Three Genes Are Required for truns - Activation of Ty Transcription in Yeast Fred Winston , ’ ”

Mutations in the SPT3 gene were isolated as one class of suppressors of T y and solo 6 insertion mutations in Saccharomyces cerevisiae. Previous work has shown that null mutations in SPT3 abolish the normal T y 6-6 transcript; instead, a transcript that initiates 800 bases farther downstream is made, suggesting that SPT3 is required for transcription initiation in 6 sequences. We have selected for new spt mutations and have screened for those with the unique suppression pattern of spt3 mutations with respect to two insertion mutations. Our selection and screen has identified two additional genes, SPT7 and SPT8, that are also required for transcription initiation in 6 sequences. We show that mutations in SPT7 or SPT8 result in the same alteration of Ty transcription as do mutations in SPT3. In addition, mutations in all three genes cause a sporulation defect. By assay of a Ty-lac2 fusion we have shown that spt3, spt7 and spt8 mutations reduce transcription from a 6 sequence by 10-25-fold. Finally, we show that SPT3 mRNA levels are unaffected in either spt7 or spt8 mutants, suggesting that these two genes do not regulate transcription .of SPT3. HE T y elements of the yeast Saccharomyces cereT visiae are a dispersed set of repetitive transposable genetic elements. They are 5.9 kb long, flanked by direct terminal repeats called 6 sequences (CAMERON, LOH and DAVIS 1979). T y elements are a member of a group of eukaryotic transposable elements termed retrotransposons (BOEKE et al. 1985). This group also includes copia-like elements of Drosophila and retroviral proviruses of mammals. T y elements share several structural characteristics with these other elements. They contain two open reading frames, analogous to the gag and pol open reading frames of retroviruses (CLARE and FARABAUCH 1985; WARMINCTON et al. 1985; HAUBER, NELBOCK-HOCHSTETTER and FELDMANN 1985); T y transposition occurs via an RNA intermediate (BOEKE et al. 1985); and T y elements encode a reverse transcriptase (GARFINKEL, BOEKE and FINK 1985; MELLOR et al. 1985). Insertion of T y elements or their solo 6 derivatives in the 5‘ noncoding region of genes can inhibit or otherwise alter adjacent gene expression [for a review, see ROEDER and FINK (1983)J. T y transcription initiates in the 5’ 6 (LTR) and proceeds across the element, terminating in the 3’ 6 (ELDER, LOH and DAVIS 1983). T y transcription and the effect of T y insertion mutations on expression of adjacent genes can be affected by the state of several different yeast genes, including those that affect mating (MAT and some STE genes; ERREDE et al. 1980) and several genes designated SPT, To whom correspondence should be addressed. * Present address: Department of Biology, MIT, Cambridge, Massachusetts 02139. TYE and ROC (WINSTON et al. 1984; CIRIACY and WILLIAMSON 198 1 ; DUBOIS, JACOBS and JAUNIAUX 1982). We have previously shown that the yeast SPT3 gene product is necessary for transcription initiation in 6 sequences of T y elements (WINSTON, DURBIN and FINK 1984). In the absence of SPT3, T y transcription initiates 800 bp farther downstream, in the internal region of T y elements. A similar transcriptional effect on solo 6 insertions in the 5’ noncoding region of genes also has been observed in spt3 mutants. In wildtype strains, transcription initiates in the solo 6; in spt3 mutants transcription initiates farther downstream. The virtual abolition of normal length T y transcripts in an spt3 mutant results in elimination of T y transposition, since T y RNA is an essential intermediate in the transposition process (BOEKE, STYLES and FINK 1986). Mutations in SPT3 cause several other striking mutant phenotypes, including suppression of T y and 6 insertion mutations and defects in mating and sporulation (WINSTON et al. 1984; WINSTON, DURBIN and FINK 1984). This variety of phenotypes suggests that SPT? is important for normal transcription of cellular sequences in addition to T y elements or that T y transcription is required for these functions. In this paper, we describe the identification of two additional yeast genes, SPT7 and SPT8, that are also required for transcription from 6 sequences. Our results show that spt7 and spt8 mutants display many of the same phenotypes as spt3 mutants. These include suppression of insertion mutations and a sporulation Genetics 1 1 5 649-656 (April, 1987) 650 F. Winston et al. defect. Furthermore, we show that SPT7 and SPT8 do not regulate transcription of SPT3. These results indicate that these three genes act together in regulation of T y transcription. MATERIALS AND METHODS Yeast strains: The designations for all yeast strains are standard (SHERMAN 1981). The yeast strains used in this study are listed in Table 1. Genotypes listed in brackets indicate integrated plasmids. General genetic methods: Standard yeast genetic procedures of crossing, sporulation and tetrad analysis were followed as described by MORTIMER and HAWTHORNE (1969) and SHERMAN, FINK and LAWRENCE (1978). For measurement of sporulation frequency, cultures were sporulated for 1 day at 23 O and 2 days at 30 O on solid sporulation medium. Sporulated cultures were examined by light microscopy in a hemacytometer. Sporulation frequency is the number of tetrads divided by the sum of tetrads and unsporulated cells. Media: All media were made as described by SHERMAN, FINK and LAWRENCE (1978). These include rich media (YPD), minimal media (SD) and sporulation media. SD complete refers to SD media containing the amino acid requirements of the particular strain being grown. SC-his and SC-lys are complete synthetic media (SHERMAN, FINK and LAWRENCE 1978) lacking histidine and lysine, respectively. Solid media contained 2% agar. Isolation of mutants: All of the mutations described in this paper were isolated in strains FW667 and L37 (Table 1). Both of these strains have a HisLys+ phenotype due to the insertions his4-9176 and lys2-173R2. The spt mutants were isolated essentially as described previously by making patches from single colonies on YPD plates and then replica plating them to SC-his plates (WINSTON et al. 1984). For mutants isolated from strain FW667, we picked a single His+ colony from each patch, purified it on SC-his plates and then scored the His and Lys phenotypes. For mutants isolated from strain L37, we replica plated the SC-his selection plates directly onto SC-lys and SC-his plates. Then, we picked one His+ Lyscandidate from each patch for colony purification and further analysis. Complementation and dominance analysis: Analysis of His+ mutants to determine if they contained spt3 mutations was done by a replica plating test. For mutants isolated from strain FW667, we replica plated patches of the candidates to YPD plates. We then replica plated confluent lawns of strains L6 and L37 to the same YPD plates and incubated the plates overnight. The following day, the YPD plates were replica plated to SD plates containing histidine and lysine to score for diploid formation and to SD plates containing only lysine to score for dominance (in the tests with strain L37) or complementation of the spt3 mutation (in the tests with strain L6). For mutants isolated from strains L37 we followed the same procedure, using strains FW508 and FW667 as the testers for complementation and dominance, respectively. Complementation tests between the new mutants were initially done by cross replica plating stripes of the different candidates as previously described (WINSTON et al. 1984). We then constructed heterozygous diploids using representative alleles from each group and the ability of each purified diploid to grow on plates without histidine was determined. Northern hybridization analysis: Analysis of transcription by Northern hybridization experiments was done as previously described (WINSTON, DURBIN and FINK 1984). In some cases, instead of baking the filter at 80' for 2 hr, RNA was cross-linked to Genescreen (New England Nuclear) by irradiation of the filter for 2 min with 1200 pW/cm' of UV light (CHURCH and GILBERT 1984). T y mRNA was measured with the probe B 16 l , an internal T y BglII restriction fragment in pBR322 (R. SUROSKY, B. TYE and G. R. FINK, unpublished data). SPT3 mRNA was detected with the probe pFW42, an EcoRI-XhoI SPT3 fragment in pBR322. This probe also faintly detects a second RNA slightly larger than and unrelated to SPT3 mRNA (WINSTON and MINEHART 1986). The intensities of bands on the autoradiograms were normalized to those for pyruvate kinase mRNA (BURKE, TEKAMP-OLSON and NAJARIAN 1983), detected with the plasmid pFR2, provided by Dr. P. SINHA. SI nuciease protection analysis: S 1 protection experiments were done as previously described (WINSTON, DURBIN and FINK 1984). The probe was B 163, a Ty BglII restriction fragment in pBR322 (R. SUROSKY, B. TYE and G. R. FINK, unpublished data). Construction of Ty-lac2 fusions: To analyze a Ty-lacZ fusion, we constructed strains that contain a single copy of the Ty9 12A44-ZacZ fusion integrated into the yeast genome at the URA3 locus. To construct such an integrant, we began with plasmid pJCA44, a 2-pm circle containing plasmid that contains 388 bp of Ty912 fused to the Escherichia coli lac2 gene, encoding a hybrid tya-@-galactosidase protein (J. CLARE and P. FARABAUCH, unpublished data). From this, we constructed plasmid pFW82, which lacks the yeast 2-pm circle EcoRI fragment. To integrate a single copy of pFW82 into the genome, it was digested with the restriction enzyme StuI, which cuts the plasmid once, in the URA3 gene. Such a linearized plasmid, upon transformation into yeast, should integrate at the URA3 locus (ORR-WEAVER, SZOSTAK, and ROTHSTEIN 198 1). We transformed the linearized plasmid into strain FW1094 (Table 1 ) and selected Ura+ transformants. Transformants were confirmed to contain a single copy of the plasmid integrated at URA3 by two tests. First, we crossed the transformants by a URA3 strain and observed 4:O segregation for Ura+:Urain all tetrads. Second, to ensure that only a single copy of the plasmid had integrated, we did Southern hybridization analysis of BamHI digested DNA prepared from the transformants. By these criteria, all transformants contained a single copy of the plasmid integrated at URA3. One transformant of strain FW1094, strain FWll54, was used in crosses to generate all strains subsequently used for analysis of the Ty-lac2 fusion. All progeny of these crosses used in the @-galactosidase assays were rechecked by Southern hybridization analysis to verify that they contained a single copy of the Ty-lacZ fusion. @-Galactosidase Assays: P-Galactosidase assays were performed as described by ROSE, CASADABAN and BOTSTEIN (198 1). &Galactosidase levels were normalized to total cellular protein (BRADFORD 1976). To grow cells for preparation of extracts, overnight cultures were first grown in SD complete media. These cultures were used to inoculate fresh SD complete media and cultures were grown to approximately 2 X IO7 cells/ml. For wild type and for each spt mutant, we assayed four different strains. All assays were repeated at least three times.