Real-time cell metabolic flux assay measures the oxygen consumption rate and extracellular acidification rate, which corresponds to mitochondrial and glycolytic adenosine triphosphate production, using pH and oxygen sensors. The manuscript explains a method to understand the energy status of osteoblasts and the characterization and interpretation of the cellular bioenergetic status.
Bone formation by osteoblasts is an essential process for proper bone acquisition and bone turnover to maintain skeletal homeostasis, and ultimately, prevent fracture. In the interest to both optimize peak bone mass and combat various musculoskeletal diseases (i.e., post-menopausal osteoporosis, anorexia nervosa, type 1 and 2 diabetes mellitus), incredible efforts have been made in the field of bone biology to fully characterize osteoblasts throughout their differentiation process. Given the primary role of mature osteoblasts to secrete matrix proteins and mineralization vesicles, it has been noted that these processes take an incredible amount of cellular energy, or adenosine triphosphate (ATP). The overall cellular energy status is often referred to as cellular bioenergetics, and it includes a series of metabolic reactions that sense substrate availability to derive ATP to meet cellular needs. Therefore, the current method details the process of isolating primary, murine bone marrow stromal cells (BMSCs) and monitoring their bioenergetic status using the Real-time cell metabolic flux analyzer at various stages in osteoblast differentiation. Importantly, these data have demonstrated that the metabolic profile changes dramatically throughout osteoblast differentiation. Thus, using this physiologically relevant cell type is required to fully appreciate how a cell’s bioenergetic status can regulate the overall function.
The formation of bone by the osteoblast is accompanied by coordinated destruction or resorption of bones by osteoclasts. The balance between osteoblastic bone formation and osteoclast resorption is a coupled process describing bone turnover or remodeling, which is essential for skeletal homeostasis. Osteoblast dysfunction leads to impaired bone formation and results in various diseases, including osteoporosis1,2,3. Ex vivo/in vitro differentiation of bone marrow stromal stem cells (BMSCs) to osteoblast precursors and mature osteoblasts results in the formation and deposition of the mineralized bone matrix in the culture vessel over time4,5,6. This bone formation by the osteoblast requires a significant amount of cellular energy. Specifically, collagen synthesis and secretion have been shown to rely heavily on cellular ATP: ADP ratios, and presumably, mineralized vesicle trafficking and secretion require additional ATP7,8,9,10,11. Many researchers have demonstrated that the process of osteoblastogenesis and osteoblast function requires an adequate supply of energy to meet the metabolic demand of bone formation12,13,14,15,16. Therefore, the goal of this method is to characterize the bioenergetic status of primary, murine stromal cells throughout osteoblast differentiation using the real-time cell metabolic flux analyzer. These techniques aid in developing a better understanding of skeletal homeostasis, which may ultimately lead to the development of novel therapeutic options capable of improving skeletal disorders.
The real-time cell metabolic flux analyzer can be used to measure the oxygen consumption rate (OCR) and extracellular acidification rate (ECAR) of live osteoblasts, which corresponds to mitochondrial and glycolytic ATP production, respectively. Fundamental to this methodology is the fact that one H+ ion per lactate is released during glycolysis in the conversion of glucose to lactate, which alters media pH reflected in the ECAR values. Conversely, during the TCA (tricarboxylic acid) cycle, oxidative phosphorylation via the mitochondria produces CO2 by utilizing or consuming oxygen, and therefore monitoring OCR is reflective of this metabolic process. The analyzer measures both OCR and ECAR in the extracellular microenvironment simultaneously and in real-time, which allows for tremendous potential when studying cellular bioenergetics6,17. Additionally, performing these assays is relatively straightforward and easily customizable depending on the experimental goal. Similar techniques have been employed to further understand T-cell metabolic regulation of the immune system18,19, cancer initiation, and progression20, along with multiple other cell types contributing to metabolic syndromes21,22.
The advantages of Real-time metabolic flux analyzer over alternative techniques include (1) the capability to measure cellular bioenergetics of live cells in real-time, (2) ability to perform assay with a relatively small number of cells (requires as low as 5,000 cells), (3) injection ports to parallelly manipulate multiple treatments in a high-throughput 96-well system, (4) use of radioactive label-free automated cell imager for normalization18,23,24. The following methods aim to provide a generalized but detailed description of monitoring cellular bioenergetics in murine BMSCs throughout osteoblast differentiation using the analyzer. It will include routinely performed assays; however, as with many techniques and methods, it is highly encouraged that individual labs determine specific details for their experiments.
Selection of assay and different types of assays available: A wide variety of assay kits and reagents are available to study the bioenergetics of cells while ensuring the reliability and consistency of the experimental results. Additionally, the desktop software also offers assay templates that can be easily customized. The assay can be defined based on the user's needs to measure different metabolic parameters. These assays can be modified in various ways based on the experimental goal and/or scientific question. For example, with four injection ports, multiple compounds can be injected into the assay media to analyze the cellular response specific to each metabolic pathway.
Cell energy phenotype test: This assay measures the live cells' metabolic phenotype and metabolic potential. This assay is also recommended as the first step to get a generalized idea of pathway-specific metabolism. A mixture of oligomycin A-an inhibitor of ATP synthase and Carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP)-a mitochondrial uncoupling agent is injected to understand the cell energy potential. The injection of oligomycin A inhibits the synthesis of ATP, resulting in an increase in the rate of glycolysis (ECAR) to enable the cells to meet their energy demands; on the other hand, the injection of FCCP results in higher OCR due to depolarization of the mitochondrial membrane. Essentially, this assay depicts basal metabolic respiration, and following the dual injections, pushes, or stresses, the metabolic response. Based on these parameters, the software then plots OCR and ECAR of the cells by classifying the cells as aerobic, quiescent, glycolytic, or energetic state over time25,26.
ATP real-time production rate assay: This measures the cellular ATP production simultaneously from glycolysis and mitochondrial respiration. This assay quantitatively measures the metabolic shifts from the two energy pathways and provides data on the mitochondrial and glycolytic ATP production rates over time. The assay obtains basal OCR and ECAR data followed by calculating mitochondrial ATP production rate through injection of oligomycin A and glycolytic ATP production rate through injection of rotenone + antimycin A mixture (total inhibition of mitochondrial function), resulting in mitochondrial acidification17,27.
Cell mitochondria stress test (or cell mito stress test): This measures the mitochondrial function through ATP-linked respiration, quantifies cellular bioenergetics, identifies mitochondrial dysfunction, and measures cells' response to stress. Various parameters, including basal and spare respiratory capacity, ATP-linked respiration, maximal respiration, and non-mitochondrial oxygen consumption, can be obtained in one assay. This assay involves sequential injections of oligomycin A, FCCP (mitochondrial uncoupling agent), a mixture of rotenone/antimycin A inhibitors to efficiently analyze the effect of these on the mitochondrial function28.
Flexibility mito fuel flex test: This measures the mitochondrial respiration rate by the oxidation of the three primary mitochondrial fuels by the presence and absence of their inhibitors. The sequential inhibition of glucose, glutamine, and fatty acids aids in measuring the dependency, capacity, and flexibility of cells and the dependency of the cells in various cellular pathways to meet the energy demand. When the mitochondria cannot meet the demands of the blocked pathway of interest by oxidizing other fuels, the cells enter a dependency state. The capacity of the cells is calculated by inhibition of the other two alternative pathways followed by the inhibition of the pathway of interest. The flexibility of cells helps in understanding the ability of mitochondria to compensate and meet the fuel needs of the inhibited pathway. It is calculated by subtracting the dependency of cells from the capacity of cells. Three different inhibitors are used independently or as a mixture of two to effectively calculate the assay parameters. 2-cyano-3-(1-phenyl-1H-indol-3-yl)-2-propenoic acid (UK5099) inhibits the oxidation of glucose by blocking the pyruvate carrier in glycolysis. Bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl) (BPTES) ethyl sulfide inhibits the glutamine oxidation pathway, and etomoxir inhibits the oxidation of long-chain fatty acids29.
Figure 1: Schematic representation of the methodology for culturing and preparing osteoblasts for analysis. Murine BMSCs are isolated from long bones, cultured, and seeded in 96-well plates at 25,000 cells/well density. Culturing these cells in Osteoblast specific media is started when they reach 80%-100% confluency to start their differentiation. The assays are performed at different stages of differentiation. The cartridge plates are hydrated one day prior to the assay. On the day of assay, different inhibitors are injected in the ports of the sensor cartridges based on the assay requirements, and a calibration buffer is added to the 96-well calibration plate. After calibration, the real-time cell metabolic flux assay is performed, followed by imaging the cell culture microplate using the microplate imager to normalize real-time cell metabolic flux analyzer data with cell count. Please click here to view a larger version of this figure.
All the procedures were based on the guidelines and approval of the Institutional Animal Care and Use Committee at Vanderbilt University Medical center.
1. Preparation of reagents and assay setup
Figure 2: The cell culture microplate, specifically designed for the analyzer. (A) The four background correction wells, A1, A12, H1, H12, are highlighted. These wells only contain assay media without any cells. (B) The barcode on the side of the plate to scan the plate using the imaging reader and analyzer. Please click here to view a larger version of this figure.
2. Preparation of sensor cartridge for extracellular flux calibration
3. Real-time cell metabolic flux analyzer media preparation
4. Preparation of compounds for the sensor cartridges
5. Prepare cell culture microplate for assay
6. Setting up the assay and imaging
Figure 3: The controller software. The software verifies the equipment is connected and is set to 37 °C. The template files for different assays that can be performed with the extracellular flux analyzer can be selected to customize the assay further based on the experimental goals. Please click here to view a larger version of this figure.
7. Obtain brightfield images
NOTE: This step is optional. If no imaging equipment is available, skip to step 8.
Figure 4: The cell imaging software communicates to the imaging reader through the computer. The cells in the microplate can be imaged before and after the assay, and the cell count/well is obtained after the assay to normalize the data. Please click here to view a larger version of this figure.
8. Running the assay
9. Obtain fluorescence images and normalize
NOTE: This step is an optional but preferred method for the normalization of BMSCs and osteoblasts. If no imaging equipment is available, another normalization method needs to be performed, such as protein or DNA isolation and quantification.
Figure 5: Representative images from the imaging software used for normalization of data from the assay. (A) Stitched bright field image showing the cell confluence throughout the entire well. (B) Stitched fluorescence image showing Hoechst-stained nuclei of osteoblasts used for counting cell numbers to normalize the assay results. These are osteoblasts after 7 days of differentiation. Please click here to view a larger version of this figure.
Figure 6: Representative graphs for routinely performed assays to understand the cellular bioenergetic profile of control vs. treatment group with their respective standard errors. (A) The cell energy phenotype test. The plot represents the glycolysis (ECAR) vs. mitochondrial respiration (OCR) of the control vs. two treatment groups (n = 3). The injection of oligomycin A and FCCP stressors elevates the baseline activity, indicated by open symbols, whereas closed symbols indicate the response of cells to different stressors. (B) The real-time ATP rate assay. The ATP rate assay indicates that both the control and treatment groups (n = 2) produce more ATP through glycolysis compared to oxidative phosphorylation. (C) The mito stress test. The mito stress test provides the mitochondrial respiration rate of control vs. treatment cells (n = 2) over time and the effect of inhibitors on the cells after their respective injections. (D) The mito fuel flex test. The mito fuel flex test measures the percentage oxidation of control and treatment groups (n = 2) with respect to glucose, glutamine, and fatty acid pathways and the dependency and flexibility of the cells on these mitochondrial fuels. Please click here to view a larger version of this figure.
The protocol describes a generalized description of how the extracellular flux assays aids in understanding the cellular bioenergetics of osteoblasts derived from murine BMSCs. We have detailed these routinely performed assays and important notes to be considered before, during, and after the assay. The two major ATP production pathways, glycolysis, and mitochondrial oxidative phosphorylation, are widely discussed to better understand the capability of cells to interchange between the pathways, thereby meeting the energy demands of the cells. Once the assay is complete, the assay results are normalized based on the cell count and exported to the respective assay report generator file. The report generator automatically calculates the assay parameters and provides a summary report of the assay. Figure 6 illustrates the representative results of routinely performed assays to better understand how mature osteoblasts control versus treatment groups react when different inhibitors are injected.
Generic, representative images of possible expected results are shown in Figure 6. For example, in Figure 6A, the cell energy phenotype monitors the OCR vs. ECAR by calculating the cells' baseline phenotype, stressed phenotype, and metabolic potential. The injection of the oligomycin A and FCCP stressor mix increases the control group's baseline activity (open symbols) by increasing the utilization of both pathways. In response to the stressors, a significantly high energy level is noticed in the control and treatment 1 group (closed symbols). On the other hand, treatment 2 had a comparatively lower baseline activity, and the cells became more aerobic. This assay aids in understanding the bioenergetics of the cells in response to different stressors.
Real-time ATP rate assay calculates the total cellular ATP production rate based on the sum of glycolytic and mitochondrial ATP production rates.
ATP Production Rate (pmol ATP/min) = glycoATP Production Rate (pmol ATP/min) + mitoATP Production Rate (pmol ATP/min).
Figure 6B indicates that both the control and treatment groups produce more ATP through glycolysis compared to that of oxidative phosphorylation. While the treatment group exhibits significantly higher total ATP production, the cells have consistently shifted from glycolytic to oxidative metabolism. This comparison of the control and treatment group indicates that this specific treatment exhibits a different bioenergetic profile compared to the control group.
Figure 6C is an example of the mitochondrial respiration rate over time, which is detailed. The basal respiration rate in the treatment group is comparatively less than in the control group. The respiration and ATP production rates in both groups are decreased along with the proton leak followed by an oligomycin A injection. The respiration rates of the cells rise back again to their maximal respiration after the injection of FCCP. The final injection of rotenone/antimycin A decreases the OCR again, resulting in spare respiration, which is measured by the differences in maximal and basal respiration. After the third injection, the mitochondrial respiration is shut down by the combination of rotenone/antimycin A, which targets and inhibits the electron transport chain (ETC) complex I and III, thereby enabling us to calculate the non-mitochondrial respiration. Compared to the control group, both the basal and maximal respiration in the treatment groups is relatively lower; this suggests that the treatment could affect the mitochondrial respiration of osteoblasts.
The mito fuel flex test measures the potential of mitochondria to oxidize the three different mitochondrial fuels, glucose, glutamine, and fatty acids. The cell's dependency, flexibility, and capacity on different pathways that fuel the mitochondrial respiration are calculated based on the oxidation of the respective mitochondrial fuel. Figure 6D shows that the control group is highly dependent on the glucose pathway. At the same time, the treatment enhanced the capacity of glucose oxidation of the cells to actively meet the fuel needs of the inhibited pathway. On the other hand, the oxidation of glutamine was efficient in the control group, resulting in comparatively higher dependency of cells to that of the treatment group, thereby increasing the overall capacity of the control group. The fatty acid pathway shows that the treatment has increased the overall capacity of the cells due to the higher dependency on mitochondrial fuel while compensating the fuel needs.
The real-time cell metabolic flux analyzer can be used to explore cellular energetics under different conditions. The protocol illustrates the efficient isolation of BMSCs, culturing cells in appropriate cell culture plates, and their differentiation to mature osteoblasts, which can be used for various assays using the extracellular flux analyzer. Further, the critical steps of real-time cell metabolic flux assay, including hydration of sensor cartridges, loading of the injection ports, performing the assay, normalization of the data, and data analyses, are also explained in detail. This assay evaluates the response of osteoblasts to different mitochondrial and glycolytic inhibitors to understand the bioenergetics of the cells. This protocol is optimized specifically for osteoblasts based on the cell type, and this method offers a more robust and precise guide compared to the manufacturer's standard protocol. Several parameters in this protocol, including the cell seeding density, the concentration of the compounds, addition of exogenous substrate to the media, buffering capacity, need to be optimized based on the specific cell type and its background.
The extracellular flux analyzer provides quick and reliable measures of cellular metabolic functions in ex vivo cultured cells. Compared to other biological assays, the total assay time is typically between 60 to 120 min. The equipment maintains normal cellular and physiological conditions by maintaining the preset temperature of 37 °C, facilitating efficient and reproducible assay results with reduced complexities. Typically, baseline OCR and ECAR are recorded three to four times before adding inhibitors. The inhibitors and treatments are also sequentially injected, facilitating the cell's metabolic response measurement over time. The analyzer allows researchers to achieve standardized results under optimized conditions. The analyzer also offers the ability to inject different compounds during the assay to observe real-time changes in the respiration of cells. For example, for ATP rate assay, use a 2 µM oligomycin A and 1 µM rotenone/1 µM antimycin A mixture as the default compound injected-Port A: 2 µM Oligomycin A; Port B: 1 µM Rotenone/1 µM Antimycin A.
The assay media is different from the typical growth media by a few factors, including the absence of bicarbonate to better detect changes in pH, the absence of glucose, glutamine, and pyruvate supplements allows the experimenter to customize the exogenous substrates added, and the media does not contain phenol red to precisely calculate the pH values.
Glucose, L-glutamine, and sodium pyruvate are the most used exogenous substrates. The use of these substrates is specific to cell type and experimental questions. In this context, it is recognized that these assays, while valuable, are performed using an artificial system. For example, it is recommended to use 10 mM glucose; however, this is supraphysiological levels, and researchers should consider whether other glucose concentrations would be more appropriate. To this point, it is also essential to consider whether glucose uptake in BMSCs and osteoblasts depends on insulin, and in that case, insulin should also be added. Since this remains a topic of debate within the field31,32, the inclusion of insulin is preferred. Along a similar line, if fatty acid utilization is relative to the experimental question, the assay media should be further supplemented with fatty acids. The importance of these exogenous substrates plays a critical role in the assays and data interpretation; therefore, concentrations of these substrates should be optimized based on the cell type and research question.
One of the primary advantages of the extracellular flux analyzer is that it requires a minimal number of cells to run the assays, with a high n given the 96-well format. The sensor cartridges also enable the ability to inject different inhibitors and compounds through the four injection ports. The flexibility to modify the assay, to perform acute injections with additional inhibitors and treatments of interest is an added advantage of this technique. These features make the analyzer capable of doing high throughput, real-time assays on a minimal number of cells and hence preferable over the classical Clark oxygen electrode method. Although the electrode method is inexpensive and simple for measuring mitochondrial respiration, it gives far higher background noise and has a lower resolution than the analyzer24,33.
Additionally, in the workflow described, the cell bioenergetic profile obtained from the analyzer can efficiently be normalized using the microplate imager by counting the cells in the microplates after the assay34. The number of viable cells may vary from well to well after the assay; use of cell count is preferable over other methods, including the protein or DNA normalization as the protein or DNA content of cells does not remain constant under different conditions or treatments for which the cellular energetics are being compared. The protein or DNA content may vary under the influence of different treatments and is a major disadvantage of normalization using this method, for example, the protein content of osteoblast changes during differentiation. As the cells mature during osteoblastogenesis, they start secreting extracellular matrix protein increasing the total cellular protein content. This can be problematic when comparing the assay results of cells under different growth phases or time points of osteoblastogenesis.
Given the multitude of advantages, the analyzer to monitor cellular bioenergetics has been a tremendous addition to advancing the field. However, limitations do exist. For example, one of the major limitations of these assays is that they can only be performed at the cellular or organoid level, and the data is not sufficient to provide us with a complete idea of what would happen in vivo under different physiological conditions. As mentioned previously, the cellular environment created during the assays is also far from the original physiological niche. The artificial micro-environment created by supplementing different nutrients like glucose, pyruvate, glutamine, and oleic or palmitic acid are often higher than the normal physiological levels of the osteoblasts. The addition of fatty acid as one of the nutrients for the assay is limited, as most of the higher carbon chain length fatty acids present in our body are very hydrophobic and highly insoluble. This results in technical difficulties in adding them to the assay media. The metabolic response of cells to single fatty acid like oleic or palmitic acid is different under other physiological conditions, where the cells experience a cocktail of various fatty acids and binding proteins. Finally, the measurement of ATP using this method, while arguably superior to static luminescent assays, remains indirect as this technique only measures oxygen. While it is not a direct measurement, it is true that the addition of the ATP synthase inhibitor, oligomycin A, followed by electron transport chain inhibitors while measuring OCR provides strong evidence of changes occurring in ATP. Therefore, techniques like mass spectroscopy can be used to support these data. Nonetheless, while these limitations should be carefully considered, this tool provides invaluable information about cellular bioenergetics.
Finally, it is noted that to understand how the said alterations in bioenergetics impact cellular functionality, specifically of the BMSCs and osteoblasts, complementary experiments are needed to be included in vitro culturing followed by alkaline phosphatase (ALIP), Von Kossa, or Alizarin staining.
In conclusion, the methods described above provide the basis for monitoring various metabolic pathways in osteoblasts using the real-time cell metabolic flux analyzer. Performing such experiments will lead to a deeper understanding of how cellular bioenergetics in primary, murine BMSCs throughout osteoblast differentiation can be modulated to improve skeletal-related outcomes.
The authors have nothing to disclose.
This work was supported by the National Institute of Health (NIH) National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) Grant AR072123 and National Institute on Aging (NIA) Grant AG069795 (to ERR).
0.25% Trypsin EDTA | Sigma-Aldrich | T4049 | |
2-cyano-3-(1-phenyl-1H-indol-3-yl)-2-propenoic acid | Sigma – Aldrich | PZ0160 | UK5099 |
Antimycin A | Sigma – Aldrich | A8674 | |
Ascorbic acid | Sigma-Aldrich | A4544-100G | |
Bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl sulfide | Sigma – Aldrich | SML0601 | BPTES |
Carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone | Sigma – Aldrich | C2920 | FCCP |
Cytation 5 imaging reader | BioTek | N/A | Microplate imager |
Etomoxir sodium salt hydrate | Sigma – Aldrich | E1905 | |
Hoechst 33342 Solution (20 mM) | Thermo Scientific | 62249 | |
Insulin | Sigma – Aldrich | I6634 | |
Oleic Acid-Albumin from bovine serum | Sigma – Aldrich | O3008 | |
Oligomycin A – 5 mg | Sigma – Aldrich | 75351 | |
Rotenone | Sigma – Aldrich | R8875-1G | |
Seahorse XF 1.0 M Glucose Solution | Agilent Technologies | 103577-100 | |
Seahorse XF 100mM Pyruvate Solution | Agilent Technologies | 103578-100 | |
Seahorse XF 200mM Glutamine solution | Agilent Technologies | 103579-100 | |
Seahorse XF DMEM media | Agilent Technologies | 103575-100 | DMEM assay media eith 5mM HEPES, pH 7.4, without phenol red, sodium bicarbonate, glucose, pyruvate, and L-glutamine |
Seahorse XFe96 Analyzer | Agilent Technologies | S7800B | Real- Time Metabolic flux analyzer |
Seahorse XFe96 FluxPak | Agilent Technologies | 102416-100 | Includes XFe96 Sensor cartridges, Cell culture microplates, and Seahorse XF Calibrant solution |
The Cell imaging 1.1.0.11 software | Agilent Technologies – BioTek | ||
Wave software 2.6.1 | Agilent Technologies | ||
β-glycerol phosphate | Sigma-Aldrich | G9422-50G |