(A) Upc2p and Ecm22p show strong sequence similarity. They are 70% identical in the DNA-binding domains and 76% identical in the C-terminal regions. Both proteins also have glutamine-rich regions. (B) The C-terminal region of Upc2p contained an activation domain, and fusions containing this region could be induced by lovastatin. Plasmids encoding the indicated protein fusions were transformed into the pGAL7::lacZ reporter strain, PJ69-4a. The resulting strains were grown in the presence or absence of 5 μg/ml lovastatin. Strains were grown to mid-log phase and harvested, and β-galactosidase activity was tested as described in Materials and Methods. Values for all β-galactosidase assays in this study represent averages of at least three experiments each. Values are shown as increases (n-fold) over that of the Gal4p DNA-binding domain alone grown under noninducing conditions. β-Galactosidase activity of the Gal4p DNA-binding domain averaged 21.6 Miller units across all experiments. (C) The Ecm22p C-terminal region also contained an activation domain and could be induced by lovastatin. Plasmids encoding the indicated protein fusions were transformed into PJ69-4a and tested as described for panel B.

(A) Levels of Upc2p increased and levels of Ecm22p decreased in response to lovastatin. Whole-cell extracts were prepared from JRY7865 (UPC2-TAP) and JRY7866 (ECM22-TAP) grown in the presence or absence of 30 μg/ml lovastatin. Protein levels were analyzed by SDS-PAGE followed by immunoblotting as described in Materials and Methods. Arrows indicate tagged proteins. Relative protein levels were quantified using the Li-Cor Odyssey imaging system. Numbers indicate changes (n-fold) in induced protein levels relative to uninduced protein levels. (B) In the presence of lovastatin, the levels of Upc2p-TAP increased at ERG3 promoters, whereas the levels of Ecm22p-TAP decreased. Chromatin immunoprecipitations were performed as described in Materials and Methods. Real-time PCR was used to quantitate the levels of ERG3 promoter DNA and ACT1 control DNA. The ratio of ERG3 promoter DNA to ACT1 DNA was calculated for each sample. Bars indicate the averages of three reactions and the increases in the levels of enrichment seen in Upc2p-TAP and Ecm22p-TAP experiments over the untagged control level. (C) The carboxy-terminal region of Upc2p is important for the activation of ERG2 and its induction by lovastatin. Plasmids containing full-length UPC2 (pJR2564) or UPC2 coding only for aa 1 to 582 (pJR2653) were transformed into a upc2Δ ecm22Δ strain (JRY7181) that contained a pERG2::lacZ reporter (pJR2316). Strains were grown in the presence or absence of 5 μg/ml lovastatin, harvested, and assayed for β-galactosidase activity as described in Materials and Methods. (D) In the presence of lovastatin, the levels of a Gal4p/Upc2p fusion increased at the GAL7 promoter. Chromatin immunoprecipitations were performed as described in Materials and Methods. Bands were quantified using the Li-Cor Odyssey imaging system, and the ratio of GAL7 promoter DNA to control DNA (Ty1) was calculated for each PCR. Bars indicate the enrichment of these ratios from the immunoprecipitations over that of the input DNA and are the average values from at least three PCRs. (E) Treatment with lovastatin did not increase the levels of fusion protein. The Gal7::lacZ reporter strain (PJ69-4a) transformed with the indicated plasmids was grown to log phase in the presence or absence of 5 μg/ml lovastatin. Whole-cell extracts were prepared, subjected to SDS-PAGE, and immunoblotted as described in Materials and Methods. The arrow indicates fusion proteins.

(A) The UPC2-1 and ECM22-1 mutations increased uninduced levels of ERG2 expression. UPC2 (pJR2564), UPC2-1 (pJR2602), ECM22 (pJR2338), and ECM22-1 (pJR2644) were each transformed into a upc2Δ ecm22Δ strain (JRY7181) that contained a pERG2::lacZ reporter (pJR2316). Strains were grown in the presence or absence of 30 μg/ml lovastatin, harvested, and assayed for β-galactosidase activity as described in Materials and Methods. (B) The UPC2-1 mutation increased transcriptional activation. Plasmids encoding the indicated protein fusions were tested as described for Fig. . (C) The ECM22-1 mutation also increased transcriptional activation. Plasmids encoding the indicated protein fusions were tested as described for Fig. . (D) The UPC2-1/ECM22-1 mutation did not increase protein levels. Plasmids encoding the indicated protein fusions and transformed into PJ69-4a were grown to mid-log phase. Whole-cell extracts were prepared, subjected to SDS-PAGE, and immunoblotted as described in Materials and Methods. Arrows indicate fusion proteins. (E) The UPC2-1/ECM22-1 mutation did not increase the level of transcription factor at the promoter. Chromatin immunoprecipitations were performed as described in Materials and Methods. Bands were quantified using the Li-Cor Odyssey imaging system, and the ratio of GAL7 promoter DNA to control DNA (Ty1) was calculated for each PCR. Bars indicate the enrichment of these ratios for the immunoprecipitations compared to that of the input DNA. Each bar represents at least three reactions.

Substitution of any amino acid larger than alanine at glycine 888 resulted in increased transcription. UPC2 (pJR2564) and UPC2 with mutations resulting in the indicated amino acid substitutions (pJR2602 to pJR2608) were each transformed into JRY7181 (upc2Δ ecm22Δ) along with a pERG2::lacZ reporter (pJR2316). Strains were grown to mid-log phase and assayed for β-galactosidase activity as described for Fig. .

(A) The ergosterol biosynthetic pathway. The first half of the pathway, shown on the left, leads to the production of farnesyldiphosphate, which, in addition to sterol production, also contributes to the biosynthesis of several nonsterol products, including heme and ubiquinone, whereas the second half of the pathway, shown on the right, is specific to sterol production. Protein products of nonessential genes are underlined. The action of lovastatin is also represented. (B) Deletion of nonessential ERG genes led to various degrees of increased ERG2 expression. A pERG2::lacZ reporter (pJR2316) was transformed into the indicated strains. The strains were grown to mid-log phase and assayed for β-galactosidase activity. (C) Increased ERG2 expression in erg3Δ strains required Upc2p or Ecm22p. A pERG2::lacZ reporter (pJR2316) was transformed into the indicated strains. The strains were then assayed for β-galactosidase activity.

A model for Upc2p and Ecm22p regulation. Sterol depletion and the ECM22-1 mutation caused decreased promoter occupancy by Ecm22p but increased transcriptional activation. In contrast, low sterol levels caused increases in Upc2p protein levels, an increase in the probability of each protein to be at a promoter, and greater potency of each bound Upc2p.