Efficient sterol influx in the yeast Saccharomyces cerevisiae is restricted to anaerobiosis or to haem deficiency resulting from mutations. Constitutive expression of SUT1, an hypoxic gene encoding a transcriptional regulator, induces sterol uptake in aerobiosis. A genome-wide approach using DNA microarray was used to identify the mediators of SUT1 effects on aerobic sterol uptake. A total of 121 ORFs (open reading frames) were significantly and differentially expressed after SUT1 overexpression, 61 down-regulated and 60 up-regulated. Among these genes, the role of the putative ABC transporter (ATP-binding-cassette transporter) Aus1, and of the cell-wall mannoprotein Dan1, was characterized better. These two genes play an essential role in aerobic sterol uptake, since their deletion compromised the SUT1 effects, but individual overexpression of either of these genes in a wild-type background was not sufficient for this process. However, constitutive co-expression of AUS1 and DAN1 in a wild-type background resulted in sterol influx in aerobiosis. These results suggest that the corresponding proteins may act synergistically in vivo to promote sterol uptake.
- ATP-binding-cassette transporter (ABC transporter)
- sterol uptake
Budding yeast Saccharomyces cerevisiae can alternatively utilize respiration or fermentation for its metabolic requirements. Its adaptation to aerobiosis and anaerobiosis is achieved by the differential expression of a large number of genes whose regulation depends on oxygen availability [1–3].
Biosynthesis of sterols and unsaturated fatty acids are strict aerobic processes in this organism. Although sterol synthesis is an energy-consuming process, under aerobiosis, yeast does not take up a significant amount of exogenous sterols. In contrast, under anaerobiosis, when sterol biosynthesis is compromised, cells become capable of importing these essential molecules, provided that they are present in the medium. This physiological phenomenon is generally referred to as aerobic sterol exclusion and its regulation is at least partly mediated by haem availability .
In contrast with lipoprotein-mediated sterol uptake, free sterol influx by eukaryotic cells is still poorly understood. This saturable process depends on many factors that are not yet formally identified. However, de-regulation of two transcriptional regulators has been described as favouring exogenous sterol accumulation in aerobiosis.
Indeed, Lewis et al.  isolated a yeast mutant, upc2-1 (uptake control), which leads to aerobic sterol uptake in haem- and ergosterol-competent cells. It was suggested that the upc2-1 allele corresponds to a gain-of-function mutation, since deletion of UPC2 does not result in sterol uptake. Upc2 contains a Zn(II)2Cys6 dinuclear cluster DNA-binding domain, a motif unique to fungal proteins shared by transcriptional regulators. Upc2, and its paralogue Ecm22, activate transcription through binding to sterol-regulatory elements of ergosterol biosynthesis genes [6,7]. Therefore Upc2-mediated sterol uptake in yeast results from altered transcriptional regulation of one or more target genes of Upc2. Recently, Wilcox et al.  used a genome-wide approach to identify all the putative Upc2 target genes. Starting from 82 genes induced in the upc2-1 mutant, only nine genes were identified as putative actors of sterol uptake. Among these candidates, deletion of DAN1, a gene of the DAN/TIR family, or inactivation of YOR011W (subsequently named as AUS1), and its close paralogue PDR11, which encode putative ABC transporters (ATP-binding-cassette transporters), significantly decreased aerobic sterol influx in the upc2-1 background, although to variable extents. Most interestingly, the combination of the two mutations aus1Δpdr11Δ severely compromised growth in anaerobiosis, thereby suggesting that they play a key role in exogenous sterol uptake .
A second transcription factor involved in sterol uptake has been identified. Assuming that the genes involved in sterol uptake are down-regulated in aerobiosis, a high-copy-number library was screened in our laboratory to identify genes whose overexpression may overcome aerobic sterol exclusion. The only gene identified using this approach was SUT1 . Deletion of this hypoxic gene does not compromise growth, irrespective of oxygen availability. Its overexpression under the control of a strong constitutive promoter results in a high increase (up to 18-fold) in sterol influx in aerobiosis . SUT1 encodes a nuclear protein which, similar to Upc2, belongs to the Zn(II)2Cys6 family. However, in contrast with Upc2, Sut1 does not appear to bind to DNA directly. Rather, this protein acts at the transcriptional level by relieving hypoxic genes from Cyc8–Tup1 repression through its physical interaction with Cyc8 . Cyc8–Tup1 acts as a co-repressor complex that is recruited to many promoters via specific interactions with DNA-binding regulatory proteins , such as Rox1, involved in the repression of many hypoxic genes . SUT1 transcription is induced in anaerobiosis or in haem-deficient strains; although putative Rox1-binding sites reside in its promoter, SUT1 appears to be only moderately down-regulated by Rox1 .
It has been reported that in some cases, the Cyc8–Tup1 protein complex can shift from a transcriptional co-repressor to a transcriptional co-activator [14–16]. In a previous work, we have reported that in the presence of Sut1, the Cyc8–Tup1 complex is also converted into a transcriptional co-activator, resulting in the induction of several hypoxic genes . The present study allowed us to characterize better the transcriptional effects of Sut1, but we failed to identify components of free sterol influx, with the exception of CSR1, a SEC14 homologue, whose overexpression promotes increased sterol influx in aerobiosis .
In the present work, we performed a genome-wide transcriptional analysis to identify all the genes differentially transcribed after SUT1 constitutive expression by comparison with the wild-type in aerobiosis. We show that co-expression of DAN1 and AUS1, already identified as being involved in sterol uptake , is sufficient to promote the uptake of sterol in aerobiosis, indicating that they may act synergistically in vivo to promote sterol entry. However, the massive sterol influx observed in anaerobiosis, in haem-depleted cells or as a result of upc2-1 mutation or SUT1 overexpression, may still require other as yet uncharacterized components.
Strains and plasmids
The yeast strains and plasmids used throughout the present study are listed in Tables 1 and 2 respectively. Strains were grown in standard minimal media [YNB: 1.7 g/l Difco™ yeast nitrogen base without amino acids and ammonium sulphate (Difco Laboratories), 5 g/l ammonium sulphate and 1% D-glucose], supplemented with the appropriate amino acids and/or bases for plasmid selection. Fenpropimorph resistance was analysed on YPD solid medium, containing 1 μg/ml (31.5 μM) fenpropimorph and 80 μg/ml ergosterol solubilized in Tergitol NP40 (Nonidet P40)/ethanol (1:1, v/v). When ergosterol was omitted, Tergitol NP40/ethanol (1:1, v/v) was added to a final concentration of 2%. The media used for anaerobic growth were supplemented with 1% (v/v) Tween 80 (polyethylene sorbitan mono-oleate) and 80 μg/ml ergosterol in Tergitol NP40/ethanol (1:1, v/v). Anaerobic conditions were obtained with the BBL GasPak Anaerobic System (Becton Dickinson) in an anaerobic chamber.
RNA extraction and mRNA purification
Cells were grown in selective medium, harvested in mid-exponential phase, washed with RNase-free water and immediately frozen in liquid nitrogen. Cells were broken using TRIzol® reagent (Invitrogen). After extraction, RNAs were precipitated with isopropyl alcohol, washed with 75% (v/v) ethanol and resuspended in RNase-free water. Poly(A)+ (polyadenylated) RNAs were purified using PolyATract mRNA Isolation System IV (Promega). For Northern-blot analysis, we performed electrophoresis, blotting and hybridization with total RNA as described previously .
Generation of Cy3- and Cy5-labelled cDNA probes
Double-strand labelled cDNA was obtained by in vitro transcription using CyScribe first-strand cDNA-labelling kit (Amersham Biosciences) with 1.5 μg of poly(A)+ RNA. Reverse transcription was performed in the presence of CyScript reverse transcriptase. RNAs were hydrolysed with 2.5 M NaOH at 37 °C for 15 min. Probes were purified using the QIAquick nucleotide removal kit (Qiagen). For each experimental condition, an independent set of probes was generated.
Microarray hybridization and scanning
S. cerevisiae microarray slides (Eurogentec, Angers, France) were used. For hybridization, 10 μl of probe, 30 μl of DIG Easy Hyb buffer (Boehringer Mannheim, Mannheim, Germany) and 2 μl (10 mg/ml) of salmon sperm DNA were mixed. Microarray experiments with FY1679α+[pNEV-N] and FY1679α+[pNF1] were performed twice with each pair of probes (Cy3 or Cy5 label), which were reversed for each experiment (generating four sets of data). The mixture was heated for 5 min at 95 °C and centrifuged at 12000 g for 2 min. Hybridization was performed at 37 °C overnight. After hybridization, the arrays were washed, dried and subsequently scanned with a GenePix 4000B microarray scanner (Axon Instruments, Foster City, CA, U.S.A.).
The GenePix 4000B software package (Axon Instruments) was used to locate spots in the microarrays. Taking into account the numerous intrinsic variations of transcriptional profiling on arrays, a statistical analysis was performed to ensure the significance of the conclusions drawn from the data. Signal intensities of each hybridization were first normalized by the method of Yang et al.  to remove the variation due to differences in the labelling between the two fluorescent dyes. For the statistical analysis, the normalized data for four independent hybridizations were compared using the tpWY software, allowing us to calculate an adjusted P value .
Exogenous respiration was determined with cells grown on selective medium (YNB) for 24 h. Cells were suspended in sterile distilled water to a density of 1.6×108 cells/ml. Enumeration of cells was performed by hematocytometer counts. For each measurement, 1.6×107 cells were suspended in 2 ml of oxygen-saturated water. Oxygen consumption was measured at 30 °C using an oxygraph (Hansatech Instruments, Norfolk, U.K.). Respiration rates are expressed as nmol of oxygen consumed ml−1·min−1. Statistical significance of differences in oxygen consumption was assessed by the Mann–Whitney test.
Analysis of sterol uptake
Sterol uptake was measured essentially as described in , from independent cultures (n=7) grown for 24 h in 5 ml of selective medium containing 26 μM [4-14C]cholesterol (2600 d.p.m./μg) prepared from a stock supplied by NEN (Boston, MA, U.S.A.; at 46 mCi/mmol). Samples were collected by removing 4 ml of culture. After centrifugation, the cells were washed four times with 0.5% (v/v) Tergitol NP40 and once with water. The net accumulation of [4-14C]cholesterol in the cells was measured by scintillation counting and is expressed as μg of cholesterol/mg dry weight.
UPC2 inactivation does not compromise the SUT1 effect on sterol uptake
Since strains bearing either the gain-of-function upc2-1 mutation, or constitutively expressing SUT1, both promote free sterol influx in aerobiosis, we wondered whether SUT1 may indirectly promote sterol uptake through activation of the UPC2 gene. To rule out this possibility, plasmid pNF1 (SUT1) was introduced both in a upc2Δ strain and in the isogenic wild-type. As a control, the same strains were transformed with the empty plasmid (pNEV-N). Sterol import was assayed by spotting serial dilutions of transformants on a fenpropimorph-containing medium supplemented with ergosterol, under standard aerobic conditions. This assay has been widely used in previous studies to evaluate the ability of the cells to import exogenous sterol in aerobiosis, thereby enabling them to resist the inhibitor [9,10,17]. As a control, cell suspensions were also plated on to a medium containing the inhibitor, but no ergosterol, to assess that growth was not due to an ergosterol-independent resistance to the inhibitor. In addition, this control allowed us to evaluate spontaneous fenpropimorph-resistant strains that arise commonly within the most concentrated suspensions: however, since the phenotype of these mutants is ergosterol-independent, they can be easily discriminated from cells that do import ergosterol (Figure 1). From these experiments, we observed that UPC2 inactivation does not impair SUT1-induced sterol import under these conditions significantly (Figure 1).
This result indicates that the effect on sterol uptake induced by constitutive expression of SUT1 is not mediated by UPC2. Hence, we decided that it will be more appropriate to characterize the SUT1 effects on transcription on a genome-wide level.
Genome-wide analysis of SUT1 effects on transcription
We have reported that Sut1 is a nuclear protein whose constitutive expression alters the transcriptional regulation of several genes [10,11,17]. In particular, we observed that hypoxic genes were up-regulated in aerobiosis as a result of SUT1 overexpression. To achieve an exhaustive identification of all the genes whose transcription is altered after SUT1 overexpression, we performed a DNA microarray-assisted screening. For this purpose, we compared the transcriptional profile of a wild-type strain (FY1679α) with the same strain transformed with pNF1, a high-copy-number plasmid in which SUT1 expression is driven by the strong constitutive PMA1 promoter . Both strains were grown under standard aerobic conditions and RNAs were isolated from mid-exponential phase cultures. ORFs (open reading frames) with differential expressions were detected from four experiments with reverse labelling. Only ORFs with a significant transcriptional alteration (adjusted P<0.05) were selected, irrespective of the signal intensity relative to the control condition (Tables 3 and 4, repressed and induced ORFs respectively). A total of 121 ORFs were differentially regulated after SUT1 overexpression; 61 of them were down-regulated and 60 were up-regulated. Changes in expression level varied from 1.4-fold up to 298-fold, but most genes exhibited a ratio of differential expression between 2 and 10. Many of the ORFs whose transcription is decreased are directly or indirectly related to mitochondrial functions (Table 3). Down-regulation of these genes may explain why the growth of SUT1-overexpressing strains is slower compared with the wild-type, as a consequence of a defect in the respiratory metabolism.
To address this question, cell suspensions of the wild-type strain (YPA3), transformed with pNF1 (SUT1) or with the control plasmid (pNEV-N), were spotted on to a medium containing a non-fermentable carbon source (glycerol/ethanol selective medium). As seen in Figure 2(A), non-fermentative growth of the SUT1-overexpressing strain was severely impaired. To confirm this result, oxygen consumption of the cells was measured. The respiration rate of the cells transformed with pNF1 was decreased by approx. 2.2-fold (Mann–Whitney test, P <0.05, n=3) when compared with the control strain (Figure 2B). Taken together, these results, which indicate that respiration is impaired after SUT1 overexpression, are in good agreement with the microarray data.
Other essential metabolic function(s) may be affected by SUT1-induced de-regulation of the ORFs listed in Table 3. We have already reported that SUT1 constitutive expression has a general negative effect on transcription . Therefore we decided to focus on the genes that were activated (listed in Table 4).
Our results were compared with those obtained by Kwast et al.  in a study aimed at investigating the specific role of Rox1 in the adaptation to anaerobiosis , because our experimental conditions (batch cultures) were similar to those used by these authors. Among the genes induced by SUT1 overexpression, 29 genes were identified as hypoxic genes, of which 13 were negatively regulated by Rox1 . Altogether, these results are in full agreement with our previous report in which we have shown that SUT1 constitutive expression relieved hypoxic genes from Cyc8–Tup1-mediated repression in aerobiosis . It is noteworthy that, because of the stringency of our statistical analysis, some genes may have been discarded from our selection. Indeed, in a previous study, we identified CSR1 as a clear SUT1 target , besides HES1 (M. Régnacq, unpublished work), but neither of these genes were selected in our microarray screening.
Among the up-regulated ORFs, the DAN (delayed anaerobic, cell wall mannoprotein induced during anerobic growth)/TIR (TIP1-related, cell wall mannoprotein) mannoproteins are widely represented (approx. 1/4). At least five other ORFs are also related to cell-wall modelling. Indeed, YGL028c (SCW1) and YEL040w (UTR2) encode glucanases, and YNR067c is similar to another bacterial glucanase, whereas YKL008c (LAC1) is involved in the transport of GPI (glycosylphosphatidylinositol)-anchored proteins. The function of YLR413w is still unknown, but it is highly similar to other members of a family of proteins involved in the elongation of fatty acids, which are themselves components of sphingolipids and GPI anchors. Several characterized or putative transporters are also up-regulated. YLL061w (MMP1), YPL274w (SAM3), YBR294w (SUL1), YHL016c (DUR3) and YLL055w belong to the major facilitator superfamily, whereas YOR011w (AUS1) belongs to the ABC transporter superfamily. So far, with the exception of Aus1 (see below), the function of these proteins has never been attributed to sterol uptake and/or metabolism.
A similar approach has been reported using upc2-1 strains in which free sterol influx is stimulated in aerobiosis . We reasoned that the genes that are most probably involved directly in sterol uptake should lie among those being regulated by upc2-1 and SUT1. We thus compared their respective effects on transcription. A list of the ORFs regulated by both SUT1 and upc2-1 is presented in Table 5. All the 15 candidates have been shown to be up-regulated by the upc2-1 gain-of-function mutation . Most of them correspond to DAN/TIR mannoproteins. Wilcox et al.  identified Aus1 and Pdr11, members of the ABC transporter superfamily, and Dan1, a member of the DAN/TIR mannoprotein family, as being involved in sterol influx, since deletion of either of theses genes impairs sterol uptake. Interestingly, AUS1 and DAN1, but not PDR11, are also significantly up-regulated after SUT1 constitutive expression in our screening (adjusted P <0.001 for both of them). This result was also confirmed by Northern-blot analysis (results not shown). Among the ORFs whose expression is commonly perturbed by upc2-1 and SUT1, CYB5-encoding cytochrome b5 and YPL272c (unknown function) are other potentially interesting candidates, since they were reported in a recent study to be also induced after azole exposure . Their putative function in sterol uptake has not been examined so far.
AUS1 and DAN1 are required for sterol uptake induced by SUT1
To characterize further the effects of AUS1 and DAN1 on sterol import, SUT1 constitutive expression was assayed in an AUS1-deleted strain and in a DAN1-deleted strain. These strains were transformed with Yep13-SUT1 and tested for fenpropimorph resistance in the presence of ergosterol; the strains were then compared with the isogenic wild-type harbouring the same plasmid. As shown in Figure 3 AUS1 or DAN1 deletion abolished SUT1 effects on sterol import in aerobiosis. As a control, the same strains were transformed with a plasmid carrying the wild-type AUS1 or DAN1 allele. The corresponding strains recovered their resistance to fenpropimorph. Therefore both AUS1 and DAN1 are required for SUT1-induced sterol import in aerobiosis.
AUS1 constitutive expression mediates sterol uptake in a rox1Δ strain
AUS1 encodes a putative ABC transporter. As shown by Wilcox et al. , this gene is necessary for sterol uptake and inhibition of ATP production compromises free sterol influx. However, so far, no direct positive role in sterol uptake has been reported for this gene. If Aus1 protein is directly involved in exogenous sterol uptake, its constitutive expression should result in enhanced sterol influx in aerobiosis. To test this hypothesis, we made a construct in which the AUS1 ORF is under the control of the constitutive PMA1 promoter, on a replicative plasmid (pNEV-AUS1). This construct was introduced into a wild-type strain (YPA3), which was subsequently challenged for growth in the presence of fenpropimorph and ergosterol. The sensitivity to the inhibitor did not differ from that of the control untransformed strain (results not shown). These results indicate that Aus1 is necessary but not sufficient to promote sterol uptake and Aus1 may require additional protein(s). Since AUS1 does not seem to be down-regulated by Rox1 in contrast with several of the SUT1-regulated genes , we tested the possibility that the proteins co-operating with Aus1 for sterol uptake could be negatively regulated by Rox1. Therefore we introduced pNEV-AUS1 into an ROX1-deleted strain. Interestingly, in this genetic background, a moderate but significant fenpropimorph resistance related to sterol entry was observed, whereas the untransformed isogenic strain remains sensitive to the inhibitor (Figure 4). This is observed only if ergosterol is added to the medium. Notably, this effect is lost in a rox1Δdan1Δ background (Figure 4), indicating that DAN1 is required for assisting AUS1 to induce sterol uptake in aerobiosis.
AUS1 and DAN1 co-expression results in enhanced sterol influx
We then questioned whether constitutive co-expression of AUS1 and DAN1 might be sufficient to mediate sterol uptake in aerobiosis. For this purpose, we co-transformed the wild-type YPA3 strain with two plasmids, each containing one of the two genes under the control of the PMA1 constitutive promoter. In a first step, sterol entry was tested using the fenpropimorph assay. Colonies were obtained, whereas the control untransformed strain did not grow at all; however, the size of the colonies was much smaller than those obtained with pNF1 (SUT1 overexpressed), which was used as a positive control (results not shown). Quantification of sterol uptake using [4-14C]cholesterol confirmed that co-expression of AUS1 and DAN1 in aerobiosis increases sterol influx approx. 1.7-fold compared with the wild-type (Table 6). Although moderate, this sterol influx is highly reproducible and its significance was confirmed by a statistical analysis (Mann–Whitney test, P <0.05, n=7). However, exogenous sterol accumulation remained much lower than the SUT1 effect (>18-fold stimulation), suggesting that additional factors are required to promote full permeation.
Despite intensive investigations performed in several laboratories, the molecular mechanisms allowing sterol uptake in anaerobiosis, and their exclusion in aerobiosis, are still poorly understood. So far, only a few genes involved in sterol influx have been characterized, two of them encoding transcriptional regulators, namely UPC2 and SUT1. Their de-regulation, as a result of upc2-1 gain-of-function mutation  or SUT1 constitutive overexpression , is capable of promoting free sterol influx. Since both of them perturb the transcription of several genes, this suggests that exogenous sterol accumulation probably requires a cell competency ensured by several proteins whose genes are normally down-regulated in aerobiosis. Using a genome-wide transcriptional analysis, Wilcox et al.  have identified three Upc2 target genes whose deletion partly abolished upc2-1 effects on free sterol influx, indicating that they are either directly or indirectly involved in this process . One of these genes, DAN1, encodes a mannoprotein that belongs to a family of stress proteins, and its role in exogenous sterol accumulation is still puzzling. Dan1 is a putative cell-wall protein with a signature of a GPI anchor at its C-terminus. The putative role of AUS1, and of its close homologue PDR11, seems easier to understand. Indeed, their primary sequence indicates that the corresponding proteins belong to the family of ABC transporters. These proteins mediate the ATP-dependent efflux of a wide variety of molecules such as xenobiotic substrates (MDR, multi-drug resistance protein; ) or even lipids (ABC transporter A1, involved in cellular cholesterol efflux in mammals; ). Although it was reported that, consistent with the putative mode of action of Aus1 and Pdr11, the use of energy-depleting molecules abolished upc2-1 effect on sterol influx , the direct implication of these two proteins in sterol uptake has not been formally proven so far. In two recent reports, it has been proposed that ABC transporters are involved in the influx of lipophilic molecules as well. Indeed CjMDR1, an MDR-type ABC transporter isolated from the plant Coptis japonica, was capable of transporting the alkaloid berberine in the inward direction when expressed in a Xenopus oocyte expression system, and this observation is in good agreement with its putative role in planta . It has also been reported that Cdr3, an ABC transporter of Candida albicans, mediated out-to-in translocation of a phospholipid analogue between the two bilayers of the plasma membrane . However, the in vivo function of these proteins has not been clearly established so far.
Since UPC2 and SUT1 have almost similar effects on exogenous sterol accumulation in aerobiosis, we first examined whether SUT1 effect on sterol influx might be mediated indirectly by UPC2. This does not appear to be the case since deletion of UPC2 did not abolish SUT1 effects (Figure 1). Therefore we decided to identify the genes whose transcription is modified after SUT1 overexpression, in a genome-wide screening using microarrays. For this purpose, we compared the transcriptional profiling of a strain overexpressing SUT1 with that of the wild-type, both of which were cultivated in aerobiosis. This led us to identify 121 ORFs differentially regulated in the two strains. This relatively large number might be explained by the fact that the analysis was performed with RNA extracted at a single time point, when the culture had reached the mid-exponential phase. Thus it is not excluded that the differential regulation of at least some of these genes is the result of indirect cumulative effects of SUT1 overexpression. However, we noticed that among these genes, 15 are also differentially regulated in the upc2-1 strain (Table 5; ). Considering the similar effects of these two genes on free sterol influx, we reasoned that the most interesting candidates for controlling this process were genes that are commonly regulated by both the transcription factors. We thus focused more particularly on DAN1 and AUS1 to characterize further their role.
We confirmed that they do mediate SUT1 effect on exogenous sterol accumulation, since the individual inactivation of either of these genes abolished SUT1-induced fenpropimorph resistance on a medium containing ergosterol, in aerobiosis (Figure 3). These results are in perfect agreement with the observations of Wilcox et al. . No effect could be observed on exogenous sterol accumulation when these genes were overexpressed individually, indicating that this process requires more than a single protein. This is in good agreement with the results of genetic screenings performed so far.
Many but not all (e.g. AUS1) hypoxic genes are down-regulated by Rox1 in aerobiosis. Based on the fact that a ROX1-deleted strain is unable to import sterols in aerobiosis, we hypothesized that this phenomenon requires both Rox1-independent and -dependent proteins. Interestingly, MOT3, a gene encoding a transcriptional regulator which has been shown recently to act in synergy with Rox1 to repress anaerobic genes , was negatively regulated after SUT1 overexpression (Table 3). We decided to overexpress AUS1 in a ROX1-deleted background. All the clones tested were capable of growing on a fenpropimorph-containing medium, supplemented with ergosterol, demonstrating an efficient exogenous sterol accumulation in aerobiosis (Figure 4). This phenotype required the presence of a wild-type copy of DAN1.
This latter observation prompted us to overexpress both AUS1 and DAN1 in a single wild-type strain. Although growth on fenpropimorph was low, radiolabelled cholesterol accumulated significantly and reproducibly in this strain (Table 6). The level of sterol influx does not reach that obtained in a haem-incompetent strain or in a strain overexpressing SUT1, indicating that the cells are not fully permeable. However, it must be stressed that the uptake (1.7-fold increase) is close to that obtained when SUT1 expression is directed by its own promoter , a level which allowed the cloning of this gene .
In conclusion, both deletion and overexpression of AUS1 and DAN1 support the notion that these proteins are key components of sterol import in yeast. However, their precise contribution is still not clear. Particularly, it is not yet proven that any of the actors identified correspond to a sterol transporter; so far, with the exception of transcriptional regulators such as SUT1 or UPC2, no single gene capable of promoting full permeation in aerobiosis has been identified. This shows that this process requires additional components. Further investigation on the results of de-regulating the genes identified in the present study would be helpful to address their functional implications in sterol import.
We thank Dr A. Maitournam (Institut Pasteur, Paris, France) for his help with the statistical analysis of the microarray data and Dr R. Zitomer (Albany, New York, U.S.A.) for the gift of the RZ53-6Δrox1 strain. We acknowledge V. Ansaney (UMR Sciences Pour l'œnologie, Montpellier, France) for her help in performing the microarray experiments and F. Agasse (UMR 6187, Laboratoire de Biomembranes et Signalisation Cellulaire, Poitiers, France) and M. Faucher (UMR 6161, Poitiers, France) for their help and advice. This work was supported by Ministère de l'Education Nationale, de la Recherche et de la Technologie, France and CNRS. P.A. was supported by a grant from Region Poitou-Charentes.
Abbreviations: ABC transporter, ATP-binding-cassette transporter; GPI, glycosylphosphatidylinositol; NP40, Nonidet P40; ORF, open reading frame
- The Biochemical Society, London