Portal de Periódicos da CAPES

Stable association of chloroplastic precursors with protein translocation complexes that contain proteins from both envelope membranes and a stromal Hsp100 molecular chaperone

1997; Springer Nature; Volume: 16; Issue: 5 Linguagem: Inglês

10.1093/emboj/16.5.935

1460-2075

Erik Nielsen,

Photosynthetic Processes and Mechanisms

Article1 March 1997free access Stable association of chloroplastic precursors with protein translocation complexes that contain proteins from both envelope membranes and a stromal Hsp100 molecular chaperone Erik Nielsen Erik Nielsen MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA Search for more papers by this author Mitsuru Akita Mitsuru Akita MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA Search for more papers by this author Jennifer Davila-Aponte Jennifer Davila-Aponte MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA Search for more papers by this author Kenneth Keegstra Corresponding Author Kenneth Keegstra MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA Search for more papers by this author Erik Nielsen Erik Nielsen MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA Search for more papers by this author Mitsuru Akita Mitsuru Akita MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA Search for more papers by this author Jennifer Davila-Aponte Jennifer Davila-Aponte MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA Search for more papers by this author Kenneth Keegstra Corresponding Author Kenneth Keegstra MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA Search for more papers by this author Author Information Erik Nielsen1, Mitsuru Akita1, Jennifer Davila-Aponte1 and Kenneth Keegstra 1 1MSU-DOE Plant Research Laboratory, Michigan State University, East Lansing, MI, 48824 USA The EMBO Journal (1997)16:935-946https://doi.org/10.1093/emboj/16.5.935 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Cytoplasmically synthesized precursors interact with translocation components in both the outer and inner envelope membranes during transport into chloroplasts. Using co-immunoprecipitation techniques, with antibodies specific to known translocation components, we identified stable interactions between precursor proteins and their associated membrane translocation components in detergent-solubilized chloroplastic membrane fractions. Antibodies specific to the outer envelope translocation components OEP75 and OEP34, the inner envelope translocation component IEP110 and the stromal Hsp100, ClpC, specifically co-immunoprecipitated precursor proteins under limiting ATP conditions, a stage we have called docking. A portion of these same translocation components was co-immunoprecipitated as a complex, and could also be detected by co-sedimentation through a sucrose density gradient. ClpC was observed only in complexes with those precursors utilizing the general import apparatus, and its interaction with precursor-containing translocation complexes was destabilized by ATP. Finally, ClpC was co-immunoprecipitated with a portion of the translocation components of both outer and inner envelope membranes, even in the absence of added precursors. We discuss possible roles for stromal Hsp100 in protein import and mechanisms of precursor binding in chloroplasts. Introduction Most chloroplastic proteins are encoded by nuclear genes and are translated on cytoplasmic ribosomes. To reach their correct position, these proteins must be transported across the double membrane system surrounding chloroplasts (Chua and Schmidt, 1978; Highfield and Ellis, 1978). These proteins are synthesized as precursors containing an N-terminal transit peptide responsible for their targeting (Schmidt et al., 1979). Translocation of precursors into chloroplasts can be divided into two discernible steps based on their differing energy requirements. The first is association of a precursor with the chloroplastic translocation apparatus, and the second is transport across the membranes. Stable association of precursors with the translocation apparatus requires low levels of ATP or other NTPs (Olsen et al., 1989), and results in the irreversible interaction of precursors with the chloroplastic envelopes. At this stage, the precursor remains susceptible to exogenous protease and the transit peptide is not cleaved by the stromal processing peptidase, indicating that the precursor has not completely traversed the envelope membranes (Cline et al., 1985). Translocation of precursors across the envelope membranes can be initiated by raising stromal ATP concentrations (Pain and Blobel, 1987; Theg et al., 1989). After a precursor has traversed the envelope membranes, the transit peptide is proteolytically removed by a stromal processing peptidase, producing a mature-sized protein in the stromal compartment (Reed et al., 1990). Translocation of precursors across the two chloroplastic envelope membranes is thought to occur simultaneously at 'contact sites' (Schnell and Blobel, 1993), a term given to regions where both envelope membranes are found in close physical proximity. By analogy with mitochondria, where precursors must also cross two membranes, precursors at contact sites are thought to interact with proteinaceous complexes from both the inner and outer membranes (for review, see Schatz and Dobberstein, 1996). In mitochondria, translocation complexes from the outer and inner membranes can act independently from one another, forming contact sites only when precursors associate with both complexes simultaneously (Segui-Real et al., 1993; Horst et al., 1995). Whether simultaneous engagement is required in chloroplasts is presently unknown. Recent work on the chloroplastic protein import apparatus has resulted in the identification of several components of the envelope-based translocation complex (for reviews, see Gray and Row, 1995; Schnell, 1995). Several of these translocation complex members including the outer envelope membrane proteins OEP86, OEP75 and OEP34, and the inner envelope-membrane protein IEP110, have been identified, and their corresponding cDNAs have been cloned (Hirsch et al., 1994; Schnell et al., 1994; Seedorf et al., 1995; Tranel et al., 1995; Luübeck et al., 1996). All of these proteins are integral membrane proteins, and only OEP86 and OEP34 show any sequence homology with other proteins that could provide insight into their functions. Both OEP86 and OEP34 contain GTP-binding motifs and can bind GTP (Hirsch et al., 1994; Kessler et al., 1994; Seedorf et al., 1995). The exact functions of these proteins have yet to be determined but, based on biochemical and structural features, OEP86 has been proposed to be the transit peptide receptor protein, and OEP75 may represent a translocation pore (Perry and Keegstra, 1994). Chloroplasts also contain soluble molecular chaperones that may play a role in protein import. A cDNA encoding the major stromal Hsp70 (S78) has been isolated and shows significant sequence similarity with both bacterial and mitochondrial Hsp70 homologs (Marshall and Keegstra, 1992). Also, a cDNA clone encoding a stromal Hsp100 (ClpC) has been isolated and shows high sequence similarity with prokaryotic Hsp100s (Moore and Keegstra, 1993). Given the central role chaperones play in protein translocation in other systems (for review, see Schatz and Dobberstein, 1996), we sought to evaluate the hypothesis that stromal chaperones were involved in protein transport. We identified interactions of both stromal ClpC and S78 with translocation complexes. However, only complexes with ClpC were effectively solubilized in non-ionic detergent. This complex formed under binding and import conditions, and was influenced by the presence of ATP. The implications of these results for the formation of translocation complexes during binding is considered. Results Translocation components of the outer membrane, inner membrane and stroma form a stable complex with the precursor under binding conditions In vitro import of precursors into intact chloroplasts can be halted at the envelope membranes in the presence of low ATP concentrations (Olsen et al., 1989). These bound precursors form a stable interaction with the chloroplastic protein transport machinery, but can be fully imported when internal ATP concentrations are increased to adequate levels (Cline et al., 1985; Olsen and Keegstra, 1992). We sought to determine whether precursor proteins trapped at this early stage of translocation were sufficiently stably associated with translocation components to survive detergent solubilization and analysis. Three different methods of analysis were used to detect complexes. The first was co-immunoprecipitation of radiolabeled precursors with antibodies directed against individual translocation components. To accomplish this, radiolabeled precursor to the small subunit (prSS) of ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) was allowed to interact with isolated, intact chloroplasts in the presence of 100 μM ATP. After re-isolation, intact chloroplasts were lysed hypotonically and separated into membrane and supernatant fractions by centrifugation. Chloroplastic membranes were solubilized with buffer containing decylmaltoside and subjected to centrifugation to remove aggregates and incompletely solubilized membranes. Using decylmaltoside, ∼80–90% of precursors and translocation components remained in the supernatant after centrifugation (data not shown). The solubilized membrane proteins were subjected to immunoprecipitation using antibodies to specific components of the chloroplastic protein translocation machinery. Sufficient antiserum was added in each case to ensure that >80% of each translocation component was immunoprecipitated (data not shown). Immunoprecipitates were analyzed by SDS–PAGE and fluorography to detect co-immunoprecipitation of radiolabeled prSS (Figure 1). An aliquot of the solubilized membrane fraction was analyzed by SDS–PAGE prior to immunoprecipitation to assess the amount of prSS that was bound to chloroplasts (Figure 1, lane 1). Antibodies specific for OEP75 and OEP34, two outer membrane translocation components, were capable of co-immunoprecipitating prSS (Figure 1, lanes 3 and 4), while pre-immune sera to OEP75 (Figure 1, lane 2) and OEP34 (data not shown) were not. In addition, antibodies specific to IEP110, an inner membrane translocation component, were also capable of co-immunoprecipitating prSS (Figure 1, lane 5), whereas the corresponding pre-immune serum was not (data not shown). As the amount of protein present in the total solubilized membrane fraction is equivalent to one-fifth of that added to each immunoprecipitation, we estimated that ∼10% of the prSS added to each immunoprecipitation was associated with translocation complexes (compare lane 1 with lanes 3–5). These co-immunoprecipitation efficiencies are comparable with those observed with mitochondrial transport complexes (Manning-Krieg et al., 1991; Ungermann et al., 1994). Figure 1.Association of prSS with known translocation components. Gel-filtered, 35S-labeled, prSS was incubated with isolated chloroplasts in 100 μM ATP. After incubation for 10 min at room temperature, intact chloroplasts were repurified, lysed hypotonically, and the chloroplastic membranes were resuspended in buffer containing decylmaltoside. Immunoprecipitation reactions were performed with anti-OEP75 (lane 3), anti-OEP34 (lane 4), anti-IEP110 (lane 5) or anti-ClpC (lane 7) antibodies, and OEP75 pre-immune control (lane 2). Ten per cent of each reaction was removed before immunoprecipitation (R, a representative sample is shown in lane 1); the remaining 90% was split into equal portions to which immune (I lanes) or pre-immune sera (P lanes) to OEP75, OEP34, IEP110 or ClpC were added. Only a representative pre-immune control for OEP75 is shown (lane 2); other pre-immune controls were similar. Instead of a pre-immune control, an anti-IEP35 (lane 6) immunoprecipitation was performed on the other half of the anti-LHCP (lane 8) immunoprecipitation. Samples were analyzed by SDS–PAGE and fluorography. Download figure Download PowerPoint To evaluate the possibility that co-immunoprecipitation of prSS was due to non-specific interaction of precursors with membrane proteins, immunoprecipitation was performed using antibodies raised against two membrane proteins that are not part of the translocation apparatus. Antibodies raised against either the inner envelope membrane protein, IEP35 (Schnell et al., 1994), or the thylakoid membrane light-harvesting complex protein (LHCP) family (Payan and Cline, 1991) were incapable of co-immunoprecipitating prSS (Figure 1, lanes 6 and 8), demonstrating the specificity of the co-immunoprecipitation procedure. Because soluble molecular chaperones have been shown in other systems to interact with translocation complexes (Schatz and Dobberstein, 1996), we investigated whether any chloroplastic chaperones were present in this complex. Antibodies to ClpC, a chloroplastic molecular chaperone of the Hsp100 family, were capable of co-immunoprecipitating prSS (Figure 1, lane 7), but no association was detected with pre-immune serum (data not shown). S78 could also be detected in association with translocation complexes, but this interaction displayed characteristics that required further investigation (see below). Collectively, the results described above suggested that prSS forms a stable association with known translocation components from both the outer envelope membrane and the inner envelope membrane, as well as a stromally localized chaperone. The most likely explanation for the results described above was that all of these translocation components were associated with one another in a large complex. To evaluate this prediction, the solubilized complex was analyzed via a second strategy. The putative complexes were first immunoprecipitated with anti-OEP75 or anti-ClpC antibodies. Again, sufficient antibodies were added to ensure that >80% of these components were immunoprecipitated (data not shown). The composition of the immunoprecipitates was then analyzed by SDS–PAGE and immunoblotting (Figure 2). Antibodies against these two proteins were chosen because, based on the known location of OEP75 and ClpC within chloroplasts, they should be situated at opposite sides of a putative translocation complex. To demonstrate the relative amounts of the proteins present before immunoprecipitation, a sample of the solubilized membranes from the chloroplast-binding reaction was analyzed (Figure 2, R lanes). The complexes associated with both OEP75 and ClpC contained IEP110 (Figure 2A, I lanes), OEP34 (Figure 2D, I lanes) and OEP86 (Figure 2B, I lanes). In addition, complexes associated with ClpC contained OEP75 (Figure 2B, I lanes), and OEP75 was found in complexes containing ClpC (Figure 2C, I lanes). By comparing the amounts of the translocation components co-immunoprecipitated with amounts present in the total soluble membrane fractions, we estimated that between 10 and 20% of most translocation components remained associated with ClpC, or OEP75 after detergent solubilization and immunoprecipitation. Lower levels of OEP86 were associated with immunoprecipitated ClpC complexes, and both OEP86 and OEP34 were co-immunoprecipitated with OEP75 at higher levels. Figure 2.Complexes immunoprecipitated by anti-OEP75 and anti-ClpC contain other translocation components. Radiolabeled prSS was incubated with isolated chloroplasts in 100 μM ATP. After a 10 min incubation, intact chloroplasts were repurified, lysed hypotonically, and isolated chloroplastic membranes were solubilized in buffer containing decylmaltoside. Immunoprecipitation reactions were performed with anti-OEP75 or anti-ClpC antibodies, or their corresponding pre-immune controls. Ten per cent of each reaction was removed for direct analysis on SDS–PAGE (R lanes); the remaining 90% was split into two equal fractions and immunoprecipitated with either anti-OEP75 serum (I lanes) or the corresponding pre-immune control (P lanes). Similar reactions were analyzed with anti-ClpC antibodies. Three replicates of the immunoprecipitations were analyzed by SDS–PAGE and transferred to Immobilon-P membrane. After immunoblotting, each membrane was divided into two parts, above (A–C) and below (D–F) the position of IgG, and probed with antibodies against IEP110 (A), OEP75 and OEP86 (B), ClpC and S78 (C), OEP34 (D), LHCP (E) and IEP35 and ClpP (F). Download figure Download PowerPoint These immunoprecipitations were specific, as the samples immunoprecipitated with the corresponding pre-immune sera did not contain any of these proteins (Figure 2, P lanes). No association of LHCP or IEP35 was found in immunoprecipitates of OEP75 or ClpC (Figure 2E and F, respectively). Several proteins of differing molecular weights were detected by the LHCP antiserum, reflecting the presence of members of an antigenically related family of proteins (Payan and Cline, 1991). ClpC is only one class of stromal chaperones identified in chloroplasts (Gething and Sambrook, 1992; Moore and Keegstra, 1993; Shanklin et al., 1995; Marshall et al., 1996). Thus, we sought to determine whether molecular chaperones of the Hsp70 family were associated with the translocation complex, because they have been shown to interact with translocating precursors in other protein transport systems (Kang et al., 1990; Scherer et al., 1990; Vogel et al., 1990). Antibodies raised against the stromal Hsp70, S78, were used to probe the immunoblot of the immunoprecipitated complexes. The presence of S78 in the solubilized membrane fraction demonstrated that some stromal S78 was present in the isolated membranes (Figure 2C, R lane). However, no association of S78 with the complexes immunoprecipitated with anti-OEP75 or anti-ClpC antibodies could be detected (Figure 2C, I lanes). We concluded that S78 was not stably associated with the solubilized translocation complex in these immunoprecipitations. In prokaryotes, the Hsp100 chaperone family can function as subunits of the Ti protease (for review, see Squires and Squires, 1992). This protease is active as a hetero-oligomer containing the Hsp100 homolog and a separate subunit, ClpP, which contains protease activity (Hwang et al., 1988; Maurizi et al., 1990). Because both of these proteins have homologs in chloroplasts (Shanklin et al., 1995), it was possible that the association between ClpC and translocation complex members reflected association with a protease, and did not indicate that ClpC itself was a translocation component. To evaluate this possibility, complexes immunoprecipitated by anti-OEP75 and anti-ClpC antibodies were probed with antibodies to stromal ClpP (Figure 2F). A reactive band at the correct molecular weight (Figure 2F, R lane) demonstrated the presence of ClpP in the membranes, presumably from stromal contamination. No significant association of this protein with the ClpC immunoprecipitate was detected (Figure 2F, I lanes) even when longer exposures were examined (data not shown). We concluded that the majority of the ClpC associated with translocation components was not associated with stromal ClpP and, therefore, the association was not involved in proteolysis of these complexes. To obtain further evidence that solubilized translocation components were present in a complex, a third method of analysis was used. Solubilized chloroplastic envelope membranes containing translocation components and prSS were layered over a sucrose density gradient and fractionated to enable study of their sedimentation patterns (Figure 3). Radiolabeled prSS, analyzed by scintillation counting, appeared in two peaks, with the majority (∼80% of total) found near the top of the gradient in fractions 1–9, and a smaller peak (∼20% of total) migrating further into the gradient at fraction 23. Because prSS is rapidly processed and assembled into Rubisco holoenzyme after import into isolated chloroplasts (Archer and Keegstra, 1993), it was necessary to determine whether both peaks contained full-length precursor. This was measured by SDS–PAGE followed by fluorography, and all radioactivity was found in prSS (Figure 3B). Additionally, silver staining indicated that some Rubisco holoenzyme and Cpn60/10 could be detected in the gradient fractions, presumably due to stromal contamination of the isolated membranes, but these complexes sedimented primarily in fractions 11 and 15, respectively (data not shown). To determine the sedimentation patterns of the translocation components, OEP86, OEP75, OEP34 and IEP110, and of the chaperone ClpC, the gradient fractions were analyzed by SDS–PAGE, followed by immunoblotting with specific antibodies (Figure 3C). These components sedimented in two distinct peaks, similar to the sedimentation pattern of prSS (compare Figure 3B and C). However, two control proteins, IEP35 and LHCP, were observed to have different sedimentation patterns (Figure 3C). From these results, we conclude that a significant portion of the radiolabeled precursor had been incorporated into a large complex containing components of the translocation apparatus from the outer and inner membranes as well as the stromal Hsp100 chaperone ClpC. This complex sedimented far into the gradient at fraction 23, with significant amounts of the various translocation components being present in this fraction. Interestingly, OEP86 and OEP34 migrated no further into the gradient than fraction 23, while OEP75, IEP110 and ClpC were also found in fraction 25. Whether these differences represented different complexes was beyond the resolution of the sucrose gradient and could not be determined. Figure 3.Translocation components and prSS co-sediment as a complex. Radiolabeled prSS was incubated with isolated chloroplasts in 100 μM ATP. After incubating for 10 min, intact chloroplasts were repurified, lysed hypotonically, and isolated membranes were solubilized in buffer containing decylmaltoside. The solubilized membrane fraction was layered over a 10–30% linear sucrose density gradient and sedimented at 150 000 g for 18 h. Fractions were removed and the sedimentation patterns were analyzed by scintillation counting (A), SDS–PAGE and fluorography (B) and SDS–PAGE and transferring to Immobilon-P for immunoblotting with antibodies against the indicated proteins (C). Download figure Download PowerPoint A significant portion of both the radiolabeled precursors and translocation components was observed at the top of the gradient. One possible explanation for the presence of translocation components near the top of the gradient is that they represent individual components that were not part of translocation complexes. Another, more likely, explanation is that some of the translocation complexes dissociated into individual components during experimental manipulation and could have been maintained as complexes if milder or more stabilizing conditions had been used. Further work will be needed to distinguish between these two possibilities. ClpC and S78 both interact with translocation complexes, but only the association with ClpC is stable in solubilized complexes ClpC was detected in a complex with other translocation components and prSS but, using similar conditions, association of the stromal Hsp70, S78, could not be detected (Figure 2). Because Hsp70 homologs have been observed to be important members of other protein translocation systems (Kang et al., 1990; Scherer et al., 1990; Vogel et al., 1990), their possible involvement was examined in more detail. Our previous experiments had been performed with complexes prepared from isolated membranes. Because lysis and fractionation of chloroplasts during membrane isolation might disrupt interactions between chaperones and translocation complexes, whole chloroplasts were solubilized and the resulting complexes examined for the presence of S78 (Figure 4). The absence of processed mature-sized Rubisco small subunit (mSS) in aliquots of the solubilized chloroplasts indicated that import had not occurred (Figure 4A, lanes 1, 4 and 7). Antibodies to OEP75, ClpC and S78 were able to co-immunoprecipitate prSS from the solubilized chloroplasts (Figure 4A, lanes 2, 5 and 8, respectively). No prSS could be detected with the corresponding pre-immune controls (Figure 4A, lanes 3, 6 and 9), demonstrating the specificity of the immunoprecipitation reactions. Centrifugation was not performed prior to these immunoprecipitations; thus, it was possible that S78 had not been detected in earlier studies (Figure 2) either because its association with translocation complexes was disrupted during chloroplast lysis and fractionation or because S78 was in a complex which was pelleted during the centrifugation step. To examine the latter possibility, detergent-solubilized chloroplasts were subjected to centrifugation before immunoprecipitation was performed on the supernatant. In the soluble complexes, prSS could still be co-immunoprecipitated with anti-OEP75 and anti-ClpC antibodies, but not with anti-S78 antibodies (Figure 4B, compare lanes 2 and 6, with lane 10). When the pellet fraction was analyzed, significant amounts of prSS were present (Figure 4B, lanes 4, 8 and 12). Immunoblotting with antibodies to OEP75, ClpC and S78 demonstrated that these proteins were present in the pellet, but other proteins not involved in protein translocation were also present (data not shown). Figure 4.Translocation complexes containing S78 are insoluble. Radiolabeled prSS was incubated with isolated chloroplasts in 100 μM ATP. After a 10 min incubation, intact chloroplasts were repurified and immediately solubilized in buffer containing decylmaltoside. (A) Immunoprecipitation reactions were performed directly after solubilization with anti-OEP75 (lane 2), anti-ClpC (lane 5) or anti-S78 (lane 8) antibodies, and their corresponding pre-immune controls (lanes 3, 6 and 9). Ten per cent of each reaction was removed for direct analysis by SDS–PAGE (R, lanes 1, 4 and 7); the remaining 90% was split into two equal fractions and immunoprecipitated with immune (I lanes) or their corresponding pre-immune sera (P lanes). (B) Solubilized chloroplasts were centrifuged at 150 000 g for 5 min before immunoprecipitation with anti-OEP75 (lane 2), anti-ClpC (lane 6) or anti-S78 (lane 10) antibodies, and their corresponding pre-immune controls (lanes 3, 7 and 11). Ten per cent of each reaction was removed for direct analysis by SDS–PAGE (R, lanes 1, 5 and 9); the remaining 90% was split into equal fractions and immunoprecipitated with immune (I lanes), or their corresponding pre-immune sera (P lanes). After centrifugation, the pellet was resuspended in SDS–PAGE buffer and analyzed by SDS–PAGE (Pell., lanes 4, 8 and 12). Samples were analyzed by SDS–PAGE and fluorography. Download figure Download PowerPoint The interaction of ClpC and S78 with translocation complexes was characterized further by investigating how the interactions changed during an import timecourse (Figure 5). First, radiolabeled precursors were allowed to bind to isolated chloroplasts in the presence of low levels of ATP. After removal of unbound precursors, chloroplasts were resuspended in the presence of high levels of ATP, thereby allowing import to occur. Analysis of the import timecourses demonstrated that mSS accumulated at significant levels only after a 2.5 min lag, but accumulation then continued for 30 min (Figure 5A and D). Precursor, but not mSS, was detected in the anti-ClpC immunoprecipitated fractions whether or not a centrifugation step was performed prior to immunoprecipitation (Figure 5B and E). Complexes immunoprecipitated by anti-ClpC antibodies were present at high levels at time 0, and decreased as the import reaction progressed (Figure 5B and E). Precursors were detected in the anti-S78-immunoprecipitated fractions if centrifugation was not performed (Figure 5C) but, after centrifugation, precursors were not observed in the anti-S78 immunoprecipitates (Figure 5F). Interestingly, no mSS was found in association with the anti-ClpC- or anti-S78-precipitated fractions. Analysis of the immunoprecipitation supernatant fractions showed that the mSS was not degraded by proteolysis (data not shown). We concluded that both ClpC and S78 interacted with precursors that were productively bound because these complexes disappeared in later timepoints, when mSS appeared. Because complexes precipitated by anti-S78 antibodies sedimented by centrifugation, we concluded that they were either very large or represented an aggregation of translocation complexes or incompletely solubilized membrane fragments. At present we are unable to distinguish between these possibilities. Figure 5.Association of ClpC and S78 with prSS-containing translocation complexes during import. Radiolabeled prSS was incubated with isolated chloroplasts in 100 μM ATP. After a 10 min incubation, intact chloroplasts were repurified, import was initiated by addition of 4 mM ATP at time = 0, and aliquots were removed at the given timepoints. Intact chloroplasts were again repurified, and then solubilized in buffer containing decylmaltoside. The solubilized chloroplasts from each timepoint were either immunoprecipitated directly (A–C), or subjected to centrifugation for 5 min at 150 000 g (D–F). Ten per cent of each timepoint was removed for direct analysis on SDS–PAGE (A and D). The remaining 90% of each timepoint was split into equal fractions and immunoprecipitated with either anti-ClpC (B and E) or anti-S78 (C and F) antibodies. Samples were analyzed by SDS–PAGE and fluorography.