Metabolic analysis of mouse bone-marrow-derived dendritic cells using an extracellular flux analyzer

Summary Dendritic cell (DC) maturation induced by Toll-like receptor (TLR) agonists requires the activation of downstream metabolic changes. Here, we provide a detailed protocol to measure glycolysis, mitochondrial respiration, and fatty acid oxidation in mouse bone-marrow-derived DCs with the Seahorse XF24 extracellular flux (XF) analyzer. XF analysis with the Seahorse bioanalyzer has become a standard method to measure bioenergetic functions in cells, and this protocol can be adapted to other immune cells. For complete information on using this protocol, please refer to Gotoh et al. (2018).


SUMMARY
Dendritic cell (DC) maturation induced by Toll-like receptor (TLR) agonists requires the activation of downstream metabolic changes. Here, we provide a detailed protocol to measure glycolysis, mitochondrial respiration, and fatty acid oxidation in mouse bone-marrow-derived DCs with the Seahorse XF24 extracellular flux (XF) analyzer. XF analysis with the Seahorse bioanalyzer has become a standard method to measure bioenergetic functions in cells, and this protocol can be adapted to other immune cells. For complete information on using this protocol, please refer to Gotoh et al. (2018).

BEFORE YOU BEGIN
Immunometabolism is an emerging field of investigation at the interface between the historically distinct disciplines of immunology and metabolism (Mathis and Shoelson, 2011). Recent studies on immunometabolism in myeloid dendritic cells (mDCs) provide new insights on the mechanism of the critical controllers of innate and adaptive immunity (Pearce and Everts, 2014) (O'Neill and Pearce, 2015). In particular, extracellular flux (XF) analysis has become a standard method to measure bioenergetic functions in DCs (Everts et al., 2012) (Pantel et al., 2014) (Pelgrom et al., 2016).
This protocol comprises several methods to quantify the energy utilization of DCs in real-time using a Seahorse extracellular flux analyzer. The basic protocol describes a standard test with an XFe24 analyzer. If you use an XFe96 analyzer, you should adjust the number of DCs and the volume of buffer used according to the XFe96 analyzer. Please check the latest information on the Agilent website (https://www.agilent.com/en/product/cell-analysis/real-time-cell-metabolic-analysis) before your experiments.
In this protocol, we mainly described the analytical methods using mouse bone marrow-derived DCs (BMDCs). If you are using other cell types, please refer to Table 1. Although we show the number of cells that we could analyze so far in Table 1, we recommend that you consider the appropriate number of cells before starting the experiments.
CRITICAL: To obtain better results, it is important to seed more cells by monolayers KEY RESOURCES TABLE  (Gotoh et al., 2018) Mouse splenic dendritic cells 200,000 RPMI (Gotoh et al., 2018) Human monocyte derived dendritic cells 200,000 RPMI n/a Mouse embryonic fibroblasts (MEF) 20,000 DMEM (Monji et al., 2016) Mouse primary neurons 40,000 DMEM  Mouse primary oligodendrocytes 40,000 DMEM  Mouse  (Gotoh et al., 2020) Mouse bone marrow triple negative cells 200,000 DMEM (Gotoh et al., 2020) Mouse bone marrow stroma cells 10,000 DMEM n/a Warm the assay medium to 37 C without CO2. Adjust the pH with HCl and NaCl solutions to pH 7.4. Filter using polycarbonate membrane filter.
Keep at 37 C until ready to use without CO2.
CRITICAL: The value of ECAR is affected by the pH of assay medium. Therefore, to accurately measure the ECAR value, it is necessary to adjust the pH to 7.4 4 h before assay under 37 C.
Note: We recommend using within 4 h of medium preparation to avoid reagents degradation over the time and pH changes. In addition, we recommend warming the assay medium in a 37 C incubator without CO2 1 h before its use on cells. Preparation of the XF assay medium Once 8.4 g of RPMI-1640 medium powder, D-glucose, fetal bovine serum, and sodium pyruvate are dissolved, add dH20 to bring the total volume to 1000 mL. Adjust the pH with HCl and NaCl solutions to pH 7.4. Sterilize the cell culture media with Bottle Tops and Filter Units and store at 2 C-8 C.
CRITICAL: Adjust pH as precisely as possible; the final pH will affect the results of ECAR data. Because glycolysis is measured by the changes in extracellular pH, the XF assay medium should not contain any buffering reagents.
Note: If you use other cells, prepare a DMEM-based medium by referring to Table 1. Preparation of the BSA solution ( Figure 1A)

Reagent
Add 100 mL of 150 mM NaCl in a 250 mL glass beaker. Warm the beaker with a stir bar in a 37 C incubator. Add 2.267 g of BSA to 100 mL of 150 mM NaCl in a 250 mL glass beaker. Stir until the BSA is completely dissolved. Preparation of the PA solution ( Figure 1B) Add 44 mL of 150 mM NaCl to a 100 mL glass beaker. Warm the beaker with a stir bar in a 70 C incubator. Add 30.6 mg PA to 44 mL of 150 mM NaCl in a 100 mL glass beaker. The PA solution may appear increasingly cloudy as the temperature reaches 50-60 C but will clarify approaching 70 C. Stir until the PA is completely dissolved.
CRITICAL: PA is difficult to dissolve at temperatures of <50 C. Therefore, take care not to lower the PA solution's temperature to <50 C until PA binds to BSA. If PA do not conjugate to BSA, PA is not incorporated into cells. Therefore, this section is important.
Conjugating PA to BSA ( Figure 1D) CRITICAL: If PA do not conjugate to BSA, PA is not incorporated in DCs. Therefore, this section is important. In particular, because PA is difficult to dissolve at temperatures of <50 C, be careful the temperature of PA solution.
Add 50 mL of 150 mM NaCl to a 250 mL glass beaker. Warm the beaker with a stir bar in a 37 C incubator. Transfer 5 mL of the PA solution with a 10 mL pipette into a 250 mL glass beaker. Add total of 40 mL of PA solution into 250 mL glass beaker in 8 batches of 5 mL each. Stir a 250 mL glass beaker at 37 C for 1 h. Add 10 mL of 150 mM NaCl to a 250 mL glass beaker and adjust the final volume to 100 mL. Check the solution's pH and adjust it to pH 7.4. Separate into 2-5 mL each and freeze at À20 C.

STEP-BY-STEP METHOD DETAILS Bone marrow cell isolation
Timing: 3 h Note: Ensure that all the reagents and samples are kept on ice during the entire procedure. Perform all steps in a laminar flow hood with sterile equipment to maintain sterility.
For 2-4 whole Seahorse XF24 cell culture microplates, you will need BMDCs from one mouse.
Note: We recommend using male 6-10 week-old mice. 1. Sterilize the dissection kit and bench area with 70% (vol/vol) ethanol spray. Euthanize the mice with CO 2 or inhaled anesthetics, such as isoflurane or sevoflurane, followed by cervical dislocation. 2. Pin down the mouse to expose its abdomen. Spray the euthanized mice with 70% (vol/vol) ethanol to sterilize its skin. 3. Use scissors to cut along the midline of the abdomen until exposing the femurs. Remove the femurs and tibias and place them in a dish containing PBS. (Methods video S1) 4. Remove the muscle and as much connective tissue as possible from the femurs and tibias. Place the harvested bones on a 60-mm dish filled with PBS.
CRITICAL: If the bone is broken or the soft tissue is inadequately removed, the number of bone marrow cells harvested may be decreased.
This section is shown in Methods video S2.
5. Trim both ends of the femurs and tibias carefully using sterile scissors to expose the interior marrow shaft (Methods video S2). 6. Use a 5 mL syringe to draw up to 5 mL of fresh culture medium. Attach a 21-gauge needle to the syringe. Hold the bone over a fresh 60-mm dish, with its narrow end pointing down. Flush the marrow out of the bone with 5 mL of PBS (Methods video S3).
7. Aspirate the marrow and the medium from the 60-mm dish and pipette up and down three times, rinsing the sides of the dish each time to disperse the marrow. 8. Collect the cell suspension with the syringe and add the suspension to a 15 or 50 mL tube.
Optional: Pass the cell suspension through a 70-mm cell strainer to remove any remaining bone or muscle fragments.
9. Centrifuge the cell suspension at 3,0003 rpm (800 3 g) at 4 C for 5 min to pellet the cells.
Discard the supernatant by gently tilting the tube and pouring the media into a waste disposal beaker. Recap the tube and gently tap to disperse the cell pellet. 10. Add 1.0 mL / body of RBC lysis buffer to the cells and gently tap the tube with fingers to mix the lysis solution for 30-60 s. Add 10 mL of RPMI medium to dilute the buffer after RBC lysis. 11. Centrifuge the cell suspension at 3,0003 rpm (800 3 g) at 4 C for 5 min to pellet the cells.
Discard the supernatant by gently tilting the tube and pouring the media into a waste disposal beaker. Recap the tube and gently tap to disperse the cell pellet.
CRITICAL: If red blood cell lysis fails, return to step 10.
12. Resuspend the cells in 10 mL of culture medium and place on ice. Mix the cell culture medium well and count the cells; 2-6 3 10 7 bone marrow cells can be collected from two tibias and two femurs. 13. Day 0: Seed 2 3 10 6 bone marrow cells in 2 mL of culture medium onto a 12-well culture plate with 10-25 ng/mL of GM-CSF in a 37 C and humidified 5% CO 2 incubator.
CRITICAL: Be careful to avoid collecting the highly adherent cells.
Important: Non-BMDCs can be eliminated by early washing steps, discarding highly adherent cells, and enriching or sorting for CD11c + cells (Helft et al., 2015).
Note: If the culture media is not added properly, the survival rate and number of the BMDCs will be reduced.

Purification of myeloid dendritic cell
Timing: 2 h Note: The percentage of CD11c + BMDCs is approximately 60%-90%. Use the following process to obtain a more concentrated population of CD11c + BMDCs. After this step, we speculate that 0.5-1 3 10 7 DCs / body can be recovered. Optional: If you use the autoMACSâ Pro Separator, refer to the corresponding user's manual on using the autoMACSâ Pro Separator.

OPEN ACCESS
Note: After collecting the CD11c high mDCs, we recommend checking the purity of BMDCs by flow cytometry. If you will check the purity of BMDCs, refer https://www.biolegend.com/ en-us/protocols/cell-surface-flow-cytometry-staining-protocol.
Hydration of a Seahorse XFe sensor cartridge 29. Place the assay cartridge upside down next to a 24-well utility plate. Add 1.0 mL of XF calibrant solution to each well of the 24-well utility plate. 30. Put the cartridge back onto the utility plate and the sensor cartridge with the lid. 31. Place it in a 37 C incubator without CO 2 to hydrate for 4-48 h. To prevent evaporation of the water, verify that the incubator is properly humidified.
Note: The sensor cartridge should be hydrated for at least 4 h before assay. However, we do not recommend using an XFe sensor cartridge after >48 h of hydration.

Preparation of poly-L-lysine-coated microplates
Note: If you use non-adherent cells including BMDCs, we recommend this step. The adhesion of cells to a plate affects the accuracy of the experiment.
32. Apply 50 mL of 0.01% poly-L-Lysine (PLL) or poly-D-Lysine (PDL) to the wells of the 24-well XF cell culture microplate. Tap the microplate to ensure that the liquid completely covers the bottom of the well. 33. Incubate the microplate with the PLL or PDL solution for at least 5 min at room temperature. 34. Remove the PLL or PDL solution by aspiration and thoroughly rinse the bottom of the plate with sterile water. Dry for at least 2 h before seeding cells. 36. Centrifuge the cell suspension at 3,0003 rpm (800 3 g) at 4 C for 5 min to pellet the cells. Discard the supernatant by gently tilting the tube and pouring the media into a waste disposal beaker. Recap the tube and gently tap to disperse the cell pellet. 37. Resuspend the cell pellet in XF assay medium at a density of approximately 2 3 10 6 /mL. Mix the cell suspension well and seed 100 mL of it on a PLL-coated 24-well XF microplate without a blank well.
Note: This protocol describes measuring oxygen consumption rates (OCR) and ECAR at 2 3 10 5 BMDCs/well using an XF24 analyzer. If you would like to analyze only OCR, you will obtain the results by reducing the number of mDCs used. If you would like to analyze splenic dendritic cells or plasmacytoid dendritic cells, you should seed 2-5 3 10 5 mDCs/well. Note: BMDCs are non-adherent cells. Therefore, to use swinging buckets or plates attach BMDCs to the bottom of the plate.
39. Slowly add 400 mL of XF assay medium into the sample to avoid disrupting the cell monolayer. Add 500 mL of XF assay medium into the blank well. Normally, 4 wells (1A, 3C, 4B, 6D) are used as a blank well ( Figure 2B).

CRITICAL: No cells should be placed in the blank wells.
40. Incubate the microplate for 30-60 min at 37 C without CO 2 until you are ready to load the plate onto the XF24 extracellular flux analyzer.
CRITICAL: This session is critical for obtaining accurate results of the Seahorse assay. The number of cells directly affects the values of OCR and ECAR. If you seed too many cells, the cells will be seeded in multiple layers. You should observe the cells under an inverted microscope to confirm that a monolayer of cells is present in all the wells.

XF assay
41. Prepare 2 mL of 103 TLR agonist, mitochondrial inhibitor, or uncoupler injection solutions and an XFe sensor cartridge. 42. Add 56-75 mL of each 103 injection solution into injection ports A, B, C, and D, respectively (Figure 2A). Return the hydration cartridge to the 37 C incubator without CO 2 before setting up the run. 43. Design a study protocol in the XF24 extracellular flux analyzer software provided by the manufacturer ( Figures 3A-3D). Click the ''I'm Ready'' button and place the hydrated cartridge from the 37 C incubator without CO 2 on the XF24 extracellular flux analyzer ( Figures 2B and 3E).
Note: We recommend 4 blank ( Figure 3B) and 3 or more replicas wells for 1 plate.
44. Wait for the machine to calibrate the sensors. When the calibration is finished, keep the cartridge on the machine while the calibrant plate in the load position is sent out. At this point, take the XF24 cell culture microplate from the 37 C incubator without CO 2 and put it into the load position and click the ''CONTINUE'' button. 45. After the XF24 extracellular flux analyzer run is finished, remove the assayed XF24 cell culture microplate, place it in a 37 C incubator to determine the cell counts or the protein concentration, and click the ''CONTINUE'' button to end the program.

After XF assay (optional)
Note: Normalization is an important component in the workflow for performing analysis of raw data to ensure accurate and consistent interpretation of results. We show a method using the number of cells. If you will use the method of normalization (e.g., using protein concentration, cell count, DNA content), refer https://www.agilent.com/cs/library/technicaloverviews/ public/Methods_and_Strategies_for_Normalizing_Tech_Overview_022118.pdf Note: If XF plate is coated with protein containing cellular adherents (e.g., collagen, laminin, Matrigel), normalization using total protein is also not applicable.
46. Carefully remove the supernatant and add PBS 100 mL. 47. Strip off the cells on bottom and count the cells. 48. Standardize using Wave software.

EXPECTED OUTCOMES
Measuring TLR-induced metabolic changes: The binding of TLR and TLR agonists leads to rapid activation of DCs (Kawai and Akira, 2011). Activated DCs by TLR agonist, including LPS (TLR4 ligand) and CpG-DNA (TLR9 ligand), exhibit a rapid increase in glucose consumption and lactate production, as indicated by the real-time changes in extracellular acidification (ECAR) and OCR Gotoh et al., 2018). DC activation is dependent on an early increase in glycolysis. Conversely, LPS and CpG-DNA do not affect the mitochondrial respiratory chain in the DCs. Imiquimod, a TLR7 agonist, also enhances glycolysis of DCs. Unlike LPS and CpG, imiquimod impairs the mitochondrial respiratory chain by binding and inhibiting the quinone oxidoreductases NQO2 (Gross et al., 2016). Figure 4 depicts ECAR and OCR measurements to detect metabolic changes by TLR stimulation.
Measuring post TLR stimulation changes: Several studies have also shown that immature DCs use mitochondrial oxidative phosphorylation (OXPHOS) as a core metabolic process, rather than glycolysis. Conversely, mature DCs induced by TLR ligands display increased glycolysis and inactivated OXPHOS (Everts et al., 2012;Gotoh et al., 2018). Analysis can be done to measure the difference in mitochondrial and glycolytic stress between the immature and mature DCs ( Figure 5).
Measuring fatty acid metabolism in DCs: Fatty acid oxidation has critical roles in regulating adaptive and innate immune responses (O'Neill et al., 2016). Tolerogenic and mature DCs showed substantially different levels of proteins and metabolites involved in the fatty acid oxidation (FAO) pathway (Malinarich et al., 2015) ( Figure 6).
The individual protocols to measure each of these metabolic changes are outlined in ''quantification and statistical analysis.''   2. Centrifuge the cell suspension at 3,0003 rpm at 4 C for 5 min to pellet the cells. Discard the supernatant by gently tilting the tube and pouring the media into a waste disposal beaker. Recap the tube and gently tap to disperse the cell pellet. 3. Resuspend the cell pellet in XF assay medium at a density of approximately 2 3 10 6 /mL. Mix the cell suspension well and seed 100 mL of it on a PLL-coated 24-well XF microplate without a blank well. 4. Centrifuge the microplate for 2-5 min at 2000 rpm at room temperature to allow the cells to settle into a monolayer at the bottom of the plate. 5. Slowly add 400 mL of XF assay medium with PA (0.3 mM) +BSA (0.05 mM), BSA (0.05 mM), or 200 mM Etomoxir into the sample to avoid disrupting the cell monolayer. Add 500 mL of XF assay medium into the blank well. 6. Incubate the microplate for 30-60 min at 37 C without CO2 until you are ready to load the plate onto the XF24 extracellular flux analyzer ( Figure 6).

Note:
Recently, several studies showed that Etomoxir induces Cpt1a-independent off-target effects at concentrations >10 or 100 mM (Divakaruni et al., 2018) (Raud et al., 2018. Therefore, if you want to analyze only the CPT1a-specific inhibitory effect, we recommend to analysis with Etomoxir at concentrations < 10 mM.

LIMITATIONS
In these protocols, we have described several analytic methods using mouse BMDCs. Because we have analyzed similar assays on various cells, we consider these protocols to be beneficial. Conversely, there are several new reagents and protocols. Therefore, when you perform analyses with a Seahorse extracellular flux analyzer, we recommend checking for new protocols at https:// www.agilent.com/en/product/cell-analysis/real-time-cell-metabolic-analysis.

Problem 1
RBCs are not hemolysed by RBC lysis buffer. Refer to Figure 9 (step 10).

Potential solution
Extend the mixing time to 60-120 s. Add 2 mL / body of RBC lysis buffer to the cells. Remake RBC lysis buffer.

Problem 2
Bone marrow cells are not properly differentiated into BMDCs (step 15).

Potential solution
Reduce the number of bone marrow cells seeded onto the culture plate. Extend the culturing period of the BMDCs. Increase the concentration of GM-CSF. Use a different lot of fetal bovine serum.

Problem 3
The measured value of OCR or ECAR is small or lower than the sensitivity (step 45).

Potential solution
Check the viability of BMDCs.

OPEN ACCESS
Increase the number of BMDCs seeded into an XF24 cell culture microplate. Improve the condition of the BMDCs. Prepare the assay medium again.

Problem 4
ECAR does not increase after the administration of TLR agonist (step 45).

Potential solution
Increase or decrease the concentration of the TLR agonist. ECAR may not increase in DCs whose energy metabolism has shifted from mitochondrial respiration to glycolysis by the inhibitor or genetic modification. Reduce the FCS concentration to 1%-5% because FCS has some buffering capacity.  Problem 5 The value of OCR and ECAR is not stable (step 45).

Potential solution
CRITICAL: We recommend checking raw data for pO 2 and pH.
Check the levels of pO 2 and pH. When a large number of cells are seeded in a well, an artifact of OCR may be generated due to the hypoxia. If pO2 or pH decreases over time, we recommend to reduce the number of cells or increase the rest time of assay (Gerencser et al., 2009).

RESOURCE AVAILABILITY
Lead contact Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Kazuhito Gotoh (gotou.kazuhito.712@m.kyushu-u.ac.jp).

Materials availability
All reagents generated in this study are available from the lead contact with a completed Materials Transfer Agreement.

Data and code availability
The data that support the findings of this study are available from the lead contact upon reasonable requests.