Peroxisome reintroduction in Hansenula polymorpha requires Pex25 and Rho1

For decades peroxisomes have been considered to be autonomous organelles that multiply by growth and division (Lazarow and Fujiki, 1985). Recent studies have however revealed that peroxisomes may also form from the endoplasmic reticulum (ER), a phenomenon that was convincingly demonstrated upon functional complementation of PEX3- or PEX16-deficient strains of various organisms (Hoepfner et al., 2005; Kragt et al., 2005; Tam et al., 2005; Haan et al., 2006; Kim et al., 2006). These cells are devoid of peroxisomal membrane structures, but form new organelles upon reintroduction of the corresponding deleted genes.

Peroxisomes may also multiply by fission and several proteins that are involved in this process have been identified (e.g., Pex11 and dynamin-related proteins; Thoms and Erdmann, 2005; Fagarasanu et al., 2007). Recent studies suggested that in yeast the bulk of the organelles are formed by fission (Kuravi et al., 2006; Motley and Hettema, 2007; Nagotu et al., 2008b). For instance, mutations that completely block peroxisome fission result in the presence of a single enlarged peroxisome per cell, also after prolonged cultivation at peroxisome-inducing cultivation conditions. In these mutants peroxisome formation from the ER is not affected (e.g., in dnm1, vps1, or dnm1 vps1 mutants), but generation of additional organelles was never observed (Kuravi et al., 2006; Motley and Hettema, 2007; Motley et al., 2008; Nagotu et al., 2008b). These observations suggest that at normal physiological conditions peroxisome formation from the ER may not prominently contribute to the total organelle population in yeast cells.

Several observations however indicate that the ER does play a role in peroxisome formation in wild-type cells and various peroxisomal membrane proteins (PMPs), if not all (van der Zand et al., 2010), are proposed to traffic to peroxisomes via the ER. Examples include Yarrowia lipolytica Pex2 and Pex16 (Titorenko and Rachubinski, 1998), Saccharomyces cerevisiae Pex3 (Hoepfner et al., 2005; Kragt et al., 2005), plant Pex16 (Karnik and Trelease, 2005, 2007), mammalian Pex16 (Kim et al., 2006), Pichia pastoris Pex30 and Pex31 (Yan et al., 2008), and S. cerevisiae Pex11 (Knoblach and Rachubinski, 2010). Generally, at steady-state conditions the bulk of these PMPs are localized to peroxisomes and difficult to detect at the ER. Exceptions are plant Pex16 (Karnik and Trelease, 2005) and P. pastoris Pex30 and Pex31 (Yan et al., 2008), which were shown to invariably have a dual localization at the ER and peroxisomes. Also, proteins of the endomembrane system have been implicated to serve a role in peroxisome biogenesis, such as Arf, coatomer (Lay et al., 2006), Sec20, and Sec39 (Perry et al., 2009). The molecular details of the role of these proteins in peroxisome biogenesis and proliferation need to be further elucidated.

Important players in peroxisome fission include dynamin-like proteins, such as Vps1 in S. cerevisiae (Hoepfner et al., 2001), Dnm1 in S. cerevisiae and Hansenula polymorpha (Kuravi et al., 2006; Nagotu et al., 2008b), and Dlp1 in mammals (Koch et al., 2003; Li and Gould, 2003). These GTPases are most likely involved in the actual organelle fission process. Another key protein in fission is the highly conserved peroxisomal membrane protein Pex11, which was recently shown to be responsible for tubulation of the peroxisomal membrane before fission (Opaliński et al., 2011).

All eukaryotes studied so far contain several proteins that show similarity to Pex11 (Kiel et al., 2006). For instance, S. cerevisiae contains Pex25 and Pex27 in addition to Pex11 (Smith et al., 2002; Rottensteiner et al., 2003; Tam et al., 2003). In the yeast H. polymorpha the additional members of the Pex11 protein family are Pex11C and Pex25.

Here we study the role of all three members of the H. polymorpha Pex11 protein family in peroxisome formation. We show that that Pex25 plays a crucial role in the formation of peroxisomes upon reintroduction of PEX3 in H. polymorpha pex3 cells. We also demonstrate that the pex11 pex25 double-deletion strain is peroxisome deficient. Most likely this is caused by the simultaneous block in fission and peroxisome reintroduction.

As shown in Fig. 1, A–C, all three members of the Pex11 protein family are localized to peroxisomes. The fluorescence signal observed for Pex11C-GFP is low relative to Pex11-GFP and Pex25-GFP, which is most likely due to relatively low expression of PEX11C as also is suggested by transcriptomics data (van Zutphen et al., 2010).

The role of the three Pex11 family proteins in peroxisome formation was analyzed in various deletion strains. Cells of single deletion strains (designated pex11, pex25, and pex11C) grew like wild-type (WT) controls on glucose. However, on methanol the doubling times of pex11 and pex25 cultures (t_d = 8 h) were twice that of identically grown WT or pex11C cultures (t_d = 4 h). Quantitative analyses of peroxisome numbers (Fig. 2; Table I) revealed that deletion of PEX11 resulted in a strong reduction in peroxisome numbers in methanol-induced cells, whereas in pex25 cells a slight increase was observed. Deletion of PEX11C had no effect on peroxisome numbers (Fig. 2; Table I). Deletion of PEX11C in pex11 or pex25 cells also had no effect on the phenotype of the initial single mutants (Figs. 1 and 2; Table I). These data confirm the role of Pex11 in peroxisome proliferation, whereas Pex25 has a slightly negative effect in this process.

Surprisingly, deletion of PEX25 in pex11 cells (strain pex11 pex25) resulted in the mislocalization of the peroxisomal matrix marker protein DsRed-SKL to the cytosol (Fig. 1 D; Fig. 2; Table I).

To study the apparent peroxisome-deficient phenotype of pex11 pex25 cells in more detail, electron microscopy was performed. These studies failed to resolve any peroxisomal membrane remnants in pex11 pex25 cells (Fig. 3 B), akin to H. polymorpha pex3 or pex19 cells, at conditions that WT cells contained multiple peroxisomes (Fig. 3 A).

Western blot analysis indicated that the levels of the two PMPs Pex3 and Pex14 were only slightly reduced in pex11 pex25 cells, whereas they were normal in the single-deletion strains (Fig. 1 G). Localization studies by fluorescence microscopy revealed that Pex3 was mislocalized to the cytosol (Fig. 3 C, Pex3-GFP). This observation is consistent with the view that peroxisomal membrane remnants are absent in pex11 pex25 cells.

The observation that pex11 pex25 cells were peroxisome deficient raised the question of whether in these cells both the processes of peroxisome fission (due to the absence of Pex11) and reintroduction (due to the absence of Pex25) were blocked. This hypothesis was first addressed by analyzing if, in a pex3 pex25 double-deletion strain, peroxisome formation is restored after reintroduction of PEX3-GFP, similar as in H. polymorpha pex3 cells (Haan et al., 2006; Nagotu et al., 2008b).

To this end we constructed a pex3 pex25 strain that contained PEX3-GFP under control of the inducible amine oxidase promoter (P_AMO) using a pex3 strain as a control. After precultivation of the strains on glucose/ammonium sulfate, thus repressing P_AMO, Pex3-GFP protein was invariably undetectable (not depicted). Upon a shift of pex3 pex25 P_AMOPEX3 cells to fresh glycerol/methanol/methylamine-containing media, Pex3-GFP fluorescence was generally first detected after 2 h of cultivation (Fig. 4 D) and frequently observed as a single spot per cell that did not develop into a peroxisome upon further cultivation. Even after 20 h of cultivation on glycerol/methanol/methylamine media peroxisomes were absent and Pex3-GFP was still infrequently observed in spots or had accumulated in the vacuole. Subsequent electron microscopy analyses also failed to resolve any peroxisome structures in these cells at any time of cultivation (not depicted). Under the same conditions peroxisomes were readily formed in pex3 controls (Fig. 4 A). In identical experiments, using pex3 pex11C or pex3 pex11 cells, peroxisomes were reintroduced like in the pex3 control (Fig. 4, B and C), with the exception that in pex3 pex11 cells relatively low numbers of enlarged peroxisomes were formed, as expected for H. polymorpha pex11 cells (Fig. 4 F).

Western blot analysis (Fig. 4 E) revealed that until 8 h after induction of PEX3-GFP expression the Pex3-GFP levels were in the same range as endogenous Pex3 levels observed in identically grown WT cells, indicating that the cells did not experience Pex3 overproduction when peroxisomes were reintroduced. At 20 h after induction, Pex3-GFP levels were enhanced in all strains relative to Pex3 in the WT control, except for strain pex3 pex25 where Pex3-GFP levels were below the level of detection. This is in line with the reduction of GFP fluorescence in this strain at late time points.

Because Pex25 is required for reintroduction of peroxisomes in pex3 cells, we hypothesized that Pex3 may not properly sort to the ER in the absence of Pex25 to form new peroxisomes. To address this question, we constructed a pex3 pex25 strain, in which Pex3-mCherry was artificially sorted to the ER. To this end we constructed a gene encoding a fusion protein containing the first N-terminal 30 amino acids of the ER protein BIP (BiP_N30) and full-length Pex3 (lacking the start codon) fused to mCherry under control of the inducible alcohol oxidase promoter (P_AOXBIP_N30PEX3-mCherry). A similar construct was previously reported to functionally complement S. cerevisiae pex3 cells (Kragt et al., 2005). Indeed, upon synthesis of this fusion protein in H. polymorpha pex3 control cells peroxisomes were readily formed (Fig. 5 A). Essentially similar results were obtained when the fusion protein was introduced in pex3 pex11 cells (Fig. 5 B). In contrast, however, peroxisomes were not detected when the construct was expressed in pex3 pex25 or pex3 pex11 pex25 cells (Fig. 5, C and D). In these cells large cytosolic alcohol oxidase crystals were formed (Fig. 5, C and D, asterisk), akin to pex3 cells, demonstrating that these cells are indeed peroxisome deficient. These data suggest that the failure of pex3 pex25 cell to form peroxisomes from the ER cannot be restored by artificial targeting of Pex3 to the ER.

We next analyzed whether peroxisomes were formed in pex11 pex25 cells upon reintroduction of either PEX25-mCherry (Fig. 6, A–C) or PEX11-mCherry (Fig. 6, D–F). Separate strains were constructed in which either the ER marker protein BiP_N30-GFP-HDEL (Fig. 6, A and D) or Pex3-GFP (Fig. 6, B and E) or GFP-SKL (Fig. 6, C and F) were produced. Upon a shift of pex11 pex25 P_AOX PEX25-mCherry cells from P_AOX-repressing to P_AOX-inducing conditions (shift from glucose/ammonium sulfate to methanol/glycerol/ammonium sulfate), peroxisomes were readily formed (Fig. 6 C). After 4 h of incubation in the presence of methanol, Pex25 was initially localized in a spot at the ER (Fig. 6 A). Pex3 colocalized with Pex25 at this spot (Fig. 6 B). At later time points (16 h after induction) the cells grew on methanol like pex11 cells and contained peroxisomes harboring the peroxisomal matrix marker protein GFP-SKL (Fig. 6 C).

In pex11 pex25 P_AOX PEX11-mCherry cells peroxisomes were never formed (Fig. 6 F) and growth on methanol was not restored. In these cells Pex11 was present at the ER and the nuclear envelope (Fig. 6 D). Pex3 was also present at these membranes, but concentrated in a spot that also showed a slightly enhanced Pex11-mCherry fluorescence intensity (Fig. 6 E, arrow). GFP-SKL remained mislocalized to the cytosol in these cells, also upon prolonged induction in the presence of methanol (Fig. 6 F, 16 h of induction). These data indicate that the absence of Pex25 apparently also prevents reintroduction of peroxisomes in pex11 pex25 cells.

As Pex3 is initially sorted to the ER upon reintroduction in pex3 cells, we analyzed the localization of Pex25 in pex3 cells. As shown in Fig. 7 B, synthesis of Pex25-mCherry in pex3 cells results in localization of the protein in structures adjacent to the ER, marked by BiP_N30-GFP-HDEL. Pex11-mCherry colocalized with the ER marker in pex3 pex11 cells (Fig. 7 A), but was never observed in spots.

In pex3 cells Pex25 is localized in spots adjacent to the ER. As reintroduction of peroxisomes in these cells requires Pex25 and involves the initial targeting of Pex3 to the ER, we tested whether Pex3 interacts with Pex25 using a yeast two-hybrid assay. As shown in Fig. 8, no interaction between Pex25 and Pex3 was detected. Also, no interaction between Pex11 and Pex3 was observed. As reported previously for Pex11 proteins from other species, H. polymorpha Pex11 interacts with itself and hence most likely forms oligomers (Rottensteiner et al., 2003; Tam et al., 2003). The same was observed for H. polymorpha Pex25. We could not detect an interaction between H. polymorpha Pex11 and Pex25. These data indicate that Pex25 and Pex3, which are both involved in reintroduction of peroxisomes at the ER, most likely do not show a direct physical interaction.

In a series of experiments aimed at the identification of essential genes involved in reintroduction of peroxisomes in H. polymorpha pex3 cells, we identified a temperature-sensitive mutation in RHO1. We previously showed that synthesis of a protein consisting of the first 50 residues of Pex3 fused to GFP (N₅₀.Pex3-GFP; Faber et al., 2002a) is sorted to the ER/nuclear envelope in H. polymorpha pex3 cells and leads to the formation of membrane vesicles that develop into normal peroxisomes upon synthesis of full-length Pex3 (Faber et al., 2002a). We reasoned that mutants defective in sorting of N₅₀.Pex3-GFP to the ER or the formation of the N₅₀.Pex3-GFP–containing vesicles would also be defective in peroxisome formation from the ER.

Cells producing N₅₀.Pex3-GFP under control of the amine oxidase promoter were mutagenized with nitrosoguanidine. Subsequently, mutants were selected that showed a temperature-sensitive (ts) growth phenotype on rich glucose media (YPD). In total, 65 mutants were isolated and subsequently analyzed by fluorescence microscopy on mineral media containing glucose/methylamine. Of these strains, 21 showed mislocalization of N₅₀.Pex3-GFP to the cytosol after a shift of the cells to the restrictive temperature (44°C), but not at the permissive temperature (35°C).

One of these mutants, 3-34-ts, was used for further analysis. This strain showed GFP fluorescence in the cytosol at restrictive temperature (Fig. 9 D), whereas at permissive temperature (35°C) fluorescent spots were observed (Fig. 9 C). As expected, pex3 control cells producing N₅₀.Pex3-GFP formed normal fluorescent spots at both temperatures (Fig. 9, A and B).

By functional complementation of mutant 3-34-ts with a gene library, we identified the RHO1 gene. Indeed, when grown at restrictive temperatures the complemented cells contained normal fluorescent spots as in control cells (not depicted). These data suggest that Rho1 is involved in the reintroduction of peroxisomes in pex3 cells. Alignment of the sequences of WT and mutant Rho1 protein in strain 3-34-ts revealed an amino acid substitution at amino acid position 164 (Ala into Val). Alanine 164 is located in the highly conserved SAK motif of Rho1, which participates in interactions with the guanine of GTP in the active site.

Marelli et al. (2004) previously showed that Rho1 controls peroxisome membrane dynamics and biogenesis in S. cerevisiae. In addition, these authors showed that Rho1 binds several peroxins in vitro, including Pex25. This led us to investigate whether Rho1 is also involved in Pex25-induced peroxisome reintroduction in pex11 pex25 cells.

We first analyzed the localization of Rho1 in H. polymorpha WT cells producing GFP-Rho1 and DsRed-SKL to mark peroxisomes. The data presented in Fig. 9 E convincingly show that GFP-Rho1 colocalizes with peroxisomes as well as with the vacuole and plasmamembrane, like in S. cerevisiae (Marelli et al., 2004).

We subsequently introduced the temperature-sensitive Rho1 mutation in the pex11 pex25 strain containing PEX25-mCherry under control of P_AOX (strain pex11 pex25 P_RHO1 RHO1^ts P_AOXPEX25 mCherry). Cells were pregrown on glucose at the permissive temperature (35°C) to the late exponential growth phase and subsequently shifted to fresh glycerol/methanol-containing media and grown at the permissive or restrictive conditions. When grown at 35°C, peroxisomes were readily formed and marked by Pex25 mCherry (Fig. 9 H). The organelles were also readily detectable by electron microscopy (Fig. 10 A). However, at 44°C Pex25-mCherry initially (3–5 h of cultivation) was observed as a distinct spot (Fig. 9 I), most likely located at the ER/nuclear membrane, which disappeared again after further cultivation (not depicted). Synthesis of Pex25-mCherry in pex11 pex25 cells without the temperature-sensitive mutation in RHO1 resulted in peroxisome formation both at 35°C (Fig. 9 F) and 44°C (Fig. 9 G).

Careful electron microscopical analysis of these cells failed to resolve peroxisomal structures at any time of cultivation. These data suggest that Rho1 is involved in peroxisome reintroduction in H. polymorpha pex11 pex25 cells (Fig. 10 B). Instead, these cells contained various tubular-shaped structures. Examples of longitudinal and cross sections through these structures are shown in Fig. 10, B–D. Possibly, these structures represent peroxisomal prestructures which were unable to develop into normal organelles due to the absence of functional Rho1 protein. Using yeast two-hybrid analysis we could not detect interaction of H. polymorpha Rho1 with Pex25, Pex11, or Pex3 (Fig. 8).

We have analyzed the function of the H. polymorpha Pex11 protein family in peroxisome formation. These studies identified Pex25 as the first protein specifically involved in the reintroduction of peroxisomes in cells lacking preexisting ones.

Our data indicate that Pex25 as well as Rho1 act in the process of reintroduction of peroxisomes in cells lacking preexisting peroxisomes.

Remarkably, H. polymorpha pex25 cells are not peroxisome deficient. We anticipate that this is related to the fact that peroxisome fission is not blocked in these cells. We therefore propose that peroxisomes in pex25 cells are most likely formed by fission of preexisting ones. Because pex25 cells show a twofold increase in doubling time during growth on methanol, the defect in the process of peroxisome formation from the ER may affect optimal peroxisome biogenesis and/or function, thereby reducing growth on methanol.

The most surprising finding of our studies was that deletion of both PEX11 and PEX25 results in peroxisome deficiency. A simple explanation would be that both processes of peroxisome proliferation (fission, which requires Pex11, and peroxisome formation from the ER, which requires Pex25) are blocked in the double-deletion strain. As a consequence, peroxisomes are absent. Indeed, the phenotype of pex11 pex25 cells is reminiscent of H. polymorpha pex19 cells (Otzen et al., 2004), in which Pex3 is also cytosolic.

In pex11 pex25 cells, Pex3 is not targeted to the ER and peroxisomes are not formed. Evidently, Pex25 is not required for targeting of Pex3 to the ER, as artificial targeting of Pex3 to the ER in pex3 pex25 cells did not restore peroxisome formation. What the molecular function of Pex25 is during peroxisome formation from the ER remains speculative. Most likely it acts at the ER, as it is localized in spots adjacent to the ER in pex3 cells. Different from Pex11, Pex25 is not evenly distributed over the ER, but present in spots, which might represent the sites of peroxisome reintroduction.

Interestingly, previously Marelli et al. (2004) showed that in S. cerevisiae Pex25 recruits the small GTPase Rho1 to peroxisomes. This led us to speculate that in addition to Pex25, Rho1 also has a function in peroxisome reintroduction in H. polymorpha pex11 pex25 cells. We show that, like in S. cerevisiae, Rho1 colocalizes with peroxisomes in H. polymorpha. Moreover, our data indicate that functional Rho1 is required for peroxisome formation in pex11 pex25 cells during reintroduction of PEX25. The function of the protein in this reintroduction process is still speculative. At the restrictive temperature pex11 pex25 RHO^ts cells producing Pex25 contained various tubular structures instead of normal peroxisomes, which may suggest a failure in membrane fusion processes that are required for peroxisome formation. A similar function for Rho1 has been reported for vacuole membrane fusion (Logan et al., 2010) and membrane fusion at the plasmamembrane during exocytosis (Yamashita et al., 2010).

The phenotype of the temperature-sensitive mutant 3-34-ts at the restrictive temperature suggests that in the absence of functional Rho1, Pex3 is not properly sorted to the ER. Similarly, Pex3 is cytosolic in pex11 pex25 cells, but sorts to the same location as Pex25 upon reintroduction of PEX25. Likely, Pex3, Pex25, and Rho1 are all required to form a functional preperoxisomal vesicle, a process that is disturbed when one of the three components is missing or defective.

Our two-hybrid studies did not reveal direct physical interactions between Pex3, Pex25, and Rho1. Previously, Marelli et al. (2004) showed that Escherichia coli produced GST-Rho1, immobilized on glutathione resin, is capable to bind TAP-tagged S. cerevisiae Pex25 during incubation with a yeast extract containing this tagged protein. In this experiment, 5 out of 20 tested peroxins bound to Rho1, of which Pex25 and Pex30 showed the strongest interaction. The fact that we do not observe an interaction between H. polymorpha Pex25 and Rho1 is possibly related to either the lower sensitivity of the two-hybrid assay or to the fact that the interaction may not be direct and requires additional H. polymorpha proteins, which are absent in the S. cerevisiae–based two-hybrid assay.

At present, Pex3 is generally considered to represent the key player in peroxisome formation from the ER (Smith and Aitchison, 2009). Our current data confirm this crucial role of Pex3, but show that Pex25 and Rho1 are also essential for reintroduction of peroxisomes in H. polymorpha cells lacking peroxisomal membrane structures.

The common view on the function of Pex11 in peroxisome proliferation is that it is involved in elongation of preexisting organelles before the actual fission process. Our recent data on H. polymorpha Pex11 are in line with this view (Opaliński et al., 2011). We previously showed that in this organism peroxisome fission is the major pathway of peroxisome proliferation (Nagotu et al., 2008b). Hence, in yeast deletion of a gene in peroxisome fission does result in a major reduction in peroxisome numbers. As a consequence, in H. polymorpha pex11 cells peroxisomes are most likely invariably formed from the ER. This is in line with our observation that, in the absence of Pex11, peroxisomes can be formed from the ER upon reintroduction of Pex3 in pex3 pex11 cells.

Detailed fluorescence microscopy including live-cell imaging (unpublished data) never revealed detectable amounts of Pex11 protein at the ER in WT cells, as recently was reported for Pex11 in S. cerevisiae (Knoblach and Rachubinski, 2010). However, we detected H. polymorpha Pex11 at the ER in pex3 cells and in pex11 pex25 lacking peroxisomal structures.

Yeast strains used in this study are listed in Table II. Yeast cultures were grown at 37°C on (a) YPD media containing 1% yeast extract, 1% peptone, and 1% glucose; (b) selective media containing 0.67% yeast nitrogen base without amino acids, supplemented with 1% glucose (YND) or 0.5% methanol (YNM); or (c) mineral media (MM; van Dijken et al., 1976) supplemented with 0.5% glucose, or 0.5% methanol as carbon source and 0.25% ammonium sulfate or 0.25% methylamine as nitrogen source. In the case of peroxisome-deficient cells, 0.1% glycerol was added to the methanol-containing media. If required, amino acids, uracil, or leucine was added to a final concentration of 30 µg/ml. For growth on agar plates the medium was supplemented with 2% agar. For the selection of resistant transformants, YPD plates containing 100 µg/ml zeocin or 100 µg/ml nourseothricin (Invitrogen) were used.

For cloning purposes, E. coli DH5α was used. Cells were grown at 37°C in LB media supplemented with 100 µg/ml ampicillin or 50 µg/ml kanamycin when required. Cells were grown in shake flask cultures as described previously (Nagotu et al., 2008b).

The plasmids and primers used in this study are listed in Tables III and IV. All integrations and deletions were confirmed by Southern blotting.

A pex25 deletion strain was constructed by replacing the genomic region of PEX25 comprising nucleotides +198 to +784 by the antibiotic marker nourseothricin (NAT). To this end, pRSA018 was made using Gateway Technology (Invitrogen). Two DNA fragments comprising the regions −674 to +197 and +785 to +2109 of the PEX25 genomic region were obtained by PCR using primers RSAPex25-1/RSAPex25-2 and RSAPex25-3/RSAPex25-4, respectively, and H. polymorpha genomic DNA as a template. The PCR fragments were cloned into the vectors pDONR-P4-1R and pDONR-P2R-P3, respectively, resulting in the entry vectors pENTR-PEX25 5′ and pENTR-PEX25 3′. From plasmid pAG25, the NAT fragment of pAG25 NcoI (partial digestion)–EcoRV was cloned into pHIPZ4 (Asp718I (klenow fill-in)-NcoI), resulting in pHIPN4 (NAT). PCR amplification was performed using primers attB1-Ptef1-forward and attB2-Ttef1-reverse using pHIPN4 as the template. The resulting PCR fragment was recombined into vector pDONR-221 yielding entry vector pENTR-221-NAT.

Recombination of the entry vectors pENTR-PEX25 5′, pENTR-221-NAT, and pENTR-PEX25 3′, and the destination vector pDEST-R4-R3, resulted in pRSA018. Subsequently, H. polymorpha WT leu1.1 ura3 cells were transformed with the 2912-bp PEX25 deletion fragment, which was obtained by PCR using primers RSAPex25-5 and RSAPex25-6 and pRSA018 as a template. The resulting strain was designated pex25.

A pex11C deletion strain was constructed by replacing the genomic region of PEX11C comprising nucleotides +186 to +595 by the antibiotic marker hygromycin B, HPH. Two DNA fragments comprising the regions −678 to +185 and +596 to +1454 of the PEX11C genomic region were obtained by PCR using primers RSAPex11C-1/RSAPex11C-2 and RSAPex11C-3/RSAPex11C-4, respectively, and H. polymorpha genomic DNA as a template. The PCR fragments were cloned into the vectors pDONR-P4-1R and pDONR-P2R-P3, respectively, resulting in the entry vectors pENTR-PEX11C 5′ and pENTR-PEX11C 3′. From plasmid pAG32 the HPH fragment of pAG32 NcoI (partial digestion)–EcoRV was cloned into pHIPZ4 (Asp718I(klenow fill-in)-NcoI), resulting in pHIPH4 (HPH, hygromycin B). PCR amplification was done with primers att B1-Ptef1-forward and attB2-Ttef1-reverse using pHIPH4 as the template. The resulting PCR fragment was recombined into vector pDONR-221 yielding entry vector pENTR-221-HPH. Recombination of the entry vectors pENTR-PEX11C 5′, pENTR-221-HPH, and pENTR-PEX11C 3′, and the destination vector pDEST-R4-R3, resulted in pRSA019. Subsequently, H. polymorpha WT leu1.1 ura3 cells were transformed with the 3560-bp PEX11C deletion fragment, which was obtained by PCR using primers RSAPex11C-5 and RSAPex11C-6 and pRSA019 as a template. The resulting strain was designated pex11C.

To enable Pex11C localization in H. polymorpha WT cells, an in-frame fusion was constructed of the C terminus of the PEX11C gene with the GFP gene, under the control of its homologous PEX11C promoter. The PEX11C gene was amplified using primers RSAPex11Cfusfw and RSAPex11Cfusrev, resulting in a product lacking the stop codon. This PCR product was then digested with BglII and HindIII and ligated in pSNA10, resulting in plasmid pAMK24. Plasmid pAMK24 was linearized with BstBI and integrated into H. polymorpha WT cells.

Integration of pHIPZ4-DsRed-T1-SKL into the P_AOX region of the H. polymorpha genome was achieved by transforming SphI-linearized plasmid DNA. Random integration of pHIPX4-GFP-SKL into the H. polymorpha genome was obtained by transforming NotI-linearized plasmid DNA.

The pex11 pex25 and pex3 pex11 mutants were obtained by crossing the pex11 and pex25 or pex3 and pex11 single mutants (Sudbery et al., 1988). Diploids were subjected to random spore analysis, and prototrophic segregants were subjected to complementation analysis to determine their genotypes. The pex11 pex11C, pex25 pex11C, and pex3 pex11C double mutants were obtained by making a knockout of PEX11C in pex11, pex25, and pex3 (leu-), respectively. Strain pex3 pex25 was made by making a deletion of PEX25 in pex3 (leu-). The triple mutant pex11 pex25 pex11C was made by a deletion of PEX11C in pex11 pex25. The triple mutant pex3 pex11 pex25 was made by crossing pex3 with pex11 pex25.

The pex3 pex11, pex3 pex25, and pex3 pex11C strains producing Pex3-GFP were obtained by transforming pHIPZ5-PEX3-GFP in these strains. The pex3 pex11 strain producing Pex11-mCherry was obtained by transforming pRSA03 and pRSA017 in pex3 pex11.

The pex11 pex25 double mutant producing Pex3-GFP under the endogenous promoter was obtained by transforming linearized pHOR46 in pex11 pex25. The pex11 pex25 strain producing Pex11mCherry or Pex25mCherry was made by transforming pRSA022 or pRSA08 in pex11 pex25.

For the construction of pENTR-P4-P1R-P_AOXBiP_N30, a PCR fragment of 1664 bp was obtained by primers RSAatt P_AOX F and RSAattBIPrev on pRSA017. The PCR fragment was cloned into the vector pDONR-P4-P1R, resulting in the entry vector pENTR-P4-P1R-P_AOXBiP_N30. For the construction of entry vector pENTR-221-PEX3-ATG, PCR amplification was performed with primers RSAattB1Pex3fwA and RSAattB2Pex3rev on H. polymorpha genomic DNA. The PCR fragment 1429 was cloned in entry vector pDONR-221 resulting in the entry vector pENTR-221-PEX3-ATG. pEXP-BiP_N30-PEX3-mCherry was obtained by recombination of the entry vectors pENTR-P4-P1R-P_AOXBiP_N30, pENTR-221-PEX3-ATG, and pRSA02, and the destination vector pDEST-R4-R3-NAT. The SphI-linearized plasmid was transformed in pex3, pex3 pex11, pex3 pex25, or pex3 pex11 pex25.

For the construction of plasmid pR6-5, a PCR fragment of 1119 bp was obtained using primers EMK11 and EMK12 and genomic DNA of a H. polymorpha mutant strain containing a temperature-sensitive mutation in the RHO1 gene. The resulting NaeI–SacII fragment was inserted between the Eco47III and SacII of pBSK URA3. For the construction of plasmid pHIPH-Rho1, a PCR fragment of 534 bp was obtained by primers RSARho1Not1fw and RSARho1HindIIIrev on pR6-5. The resulting HindIII–NotI fragment was inserted between the HindIII and NotI of pHIPH4. pHIPH-Rho1 was linearized by NarI and integrated into the genome of pex11 pex25 P_AOX PEX25-mcherry. Correct integration was confirmed by PCR and the fragment was sequenced to check for the mutation in the RHO1.

For the construction of entry vector pENTR-221-RHO1, PCR amplification was performed with primers GFP-Rho1fw and GFP-Rho1rev on H. polymorpha genomic DNA. The PCR fragment of 652 bp was cloned in entry vector pDONR-221 resulting in the entry vector pENTR-221-RHO1. pEXP-GFPRho1 was obtained by recombination of the entry vectors pENTR/41-PAMO-GFP, pENTR-221-RHO1, pENTR/23-TAMO, and the destination vector pRSA07. The DraIII-linearized plasmid was transformed to H. polymorpha WT. To mark the peroxisomes, pSNA03 linearized by KpnI was transformed into the same strain.

For the construction of plasmid pRSA01, a PCR fragment of 700 bp was obtained by primers RSA10fw and RSA11rev on pCDNA3.1mCherry. The resulting BglII–SalI fragment was inserted between the BglII and SalI of pANL31. For construction of plasmid pRSA02, PCR was done with primers RSA12Fw and RSA13Rev on pRSA01. The PCR fragment was cloned into the vector pDONR-P2R-P3, resulting in the entry vector pRSA02. For the construction of entry vector pENTR-P4-P1R-P_AOX, PCR amplification was done with primers att P_AOX F and att P_AOX R on pANL29. The PCR fragment was cloned in entry vector pDONR-P4-P1R resulting in the entry vector pENTR-P4-P1R-P_AOX. pRSA03 was obtained by recombination of the entry vectors pENTR-P4-P1R-P_AOX, pENTR-221-PEX11, and pRSA02, and the destination vector pDEST-R4-R3-NAT. For stable integration of the plasmid pRSA03 into the H. polymorpha genome, the plasmid was linearized with SacII in the P_AOX region and transformed to pex11, pex3, pex11 strains.

The PEX25 coding sequence lacking a stop codon was amplified using the primers BB-JK-037 and BB-JK-038 and cloned into the vector pDONR-221 resulting in plasmid pENTR-221-PEX25. Plasmid pRSA06 was obtained by recombination of the entry vectors pENTR-P4-P1R-P_AOX, pENTR-221-PEX25, and pRSA02, and the destination vector pDEST-R4-R3-NAT. For stable integration of the plasmid into the H. polymorpha genome, the plasmid was linearized with SacII in the P_AOX region and transformed to H. polymorpha WT.

For the construction of pRSA07, a 519 bp BamHI–NcoI fragment from pREMI-Z was inserted between the BamHI and NcoI of pHIPZ4-Nia to get plasmid pDEST-Zeo-tussen. The 1143 bp HindIII–Asp718I fragment (blunted) from pDEST-Zeo-tussen was ligated with pDEST-R4-R3 (digested with SfoI) to obtain pRSA07. pRSA08 was obtained by recombination of pENTR-P4-P1R-P_AOX, pENTR-221-PEX25, and pRSA02 and destination vector pRSA07. pRSA022 was obtained by recombination of pENTR-221-PEX11 pENTR-P4-P1R-P_AOX, pRSA02, and destination vector pRSA07.

For the construction of plasmid pBS-BiP, a PCR fragment of 100 bp was obtained by primers KN18 and KN19 on genomic DNA, and the resulting BamHI–HindIII fragment was inserted between the BamHI and HindIII of pBlueScript II. To obtain plasmid pBS-BiP_N30-GFP-HDEL, PCR fragment of 700 bp was obtained by primers KN14 and KN17 on pANL29, and the resulting SalI–BglII fragment was inserted between the SalI and BglII of pBS-BiP. Subsequently, pBS-BiP_N30-GFP-HDEL was digested with BamHI–SalI and ligated with BamHI–SalI-digested pHIPX7 to obtain pHIPX7-BiP_N30-GFP-HDEL. BamHI–EcoRI fragment (sticky ends filled in) of pHIPX7-BiP_N30-GFP-HDEL was ligated with HindIII–EcoRI fragment (sticky ends filled in) of pHIPX4 to obtain pHIPX4-BiP_N30-GFP-HDEL. To obtain plasmid pRSA017, NotI–SalI fragment of pHIPX4-BiP_N30-GFP-HDEL was ligated with NotI–SalI fragment of pHIPZ4-DsRed-T1-SKL. For random integration of the plasmids pHIPX4-BiP_N30-GFP-HDEL and pRSA017 into the H. polymorpha genome, the plasmid was linearized with KpnI and transformed to various strains.

The LexA system was used for screening interactions between H. polymorpha proteins using derivatives of the reporter strain S. cerevisiae L-40 (MATa leu2 his3 trp1 ade2 GAL4 gal80 LYS2::(lexAop)₄-HIS3 URA3::(lexAop)_s-lacZ; Takara Bio Inc.).

For PEX11, PEX25, and RHO1, a 799-bp DNA fragment comprising the entire PEX11 coding sequence, a 1297-bp DNA comprising the entire PEX25 coding sequence, and a 611-bp fragment were amplified with primer combinations RSAPex11fwbamhi + RSAPex11revsali, RSAPex25fwbamhi + RSAPex25revsali, and RSARho1fwbamhi + RSARho1 revsali, respectively, using genomic H. polymorpha WT DNA as template. PCR fragments were digested with BamHI and SalI and inserted between the BamHI and SalI sites of the vectors pBTM116-C and pVP16-C, respectively. This yielded plasmids pBTM116-PEX11, pVP16-PEX11, pBTM116-PEX25, pVP16-PEX25, pBTM116-RHO1, and pVP16-RHO1, respectively.

S. cerevisiae L-40 was cotransformed with the indicated pVP16- and pBTM116-derived fusion constructs. Subsequently, β-galactosidase filter lift assays were performed according to the manufacturer’s instructions (Takara Bio Inc.). From each cotransformation three independent transformants were tested. Empty vectors were used to check for reporter self-activation. The well-established HpPex3–HpPex19 interaction was used as a positive control.

Standard recombinant DNA techniques and transformation of H. polymorpha was performed as detailed previously (Faber et al., 1994). Crude cell extracts of TCA-precipitated cells were prepared as detailed previously (Baerends et al., 2000). SDS-PAGE and Western blotting were performed by established methods. Western blots were probed with polyclonal antibodies raised in rabbit against various H. polymorpha proteins.

All images were made at room temperature using a 100x 1.30 NA Plan Neofluar objective (Carl Zeiss). Images were captured in the media in which the cells were grown.

Wide-field images were captured using a fluorescence microscope (Axioskop50; Carl Zeiss) using MetaVue software and a digital camera (model 1300Y; Princeton Instruments). GFP signal was visualized with a 470⁄40-nm bandpass excitation filter, a 495-nm dichromatic mirror, and a 525⁄50-nm bandpass emission filter. DsRed fluorescence was visualized with a 546⁄12-nm bandpass excitation filter, a 560-nm dichromatic mirror, and a 575–640-nm bandpass emission filter. mCherry fluorescence was visualized with a 587/25-nm bandpass excitation filter, a 605-nm dichromatic mirror, and a 647/70-nm bandpass emission filter. Confocal images were captured using a confocal microscope (LSM510; Carl Zeiss), using photomultiplier tubes (Hamamatsu Photonics); images were acquired using AIM 4.2 software (Carl Zeiss). GFP fluorescence was analyzed by excitation of the cells with a 488-nm argon ion laser (Lasos), and emission was detected using a 500–550-nm bandpass emission filter. The DsRed signal was visualized by excitation with a 543-nm helium neon laser (Lasos) and emission was detected using a 565–615-nm bandpass emission filter.

Intact cells were collected by centrifugation and subsequently washed with distilled water to remove excess cultivation media before fixation in 1.5% (wt/vol) KMnO₄ for 20 min at room temperature. Poststaining was in 1% uranyl acetate (wt/vol) overnight. After dehydration in a graded ethanol series the samples were embedded in Epon 812. Ultrathin sections were cut with a diamond knife and examined in an electron microscope (model CM-12; Philips). Image analysis was performed using ImageJ (http://rsbweb.nih.gov/ij) and figures were prepared using Adobe Photoshop CS3.

We thank Rinse de Boer for skillful assistance during electron microscopy studies. We thank Elena Kurbatova for isolation of the temperature-sensitive Rho1 mutant.

This project was carried out within the research program of the Kluyver Centre for Genomics of Industrial Fermentation, which is part of the Netherlands Genomics Initiative/Netherlands Organization for Scientific Research.