Latest on yaest Psi fibrils
May 29, 1997 University of Chicago Medical Center press releaseResearchers at the University of Chicago's Howard Hughes Medical Institute have found that a protein molecule able to transmit a genetic trait without DNA or RNA in yeast is able to string itself together into long fibers much like those found in the brain in "mad cow" and human Creutzfeldt-Jakob diseases.
The finding is reported in the May 30 issue of the journal Cell. Scientists have suspected that in the neurodegenerative diseases of mammals such as sheep scrapie, mad cow disease (or bovine spongiform encephalopathy) and the kuru disease of the Papua New Guinea tribes, a normal protein in the brain can somehow become twisted and then corrupt other, healthy molecules of the same protein to do likewise a process much like the seeding of a crystal. The improperly folded protein molecules seem to spin themselves together into fibers, which grow as other molecules are recruited.
The infectious protein particles are called prions, and their existence has been hotly debated for 30 years, since researchers showed that diseased brain tissue remained infectious even after treatment with radiation that would have destroyed any DNA or RNA.
Last year the Chicago team led by Susan Lindquist, Ph.D., professor of molecular genetics and cell biology, showed that prion-like proteins exist in yeast. In the mammalian brain, whose cells do not divide, prions pass between cells and function as infectious agents; in yeast, they produce heritable changes in metabolism from one generation to the next as the cells divide. The change is easy to see, because in one case the cells are red and in the other white.
"That a genetic property carried by protein shape can be responsible for inheritance from generation to generation or for an infection is a revolutionary concept," Lindquist said.
Lindquist's group focused on a yeast protein called sup35, part of the normal yeast machinery for making all the other proteins in the cell. In certain strains which appear to have identical DNA to normal strains the sup35 protein doesn't work. They showed that the defective trait can be propagated by this faulty protein, without any DNA or RNA serving as the genetic blueprint. They now show that even in the test tube, the purified yeast protein can knit together into fibers that have the same staining properties and molecular architecture as the amyloid plaques seen at autopsy in the brains of animals and humans that have died of transmissible spongiform encephalopathies. They also show that the formation of fibers from normal protein molecules is greatly speeded up by the presence of defective ones.
"Instead of a vague conceptual model for this new type of inheritance, we now have a detailed molecular mechanism for this mysterious process," Lindquist says, "and this seems to be closely related to the mechanism behind these devastating neurodegenerative diseases."
Although the yeast sup35 protein and the mammalian prion protein are not at all related to each other the yeast pose no risk to consumers of bread or beer the researchers think that in-depth analysis of the yeast prion-like elements and other proteins that help them fold up may lead to new approaches to therapies for neurodegenerative diseases. "From the molecular standpoint, this looks like the changes you get in the mammalian prion," said research associate John Glover, Ph.D., lead author on the Cell paper. "This gives us a clear structural basis for understanding how these things behave in the cell," he said. The researchers said it is much cheaper and easier to study genetic mechanisms in yeast than in animals.
Lindquist said that the ability of certain proteins to confer heritable properties by changing their shape may underlie other unexplained genetic phenomena. A similar protein misfolding that is not infectious seems to cause Alzheimer's disease.
Other authors on the Cell paper include electron microscopist Anthony Kowal, graduate students Eric Schirmer and Jia-Jia Liu, and research associate Maria Patino, Ph.D. The research was funded by the Howard Hughes Medical Institute and the National Institutes of Health.
Cell, Vol. 89, 811-819, May 30, 1997 John R. Glover, Anthony S. Kowal, Eric C. Schirmer, Maria M. Patino, Jia-Jia Liu, Susan LindquistThe Psi+ factor of S. cerevisiae represents a new form of inheritance: cytosolic transmission of an altered phenotype is apparently based upon inheritance of an altered protein structure rather than an altered nucleic acid. The molecular basis of its propagation is unknown. We report that purified Sup35 and subdomains that induce Psi+ elements in vivo form highly ordered fibers in vitro. Fibers bind Congo red and are rich in þ sheet, characteristics of amyloids found in certain human diseases, including the prion diseases. Some fibers have distinct structures and these, once initiated, are self-perpetuating. Preformed fibers greatly accelerate fiber formation by unpolymerized protein. These data support a "protein-only" seeded polymerization model for the inheritance of Psi+.
Psi+, a genetic element found in the budding yeast Saccharomyces cerevisiae, is inherited in a dominant, cytoplasmic fashion (hence the brackets and capital letters in its name). Psi+ was discovered 30 years ago (Cox, 1965 ), but only recently has its remarkable molecular nature been uncovered. Several lines of evidence suggest that Psi+ consists solely of protein. This nuclear encoded protein, produces a heritable change in phenotype by switching to an altered and self-perpetuating structural form (Wickner, 1994 ; Lindquist, 1997 ).
As revolutionary as this hypothesis may be for the inheritance of a genetic trait, yet another cytoplasmically inherited genetic element in yeast, Ure 3, appears to propagate by a similar mechanism (Wickner, 1994 ; Lindquist, 1997 ). Moreover, there is reason to believe that the behavior of these unconventional genetic elements points to a molecular process that is broadly distributed and functions in a wide variety of biological contexts. Indeed, Psi+ and Ure 3 have been called "yeast prions" because of the striking similarity between their proposed mechanism of propagation and that of the mammalian prion diseases (transmissible spongiform encephalopathies; Prusiner, 1994 ; Horwich and Weissman, 1997 ).
The infectious agent in the transmissible spongiform encephalopathies produces rapidly progressing, inexorably fatal neurodegeneration after a long latent period (see Prusiner, 1994 ; Caughey and Chesebro, 1997 ; Horwich and Weissman, 1997 ). This agent, the prion, appears to consist solely of a plasma-membrane protein, PrPC, that has acquired an altered, pathological conformation, PrPSc. PrPSc molecules are thought to interact with PrPC molecules and to induce the latter to switch conformation, producing a chain reaction that kills neurons and generates new infectious protein. Thus, with the mammalian prions an infectious disease propagates through a self-perpetuating change in the structure of a normal cellular protein, apparently without an accompanying nucleic acid vector. With the yeast prions, a similar process produces a heritable new metabolic state, apparently without an accompanying change in any nucleic acid.
The phenotype conferred by the yeast Psi+ factor is an increase in the suppression of nonsense mutations, brought about by an increased tendency of ribosomes to read through stop codons (Cox et al., 1988 ). Sup35, the protein determinant of Psi+, is an essential subunit of the translation termination factor (Stansfield et al., 1995 ; Zhouravleva et al., 1995 ). In normal yeast strains, most Sup35 protein is soluble and functions in termination in a complex with its partner protein Sup45. In Psi+ strains, most Sup35 is insoluble, and this insolubility is inherited from generation to generation (Patino et al., 1996 ; Paushkin et al., 1996 ).
The reduced concentration of functional termination factor is believed to be responsible for the occasional failure of ribosomes to terminate at nonsense codons in Psi+ strains. Mutations in Sup35 that impair its translation-termination function produce similar suppressor phenotypes, but these are recessive and inherited in a simple Mendelian fashion (Cox et al., 1988 ; Stansfield and Tuite, 1994 ). That Psi+ is due to a change in the conformation of Sup35 rather than a mutation is supported by the observation that Hsp104, a protein whose only known function is to alter the conformational state of other proteins (Parsell et al., 1994 ), plays a determining role in Psi+ inheritance (Chernoff et al., 1995 ; Patino et al., 1996 ).
Sup35 is composed of three distinct sequence elements (Kushnirov et al., 1990 ; Ter-Avanesyan et al., 1993 ; Paushkin et al., 1997 ). The C-terminal region contains highly conserved GTP-binding consensus sites, binds Sup45, and provides the essential termination function of Sup35. The N-terminal and middle regions are less conserved and are demarcated by their unusual amino acid compositions. The middle region is highly charged and its function is unclear, although it, too, appears to interact with Sup45 (Paushkin et al., 1997 ). The N-terminal region is extremely rich in glutamine and asparagine (51 of 123 amino acids) and contains several imperfect repeats of the nonapeptide PQGGYQQYN. This region is resistant to proteolytic digestion in Psi+ cells (Paushkin et al., 1997 ) and plays a critical role in Psi+ metabolism. First, it is required for the maintenance of Psi+ (Doel et al., 1994 ; Ter-Avanesyan et al., 1994 ). Deletion of this segment has no effect on Psi- cells, but causes Psi+ cells to revert to Psi-.
Second, and more remarkably, transient overexpression of this segment induces at high frequency the formation of new, heritable Psi+ elements in Psi- strains.
The protein-only hypothesis for the inheritance of Psi+ predicts that the alternative, insoluble form of Sup35 in Psi+ cells facilitates conversion of newly synthesized Sup35 to the same state. This process is thought to produce a dominant and cytoplasmically inherited reduction in the efficiency of translation termination. Recently, we tested this hypothesis by comparing the fate of newly synthesized protein in Psi+ and Psi- cells (Patino et al., 1996 ).
The critical N-terminal region of Sup35, together with the middle region, was fused to green fluorescent protein (GFP), allowing its distribution to be monitored in real time in living cells. As soon as the protein had accumulated at levels high enough to be visualized, it coalesced into a small number of intense foci in Psi+ cells, but remained freely distributed in Psi- cells. Additional experiments established that this coalescence was a true marker of the Psi+ state. For example, when cells were cured of [PSI+] by a variety of genetic manipulations, coalescence was not observed. Moreover, when expression of the GFP fusion protein was continued at a higher level for a longer period, coalescence was observed even in Psi- cells, and platings on selective media demonstrated the concomitant appearance of new Psi+ elements in the culture (Patino et al., 1996 ).
While such experiments provide strong support for the protein-only theory of Psi+ inheritance, the specific molecular nature of the altered state of Sup35 and the mechanism by which this state is transferred to new protein is completely unknown. Here we have examined the capacity of Sup35 and various subdomains to self-associate in a purified system in vitro. The whole protein and derivatives containing the N-terminal region, which plays such a critical role in Psi+ metabolism, produce highly ordered aggregates in the form of long, rigid fibers. These fibers bind Congo red and have substantial þ-sheet structure, both diagnostic of the amyloid fibers associated with certain human diseases (Sipe, 1994 ; Kelly, 1996 ), including the transmissible spongiform encephalopathies. By electron microscopy, these fibers have at least two distinct structures. Once a particular structure is initiated, it is apparently self-perpetuating, continuing along the entire length of the fiber. The time course of fiber formation indicates that this is a self-seeded process. These data provide a convincing molecular explanation for the self-propagating nature of the Psi+ genetic element.
Maintenance of Psi+ elements in yeast cells requires not only the N-terminal domain of Sup35, but also the protein chaperone Hsp104 (Chernoff et al., 1995 ; Patino et al., 1996 ). We postulated that Hsp104 promotes a conformational change in Sup35 that modulates its aggregation propensity and facilitates maintenance of the Psi+ state (Patino et al., 1996 ). Prior to initiating investigations into the molecular properties of Sup35 in vitro, we asked if the previously perceived requirement for Hsp104 in the formation of Sup35 aggregation was absolute, or could be overcome by high-level expression of Sup35. The coalescence of a copper-regulated Sup35 protein fused to GFP was examined at different levels of induction. At low levels, fluorescent foci were observed only in Psi+ HSP104 cells; at high levels, foci were also observed in Psi- hsp104 cells (not shown). We conclude that Hsp104 is not absolutely required for the coalescence of Sup35 into [PSI+]-like foci when this protein is expressed at a high level.
Although most NM fibers were straight, some had a strikingly different, wavy appearance with a mean periodicity of 40.2 ± 3.7 nm along the fiber axis (Figure 2C). The mean width across the narrowest parts of the wavy fibers was 10.3 ± 1.5 nm. Wavy fibers were observed under several buffer conditions and protein concentrations, but they were always in the minority. Reexamination of fibers formed by whole Sup35 demonstrated that the central core of the Sup35 filament was also occasionally wavy (not shown). Transitions between smooth and wavy structures within a single fiber were not observed. Thus, fibers can have different structures. But once a particular conformation is initiated, it is maintained along the entire traceable length of the fiber.
(A) Difference spectrum of Congo red and Congo red bound to NM fibers. (B) Scatchard analysis of Congo red binding to NM fibers. (C) Circular dichroism of NM fibers. Three-week-old NM fibersThe formation of fibers with a high þ-sheet content is another diagnostic feature of amyloid proteins. The circular dichroism spectrum of mature NM fibers exhibited a minimum at 218 nm, which is characteristic of structures rich in þ sheet (Figure 3C). Conversion of CD spectra to units of molar ellipticity was based upon protein concentrations of the same sample calculated from total amino acid composition. Deconvolution programs use structural information based upon known crystallographic structures. Insufficient data are available for fibrous, þ sheet-rich proteins to provide a reliable basis for the deconvolution of this spectrum (for review, see Nesloney and Kelly, 1996 ).
The time course of change in NM secondary structure, as determined by circular dichroism, revealed a similar kinetic pattern. In unseeded reactions, a progressive change in secondary structure, suggestive of a loss of random coil with an increase in þ sheet, was initiated after a long lag phase (Figure 6A). In seeded reactions, the same change in secondary structure occurred, but the lag phase was shortened or eliminated (Figure 6B). We noted that the change in structure appeared to plateau with a stronger random coil signal than was obtained with mature, aged fibers (Figure 3C). This suggests that an additional adjustment of structure takes place on a longer time scale, as fibers are aged. Because the signal produced by random coil is much stronger than that produced by þ sheet (on a residue-for-residue basis), this adjustment likely represents a rather small change in secondary structure. By standard electron microscopy, newly formed fibers were indistinguishable from aged fibers.
We also examined fiber formation with a mutant protein lacking amino acids 22-69 (including two of the four nonapeptide repeats that characterize this domain) that is defective in Psi+ induction in vivo (Ter-Avanesyan et al., 1993 ; Derkatch et al., 1996 ), when this mutation is incorporated either into the whole protein or the NM fragment. NM22-69 and Sup3522-69 formed fibers that bound Congo red and were similar in appearance to those formed by wild-type NM and Sup35 by electron microscopy (not shown). However, the mutant proteins formed fibers much more slowly than did the wild-type proteins, in both unseeded reactions and in reactions seeded with NM fibers (Figure 4C).
Sup35 is the protein determinant of the unconventional genetic element known as Psi+. We have shown that this protein, and derivatives containing the N-terminus, have an intrinsic capacity to form highly ordered fibrous structures. These fibers bind Congo red and are rich in þ-sheet secondary structureproperties common to protein amyloids. Fibers form after a lag phase, whose duration depends upon the protein concentration. The ability of a small quantity of preformed fibers to shorten or eliminate the lag phase confirms that fiber formation is a nucleated process. Our data do not prove the "protein-only" hypothesis of [PSI+] inheritance. However, striking parallels between the process of fiber formation in vitro and the behavior of Psi+ elements in vivo make this self-perpetuating, ordered assembly process a likely molecular mechanism for the inheritance of this unconventional genetic element.
We have, however, begun to acquire some understanding of the structures of the fibers themselves. Fibers formed by whole Sup35 are long, semirigid, and nonbranching. Under some buffer conditions they have a smooth appearance but, more typically, a central backbone with amorphous material splayed out along its sides. Although the N region is necessary and sufficient for fiber formation, the fibers it forms are thinner and generally shorter than those formed by NM. Neither the M nor the C fragments form fibers on their own. The width of NM fibers is similar to that of the central backbone in whole Sup35 fibers and both NM fibers and the backbone of Sup35 can assume distinct structural forms, "straight" or "wavy." These data indicate that (1) self-association of N dictates the formation of the fiber axis, (2) M packs against the exterior of this core, and (3) C is located on the periphery and orderly packing of this domain is not critical for the stability of the fiber.
The regular periodicity of wavy fibers indicates that an orderly packing of substructures must underlie the formation of these fibers. Atomic force microscopy (S. Xu, B. Bevis, and S. Lindquist, unpublished data) suggests that smooth fibers are also composed of a regularly repeating substructure and that additional forms with more subtle distinctions, not resolvable by standard electron microscopy, may be present. Most remarkably, transitions between wavy and straight fibers were never observed. Thus, once a specific type of packing is established in the fiber, it is self-perpetuating along the length of the fiber.
Self-perpetuating, alternatively packed conformations have been observed in seeded polymerization of bacterial flagellin (Akasura, 1970 ) as well as in fibers formed from Aþ (Harper et al., 1997 ). Recently they have been suggested to account for one of the most perplexing aspects of mammalian prion biology, the existence of different "strains" (Bessen and Marsh, 1994 ; Bessen et al., 1995 ; Lansbury and Caughey, 1995 ). These strains produce diseases with different rates of progression and distinct pathologies. Their very existence has been used by others to argue that a nucleic acid must be involved in prion propagation (see Horwich and Weissman, 1997 , for review). Remarkably, strain variation is also observed in yeast Psi+ elements, which have different stabilities and strengths of nonsense-codon suppression (Derkatch et al., 1996 ). We suggest that the alternative conformational states we have observed in Sup35 fibers may provide a "protein only" structural basis for Psi+ strain variation.
Most importantly, our data suggest that a common biochemical process underlies the transmission of a cytoplasmically inherited trait in yeast and the pathogenesis of some of the most perplexing and intractable human diseases, including the spongiform encephalopathies and Alzheimer's disease. This process is the seeded formation of highly ordered amyloid (or amyloid-like) fibers that has been shown by others to be a characteristic of both peptides derived from mammalian PrP protein of transmissible spongiform encephalopathies.
Genetic analysis in yeast should provide a powerful new tool for elucidating the mechanisms by which such protein structures are established and propagated and may provide a new avenue for testing therapeutic strategies. The remarkably similar structural characteristics of the yeast Sup35 and mammalian amyloid proteins, and their connection to such divergent biological processes as heritable changes in metabolism and fatal neurodegenerative diseases, suggest that self-perpetuating changes in protein structure are widespread and figure in a broader range of biological processes than previously suspected.
Akasura, S. (1970). Polymerization of flagellin and polymorphism of flagella. Adv. Biophys. 1, 99-155. Bessen, R.A., and Marsh, R.F. (1994). Distinct PrP properties suggest the molecular basis of strain variation in transmissible mink encephalopathy. J. Virol. 68, 7858-7868. Bessen, R.A., Kocisko, D.A., Raymond, G.J., Nandan, S., Lansbury, P.T., and Caughey, B. (1995). Non-genetic propagation of strain-specific properties of scrapie prion protein. Nature 375, 698-700. Caughey, B., and Chesebro, B. (1997). Prion protein and the transmissible spongiform encephalopathies. Trends Cell Biol. 7, 56-62. Caughey, B., Kocisko, D.A., Raymond, G.J., and Lansbury, P.T., Jr. (1995). Aggregates of scrapie associated prion protein induce the cell free conversion of protease-sensitive prion protein to the protease-resistant state. Chem. Biol. 2, 807-817. Chernoff, Y.O., Derkatch, I.L., and Inge-Vechtomov, S.G. (1993). Multicopy SUP35 gene induces de novo appearance of psi-like factors in the yeast Saccharomyces cerevisiae. Curr. Genet. 24, 268-270. Chernoff, Y.O., Lindquist, S.L., Ono, B.-I., Inge-Vechtomov, S.G., and Liebman, S.W. (1995). Role of the chaperone protein Hsp104 in propagation of the yeast prion-like factor Psi+. Science 268, 880-884. Clos, J., and Brandau, S. (1994). pJC20 and pJC40two high-copy-number vectors for T7 RNA polymerase-dependent expression of recombinant genes in Escherichia coli. Prot. Exp. Purif. 5, 133-137. Come, J.H., Fraser, P.E., and Lansbury, P.T., Jr. (1993). A kinetic model for amyloid formation in the prion diseases: importance of seeding. Proc. Natl. Acad. Sci. USA 90, 5959-5963. Cox, B.S. (1965). , a cytoplasmic suppressor of super-suppression in yeast. Heredity 20, 505-521. Cox, B.S., Tuite, M.F., and McLaughlin, C.S. (1988). The psi factor of yeast: a problem of inheritance. Yeast 4, 159-178. Derkatch, I.L., Chernoff, Y.O., Kushnirov, V.V., Inge-Vechtomov, S.G., and Liebman, S.W. (1996). Genesis and variability of [PSI] prion factors in Saccharomyces cerevisiae. Genetics 144, 1375-1386. Didichenko, S.A., Ter-Avanesyan, M.D., and Smirnov, V.N. (1991). Ribosome-bound EF1-like protein of yeast Saccharomyces cerevisiae. Eur. J. Biochem. 198, 705-711. Doel, S.M., McCready, S.J., Nierras, C.R., and Cox, B.S. (1994). The dominant PNM2- mutation that eliminates the factor of Saccharomyces cerevisiae is the result of a missense mutation in the SUP35 gene. Genetics 137, 659-670. Harper, J.D., Wong, S.S., Lieber, C.M., and Lansbury, P.T., Jr. (1997). Observation of metastable Aþ amyloid protofibrils by atomic force microscopy. Chem. Biol. 4, 119-125. Horwich, A.L., and Weissman, J.S. (1997). Deadly conformationsprotein misfolding in prion disease. Cell 89, 499-510.[Full Text] Jarret, J.T., and Lansbury, P.T., Jr. (1993). Seeding "one dimensional crystallization" of amyloid: a pathogenic mechanism in Alzheimer's disease and scrapie? Cell 73, 1055-1058.[Summary] Jarret, J.T., Berger, E.P., and Lansbury, P.T., Jr. (1993). The carboxy terminus of the þ amyloid protein is critical for the seeding of amyloid formation: implications for the pathogenesis of Alzheimer's disease. Biochemistry 32, 4693-4697. Kelly, J.W. (1996). Alternative conformations of amyloidogenic proteins govern their behavior. Curr. Opin. Struct. Biol. 6, 11-17. Klunk, W.E., Pettegrew, J.W., and Abraham, D.J. (1989). Two simple methods for quantitating low affinity dye-substrate binding. J. Hystochem. Cytochem. 37, 1293-1279. Kushnirov, V.V., Ter-Avanesyan, M.D., Didichenko, S.A., Smirnov, V.N., and Chernoff, Y.O. (1990). Divergence and conservation of SUP2 (SUP35) gene of yeasts Pichia pinus and Saccharomyces cerevisiae. Yeast 6, 461-472. Lansbury, P.T., Jr., and Caughey, B. (1995). The chemistry of scrapie infection: the `ice 9' metaphor. Chem. Biol. 2, 1-5. Lindquist, S. (1997). Mad cows meet psi-chotic yeast: the expansion of the prion hypothesis. Cell 89, 495-498.[Full Text] Nesloney, C.L., and Kelly, J.W. (1996). Progress toward understanding þ-sheet structure. Bioorganic Med. Chem. 4, 739-766. Parsell, D.A., Kowal, A.S., Singer, M.A., and Lindquist, S. (1994). Protein disaggregation mediated by heat-shock protein Hsp104. Nature 372, 475-478. Patino, M.M., Liu, J.J., Glover, J.R., and Lindquist, S. (1996). Support for the prion hypothesis for inheritance of a phenotypic trait in yeast. Science 273, 622-626. Paushkin, S.V., Kushnirov, V.V., Smirnov, V.N., and Ter-Avanesyan, M.D. (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. Paushkin, S.V., Krusnirov, V.V., Smirnov, V.N., and Ter-Avanesyan, M.D. (1997). Interaction between yeast Sup45p (eRF1) and Sup35p (eRF3) polypeptide chain release factors: implications for prion-dependent regulation. Mol. Cell. Biol. 17, 2798-2805. Prusiner, S.B. (1994). Biology and genetics of prion diseases. Annu. Rev. Microbiol. 48, 655-658. Sambrook, J., Fritsch, E.E., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual2nd edition., Cold Spring Harbor, New York: Cold Spring Harbor Laboratory. Scott, M., Foster, D., Mirenda, C., Serban, D., Coufal, F., Walchli, M., Torchia, M., Groth, D., Carlson, G., DeArmond, S.J., and Prusiner, S.B. (1989). Transgenic mice expressing hamster prion protein produce species-specific scrapie infectivity and amyloid plaques. Cell 59, 847-857.[Summary] Sherman, F., Fink, G.R., and Hicks, J.B. (1986). Manual for Methods in Yeast Genetics., Cold Spring Harbor, New York: Cold Spring Harbor Laboratory. Sikorski, R.S., and Hieter, P. (1989). A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 122, 19-27. Sipe, J.D. (1994). Amyloidosis. Crit. Rev. Clin. Lab. Sci. 31, 325-354. Spiess, E., Zimmerman, H.P., and Lunsdorf, H. (1987). In Electron Microscopy in Molecular Biology: A Practical Approach, J. Sommerville and U. Scheer, eds. (Oxford: IRL Press Ltd.), pp. 147-166. Stansfield, I., and Tuite, M.F. (1994). Polypeptide chain termination in Saccharomyces cerevisiae. Curr. Genet. 25, 385-395. Stansfield, I., Jones, K.M., Kushnirov, V.V., Dagkesamanskaya, A.R., and Poznyakovski, A.I. (1995). The products of the SUP45 (eRF1) and SUP35 genes interact to mediate translation termination in Saccharomyces cerevisiae. EMBO J. 14, 4365-4373. Ter-Avanesyan, M.D., Kushnirov, V.V., Dagkesamanskaya, A.R., Didchenko, S.A., and Chernoff, Y.O. (1993). Deletion analysis of the SUP35 gene of the yeast Saccharomyces cerevisiae reveals two non-overlapping functional regions in the encoded protein. Mol. Microbiol. 7, 683-692. Ter-Avanesyan, M.D., Dagkesamanskaya, A.R., Kushnirov, V.V., and Smirnov, V.N. (1994). The SUP35 omnipotent suppressor gene is involved in the maintenance of the non-Mendelian determinant Psi+ in the yeast Saccharomyces cerevisiae. Genetics 137, 671-676. Thiele, D.J. (1988). ACE1 regulates expression of the Saccharomyces cerevisiae metallothionein gene. Mol. Cell. Biol. 8, 2745-2752. Wickner, R.B. (1994). Ure 3 as an altered URE2 protein: evidence for a prion analog in Saccharomyces cerevisiae. Science 264, 566-569. Zhouravleva, G., Frolova, L., Legoff, X., Leguellec, R., and Inge-Vechtomov, G. (1995). Termination of translation in eukaryotes is governed by two interacting polypeptide chain release factors, eRF1 and eRF3. EMBO J. 14, 4065-4072.