Human Spt6 stimulates transcription elongation by RNA polymerase II in vitro

Abstract

Recent studies have suggested that Spt6 participates in the regulation of transcription by RNA polymerase II (RNAPII). However, its underlying mechanism remains largely unknown. One possibility, which is supported by genetic and biochemical studies of Saccharomyces cerevisiae, is that Spt6 affects chromatin structure. Alternatively, Spt6 directly controls transcription by binding to the transcription machinery. In this study, we establish that human Spt6 (hSpt6) is a classic transcription elongation factor that enhances the rate of RNAPII elongation. hSpt6 is capable of stimulating transcription elongation both individually and in concert with DRB sensitivity-inducing factor (DSIF), comprising human Spt5 and human Spt4. We also provide evidence showing that hSpt6 interacts with RNAPII and DSIF in human cells. Thus, in vivo, hSpt6 may regulate multiple steps of mRNA synthesis through its interaction with histones, elongating RNAPII, and possibly other components of the transcription machinery.

FIG. 1.

Immunological analysis of hSpt6. (A) Schematic representation of hSpt6. The acidic, Tex homology, HhH, S1, and STPG/Q-rich domains are indicated. Thin lines show locations of peptides 1 and 2. (B) Immunoblot analysis of HeLa cell extracts using monoclonal antibodies against hSpt6. WCE (7.5 μl; lanes 1) and NE (3 μl; lanes 2) were subjected to immunoblot analysis. Shown at right is the immunoblot analysis performed without the first antibody. (C) Immunoblot analysis of anti-peptide 1-1 and anti-peptide 2-1 immunocomplexes. Aliquots (2.0 μl) of the eluates from the anti-peptide 1-1 and anti-peptide 2-1 immunocomplexes were subjected to immunoblot analysis using anti-peptide 1-1 (hSpt6), anti-CTD (RNAPII; Covance Inc.), and anti-DSIF p160 (hSpt5) (22) antibodies. (D) Immunodepletion of hSpt6 from HeLa NE. NE (200 μl) were treated four times with anti-peptide 2-1 (lanes 2 to 5) or anti-peptide 1-1 (lanes 6 to 9) antibodies coupled to protein G-Sepharose; aliquots (2.0 μl) of the flowthrough fractions were subjected to immunoblot analysis using anti-peptide 1-1 (hSpt6), anti-cyclin T1 (cyclin T1; kindly provided by David Price), anti-CDK9 (CDK9; Santa Cruz Biotechnology, Inc.), anti-CTD (RNAPII), anti-DSIF p160 (hSpt5), anti-NELF-A (NELF-A) (33), and anti-RD (NELF-E) (32) antibodies.

FIG. 2.

hSpt6 is important for transcription on a naked DNA template in vitro. (A) Immunodepletion of hSpt6 causes a transcriptional defect. Transcription assays were carried out using the NE that were immunodepleted using either anti-peptide 1-1 (NE) or anti-peptide 2-1 (dNE) as shown in Fig. 1. The NE (2 μl) were incubated with 125 ng of the supercoiled DNA template pTF3-6C₂AT for 45 min at 30°C. NTPs (final concentrations of 80 μM 3′-OMe-GTP, 60 μM ATP, 600 μM CTP, and 1 μM UTP containing 5 μCi of [α-³²P]UTP) were added to the reaction mixtures, which were then incubated for the times indicated. An arrow indicates the position of 380-nt, full-length transcripts. (B) Add-back of the immunoprecipitates. Reactions were carried out as described for panel A, except that aliquots (2.0 μl) of the eluates from the anti-peptide 2-1 and anti-peptide 1-1 immunoprecipitates were added to the reactions carried out using dNE. The immunoprecipitates and dNE were incubated for 45 min at 30°C with 125 ng of the supercoiled DNA template pTF3-6C₂AT before addition of NTPs. An arrow indicates the position of the full-length transcripts.

FIG. 3.

Preparation of recombinant hSpt6 and its mutants. (A) Schematic representation of wild-type and mutant Spt6 proteins used in this study. Letters A to E above the boxes stand for the conserved domains of hSpt6. Numbers indicate amino acid positions at the N and C termini of each mutant. (B) Silver staining. Purified recombinant hSpt6 proteins (ca. 10 ng) were subjected to SDS-PAGE and visualized by silver staining. (C) Immunoblot analysis using anti-peptide 2-1 antibody. HeLa NE (12 μg; lane 1) and purified recombinant hSpt6 proteins (ca. 10 ng; lanes 2 to 5) were analyzed by immunoblotting.

FIG.4.

hSpt6 is important for transcription elongation in vitro. Transcription assays were carried out as described for Fig. 2B except that three different templates, pTF3-6C₂AT (12.5 ng), pTF3-6C₂AT-100 (22.5 ng), and pTF3-6C₂AT-50 (90 ng) were used; they yielded 380-, 100-, and 50-nt transcripts, respectively. Purified wild-type r-hSpt6 (A and B), mutant 1 (C and D), mutant 2 (E), or mutant 3 (F) was included in the reactions where indicated. Transcription was allowed to proceed for the indicated times; then the reactions were processed as described in Materials and Methods. Arrows indicate the positions of the 380-, 100-, and 50-nt transcripts. In panels B, D, E, and F, the counts of each transcript at the 9-min time point were quantified and normalized against the counts obtained from a reaction carried out using mock-depleted NE (mean ± standard deviation [SD], P = 0.05, n = 4).

FIG. 5.

Possible interaction between hSpt6 and DSIF. (A) Distribution of various transcription factors in phosphocellulose P11 column fractions was analyzed by immunoblotting. The results are summarized schematically at top. (B) Transcription assays were carried out using the concentrated P1.0 fraction as described for Fig. 4. Purified recombinant hSpt6 and DSIF were added where indicated. NTPs (final concentrations of 80 μM 3′-OMe-GTP, 60 μM ATP, 600 μM CTP, and 5 μM UTP containing 10 μCi [α-³²P]UTP) were added to the reactions, and the reactants were then incubated for 9 min. Arrows indicate the positions of the 380-, 100-, and 50-nt transcripts. (C) The counts of each transcript at the 9-min time point were quantified and normalized against counts obtained from a reaction without recombinant proteins (mean ± SD, P = 0.05, n = 3).

FIG. 6.

Evidence that hSpt6 affects already initiated transcription complexes. (A) Transcription assays were carried out as described for Fig. 4. Purified r-hSpt6 (50 fmol) was added during preincubation (lane 4) or 9 min after NTP addition (lanes 5, 7, 9, 11, and 13). The reactions were terminated at various time points as indicated. Arrows indicate the positions of the 380-, 100-, and 50-nt transcripts. (B) The scheme for this experiment. (C) Counts of each transcript at the 9-min time point were quantified and normalized against counts obtained in lane 2 of panel A (mean ± SD, P = 0.05, n = 3).

FIG. 7.

hSpt6 enhances the rate of transcription elongation. (A) An outline of transcription assays using an immobilized DNA template is presented with a partial nucleotide sequence of the template. (B) The +42 complex was isolated and incubated with buffer (lane 2) or 16, 32, or 160 ng (lanes 3 to 5) of r-hSpt6 mutant 1 for 10 min at 30°C; the reactions were chased for 60 s by addition of 30 μM NTPs. (C) The +42 complex was incubated with (lanes 5 to 8) or without (lanes 1 to 4) purified r-hSpt6 mutant 1 for 10 min at 30°C, and the reactions were chased for 20, 60, or 180 s, or not, as indicated. (D) PICs were formed by incubating the DNA template on magnetic beads, GTFs, and RNAPII in the absence (lane 1) or presence of 8, 16, and 160 ng (lanes 2 to 4) of r-hSpt6 mutant 1 for 20 min. Transcription was allowed to proceed to position +24 by incubating the reaction mixtures with NTPs lacking GTP for 8 min at 30°C.

FIG. 8.

Interactions of hSpt6 with DSIF and RNAPII. Clonal cell lines expressing FH-RNAPII (9) (A) and FLAG-DSIF (31) (B) were treated with 1% formaldehyde for 10 min at room temperature. Cell extracts were prepared and incubated with anti-FLAG M2-agarose gel, anti-Myc-agarose gel, and anti-HA-agarose gel, and immunoprecipitated materials were recovered as described in Materials and Methods. Input (1%), unbound (1%), wash (10%), and eluate (20%) materials were subjected to immunoblot analysis using anti-peptide 1-1 (hSpt6), anti-hSpt5, anti-E4TF1-p53 (22), anti-hSpt4 (14), and anti-phosphoserine-2 of the RNAPII CTD (Covance Inc.). We quantified band density and calculated the recovery rate of each protein.