Abstract
We demonstrated a neutral Mg-ATPase activity in human peroxisomal membranes. To establish the precise experimental conditions for detection of this ATPase, both cytochemical and biochemical characterizations were first carried out in liver peroxisomes from control and cipofibrate-treated rats. The results demonstrated an Mg-ATPase reaction in both normal and proliferated peroxisomes. The nucleotidase activity, with marked preference for ATP, was sensitive to the inhibitors N-ethylmaleimide and 7-chloro-4-nitrobenzo-2-oxadiazole (NBDCl). An ultrastructural cytochemical analysis was developed to evaluate the peroxisomal localization, which localized the reaction product to the peroxisomal membrane. These characteristics can help to differentiate the peroxisomal ATPase from the activity found in mitochondria and endoplasmic reticulum. The conditions established for detecting the rat peroxisomal ATPase were then applied to human peroxisomes isolated from liver and skin fibroblasts in culture. A similar Mg-ATPase activity was readily shown, both cytochemically and biochemically, in the membranes of human peroxisomes. These results, together with previous evidence, strongly support the presence of a specific ATPase in the human peroxisomal membrane. This ATPase may play a crucial role in peroxisome biogenesis.
Peroxisomes are essential and ubiquitous subcellular organelles, surrounded by a single membrane and containing a granular matrix (de Duve 1983). Peroxisomes contain matrix and membrane-bound enzymes involved in several metabolic pathways, particularly those concerned with lipid metabolism (van den Bosch et al. 1992). Peroxisomes appear to form by growth and division of pre-existing organelles (Lazarow and Fujiki 1985). Peroxisomal matrix and membrane proteins, containing peroxisomal targeting sequences, are synthesized on cytoplasmic free ribosomes and are imported post-translationally into the organelle. Peroxisomes contain the machinery for recognizing and importing these proteins into the organelle (reviewed in Subramani 1998). ATP is required for in vitro (Imanaka et al. 1987) and in vivo (Soto et al. 1993) import of matrix proteins, and an ATPase is likely to perform a crucial role in this import process (Imanaka et al. 1987). One major peroxisomal membrane protein (PMP) of 70 kD (PMP70) contains an ATP-binding domain and belongs to the P-glycoprotein type of transporters (Kamijo et al. 1990). Several mutations affecting peroxisome biogenesis have been found in yeast and animals (for review see Subramani 1997). In humans, such mutations can cause severe (even lethal) inherited metabolic disorders, such as the Zellweger syndrome (Santos et al. 1988b), the prototype of the group of the peroxisome biogenesis disorders (PBDs) (Lazarow and Moser 1995). The gene defects have been elucidated for several such disorders (Chang et al. 1999). Among the genes involved in peroxisome biogenesis (Pex genes), some of their products (peroxins) show ATPase activity (Gould and Valle 2000). In addition, ATP has been related to other peroxisomal functions, such as the ATP requirement for peroxisomal fatty acid oxidation in isolated hepatocytes (Leighton et al. 1987), or with the apparent need for ATP to keep structure-linked latency in peroxisomes in situ (Wolvetang et al. 1990a).
Biochemical evidence for the presence of a neutral Mg-ATPase activity co-sedimenting with rat liver peroxisomes was presented by del Valle et al. (1988) and has been supported by cytochemical data (Makita et al. 1990; Makita 1995). The peroxisomal activity, localized in the membrane of the organelle, exhibits properties characteristic of the vacuolar type of ATPase (del Valle et al. 1988). Some properties common to the mitochondrial F1FO-ATPases have also been detected for the peroxisomal activity (Wolvetang et al. 1990b; Imanaka et al. 1993). Shimizu et al. (1992a) were able to characterize two types of rat liver peroxisomal ATPases. These activities were largely induced by the peroxisome proliferator clofibrate. Makita and Hakoi (1995) reported similar findings and a reduction in ATPase activity in rats treated with acetylsalicylic acid. Malik et al. (1991) found ATPase activity in bovine liver. In yeast, evidence of an alkaline peroxisomal proton-translocating ATPase has also been presented (Douman et al. 1987; Whitney and Bellion 1991).
In this report we show unequivocal evidence for the presence of an Mg-ATPase located in membranes from human peroxisomes. By using biochemical and cytochemical conditions standarized to detect a rat liver peroxisomal ATPase, we were able to detect this ATPase activity in peroxisomes isolated from human liver and skin fibroblasts in culture.
Materials and Methods
Human Samples
Human liver and skin biopsy specimens were obtained from patients undergoing surgery for uncomplicated gallstone disease. Informed consent from the patients was obtained by following procedures approved by the Ethics Committee of the Medical School of the Catholic University of Chile. Liver function tests were normal in all the cases. Liver samples were kept frozen at — 70C and, when needed, a portion of the frozen sample was sectioned on dry ice and analyzed exactly as described by Alvarez et al. (1992). Skin fibroblasts from control subjects were cultured in Dulbecco's MEM containing 10% fetal calf serum and antibiotics in a humidified incubator at 37C and 5% CO2 as previously described (Santos et al. 1985).
Rat Liver
Male Sprague-Dawley rats (200–250 g) were employed, fed with standard diet (control rats) or fed for 2 weeks with standard diet containing 50 mg/kg of ciprofibrate to induce peroxisome proliferation (treated rats).
Biochemical Assays
L-fractions, which contain mainly light mitochondria, peroxisomes, lysosomes, and some microsomes, were prepared from human and rat liver homogenates by standard subcellular methodology (de Duve et al. 1955). To obtain purified peroxisomes from human liver, L-fractions were subfractionated in continuous Nycodenz density gradients (1.05–1.30 g/ml) as previously described (Alvarez et al. 1992). For rat liver, the fractionation was performed in Nycodenz following the conditions described previously (Bronfman et al. 1979), except that centrifugation was performed employing a VTi-65 Beckman rotor at 40000 rpm for 50 min at 8C. Fractions of 0.4 ml were collected and their density determined in a calibrated organic density gradient column. The purity of the peroxisomal fractions was controlled by protein and marker enzyme determinations: catalase (peroxisomes), glutamate dehydrogenase or cytochrome c oxidase (mitochondria), and NADPH cytochrome-c reductase (endoplasmic reticulum) (Leighton et al. 1977).
A postnuclear supernatant, containing most membranous organelles, was prepared from fibroblast homogenates and subfractionated in continuous Nycodenz density gradients exactly as previously described (Santos et al. 1988a). The peroxisomal fractions were characterized using marker enzymes as described (Santos et al. 1988a).
The ATPase activity was assayed biochemically as previously described (del Valle et al. 1988), but without oligomycin, measuring inorganic phosphate released from ATP (Lanzetta et al. 1979). The assay mixture contained Tris-HCl 100 mM, pH 7.0, and 100–200 μg peroxisomal or 60 μg mitochondrial protein. The reaction was initiated by the addition of 2 mM sodium ATP (or other substrate), 2 mM MgCl2, and 20 mM KCl in a final volume of 0.5 ml. One unit corresponds to 1 μmol of inorganic phosphate released in 1 min. Substrate analogues and inhibitors were tested under these conditions.
To control biochemically the effect of fixation on rat liver ATPase activity, the enzyme was measured in both freshly prepared and fixed L-fractions and in purified peroxisomes. The following fixatives were tried: 4% paraformaldehyde, 2% paraformaldehyde, 2% paraformaldehyde plus 0.5% glutaraldehyde, and 4% paraformaldehyde plus 0.5% glutaraldehyde. The latter mixture gave the best preservation of the peroxisomal ATPase activity and was used for the cytochemical studies. The samples were washed for 30 min in 0.01 M PIPES buffer, pH 7.0, with 0.15 M sucrose after fixation and then were resuspended for the assays. For the biochemical controls, the activity of the enzyme was measured after incubation of the suspensions in the presence or absence of lead nitrate under the conditions employed for cytochemistry.
Electron Microscopy
Pellets of subcellular fractions were processed as reported previously to obtain homogeneous sampling (Leighton et al. 1975). Briefly, after decantation of the supernatant and gentle mixing of the pellet with a glass rod, a sample was suctioned with a Pasteur pipette. Then the tip was introduced into cold fixative and part of its content was slowly released into the fixative. After 60 min the extruded filament was sliced into fragments that could be handled as tissue blocks. After fixation the pellets were washed for 30 min at 2C in 0.01 M PIPES buffer, pH 7.0, with 0.15 M sucrose added, and were further processed for histochemical or ultrastructural analysis.
The cytochemical medium for ATPase was prepared as previously reported (Koenig and Vial 1973; Koenig et al. 1987) with some modifications to improve detection of the peroxisomal ATPase. It consisted of 0.1 M Tris-maleate buffer, pH 7.0, 0.15 M sucrose, 2 mM magnesium sulfate, 4 mM ATP 10, and 3.6 mM Pb (NO3)2 (standard medium).
To control the cytochemical procedure, various conditions were employed: absence of substrate; substitution of adenosine monophosphate or glucose-6-phosphate for ATP; addition of NEM or NBDCl as inhibitors; and preincubation of the specimens for 10 min at 90C instead of 2C before addition of substrate and lead nitrate. Incubations were carried out for 120–180 min at 37C. The specimens were rinsed in distilled water and postfixed with 1% osmium tetroxide for 30 min at 2C, dehydrated in ethanol and acetone, and embedded in epon. Thin sections were cut with an MT2 Porter Blum ultramicrotome and were observed (with and without lead nitrate staining) under a Siemens Elmiskop IA electron microscope.
Cytochemical demonstration of ATPase activity was also performed using the cerium chloride method for capturing inorganic phosphate produced by the ATPase activity (Robinson and Karnovsky 1983).
Cytochemistry for catalase was performed exactly as described by Alvarez et al. (1992), using a modification of the alkaline diaminobenzidine method of Roels and Goldfischer (1979). Incubations were performed at 37C for 15 hr in the presence and absence of H2O2.
Nycodenz was purchased from Nyegaard, Oslo; ciprofibrate was a gift from Sterling-Winthrop Research Institute (Rensselaer, NY). Other biochemical reagents were from Sigma Chemical (St Louis, MO).
Results
To establish the experimental conditions for detection of a human peroxisomal ATPase, several biochemical and cytochemical studies were first carried out using peroxisomes isolated from rat liver.
Rat Liver Peroxisomes
Biochemical Assays. Peroxisomes were purified from control and cipofibrate-treated rats after subfraction-ation of an L-fraction in a continuos Nycodenz gradient (Figure 1). Peroxisomal fractions in this type of fractionation experiment are mainly contaminated by mitochondria. An Mg-ATPase activity was found in peroxisomal fractions. Several substrate analogues were tested in peroxisomes and in mitochondrial fractions. Table 1 shows the phosphate-releasing activity of the purified organelles with different substrates at 2 mM concentration. The peroxisome-associated ATPase activity is specific for ATP, and less active with other substrates: ATP>CTP>GTP>UTP>ADP> UDP = AMP. The pattern of activity for the mitochondrial enzyme is different. It shows similar activity with ATP, GTP, and UTP. The ATPase activity in the presence of inhibitors was measured in mitochondria and peroxisomes. These results are shown in Table 2. Strong inhibition of the peroxisomal enzyme with 50 and 100 μM NBDCl was observed, whereas the mitochondria-associated enzyme showed no inhibition. The concentration of NBDCl employed was established from a concentration curve that gave an IC50 of 20 μM for the peroxisomal enzyme and only slight inhibition for the mitochondrial activity with up to 20 μM NBDC1, employing 100–200 μg peroxisomal or 60 mitochondrial protein per test (data not shown). NEM 1 mM, as described previously (del Valle et al. 1988), also markedly inhibited the peroxisomal ATPase, with only a small inhibitory effect on the mitochondrial enzyme. Vanadate and ouabain had no effect on the peroxisomal ATPase activity. Sodium azide inhibited the mitochondrial ATPase.
As shown in Figure 1, most of the ATPase activity follows the mitochondrial and endoplasmic reticulum markers, with only small activity apparently associated with peroxisomes. The ATPase-specific activities for the experiment illustrated were 58, 101 and 4.1 mU/mg protein for the L-fraction, the mitochondrial peak, and the peroxisomal peak, respectively. The ATPase + NBDCl graph illustrates the relative inhibition by NBDCl of the activity present in each fraction from the gradient. The peroxisomal peak shows 77% inhibition and the mitochondrial peak none, in agreement with the results shown in Table 2. The ATPase activity in the low-density fractions also shows some sensitivity to NBDCl. As discussed below, this might correspond to fragments from damaged mitochondria.
Subfractionation by isopycnic equilibrium in a continuous Nycodenz gradient of an L-fraction from the liver of a ciprofibrate-treated rat. Enzyme markers were employed to discriminate peroxisome-associated ATPase activity: catalase as peroxisomal marker, glutamate dehydrogenase for mitochondria, and NADPH cytochrome c reductase for endoplasmic reticulum. The total ATP-ase activity and the NBDCl-sensitíve ATPase are shown. The NBDCl-sensitive ATPase is expressed as the percent of inhibition observed in each subcellular fraction. For the marker enzymes the distribution is expressed as frequency vs the density limits of each fraction. Frequency corresponds to the product of the fractional amount of enzyme recovered in the fraction and the differences in the density limits of the fraction collected.
Rat liver ATPase activity with different substratesa
Peroxisomes and mitochondria were obtained after subfractionation of an L-fraction from a peroxisome-proliferated rat liver in a continuous Nycodenz gradient. The ATPase activity of both organelles was measured biochemically in the presence of different substrates. The activity is expresed as nmoles of inorganic phosphate released per minute per mg protein. Substrates were employed at 2 mM concentration in all cases. Similar results were obtained in two independent experiments.
Cytochemical Localization. Pellets of proliferated peroxisomal fractions, similar to those characterized in Figure 1, were fixed and incubated in cytochemical media. Peroxisomes were identified using the standard cytochemical reaction for catalase. The DAB peroxidation product was confined to the peroxisomal matrix, which characteristically (Leighton et al. 1975) appeared partially extracted in the smaller peroxisomes and markedly so in the larger ones (Figure 2A. Occasional membranous profiles and mitochondria were found as minor contaminants in these fractions (data not shown). After incubation in the Mg-ATPase standard medium, cytochemical staining was seen in peroxisomes. The reaction product outlined the peroxisomal membrane (Figure 2B. When membranes were sectioned perpendicularly, the peroxisomal matrix was practically free of reaction and aggregates of reaction product were found only on the cytoplasmic side (Figure 2B. After fixation in 4% paraformaldehyde-0.5% glutaraldehyde, approximately 10% of the Mg-ATPase activity present in the peroxisomal fractions was preserved, and over 9% when lead ions were present in the incubation medium. In fixed proliferated peroxisomal fractions, the Mg-ATPase activity showed substrate specificity and sensitivity to inhibitors similar to those of unfixed fractions. The enzyme activity decreased by more than 50% in the presence of both NEM and NBDCl, and ATP was the best substrate. The positive reaction associated with the peroxisomal membrane was markedly reduced if ATP was replaced by another substrate or in the presence of specific inhibitors.
Effect of inhibitors on activities of rat peroxisomal and mitochondrial ATPasesa
The average specific activity of the control measurements was 9.1 ± 1.9 mU/mg protein for the peroxisomal ATPase and 82.7 ± 21.4 for the mitochondrial ATPase. For each inhibitor, the activities are expressed relative to the controls run in parallel. Numbers of separate determinations are shown in parentheses.
In the presence of glucose-6-phosphate, an intense positive reaction was observed only in the membranous profiles (Figure 2C, apparently corresponding to the activity of glucose-6-phosphatase in endoplasmic reticulum fragments. Addition of NBDCl to the ATPase standard medium specifically inhibited the ATPase reaction associated with peroxisomes and had no effect on the ATPase present in the endoplasmic reticulum membranes (Figure 2D. However, in the presence of NEM the ATPase cytochemical reaction in both the membranous profiles and the peroxisomes was greatly reduced (data not shown).
The results obtained with purified peroxisomes were confirmed in L-fractions obtained from control rat livers. The Mg-ATPase reaction product was also found on the cytoplasmic side of the peroxisomal membranes (Figure 3). A strong ATPase reaction was present in the membranes of the endoplasmic reticulum, whereas few aggregates of product were associated with mitochondria (data not shown). When glucose-6-phosphate was used as substrate, a positive reaction was found only in the endoplasmic reticulum vesicles (data not shown). Addition of NBDCl resulted in a strong specific inhibition of the Mg-ATPase associated with the peroxisomal membrane (data not shown).
Human Peroxisomes
Biochemical Assays. The presence of an Mg-ATPase in human peroxisomes was evaluated biochemically and cytochemically using the standarized conditions for rat liver peroxisomes. Peroxisomal fractions isolated from frozen human liver samples showed easily detected Mg-ATPase activity. As shown in Table 3, the specific activity of the human liver Mg-ATPase was about half that of the rat liver. This activity showed a similar pattern of substrate analogues ATP>CTP> GTP>UTP>ADP>UDP = AMP to rat liver (data not presented) and showed a similar pattern of inhibition with NBDCl (Table 3).
Cytochemical reactions in purified peroxisomal fractions from ciprofibrate-treated rats. (
Figure 4 shows the distribution pattern of the enzyme markers in a Nycodenz density gradient after subfractionation of an L-fraction prepared from frozen human liver. Similar to rat liver, most of the ATPase activity follows the mitochondrial distribution, with only very small activity associated with peroxisomal fractions. The NBDCl pattern of inhibition of the ATPase activity in each fraction of the gradient shows that the peroxisomal peak exhibited about 90% inhibition. The ATPase activity in the low-density fractions also shows some sensitivity to NBDCl.
Peroxisomes were also isolated from normal human fibroblasts in culture. These fibroblasts were homogenized and the postnuclear supernatant was sub-fractionated in a Nycodenz density gradient designed to separate peroxisomes from the rest of the membranous organelles (Santos et al. 1988a). Figure 5 shows peroxisomes that were isolated at the high-density region of the gradient (density of 1.18 g/ml) where there is little contamination by mitochondria. These high-density isolated peroxisomes contained Mg-ATPase activity lower than that in human liver (Table 3) but which was inhibited by the NBDCl inhibition pattern (Table 3).
Cytochemical Localization. Pellets of peroxisomal fractions obtained from human liver, similar to those characterized in Figure 4, were fixed and incubated in cytochemical media. Peroxisomes were identified using the standard DAB cytochemical reaction for catalase (Figure 5A. A nucleoid type of structure is seen in these human peroxisomes, as previously shown for peroxisomes isolated from frozen human livers (Alvarez et al. 1992). Incubation of peroxisomes in the Mg-ATPase standard medium showed cytochemical staining of peroxisomal membranes (Figure 5B. This staining was markedly reduced in the presence of NB-DCl (Figure 5D. The reaction seen in the presence of glucose-6-phosphate (Figure 5C apparently corresponds to the activity of glucose-6-phosphatase in endoplasmic reticulum. As shown for rat liver peroxisomes, the human peroxisome ATPase is strongly inhibited by NBDCl.
Cytochemical ATPase reaction in an L-fraction from normal rat liver. The reaction product is associated with the peroxisomal membranes. In perpendicular sections, the reaction product appears to be located on the cytoplasmic side of the peroxisomal membrane. A positive Mg-ATPase reaction is also seen in the membranous profiles present in this fraction. (
The presence of Mg-ATPase was also detected in peroxisomes isolated from human fibroblasts in Nycodenz density gradients, as seen in Figure 6. Peroxisomes were identified using the cytochemical DAB method for catalase (Figure 7A. An Mg-ATPase was also detected in the peroxisomal membrane (Figure 7B. This activity was drastically reduced in the presence of the inhibitor NBDCl (Figure 7D. As previously shown for human liver, a strong cytochemical reaction was seen in the endoplasmic reticulum when the cytochemical medium for ATPase was modified by replacing ATP with glucose-6-phosphate (Figure 7C. To confirm our cytochemical results on peroxisomes isolated from human fibroblasts, we performed additional experiments to demonstrate the presence of an ATPase activity in situ. For this, we fixed human skin fibroblasts and carried out the cytochemical cerium chloride method for detecting ATPase activity. As shown in Figure 8B, finer and clearer reaction products are located on peroxisomal membranes. For comparison, the cytochemical demonstration of catalase is also shown (Figure 8A.
Specific activity of human and rat peroxisomal ATPasesa
Peroxisomes were obtained after subfractionation of an L-fraction from a peroxisome-proliferated rat liver, from a frozen human liver, and from a post-nuclear supernatant obtained from human fibroblast homogenates in a continuous Nycodenz gradient. The ATPase activity is expressed as nmoles of inorganic phosphate released per minute per mg protein. Similar results were obtained in two independent experiments.
Subfractionation by isopycnic equilibrium in a continuous Nycodenz gradient of an L-fraction from a frozen human liver. Enzyme markers were employed to discriminate peroxisome-associated ATPase activity: catalase as peroxisomal marker, cytochrome c oxidase for mitochondria, and NADPH cytochrome c reductase for endoplasmic reticulum. The total ATPase activity and the NBDCl-sensitíve ATPase are shown. The NBDCl-sensitive ATPase is expressed as the percent of inhibition observed in each subcellular fraction. For the marker enzymes the distribution is expressed as frequency vs the density limits of each fraction. Frequency corresponds to the product of the fractional amount of enzyme recovered in the fraction and the differences in the density limits of the fractions collected.
Cytochemical reactions in purified peroxisomal fractions from human liver. (
Discussion
The experiments presented here demonstrate, for the first time, the presence of an ATPase activity in the membrane of human peroxisomes. The cytochemical reaction product is localized to the peroxisomal membrane. Specifically, the topology of the cytochemical reaction product in subcellular fractions excludes the possibility that the ATPase activity is also localized in the peroxisomal matrix. The cytochemical results, in addition to the detection of peroxisomal and mitochondrial ATPase activities, also show activity associated with the endoplasmic reticulum. Because microsomal profiles can contaminate peroxisomal fractions, this fact should be taken into consideration when results from biochemical experiments in which the analysis is centered mainly on the peroxisomal and mitochondrial activities are interpreted (del Valle et al. 1988; Wolvetang et al. 1990a; Malik et al. 1991).
The evidence available from subcellular fractionation data does not support the presence of a functional proton-translocating ATPase in peroxisomes, nor does it indicate a putative function. However, evidence for an ATP-dependent structure-linked latency of peroxisomes in situ has been found in fibroblasts (Wolvetang et al. 1990b), thus raising the possibility that the observations on permeability made with isolated organelles or membrane fragments do not reflect the membrane properties of the organelle in situ. Kim et al. (1998) found a pH of 6.6 in the matrix of human peroxisomes and suggested that a proton-translocating ATPase must exist to maintain this pH gradient across the human peroxisomal membrane. In contrast to these results, Dansen et al. (2000) found a rather basic pH of 8.2 in peroxisomes from human fibroblasts. Furthermore, these authors found that ATP depletion did not result in peroxisomal pH change, arguing against the involvement of an ATP-driven proton pump.
Subfractionation by isopycnic equilibrium in a continuous Nycodenz gradient of a postnuclear supernatant fraction from human fibroblasts. Enzyme markers were employed to discriminate peroxisome-associated ATPase activity: catalase as peroxisomal marker, cytochrome c oxidase for mitochondria, and NADPH cytochrome c reductase for endoplasmic reticulum. The total ATPase activity and the NBDCl-sensitíve ATPase are shown. The NBDCl-sensitive ATPase is expressed as the percent of inhibition observed in each subcellular fraction. For the marker enzymes the distribution is expressed as frequency vs the density limits of each fraction. Frequency corresponds to the product of the fractional amount of enzyme recovered in the fraction and the differences in the density limits of the fractions collected.
Cytochemical reactions in purified peroxisomal fractions isolated from human fibroblasts fractionated in a continuos Nycodenz density gradient. (
Initially, the properties of the rat peroxisomal ATP-ase suggested that it belongs to the vacuolar or V-type ATPases: neutral, NEM sensitive, low activity when Mg2+ is replaced by Ca2+, and insensitive to oligomycin and vanadate (del Valle et al. 1988). It is also resistant to azide, an inhibitor of F-type ATPases, and to the Na+,K+-ATPase inhibitors ouabain and vanadate (del Valle et al. 1988; Wolvetang et al. 1990a; Malik et al. 1991). However, on the basis of its resistance to bafilomycin, apparently a specific V-type inhibitor, and its moderate sensitivity to oligomycin, Wolvetang et al. (1990a) proposed that the peroxisomal enzyme also has F-type properties. In addition, Imanaka et al. (1993) reported the immunoreactive presence of the F1-ATPase γ-subunit in rat liver peroxisomes. NBDCl, an alkylating agent, is known to inhibit various ATPases in variable proportions, apparently depending on the enzyme and the conditions employed (Bowman and Bowman 1986; Rudnick 1986). Under the conditions used in the present study, NBDCl selectively inhibited the peroxisomal activity in fresh and fixed fractions. In platelets, NBDCl also acts as a specific inhibitor. It inhibits the V-type ATPase more strongly than the F1F0 type or the phosphoenzyme type (Dean et al. 1984). Our assay conditions, specially developed to measure the peroxisomal ATPase activity (del Valle et al. 1988), might explain why we did not find an effect of NBDCl on normal mitochondria. The activity in the low-density fractions of the gradients, where a non-peroxisomal NBDCl-sensitive activity was detected, might correspond to damaged mitochondrial fragments, which are known to equilibrate at low densities (Leighton et al. 1977) and might be more susceptible to inhibition by NBDCl than the intact organelles that band at the peak of the mitochondrial marker.
In situ cytochemical reactions of catalase and ATPase in fixed skin fibroblasts. (
Systematic characterization of the peroxisomal ATP-ase, its activity, and the extent of its functional similarities with other ATPases requires its purification. The fact that the 70-kD peroxisomal membrane protein is an ATP-binding protein, presumably involved in protein transport (Kamijo et al. 1990), suggests that it might be related to the ATPase activity. In fact, the consensus sequences for ATP binding are exposed to the cytosol and are susceptible to protease, as previously shown for rat peroxisomal ATPase activity (del Valle et al. 1988). However, a detailed study performed by Shimizu et al. (1992b) showed no evidence of ATPase activity for this protein.
It has been suggested that an ATPase activity is involved in peroxisome biogenesis, particularly at the level of import of peroxisome proteins to pre-existing mammalian peroxisomes. In vitro and in vivo evidence suggests that ATP hydrolysis is required to efficiently import peroxisomal matrix proteins (Imanaka et al. 1987; Soto et al. 1993). Furthermore, two ATP-ases have been implicated in the basic import defect of some peroxisome biogenesis disorders in humans: (a) a 143-kD AAA ATPase, product of the Pex 1 gene, and (b) a 104-kD AAA ATPase, product of the Pex 6 gene (Gould and Valle 2000). Although these AAA proteins have a specific role in peroxisomal membrane fusion (Titorenko and Rachubinski 2000), they may also have a general function in the assembly, organization, and disassembly of protein complexes (Gould and Valle 2000).
Finally, a precise characterization of the role of this novel human peroxisomal ATPase is required, especially because the short-term regulatory mechanisms of this organelle have not yet been characterized.
Footnotes
Acknowledgements
Supported by project FONDECYT 1980978.