Distinct amino acid compositional requirements for formation and maintenance of the [PSI⁺] prion in yeast

Abstract

Multiple yeast prions have been identified that result from the structural conversion of proteins into a self-propagating amyloid form. Amyloid-based prion activity in yeast requires a series of discrete steps. First, the prion protein must form an amyloid nucleus that can recruit and structurally convert additional soluble proteins. Subsequently, maintenance of the prion during cell division requires fragmentation of these aggregates to create new heritable propagons. For the Saccharomyces cerevisiae prion protein Sup35, these different activities are encoded by different regions of the Sup35 prion domain. An N-terminal glutamine/asparagine-rich nucleation domain is required for nucleation and fiber growth, while an adjacent oligopeptide repeat domain is largely dispensable for prion nucleation and fiber growth but is required for chaperone-dependent prion maintenance. Although prion activity of glutamine/asparagine-rich proteins is predominantly determined by amino acid composition, the nucleation and oligopeptide repeat domains of Sup35 have distinct compositional requirements. Here, we quantitatively define these compositional requirements in vivo. We show that aromatic residues strongly promote both prion formation and chaperone-dependent prion maintenance. In contrast, nonaromatic hydrophobic residues strongly promote prion formation but inhibit prion propagation. These results provide insight into why some aggregation-prone proteins are unable to propagate as prions.

FIG 1

Prion formation library experiment. (A) Schematic of Sup35. The PFD is enlarged below, showing the ND and ORD. (B) Sequence of Sup35. The oligopeptide repeats are underlined. The region of the ND and ORD targeted for mutagenesis are in bold italics. (C) Experimental scheme for prion formation library experiments. [psi^–] cells in which Sup35C was expressed from a URA3 plasmid as the sole copy of Sup35 were transformed with a randomly mutated version of Sup35 and then selected for loss of the wild-type plasmid (step 1). Cells were screened to remove clones in which the mutant Sup35 had compromised activity, and randomly selected clones were sequenced to generate the naive library. The library was then screened for clones that could form and propagate prions (step 2).

FIG 2

Nonaromatic hydrophobic residues show different prion formation and maintenance propensities. Comparisons of the previously determined log odds ratios based on mutagenesis of Sup35-27 (13) were undertaken to the log odds ratios from the ND (A) or ORD (B) prion formation library experiments, the ORD prion maintenance library experiment (C), or the prion propagation library experiment in which an additional step was added to remove mutants that were not efficiently recruited into wild-type prion aggregates (D). While the odds ratios for charged, aromatic, polar residues (filled diamonds) showed similar trends in each library, nonaromatic hydrophobic residues (open diamonds) scored substantially worse in the ORD prion formation and maintenance libraries. Charged residues are Asp, Glu, Lys, and Arg. Polar residues are Ser, Thr, Asn, and Gln. Aromatic residues are Trp, Tyr, and Phe. Nonaromatic hydrophobic residues are Leu, Ile, Val, and Met. Error bars indicate standard errors, calculated according to equation 5.

FIG 3

Prion maintenance library experiments. (A) Experimental scheme. [psi^–] cells in which Sup35C was expressed from a URA3 plasmid as the sole copy of Sup35 were transformed with a randomly mutated version of Sup35 and then selected for loss of the wild-type plasmid (step 1). These cells were mated with wild-type [PSI⁺] cells in which the sole copy of Sup35 was expressed from a URA3 plasmid (step 2). After selection for loss of the URA3 plasmid (step 3), red and white clones were sequenced. In the modified protocol to select against mutants with a defect in adding onto wild-type aggregates, selection for diploid cells in step 2 was done in the absence of adenine. (B) Plasmids expressing mutant Sup35s from individual red prion maintenance library isolates were transformed into [PSI⁺] cells in which the sole copy of Sup35 was expressed from a URA3 plasmid. To test whether the mutant Sup35s were inactivated in the presence of wild-type [PSI⁺], cells were plated on limiting adenine medium, selecting for both the wild-type and mutant Sup35-expressing plasmids (left). Cells were then retested on YPD after selection for loss of the wild-type plasmid (right). Representative examples are shown, with the sequences of the mutagenized regions indicated. Sup35-27, a scrambled version of Sup35 that is not incorporated into wild-type [PSI⁺] aggregates, and wild-type Sup35 are shown as controls.

FIG 4

Receiver operator characteristic (ROC) plot (57) assessing the ability of PMP scores and PAPA to predict the prion-propagating library mutants. A leave-one-out method of cross-validation was used to assess whether PMP scores from the prion maintenance library experiment were sufficient to predict which library members would successfully propagate [PSI⁺]. PMP scores showed reasonable prediction accuracy (area under the curve [AUC], 0.79); the star indicates the point on the ROC plot for a PMP score of zero. PAPA showed virtually no ability to distinguish between red and white isolates (AUC, 0.56), with prediction accuracy barely above what would be expected by random chance (dotted line). False positive rate = (number of red isolates scored as prion propagating)/(total number of red isolates). True positive rate = (number of white isolates scored as prion propagating)/(total number of red isolates).

FIG 5

Successful design of prion-propagating sequences. A library of random 10-amino-acid sequences was built in silico. The library was screened using the PMP scores from the ORD prion propagation library experiment. Six high-scoring sequences (left side of each panel) and six low-scoring sequences (right side) were selected and inserted into Sup35 in place of the third repeat of the ORD. Mutants were introduced to wild-type [PSI⁺] cells. Transformants were spotted onto 5-FOA to select for loss of the plasmid expressing wild-type Sup35, and either streaked onto YPD medium to test for loss of [PSI⁺] (A) or streaked onto SC medium plus 4 mM guanidine-HCl and then streaked onto YPD medium to test for loss of [PSI⁺] (B). Untreated wild-type [PSI⁺] and [psi⁻] cells are shown as controls.

FIG 6

Aromatic residues in the ORD are critical for prion propagation. (A) Prion maintenance by tyrosine substitution mutants. The five tyrosines in repeats 3 to 5 of the Sup35 ORD were replaced with Ala, Val, Ile, Leu, Met, Phe, or Trp. These mutants were introduced into wild-type [PSI⁺] cells expressing wild-type Sup35 from a plasmid. After selection for loss of the wild-type plasmid, cells were streaked onto YPD medium to test for the ability to maintain [PSI⁺]. Strain YER632/pJ533 was included as a [psi⁻] control (632). (B) Tyrosine substitution mutants were efficiently incorporated into wild-type [PSI⁺] aggregates. Plasmids expressing GFP fusions of each tyrosine substitution mutant PFD under the control of the GAL1 promoter were transformed into wild-type [PSI⁺] and [psi⁻] strains. Cells were grown in galactose/raffinose dropout medium for 2 h and visualized by confocal microscopy. Foci were observed for each fusion in [PSI⁺] cells but not [psi⁻] cells. (C) Prion formation by tyrosine substitution mutants. [psi⁻] strains expressing each mutant as the sole copy of Sup35 were transformed either with an empty vector (left) or with a plasmid expressing the matching Sup35 mutant under the control of the GAL1 promoter (right). All strains were cultured for 3 days in galactose/raffinose dropout medium, and then 10-fold serial dilutions were plated onto medium lacking adenine to select for [PSI⁺]. (D) Tryptophan, alanine, and phenylalanine substitution mutants form stable, curable prions. Ade⁺ isolates from panel B were streaked onto either SC medium (−) or SC plus 3 mM guanidine-HCl (+) and then restreaked onto YPD to test for prion loss. Two representative Ade⁺ isolates are shown for each mutant. (E) Overexpression of the tyrosine substitution mutants induced wild-type [PSI⁺] formation. Yeast expressing wild-type Sup35 were transformed with either an empty vector (vector) or the vector modified to express either the wild-type Sup35 NM domain (wild-type) or the NM domain of the ORD tyrosine substitution mutants under the control of the GAL1 promoter. Cells were then tested for [PSI⁺] formation.

FIG 7

Aromatic residues are overrepresented and nonaromatic hydrophobics are underrepresented among domains with prion activity. Alberti et al. (12) tested 100 prion-like domains in four assays for prion-like activity. Three of the assays tested aggregation activity, while a fourth tested the ability of the domains to support prion activity when inserted in place of the Sup35 PFD. Box-and-whisker plots show the frequency of Q/N residues (A), aromatic residues (B), and nonaromatic hydrophobic residues (C; Ile, Leu, Met, and Val) among each of the Alberti proteins that passed all tests (white bars) or that passed all tests except the Sup35-fusion protein assay (gray bars).