Analysis of Spt7 function in the Saccharomyces cerevisiae SAGA coactivator complex

Abstract

The Saccharomyces cerevisiae SAGA complex is required for the normal transcription of a large number of genes. Complex integrity depends on three core subunits, Spt7, Spt20, and Ada1. We have investigated the role of Spt7 in the assembly and function of SAGA. Our results show that Spt7 is important in controlling the levels of the other core subunits and therefore of SAGA. In addition, partial SAGA complexes containing Spt7 can be assembled in the absence of both Spt20 and Ada1. Through biochemical and genetic analyses of a series of spt7 deletion mutants, we have identified a region of Spt7 required for interaction with the SAGA component Spt8. An adjacent Spt7 domain was found to be required for a processed form of Spt7 that is present in a previously identified altered form of SAGA called SLIK, SAGA(alt), or SALSA. Analysis of an spt7 mutant with greatly reduced levels of SLIK/SAGA(alt)/SALSA suggests a subtle role for this complex in transcription that may be redundant with a subset of SAGA functions.

FIG. 1.

Levels of Spt20 and Ada1 are dependent on Spt7. (A) Western analysis was performed to measure the levels of Spt7, Spt20, and Ada1. Wild-type (strain FY1977), spt7Δ (FY2026), spt20Δ (FY1291), and ada1Δ (FY2027) strains were grown in YEPD medium, and whole-cell extracts were prepared and subjected to Western blot analysis, using antibodies to Spt7, Ada1, and HA (to detect the HA-Spt20 fusion protein). Taf1 levels were measured as a loading control. (B) Northern analysis of mRNA levels in spt7Δ, spt20Δ, and ada1Δ mutants. The same wild-type, spt7Δ, spt20Δ, and ada1Δ strains were grown in YEPD medium, and RNA was isolated and analyzed by Northern analysis. ACT1 mRNA levels were measured as a loading control.

FIG. 2.

Characterization of SAGA purified from wild-type and SAGA mutant strains. TAP-purified complexes from wild-type (strain FY2031), spt3Δ (FY2040), gcn5Δ (FY2034), spt8Δ (FY2037), ada1Δ (FY2035), spt20Δ ada1Δ (FY2041), and spt20Δ (FY2036) strains were prepared, run on SDS-PAGE (5 to 20% acrylamide) gels, and silver stained. The amount of each complex loaded was adjusted to have an equal level of Spt7 in each lane. All samples were run on one gel.

FIG. 3.

Western analysis of TAP-purified mutant SAGA complexes. TAP-purified complexes from the same extracts as in Fig. 2 were run on SDS-PAGE gels and Western blotted using the following antibodies: HA (12CA5), Ada1, Spt20, Taf6, Taf10, Taf12, Spt3, Spt8, Ada2, and Gcn5.

FIG. 4.

Spt7 deletion analysis. (A) Diagram of Spt7 deletion mutants described in this study. Deletion mutations were constructed as described in Materials and Methods. (B) Phenotypic and biochemical analyses of spt7 deletion mutants. The strains used are as follows: wild-type (strain FY2031), spt7Δ (FY1093), spt7-873 (FY2042), spt7-1125 (FY2032), spt7-1180 (FY2033), spt7-300 (FY2044), and spt7-200 (FY2043). All strains contain the his4-917δ and lys2-173R2 alleles for determining the Spt phenotype (see Materials and Methods). An Spt⁻ phenotype corresponds to growth on media lacking histidine and no growth on media lacking lysine. Cells were grown to saturation overnight in YEPD medium and spotted in a dilution series from 10⁸ to 10⁴ cells/ml, and the plates were incubated at 30°C. The plates shown were incubated for the following times: YEPD, 2 days; SC-Lys, 3 days; SC-His, 4 days; YPgal, 3 days; SC, 2 days; and SC-Ino, 2 days.

FIG. 5.

Silver stain of SAGA purified from wild-type and spt7 mutant strains. TAP-purified SAGA complex from strains analyzed for phenotypes in Fig. 4B were visualized on an SDS-PAGE (5 to 20% acrylamide) gel by silver stain. The amount of each complex loaded was adjusted to have an equal level of Spt7 in each lane. All samples were run on one gel. The additional faint bands present in lanes 4 and 5 were not observed reproducibly and are not part of SAGA.

FIG. 6.

Analysis of intragenic complementation between spt7-100 and spt7-873. (A) Comparison of the phenotypes of wild-type (strain FY632), spt7Δ/spt7Δ (FY2065) spt7-100/spt7-100 (FY2063), spt7-873/spt7-873 (FY2064), and spt7-873/spt7-100 (FY2062) diploids. Cells were grown to saturation overnight in YEPD medium and spotted in a dilution series from 10⁸ to 10⁴ cells/ml, and the plates were incubated at 30°C for 2 days. (B) Western analysis of TAP-purified complexes from wild-type (FY2031), spt7-100 (FY2029), and spt7-873 (FY2042) haploids as well as spt7-873/spt7-100 (FY2062) heterozygous diploids. Purified complexes were run on SDS-PAGE (8% acrylamide) gels and probed using the HA (12CA5) antibody.

FIG. 7.

Spt7 in SLIK/SALSA lacks part of the carboxy terminus. Mono Q fractionation of yeast whole-cell extract from wild-type (strain FY2025), spt7-200 (FY2045), and spt7-300 (FY2046) strains expressing HA- and Myc-tagged Spt7. Western analyses of Mono Q fractions were probed with antibodies to HA or Myc (to detect Spt7) as well as antibodies to Ada1. The diagram depicts the doubly-tagged Spt7 protein; Spt7-200 lacks the region in black, and Spt7-300 lacks the region in gray. The faster-migrating band in the panel showing the anti-HA antibody-probed blot of Mono Q fractions from a wild-type strain represents the carboxy-terminally processed form of Spt7. The presence of Ada1 in the fractions containing SLIK/SALSA in the spt7-200 panel is likely due to the loss of Spt8 from SAGA upon fractionation.

FIG. 8.

Analysis of yeast whole-cell extract from untagged and tagged wild-type (strain FY2025), spt7-200 (FY2045), spt7-300 (Y2046), and spt8Δ (FY2038) strains on Western blots probed with HA (12CA5) or Myc (A14) antibodies. Equal amounts of protein were loaded in each lane.

FIG. 9.

Northern analysis of spt7 internal deletion mutants. Wild-type (strain FY3), gcn4Δ (FY971), spt7Δ (FY2028), spt7-200 (FY2047), spt7-300 (FY2048), spt7-400 (FY2049), and spt8Δ (FY2039) mutants were grown in SC-His to approximately 1 × 10⁷ to 2 × 10⁷ cells/ml and induced with 40 mM 3-aminotriazole for 2 h. RNA was then made from uninduced and induced samples. Northern blots were probed separately with HIS3, TRP3, and ACT1 DNA probes. Shown is a representative Northern blot. The bar graphs below the Northern blot show the average and standard error measured for each mRNA level, with the wild-type, uninduced level equal to 1.0. Each value was determined three to six times.