A Role for Cytosolic Hsp70 in Yeast [PSI⁺] Prion Propagation and [PSI⁺] as a Cellular Stress

Corresponding author: Daniel C. Masison, Bldg. 8, Rm. 407, 8 Center Dr., MSC 0851, National Institutes of Health, Bethesda, MD 20892-0851., masisond{at}helix.nih.gov (E-mail)

Communicating editor: A. P. MITCHELL

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

[PSI⁺] is a prion (infectious protein) of Sup35p, a subunit of the Saccharomyces cerevisiae translation termination factor. We isolated a dominant allele, SSA1-21, of a gene encoding an Hsp70 chaperone that impairs [PSI⁺] mitotic stability and weakens allosuppression caused by [PSI⁺]. While [PSI⁺] stability is normal in strains lacking SSA1, SSA2, or both, SSA1-21 strains with a deletion of SSA2 cannot propagate [PSI⁺]. SSA1-21 [PSI⁺] strains are hypersensitive to curing of [PSI⁺] by guanidine-hydrochloride and partially cured of [PSI⁺] by rapid induction of the heat-shock response but not by growth at 37°. The number of inheritable [PSI⁺] particles is significantly reduced in SSA1-21 cells. SSA1-21 effects on [PSI⁺] appear to be independent of Hsp104, another stress-inducible protein chaperone known to be involved in [PSI⁺] propagation. We propose that cytosolic Hsp70 is important for the formation of Sup35p polymers characteristic of [PSI⁺] from preexisting material and that Ssa1-21p both lacks and interferes with this activity. We further demonstrate that the negative effect of heat stress on [PSI⁺] phenotype directly correlates with solubility of Sup35p and find that in wild-type strains the presence of [PSI⁺] causes a stress that elevates basal expression of Hsp104 and SSA1.

Sup35p is a Saccharomyces cerevisiae protein involved in termination of translation. In a state referred to as [PSI⁺], a significant portion of the Sup35p in the cell coalesces into nonfunctional, self-propagating, amyloid-like polymers (PATINO et al. 1996 ; PAUSHKIN et al. 1996 ; GLOVER et al. 1997 ; KING et al. 1997 ). Once present, [PSI⁺] propagates by recruitment of the soluble form of Sup35p into the aggregate in a manner analogous to that of mammalian prions (WICKNER 1994 ; LANSBURY and CAUGHEY 1995 ; PAUSHKIN et al. 1997 ). [PSI⁺] propagates in the cytoplasm and is virtually infallibly transmitted to all progeny during mitosis and meiosis. Depletion of soluble Sup35p by [PSI⁺] reduces translation termination efficiency causing nonsense suppression. Despite this and the requirement of Sup35p for growth, [PSI⁺] has no overt effect on the fitness of strains that carry it.

The mechanisms balancing the levels of soluble and aggregated Sup35p, essential for cell growth and for [PSI⁺], respectively, are not well understood. The protein chaperone Hsp104, which is induced by environmental stress and resolubilizes stress-denatured proteins (PARSELL et al. 1994 ), influences Sup35p solubility in [PSI⁺] strains. Deletion of HSP104 is incompatible with [PSI⁺] maintenance and plasmid-based overexpression of Hsp104 at optimal growth temperature increases Sup35p solubility, reducing both [PSI⁺]-mediated nonsense suppression and [PSI⁺] mitotic stability (CHERNOFF et al. 1995 ). A new role for Hsp104 in assisting the aggregation of Sup35p has been proposed to explain the requirement of Hsp104 for [PSI⁺] maintenance (PATINO et al. 1996 ). Both effects have also been suggested to be due to the disaggregating activity of Hsp104 that generates new inheritable [PSI⁺] "seeds" by breaking up Sup35p aggregates; lack of Hsp104 results in failure to generate new seeds, and its overabundance favors complete solubilization of Sup35p (PAUSHKIN et al. 1996 ). Unlike overexpression under optimal growth conditions, physiologically elevated expression of Hsp104 in the context of heat shock, sporulation, and entry into stationary phase does not cause elimination of [PSI⁺]. A simple explanation for this may be that [PSI⁺] escapes complete dissolution because much of the additional Hsp104, whose abundance is regulated by the need for its activity, is occupied with other substrates under conditions that induce its expression. Alternatively, other Hsps induced by stress may affect [PSI⁺] metabolism under these conditions.

The [PSI⁺]-curative effect of artificially overexpressing Hsp104 at optimal growth temperature is partially inhibited by simultaneous overexpression of Ssa1p, a constitutively expressed Hsp70 whose expression is further induced by stress (NEWNAM et al. 1999 ). This inhibition correlates with a decreased shift in Sup35p solubility while the elevated level of Hsp104 remains unchanged. This led to the suggestion that resistance of [PSI⁺] to the elevated level of Hsp104 during stress may be due to a protective effect of a concomitant elevated level of Ssa1p that inhibits Hsp104 activity (NEWNAM et al. 1999 ). However, overexpressing Ssa1p has no effect on Hsp104 activity after a shift to elevated temperature, the physiological condition that induces Hsp104 expression where [PSI⁺] is stable (NEWNAM et al. 1999 ).

This study also found that overexpression of SSA1 increases nonsense suppression by [PSI⁺], further suggesting a positive effect of Ssa1p on [PSI⁺]. On the other hand, it was reported that overexpression of SSA1 antagonizes the ability of [PSI⁺] to mediate nonsense suppression, leading to the opposite conclusion (CHERNOFF et al. 1995 ). Clearly, the role of Ssa1p in [PSI⁺] metabolism is not well understood.

Assigning a specific role for Ssa1p in [PSI⁺] metabolism is difficult since it belongs to a conserved, functionally redundant subfamily (SSA) of four cytosolic Hsp70 chaperones in yeast (BOORSTEIN et al. 1994 ). Also, in addition to its protein-folding function, Ssa1p is a negative regulator of its own expression and possibly that of other stress response proteins (STONE and CRAIG 1990 ; CRAIG and GROSS 1991 ). Ssa2p, the major constitutively expressed Hsp70, is identical to Ssa1p at 624 of its 639 residues. SSA3 and SSA4, whose expression is repressed under optimal growth conditions, are 88% identical to each other and 80% identical to SSA1/SSA2 (BOORSTEIN et al. 1994 ). Deleting both constitutively expressed SSA1 and SSA2 results in derepression of SSA3 and SSA4, a response to maintain essential levels of cytosolic Hsp70 activity (WERNER-WASHBURNE et al. 1987 ). The conservation and functional redundancy of the differentially regulated SSA genes has led to the implication that rather than providing distinct functions, the multiple genes provide appropriate Hsp70 expression patterns under various environmental conditions (BOORSTEIN and CRAIG 1990 ). On the other hand, there are specific regions of divergence within this family and the importance of the most highly conserved SSA1/SSA2 pair is evident in that the double deletion has pleiotropic phenotypes despite compensatory expression of other SSAs (CRAIG 1992 ).

In a screen designed to identify cellular factors important for maintaining the balance between soluble and aggregated forms of Sup35p in [PSI⁺] strains, we isolated a mutant allele of SSA1 (SSA1-21) that dramatically impairs [PSI⁺] metabolism. The effects of this mutation provide new insight into the role of Ssa1p in [PSI⁺] metabolism, suggesting that Hsp70 plays an essential role in [PSI⁺] prion propagation. We find that in SSA1-21 strains the number of inheritable [PSI⁺] seeds is significantly reduced and that, while [PSI⁺] stability is normal in strains lacking SSA1, SSA2, or both, deleting SSA2 in an SSA1-21 mutant abolishes its ability to propagate [PSI⁺]. Our data support a model that Hsp70 assists in the generation of inheritable Sup35p polymers in [PSI⁺] cells.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and media:
Yeast strains are listed in Table 1. [PSI⁺] strains are designated by a superscripted plus (⁺) and are identical to [psi^-] strains of the same name except for the presence of [PSI⁺]. The SSA1-21 allele was generated in 628-3A⁺ by mutagenesis using ethyl methanesulfonic acid (Sigma, St. Louis) as described (LAWRENCE 1994 ). 704-7C is an SSA1-21 strain congenic with 628-3A from a fourth backcross. Strain 704-7C crossed with 628-3A produced the diploid strain 707. 707-3C is a meiotic segregant of strain 707. Strain D320 (a meiotic segregant of strain MW63; WERNER-WASHBURNE et al. 1987 ) crossed with 628-2D (congenic with 628-3A) produced strain 668, which is the source of the SSA disruption alleles ssa1::HIS3 and ssa2::LEU2. Because these strains are not isogenic, direct comparisons between them cannot be made with complete conviction. To increase confidence in interpreting the results, several strains with the same genotype were used in experiments as indicated. In every case, similar results were obtained with the additional strains. YPD, YPAD, and synthetic media are as described (SHERMAN 1994 ). Synthetic galactose media contain 2% raffinose and 2% galactose.

Plasmids:
The SSA1 gene was cloned from strain 628-3A. S1-2, the originally selected mutant, using the polymerase chain reaction (PCR) with Taq polymerase (Promega, Madison, WI) and the following primers: (1) 5'-GGGCCCGGATTCCACCTGCAGGGTCTGAGCCC-3' and (2) 5'-GGGCCCAAGCCTCGTAGTCTAAATGAGTTACG-3'. Two independently amplified PCR products were cut with restriction enzymes HindIII and BamHI and inserted into pRS315 (SIKORSKI and HIETER 1989 ) cut by the same enzymes, generating plasmids pRDW3 and pRDW5. Plasmid pRDW50 contains the same fragment from pRDW5 inserted into YCp50 (ROSE et al. 1987 ). Plasmid pJ101 was constructed by isolating the 7.2-kbp BamHI-HindIII fragment containing SSA1 from plasmid pYG100BH, from Elizabeth Craig (INGOLIA et al. 1982 ), and inserting it into BamHI-HindIII-digested pRS315 (SIKORSKI and HIETER 1989 ). Plasmid pJ110 is identical to pJ101 with the exception that nucleotide +1448 of the SSA1 coding region (where +1 marks the A of the initiator methionine) was changed from thymine to guanine using the Stratagene (La Jolla, CA) QuickChange Site-Directed mutagenesis kit. The oligonucleotides used to do this were 5'-CGACTCTAACGGTATTTGGAATGTTTCCGCCGT-3' and 5'-ACGGCGGAAACATTCCAAATACCGTTAGAGTCG-3', where codon 483 is underlined and the altered base is boldface. Plasmid pZFO, from Elizabeth Craig (STONE and CRAIG 1990 ), is a YCp50-based plasmid containing the SSA1 promoter and the first 30 bases of SSA1 coding DNA fused in frame to the Escherichia coli lacZ gene. Plasmids pW4UGG and pW4UAA, from David Bedwell (BONETTI et al. 1995 ), contain a leader peptide with a tryptophan (W4UGG) or an ochre codon (W4UAA) at the fourth amino acid position fused in frame to the E. coli lacZ gene, under control of the Gal1 promoter. Plasmid pH219 (from H. Edskes) was constructed by inserting the NotI-SalI fragment from pFl44L-HSP104 (M. Bogota, Warsaw) into NotI-SalI-digested pRS314 (SIKORSKI and HIETER 1989 ).

Genetic methods:
The presence of [PSI⁺] allows the weak suppressor tRNA, SUQ5 (SUP16), to suppress a UAA nonsense codon within ade2-1 (COX 1965 ; LIEBMAN et al. 1975 ). Nonsuppressed [psi^-] cells require adenine for growth and are red when adenine is limiting (i.e., on YPD medium) due to accumulation of a pigmented form of the substrate of Ade2p, aminoamidizole ribotide (AIR; SILVER and EATON 1969 ). Suppressed [PSI⁺] cells grow in the absence of adenine and are white on YPD. However, any mutation or condition that affects mRNA stability or translation efficiency can affect nonsense suppression and influence color development or auxotrophic growth. Additionally, certain metabolic conditions, such as respiratory deficiency, result in failure to become pigmented despite accumulation of AIR. Therefore, in addition to its ability to confer adenine prototrophy and white color, we routinely verify the presence of [PSI⁺] by assaying both guanidine curability and either its cytoducibility or its dominant phenotype when crossed, followed by guanidine curability of resulting diploids. Our [PSI⁺] strains grown on nonselective indicator medium at elevated temperatures accumulate pigment, making it difficult to monitor the presence of [PSI⁺]. [PSI⁺] colonies on such plates are readily identified by the development of white halos of cells at their growing edge upon subsequent incubation at 25°.

Cytoduction assays were performed as described (MASISON et al. 1997 ). Briefly, [rho^-] recipient cells are mixed with [rho⁺] donor cells of opposite mating type (at least one partner having the kar1-1 mutation; CONDE and FINK 1976 ) and incubated on YPAD plates for 6–14 hr. Cells are then spread onto medium selecting for recipient cells. Cytoduction recipients, which become [rho⁺], are distinguished from infrequent diploids by their genotype.

Thermotolerance assays:
Assays were performed as described (NICOLET and CRAIG 1991B ) with minor modifications. First, cells from 23° YPAD cultures maintained at OD₆₀₀ < 0.6 for 20 hr were diluted in YPAD to OD₆₀₀ = 0.3. To assay basal thermotolerance, 500 µl was then transferred to a 13 x 100 mm borosilicate tube and 100 µl was removed and transferred to ice for a zero time point. The remainder was placed in a 52° water bath, and 100-µl aliquots were transferred to ice after indicated times. To assay acquired thermotolerance, 1 ml of the same diluted culture (OD₆₀₀ = 0.3) was first transferred to a 15-ml prewarmed tube and incubated at 39° for 50 min before removing 500 µl for the assay. Collected samples were titered by serially diluting in water and plating on YPD. Colonies were counted after incubating 2¹/₂ days at 30°.

Guanidine curing:
Routine curing of [PSI⁺] was done by growing cells to colonies on YPD containing 3 mM guanidine-hydrochloride (Gdn-HCl; Sigma). Cells were then restreaked onto YPD and red colonies isolated. For quantitative assays, cells from -ade selection plates were grown at 30° in YPAD to OD₆₀₀ 1.0, diluted to OD₆₀₀ = 0.1 in YPAD containing final concentrations of 1 mM or 3 mM Gdn-HCl and incubated with agitation at 30°. Cultures were maintained at OD₆₀₀ 1.0 by repeated dilution in appropriate medium. Cells from the cultures were assayed for the presence of [PSI⁺] by spreading 300–500 cells onto YPD plates and incubating for 3 days at 30°, followed by 3 or more days at 25°, and scoring for red ([psi^-]) or white ([PSI⁺]) color. Sectored colonies, which were observed at low frequencies in cultures containing 3 mM Gdn-HCl, were scored as [PSI⁺]. Since Gdn-HCl is a potent inducer of mitochondrial petites that remain white when [psi^-], scoring was further verified by mating the entire plate of cells and assaying for [PSI⁺] as described above.

Quantitation of promoter activity and readthrough of translation termination:
For all assays, cells at OD₆₀₀ = 0.8 growing in medium selecting for appropriate plasmids were collected by centrifugation and assayed for ß-galactosidase as described (GUARENTE 1983 ).

Protein blots:
Overnight YPAD cultures were diluted to OD₆₀₀ = 0.1–0.2, grown to OD₆₀₀ = 0.5–0.6, and cell extracts were prepared as described (DEPACE et al. 1998 ; NEWNAM et al. 1999 ) with minor modifications. Cells were collected by centrifugation, suspended in an equal volume of lysis buffer (25 mM Tris-HCl, pH 7.5, 50 mM KCl, 5 mM MgCl₂, 0.5 mM DTT, 5 mM phenylmethylsulfonyl fluoride (PMSF), 2 µg/ml pepstatin, 2 µg/ml leupeptin, and 100 µg/ml RNase A), and broken using a mini-bead beater (Biospec Products, Bartlesville, OK) by agitating with glass beads twice for 20 sec at maximum speed, incubating on ice for at least 30 sec between agitations. Cell debris was removed by centrifugation at 5000 x g for 5 min and the supernatant (total lysate) collected. Protein concentrations were determined using the Bio-Rad (Hercules, CA) protein assay kit, and 10 µg of protein for each sample was separated on 10% SDS/PAGE gels in Tris-HCl buffer (Bio-Rad). For fractionation experiments, a portion of the total lysate was centrifuged at 16,000 x g for 30 min. After removing the supernatant, the pellet was suspended in a volume of lysis buffer equivalent to that of the lysate from which it was derived. Equal volumes of sample were used when comparing supernatant and pellet fractions.

Proteins separated in the gels were then electrophoretically transferred to PVDF membranes (Millipore, Bedford, MA) and processed for immunoblotting according to instructions for use with the Bio-Rad minicell transfer apparatus. Antisera to Ssa1/2p and Ssa3/4p were gifts from Elizabeth Craig, anti-Sup35p antibodies were from Susan Lindquist and Mick Tuite, and anti-Hsp104 (SPA-1040) was purchased from Stressgen (Victoria, BC, Canada). Bio-Rad alkaline phosphatase-conjugated goat anti-rabbit IgG (#170-6518) was used as secondary antibody. Chemiluminesence using Bio-Rad Immun-Star substrate (#170-5012) and exposure to X-ray film (Kodak X-AR) was used to detect reacting antigen according to manufacturer's instructions. Quantitation of the density of bands on the blots presented in Fig 4 (using exposures giving maximum OD 1.0 per band) was done using a Scanmaster III Plus scanning densitometer (Howtek, Inc.) and Diversity One (Protein-Database, Inc.) software.

View larger version (149K): In this window In a new window Download PPT slide   Figure 1. Mitotic stability and suppressor phenotype of [PSI+]. SSA1-21 [PSI+] (704-7C+), wild-type (WT) [PSI+] (628-3A+), and wild-type [psi-] (628-3A) strains were grown on YPD for 2 days at 30° followed by 2 days at 25°. Small white colonies on all streaks are petites.

View larger version (115K): In this window In a new window Download PPT slide   Figure 2. Effects of temperature on [PSI+] phenotype. Patches of cells [top left, wild-type [psi-] (628-3A); top right, wild-type [PSI+] (628-3A+); bottom left, SSA1-21 [psi-] (704-7C); bottom right, SSA1-21 [PSI+] (704-7C+)] grown on YPD at 30° were velveteen replica-plated onto a series of YPD and -ade plates that were then incubated for 4 days at the indicated temperatures.

View larger version (68K): In this window In a new window Download PPT slide   Figure 3. Solubility of Sup35p at different temperatures. Whole cell lysates were prepared from [PSI+] wild-type (628-3A+) and SSA1-21 (704-7C+) cells grown at the indicated temperatures and separated by centrifugation into soluble (S) and pellet (P) fractions. Proteins in each fraction were then separated in denaturing polyacrylamide gels and immunoblotted using antiserum specific for Sup35p. For both strains, exposures were selected in which the intensity of the signal of Sup35p in the pellet fractions is similar. Overall levels of Sup35p do not change with respect to temperature (not shown). Similar results were obtained in two replicate experiments using these and another SSA1-21 strain (707-4C, data not shown).

View larger version (60K): In this window In a new window Download PPT slide   Figure 4. Basal and heat-induced levels of stress response proteins. Lysates prepared from strains grown at 27° (basal) and from portions of the same cultures after a shift to 37° for 1 hr (indicated as - and + heat shock, respectively) were separated in denaturing polyacrylamide gels and immunoblotted with antisera specific for Hsp104 (top), Ssa1/2p (middle), or Ssa3/4p (bottom). Strains are isogenic [PSI+] and [psi-] variants of wild type (WT, 628-3A), ssa1 (A1, 668-2D), SSA1-21 (A1-21, 704-7C), ssa2 (A2, 668-19B), ssa1 ssa2 (A1 A2, 668-33A), and SSA1-21 ssa2 (A1-21 A2, 674-1A). There is no [PSI+] variant of SSA1-21 ssa2 (see text). Identical samples from this strain were loaded on the [PSI+] and [psi-] blots as a control for selecting similar film exposures to allow a more direct comparison of expression level between [PSI+] and [psi-] strains. The Ssa1/2p data are from the same blots as those used for Hsp104, which were stripped and reprobed. Due to a very high amount of Hsp104 in the [psi-] ssa1 ssa2 cells, one-fifth of the amount of sample was loaded in these lanes as indicated. The numbers above the Hsp104 and Ssa1/2p blots represent densitometric quantitation of the bands on the [PSI+] blots. The values are normalized to actin levels (not shown) and relative to that of the wild-type basal sample set at 1.0. A similarly reduced level of Ssap abundance compared with wild-type cells was also observed for two other SSA1-21 strains using a nonspecific Hsp70 antibody (not shown).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A dominant mutant allele of SSA1 impairs mitotic transmission of [PSI⁺]:
In a screen to identify cellular factors involved in [PSI⁺] metabolism, pink colonies, indicative of an impaired ability of [PSI⁺] to confer nonsense suppression, were isolated from mutagenized [PSI⁺] cells (see MATERIALS AND METHODS). One dominant mutation was mapped to chromosome I within 4 cM of cdc15. Suspecting the mutation to be within SSA1, we isolated SSA1 from a mutant strain using PCR. Two independent PCR products conferred the dominant effects on suppression and [PSI⁺] stability (see below) when expressed from plasmids in a wild-type strain (data not shown).

Sequence analysis of the amplified genes revealed a base change resulting in a substitution of tryptophan for leucine at amino acid residue 483 (L483W). A wild-type allele of SSA1 modified to contain this single alteration conferred the dominant antisuppressor and [PSI⁺] loss phenotypes (see below) when expressed from a single-copy plasmid (pJ110) in [PSI⁺] cells (data not shown). For transformants that remained [PSI⁺], loss of the plasmid restored a normal [PSI⁺] phenotype. We have designated this allele SSA1-21. L483, within the peptide binding domain, is very highly conserved among Hsp70 homologs within and across species and is present in 9 of the other 12 Hsp70 homologs in S. cerevisiae, including all members of the SSA subfamily.

When restreaked onto YPD at 30°, cells from pink SSA1-21 colonies repeatedly gave rise to red colonies at frequencies of 2–12% (Fig 1). All cells from the red colonies remained red. For the wild-type strain, all cells from white colonies gave rise to only white colonies while those from red colonies gave rise to only red colonies. Dominant suppressor activity and guanidine curability (see MATERIALS AND METHODS) demonstrated that cells from white and pink colonies were [PSI⁺] and those from red colonies were [psi^-]. Thus, [PSI⁺] is considerably less stable in SSA1-21 cells compared with wild-type cells.

Despite the high frequency of [psi^-] cells in mature [PSI⁺] colonies, these colonies were uniformly pink rather than sectored. We reasoned that the actual rate of [PSI⁺] loss is low and that most [psi^-] cells appear late in the growth of the colony. We tested this by restreaking cells from colonies grown for 15–96 hr. No [psi^-] cells were found in colonies of <1 mm and thereafter [psi^-] cells appeared in proportion to colony size to a maximum of 10%. When grown in YPAD liquid medium at 30° from inocula grown in medium lacking adenine, >99% of SSA1-21 cells (strains 704-7C⁺, 707-3C⁺) remained [PSI⁺] after 19 generations of log-phase growth. Although the mitotic loss of [PSI⁺] from SSA1-21 cells was much higher than that for wild-type strains, it was low enough that without a direct selection for [psi^-] cells we could not adequately quantitate it.

An alternative explanation for the late appearance of [psi^-] cells in SSA1-21 colonies is that abnormal activity of Ssa1p, which is expressed at elevated levels as cells enter stationary phase (WERNER-WASHBURNE et al. 1989 ; WERNER-WASHBURNE and CRAIG 1989 ), leads to metabolic conditions incompatible with [PSI⁺] stability as colonies age. The appearance of [psi^-] cells was monitored in cultures of wild-type (628-3A⁺) and SSA1-21 (704-7C⁺) strains grown in liquid YPAD at 30° from a low starting density until 1 wk after cells stopped dividing. We saw no [psi^-] cells at any time in the wild-type culture and the proportion of [psi^-] cells in the SSA1-21 culture was essentially constant at 0.2% throughout the course of the experiment. Thus, sustained stationary phase conditions do not affect [PSI⁺] stability in our wild-type or SSA1-21 strains.

SSA1-21 impairs [PSI⁺]-mediated allosuppression:
The accumulation of red pigment by the SSA1-21 mutant at 30° correlated with inability to grow without adenine at this temperature (Fig 2). SSA1-21 strains were also pink and grew slower than wild type on medium lacking adenine at 23°. The uniform pink color of SSA1-21 colonies therefore reflects a weakened ability of [PSI⁺] to suppress ade2-1 rather than a mixture of red and white cells.

A strikingly similar weakened [PSI⁺] phenotype has been described among variants of [PSI⁺] induced by overexpression of Sup35p (DERKATCH et al. 1996 ). The weak phenotype was suggested to be due to propagation of a less stable structure of Sup35p polymer representing a new "strain" of [PSI⁺]. If SSA1-21 was inducing an alteration in the characteristic structure of [PSI⁺], then this new form should likewise be inheritable. However, restoration of the normal [PSI⁺] phenotype in wild-type strains after loss of a plasmid bearing SSA1-21 argues against this. Also, wild-type cytoduction recipients of [PSI⁺] from SSA1-21 donors displayed normal [PSI⁺] phenotypes, and although diploids heterozygous for SSA1-21 displayed a weakened [PSI⁺], all wild-type meiotic segregants had a normal [PSI⁺] phenotype (data not shown). Thus, SSA1-21 is affecting the stability of a characteristically normal form of [PSI⁺] rather than generating a new inheritable, unstable [PSI⁺].

SSA1-21 strains lacking Ssa2p are Psi-no-more:
Because SSA1 and SSA2 are nearly identical, have overlapping functions, and are both constitutively expressed, we analyzed the effects of combinations of mutations in these genes on [PSI⁺]. The mitotic stability of [PSI⁺] in related SSA deletion mutants (ssa1, 668-2D⁺; ssa2, 668-19B⁺; and ssa1 ssa2, 668-33A⁺) was assayed by monitoring the appearance of [psi^-] cells on YPD medium at temperatures ranging from 23° to 37° (data not shown). The presence of [PSI⁺] had no effect on growth of any of the strains at any temperature. We saw no indication of [PSI⁺] loss at any temperature for the wild-type or any of the deletion strains. Moreover, in six additional meiotic segregants of strain 668 for each combination of SSA deletion alleles, [PSI⁺] stability was normal under optimal growth conditions. Thus, the absence of Ssa1p, Ssa2p, or both has no effect on the mitotic stability of [PSI⁺] in our strains, and the Ssa1-21p defect is not a simple loss of function.

While [PSI⁺] is transmitted to all meiotic segregants of diploids homozygous for SSA1-21 or ssa2 (data not shown), it displayed irregular segregation among meiotic progeny of diploids heterozygous for both mutations. Among 38 tetrads from doubly heterozygous diploids, all of the 40 segregants inheriting both SSA1-21 and ssa2, and only these segregants, were [psi^-]. A Psi-no-more (YOUNG and COX 1971 ) effect of the double mutant was confirmed by cytoduction assays (Table 2). All SSA1-21 ssa2 haploid recipients of cytoplasm from [PSI⁺] strains remained nonsuppressed, while recipients of all other strains became [PSI⁺]. Many diploids formed by the cytoduction crosses with SSA1-21 ssa2 strains were also [psi^-]. We ruled out the possibility that [PSI⁺] was present but unable to cause suppression in SSA1-21 ssa2 cells with further crosses. Using the SSA1-21 ssa2 cytoductants as donors, all cytoduction recipients (628-1D) remained [psi^-] (Table 2). Also, all diploids made from crosses of 25 independent SSA1-21 ssa2 cytoductants with wild-type [psi^-] mating partners were [psi^-] (data not shown). Thus, Ssa2p is essential for [PSI⁺] propagation in the SSA1-21 background.

To determine if Ssa1p could functionally replace Ssa2p for [PSI⁺] propagation in an SSA1-21 strain, an ssa2 strain (668-19B⁺) was transformed with a plasmid carrying SSA1-21 (pRDW50). As controls, ssa1 and ssa1 ssa2 strains (668-2D⁺ and 668-33A⁺) were similarly transformed. As expected, the ssa1 ssa2 transformants became [psi^-] (data not shown). The ssa1 and ssa2 transformants both remained [PSI⁺] and displayed the SSA1-21 weakened allosuppression and increased mitotic loss of [PSI⁺] (data not shown). This shows that either Ssa1p or Ssa2p can support [PSI⁺] propagation in the presence of Ssa1-21p.

In ssa1 ssa2 strains, the derepression of functionally redundant Ssa3/4p (CRAIG 1992 ; Fig 4), presumably compensates for lack of Ssa1/2p in [PSI⁺] propagation. When these strains carry SSA1-21 on a plasmid, Ssa1-21p represses Ssa3/4p expression, essentially replacing all of the Ssa3/4p with Ssa1-21p. Since Ssap function is essential, it is impossible to assay [PSI⁺] stability in the absence of Ssap. These results, however, suggest that the presence of wild-type Ssap is necessary for [PSI⁺] propagation.

[PSI⁺] is similarly affected by heat stress in wild-type and SSA1-21 strains:
On rich medium (YPD), wild-type cells displayed an increase in accumulation of red pigment with increasing temperature. This correlated with a decrease in growth without adenine (Fig 2). SSA1-21 cells exhibited a very similar temperature-dependent weakening of [PSI⁺]-mediated effects but at lower temperature (Fig 2). Quantitated levels of suppression for this temperature effect using a translational readthrough assay correlated with the color and growth phenotypes (Table 3). Nonsense suppression was lower in the SSA1-21 mutant than in wild type at all temperatures, and for both strains it decreased with increasing temperature. Notably, the proportional decrease in suppression with increasing temperature was very similar for both strains. Thus, [PSI⁺]-mediated nonsense suppression was similarly reduced by elevated growth temperature in both wild-type and SSA1-21 cells.

When cells from [PSI⁺] colonies that were red when grown on YPD at 37° were regrown at 30°, no red [psi^-] colonies were found. As with wild type, when grown on YPD at 37°, SSA1-21 cells (from five SSA1-21 meiotic segregants of strain 707) gave rise to the same ratio of [psi^-]/[PSI⁺] colonies as when grown at 30°. Thus, for both wild-type and SSA1-21 strains, heat stress reduced [PSI⁺]-mediated nonsense suppression but did not affect [PSI⁺] mitotic stability.

Solubility of Sup35p in [PSI⁺] strains is increased at elevated growth temperature:
The ratio of soluble to insoluble Sup35p in both wild-type and SSA1-21 strains increased with increasing temperature in parallel with the suppressor phenotypes (Fig 3). Therefore, the weakened [PSI⁺]-mediated allosuppression caused by heat stress directly correlated with an increase in the relative amount of soluble Sup35p in [PSI⁺] cells.

The SSA1-21 effects are not due to Hsp104 overabundance:
Since the level of Hsp104 expression under optimal growth conditions critically influences [PSI⁺] metabolism (CHERNOFF et al. 1995 ; PATINO et al. 1996 ; PAUSHKIN et al. 1996 ; NEWNAM et al. 1999 ), we compared the levels of constitutive expression of Hsp104 in the wild-type and SSA mutant strains. In the [PSI⁺] strains, we saw only a modest increase in the basal level of Hsp104 in SSA1-21 cells compared with wild-type cells (Fig 4), suggesting that the SSA1-21 effects were not due to derepression of Hsp104 expression.

We also examined the effects of elevated expression of Hsp104 (data not shown). Hsp104, under control of its own promoter, was expressed from a single-copy plasmid (pH219) in wild-type (628-3A⁺) and SSA1-21 (707-3C⁺) strains. There was no effect of the control plasmid (pRS314) on the [PSI⁺] phenotype of either strain. Wild-type cells carrying pH219 were pinker than pRS314 transformants on indicator medium, grew more slowly without adenine, and showed elevated loss of [PSI⁺] (2–3% of cells). SSA1-21 transformants carrying pH219 also accumulated more pigment, grew more slowly without adenine, and displayed elevated loss of [PSI⁺] (30–35% of cells). Therefore, as with elevated temperature, [PSI⁺] in SSA1-21 and wild-type strains was similarly affected by elevated levels of Hsp104.

SSA1-21 does not affect thermotolerance:
Overproduction of Ssa1p reduces basal thermotolerance conferred by basal expression of Hsp104 (NEWNAM et al. 1999 ). This led to the suggestion that the ability of overexpressed Ssa1p to interfere with the Sup35p solubilizing effects of overexpressed Hsp104 in [PSI⁺] strains at optimal growth temperature is conferred by an activity that interferes with Hsp104 (NEWNAM et al. 1999 ). The effects of SSA1-21 on [PSI⁺] may be caused by a defect in this putative Hsp104-inhibitory activity.

We measured basal thermotolerance in wild-type and SSA1-21 strains as the ability to survive a shift in temperature from 23° to 52° (NICOLET and CRAIG 1991B ). Basal thermotolerance of five SSA1-21 strains (meiotic segregants of strain 707) was not significantly different from that of wild type (data not shown). When pretreated at 39° for 50 min to induce Hsp104 and the heat-shock response, wild-type and SSA1-21 strains displayed similar levels of acquired tolerance to a subsequent shift to 52° (data not shown). This indicates that the constitutive and inducible activity of Hsp104 that confers thermotolerance was neither stimulated nor inhibited in the SSA1-21 mutant strains.

[PSI⁺] is partially cured from SSA1-21 strains by rapid induction of the heat-shock response:
We observed an elevated frequency of [psi^-] cells in the SSA1-21 cultures that underwent the 39° pretreatment. Closer examination of this effect revealed that a rapid shift from 23° to 39° induced a partial loss of [PSI⁺] from SSA1-21 cultures after 10 min, an effect that diminished after 1 hr (Table 4). Many SSA1-21 heat-shocked cells gave rise to colonies containing numerous deep sectors, reflecting loss of [PSI⁺] from cells in the early growth of these colonies. For wild-type and SSA deletion strains, [PSI⁺] stability was unaffected by this type of heat shock. Thus, metabolic conditions in SSA1-21 cells become transiently incompatible with [PSI⁺] propagation upon a rapid temperature upshift.

SSA1-21 strains are hypersensitive to curing by guanidine-hydrochloride:
By an unknown biochemical mechanism, the presence of millimolar concentrations of Gdn-HCl in the growth medium causes efficient curing of [PSI⁺] (TUITE et al. 1981 ; COX et al. 1988 ; EAGLESTONE et al. 2000 ). We compared susceptibility of [PSI⁺] in wild-type (628-3A⁺) and SSA1-21 (704-7C⁺) strains to guanidine curing using concentrations of 1 mM and 3 mM Gdn-HCl (Fig 5). The generation times of both wild-type and SSA1-21 strains, with and without the Gdn-HCl, were 112 ± 5 min. After growing for 22 generations in the absence of Gdn-HCl, no [psi^-] cells were detected in the wild-type culture and <1% of cells in the SSA1-21 culture were [psi^-]. When grown in 1 mM Gdn-HCl, [PSI⁺] loss began immediately from cells of the SSA1-21 culture but was stable in wild-type cells throughout the duration of the experiment. Thus, the SSA1-21 mutant was hypersensitive to the [PSI⁺]-curative effect of Gdn-HCl. Both strains grown in 3 mM Gdn-HCl lost [PSI⁺] at similar rates. However, while the wild-type cells divided several times before [psi^-] cells began to appear, [PSI⁺] loss began immediately in the SSA1-21 population. This indicates that SSA1-21 cells have fewer inheritable [PSI⁺] particles per cell than wild-type cells (see DISCUSSION).

View larger version (15K): In this window In a new window Download PPT slide   Figure 5. Guanidine curing of [PSI+]. [PSI+] cultures of wild-type (628-3A+, circles) or SSA1-21 (704-7C+, squares) strains grown at 30° were maintained at OD600 < 1 in YPAD containing 1 mM (open symbols) or 3 mM (solid symbols) guanidine-hydrochloride. Aliquots from the cultures were assayed at various times to determine the percentage of [PSI+] cells remaining as a function of cell doublings. Error bars indicate the difference in results of two independent experiments.

Ssa1-21p retains activity in other cellular functions:
The highly conserved nature of the L483 residue suggests that it is functionally important. We looked for a phenotypic effect of the L483W mutation in an attempt to gain insight into this function. Virtually all of the Ssap in SSA1-21 ssa2 and Ssa1⁺ ssa2 strains is Ssa1-21p or Ssa1p, respectively (Fig 4). Growth of these strains to colonies on various fermentable and nonfermentable carbon sources at temperatures ranging from 15° to 37° was indistinguishable. This indicates that Ssa1-21p retains a significant amount of Ssa1p function.

We measured the ability of Ssa1-21p to regulate its own expression using a transcriptional fusion assay (Table 5). SSA1 promoter activity in the [psi^-] and [PSI⁺] wild-type strains was 33 and 41% higher (on average) than in the corresponding SSA1-21 strains. Thus, the negative regulatory activity of Ssa1-21p is not only functional but also modestly hyperactive. Consistent with this, the level of Ssa1/2p in SSA1-21 strains was modestly and reproducibly reduced compared with that of the wild-type strain (Fig 4).

We also analyzed the effect of SSA1-21 on microtubule metabolism. An SSA1 mutant, impaired in its ability to interact with its cochaperone Ydj1p, is defective in microtubule function, causing hypersensitivity to the microtubule depolymerizing drugs thiabendazole and benomyl (OKA et al. 1998 ). We found that SSA1-21 mutants were no more sensitive to the effects of these drugs than wild-type strains (data not shown). This may indicate that the ability of Ssa1-21p to interact with Ydj1p is unaltered.

[PSI⁺] induces expression of heat-shock proteins:
In wild-type strains, [PSI⁺] elevated the basal expression of Hsp104 and increased the levels of heat-induced expression of both Hsp104 and Ssa3/4p (Fig 4). In addition, [PSI⁺] elevated SSA1 promoter activity in all of the strains we examined (on average; Table 5). These results indicate that [PSI⁺] causes a stress to cells. In the SSA1-21 strain [PSI⁺] had very little effect on Hsp104 expression (Fig 4). This was predictable since the phenotypic effects of [PSI⁺] in this mutant were much weaker than in the wild-type strain.

With the caveat that the strains are not isogenic, expression of Hsp104 was elevated in all of the ssa mutant strains (Fig 4). In the [psi^-] SSA1-21 mutant, the modest induction of Hsp104 may reflect an accumulation of protein aggregates due to impairment in the chaperone function of Ssa1-21p or to a reduced Hsp70 abundance caused by overactive Ssa1-21p regulatory function. The derepression of Hsp104 expression in the ssa1 and ssa2 strains may indicate that loss of specific chaperone function of either of these Hsp70s similarly triggers Hsp104 induction, suggesting that these functionally redundant Ssaps may possess distinct activities.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have demonstrated that intact Hsp70 function is required for propagation of the yeast [PSI⁺] element. While strains with deletion of SSA1, SSA2, or both propagate [PSI⁺] normally, substitution of a highly conserved residue in Ssa1p (L483W) dramatically impairs [PSI⁺] metabolism. [PSI⁺] cannot propagate in strains expressing Ssa1-21p unless another wild-type, functionally redundant Hsp70 is also expressed. The presence of Ssa1-21p reduces the number of inheritable Sup35p polymers normally found in [PSI⁺] cells, and this is manifested as a weakened, mitotically unstable [PSI⁺]. In wild-type and SSA1-21 cells [PSI⁺] is similarly sensitive to the effects of thermal stress and elevated expression of Hsp104.

Ample evidence supports the notion that Hsp104 plays a primary role in regulating the solubility of Sup35p in [PSI⁺] strains (CHERNOFF et al. 1995 ; PATINO et al. 1996 ; PAUSHKIN et al. 1996 ; EAGLESTONE et al. 1999 ; NEWNAM et al. 1999 ). Hsp104 has been suggested to function in [PSI⁺] propagation by assisting the formation of Sup35p aggregates through generation of transient-folding intermediates (PATINO et al. 1996 ) or by splitting up Sup35p polymers, which increases the number of inheritable [PSI⁺] seeds (PAUSHKIN et al. 1996 ). Our data suggest that the SSA1-21 effects on [PSI⁺] propagation are not mediated through an effect on Hsp104, indicating that Hsp70 plays a major role in [PSI⁺] metabolism. We propose that Ssa1-21p lacks a specific Hsp70 function that facilitates the generation of inheritable [PSI⁺] particles from preexisting material and can interfere with a similar activity possessed by homologous Ssaps.

Effects of heat-shock factors on [PSI⁺] metabolism:
The inability of SSA1-21 ssa2 strains to propagate [PSI⁺] supports a model that a specific nonessential Hsp70 activity is necessary for [PSI⁺] propagation. The incomplete complementation by Ssa1p or Ssa2p suggests that Ssa1-21p also interferes with this activity of these proteins. The dominant effect of SSA1-21 may be explained by enhanced repression of basal expression of other Hsps required for [PSI⁺] propagation, but we favor the model that inhibition of other Ssaps occurs through direct physical interaction. Evidence exists that Ssa1p acts as a multimer (NICOLET and CRAIG 1991A ), and the near identity with Ssa2p argues that Ssa1p also physically interacts with Ssa2p. Additionally, overexpressing Ssa1p represses basal Hsp expression (STONE and CRAIG 1990 ) but has the opposite effect on [PSI⁺] stability as does the presence of Ssa1-21p (NEWNAM et al. 1999 ). Moreover, we have found that the L483W mutation in Ssa2p (SSA2-21) causes the same weak suppression and [PSI⁺] mitotic instability as SSA1-21 (G. JUNG and D. C. MASISON, unpublished observations).

This model may also explain why some diploids formed in crosses with SSA1-21 ssa2 mutants become [psi^-]. Under optimal growth conditions only 20% of the Ssap in the cell is Ssa1p (CRAIG 1992 ). Nonetheless, Ssa1-21p, which would be expected to be even less abundant because of the reduced activity of its promoter in SSA1-21 cells, significantly weakens [PSI⁺]. The very high relative abundance of Ssa1-21p in SSA1-21 ssa2 cells may sufficiently interfere with propagation of the [PSI⁺] introduced upon mating that some of the resulting diploids lose [PSI⁺].

Alternatively, the L483W mutation may confer a gain of function that negatively affects [PSI⁺] propagation and that is partially inhibited by wild-type Ssap. Such a new function could allow Ssa1p to solubilize Sup35p polymers or affect other factors leading to this effect.

In wild-type and SSA1-21 strains, elevated growth temperature, which increased Sup35p solubility, had the same effect of reducing [PSI⁺] "strength" without altering [PSI⁺] stability. A weakening of [PSI⁺] by stressful growth conditions that do not cure [PSI⁺] was previously reported by others, who postulate that Hsp104 is the primary factor regulating solubility of Sup35p in [PSI⁺] strains in response to environmental stress (EAGLESTONE et al. 1999 ). This suggests that during stress [PSI⁺] is similarly affected by the solubilizing activity of Hsp104 in both wild-type and SSA1-21 strains, an interpretation supported by the effects of an extra copy of HSP104 in these strains. Consequently, the instability of [PSI⁺] in the SSA1-21 strain suggests that the reduced allosuppression in this mutant at optimal growth temperature is due to a different mechanism from that caused by heat stress. Results of the guanidine-curing experiment suggest that heat stress reduces only the size of inheritable Sup35p aggregates, and in contrast the SSA1-21 mutation significantly reduces the number of these aggregates.

Guanidine does not induce an active process that eliminates [PSI⁺], but somehow blocks its propagation. Curing then results from subsequent dilution of inheritable [PSI⁺] seeds among daughter cells as cells divide (EAGLESTONE et al. 2000 ). Wild-type cells have an average of 60 seeds (MCCREADY et al. 1977 ; EAGLESTONE et al. 2000 ), assuming that one is enough to regenerate the [PSI⁺] state. This starting number of seeds must therefore be diluted through several cell divisions before [psi^-] (i.e., seed-free) cells arise. When guanidine curing is done under conditions of stress that reduce [PSI⁺]-mediated allosuppression, the number of cell divisions spanned by this lag (four to five) does not change (EAGLESTONE et al. 1999 , EAGLESTONE et al. 2000 ). This implies that the elevated solubility of Sup35p in wild-type cells grown under stress results from a reduction in the size but not the number of starting seeds. This is consistent with the inefficiency by which stress alone cures [PSI⁺]. If Hsp104 is regulating the solubility of Sup35p in environmentally stressed cells, then this indicates that Hsp104 acts by progressively reducing the size of Sup35p polymers rather than by splitting them into more numerous inheritable seeds.

In the presence of guanidine, our wild-type strain divided four to five times before [psi^-] cells appeared but there was no lag in the appearance of [psi^-] cells for the SSA1-21 strain. This indicates that there are 16-fold fewer [PSI⁺] seeds in the SSA1-21 cells. The weak state of [PSI⁺] in SSA1-21 strains is thus due to a reduction in number of inheritable particles, which is consistent with the high mitotic loss of [PSI⁺] from SSA1-21 cells. We agree with the supposition that formation of new inheritable [PSI⁺] seeds arises through disruption of preexisting Sup35p polymers and propose that this process is mediated by Hsp70 and impaired by the L483W mutation. A defect in spontaneous nucleation of inheritable seeds, which occurs very rarely in wild-type cells, would essentially be undetectable.

Despite this large difference in number of seeds, the difference in allosuppression between SSA1-21 and wild-type cells at 30° is less than threefold. Therefore, the size of these particles in SSA1-21 cells at optimal temperature may actually be larger than those in wild-type cells. If so, then the function of cytosolic Hsp70 in [PSI⁺] propagation may be to break long Sup35p polymers or to prevent bundling or branching of the fibers.

Although the Hsp70 defect reduces the number of seeds, there is no significant increase in [PSI⁺] loss from SSA1-21 strains by growth at elevated temperature. This argues against a specific role for Ssa1p in protecting [PSI⁺] under stressful growth conditions and suggests that [PSI⁺] is simply evading complete dissolution. One way that this may occur is that while remaining capable of regenerating [PSI⁺], seeds are reduced to a size that is not efficiently recognized as a substrate for Hsp104. At optimal growth temperature the scarcity of other substrates allows overexpressed Hsp104 to solubilize these seeds.

Effects of heat-shock factors on Hsp expression:
Our results indicate that Ssa1-21p is modestly hyperactive in its negative regulatory role. One effect of the L483W mutation, which lies within the peptide-binding domain, may be a reduced affinity for substrates. This may explain both the elevated level of Hsp104 in the SSA1-21 [psi^-] strains and the hyperactive regulatory activity of Ssa1-21p. Reduced ability to bind substrates should lead to accumulation of protein aggregates, stimulating induction of Hsp104, and to an elevated effective concentration of substrate-free Ssa1-21p, which acts in repressing gene expression (CRAIG and GROSS 1991 ).

All of the SSA deletion strains had a high basal level of Hsp104 at optimal growth temperature. Despite this, [PSI⁺] stability was unaffected in all of these mutants. Since the SSA deletions were crossed into our strains from an unrelated background, interpreting the results of experiments with these strains is complicated by the possibility that strain differences are affecting Hsp expression. However, any contribution to the variation in the observed levels of Hsp104 and Hsp70 that is due to strain differences is irrelevant with regard to [PSI⁺] since [PSI⁺] was not affected by any combination of SSA deletions. Furthermore, [PSI⁺] was most affected in SSA1-21 cells that have a much lower level of Hsp104. These results indicate that the abundance of Hsp104 is not necessarily an accurate parameter for predicting the solubility of Sup35p in [PSI⁺] strains and support our suggestion that cytosolic Hsp70 is playing a significant role in regulating propagation of the aggregated form of Sup35p in [PSI⁺] strains.

Repeating the caveat that the strains are not isogenic, although there are similar levels of total Ssa1/2p in wild-type, ssa1, and ssa2 strains, the high basal levels of Hsp104 in both mutant strains might indicate that simultaneous expression of both proteins is required for full cytosolic Hsp70 function. If so, then despite their near identity this would reflect a functional distinction between Ssa1p and Ssa2p, suggesting that their roles in the cell do not completely overlap.

Effects of [PSI⁺] on heat-shock metabolism:
The presence of [PSI⁺] caused a mild stress in our wild-type strain. This could be due to the accumulation of translation readthrough products having C-terminal extensions that are unable to fold properly, the presence of the Sup35p aggregates, or both. Translational misreading caused by paromomycin induces a stress response that confers thermotolerance (GRANT et al. 1989 ). If such misreading is the cause of stress in our [PSI⁺] cells, then the weak tRNA suppressor SUQ5, which is present in all of our strains, may be required in addition to [PSI⁺] to observe these effects. With mammalian prions, the presence of protein aggregates alone seems to affect the stress response (TATZELT et al. 1995 ).

[PSI⁺] also enhanced induction of Ssa3/4p in wild-type and ssa1 strains at elevated temperature. A previous study found that exposure to nonlethal elevated temperature causes some [PSI⁺] strains to become more tolerant to extended lethal heat treatment than isogenic [psi^-] variants (EAGLESTONE et al. 1999 ). This led to the suggestion that under some conditions [PSI⁺] provides a benefit. In those strains, Ssa3/4p expression was not analyzed and Hsp104 abundance was reported to be unaltered by the presence of [PSI⁺]. Our results may provide a physiological explanation for the observed thermotolerance and we favor the notion that [PSI⁺] is merely a pathologic condition that presents a stress.

We show that cytosolic Hsp70 can significantly affect [PSI⁺] metabolism and demonstrate the dependence of a particular aspect of yeast prion propagation on a specific cellular activity: efficient generation of inheritable [PSI⁺] particles from preexisting material requires intact Hsp70 function. The biochemical mechanism underlying this requirement remains to be established. The very high conservation of the L483 residue signifies its evolutionary importance. However, aside from its effects on [PSI⁺], we have not yet identified a metabolic defect in the SSA1-21 mutant that distinguishes it from wild type. It is possible that Ssa1-21p affects the expression or activity of factors other than Hsp70 required by [PSI⁺], and it is feasible that a yet-uncharacterized activity of Ssa1p remains to be discovered.

	FOOTNOTES

¹ Present address: School of Biology, Georgia Institute of Technology, Atlanta, GA 30332-0230.

	ACKNOWLEDGMENTS

We thank Elizabeth Craig (Madison, WI) for generously providing plasmids, strains, and antibodies; David Bedwell (Birmingham, AL) and Herman Edskes (National Institutes of Health, Bethesda, MD) for plasmids; and Susan Lindquist (Chicago, IL) and Mick Tuite (Canterbury, UK) for antibodies.

Manuscript received April 11, 2000; Accepted for publication June 16, 2000.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

BONETTI, B., L. FU, J. MOON, and D. M. BEDWELL, 1995  The efficiency of translation termination is determined by a synergistic interplay between upstream and downstream sequences in Saccharomyces cerevisiae.. J. Mol. Biol. 251:334-345[Medline].

BOORSTEIN, W. R. and E. A. CRAIG, 1990  Structure and regulation of the SSA4 HSP70 gene of Saccharomyces cerevisiae.. J. Biol. Chem. 265:18912-18921[Abstract/Free Full Text].

BOORSTEIN, W. R., T. ZIEGELHOFFER, and E. A. CRAIG, 1994  Molecular evolution of the HSP70 multigene family. J. Mol. Evol. 38:1-17[Medline].

CHERNOFF, Y. O., S. L. LINDQUIST, B. ONO, S. G. INGE-VECHTOMOV, and S. W. LIEBMAN, 1995  Role of the chaperone protein Hsp104 in propagation of the yeast prion-like factor. Science 268:880-884. [psi+][Abstract/Free Full Text].

CONDE, J. and G. R. FINK, 1976  A mutant of Saccharomyces cerevisiae defective for nuclear fusion. Proc. Natl. Acad. Sci. USA 73:3651-3655[Abstract/Free Full Text].

COX, B. S., 1965  "" a cytoplasmic suppressor of super-suppressor in yeast. Heredity 20:505-521.

COX, B. S., M. F. TUITE, and C. S. MCLAUGHLIN, 1988  The psi factor of yeast: a problem in inheritance. Yeast 4:159-178[Medline].

CRAIG, E. A., 1992 The heat-shock response of Saccharomyces cerevisiae., pp. 501–573 in The Molecular and Cellular Biology of the Yeast Saccharomyces, edited by E. W. JONES, J. R. PRINGLE and J. R. BROACH. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

CRAIG, E. A. and C. A. GROSS, 1991  Is hsp70 the cellular thermometer? Trends Biochem. Sci. 16:135-140[Medline].

DEPACE, A. H., A. SANTOSO, P. HILLNER, and J. S. WEISSMAN, 1998  A critical role for amino-terminal glutamine/asparagine repeats in the formation and propagation of a yeast prion. Cell 93:1241-1252[Medline].

DERKATCH, I. L., Y. O. CHERNOFF, V. V. KUSHNIROV, S. G. INGE-VECHTOMOV, and S. W. LIEBMAN, 1996  Genesis and variability of [PSI] prion factors in Saccharomyces cerevisiae.. Genetics 144:1375-1386[Abstract].

EAGLESTONE, S. S., B. S. COX, and M. F. TUITE, 1999  Translation termination efficiency can be regulated in Saccharomyces cerevisiae by environmental stress through a prion-mediated mechanism. EMBO J. 18:1974-1981[Medline].

EAGLESTONE, S. S., L. W. RUDDOCK, B. S. COX, and M. F. TUITE, 2000  Guanidine hydrochloride blocks a critical step in the propagation of the prion-like determinant [PSI⁺] of Saccharomyces cerevisiae.. Proc. Natl. Acad. Sci. USA 97:240-244[Abstract/Free Full Text].

GLOVER, J. R., A. S. KOWAL, E. C. SCHIRMER, M. M. PATINO, and J. J. LIU et al., 1997  Self-seeded fibers formed by Sup35, the protein determinant of [PSI+], a heritable prion-like factor of S. cerevisiae.. Cell 89:811-819[Medline].

GRANT, C. M., M. FIROOZAN, and M. F. TUITE, 1989  Mistranslation induces the heat-shock response in the yeast Saccharomyces cerevisiae. Mol. Microbiol. 3:215-220[Medline].

GUARENTE, L., 1983  Yeast promoters and lacZ fusions designed to study expression of cloned genes in yeast. Methods Enzymol. 101:181-191[Medline].

INGOLIA, T. D., M. R. SLATER, and E. A. CRAIG, 1982  Saccharomyces cerevisiae contains a complex multigene family related to the major heat shock-inducible gene of Drosophila. Mol. Cell. Biol. 2:1388-1398[Abstract/Free Full Text].

KING, C. Y., P. TITTMANN, H. GROSS, R. GEBERT, and M. AEBI et al., 1997  Prion-inducing domain 2-114 of yeast Sup35 protein transforms in vitro into amyloid-like filaments. Proc. Natl. Acad. Sci. USA 94:6618-6622[Abstract/Free Full Text].

LANSBURY, P. T., JR. and B. CAUGHEY, 1995  The chemistry of scrapie infection: implications of the `ice 9' metaphor. Chem. Biol. 2:1-5[Medline].

LAWRENCE, C. W., 1994  Classical mutagenesis techniques. Methods Enzymol. 194:273-281.

LIEBMAN, S. W., J. W. STEWART, and F. SHERMAN, 1975  Serine substitutions caused by an ochre suppressor in yeast. J. Mol. Biol. 94:595-610[Medline].

MASISON, D. C., M. L. MADDELEIN, and R. B. WICKNER, 1997  The prion model for [URE3] of yeast: spontaneous generation and requirements for propagation. Proc. Natl. Acad. Sci. USA 94:12503-12508[Abstract/Free Full Text].

MCCREADY, S. J., B. S. COX, and C. S. MCLAUGHLIN, 1977  The extrachromosomal control of nonsense suppression in yeast: an analysis of the elimination of [psi+] in the presence of a nuclear gene PNM.. Mol. Gen. Genet. 150:265-270[Medline].

MILLER, J. F., 1972 Experiments in Molecular Genetics. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

NEWNAM, G. P., R. D. WEGRZYN, S. L. LINDQUIST, and Y. O. CHERNOFF, 1999  Antagonistic interactions between yeast chaperones Hsp104 and Hsp70 in prion curing. Mol. Cell. Biol. 19:1325-1333[Abstract/Free Full Text].

NICOLET, C. M. and E. A. CRAIG, 1991a  Functional analysis of a conserved amino-terminal region of HSP70 by site-directed mutagenesis. Yeast 7:699-716[Medline].

NICOLET, C. M. and E. A. CRAIG, 1991b  Inducing and assaying heat-shock response in Saccharomyces cerevisiae.. Methods Enzymol. 194:710-717[Medline].

OKA, M., M. NAKAI, T. ENDO, C. R. LIM, and Y. KIMATA et al., 1998  Loss of Hsp70-Hsp40 chaperone activity causes abnormal nuclear distribution and aberrant microtubule formation in M-phase of Saccharomyces cerevisiae.. J. Biol. Chem. 273:29727-29737[Abstract/Free Full Text].

PARSELL, D. A., A. S. KOWAL, M. A. SINGER, and S. LINDQUIST, 1994  Protein disaggregation mediated by heat-shock protein Hsp104. Nature 372:475-479[Medline].

PATINO, M. M., J.-J. LIU, J. R. GLOVER, and S. LINDQUIST, 1996  Support for the prion hypothesis for inheritance of a phenotypic trait in yeast. Science 273:622-626[Abstract].

PAUSHKIN, S. V., V. V. KUSHNIROV, V. N. SMIRNOV, and M. D. TER-AVANESYAN, 1996  Propagation of the yeast prion-like [psi+] determinant is mediated by oligomerization of the SUP35-encoded polypeptide chain release factor. EMBO J. 15:3127-3134[Medline].

PAUSHKIN, S. V., V. V. KUSHNIROV, V. N. SMIRNOV, and M. D. TER-AVANESYAN, 1997  In vitro propagation of the prion-like state of yeast Sup35 protein. Science 277:381-383. [see comm