|
|
|
|
|
||
|
|||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | |||||
First published online 16 June 2009
doi: 10.1242/jcs.049460
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Article |
1 CNR Institute for Neuroscience and Department of Pharmacology, Università degli Studi di Milano, Italy
2 CNR Institute of Chemistry of Molecular Recognition, Milano, Italy
3 Department of Pharmacobiological Science, University of Catanzaro `Magna Graecia', Roccelletta di Borgia (CZ), Italy
* Author for correspondence (e-mail: n.borgese{at}in.cnr.it)
Accepted 31 March 2009
 |
Summary |
|---|
 Top Summary IntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
Key words: Cytochrome b5, Endoplasmic reticulum, Liposomes, Post-translational translocation, Rabbit reticulocyte lysate, Unassisted insertion
|
Introduction |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
). The well characterized co-translational pathway is, however, not accessible to the large and diverse group of membrane proteins that are classified as tail-anchored (TA) proteins.
TA proteins consist of a cytosolic N-terminal domain anchored to the membrane by a single transmembrane domain (TMD) close to the C-terminus. Different members of this group of proteins are found on essentially all membranes of the cell, where they have crucial roles in processes such as vesicular trafficking, protein translocation, and regulation of apoptosis (Borgese et al., 2003). A recent bioinformatic analysis of the human genome has revealed the presence of 325 genes that code for 411 putative TA proteins (Kalbfleisch et al., 2007), confirming the large size and functional importance of this group of membrane proteins.
To reach their correct destination within the secretory pathway, TA proteins, similarly to other membrane proteins, must first insert into the ER membrane (Jäntti et al., 1994; Kutay et al., 1995; Linstedt et al., 1995; Pedrazzini et al., 1996). However, unlike the TMDs of other ER-directed membrane proteins, the TMD of TA proteins, because of its C-terminal position, remains sequestered within the ribosome and inaccessible to SRP and to the lipid bilayer until release of the completed polypeptide chain. Thus, TA proteins are obliged to use a post-translational mechanism for their insertion into the ER membrane (Borgese et al., 2007; Borgese et al., 2003).
In recent years, the introduction of new translocation assays, which distinguish between true transmembrane integration of TA proteins and their salt-resistant association to the cytosolic leaflet of the bilayer (Brambillasca et al., 2005), has led to important advances in our understanding of the mechanisms used by this class of proteins to insert into the ER membrane. The analysis of the membrane components involved and of the energy requirements has revealed the existence of at least two insertion pathways. One group of TA proteins, which includes the translocon component Sec61β and the SNARE protein synaptobrevin 2, insert into the membrane by an assisted mechanism, meaning that they require energy and at least one proteinaceous component of the ER membrane (Abell et al., 2004; High and Abell, 2004; Kim et al., 1997; Kutay et al., 1995). Other TA proteins, such as the electron carrier cytochrome b5 (b5) and the protein tyrosine phosphatase (PTP)-1B, are instead able to translocate at nanomolar ATP concentrations into pure lipid vesicles, provided that these have low levels of cholesterol (Brambillasca et al., 2006; Brambillasca et al., 2005). We have called this latter mechanism unassisted, meaning that the insertion does not depend on, and is not even facilitated by, any membrane protein, but not taking into consideration a possible requirement for cytosolic proteins. Access to, or exclusion from, the unassisted pathway is determined by the TMD of the TA substrate. Moderately hydrophobic and short TMDs can be inserted in an unassisted manner, whereas more hydrophobic ones require assistance (Brambillasca et al., 2006).
In addition to investigations on the required membrane components, complementary studies have addressed the possible involvement of cytosolic proteins in the delivery of TA proteins to the ER. For the assisted pathway, a soluble receptor, named Asna1 or transmembrane recognition complex subunit of 40 kDa (TRC40) has recently been identified in the rabbit reticulocyte lysate (RRL). Asna1/TRC40 is an ATPase that recognizes the TMD of assisted TA proteins and delivers them to the ER membrane in an ATP-dependent manner (Favaloro et al., 2008; Schuldiner et al., 2008; Stefanovic and Hegde, 2007). However, it interacts poorly with the TMD of b5, and a dominant-negative mutant of the chaperone is ineffective on unassisted insertion (Stefanovic and Hegde, 2007). Other studies have reported the involvement of Hsc70 and Hsp40 (Abell et al., 2007; Rabu et al., 2008) and SRP (Abell et al., 2004), suggesting that more than one assisted pathway might function in parallel.
The goal of the present study was to investigate the involvement of cytosolic proteins in the unassisted pathway of TA protein insertion. Indeed, a possible role for cytosolic chaperone(s), which could be required to keep `unassisted' TA substrates in a translocation-competent conformation, was not excluded in previous studies (Brambillasca et al., 2006; Brambillasca et al., 2005), and was in fact suggested in a recent report (Rabu et al., 2008). To address this question, we have used b5 as representative of the unassisted class of TA proteins, and performed experiments both on the in vitro translated protein and on bacterially expressed purified constructs. Our results demonstrate that, although b5 binds to proteins present in the RRL, these proteins are neither necessary for, nor do they facilitate, transmembrane integration of b5. Nonetheless, oxidation of cytosolic proteins inhibits insertion both in vitro and in vivo, suggesting a regulatory role of chaperones in targeting and insertion. Quite remarkably, an 85-residue polar domain, appended to the C-terminus of b5, can be translocated into protein-free liposomes without assistance from any cytosolic protein or other lysate component.
|
Results |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
), we developed a stringent assay that allowed us to evaluate true transmembrane integration of TA proteins. Briefly, b5 tagged with a C-terminal epitope derived from the N-terminus of bovine opsin (b5-ops28), is translated in vitro, incubated post-translationally with vesicles, and finally digested with proteinase K (PK). The generation of a 
5 kDa protected fragment, consisting of the b5 TMD plus the 28 downstream polar residues, reports the translocation of the C-terminus of the tagged construct. In the case of insertion into ER-derived vesicles, translocation is confirmed by the N-glycosylation of an acceptor site within the opsin sequence.
Using this assay, we demonstrated that b5-ops28 can efficiently integrate into pure lipid vesicles (unassisted translocation) (Brambillasca et al., 2006; Brambillasca et al., 2005). To investigate the possible role of cytosolic proteins in the process, we first tested the effect of dilution of the RRL on translocation efficiency (Fig. 1A). After translation of b5-ops28, the sample was diluted either in RRL or in a buffer (TB) permissive for translocation, with a resulting increase (lanes 1-3), or five (lanes 4-6) or ten (lanes 7-9) times lower RRL protein concentration during the subsequent incubation with or without liposomes. After 1 hour of translocation, an aliquot of each sample was directly analyzed by SDS-PAGE and autoradiography (–PK), whereas the remaining sample was digested with proteinase K (+PK) and immunoprecipitated. When liposomes were present, a protected fragment was generated that was destroyed if detergent was present during exposure to PK. The quantification of protected fragment band intensity normalized to protein load (–PK) indicated that reduction of RRL concentration did not affect translocation efficiency. Furthermore, the kinetics of b5-ops28 insertion under conditions of diluted or concentrated RRL were indistinguishable (Fig. 1B).
|
The results of Fig. 1 suggest that, if a lysate protein is involved in b5 integration, it is either very abundant or has such a low dissociation rate from b5 as to be insensitive to dilution. To distinguish between these possibilities, we turned to a purified b5-ops28 whose insertion into liposomes could be followed in the absence of any RRL component. b5-ops28, produced as a glutathione-S-transferase (GST) fusion protein (see supplementary material Fig. S1) and bound to glutathione (GSH)-Sepharose, was eluted from the resin by thrombin cleavage either in the presence or in the absence of RRL. As shown in supplementary material Fig. S2A, the resulting b5-ops28 sample was of high purity and was devoid of visible contamination by bacterial proteins. To evaluate translocation, we adapted the protease protection assay to the bacterially expressed protein, revealing the protected fragment by western blotting with anti-opsin monoclonal antibodies (mAbs). After incubation of b5-ops28 with liposomes, and digestion in the absence of detergent, a protected fragment was detected, both in the presence and in the absence of RRL (Fig. 2A, lanes 2 and 5). The percentage of translocation (Fig. 2A) and the kinetics of the insertion (Fig. 2B) of b5-ops28 in the absence or presence of RRL were indistinguishable. Thus, b5-ops28 inserts efficiently into vesicles of pure lipids independently of any protein, nucleotide or other lysate component.
|
An 85-residue peptide, appended to the TMD of b5, can be translocated across protein-free liposomes in the absence of any cytosolic protein
In a previous study, we investigated the limits of the unassisted insertion pathway and found that polar sequences of up to nearly 100 amino acids placed downstream to the TMD of b5 could be post-translationally translocated into protein-free liposomes after in vitro translation in the RRL (Brambillasca et al., 2006). We asked whether such long polar domains could also be translocated in the absence of cytosolic proteins. To this end, we expressed in bacteria a fusion protein of GST and b5-ops85 (see supplementary material Fig. S1), a construct derived from b5, in which the C-terminal polar domain was increased to 85 residues (Brambillasca et al., 2006). b5-ops85, cleaved from GST in the presence of RRL or buffer (Fig. 3A), was incubated either with pig pancreas microsomes (MRs) or liposomes and then subjected to the protease protection assay. Remarkably, as demonstrated by a protected fragment of the expected size that was digested in the presence of detergent, b5-ops85 was able to translocate into both MRs and liposomes, regardless of the presence or absence of lysate proteins. After incubation with MRs, a minor, higher Mr band was generated (Fig. 3A, asterisk in lanes 2 and 7) that corresponds to the glycosylated protected fragment, as demonstrated by its disappearance when an inhibitory tripeptide was present during the translocation reaction (not shown). The low glycosylation efficiency compared with that observed with the in vitro translated protein (Brambillasca et al., 2006) can be explained by the higher substrate load imposed with the purified protein used in the assay.
|
Finally, we compared the insertion efficiency of b5-ops85 with that of b5-ops28 and established the protein to lipid molar ratio in the resulting proteoliposomes. Translocation reactions were carried out with known amounts of the two proteins and of lipid vesicles, and translocation efficiency was calculated taking into account differences in the efficiency of blotting or of immunoprecipitation between the intact protein and the protected fragments and possible incomplete protection from the PK digestion (see legend to supplementary material Fig. S2). As shown in Fig. 3C, b5-ops28 had a concentration in reconstituted proteoliposomes comparable with that of endogenous b5 in liver MRs, where the cytochrome represents 1% of all protein (Omura and Takesue, 1970). b5-ops85 showed an insertion efficiency about five times lower than b5-ops28, in agreement with our previous results on the in vitro translated proteins (Brambillasca et al., 2006).
b5 interacts with cytosolic proteins via its TMD
Although the above results demonstrate that cytosolic proteins are not required for insertion of b5, we hypothesized that it might nonetheless bind to chaperones that could be involved in its specific in vivo targeting (Borgese et al., 2001; D'Arrigo et al., 1993; De Silvestris et al., 1995). To evaluate possible interactions of b5-ops28 with cytosolic proteins, we first compared the sucrose gradient sedimentation profiles of the protein eluted in the presence or absence of RRL (Fig. 4A). Under both conditions, b5-ops28 was recovered in high Mr complexes. However, the patterns obtained after elution under the two conditions were different. In agreement with previous reports on b5 (Spatz and Strittmatter, 1971), b5-ops28 eluted in buffer alone self-associated into oligomers of heterogeneous size up to 240 kDa. In the presence of RRL, the sedimentation profile was shifted to the lower Mr region of the gradient, with two major peaks (fractions 3 and 6) in the 60,000 to 10,0000 region, indicating an association of b5-ops28 with one or more RRL proteins.
|
In an attempt to identify these proteins, we probed for the binding of known chaperones reported to be involved in the insertion of TA proteins that follow the assisted pathway (Abell et al., 2004; Abell et al., 2007; Stefanovic and Hegde, 2007). GST-b5 or a truncated construct lacking the TMD (GST-b5-
TMD) (supplementary material Fig. S1) immobilized on GSH-Sepharose were incubated with RRL, and fractions of bound and unbound proteins were separated. These were probed with antibodies against Hsc70, Hsp40, SRP54 and TRC40. As shown in Fig. 4B, Hsc70 strongly bound to b5 and to a lesser extent to b5-TMD. Binding of SRP54 and Hsp40 was close to undetectable, whereas TRC40 showed clear, TMD-dependent binding, in agreement with a previously observed weak crosslinkage of this chaperone to b5 (Stefanovic and Hegde, 2007).
|
), translocation of the in vitro translated protein was almost completely blocked (Fig. 5A, lane 3). The effect of oxidation could be neutralized when GSH was added during exposure to DIA (Fig. 5A, lane 5, GSH-co) and, importantly, was reversed if GSH was added after treatment with DIA (Fig. 5A, lane 4, GSH-post). By contrast, addition of ATP did not reverse the inhibition, and incubation with DIA after translocation had no effect on generation of the protected fragment (not shown), indicating that oxidation did not affect the integrity of the liposomes.
A pronounced but weaker inhibition (60%) was also obtained by alkylation with N-ethylmaleimide (NEM), which was neutralized by addition of dithiothreitol (DTT), together with NEM (Fig. 5B). As in the case of DIA, addition of ATP after NEM treatment did not reverse the inhibition (not shown).
To investigate whether DIA and NEM are also inhibitory on other unassisted proteins, we tested the effect of these compounds on PTP-1B, which inserts by the same pathway as b5 (Brambillasca et al., 2006). After in vitro translation, insertion of PTP-1B was indeed inhibited by DIA (supplementary material Fig. S3) and by NEM (not shown), indicating that the effects of oxidation and alkylation are not a peculiarity restricted to b5 insertion.
We next confirmed the lysate-dependence of the NEM and DIA effects on insertion by using bacterially expressed b5-ops28. b5-ops28, cleaved from GST in the presence of RRL, bovine serum albumin (BSA) or buffer, was treated either with NEM or with DIA. DIA inhibited translocation in the presence of RRL, but not when b5-ops28 was eluted and translocated in the presence of buffer alone, or when BSA was present during both elution and the subsequent incubation with liposomes (Fig. 6A). Interestingly, DIA was ineffective if allowed to act before incubation with b5-ops-28 (pre-treated lysate) (Fig. 6B), indicating that the oxidant is effective on pre-formed complexes involving the cytosolic partners in the reduced state (Fig. 6B). Analogously to DIA, the inhibitory effect of NEM was observed only when RRL was present and not when the protein was eluted in PBS (Fig. 6C).
|
To further characterize the trapping phenomenon, we investigated whether the chaperones analyzed in the pull-down assays of Fig. 4 showed increased binding to b5 in the presence of DIA. We found no difference for Hsp40 and SRP54 (not shown), but a slight increase in TRC40 binding (supplementary material Fig. S5). We therefore immunodepleted TRC40 and Hsc70 (which shows strong binding also in the absence of DIA) (Fig. 4B) from the RRL and then compared the inhibitory effects of DIA in the presence of untreated or depleted lysate. As shown in Fig. 7A (left panel), our antibodies were efficient in depleting TRC40 from the RRL under native conditions. Hsc70 depletion was also significant (70%), although less complete (Fig. 7C). To assess the functional significance of the TRC40 depletion, we compared the translocation efficiency of in vitro translated b5 and a b5 construct (b5-Syb2-ops28) (supplementary material Fig. S1) with the TMD of synaptobrevin 2 (Syb2 or VAMP2). This construct, like Syb2 itself, follows the assisted pathway (Brambillasca et al., 2006). As expected from the work of (Stefanovic and Hegde, 2007), TRC40 depletion had no effect on b5-ops28 transmembrane integration, but caused reduced (by 45%) integration of the chimera (Fig. 7A, right).
|
We then compared the effect of oxidation of both the in vitro translated and of the bacterially produced b5-ops28 in untreated and depleted lysates. As shown in Fig. 7B, DIA was equally effective in TRC40 and undepleted RRL. Likewise, the 70% depletion of Hsc70 did not affect DIA-mediated inihibition. These results suggest that proteins other than TRC40 and Hsc70 mediate the inhibitory effect of the oxidant on b5 insertion.
DIA also inhibits b5 insertion in vivo
To reveal a possible regulatory role of redox conditions on TA protein insertion, we investigated the effect of DIA on b5 insertion in intact cells. Since protein synthesis is inhibited by DIA, we injected the purified bacterially expressed protein into the cytosol of cultured cells. Cells were then returned for 15 minutes to the incubator, in the presence or absence of DIA and then fixed. In the absence of DIA, b5-ops28 showed the typical ER immunostaining pattern for b5, which coincided with that of calnexin (Fig. 8Aa-c) and not with mitochondrial staining (Fig. 8Ba-c). Thus, microinjected b5 is targeted in the same way as the protein produced within cells. DIA treatment caused a dramatic shift in the localization of microinjected b5, which now yielded a diffuse staining pattern, consistent with a cytosolic localization (Fig. 8A,Bd-f), thus recapitulating the insertion inhibition observed in vitro.
|
|
Discussion |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
). There is now general agreement that different TA proteins have different requirements for insertion (High and Abell, 2004) depending on features of the TMD (Brambillasca et al., 2006; Rabu et al., 2008). More specifically, most TMDs require an ER membrane protein for integration (assisted pathway), but moderately hydrophobic and short ones, similar to that of b5, can translocate their C-terminus across protein-free lipid bilayers (Brambillasca et al., 2006; Brambillasca et al., 2005). Although we refer to this latter mechanism as non-assisted, our previous studies did not exclude a role in this pathway for cytosolic proteins, which might be needed to keep the TA substrate in a state competent for delivery to the bilayer. This question has been addressed in the present investigation.
b5 binds to cytosolic proteins, but these do not facilitate its transmembrane integration
Recent studies have identified a cytosolic ATPase (Asna1/TRC40) in the ATP-dependent pathway of delivery of a number of TA substrates to the ER membrane, both in mammals (Favaloro et al., 2008; Stefanovic and Hegde, 2007) and in yeast [where it is called Get3 (Schuldiner et al., 2008)]. However, this protein appeared not to be involved in the delivery of b5 to the ER membrane (Stefanovic and Hegde, 2007), and whether other cytosolic proteins were involved was unclear. A recent study (Rabu et al., 2008) reported that in vitro insertion of puromycin-released truncated TA polypeptides with moderately hydorophobic TMDs (b5, PTP-1B, and bcl2) is strongly facilitated by Hsc70/Hsp40. However, transmembrane integration of b5-ops28 in the RRL occurs at ATP concentrations as low as 3 nM and is not stimulated by re-addition of nucleotides (Brambillasca et al., 2006). This observation is difficult to reconcile with an insertion mechanism mediated by DnaK proteins, whose Km for ATP is in the micromolar range (Braell et al., 1984).
To resolve the question of the role of chaperones in b5 insertion, we turned to a simplified system consisting of a purified preparation of b5 produced in E. coli. The purified protein could be integrated into liposomes with high efficiency, and, importantly, the efficiency and time course of its insertion were the same in buffer alone or in the presence of RRL. Notably, we found that even a polar stretch of 85 residues appended to the TMD of b5 could be translocated across protein-free bilayers in the absence of any cytosolic factor, extending the significance of our previous study on the in vitro or in vivo expressed lengthened construct (Brambillasca et al., 2006). We conclude that b5 [and presumably other unassisted proteins, such as PTP-1B (Brambillasca et al., 2006)], can integrate into lipid bilayers without any assistance from, or even facilitation by, either membrane or cytosolic proteins,
Notwithstanding the absence of any requirement for chaperones, several observations of this study indicate that b5 does interact, via its TMD, with more than one cytosolic protein. Sucrose gradient analysis demonstrated a change in the sedimentation profile of purified b5-ops-28 in the presence of RRL, and pull-down assays showed that b5 binds both Hsc70 and TRC40. Furthermore, in the presence of RRL, alkylating (NEM) and oxidative (DIA) agents inhibited b5 insertion. A recent paper has reported inhibition of transmembrane integration of the assisted TA proteins, RAMP4 and Sec61β, but not of b5, by oxidation and alkylation (Favaloro et al., 2008). The discrepancy between these and our results might be explained by the type of oxidants and the NEM concentration used in that study, as well as in other differences in experimental procedure.
Of the two tested inhibitory compounds, DIA had the most striking effect, inducing a near complete, RRL-dependent translocation block, which could not be correlated with gross aggregation of the TA substrate and which was nearly fully reversible. The effect of NEM was less pronounced, and we do not know at present whether or not alkyation and oxidation affect the same protein. Very interestingly, we found that DIA also acts within living cells, indicating that the observed effect does not specifically require reticulocyte proteins. In an attempt to identify the cytosolic proteins responsible for the inhibition, we considered TRC40 and Hsc70, because pull-down experiments had shown that they both bind to the b5 tail. However, strong depletion of these chaperones from the RRL did not affect the translocation block obtained with DIA, suggesting that they do not mediate the inhibition.
Although the protein(s) through which DIA mediates its inhibition remain to be identified, the most likely mechanism consistent with the cytosol-independence of b5 insertion is a redox-induced trapping phenomenon. We hypothesize that an altered conformation of one or more of the binding partners of b5 results in the formation of an unproductive complex incapable of releasing the substrate to the membrane. A similar trapping mechanism might also explain the recent results of Rabu and colleagues (Rabu et al., 2008). These authors observed inhibition of b5 (as well as of PTP-1B and bcl2) insertion by small molecule inhibitors of Hsp40/Hsc70, in parallel with the formation of b5-containing Hsp40/Hsc70 aggregates.
Fig. 9 depicts three pathways by which b5 can insert into the lipid bilayer, revealed by our results. In the absence of cytosol (Fig. 9, pathway 1), b5 self-associates via its TMD (as demonstrated by sucrose gradient analysis), but we suggest that it can be released from these oligomers in an insertion-competent form, which we have depicted as a monomer. In the presence of RRL, de novo synthesized b5 interacts with one or more cytosolic proteins (Fig. 9, pathway 2) from which it can be released at very low ATP concentrations [3 nM (Brambillasca et al., 2006)] to converge onto the spontaneous insertion pathway followed in the absence of cytosol. Oxidation of the interacting protein(s), both in vivo and in vitro, causes a conformational transition that results in trapping of the TA substrate in a non-productive complex (Fig. 9, pathway 3). Thus, cytosolic factors might have a regulatory role, modulating the delivery of non-assisted TA proteins to the appropriate lipid bilayer in response to altered conditions in the cytosol.
|
The targeting problem
Unassisted insertion has been reported not only for ER-targeted TA proteins, but also for some that insert into the outer mitochondrial membrane and peroxisomes (Kemper et al., 2008), as well as in the outer chloroplast envelope (Qbadou et al., 2003). For instance, in vitro synthesized Fis1, a fission protein shared by mitochondria and peroxisomes (Mozdy et al., 2000; Schrader, 2006), can insert similarly to b5, into protein-free lipid vesicles (Kemper et al., 2008). Furthermore, cytosolic proteins appear to be dispensable for insertion of a number of outer mitochondrial membrane TA proteins, both in cell-free systems with purified mitochondria (Lan et al., 2000) and in digitonin-permeabilized cultured cells (Setoguchi et al., 2006). How then is specificity in targeting of these proteins achieved in vivo? Lipid composition, such as low sterol content (Brambillasca et al., 2005; Kemper et al., 2008) might broadly determine which membranes are competent for spontaneous TA protein integration; however, because the sterol content of the ER, the OMM and the peroxisomal membrane are similarly low, additional factors must determine the discrimination between these two organelles. It is plausible that cytosolic targeting factors play this role. In agreement, a recent study has revealed that Fis1 targeting to peroxisomes is mediated by the peroxisomal membrane protein import factor Pex19 (Delille and Schrader, 2008). On the basis of the results presented in this study, however, in the case of the non-assisted TA proteins, such targeting factors must have a dual role, preventing spontaneous insertion and favouring delivery to the correct target (hypothetical pathway 4 in Fig. 9).
Much of the difficulty in understanding the targeting mechanisms of unassisted TA proteins has been due to the lack of recapitulation of their in vivo targeting specificity in in vitro systems (e.g. Borgese et al., 2001). The lack of specificity in cell-free systems can now be rationalized by postulating that, under in vitro conditions, binding of unassisted TA substrates to targeting factors is too weak to be effective, so that integration into any bilayer of appropriate lipid composition can occur (pathway 2 in Fig. 9). Appropriate in vivo and in vitro assays, based on the knowledge of the totally unassisted insertion pathway described here, will hopefully lead to the unravelling of the mechanisms underlying the specific targeting of this class of TA proteins.
|
Materials and Methods |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
). A polyclonal antibody against rabbit TRC40 was generated by immunizing rabbits with a peptide corresponding to the C-terminal region (from residues 84 to 100) conjugated to Keyhole Limpet Hemocyanin. A specific 40 kDa band, absent in blots probed with pre-immune serum, was recognized by the anti-TRC40 antibody in western blots both of RRL and of rat liver MRs. The antibody was also efficient in immunoprecipitating TRC40 under native conditions. Anti-b5 polyclonal antibody was present in our laboratory (Borgese et al., 2001). Anti-bovine opsin mAb R2-15 and anti-SRP54 polyclonal antibodies were from Paul Hargrave (Adamus et al., 1991) and Bernhard Dobberstein (University of Heidelberg, Heidelberg, Germany), respectively. Hsc70 rat mAbs and Hsp40 polyclonal antibodies were from Stressgen.
Translocation assay of in vitro synthesized b5-ops28
Transcription and translation of pGEM4-b5-ops28 in the RRL (Promega), translocation reactions with total rat liver MRs, prepared by differential centrifugation, pig pancreas MRs (Brambillasca et al., 2005) or liposomes, prepared by extrusion of bovine liver phosphatidylcoline (PC) (Avanti Polar Lipids) suspensions, were performed as previously described (Brambillasca et al., 2005). For a standard translocation reaction 2.5-5 µl translated product was incubated with 9 µg PC liposomes in a 10 µl final reaction. In some experiments, translation products were diluted in a buffer (translocation buffer, TB) containing 50 mM HEPES-K+ pH 7.2, 250 mM sorbitol, 70 mM potassium acetate, 5 mM K+EGTA, 2.5 mM magnesium acetate and 4 mM DTT before carrying out the translocation reaction. The concentration of PC liposomes was determined by including a known amount of radioactive PC as tracer. The protease protection assay to detect the translocated portion of the opsin-tagged constructs, immunoprecipitation of the intact protein or the protected fragment with anti-opsin mAbs, and analysis of the protected fragments on Tris-Tricine gels have been described in detail (Brambillasca et al., 2005).
To block sulfhydryl groups of RRL proteins, in vitro translated samples were incubated with 20 mM NEM (diluted from a freshly made 0.5 M solution in ethanol) for 30 minutes at 32°C. After quenching the NEM with 40 mM DTT for 10 minutes at 32°C, liposomes were added and incubation was continued for 1 hour at 32°C. To assay for translocation under oxidizing conditions, the translation mixture was incubated for 30 minutes with 5 mM DIA at 32°C before addition of vesicles. To revert the effect of DIA, 20 mM GSH was added either during (GSH-co) or after (GSH-post) exposure to DIA. In the latter case, incubation was continued for 20 minutes before the addition of vesicles.
Radioactive gels were imaged with the Storm phosphoimager (GE-Healthcare) and band intensities quantified with ImageQuant software. The percentage of translocation in liposomes was calculated from the quantification of the percentage of glycosylation of a parallel sample translocated into MRs and comparison between the protected fragments generated with liposomes and MRs, as previously described (Brambillasca et al., 2005).
Purification of b5, b5-TMD, b5-ops28 and b5-ops85 expressed as fusion proteins in bacteria
bl21 bacteria, transformed with plasmids coding for GST fusion proteins with the different b5 forms, and induced for 4 hours with 0.5 mM isopropyl-β-D-1-thiogalactopyranoside (IPTG), were collected from 1 liter cultures and lysed with 80 ml of buffer containing 2% Triton-X 100, 150 mM NaCl, 25 mM Tris-HCl pH 7.4, 10 mM EDTA, 0.2 mg/ml lysozyme, 10 mg/ml DNase, plus a protease inhibitor mixture. After incubation for 3-5 hours at room temperature, the lysate was clarified by centrifugation (10,000 x g for 40 minutes at 4°C) and the soluble fraction was aliquoted and conserved at –80°C. To purify the b5 proteins, the corresponding soluble fraction was incubated overnight at 4°C on a rotating wheel with GSH-Sepharose 4B beads (GE-Healthcare, 40 µl of a 50% resin suspension per 300 µl soluble fraction plus 700 µl lysis buffer). After removing unbound material, the beads were incubated with 100 µl of a solution containing 2 mM ATP, 10 mM MgSO4, 50 mM Tris-HCl (pH 7.4) for 10 minutes at 37°C, to remove associated chaperones. The beads were then washed once with PBS plus 1% Triton X-100 followed by ten washes with PBS without detergent. The b5 constructs were then released by incubation for 40 minutes at room temperature with 0.5 U thrombin diluted in 20 µl of either PBS, RRL or BSA solution (100 mg/ml in PBS). The yield of the thrombin cleaved protein was evaluated by the Lowry assay or by comparison of its Coomassie blue staining with that of known amounts of BSA after SDS-PAGE analysis.
Translocation of b5 constructs purified from bacteria
For a standard translocation reaction 1.5-3 µl of the purified protein described in the preceding paragraph (corresponding to 50-500 ng) was incubated with 9 µg PC liposomes in a 10 µl final volume. The translocation and protease protection assay followed by antiopsin immunoprecipitation was performed as described for the radioactive translation product, except that the PK concentration was 0.5 mg/ml. NEM or DIA treatment were performed as described for the in vitro translated product.
After translocation, the samples were analyzed by SDS-PAGE on 13% Tris-tricine gels, followed by blotting onto nitrocellulose (0.22 pore size, GE Healthcare) and the intact protein or the protected fragment were revealed by enhanced chemiluminescence with Pico reagents (Pierce) using the antiopsin mAb or anti-b5 polyclonals. The films were digitized and band intensities determined with the ImageJ software after calibration with the optical density calibration step table (Stouffer Graphics Arts, Mishawaka, IN). Since the opsin tag is present both on the intact protein (–PK) and on the protected fragment (+PK) the ratio of band intensities in the digested versus the undigested polypeptide reflects the molar percentage of translocation (see supplementary material Fig. S2 for details).
Velocity sucrose gradient centrifugation
b5-ops28 was produced in bacteria and eluted from GSH-Sepharose beads with thrombin in PBS or in RRL. A 15 µl sample was diluted in 185 µl of a buffer containing 20 mM NaCl and 25 mM Tris-HCl pH 7.4, and loaded on top of a 12 ml 5-20% linear sucrose gradient containing the same salts, in tubes of the SW40 rotor (Beckman Instruments), which were centrifuged at 100,000 x g for 16 hours at 4°C. Markers for sedimentation rates (cytochrome c, BSA, rat IgG and catalase) were run on a separate gradient centrifuged in parallel. 15 fractions were collected from the top with an Auto Densiflow probe (Buchler Instruments) and subjected to precipitation with TCA in the presence of 75 µg cytochrome c as carrier. The precipitated proteins were analyzed by SDS-PAGE on 13% Tris-glycine gels followed by western blotting, using the R2-15 mAb against the opsin tag.
Pull-down assays
GSH resin (30 µl) was incubated overnight at 4°C on a rotating wheel with 500 µg GST-b5 or GST-b5 TM. After removing unbound material, the beads were treated to detach associated chaperones and washed to remove any detergent trace. The beads were then incubated with 30 µl RRL in the presence or absence of 5 mM DIA for 30 minutes at 32°C. After the removal of the unbound proteins (U), the beads containing bound (B) RRL proteins, were subjected to thrombin cleavage to avoid interference of the fusion proteins with the subsequent western blotting analysis. Equivalent aliquots of the bound and unbound fractions as well as of total (T) RRL were analyzed by immunoblotting.
TRC40 and Hsc70 lysate depletion
Protein-G-Sepharose (35 µl) was incubated with 15 µl anti-TRC40 antibody (IgG purified 20 mg/ml) or 40 µl Hsc70 mAb (0.5 mg/ml) plus 1 µl RNasin (Promega) for 1 hour at 4°C. After removing the supernatant, the resin was incubated with 35 µl RRL and 1 µl RNasin for 1 hour at 4°C, and the depleted lysate was recovered after centrifugation.
Microinjection and treatment of cells with DIA
Purified bacterially produced b5-ops28 (250 ng/µl) eluted in PBS was injected into the cytosol of CV1 cells with a microinjector (Eppendorf 5200), applying a pressure of 80-90 hPa. The microinjected cells were returned to the incubator in complete medium supplemented or not with 2 mM DIA and fixed 15 minutes later.
Immunofluorescence
Cells were fixed with paraformaldehyde, incubated with primary and secondary antibodies as previously described (De Silvestris et al., 1995), and mounted in Vectashield (Vector laboratories). Immunostained cells were imaged with a Zeiss confocal microscope (LSM 510 Meta), with the use of a 63x magnification PlanApo lens (NA 1.4). Single confocal sections are shown in the illustrations, which were prepared with Photoshop software.
|
Footnotes |
|---|
We thank Silvia Brambillasca for critical discussion of our work, Elisa Fasana for carrying out the experiment shown in supplementary material Fig. S3 and Bernhard Dobberstein for the anti-SRP54 antibodies. Special thanks to the Monzino Foundation (Milano, Italy) for their generous gift of the LSM 510 Meta confocal microscope. This research was supported by Telethon, grants GGP04129 and GGP07010, and by an institutional CNR grant (ME.P02.007). S.F.C. was supported by a research contract of the University of Milano.
|
References |
|---|
|
TopSummaryIntroductionResultsDiscussionMaterials and MethodsReferences |
|---|
Abell, B. M., Pool, M. R., Schlenker, O., Sinning, I. and High, S. (2004). Signal recognition particle mediates post-translational targeting in eukaryotes. EMBO J. 23, 2755-2764.[CrossRef][Medline]
Abell, B. M., Rabu, C., Leznicki, P., Young, J. C. and High, S. (2007). Post-translational integration of tail-anchored proteins is facilitated by defined molecular chaperones. J. Cell Sci. 120, 1743-1751.
Adamus, G., Arendt, A. and Hargrave, P. A. (1991). Genetic control of antibody response to bovin rhodopsin in mice: epitope mapping of rhodopsin structure. J. Neuroimmunol. 34, 89-97.[CrossRef][Medline]
Borgese, N., Gazzoni, I., Barberi, M., Colombo, S. and Pedrazzini, E. (2001). Targeting of a tail-anchored protein to endoplasmic reticulum and mitochondrial outer membrane by independent but competing pathways. Mol. Biol. Cell 12, 2482-2496.
Borgese, N., Colombo, S. and Pedrazzini, E. (2003). The tale of tail-anchored proteins: coming from the cytosol and looking for a membrane. J. Cell Biol. 161, 1013-1019.
Borgese, N., Brambillasca, S. and Colombo, S. (2007). How tails guide tail-anchored proteins to their destinations. Curr. Opin. Cell Biol. 19, 368-375.[CrossRef][Medline]
Braell, W. A., Schlossman, D. M., Schmid, S. L. and Rothman, J. E. (1984). Dissociation of clathrin coats coupled to the hydrolysis of ATP: role of an uncoating ATPase. J. Cell Biol. 99, 734-741.
Brambillasca, S., Yabal, M., Soffientini, P., Stefanovic, S., Makarow, M., Hegde, R. S. and Borgese, N. (2005). Transmembrane topogenesis of a tail-anchored protein is modulated by membrane lipid composition. EMBO J. 24, 2533-2542.[CrossRef][Medline]
Brambillasca, S., Yabal, M., Makarow, M. and Borgese, N. (2006). Unassisted translocation of large polypeptide domains across phospholipid bilayers. J. Cell Biol. 175, 767-777.
Ceppi, P., Colombo, S., Francolini, M., Raimondo, F., Borgese, N. and Masserini, M. (2005). Two tail-anchored protein variants, differing in transmembrane domain length and intracellular sorting, interact differently with lipids. Proc. Natl. Acad. Sci. USA 102, 16269-16274.
D'Arrigo, A., Manera, E., Longhi, R. and Borgese, N. (1993). The specific subcellular localization of two isoforms of cytochrome b5 suggests novel targeting pathways. J. Biol. Chem. 268, 2802-2808.
De Silvestris, M., D'Arrigo, A. and Borgese, N. (1995). The targeting information of the mitochondrial outer membrane isoform of cytochrome b5 is contained within the carboxyl-terminal region. FEBS Lett. 370, 69-74.[CrossRef][Medline]
Delille, H. K. and Schrader, M. (2008). Targeting of hFis1 to peroxisomes is mediated by Pex19p. J. Biol. Chem. 283, 31107-31115.
Favaloro, V., Spasic, M., Schwappach, B. and Dobberstein, B. (2008). Distinct targeting pathways for the membrane insertion of tail-anchored (TA) proteins. J. Cell Sci. 121, 1832-1840.
High, S. and Abell, B. M. (2004). Tail-anchored protein biosynthesis at the endoplasmic reticulum: the same but different. Biochem. Soc. Trans. 32, 659-662.[CrossRef][Medline]
Jäntti, J., Keränen, S., Toikkanen, J., Ehnholm, C., Södderlund, H. and Olkkonen, V. M. (1994). Membrane insertion and intracellular transport of yeast syntaxin Sso2p in mammalian cells. J. Cell Sci. 107, 3623-3633.[Abstract]
Kalbfleisch, T., Cambon, A. and Wattenberg, B. W. (2007). A bioinformatics approach to identifying tail-anchored proteins in the human genome. Traffic 8, 1687-1694.[CrossRef][Medline]
Kemper, C., Habib, S. J., Engl, G., Heckmeyer, P., Dimmer, K. S. and Rapaport, D. (2008). Integration of tail-anchored proteins into the mitochondrial outer membrane does not require any known import components. J. Cell Sci. 121, 1990-1998.
Kim, P. K., Janiak-Spens, F., Trimble, W. S., Leber, B. and Andrews, D. W. (1997). Evidence for multiple mechanisms for membrane binding and integration via carboxyl-terminal insertion sequences. Biochemistry 36, 8873-8882.[CrossRef][Medline]
Kosower, N. S. and Kosower, E. M. (1995). Diamide: an oxidant probe for thiols. Methods Enzymol. 251, 123-133.[Medline]
Kutay, U., Ahnert-Hilgen, G., Hartmann, E., Wiedenmann, B. and Rapoport, T. A. (1995). Transport route for synaptobrevin via a novel pathway of insertion into the endoplasmic reticulum membrane. EMBO J. 14, 217-223.[Medline]
Lan, L., Isenmann, S. and Wattenberg, B. W. (2000). Targeting and insertion of C-terminally anchored proteins to the mitochondrial outer membrane is specific and saturable but does not strictly require ATP or molecular chaperones. Biochem. J. 349, 611-621.[CrossRef][Medline]
Linstedt, A. D., Foguet, M., Renz, M., Seelig, H. P., Glick, B. S. and Hauri, H.-P. (1995). A C-terminally-anchored Golgi protein is inserted into the endoplasmic reticulum and then transported to the Golgi apparatus. Proc. Natl. Acad. Sci. USA 92, 5102-5105.
Mozdy, A. D., McCaffery, J. M. and Shaw, J. M. (2000). Dnm1p GTPase-mediated mitochondrial fission is a multi-step process requiring the novel integral membrane component Fis1p. J. Cell Biol. 151, 367-380.
Omura, T. and Takesue, S. (1970). A new method for simultaneous purification of cytochrome b5 and NADPH-cytochrome c reductase from rat liver microsomes. J. Biochem. Tokyo 67, 249-257.
Ozols, J., Carr, S. A. and Strittmatter, P. (1984). Identification of the NH2-terminal blocking group of NADH-cytochrome b5 reductase as myristic acid and the copmlete amino acid sequence of the membrane-binding domain. J. Biol. Chem. 259, 13349-13354.
Pedrazzini, E., Villa, A. and Borgese, N. (1996). A mutant cytochrome b5 with a lengthened membrane anchor escapes from the endoplasmic reticulum and reaches the plasma membrane. Proc. Natl. Acad. Sci. USA 93, 4207-4212.
Qbadou, S., Tien, R., Soll, J. and Schleiff, E. (2003). Membrane insertion of the chloroplast outer envelope protein, Toc34: constrains for insertion and topology. J. Cell Sci. 116, 837-846.
Rabu, C., Wipf, P., Brodsky, J. L. and High, S. (2008). A precursor-specific role for Hsp40/Hsc70 during tail-anchored protein integration at the endoplasmic reticulum. J. Biol. Chem. 283, 27504-27513.
Rapoport, T. A. (2007). Protein translocation across the eukaryotic endoplasmic reticulum and bacterial plasma membranes. Nature 450, 663-669.[CrossRef][Medline]
Schrader, M. (2006). Shared components of mitochondrial and peroxisomal division. Biochim. Biophys. Acta 1763, 531-541.[Medline]
Schuldiner, M., Metz, J., Schmid, V., Denic, V., Rakwalska, M., Schmitt, H. D., Schwappach, B. and Weissman, J. S. (2008). The GET complex mediates insertion of tail-anchored proteins into the ER membrane. Cell 134, 634-645.[CrossRef][Medline]
Setoguchi, K., Otera, H. and Mihara, K. (2006). Cytosolic factor- and TOM-independent import of C-tail-anchored mitochondrial outer membrane proteins. EMBO J. 25, 5635-5647.[CrossRef][Medline]
Spatz, L. and Strittmatter, P. (1971). A form of cytochrome b5 that contains an additional hydrophobic sequence of 40 amino acids. Proc. Natl. Acad. Sci. USA 68, 1042-1046.
Stefanovic, S. and Hegde, R. S. (2007). Identification of a targeting factor for post-translational membrane protein insertion into the ER. Cell 128, 1147-1159.[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Rabu, V. Schmid, B. Schwappach, and S. High Biogenesis of tail-anchored proteins: the beginning for the end? J. Cell Sci., October 15, 2009; 122(20): 3605 - 3612. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
