- Email: [email protected]
Separation by Blue Native and Colorless Native Polyacrylamide Gel Electrophoresis of the Oxidative Phosphorylation Complexes of Yeast Mitochondria Solubilized by Different Detergents: Specific Staining of the Different Complexes
ANALYTICAL BIOCHEMISTRY ARTICLE NO.
242, 248–254 (1996)
0460
Separation by Blue Native and Colorless Native Polyacrylamide Gel Electrophoresis of the Oxidative Phosphorylation Complexes of Yeast Mitochondria Solubilized by Different Detergents: Specific Staining of the Different Complexes Xavier Grandier-Vazeille and Martine Gue´rin1 Institut de Biochimie et de Ge´ne´tique Cellulaires du CNRS, Universite´ de Bordeaux II, 1 Rue Camille Saint-Sae¨ns, 33077 Bordeaux, France Received June 27, 1996
Blue native polyacrylamide gel electrophoresis (BN-PAGE) or colorless native polyacrylamide gel electrophoresis (CN-PAGE) allowed separation of the oxidative phosphorylation complexes of yeast mitochondria. These complexes were characterized by specific staining related to their enzymatic activity. Solubilization of mitochondria by different nonionic detergents such as Triton X-100, dodecyl maltoside, Nonidet P-40, Lubrol, octyl glucoside, or Hecameg led to the separation of F1 –FO ATPase complexes exhibiting distinct apparent molecular masses related to different contaminating proteins and lipids. All these different forms were active in ATP hydrolysis as revealed directly on the gel. Analysis of the subunit composition of these complexes was carried out by a twodimensional Tricine–SDS–PAGE and showed that the purest F1 –FO ATPase complex was obtained with Lubrol, whereas with Hecameg and octyl glucoside, only the F1 part of ATPase was solubilized. q 1996 Academic Press, Inc.
Separation of enzymatically active heteromultimeric complexes of the inner mitochondrial membrane was greatly improved by the use of either blue native polyacrylamide gel electrophoresis (BN-PAGE)2 or colorless native polyacrylamide gel electrophoresis (CN-PAGE). 1 To whom correspondence should be addressed. Fax: (33) 56 99 90 59. E-mail: [email protected]. 2 Abbreviations used: BN-PAGE, blue native polyacrylamide gel electrophoresis; CN-PAGE, colorless native polyacrylamide gel electrophoresis; Bis–Tris, 2,2-bis(hydroxymethyl)-2,2 *,29-nitrilotriethanol [bis(2-hydroxyethyl)amino–tris(hydroxymethyl)methane]; OXPHOS, oxidative phosphorylation.
Coomassie blue dye, added to sample and cathode buffers for BN-PAGE, induces a charge shift on the proteins so that the electrophoretic mobility of the proteins is mainly determined by the negative charges of bound Coomassie dye. In CN-PAGE, performed at a determined pH, the intrinsic charge of the proteins allows their separation (1). Further analysis of the polypeptide composition of each of the complexes was performed in a second dimension by Tricine–SDS–PAGE (1, 2). This system was essentially used in analyzing mammalian mitochondria (1, 2) and allowed to screen for OXPHOS defects in human diseases (3–5). We attempted to apply this system to yeast mitochondria and more particularly to the analysis of the F1 –FO ATPase complex. This membrane-bound enzyme catalyzes the synthesis of ATP from ADP and Pi in a reaction driven by the electrochemical proton gradient maintained across the membrane by the respiratory chain. This enzyme is structurally conserved among the different organisms and is composed of two structurally and functionally distinct moieties, F1 and FO . The F1 part, which is composed of five different subunits termed a, b, g, d, and e, carries the catalytic sites. The FO sector is embedded in the membrane and forms a specific proton channel. It is composed of at least six hydrophobic subunits, the subunits 6, 8, and 9 which are mtDNA encoded and the subunits 4, d, and OSCP which are nDNA encoded (for review see Ref. 6). However, according to Cox et al. (7), the FO sector may comprise only the three mitochondrial hydrophobic proteins (subunits 6, 8, and 9) which are presumably membrane integral proteins and a further panel of subunits, whose interaction with the membrane is less well defined, may form the FA sector. This FA sector may comprise subunits b (also denoted subunit 4), d, and OSCP
248
AID
0003-2697/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AB 9773
/
6m1f$$$241
10-17-96 07:31:40
aba
AP: Anal Bio
ELECTROPHORETIC SEPARATION OF YEAST MITOCHONDRIA
and the proteins involved in the regulation of the activity (8). In an attempt to answer the question of structural signification of FA , we analyzed the subunit composition of active ATPase complexes obtained from yeast mitochondria solubilized with different nonionic detergents, using the nondenaturing polyacrylamide gel electrophoresis techniques (1). MATERIALS AND METHODS
Chemicals SDS, TEMED, Tricine, Bis–Tris, Triton X-100, Nonidet P-40, Lubrol WX, ATP, and oligomycin were purchased from Sigma; Hecameg was from Vegatec; and octyl glucoside (n-octyl b-D-glucopyranoside) and dodecyl maltoside (n-dodecyl-b-D-maltoside) were from Boehringer. Serva blue G (Coomassie blue G-250) and eaminocaproic acid were from Serva. All other chemicals were purchased from Merck. Yeast Strains Strains of Saccharomyces cerevisiae used are the wild-type diploid strain yeast foam and the wild-type haploid strain AB1-4A/8 (MATa, his 4). Cells were grown on a complete medium supplemented with 2% lactate and harvested in mid-exponential growth phase as already described (9).
249
F1 –FO ATPase was a gift of Dr. J. Velours. Two-dimensional (2-D) Tricine–SDS–PAGE was performed overnight at 157C on 14% acrylamide slab gel (28 1 16.5 1 0.075 or 15.5 1 17.5 1 0.075 cm), with pieces or lanes cut out from the one-dimensional gels. These pieces were soaked in 1% SDS, 1% b-mercaptoethanol for 2 h at room temperature, washed four times in water to remove b-mercaptoethanol solution, and loaded at the usual position for stacking gel (1). BN-PAGE and CN-PAGE gels were fixed and stained with 0.025% (mass/vol) Serva blue G in 50% methanol, 10% acetic acid. Two-dimensional gels were fixed and stained with Serva blue G and further by silver staining according to the procedure of Ansorge (11). Specific Staining of the OXPHOS Complexes After CN-PAGE the ATPase activity was located on the gel by incubation in a 50 mM glycine/NaOH buffer, pH 8.6, containing 5 mM ATP, 5 mM MgCl2 , 0.05% lead acetate as described by Yoshida et al. (12). Staining for dehydrogenase activities was performed in a 100 mM Tris/glycine buffer at pH 7.4 containing 1 mg/ml nitro blue tetrazolium and either 100 mM bNADH or 1 mM sodium succinate as appropriate (13). Cytochrome-containing bands were visualized by staining with 2 mM 3,3*,5,5*-tetramethylbenzidine and 30 mM H2O2 in 100 mM sodium acetate buffer at pH 5 (14).
Sample Preparation and Electrophoresis Techniques Mitochondria isolated from spheroplasts according to Gue´rin et al. (10) were sedimented by centrifugation (12,500g 1 10 min), suspended in 50 mM Bis-Tris buffer, pH 7.0, containing 750 mM e-aminocaproic acid, washed twice in this buffer, and pelleted by centrifugation (12,500g 1 10 min). Mitochondria were resuspended at 10 mg/ml in the same buffer and solubilized for 45 min at 47C by adding detergent in a detergent/ protein ratio of 2.5 for dodecyl maltoside, 0.7 for Triton X-100, 1 for Nonidet P-40, 2 for Hecameg and octyl glucoside, and 3 for Lubrol. Insoluble material was removed by centrifugation (108,000g 1 10 min). Separation of complexes by BN-PAGE or CN-PAGE was performed on linear gradient (5–14%) polyacrylamide slab gels (14.5 1 16.5 1 0.075 cm). For BN-PAGE, sample and cathode buffers were added with Serva Blue as described by Scha¨gger and von Jagow (2). For CNPAGE, ponceau S was added to samples before electrophoresis to follow the migration front. BN-PAGE and CN-PAGE were run at 47C in a vertical apparatus as described by Scha¨gger et al. (1). Apparent molecular masses (Mapp) were determined by using molecular mass markers: urease hexamer (545 kDa) and trimer (272 kDa); bovine albumin dimer (132 kDa) and monomer (66 kDa); chicken egg albumin (45 kDa); carbonic anhydrase (29 kDa); a-lactalbumin (14.2 kDa). Pure
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
Other Assays Protein content of mitochondria was measured using the Biuret method. The protein content of samples containing e-amino caproic acid and detergents was measured by a modified micro-Bradford procedure (15) and in other samples with the Pierce BCA protein assay using bovine serum albumin as standard. Mitochondrial and plasmic ATPase activities were measured on solubilized mitochondria according to Somlo (16) and Serrano (17), respectively. RESULTS AND DISCUSSION
Solubilization of Membrane Complexes by Detergents Preliminary experiments were carried out on mitochondria isolated from two yeast strains, yeast foam and AB1-4A/8. Proteins were solubilized by the nonionic detergent dodecyl maltoside and loaded on a nondenaturing gradient polyacrylamide gel (BN-PAGE). Figure 1 shows that although identical solubilization conditions were used, different profiles of separation of the OXPHOS complexes of yeast mitochondria were obtained. In yeast foam mitochondria, essentially one band with an Mapp of about 820 kDa was observed (lane f), whereas in AB1-4A/8, five bands were visualized: among them the three major bands had apparent mo-
aba
AP: Anal Bio
´ RIN GRANDIER-VAZEILLE AND GUE
250
FIG. 1. Resolution of the OXPHOS complexes by one-dimensional blue native PAGE. Mitochondrial membrane proteins from yeast foam (210 mg) and AB1-4A/8 (136 mg) were solubilized by dodecyl maltoside (detergent/protein Å 2 mg/mg protein) for 10 min (b), 30 min (c), 60 min (d, f), and 120 min (e). After centrifugation (108,000g 1 10 min) the supernatant from AB1-4A/8 (b, c, d, e) and yeast foam (f) and the molecular weight standards (a) were resolved by blue native PAGE (5–14% gradient polyacrylamide slab gel) as described under Materials and Methods.
lecular masses of about 820, 440, and 245 kDa (lane b). Increasing incubation time of yeast foam mitochondria in the presence of detergent did not improve solubilization of the OXPHOS complexes (data not shown), whereas in AB1-4A/8 mitochondria, it led to a splitting of the Mapp 820-kDa band (lanes c, d, and e); both bands corresponded to the ATPase complex as seen in an additional 2-D Tricine–SDS–PAGE (data not shown). Since mitochondria from both strains contained equivalent amounts of cytochrome c oxidase as determined by spectrophotometry and equivalent ATPase activities (data not shown), it was hypothesized that both strains had the same amounts of OXPHOS complexes. Thus, the difference observed between the strains could be due to a different lipid composition of the inner membrane. Consequently, mitochondria from both yeast strains were solubilized, using different detergent/protein ratios and varying incubation times. Whatever the conditions used, solubilization of yeast foam mitochondria was not improved (data not shown). With regard to AB1-4A/8 mitochondria, after BN-PAGE, fixation, and restaining of proteins with Coomassie blue (Fig. 2), the densitometric quantification of the stained bands revealed that, for a given dodecyl maltoside/protein ratio, increasing incubation time did not significantly modify the relative amounts of the three major bands, whereas increasing the detergent/protein ratio enhanced the band 1/band 3 ratio from 1.0 to 1.5. In a second step, characterization of the OXPHOS complexes separated by nondenaturing PAGE was per-
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
FIG. 2. Analysis of dodecyl maltoside-solubilized mitochondrial membrane proteins of AB1-4A/8 by blue native PAGE. Mitochondria from AB1-4A/8 (130 mg) were treated with different concentrations of dodecyl maltoside [detergent/protein ratio of 5 (A), 2 (B), and 1 (C)] for 10 min (a, j, o), 30 min (b, i, n), 60 min (c, h, m), 120 min (d, g, l), and 180 min (e, f, k). After a clarifying spin (108,000g 1 15 min) the supernatants were subjected to blue native PAGE (5–14% gradient polyacrylamide) as detailed under Materials and Methods. The gel was fixed and stained with Coomassie brilliant blue R.
formed by a two-dimensional Tricine–SDS–PAGE of individual lanes from the BN-PAGE gel (Fig. 3): it appeared that band 1 (cf. band 1 in Fig. 2) contained most of the characteristic polypeptides of the F1 –FO ATPase complex (a, b, g, 4, OSCP, d, and 9 subunits). It should
FIG. 3. Two-dimensional electrophoresis of the OXPHOS complexes from yeast mitochondria solubilized by dodecyl maltoside. Supernatant from yeast mitochondria (AB1-4A/8) solubilized by dodecyl maltoside (detergent/protein Å 2.5 mg/mg) was separated in the first dimension by blue native PAGE and in the second dimension by Tricine–SDS–PAGE (14% T, 3% C). The gel was fixed and stained with Coomassie brilliant blue R and the complexes were characterized by their polypeptide patterns. 1, 2, and 3 correspond to bands 1, 2, and 3, respectively, in Fig. 2.
aba
AP: Anal Bio
ELECTROPHORETIC SEPARATION OF YEAST MITOCHONDRIA
be noted that resolution of the other bands revealed in the first dimension was very poor and thus it was difficult to define them. In an attempt to improve the solubilization of mitochondrial complexes, other nonionic detergents were used. To allow a standardization of the different assays, the detergent/protein ratio was chosen instead of the only detergent concentration since varying the protein and detergent concentrations for a given detergent/protein ratio did not improve the solubilization. Different detergent/protein ratios and incubation times were assayed on mitochondria isolated from the two yeast strains. From the different assays, it appeared that the best detergent/protein ratios were 0.7 for Triton X-100, 1 for Nonidet P-40, 2 for Hecameg and octyl glucoside, 2.5 for dodecyl maltoside, and 3 for Lubrol. These ratios corresponded to 10 mM Triton X-100, 13 mM Nonidet P-40, 50 mM Hecameg, 48 mM dodecyl maltoside, and 60 mM Lubrol, respectively. From Fig. 4, several lines of evidence can be drawn: (i) whatever the yeast strain, the modifications observed in the migration of the ATPase complex as a function of the detergent used were the same; i.e., Hecameg (lanes b and h) and octyl glucoside (lanes c and g) led to a complex with a lower apparent molecular mass; (ii) starting from the same amount of mitochondria, dodecyl maltoside solubilized a higher amount of ATPase than the other detergents; (iii) in AB1-4A/8 mitochondria (lanes a–d), Triton X-100 (lane a) or Lubrol (not shown) solubilized one additional band with a higher Mapp , compared to solubilization with dodecyl maltoside (lane d); (iv) Hecameg (lane b) or octyl glucoside (lane c) did not reveal the Mapp 820kDa band, whereas the intensity of the Mapp 440-kDa band was increased; (v) a weaker band migrating slightly above Mapp 440-kDa band was observed when
FIG. 4. Effect of different detergents on the solubilization of yeast mitochondria OXPHOS complexes analyzed by blue native electrophoresis. Mitochondria from AB1-4A/8 (a, b, c, d) and from yeast foam (e, f, g, h, i, j) were solubilized by Triton X-100 (a, e), Hecameg (b, h), octyl glucoside (c, g), dodecyl maltoside (d, i), Nonidet P-40 (f), or Lubrol (j) and subjected to blue native PAGE (5–14% gradient polyacrylamide). The gel was fixed and stained with Coomassie blue.
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
251
FIG. 5. Resolution of OXPHOS complexes by one-dimensional colorless native PAGE. Dodecyl maltoside-solubilized yeast mitochondrial membrane proteins (AB1-4A/8) were separated by colorless native electrophoresis (5–14% polyacrylamide gradient). One part of the gel was stained by Coomassie brilliant blue R (A), and the other part was analyzed by specific ATPase activity staining (B) as described under Materials and Methods. (a, c) and (b, d): supernatants after solubilization of 136 and 220 mg mitochondria, respectively.
mitochondria were solubilized with Triton X-100 or dodecyl maltoside (this band did not correspond to the same complex as the one depicted with Hecameg (see Figs. 2 and 3, band 2); (vi) in all cases, the 245-kDa band remained unchanged; (vii) in yeast foam mitochondria (lanes e–j), changing the detergent did not improve the solubilization of the other OXPHOS complexes. Specific Stainings and Characterization of the OXPHOS Complexes Since it was difficult after BN-PAGE to define the different complexes obtained with dodecyl maltosidesolubilized mitochondria of AB1-4A/8 (cf. Fig. 1), we used CN-PAGE and after migration, the complexes were revealed on one hand by a specific staining related to their peculiar enzymatic activity and on the other hand by Coomassie blue staining. As already shown with mammalian mitochondria (2), the migration on CN-PAGE of the OXPHOS complexes of yeast mitochondria differed from that obtained on BN-PAGE, and under these former conditions of electrophoresis, protein bands were defined by their Rf . Separation by CNPAGE of dodecyl maltoside solubilized AB1-4A/8 mitochondria showed five bands with a major complex at Rf Å 0.27 (Fig. 5A). ATPase activity was revealed directly on the gel using the method of Yoshida et al. (12): at the ATPase complex level, hydrolysis of ATP in the presence of lead acetate led to a white lead phosphate precipitate, allowing visualization of this complex. It can be seen in Fig. 5B that the major band (Rf Å 0.27) corresponded to the ATPase complex; in
aba
AP: Anal Bio
´ RIN GRANDIER-VAZEILLE AND GUE
252
ton X-100 solubilized more cytochrome-containing complexes than did dodecyl maltoside, whereas these differences in solubilization were not observed with ATPase. Resolution of Different Forms of ATPase Complex
FIG. 6. Colorless native electrophoresis and activity staining of detergent-solubilized yeast mitochondrial membrane proteins. Mitochondria of AB1-4A/8 were solubilized by Triton X-100 (a, c, e, g) or by dodecyl maltoside (b, d, f, h) and complexes were separated by colorless native electrophoresis (5–14% polyacrylamide gradient). Strips were then cut out from the gel and stained for ATPase activity (A), NADH dehydrogenase (B), succinate dehydrogenase (C), and cytochrome content (D) as described under Materials and Methods.
addition, a minor band (Rf Å 0.44) was also detected when the gel was overloaded with proteins (lane d). The same experiment was carried out using Triton X100 or dodecyl maltoside as detergent (Fig. 6). To allocate the other bands, the gel was cut out in four strips of two lanes, corresponding to the Triton X-100 or to the dodecyl maltoside supernatant, respectively, and each gel strip was specifically stained. In Fig. 6A, it can be seen that the use of Triton X-100 led to a large ATPase band between Rf Å 0.11 and Rf Å 0.24, whereas with Dodecyl maltoside we observed a well-defined band at Rf Å 0.27. NADH dehydrogenase and succinodehydrogenase, when reducing nitro blue tetrazolium, led to a violet staining of the gel bands according to the substrate added (13). Figure 6B shows the bands revealed when using NADH as substrate. Whatever the detergent used, several bands were revealed: the bands with the lower mobility (Rf Å 0.10) and the higher mobility (Rf Å 0.9) were observed with both detergents. Among additional bands, one with a high activity (Rf Å 0.52) was observed only in Triton X-100 supernatant. With succinate as substrate (Fig. 6C), an active spot (split in three bands) was obtained only with Triton X-100. Finally, Fig. 6D shows the four cytochrome-containing proteins or complexes (bc1 complex, cytochrome c reductase, cytochrome c oxidase, lactate dehydrogenase (cytochrome b2)), stained in blue by tetramethylbenzidine in the presence of hydrogen peroxide (14). To date, assignment of these different bands has not been done. Although this last staining was very labile, it appeared that, as for NADH and succinate dehydrogenases, Tri-
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
When mitochondria from AB1-4A/8 were solubilized using different detergents, analysis of the OXPHOS complexes by BN-PAGE or CN-PAGE showed that the Mapp or the Rf respectively of the major band was different following the detergent used. Therefore, the ATPase activity and the oligomycin sensitivity were determined in the different supernatants after solubilization of mitochondria by the different detergents (Table 1). It should be noted that even though a contamination by plasmic H/-ATPase was observed in some mitochondrial preparations which represented about 15% of the total activity, after solubilization by the different detergents, essentially the mitochondrial ATPase was detected in the supernatants (data not shown). Starting from the same amount of mitochondria, extraction efficiency, as assessed by measuring the total ATPase activity, decreased in the following order: Triton X-100 ú Nonidet P-40 ú dodecyl maltoside ú Lubrol ú Hecameg. It should be noted that only extraction with Hecameg led to an oligomycin-insensitive ATPase. OXPHOS complexes, solubilized by the different detergents, were separated on CN-PAGE and stained for the ATPase activity; in parallel, the gel was stained with Coomassie blue. From Fig. 7 it appeared that solubilized mitochondria revealed an ATPase activity at different Rf depending on the detergent used: with Nonidet P-40 (lane a) or Triton X-100 (lane b) only one large band between Rf Å 0.10 and 0.24 was observed which encompassed the two bands at Rf Å 0.10 and
TABLE 1
Measurements of the ATPase Activities in Detergent-Treated Mitochondria Detergent
Total ATPase activity
% Oligomycin sensitivity
Without Triton X-100 Nonidet P-40 Dodecyl maltoside Lubrol Hecameg
4500 3600 3000 2700 2400 750
95 95 n.d. 95 95 5
Note. Mitochondria were pelleted and then suspended at 10 mg/ ml in a 50 mM Bis–Tris buffer, pH 7, containing 750 mM e-aminocaproic acid and split into six aliquots to which detergents were added in the appropriate detergent/protein ratio. After a clarifying spin (108,000g 1 15 min) of the detergent treated mitochondria, ATPase activity (measured in triplicate and expressed in nmol Pirmin01) was measured on the supernatants.
aba
AP: Anal Bio
ELECTROPHORETIC SEPARATION OF YEAST MITOCHONDRIA
FIG. 7. Action of different detergents on the solubilization of the ATPase complex separated by colorless native PAGE. AB1-4A/8 mitochondria were solubilized either by Nonidet P-40 (a) or Triton X-100 (b) or Lubrol (c) or Dodecyl maltoside (d) or Hecameg (e) or Octyl glucoside (f) and subjected to colorless native electrophoresis (5–14% polyacrylamide gradient). The gel was analyzed by specific ATPase activity staining.
0.24 as revealed by Coomassie blue (data not shown); in the same way, solubilization with Lubrol (lane c) led to a major ATPase active band between Rf Å 0.11 and 0.23. This result differed from the observation carried out with potato tuber mitochondria in which Lubrol failed to solubilize a complex having an ATPase activity (18). With dodecyl maltoside (lane d) only a single band at Rf Å 0.27 was observed, whereas with Hecameg (lane e) or octyl glucoside (lane f), the major complex migrated at Rf Å 0.44. Solubilization of mitochondria by Hecameg led also to a weak activity at Rf Å 0.61: this minor band was not analyzed. Similar results were obtained with yeast foam mitochondria (data not shown). It should be noted that, depending on the detergent used, the same differences in migration were observed in BN-PAGE and CN-PAGE. In order to know if each of the bands, exhibiting an ATPase activity corresponded to different types of ATPase complexes, their subunit composition was analyzed by a two-dimensional Tricine–SDS–PAGE. To reveal all the subunits of the complexes, a number of electrophoresis strips (stained bands) cut out from 1-D BN-PAGE gels were stacked in the 2-D wells. This method was chosen because attempts to increase the protein loading (or the detergent/protein ratio) on the one-dimensional gel caused extensive smearing and decreased the resolution of the system. From Fig. 8A, several lines of evidence can be drawn: (i) After solubilization by Triton X-100 the major band of Mapp 820 kDa (lane a) and the band with the higher Mapp (lane b) (cf. Fig. 4) showed all the subunits of the F1 and FO sector and, in addition, several associated proteins; (ii) the same kind of profile was observed when mitochondria were solubilized by
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
253
Nonidet P-40 (lane e) or dodecyl maltoside (lane f); (iii) solubilization with Lubrol (lane c) led to the purest ATPase, compared to the other detergents, only the e subunit was not revealed on the gel as already described by Arselin et al. (19). In Fig. 8A (lane c), d subunit of the ATPase solubilized with Lubrol appeared as a faint band on the gel; the same experiment was performed again (Fig. 8B) overloading the gel with Lubrol (lane i) and dodecyl maltoside (lane j) supernatant strips, and using as control a pure F1 –FO ATPase (lane h) (all the subunits of this ATPase were assigned using antibodies, Dr. J. Velours, personal communication). Although in this last experiment the four subunits 4, OSCP, d, and 6 were seen as a triplet, the d subunit was observed in both preparations and as already described, Lubrol extract was the purest; (iv) after solubilization of mitochondria by Hecameg (Fig. 8A, lane d), the major band of Mapp 440 kDa exhibited only some of the subunits of the F1 part of the enzyme (a, b, g) and some of the contaminant proteins already observed with the other detergents. Despite overloading the gel, the d subunit was not detected. CONCLUSION
This paper deals with the solubilization by different nonionic detergents of the OXPHOS complexes of yeast
FIG. 8. Two-dimensional electrophoretic analysis of major stained bands obtained after BN-PAGE of AB1-4A/8 solubilized by different detergents. Mitochondria (130 mg) solubilized by Triton X-100 (a, b), Lubrol (c, i), Nonidet P40 (e), Hecameg (d), or dodecyl maltoside (f, j) were separated in the first dimension by BN-PAGE. The major stained bands were separated in the second dimension by Tricine– SDS–PAGE (14% T; 3% C) as described under Materials and Methods. (a) corresponds to the higher Mapp band of Triton X-100 (cf. Fig. 4, lane a); (h) pure F1 –FO ATPase; (g) molecular weight markers (phosphorylase b, bovine serum albumin, ovalbumin, carbonic anhydrase, soybean trypsin inhibitor, and a-lactalbumin). A and B correspond to two different patterns where d and 6 subunits were more or less resolved.
aba
AP: Anal Bio
´ RIN GRANDIER-VAZEILLE AND GUE
254
mitochondria, their separation on gel by nondenaturing PAGE (1, 2) and their characterization directly on the gel by specific stainings related to their enzymatic activities. On the other hand we focused on the F1 –FO ATPase, studying the degree of purity and the composition of the complex solubilized as a function of the detergent used. First, it appeared that using detergent/ protein ratios equivalent to that described for mammalian mitochondria led to a less efficient solubilization of the OXPHOS complexes in yeast mitochondria. This result could be due to a different lipid composition of the inner membrane (20). This explanation could account for the observation that (i) depending on the yeast strain, different patterns of solubilized OXPHOS complexes were obtained, and (ii) it seemed difficult to solubilize cytochrome c oxidase from yeast foam mitochondria with nonionic detergents. Separation of the OXPHOS complexes on CN-PAGE allowed use of specific staining to visualize the different complexes according to their enzymatic activity. It appeared that, although all the detergents solubilized ATPase, Triton X-100 was much more efficient in solubilizing NADH dehydrogenase, succinate dehydrogenase, and cytochrome-containing proteins than dodecyl maltoside. With regard to the F1 –FO ATPase complex, different experimental conditions (detergents, detergent/protein ratio, incubation time) were used to improve the enzyme extration and it appeared that depending on the detergent used, different forms of the complex were obtained. Using optimal conditions, it was observed that Triton X-100, Lubrol, and Nonidet P-40, which all possess a long ethylene glycol ether chain and the same CMC (0.2 mM), solubilized the ATPase complex as two bands with high Mapp , leading to a diffuse staining of the complex after CN-PAGE (Fig. 7); dodecyl maltoside, which differs from the three other detergents in its structure, but which exhibits the same CMC, solubilized only a single form of ATPase complex with a slightly lower Mapp . Among these four detergents, Lubrol raised to the purest F1 –FO ATPase complex. On the other hand, Hecameg and octyl glucoside, which are small molecules compared to the other detergents, and which have a higher CMC (20 mM) solubilized essentially a subcomplex (a, b, and g subunits of F1) which however exhibited an activity as revealed directly on the gel. With these latter detergents, it was never possible to solubilize the whole F1 –FO ATPase complex. In our hands and under the conditions we used, it was not possible to separate FA sector from the FO sector. The fact that we obtained an active F1 AT-
AID
AB 9773
/
6m1f$$$243
10-17-96 07:31:40
Pase without any protein corresponding to the FA part (subunits 4, d, and OSCP) nor to the FO sector (subunits 6, 8, and 9) as defined by Cox et al. (7) suggested a relatively weak interaction between F1 and FA / FO which can be destabilized by small detergent molecules. In conclusion, we obtained either a complete, more or less pure F1 –FO ATPase or only the F1 part, but never an intermediate form. ACKNOWLEDGMENTS The authors are grateful to Dr. J. Velours for the gift of the pure ATPase. This work was supported by grants from the University of Bordeaux II, the Centre National de la Recherche Scientifique, and the Conseil Re´gional d’Aquitaine.
REFERENCES 1. Scha¨gger, H., Cramer, W. A., and von Jagow, G. (1994) Anal. Biochem. 217, 220–230. 2. Scha¨gger, H., and von Jagow, G. (1991) Anal. Biochem. 199, 223–231. 3. Bentlage, H., de Coo, R., ter Laak, H., Sengers, R., Trijbels, F., Ruitenbeek, W., Schlote, W., Pfeiffer, K., Gencic, S., von Jagow, G., and Scha¨gger, H. (1995) Eur. J. Biochem. 227, 909–915. 4. Scha¨gger, H., and Ohm, T. G. (1995) Eur. J. Biochem. 227, 916– 921. 5. Klement, P., Nijtmans, L., Van der Bogert, C., and Houstek, J. (1995) Anal. Biochem. 231, 218–224. 6. Senior, A. E. (1988) Physiol. Rev. 68, 177–231. 7. Cox, G. B., Devenish, R. J., Gibson, F., Howitt, S. M., and Nagley, P. (1992) Molecular Mechanisms in Bioenergetics (Ernster, L., Ed.), pp. 283–315, Elsevier. 8. Hashimoto, T., Yoshida, Y., and Tagawa, K. (1990) J. Bioenerg. Biomemb. 22, 27–38. 9. Grandier-Vazeille, X., Ouhabi, R., and Guerin, M. (1994) Eur. J. Biochem. 223, 521–528. 10. Guerin, B., Labbe, P., and Somlo, M. (1979) Methods Enzymol. 55, 149–159. 11. Ansorge, W. (1983) Electrophoresis 82, 235–242. 12. Yoshida, M., Sone, N., Hirata, H., and Kagawa, Y. (1975) J. Biol. Chem. 250, 7910–7916. 13. Kuonen, D., Roberts, P. J., and Cottingham, I. R. (1986) Anal. Biochem. 153, 221–226. 14. Thomas, P. J., Ryan, D., and Levin, W. (1976) Anal. Biochem. 75, 168–176. 15. Pande, S. V., and Murthy, S. R. (1994) Anal. Biochem. 220, 424– 426. 16. Somlo, M. (1968) Eur. J. Biochem. 5, 276–284. 17. Serrano, R. (1978) Mol. Cell Biochem. 22, 51–63. 18. Valerio, M., Diolez, P., and Haraux, F. (1993) Eur. J. Biochem. 216, 565–571. 19. Arselin, G., Gandar, J. C., Gue´rin, B., and Velours, J. (1991) J. Biol. Chem. 266, 723–727. 20. Daum, G. (1985) Biochim. Biophys. Acta 822, 1–42.
aba
AP: Anal Bio
242, 248–254 (1996)
0460
Separation by Blue Native and Colorless Native Polyacrylamide Gel Electrophoresis of the Oxidative Phosphorylation Complexes of Yeast Mitochondria Solubilized by Different Detergents: Specific Staining of the Different Complexes Xavier Grandier-Vazeille and Martine Gue´rin1 Institut de Biochimie et de Ge´ne´tique Cellulaires du CNRS, Universite´ de Bordeaux II, 1 Rue Camille Saint-Sae¨ns, 33077 Bordeaux, France Received June 27, 1996
Blue native polyacrylamide gel electrophoresis (BN-PAGE) or colorless native polyacrylamide gel electrophoresis (CN-PAGE) allowed separation of the oxidative phosphorylation complexes of yeast mitochondria. These complexes were characterized by specific staining related to their enzymatic activity. Solubilization of mitochondria by different nonionic detergents such as Triton X-100, dodecyl maltoside, Nonidet P-40, Lubrol, octyl glucoside, or Hecameg led to the separation of F1 –FO ATPase complexes exhibiting distinct apparent molecular masses related to different contaminating proteins and lipids. All these different forms were active in ATP hydrolysis as revealed directly on the gel. Analysis of the subunit composition of these complexes was carried out by a twodimensional Tricine–SDS–PAGE and showed that the purest F1 –FO ATPase complex was obtained with Lubrol, whereas with Hecameg and octyl glucoside, only the F1 part of ATPase was solubilized. q 1996 Academic Press, Inc.
Separation of enzymatically active heteromultimeric complexes of the inner mitochondrial membrane was greatly improved by the use of either blue native polyacrylamide gel electrophoresis (BN-PAGE)2 or colorless native polyacrylamide gel electrophoresis (CN-PAGE). 1 To whom correspondence should be addressed. Fax: (33) 56 99 90 59. E-mail: [email protected]. 2 Abbreviations used: BN-PAGE, blue native polyacrylamide gel electrophoresis; CN-PAGE, colorless native polyacrylamide gel electrophoresis; Bis–Tris, 2,2-bis(hydroxymethyl)-2,2 *,29-nitrilotriethanol [bis(2-hydroxyethyl)amino–tris(hydroxymethyl)methane]; OXPHOS, oxidative phosphorylation.
Coomassie blue dye, added to sample and cathode buffers for BN-PAGE, induces a charge shift on the proteins so that the electrophoretic mobility of the proteins is mainly determined by the negative charges of bound Coomassie dye. In CN-PAGE, performed at a determined pH, the intrinsic charge of the proteins allows their separation (1). Further analysis of the polypeptide composition of each of the complexes was performed in a second dimension by Tricine–SDS–PAGE (1, 2). This system was essentially used in analyzing mammalian mitochondria (1, 2) and allowed to screen for OXPHOS defects in human diseases (3–5). We attempted to apply this system to yeast mitochondria and more particularly to the analysis of the F1 –FO ATPase complex. This membrane-bound enzyme catalyzes the synthesis of ATP from ADP and Pi in a reaction driven by the electrochemical proton gradient maintained across the membrane by the respiratory chain. This enzyme is structurally conserved among the different organisms and is composed of two structurally and functionally distinct moieties, F1 and FO . The F1 part, which is composed of five different subunits termed a, b, g, d, and e, carries the catalytic sites. The FO sector is embedded in the membrane and forms a specific proton channel. It is composed of at least six hydrophobic subunits, the subunits 6, 8, and 9 which are mtDNA encoded and the subunits 4, d, and OSCP which are nDNA encoded (for review see Ref. 6). However, according to Cox et al. (7), the FO sector may comprise only the three mitochondrial hydrophobic proteins (subunits 6, 8, and 9) which are presumably membrane integral proteins and a further panel of subunits, whose interaction with the membrane is less well defined, may form the FA sector. This FA sector may comprise subunits b (also denoted subunit 4), d, and OSCP
248
AID
0003-2697/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
AB 9773
/
6m1f$$$241
10-17-96 07:31:40
aba
AP: Anal Bio
ELECTROPHORETIC SEPARATION OF YEAST MITOCHONDRIA
and the proteins involved in the regulation of the activity (8). In an attempt to answer the question of structural signification of FA , we analyzed the subunit composition of active ATPase complexes obtained from yeast mitochondria solubilized with different nonionic detergents, using the nondenaturing polyacrylamide gel electrophoresis techniques (1). MATERIALS AND METHODS
Chemicals SDS, TEMED, Tricine, Bis–Tris, Triton X-100, Nonidet P-40, Lubrol WX, ATP, and oligomycin were purchased from Sigma; Hecameg was from Vegatec; and octyl glucoside (n-octyl b-D-glucopyranoside) and dodecyl maltoside (n-dodecyl-b-D-maltoside) were from Boehringer. Serva blue G (Coomassie blue G-250) and eaminocaproic acid were from Serva. All other chemicals were purchased from Merck. Yeast Strains Strains of Saccharomyces cerevisiae used are the wild-type diploid strain yeast foam and the wild-type haploid strain AB1-4A/8 (MATa, his 4). Cells were grown on a complete medium supplemented with 2% lactate and harvested in mid-exponential growth phase as already described (9).
249
F1 –FO ATPase was a gift of Dr. J. Velours. Two-dimensional (2-D) Tricine–SDS–PAGE was performed overnight at 157C on 14% acrylamide slab gel (28 1 16.5 1 0.075 or 15.5 1 17.5 1 0.075 cm), with pieces or lanes cut out from the one-dimensional gels. These pieces were soaked in 1% SDS, 1% b-mercaptoethanol for 2 h at room temperature, washed four times in water to remove b-mercaptoethanol solution, and loaded at the usual position for stacking gel (1). BN-PAGE and CN-PAGE gels were fixed and stained with 0.025% (mass/vol) Serva blue G in 50% methanol, 10% acetic acid. Two-dimensional gels were fixed and stained with Serva blue G and further by silver staining according to the procedure of Ansorge (11). Specific Staining of the OXPHOS Complexes After CN-PAGE the ATPase activity was located on the gel by incubation in a 50 mM glycine/NaOH buffer, pH 8.6, containing 5 mM ATP, 5 mM MgCl2 , 0.05% lead acetate as described by Yoshida et al. (12). Staining for dehydrogenase activities was performed in a 100 mM Tris/glycine buffer at pH 7.4 containing 1 mg/ml nitro blue tetrazolium and either 100 mM bNADH or 1 mM sodium succinate as appropriate (13). Cytochrome-containing bands were visualized by staining with 2 mM 3,3*,5,5*-tetramethylbenzidine and 30 mM H2O2 in 100 mM sodium acetate buffer at pH 5 (14).
Sample Preparation and Electrophoresis Techniques Mitochondria isolated from spheroplasts according to Gue´rin et al. (10) were sedimented by centrifugation (12,500g 1 10 min), suspended in 50 mM Bis-Tris buffer, pH 7.0, containing 750 mM e-aminocaproic acid, washed twice in this buffer, and pelleted by centrifugation (12,500g 1 10 min). Mitochondria were resuspended at 10 mg/ml in the same buffer and solubilized for 45 min at 47C by adding detergent in a detergent/ protein ratio of 2.5 for dodecyl maltoside, 0.7 for Triton X-100, 1 for Nonidet P-40, 2 for Hecameg and octyl glucoside, and 3 for Lubrol. Insoluble material was removed by centrifugation (108,000g 1 10 min). Separation of complexes by BN-PAGE or CN-PAGE was performed on linear gradient (5–14%) polyacrylamide slab gels (14.5 1 16.5 1 0.075 cm). For BN-PAGE, sample and cathode buffers were added with Serva Blue as described by Scha¨gger and von Jagow (2). For CNPAGE, ponceau S was added to samples before electrophoresis to follow the migration front. BN-PAGE and CN-PAGE were run at 47C in a vertical apparatus as described by Scha¨gger et al. (1). Apparent molecular masses (Mapp) were determined by using molecular mass markers: urease hexamer (545 kDa) and trimer (272 kDa); bovine albumin dimer (132 kDa) and monomer (66 kDa); chicken egg albumin (45 kDa); carbonic anhydrase (29 kDa); a-lactalbumin (14.2 kDa). Pure
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
Other Assays Protein content of mitochondria was measured using the Biuret method. The protein content of samples containing e-amino caproic acid and detergents was measured by a modified micro-Bradford procedure (15) and in other samples with the Pierce BCA protein assay using bovine serum albumin as standard. Mitochondrial and plasmic ATPase activities were measured on solubilized mitochondria according to Somlo (16) and Serrano (17), respectively. RESULTS AND DISCUSSION
Solubilization of Membrane Complexes by Detergents Preliminary experiments were carried out on mitochondria isolated from two yeast strains, yeast foam and AB1-4A/8. Proteins were solubilized by the nonionic detergent dodecyl maltoside and loaded on a nondenaturing gradient polyacrylamide gel (BN-PAGE). Figure 1 shows that although identical solubilization conditions were used, different profiles of separation of the OXPHOS complexes of yeast mitochondria were obtained. In yeast foam mitochondria, essentially one band with an Mapp of about 820 kDa was observed (lane f), whereas in AB1-4A/8, five bands were visualized: among them the three major bands had apparent mo-
aba
AP: Anal Bio
´ RIN GRANDIER-VAZEILLE AND GUE
250
FIG. 1. Resolution of the OXPHOS complexes by one-dimensional blue native PAGE. Mitochondrial membrane proteins from yeast foam (210 mg) and AB1-4A/8 (136 mg) were solubilized by dodecyl maltoside (detergent/protein Å 2 mg/mg protein) for 10 min (b), 30 min (c), 60 min (d, f), and 120 min (e). After centrifugation (108,000g 1 10 min) the supernatant from AB1-4A/8 (b, c, d, e) and yeast foam (f) and the molecular weight standards (a) were resolved by blue native PAGE (5–14% gradient polyacrylamide slab gel) as described under Materials and Methods.
lecular masses of about 820, 440, and 245 kDa (lane b). Increasing incubation time of yeast foam mitochondria in the presence of detergent did not improve solubilization of the OXPHOS complexes (data not shown), whereas in AB1-4A/8 mitochondria, it led to a splitting of the Mapp 820-kDa band (lanes c, d, and e); both bands corresponded to the ATPase complex as seen in an additional 2-D Tricine–SDS–PAGE (data not shown). Since mitochondria from both strains contained equivalent amounts of cytochrome c oxidase as determined by spectrophotometry and equivalent ATPase activities (data not shown), it was hypothesized that both strains had the same amounts of OXPHOS complexes. Thus, the difference observed between the strains could be due to a different lipid composition of the inner membrane. Consequently, mitochondria from both yeast strains were solubilized, using different detergent/protein ratios and varying incubation times. Whatever the conditions used, solubilization of yeast foam mitochondria was not improved (data not shown). With regard to AB1-4A/8 mitochondria, after BN-PAGE, fixation, and restaining of proteins with Coomassie blue (Fig. 2), the densitometric quantification of the stained bands revealed that, for a given dodecyl maltoside/protein ratio, increasing incubation time did not significantly modify the relative amounts of the three major bands, whereas increasing the detergent/protein ratio enhanced the band 1/band 3 ratio from 1.0 to 1.5. In a second step, characterization of the OXPHOS complexes separated by nondenaturing PAGE was per-
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
FIG. 2. Analysis of dodecyl maltoside-solubilized mitochondrial membrane proteins of AB1-4A/8 by blue native PAGE. Mitochondria from AB1-4A/8 (130 mg) were treated with different concentrations of dodecyl maltoside [detergent/protein ratio of 5 (A), 2 (B), and 1 (C)] for 10 min (a, j, o), 30 min (b, i, n), 60 min (c, h, m), 120 min (d, g, l), and 180 min (e, f, k). After a clarifying spin (108,000g 1 15 min) the supernatants were subjected to blue native PAGE (5–14% gradient polyacrylamide) as detailed under Materials and Methods. The gel was fixed and stained with Coomassie brilliant blue R.
formed by a two-dimensional Tricine–SDS–PAGE of individual lanes from the BN-PAGE gel (Fig. 3): it appeared that band 1 (cf. band 1 in Fig. 2) contained most of the characteristic polypeptides of the F1 –FO ATPase complex (a, b, g, 4, OSCP, d, and 9 subunits). It should
FIG. 3. Two-dimensional electrophoresis of the OXPHOS complexes from yeast mitochondria solubilized by dodecyl maltoside. Supernatant from yeast mitochondria (AB1-4A/8) solubilized by dodecyl maltoside (detergent/protein Å 2.5 mg/mg) was separated in the first dimension by blue native PAGE and in the second dimension by Tricine–SDS–PAGE (14% T, 3% C). The gel was fixed and stained with Coomassie brilliant blue R and the complexes were characterized by their polypeptide patterns. 1, 2, and 3 correspond to bands 1, 2, and 3, respectively, in Fig. 2.
aba
AP: Anal Bio
ELECTROPHORETIC SEPARATION OF YEAST MITOCHONDRIA
be noted that resolution of the other bands revealed in the first dimension was very poor and thus it was difficult to define them. In an attempt to improve the solubilization of mitochondrial complexes, other nonionic detergents were used. To allow a standardization of the different assays, the detergent/protein ratio was chosen instead of the only detergent concentration since varying the protein and detergent concentrations for a given detergent/protein ratio did not improve the solubilization. Different detergent/protein ratios and incubation times were assayed on mitochondria isolated from the two yeast strains. From the different assays, it appeared that the best detergent/protein ratios were 0.7 for Triton X-100, 1 for Nonidet P-40, 2 for Hecameg and octyl glucoside, 2.5 for dodecyl maltoside, and 3 for Lubrol. These ratios corresponded to 10 mM Triton X-100, 13 mM Nonidet P-40, 50 mM Hecameg, 48 mM dodecyl maltoside, and 60 mM Lubrol, respectively. From Fig. 4, several lines of evidence can be drawn: (i) whatever the yeast strain, the modifications observed in the migration of the ATPase complex as a function of the detergent used were the same; i.e., Hecameg (lanes b and h) and octyl glucoside (lanes c and g) led to a complex with a lower apparent molecular mass; (ii) starting from the same amount of mitochondria, dodecyl maltoside solubilized a higher amount of ATPase than the other detergents; (iii) in AB1-4A/8 mitochondria (lanes a–d), Triton X-100 (lane a) or Lubrol (not shown) solubilized one additional band with a higher Mapp , compared to solubilization with dodecyl maltoside (lane d); (iv) Hecameg (lane b) or octyl glucoside (lane c) did not reveal the Mapp 820kDa band, whereas the intensity of the Mapp 440-kDa band was increased; (v) a weaker band migrating slightly above Mapp 440-kDa band was observed when
FIG. 4. Effect of different detergents on the solubilization of yeast mitochondria OXPHOS complexes analyzed by blue native electrophoresis. Mitochondria from AB1-4A/8 (a, b, c, d) and from yeast foam (e, f, g, h, i, j) were solubilized by Triton X-100 (a, e), Hecameg (b, h), octyl glucoside (c, g), dodecyl maltoside (d, i), Nonidet P-40 (f), or Lubrol (j) and subjected to blue native PAGE (5–14% gradient polyacrylamide). The gel was fixed and stained with Coomassie blue.
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
251
FIG. 5. Resolution of OXPHOS complexes by one-dimensional colorless native PAGE. Dodecyl maltoside-solubilized yeast mitochondrial membrane proteins (AB1-4A/8) were separated by colorless native electrophoresis (5–14% polyacrylamide gradient). One part of the gel was stained by Coomassie brilliant blue R (A), and the other part was analyzed by specific ATPase activity staining (B) as described under Materials and Methods. (a, c) and (b, d): supernatants after solubilization of 136 and 220 mg mitochondria, respectively.
mitochondria were solubilized with Triton X-100 or dodecyl maltoside (this band did not correspond to the same complex as the one depicted with Hecameg (see Figs. 2 and 3, band 2); (vi) in all cases, the 245-kDa band remained unchanged; (vii) in yeast foam mitochondria (lanes e–j), changing the detergent did not improve the solubilization of the other OXPHOS complexes. Specific Stainings and Characterization of the OXPHOS Complexes Since it was difficult after BN-PAGE to define the different complexes obtained with dodecyl maltosidesolubilized mitochondria of AB1-4A/8 (cf. Fig. 1), we used CN-PAGE and after migration, the complexes were revealed on one hand by a specific staining related to their peculiar enzymatic activity and on the other hand by Coomassie blue staining. As already shown with mammalian mitochondria (2), the migration on CN-PAGE of the OXPHOS complexes of yeast mitochondria differed from that obtained on BN-PAGE, and under these former conditions of electrophoresis, protein bands were defined by their Rf . Separation by CNPAGE of dodecyl maltoside solubilized AB1-4A/8 mitochondria showed five bands with a major complex at Rf Å 0.27 (Fig. 5A). ATPase activity was revealed directly on the gel using the method of Yoshida et al. (12): at the ATPase complex level, hydrolysis of ATP in the presence of lead acetate led to a white lead phosphate precipitate, allowing visualization of this complex. It can be seen in Fig. 5B that the major band (Rf Å 0.27) corresponded to the ATPase complex; in
aba
AP: Anal Bio
´ RIN GRANDIER-VAZEILLE AND GUE
252
ton X-100 solubilized more cytochrome-containing complexes than did dodecyl maltoside, whereas these differences in solubilization were not observed with ATPase. Resolution of Different Forms of ATPase Complex
FIG. 6. Colorless native electrophoresis and activity staining of detergent-solubilized yeast mitochondrial membrane proteins. Mitochondria of AB1-4A/8 were solubilized by Triton X-100 (a, c, e, g) or by dodecyl maltoside (b, d, f, h) and complexes were separated by colorless native electrophoresis (5–14% polyacrylamide gradient). Strips were then cut out from the gel and stained for ATPase activity (A), NADH dehydrogenase (B), succinate dehydrogenase (C), and cytochrome content (D) as described under Materials and Methods.
addition, a minor band (Rf Å 0.44) was also detected when the gel was overloaded with proteins (lane d). The same experiment was carried out using Triton X100 or dodecyl maltoside as detergent (Fig. 6). To allocate the other bands, the gel was cut out in four strips of two lanes, corresponding to the Triton X-100 or to the dodecyl maltoside supernatant, respectively, and each gel strip was specifically stained. In Fig. 6A, it can be seen that the use of Triton X-100 led to a large ATPase band between Rf Å 0.11 and Rf Å 0.24, whereas with Dodecyl maltoside we observed a well-defined band at Rf Å 0.27. NADH dehydrogenase and succinodehydrogenase, when reducing nitro blue tetrazolium, led to a violet staining of the gel bands according to the substrate added (13). Figure 6B shows the bands revealed when using NADH as substrate. Whatever the detergent used, several bands were revealed: the bands with the lower mobility (Rf Å 0.10) and the higher mobility (Rf Å 0.9) were observed with both detergents. Among additional bands, one with a high activity (Rf Å 0.52) was observed only in Triton X-100 supernatant. With succinate as substrate (Fig. 6C), an active spot (split in three bands) was obtained only with Triton X-100. Finally, Fig. 6D shows the four cytochrome-containing proteins or complexes (bc1 complex, cytochrome c reductase, cytochrome c oxidase, lactate dehydrogenase (cytochrome b2)), stained in blue by tetramethylbenzidine in the presence of hydrogen peroxide (14). To date, assignment of these different bands has not been done. Although this last staining was very labile, it appeared that, as for NADH and succinate dehydrogenases, Tri-
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
When mitochondria from AB1-4A/8 were solubilized using different detergents, analysis of the OXPHOS complexes by BN-PAGE or CN-PAGE showed that the Mapp or the Rf respectively of the major band was different following the detergent used. Therefore, the ATPase activity and the oligomycin sensitivity were determined in the different supernatants after solubilization of mitochondria by the different detergents (Table 1). It should be noted that even though a contamination by plasmic H/-ATPase was observed in some mitochondrial preparations which represented about 15% of the total activity, after solubilization by the different detergents, essentially the mitochondrial ATPase was detected in the supernatants (data not shown). Starting from the same amount of mitochondria, extraction efficiency, as assessed by measuring the total ATPase activity, decreased in the following order: Triton X-100 ú Nonidet P-40 ú dodecyl maltoside ú Lubrol ú Hecameg. It should be noted that only extraction with Hecameg led to an oligomycin-insensitive ATPase. OXPHOS complexes, solubilized by the different detergents, were separated on CN-PAGE and stained for the ATPase activity; in parallel, the gel was stained with Coomassie blue. From Fig. 7 it appeared that solubilized mitochondria revealed an ATPase activity at different Rf depending on the detergent used: with Nonidet P-40 (lane a) or Triton X-100 (lane b) only one large band between Rf Å 0.10 and 0.24 was observed which encompassed the two bands at Rf Å 0.10 and
TABLE 1
Measurements of the ATPase Activities in Detergent-Treated Mitochondria Detergent
Total ATPase activity
% Oligomycin sensitivity
Without Triton X-100 Nonidet P-40 Dodecyl maltoside Lubrol Hecameg
4500 3600 3000 2700 2400 750
95 95 n.d. 95 95 5
Note. Mitochondria were pelleted and then suspended at 10 mg/ ml in a 50 mM Bis–Tris buffer, pH 7, containing 750 mM e-aminocaproic acid and split into six aliquots to which detergents were added in the appropriate detergent/protein ratio. After a clarifying spin (108,000g 1 15 min) of the detergent treated mitochondria, ATPase activity (measured in triplicate and expressed in nmol Pirmin01) was measured on the supernatants.
aba
AP: Anal Bio
ELECTROPHORETIC SEPARATION OF YEAST MITOCHONDRIA
FIG. 7. Action of different detergents on the solubilization of the ATPase complex separated by colorless native PAGE. AB1-4A/8 mitochondria were solubilized either by Nonidet P-40 (a) or Triton X-100 (b) or Lubrol (c) or Dodecyl maltoside (d) or Hecameg (e) or Octyl glucoside (f) and subjected to colorless native electrophoresis (5–14% polyacrylamide gradient). The gel was analyzed by specific ATPase activity staining.
0.24 as revealed by Coomassie blue (data not shown); in the same way, solubilization with Lubrol (lane c) led to a major ATPase active band between Rf Å 0.11 and 0.23. This result differed from the observation carried out with potato tuber mitochondria in which Lubrol failed to solubilize a complex having an ATPase activity (18). With dodecyl maltoside (lane d) only a single band at Rf Å 0.27 was observed, whereas with Hecameg (lane e) or octyl glucoside (lane f), the major complex migrated at Rf Å 0.44. Solubilization of mitochondria by Hecameg led also to a weak activity at Rf Å 0.61: this minor band was not analyzed. Similar results were obtained with yeast foam mitochondria (data not shown). It should be noted that, depending on the detergent used, the same differences in migration were observed in BN-PAGE and CN-PAGE. In order to know if each of the bands, exhibiting an ATPase activity corresponded to different types of ATPase complexes, their subunit composition was analyzed by a two-dimensional Tricine–SDS–PAGE. To reveal all the subunits of the complexes, a number of electrophoresis strips (stained bands) cut out from 1-D BN-PAGE gels were stacked in the 2-D wells. This method was chosen because attempts to increase the protein loading (or the detergent/protein ratio) on the one-dimensional gel caused extensive smearing and decreased the resolution of the system. From Fig. 8A, several lines of evidence can be drawn: (i) After solubilization by Triton X-100 the major band of Mapp 820 kDa (lane a) and the band with the higher Mapp (lane b) (cf. Fig. 4) showed all the subunits of the F1 and FO sector and, in addition, several associated proteins; (ii) the same kind of profile was observed when mitochondria were solubilized by
AID
AB 9773
/
6m1f$$$242
10-17-96 07:31:40
253
Nonidet P-40 (lane e) or dodecyl maltoside (lane f); (iii) solubilization with Lubrol (lane c) led to the purest ATPase, compared to the other detergents, only the e subunit was not revealed on the gel as already described by Arselin et al. (19). In Fig. 8A (lane c), d subunit of the ATPase solubilized with Lubrol appeared as a faint band on the gel; the same experiment was performed again (Fig. 8B) overloading the gel with Lubrol (lane i) and dodecyl maltoside (lane j) supernatant strips, and using as control a pure F1 –FO ATPase (lane h) (all the subunits of this ATPase were assigned using antibodies, Dr. J. Velours, personal communication). Although in this last experiment the four subunits 4, OSCP, d, and 6 were seen as a triplet, the d subunit was observed in both preparations and as already described, Lubrol extract was the purest; (iv) after solubilization of mitochondria by Hecameg (Fig. 8A, lane d), the major band of Mapp 440 kDa exhibited only some of the subunits of the F1 part of the enzyme (a, b, g) and some of the contaminant proteins already observed with the other detergents. Despite overloading the gel, the d subunit was not detected. CONCLUSION
This paper deals with the solubilization by different nonionic detergents of the OXPHOS complexes of yeast
FIG. 8. Two-dimensional electrophoretic analysis of major stained bands obtained after BN-PAGE of AB1-4A/8 solubilized by different detergents. Mitochondria (130 mg) solubilized by Triton X-100 (a, b), Lubrol (c, i), Nonidet P40 (e), Hecameg (d), or dodecyl maltoside (f, j) were separated in the first dimension by BN-PAGE. The major stained bands were separated in the second dimension by Tricine– SDS–PAGE (14% T; 3% C) as described under Materials and Methods. (a) corresponds to the higher Mapp band of Triton X-100 (cf. Fig. 4, lane a); (h) pure F1 –FO ATPase; (g) molecular weight markers (phosphorylase b, bovine serum albumin, ovalbumin, carbonic anhydrase, soybean trypsin inhibitor, and a-lactalbumin). A and B correspond to two different patterns where d and 6 subunits were more or less resolved.
aba
AP: Anal Bio
´ RIN GRANDIER-VAZEILLE AND GUE
254
mitochondria, their separation on gel by nondenaturing PAGE (1, 2) and their characterization directly on the gel by specific stainings related to their enzymatic activities. On the other hand we focused on the F1 –FO ATPase, studying the degree of purity and the composition of the complex solubilized as a function of the detergent used. First, it appeared that using detergent/ protein ratios equivalent to that described for mammalian mitochondria led to a less efficient solubilization of the OXPHOS complexes in yeast mitochondria. This result could be due to a different lipid composition of the inner membrane (20). This explanation could account for the observation that (i) depending on the yeast strain, different patterns of solubilized OXPHOS complexes were obtained, and (ii) it seemed difficult to solubilize cytochrome c oxidase from yeast foam mitochondria with nonionic detergents. Separation of the OXPHOS complexes on CN-PAGE allowed use of specific staining to visualize the different complexes according to their enzymatic activity. It appeared that, although all the detergents solubilized ATPase, Triton X-100 was much more efficient in solubilizing NADH dehydrogenase, succinate dehydrogenase, and cytochrome-containing proteins than dodecyl maltoside. With regard to the F1 –FO ATPase complex, different experimental conditions (detergents, detergent/protein ratio, incubation time) were used to improve the enzyme extration and it appeared that depending on the detergent used, different forms of the complex were obtained. Using optimal conditions, it was observed that Triton X-100, Lubrol, and Nonidet P-40, which all possess a long ethylene glycol ether chain and the same CMC (0.2 mM), solubilized the ATPase complex as two bands with high Mapp , leading to a diffuse staining of the complex after CN-PAGE (Fig. 7); dodecyl maltoside, which differs from the three other detergents in its structure, but which exhibits the same CMC, solubilized only a single form of ATPase complex with a slightly lower Mapp . Among these four detergents, Lubrol raised to the purest F1 –FO ATPase complex. On the other hand, Hecameg and octyl glucoside, which are small molecules compared to the other detergents, and which have a higher CMC (20 mM) solubilized essentially a subcomplex (a, b, and g subunits of F1) which however exhibited an activity as revealed directly on the gel. With these latter detergents, it was never possible to solubilize the whole F1 –FO ATPase complex. In our hands and under the conditions we used, it was not possible to separate FA sector from the FO sector. The fact that we obtained an active F1 AT-
AID
AB 9773
/
6m1f$$$243
10-17-96 07:31:40
Pase without any protein corresponding to the FA part (subunits 4, d, and OSCP) nor to the FO sector (subunits 6, 8, and 9) as defined by Cox et al. (7) suggested a relatively weak interaction between F1 and FA / FO which can be destabilized by small detergent molecules. In conclusion, we obtained either a complete, more or less pure F1 –FO ATPase or only the F1 part, but never an intermediate form. ACKNOWLEDGMENTS The authors are grateful to Dr. J. Velours for the gift of the pure ATPase. This work was supported by grants from the University of Bordeaux II, the Centre National de la Recherche Scientifique, and the Conseil Re´gional d’Aquitaine.
REFERENCES 1. Scha¨gger, H., Cramer, W. A., and von Jagow, G. (1994) Anal. Biochem. 217, 220–230. 2. Scha¨gger, H., and von Jagow, G. (1991) Anal. Biochem. 199, 223–231. 3. Bentlage, H., de Coo, R., ter Laak, H., Sengers, R., Trijbels, F., Ruitenbeek, W., Schlote, W., Pfeiffer, K., Gencic, S., von Jagow, G., and Scha¨gger, H. (1995) Eur. J. Biochem. 227, 909–915. 4. Scha¨gger, H., and Ohm, T. G. (1995) Eur. J. Biochem. 227, 916– 921. 5. Klement, P., Nijtmans, L., Van der Bogert, C., and Houstek, J. (1995) Anal. Biochem. 231, 218–224. 6. Senior, A. E. (1988) Physiol. Rev. 68, 177–231. 7. Cox, G. B., Devenish, R. J., Gibson, F., Howitt, S. M., and Nagley, P. (1992) Molecular Mechanisms in Bioenergetics (Ernster, L., Ed.), pp. 283–315, Elsevier. 8. Hashimoto, T., Yoshida, Y., and Tagawa, K. (1990) J. Bioenerg. Biomemb. 22, 27–38. 9. Grandier-Vazeille, X., Ouhabi, R., and Guerin, M. (1994) Eur. J. Biochem. 223, 521–528. 10. Guerin, B., Labbe, P., and Somlo, M. (1979) Methods Enzymol. 55, 149–159. 11. Ansorge, W. (1983) Electrophoresis 82, 235–242. 12. Yoshida, M., Sone, N., Hirata, H., and Kagawa, Y. (1975) J. Biol. Chem. 250, 7910–7916. 13. Kuonen, D., Roberts, P. J., and Cottingham, I. R. (1986) Anal. Biochem. 153, 221–226. 14. Thomas, P. J., Ryan, D., and Levin, W. (1976) Anal. Biochem. 75, 168–176. 15. Pande, S. V., and Murthy, S. R. (1994) Anal. Biochem. 220, 424– 426. 16. Somlo, M. (1968) Eur. J. Biochem. 5, 276–284. 17. Serrano, R. (1978) Mol. Cell Biochem. 22, 51–63. 18. Valerio, M., Diolez, P., and Haraux, F. (1993) Eur. J. Biochem. 216, 565–571. 19. Arselin, G., Gandar, J. C., Gue´rin, B., and Velours, J. (1991) J. Biol. Chem. 266, 723–727. 20. Daum, G. (1985) Biochim. Biophys. Acta 822, 1–42.
aba
AP: Anal Bio