Live-imaging of Mitochondrial System in Cultured Astrocytes

The present protocol has been performed with the approval of an ethical committee (agreement VD3602, Lausanne, Switzerland) and follows European guidelines for the use of animals.

1. Rat hippocampal astrocyte primary culture

1. Sacrifice five rat pups (Wistar IGS Rat) by decapitation at postnatal day 1-2.
2. Remove the brain and keep it in a Petri dish containing 5 mL of fresh astrocyte medium (DMEM with GlutaMAX supplemented with 1% Penicillin/Streptomycin and 10% fresh Horse Serum).
3. Isolate the hippocampus. Dissociate in 5 mL of the astrocytic medium by three passages through a 21 G needle and three passages through a 25 G needle.
4. Transfer the dissociated cells to a 15 mL centrifuge tube and count in a hemocytometer.
5. Plate 20,000-25,000 cells/cm² in multiwell dishes (9.4 mm x 10.7 mm x 9.3 mm) and store them at 37 °C under an atmosphere containing 5% CO₂ for the rest of the experiment.
6. Replace the medium entirely at 3 days in vitro (DIV3).
7. At DIV8, add 0.6 pg of p24 antigen per cell of lentiviral vector coding for mitochondrial biosensor MitoTimer (LV-G1-MitoTimer²²) diluted in phosphate-buffered saline (PBS + 0.01% BSA).
8. At DIV9, wash with pre-warmed 1x sterile PBS (37 °C) and add fresh astrocyte medium.
9. At DIV11, perform 2 washes with pre-warmed 1x sterile PBS (37 °C) and add fresh astrocyte medium without phenol red.
   ​NOTE: Choice of the coating depends on the intended assay. As a standard coating for astrocyte primary culture, it is recommended to use 0.2 mg/mL poly-D-lysine or 8.7 µg/cm² of basement membrane matrix (e.g., Matrigel). To visualize the mitochondrial system of astrocytes, it is essential to work on flattened and stretched cells. In this context, the combination of basement membrane matrix on IBIDI u-slides is the most suitable. It is also crucial not to work with phenol red, which is toxic for the cell under repeated exposure to light.

2. Long-term monitoring of the mitochondrial system.

1. Assess the astrocytic mitochondrial system a minimum of 3-5 days after the lentiviral infections with LV-G1-MitoTimer (to allow a sufficient level of fluorescence in the mitochondria).
2. Image the cells with an inverted microscope controlled by the acquisition and analysis software and a module for full automation of acquisition.
   NOTE: Refer to the Table of Materials for the details of the software and module used in this study.
3. Ensure that the microscope is equipped with a cage incubator (see Table of Materials) to maintain astrocyte culture at 5% CO₂ and 37 °C throughout the experiment.
4. Capture fluorescence images using sequential excitation at 490 nm (for green channel) and 550 nm (for red channel) with detection of green (500-540 nm for green channel) and red (550-600 nm for red channel) fluorescence signals.
5. Select 5 astrocytes per well with a mitochondrial network expressing sufficient levels of LV-G1-MitoTimer using a magnification of 40x. Take care to select astrocytes as flat and large as possible and not located in clusters of cells (to work on individual cells).
6. Save the coordinates of the 5 selected cells in an xlm file (map.xlm). This map.xlm allows the user to return to the same cells over time.
7. Acquire image sequences (1 image/s for 60 s) using a magnification of 150x (100x oil immersion objective, 1.5x intermediate magnification) for each coordinates.
8. Repeat image acquisition (1 image/s for 60 s) for the same astrocytes 6 h, 12 h, and 24 h after treatments.
   ​NOTE: Use a microscope equipped with a fast and efficient autofocus system. In this context, the hardware solution called Perfect Focus System (PFS) was used. PFS uses near-infrared 870-nm LED and CCD line sensors to combat axial focus fluctuations in real-time during long-term imaging investigations.

3. Analysis of individual mitochondrial morphology and LV-G1-MitoTimer ratio

NOTE: NIS General Analysis 3 (GA3) from Nikon was used to automatize morphometric analysis in this study.

1. For each image sequence (baseline, 6 h, 12 h, 24 h), select the first frame for red and green channels by clicking on ND Processing > Select Frame.
2. Merge the red and the green channels by selecting Conversions > Merge Channel.
3. To correct image shading, select Preprocessing > Auto Shading Correction.
4. Apply the Rolling Ball algorithm by selecting Preprocessing > Rolling Ball.
5. To generate binary masks for each mitochondrion, select Segmentation > Threshold.
6. Remove any objects truncated by the border by selecting Binary processing > Touching Border.
7. To measure the surface area, select Measurement > Object Area.
8. To measure the diameter, select Measurement > Eq Diameter.
9. To measure the length, select Measurement > Length.
10. To measure the width, select Measurement > Width.
11. To measure roughness, select Measurement > Roughness.
12. To measure circularity, select Measurement > Circularity.
13. To measure elongation, select Measurement > Elongation.
14. Compose a group (right-click) with the above measurement and rename it as MorphoData.
15. Measure mean green intensity by selecting Measurement > Mean Intensity.
16. Measure mean red intensity by selecting Measurement > Mean Intensity.
17. To measure the Red/Green ratio, select Measurement > Ratio.
18. Compose a group (right-click) with the above measurement and rename it as RatioData.
19. Export the table to a CSV file by selecting Reference > Table to CSV.
20. Save the GA3 script of analysis by selecting Save As
    NOTE: A non-exhaustive list of the criteria used in this procedure is summarized in Figure 1, Table 1, and Table 2. The analysis GA3 script file is available in supplemental (Supplemental Coding File 1 and Figure 2). The thresholds used were adapted to individualize as many mitochondria as possible (excluding large mitochondrial networks where possible). Wherever possible, mitochondrial networks are either individualized manually or excluded from the analyses. Due to intercellular variability, perform these analyses on at least 20-25 cells per condition (with a minimum of 50 mitochondria by cell).

4. Analysis of mitochondrial motility

NOTE: Due to the high complexity of mitochondrial movements, manual motility analysis is preferred. Here, Nikon's NIS Element system was used to manually track mitochondria.

1. Open the tracking module in NIS, select View > Analysis > Tracking.
2. Click on Define New ROI.
3. With the Automatic Detection Tool, select 25-50 mitochondria on the first image of the image sequence.
4. Click on Track Autodetected ROIs Analyze.
5. If necessary, delete the incorrect ROI tracks.
6. Export the table to a CSV file.
   ​NOTE: A non-exhaustive list of the criteria used in this procedure is summarized in Figure 1 and Table 3. A tracking analysis generally gives a path depicting the movements of the center of every tracked object. The different tracking options must be set in the tracking option. Favor the selection of isolated mitochondria that are sufficiently distant from each other to facilitate tracking. Keep only the objects that could consistently be tracked over the whole sequence. Proceed the same way to remove outliers due to bad quality tracking. Consequently, the set of objects analyzed in this section is different from the one analyzed for static morphological analyses and will generally be smaller.

5. Data transformation, normalization, and statistical analysis

NOTE: Mainly due to the high heterogeneity of the mitochondria, data generated often have a non-normal distribution.

1. Manually log-transform the measurements prior to processing.
2. For each analysis and time frame, manually normalize the resulting data by the mean of those obtained in the reference acquisition.
3. In addition, consider a second/double normalization to a control condition, for example, to evaluate the effect of a treatment.
4. Then, perform statistical analysis using a two-way matched ANOVA. Here GraphPad Prism V8 was used.