The permeabilized cells and muscle fibres technique allows one to study the functional properties of mitochondria without their isolation, thus preserving all of the contacts with cellular structures, mostly the cytoskeleton, to study the whole mitochondrial population in the cell in their natural surroundings and it is increasingly being used in both experimental and clinical studies. The functional parameters (affinity for ADP in regulation of respiration) of mitochondria in permeabilized myocytes or myocardial fibres are very different from those in isolated mitochondria in vitro. In the present study, we have analysed the data showing the dependence of this parameter upon the muscle contractile state. Most remarkable is the effect of recently described Ca2+-independent contraction of permeabilized muscle fibres induced by elevated temperatures (30–37°C). We show that very similar strong spontaneous Ca2+-independent contraction can be produced by proteolytic treatment of permeabilized muscle fibres that result in a disorganization of mitochondrial arrangement, leading to a significant increase in affinity for ADP. These data show that Ca2+-insensitive contraction may be related to the destruction of cytoskeleton structures by intracellular proteases. Therefore the use of their inhibitors is strongly advised at the permeabilization step with careful washing of fibres or cells afterwards. A possible physiologically relevant relationship between Ca2+-regulated ATP-dependent contraction and mitochondrial functional parameters is also discussed.
With great interest, we have read the recent paper in the Biochemical Journal by Christopher Perry et al.  entitled ‘Inhibiting myosin-ATPase reveals a dynamic range of mitochondrial respiratory control in skeletal muscle’. In this paper, the authors used the PmFBs (permeabilized fibre bundles) technique for studies of regulation of respiration both in skeletal muscle of rat (either red or white gastrocnemius) and biopsy samples taken from human skeletal muscle (vastus lateralis) . Since we were among the first to suggest and widely use PmFBs to study delicate mechanisms of energy metabolism and energy transfer in various types of muscles  and have gained a lot of experience in this field, in the present paper, we make several comments on this paper. Importantly, the permeabilized cells and fibres technique allows one to study the functional properties of mitochondria without their isolation, thus preserving all of the interactions with other cellular structures intact (mostly with the cytoskeleton and endoplasmic reticulum) and to study the whole population of mitochondria in the cell in their natural surroundings [2,3]. Another advantage is that this approach allows the analysis of very small limited tissue samples (a few milligrams) such as in human biopsies. This is why the PmFBs technique is increasingly used in many experimental and clinical studies [1–5]. Perry et al.  have made a detailed and interesting study of the influence of several factors, such as myosin ATPase inhibition, temperature, oxygenation and contraction on the functional parameters of mitochondrial oxidative phosphorylation, the apparent Km for ADP in regulation of respiration and maximal respiration rates . The most exiting finding of this study is the important role of fibre contraction in all of these processes, which was found to be rather mysterious for both authors and readers. Very strong contraction was induced simply by placing fibres into solution which contained a high concentration of EGTA and neither Ca2+ nor ATP added, at 30°C and at higher temperatures . Thus the muscle contraction was independent of Ca2+ even if there were trace amounts of adenine nucleotides left in the cells. This strong shortening, a supercontraction, was rather rapid and significantly decreased the apparent Km values for ADP .
It is most interesting to understand what is behind this unusual, but very strong, muscle shortening excluding Ca2+-regulated actomyosin contraction cycle. In the present study, we have made one observation that could help to solve the mystery of Ca2+-insensitive contraction seen by Perry et al.  and even explain it, as described below.
MATERIALS AND METHODS
Cardiomyocytes were isolated as described in [6,7]. The permeabilized fibres were prepared from rat soleus muscle as described in . The dissection of the muscle strips immersed in ice-cold solution A  was carried out under a microscope or magnifying glass, through diverse smooth horizontal moves in order to obtain thin muscle fibres bundles. It is important to take into account that this procedure has to be carried out at 4°C and under a microscope using a cold light source [optical fibres or LED (light-emitting diode)] to avoid damage to the muscle fibres due to heat. Another important point is dissection time, because spending more than 20 min to obtain the fibre bundles could trigger hypoxic effects, damaging mitochondrial oxidative phosphorylation. The permeabilization of the fibre bundles was performed in a vial containing 3 ml of solution A (supplemented with 2 mg·ml−1 BSA) and 50 μg·ml−1 saponin (in some experiments, 100 μg·ml−1 was used) and shaking gently at 4°C for 30 min. After this step, permeabilized fibres were transferred to a tube containing Mitomed  (supplemented with 2 mg·ml−1 BSA and 1 μM leupeptin) and were shaken gently at 4°C for 10 min to wash out saponin, other metabolites, especially traces of ADP or ATP, and proteases released during fibre preparation. This washing procedure was repeated three times. Mitochondrial respiration of permeabilized cells was measured in Mitomed solution in the presence of leupeptin .
Mitochondria were visualized with MitoTracker® Green or with TMRM (tetramethylrhodamine methyl ester) as described in . The digital images of TMRM and MitoTracker® Green fluorescence were acquired with an inverted confocal microscope (Leica DM IRE2) with a ×63 water-immersion lens. The MitoTracker® Green fluorescence was excited with a 488 nm argon laser, using 510–550 nm for emission. TMRM fluorescence was measured using 543 nm for excitation (helium–neon laser) and at least 580 nm for emission.
RESULTS AND DISCUSSION
We have been using the permeabilized cells technique for several decades [2,4,5], and even studied directly the effect of contraction on the parameters of respiration regulation, but that contraction was regulated by Ca2+, dependent on ATP and totally reversible, as normal contraction should be . Supercontraction of permeabilized cardiomyocytes, when their length was shortened by a factor of approximately 3 (1–3 μM Ca2+ and 1 mM ATP) resulted in a decrease in apparent Km for ADP from 320±20 μM to 17±3 μM, and the ordered crystal-like arrangement of mitochondria was lost . A return to 0.1 μM Ca2+ restored both the initial length and high apparent Km for ADP (to an average of 300 μM) . No effect of Ca2+ on the high value of apparent Km for ADP (approximately 350 μM) was seen in ghost cardiomyocytes from which myosin was removed by treatment with KCl . Thus the high value of apparent Km for ADP is a parameter characteristic of relaxed cardiomyocytes. In the regular relaxing solution without Ca2+, which is traditionally used in multiple studies, the apparent Km for ADP was always equally high both in permeabilized cardiomyocytes and fibres [4,5]. This is in agreement with the fact that no shortening was usually observed, when very many preparations of cardiomyocytes permeabilized by saponin were visualized by confocal microscopy (one example is shown in Figures 1A and 1B) at room temperature (22–25°C) even for a long duration [4–11]. Similar results were obtained in many other laboratories (see references in ).
Proteolytic treatment induces strong contraction and change of shape of permeabilized cardiomyocytes
Nevertheless, in the present study, we have made one observation that could help to solve the mystery of Ca2+-insensitive contraction seen by Perry et al.  and even explain it. Most probably, it is the collapse of the cytoskeleton that makes fibres become rapidly shorter in their study. We have taken advantage of the situation to present again our data recorded 7 years ago and even published (Figures 2 and 4 in reference ), and they are also mentioned briefly in another of our publications . This surprising phenomenon, i.e. Ca2+-independent contraction, has already been seen in experiments with the proteolytic treatment of permeabilized cardiomyocytes and is shown in Figure 1(C). In Supplementary Movie S1 (at http://www.BiochemJ.org/bj/445/bj4450333add.htm), one can see this spectacular phenomenon: in the absence of Ca2+ and ATP, the addition of trypsin at a 1 μM concentration results in a sudden shortening of permeabilized cardiomyocytes to approximately one-tenth of the initial length(!), and then a surprising spontaneous increase again in size, but with a chaotic mitochondrial arrangement, but the initial regular rod-like cell shape and the intracellular structural organization are lost (Figure 1C). Interestingly, mitochondria remain attached to some cellular structures. This amazing effect shows that the rod-like shape of cardiomyocytes and very regular crystal-like arrangement of mitochondria are maintained by multiple interactions and forces within cellular cytoskeletal and sarcomere structures which are destroyed by proteolysis. Very sensitive to the action of proteolytic enzymes are tubulins, the microtubular system, also plectin, titin and Z-line-associated proteins and finally myosin itself [9–11].
Quality test of the permeabilized soleus muscle fibres
Interestingly, the Ca2+-induced ATP-dependent supercontraction also decreased the effect of trypsin on the apparent Km for ADP , obviously by hindering the access of trypsin molecules to some important cytoskeletal proteins. A decrease in apparent Km for ADP may mean that strong contraction of myofibrils evidently deforms the intermyofibrillar mitochondria, probably resulting in the dissociation of some important cytoskeletal proteins (for example tubulin) from the mitochondrial outer membrane and increasing the VDAC (voltage-dependent anion channel) permeability [4,5]. The influence of myofibrillar contraction on the shape of mitochondria has also been recorded by Yaniv et al. .
The connection between contractile state and mitochondrial functional parameters may indeed be important, as supposed both by Yaniv et al.  and Perry et al. . Under physiological conditions, an increase in the heart muscle contractile force is caused by increased ventricle filling, stretch of cardiomyocytes, increase in sarcomere length and decrease in the lattice spacing in myofibrils (length-dependent activation, the Frank–Starling law of the heart) [13,14]. The results described above allow us to suppose that this may even increase the apparent Km for ADP and thus increase the flux of phosphocreatine into cytoplasm due to increased recycling of ADP in mitochondria coupled to creatine phosphorylation [4,5]. In skeletal muscle, the effect of contraction on mitochondria may depend on whether the contraction is isotonic or isometric.
The effects of proteolytic enzymes described above made it necessary to use their inhibitors during cell permeabilization to protect the cell cytoskeleton from the eventual action of intracellular proteases which may be released or activated by detergents, and to perform multiple steps of washing before using the permeabilized cells for functional measurements. According to our experience in testing various inhibitors, the combination of STI (soya bean trypsin inhibitor) and leupeptin gave the best results. ‘Double-headed’ [15,16] STI could simultaneously inhibit both trypsin- and chymotrypsin-type serine proteinases, whereas leupeptin efficiently inhibits a number of serine proteinases and threonine proteinases, as well as thiol proteinases , which originate from lysosomes or the proteasome (cf. [17–19]).
Not only biochemical, but also mechanical, damage can be a reason for variable Km and contracted fibres with collapsed cytoskeleton. The scissors or forceps used must be extremely sharp and antimagnetic to prevent mechanical damage of the fibres. Incorrect dissection of fibres, too heavy-handed treatment and vigorous stirring during washing and permeabilization are frequently the causes of damage to the cytoskeletal structures, intracellular position, and organization and surrounding of mitochondria within the cell. Moreover, insufficient mechanical dissection and permeabilization can lead to insufficient quality of the preparation, and thus to low maximal State 3 mitochondrial respiration, low RCR (respiratory control ratio) and low accessibility of mitochondria to various agents, inhibitors, etc. . To prevent this situation, it is recommended to use approximately 10 mm long and 100–250 μm thick bundles (each containing between two and five single fibres). The quality of fibres prepared could be evaluated before the experiments by using an oxygraphic quality test . The RCR (the ratio of respiration rates in States 3 and 2) with ADP (2 mM) was higher than 5 and a cytochrome c effect  was not seen. Among these tests, the most sensitive is the PK (pyruvate kinase)/PEP (phosphoenolpyruvate)/creatine test shown in Figure 2.
These necessary steps are included in the latest versions of experimental protocols that we recommend [2,11]. Also, we recommend using for oxygraphic measurements the Mitomed solution of the following composition: 110 mM sucrose, 60 mM potassium lactobionate, 0.5 mM EGTA, 3 mM MgCl2, 0.5 mM dithiothreitol, 20 mM taurine, 3 mM KH2PO4 and 20 mM potassium Hepes (pH 7.1), with 5 mM glutamate, 2 mM malate and 2 mg/ml essentially fatty-acid-free BSA; this solution contains lactobionate and taurine known to have a membrane-stabilizing action [2,7]. In this solution, the permeabilized cells and fibres are more stable than in salt solutions used previously.
Also, all reagents have to be of a high degree of purity. The use of Mitomed solution allows the avoidance of using high EGTA concentrations and serious problems related to the contaminations in the commercial preparations of EGTA .
Accounting for all of these important details, the collapse of the cytoskeleton, the mysterious contraction and fibre inactivation should be avoided.
Andrey Kuznetsov and Rita Guzun carried out the confocal microscopic studies. Rita Guzun, Rafaela Bagur and Tuuli Kaambre carried out isolation of cardiomyocytes, preparation of fibres and oxygraphic studies. François Boucher and Valdur Saks participated in the planning of the study and writing the paper.
This work was supported by project SYBECAR from Agence de la Recherche Nationale, France, the Estonian Science Foundation [grant numbers 7823 and 8987], the Estonia Ministry of Education and Science [grant number SF0180114Bs08] and a research grant from the Austrian Science Fund (FWF) [grant number P 22080-B20].