Yeast mutants resistant to E1A

Although expression of Gal4p-E1A_1-82 inhibited cell growth with the same G₁ phenotype as full length E1A proteins, it was formally possible that this effect was a synthetic function created by the fusion of the E1A residues to the DNA binding domain of Gal4p. Such a synthetic toxicity has been noted previously for fusions of Gal4p with the transcription activation domain of VP16 (6). To address this possibility we undertook a screen for yeast mutants that were resistant to expression of Gal4p-E1A_1-82. We reasoned that if the toxicity of Gal4p-E1A_1-82 was a novel synthetic function of the fusion protein, then yeast mutants selected for resistance to Gal4p-E1A_1-82 would remain sensitive to inhibition by full length E1A proteins. On the other hand, if such mutants were also resistant to full length E1A, it would provide compelling evidence that the growth inhibition is a normal function of the E1A_1-82 polypeptide.

Mutants resistant to Gal4p-E1A_1-82 expression were isolated as described in Materials and Methods. We refer to the genes defined by these mutants as ``NET'' genes, for N-terminal E1A toxicity, the phenotype of the wild type allele. The mutant net1-1 was isolated by its resistance to high copy expression of Gal4p-E1A_1-82. This resistance segregated as a single nuclear gene in genetic crosses and secondary screening by Western blot analysis showed that Gal4p-E1A_1-82 continued to be expressed in the mutant cells. The resistance of net1-1 to E1A expression is illustrated Fig. 6. The wild type yeast strain transformed efficiently with the pMA424 vector alone (Fig. 6A) but failed to give colonies when transformed with the plasmid expressing Gal4p-E1A_1-82 because expression of the fusion protein killed the transformants (Fig. 6B). In contrast, net1-1 was efficiently transformed by plasmids expressing the fusion (Fig. 6D).

The net1-1 mutant was then tested for its resistance to expression of full length E1A₂₄₃. For these experiments, wild type and net1-1 cells were transformed with either vector alone or with plasmids expressing E1A₂₄₃, and transformants were grown on media containing glucose to maintain repression of the E1A gene. Serial dilutions of cultures were then spotted on plates containing either glucose, or galactose to induce the expression of E1A₂₄₃ (Fig. 7).

Figure 7 Cells that are net1-1 are resistant to expression of full length E1A₂₄₃. MSY596 (wild type) or MSY826 (net1-1) cells were transformed with either control plasmid ('vector'), or plasmids expressing full length E1A₂₄₃ protein under control of the GAL1 promoter ('E1A'), and selected on glucose medium. Serial dilutions of transformants were then plated on either glucose or galactose plates as indicated and examined for growth. (A) MSY596; (B) MSY826 (net1-1)

As can be seen in Fig. 7A, the E1A₂₄₃ plasmid inhibited the growth of wild type cells on galactose, where expression of E1A₂₄₃ is induced, but not on glucose, where it is repressed. However, net1-1 was insensitive to the expression E1A₂₄₃ as both control cells and those transformed with E1A₂₄₃ grew at similar rates on galactose and glucose media (Fig. 7B). Thus, the net1-1 mutant, selected for its resistance to expression of Gal4p-E1A_1-82, is also resistant to full length E1A₂₄₃. These results provide strong genetic evidence that growth inhibition by Gal4p-E1A_1-82 reflects the normal functions of the E1A domains and not a synthetic function of the Gal4p fusion.

We tested the specificity of net1-1 for resistance to E1A by challenging the cells with PadhGV16, a plasmid expressing a Gal4p-VP16 fusion that is toxic to wild type cells (6). As expected, this plasmid failed to give viable colonies when used to transform the wild type yeast strain (Fig. 6C). However, net1-1 also remained highly sensitive to expression of Gal4p-VP16 and failed to give viable colonies (Fig. 6E). Thus, the resistance of net1-1 for Gal4p-E1A_1-82 expression cannot be a general resistance to the toxic properties of transcriptional activators fused to Gal4p. Further, these results indicate that the molecular mechanism of E1A toxicity must be distinct from that of Gal4p-VP16.

Next:Identification of two NET genes: adenylate cyclase and the Ras GTP-GDP exchange factor Cdc25p Up:TOC Previous:Residues 1-82 are sufficient for function

mms7r@Virginia.EDU
November 26, 1995 at 9:52 PM