Microbiologie et Génétique Moléculaire, Institut National Agronomique Paris-Grignon, Institut National de la Recherche Agronomique UMR1238, Centre National de la Recherche Scientifique UMR2585, 78850 Thiverval-Grignon, France
Correspondence
Sylvie Blanchin-Roland
blanchin{at}grignon.inra.fr
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
The GenBank/EMBL/DDBJ accession number for the sequence reported in this paper is AJ509167.
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In Y. lipolytica, screening of an insertional library for altered expression of either acidic or alkaline protease genes, or both, revealed besides components of the Rim pathway, other potential regulators of protease expression, in particlular YlVps28p (Gonzalez-Lopez et al., 2002). Mutations in YlVPS28 conferred an Xpr2 phenotype like Ylrim mutations. But, in contrast to Ylrim mutations, they clearly interfered with the capacity of the strains to execute the yeasthyphae transition (Gonzalez-Lopez et al., 2002
).
In S. cerevisiae, Vps28p is a class E component of the endosomal-sorting complex ESCRT-I (Katzmann et al., 2001). This complex participates in the recognition of both ubiquitinated endocytic and biosynthetic cargos that are targeted to the multivesicular body (MVB) vesicles to be ultimately delivered to the lumen of the vacuole when the MVB fuses with this organelle (Futter et al., 1996
). In contrast, transmembrane proteins that remain in the limiting membrane of the MVB are either delivered to the limiting membrane of the vacuole, or recycled to the plasma membrane or the Golgi (Lemmon & Traub, 2000
). The MVB-sorting pathway thus affects both endocytosis and exocytosis of membrane-anchored proteins. It is conserved from yeast to higher eukaryotes, and is required for a growing list of cellular functions that includes downregulation of receptors and transporters, regulation of the immune response and even the budding of certain viruses like human immunodeficiency virus (for a review see Katzmann et al., 2002
).
Therefore, the question was whether Ylvps28 mutations affected expression of protease genes through an indirect effect on protein expression, or through a specific interference with the Rim signalling pathway. This last hypothesis was particularly attractive since interactions had been detected during a whole-genome screen of two-hybrid interactants in S. cerevisiae, between Rim20p and two other Vps components, Snf7p and Vps4p, and Rim13p and Snf7p (Ito et al., 2001). Snf7p is a component of the ESCRT-III complex (Babst et al., 2002a
), which acts downstream from ESCRT-I and ESCRT-II complexes in the MVB pathway, whereas the Vps4p/End13p AAA ATPase acts at the end of the endocytic cycle to dissociate the ESCRT-III complex from the endosomal membrane (Babst et al., 1998
). Subsequent observations in A nidulans, S. cerevisiae and C. albicans confirmed that Snf7p indeed interacted physically with Rim20p/PalA (Bowers et al., 2004
; Ito et al., 2001
; Vincent et al., 2003
; Xu et al., 2004
). Analysis of mutations affecting the interaction domain of Rim20p showed that both Rim20pRim101p and Rim20pSnf7p interactions were required for Rim101p processing (Xu et al., 2004
). The interaction between Rim20p and Vps4p was also observed in another two-hybrid analysis (Bowers et al., 2004
), but its in vivo significance was later questioned since Vps4p does not appear to be required for Rim101p processing in S. cerevisiae (Kullas et al., 2004
). Indeed, both Snf7p and Vps4p were shown to interact with Bro1p/Vps31p (Gavin et al., 2002
), a Rim20p/PalA paralogue, which is another soluble class E factor required for MVB sorting but not for pH sensing (Odorizzi et al., 2003
). The interaction between Rim13p and Snf7p has not been confirmed by other studies. Finally, a systematic phenotypic analysis of a nearly complete collection of gene-deletion mutants of S. cerevisiae indicated that several vps mutant strains exhibited impaired growth at alkaline pH or in the presence of high concentrations of monovalent cations (Giaever et al., 2002
), two phenotypes that are characteristic of rim mutants in this organism (Lamb et al., 2001
).
More recently, and while our work was in progress, definite evidence for a functional link between the Rim and Vps pathways was obtained in the yeasts S. cerevisiae and C. albicans. In C. albicans, a mutant strain inactivated for the CaSNF7 gene displayed all the phenotypes expected for a Carim mutant strain (pH and ion sensitivity during growth, and the absence of alkaline-induced filamentation and Rim101p processing) and was suppressed by a constitutively active form of CaRim101p (Kullas et al., 2004). Other vps mutants affecting different ESCRT-I, -II and -III components were shown, in a separate study, to exhibit morphological defects characteristic of rim mutants (Xu et al., 2004
). In S. cerevisiae, mutations in several ESCRT-I, -II and III components affected processing of the transcriptional activator Rim101p, as would do, for example, a rim20 mutation (Xu et al., 2004
). Interestingly in this last study, not all ESCRT-encoding genes were required for the processing of Rim101p, and none of the non-ESCRT vps genes were. The theory that not the whole endocytic pathway was involved in pH signalling was further confirmed by the observation that Vps4p defects had no drastic effect on Rim101p processing either in S. cerevisiae or in C. albicans (Kullas et al., 2004
; Xu et al., 2004
).
This paper reports on the characterization of vps mutant strains in Y. lipolytica. It confirms the conservation of a functional link between the components of the endocytic pathway and pH signalling, and the absence of an involvement of Vps4p, over the full range of ascomycetous yeasts. We further show that SNF7 is an essential gene in Y. lipolytica. This last observation may explain why ESCRT components were not easily, or not at all, identified in exhaustive screens for genes affecting pH signalling in Y. lipolytica and A. nidulans.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Culture media and phenotypic tests.
Complete YPD medium, minimal YNB-glucose medium, derepressing (Y) medium and protease-inducing medium (YPDm) have been described previously (Blanchin-Roland et al., 1994; Lambert et al., 1997
). Solid and liquid media were buffered at pH 4·0 (with 0·2 M sodium citrate buffer), pH 7·0 (with 0·2 M sodium phosphate buffer) or pH 8·0 (with 25 mM Tris/HCl) and supplemented with 0·1 g uracil l1. The pH of liquid media was stable within 0·3 pH units during growth. The Lac+ phenotype was screened on Y medium buffered at pH 7·0, as previously described (Gonzalez-Lopez et al., 2002
).
Growth at various pHs was assayed by spotting serial threefold dilutions of exponential-phase cultures in YPD liquid medium, on plates buffered either with 0·2 M citrate-phosphate buffer at pH 3·5, or with 0·2 M MOPS for assays at pH 7·08·5. Crossing of compatible Y. lipolytica auxotrophs, diploid sporulation and spore isolation was done as described previously (Gonzalez-Lopez et al., 2002). Dilutions of spore suspensions were plated on complete medium and colonies were tested on appropriate media.
DNA and RNA techniques.
Standard recombinant DNA techniques were performed essentially as previously described (Gonzalez-Lopez et al., 2002). Sequences were obtained from the DNA sequencing department of Eurogentec. They were assembled and annotated using the GCG package. All transformation events were checked by colony PCR using PuRe Taq Ready-To-Go PCR beads (Amersham) and confirmed by Southern blotting.
Gene expression was determined by real-time quantitative RT-PCR using a LightCycler (Roche Molecular Biochemicals). Cells were grown at 28 °C in Y medium buffered either at pH 4 or 8, and harvested at the exponential phase (OD600 about 0·6). Total RNA was extracted with the RNeasy mini-kit from Qiagen and then treated by DNase I (Qiagen). The cDNA was synthesized from 1 µg total RNA according to the manufacturer's protocol of the SuperScript II RNase H reverse transcriptase kit (Invitrogen). Primer sets for the genes were as follows: ACT20-ACT21r for ACT1, XPR210-XPR211r for XPR2, PHR112-PHR113r for PHR1, RIM104-RIM105r for RIM101 and PHR203-PHR204r for PHR2 (Table 2). The PCR parameters were 95 °C for 8 min, followed by 45 cycles at 95 °C for 10 s, 55 °C for 7 s, and 72 °C for 10 s. A negative control with sterile water was performed for each primer set. The threshold cycle (Cp) was determined as the cycle above which the fluorescence signal produced by the SYBRGreen I dye, reached a baseline level. The expression levels of the genes were determined relative to the expression of the ACT1 gene, which was uniform, as confirmed by Northern experiments (data not shown). For each gene, experiments were carried out at least four times, using two cDNA samples from two (or in a few cases three) independent cultures. Statistical analysis was performed using Student's t-test. A P value lower than 0·05 was considered as significant.
|
Production of ATP-binding site mutant YlVps4p-(K172A).
Site-directed mutagenesis of the ATP-binding site of Vps4p was performed by the overlapping PCR technique to create a mutated Vps4p with a Lys172Ala substitution. Two PCR products were amplified with SY12 genomic DNA as a template, using primers V4BP1-VPS4ka3r and VPS4ka5-V4XT2. These PCR products, containing the mutation in their overlapping regions, were mixed and amplified with the flanking primers to generate a full-length PCR product containing the internal mutation. This PCR product was digested by BglII/XbaI and then inserted into pINA300' digested by BamHI/NheI, to generate plasmid pINA1357, which was confirmed by sequencing. The NheI-digested pINA1357 was targeted to the promoter of VPS4 in strain SBR114u to give strain SY1364.
Cloning of the YlSNF7 gene.
The Y. lipolytica wild-type SNF7 gene was amplified from SY12 genomic DNA by the Expand High Fidelity PCR system from Roche with primers V32XP1/V32BT2. The PCR product carrying SNF7 was digested with XbaI, and then inserted into the replicative LEU2 plasmid pINA240 digested by NruI/NheI to give pINA1355, which was checked by DNA sequencing.
Introduction of the YlRIM101-1119 truncated allele into Ylvps mutant strains.
YlURA3 was excised from strains CGL-U15 (Ylvps23-U15 allele), CGL-Y2 (Ylvps2-Y2 allele) and SBR114 (Ylvps4-14 allele), by using plasmid pUB4-CRE as previously described (Fickers et al., 2003) to give strains CGL-U15u, CGL-Y2u and SBR114u, respectively. The DraIII-digested pINA1119 (Lambert et al., 1997
) was targeted to the terminator of YlRIM101 in these resulting Ura strains, and in LAM26-03 strain (rim13-26 allele). Secondary Ura clones were selected on 5-fluoroorotic acid medium, and recombination events leaving the YlRIM101-1119 truncation were selected (strains SBR116SBR119, see Table 1
).
Staining with FM4-64.
FM4-64 localization experiments were done using a modification of the protocol described by Vida & Emr (1995). Yeast cells were grown in YPD to an OD600 of about 0·6. Cells (5 ml culture volume) were harvested, incubated in 150 µl YPD containing 40 µM FM4-64 (Molecular Probes) for 30 min at 0 °C, washed three times on ice, and further incubated in 200 µl YPD for 20 min at 18 °C. An aliquot of cells was centrifuged, resuspended in water and visualized by differential interference contrast (DIC) and fluorescence microscopy using an Olympus U-RFL-T microscope with a CoolSNAP camera.
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
This mutant strain was affected in a gene orthologous to S. cerevisiae VPS23, which encodes a component of the ESCRT-I complex required for sorting membrane proteins into the MVB pathway (Katzmann et al., 2001), as was YlVPS28 previously identified in the same screening (Gonzalez-Lopez et al., 2002
). YlVPS23 encodes a predicted protein of 378 aa (accession no. AJ509167) displaying 36 % identity with Homo sapiens TSG101 (accession no. AAC52083), 34 % with Drosophila melanogaster TSG101 (accession no. AF315343), 28 % with S. cerevisiae Vps23p/Stp22p (Li et al., 1999
) and with C. albicans CA2343/orf19.2343 (http://genolist.pasteur.fr/CandidaDB/). Comparison of the Y. lipolytica amino acid sequence with its homologues showed a conservation of the domain structure predicted for ScVps23p and TSG101 (Katzmann et al., 2001
): a N-terminal domain homologous to ubiquitin-conjugating enzymes, but unlikely to catalyse ubiquitination because of the substitution for the C residue that binds ubiquitin, a Pro-rich region, a coiled-coil region (positions 218251 and 265299) and a conserved C-terminal portion. The YlmTn1 insertion truncates YlVps23p at aa 150 (Ylvps23-U15).
Role of YlVPS23 and YlVPS28 in alkaline- or acidic-pH responses
To establish whether YlVPS genes are involved in ambient pH signalling, we assessed the effects of Ylvps mutations on the transcription of different pH-regulated genes by real-time quantitative PCR on total RNA extracted from cultures grown in Y medium at pH 4·0 and 8·0, using actin transcript as a reference. We chose the alleles Ylvps23-U15 (the only mutation available affecting this gene) and Ylvps28-Y2, which truncates Vps28p at aa 87. As ambient alkaline-pH-responding genes, we chose, as previously, XPR2 and YlPHR1 (accession no. YALI0D04851g) (Gonzalez-Lopez et al., 2002) plus YlRIM101. Since expression of the acidic-protease gene AXP1 was shown to be independent of the Rim pathway, we chose as acidic pH-responding gene YlPHR2 (accession no. YALI0D06039g), a homologue of CaPHR2 known to be expressed at acidic pH only in C. albicans (Muhlschlegel & Fonzi, 1997
). As a control of the effect of rim mutations on the expression of these genes, we chose one Ylrim mutant, carrying the Ylrim13-26 mutation (see Methods) that displays a typical Rim phenotype (Gonzalez-Lopez et al., 2002
).
At pH 8·0, both Ylvps23-U15 and Ylvps28-Y2 mutations significantly decreased YlRIM101 expression (more than 2 times compared to wild-type, P values 0·01), reduced YlPHR1 expression more than 20 times and nearly abolished XPR2 transcription (Fig. 1
a). These observations were fully confirmed by measuring the expression of the XPR2':lacZ fusion integrated at the XPR2 locus of these strains and by Northern blotting (data not shown). At pH 4·0, no or very little transcription of these alkaline reporter genes could be detected (Fig. 1b
). Thus, like YlRIM13 and other YlRIM genes (Gonzalez-Lopez et al., 2002
; Lambert et al., 1997
), YlVPS23 and YlVPS28 are absolutely required for transcriptional activation of alkaline-induced genes at pH 8·0. In both Ylvps mutant strains, as in the Ylrim13-26 mutation strain, at alkaline pH, expression of the acid-induced gene YlPHR2 derepressed more than 20 times over the wild-type level at the same pH, reaching its level at acidic pH in the reference strain (Fig. 1a, b
).
|
|
Expression of the YlRIM101-1119 allele rescues the alkaline-sensitive growth phenotype of the Ylrim13-26 mutant and generates a mild acid-sensitive phenotype (Fig. 2a, RimSL or rim13+RimSL). Both effects were observed when this activated form was expressed in Ylvps28-Y2 and Ylvps23-U15, the acid-sensitive phenotype being exacerbated in these strains compared to Ylrim13-26 (Fig. 2a
).
All these results confirmed that (i) like Ylrim mutations, Ylvps23 and Ylvps28 mutations are suppressed by a constitutively active form of YlRim101p, suggesting that their effects at alkaline pH are mediated through YlRim101p, (ii) unlike Ylrim mutant strains, Ylvps mutant strains display a strong alkaline-sensitive growth phenotype and an acid-sensitive phenotype that is exacerbated when the alkaline response is turned on at acidic pH, suggesting that their response to ambient pH is affected both in Rim-dependent and in Rim-independent ways.
YlSNF7 is an essential gene in Y. lipolytica
In S. cerevisiae, Vps23p and Vps28p are both components of the ESCRT-I complex (Katzmann et al., 2001). We wondered whether Snf7p, which is known to interact with Rim20p in S. cerevisiae, A. nidulans and C. albicans (see Introduction), was also involved in pH regulation. The YALI0C16027g sequence encodes a predicted protein of 215 aa that matches putative orthologous gene products in other species, with the following identity scores: 49 % with S. cerevisiae Snf7p, 50 % with C. albicans Snf7p (CA4903 http://genolist.pasteur.fr/CandidaDB/), 43 % with Schizosaccharomyces pombe Spac1142·07c (CAB77014.1), 47 % with Neurospora crassa hypothetical protein (CAD70846·1) and 41 % with H. sapiens CHMP4b (CAC14088.1). Like its homologues, YlSnf7p presents several putative coiled-coil regions (positions 1448, 5385 and 149173).
As SNF7 is non-essential in S. cerevisiae and C. albicans, we first tried to delete the YlSNF7 coding sequence in the haploid strain SY12 (see Methods). We got only two disruptants from several transformation assays, most transformants resulting from ectopic integrations. The two clones were not affected in endocytosis and displayed no pH-response defect (data not shown). In order to rule out that a functionally redundant protein may compensate for the loss of YlSnf7p, we searched for possible YlSnf7p homologues within the Y. lipolytica genome. Only two distant paralogues YlVps20p (28 % identity) and Mos10p/Vps60p (33 % identity) were identified, as would be the case in S. cerevisiae for a search of ScSnf7p paralogues. We wondered if the presence of suppressing mutation(s) in the YlRIM101 gene may account for the lack of phenotypes for Ylsnf7 strains. The sequence of the YlRIM101 locus was checked entirely in these strains, but no mutation could be detected. We thus assumed that YlSNF7 might be essential. Therefore, two strategies were developed to disrupt this gene. (i) One YlSNF7 allele of the diploid strain AM3/AM4 was disrupted as before, with selection for Ura+ isolates. Deletion of one YlSNF7 allele was confirmed by PCR and Southern blotting for several transformants. Two independent diploids were sporulated and their progeny was studied by random spore analysis. Among a total of 153 fast-growing auxotrophic segregants, none were Ura+ (i.e. disrupted for YlSNF7), whereas the other auxotrophic markers segregated normally. Twelve very slow-growing colonies were later observed on the plates; they were all Ura+, and upon a PCR test were shown to be disomic, carrying both a normal and a deleted copy of YlSNF7 (data not shown). We concluded, therefore, that the YlSNF7 gene was essential for spore germination and/or vegetative growth. (ii) The plasmid pINA1356, carrying an internal deletion of the whole YlSNF7 coding sequence, was integrated by a single crossing-over event into the promoter of the YlSNF7 gene in the haploid strain E122, with selection for Ura+. This created a wild-type copy and a deleted copy integrated in tandem in the genome and flanking the URA3 marker. The resulting strain was then transformed with the Leu+ replicative plasmid pINA1355 carrying the wild-type YlSNF7 gene, with selection for unstable Leu+ clones. Secondary Ura clones were selected on 5-fluoroorotic acid medium and screened for the occurrence of recombination events leaving the Ylsnf7-1356 deletion. Six Ura Leu+ transformants, carrying the expected deletion of the genomic YlSNF7 allele and a free pINA1355 plasmid (as shown by E. coli transformation), were tested for the stability of the Leu+ marker. No loss of the replicative plasmid was observed among more than 600 clones tested, after 30 generations on non-selective medium, thus confirming that the YlSNF7 gene was essential for the viability of the cells during vegetative growth.
YlVPS4 does not appear to be required for the Rim101-dependent pH response
We wondered whether Vps4p (Babst et al., 1998), which acts downstream from the three ESCRT complexes and has been reported to interact with Rim20p (Bowers et al., 2004
), was also involved in pH regulation in Y. lipolytica. A putative homologue (accession no. YALI0B16368g) of this gene was identified in the Y. lipolytica genome. The predicted 428 aa YlVps4p protein matches putative homologues in other species with the following identity scores: 77 % with C. albicans Vps4p (CA1340; http://genolist.pasteur.fr/CandidaDB/), 73 % with S. cerevisiae Vps4p, 67 % with A. nidulans hypothetical protein AN3061.2 (EAA63632.1), 63 % with Schizosaccharomyces pombe Spac2G11.06 (CAA91171.1), 61 % with H. sapiens Vps4-2 (AAG33022.1) and 60 % with Mus musculus Skd1p (AAD47570.1). Like its homologues, YlVps4p displays the three-domain structure postulated for ScVps4p (Babst et al., 1998
): the N-terminal domain with a putative coiled-coil motif (position 5180), one central AAA domain containing the AAA-protein family signature (position 263281), two consensus ATP-binding sites consisting of a Walker A box motif (position 172179) and a version of the Walker B motif called the DEAD box (position 229242) and the highly charged C-terminal domain.
To assess the effects of YlVPS4 on pH response, the phenotype of the SBR114 strain, carrying the null mutation Ylvps4-14, was established (see Methods). The effects of the YlVPS4 deletion on the expression of alkaline- and acid-induced genes were assessed by real-time quantitative PCR, at pH 4·0 and 8·0 (Fig. 1
). At alkaline pH, in Ylvps4
-14, the transcript levels of XPR2, YlPHR1, YlRIM101 and YlPHR2 varied by a factor of less than two compared to the wild-type strain, less than observed in the rim13, vps23 and vps28 mutants (see above). At acidic pH, XPR2 and YlPHR1 showed significantly more expression than in these mutants (P<0·01), whereas the level of expression of the acid-induced gene YlPHR2 was barely affected (less than twofold, P 5x105). These results indicate that the Ylvps4
mutation affects the pH response much less severely than the Ylrim13, Ylvps23 and Ylvps28 mutations at alkaline pH, and affects the pH response in an opposite way with respect to alkaline-induced genes at acidic pH. These complex effects might partly relate to the slow growth of this mutant. Indeed, the deletion of YlVPS4 conferred a pronounced growth defect on cells grown on YPD medium at all pH values tested (Fig. 2b
).
Next, we checked whether expression of the constitutively active, truncated form YlRIM101-1119 was able to suppress the defects of the Ylvps4-14 mutant, as it did for the rim13, vps23 and vps28 mutants. To address this question, we replaced the wild-type YlRIM101 allele by the most truncated allele, YlRIM101-1119 (Lambert et al., 1997
), in the Ylvps4
-14 mutant strain (see Methods). As shown in Fig. 2(b)
, expression of the YlRIM101-1119 allele did not rescue the growth defects of Ylvps4
-14, at all pH values tested. At both alkaline and acidic pHs, the YlRIM101-1119 allele failed to restore the expression of the two alkaline-induced genes, XPR2 and PHR1, in the Ylvps4
-14 mutant (Fig. 1a, b
).
Vps4p exhibits an ATPase activity, which is required in S. cerevisiae for the dissociation of ESCRT complexes at the end of the MVB cycle (Babst et al., 1998), and has been proposed as an interactant of Rim20p (see Introduction). In order to check whether the growth defect of Ylvps4
-14 was due to the loss of the ATPase activity or to the loss of the protein itself, we constructed the SY1364 strain (see Methods; Table 1
) where a K172A mutation specifically destroys the ATP-binding site of Vps4p. Correct expression of the mutated allele, which is expected to display a dominant negative phenotype on endocytosis (Babst et al., 1997
), was checked by its ability to block endocytosis in a YlVps4+ background, as checked by FM4-64 staining (data not shown). As shown in Fig. 2(b)
, the growth of the Ylvps4K172A mutant was as severely affected as the growth of Ylvps4
-14 at all pH values tested. Taken together, these results suggest that effects of the Ylvps4
-14 mutation mostly result from an endocytic Rim-independent defect on cell physiology.
Ylvps mutants exhibit endocytic defects not suppressed by the C-terminal truncated allele of RIM101
In order to confirm that the vps mutations indeed affected the MVB pathway in Y. lipolytica, the endocytic defects of Ylvps4-14, Ylvps23-U15 and Ylvps28-Y2 mutant strains were visualized with the fluorescent dye FM4-64, a lipophilic dye that binds to the plasma membrane, is internalized and then delivered to the vacuolar membrane (Vida & Emr, 1995
). Cells were labelled with FM4-64 at 0 °C, then warmed to 18 °C and assessed for the distribution of FM4-64. By 20 min, FM4-64 could be seen on the vacuole limiting membrane in the wild-type cells. In contrast, in all the mutant strain cells, the bulk of FM4-64 remained in small compartments adjacent to the vacuole (Fig. 3
). These structures are highly reminiscent of class E compartments, which accumulate proteins destined to the vacuole in class E vps mutants in S. cerevisiae (Rieder et al., 1996
; Vida & Emr, 1995
). When the constitutively active YlRIM101-1119 allele was expressed in Ylvps4
-14, Ylvps23-U15 and Ylvps28-Y2 cells, the bulk of FM4-64 remained in small compartments adjacent to the vacuole (Fig. 3
), as in the parental mutant strains. These results show that mutations in YlVPS4, YlVPS23, YlVPS28 indeed lead to a membrane-trafficking defect that, as expected, is independent of the status of the Rim pathway.
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Here we report on the characterization of Ylvps mutations affecting the ESCRT-I complex of the MVB pathway in Y. lipolytica. We show that Ylvps23 and Ylvps28 mutations not only affect the endocytic pathway, as expected from the function of the proteins encoded by their homologues in S. cerevisiae and C. albicans (Rieder et al., 1996; Vida & Emr, 1995
; Kullas et al., 2004
; Cornet et al., to be published), but also that the mutant strains have a bona fide Rim phenotype, as described in these two other yeasts (Kullas et al., 2004
; Xu et al., 2004
; unpublished data). The proteins are required for growth at alkaline pH (Fig. 2
), activation of alkaline-induced genes and repression of an acid-induced gene, like YlPHR2 (Fig. 1
). All the defects in the alkaline response are suppressed at least partially by the constitutively active RIM101-1119 allele expressing a truncated form of the downstream transcriptional activator Rim101p. The suppressive effect of the truncated form of Rim101p suggests that these MVB components act upstream from Rim101p. In agreement with findings reported in S. cerevisiae and C. albicans, we showed that Ylvps4 does not appear to be required for activation of the Rim pathway in Y. lipolytica. This suggests that part of the MVB pathway only is required for ambient pH signalling in all ascomycetous yeasts. An attractive hypothesis would be that early steps of endocytosis including formation, but not dissociation, of the ESCRT complexes mediate signalling by the postulated membrane receptors Rim21p/PalH and/or Rim9p/PalI (Peñalva & Arst, 2004
), and lead to recruitment by Snf7p of the cytoplasmic Rim machinery (Kullas et al., 2004
; Xu et al., 2004
).
During this work, we observed that YlPHR2 behaved as expected for a bona fide acid-induced gene. YlPHR2 was derepressed at alkaline pH in YlRim13, Ylvps23 or Ylvps28 mutant strains to identical levels, close to those observed at pH 4·0 in the wild-type strain (Fig. 1). It was repressed in these three backgrounds to similar levels at pH 4·0 upon expression of the artificially truncated form of YlRim101p. This repression was incomplete in all cases, suggesting that our construct YlRim101-1119p does not perfectly mimic the physiologically active form of YlRim101p, as previously reported (Gonzalez-Lopez et al., 2002
; Lambert et al., 1997
). These results show that, like C. albicans, Y. lipolytica expresses two isoforms of the surface proteins Phr1p/Phr2p with inverted pH-regulation patterns, both YlRim101p-dependent (Muhlschlegel & Fonzi, 1997
). A search in the promoters of YlPHR1 and YlPHR2 for GCCARG, the putative consensus binding site for YlRim101p (Blanchin-Roland et al., 1994
; Lambert et al., 1997
), identified one such site in YlPHR1 (position 139 from ATG) and three in YlPHR2 (positions 302, 296 and 224 from ATG). Further work is needed to confirm the role of the putative YlRim101p-binding sites identified so far in the promoters of these acid- and alkaline-induced genes.
Besides effects on endocytosis and alkaline-pH sensing, our results evidenced other effects associated with mutations in VPS genes in Y. lipolytica. Ylvps mutants appear more sensitive to acidic and alkaline pH than Ylrim mutants (Fig. 2), and the induction of an alkaline response, by a constitutively active form of the Rim101p regulator, is more deleterious at acidic pH for Ylvps23 and Ylvps28 than for Ylrim mutants (Fig. 2
). This exacerbated defect could be explained by the fact that the growth of Ylvps mutant strain was already affected at acidic pH, whereas the growth of Ylrim mutants was not. These defects suggest that ambient pH might influence MVB pathway activity in vivo, as has been suggested by Xu et al. (2004)
.
A puzzling observation is that only two ESCRT-I component encoding genes were identified in our initial mutagenesis. This may seem surprising in view of the different VPS genes reported in S. cerevisiae, from the screening of a haploid deletion-strain library, as being required for growth at mild alkaline pH (Serrano et al., 2004), or functioning in the Rim101p pathway (Barwell et al., 2005
). This last exhaustive screening revealed the genes already shown to be involved in Rim101p processing (Xu et al., 2004
), as encoding components of either the ESCRT-I (Vps23, Vps28, Vps37), the ESCRT-II (Vps22, Vps25, Vps36) or the ESCRT-IIIA sub-complex (Vps20, Vps32). Although our screen was not exhaustive, it identified most (i.e. five out of six) of the RIM genes, but only two out of the eight VPS genes, so far identified in S. cerevisiae as affecting Rim101p processing. It is even more surprising to observe that no vps mutant at all was identified in A. nidulans, in exhaustive screens for defects in pH signalling (Peñalva & Arst, 2004
). Recent reports suggest that fungi do have homologues of the VPS genes (Read & Kalkman, 2003
). The fact that YlSNF7 was found to be essential, the Ylvps4
mutant strain grows very poorly, and Ylvps23 and Ylvps28 mutations mildly affect growth at all pH conditions tested (Fig. 2
), even at pH 7·0, may offer a general clue to this apparent paradox. A likely hypothesis is that most, if not all, VPS genes are essential, or are required to avoid major growth defects in A. nidulans and to a lesser extent in Y. lipolytica. In S. cerevisiae, although vps mutant strains were reported to be viable, a careful analysis shows that many, including vps23 and snf7, may present severe growth defects on several types of media, including complete medium, after 20 or more generations (Giaever et al., 2002
; see also http://www.yeastgenome.org/). In C. albicans, the growth of the snf7
/snf7
mutant is significantly reduced on complete medium and drastically impaired at alkaline pH (Kullas et al., 2004
). Since pH signalling is not an essential function in fungi under most circumstances, it might be that endocytosis is required for cellular growth.
In Y. lipolytica, transition from unicellular yeast-type growth to filamentous hyphal growth is not pH dependent, contrary to what is observed in C. albicans (Fonzi, 2002). We notice that all Ylvps mutants tested show a strong defect in hyphal formation not suppressed by expression of the truncated, active form of YlRim101p, contrary to Ylrim mutants, which display near normal morphogenesis at all pH conditions tested (Gonzalez-Lopez et al., 2002
; data not shown). In contrast to this, all C. albicans rim or vps mutant strains present defects in hyphal formation and these defects are either totally, in Carim mutants, or at least partially, in Cavps mutants, bypassed by the expression of the constitutive form of CaRim101p (Kullas et al., 2004
; Xu et al., 2004
). Taken together, these results suggest that VPS genes might be required directly for hyphal growth in Y. lipolytica and probably also in other fungi, but that this function has been taken over by the Rim pathway in C. albicans. This may further explain why VPS genes were not identified during hunts for mutant strains defective in pH response in A. nidulans. Further work is thus needed to assess the involvement of VPS genes in pH sensing and control of cellular morphology in diverse fungal species.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Babst, M., Wendland, B., Estepa, E. J. & Emr, S. D. (1998). The Vps4p AAA ATPase regulates membrane association of a Vps protein complex required for normal endosome function. EMBO J 17, 29822993.
Babst, M., Katzmann, D. J., Estepa-Sabal, E. J., Meerloo, T. & Emr, S. D. (2002a). ESCRT-III: an endosome-associated heterooligomeric protein complex required for MVB sorting. Dev Cell 3, 271282.[CrossRef][Medline]
Babst, M., Katzmann, D. J., Snyder, W. B., Wendland, B. & Emr, S. D. (2002b). Endosome-associated complex, ESCRT-II, recruits transport machinery for protein sorting at the multivesicular body. Dev Cell 3, 283289.[CrossRef][Medline]
Barth, G. & Gaillardin, C. (1997). Physiology and genetics of the dimorphic fungus Yarrowia lipolytica. FEMS Microbiol Rev 19, 219237.[CrossRef][Medline]
Barwell, K. J., Boysen, J. H., Xu, W. & Mitchell, A. P. (2005). Relationship of DFG16 to the Rim101p pH response pathway in Saccharomyces cerevisiae and Candida albicans. Eukaryot Cell 4, 890899.
Blanchin-Roland, S., Cordero Otero, R. R. & Gaillardin, C. (1994). Two upstream activation sequences control the expression of the XPR2 gene in the yeast Yarrowia lipolytica. Mol Cell Biol 14, 327338.[Abstract]
Bowers, K., Lottridge, J., Helliwell, S. B., Goldthwaite, L. M., Luzio, J. P. & Stevens, T. H. (2004). Protein-protein interactions of ESCRT complexes in the yeast Saccharomyces cerevisiae. Traffic 5, 194210.[CrossRef][Medline]
Caddick, M. X., Brownlee, A. G. & Arst, H. N., Jr (1986). Regulation of gene expression by pH of the growth medium in Aspergillus nidulans. Mol Gen Genet 203, 346353.[CrossRef][Medline]
Dujon, B., Sherman, D., Fischer, G. & 64 other authors (2004). Genome evolution in yeasts. Nature 430, 3544.[CrossRef][Medline]
Fickers, P., Le Dall, M. T., Gaillardin, C., Thonart, P. & Nicaud, J. M. (2003). New disruption cassettes for rapid gene disruption and marker rescue in the yeast Yarrowia lipolytica. J Microbiol Methods 55, 727737.[CrossRef][Medline]
Fonzi, W. A. (2002). Role of pH response in Candida albicans virulence. Mycoses 45, 1621.[Medline]
Futter, C. E., Pearse, A., Hewlett, L. J. & Hopkins, C. R. (1996). Multivesicular endosomes containing internalized EGF-EGF receptor complexes mature and then fuse directly with lysosomes. J Cell Biol 132, 10111023.[Abstract]
Gavin, A. C., Bösche, M., Krause, R. & 35 other authors (2002). Functional organization of the yeast proteome by systematic analysis of protein complexes. Nature 415, 141147.[CrossRef][Medline]
Giaever, G., Chu, A. M., Ni, L. & 70 other authors (2002). Functional profiling of the Saccharomyces cerevisiae genome. Nature 418, 387391.[CrossRef][Medline]
Gonzalez-Lopez, C. I., Szabo, R., Blanchin-Roland, S. & Gaillardin, C. (2002). Genetic control of extracellular protease synthesis in the yeast Yarrowia lipolytica. Genetics 160, 417427.
Ito, T., Chiba, T., Ozawa, R., Yoshida, M., Hattori, M. & Sakaki, Y. (2001). A comprehensive two-hybrid analysis to explore the yeast protein interactome. Proc Natl Acad Sci U S A 98, 45694574.
Katzmann, D. J., Babst, M. & Emr, S. D. (2001). Ubiquitin-dependent sorting into the multivesicular body pathway requires the function of a conserved endosomal protein sorting complex, ESCRT-I. Cell 106, 145155.[CrossRef][Medline]
Katzmann, D. J., Odorizzi, G. & Emr, S. D. (2002). Receptor downregulation and multivesicular-body sorting. Nat Rev Mol Cell Biol 3, 893905.[CrossRef][Medline]
Kullas, A. L., Li, M. & Davis, D. A. (2004). Snf7p, a component of the ESCRT-III protein complex, is an upstream member of the RIM101 pathway in Candida albicans. Eukaryot Cell 3, 16091618.
Lamb, T. M., Xu, W., Diamond, A. & Mitchell, A. P. (2001). Alkaline response genes of Saccharomyces cerevisiae and their relationship to the RIM101 pathway. J Biol Chem 276, 18501856.
Lambert, M., Blanchin-Roland, S., Le Louedec, F., Lepingle, A. & Gaillardin, C. (1997). Genetic analysis of regulatory mutants affecting synthesis of extracellular proteinases in the yeast Yarrowia lipolytica: identification of a RIM101/pacC homolog. Mol Cell Biol 17, 39663976.[Abstract]
Lemmon, S. K. & Traub, L. M. (2000). Sorting in the endosomal system in yeast and animal cells. Curr Opin Cell Biol 12, 457466.[CrossRef][Medline]
Li, Y., Kane, T., Tipper, C., Spatrick, P. & Jenness, D. D. (1999). Yeast mutants affecting possible quality control of plasma membrane proteins. Mol Cell Biol 19, 35883599.
Muhlschlegel, F. A. & Fonzi, W. A. (1997). PHR2 of Candida albicans encodes a functional homolog of the pH-regulated gene PHR1 with an inverted pattern of pH-dependent expression. Mol Cell Biol 17, 59605967.[Abstract]
Odorizzi, G., Katzmann, D. J., Babst, M., Audhya, A. & Emr, S. D. (2003). Bro1 is an endosome-associated protein that functions in the MVB pathway in Saccharomyces cerevisiae. J Cell Sci 116, 18931903.
Ogrydziak, D. M. (1993). Yeast extracellular proteases. Crit Rev Biotechnol 13, 155.[Medline]
Ogrydziak, D. M., Demain, A. L. & Tannenbaum, S. R. (1977). Regulation of extracellular protease production in Candida lipolytica. Biochim Biophys Acta 497, 525538.[Medline]
Peñalva, M. A. & Arst, H. N., Jr (2002). Regulation of gene expression by ambient pH in filamentous fungi and yeasts. Microbiol Mol Biol Rev 66, 426446.
Peñalva, M. A. & Arst, H. N., Jr (2004). Recent advances in the characterization of ambient pH regulation of gene expression in filamentous fungi and yeasts. Annu Rev Microbiol 58, 425451.[CrossRef][Medline]
Ramon, A. M., Porta, A. & Fonzi, W. A. (1999). Effect of environmental pH on morphological development of Candida albicans is mediated via the PacC-related transcription factor encoded by PRR2. J Bacteriol 181, 75247530.
Read, N. D. & Kalkman, E. R. (2003). Does endocytosis occur in fungal hyphae? Fungal Genet Biol 39, 199203.[CrossRef][Medline]
Rieder, S. E., Banta, L. M., Kohrer, K., McCaffery, J. M. & Emr, S. D. (1996). Multilamellar endosome-like compartment accumulates in the yeast vps28 vacuolar protein sorting mutant. Mol Biol Cell 7, 985999.[Abstract]
Serrano, R., Bernal, D., Simon, E. & Arino, J. (2004). Copper and iron are the limiting factors for growth of the yeast Saccharomyces cerevisiae in an alkaline environment. J Biol Chem 279, 1969819704.
Tréton, B., Blanchin-Roland, S., Lambert, M., Lépingle, A. & Gaillardin, C. (2000). Ambient pH signalling in ascomycetous yeasts involves homologues of the Aspergillus nidulans genes palF and paIH. Mol Gen Genet 263, 505513.[CrossRef][Medline]
Vida, T. A. & Emr, S. D. (1995). A new vital stain for visualizing vacuolar membrane dynamics and endocytosis in yeast. J Cell Biol 128, 779792.[Abstract]
Vincent, O., Rainbow, L., Tilburn, J., Arst, H. N., Jr & Peñalva, M. A. (2003). YPXL/I is a protein interaction motif recognized by aspergillus PalA and its human homologue, AIP1/Alix. Mol Cell Biol 23, 16471655.
Xu, W. & Mitchell, A. P. (2001). Yeast PalA/AIP1/Alix homolog Rim20p associates with a PEST-like region and is required for its proteolytic cleavage. J Bacteriol 183, 69176923.
Xu, W., Smith, F. J., Jr, Subaran, R. & Mitchell, A. P. (2004). Multivesicular body-ESCRT components function in pH response regulation in Saccharomyces cerevisiae and Candida albicans. Mol Biol Cell 15, 55285537.
Received 13 May 2005;
revised 20 July 2005;
accepted 27 July 2005.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
J MED MICROBIOL | ALL SGM JOURNALS |