| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 281, Issue 45, 34492-34502, November 10, 2006
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1123
From the
Instituto de Biologia Molecular e Celular (IBMC) Rua do Campo Alegre, 823, 4150-180 Porto, Portugal, the Instituto de Ciências Biomédicas de Abel Salazar (ICBAS), Universidade do Porto, Largo do Professor Abel Salazar, 2, 4099-003 Porto, Portugal, and the ¶Katholieke Universiteit Leuven, Faculteit Geneeskunde, Departement Moleculaire Celbiologie, Herestraat 49, B-3000 Leuven, Belgium
Received for publication, July 28, 2006 , and in revised form, September 6, 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
During recent years, a large amount of data has been collected regarding the biological properties of Pex19p and Pex3p. Pex19p is an acidic largely disordered protein displaying a dual subcellular localization (cytosolic and peroxisomal) with the ability to interact with most peroxisomal membrane proteins (PMPs)4 both in vitro and in vivo (13-15). The Pex19p domain involved in these interactions resides in the C-terminal two-thirds of the peroxin (16-18). Pex19p also interacts with Pex3p, another intrinsic protein of the peroxisomal membrane. In this case, however, the binding interface in Pex19p is more complex. Indeed, there are two Pex3p-binding domains in Pex19p. The first one resides in the 50 N-terminal amino acid residues of Pex19p and interacts strongly with Pex3p; the second one overlaps with the PMP-binding domain and binds Pex3p with a lower affinity (16, 17). This different mode of binding is illustrated by the observation that trimeric complexes comprising Pex19p, Pex3p, and a PMP can be obtained in vitro (19, 20) and is related to the fact that the Pex19p-Pex3p interaction serves a different purpose. Indeed, it has been shown that the intracellular localization of Pex19p is largely dictated by the localization of Pex3p (21, 22). For instance, when Pex3p is artificially missorted to mitochondria, Pex19p is also found in this compartment (21). Since these two proteins are required for the biogenesis of PMPs, these results indicate that Pex19p and Pex3p constitute, at least transiently, a functional/structural unit. Our knowledge on the mechanistic role of Pex16p in the biogenesis of the peroxisomal membrane is still scarce. However, increasing evidence points to a potential role for this peroxin in the very early steps of this process (7, 23).
What is exactly the role of Pex19p? Presently, there are no doubts that Pex19p can bind newly synthesized PMPs. In one study, overexpression of a Pex19p version containing a nuclear localization signal in mammalian cells resulted in the nuclear localization of several newly synthesized PMPs (10). In another work, using an in vitro translation system, it was shown that Pex19p interacts with newly synthesized PMPs (20). Thus, Pex19p has the capacity to bind unfolded peroxisomal membrane proteins. However, the following question remains regarding the subcellular localization of the Pex19p-PMP interaction. Under physiological conditions, does this interaction occur in the cytosol or at the peroxisomal membrane? This uncertainty is at the basis of the two different models regarding the role of Pex19p in peroxisomal biogenesis. According to some authors, Pex19p interacts with newly synthesized PMPs in the cytosol and, acting both as a chaperone and as a mobile receptor, delivers these proteins to the peroxisomal membrane via Pex3p (10, 19, 21, 24). Others have proposed that Pex19p acts as membrane chaperone promoting the assembly/disassembly of PMPs at the peroxisomal membrane (25-29).
Understanding the mechanism of any biological process requires a kinetic perspective and is highly facilitated if an open experimental system can be applied to its characterization. In vitro import systems offer these two advantages, because experimental time windows of minutes can be used and components can be added or even removed/neutralized in an easy way. Such a strategy was in fact applied to the study of the biogenesis of peroxisomal intrinsic membrane proteins several years ago. Specific insertion of PMP22 and PMP70 into the peroxisomal membrane was demonstrated (30-32). The process was shown to occur post-translationally, to be ATP-independent, and to require cytosolic components. However, at that time, the identities of the components involved in the targeting and insertion of this class of proteins into the organelle membrane were unknown. Presently, most, if not all, peroxins involved in this process are already known, allowing us to use many more tools (e.g. recombinant proteins and antibodies) to dissect its mechanism. Here, using a postnuclear supernatant (PNS) from rat liver, we show that a fusion protein containing a fragment of PMP24, a peroxisomal intrinsic membrane protein (33), is specifically inserted into the peroxisomal membrane despite the presence of all of the other cellular organelles in the import reactions. The involvement of Pex19p and Pex3p in this process was addressed. Our results indicate that the import competence of the PMP24 reporter protein is predetermined by Pex19p during or immediately after the translation step. Data suggesting the existence of a cargo-induced peroxisomal targeting mechanism for Pex19p are also provided.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
In Vitro Import Experiments—Rat liver PNS were prepared as described before (36) in SEM buffer (0.25 M sucrose, 20 mM MOPS-KOH, pH 7.4, 1 mM EDTA-NaOH, pH 7.4, 2 µg/ml N-(trans-epoxysuccinyl)-L-leucine 4-guanidinobutylamide). Import reactions containing 450 µg (protein) of rat liver PNS were performed at 37 °C (unless indicated otherwise) in 100 µl of import buffer (0.25 M sucrose, 50 mM KCl, 20 mM MOPS-KOH, pH 7.4, 3 mM MgCl2, 0.2% (w/v) lipid-free bovine serum albumin, 80 µM methionine, 2 µg/ml N-(trans-epoxysuccinyl)-L-leucine 4-guanidinobutylamide) using 0.15-0.25 µl of a reticulocyte lysate containing the 35S-labeled reporter protein. Where indicated, nucleotides were added to 5 mM (ATP or ATP
S) or 0.5 mM (GTP and GTPS) final concentrations. Treatment of reticulocyte lysates and PNS fractions with apyrase (20 units/ml; grade VII; Sigma) was performed in import buffer for 5 min at 37 °C. Control reactions with heat-inactivated apyrase (10 min at 95 °C) were treated in the same way. In antibody inhibition experiments, purified IgGs (8 µg) in SEM buffer were added to PNS fractions in import buffer and incubated for 20 min on ice before starting the import reaction with the 35S-labeled reporter protein. Proteinase K treatment (400 µg/ml final concentration) of import reactions in the absence or presence of detergents (1% (w/v) Triton X-100, 0.5% (w/v) sodium deoxycholate), sedimentation of organelles by centrifugation, treatment of organelle pellets with Na2CO3, pH 11.5 (37), or with solutions of high and low ionic strengths, protein precipitation with trichloroacetic acid, and SDS-PAGE analysis were done exactly as described before (38). Routinely, the gels were blotted onto a nitrocellulose membrane, exposed to an x-ray film, and afterward probed with several antibodies. In the density gradient centrifugation analysis of import reactions, a 4-fold scale-up of the standard import reaction was used. After proteinase K treatment, the complete import mixture was diluted to 1.5 ml with SEM and applied onto the top of step Nycodenz gradients (1.5 ml of 45% (w/v), 6 ml of 30% (w/v), 2 ml of 25% (w/v), and 2 ml of 20% (w/v) Nycodenz in 5 mM MOPS-KOH, pH 7.2, and 1 mM EDTA-NaOH, pH 7.2). The tubes were centrifuged in a vertical rotor (STEPSAVERTM 65V13; Sorvall®) at 48,000 x g for 2 h at 4°C. Fourteen equal fractions (
920 µl) were collected from the bottom of the gradient, and a 250-µl aliquot of each fraction was subjected to trichloroacetic acid precipitation and SDS-PAGE. All of the in vitro import experiments reported in this work were performed at least five times.
Construction of Plasmids Encoding the Reporter Peroxisomal Membrane Protein—To construct an expression plasmid encoding a fusion protein comprising (from the N to the C terminus) GFP, a linker of 12 amino acids (SGLRSRAQASNS), amino acid residues 1-175 of human PMP24 (33), and three copies of the c-Myc epitope, the plasmid yf48b10.r1 (IMAGE Consortium ID 25369) was used as a template in a PCR using the forward primer 5'-GGGGGAATTCTATGGCAGCCCCGCCGCAG-3' and the reverse primer 5'-CGGAGATCTCAGGTCGACGCGGCCGCGGATCCCCTTACGACCTCGGTGATACTCAAAGAGCCAC-3'. The amplified fragment was digested with EcoRI and BglII and inserted into EcoRI/BamHI-digested pEGFP-C1 (Clontech), yielding pMP13. This plasmid was digested with BamHI and SalI and ligated to a DNA fragment obtained by annealing the primers 5'-GATCATGGGACAGAAGCTGATCTCAGAGGAGGACCTGGAGCAGAAACTCATCTCTGAAGAAGATCTGGAACAAAAGTTGATTTCAGAAGAAGATCTG-3' and 5'-TCGACAGATCTTCTTCTGAAATCAACTTTTGTTCCAGATCTTCTTCAGAGATGAGTTTCTGCTCCAGGTCCTCCTCTGAGATCAGCTTCTGTCCCAT-3' (encoding the c-Myc epitopes). This plasmid (pMP15), encoding the desired PMP24-fusion protein, was used in the transfection experiments. In order to obtain a pGEM-4-based plasmid capable of expressing this reporter protein upon in vitro transcription/translation reactions, pMP15 was subjected to PCR using the forward primer 5'-GCGCGCGTCGACCACCATGGTGAGCAAG-3' and the reverse primer 5'-TCCCCCCGGGCTACAGATCTTCTTCTGAAATC-3'. The amplified fragment was digested with SalI and XmaI and subcloned into the SalI/XmaI-digested pGEM-4 vector (Promega), originating pMP17. This plasmid was also used in a PCR with the primers 5'-AGTCAGTGAGCGAGGAAGCGGAAGAGC-3' and 5'-CGCCACGCCTAAGAATTCGAAGCTTGAG-3' in order to produce a DNA fragment encoding GFP plus the linker of 12 amino acids (see above) preceded by the T7 RNA polymerase promoter.
Expression and Purification of Recombinant Proteins—The plasmids encoding His6-HsPex19p (pMF119) and His6-HsPex19p-(31-299) (pTW151) are described elsewhere (26). To generate a bacterial expression construct coding for His6-HsPex19p-(1-124) (pMF956), the corresponding cDNA was amplified by PCR (template pMF119; forward primer, 5'-GCATGCCATGGCCTGGGGAATGGCCGCCGCTGAGGAA-3'; reverse primer, 5'-AAGAGGGCGGCCGCTCATTCTTGTTGGGAGGTCAT-3'), digested with NcoI and NotI, and cloned into the NcoI/NotI-digested pETM11 vector (kindly provided by Dr. G. Stier, EMBL, Heidelberg, Germany). His6-HsPex19p, His6-HsPex19p-(1-124), and His6-HsPex19p-(31-299) were expressed in the M15, BL21 (DE3), and TOP10 strains of Escherichia coli, respectively. One hundred-ml cultures were induced with 1 mM isopropyl 1-thio-
-D-galactopyranoside (for His6-HsPex19p and His6-HsPex19p-(1-124)) or 2% (w/v) L-arabinose (for His6-HsPex19p-(31-299)) for 3 h at 37 °C. Pelleted cells were cooled on ice and lysed by sonication in 1.5 ml of 50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1 mM DTT, 0.1 mg/ml phenylmethylsulfonyl fluoride, and 1:500 (v/v) mammalian protease inhibitor mixture (Sigma). Cell debris was removed by centrifugation (15 min at 10,000 x g), and the clarified supernatant was incubated with 100 µl (bed volume) of HIS-SelectTM nickel affinity gel (Sigma) for 2 h at 4 °C. Thegel was washed three times with 1.5 ml of 50 mM sodium phosphate, pH 8.0, 150 mM NaCl, and the His6-tagged proteins were eluted by washing the beads three times with 200 µl of 50 mM sodium phosphate, pH 8.0, 150 mM NaCl, 100 mM imidazole. The eluted proteins were concentrated to 100 µl using a Vivaspin 10,000 molecular weight cut-off PES concentrator (Vivascience), diluted to 600 µl with SEM, 1 mM DTT, and concentrated again. This procedure was repeated three times. The recombinant proteins were further purified by anionic exchange chromatography using 1-ml Econo-Pac® high capacity ion exchange cartridges (Bio-Rad) and a linear gradient of 25-500 mM NaCl in 50 mM Tris-HCl, pH 8.0. Fractions containing the purified proteins were pooled and subjected to five cycles of concentration/dilution with SEM supplemented with 1 mM DTT using the Vivaspin columns. The proteins (5-10 mg/ml) were frozen in liquid N2 and stored at -70 °C.
To generate a bacterial expression construct coding for His6-TEV-HsPex3p-(34-373) (pMF1259), the corresponding cDNA was amplified by PCR using pMF158 (26) with the forward primer 5'-GGGGGCCATGGCTCAGAAGAAAATCAGAGAAATACAG-3' and the reverse primer 5'-GAGGGGTACCTCATTTCTCCAGTTGCTG-3'. The DNA fragment was digested with NcoI and KpnI and cloned into the NcoI/KpnI-digested pETM11 vector. Expression of His6-TEV-HsPex3p-(34-373) was done in the BL21 (DE3) E. coli strain as described (20). The protein was purified using the HIS-SelectTM nickel affinity gel as described above but using a lower concentration of NaCl (75 mM) in all buffers. The concentration and dilution of His6-TEV-HsPex3p-(34-373) with SEM buffer containing 1 mM DTT was done as described above.
A cDNA encoding the 373 amino acids of human Pex3p (39) was amplified from human skin fibroblast total RNA by reverse transcription-PCR exactly as described before (38) using the primers 5'-CGCGTCGACATGCTGAGGTCTGTATG-3' and 5'-CGCGTCGACTCATTTCTCCAGTTGCT-3'. The 1.1-kilobase pair cDNA fragment was cloned into the pGEM®-T Easy vector according to the manufacturer's instructions (Promega). The recombinant plasmid was digested with EcoRI and SalI, and the insert was cloned into the expression vector pMAL-c2 (New England Biolabs). The human Pex3p fused to the maltose-binding protein was expressed in the TB1 strain of E. coli and obtained as inclusion bodies.
The same strategy was used to obtain a cDNA encoding human Pex16p (12) using the primers 5'-CGCGAATTCAGGATGGAGAAGCTGCGG-3' and 5'-CGCGTCGACGGAGGTCTGTCAGCCCCA-3'. After cloning the 1.1-kilobase pair cDNA fragment into the pGEM®-T Easy vector, the plasmid was digested with HindIII and SalI, and the insert was cloned into the expression vector pGEX-5X-1 (Amersham Biosciences). The human Pex16p fused to the glutathione S-transferase was expressed in the XL1 strain of E. coli and obtained as inclusion bodies.
Native Polyacrylamide Gel Electrophoresis—The recombinant proteins were incubated in 20 µl of 50 mM Tris-HCl, pH 8.0, 2 mM DTT for 30 min at room temperature. After the addition of 2 µl of 0.17% (w/v) bromphenol blue, 50% (w/v) sucrose, the samples were centrifuged at 15,000 x g for 10 min and loaded onto Tris nondenaturating discontinuous 9 or 10% polyacrylamide gels (40). The gels were run at 250 V at 4 °C and stained with Coomassie Brilliant Blue R-250. In the experiments involving 35S-labeled proteins, the gels were blotted onto nitrocellulose membranes and exposed to an x-ray film or fixed, incubated with AmplifyTM Fluorographic Reagent (Amersham Biosciences), and dried according to the manufacturer's protocol.
Antibodies—Antibodies directed to human Pex3p and Pex16p were produced in rabbits after immunization with the maltose-binding protein-Pex3p and the glutathione S-transferase-Pex16p fusion proteins. The antibodies directed to GFP (26), Pex13p (41), Pex14p (42), and PMP24 (33) were described before. The anti-PMP70 antibody (43) was kindly provided by Dr. Wilhelm W. Just (University of Heidelberg, Heidelberg, Germany). The antibodies directed to catalase (catalog no. RDI-CATALASEabr; Research Diagnostics, Inc.), KDEL (catalog no. ab12223; Abcam), cytochrome c (catalog no. 556433; BD Pharmingen), and c-Myc (catalog no. 11667149001; Roche Applied Science) were purchased. Rabbit and mouse antibodies were detected on Western blots using alkaline phosphatase-conjugated anti-rabbit and anti-mouse antibodies (Sigma), respectively. Total IgGs were purified from rabbit sera using Protein A-Sepharose beads according to the manufacturer's protocol (Sigma). Anti-Pex3p and anti-Pex16p IgGs were immunopurified by affinity chromatography using 2-ml columns containing the respective recombinant proteins covalently attached to Sepharose 4B exactly as described before (44). IgG solutions were subjected to five cycles of dilution/concentration with SEM using the Vivaspin columns as described above.
Miscellaneous—35S-Labeled proteins were synthesized using the TNT® quick coupled transcription/translation system (Promega) in the presence of RedivueTM L-[35S]methionine (specific activity >1000 Ci/mmol) following the manufacturer's instructions. Where indicated, purified Pex19p recombinant proteins in SEM buffer were added to the transcription/translation reactions to 2.4 µM final concentrations. When different reticulocyte lysates were used in the same experiment, the amounts of the 35S-labeled reporter protein per volume unit were previously quantified by SDS-PAGE and autoradiography. Lysates displaying more than 20% variation in the yield of the 35S-labeled reporter protein were not used in these experiments.
For immunoprecipitations, reticulocyte lysates containing the 35S-labeled reporter protein or organelle suspensions subjected to import reactions were diluted with immunoprecipitation buffer (50 mM Tris-HCl, pH 8.0, 1 mM EDTA-NaOH, pH 8.0, 150 mM NaCl, 1% (w/v) Nonidet P-40, 0.5% (w/v) sodium deoxycholate, 500 µg/ml phenylmethylsulfonyl fluoride, and 1:100 (v/v) mammalian protease inhibitor mixture) for 30 min at 4 °C. Clarified supernatants (15,000 x g, 15 min) were subjected to immunoprecipitations using Protein G-Sepharose 4 Fast Flow (Amersham Biosciences) preincubated with the relevant antisera, as described (44).
Protein concentrations were determined from the absorbance at 280 nm in a buffer containing 6 M guanidine, 20 mM sodium phosphate, pH 6.5, and by using the extinction coefficients calculated by ProtParam (available on the World Wide Web at www.expasy.org).
Densitometric analysis of autoradiographed gels was performed using the UN-SCAN-IT automated system. Apparent KD and IC50 values were determined with GraphPad Prism 4 software from the results of three and five independent experiments, respectively.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
|
15-20% of the labeled GFP-P24 protein added to the assay was converted into this 14-kDa fragment. Finally, proteinase K treatment of import reactions performed in the presence of detergents leads to the degradation of this 14-kDa fragment, indicating that this polypeptide is not intrinsically resistant to the protease used (lane 12). Altogether, these results suggest that the GFP-P24 was inserted into some membrane system via its PMP24 moiety, exposing at the same time the GFP domain into the cytosol. It should be mentioned that we were unable to immunoprecipitate the 14-kDa fragment using the c-Myc antibody. However, we also failed in immunoprecipitating GFP-P24 with this antibody directly from the reticulocyte lysate. The reason for this phenomenon is not known. Thus, we cannot be sure that the C terminus of the reporter protein has reached the lumen of the organelles in which it was inserted, as the in vivo analysis could suggest. Regardless of this uncertainty, the 14-kDa fragment cannot be extracted from the membrane fraction by sonication of the organelles in the presence of solutions with high or low ionic strengths nor by treatment with alkali (see supplemental Fig. 1SB). This behavior strongly suggests that insertion of GFP-P24 into some membrane system(s) did occur. Finally, the acquisition of a protease-resistant status by this 14-kDa fragment is temperature- and time-dependent and does not require hydrolysis of ATP or GTP (see supplemental Fig. 1S, C and D).
|
|
Quite different results were obtained when the recombinant Pex19p protein was included in these in vitro import assays not just during the import reaction itself but rather from the very beginning of our protocol (i.e. during the in vitro translation of GFP-P24). In the experiment described below, the import efficiencies obtained with GFP-P24 translated in the presence of recombinant full-length Pex19p, a recombinant fragment comprising amino acid residues 31-299 (Pex19p-(31-299)) or a protein containing the first 124 amino acid residues of this peroxin (Pex19p-(1-124)) are compared with the import efficiencies obtained with a standard GFP-P24-containing lysate (i.e. no recombinant proteins were added during translation of GFP-P24) in import reactions supplemented with the corresponding recombinant proteins. The final concentration of the recombinant proteins in the import reactions was 6 nM in all cases. As shown in Fig. 5, inclusion of recombinant full-length Pex19p during the translation of GFP-P24 results in a 2.5-fold increase in the amount of protease-protected 14-kDa fragment (Fig. 5, compare lanes 1 and 2). In contrast, when Pex19p-(31-299) is used in these experiments, a strong inhibition of the process is observed but only when the recombinant protein was added during the translation step of GFP-P24 (lanes 3 and 4). This recombinant protein still contains the cargo-binding domain of Pex19p but lacks the Pex3p-interacting domain presumably necessary for the docking step at the peroxisomal membrane (16). Inclusion of Pex19p-(1-124) in the import reactions before or after translation of GFP-P24 has little or no effect on the amount of the 14-kDa fragment present in the peroxisomal membrane. This Pex19p fragment is still able to interact with Pex3p but lacks the cargo protein-binding domain (16). Taken together, these observations suggest that Pex19p interacts with GFP-P24 during or immediately after its synthesis and mediates the insertion of the reporter PMP into the peroxisomal membrane.
|
|
Altogether, these experiments suggest that, on an experimental time scale of minutes, Pex19p, its strongest Pex3p-interacting domain, and Pex3p are involved in the insertion of GFP-P24 into the peroxisomal membrane.
The Pex3p-binding Affinity of Pex19p Is Increased in the Presence of GFP-P24—As shown above, inclusion of recombinant Pex19p during the translation of GFP-P24 leads to a stimulatory effect on the insertion of GFP-P24 into the peroxisomal membrane, whereas Pex19p-(31-299) has the opposite effect. These observations strongly suggest that protein complexes involving the reporter PMP on one side and the recombinant proteins on the other are efficiently formed under these conditions. This prompted us to analyze such protein complexes, the aim being to compare the binding properties of Pex19p for Pex3p in the absence and in the presence of the cargo protein. We first asked whether or not the bacteria-expressed proteins used in this work display the reported binding properties (16). Native PAGE analysis of recombinant Pex19p, Pex19p-(31-299), or Pex19p-(1-124) preincubated with increasing amounts of Pex3p-(34-373) show that indeed Pex19p and its Pex19p-(1-124) fragment interact with the recombinant cytosolic domain of Pex3p (see supplemental Fig. 2S). The apparent KD values of these interactions as estimated by this technique suggest that Pex19p binds Pex3p-(34-373) with a higher affinity (KD 3.4 µM) than Pex19p (1-124) does (KD 26 µM), reflecting the existence of a second Pex3p binding site in the full-length version of this peroxin that is absent in the Pex19p-(1-124) protein (16). However, this second Pex3p-binding site on Pex19p per se is quite weak under these experimental conditions, because no complex comprising Pex3p-(34-373) and Pex19p-(31-299) could be detected in this analysis (but see below).
We next subjected reticulocyte lysates containing GFP-P24 to the same electrophoretic technique. GFP-P24 translated in the absence of recombinant proteins is detected mainly as a group of sharp bands appearing at the top of the gel (see Fig. 6A, arrow a). The nature of these bands is presently not known. They could represent precipitated GFP-P24 or GFP-P24 in complex with some rabbit proteins (e.g. the TCP1 ring complex) (see Ref. 32). In addition, a very diffuse and weak radioactive smear displaying higher electrophoretic mobility is also visible. The presence of radiolabeled GFP-P24 in these two zones of the native gel was confirmed by subjecting an entire lane from a similar native gel to a second electrophoretic separation on an SDS-containing gel (data not shown). When GFP-P24 translated in the presence of recombinant Pex19p is analyzed in these gels, a different result is obtained. The radioactive bands displaying low electrophoretic mobility contain now a minor fraction of GFP-P24. The majority of the radiolabeled protein is detected as a smear followed by a weak but sharp band (Fig. 6A, arrow c). Interestingly, the addition of recombinant Pex19p to both GFP-P24-containing lysates prior to electrophoresis increases significantly the amount of GFP-P24 in this sharp band. Apparently, the amount of recombinant Pex19p present in all of these samples also determines the amount of GFP-P24 that can be detected in this zone of the gel. This suggests that the GFP-P24-Pex19p complexes present in the original reticulocyte lysates (presumably involving endogenous rabbit Pex19p in one case and this rabbit protein plus recombinant Pex19p in the other) are relatively labile when analyzed by this electrophoretic technique. The presence of a vast excess of recombinant Pex19p during electrophoresis ensures that GFP-P24 that dissociates from the original Pex19p-containing complexes is rapidly recaptured by other Pex19p molecules. It should be noted that although equal amounts of radioactively labeled GFP-P24 were processed for each lane (as judged by SDS-PAGE and autoradiography), the levels of labeled GFP-P24 recovered in the sharp band are much higher for the Pex19p-supplemented lysate than for the standard one. These results indicate that the fraction of GFP-P24 available to enter these gels is also determined by the amount of Pex19p present during the translation step of GFP-P24. Data indicating that this sharp band indeed corresponds to a GFP-P24-Pex19p complex come from the two following observations. First, the addition of recombinant Pex3p-(34-373) to the two reticulocyte lysates results in a GFP-P24-containing protein complex displaying a smaller electrophoretic mobility (Fig. 6A, arrow b). This observation strongly suggests that a trimeric complex containing GFP-P24, Pex19p, and Pex3p-(34-373) was formed. Second, when these experiments are performed with recombinant Pex19p-(31-299) instead of full-length Pex19p, a sharp band that migrates slightly slower than the GFP-P24-Pex19p complex is also observed. However, the addition of recombinant Pex3p-(34-373) to these samples has only a partial effect on the electrophoretic mobility of the GFP-P24-containing protein complex: the majority of the radiolabeled protein retains its electrophoretic mobility in agreement with the fact that Pex19p-(31-299) lacks the strongest Pex3p-binding domain present in the full-length peroxin; a minor fraction of the GFP-P24-containing complex, however, was shifted to an upper region of the gel, suggesting that this truncated version of Pex19p does have the capacity to interact with Pex3p, although with a lower affinity. Please note that we were unable to detect formation of a protein complex comprising only these two recombinant proteins (see above and supplementary Fig. 2S). Apparently, this result is related to the presence of GFP-P24 in these experiments.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
Considering that Pex19p is involved in the biogenesis of PMPs, the finding that the addition of high amounts of recombinant Pex19p to our in vitro import assays not only does not stimulate the insertion process of our reporter protein into the peroxisomal membrane but also has exactly the opposite effect may seem surprising. There are probably two reasons behind this observation. The first is related to the properties of the Pex19p-PMP protein complex. The available data strongly suggest that this interaction has to occur very early in the life of a PMP, during or immediately after its translation (20); otherwise, these hydrophobic proteins probably precipitate, giving rise to forms that are no longer import-competent. Thus, the subsequent addition of Pex19p to our import reactions has no positive effect. The second reason derives from the fact that recombinant Pex19p interacts with Pex3p-(34-373) even in the absence of cargo proteins. Considering that this domain of peroxisomal Pex3p is exposed into the cytosol (56), there is no reason to believe that this interaction does not occur also in our in vitro import assays. It should be noted that the KD value estimated for the Pex19p-Pex3p interaction by native PAGE is 13-fold higher than the IC50 value obtained for Pex19p in the in vitro import assays. However, KD values estimated by this technique may be overestimated; native-PAGE is a nonequilibrium technique in which proteins are subjected to nonphysiological conditions. Thus, the inhibitory effect observed in the presence of high concentrations of Pex19p in the import assays may be due to the titration of peroxisomal Pex3p. If this is so, it could be argued that the capacity of free Pex19p (i.e. with no cargoes) to interact with peroxisomal Pex3p is a property unique to the recombinant version of Pex19p used here. Data from two different laboratories strongly suggest that this is not the case. Pex19p is a low abundance protein in mammalian cells, which under steady-state conditions cannot be detected in peroxisomes by current immunofluorescence techniques (10). However, overexpression of N-terminally tagged versions of Pex19p in these cells results in a Pex3p-dependent peroxisomal localization of the engineered peroxin despite the fact that no peroxisomal proliferation occurs in these cells (8, 21, 22). More importantly, exactly the same findings were made when using a truncated molecule lacking the PMP-binding domain of Pex19p (21). Thus, it seems that also in vivo peroxisomal Pex3p has the capacity to interact with Pex19p molecules that carry no cargo proteins whenever the concentration of this peroxin is well above its normal physiological values.
One of the major conclusions reported in this work derives from the experiments in which GFP-P24 was first translated in the presence of different versions of recombinant Pex19p and afterward used in the in vitro import assays. Our results indicate that full-length Pex19p stimulates the insertion process of the reporter protein into the peroxisomal membrane, whereas Pex19p-(31-299) has exactly the opposite outcome. No such effects were observed when the recombinant proteins were included at the same concentration in the import assays programmed with a standard GFP-P24 reticulocyte lysate. These findings indicate that the import competence of GFP-P24 is dictated by the Pex19p molecules that predominate during its translation and not by the ones that predominate during its peroxisomal import (be it a recombinant version of Pex19p or the endogenous rat liver peroxin). Thus, we can only conclude that recombinant Pex19p or Pex19p-(31-299) formed a complex with GFP-P24 during its translation and that it is as such that GFP-P24 is presented to the peroxisomal docking/insertion machinery. Whether or not collision of these GFP-P24-containing complexes with the peroxisomal machinery results in the insertion of the protein into the organelle membrane depends on the functionality of the Pex19p molecules present in these complexes. In contrast to recombinant full-length Pex19p, Pex19p-(31-299) fails in promoting the insertion of GFP-P24 into the peroxisomal membrane presumably because it does not interact with peroxisomal Pex3p as strongly as the full-length peroxin does. The observation that anti-Pex3p IgGs block the insertion of GFP-P24 into the peroxisomal membrane is compatible with this interpretation.
We note that, presently, the in vivo concentration of Pex19p is not known for any organism. Thus, we cannot compare the concentration of recombinant Pex19p used in our assays with its intracellular values. There are, however, some figures that may be relevant here. According to the morphometric data of the rat hepatocyte (57), if there were one molecule of Pex19p per peroxisome, its cytosolic concentration would be 0.23 nM. Thus, a 6 nM concentration of Pex19p in the cell cytosol would correspond to 26 molecules of Pex19p per peroxisome. It may also be important to note that rat Pex3p is as abundant as Pex14p (58), and we have shown previously that this peroxin corresponds to 0.25% of total peroxisomal rat protein (44). A value of 0.25% for Pex3p is also compatible with the observation that a 38-kDa component of rat liver PMPs comprising Pex3p and another peroxisomal membrane protein corresponds to 0.37% of total peroxisomal proteins (41). Considering that peroxisomes comprise 2.5% of total liver protein and that we use in our assays 450 µg of PNS protein in 100 µl, then the concentration of rat Pex3p (Mr = 42,209) in our import reactions should be around 6.6 nM. Thus, the recombinant Pex19p used in the experiment described above does not seem to be present in a molar excess in relation to the endogenous Pex3p.
Insertion of GFP-P24 into the peroxisomal membrane was insensitive to the inclusion of apyrase or high concentrations of nonhydrolyzable ATP/GTP analogs in the import reactions, indicating that docking of the Pex19p-cargo protein complex at the organelle membrane and the subsequent insertion of the reporter PMP into the lipid bilayer do not require ATP/GTP hydrolysis. These data are in agreement with previous results on the energetics of import of PMP22 and PMP70 (30, 31) but are in stark contrast with the recent proposal that targeting of Pex19p into the peroxisome is an ATP-dependent process (19). It should be noted, however, that this ATP-dependent process was observed in an in vitro import system with a different configuration. Instead of following the fate of a labeled reporter PMP, the authors monitored the behavior of Pex19p. Presumably, the postnuclear supernatant used in those experiments was the source of the cargo proteins for the labeled Pex19p. Unfortunately, the mechanistic details of the Pex19p-PMP interaction are still unknown. (Does it occur co-translationally? Is it mediated by ATP-dependent chaperones?) Thus, the exact step of this pathway that requires hydrolysis of ATP remains to be determined.
Another interesting observation presented here regards the properties of the Pex19p-Pex3p interaction. Our results indicate that the binding affinity of Pex19p for Pex3p is increased in the presence of the cargo protein, GFP-P24. We do not know if this property derives from some cargo-induced conformational alterations in Pex19p or if Pex3p also interacts with the cargo protein, giving rise to a divalent interaction involving the Pex19p-cargo protein on one side and Pex3p on the other. Different methodologies will have to be used in order to clarify this point. However, regardless of the molecular interfaces involved in these interactions, it is evident that an increase in the Pex3p-binding affinity provides the means to direct Pex19p to the docking/insertion machinery only after Pex19p is preloaded with a cargo protein.
Altogether, the data presented here shed new light into the Pex19p-mediated protein import pathway. Our results strongly suggest that Pex19p-PMP complexes formed in the absence of peroxisomes are indeed the substrates for the peroxisomal docking/insertion machinery. Whether or not Pex19p also participates in subsequent steps of the import pathway of PMPs (e.g. in the membrane insertion step or in protein complex assembly/disassembly) will be the aim of future work. The tools and experimental strategies described here will surely help in addressing these issues.
|
FOOTNOTES |
|---|
The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. 1S and 2S.
1 Supported by Fundação para a Ciência e Tecnologia.
2 Present address: Instituto de Genética Médica Jacinto Magalhães, Praça Pedro Nunes, 4050-466 Porto, Portugal.
3 To whom correspondence should be addressed: Instituto de Biologia Molecular e Celular, Universidade do Porto, Rua do Campo Alegre, 823, 4150-180 Porto, Portugal. Tel.: 351-226074900; Fax: 351-226099157; E-mail: jazevedo{at}ibmc.up.pt.
4 The abbreviations used are: PMP, peroxisomal membrane protein; GFP, green fluorescent protein; PNS, postnuclear supernatant; MOPS, 4-morpholinepropanesulfonic acid; DTT, dithiothreitol; ATPS, adenosine 5'-O-(thiotriphosphate); GTPS, guanosine 5'-O-(thiotriphosphate).
5 C. P. Guimarães, A. M. Gouveia, and J. E. Azevedo, unpublished results.
6 C. Reguenga and J. E. Azevedo, unpublished results.
7 M. P. Pinto, C. P. Grou, I. S. Alencastre, M. E. Oliveira, C. Sá-Miranda, M. Fransen, and J. E. Azevedo, unpublished observations.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
+ refers to import reactions that received the highest amount of the recombinant proteins previously subjected to an incubation at 95 °C for 5 min. After protease treatment, samples were processed for SDS-PAGE and blotted onto a nitrocellulose membrane. After autoradiography, the membrane was probed with anti-catalase antibodies (CAT) to control loadings. The numbers indicate the concentrations of the recombinant proteins in nM present in each import reaction. B, PNS fractions were preincubated 20 min on ice with 8 µg of immunopurified IgGs directed to Pex3p (lane 1) or Pex16p (lane 2) or with total IgGs isolated from an anti-PMP70 serum (lane 3) or from a nonimmune serum (lane 4) and afterward subjected to an import reaction with 35S-labeled GFP-P24. A control import reaction preincubated under the same conditions but with no IgGs (lane 5) was also performed. Samples were analyzed as described above. 14F, 14-kDa protease-resistant fragment of GFP-P24.
