Evidence that Spt4, Spt5, and Spt6 control transcription elongation by RNA polymerase II in Saccharomyces cerevisiae

Abstract

Previous characterization of the Saccharomyces cerevisiae Spt4, Spt5, and Spt6 proteins suggested that these proteins act as transcription factors that modify chromatin structure. In this work, we report new genetic and biochemical studies of Spt4, Spt5, and Spt6 that reveal a role for these factors in transcription elongation. We have isolated conditional mutations in SPT5 that can be suppressed in an allele-specific manner by mutations in the two largest subunits of RNA polymerase II (Pol II). Strikingly, one of these RNA Pol II mutants is defective for transcription elongation and the others cause phenotypes consistent with an elongation defect. In addition, we show that spt4, spt5, and spt6 mutants themselves have phenotypes suggesting defects in transcription elongation in vivo. Consistent with these findings, we show that Spt5 is physically associated with RNA Pol II in vivo, and have identified a region of sequence similarity between Spt5 and NusG, an Escherichia coli transcription elongation factor that binds directly to RNA polymerase. Finally, we show that Spt4 and Spt5 are tightly associated in a complex that does not contain Spt6. These results, taken together with the biochemical identification of a human Spt4-Spt5 complex as a transcription elongation factor (Wada et al. 1998), provide strong evidence that these factors are important for transcription elongation in vivo.

Figure 1

Northen blot analysis of SPT5⁺ and spt5 Cs⁻ strains. SPT5⁺, spt5-242, and spt5-276 strains were grown at 30°C or 13°C for 48 hr prior to isolation of RNA for Northern analysis. The blots were probed with HIS4 or TPI1 as indicated, and also with TUB2 as a normalization control. Note that the order of the two mutants is different in the two experiments shown. The blot on the left contains RNA from strains FY2, FY1672, and FY1673. The blot on the right contains RNA from strains FY120, FY1634, and FY1635.

Figure 2

Model of the RNA Pol II transcription elongation cycle (for review, see Uptain et al. 1997).

Figure 3

spt5-242 ppr2Δ double mutants. The indicated strains derived from a genetic cross of FY1645 × FY1670. Cells were grown at 37°C on a YPD plate and then replica plated to YPD plates that were incubated at 30°C or 37°C for 2 days.

Figure 4

The cold-sensitive growth defect of the spt5 Cs⁻ mutants is suppressed by decreasing the rate of transcription elongation. (A) Suppression of spt5-242 by rpb2-10, a mutation known to decrease the rate of elongation by RNA Pol II (Powell and Reines 1996). Growth of the indicated strains (FY1649, FY1650, FY1651, FY1648, FY1652, and FY1653) on SC − Ura plates at 30°C and 15°C was assayed. The photograph of the 30°C plate was taken after 2 days; the photograph of the 15°C plate was taken after 8 days. The rpb2-10 mutation also suppressed the other spt5 Cs⁻ mutation, spt5-276 (data not shown). (B) Suppression of spt5-276 and spt5-242 by 6AU. Growth of the indicated strains (FY267, FY1647, and FY1645) on SC − Ura plates at 15°C was assayed in the presence or absence of 6AU. The plates were photographed after 8 days of incubation.

Figure 5

A ppr2Δ mutation causes conditional lethality in combination with spt4, spt5, and spt6 mutations. Strains of the indicated genotypes were grown as patches on YPD plates at 30°C and were then replica plated to duplicate YPD plates. These plates were incubated at 30°C or 37°C for 2 days and then photographed. The strains were derived from single tetrads obtained by crossing an spt strain with a ppr2Δ strain as indicated. (A) spt4Δ ppr2Δ (FY1644 × FY1646); (B) spt5-194 ppr2Δ (FY1585 × FY366); (C) spt6-14 ppr2Δ (FY1644 × FY1655).

Figure 6

Identification of an Spt4–Spt5 complex. (A) Strains expressing the indicated GST fusion protein were used to prepare whole cell extracts. The GST fusion and associated proteins were isolated by affinity chromatgraphy, separated by SDS-PAGE, and immunoblotted, as described in Materials and Methods. Shown is a blot probed with anti-GST (top) and anti-Spt5 (bottom) antibodies. (B) Spt4 coimmunoprecipitates with Spt5. Extracts were prepared from wild-type (FY1654) and Spt4–HA1-tagged (FY1643) strains and used for immunoprecipitations of Spt5. Western blots of the immunoprecipitates were probed with anti-Spt5 (top), anti-HA1 (middle), or anti-Spt6 (bottom) antibodies. (C) Spt5 coimmunoprecipitates with Spt4. Anti-HA1 immunoprecipitations were carried out by use of extracts of wild-type (FY120) or HA1–Spt4-tagged (FY1643) strains. Western blots of the immunoprecipitates were probed with anti-HA1 (top) and anti-Spt5 (bottom) antibodies.

Figure 6

Identification of an Spt4–Spt5 complex. (A) Strains expressing the indicated GST fusion protein were used to prepare whole cell extracts. The GST fusion and associated proteins were isolated by affinity chromatgraphy, separated by SDS-PAGE, and immunoblotted, as described in Materials and Methods. Shown is a blot probed with anti-GST (top) and anti-Spt5 (bottom) antibodies. (B) Spt4 coimmunoprecipitates with Spt5. Extracts were prepared from wild-type (FY1654) and Spt4–HA1-tagged (FY1643) strains and used for immunoprecipitations of Spt5. Western blots of the immunoprecipitates were probed with anti-Spt5 (top), anti-HA1 (middle), or anti-Spt6 (bottom) antibodies. (C) Spt5 coimmunoprecipitates with Spt4. Anti-HA1 immunoprecipitations were carried out by use of extracts of wild-type (FY120) or HA1–Spt4-tagged (FY1643) strains. Western blots of the immunoprecipitates were probed with anti-HA1 (top) and anti-Spt5 (bottom) antibodies.

Figure 6

Identification of an Spt4–Spt5 complex. (A) Strains expressing the indicated GST fusion protein were used to prepare whole cell extracts. The GST fusion and associated proteins were isolated by affinity chromatgraphy, separated by SDS-PAGE, and immunoblotted, as described in Materials and Methods. Shown is a blot probed with anti-GST (top) and anti-Spt5 (bottom) antibodies. (B) Spt4 coimmunoprecipitates with Spt5. Extracts were prepared from wild-type (FY1654) and Spt4–HA1-tagged (FY1643) strains and used for immunoprecipitations of Spt5. Western blots of the immunoprecipitates were probed with anti-Spt5 (top), anti-HA1 (middle), or anti-Spt6 (bottom) antibodies. (C) Spt5 coimmunoprecipitates with Spt4. Anti-HA1 immunoprecipitations were carried out by use of extracts of wild-type (FY120) or HA1–Spt4-tagged (FY1643) strains. Western blots of the immunoprecipitates were probed with anti-HA1 (top) and anti-Spt5 (bottom) antibodies.

Figure 7

Purification of Spt5–Flag and associated proteins. An extract of a wild-type strain in which Spt5 was tagged with the Flag epitope was mixed with anti-Flag beads, washed, and then bound proteins were competitively eluted with a Flag peptide as described in Materials and Methods. Approximately 1.7% of the material loaded on the anti-Flag beads, unbound material, each of the six washes, and 20% of each of the three eluted fractions were separated on an SDS–polyacrylamide gel and blotted. (Left) This blot was probed with antibodies that recognize Spt5, Spt4, Spt6, and the CTD of Rpb1. (Right) Western blots of a control experiment carried out identically and in parallel in which the extract used contained Spt5 that was not Flag tagged.

Figure 8

A repeated element in Spt5 is homologous to a region of NusG. Alignments of a conserved, repeated element in Spt5 proteins along with the consensus sequence developed by the program MEME. This consensus is similar to a motif of unknown function, KOW (Kyrpides et al. 1996), found in NusG, ribosomal protein L24, and their homologs. For comparison, the homologous sequences of the highest scoring NusG and L24 proteins are included. The accession number for S. cerevisiae Spt5 is M62882; for human, U56402; for Caenorhabditis elegans, Z68316; for Schizosaccharomyces pombe, Z99753; for Tetrahymena thermophila NusG, P35872; and for Mycoplasma capricolum L24, P10141.