Remodeling of Nutrient Homeostasis by Unfolded Protein Response

Nature is unlikely to maintain systems that are detrimental to survival. Any pathway that causes illness must also have a meaningful role. Endoplasmic reticulum (ER) stress, which mediates obesity-induced insulin resistance, is no exception as it helps maintain nutrient homeostasis.

About one-third of all proteins are synthesized and folded into their native conformation in ER (1). Unfortunately, as much as 30% of all newly synthesized proteins, especially those abundant in cells, are misfolded. These are generally repaired or cleared during the maturation processes (2). ER is also responsible for protein quality control (3). Unfolded proteins in the ER lumen are identified by several ER stress sensors, including activating transcription factor (ATF) 6, protein kinase RNA-like ER kinase (PERK), and inositol-requiring enzyme 1 (IRE-1). These ER stress sensors trigger cellular adaptation to unfolded protein accumulation that restores normal cell function. This is called the unfolded protein response (UPR). The UPR initially 1) activates a signaling pathway that induces a large number of ER chaperones that assist protein folding, 2) halts protein translation and transcription, and 3) ubiquitinates and degrades unfolded proteins by 26S proteasome and through autophagy. The latter process is called ER-associated degradation. If these adaptations fail to maintain cellular homeostasis, the UPR leads the cell toward apoptosis or dedifferentiation (3).

Chronic UPRs are causally linked to the pathogenesis of human metabolic disease including obesity and type 2 diabetes. Accumulating evidence suggests that obesity promotes ER stress, which is detected as enhanced UPR signaling. In turn, this activates c-Jun NH₂-terminal kinase (JNK) and impairs insulin signaling at the level of insulin receptor substrates (IRSs) in the liver and adipose tissue (4). However, the signals that activate UPR under this condition are still unknown. Defective autophagy (5) and proteasome dysfunction (6) have recently been proposed as molecular links between obesity and ER stress. In the setting of obesity, ER stress in the liver and adipose tissue are assumed to cause insulin resistance and associated hyperinsulinemia (Fig. 1). Conversely, it is possible to postulate that the hyperinsulinemia that is often observed in obesity causes ER stress. Indeed, previous in vitro studies suggest that insulin upregulates UPR markers in cultured murine peritoneal macrophages (7) and human neuroblastoma cells (8).

In this issue, Boden et al. (9) show in vivo data supporting a role for insulin as a cause of ER stress in human adipose tissue. The new report focuses on 4 h and 8 h euglycemic (5.5 mmol/L) clamps using three different insulin concentrations (basal ∼70 pmol/L, medium postprandial ∼480 pmol/L, and high postprandial ∼1,450 pmol/L) and examined subcutaneous adipose tissue biopsies before and after the clamps in 16 healthy overweight and obese subjects. The highest postprandial insulin levels (∼1,450 pmol/L) increased UPR markers including glucose-regulated protein 78 (GRP78), spliced form of X-box–binding protein 1 (Xbp-1s), ATF-4, nuclear ATF-6, and phosphoeukaryotic initiation factor 2α (phospho-e1F2α) at both time points, whereas acutely lowering insulin to below basal levels decreased the markers. Because insulin is an anabolic hormone that enhances global protein synthesis, Boden et al. hypothesized that it increases protein synthesis enough to exceed ER folding capacity, thereby producing ER stress via accumulation of misfolded/unfolded and ubiquitinated proteins. This hypothesis was supported by additional experimental data. Among 25 proteins that increased after 4 h of euglycemic-hyperinsulinemia, 5 ubiquitination pathway proteins increased by 1.8–3.0 fold, ubiquitinated proteins accumulated in adipose tissue after 4 h of euglycemic-hyperinsulinemia, the ER stress pathway and the ubiquitination pathway were both upregulated in a canonical pathway analysis of mRNA and protein data, and insulin-induced posttranslational protein modifications, including acetylations, methylations, nitrosylations, succinylations, and ubiquitinations, were identified in 8 proteins by mass spectrometry.

Hyperinsulinemia increases glucose uptake and intracellular glucose metabolism, leading to the generation of reactive oxygen species in mitochondria. However, Boden et al. ruled out the possible involvement of glucose metabolism–associated oxidative stress in the insulin-induced ER stress by using clamp studies with identical degrees of hyperinsulinemia and different rates of glucose infusion, hyperglycemia, and euglycemia. Under the same degree of hyperinsulinemic clamp, steady-state glucose levels did not significantly affect induction of mRNA for UPR markers. Although increased glucose infusion rates resulted in enhanced glucose uptake into the adipose tissue, this did not upregulate oxidative stress markers, such as urinary excretion of 8-iso-prostaglandin-2α (8-iso-PGF-2α), mRNA levels of nuclear factor erythroid 2–related factor 2 (Nrf-2), heme oxygenase 1 (HO-1), vascular endothelial growth factor (VEGF), or inflammatory markers involved in the nuclear factor-κB (NF-κB) and JNK pathways, or phosphorylation of JNK1 in adipose tissue. Additional studies that were conducted in vitro in cultured 3T3-L1 adipocytes showed that insulin at a dose of 10 nmol/L induced mRNA and protein levels of the UPR markers.

Based on these findings, the new report concluded that acute physiological increases in circulating insulin produces ER stress and UPR possibly via enhancement of protein biosynthesis and posttranslational protein modification that leads to accumulation of unfolded/misfolded and ubiquitinated proteins, rather than via glucose uptake and subsequent oxidative stress. Perhaps the most notable aspect of this report is that these conclusions were reached using human tissues. These observations are of great value because they trigger a discussion concerning the significance of ER stress in energy homeostasis.

Despite the merits of the new report, it also has some notable limitations. One of these relates to the relatively small sample size used in this study, an issue that suggests these results should be confirmed in larger-scale studies. Further, the molecular basis for some of the new observations could be addressed by additional cellular experiments to further support the new findings.

Concerning the pathway involved in insulin-induced ER stress, the observations of enhanced protein biosynthesis of several insulin-responsive proteins and protein modifications do not offer comprehensive proof regarding the burden of misfolded/unfolded proteins. More reliable evidence of protein burden leading to ER stress could be provided by Western blotting of ubiquitinated proteins and electron microscopy of the expansion and disorganization of ER membranes. Interestingly, insulin-induced accumulation of ubiquitinated proteins in the adipose tissue seemed both marginal and variable in this study. This suggests that, besides unfolded protein accumulation, other pathways directly altering ER homeostasis should also be considered. Such pathways, perhaps involving oxidative stress (10), hypoxia (11), and toxic lipid (12), may also be candidates that mediate ER stress in obesity. Moreover, to exclude involvement of glucose uptake in insulin-induced ER stress, it should be confirmed that glucose transport inhibitors do not cancel the insulin-induced ER stress in 3T3-L1 adipocytes.

If the insulin-mediated UPR is dependent on enhanced protein biosynthesis, accumulation of ubiquitinated protein should precede the onset of UPR in the 3T3-L1 adipocytes cultured with insulin, and the insulin-induced ER stress must be canceled with protein biosynthesis inhibitors. In addition to enhanced protein biosynthesis, autophagy (5) and proteasome (6) are candidate targets for insulin to induce ER stress because insulin impairs their function as observed in the state of obesity (Fig. 1).

Boden et al. postulate that insulin-induced UPR may relieve ER stress and its associated insulin resistance. On the other hand, they caution about unrelieved ER stress and insulin resistance caused by long-lasting hyperinsulinemia caused by excessive calorie intake or in diabetic patients treated with high-dose insulin (Fig. 1). An important issue that should be proven is the net effect of insulin on the balance between UPR-mediated relief of ER stress and UPR-induced insulin resistance. Evaluation of some signs of UPR-induced insulin resistance, for example, JNK activation and decreased levels of insulin receptor substrates (6), might help clarify the direction in which UPR remodels a cell—toward recovery or illness.

Finally, insulin is one of the key players in accelerating aging (13). In this regard, it could be postulated that insulin resistance itself is a compensatory adaptation to avoid cellular aging caused by excessive insulin signaling. The new study by Boden et al. (9) pointed to several “black boxes” that warrant further investigation to understand both pathological and physiological ER stresses for remodeling nutrient homeostasis.