Characterizing Sterol Defect Suppressors Uncovers a Novel Transcriptional Signaling Pathway Regulating Zymosterol Biosynthesis *

erg26-1 ts cells harbor defects in the 4 (cid:1) -carboxysterol-C3 dehy-drogenase activity necessary for conversion of 4,4-dimethylzymos-terol to zymosterol. Mutant cells accumulate toxic 4-carboxysterols and are inviable at high temperature. A genetic screen aimed at cloning recessive mutations remediating the temperature sensitive growth defect has resulted in the isolation of four complementation groups, ets1-4 ( erg26-1 ts temperature sensitive suppressor). We describe the characterization of ets1-1 and ets2-1 . Gas chromatog-raphy/mass spectrometry analyses demonstrate that erg26-1 ts ets1-1 and erg26-1 ts ets2-1 cellsdonotaccumulate4-carboxysterols, rather these ets1-1 and ets2-1 intermediates. Chromosomal integration of ERG1 and ERG7 their loci in erg26-1 ts ets1-1 and erg26-1 ts ets2-1 mutants, in the loss of accumulation of squalene and squalene epoxide, re-accumulation of 4-carboxysterols and cell inviability high temperature. Enzymatic assays demonstrate that mutants

Sterols play crucial roles in the physiology of all eukaryotes. They are essential for membrane fluidity and function (1,2) and have emerged as important second messenger lipids involved in proper developmental signaling (3,4). The vertebrate sterol end product, cholesterol, can be synthesized de novo from acetyl coenzyme A through a multistep pathway (2), or obtained through the uptake of exogenous dietary choles-terol via the low density lipoprotein receptor (5,6). An exquisite transcriptional signaling pathway comprised of a specific promoter SRE 3 and various SREBP isoforms regulates mammalian sterol gene expression in response to changes in sterol levels, thus maintaining cholesterol homeostasis (7). HMG-CoA reductase is the rate-limiting enzyme in higher eukaryotic sterol biosynthesis (2). It is regulated both at the transcriptional and protein stability levels in response to changes in sterol levels. The protein is rapidly degraded by a proteasome-dependent mechanism when sterol levels are high (8,9).
The sterol biosynthetic pathway in the budding yeast Saccharomyces cerevisiae is highly conserved with mammals (10,11). The only difference is that ergosterol is synthesized as the end product sterol rather than cholesterol. All of the genes have been cloned that are required for the biosynthesis of yeast ergosterol (11)(12)(13). Additional regulators of sterol metabolism and transport have been isolated and characterized, such as Erg28p (14) and the putative lipid transporter, Arv1p (15,16). Physiological characterization of various erg mutants has revealed roles for yeast sterols in endocytosis (17), lipid raft formation and function (18,19), cation and amino acid uptake (19,20), and cell cycle regulation (21).
Yeast sterols have the ability to regulate endogenous sterol biosynthesis, presumably by a transcriptional feedback mechanism. Genes regulated by sterol availability include those encoding HMG-CoA reductase (HMG1), squalene synthase (ERG9), sterol C-5 desaturase (ERG3), and acetoacetyl-CoA thiolase (ERG10) (10) (Fig.  1). The ERG10 promoter contains an active SRE-like element that has seven of the eight nucleotides that are found in the core mammalian SRE-1 consensus sequence (22). In addition, the ERG3 promoter contains a 22-bp region that is involved in the sterol regulation of transcription (23). Vik and Rine (24) have uncovered a 7-bp SRE that regulates ERG2 (sterol C-8 isomerase) and ERG3, which is shared among several ERG genes, as well as other genes needed for lipid metabolism, including the LCB1/2 serine palmitoyltransferase genes, which are required for the first step in sphingolipid biosynthesis (10). The Upc2p and Ecm22p transcription factors bind to this SRE and most likely regulate sterol responsive gene transcription (24). Thus, the paradigm of transcriptional feedback regulation by a sterol product extends to S. cerevisiae, making this single cell eukaryote a highly useful tool in which to study the transcriptional regulation of sterol gene expression in eukaryotes.
We recently isolated a yeast strain, designated erg26-1 ts , which harbors a ts allele of ERG26 (25). Erg26p (4␣-carboxysterol-C3 dehydrogenase) (26), along with Erg25p (sterol-C-4 methyl oxidase) (27) and Erg27p (C-3 ketoreductase) (28), is required for the conversion of 4,4-dimethylzymosterol to zymosterol in the sterol biosynthetic pathway (2,10). erg26-1 ts cells accumulate 4-carboxysterols, but more importantly also harbor defects in LCB and complex sphingolipid metabolism (25), which suggests that sterol and sphingolipid biosyntheses may be coordinately regulated in yeast. Multiple studies in mammalian cell systems strongly suggest cross-talk regulation between sterol and sphingolipid pathways in higher eukaryotes (29 -33). To further explore whether yeast have transcriptional and posttranslational coordinate regulatory pathways governing sterol and sphingolipid homeostasis, we isolated and characterized recessive mutations that would rescue the inviability of erg26-1 ts cells at high temperature. Four complementation groups were isolated, designated ets1-4. Our data demonstrate that ets1-1 and ets2-1 gene groups harbor recessive mutations in the ERG1 squalene epoxidase and ERG7 oxidosqualene cyclase genes, respectively. Using these mutant strains, we present evidence of a novel transcriptional signaling pathway, which coordinately regulates the transcription of several ERG genes at multiple points in the sterol pathway.

EXPERIMENTAL PROCEDURES
Strains, Media, and Miscellaneous Microbial Techniques-The parental yeast strains used in this study are W303 (MATa leu2-3,112 trp1-1 ura3-1 his3-11,15 can1-100) or BY4741 (MATa his3 leu2 met15 ura3). Yeast strains were grown in either YEPD (1% yeast extract, 2% bacto-peptone, 2% glucose) or in synthetic minimal medium containing 0.67% Yeast Nitrogen Base (Difco) supplemented with the appropriate amino acids, adenine, and uracil. Antifungal drugs were either added directly to liquid YEPD medium or added to YEPD plates at the indicated concentrations. Lovastatin treatment was for 16 h, fluconazole, amphotericin B, and nystatin treatment was for 3 h. Yeast transformation was performed using the procedure described by Ito et al. (34). For routine propagation of plasmids, Escherichia coli XL1Blue cells were used and grown in LB medium supplemented with ampicillin (150 g/ml). Bacterial transformations were carried out by electroporation methods.
Isolation of erg26-1 ts ts Suppressors-To obtain recessive ts suppressors, MATa and MAT␣ haploid erg26-1 ts cells were streaked onto prewarmed YEPD plates and grown for 3-6 days at 37°C. 29 MATa and 30 MATa ts revertants were obtained. All individual MATa revertants were mated to each MATa and diploids were tested for growth at 27°C and 37°C. MATa/MAT␣ diploids that grew at 37°C were considered homozygous for a single mutation suppressing the ts phenotype. Complementation groups were then assigned based on the results of the diploid growth assays. To obtain haploid strains harboring only the erg1-1/ets1-1 or erg7-1/ets2-1 allele, first a single haploid from each complementation group was selected, and the erg26-1 ts allele was replaced with erg26-1 ts ::erg26-1 ts ::URA3 in order to be able to track the ts allele through genetic crosses. Second, the resulting haploid strain was backcrossed to the opposite wild-type mating partner, the diploid was then induced to sporulate and haploid spores were selected based on their harboring the erg26-1 ts ::erg26-1 ts ::URA3 allele and showing growth at 37°C. This selection process was repeated five times. Spores harboring erg1-1/ets1-1 or erg7-1/ets2-1 alleles were then obtained from ascicontaining tetratype spores. Metabolic Radiolabeling and Analysis of Sterols and Neutral Lipids-For steady state labeling of sterols, cultures were incubated at 30°C with 5mCi/ml [ 14 C]acetate (51 mCi/mmol) for 3 h. Radiolabeled sterols were extracted using chloroform, methanol (2:1) and analyzed by one-dimensional TLC using hexane, diethyl ether, acetic acid (80:20:2). Neutral lipid radiolabeling and TLC analysis using petroleum ether, diethyl ether, acetic acid (70:30:1) were performed as described previously (25). Sterol intermediates were visualized by autoradiography using Kodak XAR5 film.
GC/MS Analysis of Sterols-Sterols were extracted as part of the nonsaponifiable fraction essentially as described (36). Washed cells were resuspended in 1.5 ml of methanol, 1.0 ml of 60% potassium hydroxide, 1.0 ml 0.5% pyrogallol, and heated for 2 h at 85°C. The nonsaponified material was extracted two times with petroleum ether, hexane (1:1). Cholesterol (20 l of 0.5 mg/ml stock) was added to each sample for use as an internal standard. Samples were dried down under nitrogen and resuspended in 200 l of a 1:1 (v/v) mixture of pyridine, bis(trimethylsilyl)trifluoroacetamide, 1% trimethylchlorosilane for trimethylsilylation (TMS) of free hydroxyl groups (37,38). Each TMS treatment extract was diluted by addition of 0.5 ml of petroleum ether, hexane (1:1) and passed over a silica gel minicolumn that was pretreated with hexane, petroleum ether (1:1, v/v) (39). Sterols were eluted with hexane, ethyl acetate (1:1, v/v). The pooled flow-through and eluant fractions were dried down under nitrogen and resuspended in 200 l of cyclohexane, dichloromethane (1:1, v/v). Sterol samples were analyzed by gas chromatography on a Hewlett-Packard model 6890 gas chromatograph using a 30-m HP-5 column. Standard instrument conditions were as follows: injection port temperature 250°C, detector temperature 300°C, initial oven temperature at 250°C. After a 30-min hold at 250°C, the oven temperature was increased at 2°C/minute up to 280°C, held for 15 min, and then increased at 5°C/min to 310°C and held for 5 min. Cellular sterol species were identified by comparison of peak position retention times to those of standard compounds (ergosterol, lanosterol, cholesterol (internal standard), and squalene) and/or relative to the cholesterol internal standard peak position (40 -42) and by GC/MS analysis using data base files of sterol spectra. GC/MS analysis was performed on a Hewlett-Packard model 6890 gas chromatograph equipped with a HP-5MS capillary column and Hewlett-Packard model 5973 Mass Selective Detector. GC/MS run conditions were essentially the same as that described above for analysis of sterols by gas liquid chromatography.
Squalene Epoxidase Assays-Yeast strains were grown aerobically in YEPD medium at 25°C to late-log phase (A 600 between 1.5 and 2). Cells were harvested, washed with 10 mM Tris⅐HCl, pH 7.4, and disrupted in a glass bead mill. Cell extracts were prepared by centrifugation at 12,000 ϫ g for 20 min at 4°C. Protein concentration was determined by the method of Bradford using bovine serum albumin as standard. 3 mg of total protein were incubated with the cofactor-substrate mix (5 mM NADPH, 10 mM sodium phosphate buffer, pH 7.4, containing 25 mM glucose 6-phosphate, 0.04 mM FAD, 2 units of glucose-6-phosphate dehydrogenase, 10 mM MgCl 2 , 5 mM mevalonic acid lactone, and 0.5 Ci of [ 14 C]mevalonic acid) at 30°C for 2 h. The enzymatic reaction was stopped by addition of 15% (v/v) KOH in methanol and incubation at 60°C for 15 min. Lipids were extracted with hexane and analyzed by TLC on silica gel 60 plates using CHCl 3 as developing solvent. TLC plates were scanned in a PhosphorImager SI (Molecular Dynamics) and analyzed with Image Quant 5.1. Lipid species were identified by using respective standards on the same silica plates. Values of each lipid species represent percent of the total intensity/lane. The values represent the average of three independent experiments.
[2-14 C]Acetate Incorporation into Sterols for Oxidosqualene Cyclase Activity-Sterol biosynthesis in whole yeast cells was followed by incorporation of [2-14 C]acetate into nonsaponifiable lipids as previously described (43). Briefly, washed cells (2-4 ϫ 10 6 cells) were resuspended in 1 ml of YPD, containing 0.1 mg/ml Tween 80, incubated with 0.2 Ci of [2-14 C]acetate (50 mCi/mM) and shaken for 30 min or 3 h at 30°C. Then, cells were harvested by centrifugation and saponified in 50% ethanol containing 15% KOH for 2 h at 80°C. Nonsaponifiable lipids were extracted twice with 1 ml of light petroleum and separated on silica gel plates (Merck) using n-hexane/ethyl acetate (85:15, v/v) as developing solvent. Authentic standards of ergosterol, lanosterol, dioxidosqualene, oxidosqualene, and squalene were run with each sample. Radioactivity in separated bands was measured using a System Imaging Scanner (Packard).
Oxidosqualene Cyclase Assays-3S-2,3-[ 14 C]Oxidosqualene was prepared by incubating a pig liver S 10 preparation with [ 14 C]mevalonate as described (44), in the presence of 0.1 mM of the oxidosqualene cyclase (OSC) inhibitor U14266A (U14, 3␤-(2-dimethylaminoethoxy)-androst-5-on-17-one) (45). After 3 h of incubation, the reaction was stopped by adding 1 volume of 15% KOH in methanol, and lipids were saponified at 80°C for 30 min. The nonsaponifiable lipids were extracted three times with two volumes of light petroleum and separated by TLC on silica gel plates. The plates were first developed in light petroleum to a height of ϳ10 cm above the origin. After drying, the plates were developed to 15 cm above the origin with n-hexane/ethyl acetate (90:10, v/v). Radioactive bands corresponding to 2,3-oxidosqualene were scraped off and eluted with dichloromethane. The extract was brought to dryness under a stream of nitrogen and 3S-2,3-[ 14 C]oxidosqualene was dissolved in benzene.
OSC activity in total yeast homogenate (6 mg of protein/ml) was determined as described elsewhere (43). Yeast cells were grown at 30°C to early stationary phase in YPD medium. Cells were then pelleted at 3000 ϫ g for 10, resuspended (0.15 g of wet cells/ml) in 0.02 M KH 2 PO 4 buffer, pH 7.4 containing 1.2 M sorbitol, and treated with lyticase (1 mg/g wet cells) for 60 min at 30°C. The spheroplasts obtained after lysis were pelleted at 3000 ϫ g for 10 min, washed twice in 0.02 M KH 2 PO 4 buffer, pH 7.4 containing 1.2 M sorbitol, and homogenized with a Potter device in 10 mM MES/Tris buffer, pH 6.9 containing 0.2 mM EDTA and 1 mM phenylmethylsulfonyl fluoride. 0.5 ml of the homogenate were incubated with 3S-2,3-[ 14 C]oxidosqualene (5000 cpm, 25 M) in the presence of Tween 80 (0.2 mg/ml) and Triton X-100 (1 mg/ml) in 10 mM MES/Tris, pH 6.9, and 0.2 mM EDTA for 30 min at 30°C under vigorous shaking in a water bath. The reaction was stopped by adding 1 ml of 15% KOH in methanol, and lipids were saponified at 80°C for 30 min. The nonsaponifiable lipids were extracted twice with 1 ml of light petroleum and separated on silica gel plates using dichloromethane as developing solvent. Radioactivity in 2,3-oxidosqualene and lanosterol was counted using a System Scanner (Packard). The amount of product formed was used for the calculation of the enzyme activity.
Liquid ␤-Galactosidase Assays-LacZ assays were performed as described in the Yeast Protocols Handbook (BD Biosciences Clontech) using the substrate, CPRG. Briefly, five independent colonies from a single strain were inoculated in minimal medium containing the appropriate amino acids and grown to exponential phase (A 600 ϳ 0.5-0.8). Cells were pelleted from 1.5 ml of the growing culture and resuspended in 300 l of buffer 1 (100 mM HEPES, pH 7.3 containing 150 mM NaCl, 4.5 M L-aspartate (hemimagnesium salt), 2.0 g of bovine serum albumin, and 100 l of Tween 20). 100 l of cells in buffer 1 were then subjected to three rounds of freeze/thawing using a dry ice/ethanol water bath. 700 l of buffer 2 (20 ml of buffer 1 containing 27.1 mg of CPRG) were added, and the reaction was allowed to proceed until color development became visible. The reaction was terminated by the addition of 500 l of 3.0 mM ZnCl 2 . Cellular debris was pelleted by centrifugation, and ␤-galactosidase activity was measured at A 578 using the resulting supernatant. Miller units were calculated as described (46). The Miller units represented are the average values of five independent experiments.
mRNA Expression Analysis-Cells were grown to exponential phase at 30°C in YEPD. Total RNA was extracted from 3 ϫ 10 7 cells using glass beads, SDS and buffered-saturated hot phenol. Total RNA was denatured in formaldehyde-SSPE (10 mM NaH 2 PO 4 buffer, pH 7.7, containing 0.18 M NaCl and 1 mM EDTA) for 15 min at 65°C. 20 g of denatured RNA was applied to a nylon membrane using a slot blot apparatus (Bio-Rad). Membranes were washed in SSPE, UV crosslinked and hybridized to gel-purified PCR products that were obtained using pfu polymerase. The hybridization buffer consisted of SSPE containing Denhardt's solution. Post-hybridization, membranes were washed several times with SSPE containing 0.1% SDS. Hybridization and all washes were performed at 65°C. PCR products were [ 32 P i ] radio-labeled by random priming (Takara, Madison, WI). The level of U2 mRNA expression was used as a loading control. ERG gene expression levels were determined by autoradiography (Kodak XAR5) followed by densitometry using a Bio-Rad model GS-670 Imaging Densitometer and Molecular Analyst software, version 1.4.1, or using a Storm 840 Phos-phorImaging system and Storm Scanner Control software (Fig. 5E).
Western Analysis of Erg1p Levels-Strains were grown at 25°C in YEPD medium to late-log phase and subsequently disrupted in a glass bead mill. Lysed cells were centrifuged at 12,000 ϫ g for 20 min to obtain a total cell-free extract. 20 g of total cell extract protein from various ERG strains were resolved by SDS-PAGE and transferred onto a polyvinylidene difluoride membrane. Western blot analysis was performed using polyclonal anti-Erg1p antibodies and anti-Sui2p antibodies, respectively, and horseradish peroxidase-conjugated secondary antibodies as previously described (47). Erg1p migrates at 55 kDa and Sui2p, which serves as a loading standard, migrates at ϳ35 kDa.
We grew ets1-1 and ets2-1 cells in the absence and presence of various antifungal drugs and growth rate was determined over time. Both strains had an increased sensitivity to azole drugs (fluconazole and itraconazole) targeting the lanosterol C-14 demethylase (Fig. 1), Erg11p (Fig. 3, A and B), whereas they were resistant to the ergosterol binding agent, amphotericin (Fig. 3C). ets 1-1 cells also were sensitive to the allyamine, terbinafine (Fig. 3D), which targets the squalene epoxidase, Erg1p, but were resistant to the morpholine antifungal drug, fenpropimorph (Fig. 3E). ets2-1 cells were mildly resistant to terbinafine (Fig.  3D), but showed normal sensitivity to fenpropimorph (Fig. 3E). Thus, ets1-1 and ets2-1 cells had antifungal drug resistance profiles that were different than wild-type cells.
erg mutations may also cause the accumulation of specific sterol intermediates, so we analyzed sterol and neutral lipid compositions of 4 M. Germann, C. Gallo, and J. T. Nickels, Jr., manuscript in preparation. ets1-1 and ets2-1 cells. Using GC/MS, we observed that erg26-1 ts ets1-1 and ets1-1 cells accumulated squalene, whereas they contained lower levels of most other sterol intermediates and the sterol end product, ergosterol (TABLE ONE). erg26-1 ts ets2-1 and ets2-1 cells accumulated squalene epoxide and also contained much lower levels of ergosterol than wild-type cells. Neutral lipid composition demonstrated that ets2-1 cells synthesize less sterol esters than wild type, whereas the sterol esters produced by ets1-1 cells migrated differently than those seen in either wild-type or erg26-1 ts cells (Fig. 2C). ets2-1 also accumulated two lipid species not seen in wild type, one of which migrated similar to fatty acid alcohols (Fig. 2C, asterisk, ets2-1). ets1-1 cells accumulated a minor species that migrated similar to fatty acids (Fig. 2C, asterisk, ets1-1). The exact molecular species of these lipids is not known. Thus, ets1-1 and ets2-1 cells harbor defects in sterol metabolism.
ets1-1 and ets2-1 cells may accumulate squalene and squalene epoxide, respectively, because they harbor recessive mutations in the squalene epoxidase, Erg1p, and oxidosqualene cyclase, Erg7p. To test this, we assayed Erg1p or Erg7p enzyme activities in wild-type and mutant cells. For squalene epoxidase activity, we determined the percent conversion of [ 14 C]mevalonic acid to squalene, squalene epoxide, lanosterol, and ergosterol in cell extracts (TABLE TWO). For oxidosqualene cyclase, we determined the conversion of [ 14 C]acetate to squalene, oxidosqualene, dioxidosqualene, and lanosterol in vivo (TABLE  THREE), and we directly assayed for oxidosqualene cyclase activity using [ 14 C]oxidosqualene as a substrate.
Our data suggested that ets1-1 and ets2-1 were allelic to ERG1 and ERG7, respectively. If so, replacing the ERG1 chromosomal locus in erg26-1 ts ets1-1 and the ERG7 chromosomal locus in erg26-1 ts ets2-1 with the corresponding wild-type genes should revert the temperature resistance and cause the re-accumulation of 4-carboxysterol intermediates in these cells. We found that this was the case. Thus, we conclude that ets1-1 and ets2-1 are recessive alleles of ERG1 and ERG7, respectively. Therefore, we have renamed ets1-1 as erg1-1 and ets2-1 as erg7-1 for the remainder of the report. erg1-1 and erg7-1 Alleles Harbor Mutations That Affect Activity-To gain more insight into domains required for Erg1p and Erg7p activity, we sequenced erg1-1 and erg7-1 alleles. erg1-1 carried a single G-A nucleotide change at ϩ740 (G247N), resulting in a glycine to aspartic acid change. erg7-1 harbored two mutations, G-A at ϩ1588 (D530N) and T-A at ϩ1844 (V615E), causing aspartic acid to asparagine and valine to glutamic changes, respectively. We constructed plasmids carrying these alleles (pRS416-erg1-1 and pRS416-erg7-1) and characterized their activities in erg1-and erg7-null strains. An erg1 pRS416-erg1-1 and erg7 pRS-erg7-1 cells were viable and accumulated squalene (TABLE TWO, BY4741) and squalene epoxide, respectively, to levels similar to those seen in erg1-1 and erg7-1 cells. These mutant alleles were also able to suppress the ts phenotype of erg26-1 ts cells, as erg26-1 ts erg1 pRS416-erg1-1 and erg26-1 ts erg7 pRS416-erg7-1 cells were viable at high temperature.
ERG1, ERG7, and ERG25/26/27 Expression Levels Are Coordinately Regulated in erg Mutants-The fact that we observed elevated Erg1p levels in erg1-1 strains suggests that sterol gene expression may be regulated in response to squalene accumulation (Fig. 4). To test this idea and also determine whether cells respond to the accumulation of squa- lene epoxides, we examined ERG1 and ERG7 expression levels in wild type, erg1-1, and erg7-1. To determine whether this putative sterol responsive pathway regulates other sterol genes, we also examined the expression levels of ERG25/ERG26/ERG27. ERG25/26/27 together are required for the conversion of 4,4-dimethylzymosterol to zymosterol (Fig. 1) (10). To further define what sterol intermediates may potentially activate the pathway, we looked at expression levels in erg2-, erg6-, and erg24-null cells, as well as wild type cells treated with lovastatin (HMG-CoA reductase inhibitor), fluconazole (Erg11p lanosterol C14-demethylase inhibitor), and amphotericin B or nystatin (ergosterol binding agents) (Fig. 1). ERG expression levels were determined using promoter-LacZ fusion constructs and/or Northern analysis.
Sterol Responsive ERG Expression Induction Requires Upc2p and Ecm22p Transcription Factors-We wanted to begin to define the cell factors functioning to regulate ERG expression in response to blocks in sterol biosynthesis. We initiated these studies by examining lovastatin-    (Fig. 6). The loss of Mot3p, Upc2p, or Ecm22p functions alone had no effect. Identical results were obtained when we examined fluconazole-induced expression induction. Thus, the loss of both Upc2p and Ecm22p functions, and not either alone, was required to eliminate sterol responsive expression regulation.

DISCUSSION
Our results demonstrate that the expression of yeast ERG1, ERG7, and ERG25/26/27 genes are coordinately regulated in response to multiple blocks in sterol biosynthesis. Strikingly, Northern analysis revealed that the degree of induction was not the same for all mutant strains tested. These results suggest the possibility that the pathway controlling expression of these and other ERG genes may have the ability to be activated to different levels, depending on what specific sterol intermediate accumulates, or to what extent ergosterol levels are depleted. By using the HMG-CoA reductase inhibitor, lovastatin, we showed that the transcription factors, Upc2p and Ecm22p are required for expression regulation.
The genes and/or cDNAs for human ERG1 and ERG7 have been cloned and the Erg7 oxidosqualene cyclase protein has been crystallized (53)(54)(55)(56)(57). The hERG1 promoter contains a consensus SRE that acts as an activator sequence in cells overexpressing SREBP, and a nuclear factor Y binding site that is required for sterol-dependent regulation (58). Whether these sequences act together in response to sterol levels or independently is not known. M'Baya et al. (59) has shown that yeast ERG1 is regulated in response to changes in sterol levels. ERG1 contains a Vik and Rine (24) 7-bp SRE, and a sterol responsive promoter sequence, comprised of two 6-bp direct repeats, that is required for terbinafine-and ketoconazole-dependent induction of ERG1 expression (60). Presently, little is known about the transcriptional regulation of ERG7 in yeast. Interestingly, the ERG7 promoter does not contain a "putative" consensus yeast SRE element.
The cDNAs or genes for human and mouse ERG25/26/27 have been cloned (61)(62)(63)(64)(65). There is a report that ERG25 is regulated through an SREBP-dependent mechanism in the vascular wall of pigs, but little else is known about expression regulation of mammalian ERG25/26/27 (66). We showed that a sterol responsive pathway coordinately regulates the expression of these genes. ERG25 does contain three yeast SRE sites (24), while ERG26 and ERG27 lack this sequence. The transcription factors, Ecm22p and Upc2p were shown to bind to the SREs of ERG2 and ERG3, and they most likely regulate the expression of these genes in response to lovastatin treatment and inhibition of HMG-CoA reductase activity (24).
These results suggest that ERG25 expression should be regulated in the same manner, through the binding of Upc2p/Ecm22p to SREs in the ERG25 promoter. So then how can we reconcile the fact that ERG26 and ERG27 lack SREs, yet they are coordinately regulated along with ERG25 in response to changes in sterol levels? One possibility is that ERG26 and ERG27 contain a novel Ecm22p and/or Upc2 binding site(s) that has yet to be elucidated. Another explanation is that two independent pathways exist that both converge on ERG genes to regulate their expression in response to changes in sterol levels. Recall that hERG1 contains both SRE and nuclear factor Y binding sites that may act independently to regulate expression in response to changes in sterol levels. We are currently determining the promoter sequences within ERG26 and ERG27 required for sterol responsive expression regulation.
The Erg1p squalene epoxidase converts squalene to 2,3-oxidosqualene (2,11). The reaction requires oxygen and FAD. The Erg7p oxidosqualene cyclase then converts the Erg1p product, 2,3-oxidosqualene, to lanosterol (2,11). We mapped a single mutation in the yeast ERG1 gene (G247D) that causes increased ERG1 expression, elevated levels of Erg1p, altered SBI sensitivities and weakened enzymatic activity. Some of these phenotypes have been seen in other erg1 mutants or in cells treated with terbinafine (47,48). The G247D mutation resides within a conserved domain of Erg1p that is necessary for monoxygenase activity (67,68). Interestingly, a terbinafine resistant mutant of Erg1p (L251F) in S. cerevisiae has been isolated, which carries a mutation within this same region (67). Thus, terbinafine resistance and sensitivity can map within the same conserved region. We mapped two mutations in Erg7p (D530N, V615E) that together were necessary for suppressing the ts and 4-carboxysterol accumulation of erg26-1 ts cells. erg7 alleles harboring either mutation alone were unable to suppress these phenotypes. The mutations reside within a large conserved domain required for sterol cyclization (57).
Our in vitro enzymatic assays demonstrated that erg26-1 ts cells had weakened Erg1p and Erg7p activities (TABLES TWO and THREE). In the case of Erg1p, we found that mutant cell extracts accumulated ϳ2-fold higher levels of the Erg1p substrate, squalene, and produced less lanosterol than wild-type cells. Mammalian mevalonate kinase is inhibited by specific cholesterol biosynthetic and nonsterol isoprene intermediates (69). Multiple studies have demonstrated that yeast and mammalian phospholipid biosynthetic enzyme activities are regulated by numerous lipid species (70 -75). Thus, the 4-carboxysterols accumulating in erg26-1 ts cells could be directly interacting with and competitively or non-competitively inhibiting Erg1p activity. Our results bring up the possibility that regulation of ERG26 mRNA levels and/or Erg26p enzymatic activity may represent a temporally efficient way for yeast cells to regulate other sterol biosynthetic enzymes through changes in the levels of 4-carboxysterols in cells. Lipid-dependent regulation of Ergp activities adds a second level of sophistication to mechanisms reg- FIGURE 5. Multiple ERG genes are coordinately regulated by a sterol responsive pathway. A, steady state levels of ERG1-LacZ and ERG7-LacZ promoter activities were assayed in the indicated strains using the substrate, CPRG. B, steady state levels of ERG25-LacZ, ERG26-LacZ, and ERG27-LacZ promoter activities were assayed in the indicated strains using the substrate, CPRG. C and D, mRNA expression levels were determined for the indicated ERG genes by Northern slot blot hybridization. E, mRNA expression levels were determined in wild-type cells grown in the absence and presence of 50 g/ml lovastatin for 12 h, or 16 g/ml fluconazole, 0.25 g/ml amphotericin B, or 10 g/ml nystatin for 3 h. Wild type, hatched bars; erg26-1 ts , black bars; erg1-1, gray bars; erg2-1, white bars; U2, loading control. The numbers represent relative density values for U2, where all values are compared with the wild-type control. erg24-null cells require 200 mM CaCl 2 for viability.
ulating the sterol pathway. It may also help to explain why ERG25/26/27 are coordinately regulated with other sterol biosynthetic genes.
Gachotte et al. (28) showed that erg27 mutants do not accumulate the predicted 3-ketosterols, but rather those intermediates that are seen in cells harboring defects in Erg7p activity. In fact, erg27 mutants have decreased levels of Erg7p activity, which can be partially suppressed by ectopic expression of an Erg27p-HA chimera (76). Moreover, Erg7p associates with Erg27p in lipid particles, and this association seems to be necessary for organelle localization and for protein stability. Thus, Erg27p and Erg7p may form a complex in vivo, and that decreased Erg7p activity in erg27 mutants arises from mislocalization and proteolytic degradation of the protein because of the loss of this complex (76). Studies have demonstrated physical interactions between Erg25p/ Erg26p/Erg27p (25,77). The erg26-1 ts allele could interfere in some way with the interaction between Erg27p and Erg7p, causing decreased activity. However, erg26 cells do not accumulate those sterols seen in erg27 cells (25,26). An alternative explanation is that the 4-carboxysterols in erg26-1 ts cells directly interact with Erg7p and inhibit its activity. If this type of lipid-dependent regulation is subtle, it may only be unmasked using in vitro enzymatic assays. Enzymatic assays using purified Erg1p and Erg7p preparations will help to resolve these issues.