Deterioration of Rat -Liver Mitochondria during Isopycnic Centrifugation in an Isoosmotic Medium

Eur. J. Biochem. 51, 603-608 (1975) Deterioration of Rat -Liver Mitochondria during Isopycnic Centrifugation in an Isoosmotic Medium Michkle COLLOT, Simone WATTIAUX-DE CONINCK, and Robert WATTIAUX Laboratoire de Chimie Physiologique, Facultes Universitaires Notre-Dame de la Paix, Namur (Received August 26/0ctober 9, 1974) We have investigated the effect of the centrifugation speed on the behavior of rat-liver mitochondria during isopycnic centrifugation in an isoosmotic medium. The gradient was made with a macromolecular compound, glycogen dissolved in 0.25 M aqueous sucrose. The distribution curves of several mitochondrial enzymes change when the centrifugation reaches a certain speed : they are shifted toward regions of lower density. The results are plausibly explained by supposing that the inner mitochondrial membrane becomes permeable to sucrose at high centrifugation speeds, and that the granules swell. The main causal agent of the phenomenon is the hydrostatic pressure the mitochondria are subjected to during centrifugation. Morphological observations show that mitochondria are markedly deteriorated when centrifuged at high speed in the glycogen gradient; they are swollen and the outer membrane is broken; also frequently, a large electron-dense granule is seen in the matrix near the inner membrane. Rat-liver mitochondria are severely deteriorated when centrifuged in a sucrose gradient under too high a hydrostatic pressure [l]. In such experiments, the mitochondria are constantly exposed to a hypertonic medium that could affect their structure. Therefore, we have investigated whether deterioration of these organelles would also occur when the centrifugation is performed in an isoosmotic medium. We made use of a kind of density gradient like that described by Beaufay et al. [2]; in this gradient, the density variations are ensured by varying the concentration of a macromolecule, glycogen, the solvent being aqueous 0.25 M sucrose. In these conditions, the granules are constantly kept in an isoosmotic medium during the centrifugation. The behavior of the mitochondria has been assessed by establishing the distribution of reference enzymes selected for their different submitochondrial locations as described in our previous papers [I, 3,4]. Enzymes. Ferro cytochrome c : oxygen oxidoreductase or cytochrome oxidase (EC 1.9.3.1); amine : oxygen oxidoreductase (deaminating) (flavin-containing) or monoamine oxidase (EC 1.4.3.4); sulfite : oxygen oxidoreductase or sulfite cytochrome c reductase (EC 1.8.3.1); L-malate : NAD oxidoreductase or malate dehydrogenase (EC 1.1.1.37). MATERIALS AND METHODS Centrifugation Experiments All experiments were performed on the mitochondrial fraction from rat-liver corresponding to the sum of fractions M and L of de Duve et al. [5]. Density gradient centrifugation was carried out according to a method similar to that of Beaufay et al. [2] with the equipment described by de Duve et al. [6]. In all experiments, granule preparations were layered above the gradient. The gradient solutions were made by dissolving glycogen in 0.25 M sucrose, the ph being adjusted near neutrality by addition of NaOH. Some additional experiments details will be given in the legends of the figures. Enzyme Assays Cytochrome oxidase was assayed by the method of Appelmans et al. [7]. Monoamine oxidase was measured by the procedure of Schnaitman et al. [S]. Malate dehydrogenase was assayed spectrophotometrically at 340 nm and 25 "C in a medium containing 25 mm Tris buffer ph 7.4, 0.15 mm NADPH, 0.1 %

604 Isopycnic Centrifugation of Rat-Liver Mitochondria Triton X-100 and 0.25 mm oxalacetate. Sulfite cytochrome c reductase was measured following the method of Wattiaux-De Coninck and Wattiaux [9]. Proteins were measured according to Lowry et al. [lo]. Morphological Examinat ions Morphological examinations were performed as previously described according to a procedure similar to that of Baudhuin et al. [ll]. RESULTS Fig. 1 shows the distribution of the four mitochondrial enzymes after isopycnic centrifugation at different speeds in a glycogen gradient with 0.25 M sucrose as solvent. The time integral of the square angular velocity is 144 rad2/ns. After centrifugation at 39000 rev./min, the four enzymes are recovered in the bottom subfraction of the gradient. Such a distribution is comparable with the distribution found for cytochrome oxidase by Beaufay et al. [2] in a similar gradient (series B of these authors). At 45000 rev./min there are no significant modifications ; on the contrary when the centrifugation speed reaches 51 000 rev./min the enzyme distributions are strikingly different. Monoamine oxidase (outer membrane), cytochrome oxidase (inner membrane) and malate dehydrogenase (mitochondrial matrix) are chiefly found in the upper part of the gradient. Some activity is still recovered in the bottom subfraction; it is higher for malate dehydrogenase than for membrane enzymes. Sulfite cytochrome c reductase (intermembrane space) is distributed equally amongst the bottom subfraction and the fractions of low density. The distributions observed after centrifugation at 56000 and 65 000 rev./ min resemble those obtained after centrifugation at 51 000 rev./min. The following experiments describe the free and the total activity of the sulfite cytochrome c reductase and the malate dehydrogenase in the gradient fractions. The free activity is assayed in an isoosmotic medium, without adding Triton X-100 in the incubation mixture. It allows an appraisal of the integrity of the mitochondrial membranes. The centrifugation was performed at 39000 and 65000 rev./min. As shown by Fig. 2 the distributions of the total activities are different at the two centrifugation speeds as expected from former experiments. The free activity of the enzymes is relatively low when the mitochondrial fraction is centrifuged at 39000 rev./min; it is the highest in the bottom subfraction. After centrifugation at 65000 rev./min, the free activity of malate dehydrogenase is slightly increased, that of sulfite d 6 0-c? 0, 40- z1 C U 20- E LL 0-60. 40-20. 20 1.03 1.08 1.03 1.08 1.03 1.08 1.03 1.08 1.03 1.08 Density (g/rni) Fig. 1. Influence of the speed of centrifugation on distribution patterns of mitochondria1 enzymes. Isopycnic centrifugation of a rat-liver mitochondrial fraction in a 4 to 20 % (w/w) glycogen gradient with 0.25 M sucrose in water as solvent. Time integral of the square angular velocity 144 radz/ns. Particles suspended in 0.25 M sucrose were initially layered at the top of the gradient. (A) Centrifugation at 39000 rev./min; (B) centrifugation at 45000 rev./min; (C) centrifugation at 51000 rev./min; (D) centrifugation at 55000 rev./min; (E) centrifugation at 65000 rev./min in the SW 65 Spinco rotor of the Spinco preparative ultracentrifuge. In plotting the curves we calculated the average frequency of the components for each fraction Q/ZQ. A@, where Q represents the activity found in the fraction, CQ represent the total recovered activity, and represents the increment of density from top to bottom of the fraction. These values were plotted against density in histogram form. Unshaded blocks (0) represent to scale amount present in the bottom subfraction and to facilitate comparison an identical abscissa value has been arbitrarily chosen in all cases. (a) Cytochrome oxidase; (b) monoamine oxidase ; (c) malate dehydrogenase; (d) sulfite cytochrome c reductase (e) proteins cytochrome c reductase is strikingly higher, particularly in the bottom subfraction where the free activity of this enzyme is equal to the total activity. Hydrostatic pressure appears to be the principal cause of the deterioration of mitochondria in a sucrose

M. Collot, S. Wattiaux-De Coninck, and R. Wattiaux 605 40 t t 8 Monoarnine n oxidase - 60 P 4 0 5 40 9. Malate n dehydrogenase r 60 Sulf ite cytochrome c reductase I J 1.03 I.08 1.03 1.06 1.13 Density (g/rno Fig. 2. Influence of the speed of centrifugation on distribution of free activity of malate dehydrogenase andsulfite cytochrome c reductase. Isopycnic centrifugation of a rat-liver mitochondrial fraction in a 4 to 20% (wiw) glycogen gradient with 0.25 M sucrose in water as solvent. Integral of the square angular velocity: 144 rad2/ns. (A) Centrifugation at 39000 rev./min; (B) centrifugation at 65000 rev./min in the rotor SW 65 of the Spinco preparative ultracentrifuge. (a) Malate dehydrogenase; (b) sulfite cytochrome c reductase. Hatched area (@) free activity of the enzymes in each fraction, as a percentage of the total activity supposed equal to hundred. For explanations of the graph, see legend of Fig. 1. The free activity of malate dehydrogenase and of sulfite cytochrome c reductase in the mitochondrial preparation before centrifugation was 5.5% and 9% respectively gradient when the centrifugation speed is increased [l]. This factor plays also a determinant role when centrifugation is performed in 0.25 M sucrose as shown by the next experiment. Mitochondria were subjected to the same centrifugal field in a glycogen gradient but under two different hydrostatic pressures. The following experimental procedure was followed. A glycogen gradient of 5 ml was set up in two tubes of the Spinco rotor SW 41, these tubes having a volume of 12 ml. After layering of 0.6 ml of the mitochondrial fraction, one tube (tube A) was cut just above the granule suspension, the second tube (tube B) was filled with 5.4 ml 0.25 M sucrose. Centrifugation was then performed at 30000 rev./min during 120 min. In these conditions, the hydrostatic pressure exerted on the bottom of the tube was of about 600 kg/cm2 40 20 0 - I I I 1.03 1.07 1.11 1.03 1.07 1.ll Density (g/rnl) Fig. 3. Influence of hydrostatic pressure on distribution of mitochondrial enzymes. Isopycnic centrifugation of a rat-liver mitochondrial fraction in a 4 to 20 (w/w) glycogen gradient with 0.25 M sucrose in water as solvent. 0.6 ml of the mitochondrial fraction was layered above a glycogen gradient of 5 ml in two tubes of the Spinco rotor SW 41, (A) and (B). After layering tube (A) was cut just above the granule suspension and tube (B) was filled with 5.4 ml 0.25 M sucrose, centrifugation was then performed at 30000 rev./min during 120 min. During that run, the hydrostatic pressure exerted on the bottom of the tube was about 600 kg/cm2 for the tube (A) and 900 kg/cm2 for the tube (B). For explanation of the graph, see legend of Fig. 1 for the tube A and 900 kg/cm2 for the tube B. The effect of hydrostatic pressure is evident (Fig. 3). After centrifugation, in tube A, the mitochondrial enzymes are chiefly recovered at the bottom of the gradient; after centrifugation in tube B, one observes a shift of the distribution toward a region of low density and a certain dissociation between the distribution of sulfite cytochrome c reductase and those of the three other mitochondrial enzymes. The change produced by increasing the hydrostatic pressure at constant centrifugation speeds is similar to that seen when centrifugation speed is raised. In another experi- Eur. J. Biochem. 5f (1975)

606 Isopycnic Centrifugation of Rat-Liver Mitochondria ment, the mitochondrial fraction was homogeneously distributed throughout the whole gradient before centrifugation according to Beaufay et al. [2]. Then, a procedure similar to that described in a previous publication [l] was followed. A first centrifugation was performed at 39000 rev./min (Spinco rotor SW 65, Jco2 dt = 40 rad2/ns), to eliminate the mitochondria from the upper part of the gradient. Next, the centrifuge was stopped; one of the tubes (tube A) was cut at 45 % of the height of the liquid column and the sucrose-glycogen solution removed ; the second tube (B) remained unchanged. A second run was performed at 65000 rev./min, the time integral of the square angular velocity being 144 rad2/ns. Fractions were then collected and analysed. Results were similar to that of the preceding experiment. The mitochondrial enzymes are mainly recovered at the bottom of the gradient in tube A and in the upper part of the gradient in tube B. Electron micrographs were taken of the mitochondria before and after centrifugation at 39000 and 65000 rev./min in a glycogen gradient with 0.25 M sucrose as solvent. Fig. 4 shows the morphological aspect of the preparation. Mitochondria suspended in 0.25 M sucrose are mostly in a condensed form with apparently intact double membrane. When glycogen is present in the suspension medium, more swollen mitochondria are apparent in the preparation. After centrifugation at 39000 rev./min, the major part of the mitochondria are swollen; after centrifugation at 65 000 rev./min, all the mitochondria seem strikingly enlarged and their outer membrane is broken. A dark body is apparent in the granule matrix, frequently near the inner membrane. DISCUSSION Our results show that the distribution of mitochondrial enzymes after isopycnic centrifugation in an isoosmotic medium, depends on the speed of centrifugation. The distribution of the major part of cytochrome oxidase (inner membrane), monoamine oxidase (outer membrane) and malate dehydrogenase (matrix) is markedly shifted toward the regions of low density when the centrifugation speed becomes too high. The same phenomenon occurs for sulfite cytochrome c reductase (intermembrane space), however, a large amount of the enzyme is still recovered at the bottom of the gradient where the mitochondria are located after centrifugation at 39 000 rev./min. It is generally admitted that the mitochondrial outer membrane is permeable to sucrose in vitro, therefore our results are best explained by supposing that at high centrifugation speeds, the mitochondrial inner membrane becomes permeable to sucrose and consequently the granules accumulate water and swell ; as the macromolecule glycogen cannot pass through the mitochondrial membrane, the density of the mitochondria becomes lower than that of the medium and the organelles migrate to equilibrate in a zone of lower density in the gradient. The swelling of the mitochondria disrupts to some extent the outer membrane and causes a release of the intermembrane space components in the medium already before the granules go to their new equilibrium density. The large increase of sulfite cytochrome c reductase free activity observed after centrifugation at 65000 rev./min is in good agreement with this interpretation; it shows that the outer mitochondrial membrane is considerably damaged at that centrifugation speed. The free activity of malate dehydrogenase appears to be less increased in these conditions. This does not disagree with our hypothesis that the inner mitochondrial membrane becomes more permeable to sucrose when the centrifugation speed is too high. Indeed it is possible that the increase of permeability is not sufficient to allow NADH to have a free access to malate dehydrogenase inside the mitochondrial matrix. As in the sucrose gradient experiments, the hydrostatic pressure seems to be the main causal agent of the deterioration of mitochondria when the centrifugation speed is increased. However, when the centrifugation is performed in a sucrose gradient, the mitochondria not only swell but become leaky in respect of their macromolecular components when the hydrostatic pressure is high enough [l]. The latter phenomenon is not apparent during centrifugation in a glycogen gradient. This difference may possibly result from the fact that in a glycogen gradient with Fig. 4. Pellicle of a mitochondrial fraction cut perpendicular to its suvface. (A) Mitochondria1 fraction in 0.25 M sucrose; the main components are intact mitochondria in a condensed form; (B) 0.6 ml of a mitochondrial fraction isolated in 0.25 M sucrose is added to a 5 ml glycogen gradient (4 to 20% w/w) with 0.25 M sucrose as solvent; then the content of the tube is carefully mixed and an aliquot of the homogeneous suspension taken for examination. The main components are intact mitochondria but swollen granules seem more frequent that in (A) ; (C) mitochondrial fraction after centrifugation at 39000 rev./min at 0 "C, in a glycogen gradient (4 to 20% w/w) with 0.25 M sucrose as solvent. Time integral of the square angular velocity: 144 rad2/ns, Spinco rotor SW 65. Generally, the matrix content of the mitochondria is less dense, the intracristae space more shrunken ; numerous granules are swollen and the outer membrane broken; (D) mitochondrial fraction after centrifugation like in (C) but at 65000 rev./min. Mitochondria are markedly altered; they are swollen, the matrix content is clear; a large granule (+) is frequently seen inside the matrix, near the inner membrane; the outer membrane is broken

M. Collot, S. Wattiaux-De Coninck, and R. Wattiaux 607 Fig. 4

608 M. Collot, S. Wattiaux-De Coninck, and R. Wattiaux : Isopycnic Centrifugation of Rat-Liver Mitochondria 0.25 M sucrose as solvent, the swelling of the granules caused by a permeation of the inner membrane to sucrose decreases their equilibrium density; as a result the granules go to equilibrate in a region subjected to a lower hydrostatic pressure after swelling. On the contrary, in a sucrose gradient, the mitochondria made more permeable to sucrose become denser and therefore migrate to a region subjected to a higher hydrostatic pressure. Morphological examinations show that after centrifugation at 65 000 rev./min the mitochondria are deeply altered; as is expected from the biochemical results, most are very swollen and have their outer membrane broken. Until now, we do not know the significance of the dense bodies frequently seen in the matrix of these mitochondria. They have also been observed when mitochondria were subjected to highspeed pelleting in 0.25 M sucrose [12]. It is probable that they are caused by the hydrostatic pressure exerted on the granules during centrifugation, Indeed, they are observed when mitochondria are subjected to a high hydrostatic pressure in a hydraulic press [13]; recently, Morton et al. [14] have shown that a large matrix granule of 300-400 nm diameter is apparent in most mitochondria of sheep-liver fragments submitted to a hydrostatic pressure of 1000 kg/cm2 during 60 min. This work was supported by the Fonds National de la Recherche ScientiJique. The authors wish to thank Mrs M. J. Dehasse, Miss N. Henry and Mr J. Cpllet for their skilful technical assistance. REFERENCES 1. Wattiaux, R., Wattiaux-De Coninck, S. & Ronveaux- Dupal, M. F. (1971) Eur. J. Biochem. 22, 31-39. 2. Beaufay, H., Jacques, P., Baudhuin, P., Sellinger, 0. Z., Berthet, J. & de Duve C. (1964) Biochem. J. 92, 184-205. 3. Wattiaux, R. & Wattiaux-De Coninck, S. (1970) Biochem. Biophys. Res. Commun. 40, 1185-1188. 4. Wattiaux-De Coninck, S., Ronveaux-Dupal, M. F., Dubois, F & Wattiaux, R. (1973) Eur. J. Biochem. 39, 93-99. 5. de Duve, C., Pressman, B. C., Gianetto, R., Wattiaux, R. & Appelmans, F. (1955) Biochem. J. 60, 604-617. 6. de Duve, C., Berthet, J. & Beaufay, H. (1959) Prog. Biophys. Chem. 9, 325-369. 7. Appelmans, F., Wattiaux, R. & de Duve, C. (1955) Biochern. J. 59,438-445. 8. Schnaitman, C., Erwin, V. G. & Greenawalt, J. W. (1967) J. Cell Biol. 32, 719-735. 9. Wattiaux-De Coninck, S. & Wattiaux, R. (1971) Eur. J. Biochem. 19,552-556. 10. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J. (1951) J. Biol. Chem. 193, 265-275. 11. Baudhuin, P., Evrard, P. & Berthet, J. (1967) J. Cell Biol. 32,181-191. 12. Ronveaux-Dupal, M. F., Collot, M., Wattiaux-De Coninck, S. & Wattiaux, R. (1972) Arch. Int. Physiol. Biochim. 80,406-407. 13. Beaufay, H. (1973) Spectra, 4, 51-61. 14. Morton, D. J., Rowe, R. W. D. & MacFarlane, J. J. (1973) J. Bioenerg. 4, 445-453. M. Collot, S. Wattiaux-De Coninck, and R. Wattiaux, Laboratoire de Chimie Physiologique, Facultes Universitaires Notre-Dame de la Paix, Rue de Bruxelles 61, B-5000 Namur, Belgium