A critical role for eukaryotic elongation factor 1A-1 in lipotoxic cell death

Abstract

The deleterious consequences of fatty acid (FA) and neutral lipid accumulation in nonadipose tissues, such as the heart, contribute to the pathogenesis of type 2 diabetes. To elucidate mechanisms of FA-induced cell death, or lipotoxicity, we generated Chinese hamster ovary (CHO) cell mutants resistant to palmitate-induced death and isolated a clone with disruption of eukaryotic elongation factor (eEF) 1A-1. eEF1A-1 involvement in lipotoxicity was confirmed in H9c2 cardiomyoblasts, in which small interfering RNA-mediated knockdown also conferred palmitate resistance. In wild-type CHO and H9c2 cells, palmitate increased reactive oxygen species and induced endoplasmic reticulum (ER) stress, changes accompanied by increased eEF1A-1 expression. Disruption of eEF1A-1 expression rendered these cells resistant to hydrogen peroxide- and ER stress-induced death, indicating that eEF1A-1 plays a critical role in the cell death response to these stressors downstream of lipid overload. Disruption of eEF1A-1 also resulted in actin cytoskeleton defects under basal conditions and in response to palmitate, suggesting that eEF1A-1 mediates lipotoxic cell death, secondary to oxidative and ER stress, by regulating cytoskeletal changes critical for this process. Furthermore, our observations of oxidative stress, ER stress, and induction of eEF1A-1 expression in a mouse model of lipotoxic cardiomyopathy implicate this cellular response in the pathophysiology of metabolic disease.

Figure 1.

eEF1A-1 expression is disrupted in mutant CHO cells. (A) CHO sequence upstream of the retroviral insertion was obtained by 5′RACE of mutant cDNA (shaded) and corresponds to bases 1-41 of the 5′UTR of hamster eEF1A-1. Rat, mouse, and human sequences are aligned below. (B) Directed PCR for eEF1A-1 and eEF1A-2 expression in wild-type CHO and mutant cDNA. Control reactions contained no cDNA (lanes 5 and 6). Lanes 1, 3, and 5 include forward (F) and reverse (R) primers for either eEF1A-1 (top) or eEF1A-2 (bottom). Lanes 2, 4, and 6 include forward primers for either eEF1A-1 (top) or eEF1A-2 (bottom) and reverse primers for the retroviral sequence (ROSAβgeo). (C and D) eEF1A-1 expression in wild-type CHO and mutant cells was assessed by Northern blotting (C) and immunoblotting (D) using three independent RNA and cell lysate preparations.

Figure 2.

Disruption by insertional mutagenesis or targeted knockdown of eEF1A-1 in CHO cells confers palmitate-resistance. (A) Wild-type and eEF1A-1 mutant CHO cells were incubated with 500 μM palmitate (Palm), 80 nM staurosporine (Staur), 2 μM actinomycin D (Act D), 20 μM cycloheximde (Cyclo), or 10 μM camptothecin (Camp) for 24 h. Cell death was determined by propidium iodide staining and flow cytometry. (B) Cells were incubated as described in A. Apoptosis was determined by FragEL and flow cytometry. (C) Basal eEF1A-1 protein levels in whole cell lysates from wild-type CHO, eEF1A-1 mutant CHO, and stable CHO-derived cell lines expressing control siRNA (siRNA1) or siRNA directed against eEF1A-1 (siRNA2 and siRNA3) were detected by immunoblotting and quantified by densitometry. Inset is a representative blot. (D) Cell lines from C were incubated for 12 h with 500 μM palmitate. Cell death was determined by propidium iodide staining and flow cytometry. All data expressed as mean ± SEM for five independent experiments, ^*p < 0.05.

Figure 3.

Targeted knockdown of eEF1A-1 expression confers palmitate-resistance in H9c2 rat cardiomyoblasts. (A) Myoblasts were incubated for 5 h with palmitate, followed by 30-min incubation with H₂DCFDA. Mean DCF fluorescence, indicative of relative cellular ROS level, was measured by flow cytometry. (B) Cells were incubated for 24 h with palmitate. Apoptosis was determined by FragEL and flow cytometry. (C) Cells were incubated as described in B, and cell death was determined by propidium iodide staining and flow cytometry. (D) Basal eEF1A-1 protein levels in whole cell lysates from wild-type H9c2 myoblast and stable H9c2-derived cell lines expressing control siRNA (siRNA1) or siRNA directed against eEF1A-1 (siRNA2 and siRNA3) were detected by immunoblotting and quantified by densitometry. Inset is a representative blot. (E) Cell lines from D were incubated for 24 h with 500 μM palmitate. Cell death was determined by propidium iodide staining and flow cytometry. (F) Basal eEF1A-1 expression plotted against cell death in response to 500 μM palmitate for CHO-derived lines (wild type, mutant, and siRNA, closed symbols) and for H9c2-derived cell lines (wild type and siRNA, open symbols). All data expressed as mean ± SEM for five independent experiments, ^*p < 0.05.

Figure 4.

eEF1A-1 expression is rapidly induced in response to palmitate and oxidative stress. (A) Wild-type CHO and mutant cells were incubated for 24 h with 2.5 mM H₂O₂ and cell death was quantified by propidium iodide staining and flow cytometry. (B) Wild-type CHO cells were incubated with 500 μM palmitate, followed by detection of eEF1A-1 protein levels in whole cell lysates by immunoblotting. (C) Wild-type CHO cells were incubated with 2.5 mM H₂O₂, followed by detection of eEF1A-1 protein levels in whole cell lysates by immunoblotting. (D) H9c2 cardiomyoblasts were incubated as described in B, followed by detection of eEF1A-1 protein levels in whole cell lysates by immunoblotting. Insets are representative blots. All data expressed as mean ± SEM for five independent experiments, ^*p < 0.05.

Figure 5.

Palmitate-induced ROS activate an ER stress response and induce eEF1A-1 expression and cell death in cardiomyoblasts. (A) Myoblasts were incubated for 1 h with ethanol (vehicle control) or 200 μM α-tocopherol, followed by 5 h with 500 μM palmitate. Expression of ER stress response proteins (GRP78 and CHOP-10) and eEF1A-1 were detected in whole cell lysates by immunoblotting. Cells incubated for 5 h with 2.5 μg/ml tunicamycin were included as a positive control. Relative densitometry values for the representative blots shown are given below each band. Open arrow indicates a nonspecific protein band. (B) Cells were incubated as described in A, followed by incubation for 30 min with H₂DCFDA. Mean DCF fluorescence, indicative of relative cellular ROS level, was measured by flow cytometry. (C) Cells preincubated as described in A were incubated for 24 h with or without 500 μM palmitate. Cell death was determined by propidium iodide staining and flow cytometry. (D) Wild-type (closed symbols) and eEF1A-1 null mutant (open symbols) CHO cells were incubated with 500 μM palmitate. GRP78 protein levels in whole cell lysates were detected by immunoblotting. Insets are representative blots. (E) Wild-type H9c2 myoblast and stable H9c2-derived cell lines expressing control siRNA (siRNA1) or siRNA directed against eEF1A-1 (siRNA2 and siRNA3) were incubated for 48 h with either 2.5 μg/ml tunicamycin or 1 μM thapsigargin. Cell death was determined by propidium iodide staining and flow cytometry. For B-E, data expressed as mean ± SEM for at least three independent experiments, ^* p < 0.05.

Figure 6.

Disruption of eEF1A-1 prevents palmitate-induced changes in the actin cytoskeleton preceding cell death. (A) To assess total protein synthesis, wild-type CHO and mutant cells were pulsed for 30 min with [³⁵S]methionine, and TCA-precipitable proteins were collected. Radiolabel incorporation was quantified by scintillation counting and normalized to total cellular protein. (B) Initial uptake of BODIPY-labeled palmitate analogue was measured in wild-type CHO and mutant cells under conditions of either low (0.6 μM) or lipotoxic (500 μM) free FA availability. Mean fluorescence for each cell type was measured by flow cytometry. (C-F) Changes in actin cytoskeleton distribution (red) were assessed by rhodamine-phalloidin cytochemical staining and fluorescence microscopy of wild-type CHO cells under basal conditions (C) and after 5-h incubation with palmitate (D). Identical experiments were performed in eEF1A-1-deficient mutant CHO cells under basal conditions (E) and after 5-h palmitate treatment (F). Cells were counterstained with DAPI (blue). Bars, 20 μm. (G and H) Subcellular fractions containing soluble actin (G) and insoluble actin (H) were isolated from wild-type and mutant CHO cells under basal conditions and after treatment with palmitate for 5 h. Actin levels were assessed by immunoblotting and quantitated by densitometry. Insets are representative blots. Data expressed as mean ± SEM for four independent experiments, ^*p < 0.05.

Figure 7.

eEF1A-1 and markers of oxidative and ER stress are increased in lipotoxic cardiac tissue. (A) Representative immunoblotting of eEF1A-1, catalase, ER chaperone proteins (GRP78 and PDI), and actin in whole tissue homogenates from ventricles of 12-wk-old wild-type or MHC-ACS mice. Blots are representative of the four animals included in each group. Open arrow indicates nonspecific protein bands. (B) Densitometry of immunoblotting, as shown in A. Data expressed as mean of the protein to actin ratio ± SEM for four animals in each group, ^*p < 0.05.

Figure 8.

Role of eEF1A-1 in palmitate-induced cell death. Under conditions of FA overload in nonadipose tissues, the cellular capacity to store FAs as triglycerides or to use FAs for energy is overwhelmed. This FFA overload can lead to the production of ROS, which can, in turn, induce ER stress. Prolonged or severe ER stress, which may occur in the presence of excess palmitate, can lead to further ROS accumulation, potentially amplifying the apoptotic/cell death response. Palmitoyl CoA, generated by esterification of palmitate as it enters the cell, may also induce ER stress directly, leading to the production and accumulation of ROS and subsequent apoptosis/cell death. In either of these scenarios, eEF1A-1 mediates actin cytoskeleton changes involved in the progression of the lipotoxic cell death response downstream of induction of oxidative and ER stress.