These authors contributed equally.
New mechanistic insights into pre-protein transport across the second outermost plastid membrane of diatoms
Article first published online: 25 MAR 2010
© 2010 Blackwell Publishing Ltd
Volume 76, Issue 3, pages 793–801, May 2010
How to Cite
Hempel, F., Felsner, G. and Maier, U. G. (2010), New mechanistic insights into pre-protein transport across the second outermost plastid membrane of diatoms. Molecular Microbiology, 76: 793–801. doi: 10.1111/j.1365-2958.2010.07142.x
- Issue published online: 23 APR 2010
- Article first published online: 25 MAR 2010
- Accepted 15 March, 2010.
Chromalveolates like the diatom Phaeodactylum tricornutum arose through the uptake of a red alga by a phagotrophic protist, a process termed secondary endosymbiosis. In consequence, the plastids are surrounded by two additional membranes compared with primary plastids. This plastid morphology poses additional barriers for plastid-destined proteins, which are mostly nucleus-encoded. Recent investigations have focused on the postulated translocon of the second outermost membrane (periplastidal membrane, PPM). These studies identified a symbiont-specific ERAD (endoplasmic reticulum-associated degradation)-like machinery (SELMA), which has been implicated in plastid pre-protein import. Despite this recent progress, key factors for protein transport via SELMA are still unknown. As SELMA substrates presumably undergo ubiquitination, a corresponding ubiquitin ligase and an enzyme for the subsequent removal of ubiquitin need to reside in the space between the second and third membrane (periplastidal compartment, PPC). Here we characterize two proteins fulfilling these criteria. We show that ptE3P (P.t ricornutumE3 enzyme of the PPC), the ubiquitin ligase, and ptDUP (P.t ricornutumde-ubiquitinating enzyme of the PPC), the de-ubiquitinase, localize to the PPM and PPC, respectively. In addition, we demonstrate their retained functionality by in vitro data.
Protein modification by the reversible conjugation of ubiquitin regulates a variety of cellular processes (Pickart and Fushman, 2004). Probably the most prominent is protein degradation in eukaryotic cytoplasms, where proteins are marked for destruction at the proteasome by polyubiquitination (Schrader et al., 2009). The addition of ubiquitin to target proteins requires three enzymatic activities: the activation of ubiquitin by an E1 enzyme (ubiquitin-activating enzyme), the transfer of ubiquitin to an E2 enzyme (ubiquitin-conjugating enzyme) and finally the transfer of ubiquitin to the substrate via the action of an E3 enzyme, a ubiquitin ligase (Jentsch, 1992).
The cytosol is the destructive compartment not only for misfolded cytosolic proteins but also for aberrant proteins of the endoplasmic reticulum (ER), since the ER possesses no proteolytic machineries itself. In a process termed ER-associated degradation (ERAD) malfolded proteins are retro-translocated from the ER into the cytosol, where proteasomal degradation occurs (Vembar and Brodsky, 2008; Hirsch et al., 2009). Depending on the type of substrate and the site of damage, several ERAD mechanisms are known. One is referred to as ERAD-L and is responsible for the ubiquitin-dependent retro-translocation of misfolded soluble ER proteins and membrane proteins with lumenal lesions, respectively (Carvalho et al., 2006; Ismail and Ng, 2006).
Recently, we and others have shown that in several organisms, which evolved by the engulfment and intracellular reduction of a red alga (i.e. heterokontophytes, haptophytes, cryptophytes and apicomplexans), duplicated ERAD-L systems exist, one for the host ER and a second one localizing to the complex plastid (Sommer et al., 2007; Kalanon et al., 2009; Spork et al., 2009). This peculiar type of plastid is surrounded by four membranes and traces back to the red algal endosymbiont of secondary endosymbiosis (Stoebe and Maier, 2002; Gould et al., 2008). The plastid-located ERAD-machinery, termed SELMA (symbiont-specific ERAD-like machinery) (Hempel et al., 2009), originated from the symbiont (Agrawal et al., 2009) and contributes, according to our hypothesis, to the import of nucleus-encoded proteins into complex plastids (Sommer et al., 2007; Agrawal et al., 2009; Hempel et al., 2009).
In the diatom Phaeodactylum tricornutum, the model organism of this article, core components of the ERAD-L machinery were identified in two copies, one for the cytosol or the ER membrane, respectively, and one equipped with targeting signals for the complex plastid. These plastid copies are located either in the second outermost membrane of the complex plastid, which originates from the cytoplasmic membrane of the eukaryotic secondary symbiont, or, if soluble, in the space between the second and third outermost plastid membrane – the remnant cytoplasm of the eukaryotic symbiont, i.e. the periplastidal compartment (PPC) (Sommer et al., 2007; Hempel et al., 2009) (Fig. 3). In the ER-specific ERAD system, ubiquitination of ERAD substrates is crucial not only for proteasomal degradation but also for the process of retro-translocation (de Virgilio et al., 1998; Yu and Kopito, 1999; Shamu et al., 2001; Jarosch et al., 2002). Since symbiont-specific ERAD factors exist in chromalveolates one can assume that their mode of action is conserved. Hence, ubiquitination of SELMA substrates should be essential for plastid pre-protein import to occur. For the symbiont-specific ERAD system of P. tricornutum (SELMA) PPC-specific E1 and E2 proteins as well as a PPC-specific ubiquitin have already been described (Sommer et al., 2007). However the E3 enzyme of the catalytic triad, the ubiquitin ligase Hrd1, which is used in the ERAD-L pathway (Bays et al., 2001; Deak and Wolf, 2001; Carvalho et al., 2006; Ismail and Ng, 2006), has so far not been identified in a symbiont-specific version. Therefore we used a more general approach to search for symbiont-specific E3 enzymes in P. tricornutum that could fulfil the postulated function. Here we report on a symbiont-specific RING finger ubiquitin ligase of the PPC, which meets the criteria for a SELMA-specific E3 enzyme and shows in vitro ubiquitination activity. According to our model, ubiquitin residues must be removed from the plastid precursors after SELMA-mediated transport across the second outermost plastid membrane to reconvert the proteins into an unmodified form for further transport or maturation. Thus, a de-ubiquitinating activity should be present in the space between the second and third outermost membrane, the PPC. By searching the genomic data of P. tricornutum for de-ubiquitinating enzymes, we identified a candidate equipped with a targeting signal for PPC localization. Moreover, the expressed protein showed de-ubiquitination activity in vitro. Thus, our data highlight further factors possibly involved in protein transport into complex plastids, giving vital hints about the functions of a minimized eukaryotic cytoplasm.
Data mining: screening for symbiont-specific ubiquitin ligases and de-ubiquitinating enzymes in P. tricornutum
Hrd1 proteins, the ubiquitin ligases of the ERAD-L system, show only weak cross-species conservation. This may explain our failure to identify symbiont-specific Hrd1 homologues in P. tricornutum using standard blast conditions. We therefore searched the P. tricornutum database for entries that meet the basic requirements of a symbiont-specific Hrd1-like protein and hence show (i) a C-terminal RING finger domain protruding into the PPC, (ii) an N-terminal bipartite targeting signal (BTS) for the PPC and (iii) at least one transmembrane domain. Using these criteria we identified three candidates (PhatrDBv2.0 ID: 46465, ID: 36427, ID: 48034), which were tested for in vivo localization as GFP fusion constructs in P. tricornutum. This analysis revealed that only one candidate (ID: 48034) localized to the PPC as indicated by a so-called ‘blob-like’ localization of the GFP signal, whereas the others showed cytosolic or ER localization, respectively. We tentatively refer to this PPC-located protein as ptE3P (P.t ricornutumE3 enzyme of the PPC).
In order to identify a PPC-located de-ubiquitinating enzyme, we searched the P. tricornutum database for similarities to sequences of de-ubiquitinases from other organisms. Candidates identified were inspected thereafter for a possible periplastidal targeting signal. One candidate fulfils these criteria. A cDNA of this gene, which corrects the predicted gene model (ID 40154), was thereafter fused with GFP and expressed in P. tricornutum. Thereby we obtained a PPC-specific GFP signal. This protein is tentatively assigned the name ptDUP (P.t ricornutumde-ubiquitinating enzyme of the PPC).
Subcellular localization of ptE3P
ptDUP is a soluble protein, located in the PPC (Fig. 1A, a). The full-length ptE3P GFP fusion protein also shows a PPC-specific GFP signal (Fig. 1A, b), but is predicted to be a membrane protein with one transmembrane spanning helix (SOSUIv1.11). In order to verify this prediction, we used carbonate extraction to separate integral from membrane associated and soluble proteins. As indicated in Fig. 1B the protein ptE3P was detected predominantly in the fraction of integral membrane proteins (Fig. 1B) demonstrating that ptE3P is indeed an integral membrane protein. Immunoelectron microscopic analyses provided further support for membrane localization. Ultra-thin cuts of P. tricornutum cells were incubated with an antibody against ptE3P and showed specific labelling at one of the plastid surrounding membranes (Fig. 1C). In order to investigate the topology of ptE3P, we applied the self-assembling GFP system (Cabantous et al., 2005), which has previously been used for in vivo topology analyses in P. tricornutum and Toxoplasma gondii (van Dooren et al., 2008; Hempel et al., 2009). In this approach GFP is split into two fragments: GFP1-10 and the C-terminal beta-strand GFP-11. Only when both fragments are localized within the same cellular compartment can GFP1-10 and GFP-11 assemble to produce GFP fluorescence. For topology analyses of ptE3P we fused GFP-11 to the C-terminus of the protein, whereas the second fragment GFP1-10 was directed to the ER or the PPC, respectively, by fusion to specific marker proteins. A schematic depiction of the experimental setup is shown in Fig. 1D. In combination with the PPC marker we observed a blob-like GFP fluorescence as is typical for the PPC (Fig. 1D, b). On the other hand no fluorescence was obtained in combination with the ER marker (Fig. 1D, a). We cannot formally rule out the possibility that ptE3P might be a protein integral to the third outermost plastid membrane, since fusion of fragment GFP-11 to the N-terminus (which should protrude to the ER lumen) is not feasible due to the N-terminally situated targeting sequence. However, an integration into the third outermost membrane is unlikely, as the N-terminal targeting signal of ptE3P is PPC-specific and does not harbour an aromatic amino acid at the first position of the transit peptide (Gruber et al., 2007). In any case, these results demonstrate that the C-terminus of ptE3P, which contains the catalytic RING finger domain, is on the periplastidal side of the membrane and therefore ptE3P meets all of the above mentioned prerequisites for a function in SELMA-mediated pre-protein translocation.
In vitro functional assays with ptE3P and ptDUP
PtE3P and ptDUP show sequence similarity to RING finger ubiquitin ligases or de-ubiquitinating enzymes, respectively. In order to investigate if these predictions are correct, we tested the enzymatic functions of both proteins in vitro.
RING finger ubiquitin ligases like the Hrd1 protein of Saccharomyces cerevisiae can catalyse in vitro transfer of ubiquitin from E2 enzymes to themselves (Bays et al., 2001; Deak and Wolf, 2001). To demonstrate that ptE3P is a functional ubiquitin ligase, we overexpressed the full-length protein as well as only the C-terminal part of ptE3P (containing the catalytic RING domain) with an N-terminal GST-tag in Escherichia coli. After affinity purification, we used the recombinant protein for in vitro ubiquitination assays including E1 and E2 enzymes, ATP and biotinylated ubiquitin in each reaction. Incubation of full-length ptE3P as well as only the C-terminal part with the RING domain caused auto-ubiquitination of ptE3P as evidenced by Western blotting with streptavidin-HRP (Fig. 2A, top, lanes 1 and 2). Hence, ptE3P is a functional ubiquitin ligase with the C-terminal RING finger domain conferring catalytic activity. Assays without recombinant ptE3P including only purified GST, as well as assays without E2 enzyme served as negative controls, and produced no detectable amounts of ubiquitinated proteins (Fig. 2A, top, lanes 3 and 4).
In order to demonstrate that ptDUP is a functional de-ubiquitinating enzyme we heterologously overexpressed and affinity-purified the mature protein with a C-terminal His-tag. Subsequently, the enzyme was used in a fluorogenic in vitro de-ubiquitination assay. Here, the ability to cleave the synthetic de-ubiquitinating enzyme substrate Ub-AFC, in which ubiquitin is attached to AFC by an isopeptide bond – chemically the same as in protein–ubiquitin conjugates – was tested and monitored over time. Upon deconjugation of the ubiquitin derivate, the chromophore – and hence fluorescence – was set free, indicating the cleavage of the isopeptide bond of Ub-AFC (Fig. 2B). Since ptDUP shows sequence similarities to proteases of the peptidase C19 family bearing a cysteine residue at the active site, and since NEM is known as a cysteine modifier and therefore an inhibitor of most de-ubiquitinases in vitro (e.g. Hartmann-Petersen et al., 2003; Guterman and Glickman, 2004; Kayagaki et al., 2007), we tested ptDUP in respect to NEM inhibition. Figure 2B shows that ptDUP activity is inhibited by NEM, most likely by chemical inhibition of the catalytic site cysteine.
Most phototrophic chromalveolates, an ancient group of protists that evolved by secondary endosymbiosis, harbour plastids which are surrounded by four membranes (complex plastids). Since the majority of plastid proteins is encoded on the nucleus of the host cell, plastid pre-proteins have to cross all four membranes to reach the stroma. Hence, mechanisms had to evolve to mediate pre-protein translocation across these membranes. Recently, we and others have shown that many components of an ERAD-L translocation machinery are encoded twice in several chromalveolates (Sommer et al., 2007; Kalanon et al., 2009; Spork et al., 2009). Localization studies on these proteins in the diatom P. tricornutum demonstrated that one set is specific for the host, whereas the other is either located in the second outermost plastid membrane or, if soluble, in the PPC. Among others, copies of Der1-1, Der1-2, Cdc48, Ufd1, E1 (namely Uba1) and E2 (Ubc4) of the ERAD-specific ubiquitination machinery, as well as ubiquitin itself, were shown to be specific for the complex plastid of diatoms (Sommer et al., 2007; Hempel et al., 2009). According to our hypothesis, these proteins are components of an ERAD-derived translocation system, called SELMA, which is acting at the second outermost membrane of complex plastids and provides the protein transport activity for nucleus-encoded plastid proteins across that membrane. Consistently, in vivo studies in P. tricornutum have shown that transit peptides of periplastidal pre-proteins interact physically with the SELMA complex at the second outermost plastid membrane (Hempel et al., 2009). Furthermore, a conditional knockout of a symbiont-specific Der1 protein of T. gondii showed impairment in apicoplast pre-protein transport (Agrawal et al., 2009).
Nonetheless, some factors implicated in ubiquitination/de-ubiquitination, which are essential for the postulated mode of protein translocation across the second outermost membrane as inferred from the common ER-specific ERAD system, have not yet been found. Transport via the ‘genuine’ ERAD-L system in S. cerevisiae depends on ubiquitination of the transported proteins (Jarosch et al., 2002) by E1 and E2 enzymes and the ubiquitin ligase Hrd1 as an E3 enzyme (Bays et al., 2001; Deak and Wolf, 2001; Garza et al., 2009; Hirsch et al., 2009). In earlier studies we have shown that symbiont-specific Uba1 (E1 enzyme), Ubc4 (E2 enzyme) and ubiquitin are encoded by the diatom P. tricornutum (Sommer et al., 2007), but we failed to detect a PPC-specific Hrd1 protein, the RING finger ubiquitin ligase of ERAD-L. Thus, we used another search strategy to identify PPC-located RING finger proteins which are expressed as pre-proteins with a BTS, the bipartite targeting sequence, which directs proteins across plastid surrounding membranes. This led to the identification of the RING finger membrane protein ptE3P, a periplastidal membrane protein which exposes the RING finger domain into the PPC, as demonstrated by self-assembling GFP assays (Fig. 1D). In addition, we showed that ptE3P has ubiquitin-ligase activity in vitro (Fig. 2A). Due to the lack of strong similarities between Hrd1 and ptE3P, we can only tentatively assign ptE3P as the PPC-located functional homologue of diatoms. However, several facts support our interpretation. First, a host-specific Hrd1 can be detected in P. tricornutum, indicating that the search criteria for Hrd1 homologues functions in the P. tricornutum data set. Second, ptE3P fulfils basic criteria for an ubiquitin ligase of the SELMA complex, as its catalytic RING finger domain is exposed to the PPC, where – according to our hypothesis – ubiquitination of pre-proteins occurs during translocation across the second outermost plastid membrane. We could detect no further PPC-specific ubiquitin ligases, suggesting that ptE3P might represent the ubiquitin ligase which functions in the SELMA context. For further investigations we performed in vitro ubiquitination assays which included GFP-tagged transit peptides of PPC- and stroma-specific proteins; however, no specific transit peptide ubiquitination was detected in such assays (data not shown). Hence, we assume that further factors might be involved in substrate ubiquitination at the second outermost plastid membrane, which are not present in standard in vitro ubiquitination reactions. Another reason might be that the E2 enzyme used in these standard assays is not compatible with substrate specific ubiquitination.
After crossing the second outermost plastid membrane via SELMA, the ubiquitin moiety(ies) has, according to our hypothesis, to be removed from the transported molecules. PtDUP, a PPC-specific de-ubiquitinating enzyme may provide this function. Hence, we include both newly identified ubiquitin-associated proteins, ptE3P and ptDUP, in our model for protein import into secondary plastids, as illustrated in Fig. 3. According to that model, PPC- and stroma-specific nucleus-encoded proteins are transported across the second outermost plastid membrane via the symbiont-specific ERAD-like machinery (SELMA). Membrane proteins sDer1-1 and sDer1-2, which were previously shown to form an oligomeric complex that interacts with periplastidal targeting signals, are believed to form the translocation channel of SELMA (Hempel et al., 2009). In vivo import studies on stroma-specific pre-proteins in a sDer1-knockout strain in T. gondii showed that stroma-specific proteins use a SELMA-specific pathway as well (Agrawal et al., 2009). Based on the fact that core components of the ‘genuine’ ERAD translocation machinery are conserved in P. tricornutum in a symbiont-specific version, we postulate that SELMA-mediated translocation occurs in a similar fashion. Pre-proteins are ubiquitinated by PPC-specific Uba1 (E1 enzyme), Ubc4 (E2 enzyme) and the newly identified ubiquitin ligase ptE3P (E3 enzyme). Such modified substrates are subsequently bound by the hexameric ATPase Cdc48 and cofactors such as Ufd1 and are extracted into the PPC. Following translocation ubiquitin moieties have to be removed to ensure correct protein maturation. This final step is possibly catalysed by the newly identified PPC-specific de-ubiquitinating enzyme ptDUP.
We cannot definitely exclude that ubiquitination/de-ubiquitination activity is used in other PPC-specific processes; however, no indications exist for such mechanisms in diatoms, haptophytes and apicomplexans. Interestingly, in cryptophytes – which have retained machineries for protein synthesis within the PPC – (poly)ubiquitination might still play a role as a signal for PPC-specific proteasomal degradation. In cryptophytes all proteasome subunits are encoded on the nucleomorph genome and suggest that degradation of misfolded proteins still occurs in the PPC.
Searching genomic databases of an additional diatom and two apicomplexans revealed host-specific Hrd1 proteins with a typical distribution of membrane spanning helices as well as a nucleomorph-encoded, hence symbiont-specific Hrd1 protein in the cryptophyte Guillardia theta. No further PPC-specific Hrd1 entries but several candidates for symbiont-specific ubiquitin ligases with structural modifications, i.e. fewer transmembrane domains, such as ptE3P were identified. Assuming that no further symbiont-specific Hrd1 candidates escaped our attention, the presence of a nucleomorph and the lack of this remnant nucleus in other plastid harbouring chromalveolates might therefore be reflected by the Hrd1 distribution. This may imply that the Hrd1-copy introduced into the cellular merger by the endosymbiontic red alga, which encodes a Hrd1 protein with six transmembrane domains, was not successfully transferred to the nucleus of the secondary host. In addition, the host copy in nucleomorph-less organisms was not duplicated for further use in the PPC. Thus, loss of the endosymbiont-specific copy of Hrd1 and transfer of its function to structurally different E3 enzymes with less transmembrane spanning helices might be one prerequisite for evolving a nucleomorph-less complex plastid of chromalveolates.
The sequence for ptE3P can be retrieved from P. tricornutum database (PhatrDBv2.0, http://genome.jgi-psf.org/Phatr2/Phatr2.home.html) with protein ID 48034. The original gene model of ptDUP (ID 40154) has to be corrected as RT-PCR demonstrated that there is no intron in the 3′ region. The correct cDNA sequence for ptDUP was annotated at NCBI (Accession No. GU474203). Sequences were amplified under standard PCR conditions using genomic DNA of P. tricornutum or cDNA as template, respectively. RNA isolation was carried out using Trizol reagent (Invitrogen). The reverse transriptase reaction was performed with SuperScriptII (Invitrogen) using random hexamer primer. For in vivo localization studies sequences of full-length ptE3P and ptDUP were cloned in front of eGFP into pPhaT1 as described previously (Sommer et al., 2007). For self-assembling GFP assays a modified pPha-T1 vector pPha-DUAL[2xNR] was used. pPha-DUAL[2xNR] contains two multiple cloning sites both under the control of the endogenous nitrate reductase promoter of P. tricornutum. For topology analyses of ptE3P the full-length sequence was fused to the N-terminus of GFP-11, whereas the full-length protein PDI (ID 44937) and the topogenic signal of sHsp70 (ID 55890) were fused to the N-terminus of GFP1-10 as an ER marker or PPC marker, respectively. Both marker proteins were described and used in previous studies (Gould et al., 2006; Hempel et al., 2009). For the overexpression of ptE3P (full-length and C-terminal amino acids 241–537) and ptDUP (full-length protein Δaa 2–42) in E. coli sequences were cloned into pGEX-5x-3 and a modified pQE60 (Qiagen) expressing LacI, respectively.
Self-assembling GFP assays
For in vivo topology analyses of ptE3P the self-assembling GFP assay was used (Cabantous et al., 2005). The sequences of ptE3P and a marker protein for PPC/ER were fused to GFP fragments GFP-11 and GFP1-10 and cloned into pPha-DUAL[2xNR] (see Plasmid construction). Constructs were transformed into P. tricornutum as described previously (Apt et al., 1996) with the exception that cells were grown under non-induced conditions with 1.5 mM NH4 as solely nitrogen source. Transformants were checked by colony PCR for genomic integration of both constructs. Positive colonies were grown in liquid culture containing 1.5 mM NH4. For microscopic analyses cultures were washed in nitrogen-free medium and transferred to f/2 medium containing 0.9 mM NO3 to induce expression of self-assembling GFP fragments. After 6 h of induction transformants were analysed with a confocal laser scanning microscope (see Electron- and fluorescence microscopy).
Integral membrane proteins and associated membrane proteins were separated by carbonate extraction using sodium carbonate buffer pH 11.5 as described previously (Hempel et al., 2009).
Protein purification and in vitro ubiquitination/de-ubiquitination assays
GST-tagged ptE3P and His-tagged ptDUP proteins were expressed in E. coli BL21-CodonPlus(DE3)-RIL cells (Stratagene) and purified using glutathione sepharose matrix (GE-Healthcare) and Ni-NTA agarose (Qiagen), respectively, virtually according to manufacturer's instructions. Expression of ptE3P was carried out under standard conditions, whereas recombinant ptDUP was expressed for 2 h at 30°C with 0.5 mM IPTG. Ubiquitination assays were carried out with components of an ubiquitination kit (Enzo Life Sciences). A standard assay contained: 4 µg of purified GST-tagged ptE3P, 1× ubiquitination buffer, 1 U inorganic pyrophosphatase, 1 mM DTT, 5 mM Mg-ATP, 100 nM E1-Enzym, 2.5 µM E2-Enzym (UbcH5b) and 5 µM biotinylated ubiquitin. The reaction was incubated for 1.5 h at 37°C and stopped with 2× non-reducing gel loading buffer. 25 µl of each reaction was separated by SDS-PAGE and subjected to Western blot using HRP-Streptavidin for detection of ubiquitin moities.
The deconjugation reaction of Ub-AFC (BostonBiochem) was monitored over 5 min with individual measurements every 5 s using a Tecan Infinite M200 microplate reader. The release of AFC fluorescence using λex = 400 nm and λem = 505 nm indicated the hydrolysis of Ub-AFC. Reactions were performed using 1 µM Ub-AFC, 12 µg of affinity-purified enzyme and reaction buffer (20 mM Tris, pH 7.5, 10 mM DTT, 5 mM EDTA and 0.1% BSA) in a final volume of 100 µl. Inhibition of de-ubiquitination was tested by 15 mM final concentration of NEM.
Electron- and fluorescence microscopy
For transmission electron microscopic analyses P. tricornutum cells were harvested (5 min, 1000 g) and fixed for 4 h with 0.02% glutaraldehyde and 4% paraformaledhyde in f/2 medium. Cells were washed in Sorensen's buffer, dehydrated in a graded ethanol series, and embedded in lowicryl resin. Samples were post-stained with uranyl acetate and lead citrate under standard conditions. Sections were cut with a diamond knife and mounted on pioloform-coated grids. Primary antibody against ptE3P was detected with a secondary antibody coupled to 30 nm gold particles (Biocell). Samples were analysed using a JEOL 2100 electron microscope.
In vivo localization of GFP fusion proteins was analysed with a confocal laser scanning microscope Leica TCS SP2 using a HCX PL APO 63×/1.32–0.6 oil Ph3 CS objective. GFP and chlorophyll fluorescence was excited at 488 nm and detected at a bandwidth of 500–520 nm and 625–720 nm, respectively.
We are grateful to the group of Geoff Waldo (Los Alamos, USA) for providing templates for self-assembling GFP fragments. For immunoelectron microscopic analyses we thank Marianne Johannsen. Furthermore, we thank Jude Przyborski for critical reading of the manuscript. This work was supported by the German Research Foundation (SFB 593 and Graduate School 1216).
- 2009) Genetic evidence that an endosymbiont-derived ERAD system functions in import of apicoplast proteins. J Biol Chem 284: 33683–33691. , , , and (
- 1996) Stable nuclear transformation of the diatom Phaeodactylum tricornutum. Mol Gen Genet 252: 572–579. , , and (
- 2001) Hrd1p/Der3p is a membrane-anchored ubiquitin ligase required for ER-associated degradation. Nat Cell Biol 3: 24–29. , , , , and (
- 2005) Protein tagging and detection with engineered self-assembling fragments of green fluorescent protein. Nat Biotechnol 23: 102–107. , , and (
- 2006) Distinct ubiquitin-ligase complexes define convergent pathways for the degradation of ER proteins. Cell 126: 361–373. , , and (
- 2001) Membrane topology and function of Der3/Hrd1p as a ubiquitin-protein ligase (E3) involved in endoplasmic reticulum degradation. J Biol Chem 276: 10663–10669. , and (
- 2008) Toxoplasma gondii Tic20 is essential for apicoplast protein import. Proc Natl Acad Sci USA 105: 13574–13579. , , , , and (
- 2009) In vitro analysis of Hrd1p-mediated retrotranslocation of its multispanning membrane substrate 3-hydroxy-3-methylglutaryl (HMG)-CoA reductase. J Biol Chem 284: 14710–14722. , , and (
- 2006) Nucleus-to-nucleus gene transfer and protein retargeting into a remnant cytoplasm of cryptophytes and diatoms. Mol Biol Evol 23: 2413–2422. , , , , , and (
- 2008) Plastid evolution. Annu Rev Plant Biol 59: 491–517. , , and (
- 2007) Protein targeting into complex diatom plastids: functional characterisation of a specific targeting motif. Plant Mol Biol 64: 519–530. , , , , , and (
- 2004) Complementary roles for Rpn11 and Ubp6 in deubiquitination and proteolysis by the proteasome. J Biol Chem 279: 1729–1738. , and (
- 2003) Ubiquitin binding proteins protect ubiquitin conjugates from disassembly. FEBS Lett 535: 77–81. , , and (
- 2009) ERAD-derived preprotein transport across the second outermost plastid membrane of diatoms. Mol Biol Evol 26: 1781–1790. , , , , and (
- 2009) The ubiquitylation machinery of the endoplasmic reticulum. Nature 458: 453–460. , , , , and (
- 2006) Have you HRD? Understanding ERAD is DOAble! Cell 126: 237–239. , and (
- 2002) Protein dislocation from the ER requires polyubiquitination and the AAA-ATPase Cdc48. Nat Cell Biol 4: 134–139. , , , , , , and (
- 1992) The ubiquitin-conjugation system. Annu Rev Genet 26: 179–207. (
- 2009) Characterization of two putative protein translocation components in the apicoplast of Plasmodium falciparum. Eukaryot Cell 8: 1146–1154. , , and (
- 2007) DUBA: a deubiquitinase that regulates type I interferon production. Science 318: 1628–1632. , , , , , , et al. (
- 2004) Polyubiquitin chains: polymeric protein signals. Curr Opin Chem Biol 8: 610–616. , and (
- 2009) Targeting proteins for degradation. Nat Chem Biol 5: 815–822. , , and (
- 2001) Polyubiquitination is required for US11-dependent movement of MHC class I heavy chain from endoplasmic reticulum into cytosol. Mol Biol Cell 12: 2546–2555. , , , , and (
- 2007) Der1-mediated preprotein import into the periplastid compartment of chromalveolates? Mol Biol Evol 24: 918–928. , , , , , and (
- 2009) An unusual ERAD-like complex is targeted to the apicoplast of Plasmodium falciparum. Eukaryot Cell 8: 1134–1145. , , , , , , et al. (
- 2002) One, two, three: nature's tool box for building plastids. Protoplasma 219: 123–130. , and (
- 2008) One step at a time: endoplasmic reticulum-associated degradation. Nat Rev Mol Cell Biol 9: 944–957. , and (
- 1998) Ubiquitination is required for the retro-translocation of a short-lived luminal endoplasmic reticulum glycoprotein to the cytosol for degradation by the proteasome. J Biol Chem 273: 9734–9743. , , and (
- 1999) The role of multiubiquitination in dislocation and degradation of the alpha subunit of the T cell antigen receptor. J Biol Chem 274: 36852–36858. , and (