Preparation of mitochondria to measure superoxide flashes in angiosperm flowers

Plant materials/plant growth

M. denudata was grown on the campus of Beijing Forestry University (40°00′02″ N, 116°20′15″, a.s.l., 60 m). Pistils and petals of 15 flowers were collected during afternoons in March and April. N. nucifera was grown in Bajia Country Park (40°00′50″N, 116°19′39″E, a.s.l., 47 m). Receptacles and petals of nearly 10 flowers were collected during afternoons in July–August.

Solutions

Method A: Grinding buffer: 0.3 M sucrose, 25 mM Na₄P₂O₄, 2 mM EDTA, 10 mM KH₂PO₄, 1% (w/v) polyvinylpyrrolidone-40, 1% (w/v) defatted bovine serum albumin (BSA), 4 mM cysteine, and 20 mM ascorbic acid were added just prior to grinding. pH was adjusted to 7.5 with KOH. Resuspension buffer: 0.3 M sucrose, 10 mM N-Tris [hydroxymethyl]-methyl-2-aminoethanesulfonic acid (TES-KOH), and 0.1% BSA, pH = 7.5. Mitochondrial basic incubation medium: 0.3 M sucrose, 10 mM TES-KOH. 10 mM NaCl, 5 mM KH₂PO₄, 2 mM MgSO₄, and 0.1% BSA, pH = 7.2.

Method B: Grinding buffer: 0.3 M sucrose, 25 mM Na₄P₂O₄, 2 mM EDTA, 10 mM KH₂PO₄, 1% (w/v) polyvinylpyrrolidone-40, 1% (w/v) defatted bovine serum albumin (BSA) and 20 mM ascorbic acid were added just prior to grinding. pH was adjusted to 7.5 with KOH. Resuspension buffer: 0.3 M sucrose and 10mM TES-KOH, pH = 7.5. Preparation of a single linear PVP-40 gradient in 28% (v/v) Percoll: 0.3 M sucros, 10 mM KH₂PO₄, 0.1% BSA, 28% (v/v) Percoll and a linear gradient of 0–10% (w/v) PVP-40 (top to bottom) in a 30 ml centrifuge tube. pH = 7.2. Mitochondrial basic incubation medium: 0.3 M sucrose, 10 mM TES-KOH. 10 mM NaCl, 5 mM KH₂PO₄, 2 mM MgSO₄, and 0.1% BSA, pH = 7.2.

Isolation of mitochondria

Method A: Our efficiency method to obtain crude, high viability mitochondria. All steps were carried at 4 °C on ice. Mitochondria of magnolia were isolated from style and petal tissues while mitochondria of lotus were isolated from receptacle and petal tissues. About 0.5 g of pistil or petal tissues were cut up from each species into 1 mm³-fragments with scissors. They were ground in 1–2 ml of grinding buffer using a pestle with a small amount of quartz. The extract was filtered through 20 µm nylon mesh and then centrifuged at 2,000 × g for 10 min to remove most of the thylakoid membranes and intact chloroplasts. The supernatant was transferred to a new tube and centrifuged at 12,000 × g for 20 min. The pellet was resuspended in 1 ml resuspension buffer and centrifuged for 5 min at 1,500 × g to remove the residual intact chloroplasts. This new supernatant was centrifuged for 20 min at 12,000 × g to yield the crude mitochondria. The crude mitochondria were suspended in mitochondrial basic incubation medium and placed on ice for further studies.

Method B: to obtain purified mitochondria as previous study. All steps were carried at 4 °C on ice. Mitochondria were isolated from style of magnolia and receptacle of lotus. About 45 g of pistil tissues were cut up with scissors (see above). Isolation of mitochondria was based on the method of Sweetlove, Taylor & Leaver (2007) with minor modification. Briefly, pistil tissues were blended in 200 ml grinding buffer (see above), filtered through 20 µm nylon mesh and then centrifuged at 1,100 × g for 10 min. The supernatant was centrifuged at 18,000 × g for 20 min. The pellet was resuspended in 30 ml resuspension buffer and centrifuged for 10 min at 1,100 × g. This new supernatant was centrifuged for another 20 min at 18,000× g. The final pellet was suspended in 1 ml resuspension buffer and layered over the linear PVP-40 gradient in 28% (v/v) Percoll (see above), and centrifuged for 45 min at 40,000 × g. The mitochondria were found in a tight white band near the bottom of tube. The mitochondria fraction was carefully removed and resuspended in 20 ml resuspension buffer, the suspension was centrifuged for 20 min at 15,000 × g. The purified mitochondria were suspended in mitochondrial basic incubation medium and place on ice for further studies.

Mitochondrial respiratory function assay

The oxygen consumption rates of mitochondria were determined with a Clark-type oxygen electrode (Strathkelvin 782 2-Channel Oxygen System version 1.0; Strathkelvin Instruments, Motherwell, UK) at 25 °C. A 10 µl aliquot of mitochondrial suspension was blended in 1 ml of mitochondrial basic incubation medium. The oxygen sensor signal was recorded on a computer at intervals of 0.5 s with Strathkelvin Instruments software (782 System version 1.0). Oxygen consumption was measured with 250 µM ADP (state 3) and with 5 mM succinate (state 4). The respiratory control ratio (RCR) was calculated as the ratio of state 3 to state 4 respiration. The mitochondrial suspensions with higher than a state 3 RCR were used in subsequent studies.

Confocal imaging of Mito-ROS and ΔΨ_m

To visualize the superoxide flashes and ΔΨ_m, isolated mitochondria from pistil and petal tissues of magnolia and lotus were immobilized on round glass cover slides (pretreatment with 0.2 mg/ml poly-L-lysine for 1 h; Sigma, St. Louis, MO, USA) by centrifugation at 2,000 × g for 5 min at 4 °C and mounted on an inverted microscope (Zeiss LSM 710; Carl Zeiss, Oberkochen, Germany) for imaging. To measure the subcellular locations of mitochondria and ROS, mitochondria were first incubated with 100 nM MitoTracker Green (Invitrogen, Carlsbad, CA, USA) for 30 min at 25 °C and washed in mitochondrial basic incubation medium, then loaded with 2.5 µM MitoSOX Red for 5 min. MitoTracker Green was excited with 488 nm and emissions were collected at 500–530 nm, while MitoSOX Red was excited with 543 nm and collected at an emission wavelength of 560–620 nm. Isolated mitochondria were labelled with 2.5 µM MitoSOX Red and 5 mM succinate as a respiration substrate to measure superoxide flashes. To understand the MitoSOX-flashes behavior in the change of respiration state and uncoupler, 0.25 mM ADP and 5µM FCCP (Carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone) was added to observe the superoxide flashes. Isolated mitochondria were loaded with 50 nM TMRM and 5 mM succinate for 1 min at 25 °C to measure the ΔΨ_m. The excitation wavelength for TMRM was 543 and the emission wavelength was 550–620 nm. A total of 100 frames of 512 × 512 pixels were collected for a typical time-series recording. The frame rate was 50–60 frames/min. All experiments were performed at room temperature (24–26 °C).

Data analysis

The images obtained by laser scanning confocal microscopy were analyzed using Image J 1.48v (Wayne Rasband, National Institutes of Health, Bethesda, MD, USA). Superoxide flashes and variations in the ΔΨ_m were identified using FlashSniper (Li et al., 2012), and their morphological, properties, and duration were measured automatically. Statistical analyses were performed using SPSS Statistics 23.0 software (IBM Corp., Armonk, NY, USA). Images were processed and assembled using Adobe Photoshop CS 5 (Adobe Systems Corp., San Jose, CA, USA).

Respiratory function and viability of isolated mitochondria

Crude mitochondria were sampled from petal and style tissues of magnolia as shown in Fig. 1A, while mitochondria from petal and receptacle tissues of lotus were sampled as shown in Fig. 1F. A signal with excitation at 488 nm was confirmed to avoid the disturbing auto-fluorescence of intact chloroplasts. As shown in Fig. 1B and Fig. 1G, no intact chloroplasts were detected in the crude isolated mitochondria. To compare the previous method (method B) (Day, Neuburger & Douce, 1985) and our efficient method (method A) to isolate mitochondria, the respiratory function of the isolated mitochondria was determined with a Clark-type oxygen electrode. As a result, the RCR did not change significantly in mitochondria isolated from flowers using method A (n = 6), but RCR declined in isolated mitochondria using method B (n = 6) (Table 1). As the viability of mitochondria is reflected by the ΔΨ_m, crude isolated mitochondria were loaded with the TMRM indicator. Highly viable and highly dense mitochondria were observed in mitochondria of magnolia (Figs. 1C, 1D) and lotus (Figs. 1H , 1I). The viability of mitochondria using method B was lower than that of method A (Figs. 1E, 1J). We assessed the time consumed, amount of sample consumed, and the viability of both methods. Using method B, mitochondria were processed in 5.28 ± 0.23 h and consumed 43.92 ± 3.78 g of flower tissues (n = 6), whereas mitochondria were isolated within 1.13 ± 0.14 h with only 0.47 ± 0.12 g tissues (n = 6) using our method A. This result shows that our mitochondrial isolation method was highly efficient to obtain highly viable mitochondria in the flower species.

ROS production in floral mitochondria

To identify the intracellular site of ROS production, mitochondrial ROS were loaded with MitoSOX Red for 5 min (2.5 µM), while mitochondria were loaded with Mito Tracker Green for 30 min (100 nM). ROS production and mitochondrial location were coincident in the mitochondria isolated from petals and styles of magnolia, suggesting that mitochondria are the primary site of ROS production in this species (Figs. 2C, 2G). The same results were found in the mitochondria isolated from receptacle and petal of lotus (Figs. 2J, 2N).

The fluorescent level of ROS increased significantly in the mitochondria isolated from style compared to the petal of magnolia (Figs. 2B, 2D, 2F) (n = 100). In addition, similar results were found in the isolated mitochondria of lotus, as the ROS level was significantly higher in the receptacle than in the petal (Figs. 2I, 2K, 2M) (n = 100). Our results confirm that mitochondrial ROS tended to accumulate in the pistil of both magnolia and lotus, indicating that mitochondrial ROS might be more involved in the electron transport chain in the pistil than in the petal.

Superoxide flashes in flowers

To investigate the nature of superoxide flashes in magnolia and lotus, isolated mitochondria were loaded with the ROS fluorescent probe MitoSOX Red with 5 mM succinate added as respiratory substrate. According to a previous study (Wang et al., 2016b), we defined the variation of fluorescence at df/F₀ > 0.2 within 10 s as a single superoxide flash event. A transient increase in MitoSOX fluorescence and variations in the trace were observed during 100 s in single mitochondrial events (Fig. 3A, B, and Video S1). Among these instantaneous traces, three types of mitochondrial superoxide traces were classified (Figs. 3D–3F): low variation slope traces (0.2 < df/F₀ < 0.5) (Fig. 3D), high variation slope traces (0.5 ≤ df/F₀) (Fig. 3E), and multi-event traces (0.2 < df/F₀) (Fig. 3F). We also compared the frequency of superoxide flashes (/100s ×1, 000 µm²) in mitochondria isolated from petals and pistils of magnolia and lotus. Notably, superoxide oxide flashes labelled with MitoSOX Red were detected at a rate of 129.18 ± 20.11 (/100 s ×1, 000 µm², n = 13) in mitochondria isolated from the magnolia style (Fig. 3G) which was significantly higher than mitochondria in the petal (75.23 ± 10.48/100 s ×1,000 µm², n = 11). In lotus (Fig. 3H), the rate of superoxide flashes was 48.24 ± 10.24 (/100 s ×1, 000 µm², n = 10) in mitochondria isolated from the style, which was also significantly higher than mitochondria in the petal (25.68 ± 4.79 /100 s ×1,000 µm², n = 10). These results indicate that superoxide flashes, together with ROS bursts, are highly autonomous and predominantly reflect the properties and physical activities of mitochondria in different tissues and species.

The MitoSOX-flashes were closely linked to functional ETC and respiratory activity. To observe the mitochondrial behavior in the change of respiration state and uncoupler, 0.25 mM ADP and 5 µM FCCP was added for the measurement of mitochondria isolated from style of Magnolia. The addition of ADP resulted in a significantly decrease in MitoSOX fluorescence (Figs. 4B, 4D) (n = 100). Mitochondrial respiratory uncoupled by FCCP also led to the significantly decrease in MitoSOX signal (Figs. 4C, 4D) (n = 100). Similar behavior was found in the measurement of MitoSOX-flashes, the rate of superoxide flashes with succinate was significantly higher than mitochondria in the addition of ADP and FCCP (Fig. 4E) (n = 6). Low flashes occurred in mitochondria upon uncoupling suggested that the electrical transmembrane gradient might modulate the superoxide flashes.

Depolarization of the mitochondrial membrane potential in flowers

To study variations in the ΔΨ_m, isolated mitochondria were labelled with TMRM, and 5 mM of succinate was added. The decline in fluorescent intensity at df/F₀ <−0.2 was defined as an event. Transient depolarization of the ΔΨm accompanied by later polarization occurred in a single mitochondrion (Fig. 5A, B and Video S2). According to the wide variation in ΔΨm, the trace ΔΨm was catalogued into three types (Figs. 5D–5F): Instantaneous loss of ΔΨm along with instant recovery (Fig. 5D), instantaneous loss of ΔΨm with a short period of stability before recovery (Fig. 5E), and multi-event trace including the above two types (Fig. 5F). The frequency of a TMRM-event in mitochondria isolated from magnolia petals (544.92  ± 56.98/100 s ×1,000 µm², n = 15) was significantly lower than the values in the style (1,009.10 ±130.10 /100 s ×1,000 µm², n = 15) (Fig. 5G). The same result was found in lotus (Fig. 5H) that the frequency of TMRM events in mitochondria isolated from the lotus petal was 51.94 ± 10.57 (/100 s ×1,000 µm², n = 10) which was lower than that in the receptacle (119.99 ± 19.00/100 s ×1,000 µm², n = 10). We conclude that transient and spontaneous depolarization of ΔΨm occurred in all tissues and the higher frequency of variation of ΔΨm in the pistils of flowers suggest that they have a higher level of mitochondrial dynamics.