The metabolic status of human retinal pigment epithelial cells (H-RPE) reflects their health and function. Presented here is an optimized protocol for examining the real-time metabolic flux of H-RPE using high-resolution respirometry.
Metabolic dysfunction of retinal pigment epithelial cells (RPE) is a key pathogenic driver of retinal diseases such as age-related macular degeneration (AMD) and proliferative vitreoretinopathy (PVR). Since RPE are highly metabolically-active cells, alterations in their metabolic status reflect changes in their health and function. High-resolution respirometry allows for real-time kinetic analysis of the two major bioenergetic pathways, glycolysis and mitochondrial oxidative phosphorylation (OXPHOS), through quantification of the extracellular acidification rate (ECAR) and oxygen consumption rate (OCR), respectively. The following is an optimized protocol for conducting high-resolution respirometry on primary human retinal pigment epithelial cells (H-RPE). This protocol provides a detailed description of the steps involved in producing bioenergetic profiles of RPE to define their basal and maximal OXPHOS and glycolytic capacities. Exposing H-RPE to different drug injections targeting the mitochondrial and glycolytic machinery results in defined bioenergetic profiles, from which key metabolic parameters can be calculated. This protocol highlights the enhanced response of BAM15 as an uncoupling agent compared to carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) to induce the maximal respiration capacity in RPE. This protocol can be utilized to study the bioenergetic status of RPE under different disease conditions and test the efficacy of novel drugs in restoring the basal metabolic status of RPE.
Retinal pigment epithelial cells (RPE) exist as a monolayer of pigmented epithelial cells strategically located between the photoreceptors and fenestrated endothelium of the choriocapillaris. RPE are highly metabolically active with numerous functions, including (1) phagocytosis of shed photoreceptor discs, (2) recycling of visual pigment to maintain the visual cycle, (3) transport of nutrients, metabolites, ions, and water, (4) absorption of light, (5) protection against photooxidation, (6) secretion of essential factors to support retinal integrity, and (7) formation of the outer blood-retinal barrier1. Degeneration of RPE is associated with metabolic dysfunction and mitochondrial defects, leading to blinding eye diseases such as age-related macular degeneration (AMD) and proliferative vitreoretinopathy (PVR)2.
Two key bioenergetic pathways include glycolysis, which occurs in the cytoplasm, and oxidative phosphorylation (OXPHOS), which occurs in mitochondria. During glycolysis, one molecule of glucose is converted to two molecules of pyruvate and a net production of two molecules of adenosine triphosphate (ATP). In contrast to glycolysis, OXPHOS produces far higher levels of ATP (~32-38 molecules of ATP per glucose molecule). Notably, OXPHOS consumes oxygen and requires functional mitochondria to occur, whereas glycolysis occurs in the cytoplasm and does not require oxygen.
Prior to the introduction of fluorescence- or phosphorescence-based techniques to examine mitochondrial respiration, oxygen levels were measured in permeabilized cell suspensions in chambers equipped with a Clark-type oxygen electrode3. While the Clark electrode is much cheaper than fluorescence-based respirometry and works in non-adherent cells, it is relatively low-throughput, with each respiratory run lasting around 15-20 min and requiring far higher quantities of cells for each sample3. Thus, the fluorescence-based respirometry technique has largely replaced the Clark electrode and has become a popular technique in the fields of metabolism and mitochondrial research.
This protocol describes a high-throughput, high-resolution, fluorescence-based respirometry technique that kinetically measures the OXPHOS and glycolytic bioenergetic profiles of living cells. Since the process of OXPHOS consumes oxygen, the bioenergetic profile for OXPHOS is produced by mapping changes in the oxygen consumption rate (OCR) over time4. In this technique, two fluorophores are embedded in the sensor cartridge sleeve which is connected to fiber optic bundles that emit light, exciting the fluorophores. Changes in fluorophore emission are measured by highly sensitive fluorescence sensors and transmitted through the fiber optic bundle to be converted to an OCR readout5. The fluorophore becomes quenched by oxygen, thus enabling the determination of extracellular oxygen levels in the assay media, known as oxygen flux or OCR. The other fluorophore is a pH sensor probe sensitive to changes in proton efflux, which is converted to a measure of extracellular acidification rate (ECAR). During measurements, the fiber optic bundles with embedded fluorophores are lowered to 200 µm above the cell monolayer, creating a transient micro-chamber that enables rapid, real-time readings. Once a 10% change in oxygen or proton levels is detected, the sensors are lifted upward, allowing a larger volume of media to mix with the transient micro-chamber media, restoring OCR and ECAR values back to baseline. Each sensor cartridge is equipped with four ports to allow for the sequential administration of up to four compounds per well during the assay. Measurements can be collected before and after injection of the compounds in each port, revealing key information about the metabolic status of the cells.
Interrogating these two distinct metabolic pathways can yield important discoveries regarding the metabolic status of RPE following exposure to different pathogenic stimuli, and thus can be used to test the efficacy of drugs in restoring the metabolic integrity of RPE6,7,8. The advent of high-throughput respirometry and the availability of specific mitochondrial inhibitors have stimulated more research in defining the bioenergetic profiles of RPE and identifying defects in metabolism and mitochondria during disease states6,7,8,9,10,11,12,13. High-resolution respirometry has highlighted the key role of metabolic reprogramming of RPE in retinal pathologies such as AMD and PVR. Two key cytokines involved in the pathogenesis of AMD and PVR are transforming growth factor-beta 2 (TGFβ2) and tumor necrosis factor-alpha (TNFα). The induction of epithelial-mesenchymal transition (EMT) by TGFβ2 is accompanied by mitochondrial dysfunction, OXPHOS suppression, and a compensatory increase in glycolytic capacity in RPE6. More recently, the pro-inflammatory cytokine, TNFα, has been shown to induce a significant upregulation of basal OXPHOS and reduced glycolysis in H-RPE7. The administration of dimethyl fumarate significantly suppressed TNFα-induced inflammation in H-RPE and restored mitochondrial morphology and basal bioenergetic profiles7. The divergent metabolic profiles induced by these two growth factors stimulate intriguing mechanistic questions regarding the involvement of metabolic reprogramming in retinal diseases. The following protocol describes the steps for assessing OXPHOS and glycolytic bioenergetic profiles in H-RPE using high-resolution respirometry.
This optimized protocol for high-resolution respirometry of H-RPE involves the use of BAM15 as the uncoupler instead of the commonly used FCCP. While previous studies on high-resolution respirometry of RPE utilized FCCP9,24, BAM15 appears to induce a more robust induction of maximal respiration levels in H-RPE compared to FCCP. While both FCCP and BAM15 are safe to use in cells, BAM15 is reported to have fewer side effects in normal cells compared to FCCP or carbonylcyanide-3-chlorophenylhydrazone (CCCP)25. Kenwood et al. showed that BAM15 depolarizes mitochondria without impacting plasma membrane potential, thus inducing a sustained maximal mitochondrial respiration rate at low cytotoxicity26. FCCP, on the other hand, depolarizes both mitochondria and the plasma membrane and displays higher cytotoxicity26.
There are several critical steps in the protocol, including ensuring the cells are properly plated in a confluent, even, and homogenous monolayer of RPE in all experimental wells of the microplate. Mature RPE are highly dependent upon OXPHOS for energy generation, and thus the cells should be allowed to mature for at least 1 month to ensure that the RPE generate proper basal and maximal OCR readings during the assay. De-gassing the cell culture plate for 1 h at 37 °C in a humidified oven (without CO2) before placing it into the instrument is crucial for accurate ECAR readings, as CO2 can impact the pH of the assay media. It is important to remember to hydrate the sensor cartridge the day before the assay to ensure that it provides reliable OCR and ECAR readings on the day of the assay. Care should be taken to properly reconstitute the drugs to be injected and aliquot the reconstituted drug stocks into smaller volumes for long-term storage to minimize freeze/thaw cycles. It is critical to prepare 10x drug solutions that are diluted in the respective assay media (e.g., dilute in Mito Stress Test assay media for the Mito Stress Test) to factor in the dilution from injection of the drug into each well filled with assay media. With each subsequent drug injection there is more media in each well, and thus the volumes of drug being loaded increase with each injection, and care should be taken to follow the volumes specified in the protocol. It is important to perform quality checks on completion of the experiment by examining the sensor cartridge to ensure that no residual drug is evident and observing the cell culture plate under a microscope to ensure that the confluent and homogeneous cell monolayer remains.
Modifications to this protocol include injecting different drugs in the ports and determining how these drugs impact OCR and ECAR readings. One popular modification is to inject an experimental drug of choice as Port A before injecting the usual Mito or Glycolytic Stress Test drugs. This type of protocol provides insights into how an acute injection of the drug of choice impacts the subsequent OXPHOS and glycolytic parameters. Other modifications include examining different cell types; this requires initial troubleshooting of the optimal seeding cell density and optimization of the assay media by ensuring that basal readings range from 50-100 pmol/min for OCR and 10-20 mpH/min for ECAR. The optimal concentrations of the injected drugs need to be determined for each new cell type examined by observing the OCR and ECAR responses to a serial dilution of the drugs.
A key limitation of the protocol is that cell viability diminishes over time, as the cells are not in a CO2 incubator in their normal growth media, and thus the assay should be completed within 3-4 h to ensure maximal cell viability. Furthermore, exposure to mitochondrial toxins injected into each well can further diminish cell viability over time. Once the assay is completed, the cells must be lysed for evaluation of the protein content to normalize OCR and ECAR readings, and thus the same cells cannot be harvested for subsequent molecular biology assays.
Alternatives to the Seahorse for bioenergetic profiling include the Oroboros Oxygraph 2k (O2k)27, BaroFuse28,29, and Resipher (Lucid Scientific)7. The Oroboros O2k is a closed two-chamber respirometer that utilizes S1 Clark-type polarographic oxygen electrodes. While the Oroboros O2k produces highly sensitive measurements of real-time metabolic flux, the device is labor-intensive as the operator is required to manually inject each drug30. The BaroFuse is a novel multichannel microfluidic perfusion system that uses gas pressure to enable multiple parallel perfusion experiments and is linked to an oxygen detection system to measure OCR. The advantage of this flow culture system is that tissue function and viability is maintained, as opposed to the Seahorse where cell viability diminishes over longer assays. The Resipher utilizes highly sensitive optical oxygen sensors to measure OCR whilst the cells are in a 96-well plate in the incubator, hence enabling continuous OCR measurements over several weeks to months. Notably, these instruments do not measure ECAR, and thus the Seahorse has the advantage of simultaneous exploration of both OXPHOS and glycolysis.
Interrogating real-time bioenergetic profiles of OXPHOS and glycolysis is emerging as a key factor in characterizing RPE health and function. High-resolution respirometry enables an efficient means of comparing the metabolic status of normal and diseased RPE, thus opening new avenues of screening drug efficacy for retinal diseases such as AMD and PVR. Future directions for high-resolution respirometry on RPE include optimizing a protocol to examine bioenergetic profiles for highly polarized RPE monolayers grown on transwell filters. Calton et al. (2016) successfully achieved this by cutting a triangular section of the polarized RPE monolayer grown on transwell filters31. Further expansion of the methodology includes examining the bioenergetic profiles of induced pluripotent stem cell-derived RPE (iPSC-RPE), isolated from patients with different retinal degenerative conditions32. Exploration of how pathogenic cytokines involved in AMD and PVR impact the dynamic nature of RPE metabolism can reveal metabolic vulnerabilities that can inform the identification of novel druggable targets.
The authors have nothing to disclose.
This study was supported in part by grants from: the Fight for Sight Leonard & Robert Weintraub Postdoctoral Fellowship (D.Y.S.); BrightFocus Foundation Postdoctoral Fellowship Program in Macular Degeneration Research (M2021010F, D.Y.S.); Department of Defense, Spinal Vision Research Program under Award no. VR180132 (M.S.-G. and L.A.K.); National Eye Institute of the National Institutes of Health under Award no. R01EY027739 (L.A.K.). Acknowledgment is made to the donors of the Macular Degeneration Research M2021010F, a program of the BrightFocus Foundation, for support of this research. Schematics were created with Biorender.com
2-Deoxy-D-glucose | Sigma | D8375-5G | Reconstituted in the Glycolytic Assay Media from its powder form and used in the Glycolytic Stress Test (Port C injection) |
96 Well Non-Treated Plate, 5/Pack, Sterile | VWR | 229597 | BCA Assay 96-well plate |
Antimycin A from Streptomyces sp. | Sigma | A8674-25MG | Inhibitor of Complex III of the electron transport chain (part of the Port C drug injection together with rotenone) |
BAM15 | Sigma | SML1760-5MG | Uncoupling agent used for the Mito Stress Test (Port B injection) |
BCA Assay | ThermoFisher | 23225 | To determine total protein content in each well |
Calibrant Solution | Seahorse Bioscience | 100840-000 | Solution used to calibrate the probes of the sensor cartridge |
Cell Lysis buffer 10x | Cell Signaling Technologies | 9803S | Dilute 10x Cell Lysis Buffer to a 1x solution using deionized water and add 1 mM PMSF before use |
D-(+)-Glucose solution | Sigma | G8644-100ML | Supplement to be added to the Mito Stress Test assay media and also used in the Glycolytic Stress Test (Port A injection) |
DMSO, Cell culture grade | Sigma-aldrich | D4540-100ML | For reconstituting all mitochondrial drugs except for 2DG |
FCCP | Sigma | C2920-10MG | Uncoupling agent used for the Mito Stress Test (Port B injection) |
GlutaMAX | Gibco | 35050061 | Supplement to be added to the assay media for both the Mito and Glycolytic Stress Tests |
HEPES solution | Sigma | H0887-100ML | Supplement to be added to the assay media for both the Mito and Glycolytic Stress Tests |
H-RPE – Human Retinal Pigment Epithelial Cells | Lonza | 194987 | Primary human fetal RPE cells |
Oligomycin – CAS 1404-19-9 – Calbiochem | Sigma | 495455-10MG | ATP synthase inhibitor used for the Mito Stress Test (Port A injection) and Glycolytic Stress Test (Port B injection) |
Parafilm | Bemis | PM996 | To seal the plate once the cells are lysed to prevent evaporation in the freezer |
Phenylmethylsulfonyl fluoride (PMSF) Protease Inhibitor | Gold Biotechnology Inc | 50-153-2823 | Used at 1 mM for the 1x Cell Lysis Buffer solution |
ReagentPack Subculture Reagents, 100 mL | Lonza | CC-5034 | Passaging reagents for primary human fetal RPE cells. Each kit contains 100 mL Trypsin/EDTA, Trypsin Neutralizing Solution, HEPES Buffered Saline |
Rotenone | Sigma | R8875-1G | Inhibitor of Complex I of the electron transport chain (part of the Port C drug injection together with antimycin A) |
RtEGM Retinal Pigment Epithelial Cell Growth Medium BulletKit – RtEBMTM Basal Medium (00195406) and RtEGMTM SingleQuots Supplements (00195407) | Lonza | 195409 | Media for primary human fetal RPE cells |
Seahorse XFe96 Analyzer | Agilent | High-Resolution Respirometry Instrument | |
Seahorse XFe96 FluxPak | Agilent | 102416-100 | This contains the 96-well Seahorse Cell Culture Microplate, the sensor cartridge and the calibrant solution |
Sodium pyruvate solution | Sigma | S8636-100ML | Supplement to be added to the Mito Stress Test assay media |
Steriflip-GP Sterile Centrifuge Tube Top Filter Unit, 50 mL process volume, 0.22 µm pore size | Millipore-Sigma | SCGP00525 | Tube top filter unit for sterile filtration of the assay media |
Synergy H1 Plate Reader | BioTek | Plate Reader for measuring absorbance at 562 nm for the BCA assay | |
XF base medium without phenol red | Agilent | 103335-100 | Base media for running the Seahorse assay |