Fusions with histone H3 result in highly specific alteration of gene expression

Abstract

Hap1 is a yeast transcriptional activator which controls expression of genes such as CYC1 and CYC7. Our results show that Hap1 activity is dependent on a functional chromatin remodeling complex SWI/SNF. Using a modified two-hybrid screen with Hap1 as bait, we recovered expression vectors encoding the Gal4 activation domain fused to histone H3 [Gal4(AD)-H3]. Hap1 activity at CYC1 or CYC7 was increased by Gal4(AD)-H3 and the effect was dependent on the presence of the activation domain of Hap1 and a functional SWI complex. Importantly, overexpression of H3 alone had no effect on Hap1 activity. Analysis of Gal4(AD)-H3 revealed that the fusion is not incorporated into the nucleosome while a functional Gal4 activation domain is dispensable. Activity of many other transcriptional activators was unchanged or slightly affected in the presence of Gal4(AD)-H3. Thus, our results identify a new class of histone H3 variants that cause highly specific alteration of gene expression. Hap1 may interact directly with H3 favoring chromatin remodeling by the SWI/SNF complex.

Figure 1

Hap1 activity is dependent on the SWI complex. (a) Reporters for CYC1 (pUAS1LacZLEU2) or CYC7 (pCYC7-5lacZLEU2) were transformed into a wild-type strain (CY337), a swi2^– strain (CY407) or a Δhap1 strain (YA1) and their activity assayed as described in Materials and Methods. GPD–HAP1 is a Hap1 expression vector under the control of the glyceraldehyde 3-phosphate dehydrogenase (GPD) promoter transformed along with the CYC1 reporter. (b) The lacZ reporters ‘CYC1’ and ‘CYC7’ are shown schematically.

Figure 2

Sequences of the junction of the fusion proteins recovered from the ‘two-hybrid’ screen. Amino acid and DNA sequences of the junction of Gal4(AD) and histone H3 are shown. Open reading frames YBR010W and YNL031C correspond to the HHT1 and HHT2 genes, respectively, and both encode histone H3.

Figure 3

Effect of Gal4(AD)–H3 on activity of Hap1 activity and derivatives. Activity of Hap1 was assayed at CYC1 or CYC7 using lacZ reporters. Reporters contain one copy of the UAS of CYC1 or CYC7 inserted upstream of the minimal CYC1 promoter driving lacZ transcription (46). Hap1 and None refer to use of the wild-type strain YPH499 and the Δhap1 strain YA1, respectively. Empty vector, plasmid pACT2 (80) used for construction of the cDNA library and expressing the activation domain of Gal4.

Figure 4

Domains of Gal4(AD)–H3 required for hyperactivation of Hap1. A lacZ reporter containing one UAS of CYC7 (46) was introduced into the HAP1⁺ strain YPH499 along with an expression vector for Gal4(AD)–H3 (or derivatives) and its activity assayed. Expression vectors (see Materials and Methods) are shown schematically on the left and their effect on Hap1 activity on the right.

Figure 5

Gal4(AD)–H3 cannot substitute for histone H3. Yeast strain MX4-22A (14) carrying deletions of all H3 and H4 genes and containing plasmid pMS329 (URA3; HHT1 HHF1, coding for H3 and H4) was transformed with a LYS2 plasmid carrying a wild-type H4 gene along with expression vectors for H3 or variants on a LEU2 plasmid. Transformants were then grown overnight in YPD and streaked on FOA plates to test whether the URA3 plasmid can be lost. WILD TYPE H3, plasmid pH3; NO H3, pACT2 (80).

Figure 6

Effect of Gal4(AD)–H3 and derivatives on various activators. Activity of various transcriptional activators was assayed with lacZ reporters. Results are an average of two to three independent transformations. Uga3 and Leu3 reporters contain one copy (20mer) of the UAS of UGA1 and LEU2, respectively, inserted upstream of the minimal CYC1 promoter (53,73). The Gal4 reporter contains four copies of UAS_GAL4 found in the GAL1-10 intergenic region (pLGSD5) (74). The ER reporter contains three estrogen-responsive elements inserted upstream of the minimal GAL1 gene (75). The Gcn4 reporter contains binding sites for Bas1, Bas2 and Gcn4 (76). The Hap2/3/4/5 reporter contains CYC1 sequences up to –265 bp (77). Uga3, a yeast activator inducible by γ-aminobutyrate (GABA) (78), was assayed in the presence (0.1%) or absence of GABA. ER, human estrogen receptor α expressed in yeast (see Materials and Methods). Activity of ER was induced with 500 nM estradiol (+hormone). Uga3, Leu3, Gal4, Gcn4 and Hap2/3/4/5 were expressed from their natural chromosomal locations. Empty vector, plasmid pACT2 (80) used for construction of the cDNA library and expressing the activation domain of Gal4.

Figure 7

Hyperactivation by Gal4(AD)–H3 is not dependent on the strength of the activator. Expression vectors for Gal4(1–147)–VP16C (or mutants) along with an expression vector for Gal4(AD)–H3 (or the empty vector) were transformed into yeast strain Y190, which contains an integrated GAL1–lacZ reporter, and β-galactosidase activity assayed. Amino acid changes in the Gal4(1–147)–VP16C mutants (85) are given on the left. + refers to the presence of the expression vector for Gal4(AD)–H3, – to the empty vector pACT2 (80).

Figure 8

Effect of Gal4(AD)–H3 on the expression of endogenous genes. Northern blot analysis of endogenous genes. Yeast strain T17 was transformed with an expression vector for Gal4(AD)–H3 or the parental vector pACT2 (80) and RNA isolated for northern blot analysis (see Materials and Methods). Twenty micrograms of total RNA were loaded per lane. GENE, probe used for northern blot analysis; +, presence of the expression vector for Gal4(AD)–H3; –, empty vector pACT2 (80).

Figure 9

Models for the mode of action of Gal4(AD)–H3. Models (a–d) for the mechanism of action of the H3 fusion protein Gal4(AD)–H3 are schematically shown. The exact position of the nucleosomes on the minimal CYC1 promoter is not known and they were drawn at arbitrary locations. TFIID is apparently constitutively bound to CYC1 (79). (a) Normal situation: Hap1 binding to target sites is favored by the SWI complex allowing interaction of Hap1 with the basic transcriptional machinery and initiation of transcription. (b) Gal4(AD)–H3 is incorporated into the nucleosome resulting in a less stable nucleosomal structure that can be more easily remodeled by the SWI complex resulting in increased activity by Hap1. (c) Gal4(AD)–H3 is not incorporated into the nucleosome and the H3 moiety of the fusion protein interacts with Hap1 allowing a classical ‘two-hybrid effect’. (d) Gal4(AD)–H3 forms a complex with Hap1 favoring chromatin remodeling by the SWI complex resulting in increased transcription.