|Home | About | Journals | Submit | Contact Us | Français|
Mitochondria are a common target of toxicity for drugs and other chemicals, and results in decreased aerobic metabolism and cell death. In contrast, mitochondrial biogenesis restores cell vitality and there is a need for new agents to induce biogenesis. Current cell-based models of mitochondrial biogenesis or toxicity are inadequate because cultured cell lines are highly glycolytic with minimal aerobic metabolism and altered mitochondrial physiology. In addition, there are no high-throughput, real-time assays that assess mitochondrial function. We adapted primary cultures of renal proximal tubular cells (RPTC) that exhibit in vivo levels of aerobic metabolism, are not glycolytic, and retain higher levels of differentiated functions and used the Seahorse Biosciences analyzer to measure mitochondrial function in real time in multi-well plates. Using uncoupled respiration as a marker of electron transport chain (ETC) integrity, the nephrotoxicants cisplatin, HgCl2 and gentamicin exhibited mitochondrial toxicity prior to decreases in basal respiration and cell death. Conversely, using FCCP-uncoupled respiration as a marker of maximal ETC activity, 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (DOI), SRT1720, resveratrol, daidzein, and metformin produced mitochondrial biogenesis in RPTC. The merger of the RPTC model and multi-well respirometry results in a single high throughput assay to measure mitochondrial biogenesis and toxicity, and nephrotoxic potential.
Many drugs and industrial and environmental chemicals are toxic because they target mitochondria . This is a particular problem in those cells/tissues that primarily rely on aerobic metabolism (e.g. kidney). The resulting mitochondrial damage leads to decreased aerobic metabolism and ATP, disrupted cellular functions, and cell injury and death. Thus, mitochondrial function is a superb proxy for cell and organ vitality.
For example, the kidney is capable of transporting and accumulating many low molecular weight chemicals and possesses enzymes that metabolize xenobiotics to toxic reactive intermediates . Most nephrotoxicants damage the tubular epithelial cells and/or the glomeruli . The tubular Na+-dependent uptake processes result in high ATP demand making them highly aerobic and particularly sensitive to such insults . In light of the fact that many known nephrotoxicants cause mitochondrial damage, the development of a high-throughput assay to measure the loss of mitochondrial respiratory capacity in RPTC would greatly enhance efforts to assess new chemicals and biological agents as potential nephrotoxicants. Perhaps more importantly, we propose that RPTC are an ideal model to measure mitochondrial toxicity in general because they exhibit in vivo levels of mitochondrial function and express transporters and biotransformation enzymes to insure uptake and metabolism of toxicants.
The renal tubular epithelia are somewhat unique among differentiated tissue in that they have some capacity for repair and regeneration. RPTC mitochondrial loss following oxidant injury results in a signaling cascade that leads to mitochondrial biogenesis . Furthermore, several compounds have been shown to produce mitochondrial biogenesis and that mitochondrial biogenesis following oxidant injury accelerates the return of mitochondrial and cellular functions [5, 6, 7]. In light of these and related observations, understanding the mechanisms of mitochondria biogenesis and the development of agents to induce it are needed. Because the master regulator PGC-1α is required for mitochondrial biogenesis, it has been the subject of many of these efforts. Unfortunately, assays of PGC-1α activation, or the activation of upstream effectors, do not consistently predict the subsequent biogenic process. Thus, a functional assay of mitochondrial biogenesis is needed.
Immortalized cell lines have been used extensively to study mechanisms of toxicity [2, 8]. Two severe limitations of these cells are the loss of differentiated functions and high rates of glycolysis with limited respiration. A number of years ago we modified the culture conditions of primary cultures of RPTC to provide polarized cells with a greater retention of differentiated functions; and the cells exhibited respiration and gluconeogenesis rates comparable to the rates measured in vivo [9, 10].
Methods for determining mitochondrial function by measuring cellular respiration have largely relied on Clark electrode chambers that lack the throughput needed in modern research. A multi-well plate-based assay platform, the Seahorse Biosciences extracellular flux (XF) analyzer, was recently introduced to address the need for higher throughput respirometric measurements. The XF instrument uses fluorescent optode detectors to measure oxygen consumption rates (OCR) from cells plated in custom 24-well plates. The XF instrument has shown potential to measure cell metabolism in primary cardiomyocytes and many cell lines [11, 12, 13]. Although the XF instrument has not been used to assess mitochondrial dysfunction or nephrotoxicity potential, it offers the possibility to be a moderate to high-throughput assay platform. To assess its potential, the primary culture of RPTC model was optimized for the XF-24 platform and tested with several nephrotoxicants and mitochondrial biogenesis activators.
Female New Zealand White Rabbits (1.5 - 2.0 kg) were purchased from Myrtle's Rabbitry (Thompson Station, TN). The basal medium is a 50:50 mixture of Dulbecco's modified Eagles’ essential medium and Ham's F12 nutrient mix without phenol red, supplemented with 15 mM NaHCO3 and 0.2 mM glycine and 6 mM sodium lactate. The medium is adjusted to pH 7.4 while gassing with 95% O2-5% CO2 and was diluted to 295 mosmol/kg H2O before filter sterilization. The isolation media is the basal medium supplemented with 2 mM heptanoic acid, penicillin G (100 U/mL), and 0.5 mM deferoxamine. The culture medium used for cell growth is the basal medium supplemented with human transferring (5 μg/mL), selenium (5 ng/mL), hydrocortisone (50 nM) and bovine insulin (10 nM). In some experiments cultures were grown in the presence of 17 mM glucose for comparison to previous studies.
In brief, rabbit renal tubules were isolated using the iron oxide perfusion method as described previously [9, 10]. The resulting proximal tubules were plated on 100-mm tissue culture grade plastic petri dishes or XF-24 V28 plates. Plates were constantly swirled on an orbital shaker at 80 rpm or held stationary. The optimal procedure was to culture RPTC on tissue culture dishes for 3 days, the cells were removed with trypsin, and then plated into XF-24 wells at 40,000 cells/well. The plates were shaken on an orbital shaker at 80 rpm for 4 days and then treated and assayed 24 h later.
The OCR measurements were performed using a Seahorse Bioscience XF-24 instrument (Seahorse Bioscience, North Billerica, MA). On the day before the experiment, the sensor cartridge was placed into the calibration buffer supplied by Seahorse Biosciences to hydrate overnight. On the day of the experiment, the cell media was removed from the wells and each well washed with PBS with Mg2+ and Ca2+ two times. The running media (growth media) was then placed into the wells and warmed to 37°C in an incubator. The injection ports of the sensors were filled with 100 μL of treatment or vehicle in buffer. The sensor was then placed into the XF-24 instrument and calibrated. After calibration, the calibration fluid plate was replaced with the cell plate. The measurement cycle consisted of a 2 min mix, 1 min wait, and a 2 min measurement. Four basal rate measurements were followed with injections and each injection is followed by four measurement cycles. The consumption rates were calculated from the continuous average slope of the O2 decreases using a compartmentalization model that accounts for O2 partitioning between plastic, atmosphere, and cellular uptake . For any one treatment, the rates from 3 - 4 wells were used. Rates for the wells were normalized for protein content. Average basal rates are the averages of the 3rd and 4th basal rates and average uncoupled rates were the averages of the 1st and 2nd rates after FCCP injection. All average rates were normalized to the vehicle control basal and t-tests between control and treatment were used to assess statistical significance.
RPTC were plated at 50,000 cells per well in black-walled 96-well plates with optically clear bottom surface, and were then treated with the toxicants. After 24 h, the medium were aspirated, and the cells were stained for 60 min in the dark with Hoechst 33342 (20 μM) and propidium iodide (7.5 μM) in phosphate buffered saline containing 6 mM lactate. Bright field and fluorescence images from each well were taken on the IN-Cell 1000 (GE Healthcare, Chalfont St. Giles, United Kingdom) to give the number of dead (red) cells and the total number of cells (blue) in each well as a function of toxicant concentration. The number of dead cells was subtracted from the total number to obtain the number of viable cells in each well, which is represented as a percentage of the total number (% viability). Control wells receiving no treatment vehicle were also included for comparison.
To assess protein content, the medium was first aspirated from the XF-24 wells and the cells were washed twice with phosphate buffered saline containing Ca2+/Mg2+. The cells were dissolved for a minimum of 4 h at room temperature in 500 μL of Triton buffer, which contained tris-base (100 mM), NaCl (150 mM), and Triton X-100 (0.05 %), adjusted to pH 7.5. The solutions were transferred to microcentrifuge tubes, vortexed, and ultrasonicated for 30 s. Aliquots were assayed for protein concentration using a bicinchoninic acid kit (Sigma).
RPTC cultured under standard conditions (stationary with 17 mM glucose), had basal OCR that were > 100-fold lower than for RPTC cultured under optimized conditions (shaking, with lactate and no glucose) (Fig. 1A). Inhibition of the F1F0-ATPase with oligomycin is a measure of the fraction of the OCR that is coupled to ATP production. In “healthy” cells, in tissue, and in humans, the oligomycin-induced decrease in OCR is typically about 70% of basal [14, 15]. RPTC OCR decreased by approximately 30% in standard conditions and by approximately 80% in optimized conditions following 1 μM oligomycin addition (Fig. 1A). As a further test of their differentiation and respiratory function, the RPTC were treated with ouabain, an inhibitor of the plasma membrane Na+/K+-ATPase, to determine the fraction of OCR that is coupled to maintenance of plasma membrane potential. Ouabain decreased OCR approximately 60% in RPTC cultured in optimized conditions while it had a minimal effect in RPTC cultured under standard conditions (Fig. 1B). Approximately 60% of the respiration is typically inhibited by ouabain in these cells, and in isolated tubules [16, 17].
Treatment of RPTC with FCCP uncouples the mitochondrial membrane potential to cause an increase in the OCR. For isolated mitochondria, the increase in OCR relative to the oligomycin treated rate is often referred to as the maximum oxidative phosphorylation capacity and it is an approximate measure of the Vmax for the electron transport chain [14, 17]. Preliminary experiments determined that 5 μM FCCP produced the maximal OCR in RPTC cultured under optimized conditions. The uncoupled rates were not increased with triple the lactate concentration (18 mM), added glucose (5.5 mM), or added palmitate (0.25 mM, BSA carrier). As shown in Fig. 1A, FCCP produced an approximately 10-fold increase in OCR in RPTC cultured under optimized conditions compared to oligomycin-OCR. The increase in OCR produced by FCCP in RPTC cultured under standard conditions was minimal.
To assess reproducibility, the experiment shown in Fig. 1 was repeated for multiple primary cell preparations. The basal rates were measured for 90 s followed by a 2 min mix period, a 2 min wait period, and the cycle was repeated four times before the injection of a pharmacological agent. For any given plate, the well-to-well variation over 20 wells at the 4th basal rate measurement was typically 12-15% and the standard deviation for the average of the last three basal rates for any one well was about 7%. The greatest source or variance was found to be among different RPTC preparations as would be anticipated for primary cell cultures.
To evaluate preparation-dependent variances and to compare respiration rates to historical values, RPTC in each well were lysed and the protein concentration was measured with the bicinchoninic acid assay. The mean ± standard deviation of the basal and uncoupled OCR among 15 preparations were 14.4 + 3.5 and 34.2 ± 7.2, respectively, and the protein-normalized rates are comparable to previously reported values .
To examine the ability of the XF-24 instrument to measure RPTC mitochondrial dysfunction, RPTC were cultured under optimized conditions and treated with nephrotoxicants or vehicle control (0.1% DMSO) for 24 h prior to assessment of OCR. After measuring basal rates, the cells were treated with FCCP to uncouple the mitochondria. To account for variations in OCR rates from different preparations, the basal rates for treated wells and the uncoupled rates (in nmol/min mg protein) were normalized to the basal rates of vehicle control wells. Separate experiments confirmed that 24 h pretreatment with 0.1% DMSO vehicle had no measurable effect on basal or uncoupled rates. It was found that the uncoupled rates, but not basal rates showed concentration-dependent decreases in RPTC treated with cisplatin, gentamicin, and HgCl2 (Fig. 2), nephrotoxicants known to cause mitochondrial damage [18, 19, 20]. The uncoupled were rates decreased by as much as 40% with no significant decrease in basal. The FCCP uncoupled rates can be used as a stress response to uncover disrupted electron transport chain activity by toxicants, even though basal metabolism is not impaired . As a control, RPTC were treated with mitomycin C, an antineoplastic agent that does not cause measurable tubular damage in rats . RPTC exposed to 10 μM mitomycin C did not exhibit changes in either basal or uncoupled OCR (data not shown).
The viability of the toxicant-treated RPTC were assessed via plasma membrane integrity as measured from automated imaging of cells stained with propidium iodide to measure loss of plasma membrane integrity. Staining with Hoechst 33342 was used to identify all cell nuclei and pink nuclei were counted as dead and divided by all nuclei. The proportion of live cells was ~95% for untreated and vehicle control cells, and the cells treated with the nephrotoxicants, mitomycin C, and metformin also exhibited ~95% live cells. The integrity of the treated cells was also evident from the intact morphology of the monolayer as seen in the corresponding bright field images (Fig. 3). Thus, the XF-24 assay provides a sensitive measure of nephrotoxicant damage of mitochondrial respiratory capacity that is evident before impairment of basal metabolism or cell death.
We recently demonstrated that several classes of compounds produce mitochondrial biogenesis in RPTC and that FCCP-uncoupled OCR is a marker of mitochondrial biogenesis [5, 6, 7]. To determine the ability of the XF-24 instrument to measure mitochondrial biogenesis, RPTC were cultured under optimized conditions, treated with agents known to produce mitochondrial biogenesis for 24 hr and basal and FCCP-uncoupled OCR determined. The AMP-kinase activator aminoimidazole carboxamide ribonucleotide (AICAR) has been reported to induce mitochondrial biogenesis  and increased FCCP OCR at 300 and 500 μM and basal OCR increased at 500 μM with a maximal increase of approximately 40% (Fig. 4). Metformin, a biguanide anti-diabetic drug, has also been reported to induce mitochondrial biogenesis  and it also increased basal and uncoupled rates to a similar extent as AICAR. The piperazine-thiazole SRT1720 was originally reported as a SIRT-1 activator and has been shown to induce mitochondrial biogenesis by Milne, et al.  and by our laboratory . Treatment with SRT1720 also revealed concentration-dependent increases in basal and uncoupled rates with maximums of ~50%. The 5-hydroxytryptamine (5-HT) receptor agonist 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane hydrochloride (DOI) has been extensively studied in the primary RPTC where it has been shown to induce mitochondrial biogenesis via activation of PGC-1α . In the respirometric assay, treatment with 10 and 20 μM DOI produced up to 40% increases in both basal and uncoupled OCR. Similarly, treatment of RPTC with resveratrol or the isoflavone daidzein (10 μM), characterized previously as biogenesis agents in RPTC , also increased basal and uncoupled respiration rates. These data confirm that the XF measurement of RPTC respiration is a sensitive, functional assay of mitochondrial biogenesis.
Scientists within the pharmaceutical industry are highly cognizant of the potential for adverse drug effects due to agents with mitochondrial liabilities. Indeed, about half of the drugs with FDA Black Box Warnings for cardiotoxicity or hepatotoxicity (circa 2007), 15 of the drugs have documented mitochondrial liabilities . Unfortunately, nephrotoxicity has not received the same level of attention despite the frequency of loss in renal function due to adverse drug effects and xenobiotic exposure. In light of the fact that many known nephrotoxicants target the mitochondria, the development of an HTS method to measure the loss of respiratory capacity in RPTC would greatly enhance efforts to assess nephrotoxic liability of new chemicals, environmental agents, and consumer products. The results presented here demonstrate that respirometric measurements of primary rabbit RPTC that involve a “stress test” such as uncoupling with FCCP provides a sensitive measure of mitochondrial damage.
A report from the National Research Council advocates the use of in vitro toxicological screening, along with advanced bioinformatics modeling, for potential initial risk assessment of drugs, environmental agents, and household chemicals [26, 27, 28]. We show here that known toxicants produced concentration-dependent changes in respiratory capacity that could be measured prior to onset of any measurable cell death. The mitochondrial damage is a phenotypic stress response and is likely more physiological that a single pathway readout given that these cells retain function that is metabolically coupled. Indeed, the observation that measurable damage is detected prior to death suggests that the RPTC model could effectively capture the potential for chronic toxicity.
At this time there is not a direct HTS assay for mitochondrial biogenesis. For example, HTS assays have used changes in mitochondrial gene expression as markers of mitochondrial biogenesis. This approach suffers from the lack of correlation of mRNA levels and mitochondrial biogenesis and leads to false negatives and false positives. Furthermore, there is not a single commonly accepted assay for mitochondrial biogenesis. Yet interest in mitochondrial biogenesis in general and specifically in the treatment of disease has increased dramatically. For example, resveratrol and SRT1720 produce mitochondrial biogenesis and they are being studied for the treatment of type 2 diabetes and aging . In addition, we have proposed that recovery of organ and cellular injury following an insult may be limited by the remaining mitochondrial function and ATP levels and that the stimulation mitochondrial biogenesis may promote recovery of organ and cellular function in the short and long term.
In the primary cultures of RPTC we used multiple mitochondrial endpoints such as basal and uncoupled respiration, ATP levels, and mitochondrial protein levels to demonstrate that daidzein, SRT1720, and the 5-HT agonist DOI produce mitochondrial biogenesis and that stimulation of mitochondrial biogenesis following oxidant injury accelerates the recovery of cellular functions. Using these validated compounds, we demonstrate that these compounds, and other compounds known to produce mitochondrial biogenesis (AICAR, metformin), increase FCCP-uncoupled respiration confirming that it is a marker of mitochondrial biogenesis. This approach represents the first HTS assay to measure phenotypic mitochondrial biogenesis and is an improved approach to measure functional mitochondrial biogenesis.
This study was supported by National Institutes of Health/National Institute of General Medical Sciences Grant [GM 084147]. Dr. Nathan Perron provided the IN-Cell images.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.