Molecular architecture of the ATP-dependent CodWX protease having an N-terminal serine active site
2003; Springer Nature; Volume: 22; Issue: 12 Linguagem: Inglês
10.1093/emboj/cdg289
ISSN1460-2075
AutoresMin Suk Kang, Soon Rae Kim, Pyeongsu Kwack, Byung Kook Lim, Sung Won Ahn, Young Min Rho, Ihn Sik Seong, S B Park, Soo Hyun Eom, Gang‐Won Cheong, Chin Ha Chung,
Tópico(s)Ubiquitin and proteasome pathways
ResumoArticle16 June 2003free access Molecular architecture of the ATP-dependent CodWX protease having an N-terminal serine active site Min Suk Kang Min Suk Kang NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Soon Rae Kim Soon Rae Kim Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea Search for more papers by this author Pyeongsu Kwack Pyeongsu Kwack Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea Search for more papers by this author Byung Kook Lim Byung Kook Lim NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Sung Won Ahn Sung Won Ahn NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Young Min Rho Young Min Rho NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Ihn Sik Seong Ihn Sik Seong NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Seong-Chul Park Seong-Chul Park Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea Search for more papers by this author Soo Hyun Eom Soo Hyun Eom Department of Life Science, Kwangju Institute of Science and Technology, Kwangju, 500-756 Korea Search for more papers by this author Gang-Won Cheong Corresponding Author Gang-Won Cheong Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea Search for more papers by this author Chin Ha Chung Corresponding Author Chin Ha Chung NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Min Suk Kang Min Suk Kang NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Soon Rae Kim Soon Rae Kim Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea Search for more papers by this author Pyeongsu Kwack Pyeongsu Kwack Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea Search for more papers by this author Byung Kook Lim Byung Kook Lim NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Sung Won Ahn Sung Won Ahn NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Young Min Rho Young Min Rho NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Ihn Sik Seong Ihn Sik Seong NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Seong-Chul Park Seong-Chul Park Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea Search for more papers by this author Soo Hyun Eom Soo Hyun Eom Department of Life Science, Kwangju Institute of Science and Technology, Kwangju, 500-756 Korea Search for more papers by this author Gang-Won Cheong Corresponding Author Gang-Won Cheong Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea Search for more papers by this author Chin Ha Chung Corresponding Author Chin Ha Chung NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea Search for more papers by this author Author Information Min Suk Kang1, Soon Rae Kim2, Pyeongsu Kwack2, Byung Kook Lim1, Sung Won Ahn1, Young Min Rho1, Ihn Sik Seong1, Seong-Chul Park2, Soo Hyun Eom3, Gang-Won Cheong 2 and Chin Ha Chung 1 1NRL of Protein Biochemistry, School of Biological Sciences, Seoul National University, Seoul, 151-742 Korea 2Division of Applied Life Science, Gyeongsang National University, Chinju, 660-701 Korea 3Department of Life Science, Kwangju Institute of Science and Technology, Kwangju, 500-756 Korea *Corresponding authors. E-mail: [email protected] or E-mail: [email protected] The EMBO Journal (2003)22:2893-2902https://doi.org/10.1093/emboj/cdg289 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info CodWX in Bacillus subtilis is an ATP-dependent, N-terminal serine protease, consisting of CodW peptidase and CodX ATPase. Here we show that CodWX is an alkaline protease and has a distinct molecular architecture. ATP hydrolysis is required for the formation of the CodWX complex and thus for its proteolytic function. Remarkably, CodX has a ‘spool-like’ structure that is formed by interaction of the intermediate domains of two hexameric or heptameric rings. In the CodWX complex, CodW consisting of two stacked hexameric rings (WW) binds to either or both ends of a CodX double ring (XX), forming asymmetric (WWXX) or symmetric cylindrical particles (WWXXWW). CodWX can also form an elongated particle, in which an additional CodX double ring is bound to the symmetric particle (WWXXWWXX). In addition, CodWX is capable of degrading EzrA, an inhibitor of FtsZ ring formation, implicating it in the regulation of cell division. Thus, CodWX appears to constitute a new type of protease that is distinct from other ATP-dependent proteases in its structure and proteolytic mechanism. Introduction ATP-dependent proteases are responsible for the degradation of the majority of proteins in cells, including misfolded proteins and short-lived regulatory proteins, thus playing essential roles in protein quality control and regulation of a variety of cellular processes (Goldberg, 1992; Gottesman, 1996). In eukaryotes, the 26S proteasome, which consists of the 20S proteasome forming a proteolytic core and the 19S regulatory complex harboring multiple ATPases, is responsible for the degradation of most of the cell proteins that are ubiquitylated (Seufert and Jentsch, 1992; Coux et al., 1996; Hochstrasser, 1996; Baumeister et al., 1998). In this process, ATP is utilized in the ubiquitylation of target proteins, unfolding of the protein substrates by the 19S complex, and translocating them into the inner proteolytic chamber of the 20S proteasome (Benaroudj and Goldberg, 2000; Benaroudj et al., 2003). Although bacteria lack the ubiquitin–proteasome pathway, they contain a number of architecturally related ATP-dependent proteases that also function in the elimination of abnormal proteins and the rapid degradation of regulatory proteins. These enzymes include the ClpAP, ClpXP and HslVU (ClpQY) proteases (Goldberg, 1992; Chung, 1993; Gottesman, 1996; Chung et al., 1997). Of these, HslVU is a homolog of the eukaryotic 26S proteasome (Seemüller et al., 1995; Missiakas et al., 1996; Rohrwild et al., 1996; Yoo et al., 1996). HslU, a member of the Clp/Hsp100 family of proteins, is an ATPase and has a chaperone activity (Seol et al., 1997; Neuwald et al., 1999; Seong et al., 2000). It interacts with and activates the proteolytic activity of HslV, which alone is a weak peptidase. Other members of the Clp/Hsp100 family include the ClpA and ClpX ATPases, which activate ClpP, their common protease (Gottesman et al., 1997, 1998). These multicomponent ATP-dependent proteases require ATP hydrolysis for the degradation of native, folded proteins, but not for the cleavage of small peptides and some unfolded polypeptides. Therefore, it has generally been known that the ATPase activity of the Clp/Hsp100 family members functions in unfolding of protein substrates and channeling them into the inner proteolytic chamber of their cognate proteases. Electron micrographs of HslV have revealed that the peptidase forms a double ring of two stacked hexamers (Rohrwild et al., 1997). On the other hand, HslU was shown to exist as both hexameric and heptameric rings. In the HslVU complex, HslV and HsIU form a cylindrical four-ring structure, in which the HslV dodecamer is flanked at each end by a HslU ring (Kessel et al., 1996; Rohrwild et al., 1997; Ishikawa et al., 2000). Recently, X-ray crystallographic studies have revealed that HslU is folded into three distinct domains: the N-terminal (N), intermediate (I) and C-terminal (C) domains (Bochtler et al., 2000), and that in the HslVU complex, two HslU hexamers bind to both ends of the HslV dodecamer in a U6V6V6U6 configuration (Sousa et al., 2000; Wang et al., 2001a). These results eliminate the possibility that HslU might also exist as a heptameric form. Electron microscopic and X-ray crystallographic studies have also revealed that two ClpA or ClpX hexamers bind to both ends of the ClpP tetradecamer in a similar configuration, thus leading to a symmetric mismatch in their complex (Wang et al., 1997; Beuron et al., 1998; Grimaud et al., 1998). The crystal structures of the HslVU complex with respect to the orientation of the HslU ATPase recently have been a matter of debate. In the earlier structure of HslVU (Bochtler et al., 2000), the ATP-binding rings of HslU are located distal to the HslV peptidase, with the I domains bridging the interstitial space between them. In contrast, two hexameric ATP-binding rings of HslU bind intimately to opposite sides of the HslV peptidase and the HslU I-domains extend outward from the complex in the structure obtained by others (Sousa et al., 2000; Wang et al., 2001b). Recent biochemical evidence showing that the activation of HslV requires its binding to the C-terminal tail region of HslU located opposite the I domain and that the deletion of seven amino acids from the C-terminus prevents the formation of the HslVU complex indicates that the HslVU structure, in which HslV is contacted by HslU through the N- and C-terminal domains, is relevant (Seong et al., 2002). Both CodW and CodX in Bacillus subtilis display >50% identity in their amino acid sequences with HslV and HslU in Escherichia coli, respectively (Slack et al. 1995). Recently, we have demonstrated that the CodW peptidase and the CodX ATPase can function together as a novel type of two-component ATP-dependent protease (Kang et al., 2001). Significantly, CodW uses the N-terminal serine hydroxyl group as the catalytic nucleophile, unlike HslV in E.coli and the β-type proteasome subunits, which use the N-terminal threonine as a nucleophile (Fenteany et al., 1995; Seemüller et al., 1995; Yoo et al., 1997b). In the present studies, we show that CodWX is a distinctive protease, not only in the use of the N-terminal serine active site but also in its molecular architecture and proteolytic mechanism, as compared with the other known ATP-dependent proteases, including HslVU. Results CodWX is an alkaline protease When comparing the rate of peptide hydrolysis using N-carbobenzoxy (Cbz)-Gly-Gly-Leu-7-amido-4-methylcoumarin (AMC) as a substrate, CodWX was found to cleave the substrate at pH 8 only ∼20% as well as HslVU. CodWX and HslVU hydrolyzed the peptide at the rates of 26 and 131 pmol/min, respectively, upon incubation with the same amount of the enzymes (i.e. 0.1 μg of the peptidases and 0.4 μg of the ATPases). Since Bacillus is known to have numerous hydrolases with alkaline pH optima, the effect of pH on the peptide hydrolysis was examined. As shown in Figure 1A, CodWX was maximally active at pH 9.5 in the hydrolysis of both Cbz-Gly-Gly-Leu-AMC and ATP. On the other hand, HslVU showed the highest activities at pH between 8.0 and 8.5. These results indicate that CodWX is an alkaline protease. Thus, all the subsequent experiments were performed at the optimum pHs for the enzymes (i.e. at pH 9.5 using sodium borate buffer for CodWX and at pH 8.0 using HEPES buffer for HslVU). Figure 1.Catalytic activities of CodWX and HslVU. (A) Hydrolysis of Cbz-Gly-Gly-Leu-AMC (left panel) and ATP (right panel) was assayed at various pHs by incubation of 0.1 μg of CodW and 0.4 μg of CodX (open circles) or 0.1 μg of HslV and 0.4 μg of HslU (filled circles) for 10 min at 37°C in the presence of 1 mM ATP as described in Materials and methods. The buffers used were 20 mM HEPES for pH from 7.0 to 8.5 and 20 mM sodium borate for pH from 8.5 to 10.5. (B) Hydrolysis of Cbz-Gly-Gly-Leu-AMC by CodWX (left panel) and HslVU (right panel) was assayed at pH 9.5 and pH 8.0, respectively, as above, but in the presence of ATP or its non-hydrolyzable analogs and in the absence (open bar) or presence of 0.2 M KCl (closed bar). The activities of the enzymes were expressed relative to those seen with ATP only, which were expressed as 100%. Download figure Download PowerPoint Requirement for ATP cleavage for peptide hydrolysis by CodWX HslVU hydrolyzes Cbz-Gly-Gly-Leu-AMC and insulin B-chain more efficiently in the presence of ATPγS, a non-hydrolyzable ATP analog, than with ATP (Yoo et al., 1997a). β,γ-imido-ATP, which also is a non-hydrolyzable ATP analog, can also support the proteolytic activity of HslVU better than ATP, but only in the presence of monovalent cations such as K+ (Huang and Goldberg, 1997). Moreover, a mutant form of HslU that is unable to cleave ATP supports the HslV-mediated hydrolysis of Cbz-Gly-Gly-Leu-AMC and insulin B-chain by forming the HslVU complex in the presence of ATP but not in its absence (Yoo et al., 1998). Therefore, it has been suggested that ATP binding but not its cleavage is essential for the proteolytic activity of HslVU at least on peptides and protein substrates that do not have appreciable secondary structure. To determine whether the peptidase activity of CodWX could also be uncoupled from ATP hydrolysis, the enzyme activity was assayed in the presence of ATP analogs. As shown in Figure 1B, CodWX did not exhibit any activity in the presence of either ATPγS or β,γ-imido-ATP whether it was treated with KCl or not. These results indicate that ATP hydrolysis is required for the peptidase activity of CodWX, unlike that of HslVU. Oligomeric nature of CodW and CodX X-ray crystallographic studies have revealed that HslU has a hexameric ring structure and HslV is a dodecamer consisting of two stacks of hexameric rings (Bochtler et al., 2000; Sousa et al., 2000; Wang et al., 2001a). Gel filtration analysis under non-denaturing conditions has shown that the purified CodW has a molecular mass of ∼250 kDa, corresponding to a size of a dodecamer of 20 kDa subunits (Kang et al., 2001). These results imply that CodW, like HslV, is composed of two stacked hexameric rings. In order to determine the oligomeric nature of CodX, the purified ATPase was also subjected to gel filtration on a Superose-6 column equilibrated with 0.1 M NaCl, which is a typical salt concentration used for gel filtration analysis under non-denaturing conditions. For comparison, HslU was also chromatographed under similar experimental conditions. As reported previously (Yoo et al., 1996), HslU was recovered in the fractions corresponding to the size of a hexamer or heptamer (i.e. 300–350 kDa) in the presence of ATP, but was present as a monomer or dimer in its absence (Figure 2A). In contrast, CodX was eluted in the fractions corresponding to about twice the size of HslU (650–700 kDa) whether ATP is present or not, suggesting that the isolated CodX may behave as a dodecamer or tetradecamer of 52 kDa subunits. Figure 2.Gel filtration analysis for oligomerization of CodX in the presence of NaCl, ATP or both. (A) Purified CodX or HslU (0.2 mg each) was incubated at 37°C for 1 h in the absence (open circles) or presence of 1 mM ATP (filled circles). After incubation, each sample was subjected to gel filtration on a Superose-6 HR 10/30 column that had been equilibrated with 20 mM sodium borate pH 9.5 for CodX or 20 mM HEPES pH 8.0 for HslU. These buffers also contained 0.1 M NaCl, 10 mM MgCl2, 1 mM DTT and 1 mM EDTA. Fractions of 0.4 ml were collected, and aliquots of them were assayed for protein concentration. (B) CodX was chromatographed as above but using sodium borate buffer containing 0.1 or 0.3 M NaCl in the absence or presence of 1 mM ATP. Fractions were collected as above, and aliquots were subjected to SDS–PAGE followed by staining with Coomassie blue R-250. The size markers used were thyroglobulin (669 kDa), β-galactosidase (443 kDa), catalase (232 kDa) and bovine serum albumin (67 kDa). Download figure Download PowerPoint To examine whether salt concentration might influence the oligomerization state of CodX, gel filtration was performed as above but in the presence of a higher salt concentration (i.e. 0.3 M NaCl). When ATP was present, CodX persistently ran as a complex of 650–700 kDa whether the salt was added or not (Figure 2B). Without ATP, however, addition of 0.3 M NaCl resulted in complete dissociation of CodX into monomeric subunits. These results suggest that ionic interaction between the CodX subunits is primarily responsible for their oligomerization. ADP and non-hydrolyzable ATP analogs, including ATPγS and β,γ-imido-ATP, also supported the oligomerization of CodX in the presence of 0.3 M NaCl, whereas AMP could not (data not shown). Thus, it appears that ATP binding, but not its hydrolysis, is required for the oligomerization of the CodX subunits. Requirement for ATP hydrolysis for CodWX complex formation To determine whether CodW and CodX can form a stable CodWX complex under in vitro conditions, cross-linking analysis was performed by incubation of CodW, CodX or both with glutaraldehyde in the absence or presence of various adenine nucleotides. The cross-linked products were then separated by SDS–PAGE on gradient gels followed by silver staining. For comparison, HslV and HslU were also treated as above (Figure 3A, left panel). Like HslV, CodW alone could form a dodecamer as well as a hexamer (Figure 3A, right panel). On the other hand, in the presence of ATP, CodX ran on the gel much more slowly than HslU. Moreover, CodX behaved as a protein even larger than the HslVU complex (V12U6 with a size of ∼500 kDa). These results are consistent with the finding that CodX alone behaves on a gel filtration column as a protein of 650–700 kDa (see Figure 2). CodX also migrated much more slowly than HslU in the presence of ADP or ATP analogs (data not shown), again indicating that the binding of the adenine nucleotides is sufficient for CodX oligomerization. These results suggest that CodX is composed of two hexameric or heptameric rings, unlike HslU consisting of a single hexameric ring. Figure 3.Cross-linking and gel filtration analyses for formation of CodWX and HslVU complexes in the presence of various adenine nucleotides. (A) In the left panel, HslV (2 μg), HslU (2 μg) or both were incubated in 20 mM HEPES pH 8.0 with 0.4% glutaraldehyde for 30 min at 37°C in the absence or presence of ATP, ADP or ATPγS. In the right panel, CodW (2 μg), CodX (2 μg) or both were also incubated as above but in 20 mM sodium borate pH 9.5. After incubation, the samples were subjected to SDS–PAGE followed by silver staining. Note that, for unknown reasons, HslV and CodX were stained much more poorly than HslU and CodW, respectively, by silver nitrate. (B) CodW (50 μg) and CodX (150 μg) were incubated with 0.1 mM ATP (a) or 0.1 mM ATPγS (b) for 30 min at 37°C. CodX with 0.1 mM ATP (c) or CodW alone (d) was also incubated for the same period. After incubation, each sample was chromatographed on a Superose-6 column that had been equilibrated with 20 mM sodium borate buffer pH 9.5 containing 10 mM MgCl2, 1 mM EDTA and 0.1 M NaCl. Note that the elution of CodX does not exactly overlap with that of CodW, most probably due to dissociation of a part of the CodWX complex during gel filtration. Aliquots of the column fractions were subjected to SDS–PAGE on 13% slab gels. Proteins in the gels were then stained with Coomassie blue R-250. Download figure Download PowerPoint When cross-linking analysis was carried out after incubation of CodW and CodX in the presence of ATP, a protein complex that ran even more slowly than CodX appeared in the gels of SDS–PAGE (Figure 3A, right panel). The changes in the relative amounts of CodW or CodX did not show any influence on the formation of the protein complex in the presence of ATP. However, no such complex was formed when CodX and CodW were incubated in the presence of ADP or ATPγS. β,γ-imido-ATP also could not support the formation of the large protein complex (data not shown). These results strongly suggest that the formation of the CodWX complex requires not only ATP binding but also its hydrolysis, unlike that of HslVU which requires ATP binding only. To determine whether the CodWX complex formed upon cross-linking in the presence of ATP might have proteolytic activity, the cross-linked sample just prior to loading onto the gels was assayed for the hydrolysis of Cbz-Gly-Gly-Leu-AMC. It cleaved the peptide ∼85% as well as the mixture of CodW and CodX incubated without glutaraldehyde. Thus, it is likely that the CodWX complex generated by incubation of CodW and CodX is a catalytically active form. In order to confirm whether ATP cleavage is essential for CodWX complex formation, CodW, CodX or both were incubated with ATP or ATPγS. After incubation, the samples were subjected to gel filtration on a Superose-6 column, followed by SDS–PAGE of the column fractions. CodW and CodX that had been incubated in the presence of ATP ran as much larger proteins (Figure 3B, a) than those incubated and chromatographed individually (Figure 3B, c and d). In contrast, CodW and CodX that had been incubated with ATPγS behaved as if each of them was incubated and chromatographed separately (Figure 3B, b). Similar results were obtained when the same experiments were performed upon incubation with β,γ-imido-ATP (data not shown). These results again indicate that ATP hydrolysis is required for CodWX complex formation. Thus, it appears that the inability of CodW and CodX to hydrolyze Cbz-Gly-Gly-Leu-AMC in the presence of ATP analogs (see Figure 1B) is due to their inability to form the CodWX complex without ATP hydrolysis. Both the cross-linking and gel filtration analyses suggest that CodX behaves as a dimer of two hexameric or heptameric rings, unlike HslU consisting of a single hexameric ring. To determine whether the CodX double ring (XX) could be dissociated into two single rings (X), CodX was treated with increasing concentrations of Triton X-100 and then subjected to cross-linking analysis. At the detergent concentrations >0.01% (w/v), CodX ran on the SDS–polyacrylamide gel to the same position as HslU (Figure 4A). Moreover, a CodWX complex that has the same size as HslVU (i.e. UVV) was formed upon incubation of CodX with CodW in the presence of 0.01% Triton X-100, suggesting that the complex is composed of a CodX single ring and a CodW double ring. However, the CodWX complex (WWX) was gradually disassembled upon increasing the detergent concentration. These results suggest that a hydrophobic interaction may be responsible for the dimerization of two CodX single rings and for the interaction between CodX and CodW. Figure 4.Effect of Triton X-100 on formation of the CodWX complex and its peptidase activity. (A) The cross-linking analysis was performed as in Figure 3A but in the presence of ATP (left panel) and also with increasing concentrations of Triton X-100 (right panel). (B) Hydrolysis of Cbz-Gly-Gly-Leu-AMC (open circles) and ATP (filled circles) was assayed by incubation of 0.1 μg of CodW and 0.4 μg of CodX for 10 min at 37°C in the presence of 1 mM ATP and increasing concentrations of Triton X-100. Download figure Download PowerPoint We then examined the effects of increasing concentrations of Triton X-100 on peptide hydrolysis. Figure 4B shows that the peptidase activity determined in the presence of the detergent up to 0.02% is nearly identical to that seen in its absence, suggesting that the CodWX complex containing a CodX single ring (WWX) is as active as that having a CodX double ring (WWXX) at least on the peptide substrate. However, the peptidase activity was gradually decreased upon increasing the detergent concentration >0.02%. Since CodX retained its single ring form (X) at all concentrations of Triton X-100 tested, it is likely that the decrease in the peptidase activity at high concentrations of Triton X-100 is due to disassembly of the CodWX complex (WWX) into a CodX single ring (X) and a CodW double ring (WW). Upon treatment with Triton X-100, the CodX double ring generated in the presence of ATPγS could also be dissociated into single rings, but was unable to form any complex with CodW, unlike HslU (data not shown). Thus, ATP hydrolysis appears essential for CodWX complex formation also in the presence of the detergent. Architecture of the CodWX complex To determine the structure of the CodWX complex as well as its components, electron microscopic analysis was performed. The electron micrographs of the negatively stained CodW oligomer showed two basic views, ring-shaped and two-layered structures, depending on the different orientation on the grid (Figure 5A). A total of 884 top-on views of well-stained CodW particles were translationally and rotationally aligned, and analyzed for rotational symmetry on the basis of eigenvector–eigenvalue and correlation average (van Heel and Frank, 1981). The average image of 884 top-on views revealed an unequivocal 6-fold symmetry (Figure 5B). No other statistically significant symmetry, in particular no 7-fold rotational symmetry, could be detected. The average of 294 side-on views of CodW showed a two-layered structure with an equal distribution of mass across the equatorial plane of the oligomer (Figure 5C). In addition, the architecture and dimensions of CodW (11 nm diameter × 10 nm height) are similar to the average of the HslV dodecamer (Rohrwild et al., 1997). These results suggest that CodW is a dodecamer consisting of two stacked hexameric rings with a dyad across the equatorial plane of the molecule. Figure 5.Electron micrographs and correlation averages of negatively stained CodW. (A) The electron micrograph was obtained by negative staining of purified CodW with 2% uranyl acetate as described in Materials and methods. (B) The correlation average of 884 top-on views shows six centers of masses arranged on a ring, which form a stain-filled cavity. The diameters of the ring and central cavity are ∼11 and 3 nm, respectively. (C) The average of 294 side-on views shows a two-layered structure with a dyad across the equatorial plane. The height and width of the particle are 10 and 11 nm, respectively. Download figure Download PowerPoint The electron micrographs of the negatively stained CodX oligomer also showed two basic views, ring-shaped and striated structures, depending on the different orientation on the grid (Figure 6A). A total of 431 top-on views of well-stained CodX particles were translationally aligned and subjected to multivariate statistical analysis (van Heel and Frank, 1981). The eigenimages obtained from translationally, but not rotationally, aligned images revealed both 6-fold (Figure 6B, a, numbers 4, 5, and 7) and 7-fold rotational symmetry (numbers 8 and 9). Using the 10 most significant eigenvectors, 16 classes were discriminated according to the similarity of features after rotational alignment but without application of any symmetrization. The class averages revealed that 57% of the total particles have 6-fold rotational symmetry, while 38% show 7-fold symmetry (Figure 6B, b). The remaining particles (5%), particularly numbers 5, 12, 14 and 16, showed heterogeneous images, perhaps due to incomplete stain embedding or to an unintentional inclination during preparation or microscopy. The average images of the top-on views show six or seven centers of mass arranged on a ring with low density in its center (Figure 6C, a and b, respectively). Figure 6.Electron micrographs and multivariate statistical analysis of CodX. (A) The purified CodX protein was incubated with 1 mM ATP and cross-linked by treatment with glutaraldehyde as described in Materials and methods. The electron micrograph of the protein was then obtained by negative staining with 2% uranyl acetate. The arrowheads indicate representative side-on views of CodX. (B) In (a), the average (Av) of translationally, but not rotationally, aligned 431 particles with end-on orientation and 10 most significant eigenimages (numbers 1–10) are shown. In (b), the non-symmetrized class averages (numbers 1–16) were derived from rotationally aligned images using the 10 most significant eigenvectors. The numerals shown below the class averages are the number of particles seen in each class. In (c), the non-symmetrized class averages were derived from the 466 negatively stained particles with side-on orientation. (C) The class averages of top-on views were grouped according to the similarity of rotational symmetry based on the graphs of the angular correlation coefficient, and then averaged. The averaged images of top-on views show six (a) and seven (b) centers of masses arranged on a ring with low density in its center. The numbers of particles used for averaging were 244 and 164 for hexameric and heptameric forms, respectively. The average of 466 side-on views is also shown after symmetrization (c). The diameter and height of CodX are ∼14 and 17.5 nm, respectively. Download fi
Referência(s)