Objectives: Dyslipidemia is a risk factor for post-transplant diabetes mellitus, especially in patients who are taking tacrolimus. Although lipotoxicity of dyslipidemia leads to ?-cell failure, the handling of lipids by ? cells is a mystery in molecular endocrinology. Likewise, lipid droplet homeostasis is appreciated as a key component of lipid metabolism in cells like hepatocytes, but its role in ? cells remains to be elucidated.
Materials and Methods: To evaluate the morphologic changes in ? cells with special focus on lipid droplets, we evaluated electron micrographs under metabolic stress conditions of glucotoxicity, lipotoxicity, and glucolipotoxicity in isolated rat insulinoma INS-1E ? cells. Cells were treated with palmitic acid (0.5 mM), glucose (33 mM), or both for 16 hours, after which morphologic changes were observed with an electron microscope.
Results: Many lipid droplets were observed in the cytoplasm of healthy ? cells in the control group (no treatment). Lipid droplets were also visible in the cytosol, and the cytoplasm was rich in organelles and insulin vesicles under high glucose stimulation. However, after treatment with palmitic acid, almost no lipid droplets were observed. Endocrine vesicles were also depleted, with severe morphologic disruption of other organelles. Under glucolipotoxic conditions, ? cells showed a decreased number of lipid droplets and insulin vesicles compared with controls.
Conclusions: Lipid droplet dynamics seemed important in the homeostasis of ?-cell metabolism. In this preliminary study, healthy ? cells appeared rich in lipid droplets under normal conditions. However, lipotoxicity depleted and glucolipotoxicity decreased the number of lipid droplets in ? cells. Because dyslipidemia causing lipotoxicity is one of the most frequent metabolic problems in transplant patients and increases risk of posttransplant diabetes mellitus, understanding the mystery of lipid droplets in ? cells and the pathophysiology of diabetes in transplant patients is important, especially for those taking tacrolimus.
Key words : Glucolipotoxicity, Lipotoxicity, Posttransplant diabetes mellitus
Posttransplant diabetes mellitus (PTDM) affects up to 30% of renal transplant recipients. It reduces survival and quality of life and increases the risk of infections and cardiovascular disease in these patients.1-5 Many risk factors have been identified for PTDM. Most of the pretransplant risk factors for PTDM are common to type 2 diabetes and involve age, race, family history of diabetes, obesity, hyperlipidemia, and abnormal glucose tolerance test. Immunosuppressive drugs, especially calcineurin inhibitors and steroids, represent the most important posttransplant risk factors.6-8 With obesity becoming a global epidemic in both adults and children, its major consequence, insulin resistance, has also become a worldwide major health problem leading to diabetes mellitus. High plasma lipids and glucose intolerance because of insulin resistance gradually decrease ?-cell function.9,10 In addition to the high prevalence of insulin resistance causing an increased pretransplant risk for PTDM, exposure to calcineurin inhibitors, especially tacrolimus, in the posttransplant period aggravates ?-cell failure, especially in those with dyslipidemia of insulin resistance.11-13 In a clinical study from Porrini and colleagues, pretransplant hypertriglyceridemia, an important marker of insulin resistance, was found to be a risk factor for PTDM in transplant recipients who were taking tacrolimus.12 In an animal study, Rodriguez-Rodriguez and colleagues showed that preexisting hypertriglyceridemia was an important risk factor for the diabetogenic side effects of tacrolimus in obese and lean Zucker rats.13 The same group later examined the effect of tacrolimus in ? cells under metabolic stress of glucolipotoxicity. The investigators found that tacrolimus exacerbated the same changes in nuclear transcription factors (like MafA and FoxO1) that are already promoted by glucolipotoxicity.14,15 Although it is clear that chronic exposure of ? cells to toxic levels of lipids, either with or without glucose, leads to functional changes in ? cells and makes the cells more susceptible to other toxic insults, the physiopathological mechanisms of lipid handling in ? cells are not yet clear. In hepatocytes, lipid droplets play a central role in both storage of fats and lipoprotein metabolism during fatty liver disease. Dysregulation of lipid droplet biosynthesis and degradation can increase intracellular lipid accumulation and promote the activation of pathogenetic mechanisms leading to steatosis, hepatocellular inflammation, and fibrosis.16 In the present study, we aimed to visualize the morphologic changes in ? cells under lipotoxicity and glucolipotoxicity, with a special focus on lipid droplets, which are a vital but overlooked organelle of ? cells. To evaluate the morphologic changes in ? cells under metabolic stress, we used electron microscopy to view isolated INS-1E rat ? cells after 16-hour incubation with glucose (33 mM) and/or palmitic acid (0.5 mM).
Materials and Methods
This study was conducted in the Department of Medical Biology Cell Culture Laboratory of Baskent University School of Medicine.
INS-1E cell culture
The rat insulinoma cell line (INS-1E) was kindly provided by P. Maechler (University Medical Center, Geneva, Switzerland). We cultured INS-1E cells at 37 °C in a humidified atmosphere (5% CO2) in RPMI 1640 medium (Sigma-Aldrich, R0883), which contained 1 mM sodium pyruvate (Sigma-Aldrich, S8636), 50 ?M 2-mercaptoethanol (Sigma-Aldrich, M3148), 2 mM L-glutamine (Sigma-Aldrich, G6392), 10 mM HEPES (pH 7.3) (Sigma-Aldrich, H3784), 100 U/mL penicillin, and 100 ?g/mL streptomycin (Sigma-Aldrich, P4333). The maintenance culture was passaged once per week by gentle trypsinization, and cells were seeded at a density of 4 × 104 cell/cm2 in 25-cm2 flasks (Orange Scientific, 4420100) with 6 mL of complete medium.
Palmitic acid and glucose preparation
Saturated palmitic acid (Sigma-Aldrich, P0500) was prepared in bovine serum albumin (BSA) using a modification of the method described previously by Li and associates.17 Briefly, palmitic acid was dissolved in ethanol at 200 mM and then mixed with 10% fatty acid-free, low-endotoxin BSA, with a final concentration of 4 mM. All solutions were adjusted so that their pH levels were 7.5; all solutions were filtered and sterilized and stored at -20 °C until use. Control solutions containing ethanol and BSA were similarly prepared. The stock of 1 M glucose solution was obtained by mixing commercially purchased glucose solution with the medium. Cells were treated with 33 mM glucose alone (33 mM), palmitic acid alone (0.5 mM), or palmitic acid plus glucose (0.5 mM palmitic acid, 33 mM glucose) and incubated for 16 hours.
Cell viability assay
Cell viability was measured by the 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide (MTT) assay, which is based on the cleavage of the yellow tetrazolium salt MTT to purple formazan crystals by metabolically active cells. We added 10 ?L of MTT labeling reagent to each well and allowed wells to incubate at 37 °C for 4 hours. We then added 100 ?L of solubilization buffer to each well, with incubation overnight in a humidified atmosphere at 37 °C and 5% CO2. The optical densities were measured at 550 nm and 690 nm (reference) by Epoch Microplate Spectrophotometer (Bio-Tek).
Electron microscopy method
We centrifuged the cell suspensions at 300g and then removed the supernatant. We fixed the cell pellets with 2% glutaraldehyde for 2 hours at 4 °C and then postfixed in osmium tetroxide. Cells were washed with phosphate-buffered saline and centrifuged. The cell pellet, when separated from the supernatant, was placed into melted 4% agaroses at 60 °C. The mixture was placed on the microscope slide and allowed to solidify. Solidified agarose was cut into 1-mm3 pieces. We dehydrated the cells through graded alcohol to 100% ethanol. Cells were immersed in propylene oxide twice for 15 minutes each. We then placed cells in a 3:1 mixture of propylene oxide and resin for 1 hour. After 1 hour, we placed cells in a 1:3 mixture of propylene oxide and resin overnight at room temperature. The samples were placed in the resin plastic on the mold. We conducted polymerization in an oven at 60 °C for 48 hours. After resin blocks cooled to room temperature, we removed the blocks from the mold. The blocks were trimmed and sectioned into 1-?m sections. Sections were observed under a light microscope and trimmed for a 60-nm sectioning procedure. We placed 60-nm-thick sections on copper grids and stained sections with uranyl acetate and lead citrate. We used an electron microscope (JEOL JEM 1400) to observe the cells. To evaluate the morphologic changes in ? cells after metabolic stress, we used the electron microscope to observe isolated INS-1E rat ? cells after 16-hour incubation with glucose (33 mM) and/or palmitic acid (0.5 mM).
Cell viability at each time point
With cytotoxic doses for palmitic acid, glucose, and glucose plus palmitic acid, we observed the maximum inhibition of cell viability at 16 hours after treatment.
Observation of control cells at hour 0
Control cells were treated only with ethanol and BSA (without palmitic acid) and were examined at hour 0. The solvents (BSA and ethanol) had no ultrastructural effects on the cells. At the beginning of the study, control cells were observed by electron microscopy to have nuclei with the nucleolus. Nuclei were observed in the euchromatic structure. The double-membrane structure of the nuclear membrane was observed to be normally organized. Nuclear pores were normal. Peroxisomes were observed in the cytoplasm. Lysosomes were seen in a normal structure. Dilated granular endoplasmic reticulum cisterns were observed. Cytoplasm membrane had a normal appearance. Many lipid droplets were observed in the cytoplasm (Figure 1). Mitochondria were mostly round, with elongated mitochondria seen in some cells. Rarely, cells with increased nucleus-to-cytoplasm ratio and cytoplasmic condensation were observed. Endocrine secretory vesicles were observed (Figure 1).
Cells treated with 33 mM glucose for 16 hours
Electron microscopy examinations showed that cytoplasmic membranes were well preserved. The cells had euchromatic nuclei with prominent nucleolus with increased nucleolus-to-nucleus ratio. The nuclear double-membrane structure was evaluated as intact. Euchromatin was observed to be increased compared with the control group. We observed no difference in terms of euchromatin in the glucose group compared with the 0.5 mM palmitic acid group treated for 16 hours. The number of secretory vesicles in the cytoplasm was increased compared with the hour 0 control group and the other groups treated with palmitic acid. The nucleus-to-cytoplasm ratio was observed to be decreased. The Golgi apparatus appeared intact, and mitochondria were round. Cytoplasm content was increased in terms of organelles. Lipid droplets were observed in the cytosol. Glycogen particles were increased in the cytosol (Figure 2).
Cells treated for 16 hours with 0.5 mM palmitic acid
In cells treated with palmitic acid for 16 hours, electron microscopy results showed that cells had euchromatic nuclei with a prominent nucleolus. The nuclear pores were normal, and the nuclear double-membrane structure appeared intact. The perinuclear cistern was also intact. Peroxisomes in the cells were significantly reduced compared with the control group. No lipid droplets were found in the cytoplasm of most cells. Endocrine vesicles were significantly reduced compared with both the control and the glucose-treated groups. Mitochondria were observed to be more elongated. Heterogeneous cisterna dilatation was noted in the Golgi apparatus (Figure 3).
Cells treated for 16 hours with both 33 mM glucose and 0.5 mM palmitic acid
We observed cells to have euchromatic nuclei and prominent nucleolus. The nuclear double-membrane structure appeared to be intact. Euchromatin was observed to be increased compared with the hour 0 control group. The number of secretory vesicles in the cytoplasm was decreased compared with the control group. Nucleus-to-cytoplasm ratio appeared to be increased in favor of cytoplasm. The Golgi apparatus was not well-developed, and cisterns of the Golgi apparatus were enlarged. Mitochondria were round. Granular endoplasmic reticulum cisterns were dilated. There was also a vacuole in the cytoplasm. We observed very few or no lipid droplets in the cells (Figure 4).
Cells treated for 16 hours with 0.5 mM palmitic acid control
We observed cells to have euchromatic nuclei with the double nucleolus. Nuclear pores were normal, and the nuclear double-membrane structure was well preserved. Some cells had an increased nucleus-to-cytoplasm ratio. The cytoplasm membranes of the cells were intact. Peroxisomes with prominent electron-dense centers were seen. Mitochondria were elongated. There were no hooded mitochondria or other mitochondrial pathologies. Lipid droplets were decreased compared with the hour 0 control group.
In our present study, we examined the morphologic changes in lipid droplets in ? cells under conditions of lipotoxicity and glucolipotoxicity. The most interesting finding of this preliminary study was that lipotoxicity decreased the number of visible lipid droplets in the ?-cell cytosol. However, the number of lipid droplets was preserved or increased under high glucose concentrations. Lipid droplets are multitasking organelles, in which, in addition to their energy storage function, they fulfill different roles depending on cell type and the cell’s physiological state.18 Despite its potential importance and being one of the most prominent organelles in human ? cells, the role of lipid droplets in ? cells has not been well-defined. Lipid droplets are known to be metabolically active and essential organelles in healthy ? cells. To our knowledge, our study is the first in the literature to document the increased number of lipid droplets in ? cells after glucose stimulation and show their disappearance after lipotoxic treatment. Under high glucose conditions, the ? cells increased their number of organelles, insulin granules, and lipid droplets in the cytoplasm. Among these organelles, mitochondria were found to be increased in number, and both the endoplasmic reticulum and Golgi apparatus were prominently and homogenously visualized with an increased number of secretory vesicles in the cytoplasm. We found the cells to be hypertrophied under chronic high glucose stimulation. Glucose is evolutionary the primary stimulus for insulin release in rodents and humans because it is the principal food component in nature. Indeed, the amplitude of insulin secretion induced by glucose is much larger compared with that stimulated by protein or fat. Glucose is the major and the most important stimulus for a mature healthy ? cell; it is taken into the cell via glucose transporter 2 and is immediately phosphorylated by glucokinase to start glycolysis.19 Glycolytic intermediates interact with transcription factors like PDX-1 and MafA for nuclear translocation, and insulin transcription is stimulated; insulin is then synthesized and packaged into the insulin vesicles. The higher the concentration of glucose that enters the cell, the more ATP is synthesized by oxidative phosphorylation in mitochondria. ATP stimulates the potassium-ATP channels on cell membranes to augment insulin secretion.19 Although the role of lipid droplets in insulin synthesis and secretion is not well-known, lipid droplet formation augments the mitochondrial ATP synthesis and protects the mitochondria from overloading with metabolites for oxidative phos-phorylation. In previous studies in cancer cells and hepatocytes, lipid droplet formation was shown to increase under oxidative stress or endoplasmic reticulum stress.20,21 Lipid droplets in the vicinity of mitochondria were proposed to facilitate the transfer of fatty acids from mitochondria for beta-oxidation for energy production and also to protect the mitochondria by sequestering fatty acids and pre-venting fatty acid overload of the mitochondria.22-24 In ? cells, the glucose signal increases the number of lipid droplets, secretory vesicles, and other organelles, including the mitochondria and pero-xisomes in the cytoplasm. That is, as a physiologic stimulus, glucose, at high concentrations for prolonged periods, causes an adaptive increase in the capacity and size of the organelles, including the lipid droplets, to synthesize, process, and store the insulin secretory vesicles in adequate amounts. Lipotoxic medium, on the other hand, caused a depletion of lipid droplets in the ?-cell cytoplasm. We found that prolonged treatment of the INS-1E ? cells with palmitic acid had the opposite effect of that encountered with high glucose stimulation. Secretory granules were significantly decreased in the cytoplasm, and no lipid droplets were shown in the cytoplasm. Under glucose stimulation, ? cells physiologically metabolize lipids for augmentation of insulin secretion. Physiologically, intracellular metabolism of free fatty acids is important for the synthesis of lipid-signaling molecules such as long-chain acyl-CoA and diacylglycerol. These 2 molecules are essential for insulin granule fusion and priming. However, glucose signaling is essential for the augmentation of insulin secretion via lipids. In mature healthy ? cells, glycolysis generates pyruvate and glycerol-3-phosphate, which is important for generating lipid metabolic-coupling factors like diacylglycerol and long-chain acyl-CoA. In the absence of glucose, lipids cannot increase insulin granule mobilization, fusion, or priming.19 We found that, in the absence of glucose stimulation, cells treated under lipotoxic conditions with palmitic acid had decreased biogenesis of both the secretory vesicles and lipid droplets. Lipotoxicity is also an important risk factor for organ transplant recipients. It is well-known that insulin resistance with or without type 2 diabetes mellitus is a metabolic disease characterized by chronically elevated plasma lipid levels and progressive loss of ?-cell function leading to elevated glucose levels.25,26 Dyslipidemia of insulin resistance is an important risk factor in transplant recipients for PTDM. Chronically elevated triglycerides with free fatty acids in circulation, which is the hallmark of insulin resistance, increase the susceptibility of ? cells to the diabetogenic effects of tacrolimus.11-15 Although fatty infiltration of both endocrine and exocrine pancreas is an increasingly appreciated consequence of ectopic fat deposition during insulin resistance, handling of excess fatty acids in ? cells has not been well studied. Glucose and lipids together may exert long-term glucolipotoxic effects. In our present study, cells treated with both glucose and palmitic acid showed a decreased number of both secretory vesicles and lipid droplets. Although the lipid droplets decreased in number compared with the control cells and those treated with high glucose media, the number of lipid droplets was more than that observed in lipotoxic media. The presence of glucose may have augmented the synthesis of lipid droplets in glucolipotoxic media more than that encountered with lipotoxic media alone.
Our study showed that glucose at high con-centrations induces the synthesis and storage of insulin vesicles and lipid droplets in ? cells. Glucose is known to be a major stimulus for mature, healthy ? cells to synthesize, store, and secrete insulin. Other stimuli like fats require the presence of glucose to augment the glucose-stimulated insulin secretion. During periods of active insulin synthesis and secretion, lipid droplets are also synthesized and increased in the cytoplasm, most probably to ensure the cellular homeostasis of metabolism. However, in lipotoxicity, palmitic acid either alone or with glucose is an important toxic metabolic insult for the ? cell and it can diminish the capacity of the ? cell to synthesize and store insulin granules and lipid droplets. Because dyslipidemia has been shown to be one of the most frequent and important risk factors for PTDM, especially in those who are taking tacrolimus, our study forms the basis for future studies to understand the molecular mechanisms that underly the role of lipids in ?-cell failure with special focus on lipid droplet dynamics in achieving metabolic homeostasis.
DOI : 10.6002/ect.2022.0269
From the 1School of Medicine Department of Medical Biology and the 2School of Medicine Department of Internal Medicine, Division of Endocrinology and Metabolism, Baskent University, Ankara; the 3University of Health Sciences, Etlik Zübeyde Han?m Women’s Health Training and Research Hospital, Ankara; and the 4Istinye University, School of Medicine, Department of Histology and Embryology, Istanbul, Turkey
Acknowledgements: The authors have no declarations of potential conflicts of interest. This study was supported by a grant from the Baskent University Research Fund. We thank the laboratory staff and colleagues for helpful suggestions and technical assistance.
Corresponding author: Yaprak Yilmaz-Yalcin, Baskent University, School of Medicine Department of Medical Biology, Ankara, Turkey
Phone: +90 5333500928
Figure 1. Electron Micrographs of the Control Group at 0 Hours
Figure 2. Electron Micrographs of the 33 mM Glucose Group at 16 Hours
Figure 3. Electron Micrographs of 0.5 mM Palmitic Acid Group at 16 Hours
Figure 4. Electron Micrographs of the 33 mM Glucose and 0.5 mM Palmitic Acid Group at 16 Hours