Catabolite Repression of Aox in Pichia pastoris Is Dependent on Hexose Transporter PpHxt1 and Pexophagy

Media, strains, and microbial techniques.The strains listed in Table 1 were used in this work. The yeasts were grown with shaking at 30°C in 1% yeast extract and 2% peptone (YP) or in a synthetic yeast nitrogen base (YNB) medium (0.67% Difco yeast nitrogen base; for growth of auxotrophic strains, requisite amino acids were added to a final concentration of 50 μg/ml). As a carbon source, 1% glucose, 1% fructose, 1% glycerol, 0.5% methanol or 2% maltose, a mixture of 1% glucose plus 0.5% methanol, or 1% fructose plus 0.5% methanol were added throughout the study, unless indicated otherwise. Escherichia coli strain Top10 (Invitrogen) was used as a host for the construction of plasmids and was grown at 37°C in LB medium (0.5% yeast extract, 1% peptone, 0.5% NaCl). For solid medium, 2% powdered agar was added.

P. pastoris and Saccharomyces cerevisiae were transformed by the electroporation method (4) and the lithium acetate method (16), respectively. Transformation and other standard recombinant DNA operations used in this study for E. coli were performed as described previously (32). Ampicillin or kanamycin was added to LB medium at a final concentration of 50 μg/ml. G418, zeocin, and hygromycin were added to YPD medium at final concentrations of 0.3, 0.1, and 0.75 mg/ml, respectively. The respiration inhibitor antimycin A was added to glucose or fructose medium at a final concentration of 3 μg/ml.

Plasmid and strain construction.Plasmids were generated by standard techniques (32), and the primers and plasmids used for this study are listed in Tables 2 and 3, respectively.

(i) Δhxt1 strain.A P. pastoris strain with the PpHXT1 gene deleted was constructed by the gene replacement method using the G418 resistance gene KAN as a marker. First, the upstream region of the PpHXT1 gene was amplified by PCR using genomic DNA as a template with Pyrobest DNA polymerase (TaKaRa, Japan). The primers for this PCR, HX1U1 and HX1U2, carried restriction sites for EcoRI and BamHI, respectively. The 0.2-kb PCR-amplified fragment was inserted into EcoRI/BamHI-digested plasmid pUC18 (Invitrogen), yielding pHXT1Up. The downstream region of the PpHXT1 gene was amplified with primers HX1D1 and HX1D2, carrying restriction sites for SalI and SphI, respectively. The 0.2-kb DNA fragment was inserted into SalI/SphI-digested pUC18, yielding pHXT1Down. Afterwards, the G418 resistance gene sequence with its own promoter and terminator (1,517 bp) was amplified by PCR using the plasmid pPIC3.5K (Invitrogen) as a template with primers KAN1 and KAN2, which included restriction sites for BamHI and SalI, respectively, and the fragment was cloned into BamHI/SalI-digested vector pHXT1Up to create vector pHXT1Up-KAN. This plasmid was digested with EcoRI and SalI to generate the 1.7-kb DNA fragment, which was then inserted into EcoRI/SalI-digested pHXT1Down, thus yielding a PpHXT1 deletion vector, pUC18-HXT1-del. The deletion cassette was released from pUC18-HXT1-del as a 1.9-kb EcoRI-SphI-digested fragment and transformed by electroporation into wild-type P. pastoris strain GS115. G418 resistance transformants were isolated on YPD medium supplemented with 0.3 mg/ml G418. The correct integration of the deletion cassette into the genome and replacement of the PpHXT1 open reading frame (ORF) in the transformants were confirmed by PCR analysis and further DNA sequencing (data not shown).

(ii) Δhxt2 strain.A strain deleted for the PpHXT2 gene was isolated by the gene replacement method in the same way as Δhxt1 except using the zeocin resistance gene Sh ble as a marker. The zeocin resistance gene sequence with its own promoter and terminator (1,321 bp) was amplified by PCR using the plasmid pPICZ A (Invitrogen) as a template with primers Zeo1 and Zeo2.

(iii) Δatg30 Δhxt1 strain.We constructed the Δatg30 Δhxt1 strain by the deletion of the PpHXT1 gene in P. pastoris strain Δatg30 (a gift from S. Subramani and Jean-Claude Farré, University of California, San Diego) (10). The Δatg30 Δhxt1 strain was isolated by the gene replacement method in the same way as that used for Δhxt1 except using the zeocin resistance gene Sh ble as a marker.

(iv) Construction of strains with fluorescently labeled peroxisomes.Vector pRDM054 (kindly provided by S. Subramani, University of California, San Diego) (45) is an E. coli-P. pastoris shuttle vector capable of expressing a peroxisomally targeted blue fluorescent protein (BFP) (BFP-SKL) under the control of P_AOX1. The vector was linearized in the unique PmeI site in the P_AOX1 sequence and transformed into wild-type (WT) GS115 and the Δhxt1, Δhxt2, and Δatg30 Δhxt1 strains. The resultant recombinant was screened by hygromycin resistance. Four strains, WT-BFP-SKL, Δhxt1-BFP-SKL, Δhxt2-BFP-SKL, and Δatg30 Δhxt1-BFP-SKL, were used for further study.

(v) Construction of GFP expression strains.The GFP-SKL coding sequence was amplified from plasmid pP-GFP (a vector capable of expressing a green fluorescent protein [GFP] under the control of P_AOX1) (50) using Pyrobest DNA polymerase with primers GFP-SKL-1 and GFP-SKL-2, carrying restriction sites for BamHI and EcoRI, respectively. The PCR product was inserted into BamHI/EcoRI-digested pPIC3.5k (Invitrogen) to create plasmid pGS (a vector capable of expressing a peroxisomally targeted green fluorescent protein [GFP-SKL] under the control of P_AOX1). pP-GFP and pGS were then linearized with SalI and transformed into P. pastoris by electroporation. Transformants expressing GFP and GFP-SKL, respectively, were isolated on glucose-containing YNB medium without histidine. The correct integration into the genome was confirmed by PCR analysis and further DNA sequencing, and the single copy of the GFP expression cassette in P. pastoris was selected according to data reported in our previous work (50).

(vi) Heterologous complementation.Vector pBM2974 (30) with the S. cerevisiae ADH1 promoter was used for gene expression in the yeast S. cerevisiae with URA3 as the auxotrophic selectable marker. For the construction of expression plasmids, the PpHXT1 coding region amplified as a 1.62-kb HindIII/EcoRI fragment with primers CE1 and CE2 and the PpHXT2 coding region amplified as a 1.58-kb HindIII-SacI fragment with primers CE3 and CE4 were ligated into vector pBM2974, resulting in plasmids pZP10 and pZP20, respectively. The expression plasmids (pZP10 and pZP20) and control plasmids pBM3135 and pBM3362, which carry S. cerevisiae SNF3 (ScSNF3) and ScHXT1, respectively (kindly provided by M. Johnston, Washington University) (30), were used to transform S. cerevisiae hxt-null strain EBY.VW4000 (kindly provided by E. Boles, Düsseldorf University, Germany) (Table 1), which is unable to grow on hexose (48), yielding several transformants on medium containing 2% maltose without uracil. These transformants were then tested for growth on glucose or fructose.

Fluorescence microscopy.Fluorescence microscopy was performed essentially as described previously (9, 42). P. pastoris strains WT-BFP-SKL, Δhxt1-BFP-SKL, Δhxt2-BFP-SKL, and Δatg30 Δhxt1-BFP-SKL, transformed with plasmid pRDM054, were pregrown in glycerol medium to log phase and then shifted into YNB medium with methanol, fructose, fructose plus methanol, glucose, or glucose plus methanol for induction, respectively. Ten microliters of culture was visualized by fluorescence microscopy (Olympus bx 51).

Fluorescence microscopy of pexophagy was performed as described previously (28). Yeast cells were incubated in methanol medium with 4 μg/ml FM4-64. The cells were then washed once with sterile water and transferred into fresh glucose or fructose medium at an optical density at 600 nm (OD₆₀₀) of 1.0 to induce pexophagy.

Cell extract preparation, enzyme assays, and Western blot analysis.To prepare cell extracts, 30 to 50 OD₆₀₀ units of cells were harvested by centrifugation at 6,000 × g for 3 min, washed twice with ice-cold 50 mM potassium phosphate buffer (pH 7.0), and then frozen at −20°C. Cells were thawed and resuspended in 1 ml lysis buffer (50 mM potassium phosphate buffer [pH 7.0], 1 mM phenylmethylsulfonyl fluoride [PMSF]). Aliquots of 1 ml were mixed with 1.8 g glass beads (Biospec Products, Bartlesville, OK) in a 2.0-ml screw-cap tube followed by disruption with a bead disrupter (Mini-BeadBeater-8; Biospec Products) for 8 cycles (1 min vibrating and 1 min resting in ice for each cycle). The lysate was centrifuged at 20,000 × g for 30 min, the pellet was discarded, and the supernatant was utilized for measuring the enzyme activity or Western blotting. The protein concentration was determined with a Bradford protein assay kit (Tiangen Ltd., Shanghai, China).

The Aox activity was assayed spectrophotometrically with peroxidase and 2,2′-azino-di-(3-ethylbenzthiazoline sulfonate) (ABTS) as described previously (44). A unit of alcohol oxidase represents 1 μmol of product/min/mg of protein at 30°C. Samples with methanol omitted were run as blanks. The Aox colony assay was performed as previously described (37).

Total proteins were separated on polyacrylamide gels under denaturing conditions (45) and then transferred onto a polyvinylidene difluoride (PVDF) membrane using the electrophoretic transfer method with rabbit anti-Aox antibody (a kind gift from Suresh Subramani, University of California, San Diego) (27) as the primary antibody and peroxidase-conjugated goat anti-rabbit immunoglobulin G (IgG; Jackson ImmunoResearch) as the secondary antibody.

Quantitative real-time reverse transcription-PCR (qRT-PCR).Total RNAs were prepared by a standard procedure according to the manual of the Pichia expression kit (Invitrogen) and were subjected to DNase I (Promega, Madison, WI) treatment to exclude the genomic DNA contaminant.

RT-PCR was carried out with the Quantscript RT kit (Tiangene, China) according to the manufacturer's instructions. RT-PCR mixtures containing 1 μl Quant reverse transcriptase, 2 μl oligo(dT)15 (10 μM), 2 μl 10× RT mix, 2 μl deoxynucleoside triphosphates (dNTPs) (2.5 mM each), 11 μl RNase-free double-distilled water (ddH₂O), and 2 μg total RNA were incubated at 37°C for 1 h. The cDNA was then subjected to RT-PCR or quantitative real-time PCR using species-specific primers to measure transcript levels.

qRT-PCR was carried out by performing three independent experiments, each in triplicate, with an FTC-200 detector (Funglyn Biotech, Shanghai, China) in 20-μl reaction mixtures containing 10 μl 2× SYBR green real-time PCR master mix (Toyob, Japan), 1 μl cDNA sample, and optimal concentrations of each primer. Transcript levels were normalized to an endogenous reference gene, actin 1 (ACT1), in each sample by the ΔΔC_T method (21). The primers for qRT-PCR listed in Table 2 were designed by using Primer Express software (Applied Biosystems, Foster City, CA), with predicted products in the 100- to 300-bp size range.

Quantification of GFP expression by FCM.GFP expression in P. pastoris was analyzed by the flow cytometry (FCM) method (49, 50). After being diluted to 1 × 10⁷ cells/ml with phosphate-buffered saline, 1 ml of sample was analyzed with a 488-nm argon laser on a FACSCalibur apparatus (Becton Dickinson, Franklin Lakes, NJ). The fluorescence values were determined by a geometric mean (Geo-Mean) method.

Nucleotide sequence accession numbers.The nucleotide sequences of PpHXT1 and PpHXT2 reported in this paper have been submitted to the GenBank database with accession numbers GU479989 and GU479990, respectively.

Sequence analysis of two hexose transporters in P. pastoris.The translated protein sequence of the P. pastoris hexose transporter homolog PpHxt1 consists of 537 amino acids, and PpHxt2 has 522 amino acids. Both transporter proteins belong to the major facilitator superfamily (MFS), which includes a variety of transport systems in eukaryotes and in prokaryotes (24). Phylogenetic analysis revealed that PpHxt1 is a member of the subgroup of hexose transporters that contains all other known yeast hexose transporters. The branch of PpHxt2 belongs to the subgroup of hexose transporters, but the relatedness to other transporters is far lower (Fig. 1), which implies that PpHxt2 is probably a species-specific transporter and plays a significant role in sugar metabolism in P. pastoris.

• Open in new tab
• Download powerpoint

FIG. 1.

Phylogenetic tree of selected hexose transporter-related proteins. The tree and bootstrap values were calculated using the neighbor-joining method implemented in MEGA, version 3.1 (18), based on the alignment of the amino acid sequences using the ClustalX program (38). Values indicate the number of times (in percentages) that each branch topology was found in 2,000 replicates of a bootstrap analysis, assuming gamma-distributed rates for amino acid substitutions. The distance between branch points shows the relatedness of the sequences. The length of the scale bar is equivalent to 0.1 substitutions of amino acids per site. The branches of PpHxt1 and PpHxt2 are highlighted. Species abbreviations for each protein name are as follows: Sc, Saccharomyces cerevisiae; Kl, Kluyveromyces lactis; Pp, Pichia pastoris; Ca, Candida albicans; Ps, Pichia stipitis; Ci, Candida intermedia; Hp, Hansenula polymorpha. The GenBank accession numbers of the sequences are as follows: NP_011962 for ScHxt1, ABW76622 for ScHxt2, NP_010632 for ScHxt3, NP_011960 for ScHxt4, ABW76628 for ScHxt5, EDN60674 for ScHxt6, NP_010629 for ScHxt7, AAU43755 for ScGal2, P53387 for KlKht2, XP_453656 for KlRag1, XP_719598 for CaHgt7, XP_001387898 for PsSut1, CAI77652 for CiGxf1, EU476006 for HpHxt1, AAR88143 for HpGcr1, CAI44932 for CiGxs1, EU476007 for HpHxs1, XP_723173 for CaHgt4, CAA75114 for KlRag4, NP_010087 for ScSnf3, and NP_010143 for ScRgt2.

PpHxt1 and PpHxt2 proteins are hexose transporters.Due to similarities between the two genes and other hexose transporters, a potential role in the transport of hexoses could be assigned to them. To confirm this hypothesis, we attempted to complement an S. cerevisiae hxt-null mutant strain, EBY.VW4000 (48), which is unable to grow in the presence of glucose or fructose due to the lack of hexose transporters. The transformants containing the hexose transporter gene PpHXT1, PpHXT2, or ScHXT1 (30) showed growth on glucose or fructose, while those containing an empty vector or the hexose sensor gene ScSNF3 (30) did not show any growth (Fig. 2). This indicated that either PpHxt1 or PpHxt2 could function as a hexose transporter in P. pastoris. However, the two PpHXT genes complemented the growth defect of this strain to different degrees, with PpHXT1 complementing better and PpHXT2 complementing only very weakly, which might be explained by the distant relatedness of PpHxt2 to the transporters of S. cerevisiae and also implied a special role for PpHxt2 in sugar fermentation in P. pastoris.

• Open in new tab
• Download powerpoint

FIG. 2.

S. cerevisiae hxt-null strain EBY.VW4000 was transformed with plasmid pZP10 carrying PpHXT1 or pZP20 carrying PpHXT2 and spotted onto YNB agar medium supplemented with the indicated levels of hexose and with the requisite amino acids (leucine, tryptophan, and histidine) and uracil. Plates were incubated at 30°C for 3 days. As a control, the strain was transformed with empty vector pBM2974, pBM3135 carrying the complete ScSNF3 gene, or pBM3362 carrying the complete ScHXT1 gene, respectively.

Regulation of transcription of the two HXT genes in P. pastoris.To determine whether the expression of the hexose transporter genes is inducible by glucose at the transcriptional level, PCR analysis was performed with total cDNA of P. pastoris wild-type strain GS115 grown in YNB medium with different carbon sources. Both genes were expressed in cells grown on a nonfermentable carbon source (glycerol) or with different levels of glucose (Fig. 3A). However, PpHXT2 expression was weak, being about 100-fold less than that with fully induced PpHXT1.

• Open in new tab
• Download powerpoint

FIG. 3.

Semiquantitative (A) and quantitative (B) analyses of transcript levels of PpHXT1 and PpHXT2 in response to different carbon substrates. Wild-type strain GS115 was grown overnight in glycerol medium and then incubated in fresh YNB medium containing 1% glycerol or the indicated levels of glucose at 30°C for 3 h. Cells were harvested, and total RNA from each culture was prepared by a standard procedure according to the manual of the Pichia expression kit (Invitrogen). RT-PCR was performed on 2 μg total RNA in a 20-μl reaction mixture. (A) PCR (25 cycles) was performed on 1 μl of the five-times-diluted cDNA template using species-specific primers for ACT1, PpHXT1, or PpHXT2. Control reactions lacking cDNA produced no products. (B) qRT-PCR was performed as described in Materials and Methods. The relative expression level indicated on the y axis (2^−ΔΔCT) for each gene at each glucose concentration was normalized for its expression in glycerol-grown cells.

To determine more precisely the dependence of the induction of each HXT gene on the external glucose concentration, we monitored the mRNA levels of the two genes in cells growing with different glucose concentrations by quantitative real-time RT-PCR (qRT-PCR) (Fig. 3B). Transcript levels of PpHXT2 are fully induced in cells grown on low levels of glucose, ranging from 0.025 to 0.1%, but become 3- to 5-fold repressed when glucose levels are high (above 0.5%). Thus, transcript levels of PpHXT2 are induced at low glucose concentrations and repressed by high levels of glucose. In contrast, transcript levels of PpHXT1 are induced by high levels of glucose and repressed by low glucose concentrations. We also found that transcript levels of PpHXT1 are not greatly different between wild-type and Δhxt2 strains, and neither are the PpHXT2 levels between wild-type and Δhxt1 strains (data not shown).

Growth characteristics of P. pastoris Δhxt1 and Δhxt2 mutants.In liquid cultures with different concentrations of glucose, the growth kinetics of the Δhxt1 strain did not differ dramatically from that of the wild-type strain. In contrast to the Δhxt1 strain, the growth of the Δhxt2 strain was impaired significantly in medium with either high (2%) or low (0.1%) concentrations of glucose medium (Fig. 4). The growth impairment of the Δhxt2 strain on fructose medium was also more significant than that of Δhxt1 (Fig. 5C). Interestingly, the final biomass of each Hxt deletion strain at stationary phase was a bit higher than that of the wild-type strain, and this difference was more obvious when strains were grown in medium containing 1% glucose or 1% fructose (Fig. 5A and C).

• Open in new tab
• Download powerpoint

FIG. 4.

Growth of the wild type and the Δhxt1 and Δhxt2 mutants of P. pastoris. Strains were grown at 30°C on YNB medium containing 0.1% glucose or 2% glucose. Growth was monitored by measuring the OD₆₀₀ at the times indicated. Data are mean values from two independent cultivations.

• Open in new tab
• Download powerpoint

FIG. 5.

Growth and catabolite repression of Aox activity in the wild-type, Δhxt1, and Δhxt2 strains on medium containing glucose (A), glucose plus methanol (B), fructose (C), or fructose plus methanol (D). Cells of each strain were preincubated overnight in glycerol medium and transferred into fresh YNB medium supplemented with the indicated levels (percentages) of carbon sources at an initial OD₆₀₀ of 0.03. Aox specific activity was measured in cell extracts prepared as described in Materials and Methods. Biomass optical density and Aox activity represent the means ± standard deviations (SD) of single measurements from two independent cultivations.

Antimycin A is a respiration inhibitor that blocks mitochondrial electron transfer between cytochromes b and c and renders the growth of the yeast cells entirely dependent upon glycolytic ATP synthesis (31). On solid medium without antimycin A, the Δhxt1 and Δhxt2 mutants showed no significant difference relative to the wild type (Fig. 6). When respiration was blocked by antimycin A, the Δhxt2 mutant could not grow on glucose or fructose (Fig. 6).

• Open in new tab
• Download powerpoint

FIG. 6.

Growth of the wild type and Δhxt1 and Δhxt2 mutants on glucose (A) or fructose (B) medium containing antimycin A. The three strains, wild-type strain GS115 and the Δhxt1 and Δhxt2 mutants, were pregrown in glycerol medium to log phase and then diluted to an OD₆₀₀ of 0.1, and 10 μl of each was spotted onto YNB agar medium containing the indicated levels of hexose without antimycin A (−Ant A) (left column, 3 days of growth) or with 3 μg/ml antimycin A (+Ant A) (right column, 5 days of growth).

Thus, the deletion of PpHXT2 has a more significant effect on hexose utilization than does the deletion of PpHXT1. In addition, PpHxt2 is especially important for growth when respiration is inhibited, which implies that PpHxt2 plays an important role in glycolysis-dependent fermentative growth.

Role of PpHXT1 and PpHXT2 in catabolite repression of Aox expression.In P. pastoris, Aox expression is tightly repressed by glucose or fructose, while methanol is absolutely necessary for the high-level induction of this gene (5, 40). To reveal the role of the two hexose transporters in the catabolite repression of Aox, the Δhxt1 mutant and the Δhxt2 mutant of P. pastoris were examined for the catabolite repression of Aox in different carbon sources. The wild-type-related strain GS115 was analyzed as a reference. In Δhxt1 cells, Aox activity could be detected in a repressible carbon source, i.e., glucose or fructose (Fig. 5), and the release of Aox repression is more distinct in the mixed carbon source of glucose or fructose plus methanol (Fig. 5B and D), which also could be confirmed by Western blot detection (Fig. 7).

• Open in new tab
• Download powerpoint

FIG. 7.

Western blot detection of the Aox protein in cells induced with different carbon sources. Cells were pregrown in glycerol medium and then washed with sterile water and shifted into YNB medium supplemented with the indicated carbon sources at an OD₆₀₀ of about 1.0. Upon induction for 10 h, the cells were harvested. Fifteen micrograms of total protein was loaded into the Δhxt1/methanol and WT/methanol lanes, while 20 μg of total protein was loaded into each of the other lanes.

Normally, in a wild-type strain, Aox activity induced by methanol increases rapidly at the exponential growth phases and then keeps relatively constant for several hours or drops slowly. In Δhxt1 cells, although Aox was derepressed in the common repressible carbon source glucose or fructose, the change in the profile of Aox activity was much different from that for the cells induced by methanol. The levels of Aox activity increased rapidly at the early exponential growth phases (from about 10 h to 20 h) but dropped sharply afterwards under all the repressible conditions in Δhxt1 cells (Fig. 5). This transient loss of catabolite repression in the Δhxt1 mutant was not caused by growth retardation since a more obvious growth retardation occurred in the Δhxt2 mutant (Fig. 5A and C).

In order to confirm that the release of Aox repression in Δhxt1 mutants on different carbon sources was taking place at the transcriptional level, we transformed strains with plasmid pRDM054 (a vector capable of expressing a peroxisomally targeted blue fluorescent protein [BFP-SKL] under the control of P_AOX1). As expected, typical punctate fluorescence (45) could be observed in Δhxt1 mutant cells grown on nonmethanol carbon sources, i.e., glucose or fructose, that failed to repress the transcription of the AOX1 promoter (Fig. 8A), and the cells of neither the Δhxt2 mutant nor the wild-type strain grown on fructose, glucose, glucose plus methanol, or fructose plus methanol exhibited fluorescence (data not shown). However, the punctuate fluorescence was transient and could not be observed at a later growth stage (data not shown), which was coincident with the transient Aox activity shown in Fig. 5.

• Open in new tab
• Download powerpoint

FIG. 8.

(A) Fluorescence microscopy images of wild-type and Δhxt1 strains expressing the peroxisome-targeted BFP-SKL fusion protein on different carbon sources. (Left) Fluorescence images of wild-type (WT), GS115 (BFP-SKL), and Δhxt1 (BFP-SKL) cells. (Right) Phase-contrast images of the same cells in visible light. Wild-type cells, induced in methanol or hexose, are presented as a positive or negative control, respectively. DIC, differential interference contrast (Normarski optics). Scale bar, 2 μm. (B) Δhxt1 and Δatg30 Δhxt1 cells expressing BFP-SKL were transferred from methanol to glucose or fructose medium to induce pexophagy. Merged images show vacuoles stained by FM4-64 (red) and peroxisomes labeled with BFP-SKL (blue). Scale bar, 2 μm. (C) Aox colony assay of wild-type (GS115), Δhxt1, and Δatg30 Δhxt1 strains under pexophagy-triggering conditions. Cells were pregrown for 2 days on YNB agar medium with methanol to induce peroxisomes and peroxisomal Aox and then replica plated onto glucose- or fructose-containing medium to induce pexophagy. After incubation for 12 h, Aox activity was visualized by overlaying the Aox activity reaction mixture with the permeabilizing agent. High Aox activity results in cells stained red.

Role of pexophagy in catabolite repression of Aox expression in the Δhxt1 strain.Pexophagy, the specific degradation of peroxisomes by autophagy-related pathways, has been best studied with methylotrophic yeasts (11, 17). P. pastoris cells grown on methanol as a carbon source induce the synthesis of peroxisomes and the enzymes required for methanol metabolism, e.g., Aox. Upon a shift of cultures to other carbon sources such as glucose or fructose, peroxisomal metabolism is no longer required. The peroxisomes become redundant for growth and are subject to degradation by the vacuole through autophagic pathways. This process is called “pexophagy,” which occurs selectively toward peroxisomes (11).

The enzyme Aox is expressed in peroxisomes in P. pastoris. To determine whether the transient derepression of Aox in glucose and fructose in the Δhxt1 mutant (Fig. 5) was caused by pexophagy, the GFP protein was expressed under the control of P_AOX1 in peroxisomes and the cytoplasm in P. pastoris, respectively. We found that the expression profile of GFP-SKL shown in Fig. 9 was similar to that of Aox shown in Fig. 5, both of which exhibited a sharp decline after an initial ascent. Interestingly, the protein level of GFP expressed in the cytoplasm (Δhxt1-GFP) was higher and more persistent than that of GFP targeted to the peroxisomes (Δhxt1-GFP-SKL) (Fig. 9B to E), whereas there were not obvious differences between them in methanol medium (Fig. 9A). Therefore, it is deduced that the sharp dropping of the sugar-induced expression levels of Aox and GFP-SKL in the Δhxt1 mutant at a later growth phase is a result of rapid degradation by pexophagy, since both Aox and GFP-SKL are peroxisomal proteins, and the Δhxt1 mutant still exhibits normal wild-type pexophagy in response to glucose or fructose (Fig. 8B and C).

• Open in new tab
• Download powerpoint

FIG. 9.

Flow cytometry analysis of GFP expression levels in the cytoplasm (GFP) and peroxisomes (GFP-SKL) based on P_AOX1 in Δhxt1 and Δatg30 Δhxt1 strains on medium containing methanol (A), fructose (B), fructose plus methanol (C), glucose (D), or glucose plus methanol (E). Cells of each strain were preincubated overnight in glycerol medium and transferred into fresh YNB medium supplemented with the indicated carbon sources at an initial OD₆₀₀ of 0.03. GFP fluorescence intensity was measured as described in Materials and Methods. Data are mean values from two independent cultivations.

When the pexophagy pathway in glucose and fructose is blocked (Fig. 8B and C) via the knocking out of the PpHXT1 gene in a pexophagy-deficient Δatg30 strain (10), the expression level of GFP-SKL remained high for a longer time and was higher than that in an Δhxt1 mutant, as expected for fructose medium (Fig. 9B and C). However, the dropping expression level of the GFP-SKL protein in response to glucose seems to be less affected in the Δatg30 Δhxt1 strain (Fig. 9D and E).