Mild protease treatment as a small-scale biochemical method for mitochondria purification and proteomic mapping of cytoplasm-exposed mitochondrial proteins

Because of its importance in basic biology and medicine, great efforts are being devoted to unraveling of the genuine mitochondrial proteome, which is the dynamic protein complement that the organelle uses to maintain its structure and functionality. Several proteomic investigations have now clearly shown that all the purification approaches we have at our disposal suffer from the problem of co-purification; therefore, it is very difficult to distinguish novel mitochondrial proteins from those that are just contaminants of the preparation. The question is further complicated by the fact that the mitochondrial proteome depends on the tissue source. Density gradient centrifugation is the most widespread purification method for obtaining highly pure mitochondrial fractions. The main disadvantage of these methods is the low yield of purified mitochondria that precludes their use in low-scale purifications. Here, we have treated small aliquots of crude mitochondria from mouse liver and from cultured hepatocytes (HEPA1-6) with trypsin under mild proteolysis conditions and have evaluated the suitability of this reaction as a small-scale purification approach. The protease removed several cytoplasmic and endoplasmic reticulum proteins, together with a fraction of mitochondrial proteins that we hypothesize to be associated with the cytosolic face of the outer mitochondrial membrane. The peculiar topology of these mitochondrial proteins could be indicative of their functional roles. Finally, our study represents an application of advanced mass spectrometry technology to the evaluation of biochemical approaches for the treatment of mitochondria.     

Isolated hepatocytes - past, present and future

The first technique for large-scale preparation of isolated hepatocytes was described in 1953 and involved perfusion of rat liver under pressure with a Ca2+-free solution containing a chelating agent. Various modifications of this technique were in use over the next ten years, until it was demonstrated that cells prepared in this manner were grossly damaged, losing most of their cytoplasmic enzymes during the preparative procedure. The successful preparation of intact isolated hepatocytes by collagenase-treatment of liver was achieved in 1967, and the widespread use of intact hepatocyte suspensions was accelerated by the development soon after of high-yield preparative techniques involving perfusion of the liver with a medium containing collagenase.The introduction of the isolated hepatocyte preparation has enabled experimental studies that otherwise would not be feasible. Important advances have been the use of cultured hepatocytes, frequently of human origin, for the investigation of the metabolism and toxicology of potential therapeutic agents. Success in this field has been achieved through the steady improvement in techniques for the maintenance in culture of differentiated hepatocytes, and in particular their cytochrome P450 complexes. Another area showing considerable promise is the employment of hepatocytes, generally from a porcine source, in temporary support systems for patients with acute liver failure. Our own studies have concentrated on the demonstration of long-range interactions between hepatocyte compartments which suggest that energy transfer between cell compartments can take place without ATP turnover.     

Induction of aconitate hydratase in hepatocytes of starving rats

Induction of the activity of aconitate hydratase (AH) was observed in rat hepatocytes under the conditions of food deprivation. The increase in AH activity after 4 days of starvation in the studied tissues was from 0.57 to 2.05 U/g crude liver weight. The induction of aconitase was associated both with the cytoplasmic and mitochondrial AH isoforms. The activities of cytosolic and mitochondrial AH isoforms in starving animals consisted of 83 and 17% of the total activity, respectively. The cytoplasmic and mitochondrial isoforms of the enzyme with specific activities 11.1 and 6.13 U/mg protein, respectively, were obtained by a five-step purification procedure that included fractionation with ammonium sulfate, ion-exchanging chromatography on DEAE-Toyopearl and gel filtration. The purified preparations of these AH isoforms were electrophoretically homogenous. The molecular weights of these isoforms were estimated and several kinetic and regulatory properties were studied.     

Bovine mitochondrial protein synthesis elongation factors. Identification and initial characterization of an elongation factor Tu-elongation factor Ts complex

Animal mitochondrial protein synthesis factors elongation factor (EF) Tu and EF-Ts have been purified as an EF-Tu.Ts complex from crude extracts of bovine liver mitochondria. The mitochondrial complex has been purified 10,000-fold to near homogeneity by a combination of chromatographic procedures including high performance liquid chromatography. The mitochondrial EF-Tu.Ts complex is very stable and cannot be dissociated even in the presence of high concentrations of guanine nucleotides. No guanine nucleotide binding to this complex can be observed in the standard nitrocellulose filter binding assay. Mitochondrial EF-Ts activity can be detected by its ability to facilitate guanine nucleotide exchange with Escherichia coli EF-Tu. The EF-Tumt exhibits similar levels of activity on isolated mammalian mitochondrial and E. coli ribosomes, but displays minimal activity on Euglena gracilis chloroplast 70 S ribosomes and has no detectable activity on wheat germ cytoplasmic ribosomes. In contrast to the bacterial EF-Tu and the EF-Tu from the chloroplast of E. gracilis, the ability of the mitochondrial factor to catalyze polymerization is not inhibited by the antibiotic kirromycin.     

Subcellular location of a soluble factor that stimulates DNA replication in permeable animal cells

The enzymatic isolation of cells with bacterial collagenase has proved to be a powerful technique for the study of a wide variety of tissues. Unfortunately, for some applications such as the isolation of cells from membranous bone, the cellular damage that results from the exposure of the cells to cytotoxic contaminants of bacterial collagenase has limited the usefulness of this approach. The use of chromatographically purified collagenase alone is often ineffective or very slow to release cells from tissue. We have found that two enzymes are necessary and sufficient to isolate cells from neonatal mouse calvaria: purified collagenase and neutral protease. These two enzymes can be chromatographically purified on a preparative scale to yield 100 mg amounts of each enzyme. The purified enzymes can be recombined in amounts which will digest calvaria at the same rate as the crude bacterial collagenase from which they were derived. The cells that are isolated using the purified enzymes are undamaged, as indicated by the measurement of their equilibrium density on gradients of Ficoll and sodium metrizoate. Cells isolated with crude collagenase never reach an equilibrium density upon isopyknic centrifugation, whereas cells isolated with the purified enzymes reach an equilibrium density of 1.074 g/ml in 90 min.     

Cytoplasmic origin of the so-called nuclear neutral histone protease.

1. We have investigated the origin of proteolytic activity which causes degradation of histones in chromatin isolated from Xenopus liver and the rat liver at neutral pH. Polyacrylamide disc gel electrophoresis was used for detection of proteolytic products of histones. 2. No proteolytic degradation of histones occurs in chromatin isolated from Xenopus erythrocytes and rat liver according to our procedure even after prolonged incubation at pH 8.0 and pH 5.0. However with chromatin isolated from Xenopus liver a high level of histone degradation is observed under similar conditions. 3. Mixing isolated nuclei from Xenopus erythrocytes with a crude cytoplasmic fraction from Xenopus liver causes histone proteolysis in isolated chromatin at pH 8.0. In similar experiments with corresponding fractions from rat liver histone proteolysis can be introduced only after repeated freezing and thawing of the cytoplasmic fraction. 4. A purified lysosomal preparation from rat liver causes a similar type of histone degradation upon incubation with chromatin from Xenopus erythrocytes and rat liver. 5. The neutral proteolytic activity that can be introduced in isolated chromatin by a crude cytoplasmic fraction and by a purified lysosomal erythrocytes and rat liver. 5. The neutral proteolytic activity that can be introduced in isolated chromatin by a crude cytoplasmic fraction and by a purified lysosomal fraction from rat liver is inhibited by sodium bisulphite. 6. We conclude that the neutral proteolytic activity which causes degradation of histones in isolated chromatin is due to a contamination with neutral protease(s) originating from cytoplasmic organelles.     

Inactivation of yeast enzymes by proteinase A and B and carboxypeptidase Y from yeast.

Changes in the activities of 15 different enzymes during incubation of a crude yeast extract with the purified yeast proteinases A and B, and carboxypeptidase Y, respectively, have been measured. The spectrum of action of the three proteinases on the enzymes measured differs significantly, increasing or decreasing their activities or having no effect. Incubation of purified cytoplasmic malate dehydrogenase or purified mitochondrial malate dehydrogenase with proteinases A and B results in selective inactivation of the cytoplasmic enzyme, whereas the mitochondrial activity is not affected. Carboxypeptidase Y has no effect on the activity of either dehydrogenase. The results support the idea of selective proteolysis as the mechanism of the earlier observed inactivation of cytoplasmic malate dehydrogenase, initiated by the addition of glucose to intact yeast cells grown on acetate as carbon source ("glucose effect").     

Schistosoma mansoni: Purification and characterization of malate dehydrogenases

Isoelectric focusing of a homogenate of Schistosoma mansoni, followed by malate dehydrogenase-specific staining, showed the presence of two major and five minor malate dehydrogenase isoenzymes (EC 1.1.1.37), with isoelectric points ranging from 7.3 to 9.5. The malate dehydrogenase isoenzymes were purified by gel filtration, followed by ion-exchange chromatography on DEAE- and CM-cellulose. The isoenzymes could be differentiated by their susceptibility to substrate inhibition. No differences in the Michaelis-Menten constants for substrate were found. One of the isoenzymes is inhibited by 5′-AMP. Further purification of this particular isoenzyme was achieved by affinity chromatography on 5′-AMP-Sepharose 4B. Analysis after subcellular fractionation indicated a mitochondrial origin for this isoenzyme. The mitochondrial isoenzyme (at a recovery of 80%) was purified 218-fold compared to the crude soluble extract, and contained about 40% of the total malate dehydrogenase activity. The enzyme has a molecular weight of 65,500 and showed absolute specificity for l-malic acid, NAD, and NADH. The final preparation has a specific activity of 451 U/mg protein. Physicochemical studies, including binding constants, substrate inhibition, thermostability, and pH optima, demonstrated differences between the mitochondrial and cytoplasmic enzymes. A role for malate dehydrogenase in Schistosoma mansoni metabolism is discussed.     

Preparation of highly purified mitochondria ofNeurospora crassa on a Percoll gradient

Mitochondria isolated from Neurospora crassa were purified by centrifugation in a Percoll density gradient. Enzyme activities and cytochrome differential spectra revealed a high purity of the mitochondria. As compared with a crude mitochondrial fraction the purified mitochondria exhibited a high respiratory activity and a fine ADP/O ratio. Electrophoresis of nucleic acids demonstrated the absence of cytoplasmic rRNA.     

Comparison of plasma membranes and endoplasmic reticulum fractions obtained from whole white adipose tissue and isolated adipocytes.

The influence of the mode of preparation upon some of the characteristics of white adipose tissue plasma membranes and microsomes has been reported. Plasma membrane fractions prepared from mitochondrial pellet were shown to have higher specific activities of (Mg2+ + Na+ + K+)-ATPase than plasma membranes originating in crude microsomes. Isolation of fat cells by collagenase treatment was found to result in a decrease in specific activity of the plasma membrane enzymes; in plasma membranes prepared from isolated fat cells, the specific activity values obtained for (Mg2+ + Na+ +k+)-ATPase and 5'-nucleotidase were only 42% and 6.3% respectively of those obtained in plasma membranes prepared from whole adipose tissue. Purification of whole adipose tissue crude microsomes by hypotonic treatment caused extensive solubilization of the endoplasmic reticulum marker enzymes, NADH oxidase and NADPH cytochrome c reductase. The lability of endoplasmic reticulum marker enzymes, however, was found to be greatly diminished in the preparations from isolated fat cells. The possibility that NADH oxidase and NADPH cytochrome c reductase activities found in the plasma membranes are microsomal enzymes adsorbed by the plasma membranes is discussed. The peptide patterns as well as the NADH oxidase and NADPH cytochrome c reductase activity patterns of plasma membranes and purified microsomes were compared by means of sodium dodecyl sulfate or Triton X-100 polyacrylamide gel electrophoresis.     

Non-ohmic proton conductance of the mitochondrial inner membrane in hepatocytes

The mitochondrial membrane potential in isolated hepatocytes was measured using the distribution of the lipophilic cation triphenylmethylphosphonium (TPMP+) with appropriate corrections for plasma membrane potential, cytoplasmic and mitochondrial binding of TPMP+, and other factors. The relationship between mitochondrial membrane potential and respiration rate in hepatocytes was examined as the respiratory chain was titrated with myxothiazol in the presence of oligomycin. This relationship was nonproportional and similar to results with isolated mitochondria respiring on succinate. This shows that there is an increased proton conductance of the mitochondrial inner membrane in situ at high values of membrane potential. From the respiration rate and mitochondrial membrane potential of hepatocytes in the absence of oligomycin, we estimate that the passive proton permeability of the mitochondrial inner membrane accounts for 20-40% of the basal respiration rate of hepatocytes. The relationship between log[TPMP+]tot/[TPMP+]e and respiration rate in thymocytes was also nonproportional suggesting that the phenomenon is not peculiar to hepatocytes. There is less mitochondrial proton leak in hepatocytes from hypothyroid rats. A large proportion of the difference in basal respiration rate between hepatocytes from normal and hypothyroid rats can be accounted for by differences in the proton permeability characteristics of the mitochondrial inner membrane.     

Complete loss of heparin-releasable triacylglycerol lipase activity after collagenase treatment of the rat liver.

Lipolytic activity at pH 8.5-9.5 was lowered by approx. 80% in homogenates from livers perfused with collagenase or heparin. No heparin-releasable lipase activity was detected in hepatocytes isolated by collagenase treatment. It is concluded that crude collagenase completely inactivates the plasma-membrane-bound heparin-releasable lipase.     

Collagenase enzymes from Clostridium: characterization of individual enzymes.

Four collagenases have been purified to apparent homogeneity from extracts of Clostridium histolyticum and partially characterized. The four purified enzymes are devoid of hydrolytic activity against casein and the synthetic substrate, benzolyarginine naphthylamide, but all retain activity against native collagen. The enzymes are initially spearated by isoelectric focusing where three of the enzymes show distinct isoelectric points: collagenase I = 5.50, collagenase II = 5.65, and collagenases IIIa and IIIb = 5.90-6.00. Collagenases IIIa and IIIb can be subsequently separated on diethylaminoethylcellulose. The four purified enzymes show single bands upon polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. Calibration of the molecular weights on the basis of migration distance shows a marked dependence on gel porosity. At high acrylamide concentration, collagenases I, II, and IIIa appear to converge to a limiting molecular weight congruent to 81 000, while collagenase IIIb has a distinctly lower value congruent to 72 000. The similarity between these molecular weight values and those derived from the sedimentation and diffusion coefficients of the native enzyme indicates that each collagenase is a single polypeptide chain. All of the collagenases have comparable catalytic activities against a series of natural and synthetic substrates and are immunologically cross-reactive. Although all four enzymes are evident upon initial electrofocusing of the crude extract, it is possible that the multiplicity of forms is, at least in part, a consequence of lysis following initial secretion from the cell.     

Guine pig liver L-asparaginase. Separation, purification, and intracellular localisation of two distinct enzymatic activities.

Two distinct L-asparaginase (EC 3.5.1.1) activities were detected in guinea pig liver: Asparaginase 1 and Asparaginase 2. Asparaginase 1 has been purified 272 fold from the crude homogenate; its molecular weight was evaluated by gel filtration to be about 150 000. The purified preparation was shown to be homogeneous by cellulose acetate strip and polyacrylamide disc-gel electrophoresis. Asparaginase 2 has been purified 63.5 fold from the crude homogenate. Its molecular weight was evaluated by gel filtration to be about 21 500. Cellulose acetate strip electrophoresis demonstrated two bands, one of which corresponded to Asparaginase 1 and the other to Asparaginase 2. Cellular fractionation in the ultracentrifuge, showed Asparaginase 1 to be present only in the cytosol fraction. Asparaginase 2 which was unstable at 105 000 X g seemed mostly localized in the mitochondria and secondarily in the cytoplasmic fraction.     

Effect of residual enzymes on degradation of radioiodinated VLDL by collagenase-dispersed hepatocytes.

Liver cells isolated with crude collagenase rapidly catabolized radioiodinated VLDL. The hepatocytes retained proteolytically active elements of the preparative enzymes. Both collagenase in the absence of cells and hepatocytes prepared with collagenase degraded the C-apoproteins of labeled VLDL. Hepatocytes isolated with a chelating agent in place of collagenase deiodinated, but did not catabolize [¹²⁵I] labeled VLDL. Similarly prepared cells degraded labeled HDL, but addition of collagenase to incubations greatly enhanced this process. Therefore, data on hepatic catabolism of serum lipoproteins obtained with hepatocytes freshly prepared with collagenase must be interpreted with caution.     

Mouse liver cell culture. I. Hepatocyte isolation

A method for isolation of mouse liver cells by a two-step perfusion with calcium and magnesium-free Hanks' salt solution followed by a medium containing collagenase is described. Several variations of the commonly used procedure for rat liver cell isolation were quantitatively compared with respect to cell yield and viability. The optimal isolation technique involved perfusion through the hepatic portal vein and routinely produced an average of 2.3 x 10(6) viable liver cells/g body weight. Optimal perfusate collagenase concentration was found to be 100 U of enzyme activity per milliliter of perfusate. Light and electron microscopic evaluation of liver morphology after several steps of the isolation showed distinct morphologic changes in hepatocytes and other liver cells during perfusion. After perfusion with Hanks' calcium- and magnesium-free solution, many hepatocytes exhibited early reversible cell injury. These changes included vesiculation and slight swelling of the endoplasmic reticulum as well as mitochondrial matrix condensation. Subsequent to perfusion with collagenase, the majority of hepatocytes appeared connected to one another only by tight junctional complexes at the bile canaliculi. Multiple evaginations were seen on the outer membrane resembling microville and probably represented the remains of cell-to-cell interdigitations between hepatocytes and sinusoidal lining cells from the space of Disse. The cytoplasmic injury seen after Hanks' perfusion was reversed after collagenase perfusion. After mechanical dispersion, isolated mouse hepatocytes were spherical in shape and existed as individual cells; many (80 to 85%) were binucleated under hase contrast light microscopy. By electron microscopy, cells appeared morphologically similar in cytoplasmic constitution to that seen in intact nonaltered liver cells.     

The interaction of purified rabbit bone collagenase with purified rabbit bone metalloproteinase inhibitor.

1. Pure rabbit bone metalloproteinase inhibitor (TIMP) bound tightly to pure rabbit bone collagenase with an apparent Kd of 1.4 X 10(-10) M. 2. The molecular weight of the enzyme-inhibitor complex was found to be 54 000, but no enzyme activity could be recovered from the complex after treatment with either mercurials or proteinases. The complex thus differed from latent collagenase in terms of size, susceptibility to mercurials and behaviour on concanavalin A-Sepharose. 3. The interaction of the purified components was compared with that of crude collagenase and crude inhibitor in culture medium. Mercurial treatment partially reversed the inhibition in the crude system, but not when the purified components were used. 4. The significance of the results is discussed in relation to the extracellular control of the activity of collagenase.     

Guinea pig pancreatic acini prepared with purified collagenase

Dispersed guinea pig pancreatic acinar cells have been used to investigate several aspects of stimulus-secretion coupling but possess the disadvantage that they are less sensitive and less responsive to secretagogues than in vitro preparations of intact pancreatic tissue (lobules). To overcome the poor responsiveness of isolated acinar cells, we have developed a new procedure for preparing dispersed, intact pancreatic acini whose sensitivity to secretagogues and morphological characteristics are similar to those of pancreatic lobules. Dispersed acini can be manipulated as suspensions of cells and full access of macromolecular probes to apical and basolateral plasmalemmal domains is obtained. Acini were prepared in good yields (~70% on a DNA basis) using only purified collagenase and mild mechanical shear in medium containing 2.0 mM Ca²⁺. Morphologically, acinar cells in the preparations retained intact junctional complexes, asymmetrical distribution of intramembranous particles between apical and basolateral plasmalemmal domains, and polarized distribution of intracellular organelles as found in intact pancreas. Dose-response curves of acini and mechanically prepared lobules to caerulein, carbachol, and bombesin were similar though acini were more sensitive to the C-terminal octapeptide of cholecystokinin. Net stimulated secretory protein discharge was ~36% over 2 h. Crude collagenase was purified for use in preparation of acini by Sephadex G-75 column chromatography which resolved collagenase from clostripain and a non-sulfhydryl-requiring protease. The purified collagenase contained at least four proteins with molecular weights between 85 000 and 110 000. Collagenase with <0.14 units of protease per unit of collagenase produced highly responsive acini; collagenase with >0.9 units of protease per unit of collagenase yielded unresponsive acini. Acini incubated with crude collagenase, chymotrypsin, or the non-sulfhydryl-requiring protease showed depressed secretory response to caerulein. Freeze-fracture electron microscopy of protease-treated acini indicated that the intramembranous particles aggregated and that many of the tight junctions had undergone a proliferation of non-cross-linked sealing strands which extended far down the basolateral plasma membrane and encircled gap junctions. Acini incubated with purified collagenase or with a clostripain-containing fraction from the Sephadex G-75 column appeared unaltered. This procedure produces acini which are morphologically and biochemically similar to the in situ pancreas and overcomes the poor response to secretagogues by isolated pancreatic acinar cells.     

Synthesis of vitamin K1 2,3-epoxide in rat liver mitochondria

Metabolism of vitamin K1 in rat liver mitochondria has been studied with succinate as the source of reducing equivalents. A metabolite was isolated that comigrated with vitamin K1 epoxide using four different chromatographic systems. The purified metabolite had an ultraviolet spectrum (200-330 nm) that was identical to that of synthetic vitamin K1 epoxide. The mass spectrum of the purified metabolite was identical to that of synthetic vitamin K1 epoxide. A comparison of production of vitamin K1 epoxide by mitochondrial and microsomal preparations indicates that the mitochondrial production of vitamin K1 epoxide was about 50% of that of the microsomes. Since the mitochondrial preparation was found to have only 3.4% of the glucose-6-phosphatase activity of the microsomal preparation, it can be concluded that the vitamin K1 epoxide isolated from the mitochondrial incubations was due primarily to mitochondrial synthesis. Epoxidation of vitamin K1 in mitochondria suggests that mitochondria might be sites for vitamin K-dependent carboxylation of protein(s).