* Oklahoma State University, Stillwater, Oklahoma 74078;
GlaxoSmithKline, Inc., Research Triangle Park, North Carolina 27709; and
Aventis Pharmaceuticals, Raleigh, North Carolina 27612
Received December 23, 2002; accepted March 7, 2003
ABSTRACT
Uncouplers of oxidative phosphorylation have relevance to bioenergetics and obesity. The mechanisms of action of chemical uncouplers of oxidative phosphorylation on biological systems were evaluated using differential gene expression. The transcriptional response in human rhabdomyosarcoma cell line (RD), was elucidated following treatment with carbonyl cyanide p-(trifluoromethoxy) phenylhydrazone (FCCP), a classical uncoupling agent. Changes in mitochondrial membrane potential were used as the biological dosimeter. There was an increase in membrane depolarization with increasing concentrations of FCCP. The concentration at 75% uncoupling (20 µM) was chosen to study gene expression changes, using cDNA-based large-scale differential gene expression (LSDGE) platforms. At the above concentration, subtle light microscopic and clear gene expression changes were observed at 1, 2, and 10 h. Statistically significant transcriptional changes were largely associated with protein synthesis, cell cycle regulation, cytoskeletal proteins, energy metabolism, apoptosis, and inflammatory mediators. Bromodeoxyuridine (BrdU) and propidium iodide (PI) assays revealed cell cycle arrest to occur in the G1 and S phases. There was a significant initial decrease in the intracellular adenosine triphosphate (ATP) concentrations. The following seven genes were selected as potential molecular markers for chemical uncouplers: seryl-tRNA synthetase (Ser-tRS), glutamine-hydrolyzing asparagine synthetase (Glut-HAS), mitochondrial bifunctional methylenetetrahydrofolate dehydrogenase (Mit BMD), mitochondrial heat shock 10-kDa protein (Mit HSP 10), proliferating cyclic nuclear antigen (PCNA), cytoplasmic beta-actin (Act B), and growth arrest and DNA damage-inducible protein 153 (GADD153). Transcriptional changes of all seven genes were later confirmed with reverse transcription-polymerase chain reaction (RT-PCR). These results suggest that gene expression changes may provide a sensitive indicator of uncoupling in response to chemical exposure.
Key Words: FCCP; uncoupling agent; gene expression; molecular marker; cell cycle; protein synthesis.
Oxidative phosphorylation is the process by which most of the energy is produced in an animal cell. In aerobic organisms, all the catabolic pathways involved in energy metabolism converge on oxidative phosphorylation or cellular respiration. Electrons are donated by NADH (reduced nicotinamide adenine dinucleotide) and flavin adenine dinucleotide, reduced (FADH2), which are produced by oxidation of nutrients including carbohydrate, fat, and protein. These electrons flow down a redox gradient, along the electron transport chain located in the inner mitochondrial membrane, and are ultimately transferred to molecular oxygen, yielding energy for generation of ATP from ADP and inorganic phosphate (Pi). Our current understanding of synthesis of ATP in energy transducing membranes, such as those of mitochondria, is based on the chemiosmotic theory, introduced in 1961 by Peter Mitchell in which transmembrane differences of proton concentrations are central to energy transduction. (Lehninger et al., 1997).
The transfer of electrons along the respiratory chain is accompanied by outward pumping of protons (H+) across the inner mitochondrial membrane, which results in transmembrane differences in proton concentration (gradient). The proton-motive force is subsequently used to drive the synthesis of ATP, as H+ flows passively back into the matrix through proton pores formed by ATP synthase, another enzyme complex system in the inner mitochondrial membrane. Thus, ATP is synthesized by coupling two reactions, electron transport and phosphorylation. Endogenous uncouplers, called uncoupling proteins (UCPs), uncouple oxidative phosphorylation by forming channels in the inner mitochondrial membranes, allowing passive movement of H+ back into mitochondrial matrix, thus disrupting proton-motive force with dissipation of oxidative energy as heat. The function of biological uncouplers is assumed to be twofold: (1) thermogenesis, in brown fat of neonates, cold adapted rodents, and hibernators; and (2) a possible protective role in the antioxidant system (Skulachev, 1998).
Certain synthetic compounds, classified as uncoupling agents and functionally akin to endogenous uncouplers, inhibit ATP synthesis by preventing the above coupling reaction by permitting H+ to bypass ATP synthesis. In the presence of an uncoupler, respiration proceeds until virtually all of the available oxygen is reduced and rapid oxidation of substrates proceeds with little or no phosphorylation of ADP. In other words, these compounds uncouple oxidation from phosphorylation. A large number of uncouplers have been discovered. Predominant classes are either hydrophobic weak acids that act as protonophores (Terada, 1990), or divalent organic cations (Shinohara et al., 1998
), capable of transferring H+ into the matrix space across inner mitochondrial membrane and thereby disrupting the membrane potential.
Although chemical uncouplers have been known from the early 1900s, and their potential use in treating clinical obesity has been understood, they have not been successfully developed as therapeutic agents. One of the important reasons is that the full spectrum of their effect on various biochemical circuitry, besides uncoupling, is not well documented. The present study was undertaken to understand the effect of chemical uncoupling on metabolic pathways and to identify potential molecular markers for uncoupling, by analyzing gene expression profiles in the RD cell line using minimally toxic levels of carbonyl cyanide p-(trifluoromethoxy) phenylhydrazone (FCCP), a classical weak acid uncoupler (Starkov, 1997).
MATERIALS AND METHODS
Chemicals.
FCCP, a classical uncoupler of oxidative phosphorylation, was selected for these studies because (1) it has been extensively employed to induce uncoupling of mitochondrial oxidative phosphorylation in many cell types, including cardiac myocytes (Hool and Arthur, 2002); (2) the mode of action of FCCP in the inner mitochondrial membrane is known (Terada, 1990
); and (3) FCCP has been used as a model compound for the development of high-throughput approaches to the detection of effects on mitochondrial membrane potential in other in vitro systems (Wong and Cortopassi, 2002
). FCCP was procured from Sigma Chemical Company (St. Louis, MO) and was at least 98% pure by thin layer chromatography. Ten mg of the compound was initially dissolved in 786.8 µl of 95% ethyl alcohol, due to its poor solubility in tissue culture media, and subsequent dilutions were made in media. Other chemicals and reagents were obtained from Clontech Laboratories (Palo Alto, CA), Molecular Probes (Eugine, OR), Calbiochem-Novabiochem Corporation (San Diego, CA), Invitrogen Life Technologies (Carlsbad, CA), CN Biosciences (San Diego, CA), and Perkin Elmer Life Sciences (Boston, MA).
Cell culture.
Skeletal muscle represents a particularly attractive target for directed uncoupling, due to the large muscle mass, which in humans, account for approximately 1520% of standard metabolic rate (SMR) (Rolfe et al., 1999). The maximal aerobic capacity of a human is generally estimated to be up to 12 times SMR. Most of this increase can be directly attributed to skeletal muscle respiratory activity, and it is clear that muscle can greatly increase its metabolic activity. Doubling metabolic rate by modestly uncoupling skeletal muscle should produce few adverse side effects, as this increase would only be comparable to mild exercise (Blaxter, 1989
). Indeed, support for this view has been obtained using proteins (UCPs 1 and 3) that may naturally uncouple mitochondria. High expression of human UCP3 in transgenic mouse skeletal muscle led to decreased weight gain despite increased food intake (Clapham et al., 2000
). Thus, skeletal muscle being the potential target tissue for anti-obesity drug development, based on availability at the time, the rhabdomyosarcoma (RD) cell line was chosen for this study. RD cells (ATCC, CCL-136), obtained from American Type Culture Collection (ATCC; Rockville, MD), were grown in Dulbecos Modified Eagles Medium (DMEM) with Glutamax and 10% fetal bovine serum (FBS), without antibiotics and incubated at 37°C with 5% CO2. They were cultured in collagen-coated "T175 Biocoat vented flasks" (Collagen I Cellware, Becton Dickinson Labware) and fed every 2 days for subculture and stock expansion. For RNA preparations, the cells were grown in Biocoat 150 mm diameter petri dishes (Collagen I Cellware, Becton Dickinson Labware) with 25 ml medium and containing a numbered collagen coated glass microscope cover slip for light microscopic evaluation of cell morphology. For RNA extraction, cells were seeded at a density of 1 x 107 in three control and three treated dishes. They were fed at 24 and 38 h, and were dosed at 48 h after seeding. For RNA extraction, cells were harvested at multiple time points, from 0 to 24 h post treatment.
Mitochondrial membrane potential assay.
Changes in mitochondrial membrane potential (m) of intact cells were measured with a J-aggregate-forming lipophilic, cationic probe, 5,5', 6,6'-tetrachloro-1, 1', 3,3'-tetraethylbenzimidazolylcarbocyanine iodide (JC-1), procured from Molecular Probes, using a FACScan flow cytometer (Becton-Dickinson, San Jose, CA), per the method described by Cossarizza et al. (1993)
. JC-1 is a carbocyanine with a delocalized positive charge and is capable of selectively entering mitochondria. It exists as a green-fluorescent monomer at lower concentrations (lower membrane potential), emitting at 530 nm upon exciting at 490 nm; and as red-fluorescent "J-aggregates" emitting at 590 nm at higher (above 0.1 µM) concentration (higher membrane potential) (Salvioli et al., 1997
; Smiley et. al., 1991
). The cells were adjusted to 1 x 106 cells/ml and suspended in DMEM. They were then dosed with 100 nM to 64 µM FCCP and incubated at 37°C for 1 h. JC-1 was added at 10 µg/ml to the cell suspension and incubated in the dark for 20 min; at the end of 20 min, they were put on ice to cool for 15 min and then washed twice and resuspended in PBS containing appropriate concentrations of FCCP. JC-1 fluorescence emission data was acquired from the flow cytometer. Differences between control and treated groups for each time point were expressed as mean (± SD), and significance was assessed using a two-tailed t-test at p < 0.05 or p < 0.01.
Experimental design.
For gene expression studies, 57 tissue culture dishes, with one cover slip each, were seeded with 1 x 107 cells. There were ten time points, with three control and three treated dishes per time point, except 0 h, which had only control dishes. A second set of 30 dishes, three per time point, was used as controls for compound concentration assay. A third set of 54 dishes, six (three control and three treated) per time point, was used for the ATP assay. The media was changed at 24 h post seeding and again 10 h prior to dosing. The treatment group was dosed with the compound at 48 h after seeding. The cultures were incubated under standard conditions, and cells harvested at the following time points for gene expression analysis: 0, 0.5, 1, 2, 4, 6, 8, 10, 12, and 24 h after dosing. For each of the applicable time points, samples were also collected for media chemistry, compound concentration, and ATP analysis.
Compound concentration assay.
Aliquots of the media were collected at the various time points from both treated dishes and a set of control dishes that contained only media with FCCP and no cells. The samples were immediately frozen and subsequently analyzed using HPLC to evaluate compound degradation over time. Differences between control and treated groups for each time point were expressed as mean (± SD), and significance was assessed using a two-tailed t-test at p < 0.05 or p < 0.01.
Media chemistry.
Media samples at each time point, for both control and treated, were centrifuged at 3200 x g for 15 min at 4°C to remove floating cells, and the supernatant stored at -20°C prior to analysis for glucose, lactate, and lactate dehydrogenase (LD) using a Hitachi 911 analyzer (Roche Diagnostics). Glucose was assayed by a hexokinase method, LD was assayed at 37°C by a modified kinetic procedure, and lactate was assayed at 37°C by a UV method catalyzed by lactate oxidase and peroxidase. Differences between control and treated group for each time point were expressed as mean (± SD), and significance was assessed using a two-tailed t-test at p < 0.01 or p < 0.05.
Morphology.
Cover slips were removed from the dishes at each time point and fixed in 95% ethanol. They were then stained with hematoxylin and eosin (H&E) and evaluated by light microscopy. Five representative 20 x fields from each of the three controls and three treated cover slips were captured with a digital camera mounted on a microscope and imported into Imagepro plusTM software. The images were then enlarged as necessary, and the cells from each 20 x field were counted to obtain separate values for cell counts and mitotic figures.
ATP assay.
Changes in intracellular ATP concentrations were measured for both control and treated cells at each time point, using Luciferin-Luciferase as per Karamohamed et al. (1999) and Ford et al. (1996)
. Bioluminescent somatic cell assay kits were purchased from Sigma Chemical Company. The cells were suspended in their original media and adjusted to 5 x 105 cells/ml. The cells were treated with ATP releasing agent, which allowed rapid release of small molecules such as ATP, while larger molecules remained trapped inside the cells, affording protection from the ATPases. The free ATP was allowed to react with Luciferin-Luciferase, and the light released was measured using a Turner Designs 20/20 luminometer. The luminescences associated with serial dilutions of known ATP standards were separately measured to generate a standard curve. A regression line fitted to the standard curve was used to derive the ATP concentrations. Differences between control and treated groups for each time point were expressed as mean (± SD), and statistical significance was assessed using a two-tailed t-test at p < 0.01 or p < 0.05.
Gene expression assays.
CCL-136 tissue culture cells, after removal of media, were lysed in situ, with Trizol ReagentTM (Gibco BRL, MD) and the lysate was stored at -80°C until use. Total RNA was isolated by chloroform/isopropanol/ethanol extraction, and RNA quality and quantity were assessed using agarose gel electrophoresis and spectrophotometric 260/280 nm absorbency. For the initial exposure time versus gene expression analysis, 33P-labeled cDNA probes were prepared with one pooled RNA sample each, from three controls and three treated dishes, and hybridized to Clontech AtlasTM human 1.2-1 cDNA microarray. From the array comparison report generated, log2 ratios of adjusted intensity were created, and 165 genes relating to energy metabolism were graphed for all the time points, using Microsoft ExcelTM to identify time points with the most transcriptional changes. Accordingly 1, 2, and 10 h time points were selected for further investigation.
For the detailed investigation, 33P-labeled cDNA probes were prepared with total RNA from each of the triplicate samples of both control and treated groups for the three time points selected, using a modified Clontech protocol, and hybridized to Clontech AtlasTM human 1.2-1, 1.2-3, and 1.2-toxicology microarrays with 1176 genes per array, with an estimated 510% overlap of genes among various arrays. Clontech AtlasTM arrays are nylon-membrane-based cDNA arrays used for broadscale differential gene expression profiling. The broad-coverage arrays profile many crucial cellular pathways and functions in a specific species. The human 1.2-1 array contains genes associated largely with cell cycle, oncogenes, ion channels, growth factors, signal transduction, DNA metabolism, cell to cell adhesion, cytokines, stress signals, and apoptosis. The 1.2-3 array contains genes associated with RNA metabolism, protein metabolism, carbohydrate metabolism, lipid metabolism, energy metabolism, cell surface antigens, transcriptional activation, cytokines, oncogenes, extracellular matrix, membrane channels, and signal transduction. The 1.2-toxicology array contains genes associated with xenobiotic metabolism, drug resistance, stress response, apoptosis, cell cycle, cell surface antigens, transcriptional activation, oncogenes, cytokines, signal transduction, cytoskeleton, energy metabolism, and DNA metabolism.
Denaturation and annealing (4 µl) was carried out at 94°C for 10 sec and 70°C for 10 min using 6 µg total RNA and 1 µl CDS Atlas specific primers (0.2 µM each). The extension reaction (22 µl, 35 min at 49°C) contained 0.5 mM each dATP, dGTP, dTTP; 50 mM Tris-HCl pH 8.3; 75 mM KCl; 3 mM MgCl2; 4.5 mM DTT; 100 µCi 33P--dCTP (3000 Ci/mmol, 10 µCi/µl, NEN); and 200 units Super Script IITM reverse transcriptase (Gibco-BRL, MD). Extension was terminated by heating to 94°C for 5 min. Unincorporated 33P-
-dCTP was removed using G50 Microspin columns (Amersham Biosciences, Inc., Piscataway, NJ). Prehybridization was carried out at 64°C for 1 h, in 6.5 ml MicroHybeTM, 3.25 µl poly-dA (1 µg/µl, ResGen), and 6.5 µl Human Cot 1 DNA (1 µg/µl, Clontech). Heat-denatured 33P-cDNA was added and hybridization carried out for 16 h. Arrays were washed at 64° following manufacturers instructions. The membranes were exposed to a phosphor imager screen for 60 h, and optical density was acquired using OptiquantTM and a Cyclone scanner (Packard Biosciences Co., Meridian, CT). Image files generated from phosphor imager scans were analyzed using Clontech Atlas SoftwareTM. After background subtraction, data were normalized and statistically analyzed using Normalization and Local Regression (NLR) software. NLR-processed text files were used to compare control with treated groups, generate p values, mean log intensity (MLI), and ratio of differences between groups. The NLR output files were again subjected to data manipulation with Microsoft ExcelTM to find changes through ranking by ratios, p values or MLI. The NLR software uses local regression analysis to estimate the normalized expression differences between two groups of arrays, such as control versus treated. The underlying mathematics of NLR, which is described in detail by Kepler et al. (2002)
, is based upon the assumption that the "large majority" of gene expression levels do not change in response to treatment. This program takes text files of transcript signal intensity data, generated in this study by the Clontech AtlasImage software, and outputs an intergroup fold change for each transcript, a p value for this change, and the MLI for each transcript. The MLI provides an indication of mean signal intensity for the gene of interest compared to the average signal of all genes in the data set, in the form of the log of the ratio of the mean intensity for the gene of interest to the mean of the intensity of the entire data set. Thus an MLI of "0" represents a ratio of "1" with the gene of interest expressed at the average level of the entire gene list examined. The MLI of more or less highly expressed transcripts will lie to the right or left of zero for the sample set, respectively. The MLI calculation is applied to all samples, including controls and treated, and thus, it only provides a crude indication of expression level for each set of genes. It is strongly recommended during more detailed analysis that one plot the normalized signal intensity for individual genes of interest, or use the group-by-group variance tables, which are also output by NLR. We routinely use Microsoft Excel spreadsheets for the examination of individual sample data as an adjunct to interpretation of statistical outputs. It is strongly recommended that the implicit assumptions and other characteristics of NLR be understood when applying this method.
Real-time quantitative reverse transcription-polymerase chain reaction (RT-PCR) (TaqManTM).
Seven genes were selected from the 10 h array data to serve as molecular markers for 80% chemical uncoupling and were analyzed by real time quantitative RT-PCR (TaqManTM) for confirmation. Total RNA was DNAase treated (Ambion DNAase I) according to the manufacturers protocol. RNA was quantified using Molecular Probes RibogreenTM assay with a Cytofluor 2350 fluorometer. Samples were diluted to 10 ng/µl prior to TaqManTM (Perkin Elmer ABI Prism 7700 Sequence Detection system) analysis. Primers were designed with the use of Perkin Elmer Primer ExpressTM software. A 96-well assay was arranged to detect mRNA expression of seven genes in three control and three treated RNA samples, using probes and primers from Keystone Biosource. The arrangement of the plate included one row for each gene and one row for 18S ribosomal RNA (rRNA) internal control. RNA samples were then arranged column-wise, with three pairs of control and three pairs of treated samples, in duplicate to fill the 12 columns of the plate. Forward and reverse primers were diluted to 9 µM and probe to 2 µM, and 20 µl of each was aliquoted to make the probe/primer master mix. The master mixes for the remaining components were prepared according to the manufacturers protocol (without probes and primers) and 35 µl was aliquoted per well. Fifteen ml of probe/primer mix was then added using a multichannel pipette, the plate was sealed with optical grade caps, and the reaction was carried out as follows: 48°C for 30 min (reverse transcriptase or RT step), 95°C for 10 min (Amplitaq activation and RT denaturation), 40 cycles of 94°C for 15 sec, and 60°C for 1 min. Values of fold change in expression were graphed for comparison purpose. Statistical significance was determined for control versus treated groups using a two-tailed t-test at p < 0.05 or p < 0.01.
Cell cycle assay.
The stages of cell cycle arrest and the proportion of dead cells at 2, 10, 24, 36, and 48 h post treatment with 20 µM FCCP were determined by flow cytometric assay with bromodeoxyuridine (BrdU) and PI. BrdU cell proliferation assay kit was purchased from CN Biosciences, and PI was obtained from Calbiochem-Novabiochem Corporation. Thirty minutes prior to each of the above time points, cells were treated with BrdU at a ratio of 1 BrdU : 100 media (to obtain a final concentration of 10 µM), incubated for 30 minutes, then trypsinized and resuspended in PBS, and fixed in 47% methyl alcohol (1 ml cell suspension in PBS + 2 ml 70% methyl alcohol). Following storage at -20°C, the cells were prepared for flow cytometric analysis. The cellular DNA was partially unwound by incubating cells in 2 N hydrochloric acid : 0.05% Triton ZX-100 for 30 min at room temperature. The acid was neutralized by washing the cells with 0.1 M sodium borate. The cells were then incubated with 100 µl anti-BrdU-FITC (diluted at 1:5 in 0.5% Tween 20, 1% BSA in PBS) per sample. PI, a fluorescent dye that intercalates into DNA, was used to measure total cellular DNA. The samples were incubated in PI at 5 µg/ml in PBS. They were then analyzed on a Becton Dickinsons FACS flow cytometer, with laser excitation at 488 nm, as per Dolbeare et al. (1983). Simultaneous measurement of the amount of BrdU incorporated into DNA and cellular DNA content were taken at various time points, up to 48 h. Cells with DNA content of 2N were considered to be in G1/G0 phase, 4N in G2/M phase, 2 to 4N in S phase and <2N was considered to be dead cells. Statistical significance of the difference between control and treated groups were assessed using a two-tailed t-test at p < 0.05 or p < 0.01.
RESULTS
Characterization of RD Cells
This human embryonal rhabdomyosarcoma cell line was established in 1968 from a malignant embryonal rhabdomyosarcoma of the pelvis of a seven-year-old girl (McAllister et al., 1969). The line consists of cells of two cytological features poorly differentiated spindle cells and larger multinucleated cells. No contractile myofilaments could be demonstrated by light or electron microscopy, and immunohistochemical staining failed to reveal the presence of significant amounts of either myosin or myoglobin (data not shown). The cells grew as a monolayer and showed occasional "whorling" patterns when grown on collagen coated plastic tissue culture dishes.
Mitochondrial Membrane Potential
There was an increase in the depolarized cell population with increasing concentrations of FCCP. Approximately 11% of cells were normally depolarized. The response to treatment was linear up to 2 µM and plateaued at around 32 µM. At 20 µM FCCP, there was approximately 75% depolarization (Fig. 1).
|
|
|
|
|
Markers for FCCP Uncoupling (Quantitative RT-PCR [TaqManTM])
Seven genes were selected from the 10 h array data to serve as molecular markers for uncoupling with FCCP. These genes were chosen to represent pathways that were transcriptionally most active and showing high levels of statistical significance. The selected genes were Ser-tRS, Glut-HAS, Mit-BMD, Mit-HSP10, PCNA, Act B, and GADD153. The fold expressions for all seven genes were statistically significant (p < 0.01; p < 0.05 ), and the changes in directions and magnitudes were in good agreement with the microarrays results (Fig. 5).
|
|
|
|
|
|
The use of microarrays for comprehensively evaluating changes in RNA expression is a promising new methodology. Technical developments that offer increased sensitivity, the prospect that large numbers of genes for a given organism could be scrutinized in this way, and a general appreciation of the need to integrate information obtained from more traditional and reductionist approaches to biology make microarray-based expression analysis a powerful tool (Bowtell, 1999).
The overall sequence of molecular events observed in this study were reminiscent of ischemic (hypoxic) cell injury and are consistent with the expected effect of chemical uncouplers on cell systems, as uncouplers essentially reverse the physiological role of oxygen, creating a "hypoxic" environment.
In living organisms, the first point of hypoxia is the cells aerobic respiration, that is, mitochondrial oxidative phosphorylation. There was progressive loss of oxidative phosphorylation with decreasing oxygen tension, resulting in widespread effects on many systems, including decreased activity of plasma membrane sodium pump, alteration of energy metabolism by switching from oxidative phosphorylation to glycolysis with increased lactic acid production, disassociation of ribosomes from rough endoplasmic reticulum leading to structural disruption of protein synthesis apparatus resulting in reduced protein synthesis, and morphological deterioration causing cytoskeletal disruption. If hypoxia continues, there is further deterioration and the cytoskeleton disperses, resulting in the loss of ultrastructural features like microvilli and formation of "blebs" at the cell surface. Myelin figures, derived from plasma as well as organellar membranes, may be seen within the cytoplasm (Cotran et al., 1999).
It is appreciated that the current study was with cells in culture and not the whole animal. Nevertheless, important changes in patterns of gene expression were evident. Dose response relationships and time courses are important parameters that must be considered in these complex analyses. The proposed sequence of the molecular events following FCCP treatment are diagrammatically represented in Figure 10. The central event in this study is the sharp decrease in ATP, which presumably led to DNA damage. Collapse of mitochondrial membrane potential is known to occur rapidly upon exposure to FCCP (Sagi-Eisenberg et al., 1983
). In the present study, membrane depolarization was followed by drop in ATP as early as 15 min post treatment.
One of the early events that followed drop in ATP was downregulation of protein synthesis, which is consistent with the effect of hypoxia (Cotran, 1999). However, whether this effect on protein synthesis is a direct result of hypoxia-mediated disassociation of ribosomes, negative feedback from cell cycle arrest, or metabolic shutdown of energy metabolism is not clear. The upregulation of GADD153 at 2 h, as per both microarray and RT-PCR, is suggestive of DNA damage during the early part of the time course. GADD153 is a negative regulator of the transcription factor C/EBP, which causes a G1 and S phase cell cycle arrest through binding with CDK2/4Cyclin D2/3 complex. (Ron and Hanener, 1992
; Wang et al., 2001
). S phase arrest has been biochemically confirmed (Fig. 6
). Examination of gene expression profile for the same period reveals that genes associated with cell proliferation show both up- and downregulation. It is easy to see that cell cycle arrest and cell proliferation activities do concurrently occur in the same population of cells, illustrating the often difficult nature of interpreting gene array data sets. This also attests to the fact that transcriptional response is never clear-cut, and such studies need to be supplemented with biochemical and/or morphologic assays to be able to draw meaningful conclusions.
Eventually, DNA damage is presumed to have led to regulatory modification of several pathways important in cell cycle regulation, as suggested by morphological evaluations (Fig. 2) and the transcriptional response of genes in the following discussion. GADD45 is a DNA damage-inducible gene that binds to PCNA and cyclin dependant kinases (CDKs), leading to S phase cell cycle arrest and stimulation of DNA excision repair (Smith et al., 1994
). GADD45 may also directly interact with p21, a cell cycle inhibitor (Kearsey et al., 1995
). Upregulation of GADD45 was supported by RT-PCR (Fig. 9
) and S phase arrest was demonstrated by the cell cycle assay (Fig. 6
). Increased activity of DNA excision repair was evident from the transcriptome. GADD45g (supported by RT-PCR in direction) and GADD45 are inducible by both cytokines and DNA damage, and are important initiators of the SPAK/JNK and p38 pathways, mediated through MAPK8 (supported by RT-PCR in direction) (Takekawa et al., 1998
). Upregulation of p21/Waf1 (supported by RT-PCR) is mediated through p53, consequent to DNA damage (El-Deiry et al., 1993
), and p21 binds with CDK 4/6 (Ahmad et al., 2001
) and PCNA (Chen et al., 1996
) to effect G1 phase cell cycle arrest. In the event that the cell is unable to repair the DNA damage in spite of the p53 mediated G1 arrest, it turns on the p53-mediated apoptotic pathway. Here, this pathway is presumed to be mediated though Fas (supported by RT-PCR in direction) signaling and caspase 3 (supported by RT-PCR in direction). Fas mediated apoptosis upon FCPP treatment has been reported by earlier workers (Linsinger et al., 1999
). P16 (supported by RT-PCR in direction) is known to specifically bind CDKs 4 and 6 to mediate G1 phase arrest (Stott et al., 1998
). Replication protein A (supported by RT-PCR) is a DNA damage responsive protein and its downregulation causes S phase arrest through downregulation of DNA polymerase (Lohrer, 1996
). PCNA (supported by RT-PCR) is involved in DNA synthesis, DNA repair, and as a cofactor for DNA polymerase
(Bravo, 1986
), and its downregulation is responsible for S phase arrest and altered DNA repair. Downregulation of the transcription factor E2F1 (supported by RT-PCR) through undetermined pathways is presumed to have contributed to S-phase arrest via DNA polymerase
(Dynlacht et al., 1994
). Downregulation of ß-catenin and N-myc (both supported by RT-PCR), also through undetermined pathways, may have contributed to G1 phase arrest through interaction with cyclins D, E, and F. The transcriptional changes found in this study are in agreement with reports in literature suggesting that the postulated DNA damage is mediated via p21, p16, PCNA, and replication protein A (Lohrer, 1996
).
A comprehensive gene expression study with FCCP, such as this, has not been attempted before, and hence extensive gene expression data is not available in literature. It is of interest to note that rotentone, which blocks electron flow through mitochondrial complex I has been found to induce dose-dependant cell cycle arrest at G2/M phase and apoptosis (Armstrong et al., 2001). Disassembly of cytoskeletal proteins subsequent to FCCP treatment, a notable event in this study, has been observed previously by other workers (Bergstrom-Porter and Shelton, 1979
; Maro et al., 1982
).
An interesting observation is the diversion from oxidative to glycolytic metabolism as oxidative phosphorylation becomes less efficient due to the effect of uncoupling. This trend is transcriptionally evident by the upregulation of glycolysis with simultaneous downregulation of pyruvate dehydrogenase, channeling pyruvate into anaerobic metabolism at 10 h. This is supported by clinical chemistry results showing increasing concentrations of media lactate in treated group, with each time point (Fig. 7). The ATP assay result (Fig. 8
) in this study was particularly unexpected. By 6 h post treatment, ATP levels in the treated cells increased to a higher level than the control. This suggests that, after the initial decline in ATP consequent to treatment, the cells responded by upregulating ATP synthetic pathways and, at the same time, downregulating pathways that expend energy. It may also be recalled that at least 60% of the compound was taken up (and presumably metabolized) in the first 4 h, and hence the reduced availability of the compound in the later time points could have potentially allowed the recovery of the cells. Also, with a population of cells that were undergoing cell cycle arrest and limited cell death, there was further reduction in utilization of ATP, contributing to an overall excess. Creatine phosphate is a major energy reserve in muscle cells and provides readily available high-energy phosphate, which can be used to generate ATP from ADP, the reaction being catalyzed by creatine kinase (Murray et al., 1996
). Creatine phosphate may also have contributed to the increase in ATP levels.
All seven genes proposed as biological markers for FCCP-mediated uncoupling of oxidative phosphorylation gave good TaqManTM confirmation, both in direction and fold changes (Fig. 5), and it was concluded that gene expression changes may be used as a sensitive indicator of uncoupling due to chemical exposure. Since DGE technology is still in the early stages and not yet a widely used technology, every effort has been made in this study to interpret cautiously after rigorous statistical analyses. Further, biochemical assays, quantitative RT-PCR, and morphological evaluations were performed to confirm transcriptional data and to test biological end points.
ACKNOWLEDGMENTS
The authors wish to thank Dr. Ron Tyler, Dr. Ruth Lightfoot, and the Division of Safety Assessment, GlaxoSmithKline, for support and funding for this work. The authors also wish to thank many others who have made this work possible, especially Dirk Springer, Tony Tong, Jackie Lee, Betty Gaskil, Heidi Colton, Warren Casey, Judy Honeycutt, and Byron Butterworth for their timely contributions and support.
NOTES
1 To whom correspondence should be addressed at GlaxoSmithKline, 5 Moore Drive, Research Triangle Park, NC 27709. Fax: (919) 483-0692. E-mail: Sabu.k.kuruvilla{at}gsk.com.
REFERENCES
Ahmad, N., Adhami, V. M., Afaq, F., Feyes, D. K., and Mukhtar, H. (2001). Resveratrol causes WAF-1/p21-mediated G(1)-phase arrest of cell cycle and induction of apoptosis in human epidermoid carcinoma A431 cells. Clin. Cancer Res. 7, 14661473.
Armstrong, J. S., Hornung, B., Lecane, P. Jones, D. P., and Knox, S. J. (2001). Rotenone-induced G2/M cell cycle arrest and apoptosis in a human B lymphoma cell line PW. Biochem. Biophys. Res. Commun. 289(5), 973978.[CrossRef][ISI][Medline]
Bergstrom-Porter, B., and Shelton, E. (1979). Effect of uncouplers of oxidative phosphorylation on microtuble location and surface structure in murine mast cells. Anat. Rec. 195, 375386.[ISI][Medline]
Blaxter, K. (1989). Energy Metabolism in Animals and Man. Cambridge University Press, Cambridge, U.K.
Bowtell, D. D. L. (1999). Options available-from start to finish-for obtaining expression data by microarray. Nat. Genet. Supplement 21, 2532.
Bravo, R. (1986). Synthesis of nuclear protein cyclin (PCNA) and its relationship with DNA replication. Exp. Cell Res. 163, 287293.[ISI][Medline]
Chen, J., Peters, R., Saha, P., Lee, P., Theodoras, A., Pagano, M., Wagner, G., and Duta, A. (1996). A 39amino acid fragment of the cell cycle regulator p21 is sufficient to bind PCNA and partially inhibit DNA replication in vivo. Nucleic Acids Res. 24, 17271733.
Clapham, J. C., Arch, J. R., Chapman, H., Haynes, A., Lister, C., Moore, G. B., Piercy, V., Carter, S. A., Lehner, I., Smith, S. A., et al. (2000). Mice overexpressing human uncoupling protein-3 in skeletal muscle are hyperphagic and lean. Nature 406, 415418.[CrossRef][ISI][Medline]
Cossarizza, A., Baccarani-Contri, M., Kalashnikova, G., and Franceschi, C. (1993). A new method for the cytofluorimetric analysis of mitochondrial membrane potential using the J-aggregate forming lipophilic cation 5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolcarbocyanine iodine (JC-1). Biochem. Biophys. Res. Commun. 197, 4045.[CrossRef][ISI][Medline]
Cotran, R. S., Kumar, V., and Collins, T. (1999). Robbins Pathologic Basis of Disease. W. B. Saunders Co., Philadelphia, PA.
Dolbeare, F., Gratzner, H. G., Pallavicini, M. G, and Gray, J. W. (1983). Flow cytometic measurement of total DNA content and incorporated bromodeoxyuridine. Proc. Natl. Acad. Sci. U.S.A. 80, 55735578.[Abstract]
Dynlacht, B. D., Brook, A., Dembski, M., Yensh, L., and Dyson, N. (1994). DNA binding and transactivation properties of DrosophilaE2F and DP proteins. Proc. Natl. Acad. Sci. U.S.A. 91, 63596363.[Abstract]
El-Deiry, W. S., Tokino, T., Velculescu, V. E., Levy, D. B., Parsons, R., Trent, J. M., Lin, D., Mercer, E., Kinzler, K. W., and Vogelstein, B. (1993). Waf 1, a potential mediator of p53 tumor suppression. Cell 75, 817825.[ISI][Medline]
Ford, S. R., Chenault, K. H., Bunton, L. S., Ginger, H. J., McCarty, J., Hall, M. S., Pangburn, S. J., Buck, L. M., and Leach, F. R. (1996). Use of firefly luciferase for ATP measurement for ATP measurement: other nucleotides enhance turnover. J. Biolumin. Chemilumin. 11, 149167.[CrossRef][ISI][Medline]
Hool, L. C., and Arthur, P. G. (2002). Decreasing cellular hydrogen peroxide with catalase mimics the effects of hypoxia on the sensitivity of the L-type Ca2++ channel to beta-adrenergic receptor stimulation in cardiac myocytes. Circ. Res. 91, 601609.
Karamohamed, S., Nordstrom, T., and Nyren, P. (1999). Real-time bioluminometric method for detection of nucleoside diphosphate kinase activity. Biotechniques 26, 728734.[ISI][Medline]
Kearsey, J. M., Coates, P. J., Prescott, A. R., Warbrick, E., and Hall, P. A. (1995). Oncogene 11, 16751683.[ISI][Medline]
Kepler, T. B., Crossby, L., and Morgan, K. T. (2002). Normalization and analysis of DNA microarray data by self-consistency and local regression. Genome Biol. 3, 112.
Lehninger, A. L., Nelson, D. L., and Cox, M. M. (1997). Principles of Biochemistry. Worth Publishers, NY.
Linsinger, G., Wilhelm, S., Wagner, H., and Hacker, G. (1999). Uncouplers of oxidative phosphorylation can enhance a Fas death signal. Mol. Cel. Biol. 19, 32993311.
Lohrer, H. D. (1996). Regulation of the cell cycle following DNA damage in normal and Ataxia telangiectasia cells. Experimentia 52, 31628.[ISI][Medline]
Maro, B., Marty, M. C., and Bornens, M. (1982). In vivo and in vitro effects of the mitochondrial uncoupler FCCP on microtubules. EMBO J. 1, 13471352.[ISI][Medline]
McAllister, R. M., Melnyk, J., Finkelstein, J. Z., Adams, E. C., and Gardner, M. B. (1969).Cultivation in vitro of cells derived from a human rhabdomyosarcoma. Cancer 24, 520526.[ISI][Medline]
Murray, R. K., Granner, D. K., Mayes, P. A., and Rodwell, V. W. (1996). Harpers Biochemistry. Appleton and Lange, Stamford, CT.
Rolfe, D. F. S., Newman, J. M., Buckingham, J. A., Clark, M. G., and Brand, M. D. (1999). Contribution of mitochondrial proton leak to respiration rate in working skeletal muscle and liver and to SMR. Am. J. Physiol. 276, C692C699.[ISI][Medline]
Ron, D., and Hanener, J. F. (1992). CHOP, a novel developmentally regulated nuclear protein that dimerizes with transcription factors C/EBP and LAP and functions as a dominant-negative inhibitor of gene transcription. Genes Dev. 6, 439453.[Abstract]
Sagi-Eisenberg, R., and Pechat, I. (1983). Membrane potential changes during IgE-mediated histamine release from rat basophilic leukemia cells. J. Membr. Biol.75, 97104.[ISI][Medline]
Salvioli, S., Ardizzoni, A., Franceschi, C., and Cossarizza, A. (1997). JC-1, but not DiOC6(3) or rhodamine 123, is a reliable fluorescent probe to assess D changes in untact cells: Implications for studies on mitochondrial functionality during apoptosis. FEBS Lett. 411, 7782.[CrossRef][ISI][Medline]
Shinohara, Y., Shidzuko, B., Shinichi, K., Seiichiro, K., Shuuji, I., and Hiroshi, T. (1998). Cationic uncouplers of oxidative phosphorylation are inducers of mitochondrial phosphorylation are inducers of mitochondrial permeability transition. FEBS Lett. 428, 8992.[CrossRef][ISI][Medline]
Skulachev, V. P. (1998). Uncoupling: new approaches to an old problem of bioenergetics. Biochim. Biophys. Acta. 1363, 100124.[ISI][Medline]
Smiley, S. T., Reers, M., Motola-Hartshorn, C., Lin, M., Chen, A., Smith, T. W., Steele, G., and Chen, L. B. (1991). Intracellular heterogeneity in mitochondrial membrane potentials revealed by a J-aggregate-forming lipophilic cation JC-1. Proc. Natl. Acad. Sci. U.S.A. 88, 36713675.[Abstract]
Smith, M. L., Chen, I. T., Zhan, Q., Bae, I., Chen, C. Y., Glimer, T. M., Kastan, M. B., OConnor, P. M., and Fornace, A. J. (1994). Interaction of the p53 regulated protein Gadd45 with proliferating cell nuclear antigen. (1994). Science 266, 13761380.[ISI][Medline]
Starkov, A. A. (1997). "Mild" uncoupling of mitochondria. Biosci. Rep. 17, 273279.[CrossRef][ISI][Medline]
Stott, F. J., Bates, S., James, M. C., McConnell, B. B., Starborg, M., Brookes, S., Palmero, I., Ryan, K., Hara, E., Vousden, K. H., et al. (1998). The alternative product from the human CDKN2A locus, p14(ARF), participates in a regulatory feedback loop with p53 and MDM2. EMBO J. 17, 50015014.
Takekawa, M., and Saito, H. (1998). A family of stress-inducible Gadd45-like proteins mediate activation of the stress-responsive MTK1/MEKK4 MAPKKK. Cell 95, 521530.[ISI][Medline]
Terada, H. (1990). Uncouplers of oxidative phosphorylation. Environ. Health Perspect. 87, 213218.[ISI][Medline]
Wang, H., Iakova, P., Wilde, M., Welm, A., Goode, T., Roesler, W. J., and Timchenko, N. A. (2001). Mol. Cell 8, 817828.[ISI][Medline]
Wong, A., and Cortopassi, G. A. (2002). High-throughput measurement of mitochondrial membrane potential in a neural cell line using a fluorescence plate reader. Biochem. Biophys. Res. Commun. 298, 750754.[CrossRef][ISI][Medline]
|