| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Article |
Correspondence to: Dago Dimster-Denk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Gene expression profiling is rapidly becoming a mainstay of functional genomic studies. However, there have been relatively few studies of how the data from expression profiles integrate with more classic approaches to examine gene expression. This study used gene expression profiling of a portion of the genome of Saccharomyces cerevisiae to explore the impact of blocks in the isoprenoid biosynthetic pathway on the expression of genes and the regulation of this pathway. Approximately 50% of the genes whose expression was altered by blocks in isoprenoid biosynthesis were genes previously known to participate in the pathway. In contrast to this simple correspondence, the regulatory patterns revealed by different blocks, and in particular by antifungal azoles, was complex in a manner not anticipated by earlier studies.—Dimster-Denk, D., J. Rine, J. Phillips, S. Scherer, P. Cundiff, K. DeBord, D. Gilliland, S. Hickman, A. Jarvis, L. Tong, and M. Ashby. Comprehensive evaluation of isoprenoid biosynthesis regulation in Saccharomyces cerevisiae utilizing the Genome Reproter Matrix(TM). J. Lipid Res. 1999. 40: 850–860.
Supplementary key words: isoprenoid pathway regulation, genome expression profiling, Saccharomyces cerevisiae
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Eukaryotic cells utilize a group of structurally related compounds, the isoprenoids, for a vast array of cellular processes. These processes include structural composition of the lipid bilayer, electron transport during respiration, protein glycosylation, tRNA modification, and protein prenylation. All isoprenoids are synthesized via a pathway known variously as the isoprenoid pathway, mevalonate pathway, or sterol biosynthetic pathway. Although the bulk end product of the pathway is sterols, there are several branches of the pathway that lead to non-sterol isoprenoids (Figure 1). Due to the involvement of isoprenoids in a variety of physiologically and medically important processes, a comprehensive understanding of the regulation of this pathway would offer many scientific and practical benefits.
|
The regulation of the isoprenoid biosynthetic pathway is known to be complex in all eukaryotic organisms examined, including the budding yeast Saccharomyces cerevisiae (1) (2) (3). The overriding principle for the regulation of this pathway is multiple levels of feedback inhibition. This feedback regulation is keyed to multiple intermediates and appears to act at numerous steps of the pathway, involving changes in transcription, translation, and protein stability. Additionally, the availability of molecular oxygen, required for sterol and heme biosynthesis, also regulates the expression of genes at key steps of the pathway (4) (5). The emerging picture is that the isoprenoid pathway has numerous points of regulation that act to control overall flux through the pathway as well as the relative flux through various branches of the pathway. Given this complexity, it can be difficult to understand the rationale for any given instance of regulation of this pathway, unless it is viewed within the context of the entire pathway (2).
We have developed the ability to perform and analyze the level of gene expression from essentially every gene of an organism's genome simultaneously. The experiments reported here use Acacia's yeast Genome Reporter Matrix(TM) (GRM) which consists of a collection of reporter gene fusions to >95% of the protein coding genes of S. cerevisiae. To examine the regulation of the isoprenoid pathway in a more comprehensive fashion, the GRM technology was used to evaluate the expression of all genes in the pathway using reporter constructions that detect changes in both the transcription and translation of genes. Because natural selection operates on a selected outcome rather than on a particular molecular mechanism, profiling strategies that detect regulatory changes at several molecular mechanism contribute to a broader view of how regulatory circuits have evolved.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Strains and media conditions
The strain of Saccharomyces cerevisiae used in this study, ABY11 (MATa leu2
1 ura3-52), was derived from S288c. GRM arrays were grown at 30°C on solid casamino acid medium (Difco) with 2% glucose and 0.5% UltraPure Agarose (Gibco BRL). The medium was supplemented with additional amino acids and adenine (Sigma) at the following concentrations: adenine and tryptophan at 30 µg/ml; histidine, methionine, and tyrosine at 20 µg/ml; leucine and lysine at 40 µg/ml. Stock solutions of the supplements were made at 100x concentrations in water. Yeast cells were transformed with the reporter plasmids by the lithium acetate method (6) (7).
The drug treatments were performed by adding the drugs directly to the liquefied medium during preparation. The HMG-CoA synthase inhibitor (L659-699) and the squalene synthase inhibitor (zaragozic acid) were provided by J. Bergstrom (Merck). The remaining drugs were purchased commercially. Stock solutions of drugs were prepared in 100% solvent (e.g., DMSO, methanol, or water). When required, the lactone forms of the statins were converted as previously described (8). The final concentration of each drug (and solvent) used is listed in Table 1.
|
Construction of reporter gene fusions
The regulatory region of each yeast gene was cloned into one of two vectors, pAB1 or pAB2. The vector pAB1 was constructed in the following manner. First, the polymerase chain reaction (PCR) was used to amplify the transcriptional terminator region from the gene PGK1 using the oligonucleotides 5'-PGKTERM (5'-GAT TGAATTCAATTGAAATCGATAG-3') and 3P-PGKTERM (5'-CCGAGGCGCCGAATTTTCGAGTTAT-3'). The amplified fragment consists of the 263 base-pair region immediately downstream of the PGK1 stop codon, and contains an EcoRI site at the 5' end and a Nar I site at the 3' end. These restriction sites were engineered into the two PCR primers (underlined sequences). The terminator was then cloned into YIplac211 (9) that had been linearized with EcoRI and Nar I, yielding pAB34. Next, the coding region of the green fluorescent protein (GFP) from Aequoria victoria (10) was amplified by PCR using the oligonucleotides 5P-GFP-ORF (5'-CATGTCTAGAGGAGAAGAACTTTTC- 3') and 3P-GFP-ORF (5'-CGCGAATTCCTATTTGTATAGTTCA-3'). Again, these oligonucleotides contain engineered XbaI and EcoRI sites at the 5' and 3' ends, respectively (underlined). This fragment was cloned into pAB34, linearized with XbaI and EcoRI, to produce pAB35. Finally, the GFP-PGK terminator fragment was moved into the episomal vector YEplac195 (9) as an XbaI/Nar I fragment, thereby producing pAB1.
The vector pAB2 is pAB1 with an altered multiple cloning site (MCS). The new MCS contains 8 base-pair recognition sites for three restriction enzymes. These larger 8 base-pair recognition sites occur less frequently throughout the yeast genome than the 6 base-pair sites present in the MCS of pAB1. Thus, the utilization of restriction enzymes that recognize 8 base-pair sequences to clone the various regulatory regions (engineered into the PCR primers used to amplify the regions) would minimize the occurrence of those sites within the regions themselves. To construct pAB2, pAB1 was linearized with XbaI and SphI, dropping out the existing MCS, and an adapter containing the new MCS was ligated in. The adapter was made by hybridizing two oligonucleotides, 8Cutter (5'-CGGCGCGCCGCGGCCGCATGGCCGGCCA AT-3') and 8CutEnd (5'-CTAGATTGGCCGGCCATGCGGCCGC GGCGCGCCGCATG-3'). This adapter has sites for the restriction enzymes FseI, NotI, and AscI (underlined).
The promoter regions were cloned utilizing PCR of genomic DNA prepared from a strain derived from S288c; JRY147 (MATa SUC2 mal mel gal2 CUP1). The promoter-specific primers were designed such that the proximal primer spanned the start codon of the specific gene and included a few (usually four) codons derived from the gene. The position of the distal primer was determined on a case-by-case basis depending on the distance to, and orientation of, the neighboring open reading frame (ORF) and the restriction sites present. Where the upstream ORF was positioned in a divergent orientation and within 1,200 base-pairs, the size of the promoter fragment amplified was adjusted such that all nucleotides up to, but not including, the start codon of the upstream ORF were present. In cases where the upstream ORF was situated in the same orientation, the amplified fragment was designed to extend into the coding region but not so as to include the start codon. The approximate sizes of the various regulatory regions present on the reporter plasmids presented in this study are given (Table 2). Both primers had restriction enzyme recognition sites engineered into the ends to allow the subsequent cloning of the PCR fragment into pAB1, or pAB2.
|
Determinations of reporter gene expression levels
The individual strains comprising the GRM were maintained as independent colonies (and cultures) in a 96-well format, in medium selecting for the URA3-containing reporter plasmid. Prior to each experiment, fresh dilutions of the reporter-containing strains were inoculated and grown overnight at 30 °C. The fresh cultures were then arrayed onto solid casamino acid medium at high density using an aspirate/dispense machine, the BioDot A/D3200 (Cartesian Technologies, Irvine CA). As stated, the drugs were added to the solid medium during preparation. Once arrayed, each plate was grown at 30°C, usually for 21 h.
The level of fluorescence expressed from each reporter gene fusion was determined using a Molecular Dynamics Fluorimager SI. AIS image analysis software (Imaging Research, Ontario CA) was used to quantitate the fluorescence of each colony in the images. Generally, the drug treatments were performed at several concentrations, with the analysis based upon the concentration producing the most informative expression profile (see below).
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
The experiments presented here used a subset of the complete Genome Reporter Matrix(TM) (GRM) consisting of 864 independent yet isogenic yeast strains, each containing a different reporter gene fusion (the 864 GRM). The 864 reporter genes included genes known or implicated to be involved in isoprenoid metabolism as well as genes diagnostic of the overall physiological state of the cell. Specifically, with regard to isoprenoid metabolism, the partial genome matrix contained reporter gene fusions for 21 genes involved in the biosynthesis and storage of ergosterol, the major sterol product of the isoprenoid pathway in yeast. Reporter genes were also included for genes involved in the utilization of isoprenoid intermediates for the synthesis of coenzyme Q , hemeA, isopentyl tRNA modifications, and protein prenylation ( Figure 1). Additionally, reporter genes were included for genes involved in cellular processes that require isoprenoids for function, such as the mating pheromone response pathway and the ras oncoprotein.
Treatments with inhibitors of isoprenoid biosynthesis
The gene expression levels of the 864 GRM were determined in response to treatments of cells with eight different inhibitors of the isoprenoid pathway. These eight drugs target four different enzymatic steps of the pathway. L659-699 inhibits HMG-CoA synthase, encoded by ERG13 (11). Three of the drugs studied (lovastatin, fluvastatin, and atorvastatin) all inhibit HMG-CoA reductase, encoded by two genes HMG1 and HMG2 (12). Zaragozic acid inhibits squalene synthase, encoded by ERG9 (13). Finally, three different azoles (miconazole, sulconazole, and fluconazole) inhibit lanosterol 14
-demethylase, encoded by ERG11 (14).
The 864 GRM was treated with these eight inhibitors at varying concentrations and in varying solvents ( Table 1). The concentration of each inhibitor was chosen to achieve a similar degree of growth inhibition. The number (and percentage) of genes among those represented in the 864 GRM exhibiting at least a 2-fold induction, or repression, of expression in response to each treatment is shown for a representative set of experiments (Table 1).
The specificity of the feedback regulation observed in this study was quite remarkable. For example, the HMG-CoA synthase inhibitor caused an induction of at least 2-fold of only 34 genes of the 864 GRM (Table 1). Of these 34 genes, 17 of them are known to be directly involved in ergosterol metabolism. Indeed, of these 17 genes induced by the inhibition of HMG-CoA synthase, 15 encode enzymes of the ergosterol biosynthetic pathway. The other two genes are involved in ergosterol storage (ARE1) and hemeA biosynthesis (HEM14). Many of the remaining 17 genes were of unknown function, and their involvement in ergosterol metabolism merits investigation. This block did not affect the genes that encode enzymes involved in the four branches off of the sterol pathway, with the exception of HEM14. Thus, the feedback regulation of the isoprenoid pathway primarily involved the central pathway leading to ergosterol biosynthesis, and not the various non-sterol branches of the pathway (Figure 1).
Interestingly, for comparable degrees of growth inhibition, the azoles caused a larger percentage of genes to change expression levels than did other blocks of the isoprenoid pathway (Table 1). This wider response to azoles may represent a lower degree of specificity of these drugs for their target, or may reflect a broader response to the inhibition of lanosterol demethylase activity in the cell (see below). Nevertheless, all agents affected the expression of 8% or more of the genes in this subset of the genome. Thus, the gene expression profiles caused by enzyme inhibitors provided a rich signal by which to compare profiles.
The results presented here focus on reporters for 50 genes involved in isoprenoid metabolism ( Table 2). Specific changes in the expression of these 50 genes in response to the inhibitor treatments are shown ( Table 3). The responses are expressed as the fold change relative to the solvent control. No change is indicated by a fold change of 1.0. Induced expression is indicated by a positive value while repressed expression has a negative value. The ratios shown are an average of at least two experiments, except miconazole and sulconazole which were single experiments. Graphic representations of these data are also presented ( Figure 2, Figure 3, and Figure 4).
|
|
|
|
Feedback regulation of ergosterol biosynthesis genes
In general, a block at all four steps of ergosterol biosynthesis examined caused a compensatory induction of the expression of genes involved in ergosterol biosynthesis (Figure 2). The expression of a subset of genes encoding enzymes of sterol biosynthesis (ERG12, ERG8, ERG19, ERG9, and ERG2) were strongly induced by all eight drugs. These genes encode enzymes that act both early in the pathway (prior to the synthesis of farnesyl diphosphate) and late in the pathway. In principle, this regulation could be due to the depletion of an intermediate(s) downstream of a given block, or the accumulation of an intermediate(s) upstream of the block. The blocks imposed at various steps of the pathway would be expected to accumulate different upstream intermediate(s). In contrast, each block would deplete some common late intermediate(s) or end product(s). Therefore, the simplest interpretation was that the regulatory response of these genes was keyed to the depletion of a common intermediate(s). The latest block examined was that of lanosterol demethylase, so by this interpretation the regulation was keyed to a pathway product(s) downstream of lanosterol. Also, ERG9 mRNA is known to be induced in cells treated with lovastatin (15), indicating that the regulatory mechanism at work here includes changes in the mRNA levels.
Several other genes of this pathway (including ERG10, HMG1, ERG20, ERG1, ERG11, and ERG6) were induced by some, but not all, blocks to ergosterol biosynthesis (Figure 2). Among these genes, some are known to respond to feedback regulation by the isoprenoid pathway. The regulated expression of ERG10 observed in this study was consistent with a previous report demonstrating that ERG10 is transcriptionally repressed by a late intermediate(s) or product(s) of the pathway (16). One of the two HMG-CoA reductase isozymes, Hmg1p, is translationally repressed by a non-sterol product of the pathway (8). In the experiments shown here, the HMG1 reporter gene was induced by an inhibitor of squalene synthase and an inhibitor of lanosterol demethylase, suggesting that HMG1 responded to the levels of a sterol product of the pathway (Figure 2). These disparate observations may be due to some procedural differences in the execution of the experiments (see below).
Three genes (ERG10, IDI1, and ERG3) exhibited decreased expression in response to the treatment of cells with azoles, but not other blocks to the isoprenoid pathway (Figure 2). It is probable that the inhibition of lanosterol demethylase by the azoles caused a buildup of some upstream intermediate (e.g., lanosterol) that is depleted by the inhibition of squalene synthase by zaragozic acid (and the two other upstream blocks). If the intracellular concentration of this intermediate were being monitored to set the expression levels of these genes, it would be consistent with the observation that treatments with azoles cause a repression of expression while zaragozic acid causes an induction. However, it has been reported that ketoconazole, an azole not examined in this study, causes increased ERG3 mRNA levels (17). Moreover, mutations in several genes that act late in the ergosterol pathway, including ERG24 and ERG5, also cause an increase in ERG3 mRNA levels. These observations indicate that ergosterol availability regulates ERG3 expression levels (17). Thus, treatments with azoles would be expected to deplete ergosterol and thereby induce ERG3 expression, as has been reported with ketoconazole. These disparate results may reflect a different biological activity of the three azoles examined here that is not shared with ketoconazole, or different treatment protocols that utilize chronic versus acute exposure regimens. Previous yeast experiments utilized higher concentrations of drugs for shorter amounts of time, thereby eliciting a more acute response. In contrast, the experiments described here used a chronic exposure to lower concentrations of drug. Considering the dynamic regulation of this pathway, it is conceivable that the cells treated with azoles in our experiments are approaching a new homeostasis reflected by the repression of some pathway genes that had been induced during the earlier acute phase of the regulatory response. This notion is consistent with the expectation that mutations of late ergosterol pathway genes, such as an ERG5 null allele, would more closely mimic an acute drug treatment as induced expression of pathway genes would not increase ergosterol biosynthesis, and a new homeostasis would not be reached.
Mechanism of lanosterol demethylase inhibitors
Azole inhibitors of lanosterol demethylase are the most widely used antifungal compounds in human medicine. Thus, there is a strong interest in determining just how these inhibitors achieve their therapeutic effect. Several observations suggest that the biological activity of fluconazole may be subtly different from that of miconazole and sulconazole. This difference was particularly evident in the regulation of the genes ERG5 and ERG4. Sulconazole, and to some extent miconazole, induced ERG5 and ERG4 expression, whereas fluconazole repressed their expression (Table 3). The two acyl-CoA:sterol acyltransferase genes, ARE1 and ARE2, also responded differently to the two classes of azoles. Both miconazole and sulconazole induced ARE1 expression and repressed ARE2 expression (Table 3). This opposing effect of a treatment on the two isozymes of an enzymatic activity has been seen before in the yeast sterol pathway (2) (4). In contrast, fluconazole treatment induced both ARE1 and ARE2 (Table 3).
Considering the breadth of the effect elicited by the azoles compared to other inhibitors of the isoprenoid pathway (Table 1), perhaps only a portion of these responses reflect inhibition of lanosterol demethylase activity per se. If this idea is correct, then part of the expression response profile of each azole reflects indirect bioactivities of that azole. Thus, it may be possible to inhibit the lanosterol demethylase enzyme with greater specificity than that of the azoles presently available. Preliminary experiments examining the GRM expression profile of an ERG11 mutant strain that encodes a conditional lanosterol demethylase allele suggest that the azoles have bioactivities in addition to the inhibition of lanosterol demethylase (data not shown).
Mating pheromone response pathway
The mating pheromone response pathway is one of five MAP kinase signaling pathways in yeast. In the case of the mating pathway, the binding of pheromone to its cell surface receptor triggers the signaling cascade, and ultimately results in the induction of the expression of a number of genes required for mating. There is a biochemical connection between isoprenoids and the yeast mating pheromone response pathway. Two components of the pathway are subject to post-translational protein prenylation (18). Yeast a-factor, encoded by the genes MFA1 and MFA2, and the 
subunit of the heterotrimeric G protein, Ste18p, are both modified by a farnesyl diphosphate moiety. Because this modification is required for a-factor function, inhibition of farnesyl diphosphate biosynthesis results in MAT a cell-type specific sterility (19) (20). The Ste18p protein, utilized by both haploid cell types, also requires prenylation for full activity (21). Thus, inhibition of isoprenoid biosynthesis, and in particular farnesyl diphosphate, would be expected to affect induced signaling via the mating pathway and consequently limit the level of induction of the pheromone responsive genes. Presumably the same effect upon basal signaling would occur, and therefore the uninduced levels of expression of these genes would be lower.
The expression of MFA1, which is itself pheromone inducible, is known to be reduced in mutants with a defective farnesyl transferase (22). This regulation is evident at the level of mRNA, and presumably reflects the decreased prenylation and hence decreased function of Ste18p. These experiments were conducted in the absence of mating pheromone and thus involve effects upon the basal level of pathway signaling and gene expression. The regulation of MFA1 by isoprenoid availability was recapitulated by the MFA1 reporter gene in the experiments described here (Figure 3). Consistent with the notion of decreased basal pathway signaling, the expression levels of FUS1 and Figure 1, two other pheromone inducible genes, also decreased in response to blocks in isoprenoid biosynthesis (Figure 3).
However, not all the effects upon the expression of mating pheromone pathway target genes could be explained by reduced farnesylation of the Ste18p. For example, late blocks in the pathway repressed the expression levels of all three genes, MFA1, FUS1, and Figure 1 (Figure 3). These late blocks would not be predicted to deplete farnesyl diphosphate pools, and thus would have no effect on prenylation of Ste18p per se. Although we cannot as yet pinpoint the basis of these responses, these late blocks are expected to fundamentally alter the composition of some of the major structural lipids of the plasma membrane. Thus, because the pheromone receptor itself as well as its cognate G-protein are intimately associated with the plasma membrane, it is not surprising that these late blocks would also have an impact, albeit indirect, on expression of the targets of the mating pheromone pathway. This is reminiscent of the sterol "sensing" enzymes such as the sterol regulatory element binding protein (SREBP) cleavage activating protein (SCAP) that appear to respond to the sterol content of the endoplasmic reticulum membrane (23). In any event, the notion that there are two modes of regulation, both isoprenoid dependent and one farnesylation dependent, is supported by the observation that a mevalonate auxotroph has lower levels of MFA1 mRNA than a mutant defective in the farnesyltransferase, although both are repressed with respect to wild-type (22).
The expression of another mating response gene, BEM1, was induced by blocks up to and including squalene synthase, but was repressed by blocks of lanosterol demethylase (Figure 3). Thus, blocking squalene synthase and blocking lanosterol demethylase were not equivalent with respect to the mating response pathway, specifically BEM1 espression. The nature of this difference was puzzling. Yet, these data revealed the power of broad-based gene expression profiling to avoid over simplistic interpretations based upon the analysis of too few genes and conditions.
RAS1 versus RAS2
The yeast S. cerevisiae has two isozymes of the Ras oncoprotein, encoded by the genes RAS1 and RAS2 (24) (25). In yeast as in humans, the Ras protein is prenylated by farnesyl diphosphate (19) (20). This modification is important for membrane localization, and therefore the function of the Ras protein. Both RAS1 and RAS2 mRNA levels have previously been shown to be repressed in response to limited isoprenoid biosynthesis (22). The 864 GRM utilized in this study allowed us to examine the expression levels of RAS1 and RAS2 in response to the various blocks in the isoprenoid pathway.
RAS1 expression was strongly repressed by all of the blocks up to and including squalene synthase (Figure 4). Because inhibition of squalene synthase caused decreased RAS1-GFP expression, the regulation of RAS1 expression presumably did not reflect the level of function of a farnesylated regulatory protein (or Ras1p itself) as inhibition of this enzyme would not result in a depletion of FPP. This is consistent with previous experiments demonstrating that reduced isoprenoid biosynthesis causes reduced RAS1 mRNA levels, and that loss of farnesyltransferase activity (through mutation) fails to elicit this response (22). This last observation was also consistent with the notion that the RAS1 regulatory response to the squalene synthase inhibitor was not due to its ability to inhibit the farnesyl transferase activity. Thus, RAS1 mRNA levels are regulated in an isoprenoid-dependent, yet farnesylation-independent manner. Although other modes of regulation cannot be ruled out, it would seem that transcription is the most likely candidate.
RAS2 expression levels remained essentially unchanged in all of the treatments examined (Figure 4). Although previous studies revealed that RAS2 mRNA levels were lowered in response to limited isoprenoid biosynthesis in a farnesylation-independent fashion, this effect reflected altered mRNA stability and was mediated by sequences within the RAS2 coding region (22). These sequences were not included in the RAS2-GFP reporter used in this study. Thus, the RAS2 reporter would not be expected to reflect this regulation. The data reported here combined with earlier studies revealed that both RAS1 and RAS2 mRNA levels were reduced in response to blocks in the isoprenoid biosynthetic pathway. However, the shared regulation was mediated by different mechanisms. This is reminiscent of the HMG1 and HMG2 isozymes. Both HMG-CoA reductase isozymes are subject to feedback regulation by the isoprenoid pathway, but by distinct mechanisms (8) (26). At present, we do not have a clear picture of whether the response of Ras gene expression to blocks in the isoprenoid pathway reflects a crude response of growth-regulating genes to blocks in the synthesis of an essential component, or whether there is a more specific regulatory response at work. Certainly the lack of a response of RAS1 and RAS2 to late blocks implies the potential for a specific regulatory axis.
Concluding remarks
The application of broad-based gene expression profiling to the regulation of the isoprenoid pathway and associated pathways provided a much richer picture of which steps are regulated and which steps are not. Several simple pictures emerged. For example, we found little evidence that blocks in isoprenoid biosynthesis affected the expression of genes encoding enzymes that prenylate proteins or tRNA. Similarly, coenzyme Q biosynthesis is largely unaffected at the level of gene expression by these blocks.
The most noteworthy regulatory effects were in opposite directions. Specifically, blocks in the isoprenoid pathway repressed expression of a large number of the genes in the mating pheromone response pathway. In contrast, the enzymes catalyzing the central isoprenoid pathway for the synthesis of sterols presented a mosaic of some genes that were induced and others that were relatively unaffected. The study identified ERG12, ERG8, ERG19, ERG9, and ERG2 as the most consistently induced genes in the pathway, and HMG-CoA synthase (ERG13) and one of the isozymes of HMG-CoA reductase (HMG2) as relatively nonresponsive members of the pathway.
Perhaps the most surprising results came from analysis of the various azole inhibitors, which had remarkably different effects. The differences in the GRM expression profiles caused by the various azoles suggested that they may have additional bioactivities other than the inhibition of lanosterol demethylase.
|
ACKNOWLEDGMENTS |
|---|
We thank J. Bergstrom (Merck) for providing L659-699 and zaragozic acid. Thanks to both Miranda Yee and Millicent Yee for their sequencing efforts. We also thank the staff at Acacia Biosciences and the members of the Rine laboratory at UC Berkeley who provided much support and many words of encouragement.
Manuscript received October 26, 1998; and in revised form January 22, 1999.
Abbreviations: GRM, Genome Reporter Matrix(TM); PCR, polymerase chain reaction; MCS, multiple cloning site; ORF, open reading frame
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-demethylase (ERG11). J. Biol. Chem. 267):2046-2056.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  G. Giaever, P. Flaherty, J. Kumm, M. Proctor, C. Nislow, D. F. Jaramillo, A. M. Chu, M. I. Jordan, A. P. Arkin, and R. W. Davis Chemogenomic profiling: Identifying the functional interactions of small molecules in yeast PNAS, January 20, 2004; 101(3): 793 - 798. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Jain, I. Akula, and T. Edlind Cyclic AMP Signaling Pathway Modulates Susceptibility of Candida Species and Saccharomyces cerevisiae to Antifungal Azoles and Other Sterol Biosynthesis Inhibitors Antimicrob. Agents Chemother., October 1, 2003; 47(10): 3195 - 3201. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Setty, A. E. Mayo, M. G. Surette, and U. Alon Detailed map of a cis-regulatory input function PNAS, June 24, 2003; 100(13): 7702 - 7707. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. W. Henry, J. T. Nickels, and T. D. Edlind ROX1 and ERG Regulation in Saccharomyces cerevisiae: Implications for Antifungal Susceptibility Eukaryot. Cell, December 1, 2002; 1(6): 1041 - 1044. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. L. Smith and T. D. Edlind Histone Deacetylase Inhibitors Enhance Candida albicans Sensitivity to Azoles and Related Antifungals: Correlation with Reduction in CDR and ERG Upregulation Antimicrob. Agents Chemother., November 1, 2002; 46(11): 3532 - 3539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Swain, K. Baudry, J. Stukey, V. McDonough, M. Germann, and J. T. Nickels Jr. Sterol-dependent Regulation of Sphingolipid Metabolism in Saccharomyces cerevisiae J. Biol. Chem., July 12, 2002; 277(29): 26177 - 26184. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. P. Kontoyiannis, G. S. May ;, M. D. De Backer, W. H. M. L. Luyten, and H. V. Bossche Identification of Azole-Responsive Genes by Microarray Technology: Why Are We Missing the Efflux Transporter Genes? Antimicrob. Agents Chemother., December 1, 2001; 45(12): 3674 - 3676. [Full Text] [PDF]
|
|
|
|
 |
|
 T. K. Van Dyk, E. J. DeRose, and G. E. Gonye LuxArray, a High-Density, Genomewide Transcription Analysis of Escherichia coli Using Bioluminescent Reporter Strains J. Bacteriol., October 1, 2001; 183(19): 5496 - 5505. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Vik and J. Rine Upc2p and Ecm22p, Dual Regulators of Sterol Biosynthesis in Saccharomyces cerevisiae Mol. Cell. Biol., October 1, 2001; 21(19): 6395 - 6405. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kalir, J. McClure, K. Pabbaraju, C. Southward, M. Ronen, S. Leibler, M. G. Surette, and U. Alon Ordering Genes in a Flagella Pathway by Analysis of Expression Kinetics from Living Bacteria Science, June 15, 2001; 292(5524): 2080 - 2083. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. D. De Backer, T. Ilyina, X.-J. Ma, S. Vandoninck, W. H. M. L. Luyten, and H. Vanden Bossche Genomic Profiling of the Response of Candida albicans to Itraconazole Treatment Using a DNA Microarray Antimicrob. Agents Chemother., June 1, 2001; 45(6): 1660 - 1670. [Abstract] [Full Text]
|
|
|
|
 |
|
 C. T. Beh, L. Cool, J. Phillips, and J. Rine Overlapping Functions of the Yeast Oxysterol-Binding Protein Homologues Genetics, March 1, 2001; 157(3): 1117 - 1140. [Abstract] [Full Text]
|
|
|
|
|
|
T. K. Van Dyk, Y. Wei, M. K. Hanafey, M. Dolan, M. J. G. Reeve, J. A. Rafalski, L. B. Rothman-Denes, and R. A. LaRossa A genomic approach to gene fusion technology PNAS, February 8, 2001; (2001) 41620498. [Abstract] [Full Text]
|
|
|
|
|
|
G. F. Bammert and J. M. Fostel Genome-Wide Expression Patterns in Saccharomyces cerevisiae: Comparison of Drug Treatments and Genetic Alterations Affecting Biosynthesis of Ergosterol Antimicrob. Agents Chemother., May 1, 2000; 44(5): 1255 - 1265. [Abstract] [Full Text]
|
|
|
|
|
|
T. K. Van Dyk, Y. Wei, M. K. Hanafey, M. Dolan, M. J. G. Reeve, J. A. Rafalski, L. B. Rothman-Denes, and R. A. LaRossa A genomic approach to gene fusion technology PNAS, February 27, 2001; 98(5): 2555 - 2560. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
