- Email: [email protected]
Pan1p, an actin cytoskeleton-associated protein, is required for growth of yeast on oleate medium
Experimental Cell Research 310 (2005) 482 – 492 www.elsevier.com/locate/yexcr
Research Article
Pan1p, an actin cytoskeleton-associated protein, is required for growth of yeast on oleate medium Joanna Kamin´ska, Monika Wysocka-Kapcin´ska, Iwona Smaczyn´ska-de Rooij, Joanna Rytka, Teresa Z˙oaa˛dek* Department of Genetics, Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Pawin´skiego 5a, 02-106 Warsaw, Poland Received 2 March 2005, revised version received 30 June 2005, accepted 30 August 2005 Available online 19 September 2005
Abstract Pan1p is a yeast actin cytoskeleton-associated protein localized in actin patches. It activates the Arp2/3 complex, which is necessary for actin polymerization and endocytosis. We isolated the pan1 – 11 yeast mutant unable to grow on oleate as a sole carbon source and, therefore, exhibiting the Oleate phenotype. In addition, mutant cells are temperature-sensitive and grow more slowly on glycerol or succinatecontaining medium but similarly to the wild type on ethanol, pyruvate or acetate-containing media; this indicates proper functioning of the mitochondrial respiratory chain. However, growth on ethanol medium is compromised when oleic acid is present. Cells show growth arrest in the apical growth phase, and accumulation of cells with abnormally elongated buds is observed. The growth defects of pan1 – 11 are suppressed by overexpression of the END3 gene encoding a protein that binds Pan1p. The morphology of peroxisomes and induction of peroxisomal enzymes are normal in pan1 – 11, indicating that the defect in growth on oleate medium does not result from impairment in peroxisome function. The pan1 – 11 allele has a deletion of a fragment encoding amino acids 1109 – 1126 that are part of (QPTQPV)7 repeats. Surprisingly, the independently isolated pan1 – 9 mutant, which expresses a truncated form of Pan1p comprising aa 1 – 859, is able to grow on all media tested. Our results indicate that Pan1p, and possibly other components of the actin cytoskeleton, are necessary to properly regulate growth of dividing cells in response to the presence of some alternative carbon sources in the medium. D 2005 Elsevier Inc. All rights reserved. Keywords: Yeast; Actin cytoskeleton; Peroxisomes; Oleic acid; Apical growth
Introduction Fatty acids are the building blocks of lipids that constitute cellular membranes, components of intracellular messengers (e.g. phosphatidylinositoles), and also modify proteins. Moreover, fatty acids are efficient fuel molecules, and yeast cells utilize exogenous free fatty acids as a source of biomass and energy. Oleic acid as a sole carbon source or added to a medium containing another nonfermentable carbon source, such as ethanol, represses genes of the fatty acid biosynthetic pathway and induces those encoding h-oxidation enzymes.
* Corresponding author. Fax: +48 22 658 46 36. E-mail address: [email protected] (T. Z˙oaa˛dek). 0014-4827/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2005.08.018
However, the signaling pathway for oleic acid sensing is unknown. In yeast, enzymes of the h-oxidation pathway are located in peroxisomes. Thus, biogenesis of peroxisomes is induced upon addition of oleic acid to the medium [1]. After uptake, oleic acid can be activated by at least four cytoplasmic acyl-CoA synthetases [2]. Acyl-CoA entry into peroxisomes is mediated by an ABC transporter [3]. The first step of h-oxidation is catalyzed by acyl-CoA oxidase, Pox1p, and is accompanied by reduction of molecular oxygen to hydrogen peroxide. The hydrogen peroxide is in turn decomposed by peroxisomal catalase A (Cta1p). Peroxisomes oxidize fatty acids to acetyl-CoA, and the reaction is accompanied by production of NADH. Further oxidation and energy production require fully functional mitochondria. Acetyl-CoA is transported to mitochondria via two alternative pathways: by acetyl carnitine transferases
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
or by glyoxylate cycle metabolites of acetyl-CoA, isocitrate and succinate [4]. It subsequently enters the citric acid cycle. NADH must also be shuttled to mitochondria. Growth on oleic acid, similarly as on other nonfermentable substrates, such as glycerol, ethanol or acetate, requires induction of gluconeogenesis genes (reviewed in [5]). Thus, coordinated expression of peroxisomal, mitochondrial and cytosolic enzymes is needed for utilization of oleic acid. Among yeast mutants defective in growth on oleic-acidcontaining media, there are many disturbed in peroxisome biogenesis or lacking peroxisomal enzymes (pas, fox, reviewed in [1]) as well as the following: odc1D odc2D which lacks mitochondrial oxodicarboxylate transporters [6]; rml2 with a defective mitoribosomal protein [7]; zwf1D idp2D which does not produce cytosolic NADPH [8]; and agc1D which lacks the mitochondrial aspartate – glutamate transporter involved in NADH shuttle [9]. Recently, we have found that kgd1 and lip5 mutations, which disturb aketoglutarate dehydrogenase, are impaired in oleate utilization due to low expression of peroxisomal enzymes in response to mitochondrial dysfunction [10]. All these mutants retain the ability to grow on ethanol, acetate or glycerol. In an effort to identify novel genes required for utilization of oleic acid, we isolated Oleate mutants defective in growth on media containing oleic acid as a sole carbon source while retaining the ability to grow on ethanol. We identified the PAN1 gene and END3, a multicopy suppressor of pan1, as required for growth on oleate. It had previously been shown that Pan1p is a multifunctional protein involved in endocytosis, actin cytoskeleton organization [11] and cell wall morphogenesis [12]. In this report, we show that expression of a DNA fragment encoding the first 859 amino acids (aa) of Pan1p is sufficient for growth of cells on oleate, although the fragment encoding (QPTQPV)7 repeats (aa 1084 –1125), if present, also plays an important role. The pan1 –11 mutant expressing the protein devoid of three of these repeats shows arrest in the apical growth phase when grown on oleate. Furthermore, glycerol and succinate in the medium inhibited the growth of this mutant. Presented results suggest that the Pan1p function in growth on oleic acid results from its role in the regulation of apical growth of the bud in response to carbon source.
Materials and methods Yeast strains, genetic methods, media and reagents Saccharomyces cerevisiae strains used were WS-K MATa leu2::P CTA1 -URA3::LEU2 arg1 ura3 trp1 [13], T81D, T37-2A (pan1– 8), TZ35 (pan1– 10), TZ81 (pan1– 9) [11], RH144-3D, RH266-1D (end3-1) [14] and MHY500 [15]. Strain WS-6 MATa leu2::P CTA1 -URA3::LEU2 pan1 – 11 arg1 ura3 trp1 was obtained by UV mutagenesis of WS-K. Strains JK48-3B MATa ura3-1 leu2 trp1 his3D-200
483
and JK48-3C MATa pan1– 11 ura3-1 leu2 trp1 lys2-801 were obtained from WS-6 MHY500 cross. Strain JK51 was obtained by integration of YIpCTA1-lacZ into ura3 locus of JK48-3C and strain MW20 by integration of pHPR131 into trp1 locus of JK48-3C. Standard genetic manipulations were performed as described by Sherman [16]. Yeast were grown on rich media containing as a carbon source 2% glucose (YPD); 2% ethanol (YPE); 2% glycerol (YPG); 0.1% glucose and 2% ethanol (YPDE); 0.25% oleic acid in Tween 20 (YPO); 2% ethanol and 0.25% oleic acid (YPEO); 0.1% glucose, 1% ethanol and 0.25% oleic acid in Tween 20 (YPDEO). The medium with acetate as a carbon source (YPA) was prepared as in [17]. Rich media with pyruvate (YPP) or succinate (YPS) as a carbon source contained 1% of the respective sodium salt. Alternatively, cells were grown on SC-ura, SC-trp or SC-leu selective media containing 2% glucose or SCG containing 2% glycerol, on SCO-trp or SCO-ura containing 0.25% oleic acid in Tween 20, on SCDE-ura-leu containing 0.1% glucose and 2% ethanol or on SCDEO-ura-leu containing 0.1% glucose, 2% ethanol and 0.25% oleic acid. The latter group of media were supplemented with the required aa, uracil or adenine. The growth of yeast strains on plates containing different carbon sources was compared by a drop test. Cells grown overnight were suspended at 1 107 cells/ml, serially diluted 1:10 and 3 Al of each dilution was spotted on the plate. Plates were incubated for 2– 4 days at the temperature indicated. Plasmids and plasmid constructions Plasmids used were: pT4 containing PAN1 gene, YIpT4 containing the EcoRI fragment of PAN1 [11], pHT837 containing PAN1 [19], pJR233 with the gene encoding green fluorescence protein (GFP) with a PTS1 (peroxisome targeting signal) at the C-terminus under the control of the MLS1 promoter [20], pHPR131 expressing PTS2-DsRed [21] and YIpCTA1-lacZ bearing PCTA1 -lacZ reporter (from M. Skoneczny, IBB PAS). Both pSR19 and pSR48 were isolated from a genomic library based on pRS314 (TRP1 marker; [22]) by complementation of the Oleate phenotype of pan1 – 11. pSR19 contains a 7.4 kb genomic fragment bearing END3, YNL083W and PMS1 genes. pSR48 contains a 6.3 kb genomic fragment bearing YIR007W and the first 2919 bp from the 5V-end of PAN1 open reading frame (orf). YEp351PAN1 bearing a genomic fragment containing PAN1 was isolated from a genomic library based on the YEp351 plasmid (LEU2 marker; M. Wysocka-Kapcin˜ska, unpublished) by complementation of the ts phenotype of pan1. YIpT4 was digested with StuI to integrate into the ura3 locus of JK48-3C and test for pan1 – 11 complementation. YIpT4DSnaBI was obtained by SnaBI digestion and religation of YIpT4 and used, after further SnaBI digestion,
484
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
for gap-repair mapping of pan1 mutations. YIpCTA1-lacZ was digested with StuI and integrated into the ura3 locus of JK48-3C strain.
a Hamamatsu ORCA 100 camera, and images were collected using Lucia G software. Actin cytoskeleton staining
Characterization of pan1 mutations The mutations pan1 – 8, – 9, –10 and –11 were mapped by gap-repair as described by Rothstein [23]. Total yeast DNA was isolated as in [24]. Fragments of pan1 alleles were amplified by PCR using specific primers. PCR products were sequenced in the DNA Sequencing and Oligonucleotide Synthesis Laboratory at IBB, PAS. The sequences obtained were compared to the sequences from Saccharomyces Genome Database using BLAST or SeqMan algorithms. Activity of b-galactosidase Yeast were grown in 5 ml YPD precultures and were inoculated into 20 ml of YPDEO. Cells grown overnight were collected, washed with water, and extracts were prepared with glass beads. Aliquots of total protein extracts were taken, and the protein concentration was estimated by the BioRad assay according to manufacturer’s instructions. The h-galactosidase activity was assayed spectrophotometrically at 420 nm using o-nitrophenyl-hd-galactopyranoside as substrate [25]. The activity was expressed in nanomole of o-nitrophenol/min/mg protein. In the experiments, five independent transformants were combined, inoculated, and their extracts were assayed in duplicates. Western blot analysis Protein extracts were analyzed by a standard Western blotting method [26] with signal development by chemiluminescence (ECL Plus, Amersham). Antibodies used were affinity-purified rabbit anti-N-term-Pan1p and anti-Cterm-Pan1p antibodies (PAN91 and PAN92, kind gift from A. Sachs, University of California at Berkeley [18,27]) and secondary anti-rabbit horseradish peroxidase-conjugated antibody (DACO).
For actin cytoskeleton staining, 10 ml cultures in SC, SCO or SCG were grown overnight to OD600 ¨0.5. Actin was stained as described [28], with some modifications. Cells were fixed for 2 h by adding to the culture concentrated formaldehyde solution to a final concentration of 3.7% and adjusting the pH to 6.5 with potassium phosphate buffer. The cells were subsequently collected by centrifugation, washed three times in phosphate-buffered saline (PBS), pH 6.5 and stained with Oregon Green 488conjugated phalloidin (Molecular Probes) for 2 h in the dark. Cells were washed five times with PBS, stained with DAPI (0.5 Ag/ml), washed twice and observed in Eclipse, a fluorescence microscope. Endocytosis assay Fluid phase endocytosis assays were carried out as described by Dulic et al. [29]. Strain pan1– 11 (JK48-3C)bearing vector or YEp351-PAN1 plasmid were grown in SC medium to OD600 of 0.5 – 0.8. About 107 cells were collected and incubated with Lucifer yellow (LY; Sigma) for 1.5 h at 30-C. An Eclipse fluorescence microscope was used to observe LY accumulation in the vacuole, and vacuolar morphology was viewed with DIC optics. Transmission electron microscopy Cells grown on YPD to the logarithmic phase were fixed with 2% glutaraldehyde followed by 2% OsO4 for 2 h. Thin sections were cut from Epon blocks and post-stained with 2% uranyl acetate followed by lead citrate [30]. Micrographs were taken on a JEOL transmission electron microscope JEM1220.
Results
Fluorescence microscopy
Isolation and genetic analysis of mutants that do not grow on media containing oleic acid as a sole carbon source
To assess inheritance of peroxisomes, respective transformants expressing PTS1-GFP or PTS2-DsRed were grown on SC-ura or SC-trp to the logarithmic phase, OD600 ¨1, and at least 120 budding cells of each strain were scored for the presence in the bud of a fluorescing spot, corresponding to a peroxisome. To observe peroxisomes on oleic-acid-containing medium, cells expressing PTS1-GFP were grown on SCE, transferred to SCEO and incubated o/n. Cells were viewed by fluorescence microscopy and with differential interference contrast (DIC) optics using an Eclipse (Nikon) microscope equipped with
To identify new factors necessary for growth on oleate, mutants were isolated which were unable to grow on oleicacid-containing medium (Oleate ) while retaining the ability to utilize ethanol. Standard UV mutagenesis of the WS-K strain, with 10% survival rate, was carried out. The UV-treated cells were plated on YPDE plates, grown for 3 days and then replica-plated on YPO medium containing oleic acid as a sole carbon source. One of the mutants, unable to grow on oleate medium, named WS-6, was further characterized. WS-6 was crossed with wild type strain MHY500, and the diploid displayed wild-type phenotype.
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
Analysis of the haploid progeny of this diploid revealed 2:2 segregation of the wild type: Oleate phenotype indicating that the loss of the ability to grow on oleate as a sole carbon source resulted from a single recessive mutation. Additionally, the mutation caused a temperature-sensitive phenotype (ts) as the mutant strain was not able to grow on YPD at 37-C. The ts phenotype segregated together with the Oleate phenotype in 30 tetrads tested. To identify the mutated gene, the WS-6 mutant strain was transformed with a yeast genomic library based on single-copy pRS314 vector with TRP1 marker [22]. The Trp+ colonies were replicaplated on synthetic medium containing oleic acid as a sole carbon source (SCO-trp) and incubated at 30-C for 5 days. From the 10,000 Trp+ transformants tested, two colonies grew on these plates. Standard isolation and retransformation of the recovered plasmids followed by phenotype testing and determination of DNA sequence were carried out. Plasmids pSR19 and pSR48 restored growth of WS-6 cells on SCO-trp medium at 30-C and on YPD medium at 36-C, albeit to a different extent (Fig. 1, see Fig. 3). Plasmid pSR19 contained the END3, YNL083W and PMS1 genes, and plasmid pSR48 contained YIR007W and the first 2919 base pairs from the 5V-end of PAN1 orf. The END3 and PAN1 genes were functionally related, both being involved in endocytosis and actin cytoskeleton organization. Furthermore, End3p interacted with Pan1p in vivo, and pan1– 4 could be suppressed by END3 overexpression [12]. There-
Fig. 1. The Oleate phenotype of pan1 – 11 is suppressed by END3 gene overexpression. Growth of wild type strain transformed with vector and pan1 – 11 strain transformed with vector, plasmid pSR19 or pSR48, bearing END3 or fragment of PAN1 respectively. Media and incubation temperatures are indicated.
485
fore, it was quite possible that the Oleate phenotype resulted from mutation in one of these genes. To determine whether the WS-6 strain bears the mutation in the END3 or PAN1 gene, a complementation test was performed. WS-6 was crossed with two temperature-sensitive mutants, pan1– 8 (T37-2A) and end3-1 (RH266-1D). The resulting diploids were grown on YPD medium at 30-C, replica-plated on YPD and incubated at 37-C. In contrast to the end3-1 mutation, pan1 –8 did not complement the ts phenotype of WS-6 strain. Hence, the WS-6 strain bears a mutation in the PAN1 gene, and END3 is a suppressor gene. This mutation, referred to as pan1– 11, was also complemented by the YEp351-PAN1 plasmid, bearing a complete PAN1 gene, and by plasmid YIpT4, bearing the EcoRI fragment of PAN1, integrated into the ura3 locus (see Fig. 3). Complementation of pan1 – 11 mutation by incomplete PAN1 orf is consistent with the previous finding that only part of the gene is necessary for its essential function(s) [11,31]. These results indicate that Pan1p and End3p are required not only for endocytosis and actin cytoskeleton function but also for growth of cells on a medium containing oleic acid as a sole carbon source. Characterization of pan1 mutant alleles and their products The pan1 –8, – 9 and – 10 mutants were isolated in an earlier screen for mutants affecting cytoplasmic – mitochondrial distribution of Mod5-Ip, a tRNA-modifying enzyme [11]. These strains were investigated, together with pan1– 11, in subsequent studies in which pan1 alleles and their products were characterized. First, total protein extracts prepared from pan1 – 8, –9, –10 and – 11 mutant strains and from parental strains were separated by SDS-PAGE and analyzed by Western blotting with anti-Pan1p antibodies. Two types of antibodies were used, one raised against the second N-terminal repeat of Pan1p, aa 392– 662, (anti-Nterm-Pan1p), and the other against aa 1059 –1113 containing QPTQPV repeats of Pan1p (anti-C-term-Pan1p). The anti-N-term-Pan1p antibody recognized a protein of about 160 kDa in parental strains and in pan1– 11 extract as well as a shorter form of Pan1p, of about 100 kDa, in pan1– 8, – 9 and – 10 extracts. The anti-C-term-Pan1p antibody recognized Pan1p in the wild type strains, weakly in pan1 – 11 but not in the pan1 – 9 mutant (Fig. 2). These results confirmed the predicted size of wild type Pan1p, indicated the size of mutant Pan1 proteins and also documented that the fragment of Pan1p containing the QPTQPV repeats is absent from all the mutants or is not recognized by the anti-C-term-Pan1p antibody. The pan1 mutations were subsequently mapped in PAN1 gene by the gap-repair method. The mutations were identified in pan1 – 8, – 9, – 10 and – 11 alleles by sequencing PCR products of reactions performed with genomic DNA from the respective strains as templates. Surprisingly, the same mutational change, a deletion of nucleotide A2582 from PAN1 orf, was found in pan1 – 8, – 9
486
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
QPTQPV repeats, polyproline sequences and the acidic tail binding to the Arp2/3 complex involved in actin polymerization [32] (Fig. 3). Molecular characterization of the pan1– 11 allele revealed a 54 base pair deletion from nucleotide 3325 to 3378 of PAN1 orf, codons 1109 –1126. This fragment encodes three QPTQPV repeats which are normally repeated seven times in wild type Pan1p. The pan1 – 9 mutant grows on oleic-acid-containing medium, whereas end3-1 is defective
Fig. 2. Characterization of Pan1 proteins synthesized by pan – 8, – 9, – 10 and – 11 mutant strains. Western blot of total protein extracts from pan1 mutant strains grown at 30-C is shown.
and – 10 alleles, even though they were isolated independently. This deletion caused frameshift and generated an adjacent stop codon. Thus, the pan1– 8, – 9 and –10 alleles encoded mutant Pan1p that was shorter than the wild type Pan1p, 859 aa versus 1480 aa, with the last four aa differing from the corresponding ones of the wild type. Pan1p1 – 859 retains the N-terminal motifs homologous to Sla1p and also the EH domains responsible for End3p and Sla1p binding and linking Pan1p to endocytic machinery. It is missing the
The pan1– 9 mutant was chosen for further characterization as a representative of pan1 –8/9/10. The growth of pan1– 9 and pan1 –11 mutants and the respective parental strains was compared on YPO and YPD media at 30-C and on YPD at 37-C. Neither pan1 mutant grew on YPD at 37-C (see Fig. 1 and not shown). Surprisingly, and in contrast to pan1 –11, the pan1 – 9 mutant was able to grow on oleic-acid-containing media similarly to the wild type strain (Fig. 4, see Fig. 1). The ts phenotype of pan1 – 9 was suppressed by the plasmid bearing a PAN1 fragment encoding Pan1p1 – 973. Thus, the pan1 mutants studied exhibited the ts phenotype, but the growth defect on oleate characterized only one of them. Interestingly, Pan1p1 – 973 was sufficient for growth at an elevated temperature, indicating that aa 860 –973 were critical for this function. Moreover, the C-terminal fragment of Pan1p, aa 860– 1480, was not absolutely required for growth on oleic acid since the pan1 –9 mutant expressing Pan1p1 – 859 did not exhibit the Oleate phenotype. However, the deletion of three QPTQPV repeats in the C-terminal fragment in pan1– 11 probably changed the interaction of Pan1p with other protein(s) and blocked its function in growth on oleate, even with an intact N-terminus. Suppression of the Oleate phenotype of pan1– 11 by an additional copy of the END3 gene indicated that End3p– Pan1p interaction is important for growth on oleate, and End3p by itself might also be required for this function. We therefore compared growth of the end3-1 mutant and the
Fig. 3. Scheme of Pan1p and its deletion variants. Pan1p contains N-terminal repeats (white), EH domains (shadowed), QPTQPV repeats (striped), a prolinerich sequence (P, dark gray), an acidic fragment (A, dotted) and three putative nuclear localization sequences (NLS, black). Truncated proteins expressed in pan1 mutants or from plasmids are shown (black bars). Complementation of growth phenotypes of pan1 – 9 and pan1 – 11 mutants by plasmids expressing truncated Pan1p is indicated by ‘‘+’’ (growth) or ‘‘ ’’ (no growth).
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
Fig. 4. The pan1 – 9 strain is not defective, and end3-1 is defective in growth on oleate medium. Growth of the wild type, pan1 – 9 and end3-1 strains on YPD and YPO at 30-C.
wild type parent on YPD and YPO media. The end3-1 mutant exhibited the ts phenotype as previously reported [14] and grew more slowly on YPO medium than the parental strain (Fig. 4). Thus, not only Pan1p but also End3p, another cytoskeletal protein, is required for growth on oleate. The Oleate phenotype of pan1– 11 mutant does not result from peroxisomal defect but from a defect in apical growth The inability to utilize oleic acid may result from a lack of peroxisome function, defective peroxisome biogenesis or defects in the synthesis and transport of peroxisomal proteins. Three experiments were performed to test whether oleate induction of peroxisomal functions is sufficient in the pan1 – 11 mutant. The activation of reporter lacZ gene expression from the PCTA1 promoter under peroxisome-inducing conditions (SCDEO) was used to measure the transcriptional activation of genes encoding peroxisomal proteins. The CTA1 promoter contains the oleate response element (ORE) and the Adr1p transcription factor-binding site which together enable oleic acid induction of transcription from this promoter [33]. The PCTA1 -lacZ construct was integrated into the ura3 locus of the pan1– 11 mutant (strain JK48-3C). The resulting strain (JK51) was also transformed with plasmid YEp351-PAN1 which complements the pan1 –11 mutation or with vector alone. The transformants were grown in SCDEO medium or in SC and SCDE for comparison, total protein extracts were prepared, and the activity of hgalactosidase was measured. A slightly lower (by 20 – 30%) level of h-galactosidase activity was observed in strain pan1 –11 [vector] compared to pan1 – 11 [PAN1] on every medium tested (Table 1), but both strains responded properly to the carbon source. The reporter gene was repressed on glucose, derepressed on ethanol and induced on oleate. Additionally, parental WS-K strain and the pan1 –11 WS-6 mutant were grown in a peroxisome-inducing medium (YPEO), and the distribution of peroxisomal enzymes, acyl-CoA oxidase and catalase A was assayed in the cytoplasmic and organellar fractions. In the pan1 – 11 mutant,
487
as in the parental strain, the two peroxisomal marker enzymes were found in the organellar fraction (data not shown). To investigate further the biogenesis of peroxisomes, their number, size and inheritance were tested in pan1– 9 and pan1 –11 mutant strains transformed with plasmidbearing PAN1 or with empty vector, and additionally with the multicopy pJR322 plasmid expressing green fluorescent protein (GFP) fused C-terminally to the peroxisome targeting signal type 1 (PTS1) motif and/or the integrative plasmid pHPR131 expressing PTS2-DsRed fusion. Transformants grown in synthetic medium containing glucose were observed for fluorescence. One to five small brightly fluorescing spots corresponding to peroxisomes were present in pan1 – 9 and pan1 – 11 cells, irrespective of the pan1 mutations being complemented with the PAN1 gene or not (Fig. 5A). Green (PTS1-GFP) and red (PTS2-DsRed) fluorescing spots colocalized in the pan1 – 11 mutant, indicating that two groups of peroxisomal proteins, depending on the PTS1 or the PTS2 signal for transport, should be properly localized in peroxisomes (Fig. 5A). To evaluate the inheritance of peroxisomes in pan1 strains, budding cells were scored for the presence of fluorescing spots in the bud. In all strains, pan1 – 9 [vector], pan1 – 9 [PAN1], pan1– 11 [vector] and pan1 –11 [PAN1], about 65% of buds contained peroxisomes, indicating that peroxisome inheritance was not disturbed in pan1 –11. When the same strains were grown on SCEO medium, 1 –10 peroxisomes were observed in all cells. Surprisingly, pan1– 11 cells grown on this medium had a striking morphology, buds of most cells were elongated, indicating that the cells were arrested in the apical growth phase (Fig. 5B). This phenotype was reversed by plasmid-expressed PAN1, the PAN1 fragment encoding Pan1p1 – 973 or END3 (Fig. 5B and not shown). These results show that peroxisomal proteins are synthesized and localized properly and that neither a defect in peroxisomal enzymes nor in the biogenesis, morphology or inheritance of peroxisomes is responsible for the lack of growth of pan1 – 11 mutant on oleate. Instead, defective is the Pan1p function necessary for the regulation of apical cell growth when oleic acid is present in the medium. The pan1– 11 mutant is defective in growth on glycerol and succinate media Since peroxisomes were normally induced by oleic acid in pan1– 11, we considered the possibility that the pan1– 11 Table 1 The pan1 – 11 mutation does not affect expression of the CTA1 gene Strain
h-gal activity (nmol/min/mg of protein) Glucose
pan1 – 11 ura3::PCTA1 -lacZ (vector) pan1 – 11 ura3::PCTA1 -lacZ (PAN1)
Ethanol
Oleate
2.6
94
996
3.4
140
1269
Representative result of three independent experiments is shown.
488
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
Fig. 5. Biogenesis of peroxisomes is normal in pan1 – 9 and pan1 – 11 mutants, but apical growth is defective in pan1 – 11 grown on oleate medium. (A) pan1 – 9 [vector] and pan1 – 9 [PAN1], both expressing PTS1-GFP from a plasmid, and pan1 – 11 [vector] and pan1 – 11 [PAN1], both expressing PTS1-GFP and PTS2-DsRed, were grown at 30-C in SC medium supplemented with required amino acids and viewed for fluorescence and by DIC optics. (B) pan1 – 11 [vector] and pan1 – 11 [PAN1], both expressing PTS1-GFP, were grown on SCEO supplemented with required amino acids and viewed for fluorescence and by DIC optics.
mutation affects some aspects of cellular metabolic adaptation involved with utilization of nonfermentable carbon sources. Hence, the pan1 –9 and pan1– 11 mutants and their parental strains were spotted on media containing glycerol, acetate, ethanol, pyruvate or succinate as carbon sources. As shown in Fig. 6, pan1 –11 mutant grew significantly slower than the parental strain on glycerol- or succinate-containing media. On the same plates, pan1 – 9 grew similarly or better (succinate) than the parental strain. There were no differences in growth on media containing ethanol, acetate or pyruvate among the two pan1 mutants and the parental strains. However, growth of pan1 – 11 on an ethanol medium was compromised when oleic acid was added (Fig. 6). The pan1 –11 cells remained inhibited but fully viable even after a 48-h incubation in YPEO (data not presented). On the other hand, there was no growth effect upon oleate addition to a glucose medium, indicating that glucose represses the inhibitory effect of oleate (Fig. 6). The pan1 – 11 strain grown on glycerol or succinate medium accumulated cells with abnormal morphology similar to that observed on oleate medium. This phenotype was reversed by PAN1 expressed from the plasmid (not shown). These
results demonstrate that mitochondrial respiration functions properly in the pan1– 11 mutant, and the defect in growth is not restricted to oleate and must be associated with carbon source-dependent regulation of apical cell growth. The pan1– 11 mutant is defective in actin cytoskeleton, fluid phase endocytosis and in cell wall biogenesis The pan1 – 9 mutant cells were characterized previously as defective in fluid phase endocytosis [11] and having an abnormal actin cytoskeleton organization [26]. In addition to regular actin patches and actin cables, the cells displayed abnormal clumps of polymerized actin. Here, we investigated whether deletion of part of the (QPTQPV)7 motif in mutant Pan1 –11p also affected actin cytoskeleton organization which could be responsible for the observed morphological changes. For actin cytoskeleton staining, the pan1 –11 [PAN1] and pan1 –11 [vector] mutant strains were grown overnight in SC, SCO or SCG. The actin cytoskeleton was stained with Oregon Green-conjugated phalloidin. There was no difference in actin cytoskeleton organization between pan1 –11 and wild type cells grown in
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
489
Fig. 8. The pan1 – 11 mutant is defective in fluid phase endocytosis. The Lucifer yellow endocytosis assay was performed on the wild type, on pan1 – 11 [vector] and pan1 – 11 [PAN1]. The same fields were viewed for fluorescence and with DIC optics.
Fig. 6. pan1 – 11 cells are defective in growth on glycerol and succinate media. Growth of wild type, pan1 – 9 and pan1 – 11 strains on the indicated media after incubation at 30-C for 3 – 4 days.
glucose medium. All dividing cells had a polarized actin cytoskeleton with actin patches concentrated in the bud and actin cables distributed along the mother-bud axis; nondividing cells exhibited evenly distributed actin patches in the cell cortex (not shown). Actin cytoskeleton polarization could not be assessed in oleate-grown cells due to high background staining; actin cables were not visible even in wild type cells. To avoid this technical problem, cells grown in glycerol medium were analyzed. Actin cytoskeleton staining of pan1-1 [vector] mutant cells grown in this medium revealed non-polarized actin patches and lack of cables (Fig. 7), indicating defects of actin cytoskeleton
polarization. Well-polarized actin cytoskeleton in pan1– 11 [PAN1] cells was observed. Fluid phase endocytosis was investigated in pan1– 11 [vector], pan1 –11 [PAN1] and wild strain by testing the ability of cells pregrown in SC medium to take up the fluorescent, water-soluble dye, Lucifer yellow (LY). After 1.5 h of incubation with LY, the wild type cells accumulated the dye in the vacuole, whereas pan1– 11 mutant cells did not. The defect in endocytosis of pan1– 11 was complemented by wild type copy of PAN1 on a plasmid, as shown by the accumulation of LY in the internal structures, vacuoles and endosomes in the transformants (Fig. 8). This indicates that pan1 – 11 mutant is defective in endocytosis, similarly to the previously described pan1 mutants. Analysis of the phenotypes of pan1 – 9 and pan1– 11 indicates that the defective growth of pan1– 11 mutant on oleate might be due to disturbed
Fig. 7. Actin cytoskeleton of pan1 – 11 cells grown on glycerol medium is not polarized. The pan1 – 9 [vector] and pan1 – 9 [PAN1] cells were grown on SCG supplemented with required aa to the logarithmic phase and fixed. Filamentous actin (F-actin) was stained with Oregon Green-conjugated phalloidin. Cells were viewed for fluorescence.
490
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
functioning of Pan1p in actin cytoskeleton organization and endocytosis. It was previously documented that the pan1 –4 mutant, similarly to several other cytoskeletal mutants (act1-1, sla1D, end3D), showed cell wall abnormalities [12]. In dividing mutant cells, the cell wall of the mother was multilayered and much thicker than that of the bud, while wild type cells displayed a single layer of cell wall throughout the cell cycle, with no apparent differences between the mother and the bud. The presence of multicopy END3 suppressed the defects in actin cytoskeleton organization but not the cell wall defects of pan1 –4 [12]. Electron microscopy was used to investigate cell wall morphology in pan1 – 9 and pan1– 11 and wild type cells grown on YPD at 30-C. All the wild type cells displayed a single layer of cell wall (Fig. 9A), whereas pan1 – 9 and pan1 – 11 cells exhibited strikingly aberrant cell wall morphologies. The cell wall of the mother cell was thicker than that of the bud in 74% of dividing cells of the pan1 –9 strain. As shown in Fig. 9B, two layers of cell wall were often separated by the dark layer of mannoproteins. In pan1– 11 strain, 83% of dividing cells showed a much thicker cell wall in the mother cell than in the bud (Fig. 9C). The structure of the thick cell wall was uniform, with no visible layers. In some cells, wall fragments were torn apart (Fig. 9D). In addition, many dividing pan1– 11 cells had two buds. Both mutants displayed abnormal cell wall morphology, but the pan1 –11 defect seems more severe, and it may contribute to the lack of growth on oleate medium. Actin cytoskeleton, endocytosis and cell wall morphogenesis are all important for polarized growth of cells.
Fig. 9. The pan1 – 9 and pan1 – 11 mutants are defective in cell wall structure. Wild type (A), pan1 – 9 (B) and pan1 – 11 (C, D) mutants grown on YPD at 30-C were analyzed by electron microscopy. Scale bar represents 1 Am.
Apparently, Pan1p as multivalent regulator of these processes is important for carbon source regulation of the apical cell growth.
Discussion In order to utilize fatty acids, yeast require a functional peroxisomal compartment containing h-oxidation enzymes. The end products of h-oxidation are converted into intermediate metabolites which are shuttled from peroxisomes via the cytosol to mitochondria for further oxidation. Although peroxisomal processes are quite well known, the relevant cytosolic processes, the mitochondrial transport of metabolites and the shuttles involved in fatty acid utilization are still being studied [9,34]. In this report, we provide the first data documenting that actin cytoskeleton-associated proteins Pan1p and End3p are required not only for growth on oleate media but also on glycerol- or succinatecontaining media. Pan1p does not affect peroxisomal enzyme synthesis or peroxisome biogenesis but rather participate in regulation of cell growth in response to carbon source in the medium. The isolated pan1– 11 mutant displayed the Oleate phenotype, which could be suppressed by overexpression of the END3 gene. The pan1 – 11 mutant shared some of the characteristics of other pan1 mutants, like being defective in endocytosis and cell wall biogenesis. Indications regarding the mechanism which may result in the Oleate phenotype came from comparison of pan1 – 11 with the pan1 – 9 mutant isolated previously [11]. The pan1 – 9 mutant, although defective in endocytosis, cytoskeleton organization and cell wall biogenesis, was able to grow on oleate medium. Thus, the Oleate phenotype of pan1 –11 is not the result of defective oleic acid uptake or cellular transport caused by defective cell wall or endocytosis. Moreover, pan1– 11 also exhibited a normal shape and number of peroxisomes as well as correct inducibility of peroxisomal enzymes. Molecular analysis of the pan1 –9 and pan1 – 11 alleles revealed deletion of one base pair in the former, resulting in premature termination after codon 859 and a 64-base-pair deletion in the latter, resulting in the synthesis of Pan1p lacking three of the seven QPTQPV repeats. Surprisingly, although the pan1– 11 deletion was localized in the region untranslated in pan1 –9, pan1– 9, in contrast to pan1– 11, grew well on oleate. One possible explanation of these phenotypes is that the QPTQPV repeats bind some protein(s) which affects the binding of other proteins at the N-terminus of Pan1p. The latter is important for regulation of Pan1p function, and deletion of some QPTQPV repeats would disturb this regulation by interfering with the exchange of binding partners. When the whole C-terminal region of Pan1p is missing, the regulation does not operate and the binding of the protein partners to the N-terminus is simplified. The hypothesis is supported by the finding that
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
overexpression of End3p, a protein which binds to the Nterminus of Pan1p [35], improves pan1 –11 growth on oleate. Pan1p binds many proteins [12,18,32], but it is not known which protein binds to the (QPTQPV)7 repeats. Recently, intramolecular regulation has been suggested in Pan1p that depends on the presence of poly-proline-rich domain at the C-terminus which cooperates with more Nterminal regions [36]. Thus, an alternative possibility is that the (QPTQPV)7 repeats contribute to the intramolecular regulation in the presence of the most distant C-part, which is not necessary, or is impossible, in the absence of the latter. In the course of our studies, we found that complete (QPTQPV)7 repeats of Pan1p are important for regulation of apical growth of cells when oleate, glycerol or succinate is present in the medium. In S. cerevisiae, the formation of polarized cellular structures is required for cell division by budding and for mating. The cell forms the bud at the surface and segregates DNA and other cellular components into the bud to form a new cell. Bud growth has two phases, apical growth and isotropic growth. The yeast cyclindependent kinase Cdc28p regulates bud morphogenesis and cell cycle progression via the antagonistic activities of Cln and Clb cyclins. Cln G1 cyclins direct polarized growth and bud emergence, whereas Clb G2 cyclins promote isotropic growth of the bud and chromosome segregation [37]. The pan1 – 4 allele was isolated as causing lethality at 37-C with the cdc28-4 allele [19]. Another pan1 mutant allele was isolated that suppressed at 37-C the elongated bud phenotype of the cdc34-2 mutant failing to degrade G1 cyclins [38]. This allele expressed the Pan1p of 869 amino acids, similar to Pan1 –9p. Thus, the PAN1 gene is important for apical growth of cells at an elevated temperature on glucose medium. Our results indicate that some fragments of Pan1p are also specifically involved in certain nonfermentative carbon source regulation of apical growth. Further experiments are required to elucidate the precise mechanism by which Pan1p responds to carbon source to regulate apical growth of yeast.
[14]
Acknowledgments
[15]
We would like to thank A. Sachs, H. Riezman and M. Skoneczny for plasmids and antibodies and R. Olkowski for technical assistance. This work was supported by grant 3P04B01624 from the State Committee for Scientific ˙ Research, Poland to T. Z.
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[16] [17]
[18]
References [19] [1] P.B. Lazarow, W.H. Kunau, Peroxisomes, in: J.R. Pringle, J.R. Broach, E.W. Jones (Eds.), The Molecular and Cellular Biology of the Yeast Saccharomyces, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1997, pp. 547 – 605. [2] D.R. Johnson, L.J. Knoll, D.E. Levin, J.I. Gordon, Saccharomyces cerevisiae contains four fatty acid activation (FAA) genes: an assess-
[20]
[21]
491
ment of their role in regulating protein N-myristoylation and cellular lipid metabolism, J. Cell Biol. 127 (1994) 751 – 762. E.H. Hettema, C.W. van Roermund, B. Distel, M. van den Berg, C. Vilela, C. Rodrigues-Pousada, R.J. Wanders, H.F. Tabak, The ABC transporter proteins Pat1 and Pat2 are required for import of longchain fatty acids into peroxisomes of Saccharomyces cerevisiae, EMBO J. 15 (1996) 3813 – 3822. E.H. Hettema, H.F. Tabak, Transport of fatty acids and metabolites across the peroxisomal membrane, Biochim. Biophys. Acta 1486 (2000) 18 – 27. H.J. Schuller, Transcriptional control of nonfermentative metabolism in the yeast Saccharomyces cerevisiae, Curr. Genet. 43 (2003) 139 – 160. A.S. Tibbetts, Y. Sun, N.A. Lyon, A.C. Ghrist, P.J. Trotter, Yeast mitochondrial oxodicarboxylate transporters are important for growth on oleic acid, Arch. Biochem. Biophys. 406 (2002) 96 – 104. P.J. Trotter, R.A. Hagerman, D.R. Voelker, A yeast strain defective in oleic acid utilization has a mutation in the RML2 gene, Biochim. Biophys. Acta 1438 (1999) 223 – 238. K.I. Minard, G.T. Jennings, T.M. Loftus, D. Xuan, L. McAlister-Henn, Sources of NADPH and expression of mammalian NADP+-specific isocitrate dehydrogenases in Saccharomyces cerevisiae, J. Biol. Chem. 273 (1998) 31486 – 31493. S. Cavero, A. Vozza, A. del Arco, L. Palmieri, A. Villa, E. Blanco, M.J. Runswick, J.E. Walker, S. Cerdan, F. Palmieri, J. Satrustegui, Identification and metabolic role of the mitochondrial aspartate – glutamate transporter in Saccharomyces cerevisiae, Mol. Microbiol. 50 (2003) 1257 – 1269. I. Smaczyn˜ska-de Rooij, A. Migdalski, J. Rytka, a-Ketoglutarate dehydrogenase and lipoic acid synthase are important for the functioning of peroxisomes of Saccharomyces cerevisiae, Cell. Mol. Biol. Lett. 9 (2004) 271 – 286. T.¯o31dek, G. Vaduva, L.A. Hunter, M. Boguta, B.D. Go, N.C. Martin, A.K. Hopper, Mutations altering the mitochondrial – cytoplasmic distribution of Mod5p implicate the actin cytoskeleton and mRNA 3V ends and/or protein synthesis in mitochondrial delivery, Mol. Cell. Biol. 15 (1995) 6884 – 6894. H.Y. Tang, J. Xu, M. Cai, Pan1p, End3p, and S1a1p, three yeast proteins required for normal cortical actin cytoskeleton organization, associate with each other and play essential roles in cell wall morphogenesis, Mol. Cell. Biol. 20 (2000) 12 – 25. I. Smaczyn˜ska de Rooij, Saccharomyces cerevisiae mutants altered in the expression of peroxisomal proteins; isolation and characterization, PhD thesis, Institute of Biochemistry and Biophysics PAS, Warsaw, Poland, 2001. S. Raths, J. Rohrer, F. Crausaz, H. Riezman, end3 and end4: two mutants defective in receptor-mediated and fluid-phase endocytosis in Saccharomyces cerevisiae, J. Cell Biol. 120 (1993) 55 – 65. P. Chen, P. Johnson, T. Sommer, S. Jentsch, M. Hochstrasser, Multiple ubiquitin-conjugating enzymes participate in the in vivo degradation of the yeast MAT alpha 2 repressor, Cell 74 (1993) 357 – 369. F. Sherman, Getting started with yeast, Methods Enzymol. 350 (2002) 3 – 41. C. De Virgilio, N. Burckert, G. Barth, J.M. Neuhaus, T. Boller, A. Wiemken, Cloning and disruption of a gene required for growth on acetate but not on ethanol: the acetyl-coenzyme A synthetase gene of Saccharomyces cerevisiae, Yeast 8 (1992) 1043 – 1051. B. Wendland, S.D. Emr, Pan1p, yeast eps15, functions as a multivalent adaptor that coordinates protein – protein interactions essential for endocytosis, J. Cell Biol. 141 (1998) 71 – 84. H.Y. Tang, M. Cai, The EH-domain-containing protein Pan1 is required for normal organization of the actin cytoskeleton in Saccharomyces cerevisiae, Mol. Cell. Biol. 16 (1996) 4897 – 4914. C. Brocard, G. Lametschwandtner, R. Koudelka, A. Hartig, Pex14p is a member of the protein linkage map of Pex5p, EMBO J. 16 (1997) 5491 – 5500. K. Stein, A. Schell-Steven, R. Erdmann, H. Rottensteiner, Interactions
492
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29] [30]
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492 of Pex7p and Pex18p/Pex21p with the peroxisomal docking machinery: implications for the first steps in PTS2 protein import, Mol. Cell Biol. 22 (2002) 6056 – 6069. M. Go´ra, K. Pluta, A. Che3stowska, T. ¯o31dek, Suppressors of translation initiation defect in hem12 locus of Saccharomyces cerevisiae, Acta Biochim. Pol. 47 (2000) 181 – 190. R. Rothstein, Targeting, disruption, replacement, and allele rescue: integrative DNA transformation in yeast, Methods Enzymol. 194 (1991) 281 – 301. M.D. Rose, F. Winston, P. Hieter, Methods in yeast genetics, A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 2004, pp. 126 – 127. L. Guarente, Yeast promoters and lacZ fusions designed to study expression of cloned genes in yeast, Methods Enzymol. 101 (1983) 181 – 191. J. Kamin˜ska, B. Gajewska, A.K. Hopper, T. ¯o31dek, Rsp5p, a new link between the actin cytoskeleton and endocytosis in the yeast Saccharomyces cerevisiae, Mol. Cell. Biol. 22 (2002) 6946 – 6948. R. Boeck, S. Tarun Jr., M. Rieger, J.A. Deardorff, S. Muller-Auer, A.B. Sachs, The yeast Pan2 protein is required for poly(A)-binding protein-stimulated poly(A)-nuclease activity, J. Biol. Chem. 271 (1996) 432 – 438. A.E. Adams, J.R. Pringle, Staining of actin with fluorochromeconjugated phalloidin, Methods Enzymol. 194 (1991) 729 – 731. V. Dulic, M. Egerton, I. Elguindi, S. Raths, B. Singer, H. Riezman, Yeast endocytosis assays, Methods Enzymol. 194 (1991) 697 – 710. E.S. Reynolds, The use of lead citrate at high pH as an electronopaque stain in electron microscopy, J. Cell Biol. 17 (1963) 208 – 212.
[31] A.B. Sachs, J.A. Deardorff, Translation initiation requires the PABdependent poly(A) ribonuclease in yeast, Cell 70 (1992) 961 – 973. [32] M.C. Duncan, M.J. Cope, B.L. Goode, B. Wendland, D.G. Drubin, Yeast Eps15-like endocytic protein, Pan1p, activates the Arp2/3 complex, Nat. Cell Biol. 3 (2001) 687 – 690. [33] H. Rottensteiner, A.J. Kal, M. Filipits, M. Binder, B. Hamilton, H.F. Tabak, H. Ruis, Pip2p: a transcriptional regulator of peroxisome proliferation in the yeast Saccharomyces cerevisiae, EMBO J. 15 (1996) 2924 – 2934. [34] B.M. Bakker, K.M. Overkamp, A.J. van Maris, P. Kotter, M.A. Luttik, J.P. van Dijken, J.T. Pronk, Stoichiometry and compartmentation of NADH metabolism in Saccharomyces cerevisiae, FEMS Microbiol. Rev. 25 (2001) 15 – 37. [35] H.Y. Tang, A. Munn, M. Cai, EH domain proteins Pan1p and End3p are components of a complex that plays a dual role in organization of the cortical actin cytoskeleton and endocytosis in Saccharomyces cerevisiae, Mol. Cell. Biol. 17 (1997) 4294 – 4304. [36] N.B. Miliaras, J.H. Park, B. Wendland, The function of the endocytic scaffold protein Pan1p depends on multiple domains, Traffic 5 (2004) 963 – 978. [37] V.J. Cid, J. Jimenez, M. Molina, M. Sanchez, C. Nombela, J.W. Thorner, Orchestrating the cell cycle in yeast: sequential localization of key mitotic regulators at the spindle pole and the bud neck, Microbiology 148 (2002) 2647 – 2659. [38] S. Bidlingmaier, M. Snyder, Large-scale identification of genes important for apical growth in Saccharomyces cerevisiae by directed allele replacement technology (DART) screening, Funct. Integr. Genomics 1 (2002) 345 – 356.
Research Article
Pan1p, an actin cytoskeleton-associated protein, is required for growth of yeast on oleate medium Joanna Kamin´ska, Monika Wysocka-Kapcin´ska, Iwona Smaczyn´ska-de Rooij, Joanna Rytka, Teresa Z˙oaa˛dek* Department of Genetics, Institute of Biochemistry and Biophysics, Polish Academy of Sciences, Pawin´skiego 5a, 02-106 Warsaw, Poland Received 2 March 2005, revised version received 30 June 2005, accepted 30 August 2005 Available online 19 September 2005
Abstract Pan1p is a yeast actin cytoskeleton-associated protein localized in actin patches. It activates the Arp2/3 complex, which is necessary for actin polymerization and endocytosis. We isolated the pan1 – 11 yeast mutant unable to grow on oleate as a sole carbon source and, therefore, exhibiting the Oleate phenotype. In addition, mutant cells are temperature-sensitive and grow more slowly on glycerol or succinatecontaining medium but similarly to the wild type on ethanol, pyruvate or acetate-containing media; this indicates proper functioning of the mitochondrial respiratory chain. However, growth on ethanol medium is compromised when oleic acid is present. Cells show growth arrest in the apical growth phase, and accumulation of cells with abnormally elongated buds is observed. The growth defects of pan1 – 11 are suppressed by overexpression of the END3 gene encoding a protein that binds Pan1p. The morphology of peroxisomes and induction of peroxisomal enzymes are normal in pan1 – 11, indicating that the defect in growth on oleate medium does not result from impairment in peroxisome function. The pan1 – 11 allele has a deletion of a fragment encoding amino acids 1109 – 1126 that are part of (QPTQPV)7 repeats. Surprisingly, the independently isolated pan1 – 9 mutant, which expresses a truncated form of Pan1p comprising aa 1 – 859, is able to grow on all media tested. Our results indicate that Pan1p, and possibly other components of the actin cytoskeleton, are necessary to properly regulate growth of dividing cells in response to the presence of some alternative carbon sources in the medium. D 2005 Elsevier Inc. All rights reserved. Keywords: Yeast; Actin cytoskeleton; Peroxisomes; Oleic acid; Apical growth
Introduction Fatty acids are the building blocks of lipids that constitute cellular membranes, components of intracellular messengers (e.g. phosphatidylinositoles), and also modify proteins. Moreover, fatty acids are efficient fuel molecules, and yeast cells utilize exogenous free fatty acids as a source of biomass and energy. Oleic acid as a sole carbon source or added to a medium containing another nonfermentable carbon source, such as ethanol, represses genes of the fatty acid biosynthetic pathway and induces those encoding h-oxidation enzymes.
* Corresponding author. Fax: +48 22 658 46 36. E-mail address: [email protected] (T. Z˙oaa˛dek). 0014-4827/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.yexcr.2005.08.018
However, the signaling pathway for oleic acid sensing is unknown. In yeast, enzymes of the h-oxidation pathway are located in peroxisomes. Thus, biogenesis of peroxisomes is induced upon addition of oleic acid to the medium [1]. After uptake, oleic acid can be activated by at least four cytoplasmic acyl-CoA synthetases [2]. Acyl-CoA entry into peroxisomes is mediated by an ABC transporter [3]. The first step of h-oxidation is catalyzed by acyl-CoA oxidase, Pox1p, and is accompanied by reduction of molecular oxygen to hydrogen peroxide. The hydrogen peroxide is in turn decomposed by peroxisomal catalase A (Cta1p). Peroxisomes oxidize fatty acids to acetyl-CoA, and the reaction is accompanied by production of NADH. Further oxidation and energy production require fully functional mitochondria. Acetyl-CoA is transported to mitochondria via two alternative pathways: by acetyl carnitine transferases
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
or by glyoxylate cycle metabolites of acetyl-CoA, isocitrate and succinate [4]. It subsequently enters the citric acid cycle. NADH must also be shuttled to mitochondria. Growth on oleic acid, similarly as on other nonfermentable substrates, such as glycerol, ethanol or acetate, requires induction of gluconeogenesis genes (reviewed in [5]). Thus, coordinated expression of peroxisomal, mitochondrial and cytosolic enzymes is needed for utilization of oleic acid. Among yeast mutants defective in growth on oleic-acidcontaining media, there are many disturbed in peroxisome biogenesis or lacking peroxisomal enzymes (pas, fox, reviewed in [1]) as well as the following: odc1D odc2D which lacks mitochondrial oxodicarboxylate transporters [6]; rml2 with a defective mitoribosomal protein [7]; zwf1D idp2D which does not produce cytosolic NADPH [8]; and agc1D which lacks the mitochondrial aspartate – glutamate transporter involved in NADH shuttle [9]. Recently, we have found that kgd1 and lip5 mutations, which disturb aketoglutarate dehydrogenase, are impaired in oleate utilization due to low expression of peroxisomal enzymes in response to mitochondrial dysfunction [10]. All these mutants retain the ability to grow on ethanol, acetate or glycerol. In an effort to identify novel genes required for utilization of oleic acid, we isolated Oleate mutants defective in growth on media containing oleic acid as a sole carbon source while retaining the ability to grow on ethanol. We identified the PAN1 gene and END3, a multicopy suppressor of pan1, as required for growth on oleate. It had previously been shown that Pan1p is a multifunctional protein involved in endocytosis, actin cytoskeleton organization [11] and cell wall morphogenesis [12]. In this report, we show that expression of a DNA fragment encoding the first 859 amino acids (aa) of Pan1p is sufficient for growth of cells on oleate, although the fragment encoding (QPTQPV)7 repeats (aa 1084 –1125), if present, also plays an important role. The pan1 –11 mutant expressing the protein devoid of three of these repeats shows arrest in the apical growth phase when grown on oleate. Furthermore, glycerol and succinate in the medium inhibited the growth of this mutant. Presented results suggest that the Pan1p function in growth on oleic acid results from its role in the regulation of apical growth of the bud in response to carbon source.
Materials and methods Yeast strains, genetic methods, media and reagents Saccharomyces cerevisiae strains used were WS-K MATa leu2::P CTA1 -URA3::LEU2 arg1 ura3 trp1 [13], T81D, T37-2A (pan1– 8), TZ35 (pan1– 10), TZ81 (pan1– 9) [11], RH144-3D, RH266-1D (end3-1) [14] and MHY500 [15]. Strain WS-6 MATa leu2::P CTA1 -URA3::LEU2 pan1 – 11 arg1 ura3 trp1 was obtained by UV mutagenesis of WS-K. Strains JK48-3B MATa ura3-1 leu2 trp1 his3D-200
483
and JK48-3C MATa pan1– 11 ura3-1 leu2 trp1 lys2-801 were obtained from WS-6 MHY500 cross. Strain JK51 was obtained by integration of YIpCTA1-lacZ into ura3 locus of JK48-3C and strain MW20 by integration of pHPR131 into trp1 locus of JK48-3C. Standard genetic manipulations were performed as described by Sherman [16]. Yeast were grown on rich media containing as a carbon source 2% glucose (YPD); 2% ethanol (YPE); 2% glycerol (YPG); 0.1% glucose and 2% ethanol (YPDE); 0.25% oleic acid in Tween 20 (YPO); 2% ethanol and 0.25% oleic acid (YPEO); 0.1% glucose, 1% ethanol and 0.25% oleic acid in Tween 20 (YPDEO). The medium with acetate as a carbon source (YPA) was prepared as in [17]. Rich media with pyruvate (YPP) or succinate (YPS) as a carbon source contained 1% of the respective sodium salt. Alternatively, cells were grown on SC-ura, SC-trp or SC-leu selective media containing 2% glucose or SCG containing 2% glycerol, on SCO-trp or SCO-ura containing 0.25% oleic acid in Tween 20, on SCDE-ura-leu containing 0.1% glucose and 2% ethanol or on SCDEO-ura-leu containing 0.1% glucose, 2% ethanol and 0.25% oleic acid. The latter group of media were supplemented with the required aa, uracil or adenine. The growth of yeast strains on plates containing different carbon sources was compared by a drop test. Cells grown overnight were suspended at 1 107 cells/ml, serially diluted 1:10 and 3 Al of each dilution was spotted on the plate. Plates were incubated for 2– 4 days at the temperature indicated. Plasmids and plasmid constructions Plasmids used were: pT4 containing PAN1 gene, YIpT4 containing the EcoRI fragment of PAN1 [11], pHT837 containing PAN1 [19], pJR233 with the gene encoding green fluorescence protein (GFP) with a PTS1 (peroxisome targeting signal) at the C-terminus under the control of the MLS1 promoter [20], pHPR131 expressing PTS2-DsRed [21] and YIpCTA1-lacZ bearing PCTA1 -lacZ reporter (from M. Skoneczny, IBB PAS). Both pSR19 and pSR48 were isolated from a genomic library based on pRS314 (TRP1 marker; [22]) by complementation of the Oleate phenotype of pan1 – 11. pSR19 contains a 7.4 kb genomic fragment bearing END3, YNL083W and PMS1 genes. pSR48 contains a 6.3 kb genomic fragment bearing YIR007W and the first 2919 bp from the 5V-end of PAN1 open reading frame (orf). YEp351PAN1 bearing a genomic fragment containing PAN1 was isolated from a genomic library based on the YEp351 plasmid (LEU2 marker; M. Wysocka-Kapcin˜ska, unpublished) by complementation of the ts phenotype of pan1. YIpT4 was digested with StuI to integrate into the ura3 locus of JK48-3C and test for pan1 – 11 complementation. YIpT4DSnaBI was obtained by SnaBI digestion and religation of YIpT4 and used, after further SnaBI digestion,
484
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
for gap-repair mapping of pan1 mutations. YIpCTA1-lacZ was digested with StuI and integrated into the ura3 locus of JK48-3C strain.
a Hamamatsu ORCA 100 camera, and images were collected using Lucia G software. Actin cytoskeleton staining
Characterization of pan1 mutations The mutations pan1 – 8, – 9, –10 and –11 were mapped by gap-repair as described by Rothstein [23]. Total yeast DNA was isolated as in [24]. Fragments of pan1 alleles were amplified by PCR using specific primers. PCR products were sequenced in the DNA Sequencing and Oligonucleotide Synthesis Laboratory at IBB, PAS. The sequences obtained were compared to the sequences from Saccharomyces Genome Database using BLAST or SeqMan algorithms. Activity of b-galactosidase Yeast were grown in 5 ml YPD precultures and were inoculated into 20 ml of YPDEO. Cells grown overnight were collected, washed with water, and extracts were prepared with glass beads. Aliquots of total protein extracts were taken, and the protein concentration was estimated by the BioRad assay according to manufacturer’s instructions. The h-galactosidase activity was assayed spectrophotometrically at 420 nm using o-nitrophenyl-hd-galactopyranoside as substrate [25]. The activity was expressed in nanomole of o-nitrophenol/min/mg protein. In the experiments, five independent transformants were combined, inoculated, and their extracts were assayed in duplicates. Western blot analysis Protein extracts were analyzed by a standard Western blotting method [26] with signal development by chemiluminescence (ECL Plus, Amersham). Antibodies used were affinity-purified rabbit anti-N-term-Pan1p and anti-Cterm-Pan1p antibodies (PAN91 and PAN92, kind gift from A. Sachs, University of California at Berkeley [18,27]) and secondary anti-rabbit horseradish peroxidase-conjugated antibody (DACO).
For actin cytoskeleton staining, 10 ml cultures in SC, SCO or SCG were grown overnight to OD600 ¨0.5. Actin was stained as described [28], with some modifications. Cells were fixed for 2 h by adding to the culture concentrated formaldehyde solution to a final concentration of 3.7% and adjusting the pH to 6.5 with potassium phosphate buffer. The cells were subsequently collected by centrifugation, washed three times in phosphate-buffered saline (PBS), pH 6.5 and stained with Oregon Green 488conjugated phalloidin (Molecular Probes) for 2 h in the dark. Cells were washed five times with PBS, stained with DAPI (0.5 Ag/ml), washed twice and observed in Eclipse, a fluorescence microscope. Endocytosis assay Fluid phase endocytosis assays were carried out as described by Dulic et al. [29]. Strain pan1– 11 (JK48-3C)bearing vector or YEp351-PAN1 plasmid were grown in SC medium to OD600 of 0.5 – 0.8. About 107 cells were collected and incubated with Lucifer yellow (LY; Sigma) for 1.5 h at 30-C. An Eclipse fluorescence microscope was used to observe LY accumulation in the vacuole, and vacuolar morphology was viewed with DIC optics. Transmission electron microscopy Cells grown on YPD to the logarithmic phase were fixed with 2% glutaraldehyde followed by 2% OsO4 for 2 h. Thin sections were cut from Epon blocks and post-stained with 2% uranyl acetate followed by lead citrate [30]. Micrographs were taken on a JEOL transmission electron microscope JEM1220.
Results
Fluorescence microscopy
Isolation and genetic analysis of mutants that do not grow on media containing oleic acid as a sole carbon source
To assess inheritance of peroxisomes, respective transformants expressing PTS1-GFP or PTS2-DsRed were grown on SC-ura or SC-trp to the logarithmic phase, OD600 ¨1, and at least 120 budding cells of each strain were scored for the presence in the bud of a fluorescing spot, corresponding to a peroxisome. To observe peroxisomes on oleic-acid-containing medium, cells expressing PTS1-GFP were grown on SCE, transferred to SCEO and incubated o/n. Cells were viewed by fluorescence microscopy and with differential interference contrast (DIC) optics using an Eclipse (Nikon) microscope equipped with
To identify new factors necessary for growth on oleate, mutants were isolated which were unable to grow on oleicacid-containing medium (Oleate ) while retaining the ability to utilize ethanol. Standard UV mutagenesis of the WS-K strain, with 10% survival rate, was carried out. The UV-treated cells were plated on YPDE plates, grown for 3 days and then replica-plated on YPO medium containing oleic acid as a sole carbon source. One of the mutants, unable to grow on oleate medium, named WS-6, was further characterized. WS-6 was crossed with wild type strain MHY500, and the diploid displayed wild-type phenotype.
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
Analysis of the haploid progeny of this diploid revealed 2:2 segregation of the wild type: Oleate phenotype indicating that the loss of the ability to grow on oleate as a sole carbon source resulted from a single recessive mutation. Additionally, the mutation caused a temperature-sensitive phenotype (ts) as the mutant strain was not able to grow on YPD at 37-C. The ts phenotype segregated together with the Oleate phenotype in 30 tetrads tested. To identify the mutated gene, the WS-6 mutant strain was transformed with a yeast genomic library based on single-copy pRS314 vector with TRP1 marker [22]. The Trp+ colonies were replicaplated on synthetic medium containing oleic acid as a sole carbon source (SCO-trp) and incubated at 30-C for 5 days. From the 10,000 Trp+ transformants tested, two colonies grew on these plates. Standard isolation and retransformation of the recovered plasmids followed by phenotype testing and determination of DNA sequence were carried out. Plasmids pSR19 and pSR48 restored growth of WS-6 cells on SCO-trp medium at 30-C and on YPD medium at 36-C, albeit to a different extent (Fig. 1, see Fig. 3). Plasmid pSR19 contained the END3, YNL083W and PMS1 genes, and plasmid pSR48 contained YIR007W and the first 2919 base pairs from the 5V-end of PAN1 orf. The END3 and PAN1 genes were functionally related, both being involved in endocytosis and actin cytoskeleton organization. Furthermore, End3p interacted with Pan1p in vivo, and pan1– 4 could be suppressed by END3 overexpression [12]. There-
Fig. 1. The Oleate phenotype of pan1 – 11 is suppressed by END3 gene overexpression. Growth of wild type strain transformed with vector and pan1 – 11 strain transformed with vector, plasmid pSR19 or pSR48, bearing END3 or fragment of PAN1 respectively. Media and incubation temperatures are indicated.
485
fore, it was quite possible that the Oleate phenotype resulted from mutation in one of these genes. To determine whether the WS-6 strain bears the mutation in the END3 or PAN1 gene, a complementation test was performed. WS-6 was crossed with two temperature-sensitive mutants, pan1– 8 (T37-2A) and end3-1 (RH266-1D). The resulting diploids were grown on YPD medium at 30-C, replica-plated on YPD and incubated at 37-C. In contrast to the end3-1 mutation, pan1 –8 did not complement the ts phenotype of WS-6 strain. Hence, the WS-6 strain bears a mutation in the PAN1 gene, and END3 is a suppressor gene. This mutation, referred to as pan1– 11, was also complemented by the YEp351-PAN1 plasmid, bearing a complete PAN1 gene, and by plasmid YIpT4, bearing the EcoRI fragment of PAN1, integrated into the ura3 locus (see Fig. 3). Complementation of pan1 – 11 mutation by incomplete PAN1 orf is consistent with the previous finding that only part of the gene is necessary for its essential function(s) [11,31]. These results indicate that Pan1p and End3p are required not only for endocytosis and actin cytoskeleton function but also for growth of cells on a medium containing oleic acid as a sole carbon source. Characterization of pan1 mutant alleles and their products The pan1 –8, – 9 and – 10 mutants were isolated in an earlier screen for mutants affecting cytoplasmic – mitochondrial distribution of Mod5-Ip, a tRNA-modifying enzyme [11]. These strains were investigated, together with pan1– 11, in subsequent studies in which pan1 alleles and their products were characterized. First, total protein extracts prepared from pan1 – 8, –9, –10 and – 11 mutant strains and from parental strains were separated by SDS-PAGE and analyzed by Western blotting with anti-Pan1p antibodies. Two types of antibodies were used, one raised against the second N-terminal repeat of Pan1p, aa 392– 662, (anti-Nterm-Pan1p), and the other against aa 1059 –1113 containing QPTQPV repeats of Pan1p (anti-C-term-Pan1p). The anti-N-term-Pan1p antibody recognized a protein of about 160 kDa in parental strains and in pan1– 11 extract as well as a shorter form of Pan1p, of about 100 kDa, in pan1– 8, – 9 and – 10 extracts. The anti-C-term-Pan1p antibody recognized Pan1p in the wild type strains, weakly in pan1 – 11 but not in the pan1 – 9 mutant (Fig. 2). These results confirmed the predicted size of wild type Pan1p, indicated the size of mutant Pan1 proteins and also documented that the fragment of Pan1p containing the QPTQPV repeats is absent from all the mutants or is not recognized by the anti-C-term-Pan1p antibody. The pan1 mutations were subsequently mapped in PAN1 gene by the gap-repair method. The mutations were identified in pan1 – 8, – 9, – 10 and – 11 alleles by sequencing PCR products of reactions performed with genomic DNA from the respective strains as templates. Surprisingly, the same mutational change, a deletion of nucleotide A2582 from PAN1 orf, was found in pan1 – 8, – 9
486
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
QPTQPV repeats, polyproline sequences and the acidic tail binding to the Arp2/3 complex involved in actin polymerization [32] (Fig. 3). Molecular characterization of the pan1– 11 allele revealed a 54 base pair deletion from nucleotide 3325 to 3378 of PAN1 orf, codons 1109 –1126. This fragment encodes three QPTQPV repeats which are normally repeated seven times in wild type Pan1p. The pan1 – 9 mutant grows on oleic-acid-containing medium, whereas end3-1 is defective
Fig. 2. Characterization of Pan1 proteins synthesized by pan – 8, – 9, – 10 and – 11 mutant strains. Western blot of total protein extracts from pan1 mutant strains grown at 30-C is shown.
and – 10 alleles, even though they were isolated independently. This deletion caused frameshift and generated an adjacent stop codon. Thus, the pan1– 8, – 9 and –10 alleles encoded mutant Pan1p that was shorter than the wild type Pan1p, 859 aa versus 1480 aa, with the last four aa differing from the corresponding ones of the wild type. Pan1p1 – 859 retains the N-terminal motifs homologous to Sla1p and also the EH domains responsible for End3p and Sla1p binding and linking Pan1p to endocytic machinery. It is missing the
The pan1– 9 mutant was chosen for further characterization as a representative of pan1 –8/9/10. The growth of pan1– 9 and pan1 –11 mutants and the respective parental strains was compared on YPO and YPD media at 30-C and on YPD at 37-C. Neither pan1 mutant grew on YPD at 37-C (see Fig. 1 and not shown). Surprisingly, and in contrast to pan1 –11, the pan1 – 9 mutant was able to grow on oleic-acid-containing media similarly to the wild type strain (Fig. 4, see Fig. 1). The ts phenotype of pan1 – 9 was suppressed by the plasmid bearing a PAN1 fragment encoding Pan1p1 – 973. Thus, the pan1 mutants studied exhibited the ts phenotype, but the growth defect on oleate characterized only one of them. Interestingly, Pan1p1 – 973 was sufficient for growth at an elevated temperature, indicating that aa 860 –973 were critical for this function. Moreover, the C-terminal fragment of Pan1p, aa 860– 1480, was not absolutely required for growth on oleic acid since the pan1 –9 mutant expressing Pan1p1 – 859 did not exhibit the Oleate phenotype. However, the deletion of three QPTQPV repeats in the C-terminal fragment in pan1– 11 probably changed the interaction of Pan1p with other protein(s) and blocked its function in growth on oleate, even with an intact N-terminus. Suppression of the Oleate phenotype of pan1– 11 by an additional copy of the END3 gene indicated that End3p– Pan1p interaction is important for growth on oleate, and End3p by itself might also be required for this function. We therefore compared growth of the end3-1 mutant and the
Fig. 3. Scheme of Pan1p and its deletion variants. Pan1p contains N-terminal repeats (white), EH domains (shadowed), QPTQPV repeats (striped), a prolinerich sequence (P, dark gray), an acidic fragment (A, dotted) and three putative nuclear localization sequences (NLS, black). Truncated proteins expressed in pan1 mutants or from plasmids are shown (black bars). Complementation of growth phenotypes of pan1 – 9 and pan1 – 11 mutants by plasmids expressing truncated Pan1p is indicated by ‘‘+’’ (growth) or ‘‘ ’’ (no growth).
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
Fig. 4. The pan1 – 9 strain is not defective, and end3-1 is defective in growth on oleate medium. Growth of the wild type, pan1 – 9 and end3-1 strains on YPD and YPO at 30-C.
wild type parent on YPD and YPO media. The end3-1 mutant exhibited the ts phenotype as previously reported [14] and grew more slowly on YPO medium than the parental strain (Fig. 4). Thus, not only Pan1p but also End3p, another cytoskeletal protein, is required for growth on oleate. The Oleate phenotype of pan1– 11 mutant does not result from peroxisomal defect but from a defect in apical growth The inability to utilize oleic acid may result from a lack of peroxisome function, defective peroxisome biogenesis or defects in the synthesis and transport of peroxisomal proteins. Three experiments were performed to test whether oleate induction of peroxisomal functions is sufficient in the pan1 – 11 mutant. The activation of reporter lacZ gene expression from the PCTA1 promoter under peroxisome-inducing conditions (SCDEO) was used to measure the transcriptional activation of genes encoding peroxisomal proteins. The CTA1 promoter contains the oleate response element (ORE) and the Adr1p transcription factor-binding site which together enable oleic acid induction of transcription from this promoter [33]. The PCTA1 -lacZ construct was integrated into the ura3 locus of the pan1– 11 mutant (strain JK48-3C). The resulting strain (JK51) was also transformed with plasmid YEp351-PAN1 which complements the pan1 –11 mutation or with vector alone. The transformants were grown in SCDEO medium or in SC and SCDE for comparison, total protein extracts were prepared, and the activity of hgalactosidase was measured. A slightly lower (by 20 – 30%) level of h-galactosidase activity was observed in strain pan1 –11 [vector] compared to pan1 – 11 [PAN1] on every medium tested (Table 1), but both strains responded properly to the carbon source. The reporter gene was repressed on glucose, derepressed on ethanol and induced on oleate. Additionally, parental WS-K strain and the pan1 –11 WS-6 mutant were grown in a peroxisome-inducing medium (YPEO), and the distribution of peroxisomal enzymes, acyl-CoA oxidase and catalase A was assayed in the cytoplasmic and organellar fractions. In the pan1 – 11 mutant,
487
as in the parental strain, the two peroxisomal marker enzymes were found in the organellar fraction (data not shown). To investigate further the biogenesis of peroxisomes, their number, size and inheritance were tested in pan1– 9 and pan1 –11 mutant strains transformed with plasmidbearing PAN1 or with empty vector, and additionally with the multicopy pJR322 plasmid expressing green fluorescent protein (GFP) fused C-terminally to the peroxisome targeting signal type 1 (PTS1) motif and/or the integrative plasmid pHPR131 expressing PTS2-DsRed fusion. Transformants grown in synthetic medium containing glucose were observed for fluorescence. One to five small brightly fluorescing spots corresponding to peroxisomes were present in pan1 – 9 and pan1 – 11 cells, irrespective of the pan1 mutations being complemented with the PAN1 gene or not (Fig. 5A). Green (PTS1-GFP) and red (PTS2-DsRed) fluorescing spots colocalized in the pan1 – 11 mutant, indicating that two groups of peroxisomal proteins, depending on the PTS1 or the PTS2 signal for transport, should be properly localized in peroxisomes (Fig. 5A). To evaluate the inheritance of peroxisomes in pan1 strains, budding cells were scored for the presence of fluorescing spots in the bud. In all strains, pan1 – 9 [vector], pan1 – 9 [PAN1], pan1– 11 [vector] and pan1 –11 [PAN1], about 65% of buds contained peroxisomes, indicating that peroxisome inheritance was not disturbed in pan1 –11. When the same strains were grown on SCEO medium, 1 –10 peroxisomes were observed in all cells. Surprisingly, pan1– 11 cells grown on this medium had a striking morphology, buds of most cells were elongated, indicating that the cells were arrested in the apical growth phase (Fig. 5B). This phenotype was reversed by plasmid-expressed PAN1, the PAN1 fragment encoding Pan1p1 – 973 or END3 (Fig. 5B and not shown). These results show that peroxisomal proteins are synthesized and localized properly and that neither a defect in peroxisomal enzymes nor in the biogenesis, morphology or inheritance of peroxisomes is responsible for the lack of growth of pan1 – 11 mutant on oleate. Instead, defective is the Pan1p function necessary for the regulation of apical cell growth when oleic acid is present in the medium. The pan1– 11 mutant is defective in growth on glycerol and succinate media Since peroxisomes were normally induced by oleic acid in pan1– 11, we considered the possibility that the pan1– 11 Table 1 The pan1 – 11 mutation does not affect expression of the CTA1 gene Strain
h-gal activity (nmol/min/mg of protein) Glucose
pan1 – 11 ura3::PCTA1 -lacZ (vector) pan1 – 11 ura3::PCTA1 -lacZ (PAN1)
Ethanol
Oleate
2.6
94
996
3.4
140
1269
Representative result of three independent experiments is shown.
488
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
Fig. 5. Biogenesis of peroxisomes is normal in pan1 – 9 and pan1 – 11 mutants, but apical growth is defective in pan1 – 11 grown on oleate medium. (A) pan1 – 9 [vector] and pan1 – 9 [PAN1], both expressing PTS1-GFP from a plasmid, and pan1 – 11 [vector] and pan1 – 11 [PAN1], both expressing PTS1-GFP and PTS2-DsRed, were grown at 30-C in SC medium supplemented with required amino acids and viewed for fluorescence and by DIC optics. (B) pan1 – 11 [vector] and pan1 – 11 [PAN1], both expressing PTS1-GFP, were grown on SCEO supplemented with required amino acids and viewed for fluorescence and by DIC optics.
mutation affects some aspects of cellular metabolic adaptation involved with utilization of nonfermentable carbon sources. Hence, the pan1 –9 and pan1– 11 mutants and their parental strains were spotted on media containing glycerol, acetate, ethanol, pyruvate or succinate as carbon sources. As shown in Fig. 6, pan1 –11 mutant grew significantly slower than the parental strain on glycerol- or succinate-containing media. On the same plates, pan1 – 9 grew similarly or better (succinate) than the parental strain. There were no differences in growth on media containing ethanol, acetate or pyruvate among the two pan1 mutants and the parental strains. However, growth of pan1 – 11 on an ethanol medium was compromised when oleic acid was added (Fig. 6). The pan1 –11 cells remained inhibited but fully viable even after a 48-h incubation in YPEO (data not presented). On the other hand, there was no growth effect upon oleate addition to a glucose medium, indicating that glucose represses the inhibitory effect of oleate (Fig. 6). The pan1 – 11 strain grown on glycerol or succinate medium accumulated cells with abnormal morphology similar to that observed on oleate medium. This phenotype was reversed by PAN1 expressed from the plasmid (not shown). These
results demonstrate that mitochondrial respiration functions properly in the pan1– 11 mutant, and the defect in growth is not restricted to oleate and must be associated with carbon source-dependent regulation of apical cell growth. The pan1– 11 mutant is defective in actin cytoskeleton, fluid phase endocytosis and in cell wall biogenesis The pan1 – 9 mutant cells were characterized previously as defective in fluid phase endocytosis [11] and having an abnormal actin cytoskeleton organization [26]. In addition to regular actin patches and actin cables, the cells displayed abnormal clumps of polymerized actin. Here, we investigated whether deletion of part of the (QPTQPV)7 motif in mutant Pan1 –11p also affected actin cytoskeleton organization which could be responsible for the observed morphological changes. For actin cytoskeleton staining, the pan1 –11 [PAN1] and pan1 –11 [vector] mutant strains were grown overnight in SC, SCO or SCG. The actin cytoskeleton was stained with Oregon Green-conjugated phalloidin. There was no difference in actin cytoskeleton organization between pan1 –11 and wild type cells grown in
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
489
Fig. 8. The pan1 – 11 mutant is defective in fluid phase endocytosis. The Lucifer yellow endocytosis assay was performed on the wild type, on pan1 – 11 [vector] and pan1 – 11 [PAN1]. The same fields were viewed for fluorescence and with DIC optics.
Fig. 6. pan1 – 11 cells are defective in growth on glycerol and succinate media. Growth of wild type, pan1 – 9 and pan1 – 11 strains on the indicated media after incubation at 30-C for 3 – 4 days.
glucose medium. All dividing cells had a polarized actin cytoskeleton with actin patches concentrated in the bud and actin cables distributed along the mother-bud axis; nondividing cells exhibited evenly distributed actin patches in the cell cortex (not shown). Actin cytoskeleton polarization could not be assessed in oleate-grown cells due to high background staining; actin cables were not visible even in wild type cells. To avoid this technical problem, cells grown in glycerol medium were analyzed. Actin cytoskeleton staining of pan1-1 [vector] mutant cells grown in this medium revealed non-polarized actin patches and lack of cables (Fig. 7), indicating defects of actin cytoskeleton
polarization. Well-polarized actin cytoskeleton in pan1– 11 [PAN1] cells was observed. Fluid phase endocytosis was investigated in pan1– 11 [vector], pan1 –11 [PAN1] and wild strain by testing the ability of cells pregrown in SC medium to take up the fluorescent, water-soluble dye, Lucifer yellow (LY). After 1.5 h of incubation with LY, the wild type cells accumulated the dye in the vacuole, whereas pan1– 11 mutant cells did not. The defect in endocytosis of pan1– 11 was complemented by wild type copy of PAN1 on a plasmid, as shown by the accumulation of LY in the internal structures, vacuoles and endosomes in the transformants (Fig. 8). This indicates that pan1 – 11 mutant is defective in endocytosis, similarly to the previously described pan1 mutants. Analysis of the phenotypes of pan1 – 9 and pan1– 11 indicates that the defective growth of pan1– 11 mutant on oleate might be due to disturbed
Fig. 7. Actin cytoskeleton of pan1 – 11 cells grown on glycerol medium is not polarized. The pan1 – 9 [vector] and pan1 – 9 [PAN1] cells were grown on SCG supplemented with required aa to the logarithmic phase and fixed. Filamentous actin (F-actin) was stained with Oregon Green-conjugated phalloidin. Cells were viewed for fluorescence.
490
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
functioning of Pan1p in actin cytoskeleton organization and endocytosis. It was previously documented that the pan1 –4 mutant, similarly to several other cytoskeletal mutants (act1-1, sla1D, end3D), showed cell wall abnormalities [12]. In dividing mutant cells, the cell wall of the mother was multilayered and much thicker than that of the bud, while wild type cells displayed a single layer of cell wall throughout the cell cycle, with no apparent differences between the mother and the bud. The presence of multicopy END3 suppressed the defects in actin cytoskeleton organization but not the cell wall defects of pan1 –4 [12]. Electron microscopy was used to investigate cell wall morphology in pan1 – 9 and pan1– 11 and wild type cells grown on YPD at 30-C. All the wild type cells displayed a single layer of cell wall (Fig. 9A), whereas pan1 – 9 and pan1 – 11 cells exhibited strikingly aberrant cell wall morphologies. The cell wall of the mother cell was thicker than that of the bud in 74% of dividing cells of the pan1 –9 strain. As shown in Fig. 9B, two layers of cell wall were often separated by the dark layer of mannoproteins. In pan1– 11 strain, 83% of dividing cells showed a much thicker cell wall in the mother cell than in the bud (Fig. 9C). The structure of the thick cell wall was uniform, with no visible layers. In some cells, wall fragments were torn apart (Fig. 9D). In addition, many dividing pan1– 11 cells had two buds. Both mutants displayed abnormal cell wall morphology, but the pan1 –11 defect seems more severe, and it may contribute to the lack of growth on oleate medium. Actin cytoskeleton, endocytosis and cell wall morphogenesis are all important for polarized growth of cells.
Fig. 9. The pan1 – 9 and pan1 – 11 mutants are defective in cell wall structure. Wild type (A), pan1 – 9 (B) and pan1 – 11 (C, D) mutants grown on YPD at 30-C were analyzed by electron microscopy. Scale bar represents 1 Am.
Apparently, Pan1p as multivalent regulator of these processes is important for carbon source regulation of the apical cell growth.
Discussion In order to utilize fatty acids, yeast require a functional peroxisomal compartment containing h-oxidation enzymes. The end products of h-oxidation are converted into intermediate metabolites which are shuttled from peroxisomes via the cytosol to mitochondria for further oxidation. Although peroxisomal processes are quite well known, the relevant cytosolic processes, the mitochondrial transport of metabolites and the shuttles involved in fatty acid utilization are still being studied [9,34]. In this report, we provide the first data documenting that actin cytoskeleton-associated proteins Pan1p and End3p are required not only for growth on oleate media but also on glycerol- or succinatecontaining media. Pan1p does not affect peroxisomal enzyme synthesis or peroxisome biogenesis but rather participate in regulation of cell growth in response to carbon source in the medium. The isolated pan1– 11 mutant displayed the Oleate phenotype, which could be suppressed by overexpression of the END3 gene. The pan1 – 11 mutant shared some of the characteristics of other pan1 mutants, like being defective in endocytosis and cell wall biogenesis. Indications regarding the mechanism which may result in the Oleate phenotype came from comparison of pan1 – 11 with the pan1 – 9 mutant isolated previously [11]. The pan1 – 9 mutant, although defective in endocytosis, cytoskeleton organization and cell wall biogenesis, was able to grow on oleate medium. Thus, the Oleate phenotype of pan1 –11 is not the result of defective oleic acid uptake or cellular transport caused by defective cell wall or endocytosis. Moreover, pan1– 11 also exhibited a normal shape and number of peroxisomes as well as correct inducibility of peroxisomal enzymes. Molecular analysis of the pan1 –9 and pan1 – 11 alleles revealed deletion of one base pair in the former, resulting in premature termination after codon 859 and a 64-base-pair deletion in the latter, resulting in the synthesis of Pan1p lacking three of the seven QPTQPV repeats. Surprisingly, although the pan1– 11 deletion was localized in the region untranslated in pan1 –9, pan1– 9, in contrast to pan1– 11, grew well on oleate. One possible explanation of these phenotypes is that the QPTQPV repeats bind some protein(s) which affects the binding of other proteins at the N-terminus of Pan1p. The latter is important for regulation of Pan1p function, and deletion of some QPTQPV repeats would disturb this regulation by interfering with the exchange of binding partners. When the whole C-terminal region of Pan1p is missing, the regulation does not operate and the binding of the protein partners to the N-terminus is simplified. The hypothesis is supported by the finding that
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492
overexpression of End3p, a protein which binds to the Nterminus of Pan1p [35], improves pan1 –11 growth on oleate. Pan1p binds many proteins [12,18,32], but it is not known which protein binds to the (QPTQPV)7 repeats. Recently, intramolecular regulation has been suggested in Pan1p that depends on the presence of poly-proline-rich domain at the C-terminus which cooperates with more Nterminal regions [36]. Thus, an alternative possibility is that the (QPTQPV)7 repeats contribute to the intramolecular regulation in the presence of the most distant C-part, which is not necessary, or is impossible, in the absence of the latter. In the course of our studies, we found that complete (QPTQPV)7 repeats of Pan1p are important for regulation of apical growth of cells when oleate, glycerol or succinate is present in the medium. In S. cerevisiae, the formation of polarized cellular structures is required for cell division by budding and for mating. The cell forms the bud at the surface and segregates DNA and other cellular components into the bud to form a new cell. Bud growth has two phases, apical growth and isotropic growth. The yeast cyclindependent kinase Cdc28p regulates bud morphogenesis and cell cycle progression via the antagonistic activities of Cln and Clb cyclins. Cln G1 cyclins direct polarized growth and bud emergence, whereas Clb G2 cyclins promote isotropic growth of the bud and chromosome segregation [37]. The pan1 – 4 allele was isolated as causing lethality at 37-C with the cdc28-4 allele [19]. Another pan1 mutant allele was isolated that suppressed at 37-C the elongated bud phenotype of the cdc34-2 mutant failing to degrade G1 cyclins [38]. This allele expressed the Pan1p of 869 amino acids, similar to Pan1 –9p. Thus, the PAN1 gene is important for apical growth of cells at an elevated temperature on glucose medium. Our results indicate that some fragments of Pan1p are also specifically involved in certain nonfermentative carbon source regulation of apical growth. Further experiments are required to elucidate the precise mechanism by which Pan1p responds to carbon source to regulate apical growth of yeast.
[14]
Acknowledgments
[15]
We would like to thank A. Sachs, H. Riezman and M. Skoneczny for plasmids and antibodies and R. Olkowski for technical assistance. This work was supported by grant 3P04B01624 from the State Committee for Scientific ˙ Research, Poland to T. Z.
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[16] [17]
[18]
References [19] [1] P.B. Lazarow, W.H. Kunau, Peroxisomes, in: J.R. Pringle, J.R. Broach, E.W. Jones (Eds.), The Molecular and Cellular Biology of the Yeast Saccharomyces, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1997, pp. 547 – 605. [2] D.R. Johnson, L.J. Knoll, D.E. Levin, J.I. Gordon, Saccharomyces cerevisiae contains four fatty acid activation (FAA) genes: an assess-
[20]
[21]
491
ment of their role in regulating protein N-myristoylation and cellular lipid metabolism, J. Cell Biol. 127 (1994) 751 – 762. E.H. Hettema, C.W. van Roermund, B. Distel, M. van den Berg, C. Vilela, C. Rodrigues-Pousada, R.J. Wanders, H.F. Tabak, The ABC transporter proteins Pat1 and Pat2 are required for import of longchain fatty acids into peroxisomes of Saccharomyces cerevisiae, EMBO J. 15 (1996) 3813 – 3822. E.H. Hettema, H.F. Tabak, Transport of fatty acids and metabolites across the peroxisomal membrane, Biochim. Biophys. Acta 1486 (2000) 18 – 27. H.J. Schuller, Transcriptional control of nonfermentative metabolism in the yeast Saccharomyces cerevisiae, Curr. Genet. 43 (2003) 139 – 160. A.S. Tibbetts, Y. Sun, N.A. Lyon, A.C. Ghrist, P.J. Trotter, Yeast mitochondrial oxodicarboxylate transporters are important for growth on oleic acid, Arch. Biochem. Biophys. 406 (2002) 96 – 104. P.J. Trotter, R.A. Hagerman, D.R. Voelker, A yeast strain defective in oleic acid utilization has a mutation in the RML2 gene, Biochim. Biophys. Acta 1438 (1999) 223 – 238. K.I. Minard, G.T. Jennings, T.M. Loftus, D. Xuan, L. McAlister-Henn, Sources of NADPH and expression of mammalian NADP+-specific isocitrate dehydrogenases in Saccharomyces cerevisiae, J. Biol. Chem. 273 (1998) 31486 – 31493. S. Cavero, A. Vozza, A. del Arco, L. Palmieri, A. Villa, E. Blanco, M.J. Runswick, J.E. Walker, S. Cerdan, F. Palmieri, J. Satrustegui, Identification and metabolic role of the mitochondrial aspartate – glutamate transporter in Saccharomyces cerevisiae, Mol. Microbiol. 50 (2003) 1257 – 1269. I. Smaczyn˜ska-de Rooij, A. Migdalski, J. Rytka, a-Ketoglutarate dehydrogenase and lipoic acid synthase are important for the functioning of peroxisomes of Saccharomyces cerevisiae, Cell. Mol. Biol. Lett. 9 (2004) 271 – 286. T.¯o31dek, G. Vaduva, L.A. Hunter, M. Boguta, B.D. Go, N.C. Martin, A.K. Hopper, Mutations altering the mitochondrial – cytoplasmic distribution of Mod5p implicate the actin cytoskeleton and mRNA 3V ends and/or protein synthesis in mitochondrial delivery, Mol. Cell. Biol. 15 (1995) 6884 – 6894. H.Y. Tang, J. Xu, M. Cai, Pan1p, End3p, and S1a1p, three yeast proteins required for normal cortical actin cytoskeleton organization, associate with each other and play essential roles in cell wall morphogenesis, Mol. Cell. Biol. 20 (2000) 12 – 25. I. Smaczyn˜ska de Rooij, Saccharomyces cerevisiae mutants altered in the expression of peroxisomal proteins; isolation and characterization, PhD thesis, Institute of Biochemistry and Biophysics PAS, Warsaw, Poland, 2001. S. Raths, J. Rohrer, F. Crausaz, H. Riezman, end3 and end4: two mutants defective in receptor-mediated and fluid-phase endocytosis in Saccharomyces cerevisiae, J. Cell Biol. 120 (1993) 55 – 65. P. Chen, P. Johnson, T. Sommer, S. Jentsch, M. Hochstrasser, Multiple ubiquitin-conjugating enzymes participate in the in vivo degradation of the yeast MAT alpha 2 repressor, Cell 74 (1993) 357 – 369. F. Sherman, Getting started with yeast, Methods Enzymol. 350 (2002) 3 – 41. C. De Virgilio, N. Burckert, G. Barth, J.M. Neuhaus, T. Boller, A. Wiemken, Cloning and disruption of a gene required for growth on acetate but not on ethanol: the acetyl-coenzyme A synthetase gene of Saccharomyces cerevisiae, Yeast 8 (1992) 1043 – 1051. B. Wendland, S.D. Emr, Pan1p, yeast eps15, functions as a multivalent adaptor that coordinates protein – protein interactions essential for endocytosis, J. Cell Biol. 141 (1998) 71 – 84. H.Y. Tang, M. Cai, The EH-domain-containing protein Pan1 is required for normal organization of the actin cytoskeleton in Saccharomyces cerevisiae, Mol. Cell. Biol. 16 (1996) 4897 – 4914. C. Brocard, G. Lametschwandtner, R. Koudelka, A. Hartig, Pex14p is a member of the protein linkage map of Pex5p, EMBO J. 16 (1997) 5491 – 5500. K. Stein, A. Schell-Steven, R. Erdmann, H. Rottensteiner, Interactions
492
[22]
[23]
[24]
[25]
[26]
[27]
[28] [29] [30]
J. Kamin´ska et al. / Experimental Cell Research 310 (2005) 482 – 492 of Pex7p and Pex18p/Pex21p with the peroxisomal docking machinery: implications for the first steps in PTS2 protein import, Mol. Cell Biol. 22 (2002) 6056 – 6069. M. Go´ra, K. Pluta, A. Che3stowska, T. ¯o31dek, Suppressors of translation initiation defect in hem12 locus of Saccharomyces cerevisiae, Acta Biochim. Pol. 47 (2000) 181 – 190. R. Rothstein, Targeting, disruption, replacement, and allele rescue: integrative DNA transformation in yeast, Methods Enzymol. 194 (1991) 281 – 301. M.D. Rose, F. Winston, P. Hieter, Methods in yeast genetics, A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 2004, pp. 126 – 127. L. Guarente, Yeast promoters and lacZ fusions designed to study expression of cloned genes in yeast, Methods Enzymol. 101 (1983) 181 – 191. J. Kamin˜ska, B. Gajewska, A.K. Hopper, T. ¯o31dek, Rsp5p, a new link between the actin cytoskeleton and endocytosis in the yeast Saccharomyces cerevisiae, Mol. Cell. Biol. 22 (2002) 6946 – 6948. R. Boeck, S. Tarun Jr., M. Rieger, J.A. Deardorff, S. Muller-Auer, A.B. Sachs, The yeast Pan2 protein is required for poly(A)-binding protein-stimulated poly(A)-nuclease activity, J. Biol. Chem. 271 (1996) 432 – 438. A.E. Adams, J.R. Pringle, Staining of actin with fluorochromeconjugated phalloidin, Methods Enzymol. 194 (1991) 729 – 731. V. Dulic, M. Egerton, I. Elguindi, S. Raths, B. Singer, H. Riezman, Yeast endocytosis assays, Methods Enzymol. 194 (1991) 697 – 710. E.S. Reynolds, The use of lead citrate at high pH as an electronopaque stain in electron microscopy, J. Cell Biol. 17 (1963) 208 – 212.
[31] A.B. Sachs, J.A. Deardorff, Translation initiation requires the PABdependent poly(A) ribonuclease in yeast, Cell 70 (1992) 961 – 973. [32] M.C. Duncan, M.J. Cope, B.L. Goode, B. Wendland, D.G. Drubin, Yeast Eps15-like endocytic protein, Pan1p, activates the Arp2/3 complex, Nat. Cell Biol. 3 (2001) 687 – 690. [33] H. Rottensteiner, A.J. Kal, M. Filipits, M. Binder, B. Hamilton, H.F. Tabak, H. Ruis, Pip2p: a transcriptional regulator of peroxisome proliferation in the yeast Saccharomyces cerevisiae, EMBO J. 15 (1996) 2924 – 2934. [34] B.M. Bakker, K.M. Overkamp, A.J. van Maris, P. Kotter, M.A. Luttik, J.P. van Dijken, J.T. Pronk, Stoichiometry and compartmentation of NADH metabolism in Saccharomyces cerevisiae, FEMS Microbiol. Rev. 25 (2001) 15 – 37. [35] H.Y. Tang, A. Munn, M. Cai, EH domain proteins Pan1p and End3p are components of a complex that plays a dual role in organization of the cortical actin cytoskeleton and endocytosis in Saccharomyces cerevisiae, Mol. Cell. Biol. 17 (1997) 4294 – 4304. [36] N.B. Miliaras, J.H. Park, B. Wendland, The function of the endocytic scaffold protein Pan1p depends on multiple domains, Traffic 5 (2004) 963 – 978. [37] V.J. Cid, J. Jimenez, M. Molina, M. Sanchez, C. Nombela, J.W. Thorner, Orchestrating the cell cycle in yeast: sequential localization of key mitotic regulators at the spindle pole and the bud neck, Microbiology 148 (2002) 2647 – 2659. [38] S. Bidlingmaier, M. Snyder, Large-scale identification of genes important for apical growth in Saccharomyces cerevisiae by directed allele replacement technology (DART) screening, Funct. Integr. Genomics 1 (2002) 345 – 356.