Post-burn hepatic insulin resistance is associated with endoplasmic reticulum (ER) stress

Abstract

Insulin resistance with its associated hyperglycemias represents one significant contributor to mortality in burned patients. A variety of cellular stress-signaling pathways are activated as a consequence of burn. A key player in the cellular stress response is the endoplasmic reticulum (ER). Here, we investigated a possible role for ER-stress pathways in the progression of insulin function dysregulation postburn. Rats received a 60% total body surface area thermal injury, and a laparotomy was performed at 24, 72, and 192 h postburn. Liver was harvested before and 1 min after insulin injection (1 IU/kg) into the portal vein, and expression patterns of various proteins known to be involved in insulin and ER-stress signaling were determined by Western blotting. mRNA expression of glucose-6-phosphatase and glucokinase were determined by reverse-transcriptase-polymerase chain reaction and fasting serum glucose and insulin levels by standard enzymatic and enzyme-linked immunosorbent assay techniques, respectively. Insulin resistance indicated by increased glucose and insulin levels occurred starting 24 h postburn. Burn injury resulted in activation of ER stress pathways, reflected by significantly increased accumulation of phospho-PKR-like ER-kinase and phosphorylated inositol requiring enzyme 1, leading to an elevation of phospho-c-Jun N-terminal kinase and serine phosphorylation of insulin receptor substrate (IRS) 1 postburn. Insulin administration caused a significant increase in tyrosine phosphorylation of IRS-1, leading to activation of the phosphatidylinositol 3 kinase/Akt pathway in normal liver. Postburn tyrosine phosphorylation of IRS-1 was significantly impaired, associated with an inactivation of signaling molecules acting downstream of IRS-1, leading to significantly elevated transcription of glucose-6-phosphatase and significantly decreased mRNA expression of glucokinase. Activation of ER-stress signaling cascades may explain metabolic abnormalities involving insulin action after burn.

Figure 1. Study design

Two different studies were performed. After deprivation of food overnight, animals included in the first study (A) were anesthetized at 24, 72 and 192 h following the burn/sham procedure and a laparotomy was performed and liver tissue was harvested for Western blot and RT-PCR analysis (n=6, each group and time point). Prior to tissue harvesting, blood was withdrawn from the right femoral artery for glucose and insulin determinations. Animals included in the second study (B) were subjected to an identical burn/sham procedure. In order to determine the effects of burn injury on hepatic insulin signaling pathways, a laparotomy was performed at 24, 72 and 192 h following the burn/sham procedure. At the respective time points, the right liver lobe was ligated, harvested and placed into liquid nitrogen for subsequent tissue preparations. Then, insulin (1 IU/kg) was injected directly into the portal vein and the remaining of the liver was harvested 1 min thereafter for subsequent tissue processing.

Figure 2. Hyperglycemia and hyperinsulinemia indicate insulin resistance following burn injury

Fasting glucose (A) and fasting insulin levels (B) were determined at various time points post-burn. Throughout the figure, histograms depict serum concentrations of glucose or insulin at fasted levels. Results shown represent six different animals per group, as indicated in the main text. Bars represent means; error bars correspond to S.E.M. Asterisks denote statistical significance: p < 0.05 for every comparison between groups.

Figure 3. Severe burn injury is associated with activation of ER stress signaling pathways and phosphorylation of IRS-1 at its serine binding site

Accumulation of phospho-PERK/PERK (A), phospho-IRE-1/IRE-1 (C), phospho-JNK/JNK (E) and phospho-IRS-1(Ser307)/IRS-1 (G) was determined in liver tissue isolated at 24, 72 and 192 h after burn or sham procedure. GAPDH served as control loading protein. Histograms depict intensities of the phosphorylated protein bands divided by the total form of the respective protein (B, D, F, H). Results shown represent three to five different animals per group, as indicated in the main text. Bars represent means; error bars correspond to S.E.M. Asterisks denote statistical significance: p < 0.05 for every comparison between groups.

Figure 4. Burn injury is associated with impaired activation of IRS-1 at its tyrosine binding site and inhibits PI3K/Akt signaling after insulin administration

Accumulation of hepatic phospho-IRS-1(Tyr612)/IRS-1(A) phospho PI3K/PI3K (C) and phospho-Akt/Akt (E) was determined in both, sham burned and burned animals before (-) and 1 min after insulin injection (+) at 24, 72 and 192 h post-burn or post-sham procedure. GAPDH served as control loading protein. Histograms depict intensities of the phosphorylated protein bands divided by the intensity of the total form of the respective protein band (B, D, F). Results shown represent three different animals per group, as indicated in the main text. Bars represent means; error bars correspond to S.E.M. Asterisks denote statistical significance: p < 0.05 for every comparison between groups.

Figure 5. Burn injury is associated with alterations in transcription of Glucokinase (GK) and Glucose-6-Phosphatase (G-6-P)

Total RNA was isolated from liver tissue of both burned and sham-burned and GK (A) and G-6-P (B) mRNA was assayed by real-time RT-PCR, with 18S ribosomal RNA (18S) used as endogenous control. Mean value of the target was divided by the mean value of the endogenous control to obtain a normalized mean quantity per sample. Results shown represent four different animals per group. Bars represent means; error bars correspond to S.E.M. Asterisks denote statistical significance: p < 0.05 for every comparison between groups.

Figure 6. Molecular mechanisms underlying insulin resistance following thermal injury

Severe burn is associated activation of ER-stress related signaling pathways, indicated by activation of PERK and IRE-1. Phosphorylation of IRE-1 causes activation of c-JNK. Activation of JNK may then lead to phosphorylation of IRS-1 at serine residues which may preclude its tyrosine phosphorylation by the insulin receptor tyrosine kinase, thus resulting in impaired PI3K/Akt signaling and insulin resistance with its associated consequences.