View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Biochimica et Biophysica Acta 1763 (2006) 6 – 17 http://www.elsevier.com/locate/bba

Identification and characterization of three peroxins—PEX6, PEX10 and PEX12—involved in glycosome biogenesis in Trypanosoma bruceii

Hanane Krazy, Paul A.M. Michels *

Research Unit for Tropical Diseases, Christian de Duve Institute of Cellular Pathology and Laboratory of Biochemistry, Universite´ catholique de Louvain, ICP-TROP 74.39, Avenue Hippocrate 74, B-1200 Brussels, Belgium

Received 10 October 2005; received in revised form 5 November 2005; accepted 7 November 2005 Available online 7 December 2005

Abstract

Protozoan Kinetoplastida such as the pathogenic trypanosomes compartmentalize several important metabolic systems, including the glycolytic pathway, in -like organelles designated glycosomes. for three involved in glycosome biogenesis of Trypanosoma brucei were identified. A preliminary analysis of these proteins, the peroxins PEX6, PEX10 and PEX12, was performed. Cellular depletion of these peroxins by RNA interference affected growth of both mammalian bloodstream-form and insect-form (procyclic) trypanosomes. The bloodstream forms, which rely entirely on glycolysis for their ATP supply, were more rapidly killed. Both by immunofluorescence studies of intact procyclic T. brucei cells and subcellular fractionation experiments involving differential permeabilization of plasma and organellar membranes it was shown that RNAi-dependent knockdown of the expression of each of these peroxins resulted in the partial mis-localization of different types of glycosomal matrix enzymes to the cytoplasm: proteins with consensus motifs such as the C-terminal type 1 peroxisomal targeting signal PTS1 or the N-terminal signal PTS2 and a for which the sorting information is present in a polypeptide-internal fragment not containing an identifiable consensus sequence. D 2005 Elsevier B.V. All rights reserved.

Keywords: Trypanosoma; Glycosome biogenesis; Peroxin; RNA interference

1. Introduction of the single mitochondrion containing DNA uniquely orga- nized and expressed [1–3], and glycosomes, Trypanosomes are parasitic organisms belonging to the which have as their most distinctive feature the presence of the protozoan order Kinetoplastida. All members of this taxonomic majority of the enzymes of the glycolytic pathway [4,5].In group possess a number of distinctive structural and metabolic other organisms, glycolysis occurs in the cytosol. In addition to properties such as the presence of peculiar organelles. The glycolytic enzymes, the glycosomes may contain enzymes or unique organelles comprise the kinetoplast, a subcompartment (parts of) other pathways, such the gluconeogenic and pentosephosphate pathways, pyrimidine biosynthesis and purine salvage, and in common with peroxisomes in other Abbreviations: ALD, aldolase; ENO, enolase; gGAPDH, glycosomal eukaryotes, enzymes of peroxide metabolism, fatty-acid glyceraldehyde-3-phosphate dehydrogenase; PEX, peroxin; PTS, peroxisome- oxidation and ether-lipid biosynthesis [4]. The classification targeting signal; PYK, pyruvate kinase; TIM, triosephosphate isomerase of glycosomes within the organelle family of peroxisomes, also i The new nucleotide sequence data reported in this paper are available in the DDJB/EMBL/GenBank databases under the accession numbers DQ226512 including yeast microbodies, plant glyoxysomes and fungal (PEX6), DQ226513 (PEX10) and DQ226514 (PEX12). (2) Slightly different Woronin bodies, is further based on the fact that all these systems are being used for abbreviations of proteins and genes in research on organelles have a single boundary membrane, contain no DNA, peroxisomes/glyoxysomes/glycosomes in mammals, plants, yeasts and proto- and follow similar routes of biogenesis [4–7]. zoa. Throughout this paper, we used the official nomenclature system for The matrix proteins of the organelles are encoded by nuclear trypanosomatid protozoa (upper case for proteins, italicized upper case for genes) [58]. genes, synthesized on free ribosomes in the cytosol and post- * Corresponding author. Tel.: +32 2 764 74 73; fax: +32 2 762 68 53. translationally imported usually without any detectable form of E-mail address: [email protected] (P.A.M. Michels). processing. Previous research, mainly performed on different

0167-4889/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.bbamcr.2005.11.002 H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17 7 yeasts and mammalian cells, has shown that the import process of glycosomal enzymes, and for trypanosomatid metabolism can conveniently be considered as comprising a number of and viability has been analyzed [17,21,22]. In contrast, little is distinct steps [6–11]. (1) Recognition of the proteins by a known about the peroxins involved in later steps of import of receptor; (2) docking of the receptor with its bound cargo on glycosomal matrix proteins. Only studies on trypanosomatid the membrane; (3) membrane translocation of the receptor– PEX2 homologues have been reported [23–26]. protein complex; (4) cycling of the discharged receptor back to Here, we present our work on the identification and the cytosol. The process is mediated by proteins called preliminary characterization of three T. brucei proteins peroxins, abbreviated PEX. In the first step, two different homologous to peroxins involved in the third and fourth steps cytosolic receptors, PEX5 and PEX7 recognize proteins to be of import of proteins into the peroxisomal matrix: PEX10 and imported through their peroxisome-targeting signals, PTS1 and PEX12, presumably responsible for the translocation process, PTS2, respectively. PTS1 is the most commonly used signal; it and PEX6, involved in receptor cycling. Moreover, we provide consists of a C-terminal tripeptide -SKL or a sequence of evidence that indeed these trypanosomal proteins are essential amino acids with similar physico-chemical properties. Yet, the for proper glycosome biogenesis. ¨10 residues preceding the tripeptide may to some extent affect the efficiency of import. PTS2 is a less frequently used 2. Materials and methods signal; it comprises a nonapeptide motif—(R/K)(L/H/ 2.1. Trypanosomes and growth conditions V)X5(QH)(LA)—close to the N-terminus of the protein. Some peroxisomal proteins do not have a PTS1 or PTS2, but may use Bloodstream and procyclic form T. brucei 427, cell line 449 [27], either a non-conserved sequence, often internal to the constitutively expressing the Escherichia coli tetracycline repressor polypeptide (I-PTS), to interact with PEX5 or PEX7, or enter integrated in its genome, were used in this study. Bloodstream forms were by forming a heteromeric complex with PTS1- or PTS2- cultured in HMI-9 medium containing 10% heat-inactivated foetal calf serum bearing proteins (Fpiggy-backing_). Indeed, it has been shown (Invitrogen) and 0.2 AgmLÀ1 phleomycin (Cayla), the selectable marker for the - that peroxisomal proteins may enter the organelle in folded and tetracycline repressor construct, at 37 C under water-saturated air with 5% CO2. Procyclic trypanosomes were grown in normal SDM-79 medium [28] even oligomeric form [12]. The cargo–receptor complex, supplemented with 10% foetal calf serum and 0.5 AgmLÀ1 phleomycin at 28 possibly even in conjunction with auxiliary proteins, interacts -C under water-saturated air with 5% CO2. Cultures were always harvested in with the membrane through a docking complex minimally the exponential growth phase, i.e., at densities lower than 2 Â 106 cells mLÀ1 comprising PEX13 and PEX14. The complex is then trans- for bloodstream forms and 1 Â 107 cells mLÀ1 for procyclic cells, by ported through the membrane, probably mediated by the so- centrifugation at 1900Âg for 10 min. called translocation complex of integral membrane RING- 2.2. Identification of sequences for the trypanosomatid homologues of finger proteins, PEX2, PEX10 and PEX12, followed by cargo peroxins PEX6, PEX10 and PEX12 release at the matrix side of the membrane. Although the notion that the receptor moves through the membrane is generally The database of the T. brucei (strain TREU927/4) genome project, when accepted, some controversy exists as to whether it effectively still at a stage of partial completion, was searched for sequences homologous to enters the matrix or only releases the cargo at the membrane’s those of various yeast and mammalian peroxins. Short fragments of putative inner surface [8]. But consensus exists that the receptors PEX5 candidates were identified in searches performed with PEX6, PEX10 and PEX12 sequences. The corresponding fragments were amplified from genomic and PEX7 return to the cytosol, mediated minimally by an DNA of T. brucei strain 427, using Taq DNA polymerase (TaKaRa), ligated in integral membrane protein (PEX15 in yeast or the non- the pGEM-T Easy vector (Promega), cloned and sequenced with a Beckmann homologous PEX26 in mammalian cells) and two soluble CEQ2000 apparatus (Beckman Instruments, Inc.) using the CEQ DTCS Dye peroxins belonging to the AAA+ protein family, PEX1 and Terminator Cycle sequencing kit (Amersham Biosciences). A purified PEX6. Several recently reported observations suggest that recombinant plasmid was then used as a probe to screen a genomic T. brucei 427 library prepared in E. coli strain MB406 with phage EGEM11 [29]. mono- or di-ubiquitylation of the import receptors functions as Positive clones were isolated, and DNA fragments subcloned and sequenced. a signal for PEX1 and PEX6 to transfer them back to the Protein sequences were aligned using the program Clustal W [30]. cytosol [reviewed in [11]]. It should be noted that an alternative function invoked for the RING-finger proteins PEX2, PEX10 2.3. Molecular biological methods and PEX12 is an involvement in this return of the receptors to the cytosol, rather than in the import of the receptor–cargo For most experiments in molecular biology standard methodologies were complex [10]. In this respect, it is important to note that RING- used [31], or protocols were followed as provided by suppliers of enzymes used for various forms of DNA and RNA manipulation (Fermentas, Roche Applied finger domains are characteristic elements of E3-ubiquitin Science, New England Biolabs, Promega, Invitrogen, TaKaRa). E. coli strain ligases. XL-1-Blue (Stratagene) was used for all plasmid cloning. The first two steps of the process by which proteins are imported into the glycosomal matrix of the kinetoplastids 2.4. Construction of expression systems, purification of recombinant Trypanosoma brucei and Leishmania donovani have already proteins, and antiserum production received some attention in previous research. Cloning and characterization of PEX5 and PEX14 of these parasites have Truncated versions of T. brucei 427 PEX6 (993 bp from positions 1956 to 2949 and 1014 bp from 1935 to 2949, both constructs comprising the AAA been reported, and the interaction between these peroxins and motifs D1 and D2) were amplified by PCR using Invitrogen Platinum Pfx DNA with PTS1 proteins has been studied in detail [13–20]. Also, polymerase, a sense oligonucleotide containing an NdeI restriction site just the importance of these peroxins for proper compartmentation upstream of the chosen (or introduced) ATG start codon and an antisense 8 H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17 oligonucleotide with a BamHI site immediately following the stop codon. PCR cold Zimmerman Post-Fusion Medium [33], and resuspended in this medium. products were purified, cloned using the vector pGEM-T Easy and, after 2–3 Â 107 cells in 0.5 mL together with 10 Ag of linearized DNA were checking their sequence, excised from the recombinant plasmid by digestion incubated for 10 min on ice, in a 4 mm electroporation cuvette. Cells were then with NdeIandBamHI and ligated to the expression plasmid pET28a subjected to a single pulse by the electroporator set for a peak discharge of 1.6 (Novagen), digested with the same enzymes. Each plasmid directs the synthesis kV, 25 Ohm and 25 AF, to give a pulse with a time constant of ¨300 As. After of a PEX construct with an N-terminal, 20 residues long extension including six electroporation, cells were diluted in 4.5 mL of SDM-79 medium and successive His residues (FHis6-tag_). E. coli strain BL21(DE3) cells were incubated overnight without selection. Subsequently, the culture was diluted transformed with these constructs. serially into 24 wells of a microtiter plate using 4.5 mL conditioned SDM-79 Cells harbouring a recombinant plasmid for expression of truncated PEX6 medium containing 30 AgmLÀ1 hygromycin. Antibiotic-resistant cells were were grown in 400 mL or 1 L TB medium [31] supplemented with 30 AgmLÀ1 detectable after about 8 days. To prepare conditioned medium, the supernatant 6 kanamycin at 37 -C. Once the culture had reached an OD600 nm of 0.7–0.9, the obtained from a procyclic culture in the exponential growth phase (7–9 Â 10 temperature was reduced to 15 -C and 1 h later isopropyl h-d-thiogalactano- cells mLÀ1) was filter sterilized and stored for a few weeks at 4 -C or frozen. pyranoside was added to a final concentration of 1 mM to induce expression of Following transfection (done in triplicate for each DNA construct) and recombinant protein and growth was continued for 15–16 h. Protein selection of positive clones, correct integration of RNAi constructs into the purifications were performed at 0–4 -C. Cells were harvested by centrifugation genome of bloodstream and procyclic forms was checked by PCR amplification (5000Âg, 15 min), resuspended in 20–40 mL of cell lysis buffer containing of the hygromycin resistance gene. Positive clones were stored at À80 -Cin 100 mM triethanolamine/HCl, pH 8, 1 M NaCl, 10% glycerol and a complete HMI-9 or SDM-79 medium containing 12 or 10% glycerol for bloodstream and protease inhibitor cocktail (Roche Applied Science). Cells were disrupted by procyclic forms, respectively. three passages through a SLM-Aminco French pressure cell (SLM Instruments, The expression of double-stranded RNA was induced by addition of Inc.) at 90 MPa. Nucleic acids were degraded by incubation of the lysate with tetracycline to the cultures at a concentration of 0.25 AgmLÀ1 for bloodstream- 100 units of Benzonase (Merck) for 30 min at 37 -C. Cell debris and insoluble form trypanosomes and 5.0 AgmLÀ1 for procyclic forms. Cultured cells were material were then removed by centrifugation (10,000Âg, 15 min). To the serially diluted each 24 h to 2 Â 105 cells mLÀ1 and 106 cells mLÀ1 for supernatant was then added 10 mM imidazole and 0.125–0.25 volumes of bloodstream and procyclic form trypanosomes, respectively. Talon Metal Affinity Resin (BD Biosciences-Clontech) and the suspension was mixed for 30 min at room temperature. The resin with bound protein was 2.6. Subcellular fractionation by treatment of cells with digitonin washed twice with five column volumes of lysis buffer containing 70 mM imidazole, followed by elution of the peroxin using three column volumes of 8 9 Procyclic trypanosomes (10 –10 cells) were taken from either a culture not the same buffer containing 200 mM imidazole. Purification was checked by induced for RNAi or an induced culture, 24 or 48 h after tetracycline addition. SDS-PAGE followed by Coomassie Blue staining or immunodetection after The cells were centrifuged, washed twice in an ice-cold buffer containing 25 mM preparation of Western blots and using a mouse anti-His antiserum (see below). HEPES, pH 7.4, 250 mM sucrose and 1 mM EDTA, and then resuspended at >1 The purified truncated PEX6 polypeptides were used to raise polyclonal mg protein mLÀ1 in 0.5 mL of the same buffer containing a complete protease antisera in rabbits (Eurogentec). The anti-PEX6 sera were highly specific, as inhibitor cocktail (Roche Applied Science). Subsequently, variable amounts of they were able to detect only a single band on Western blots of whole cell the detergent digitonin [final concentration between 0 and 1 mg (mg of lysates of T. brucei cells. 1 protein)À ] were added to aliquots of the cell suspensions (each containing 100 Attempts to express in soluble form, and/or to purify the entire PEX6 and Ag of protein), followed by 0.2 mL HBSS buffer (Invitrogen) and incubated for 4 PEX12 proteins, and a truncated version of PEX12 remained unsuccessful. min. After centrifugation of the suspensions (12,000Âg for 2 min), 0.15 mL of PEX10 was expressed and purified as a recombinant protein, and a rabbit the supernatant was taken and kept on ice for Western blot analysis of proteins antiserum was raised, but experiments conducted to determine the specificity of released from the different cell compartments, or stored at À20 -C. As controls this antiserum were non-conclusive. (no protein release and release from all compartments, respectively) were taken cells not treated with digitonin and cells to which 0.02% Triton X-100 (final 2.5. RNA interference studies concentration) had been added to assure complete protein release due to full disruption of plasma membrane and all organellar membranes. The T. brucei-specific vector pHD677 [27] was used to generate stable bloodstream and procyclic cell lines for the tetracycline-inducible expression of 2.7. Protein measurements, SDS/PAGE, Western blot analysis a double-stranded RNA of PEX6, PEX10 and PEX12. Double-stranded RNAs were produced as hairpin molecules by transcription of constructs comprising Protein concentrations were determined using the Bio-Rad (USA) protein two similar fragments of the coding regions of each gene, one somewhat longer assay, based on the Coomassie Blue-binding procedure [34], using bovine serum than the other, arranged as a direct inverted repeat. These constructs were albumin as a standard. SDS-PAGE was carried out by the Laemmli method [35]. prepared by amplification of fragments for PEX6 (sense: nucleotides from After electrophoresis, gels were either stained with Coomassie Brilliant Blue, or positions 798 to 1488, antisense: nucleotides 793 to 1406), PEX10 (sense: used for immunoblotting according to a standard procedure [36], using a nucleotides 22 to 885, antisense: nucleotides 10 to 807) and PEX12 (sense: polyvinylidene difluoride membrane (Roche Applied Science). For immunode- nucleotides 28 to 840, antisense: nucleotides 19 to 774), using appropriate tection of proteins, primary antibodies were diluted in PBS, pH 7.3, containing primers. All RNAi constructs were prepared for chromosomal integration in the 0.1% Tween 20 and 5% (w/v) low-fat milk powder. The secondary antibodies, transcriptionally silent ribosomal RNA gene repeat spacer. anti-(mouse IgG) or anti-(rabbit IgG) conjugated to horseradish peroxidase For transfection of bloodstream-form trypanosomes, 10–15 mL of culture (Rockland Immunochemicals) were diluted 1:40,000 and visualised with the was harvested at a density of ¨1.6 Â 106 cells mLÀ1, washed once with 5 mL ECLWestern blotting system, a luminol-based system (Amersham Biosciences). cytomix [32] and resuspended in the same solution. The trypanosome The primary antibodies used were mouse anti-His (1:20,000), or the following suspension (0.4 mL, containing 1 Â 107 cells mLÀ1) was incubated with 10 antisera raised in rabbits against purified proteins (either natural proteins purified Ag DNA linearized by digestion with NotI, and subsequently subjected to a from bloodstream-form trypanosomes as described [37], or recombinant proteins single discharge by a Genetronics BTX ECM630 electroporator with a 1.25 kV, purified from E. coli): anti-aldolase (ALD), 1:200,000; anti-triosephosphate 25 Ohm and 50 AF setting, to obtain a pulse with a time constant of ¨260 As. isomerase (TIM), 1:10,000; anti-glycosomal glyceraldehyde-3-phosphate dehy- The cells were diluted in 12 mL culture medium, supplemented with 20% foetal drogenase (gGAPDH), 1:100,000 [38–40]. calf serum, and incubated overnight. Subsequently, the culture was split into 24 wells of a microtiter plate. To select positive transfectants, hygromycin (Sigma) was added after 24 h at a final concentration of 5 AgmLÀ1. Antibiotic-resistant 2.8. Immunofluorescence studies cells were detectable after about 5 days. To create stable transfectants of procyclic forms, 3–8 mL of a culture at a Procyclic trypanosomes (105 –106 cells) were washed twice in ice-cold density of 4–8 Â 106 cells mLÀ1 were centrifuged, washed once in 2 mL ice- phosphate-buffered saline (PBS), pH 7.3, fixed with 6% formaldehyde in PBS .Kay ...Mces/Bohmc tBohsc ca16 20)6–17 (2006) 1763 Acta Biophysica et Biochimica / Michels P.A.M. Krazy, H.

Fig. 1. Multiple alignment of PEX6 amino-acid sequences. The trypanosomatid sequences are compared with sequences with confirmed PEX6 activity. The Walker A and B motifs of the two AAA domains, D1 and D2 are indicated as: AD1 .,BD1 o,AD2 r, and BD2 j. The AAA signature is underlined. Conserved residues are given against a black background, residues at positions with conservative substitutions are in bold and boxed. Database accession numbers for the trypanosomatid sequences have been described in the main text. Other sequences used in the alignment, with their accession codes for the Swiss-Prot database are: Homo sapiens (Q13608), Saccharomyces cerevisiae (P33760) and Yarrowia lipolytica (P36966). The figure was produced with the program ESPript 2.2 (http://espript.ibcp.fr/ESPript/cgi-bin/ESPript.cgi). 9 10 H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17 for 10 min, permeabilized by treatment with 1% Triton X-100 for 10 min. After mass of 106.0 kDa and a pI value of 5.0 (calculated using the l spreading the trypanosomes on poly- -lysine-coated slides, they were computational tools available at http://us.expasy.org/). After incubated for 45 min in PBS containing 5% BSA at 37 -C, followed by incubation in PBS containing 2% BSA and the primary antisera: mouse completion of our analysis, the near-complete genome se- monoclonal anti-T. brucei gGAPDH [41] at 1:2700–3500 dilution and rabbit quence of T. brucei TREU927/4 and that of the related polyclonal anti-T. brucei ALD (see above) at 1:1500. After washing with PBS, trypanosomatid parasites Leishmania major and T. cruzi be- cells were allowed to react with fluorescein-conjugated secondary antibodies (5 came available in the GeneDB database (http://www.geneDB. À1 AgmL Alexa Fluor 488 anti-mouse IgG and Alexa 568 anti-rabbit IgG- org/), and PEX6 homologues could be identified: T brucei, Molecular Probes), washed again and mounted in Mowiol containing 2 AM TO-PRO-3 (Molecular Probes) for DNA staining. Cells were visualised using a accession code Tb05.6E7.930; L. major, LmjF16.0060; T. Zeiss Axiovert microscope coupled to an MRC-1024 confocal scanning laser cruzi, Tc00.1047053506947.40 and Tc00.1047053504251.9). imaging system (Bio-Rad). Further genomic searches proved that the T. brucei sequence analyzed was indeed orthologous with PEX6 of other organ- 3. Results and discussion isms. The putative PEX6s of the two T. brucei strains differ in six amino acids. An alignment of the three trypanosomatid 3.1. Analysis of the sequences of the T. brucei peroxin sequences with those of human and different yeasts is homologues PEX6, PEX10 and PEX12 presented in Fig. 1. The trypanosomatid sequences are 65– 76% identical to each other, but share only 32–34 and 34– 3.1.1. PEX6 36% identity with the human and S. cerevisiae PEX6s, A fragment homologous to mammalian and yeast PEX6 was respectively. PEX6, as well as PEX1, belongs to the AAA+ found in the T. brucei (strain TREU927/4) genome project protein family (ATPases associated with various cellular database, and the corresponding full-length gene of T. brucei activities). Members of this family are characterized by the strain 427 was subsequently cloned and sequenced. The protein presence of one or more AAA modules, a conserved module of encoded by the ORF is 982 amino acids long, has a molecular 220–250 amino acids responsible for ATPase activity [42,43].

Fig. 2. Multiple alignment of PEX10 amino-acid sequences. The trypanosomatid sequences are compared with sequences with confirmed PEX10 activity. The conserved Cys and His residues of the RING motif C3HC4 in the C-terminal region are marked by j, the PEX10 signature is indicated by r. The two transmembrane segments, predicted from the hydropathy character calculated with the algorithm according to Kyte and Doolittle [57], are underlined. Conserved residues are given against a black background, residues at positions with conservative substitutions are in bold and boxed. Database accession numbers for the trypanosomatid sequences have been described in the main text. Other sequences used in the alignment, with their accession codes for the Swiss-Prot database are: H. sapiens (AB013818), (Q9M400), S. cerevisiae (Q05568), Hansenula polymorpha (Q00940). The figure was produced with the program ESPript 2.2. H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17 11

As in the previously characterized PEX6 proteins, also in the similar function of recruitment of a PEX1–PEX6 complex at trypanosomatid ones three parts can be distinguished: an N- the mammalian membrane has been proposed for PEX26, terminal region (N, residues 1–¨400) and two AAA domains which binds directly to PEX6, involving the N-terminal (denoted D1, approximately residues 400–640 and D2, domains of both peroxins, and indirectly to PEX1 [49,50]. approximately 640–980, respectively). The N-terminal domain Also, PEX1 orthologues have recently been identified in the does not match with other known structures. AAA module D2 trypanosomatid databases, each of them having the typical is relatively well conserved, and always possesses intact PEX1/PEX6 domain organization: N/D1/D2, and with the D2 Walker motifs A and B (GPPGCGKT and DELD, respectively; domain better conserved than D1 and the Walker A and B motifs Fig. 1) essential for nucleotide binding and hydrolysis. In only being fully present in D2 (Tb04.2L9.350; LmjF34.3520, contrast, module D1 of all PEX6 homologues is only a weakly Tc00.1047053505989.74 and Tc00.1047053508693.90). How- conserved AAA cassette and its Walker motifs may contain ever, PEX1 is less conserved than PEX6: the trypanosomatid substitutions or even not be recognizable as is the case in the sequences share only 49–60% identity with each other, and a trypanosomatid sequences (Fig. 1), strongly suggesting that mere 27–28 and 23% identity with the human and S. cerevisiae this module, at least in some species, may be incapable of PEX1s, respectively. binding and/or hydrolyzing ATP. It has been shown that PEX6 of mammalian cells and different yeasts form a dynamic 3.1.2. PEX10 and PEX12 hetero-oligomeric complex with PEX1, also structurally PEX10 and PEX12 are, as PEX2, integral membrane proteins organized as N/D1/D2 [44–46]. This interaction is mediated possessing a zinc-binding FRING-finger_ domain which is by the D1 domain of both peroxins, but requires ATP involved in the interaction with other peroxins [51–53]. hydrolysis by domain D2 [47]. Moreover, it has been shown Fragments of candidates for T. brucei PEX10 and PEX12 were for S. cerevisiae that PEX6 is recruited to the peroxisomal found in databases by homology searches and the corresponding membrane by the integral membrane protein PEX15 [48]. This genes of T. brucei strain 427 were subsequently cloned and PEX6 binding occurs through its domain N and requires ATP sequenced. The cloned genes appeared indeed orthologues of binding by PEX6. In mammalian peroxisome biogenesis, a PEX10 and PEX12 from other organisms and code for

Fig. 3. Multiple alignment of PEX12 amino-acid sequences. The trypanosomatid sequences are compared with sequences with confirmed PEX12 activity. The conserved Cys residues of the RING motif C3HC4 in the C-terminal region are marked by j. The two transmembrane segments, predicted from the hydropathy character calculated with the algorithm according to Kyte and Doolittle [57], are underlined. Conserved residues are given against a black background, residues at positions with conservative substitutions are in bold and boxed. Database accession numbers for the trypanosomatid sequences have been described in the main text. Other sequences used in the alignment, with their accession codes for the Swiss-Prot database are: H. sapiens (O00623), A. thaliana (Q9M841), S. cerevisiae (Q04370), Pichia pastoris (Q01961) and Caenorhabditis elegans (Q19189). The figure was produced with the program ESPript 2.2. 12 H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17 polypeptides of 298 and 395 residues, respectively, with calculated molecular masses of 32.9 and 43.1 kDa and pI values of 8.8 and 8.3. The proteins of the different trypanosomatids appeared later also in the GeneDB databases with the following accession codes: PEX10 T. brucei, Tb03.48O8.510 (with two residues differing between strains TREU927/4 and 427), L. major LmjF25.2290; T. cruzi Tc00.1047053508479.190; PEX12 T. brucei, Tb10.61.0440 (one residue difference between the two strains), L. major, LmjF19.1250; T. cruzi, Tc00. 1047053503809.20 and Tc00.1047053503641.19. Figs. 2 and Fig. 5. (A) Northern blot analysis of total RNA purified from procyclic trypanosomes before (0 h) and after (24 h and 48 h) induction of RNAi targeted 3 show alignments of the trypanosomatid putative PEX10 and against PEX10, PEX12 and PEX6. The left panel shows the depletion of the PEX12 sequences with the sequences of corresponding proteins respective peroxin mRNAs, the right panel presents the data for a not-targeted with confirmed peroxin activity from various other organisms. control transcript (enolase). Each lane contained 10 Ag of total RNA. (B) For PEX10, the trypanosomatid sequences are 53–65% Western blot analysis of lysates from procyclic trypanosomes. The lanes identical to each other, but display only 26–28 and 25–26% contain each 10 Ag of protein from trypanosomes of the wild-type T. brucei 449 cell line, or the PEX6 RNAi cell line before (0 h) and after (24 h and 48 h) identity with their human and S. cerevisiae orthologues, induction. respectively. For PEX12, these values are 40–57% identity between trypanosomatids, and 18–20% and 13–17% between trypanosomatids and human and S. cerevisiae, respectively. conserved C3HC4 consensus (C–X2–C–X(9 – 39) –C–X(1 –3) – The sequences of both T. brucei PEX10 and PEX12 contain H–X(2 – 3) –C–X2 –C–X(3 –47) –C–X2 –C, [54]), whereas the two predicted hydrophobic transmembrane segments, similar to motif in PEX12 deviates from the consensus by the absence of their counterparts from other organisms, and in line with earlier two of the seven Cys residues and the His (Figs. 2 and 3). studies indicating that the N-terminal and C-terminal parts of Furthermore, the putative trypanosomatid PEX10s have in the these proteins are exposed to the cytoplasm. The C-terminal N-terminal part a peptide sequence (TLGQEFC) very similar to parts contain the RING-finger motifs. As in the peroxins of other a motif conserved in all other known PEX10s (TLGEEYV). The organisms, the trypanosomatid PEX10 contains a perfectly function of this motif is still unknown. All these similarities with

Fig. 4. Schematic representation of the linearized constructs prepared for the creation of stable transgenic T. brucei cell lines for tetracycline-inducible RNAi of PEX6, PEX10 and PEX12. The basis of the construct is the vector pHD677 [27]. FrRNA_ represents the targeting sequence of the intergenic rRNA region, with the vector’s unique NotI site used for its linearization; PARP, tetracycline-inducible PARP (Fprocyclic acidic repetitive protein_) promoter; black circles, tetracycline repressor bound to the operator of the promoter; FSense_ and Fantisense_ represent the PEX6, PEX10 and PEX12 gene fragments amplified and inserted as an inverted repeat in the polylinker of the vector pHD677, as described in Materials and methods; VSG, constitutive promoter of a FVariant Surface Glycoprotein_ expression site; HYG, hygromycin resistance gene. H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17 13 their orthologues, together with dispersed other short segments The efficiency of mRNA destruction by RNAi in procyclic of similarities highlighted in Figs. 2 and 3, strongly suggest that cells was checked by Northern blot analysis. The RNA levels the putative trypanosomatid PEX10 and PEX12 are true in each of the cell lines were determined before induction and peroxins involved in glycosome biogenesis. 24 and 48 h after tetracycline addition to the cultures. In Fig. 5A, it is shown that 24 h after induction, the RNA levels have 3.2. Knock-down of PEX6, PEX10 and PEX12 expression by decreased to 80%, 70% and 60% for PEX10, PEX12 and RNAi leads to arrest of cell growth PEX6, respectively, and became undetectable after 48 h. The level of a control RNA, for the glycolytic enzyme enolase Bloodstream and procyclic T. brucei 449 cell lines were (ENO), remained unaltered. At the protein level too, a prepared for tetracycline-inducible expression of double- significant decrease could be observed in the case of PEX6, stranded RNA corresponding to either PEX6, PEX10 or for which an antiserum had been raised (Fig. 5B), whereas the PEX12 (Fig. 4). To this end, DNA constructs, which also levels of ENO and pyruvate kinase (PYK), taken as control directed the constitutive expression of the hygromycin resis- proteins, were not affected. Using this same antiserum, it was tance marker, were created and stably integrated in the T. observed that the expression levels of PEX6 were similar in brucei genome, in an intergenic region of a ribosomal RNA wild-type bloodstream and procyclic T. brucei 449 cells (data gene array. not shown).

Fig. 6. Effect of the intracellular depletion of different peroxins by RNAi on the growth of bloodstream form (left panels) and procyclic (right panels) T. brucei. The mRNAs targeted by RNAi were for PEX6 (panels A and D), PEX10 (panels B and E) and PEX12 (panels C and F). Shown are the 10log values of cumulative cell numbers. The growth of trypanosomes transfected with a plasmid producing double-stranded RNA to decrease specific PEX levels was determined for cells cultured in the absence (0) or presence (r) of the RNAi inducer tetracycline (0.25 AgmLÀ1 for bloodstream-form trypanosomes and 5.0 AgmLÀ1 mg for procyclic cells) from t = 0 onward. 14 H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17

The effect of depletion of each peroxin by RNAi on examined how the RNAi-induced degradation of their growth of trypanosomes was studied for both the bloodstream mRNAs affected the compartmentation of glycosomal and procyclic-form cell lines. In the bloodstream forms, enzymes with different PTS: the PTS1 protein gGAPDH, tetracycline addition resulted in reduction of cell numbers the PTS2 protein ALD, and TIM that does not have a already within 24 h for PEX10 depletion, and 24–48 h for recognizable PTS1 or PTS2 but possibly enters the glyco- PEX6 and PEX12 (Fig. 6, left panel). In each case, cell some by forming a heteromeric complex with a PTS1 or density continued to decline for several days, but increased PTS2 protein [6]. The subcellular distribution of these again after 4 to 5 days, eventually leading to the original enzymes, and that of the cytosolic marker enzyme PYK, growth rate, possibly due to the emergence of mutated was studied by treating non-induced procyclic cells, and revertants that escaped RNAi control, a phenomenon well cells cultured for 24 and 48 h in the presence of the inducer known in trypanosomes, particularly bloodstream forms tetracycline, with increasing concentrations of digitonin. This [55,56]. In procyclic trypanosomes, the effect on growth detergent forms insoluble complexes with sterols, thus was less severe, similarly to what has been reported before for permeabilizing the plasma membrane with its relatively high PEX2 and PEX14 [17,26]. A decrease in cell growth was sterol content at lower concentration than the glycosomal only visible after 2 to 3 days (Fig. 6, right panel). For PEX10 membrane that is relatively poor in sterols. Therefore, and PEX12 cell lines, slow growth continued, until after 8 to enzymes sequestered in glycosomes will only be released 10 days, cells started dying. For the PEX6 cell line, growth from the cells at higher concentrations than enzymes present arrest occurred earlier. Cells induced for depletion of PEX12 in the cytosol. Fig. 7, panels A, B and C, show how RNAi- by RNAi showed a distinct phenotype: they stopped moving, induced depletion of PEX6, PEX10 and PEX12, resulted in ballooned and formed aggregates before dying. Similar, albeit partial mis-localization of all three types of glycosomal less pronounced changes were observed for cells in which enzymes. The digitonin-dependent differential appearance of PEX10 was targeted by RNAi. No distinctive morphological enzymes outside the cells, as assayed by Western blotting, phenotype was visible in trypanosomes in which RNAi was was affected by RNAi. Whereas in non-induced cells only induced for PEX6. the cytosolic PYK was released at low digitonin concentra- tion [0.2 mg (mg protein)À1], contrary to the glycosomal 3.3. Destruction of mRNAs of PEX6, PEX10 and PEX12 by enzymes whose release required 0.7 mg digitonin (mg RNAi leads to incomplete compartmentation of glycosomal protein)À1, RNAi induction resulted in release of a part of enzymes the glycosomal proteins too at low detergent concentrations. These experiments show that, in the RNAi-treated trypano- To prove that the cloned genes are indeed peroxins somes, part of the ALD, gGAPDH and TIM is present in involved in the transport of proteins from their site of the cytosol. For each of the three cell lines similar, partial synthesis, the cytosol, into the glycosomal matrix, we relocation of PTS1, PTS2 and I-PTS proteins from glyco-

Fig. 7. Subcellular fractionation by treatment of cells with digitonin. Intact procyclic-form T. brucei cells were incubated with digitonin as indicated. The release of PYK, ALD, TIM and gGAPDH from the cells was assayed after centrifugation of the treated cell suspensions and the preparation of Western blots of the proteins in the resulting supernatants. The subcellular distribution of enzymes is shown for trypanosomes of the cell lines PEX6 RNAi (panel A), PEX10 RNAi (panel B) and PEX12 RNAi (C), before (0 h) and after (24 h and 48 h) the addition of the RNAi inducer tetracycline (Tet) at a concentration of 5.0 AgmLÀ1. For PEX12 the banding pattern at the blot 48 h after RNAi induction (not shown) was identical to that after 24 h. H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17 15 somes to cytosol was reproducibly observed, with slight are more rapidly killed after RNAi induction because the quantitative differences between different experiments and mis-localization of already minor amounts of glycolytic between the proteins within each experiment (gGAPDH not enzymes is much more detrimental than for procyclic cells shown for PEX6, ALD not for PEX10, TIM not for [4,17,21,22]. PEX12). This relocated fraction of matrix proteins presum- To further study the effect of peroxin depletion on import ably represents newly synthesized protein that could not be of glycosomal matrix proteins and glycosome morphology in imported into glycosomes, and consequently, these data may procyclic trypanosomes, immunofluorescence experiments be considered as strong evidence that indeed the three were performed using confocal microscopy. Using double cloned genes encode the authentic T. brucei peroxin labelling, ALD and gGAPDH co-localized as a congruent homologues. Similar experiments proved to be difficult to punctuate pattern in the cytoplasm of non-induced cell lines perform with bloodstream-form trypanosomes, where cells for RNAi of PEX6, PEX10 and PEX12 (Fig. 8A–C, panels labelled -Tet), similar as found in non-transfected cells (not shown) and characteristic of a glycosomal localization [17,26]. In contrast, in the PEX10 and PEX12 RNAi cell lines a diffuse fluorescence labelling along the parasite body was observed for both ALD and gGAPDH, already 24 h after addition of tetracycline (Fig. 8B and C, respectively). After 48 h, the diffuse fluorescence was even more pronounced. These results corroborate the cytosolic mis-localization of glycosomal enzymes as observed in the fractionation studies. In the case of the PEX6 RNAi cell line, induction by tetracycline for 24 h and 48 h showed the same diffuse pattern for gGAPDH as in the other cell lines, but for ALD clusters of bright punctuate structures remained visible in addition to cytoplasmic staining (Fig. 8A). Relocalization of ALD to the cytoplasm had occurred, but to a lesser extent than for gGAPDH. At present, an explanation for this quantitative difference is not known. The results of the immunofluorescence experiments and cell fractionation by digitonin titration demonstrate that the three identified T. brucei genes code for true peroxins. They are involved in glycosome biogenesis; their expression is essential for proper compartmentation of glycosomal matrix proteins. Together, the experiments presented in this paper show that RNAi-specific depletion of the transcripts of PEX6, PEX10 and PEX12 in trypanosomes is directly correlated with reduction of the level of these proteins (only proved for PEX6), mis-localization of glycosomal matrix proteins and growth defects. The observation that each of the three peroxins described in this paper is essential for the trypanosome, together with the low level of sequence conservation between the parasite peroxins and the corresponding proteins in the human host, makes them potentially good targets for trypanocidal drugs. Such drugs could be compounds that disrupt specific interactions with other peroxins involved in glycosome biogenesis without interfering with peroxisome formation in human cells. Fig. 8. Effect of peroxin depletion on the distribution of glycosomal proteins of procyclic-form T. brucei, as determined by immunofluorescence. Shown is the relocation of ALD and gGAPDH in cell lines in which PEX6 (A), PEX10 (B), Acknowledgements and PEX12 (C) was targeted by RNAi. Procyclic trypanosomes were grown in the absence or presence (for 24 or 48 h) of 5.0 AgmLÀ1 of the RNAi inducer This research was supported through grants from the FFonds tetracycline, as indicated. DNA is stained blue with TO-PRO-3. The series de la Recherche Scientifique Me´dicale_ (FRSM) of the panels at the left show the images of trypanosomes obtained by transmission FCommunaute´ franc¸aise de Belgique_ and the Interuniversity microscopy, in combination with TO-PRO staining. The panels at the right present the merged images of the ALD, gGAPDH and TO-PRO staining. (For Attraction Poles-Belgian Federal Office for Scientific, Techni- interpretation of the references to colour in this figure legend, the reader is cal and Cultural Affairs. HK acknowledges a PhD scholarship referred to the web version of this article.) from the FFonds Spe´cial de Recherche, Universite´ Catholique 16 H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17

de Louvain_ (FSR, UCL). We are grateful to Dr. Ve´ronique association involves a novel protein–protein interaction motif, Biochem. Hannaert, Nathalie Galland and Emilie Verplaetse for critical J. 391 (2005) 105–114. [21] T. Furuya, P. Kessler, A. Jardim, A. Schnaufer, C. Crudder, M. Parsons, reading of the manuscript. Glucose is toxic to glycosome-deficient trypanosomes, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 14177–14182. References [22] P.S. Kessler, M. Parsons, Probing the role of compartmentation of glycolysis in procyclic form Trypanosoma brucei: RNA interference [1] J.C. Morris, M.E. Drew, M.M. Klingbeil, S.A. Motyka, T.T. Saxowsky, Z. studies of PEX14, hexokinase, and phosphofructokinase, J. Biol. Chem. Wang, P.T. Englund, Replication of kinetoplast DNA: an update for the 280 (2005) 9030–9036. new millennium, Int. J. Parasitol. 31 (2001) 453–458. [23] J.A. Flaspohler, W.L. Rickoll, S.M. Beverley, M. Parsons, Functional [2] L. Simpson, O.H. Thiemann, N.J. Savill, J.D. Alfonzo, D.A. Maslov, identification of a Leishmania gene related to the peroxin 2 gene reveals Evolution of RNA editing in trypanosome mitochondria, Proc. Natl. Acad. common ancestry of glycosomes and peroxisomes, Mol. Cell. Biol. 17 Sci. U. S. A. 97 (2000) 6986–6993. (1997) 1093–1101. [3] J. Luke´s, D.L. Guilbride, J. Votypka, A. Zikova, R. Benne, P.T. Englund, [24] J.A. Flaspohler, K. Lemley, M. Parsons, A dominant negative mutation Kinetoplast DNA network: evolution of an improbable structure, in the GIM1 gene of Leishmania donovani is responsible for defects in Eukaryot. Cell 1 (2002) 495–502. glycosomal protein localization, Mol. Biochem. Parasitol. 99 (1999) [4] V. Hannaert, F. Bringaud, F.R. Opperdoes, P.A.M. Michels, Evolution 117–128. of energy metabolism and its compartmentation in Kinetoplastida, [25] J. Mannion-Henderson, J.A. Flaspohler, K.R. Lemley, W.L. Rickoll, M. Kinetoplastid. Biol. Dis. 2 (2003) 11 (http://www.kinetoplastids.com/ Parsons, Isolation and characterization of Leishmania mutants defective content/2/1/11). in glycosomal protein import, Mol. Biochem. Parasitol. 106 (2000) [5] M. Parsons, Glycosomes: parasites and the divergence of peroxisomal 225–237. purpose, Mol. Microbiol. 53 (2004) 717–724. [26] C. Guerra-Giraldez, L. Quijada, C.E. Clayton, Compartmentation of [6] J. Moyersoen, J. Choe, E. Fan, W.G.J. Hol, P.A.M. Michels, enzymes in a microbody, the glycosome, is essential in Trypanosoma Biogenesis of peroxisomes and glycosomes: trypanosomatid glycosome brucei, J. Cell Sci. 115 (2002) 2651–2658. assembly is a promising new drug target, FEMS Microbiol. Rev. 28 [27] S. Biebinger, L.E. Wirtz, P. Lorenz, C. Clayton, Vectors for inducible (2004) 603–643. expression of toxic gene products in bloodstream and procyclic [7] P.A.M. Michels, J. Moyersoen, H. Krazy, N. Galland, M. Herman, V. Trypanosoma brucei, Mol. Biochem. Parasitol. 85 (1997) 99–112. Hannaert, Peroxisomes, glyoxysomes and glycosomes, Mol. Membr. Biol. [28] R. Brun, M. Schonenberger, Cultivation and in vitro cloning of procyclic 22 (2005) 133–145. culture forms of Trypanosoma brucei in a semi-defined medium, Acta [8] J.H. Eckert, R. Erdmann, Peroxisome biogenesis, Rev. Physiol. Biochem. Trop. 36 (1979) 289–292. Pharmacol. 147 (2003) 75–121. [29] P.A.M. Michels, M. Marchand, L. Kohl, S. Allert, R.K. Wierenga, F.R. [9] L.A. Brown, A. Baker, Peroxisome biogenesis and the role of protein Opperdoes, The cytosolic and glycosomal isoenzymes of glyceraldehyde- import, J. Cell. Mol. Med. 7 (2003) 388–400. 3-phosphate dehydrogenase in Trypanosoma brucei have a distant [10] I. Heiland, R. Erdmann, Biogenesis of peroxisomes. Topogenesis of evolutionary relationship, Eur. J. Biochem. 198 (1991) 421–428. the peroxisomal membrane and matrix proteins, FEBS J. 272 (2005) [30] J.D. Thompson, D.G. Higgins, T.J. Gibson, CLUSTAL W: improving the 2362–2372. sensitivity of progressive multiple sequence alignment through sequence [11] R. Erdmann, W. Schliebs, Opinion: Peroxisomal matrix protein import: weighting, position-specific gap penalties and weight matrix choice, the transient pore model, Nat. Rev., Mol. Cell Biol. 9 (2005) 738–742. Nucleic Acids Res. 22 (1994) 4673–4680. [12] T. Ha¨usler, Y.D. Stierhof, E. Wirtz, C. Clayton, Import of a DHFR hybrid [31] J.E. Sambrook, E. Fritsch, T. Maniatis, Molecular Cloning: A Laboratory protein into glycosomes in vivo is not inhibited by the folate-analogue Manual, Cold Spring Harbor Laboratory Press, New York, 1989. aminopterin, J. Cell Biol. 132 (1996) 311–324. [32] M.J. van den Hoff, A.F. Moorman, W.H. Lamers, Electroporation in [13] S. de Walque, J.A.K.W. Kiel, M. Veenhuis, F.R. Opperdoes, P.A.M. Fintracellular_ buffer increases cell survival, Nucleic Acids Res. 20 (1992) Michels, Cloning and analysis of the PTS-1 receptor in Trypanosoma 2902. brucei, Mol. Biochem. Parasitol. 104 (1999) 106–119. [33] R. McCulloch, E. Vassella, P. Burton, M. Boshart, J.D. Barry, Transfor- [14] A. Jardim, W. Liu, E. Zheleznova, B. Ullman, Peroxisomal targeting mation of monomorphic and pleomorphic Trypanosoma brucei, Methods signal-1 receptor protein PEX5 from Leishmania donovani. Molecular, Mol. Biol. 262 (2004) 53–86. biochemical, and immunocytochemical characterization, J. Biol. Chem. [34] M.M. Bradford, A rapid and sensitive method for the quantitation of 275 (2000) 13637–13644. microgram quantities of protein utilizing the principle of protein-dye [15] A. Kumar, C. Roach, I.S. Hirsh, S. Turley, S. de Walque, P.A.M. Michels, binding, Anal. Biochem. 72 (1976) 248–254. W.G.J. Hol, An unexpected extended conformation for the third TPR [35] U.K. Laemmli, Cleavage of structural proteins during assembly of the motif of the peroxin PEX5 from Trypanosoma brucei, J. Mol. Biol. 307 head of bacteriophage T4, Nature 227 (1970) 680–685. (2001) 271–282. [36] H. Towbin, T. Staehelin, J. Gordon, Electrophoretic transfer of proteins [16] A. Jardim, N. Rager, W. Liu, B. Ullman, Peroxisomal targeting protein 14 from polyacrylamide gels to nitrocellulose sheets: procedure and some (PEX14) from Leishmania donovani. Molecular, biochemical, and applications, Proc. Natl. Acad. Sci. U. S. A. 76 (1979) 4350–4354. immunocytochemical characterization, Mol. Biochem. Parasitol. 124 [37] O. Misset, O.J.M. Bos, F.R. Opperdoes, Glycolytic enzymes of Trypano- (2002) 51–62. soma brucei. Simultaneous purification, intraglycosomal concentrations [17] J. Moyersoen, J. Choe, A. Kumar, F.G.J. Voncken, W.G.J. Hol, P.A.M. and physical properties, Eur. J. Biochem. 15 (1986) 441–453. Michels, Characterization of Trypanosoma brucei PEX14 and its role in [38] N. Chevalier, M. Callens, P.A.M. Michels, High-level expression of the import of glycosomal matrix proteins, Eur. J. Biochem. 270 (2003) Trypanosoma brucei fructose-1,6-bisphosphate aldolase in Escherichia 2059–2067. coli and purification of the enzyme, Protein Expression Purif. 6 (1995) [18] J. Choe, J. Moyersoen, C. Roach, T.L. Carter, E. Fan, P.A.M. Michels, 39–44. W.G.J. Hol, Analysis of the sequence motifs responsible for the interactions [39] T.V. Borchert, K. Pratt, J.P. Zeelen, M. Callens, M.E. Noble, F.R. of peroxins 14 and 5, which are involved in glycosome biogenesis in Opperdoes, P.A.M. Michels, R.K. Wierenga, Overexpression of trypano- Trypanosoma brucei, Biochemistry 42 (2003) 10915–10922. somal triosephosphate isomerase in Escherichia coli and characterisation [19] K.P. Madrid, G. De Crescenzo, S. Wang, A. Jardim, Modulation of the of a dimer-interface mutant, Eur. J. Biochem. 211 (1993) 703–710. Leishmania donovani peroxin 5 quaternary structure by peroxisomal [40] V. Hannaert, F.R. Opperdoes, P.A.M. Michels, Glycosomal glyceral- targeting signal 1 ligands, Mol. Cell. Biol. 24 (2004) 7331–7344. dehyde-3-phosphate dehydrogenase of Trypanosoma brucei and Try- [20] K.P. Madrid, A. Jardim, The Leishmania donovani PEX5–PEX14 panosoma cruzi: expression in Escherichia coli, purification, and H. Krazy, P.A.M. Michels / Biochimica et Biophysica Acta 1763 (2006) 6–17 17

characterization of the enzymes, Protein Expression Purif. 6 (1995) recruits the Pex1p–Pex6p AAA ATPase complexes to peroxisomes, Nat. 244–250. Cell Biol. 5 (2003) 454–460. [41] D.T. Hart, P. Baudhuin, F.R. Opperdoes, C. de Duve, Biogenesis of the [50] S. Weller, I. Cajigas, J. Morrell, C. Obie, G. Steel, S.J. Gould, D. Valle, glycosome in Trypanosoma brucei: the synthesis, translocation and Alternative splicing suggests extended function of PEX26 in peroxisome turnover of glycosomal polypeptides, EMBO J. 6 (1987) 1403–1411. biogenesis, Am. J. Hum. Genet. 76 (2005) 987–1007. [42] A. Beyer, Sequence analysis of the AAA protein family, Protein Sci. 6 [51] C.C. Chang, D.S. Warren, K.A. Sacksteder, S.J. Gould, PEX12 interacts (1997) 2043–2058. with PEX5 and PEX10 and acts downstream of receptor docking in [43] P.I. Hanson, S.W. Whiteheart, AAA+ proteins: have engine, will work, peroxisomal matrix protein import, J. Cell Biol. 147 (1999) 761–774. Nat. Rev., Mol. Cell Biol. 6 (2005) 519–529. [52] K. Okumoto, I. Abe, Y. Fujiki, Molecular anatomy of the peroxin Pex12p: [44] S. Tamura, N. Shimozawa, Y. Suzuki, T. Tsukamoto, T. Osumi, Y. Fujiki, ring finger domain is essential for Pex12p function and interacts with the A cytoplasmic AAA family peroxin, Pex1p, interacts with Pex6p, peroxisome-targeting signal type 1-receptor Pex5p and a ring peroxin, Biochem. Biophys. Res. Commun. 245 (1998) 883–886. Pex10p, J. Biol. Chem. 275 (2000) 25700–25710. [45] K.N. Faber, J.A. Heyman, S. Subramani, Two AAA family peroxins, [53] J.H. Eckert, N. Johnsson, Pex10p links the ubiquitin conjugating enzyme PpPex1p and PpPex6p, interact with each other in an ATP-dependent Pex4p to the protein import machinery of the peroxisome, J. Cell Sci. 116 manner and are associated with different subcellular membranous (2003) 3623–3634. structures distinct from peroxisomes, Mol. Cell. Biol. 18 (1998) 936–943. [54] P.S. Freemont, I.M. Hanson, J. Trowsdale, A novel cysteine-rich sequence [46] I. Birschmann, K. Rosenkranz, R. Erdmann, W.H. Kunau, Structural and motif, Cell 64 (1991) 483–484. functional analysis of the interaction of the AAA-peroxins Pex1p and [55] Y. Chen, C.H. Hung, T. Burderer, G.S. Lee, Development of RNA Pex6p, FEBS J. 272 (2005) 47–58. interference revertants in Trypanosoma brucei cell lines generated with a [47] I. Birschmann, A.K. Stroobants, M. van den Berg, A. Schafer, K. double stranded RNA expression construct driven by two opposing Rosenkranz, W.H. Kunau, H.F. Tabak, Pex15p of Saccharomyces promoters, Mol. Biochem. Parasitol. 126 (2003) 275–279. cerevisiae provides a molecular basis for recruitment of the AAA [56] M. Durand-Dubief, L. Kohl, P. Bastin, Efficiency and specificity of RNA peroxin Pex6p to peroxisomal membranes, Mol. Biol. Cell 14 (2003) interference generated by intra- and intermolecular double stranded RNA 2226–2236. in Trypanosoma brucei, Mol. Biochem. Parasitol. 129 (2003) 11–21. [48] J.A.K.W. Kiel, R.E. Hilbrands, I.J. van der Klei, S.W. Rasmussen, F.A. [57] J. Kyte, R.F. Doolittle, A simple method for displaying the hydropathic Salomons, M. van der Heide, K.N. Faber, J.M. Cregg, M. Veenhuis, character of a protein, J. Mol. Biol. 157 (1982) 105–132. Hansenula polymorpha Pex1p and Pex6p are peroxisome-associated [58] C. Clayton, M. Adams, R. Almeida, T. Baltz, M. Barrett, P. Bastien, S. AAA proteins that functionally and physically interact, Yeast 15 (1999) Belli, S. Beverley, N. Biteau, J. Blackwell, et al., Genetic nomenclature 1059–1078. for Trypanosoma and Leishmania, Mol. Biochem. Parasitol. 97 (1998) [49] N. Matsumoto, S. Tamura, Y. Fujiki, The pathogenic peroxin Pex26p 221–224.