We have sequenced the genomes and identified SNPs in two strains of S. cerevisiae commonly used for studying the yeast [PSI+] prion. To our knowledge these are currently the only two yeast strains sequenced that are confirmed as harboring [PSI+]. Both strains are laboratory strains and this is reflected in their close evolutionary relationship to the S. cerevisiae reference strain S288C compared to other strains analyzed as part of the SGRP (). Of the two [PSI+] strains analyzed, G600 is by far more closely related to S. cerevisiae strains commonly used within the general yeast research community. In addition to the much greater number of non-synonymous amino acid changes in 74-D694 than G600 compared to the reference strain, 74-D694 also harbors a much greater number of ISCMs throughout its genome. This finding has major significance for this strain and its use for analysis of [PSI+] related phenotypes. Potentially any internal stop codon mutation could affect protein production from the affected ORF. The presence of [PSI+] in this strain allows for the possible readthrough of any internal stop codon mutation within an ORF and could therefore have implications for analysis of [PSI+]- related phenotypes using this strain. Although harboring much fewer internal stop codon mutations, the same rationale needs to be applied to the G600 strain.
The identification of potential ISCMs in a variety of genes in strain 74-D694 may help to explain fully, or in part, some previously identified complex phenotypes associated with the presence or absence of [PSI+
. For instance, the presence of [PSI+
] has been shown to alter growth on nitrogen sources; we have identified an ISCM that could cause a major truncation in the Dal4p allantoin permease protein. The presence of [PSI+
] has been shown to alter cellular responses to exposure to metals, oxidative stress inducing agents and other stress inducing agents; we have identified ISCM in a number of genes that could be implicated in alteration of the cellular stress response. Most notably the presence of an ISCM in the MSN4
gene could have major implications for strain 74-D694's ability to deal with oxidative stress. Msn4p is a transcription factor that works in conjunction with Msn2p to induce a variety of genes in response to oxidative cellular damage. However, there are a number of genes containing ISCMs in 74-D694 () that are involved in (or potentially involved in) aspects of responding to DNA damage or stress, which could singularly or in combination contribute to complex cellular stress responses of 74-D694 in the presence or absence of [PSI+
]. Even though we currently have no data relating to the expected level of readthrough or alteration in mRNA stability of the ISCM containing genes, the identification of these potential ISCMs should be viewed as providing a framework for understanding or contributing significantly to elucidating the influence of [PSI+
] on prion-dependent phenotypes in this well-used laboratory strain.
Compared to reference strain S288C, G600 harbors far less potential ISCMs than 74-D694. In addition to the known E64TAA change in the ADE2
we identified ISCMs in ORFs YBR074W and YNL106C. The nonsense mutation in the AFI1
gene would only cause a six amino acid truncation of the translated protein, with full-length protein being 887 amino acids and deletion of AFI1
having a minimal effect on cell phenotype (Saccharomyces
Genome Database) we predict this potential 6 amino acid truncation would not have much impact on the cell. A previously identified [PSI+
] prion-dependent phenotype has been shown to be at least partly attributable to readthrough of the ISCM in the uncharacterized YBR074W ORF (). YBR074W is an uncharacterized yeast ORF that is predicted to be a metalloprotease 
. Although we show a link between this gene and protection against chemicals that induce cell wall stress, we cannot provide any insight to the possible mechanistic role this protein may play in such protection. Thus, we now identify two [PSI+
] prion dependent phenotypes associated with readthrough of ISCMs in strain G600, adenine utilization and ADE2
and cell wall stress response and YBR074W. In addition, given the common ancestry of G600 with other previously phenotypically characterized [PSI+
] strains (BSC783/4, 5V-H19 and 10B-H49) we can postulate that prion-dependent phenotypes identified in these strains may possibly result from readthrough of the two newly identified potential ISCMs in the G600 strain also being present in these strains 
We were initially surprised by the identification of what seemed a relatively large number of ISCMs in strain 74-D694. This raised the intriguing possibility that the presence of [PSI+] in this strain had allowed the accumulation of ISCMs. Such a proposal is not unreasonable given Sup35p's role in translation termination and the reduction in fidelity in [PSI+] strains. We therefore sought to assess the frequency of potential ISCMs in other sequenced S. cerevisiae strains. As part of the SGRP whole genome sequence data is available for thirty-eight S. cerevisiae strains. SNP analysis of these forty yeast strains shows a huge variation in the frequency of strain-specific ISCMs (). Among the wild yeast a number of strains have accumulated zero strain-specific ISCMs whereas others have accumulated up to around 20. Taking into account the ISCM frequency per non-synonymous SNPs it appears that ISCM accumulation can be affected by an array of complex genetic and environmental factors. We could find no common genetic or environmental factors to account for the high (or low) occurrence of ISCMs in some wild yeast. From this analysis we could find no evidence to suggest that the presence of [PSI+] could increase the rate of ISCM accumulation in evolving S. cerevisiae.
While our data suggest that the presence [PSI+
] does not influence the levels of ISCM accumulation, this proposal cannot be fully dismissed due to a number of considerations that need to be taken into account. Each non-laboratory strain sequenced as part of the SGRP is a haploid segregant from a diploid parent. Therefore the original parent strain has the potential to accumulate ISCMs without potentially altering protein function due to the extra copy of an altered gene. This fact could artificially increase the numbers of ISCMs in the wild yeast haploid segregants. Additionally the potential biological significance of an ISCM in these strains would only be observed if the original parent was homozygous for the ISCM. Such data is not available. The laboratory strains are all haploid and all have a relatively high frequency of ISCMs compared to the majority of the wild yeasts (). The potential impact of [PSI+
] on ISCM accumulation could further be masked by growth in the laboratory environment 
. Gu et al 
have demonstrated that the rate of evolution appears higher in laboratory strains compared to wild yeast. This increased rate was attributed to a relaxed selection intensity, which could also impact on the accumulation of ISCMs. To fully address whether [PSI+
] can increase ISCM accumulation a controlled long-term evolution experiment with isogenic [PSI+
] and [psi−
] cells needs to be carried out.
Evidence has accrued to support the initial proposal by True and Lindquist 
] may allow evolutionary adaptation in specific growth conditions by allowing readthrough of naturally occurring stop codons 
. To assess whether we could find evidence for this at the genome level in the form altered C-terminal regions of proteins in [PSI+
] strains, we identified all altered naturally occurring stop codons in 74-D694, G600 and all the SGRP strains compared to S288C (data not shown). We found no evidence for alteration or extensions of C-terminal regions of proteins in the [PSI+
] strains. However, this is not surprising as any potential environmental adaptation caused by C-terminal extension of a protein would be transient and would not become fixed in the population unless the environmental selection persisted.
Using strain S288C as the reference strain means that by default we set the number of potential ISCMs in this strain as zero. However, this is not the case and S288C itself has some well- documented ISCMs (Saccharomyces Genome Database). We therefore assessed whether S288C ISCMs are shared in the two [PSI+] strains and throughout the SGRP strains. Both 74-D694 and G600 harbor conserved ISCMs in the small ORF CRS5 and in FLO8 while ISCMs in ORFs PAU7 and PRP45 were present only in 74-D694. In addition an ISCM in YNR066C was present only in G600. However, absence or presence of S288C defined ISCMs in 74-D694 and G600 appears independent of the presence of [PSI+] as they are also present in various SGRP strains (data not shown).
We noted that all ORFs containing potential ISCMs identified in strain G600 and 74-D694 are in non-essential genes. Theoretically it may be expected that the presence of [PSI+] may allow the accumulation of ISCMs in essential genes also. However, given that [psi−] derivatives of such mutations will most likely be inviable and this may result in the counter-selection against such mutations in the laboratory environment, which in turn may explain the apparent bias for ISCM generation in non-essential genes in these two strains.
Following the identification of twelve polymorphisms in the Hsf1 protein as well as a number of lesser changes in chaperone proteins known to influence prion propagation (all but one being absent from G600), we decided to assess the ability of G600 and 74-D694 to grow at elevated temperatures. Given the key role that chaperones play in modulating prion propagation, as attested to by research using both these [PSI+
] strains 
, any functional differences in chaperones due to the presence of identified polymorphisms will help to accurately assess the roles of these proteins in prion propagation in a strain- specific and more general manner. Although there is a major difference in the ability of G600 and 74-D694 to grow at 39°C (), there is no significant difference in the capability of these strains to acquire thermotolerance or to induce Hsp104 (). The ability of 74-D694 to grow at elevated temperatures is potentially further compromised by the absence of a QTL associated with high temperature growth. Previous studies have identified polymorphisms present in the NCS2
genes that can influence (in a background specific manner) growth at high temperature 
. The associated QTLs are present in G600 and partly absent in 74-D694. Thus, although we identify chaperone polymorphism differences in G600 and 74-D694, these do not appear to be responsible for differences in these strains to grow at elevated temperatures. This finding suggests that overall the mechanism of heat shock response and prion propagation are similar in these two strains with respect to chaperone involvement.
Analyzing the genome of two [PSI+] strains of S. cerevisiae has identified a number of ORFs harboring potential ISCMs. The presence of ISCMs in specific genes has allowed us to develop hypotheses for the causes of previously identified prion- associated phenotypes. In addition the genome and polymorphism data has allowed us to analyze the causes of differences in heat shock responses in these two strains. This genome data will allow researchers using these strains to develop and test hypotheses based on possible involvement of [PSI+] (and perhaps other prions) in influencing phenotypic variation. Our study clearly shows that given the relative ease and low cost of obtaining yeast genome level data for specific strains, the information that can be amassed from such investment can have major impact on hypothesis driven research and experimental design.