|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Lipid Research, Vol. 48, 961-967, April 2007
Cell-based multiwell assays for the detection of substrate accumulation and oxidation1
* Department of Nutrition, Institute of Basic Medical Sciences, Faculty of Medicine, University of Oslo, Oslo, Norway
1 An abstract of part of this study was presented at the 2006 International Society for the Study of Fatty acids and Lipids, 7th Congress, Cairns, Australia, July 2328, 2006. Published, JLR Papers in Press, January 9, 2007.
2 To whom correspondence should be addressed. e-mail: stefan.hallen{at}astrazeneca.com
We describe multiwell assays for detecting the accumulation as well as the subsequent oxidation of 14C-labeled substrates in cultured cells. Accumulation is monitored in real time by an established scintillation proximity assay in which the scintillator is embedded in the plate base primarily detecting cell-associated radiolabel. The substrate oxidation assay is a novel variant of previously described experimental approaches aimed at trapping 14CO2 produced by isolated enzymes, organelles, or intact cells. This method uses a standard 96-well tissue culture plate and, on top, an inverted filter plate immersed with NaOH that are clamped into a sandwich sealed with a silicon gasket to obtain gas-tight compartments. 14CO2 is captured in the filter and quantified by conventional scintillation. We demonstrate both the accumulation and subsequent oxidation of 14C-labeled substrates in cultured human myotubes, adipocytes, and hepatocytes. Both methods are adaptable for compound screening; at the same time, these protocols provide easy-to-use and time- saving methods for in vitro studies of cellular fuel handling.
Supplementary key words fatty acids glucose uptake CO2 capture scintillation proximity assay Abbreviations: CPT1, carnitine palmitoyl transferase-1; DOG, deoxyglucose; DPBS, Dulbecco's phosphate-buffered saline; EPA, eicosapentaenoic acid; FCS, fetal calf serum; IBMX, 3-isobutyl-1-methylxanthine; LA, linoleic acid; OA, oleic acid; PA, palmitic acid; P/S, penicillin/streptomycin; SGBS, Simpson-Golabi Behmel syndrome; SPA, scintillation proximity assay
Measurement of cellular processes such as the uptake and oxidation of energy-rich substrates is important in cell biology. Often, phenotypes of metabolic abnormalities may transmit into in vitro systems with whole cells or enzymes isolated from the affected individual. It has been shown that primary skeletal muscle cells from type 2 diabetics as well as severely obese patients have reduced capacity to oxidize fatty acids compared with cells isolated from healthy controls (13). Moreover, a recent investigation demonstrated correlations between different metabolic parameters, such as respiratory quotient dynamics, insulin sensitivity, O2 consumption, body fat mass, plasma free fatty acids, and fatty acid oxidation measured in primary cells from the respective donor (4). These findings call for accessible multiwell assays for the measurement of substrate uptake and oxidation. Scintillation proximity assays (SPAs) designed to detect radioligand accumulation in adherent cell layers have been commercially available for more than a decade (e.g., Cytostar SPA and Scintiplate) and used for numerous applications (57), but there are few published studies of nutrient metabolism. We present results on the accumulation of radiolabeled fatty acids and deoxyglucose (DOG) in human cells grown on 96-well SPA plates monitored noninvasively in real time. Numerous protocols designed for the measurement of cellular metabolism use 14C-labeled substrates with simultaneous or subsequent capture of the 14CO2 produced (811). The substrate oxidation assay presented here is a novel variant of previously described experimental approaches and uses standard 96-well tissue culture and filter plates. The accessibility and high throughput of the described in vitro techniques should make them attractive tools for studies of cellular energy metabolism.
Materials DMEM, MEM, DMEM/Nutrient Mix F12, fetal calf serum (FCS), L-glutamine, biotin, D-pantothenate, human apo-transferrin, triiodo-L-thyronine, cortisol, 3-isobutyl-1-methylxanthine (IBMX), dexamethasone, sodium pyruvate, nonessential amino acid mix, BSA, Dulbecco's phosphate-buffered saline (DPBS; with Mg2+ and Ca2+), and HEPES were from Sigma (St. Louis, MO). Palmitic acid (PA), oleic acid (OA), linoleic acid (LA), and eicosapentaenoic acid (EPA) were purchased from Nu-Chek Prep, Inc. (Elysian, MN). UltimaGoldTM, OptiPhase Supermix, MicroScintTM O, and [1-14C]EPA (New England Nuclear, MA) were delivered by Perkin-Elmer (Shelton, CT). Ninety-six-well UniFilter® GF/B plates were from Whatman, Inc. (Clifton, NJ). Cytostar-T® was purchased from GE Healthcare (Chalfont St. Giles, UK), Ultroser G from Ciphergen (Cergy-Saint-Christophe, France), insulin (Actrapid®) from NovoNordisk (Bagsvaerd, Denmark), roziglitazone from Alexis Biochemicals Corp. (Lausen, Switzerland), HepG2 cells (human hepatoma cell line) from the American Type Culture Collection (Rockville, MD), and [1-14C]PA, [1-14C]OA, [1-14C]LA, and [1-14C]D-DOG from American Radiolabeled Chemicals (St. Louis, MO). [U-14C]PA and NaH14CO3 were obtained from Amersham Bioscience. Coomassie reagent was delivered by Bio-Rad (Copenhagen, Denmark), BC assay reagent from Uptima (Montluçon, France), Corning® CellBIND® tissue culture plate from Costar, and non-heat-inactivated FCS from GIBCO (Paisley, UK). Human Simpson-Golabi Behmel syndrome (SGBS) cells were kindly provided by Dr. M. Wabitsch at the University of Ulm, Germany. All other chemicals used were of standard commercial high-purity quality.
Cell culture SGBS adipocytes Preadipocytes were cultured on 96-well plates with growth medium consisting of DMEM/Nutrient Mix F12, 10% FCS (non-heat-inactivated), 8 mg/l biotin, 4 mg/l D-pantothenate, 2 mM L-glutamine, and P/S. Differentiation of confluent preadipocytes to adipocytes was initiated by washing the cells and adding serum-free growth medium supplemented with 10 mg/l human apo-transferrin, 0.2 nM triiodo-L-thyronine, 20 nM insulin, 100 nM cortisol, 500 µM IBMX, 25 nM dexamethasone, 1 µM rosiglitazone, 2 mM L-glutamine, and P/S for 4 days and then continued without IBMX, dexamethasone, and rosiglitazone for another 10 days. Human hepatoma cell line HepG2 cells were cultured on 96-well plates with MEM, 10% FCS, 1 mM sodium pyruvate, 100 µM nonessential amino acid mix, 2 mM L-glutamine, and P/S.
SPA
Fatty acid uptake Differentiated myotubes or SGBS cells were washed with PBS and added to their respective media containing 1 µCi/ml [1-14C]PA, [1-14C]OA, [1-14C]LA, or [1-14C]EPA at concentrations of 50, 100, or 200 µM adjusted by the corresponding unlabeled fatty acid (PA, OA, LA, or EPA, respectively) bound to BSA at a ratio of 2.5:1. Glucose uptake Differentiated myotubes were preincubated for 24 h with 100 µM PA or 40 µM BSA (control) and then washed with PBS before incubation in medium containing [1-14C] D-DOG (2 µCi/ml) with or without insulin (1 µM). For measurement of glucose or fatty acid uptake with the SPA technique, cells were added to 50 µl/well of radiolabeled medium and incubated (5% CO2 atmosphere at 37°C) for 240300 min. The increase in cell-associated radioactivity was measured in a six channel MicroBeta® Trilux scintillation counter (Perkin-Elmer) at several time points during incubation. Each well was counted for 30 s, giving a total time of counting one plate of only 89 min. Some empty wells were also added to 50 µl of radiolabeled medium to correct for the background. Finally, the cells were washed three times with PBS, harvested with 0.05 M NaOH (200 µl/well), and homogenized by ultrasonography. Protein was determined by either Coomassie reagent for myotubes or BC assay reagent for SGBS cells. A total of 100125 µl of cell homogenate per sample was added to scintillation fluid and counted by conventional liquid scintillation.
Substrate oxidation assay
Comparison with conventional substrate oxidation assay Trapping of metabolized CO2 was also monitored with a conventional method using HepG2 cells seeded in 12.5 cm2 flasks and cultured to confluence. During measurement of OA oxidation, cells were incubated with DMEM supplemented with 20 µM [1-14C]OA (1.1 µCi/ml), 1 mM L-carnitine, and 10 mM HEPES. The flasks were made airtight with rubber caps, and after 2 h, each flask was injected with 300 µl of phenylethylamine-methanol (1:1, v/v) through the rubber cap into a center well containing a folded filter paper. Subsequently, 300 µl of perchloric acid (1 M) was added to the cells, and the flasks were allowed to stand at room temperature to trap labeled CO2. The center wells with filters were transferred to separate counting tubes, scintillation fluid added and the samples counted in a ß-counter.
Presentation of data and results
Long-chain fatty acid accumulation in human myotubes and SGBS adipocytes Figure 3 shows the time- and concentration-dependent increases in cell-associated [14C]PA, [14C]OA, [14C]LA, and [14C]EPA measured noninvasively using SPA plates. Whereas for myotubes (Fig. 3AD) there were small or no differences between the accumulation of PA, OA, and LA, they clearly accumulated less EPA at all concentrations tested. The processes governing fatty acid accumulation in myotubes also seemed saturated at the highest concentrations. The SGBS cells, however (Fig. 3EH), did not show signs of saturation, and there were no apparent differences between the uptake rates of the different fatty acids. SGBS adipocytes also displayed two to four times greater accumulation of all fatty acids compared with myotubes.
Moreover, we compared the amounts of cell-associated fatty acids after 4 h of incubation using the SPA method with the conventional technique of washing, harvesting, and scintillation counting of the dissolved cells. The correlations between the two methods were 0.93, 0.96, 0.97, and 0.95 for PA, OA, LA, and EPA, respectively.
Insulin-stimulated glucose uptake in human primary skeletal muscle cells
Capacity and linear range of 96-well CO2 trapping The appliance for capturing metabolic CO2 (Fig. 2) was tested for capacity and linearity using NaH14CO3. Figure 5A shows the amount of 14CO2 trapped in the filters plotted against the total amount of NaH14CO3 present on the lower 96-well plate. The amount of 14CO2 captured in the filters showed a linear relation with the total amount of CO2 released (between 0 and 1 µmol), which covers the range of CO2 expected to be produced by any cultured cell type.
Comparison of the multiwell assay with the conventional "flask" method We compared our new method for trapping and quantification of 14CO2 with the conventional use of tissue culture flasks equipped with small cups. Similar amounts of 14CO2 were trapped with both methods after incubation of HepG2 cells with [14C]OA (Fig. 5B). The cells also grew equally well in both formats, as shown by measurement of cell protein (Fig. 5C).
Acute dose response of a carnitine palmitoyl transferase-1 inhibitor
Long-term incubation of myotubes with a peroxisome proliferator-activated receptor agonistThe peroxisome proliferator-activated receptor agonist GW501516 may induce genes involved in the transport and oxidation of fatty acids in cells (13, 14). We exposed myotubes to increasing concentrations of GW501516 for 3 days and then monitored the oxidation of [14C]OA. The oxidation was increased in a concentration-dependent manner up to 10 nM of the agonist (EC50 0.6 nM), with a maximum increase of 2-fold compared with the control (Fig. 7A
). Preincubation with GW501516 increased cell-associated [14C]OA similar to the oxidation (Fig. 7B), although the relative effect was markedly smaller.
We describe two multiwell in vitro assays for the detection of cellular uptake and oxidation of fatty acids and glucose. We have demonstrated the potential of the SPA technique for measurements of both fatty acid and glucose accumulation in muscle cells and SGBS adipocytes. The myotubes exhibit saturation, whereas the SGBS cells demonstrate a linear relationship between fatty acid concentration and cell-associated accumulation of radioactivity up to 200 µM concentration of fatty acids. Furthermore, the results obtained using the Cytostar-T® plates correlated very well with the total radioactivity monitored with conventional scintillation after solubilizing the cells. The noninvasiveness of the SPA procedure allows continuous monitoring of the cells and thereby makes it feasible to study the accumulation as well as the efflux of cell-associated radiolabeled substrates. In addition to high throughput, the SPA approach will also be valuable for studies of primary cells, of which there might be a limited amount available, and for other cell systems that are time-consuming to grow and differentiate (e.g., SGBS cells and myotubes). The only limitations with the detection of substrate accumulation by Cytostar-T® plates are the need for the cells to be adherent to the plate bottom and the fact that the substrate has to be labeled with a suitable ß-emitter (preferably 14C or 3H). An alternative approach for the measurement of substrate flux ex vivo/in vitro is the use of positron autoradiography, in which the accumulation and, if applicable, secretion of ß+-emitting tracers (e.g., 11C or 18F) can be followed noninvasively in tissue slices (15, 16). This method elegantly generates two-dimensional images of tracer distribution in the samples, but at the expense of low throughput and high cost.
Similar arguments presented as rationales for using SPA microplates to monitor nutrient accumulation in vitro also hold for our method for CO2 trapping. We demonstrate the capture and quantification of 14CO2 produced from cells grown on standard 96-well tissue culture plates. This method uses an inverted multiwell filter plate immersed with an alkaline solution (NaOH), placed with the wells aligned on top of the plate encompassing the cells of interest. This CO2 trap generates similar results compared with a conventional method of capturing CO2 and has a long linear range, covering the expected amount of CO2 produced by cells grown on 96-well plates. In addition, the described method enables the cells to be washed and harvested after incubation, thereby increasing the yield of data that can be obtained from the experiment. Our constructed CO2 trap also allows the exchange of filter plates during experiments, making time-course measurements very feasible. To validate the appliance for detecting CO2 trapping, myotubes and HepG2 cells were incubated with the irreversible CPT1 inhibitor etomoxir, which markedly inhibited fatty acid oxidation. Myotubes were also preincubated with the selective peroxisome proliferator-activated receptor Comparison between markers of disease in vivo and defects/changes in cellular metabolism in vitro has the potential to be a fruitful strategy for studies of metabolic abnormalities. The exploration of this field of research has to a large extent been slowed by time-consuming and cumbersome methods for studying nutrient handling. In this article, we have described two principles for the detection of substrate accumulation and oxidation that may help to overcome some of these obstacles.
This work was supported by grants to A.J.W. and C.A.D. from Lipgene, Integrated Project 6th Framework Programme, Food Quality & Safety (FOOD-CT-2003-505944) and the Johan Throne Holst Foundation for Nutrition Research at the University of Oslo, Norway. Manuscript received December 4, 2006 and in revised form January 8, 2007.
This article has been cited by other articles:
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Advertisement | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||