1. Introduction
The cell damage caused by sustained fatty acid (FA) oversupply is referred to as lipotoxicity, and it is known to play a central role in the development of certain obesity-related metabolic disorders, such as non-alcoholic fatty liver disease (NAFLD) [
1,
2], insulin resistance [
3] and type 2 diabetes mellitus [
4,
5]. Lipotoxicity induced in pancreatic β-cells is of particular importance as it hinders the compensation of insulin resistance, which intensifies the primary damage through further increase in FA release from the adipocytes. Several in vitro studies have been performed to investigate the molecular mechanism of β-cell lipotoxicity by using various insulinoma cell lines [
6,
7,
8] and the major endogenous saturated and monounsaturated FAs, i.e., palmitate (16:0) and oleate (18:1 cis-Δ9), respectively. It has been demonstrated that the enhancement of apoptosis and autophagy that leads to a reduction in β-cell mass is largely due to a destructive endoplasmic reticulum (ER) stress [
7,
9], which, in turn, is aggravated by the accumulation of toxic ceramide intermediates [
10,
11]. Lipotoxic ER stress triggers signaling pathways that lead to pro-apoptotic transcriptional changes via X box-binding protein-1 (XBP-1) mRNA cleavage and phosphorylation of eukaryotic initiation factor 2α (eIF2α). In addition, severe malfunction of the ER leads to the activation of stress-activated protein kinase (SAPK)/c-Jun N-terminal kinase (JNK) and diverse caspase cascades, which also contribute to β-cell apoptosis [
12].
Saturated palmitate and cis-unsaturated oleate have been widely investigated and a remarkably greater toxicity of the former FA has been repeatedly demonstrated [
13,
14]. In contrast, very little is known about the cellular effects of dietary trans fatty acids (TFAs), despite their implication in type 2 diabetes mellitus [
15]. We have recently compared the effects and metabolism of the two major dietary TFAs, i.e., elaidate (18:1 trans-Δ9) of industrial origin and vaccenate (18:1 trans-Δ11) derived from natural sources, with those of palmitate and oleate in RINm5F rat insulinoma cells. It has been demonstrated that the toxicity of both TFAs were much lower than that of palmitate and similar to that of oleate, which correlated well with the marked accumulation of ceramides and diglycerides (DGs) in palmitate-treated cells and with the mild elevation in ceramide and DG levels upon the administration of cis- or trans-unsaturated FAs [
16].
Besides the obvious difference between the cell damage caused by saturated and unsaturated FAs, the potentially protective effect of oleate against palmitate toxicity has also drawn growing attention in the past few years. Simultaneously added oleate alleviated palmitate induced toxicity in various cells, including hepatocytes and hepatoma cells [
17], human mesenchymal stromal cells and osteoblasts [
18], and also decreased non-alcoholic steatohepatitis (NASH) injury in rats with high fat diets [
17]. Most importantly, it has been shown that co-administration of oleate can prevent palmitate-induced activation of the unfolded protein response (UPR) in β-cells [
19]. In light of our recent observations [
16], we find it intriguing if dietary TFAs are also protective like cis-oleate when added to the cells together with palmitate.
The primary aim of the current study was to test the potential interference of elaidate and vaccenate with the palmitate toxicity in RINm5F cells. We also wanted to investigate how oleate, elaidate and vaccenate influence palmitate-induced intracellular accumulation of ceramides and DGs. To this end, we treated the cells with palmitate at toxic concentrations and co-administered one of the unsaturated FAs, then assessed cell viability and the intensity of apoptosis. We also monitored the underlying ER stress and phosphorylation of stress-activated protein kinase (SAPK)/c-Jun N-terminal kinase (JNK). The incorporation and intracellular metabolism of FAs were investigated by determining the amount of the most relevant FA derivatives, as well as of various ceramides and DGs at different times of incubation.
3. Discussion
The FA composition of dietary lipids has been long known to influence health. The advantage of monounsaturated and polyunsaturated FAs over the saturated ones is widely accepted, and extensive research has aimed to seek experimental evidence as well as to reveal the molecular mechanism of the phenomenon. Not only were the unsaturated FAs found to be less toxic in cultured or isolated cells, but they also alleviated the damage caused by saturated FAs in certain cell types, including β-cells. The spread of the industrial processing of plant oils shed light on the biological importance of configurational isomerism at the double bonds in unsaturated FAs. Some natural cis double bonds assume the trans arrangement through spontaneous isomerization during industrial hydrogenation. These TFAs have been declared to be the worst types of FAs, and are held at least partly responsible for the health risks associated to the consumption of commercial cookies and pastries or fast-food French fries. In this case, however, the need for scientific thoroughness seems to be considerably weaker, and TFAs have been banned in certain countries on the basis of partly contradictory in vivo observations and without any convincing evidence of extreme toxicity at cellular or molecular levels. It is also peculiar that TFAs ingested with ruminant meat and dairy products, which are produced by bacterial isomerization, are exonerated from the same accusations despite their similar structure.
We aimed to compare the effects of prototypical saturated, cis- and trans-unsaturated FAs in cellular systems with special attention paid to the induction of ER stress and apoptosis. Using an established model of lipotoxicity in an insulinoma cell line [
8], we demonstrated that the industrial elaidate and natural vaccenate were equally less toxic than the saturated palmitate in our short-term in vitro experiments. The only noteworthy difference between the above-mentioned TFAs and cis-oleate was in their metabolism, i.e., although the accumulation of ceramides and DGs upon the addition of any unsaturated FA remained negligible compared to that induced by palmitate, the ceramide and DG accumulation caused by either of the two TFAs was remarkably greater than that caused by oleate [
16]. The difference was clearly due to a build-up of ceramide and DG molecules containing monounsaturated FAs, and hence we speculated that either TFAs might be more favorable substrates for ceramide and DG synthesis or the ceramides and DGs containing TFA chains might be less favorable substrates for further conversions. Since TG deposition could not be seen in the electron micrographs, nor could it be detected by light microscopy after neutral fat staining, this question remains to be elucidated. Regardless, the accumulation of ceramides and DGs has a large biological significance, especially in the long-term β-cell functionality and viability, and thus in the development of diabetes mellitus.
In the present study, we continued our research by comparing the ability of elaidate and vaccenate to attenuate palmitate toxicity with that of oleate in the same cellular model. BSA-conjugated palmitate was added to the cells at toxic concentrations (250 and 500 μM), which have been seen to severely reduce cell viability (by about 60% and 85%, respectively) at 24 h, while the obvious signs of apoptosis induction and stress, including ER stress, were investigated at 500 μM palmitate concentration and after 8 h incubation. Each of the three unsaturated FAs was administered simultaneously with palmitate and at the same concentration. In spite of a double FA dose, the cellular damage caused by a palmitate and oleate combination remained far below that induced by palmitate alone, which was in accordance with previous findings. Cells were efficiently rescued by oleate co-treatment from lipotoxic death, in particular, from lipoapoptosis, and from the development of stress, as assessed by detection of JNK phosphorylation, as well as from the activation of the UPR, i.e., XBP-1 mRNA cleavage and eIF2α phosphorylation. Ceramides and DGs are biosynthetic lipid intermediates that are implicated in the deleterious consequences of FA overload, specifically in ER stress and apoptosis, and particularly in β-cells. In line with our previous findings, oleate was scarcely incorporated in ceramides; moreover, the presence of oleate largely reduced the palmitate-induced elevation in the level of those ceramides that contain saturated (palmitoyl or stearoyl) FA chains, and thus oleate was able to minimize the overall palmitate-induced ceramide accumulation. Although oleate incorporation into DGs was much more pronounced, a similar effect was seen with regards to this group of lipid intermediates, as the amount of DG species containing saturated chains, and also the overall amount of DGs, increased remarkably less when palmitate treatment was combined with oleate addition. Altogether, the expected protective effect of oleate against palmitate toxicity was evident in our experiments, and it correlated very well with the mitigation of ceramide and DG accumulation in the cells.
The two investigated TFAs were also co-administered with palmitate, and they exerted a rather similar protection against the deleterious effects of the saturated FA. Interestingly, the level of JNK phosphorylation, indicative of a general stress intensity in the cells, was not diminished by the TFAs as effectively as it was by oleate, and the attenuation of ER stress by elaidate was remarkably weaker compared to the other two unsaturated FAs; these differences were not reflected by cell viability and apoptosis assessments. The marked reduction in cell death corresponded with the obvious alleviation of ceramide build-up in the presence of unsaturated, either cis or trans, FAs. Although a detectable amount of unsaturated ceramides appeared in the TFA co-treated cells, the presence of TFAs prevented the palmitate-induced accumulation of ceramides containing saturated FA chains, and minimized the overall palmitate-induced ceramide accumulation like oleate did. A similar phenomenon can be observed with regards to DGs in the first 8 h of the incubations; however, the total DG levels in palmitate-treated and palmitate and TFA co-treated cells converge in longer incubations, partly due to a marked growth in the species containing TFA chains. These data further support the view that ceramides play a central role in the development of β-cell lipotoxicity and lipoapoptosis. They also indicate that, at least as long as the β-cells and diabetes are concerned, TFAs are not the worst and most harmful types of FAs, and they rather fall between the most deleterious saturated and the most protective cis-unsaturated ones.
The outstanding toxicity of saturated FAs is often explained by the reluctance of the cells to process fully saturated DGs, and by the consequent ineffectiveness of channeling the saturated acyl-CoA surplus towards TG synthesis [
21]. The theory that unsaturated FAs attenuate palmitate toxicity by facilitating TG synthesis is supported by experimental data obtained in certain cell types, such as CHO cells and primary mouse embryonic fibroblasts [
22], mouse myoblasts [
23], human mesenchymal stromal cells and osteoblasts [
18]. Nevertheless, a palmitate treatment was used in HepG2 hepatocarcinoma cells [
17] and controversial data have been published as to the deleterious or protective role of TG deposition in β-cell lipotoxicity [
24,
25]. A comparison of the FA profiles suggest that this proposed mechanism might not play a central role in the protective effect of the three unsaturated FAs observed in our experiments. The facilitated diversion of palmitoyl-CoA towards TG synthesis would be expected to enhance palmitate incorporation into the cells. However, co-treatments of our insulinoma cells with unsaturated FAs decreased the amount of palmitate with about as much as the increment in the amount of these unsaturated FAs. Together with the above-mentioned lack of detectable lipid droplet (TG) deposition, this finding suggests that the reduction in ceramide and DG accumulation can be attributed, at least partly, to a lower palmitate uptake in the co-treated cells. The possible interference between the cellular uptake of different FAs and/or the saturation of FA uptake in our experimental conditions deserves further investigation.
In summary, here we demonstrate the prevention of palmitate-induced excessive ceramide and DG accumulation in RINm5F cells by elaidate and vaccenate. The observed reduction in the levels of these potentially harmful biosynthetic lipid intermediates correlated well with the protective effect of the two investigated dietary TFAs.
4. Materials and Methods
4.1. Materials Used
The culture medium and supplements were purchased from Thermo Fisher Scientific (Waltham, MA, USA). Palmitate, oleate, elaidate, vaccenate, FA free bovine serum albumin (BSA), trans-vaccenic acid methyl ester, methyl oleate methyl palmitate, methyl palmitoleate methyl stearate, 1,2-dipalmitoyl-rac-glycerol (>99%), 1-palmitoyl-2-oleoyl-sn-glycerol (>99%), 1,2-dioleoyl-sn-glycerol (>97%) and 1-octadecanoyl-2-hexandecanoyl-sn-glycerol (>99%) were from Sigma Aldrich (St. Louis, MO, USA), n-hexane was purchased from Romil (Waterbeach, UK). C16:0, C17:0, C18:0, C18:1(9Z) (>99%) ceramides were purchased from Avanti Polar Lipids Inc. (Alabaster, AL, USA).Methanol (gradient grade) and acetonitrile (gradient grade) were purchased from Merck KGaA. (Darmstadt, Germany). All other chemicals used in this study were of analytical grade. All experiments and measurements were carried out by using Milipore (Darmstadt, Germany) ultrapure water.
4.2. Cell Culture and Treatment
RINm5F rat insulinoma cells were purchased from ATCC and cultured in RPMI 1640 medium, containing 2 mM L-glutamine, 1.5 g/l sodium bicarbonate, 4.5 g/l glucose, 10 mM HEPES and 1 mM sodium pyruvate and supplemented with 10% fetal bovine serum and 1% antibiotics (Thermo Fisher Scientific; Waltham, MA, USA), at 37 °C in humidified atmosphere containing 5% CO2.
Palmitate, elaidate, oleate and vaccenate were diluted in isopropanol (Molar Chemicals, Halásztelek, Hungary) to a concentration of 50 mM, conjugated with 4.16 mM FA free BSA in 1:4 ratio, at 37 °C for 1 h. The working solution for FA treatments was always prepared freshly in fetal bovine serum (FBS)-free and antibiotic-free medium at 0.25 or 0.5 mM final concentration. The culture medium had been replaced by FBS-free and antibiotic-free medium for 1 h before the cells were treated with FA for 4–24 h at 70%–80% confluence in 6-well plates (for Western blot, RT-PCR and analysis of ceramides, DGs and FA profile) or in 96-well plates (for cell viability assay and detection of apoptosis and necrosis).
4.3. Cell Viability and Apoptosis
Cell viability was assessed by using the Colorimetric (MTT) Kit for Cell Survival and Proliferation (Merck Kft.; Darmstadt, Germany) according to the manufacturer’s instructions. MTT-derived formazan was measured at 530 nm test and 630 nm reference wavelengths in a multiscan spectrophotometer (Thermo Fisher Scientific; Waltham, MA, USA). Cell viability was expressed as the percentage of viable cells in the total cell population.
Apoptotic and necrotic cells were detected by using Annexin-V-FLUOS Staining Kit (Roche; Basel, Switzerland) and fluorescence microscopy according to the manufacturer’s instructions. Cells with green fluorescence (Annexin V labeling) were considered as apoptotic while those with red or both green and red fluorescence (propidium iodide DNA staining) were considered as necrotic. A minimum of 1000 cells was counted in each experimental condition. Apoptosis index was calculated as (number of apoptotic cells) / (number of all cells counted) × 100.
4.4. Western Blot Analysis
Cells were washed twice with PBS and harvested in 100 µL lysis buffer by scraping. The lysis buffer contained 0.1% SDS, 5 mM EDTA, 150 mM NaCl, 50 mM Tris, 1% Tween 20, 1 mM Na3VO4, 1 mM PMSF, 10 mM benzamidine, 20 mM NaF, 1 mM pNPP and protease inhibitor cocktail. The lysates were centrifuged in a benchtop centrifuge (10 min, 10,000 rpm, 4 °C). Protein concentration of the supernatant was measured using Pierce BCA Protein Kit Assay (Thermo Fisher Scientific; Waltham, MA, USA), and the samples were stored at -20 °C until use.
Samples (20 μg protein) were electrophoresed in 10%–12%–15% SDS polyacrylamide gels and transferred to PVDF membranes (Millipore; Darmstadt, Germany). Primary and secondary antibodies were applied overnight at 4 °C and for 1 h at room temperature, respectively. Equal protein loading was validated by detection of glyceraldehyde 3-phosphate dehydrogenase (GAPDH), with a mouse monoclonal anti-GAPDH (Santa Cruz; Dallas, TX, USA, sc-32233) antibody, at 1:20,000 dilution, as a constitutively expressed reference protein. Primary antibodies: rabbit anti-Cleaved Caspase-3 (#9661), rabbit anti-phospho-SAPK/JNK (THR183/Tyr185) (#9251S), rabbit anti-SAPK/JNK (#9252S), rabbit anti-phospho-eIF2α (#9721), rabbit anti-eIF2α (#9722) from Cell Signaling (Danvers, MA, USA), goat anti-GRP78 (sc-1050) and rabbit anti-PDI (sc-20132) from Santa Cruz (Dallas, TX, USA). Secondary antibodies: horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG-HRP (#7074), HRP-conjugated horse anti-mouse IgG-HRP (#7076) from Cell Signaling (Danvers, MA, USA) and donkey anti-goat IgG-HRP (sc-2020) from Santa Cruz (Dallas, TX, USA). HRP was detected with chemiluminescence using SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific; Waltham, MA, USA).
4.5. RT-PCR and Endonuclease Cleavage
Total RNA was purified from the cells by using RNeasy Plus Mini Kit (Quagen; Germantown, MD, USA) following the manufacturer’s instruction. cDNA was produced by reverse transcription of 0.5 μg DNA-free RNA samples using SuperScript III First-Strand Synthesis System for RT-PCR Kit (Invitrogen; Carlsbad, CA, USA). Spliced and unspliced XBP-1 sequences (421 or 447 bp, respectively) were amplified by PCR using iProof High-Fidelity DNA Polymerase Kit (BioRad; Hercules, CA, USA) and 5’ - GCT TGT GAT TGA GAA CCA GG - 3’ SY121041268-007 XBP-1 sense (rat) and 5’ - AGG CTT GGT GTA TAC ATG G - 3’ ST00450236-001 XBP-1 antisense (mouse, rat) primers (Sigma Aldrich (St. Louis, MO, USA) at thermocycle conditions of 98 °C 3 min followed by 30 cycles of 98 °C 10 s, 57 °C 30 s and 72 °C 15 s and finished by 72 °C 10 min final extension. PCR products were purified by PEG precipitation. DNA concentration in the PCR products was measured with NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific; Waltham, MA, USA). For a better visibility, PstI restriction endonuclease cleavage of 200 ng purified PCR product was carried out by using FastDigest PstI (Thermo Fisher Scientific; Waltham, MA, USA) for 30 min at 37 °C. The unspliced XBP-1 is cut into two fragments (153 and 294 bp) by PstI while the spliced variant remains uncut [
8]. A 234 bp fragment of the GAPDH was also amplified as a reference cDNA using a GAPDH sense primer 5’ – AGA CAG CCG CAT CTT CTT GT - 3′ and a GAPDH antisense primer 5’ - CTT GCC GTG GGT AGA GTC AT - 3’. The PCR thermocycle conditions were 98 °C 1 min, then 28 cycles of 98 °C 10 s, 65 °C 30 s and 72 °C 30 s, completed by a final extension of 72 °C 10 min. Equal amounts of DNA samples (i.e., digested XBP-1 or GAPDH) were separated by electrophoresis in 2% agarose gel and visualized by EtBr staining.
4.6. Analysis of Lipid Contents
For FA, ceramide and DG analysis, cells were washed once with PBS, then harvested in 100 µL PBS by scraping. The samples were then sedimented in a benchtop centrifuge (5 min, 1,500 rpm, 24 °C), and the supernatants were discarded. The cells were suspended in PBS, and the protein concentration of the cell suspension was measured as mentioned at Western blot analysis. A total of 50 µL of each suspension was transferred to a clear crimp vial for GC-FID measurement of FAs, and 50 µL was transferred to a micro-centrifuge tube for the HPLC-MS/MS analysis of DGs and ceramides.
4.6.1. GC-FID Analysis of Fatty Acid Profiles
150 µL of methanol containing 2 W/V% NaOH was added to the 50 µL cell suspension in the crimp vials, the samples were incubated at 90 °C for 30 min, and then cooled to room temperature. A total of 400 µL of methanol containing 13%–15% of boron trifluoride was added to the samples, and the vials were incubated at 90 °C for 30 min. After cooling to room temperature, 200 µL of saturated NaCl solution and 300 µL of n-hexane were added. FA methyl esters were extracted to the upper phase containing n-hexane, and this phase was transferred to a vial for GC analysis.
Samples of 1 µL volume were separated in a Zebron ZB-88 capillary column (60 m x 0.25 mm i.d., 0.20 µm film thickness) (Phenomenex, Torrance, CA, USA) by using a Shimadzu GC-2014 gas chromatograph equipped with a Shimadzu AOC-20s autosampler and a flame ionization detector (FID) (Shimadzu, Kyoto, Japan). The carrier gas was hydrogen at 35 cm/sec velocity. The injector and detector temperature was 250 °C, and the oven temperature was ramped from 100 °C to 210 °C at a rate of 4 °C/min.
4.6.2. HPLC-MS/MS Analysis of Diglycerides and Ceramides
The cells were pelleted by centrifugation (5 min, 1,500 rpm, 24 °C) and resuspended in methanol containing ceramide 17:0 internal standard (50 ng/ml). The samples were agitated with an ultrasonic sonotrode and centrifuged (10 min, 13,400 rpm, 24 °C). The supernatants were transferred to vials for HPLC-MS/MS analysis.
Samples (10 µL) were injected in the HPLC composed of a Perkin Elmer series 200 high-pressure gradient pump, autosampler, online degasser and a thermostat (Perkin Elmer, Milan, Italy). A Kinetex® 5 µm, C8 100 Å, LC (100 x 3 mm) column (Phenomenex, Torrance, CA, USA) was used with a gradient elution of methanol (mobile phase A) and 10 mM ammonium-acetate (mobile phase B): 0 min at 90% A; 1 min at 90% A; 9 min at 95% A; 10.5 min at 98% A; 11.5 min at 98% A; 12 min at 90% A; 14 min at 90% A. Ceramide and diglyceride species were detected using a triple quadrupole mass spectrometer (Applied Biosystems MDS SCIEX 4000 Q TRAP) (Sciex, Framingham, MA, USA). The instrument was used in positive multiple reaction monitoring mode. The ion spray temperature was set to 400 °C and the voltage to 5500 V.
4.7. Statistics
The results of Western blot analyses and DNA gel electrophoresis were carried out by densitometry using ImageQuant 5.2 software and are shown as relative band densities normalized to GAPDH as a reference protein. Data are presented in the diagrams as mean values ± S.D. and were compared by ANOVA with Tukey’s multiple comparisons post-hoc test using GraphPad (San Diego, CA, USA) Prism 6 software. Differences in p values below 0.05 were considered to be statistically significant.