Identification of a class ofSaccharomyces cerevisiae mutants defective in fatty acid repression of gene transcription and analysis of thefrm2 Gene

YEAST VOL. 12: 319-331 (1996)

Identification of a Class of Saccharomyces cerevisiae Mutants Defective in Fatty Acid Repression of Gene Transcription and Analysis of the frm2 Gene MICHAEL W. McHALEt, K. DUBEAR KROENINGt A N D DAVID A. BERNLOHR*tl

fDeyurtnierit of' Biockcwiistry und jYristitLite of Hiiniun Genetics, Universiti~ of Miriticsoia, 1479 Gortrier A w i u r , St Puul. M N 5.51 08, U. S. A .

Received 7 August 1995; accepted 26 September 1995

Exogenous fatty acids transcriptionally control the expression of a wide variety of eukaryotic genes, many of which encode proteins involved in lipid metabolism. To identify gene products involved in the lipid signalling pathway, a reporter plasmid containing the 5'-upstream region of a gene demonstrated to be repressed by unsaturated fatty acids ( O L E I ) was fused in frame to the Escl7ericliiu coli gene IucZ encoding p-galactosidase. Succhuromj,ces cerevisiue mutants defective in transcriptional control by lipids were identified and this class of mutants has been named f ini (fatty acid repression mutant). The mutants were organized into six complementation groups designated ,frn71-6. Mutants from two of the complementation groups, j n i l and .frni3, were also defective in their ability to activate a reporter construct containing the 5'-upstream region of POXI . POX1 has been shown to be transcriptionally activated in the presence of unsaturated fatty acids. @in2 was rescued by a region of DNA localized to chromosome 111. This region contained an open reading frame of 579 nucleotides predicted to encode a M,. 21 116 polypeptide. The upstream region of FRMZ contained a number of potential response elements which have previously been identified as important in regulating gene expression in response to glucose and certain fatty acids. Consistent with this observation, kicZ activity driven by FRMZ or fin12 promoters was induced two- to three-fold dependent upon the carbon and fatty acid source utilized. The properties of FRM2 suggest that it functions in the fatty acid signalling pathway and that i t is itself regulated by fatty acids

KEY WORDS ~ O L E l ; lipids; oleic acid

INTRODUCTION

Lipids are essential to the proper functioning of cells, serving as basic components of membranes, as energy sources through P-oxidation, and as storage forms via triglycerides. The importance of lipids to the overall functioning of cells makes the enzymes involved in the cellular handling of lipids natural targets for regulation. Fatty acids can positively or negatively modulate the activity or function of a variety of enzymes and other proteins involved in signal transduction. Protein kinases, phospholipases, ion channels and pumps, and the enzymes of the cyclic AMP signalling pathway (reviewed by Sumida et al., 1993) are all regulated by fatty acids.

*Corresponding author.

A number of genes whose proteins are involved in fatty acid metabolism in Succ/zaron~yces cerevisiue have been shown to be regulated by the presence of fatty acids. This includes an increase in transcription of the genes encoding acyl-CoA oxidase (Einerhand et ul., 1992), the acyl-CoA binding protein (Rose et al., 1992), and thiolase (Einerhand et ul., 1991) and a decrease in the transcription of genes involved in fatty acid syn- thesis such a stearoyl-CoA desaturase (Bossie and Martin, 1989) and acetyl-CoA carboxylase (Kamiryo et a/., 1976). Analysis of the 5'-upstream regions of many of the genes known to be regulated by fatty acids in yeast has identified a number of upstream response elements (UREs) necessary for lipid regulation. Examples of UREs implicated in the positive regulation of genes in response to fatty acids include the peroxisome box (Einerhand

CCC 0749-503>(/96/0403 19-1 3 '0 1996 by John Wiley & Sons Ltd

320

et al., 1993; Filipits et al., 1993), the oleic acid response element (Sloots et al., 1991), and the ADRl binding site (Simon et al., 1991). In addition to the ADRl binding site, there are other UREs located in the 5’ regulatory region of fatty acid controlled genes that have been implicated in responsiveness to carbon sources. One of these is a repression sequence identified in the peroxisomal genes POXl and FOX3 which code for acyl-CoA oxidase and thiolase, respectively (Wang el al., 1992).

Stearoyl-CoA desaturase (SCD) is the A-9 fatty acid desaturase which catalyses the formation of a double bond between the 9th and 10th carbons of palmitoyl(16:O) and stearoyl(l8:O) CoA substrates to form palmitoleoyl (1 6: 1) and oleoyl (1 8: 1) CoA products. This enzyme is encoded by the OLEl gene which has previously been shown to be transcriptionally regulated by certain unsaturated fatty acids (Bossie and Martin, 1989; McDonough et al., 1992). The abundance of OLEI message is also controlled post-transcriptionally depending on the nature of the unsaturated fatty acid utilized (McDonough et al., 1992). The ultimate steady- state level of the OLEI message is therefore regulated by both transcriptional and post- transcriptional mechanisms. Our laboratory has focused on the transcriptional control mechanisms that regulate OLEl expression in response to unsaturated fatty acids.

In order to elucidate the pathway(s) which enable yeast cells to sense fatty acids and subse- quently regulate the expression of the appropriate genes, a screening procedure was developed to isolate S. cerevisiae mutants that do not transcriptionally regulate OLEl in response to unsaturated fatty acids. This work describes the identification of a mutant class which does not repress an OLEl-lacZ reporter in the presence of arachidonic acid (AA). Six complementation groups have been identified and designated f r m l d Cfatty acid repression mutant 1-6). Interestingly, of the six frm mutants, two Vrml and frm3) were also defective in their ability to activate gene expression by unsaturated fatty acids. These mutants were identified using a POXI-lac2 reporter construct. POXl encodes for the fatty acid inducible acyl- CoA oxidase of S. cerevisiae (Wang et al., 1992). One of the mutants,frm2, has been rescued and its gene/gene product identified. This report describes the characterization of frm2 and the description of this gene product as being required for polyunsatu- rated fatty acid control of OLEI gene expression.

M. W. McHALE ET AL.

MATERIALS AND METHODS

Strains, growth conditions and transformations The frm mutants were isolated in the S.

cerevisiae strains YPH98 ( M A Ta ura3-52, lys2- 801, ade2-101, leu2-Al, trpl-Al) and YPH102 (MATa ura3-52, lys2-801, ade2-101, leu2-Al, his3- 6200; Sikorski and Heiter, 1989; Table 1).

Undefined rich medium for yeast cultures was 2% yeast extract-1% peptone (Difco) with 2% glucose (YPD). Agar was added to 2% for plates of the same composition. Selective medium was 0.67% yeast nitrogen base without amino acids (Difco). Appropriate amino acid supplements were added as needed to 0.004%. Yeast SD, SC and SC dropout media were of standard composition (Sherman et al., 1986). Plates used for detection of expression of P-galactosidase (DX plates) were SD plates with 0-07 M-KPO, pH 7.0, and 40 pg/ml X-gal (Biosynth AG, Switzerland). Plates used to screen for mutants were DX plates with IOO~M-AA and 0.01% (v/v) Tween-40 (DXTA plates). DXTA,, plates have the same composition as DXTA except the AA is at 25 p ~ . The pH values of the DX, DXTA and DXTA,, plates were consistently between 6.9 and 7.0, well within the suggested pH range for the blue colour production from X-gal. Parental cells harbouring the OLEl-lacZ fusion plasmid (pMM2) are blue on non-repressing DX plates and white on repressing DXTA plates. Medium containing oleic acid as a sole carbon source (YPO) consisted of 2% yeast extract-1% peptone with 0.10/0 (v/v) oleic acid and 0.01% (v/v) Tween-40. All fatty acids were obtained from Nu Check Prep (Elyssian, MN) at the highest available purity. Medium used for activation of the POXI-lac2 fusion was YEOG (Wang et al., 1992), 0.3% yeast extract, 0.5% peptone (Difco), 0.05% galactose, 0.5% KPO, (pH 6.0), 0.1% (v/v) oleic acid and 0.01% (v/v) Tween-40. Plates containing 1 mg/ml 5-fluoroorotic acid (5-FOA) were used to select yeast which had shed a ura-containing plasmid (Boeke et al., 1987).

Transformation of Escherichia coli was performed using either the CaCl, method (Sambrook et ul., 1989) or by electroporation (BTX, San Diego, CA). Transformation of S. cerevisiae was performed using either the lithium acetate method (It0 et al., 1983) or by electroporation. Electroporation of S. cerevisiae was conducted according to the manufacturer’s recommendations.

FATTY ACID REGULATION 321

Table 1.

Strains Use Features

Plasmids and strains used in this study.

YPH98 YPH102

Plasmids pMM2 pMM5 pMM21 pLBGl p2MTZ

p25-3c pDK5 pDK6 pDK7 pDK8 pDK9 pDKlO pDKl1

p25- 1 c

Parental strain of frm mutants Parental strain o f f rm mutants

Low copy repression reporter Integrating repression reporter Low copy activation reporter FRMZ expression reporter frm2 expression reporter Rescue of frm2 mutant Rescue of frm2 mutant Rescue of frm2 mutant Rescue offrm2 mutant Rescue of frm2 mutant Rescue of frm2 mutant Rescue of frm2 mutant Rescue of frm2 mutant Rescue of frm2 deletion-disruption

M A Ta MATa

OLEl-lacZ fusion OLEl-lacZ fusion POX1-lucZ fusion FRM2-lucZ fusion frm2-lacZ fusion - 15 kb of chromosome I11 DNA, includes FRM2 - 15 kb of chromosome I11 DNA, includes FRM2 -5 kb of chromosome I11 DNA, includes FRM2 Fragment of FUSl and 3' UTR of FRM2 Fragment of YCL25C and FRM2 (no 3' UTR) 949 bp PCR fragment of FRM2 region 1341 bp PCR fragment of FRMZ region 2252 bp PCR fragment of FRM2 region Same as pDK10, TRP4 selectable marker

The use and salient features of each strain or plasmid used in the study are indicated. YPH98 and YPHlO2 strains generated by Sikorski and Hieter (1989). All plasmids listed except for p25-lc and p25-3c were generated in this study. p25-lc and p25-3c were identified from a yeast genomic library (Rose et al., 1987) as rescuers of thefrmZ mutant and were determined to contain regions of chromosome 111. UTR, untranslated region.

Screen for frm mutants Ethylmethane sulphonate (EMS) mutagenesis

was carried out as previously described (Lawrence, 1991) at a final concentration of 1.2-1.8%. This concentration of EMS typically resulted in 30-70% survival for both YPH98 and YPH102 parental strains. YPH98 or YPH102 cells containing an OLEl-lacZ reporter (pMM2, see below) were mutagenized with EMS. The mutagenized cells were either plated directly onto SD dropout medium and grown at 30°C until colonies became visible or were allowed to recover from 1 h to overnight in YPD medium at 30°C prior to plating. The surviving cells were replica plated to DX and DXTA plates and those cells displaying a blue colour phenotype on both plates were selected as potential frrn mutants. The mutants were further analysed by shuttling out their pMM2 reporters with growth on 5-FOA media and retransforming with new pMM2 plasmids. The retransformed mutants were screened for retention of the blue colour phenotype on both DX and DXTA plates and those mutants that still exhibited the frm phenotype were shuttled through 5-FOA a second time and transformed with an integrating reporter

construct pMM5. The pMM5 containing cells were then assayed for their colour on DX and DXTA plates to ensure retention of the frrn phenotype. The above mutant screen was carried out three independent times.

P-Galactosidase assays YPH98, YPH102 and frrn mutant strains con-

taining various reporter constructs as indicated were assayed for P-galactosidase activity essentially as described (Ausubel et al., 1987) with fatty acids dissolved in ethanol and 0.01% Tween-40 added as indicated. Two or three independent transformants were assayed in triplicate for each experiment. Cell densities were determined by measurement at A600 using a Beckman DU-70 spectrophotometer and the value of A,,, of 1 =0.5 x lo7 cells/ml. Total protein was determined by the method of Bradford (1976) using the Bio-Rad assay kit.

[3H]Oleic acid transport assays Uptake of [3H]oleic acid (Amersham, Arlington

Heights, IL) was determined as a modification of a previously described protocol (Kohlwein and

322 M. W. McHALE ET AL.

Paltauf, 1983). Cells were grown to midlog phase in selective media plus needed amino acids and 5-6 h prior to assay diluted into YPD plus 100 ~ L M - oleic acid in ethanol and 0.01% (v/v) Tween-40. The cells were harvested and washed with 50 mM- KPO, (pH 5.0), resuspended in 990 pl of the same buffer and incubated for 15 min at 30°C. To the warmed suspension, lop1 of [3H]oleic acid mix (mix=200 p1 of 100 pwoleic acid in ethanol and lop1 of [3H]oleic acid in toluene, 10 Ci/mmol specific activity; Amersham) was added. The cells and oleic acid mix were vortexed and at various times 50 pl aliquots were removed and placed into culture tubes containing 400 pl trichloroacetic acid (5% (w/v)) and 800 p1 light white mineral oil (Sigma). The suspension was vigorously vortexed and placed into Eppendorf centrifuge tubes. The cells were pelleted by brief centrifugation and the mixture was frozen on dry ice. The cell pellets were cleaved from the bottom of the tubes and the amount of radioactivity was measured.

Yeast genetic manipulations Mating, sporulation, mating type switching and

tetrad dissection were performed essentially as described (Sherman et al., 1986). Backcrosses were performed first into the laboratory strain (YPH98 or YPH102) of the opposite mating type containing pMM5, then each subsequent backcross was into the YPH98 containing pMM5 background. After each round of backcrossing, retention of the frm phenotype was scored on DXTA plates. Complementation analysis was performed by mating mutants of opposite mating type to each other as well as to the wild-type controls and growing the resultant diploids on DXTA media. Diploids were scored as blue or white. This procedure was done at least three times for each possible cross.

Plasmids In order to screen for yeast mutants which were

unable to transcriptionally regulate the OLEl gene, the plasmid pMM2 was constructed by cloning the 5’-upstream 935 bp SulI fragment of the S. cerevisiae OLEl gene which encodes SCD (Stuckey et al., 1990) into the SulI site of YEp356 (Meyers et al., 1986). This upstream fragment also encodes the first 25 amino acids of SCD which was fused in-frame to lucZ encoding P-galactosidase. An identical reporter plasmid has been previously described by others (McDonough’et al., 1992). The resulting plasmid (pDK1) was cleaved with EcoRI

and NcoI restriction enzymes (Promega) and the fragment containing the OLEl-1ucZ fusion and part of the URA3 gene was ligated into the vector pSLll90 (Brosius, 1989). This plasmid (pSLl190- pDK1) was digested with BumHI and the OLE1- lacZ fusion containing fragment was cloned into the BumHI site of YCp50 (Rose et al., 1987). This low copy reporter plasmid with P-galactosidase under the control of the OLEl regulatory region was used to screen for mutant colonies.

An integrating reporter plasmid with LEU2 as a selectable marker (pMM5) was constructed by first removing the EcoRI-NarI fragment containing l a c 2 from YIplac128 (Gietz and Sugino, 1988). The ends of the remaining portion of the plasmid were blunted by using a Klenow fill-in reaction and the plasmid was sealed by ligation (Sambrook et al., 1989). This plasmid was termed YIp128dlz. The BumHI fragment containing the OLEl-lucZ fusion from pMM2 was cloned into YIp128dlz, thus making pMM5.

An activation reporter plasmid (pMM21) was constructed placing lucZ under the transcriptional control of POXl, the S. cerevisiue gene encoding acyl-CoA oxidase (Wang et ul., 1992). The polymerase chain reaction (PCR) was employed (Perkin Elmer Cetus, Emeryville, CA) to clone 454 bp of the 5’-upstream region into the Hind111 site of pMM2 which had the OLEl upstream region removed. The 5’ primer for pMM21 was 5’-CGCAAGCTTTCGACCAAAAAAAG and the 3‘ primer was 5’-CGGGATTAAGCTTAG TACGTC and the plasmid pAD17 (Wang et al., 1992) was used as template. The pMM21 low copy reporter plasmid, containing the POXl upstream fragment also encodes the first five amino acids of acyl-CoA oxidase fused in-frame to lacZ. This construct was used to identify mutants that failed to activate transcription in response to fatty acids. A similar reporter was used previously (Wang et al., 1992).

Two genomic plasmids (p25-lc and p25-3c) were isolated from a library screening procedure used to rescue the frm2 mutant (Fig. 2). The genomic library has been described previously (Rose et al., 1987) and consists of the plasmid YCp50 containing partially Suu3A restricted S. cerevisiae genomic DNA. The cloned genomic DNA was approximately 12-20 kb in length and was cloned into the BamHI site of YCp50. A 5 kb EcoRI complementing fragment was subcloned into the EcoRI site of YCp50 and was termed pDK5. The plasmids pDK6 and pDK7 were constructed from digesting

FATTY ACID REGULATION

pDK5 with EcoRI and HindIII and subcloning the resulting fragments into the EcoRI and HindIII sites of YCp50. The plasmids pDK8, pDK9 and pDKlO were produced by using oligonucleotides as primers for PCR and the complementing plasmid p25-lc as the template. The 5' primer for pDK8 was 5'-GACTGAAGCTTTCAGTTT and the 3' primer was 5'-GCGAGGATCCTGTTC AGTG. The resulting 1839 bp PCR product contained 352 bp of the 5'-upstream region of FRM2, the 576 bp FRM2 coding region and 911 bp 3' of the translation termination of FRM2 and was cloned into the HindIII and BamHI sites of YCp50. The 5' primer for pDK9 was 5'- TAGAATTCGTGTAATGATAGACG and the 3' primer was the same as that for pDK8. The resultant 2159 bp PCR product contained 672 bp of the 5'-upstream region of FRM2, the 576 bp FRM2 coding region and 911 bp 3' of the translation termination of FRM2 and was cloned into the EcoRI and BamHI sites of YCp50. Construc- tion of pDKlO was accomplished by using the same 5' primer as that used for pDK9 while the 3' primer was 5'-GATTACGGATCCAAAGGCC. The 2237 bp PCR product was cloned into the EcoRI and BamHI sites in YCp50. This region of DNA includes 672 bp of 5'-upstream region of FRM2, the 576 bp FRM2 coding region and 989 bp 3' of the translation termination of FRM2. The plasmid pDKlOT was engineered by cleaving pDKlO with EcoRI and BamHI restriction enzymes and ligating the resulting 2237 bp fragment into YCplac22 (Gietz and Sugino, 1988). pDKl1 was identical to pDKlO except it had a TRPl selectable marker. At least three independent clones were tested for each plasmid constructed.

A reporter plasmid (pLBG1) was constructed to analyse the expression of the FRM2 gene. This was accomplished by utilizing PCR to clone the 5'- upstream region of FRM2 into YEp357 (Meyers et al., 1986) which allowed for expression of p-galactosidase under the control of the FRM2 response elements. The oligonucleotide primers, 5'-GGGAAGCTTTGGGGACATTTCTTTGGC and 5'-TAGAATTCGTGTAATGATAGACG were used as PCR primers using p25-3c as the target DNA. The resulting 685 bp product was cloned into the EcoRI and HindIII sites of Yep357. A reporter plasmid (p2MTZ) was constructed to analyse the expression of the frm2 gene. The same approach was used as for pLBGl but the template for PCR was genomic frm2 DNA.

323

Creation of a frm2 deletion-disruption Plasmid p25-lc was restricted with the enzyme

BglIl liberating a 1967 bp fragment containing the entire coding region of FRM2. This fragment was ligated into the BamHI sites of pGEM-3Zf( - ) (Promega). The resulting construct was restricted with Age1 and BclI enzymes, thereby removing a 440 bp fragment. This fragment was made up of 376 bp of the 5'-upstream region and a 64 bp of the coding region of FRM2. The restriction sites were blunted with Klenow and a 2180 bp SmaI fragment containing (IRA3 from pMM21 was ligated into the blunted sites. Finally the deletion- disruption construct was restricted with EcoRI and HindIII, releasing a 3547 bp fragment that was transformed into our wild-type strain YPH98.

DNA sequence analysis and nucleic acid blots Double-stranded plasmid templates were iso-

lated according to standard methods (Sambrook et al., 1989). DNA sequencing was performed using the dideoxy-chain-termination method of Sanger et al. (1977). Sequence analysis was performed by using the IG suite of programs (rel. 5.4) or the GCG suite of programs (ver. 7, Madison, WI). Southern blots were performed as described (Sambrook et al., 1989) to verify the chromosomal location of yeast integrating plasmids and to determine the number of copies of FRM2 in the yeast genome. Double-stranded probes were labelled either with a random primer kit (Boehringer Mannheim Biochemicals, Indianapolis, Ind.) or by the method of Feinberg and Vogelstein (1983).

RESULTS

Establishment of a screening system The OLEl system (Bossie and Martin, 1989;

Stukey et al., 1990) was chosen to establish a screening system in S. cerevisiae that would allow identification of the members of a signalling pathway that was responsive to fatty acids. The S. cerevisiue OLEl gene encodes for SCD which has been shown to be repressed by the presence of unsaturated fatty acids. A reporter plasmid (pMM2) was constructed using the 935 bp of the 5'-upstream region fused in-frame to IacZ. The OLEl-lacZ plasmid was transformed into laboratory strains YPH98 and YPH102 and cells harbouring pMM2 became blue when grown on media containing X-gal (DX plates). A plate screening system was developed taking advantage


103

80 c .- % a t $ 4

0 0 10 50 100

Arachidonic Acid (pM)

Figure 1. Arachidonic acid repression of an OLEI-lacZ reporter (pMM2) in YPH98 or YPH102 strains. YO Repression=number of white colonies per platehotal number of colonies per plate. Blue colonies indicate expression and white colonies indicate repression of the OLEl-lucZ reporter. Plates are DXTA with varying amounts of arachidonic acid as indicated. Closed circles=YPH102, open circles=YPN98. A typical repression curve is shown.

of the OLEl-lac2 fusion's repression by unsaturated fatty acids which resulted in colony colour remaining white when grown in the presence of AA and X-gal (DXTA plates). The concentration of AA that would result in 50% repression of the OLEI-lac2 reporter was less than 10 p~ (Fig. 1).

Isolation of mutants unable to repress OLEl-lacZ In order to identify mutants which were unable

to transcriptionally regulate the expression of the OLEl-lacZ fusion, yeast strains YPH98 and YPH102 containing the reporter plasmid pMM2 (OLEl-IacZ) were mutagenized with EMS and potential fatty acid repression gutants (firms) were identified. The potential frrn mutants demonstrated constitutive expression of p-galactosidase (blue colonies) under normally repressing conditions (white colonies) on DXTA plates. The mutants were further analysed by causing the cells to shed their pMM2 reporter plasmid (5-FOA plates) and then retransforming with new pMM2 plasmids to eliminate any mutants that might have arisen due to mutations in the reporter plasmid. Those cells that still exhibited the frrn phenotype on DXTA plates, were shuttled through 5-FOA a second time and transformed with an integrating reporter construct pMM5. Southern blots verified the integration of pMM5 in the 5'-upstream region of the OLE1 gene (data not shown). The pMM5- containing cells were then assayed for their lack of repression on DXTA plates to ensure retention of the frrn phenotype. This procedure identified 50 genomic mutants which were further analysed.

Analysis of the frm mutants and determination of complementation groups

The 50 frrn mutants were tested for their ability to grow on different carbon sources including oleic acid. Of the 50 f r m mutants, four did not grow on oleic acid as a sole carbon source. Kunau and colleagues had used non-viability on oleic acid as a sole carbon source to identify at least 19 different peroxisome assembly complementation groups (Kunau et al., 1993). Growth of the remaining 46 frms on medium containing oleic acid as the sole carbon source indicated that there were no gross defects in peroxisome function as the mutants grew as well as the wild-type strains.

Of the 50 f r m mutants, 15 were in the YPH98 (MATa) background and 35 were in the YPH102 (MATa) background. Mutants of opposite mating types were crossed in each possible combination and to the opposite wild-type parent to determine complementation groupings of the mutant alleles. The resulting diploids were replica plated, along with appropriate wild-type diploid, wild-type haploid and mutant haploid controls, to DX and DXTA plates. A dominant mutant was defined as one which retained its blue f r m phenotype when crossed to a wild-type parent. The phenotypes of all colonies were scored and tabulated. The procedure was carried out at least three independent times for all possible crosses. The 50 f r m mutants were divided into two groups of nine dominant and 41 recessive mutants. The recessive mutants comprise six different complementation groups, termed f r m l d . Representative members of the f r m l - f complementation groups containing pMM5 were backcrossed for five generations to eliminate mutations not directly related to the f r m phenotype to a 97.5% confidence level. The frml wild-type tetrads segregated 2:2 for the frrn phenotype, indicating a single gene or two tightly linked genes were responsible for the frrn phenotype. The dominant mutants were not analysed further.

Identijication of frms that fail to activate transcription in response to fatty acids

To identify which, if any, of the frrn strains were also defective in the ability to activate gene expression by fatty acids, representative members of each of the six frrn complementation groups were transformed with pMM21, the activation reporter construct. pMM21 is a low copy plasmid that has lac2 under the transcriptional control of P O X l , the S. cerevisiae gene encoding the fatty


and termed pDK5. Sequencing portions of pDK5 revealed that it was part of chromosome I11 of S. cerevisiae, the entire sequence of which has been deposited in GenBank (Oliver et al., 1992). The region of genomic DNA identified in pDK5 contained an open reading frame (ORF) which had been sequenced from strains XJ24-24a and AB972 and reported as part of the complete sequence of yeast chromosome I11 (Oliver et al., 1992). The putative ORF YCLX8c was renamed FRM2 according to its mutant phenotype. pDK5 also contained two partial ORFs which have been previously identified. Located 5' to FRM2 was the carboxy-terminal one-third of an ORF which has sequence similarity to a general amino acid permease (Oliver et al., 1992). The carboxy- terminal two-thirds of FUSl, a gene involved in cell fusion, was found 3' to FRM2 (Fig. 2). To rule out the possibility that either of the partial ORFs was responsible for the frm2 phenotype, plasmids pDK6 and pDK7 were constructed (Fig. 2); nei- ther of these plasmids rescued the frm2 mutant. pDK8 and pDK9 which include the entire FRM2 ORF, but differ in the amount of 5' and 3' untranslated regions they contain (Fig. 2), do not rescue frm2. pDKlO includes 672 bp of the 5'- upstream region of FRM2, the FRM2 ORF and 989 bp 3' of the translation termination of FRM2 did rescue frm2 (Fig. 2).

Creation of a frm2 deletion-disruption In order to prove that the region of DNA

identified in rescuing the frm2 mutant was allelic to the original frm2 mutant, a deletion-disruption of the FRM2 gene was generated. The deletion- disruption construct was transformed into the laboratory strain YPH98 and URA3 prototrophs were selected. A Southern blot of the transformants was performed to ensure a homologous integration event had occurred, replacing 376 bp of the 5'-upstream region and 64 bp of the coding region of FRM2 with a 2180 bp URA3 fragment (data not shown). Interestingly, the resulting frm2 deletion- disruption had an increased sensitivity to AA and was not viable when grown in the presence of 50 ~ L M (or greater) AA (data not shown). The frm2 deletion-disruption was transformed with the pMM5 integrating reporter plasmid and assayed for the frm phenotype on DXTA,, (same composition as DXTA but AA at 25 p~ rather than 100 p ~ ) plates. The frm2 deletion-disruption dis- played the same blue colour phenotype on the DXTA,, repression plates as the original frm2

Table 2. Effects of oleic acid on POXI-lacZ fusion activities in parental and frm strains.

S-Galactosidase activity (unitdmg protein)

Media Fold

Strains YPD YEOG activation

YPH98 YPH102 frml frm2 frm3 frm4 YPH98/YPH 102 frmllYPH102 frm3NPH98 frml ljirm3

5.51

3.02

0.796

16.0

16.9

12.6 30.4 41.1 76.3 84.8

20 600 3740 29 400 1840 8.32 2.75

21 400 1270 3.75 4.71

38 800 3080 21 200 697 19 800 482 16 100 21 1 18 900 223

p-Galactosidase activities of parental haploids (YPH98 or YPH102), frm haploids ( f r m l 4 ) , parental diploid (YPH98/ YPH102), frmlparental diploids (frmlNPH102 or frm3/ YPH98) and frmlfrm diploid (frmllfrm3) strains harbouring pMM21 (POXI-lucZ fusion). Strains grown in YPD (basal) or YEOG (activation) media. Fold activation equals activation activity divided by basal activity. One representative assay of three is shown.

acid inducible acyl-CoA oxidase (Wang et al., 1992). Table 2 shows that frml and frm3 failed to activate transcription in response to an unsaturated fatty acid, while frm2 and frm4 showed wild-type activation. Diploids of wild-type, frml wild-type or frmlfrrn genotype were constructed and transformed with the pMM21 activation reporter. The frmllfrm3 diploid restored wild- type activation which agreed with the fatty acid repression phenotype analysis that placed frml and frm3 in different complementation groups.

Cloning a genomic fragment which rescued the frm2 mutant

The frm2 mutant was chosen to be the first mutant rescued based on its wild-type growth characteristics. To identify the FRM2 gene, the frm2 mutant was rescued by transforming with a yeast genomic library (Rose et al., 1987) and identifying white colonies (wild-type expression) from a background of blue colonies v r m expression) on plates containing AA. Two plasmids containing genomic DNA (p25-lc and p25-3c) rescued the frm2 mutant (Fig. 2). The genomic DNA from p25-lc was subcloned to an approximately 5 kb rescuing fragment in Ycp50


A

HIS4 YCL28W YCL23C YCL22C

BIKI FUSl FRM2 YCL25C YCL24W

I i n B n n n n

B Rescues frm2

p25-3c Yes

p25-lc Yes

pDK 5 Yes

pDK6 No

pDK7 No

pDK8 No

pDK9 No

pDK 10 Yes

pDKl1 Yes

-

Figure 2. (A) Open reading frames on S. cerevisiae chromosome 111 between 65 500-83 500. Boxes indicate open reading frames and relative locations on chromosome 111. Open reading frames coding for known genes are indicated by gene name. Unknown open reading frames are named using the nomenclature of Oliver et ul. (1992). YCLX8C, has been renamed FRM2. (B)frmZ rescuing plasmids. List of plasmids identified and constructed in rescuing of the frm2 mutant. The lines indicate the region of chromosome 111 contained in each plasmid. The success or failure of each plasmid to rescue frmZ is listed on the right. Successful rescue of frm2 was indicated by loss of the f r m phenotype (blue colonies) and return to the repression phenotype (white colonies) on DXTA plates.

mutant. The frm2 deletion-disruption phenotype was rescued when transformed with pDKl1 and plated to DXTA,, media. pDKl1 also rescued the original frm2 mutant (Fig. 2). The rescued frm2 deletion-disruption did not display an increased sensitivity to AA as it was able to grow on the original DXTA plates. The frm2 deletion- disruption grew on plates with oleic acid as a sole carbon source and did not have a defect in the ability to transport [3H]oleic acid (results not shown).

A pMM5-containing frm2 deletion-disruption strain was crossed to a five times backcrossed pMM5-containing frm2 mutant strain and the resulting diploid was sporulated. Tetrad dissection

on 11 tetrads showed a 4:O segregation of the frm phenotype to wild-type phenotype when the resulting spores were replica plated to DXTA,, plates. The 4:O segregation of thefrm phenotype indicated the frrn2 deletion-disruption was allelic to the original frm2 mutant.

Nucleotide analysis of FRM2 One copy of FRM2 was detected via Southern

blotting of restriction-digested genomic DNA from the YPH98 strain. Computer analysis of the FRM2 ORF predicted a protein of 192 amino acids and a molecular mass of 21 116 Da. A number of potential phosphorylation sites in the predicted amino acid sequence of FRM2 were


TATAAA

-77

+I38

I FRM2 Stop +353

Figure 3. Putative upstream response elements in FRM2. Numbers are relative to predicted translation start for FRM2. FRM2 Start=predicted translation start site, FRM2 Stop= predicted translation stop site, TATAAA and TATATA=possible TATA boxes, OGR=overcoming glucose repression (Gancedo, 1992; Sarokin and Carlson, 1985), Abfl =ARS-binding factor (Rose et al., 1992), PBOX=peroxisome box (Einerhand et al., 1993; Filipits et al., 1993), ADRl =ADRl transcriptional activator (Chen et al., 1994), URS=upstream repression sequence (Wang et al., 1992), OAR=oleic acid response element (Simon rt ul., 1991), STRE=stress response element (Schuller et al., 1994).

identified. These include three potential protein kinase C sites at residues 17, 67 and 90, and a CAMP-dependent protein kinase phosphorylation site at residue 112. Additionally there are two potential casein kinase I1 phosphorylation sites at residues 91 and 138. The significance of these sites, if any, is currently unknown. Comparison of FRM2 with other known proteins contained in data banks using the FASTA program (Pearson and Lipman, 1988) revealed no significant simi- larities. The nucleotide sequence surrounding the putative initiating methionine in FRM2 (AAGAAAUGUCC) was found in a good consensus sequence for highly expressed genes (5‘-A[A/Y] A[A/Y]AAUGUC[U/C]-3’ where Y stands for a pyrimidine (Hamilton et al., 1987). There are two potential TATA boxes located 79 bp (TATAAA) and 46 bp (TATATA) 5’ of the predicted translational start of FRM2. There is a good poly- adenylation sequence (AATAAA) located 172 bp 3’ of the predicted translational stop codon for FRM2. The codon bias index is 0.4 for the predicted protein, thus indicating that this gene would likely only be moderately expressed (Bennetzen and Hall, 1982). Sequence analysis indicates that no introns are likely to be present in FRM2 as there is no TACTAAC box (Langford et al., 1984).

Analysis of the 5‘-upstream region of FRM2 indicated that there are a number of sequences

which are similar to UREs found to be involved in fatty acid or glucose regulation of gene expression (Fig. 3). There are eight overcoming glucose repression (OGR) elements located beginning at positions -677, - 618, - 583, - 542, - 506, - 399, - 31 1 and - 28 bp from the predicted start of translation. These have been identified in the upstream region of many of the SUC genes coding for invertase (Gancedo, 1992; Sloots et al., 1991). The OGR consensus sequence is defined as (A/C)(A/G)GAAAT. There are two potential ARS-binding factor 1 sites starting at positions - 677 and - 653. These sites have been reported to be involved in the activation of many yeast genes (Rose et al., 1992) and are defined by the sequence RTCRYNNNNNACG where R stands for a purine base, Y stands for a pyrimidine base and N represents any base. There are four sites similar to a reported peroxisome box found in a URE of genes that are activated by the presence of fatty acids (Einerhand et al., 1993; Filipits et al., 1993). They are defined by the sequence CGGNNNTNA and are located starting at positions - 579, - 523, - 490 and + 191. There are two sites similar to the reported ADRl transcriptional activator binding site (Chen et al., 1994) located starting at positions - 380 and +4. This binding site is defined by the sequence TTGG(A/G)GA. There is a site similar to the

M. W. McHALE ET AL.

Effects of fatty acids and carbon sources on (A) FRM2-lacZ and (B) frm2-lacZ fusion activities in the

328

Table 3. parental YPH98 strain.

P-Galactosidase activity (unitslmg protein)

Fatty acid Dextrose Fold activation Galactose Fold activation

(A) FRMZ-lacZ None 47.59 f 0.88 1 .o 5 1.20 f 2.33 1.0 Stearic acid 104.90 f 4.95 2.2 29.49 f 0.97 0.6 Oleic acid 128.55 f 2.92 2.7 45.25 f 3.10 0.9 Arachidonic acid 83.69 f 0.55 1.8 216.42 f 24.71 4.2

(B) fim2-lacZ None 109.13 f 5.03 1 .o 21.65 f 0.36 1 .o Stearic acid 177.60 f 2.25 1.6 26.40 f 0.90 1.2 Oleic acid 138.55 f 1.20 1.3 35.50 =k 1.17 1.6 Arachidonic acid 220.27 f 9.10 2.0 77.90 f 1.84 3.6

p-Galactosidase activities of the parental YPH98 strain containing pLBGl (FRM2-lucz) or p2MTZ Cfrrn2-lucz). Strains were grown in SC+dextrose (dextrose) or SC+galactose (galactose), f 25 pi-arachidonic acid, 500 pM-oleic acid or 500 pM-stearic acid. All media include 0.01% (v/v) Tween-40. Fold activation =activity in presence of fatty acid +carbon sourcelactivity in carbon source alone. Activities are reported as the mean + standard error.

upstream repression sequence (URS) starting at position - 133. This site has been reported to be involved in repression of yeast peroxisomal genes (Wang et al., 1992) and has the consensus sequence AGGTAAT. There are two sequences similar to the oleic acid response element (OAR) which has been found upstream of yeast genes that are activated in the presence of oleic acid (Simon et al., 1991). The OAR consensus sequence is reported to be (C/T)GGTT(A/G)TT(A/C/G). They are found starting at positions - 75 and +84. Finally, there is a stress response element (STRE) starting at position - 466. The STRE has been reported to be involved in activating a number of yeast genes in response to environmental stresses (Schuller et al., 1994). The STRE consensus sequence is (A/C/T) AGGGG(A/C/G).

Analysis of both FRM2 and frm2 expression The density of potential regulatory sites in the

putative 5' regulatory region of FRM2 suggested that this region may control expression of FRM2 in response to lipids and sugars. In order to test the hypothesis that the putative UREs upstream of the FRM2 gene were functional, reporter constructs pLBGl (FRM2-lacz) and p2MTZ Vrm2-lacz) were made which allowed for the expression of the lacZ gene under the control of the wild-type FRM2 upstream region or the upstream region from the

frm2 mutant. The P-galactosidase activities of wild-type YPH98 containing pLBGl or p2MTZ were determined and the results are summarized in Table 3. The cells were grown with either glucose or galactose as a carbon source and with or without 500 pM-stearic acid (C18:0), 500 pM-oleic acid (Cl8:1), or 25 ~ M - A A (C20:4). In general, both the wild-type fusion and the mutant upstream region fusion showed a two- to four-fold activation in the presence of fatty acids, and AA induced the highest activation when the strains were grown on galactose. However, there were subtle differences between the fusions' response to certain fatty acids and carbon sources. For example, FRM2-lacZ activity was activated by all fatty acids when the cells were grown on dextrose, yet AA was the only fatty acid to increase activity when the cells were grown on galactose. The frm2-lacZ construct had a higher basal activity in dextrose as compared to the wild-type fusion and was not activated by stearic or oleic acids to the same extent in dextrose as the FRM2-lacZ construct. The frm2-lacZ fusion did not display the inversion of activation by stearic or oleic acids that was observed with the FRM2-lacZ construct when the carbon source was changed from dextrose to galactose. Interestingly AA activated both constructs to approximately the same extent in either carbon source.


DISCUSSION

Regulation of genes involved in lipid metabolism is necessary for the proper functioning of cells. Although DNA sequences that are involved in the fatty acid regulation of gene expression have been identified, there are no reports of the genes or proteins that are responsible for this regulation. Mutants that lack the ability to alter transcription of certain genes in the presence of unsaturated fatty acids are needed to identify the proteins involved in this signal transduction pathway. This work describes the generation of a screening system for the isolation of mutants which do not repress on OLEl-lacZ reporter construct. This class of mutants has been named, fatty acid yepres- sion mutant (firm). Fifty frm mutants have been identified and classified into six different complementation groups, frml-6. The six frm complementation groups are not likely to be peroxisomal assembly mutants (Kunau et a/., 1993) as they can grow on oleic acid as a sole carbon source. Kunau and colleagues devised a yeast mutant selection in which the first step was an inability to grow on oleate as a sole carbon source. The second step involved screening the oleate non-utilizers for mislocalization of the peroxisomal matrix enzymes. This screening method has identified at least 19 different peroxisome assembly complementation groups (Kunau et a/., 1993). None of the frm complementation groups has an altered ability to transport [3H]oleic acid.

Two of the complementation groups, frml and frm3, also failed to activate transcription of a POXl-lacZ reporter construct in response to fatty acids. These mutants that fail to repress and fail to activate transcription in response to fatty acids may be extremely important in the lipid signalling pathway. We believe that these mutants may be acting on either end of a signalling pathway. They may be involved in sensing fatty acids which would stimulate the pathway or they may be acting at the opposite end and be involved in a transcriptional complex that regulates fatty acid-responsive genes. We are currently rescuing frml and frm3 with a genomic S. cerevisiae library in an attempt to determine the nature of the genes and their gene products. (firm mutants &6 are also currently being rescued.)

The region of chromosome I11 which rescued the frm2 mutant and the frm2 deletion-disruption contained an ORF (nucleotides 74 669 to 75 247) of unknown function termed YCLX8c (Oliver et al.,

1992), which has been renamed FRMZ indicating its mutant phenotype. The codon bias index, the good consensus sequence surrounding the pre- sumed initiating methionine, and the location of two potential TATA boxes upstream of FRM2 lead us to believe that the protein will be expressed as predicted. Analysis of the protein predicted to be encoded by FRM2 shows that there are potential phosphorylation sites for protein kinase C, casein kinase I1 and CAMP-dependent protein kinase. These potential phosphorylation sites may indicate FRM2 is a regulatory protein that is involved in integrating a lipid signalling pathway with the protein kinase C andlor CAMP regulatory pathways.

The 5’-upstream and 3’-downstream regions of FRM2 are extremely important for FRM2 expression. When trying to rescue the frm2 mutant (Fig. 2) several different constructs were made (pDKs 8-10) that included the entire FRM2 ORF but differed in the amount of 5’ and 3‘ untranslated regions they contained. The final rescuing plasmid (pDK10) contained all of the DNA between FUSl and YCL25C (672 bp of 5’-upstream region of FRM2 and 989 bp 3’ of the translation termination of FRM2; Fig. 2). At this time the significance of the 3‘-downstream region is unknown. No potential splice sites have been found in this region of DNA (Langford et al., 1984) and there is a transcription termination site 172 bp 3‘ of the predicted FRM2 stop codon. Within the 672 bp 5’-upstream region of FRM2 several potential UREs have been identified (Fig. 3). These include OGR and ADRl (ADR1 transcriptional activator) sequences which indicate expression of this gene may be linked to the carbon source utilized for growth. Additionally, there are other potential UREs that have been implicated in regulating gene expression in response to fatty acids. These include peroxisome box, URS and OAR sites. Finally there is a STRE site, which has been implicated in regulating transcription in response to a variety of stresses. An analysis of a further 5 kb of genomic DNA upstream and downstream of the FRM2 gene revealed no similar UREs to those found in the FRMZ 5‘-upstream region. These sites indicate FRM2 may be under complex transcriptional control, ranging from fatty acids to carbon source, which would argue that FRM2 plays a role in integrating the lipid signalling pathway into normal cellular homeostasis.

Analysis of the expression of FRM2, using the FRM2-EacZ reporter (pLBGI), and of frm2, using


the frm2-lac2 reporter (p2MTZ), indicates that their transcriptase is affected by the presence of fatty acids and carbon source (Table 3). There is clearly a difference between the response of the wild-type fusion and mutant fusion to oleic acid on both carbon sources. The other notable difference between the fusion constructs is their basal activity levels on either carbon source. If FRM2-IacZ on dextrose is taken to be baseline activity, then frm2-lacZ activity on dextrose is two-fold higher and two-fold lower on galactose. The FRM2-IacZ fusion's basal activity does not change between the two carbon sources.

Further experiments are underway to determine the significance of the potential UREs located 5' of FRMZ and if the regulation of frm2 differs from that of FRMZ. Additionally, the FRM2 protein is being expressed to analyse its cellular location and function. Each of these experiments should give us more insight into what role the FRM2 protein plays in the lipid signalling pathway.

ACKNOWLEDGEMENTS A special thanks to Dr Judith Berman and members of the Bernlohr laboratory for helpful discussions and encouragement. In addition, we are particularly indebted to Dr P. T. Magee for his excellent advice and counsel and his sharing of resources during this study. We are grateful to Ryan Berger for obtaining the PCR product of the 5'-region of frm2 for construction of pLBG1. We thank Dr R. Daniel Gietz for the YIplac128 vector and Dr Dennis M. Kinney for the YEp356 and YEp357 vectors. This research was funded by NIH Grant-5ROI GM43199.

REFERENCES Ausubel, F. M., Brent, R., Kingston, R. E., Moore,

D. D., Seidman, J. G., Smith, J. A. and Struhl, K. (1987). Current Protocols in Molecular Biology. John Wiley and Sons, New York.

Bennetzen, J. L. and Hall, B. D. (1982). Codon selection in yeast. J. Biol. Chem. 257 (6), 3026-3031.

Boeke, J. D., Trueheart, J, Natsolulis, G. and Fink, G. R. (1987). 5-fluoroorotic acid as a selective agent in yeast molecular genetics. Meth. Enz. 154, 164-175.

Bossie, M. A. and Martin, C. E. (1989). Nutritional regulation of yeast delta-9 fatty acid desaturase activity. J. Bact. 171 (12), 6409-6413.

Bradford, M. M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254.

Brosius, J. (1989). Superpolylinkers in cloning and expression vectors. DNA 8 (lo), 759-777.

Chen, C., Kacherovsky, N., Dombek, K. M., Camier, S., Thukral, S. K., Rhim, E. and Young, E. T. (1994). Identification of potential target sequences for Adrlp through characterization of essential nucleotides in UAS1. Mol. Cell. Bio. 14 (6), 3842-3852.

Cigan, A. M. and Donahue, T. F. (1987). Sequence and structural features associated with translational initiator regions in yeast-a review. Gene 59, 1-18.

Einerhand, A. W., Kos, W. T., Distel, D. and Tabak, H. F. (1993). Characterization of a transcriptional control element involved in proliferation of peroxisomes in yeast in response to oleate. Eur. J. Biochem. 214 (I), 323-331.

Einerhand, A. W. C., Van der Leij, I., Kos, W. T., Distel, B. and Tabak, H. F. (1992). Transcriptional regulation of genes encoding proteins involved in biogenesis of peroxisomes in Saccharomyces cerevisiae. Cell Biochem. and Function 10, 185-191.

Einerhand, A. W. C., Voorn-Brouwer, T. M., Erdmann, R., Kunau, W. and Tabak, H. F. (1991). Regulation of transcriptim of the gene coding for peroxisomal 3-oxoacyl-CoA thiolase of Saccharomyces cerevisiae. Eur. J. Biochem. 200, 113-122.

Feinberg, A. P. and Vogelstein, B. (1983). A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132, 6-13.

Filipits, M., Simon, M. M., Rapatz, W., Hamilton, B. and Ruis, H. (1993). A Saccharomyces cerevisiae upstream activating sequence mediates induction of peroxisome proliferation by fatty acids. Gene 132, 49-55.

Gancedo, J. M. (1992). Carbon catabolite repression in yeast. Eur. J. Biochem. 206, 297-313.

Gietz, R. D. and Sugino, A. (1988). New yeast- Escherichia coli shuttle vectors constructed with in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74, 527-534.

Hamilton, R., Watanabe, C. K. and de Boer, H. A. (1987). Compilation and comparison of the sequence context around the AUG start codon in Saccharo- myces cerevisiae mRNAs. Nucl. Acids Res. 15,

Ito, H., Fukuda, Y., Murata, K. and Kimura, A. (1983). Transformation of intact yeast cells treated with alkali cations. J. Bact. 153 (l), 163-168.

Kamiryo, T., Parthasarathy, S. and Numa, S. (1976). Evidence that acyl coenzyme A synthetase activity is required for repression of yeast acetyl coenzyme A carboxylase by exogenous fatty acids. Proc. Natl. Acad. Sci. USA 73 (2), 386-390.

Kohlwein, S. D. and Paltauf, F. (1983). Uptake of fatty acids by the yeasts, S. uvarum and Saccharomyces lipolytica. Biochim. Biophys. Acta 792, 310-317.

Kunau, W. H., et al. (1993). Two complementary approaches to study peroxisome biogenesis in

3581-3593.

FATTY ACID REGULATION 33 1

controls osmotic regulation of transcription via the stress response elements (STRE) of Saccharomyces cerevisiae CTTl gene. EMBO J. 13 (18), 4382-5489.

Sherman, F., Fink, G. R. and Hicks, J. B. (1986). Methods in Yeast Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Sikorski, R. S. and Heiter, P. (1989). A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122, 19-27.

Simon, M., Adam, G., Rapatz, W., Spevak, W. and Ruis, H. (1991). The Saccharomyces cerevisiae ADRl gene is a positive regulator of transcription of genes encoding peroxisomal proteins. Mol. Cell. Bid. 11 (2), 699-704.

Sloots, J. A., Aitchison, J. D. and Rachubinski, R. A. (199 1). Glucose-responsive and oleic acid-responsive elements in the gene encoding the peroxisomal tri- functional enzyme of Candida tropicalis. Gene 105, 129-134.

Stukey, J. E., McDonough, V. M. and Martin, C. E. (1990). The OLEl gene of Saccharomyces cerevisiae encodes the A 4 fatty acid desaturase and can be functionally replaced by the rat stearoyl-CoA desaturase gene. J. Biol. Chem. 265, 2014420149.

Sumida, C., Graber, R. and Nunez, E. (1993). Role of fatty acids in signal transduction: modulators and messengers. Prostaglandins, Leukotrienes and Essential Fatty Acids 48, 117--122.

Tanaka, S. and Isono, K. (1993). Correlation between observed transcripts and sequenced ORFs of chromosome 1 [I of Saccharomyces cerevisiae. Nucl. Acids Res. 21 (5), 1149-1153.

Wang, T. W., Lewin, A. S. and Small, G. M. (1992). A negative regulating element controlling transcription of the gene encoding acyl-CoA oxidase of Saccharo- myces cerevisiae. Nucl. Acids Res. 20 (1 3), 3495-3500.

Yoshikawa, A. and Isono, K. (1990). Chromosome I11 of Saccharomyces cerevisiae: an ordered clone bank, a detailed restriction map and analysis of transcripts suggest the presence of 160 genes. Yeast 6, 383401.

Saccharomyces cerevisiae; Forward and reverse genetics. Biochemie 75, 209-224.

Langford, C. J., Klinz, F.-J., Donath, C. and Gallwitz, D. (1984). Point mutations identify the conserved intron-contained TACTAAC box as an essential splicing signal sequence in yeast. Cell 36, 645-653.

Lawrence, C. W. (1991 ). Classical mutagenesis tech- niques. Methods Enzymol. 194, 273-281.

McDonough, V. M., Stukey, J. E. and Martin, C. E. (1992). Specificity of unsaturated fatty acid-regulated expression of the Saccharomyces cerevisiae OLEl gene. J. Biol. Chem. 267 (9), 5931-5936.

Meyers, A. M., Tzagoloff, A,, Kinney, D. M. and Lusty, C. J. (1986). Yeast shuttle and integrative vectors with multiple cloning sites suitable for construction of lac2 fusions. Gene 45, 299-3 10.

Oliver, S. G., et al. (1992). The complete sequence of yeast chromosome 111. Nature 357, 3846.

Pearson, W. R. and Lipman, D. J. (1988). Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA 85, 2444-2448.

Rose, M. D., Novick, P., Thomas, J. H., Botstein, D. and Fink, G. R. (1987). A Saccharomyces cerevisiae genomic plasmid bank based on a centromere- containing shuttle vector. Gene 60, 237-243.

Rose, T. M., Schultz, E. R. and Todaro, G. J. (1992). Molecular cloning of the gene for the yeast homolog (ACB) of diazepam binding inhibitor/endozepine/ acyl-CoA-binding protein. Proc. Natl. Acad. Sci. USA 89, 11287-11291.

Sambrook, J., Fritsch, E. F. and Maniatis, T. (1989). Molecular Cloning. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Sanger, F., Nicken, S. and Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467.

Sarokin, L. and Carlson, M. (1985). Comparison of two yeast invertase genes: conservation of the upstream regulatory region. Nucl. Acids Res. 13 (17), 6089- 6103.

Schuller, C., Brewster, J. C., Alexander, M. R., Gustin, M. C. and Ruis, H. (1994). The HOG pathway

Documents

Identification of a class ofSaccharomyces cerevisiae mutants defective in fatty acid repression of gene transcription and analysis of thefrm2 Gene