesterol biosynthesis (described in detail below). In addition, excess cholesterol is transported to the ER for esterification by acyl-coenzyme A cholesterol acyltransferase 1 (ACAT1) (Refs Reference Chang31, Reference Sakashita34). ACAT1 is an ER-membrane-bound protein that is responsible for the esterification of free cholesterol and its subsequent storage in cytoplasmic lipid droplets or use in lipoprotein assembly for transport and excretion (Ref. Reference Chang35).

Regulation of SREBP activation

The SREBP regulatory pathway (Fig. 2) was determined by the characterisation of mutant Chinese hamster ovary cell lines with defects in sterol feedback regulation (Ref. Reference Hua52). The study of sterol-resistant cells led to the identification of the SREBP-cleavage-activating protein (SCAP) (Ref. Reference Hua52). The N-terminal domain of SCAP, with its membrane-spanning segments, provides an ER membrane anchor, whereas the C-terminal domain projects into the cytosol (Ref. Reference Nohturfft, Brown and Goldstein53). Co-immunoprecipitation experiments demonstrated that the C-terminal domains of SCAP and SREBP form a complex that is required for SREBP proteolytic cleavage (Refs Reference Sakai54, Reference Sakai55).

Figure 2.

ER stress and SREBP activation. ER-stress-induced activation of SREBP can occur in cells by three potential mechanisms. (a) Phosphorylation of eIF2α can subsequently decrease INSIG1 protein levels. This allows the SCAP–SREBP complex to move to the Golgi where SREBP is activated by the S1P and S2P proteases. (b) Following ER stress, GRP78 dissociates from the SCAP–SREBP complex, which moves to the Golgi, where it is subsequently activated following its cleavage by S1P and S2P. (c) Activation of specific caspases can lead to direct cleavage of the SREBPs from the ER membrane. Abbreviations: bHLH-Zip, basic helix–loop–helix–zipper domain; eIF2, eukaryotic translation initiation factor 2; ER, endoplasmic reticulum; GRP78, glucose-related protein 78; INSIG, insulin-induced gene; IRE1, type-I ER transmembrane protein kinase; PERK, protein kinase receptor-like ER kinase; Reg, regulatory domain; S1P and S2P, site-1 and site-2 proteases; SCAP, SREBP-cleavage activating protein; SRE, sterol regulatory element; SREBP, sterol regulatory element binding protein; WD, WD-40 repeat domain.

Cholesterol causes the SCAP–SREBP complex to bind to an ER-retention protein through an interaction involving the sterol-sensing domain of SCAP (Ref. Reference Yang, Goldstein and Brown56). Mutations within the sterol-sensing domain disrupt SCAP's capacity to respond to sterols (Refs Reference Nohturfft57, Reference Nohturfft, Brown and Goldstein58). The identity of the ER-retention factor was determined to be insulin-induced gene-1 and -2 (INSIG), which binds to the sterol-sensing domain of SCAP and retains the SCAP–SREBP complex in the ER in the presence of sterols (Refs Reference Yang59, Reference Yabe, Brown and Goldstein60). Following a decline in membrane cholesterol, the INSIGs dissociate from SCAP, allowing the SCAP–SREBP complex to become incorporated into COPII vesicles for transport to the Golgi (Refs Reference Yang, Goldstein and Brown56, Reference Nohturfft57, Reference Sun61, Reference Sun62). Using recombinant COPII proteins, it was shown that in sterol-depleted cells, the COPII proteins bind to a specific six-amino-acid sequence (MELADL) on SCAP (Ref. Reference Sun61). Together, these findings suggest that membrane cholesterol blocks the binding of COPII proteins to the MELADL sequence through the interaction of INSIG with SCAP. When membrane cholesterol levels decrease, dissociation of INSIG from SCAP allows COPII proteins to bind to MELADL and ultimately transport SCAP–SREBP to the Golgi for activation.

To upregulate the transcription of numerous genes responsible for lipid biosynthesis and uptake, the SREBPs must be proteolytically cleaved to their mature forms. A two-step proteolytic process has been described in which SREBPs are released from the ER membrane (Refs Reference Hua63, Reference Sakai64). The first cleavage by S1P separates SREBP into two halves that remain bound to the membrane (Ref. Reference Sakai64). Following cleavage of SREBP within its luminal loop, a second protease S2P cleaves the N-terminal fragment within its membrane-spanning domain. It was determined that SREBP cleavage by S1P and S2P occurs in the Golgi and not as previously thought in the ER, and it was concluded that sterols regulate the cleavage of SREBPs by modifying the ability of SCAP to transport SREBPs to the Golgi, where they interact with S1P or S2P (Refs Reference Nohturfft, Brown and Goldstein58, Reference Nohturfft65).

In addition to the classic ER cholesterol-mediated regulation of SREBP-2, the predominant form of SREBP in the liver, SREBP-1c, is expressed and activated through nuclear liver X receptor and insulin signalling (Refs Reference Hegarty66, Reference Yellaturu67, Reference Yabe68). Interestingly, inhibition of endogenous cholesterol synthesis results in an expected increase in SREBP-2 activation but a paradoxical decrease in liver SREBP-1c (Refs Reference Sheng47, Reference Shimomura69). It was determined that levels of insulin reciprocally affect levels of INSIG-2, which induces SREBP-1c activation (Refs Reference Yellaturu67, Reference Yabe68). Together, regulation of the SREBPs differs between isoforms and tissues by changes in lipid and hormone signalling.

ER stress and SREBP activation

Although the accumulation of ER membrane cholesterol induces ER stress and apoptosis (Refs Reference Feng25, Reference Maxfield and Tabas30), it has been reported that disruption in ER function, leading to the activation of the UPR, induces lipid dysregulation (Refs Reference Esfandiari70, Reference Ji and Kaplowitz71, Reference Kim72, Reference Werstuck73, Reference Kammoun74, Reference Colgan75). ER-stress-induced lipid dysregulation was observed in studies with homocysteine (Hcy), a thiol-containing amino acid reported to be an independent risk factor for cardiovascular disease (Refs Reference Zhou and Austin76, Reference Outinen77). In these studies, higher Hcy levels upregulate ER-stress-response genes and SREBP expression in human umbilical vein endothelial cells (Refs Reference Outinen77, Reference Outinen78). Subsequently, it was determined that Hcy-induced ER stress promotes hepatic steatosis through the activation of SREBPs and increased expression of genes essential for the biosynthesis and uptake of cholesterol and triglycerides (Ref. Reference Werstuck73). Consistent with these findings (Refs Reference Outinen77, Reference Outinen78), cultured human hepatocytes responded to Hcy by increasing the expression of UPR-responsive genes GRP78 and GADD153/CHOP, and overexpression of GRP78 in vitro inhibited ER-stress-induced SREBP activation (Ref. Reference Werstuck73). To verify the pathological mechanism in vivo, diet-induced hyperhomocysteinaemia caused hepatic accumulation of cholesterol and triglycerides in wild-type C57 mice. The observed hepatic steatosis was associated with ER stress as well as activation of the UPR and SREBP pathways (Ref. Reference Werstuck73). Subsequent publications have confirmed that ER stress is associated with activated SREBP and lipid dysregulation (Refs Reference Esfandiari70, Reference Ji and Kaplowitz71, Reference Colgan75, Reference Woo79, Reference Bobrovnikova-Marjon80).

Although the mechanism of ER-stress-induced SREBP activation is yet to be finalised, current literature suggests three potential mechanisms by which ER stress induces SREBP activation: caspase-induced SREBP cleavage, eIF2α-phosphorylation-dependent downregulation of INSIG and GRP78 dissociation from the SCAP–SREBP complex (Fig. 2).

Experiments conducted to identify the cellular proteases that activate SREBPs originally demonstrated that caspase-3 (Ref. Reference Wang81) and caspase-7 (Ref. Reference Pai, Brown and Goldstein82) can cleave both SREBP-1 and SREBP-2 at a site different from S1P and that proteolytic cleavage occurs in the presence of cholesterol (Refs Reference Wang81, Reference Pai, Brown and Goldstein82, Reference Wang83). Using a reporter plasmid controlled by the sterol response element (SRE) of the human LDLR, it was determined that SRE-dependent gene expression is increased during the initial stages of apoptosis and can be blocked by caspase inhibitors (Ref. Reference Higgins and Ioannou84). These findings suggest that caspase-3 or other caspases might function independently of apoptotic cell death and have a role in cellular lipid metabolism. Although ER stress can lead to apoptosis and is associated with SREBP activation (Refs Reference Esfandiari70, Reference Ji and Kaplowitz71, Reference Kim72, Reference Werstuck73, Reference Kammoun74, Reference Colgan75, Reference Outinen78, Reference Bobrovnikova-Marjon80, Reference Wang81, Reference Pai, Brown and Goldstein82, Reference Wang83, Reference Higgins and Ioannou84, Reference Lee and Ye85, Reference Rao86, Reference Shiraishi87), the effect of ER stress on SREBP cleavage seems to occur through the conventional S1P–S2P proteolytic pathway (Refs Reference Colgan75, Reference Bobrovnikova-Marjon80, Reference Lee and Ye85).

Although ER-stress-induced SREBP-2 activation occurs in the presence of sterols, it might occur through the conventional pathway following intracellular cholesterol depletion. Translational inhibition following ER-stress-induced eIF2α phosphorylation causes SREBP-2 activation and is correlated with INSIG-1 degradation (Ref. Reference Lee and Ye85). In vivo deletion of PERK in mouse mammary epithelium reduced the levels of SREBP1 gene expression and showed a significant reduction in the free fatty acid content of secreted milk (Ref. Reference Bobrovnikova-Marjon80). Further experimentation using mouse embryonic fibroblasts deficient in PERK provided evidence that SREBP activation and lipid accumulation are dependent on PERK and eIF2α phosphorylation and associated with reductions in INSIG-1 protein expression (Ref. Reference Bobrovnikova-Marjon80). Because INSIG-1 binds and retains the SCAP–SREBP complex in the ER, it has been suggested that inhibition of protein synthesis during ER stress might decrease the levels of INSIG-1 protein, allowing SCAP–SREBP to exit the ER (Refs Reference Bobrovnikova-Marjon80, Reference Lee and Ye85). Alternatively, disruption of the INSIG–SCAP interaction under conditions that cause ER stress was also proposed as a mechanism to explain these findings.

Recently, an additional mechanism to eIF2α-phosphorylation-induced SREBP-2 activation under ER stress conditions has been described (Ref. Reference Kammoun74). GRP78 was overexpressed in murine liver to determine its effect on ER stress and SREBP activation. Consistent with previous studies showing that in vitro overexpression of GRP78 could attenuate ER-stress-induced SREBP activation and lipid accumulation (Ref. Reference Werstuck73), hepatic overexpression of GRP78 resulted in decreased markers of ER stress, SREBP-1 and SREBP-2 target gene expression, as well as hepatic lipid accumulation in ob/ob obese mutant mice (Ref. Reference Kammoun74). In addition to the obvious role of GRP78 in reducing ER stress, evidence for a direct effect on SREBP was presented (Fig. 2). Experiments showed that GRP78 associates with the SREBP-1 precursor complex and suggested that GRP78 might interact with SCAP and retain it in the ER under non-ER stress conditions (Ref. Reference Kammoun74). This mechanism is comparable to the conventional role of INSIG binding to SCAP and anchoring SREBP in the ER. Thus, GRP78 might sequester the SREBPs in the ER through interaction with its regulatory complex in a manner similar to its regulation of other ER membrane proteins such as ATF6 (Ref. Reference Kammoun74).

Together, these experiments provide evidence that ER-stress-induced SREBP activation probably constitutes several mechanisms. ER stress causes eIF2α-phosphorylation-induced translation inhibition, INSIG-1 degradation and release of the SCAP–SREBP complex from the ER (Ref. Reference Lee and Ye85). Experiments that overexpressed the ER chaperone GRP78 observed a decrease in ER-stress-induced SREBP activation, suggesting a possible link between GRP78 and the SCAP–SREBP complex (Refs Reference Werstuck73, Reference Kammoun74). Future experiments will be necessary to directly examine the link between GRP78 and ER-stress-induced SREBP activation.

SREBP dysregulation and disease

Various pathologies have been associated with differential regulation of the SREBP pathway. Although transgenic mice expressing a constitutive active form of SREBP-1 and SREBP-2 developed hepatic steatosis with hepatocyte cholesterol and triglyceride accumulation, peripheral white adipose tissue decreased with symptoms associated with lipodystrophy (Refs Reference Horton51, Reference Shimano88). In contrast to the in vivo overexpression of SREBP-1a in mouse adipose tissue associated with adipocyte hypertrophy, increased fatty acid secretion and hepatic steatosis (Ref. Reference Horton89), the overexpression of the SREBP-1c isoform in mouse adipose tissue inhibits adipocyte differentiation with a resulting phenotype that includes lipodystrophy, diabetes and hepatic steatosis (Ref. Reference Shimomura90). As described above, diet-induced hyperhomocysteinaemia caused ER-stress-induced SREBP activation and hepatic steatosis (Ref. Reference Werstuck73). Attenuation of ER stress by overexpression of GRP78 blocked SREBP activation and decreased the expression of genes responsible for cholesterol and fatty acid biosynthesis (Refs Reference Werstuck73, Reference Kammoun74). Acute pharmacological renal toxicity is associated with proximal tubule lipid accumulation and has been extensively studied in response to ischaemia and cytotoxicity (Refs Reference Zager and Johnson91, Reference Zager92, Reference Zager93, Reference Zager94, Reference Zager, Johnson and Hanson95). Interestingly, common to these studies is the increase in the expression of enzymes and receptors responsible for the biosynthesis and uptake of cholesterol and triglycerides (Refs Reference Zager and Johnson91, Reference Zager92, Reference Zager93, Reference Zager94, Reference Zager, Johnson and Hanson95). Because the accumulation of lipids following cytotoxic stress has been suggested to provide an advantage to cells by protecting them from further injury (Refs Reference Zager96, Reference Zager97, Reference Zager and Kalhorn98), it will be important in future experiments to determine whether the activation of SREBP protects renal tubule cells from subsequent cytotoxic injury. The importance of the UPR in protecting cells from toxicity-induced renal cell damage has been reported (Ref. Reference Kimura99). Mutant GRP78 lacking an ER retention signal was used to inhibit the usual increase in GRP78 expression following ER stress. Following chronic protein overload-induced nephrotoxicity, mice expressing the mutant GRP78 suffered tubular lesions and cell apoptosis (Ref. Reference Kimura99). This study demonstrates the importance of the UPR in the normal physiological maintenance of renal cells challenged with toxic injury. Pharmacologically induced ER stress causes injury to renal proximal tubules resembling acute tubular necrosis induced in humans by ischaemia, infection or toxins. The expression of ER chaperones in proximal tubules was found to be dramatically increased following treatment (Ref. Reference Zinszner100). This model is useful to study acute renal toxicity and future experiments should determine whether upregulation of ER chaperones is linked to SREBP activation and lipid accumulation. Atherosclerosis is a classic pathology that is characterised by higher intracellular lipid accumulation in arterial walls. Although it is well accepted that macrophages resident in atherosclerotic lesions express scavenger receptors that promote the internalisation of modified lipoproteins (Ref. Reference Moore and Freeman101), it has also been found that scavenger receptor knockout mouse models continue to develop atherosclerosis with pronounced foam cell formation (Ref. Reference Manning-Tobin102). It is unknown whether macrophages undergoing ER stress activate SREBP and promote lipid accumulation, which could offer an additional mechanism to explain macrophage foam cell formation and plaque progression. The relevant steps in the maturation of macrophages from circulating monocytes to lesion-resident foam cells include binding and migration, differentiation and lipid accumulation. It has recently been shown that SREBP activation is important during differentiation of monocytes to macrophages (Ref. Reference Ecker103). During cell differentiation, lipid regulation is required for the development of cellular membranes and organelles (Ref. Reference Ecker103) and SREBP has been implicated in vascular endothelial growth factor-induced angiogenesis (Ref. Reference Zhou104) and required for efficient cellular proliferation and migration (Ref. Reference Zhou104). In addition, increased SREBP expression has been reported in atherosclerotic lesions and associated with increased expression and secretion of the pro-inflammatory cytokine IL-8 (Ref. Reference Yeh105). Although it is not known whether these findings are related to ER stress, together they represent mechanisms of SREBP activation that could potentially contribute to atherosclerotic lesion development.

Future directions

Important questions remain to be answered pertaining to the mechanism of ER-stress-induced SREBP activation and its role in clinical pathologies characterised by lipid accumulation. These questions include: (Reference Lee1) How does ER chaperone expression influence SREBP activation? (Reference Ellgaard and Helenius2) How does general protein translation inhibition affect the SREBP–SCAP–INSIG complex? (Reference Hebert, Foellmer and Helenius3) Is ER-stress-induced SREBP activation a normal physiological prosurvival cellular pathway or activator of cell death? (Reference Sitia and Braakman4) Does ER-stress-induced SREBP activation contribute to human disease?

In vivo overexpression of GRP78 has been shown to decrease markers of ER stress, SREBP activation and lipid accumulation (Refs Reference Werstuck73, Reference Kammoun74). In addition, coimmunoprecipitation experiments found that GRP78 binds to the SREBP–SCAP–INSIG complex but could not identify a specific binding partner (Ref. Reference Kammoun74). A hypothesis has been presented in which GRP78 regulates SREBP in a similar manner to ATF6. It was suggested that under ER stress conditions, GRP78 dissociates from SREBP or its regulatory molecules SCAP and INSIG and allows the SCAP–SREBP complex to transport to the Golgi for S1P and S2P activation (Ref. Reference Kammoun74).

The evidence that ER stress is an activator of the SREBPs is overwhelming. Future studies will allow for a better understanding of the underlying mechanisms by which ER stress and SREBP activation could contribute to the development and progression of cardiovascular disease, obesity, chronic kidney disease and diabetes. Based on these important findings, novel therapeutic strategies that attenuate ER-stress-induced SREBP activation could prevent or reduce the risk of these human diseases and their complications.