Screening for Mitochondrial Biogenesis and Toxicity

A sensitive and robust cell-based screening assay

The assay takes advantage of both the high throughput and precision oxygen consumption rate measurements afforded by the XF96 platform.

Mitochondrial damage compromises ATP production and consequentially cellular viability. Given the central role that mitochondria play in regulating cellular function, drugs that undermine mitochondrial function frequently elicit toxic effects. In contrast, mitochondrial biogenesis increases cell vitality and is being pursued as a strategy to combat age-related diseases. Thus, there is growing interest in identifying agents that induce biogenesis.

Currently there are no real-time screening assays that assess multiple parameters of mitochondrial function. Methods for determining mitochondrial function by measuring cellular respiration have relied largely on Clark electrode chambers that lack the throughput needed for screening. In addition, many cultured cell lines exhibit altered mitochondrial physiology that does not correlate well to in vivo mitochondrial biogenesis and toxicity.

The XF96 Extracellular Flux Analyzer addresses the need for higher throughput respirometric measurements. To assess its utility in screening, primary cultures of renal proximal tubular cells (RPTCs) from rabbit were optimized for the XF platform and tested with well characterized nephrotoxicants and mitochondrial biogenesis activators.

Beeson et al [1] adapted primary cultures of RPTC to exhibit in vivo levels of aerobic metabolism, to be non-glycolytic, and to retain high levels of differentiated functions. The XF96 Analyzer was used to measure mitochondrial respiration of the RPTC in real time. Changes in carbonylcyanide-4- (trifluoromethoxy) -phenylhydrazone (FCCP)-uncoupled respiration, a measure of electron transport chain (ETC) integrity and a sensitive measure of mitochondrial functional capacity, were used to quantify mitochondrial toxicity and biogenesis as described by Beeson [1]. The nephrotoxicantscisplatin, HgCl2 and gentamicin induced statistically significant decreases in the FCCP-uncoupled rates prior to decreases in basal respiration and cell death. Conversely, drugs known to induce mitochondrial biogenesis such as 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (DOI), SRT1720, resveratrol, daidzein, and metformin produced statistically significant increases in FCCP-uncoupled respiration rates [1]. Decreases or increases in FCCP-uncoupled respiration rates can function as measures of mitochondrial toxicity or biogenesis, respectively.

To validate the RPTC platform as a primary screening assay, the 1280-compound LOPAC library was screened. Cells were treated with library compounds at 5 µM using one compound/well in duplicate. After 24 h treatment, the basal and uncoupled (maximal) oxygen consumption rates [OCR] were measured with the XF96 Analyzer.

Figure 1

FCCP-uncoupled rates expressed as the ratios of treated to control values for each of the wells are shown. The different colored data clusters discriminate plates and horizontal lines indicate 1 and 3 standard deviations above and below the mean. Agents that produce ratios with standard deviations ≥ 1 are potential biogenesis inducers and those with standard deviations ≤ -1 are potential toxicants.

Figure 1 shows representative data from a single pass of the primary screening assay of the LOPAC library. End-point measurements of FCCP-uncoupled rates expressed as the ratios of treated wells to vehicle control showed that most compounds had little effect, demonstrating good selectivity for active compounds. Compounds known to promote mitochondrial biogenesis showed increases in the FCCP-uncoupled rates on the order of 20–50% above controls (>1 standard deviation from the mean.) Conversely, compounds known to elicit toxicity display decreases in the FCCP-uncoupled rates on the order of 20–50% below controls. In this screen the compounds that elicit increases or decreases above or below one standard deviation are the most likely candidates for biogenesis inducers or toxicants, respectively. Positive controls for both classes of compounds consistently lie within the same ranges. The merger of the RPTC model and the Seahorse XF96 Analyzer results in an primary screening assay to simultaneously measure mitochondrial biogenesis and toxicity including nephrotoxic potential.

Discussion

Given that impairment of mitochondrial function is implicated in metabolic disorders such as diabetes, neurodegenerative diseases, heart failure, and aging in general, this assay will enhance the development of new therapeutic agents.

The validity of this screen is due to the robust mitochondrial function of the RPTC and the sensitivity of the FCCP-uncoupled respiration rates to detect dysfunction or enhanced function of mitochondria. The assay format takes advantage of the high throughput afforded by the XF96 platform and the high precision of its oxygen consumption rate measurements [OCR]. Combining the primary RPTC model with the XF96 platform produces the first high throughput phenotypic assay of mitochondrial function that can be used as a primary screen to identify stimulators of mitochondrial biogenesis or compounds with potential toxicity. It can also be employed as a secondary or tertiary screen to identify and confirm mitochondrial biogenesis inducers or mitochondrial toxicants. Given that impairment of mitochondrial function is implicated in metabolic disorders such as diabetes, neurodegenerative diseases, heart failure, and aging in general, this assay will enhance the development of new therapeutic agents.

In another study, the XF96 Analyzer was evaluated as an assay platform to screen for drug-induced mitochondrial impairment using both cell lines and primary cultures (manuscript in preparation). In addition to confirming drugs with known adverse effects on mitochondrial function and glycolysis, the authors tested several drugs including Tolcapone and Entacapone, used in the treatment of Parkinson's disease; Nilutamide and Flutamide, anti-androgens given in the treatment of prostate cancer; and the anti-diabetic drugs Troglitazone, Ciglitazone and Pioglitazone. In these experiments, it was shown that the XF96 Analyzer had intra- and inter-assay variations of less than 15%, demonstrating the broad applicability of the XF96 across different cellular screening platforms.

Gohil et al [2] used the ratio of OCR to ECAR, a measure of glycolysis, to create an ‘aerobic quotient.' They used this assay as a secondary screen to identify drugs that shift energy metabolism from mitochondrial respiration to glycolysis [extracellular acidification rate]. This very unique and informative assay is capable of screening for compounds that shift tumor cells towards a more treatable aerobic metabolic state or ones that shift cardiomyocytes or neurons to a more glycolytic state which may be more cardioprotective or neuroprotective.

Materials and Methods

Cells: RPTC were isolated using the iron oxide perfusion method as described previously [3]. The resulting proximal tubules were plated on 100-mm tissue culture plastic petri dishes. Plates were constantly swirled on an orbital shaker at 80 rpm. After 3 days, RPTC were removed with trypsin, and plated into the wells of XF96 cell culture microplates at 18,000 cells/well. The plates were shaken on an orbital shaker at 80 rpm for 4 days, treated, and assayed 24 h later. The 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, 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 diluted to 295 mosmol/kg H2O before filter sterilization. The medium is then supplemented with human transferrin (5 µg/mL), selenium (5 ng/ mL), hydrocortisone (50 nM), and bovine insulin (10 nM).

The LOPAC library was obtained from Sigma-Aldrich. Stocks are 10 mM in DMSO.

XF Analysis

XF analyses were performed in the XF96 Extracellular Flux Analyzer (Seahorse Bioscience,) a fully integrated, 96 well format instrument that measures the uptake and excretion of metabolic end products in real time. Oxygen consumption rates [OCR] were measured using XF assay kits. Assay kits each contain a disposable sensor cartridge, embedded with 96 immobilized (solid-state) dualfluorescent biosensors (oxygen and pH.) Each sensor cartridge is also equipped with drug injection chambers for delivering testing agents into wells during an assay. OCR is expressed as pmoles/min. ECAR was not utilized in this protocol.

As illustrated in Figure 2, duplicate wells were treated with 5 µM compound in 200 µL media supplemented with 10 mM HEPES. Each plate also had vehicle control, no treatment, and SRT1720 positive control wells, in duplicate. After 24 hours the plates were analyzed on an XF96 Analyzer without exchanging the culture medium. During the experiment five basal OCR were measured with a 4 minute mix and 1 minute measurement cycle. Each well was injected with 0.5 µM final concentration of FCCP in 25 µL media followed by three OCR measurements using a 2 minute mix and 1 minute measurement cycle. The average of the three uncoupled rates for each treated well are divided by the average of the three uncoupled rates for the vehicle control wells to give the ratios illustrated in Figure 1. Quality control assessments include comparisons between basal and uncoupled rates of vehicle control versus no treatment control wells, and comparisons of the control well uncoupled rates as a ratio of the averaged basal rates; the latter ratios range from 1.5–3.0 for acceptable assays.

References

Beeson CC, Beeson GC, Schnellmann RG. A high-throughput respirometric assay for mitochondrial biogenesis and toxicity. Anal Biochem. 2010; [Epub ahead of print] PubMed PMID: 20465991.
Gohil VM et al. Nutrient-sensitized screening for drugs that shift energy metabolism from mitochondrial respiration to glycolysis. Nature Biotech. 2010;28(3):249-57.
Nowak G, Schnellmann RG. L-ascorbic acid regulates growth and metabolism of renal cells: improvements in cell culture. Am J Physiol. 1996;271(6 Pt 1):C2072-80.