Analysis of new SPT16 mutants. (A) Suppression of the hht2-11 phenotypes by the SPT16 mutations. The strains used were as follows: wild type (yAAD1052), hht2-11 (yAAD1048), hht2-11 SPT16-790 (yAAD1063), and hht2-11 SPT16-857 (yAAD1061). Note that each of the strains also harbors a deletion of the HHT1-HHF1 locus such that the HHT2 allele provides the sole source of histone H3 (see Table 1). Approximately 10⁵ cells were spotted on YPD-based media and ∼10⁴ cells on SD-based media. The plates were incubated at either 30° or 14° (where indicated) for the following times: YPD 30°, 2 days; YPD 14°, 14 days; YPD + formamide (form), 4 days; SD, 2 days; and SD with no inositol (SD-ino), 3 days. (B) Nature and location of the amino acid changes present in the mutant Spt16 proteins. Shown is an alignment of several amino acids within a region spanning residues 785–862 of the S. cerevisiae Spt16 protein with those present within the corresponding region of Spt16 homologs from several other organisms. Each of the Spt16 mutants described in this study harbors a single amino acid substitution (indicated in boldface type). In both cases, the change is from a highly conserved acidic residue to a nonacidic amino acid (see the boxed residues). Note that the two mutations are relatively near each other, possibly defining a domain involved in functional interactions with histone H3.

Molecular and genetic evidence that H3-L61W affects transcription elongation. (A) Northern analysis of the FLO8 RNA transcripts in H3 WT SPT16 (yAAD1052), H3 WT SPT16-857 (yAAD1121), H3 WT SPT16-790 (yAAD1134), H3-L61W SPT16 (yAAD1049), H3-L61W SPT16-857 (yAAD1061), and H3-L61W SPT16-790 (yAAD1128) strains. In the top section, the top arrow indicates the location of the full-length transcript and the bottom arrow indicates the location of the short transcript. The intensities of the rRNA bands (bottom section) reflect the amount of total RNA loaded per sample. The data shown are representative of two independent experiments for all of the strains tested except for the H3-L61W SPT16 and H3-L61W SPT16-857 strains, which have been tested in three independent experiments, and the H3 WT SPT16-790 strain, which has been tested in a single experiment. The samples shown were analyzed on the same gel; the order of the lanes has been rearranged to clarify the presentation. (B) H3-L61W cells (strain yAAD1049) were spotted in a 10-fold dilution series from 2.8 × 10⁸ cells/ml to 2.8 × 10⁵ cells/ml (4 μl per spot) onto either synthetic complete (SC) medium (top) or SC medium containing 20 μg/ml MPA (bottom) and incubated at 14° for 28 days. (C) Approximately 1.8 × 10⁶ cells with the indicated genotypes were spotted on either SC media lacking uracil (SC −Ura) or 5-FOA media (SC +FOA) and incubated for 3 days at 30° (the hht2-11 allele encodes the H3-L61W mutant). Strains whose genotypes include pHHT1 contain a URA3-marked plasmid carrying a wild-type copy of the HHT1 gene to cover for the hht2-11 mutation, whereas the strain whose genotype includes pSPT16 (second row) contain a URA3-marked plasmid expressing wild-type Spt16 to cover for the spt16-197 mutation. Before spotting, the cells were grown on nonselective media (YPD) to allow strains containing nonlethal genomic mutations to lose the plasmids and propagate. Inability to grow on 5-FOA media indicates synthetic lethal mutant combinations. Refer to Table 1 for the full genotypes of the strains used in these experiments (yAAD2006, yAAD2000, yAAD2005, yAAD2013, yAAD2015, and yAAD994).

Analyses of the association of Spt16 and RNA Pol II across the PMA1 gene in the context of wild-type or mutant versions of histone H3 and Spt16. (A) H3-L61W causes Spt16 to accumulate at the 3′-UTR of PMA1, and the two Spt16 mutants isolated in this study alleviate this defect. ChIP experiments were performed on H3 WT (yAAD1052 and yAAD1053), H3-L61W (yAAD1048 and yAAD1049), H3-L61W Spt16-E857Q (yAAD1060 and yAAD1061), and H3-L61W Spt16-E790K (yAAD1127 and yAAD1128) using an Spt16-specific polyclonal antibody (see materials and methods). Nine primer sets were used to assess Spt16 binding across the PMA1 ORF (shaded area) as well as within the 5′- and 3′-UTRs in the different genetic backgrounds. The percentage of immunoprecipitation values (%IP) indicated in the bar graphs for each of the strains at each region across the PMA1 locus are relative to the corresponding input material. Each %IP value represents the average with standard error from three independent experiments. As a reference point, the average levels of Spt16 binding to an untranscribed region (see materials and methods) were as follows: H3 WT, 0.22% ± 0.006; H3-L61W, 0.45% ± 0.009; H3-L61W Spt16-E857Q, 0.47% ± 0.007; and H3-L61W Spt16-E790K, 0.38% ± 0.008. (B) H3-L61W does not appear to greatly affect the distribution of RNA Pol II across the PMA1 locus. ChIP experiments were performed on H3 WT (yAAD1052) and H3-L61W (yAAD1049) strains using a monoclonal antibody specific for the Rpb3 subunit of RNA Pol II (see materials and methods). The two regions assayed are indicated above the PMA1 locus diagram. The %IP values are relative to the input material and in each case represent the average of two independent experiments. As a reference point, the average levels of Rpb3 binding to the untranscribed region were as follows: H3 WT, 0.09% ± 0.01; and H3-L61W, 0.10% ± 0.01. For both sets of experiments, the analysis and quantitation of the signals were carried out using the method described by Martens and Winston (2002) (see materials and methods).

H3-L61W affects Spt16 binding across three other transcribed genes and the Spt16 mutants partially suppress these defects. ChIP experiments were performed as described in Figure 3A, except that the genomic regions tested correspond to the 5′ ORF regions and 3′-UTRs across the (A) ADH1 locus, (B) FBA1 locus, and (C) PDC1 locus. Each %IP value represents the average with standard error from three independent experiments.

The accumulation of Spt16 at the 3′-UTRs of the inducible genes GAL1 and GAL2 is dependent upon transcription. Spt16 ChIP experiments were carried out on H3 WT (yAAD 1053) and H3-L61W (yAAD1048) strains grown in the presence of either glucose (repressive condition) or galactose (activating condition). Two primer sets per gene were used to assess Spt16 binding to the 5′ ORF and to the 3′-UTR regions of the GAL1, GAL2, and PMA1 genes. The analysis and quantitation of the signals were performed using real-time PCR technology (see materials and methods). The %IP values are relative to the corresponding input material and represent the average with standard error from three independent clones. Note that accumulation of Spt16 in the 3′-UTR of the constitutively expressed gene PMA1 (bottom) is independent of the carbon source. As a reference point, in these experiments the average levels of Spt16 binding to the control untranscribed region were as follows: H3 WT (glucose), 0.49% ± 0.09; H3 WT (galactose), 0.50% ± 0.09; H3-L61W (glucose), 1.21% ± 0.13; and H3-L61W (galactose) 1.15% ± 0.11.

H3-L61W does not appear to greatly affect the distribution of nucleosomes across the PMA1 gene. ChIP experiments were performed as described in Figure 3A, except that the immunoprecipitations were performed with a histone-H3-specific antibody (see materials and methods). Each %IP value represents the average with standard error from three independent experiments. As a reference point, the average levels of histone H3 binding to the control untranscribed region were as follows: H3 WT, 2.4% ± 0.32; H3-L61W, 4.10% ± 0.57. As anticipated, the %IP of histone H3 at the untranscribed region was relatively high since nucleosomes are present throughout the genome.