Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
FEMS Microbiology Reviews 26 (2002) 257^276 www.fems-microbiology.org The viral killer system in yeast: from molecular biology to application Manfred J. Schmitt , Frank Breinig Angewandte Molekularbiologie (FR 8.3 ^ Mikrobiologie), Universita«t des Saarlandes, Im Stadtwald, Geba«ude 2, D-66123 Saarbru«cken, Germany Received 5 March 2002; received in revised form 26 March 2002; accepted 28 March 2002 First published online 8 May 2002 Abstract Since the initial discovery of the yeast killer system almost 40 years ago, intensive studies have substantially strengthened our knowledge in many areas of biology and provided deeper insights into basic aspects of eukaryotic cell biology as well as into virus^host cell interactions and general yeast virology. Analysis of killer toxin structure, synthesis and secretion has fostered understanding of essential cellular mechanisms such as post-translational prepro-protein processing in the secretory pathway. Furthermore, investigation of the receptor-mediated mode of toxin action proved to be an effective means for dissecting the molecular structure and in vivo assembly of yeast and fungal cell walls, providing important insights relevant to combating infections by human pathogenic yeasts. Besides their general importance in understanding eukaryotic cell biology, killer yeasts, killer toxins and killer viruses are also becoming increasingly interesting with respect to possible applications in biomedicine and gene technology. This review will try to address all these aspects. , 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Yeast; Prepro-protein processing ; Toxin secretion; Receptor-mediated toxicity; Endocytosis ; Retrograde transport ; Antifungal ; Killer virus; Heterologous protein secretion Contents 1. 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Molecular biology of killer yeast . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Genetic basis of killer phenotype expression: cytoplasmic RNA viruses . . . . . . . . . . . 2.2. dsRNA viruses in S. cerevisiae: L-A and its toxin-coding M satellites . . . . . . . . . . . . . 2.3. Preprotoxin processing and toxin secretion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Toxicity of the killer proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Toxin immunity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6. Virus^host cell interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7. Killer viruses in the non-conventional yeasts Hanseniaspora uvarum and Zygosaccharomyces bailii . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Killer yeasts in wine fermentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Killer toxins as potential antifungals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Killer toxin expression in transgenic plants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Heterologous protein secretion via pptox secretion and processing signals . . . . . . . . . . Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 259 259 259 263 264 266 267 Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 272 3. 4. * Corresponding author. Tel. : +49 (681) 302 4730 ; Fax: +49 (681) 302 4710. 269 269 269 270 270 270 271 E-mail address: [email protected] (M.J. Schmitt). 0168-6445 / 02 / $22.00 , 2002 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII : S 0 1 6 8 - 6 4 4 5 ( 0 2 ) 0 0 0 9 9 - 2 FEMSRE 748 1-8-02 Cyaan Magenta Geel Zwart 258 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 Fig. 1. Phenotype of a K28 killer and a non-killer strain of the yeast S. cerevisiae. A: A toxin-secreting K28 killer strain expressing functional toxin immunity (Kþ Iþ ) was streaked onto methylene blue agar (MBA; pH 4.7) that had been seeded with a lawn of a sensitive indicator non-killer strain. After 3 days of incubation at 20‡C, clear zones of growth inhibition around the streak of the killer strain indicate K28 toxin secretion. B: A non-killer strain lacking functional immunity (K3 I3 ) was streaked onto an MBA plate that had been seeded with a K28 toxin-secreting killer strain. Secreted K28 toxin di¡uses into the agar and greatly impairs cell growth of the K3 I3 non-killer strain. The dark blue colony staining indicates that the cells are being killed by the toxin. 1. Introduction In contrast to antibacterial antibiotics, bacteriophages and bacteriocins, that were described at the beginning of the last century, a similar antibiotic phenomenon in yeast was only demonstrated much later. In 1963, Bevan and Makower discovered the killer phenomenon in a Saccharomyces cerevisiae strain which was isolated as a brewery contaminant [1]. The killer phenotype they described is based on the secretion of a low molecular mass protein or glycoprotein toxin (the so-called killer toxin) which kills sensitive cells of the same or related yeast genera without direct cell^cell contact. The killer strains themselves are immune to their own toxin, but remain susceptible to the toxins secreted by other killer yeasts (Fig. 1). Besides toxin-secreting killer strains, a signi¢cant number of nonkiller yeasts can be isolated that have lost their ability to produce various killer toxins but nevertheless retain immunity. The killer phenotype, i.e. toxin production and functional immunity, is very frequent among yeasts and can be found both in natural yeast isolates and in laboratory yeast strain collections. In particular, for yeast strains living in their natural habitat it has been shown that toxin production can confer a marked advantage in the competition with sensitive yeast strains for limited available nutrients [2,3]. After the initial discovery of the killer phenomenon in S. cerevisiae, it soon became evident that killer strains are not restricted to the genus Saccharomyces but can also be found among many other yeast genera; up to now, toxin-producing killer yeasts have been identi¢ed in Candida, Cryptococcus, Debaryomyces, Hanseniaspora, Hansenula, Kluyveromyces, Metschnikowia, Pichia, Ustilago, Torulopsis, Williopsis and Zygosaccharomyces, indicating that the killer phenomenon is indeed widespread among yeasts [2,4,5]. Interestingly, the genetic basis for killer phenotype expression can be quite variable; in the few cases where killer determinants have clearly been identi¢ed, they are either cytoplasmically inherited encapsulated dsRNA viruses, linear dsDNA plasmids or nuclear genes (see below and Table 1). Among the plasmid-encoded killer toxins, the killers of Kluyveromyces lactis and Pichia acaciae are most intensively studied and have been nicely described in recent reviews [6,7]. During the past three decades, intensive investigations Table 1 Genetic basis for killer phenotype expression in yeast Yeast Genetic basis Toxin gene Ref. S. cerevisiae H. uvarum Z. bailii U. maydis K. lactis P. acaciae Pichia inositovora Pichia kluyveri Pichia farinosa P. anomala Williopsis mrakii dsRNA virus dsRNA virus dsRNA virus dsRNA virus linear dsDNA plasmid linear dsDNA plasmid linear dsDNA plasmid chromosomal chromosomal chromosomal chromosomal M1-, M2-, M28-dsRNA M-dsRNA M-dsRNA M-dsRNA pGKl1 pPac1 pPin1 not identi¢ed SMK1 not identi¢ed HMK [48,49,13,168] [9] [14] [159] [169] [26] [170] [2] [171] [172] [4] FEMSRE 748 1-8-02 Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 of the killer system particularly in S. cerevisiae resulted in substantial progress in many di¡erent ¢elds of biology, providing important insights into basic and more general aspects of eukaryotic cell biology, virus^host cell interactions and yeast virology. Moreover, since killer toxins resemble naturally secreted proteins or glycoproteins, detailed analysis of their structure and synthesis also substantially strengthened our knowledge of the mechanisms of post-translational prepro-protein processing in the eukaryotic secretion pathway. In addition, analysis of killer toxins and their receptor-mediated mode of action has proved to be an e¡ective means for investigating the molecular structure and in vivo assembly of yeast and fungal cell walls, and for providing information important to combating yeast infections caused by certain human pathogenic strains of the yeasts Candida albicans and/or Sporothrix schenkii. The main purpose of this review is to summarize the most recent data within the fast-growing ¢eld of the viral killer systems in S. cerevisiae and to focus on the molecular basis of the killer phenomenon and its possible applications, thereby stressing aspects that are equally relevant to yeast molecular biology, biomedicine and biotechnology. These are aspects of this topic which have not been thoroughly addressed in previous reviews. 259 presence of cytoplasmically inherited double-stranded RNA (dsRNA) viruses (reviewed in [7,8]). Persistent infection of yeast cells with these viruses is symptomless, and in contrast to certain fungal viruses that are associated with adverse phenotypic e¡ects (like La France disease in Agaricus bisporus or plaque formation in Penicillium), yeast dsRNA viruses have no (so far recognized) disadvantage for a single cell. In addition, yeast dsRNA viruses are considered non-infectious since no naturally occurring extracellular route of transmission has been identi¢ed. They were therefore designated cryptic viruses or ‘viruslike particles’ (VLPs). In contrast to the horizontal transfer of most pathogenic plant and animal RNA viruses, yeast dsRNA viruses only spread vertically during mating and/or heterokaryon formation in vivo and it has been suggested that this kind of virus transmission most likely represents a special adaptation since the frequent occurrence of mating and hyphal fusion in yeast and higher fungi makes an extracellular route of virus transmission dispensable [9^11]. However, since it was shown that nonkiller strains of S. cerevisiae can be successfully transfected with isolated and puri¢ed VLPs in vitro, it cannot be excluded that a natural route of VLP infection also exists in natural yeast and fungal habitats [12^14]. 2.2. dsRNA viruses in S. cerevisiae: L-A and its toxin-coding M satellites 2. Molecular biology of killer yeast 2.1. Genetic basis of killer phenotype expression: cytoplasmic RNA viruses In S. cerevisiae, the killer phenomenon is based on the Based on the lack of cross-immunity, their molecular mode of action and their killing pro¢les, toxin-producing S. cerevisiae killer strains have been classi¢ed into three major groups (K1, K2 and K28), each of them secreting a unique killer toxin as well as a speci¢c but as yet uniden- Table 2 Viral dsRNA genomes involved in killer phenotype expression in di¡erent yeasts Virus Function of virus dsRNA genome (kb) Encoded protein(s) Ref. ScV-L-A Helper virus 4.6 [24] ScV-M1 Satellite virus (‘killer’ virus) 1.6 ScV-M2 Satellite virus (‘killer’ virus) 1.5 ScV-M28 Satellite virus (‘killer’ virus) 1.8 UmV-P1 ‘Killer’ virus 1.4 UmV-P4 UmV-P6 ‘Killer’ virus ‘Killer’ virus 1.0 1.2 HuV-L Helper virus 4.6 HuV-M ZbV-L Satellite virus ‘killer’ virus Helper virus 1.0 4.6 ZbV-M Satellite virus ‘killer’ virus 2.1 Gag, major capsid protein; Pol, RNA-dependent RNA polymerase (Pol is in vivo expressed as Gag-Pol fusion protein) K1 preprotoxin (unprocessed K1 toxin precursor and immunity determinant) K2 preprotoxin (unprocessed K2 toxin precursor and immunity determinant) K28 preprotoxin (unprocessed K28 toxin precursor and immunity determinant) KP1 preprotoxin (precursor of an K/L heterodimeric protein toxin) KP4 monomeric protein toxin KP6 preprotoxin (precursor of an K/L heterodimeric protein toxin) Gag, major capsid protein; Gag-Pol, RNA-dependent RNA polymerasea Precursor of a monomeric protein toxin Gag, major capsid protein; Gag-Pol, RNA-dependent RNA polymerasea Precursor of a monomeric protein toxin a [173] [49] [39] [161] [159] [174] [9] [9] [14] [14] The dsRNA sequence has not yet been identi¢ed and, therefore, the postulated viral gene products Gag and Gag-Pol are hypothetical and only predicted by analogy to Gag and Gag-Pol of ScV-L-A. FEMSRE 748 1-8-02 Cyaan Magenta Geel Zwart 260 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 ti¢ed immunity component that renders killer cells immune to their own toxin [15^17]. The previously described killer type K3 in S. cerevisiae [18] is not clearly distinct from K2. In a killer yeast, production of either killer toxin K1, K2 and/or K28 is associated with the presence of a cytoplasmically inherited M-dsRNA satellite virus (designated ScV-M1, ScV-M2 or ScV-M28 for S. cerevisiae virus), which depends on the coexistence of L-A helper virus to be stably maintained and replicated within the cytoplasm of the infected host cell (Table 2). Cells containing neither dsRNA or only L-A are toxin-sensitive non-killers, while those also containing ScV-M1, ScV-M2 or ScV-M28 are immune killers. The killer viruses self-select by killing any virus-free segregants. Sensitive yeast strains survive mating with killers, and cytoplasmic mixing of the multiple ScV-M killer virion copies during mating accounts for the non-Mendelian inheritance pattern during subsequent meiosis [8]. The toxin-coding killer viruses are incompatible at the replicative level, i.e. they exclude one another in mixed infection and are unstable as multiple killer viruscontaining strains. This exclusion means that the viruses remain isolated in their host and might explain why ScVM killer virions remained distinct and have not been found as recombinants or hybrids [19]. L-A itself is an autonomously replicating mycovirus neither conferring a detectable phenotype upon its host (yeast) cell, nor requiring the presence of an M-dsRNA satellite virus for stable maintenance or replication. L-A belongs to the Totiviridae virus family. Viruses of this family have a genome encapsidated in an icosahedral shell that is a single segment of dsRNA which encodes a major coat protein and a RNA-dependent RNA polymerase [20]. L-A particles contain one and M particles contain two copies of the corresponding dsRNA genome. Each contains 120 copies of the 76-kDa major capsid protein Gag, and an estimated two copies of a minor 171-kDa Gag-Pol fusion protein. From cryoelectron microscopy and image reconstruction on the S. cerevisiae virus L-A as well as on the Ustilago maydis killer virus P4 it has been P diameter shown that both viruses have similar 430-A structures, consisting of 120 subunits arranged as 60 asymmetric dimers in a T = 1 lattice. The capsid wall contains P in diameter about 60 small openings each of them 10^15 A which extend throughout the virus surface and probably serve as molecular sieves which are large enough to allow in£ux of metabolites that are needed for RNA synthesis and replication and for exit of new viral transcripts, but small enough to ensure protection of the viral RNA genome from degradation [21^23]. Each L-A particle contains a single copy of a linear 4.6-kb L-A dsRNA, and the codogenic L-A plus strand (L-A(+)ssRNA) has two open reading frames (ORFs) that overlap by 130 nucleotides, with the 3P ORF in the 31 frame relative to the 5P ORF [24]. The latter encodes the major capsid protein Gag which is necessary for encapsulation and responsible for viral particle structure, while the 3P ORF shows the consensus amino acid sequence pattern typical of RNA-dependent RNA polymer- Fig. 2. Replication cycle of L-A and its toxin-coding M satellite virus. Both dsRNA viruses compete for the L-A encoded viral proteins Gag and GagPol which are essential for viral particle structure, viral (+)ssRNA packaging, in vivo RNA replication (i.e. minus-strand RNA synthesis), and (+)ssRNA synthesis on the double-stranded RNA template. FEMSRE 748 1-8-02 Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 261 Fig. 3. Comparison of the coding capacity and the potential VBS/IRE and 3P-TRE sequence elements on the killer virus transcripts M1(+)ssRNA (B) and M28(+)ssRNA (C) with those on L-A(+)ssRNA (A). Two cis-active viral binding sites (VBS1, VBS2) on the M transcripts are necessary for ssRNA binding and packaging. The 3P-TRE might be all that is required for in vivo ssRNA replication. Numbers are shown for distance from the 3P-terminus of the unpaired A residues which interrupt each VBS stem. (The indicated 5P-(+)ssRNA sequence (GAAAAA) is characteristic for yeast dsRNA viruses.) ases of plus-strand ssRNA and dsRNA viruses (Table 2 [25]). In analogy to the nomenclature of retroviral genes, the Gag encoding 5P ORF of L-A has been designated gag and the Pol coding 3P ORF pol. The gene product of the 3P ORF is expressed in vivo as a Gag-Pol fusion protein by a 31 ribosomal frameshift event during translation [4,24,26^30]. The Gag-Pol fusion itself is the key viral component in the replication of both, L-A and M virions; through its C-terminal Pol domain, it recognizes a com- FEMSRE 748 1-8-02 plex encapsulation signal of stem loop structures and an overlapping direct repeat within the 3P-terminal 400 bases of the L-A(+)ssRNA [31]. It is presumed that the N-terminal Gag domain within the Gag-Pol/ssRNA complex primes capsid (Gag) polymerization. In contrast to L-A, each of the three M-dsRNA genomes contains a single ORF encoding a preprotoxin protein (pptox) representing the unprocessed precursor of the mature and secreted killer toxin as well as a functional but so far unidenti¢ed Cyaan Magenta Geel Zwart 262 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 Fig. 4. Secretory pathway of killer toxin K28 in S. cerevisiae. After in vivo translation of the pptox coding viral M28 transcript, pptox is (post-translationally) imported into the lumen of the ER and signal peptidase (SP) cleavage removes the toxin’s N-terminal secretion signal (pre-region). In a late Golgi compartment, the Kex2p endopeptidase cleaves the pro-region, removes the intramolecular Q sequence and ¢nally leads to the secretion of the mature K/L heterodimeric protein toxin (modi¢ed after [167]). immunity component (Table 2). Interestingly, the toxincoding M genomes exclude each other at the replicative level and, therefore, simultaneous coexistence of all three ScV-M killer virions in a single yeast cell does not occur in vivo, but can be by-passed by co-expressing cDNA copies of di¡erent pptox genes [8,32]. Intensive work on L-A, mainly by the group of Reed Wickner, provided deeper insight into important cellular aspects such as RNA transcription, dsRNA replication, ssRNA packaging, ribosomal frameshifting, proteolytic precursor processing, and virus^host as well as virus^virus interactions. These aspects, which are quite common among RNA viruses showing striking similarities to certain mammalian viruses such as HIV, are nicely addressed in recent reviews [7,8,22,33]. L-A virions are icosahedral particles, about 39 nm in diameter, which replicate in vivo by a conservative mechanism. Synthesis of plus and minus RNA strands also occurs in viro, i.e. within isolated viral particles (Fig. 2 ; reviewed in [21]). After transcription of the double-stranded RNA template, the viral plus strand is extruded from the particle into the cytoplasm where it subsequently executes two essential functions. Firstly, it serves as mRNA template which is translated in vivo into the viral proteins Gag and Gag-Pol. Secondly the FEMSRE 748 1-8-02 (+)ssRNA molecule is packaged into new viral particles. Once this viral coat assembly has been completed, the RNA-dependent RNA polymerase (Gag-Pol) acts as replicase, synthesizes a new minus strand and ¢nally generates the complete dsRNA genome of the mature virus (Fig. 2) [8]. The toxin-coding M viruses are satellites of L-A, parasitizing Gag and Gag-Pol for replication [5,15]. It is not surprising, therefore, that their replication cycle is widely analogous to that of L-A, di¡ering only in one respect : not all plus-strand M transcripts are extruded from the viral particle because a mature M virion can contain up to two copies of the smaller M dsRNA genome instead of just a single copy in the case of L-A [15]. This phenomenon, which is also true for ¢lamentous viruses, has been called ‘headful replication’ to distinguish it from the ‘headful packaging model’ as occurs in some bacteriophages [21]. The (+)ssRNAs of both L-A and M viruses contain speci¢c packaging signals (so-called viral binding sites (VBS)) within their 3P-terminal region, which are essential for replication and RNA packaging (Fig. 3). These cisactive sequences consist of a stem loop structure whose stem is interrupted by an unpaired, protruding A residue [27,34,35]. Two sequence elements were found to be essen- Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 tial for the replication of L-A. An internal replication enhancer (IRE), almost indistinguishable from the VBS, and a 3P-terminal recognition element (3P-TRE) consisting of a small stem loop structure 5 bases from the 3P-terminus (Fig. 3). In vivo, these cis elements are recognized by GagPol and are responsible for both (+)ssRNA binding and RNA packaging into new virions [35]. Mutational analysis of a full-length L-A cDNA clone demonstrated that both the 5 bases of the consensus loop sequence (GAUYC) (Y = pyrimidine) and the unpaired A in its stem sequence were essential for VBS function, while the stem sequence was only necessary to provide secondary structure [31]. In contrast to L-A, the toxin-coding transcripts of ScV-M1 and ScV-M28 virions contain two potential VBS elements. Both have 5-base loops, the second identical to that in M1 (GAUUC) and the ¢rst di¡ering by 1 base (GGUUC; Fig. 3). Each is separated from a bulged A by the identical 3bp stem, resulting in an 11- or 12-bp repeat. The last 9 bp of this repeat are reiterated at 3375; the ¢rst but not the second VBS in M1 also overlaps a 10-bp imperfect direct repeat that includes the VBS loop (Fig. 3) [36,37]. The stem of the L-A VBS also overlaps a direct repeat, not including the loop; however, the signi¢cance of these repeats is still unknown. A correct 3P-end sequence and structure is required for in vitro (3)ssRNA strand synthesis; L-A, M1 and M28 3P-ends are not substantially di¡erent [38]. For conservative in vivo transcription of the (+)ssRNA from the dsRNA template, correct recognition of its very 5P-terminal sequence by Gag-Pol is required. Besides the 5P-sequence 5P-GAAAAA, there is little or no homology immediately downstream from this sequence in L-A, M28 or other dsRNAs. Therefore, this terminal recognition element (5P-TRE) seems to be su⁄cient for the e¡ective initiation of transcription [36]. 2.3. Preprotoxin processing and toxin secretion Among the virally encoded killer toxins of S. cerevisiae, K1 and K28 are the best-studied proteins. Although both toxins di¡er signi¢cantly in their amino acid composition 263 and their molecular mode of action, they share striking homologies with respect to their synthesis, processing and secretion. Each toxin is translated as preprotoxin (pptox) showing similar structures, subsequently undergoing post-translational modi¢cations via the endoplasmic reticulum, Golgi apparatus and secretory vesicles, ¢nally resulting in the secretion of a mature K/L heterodimeric protein toxin (Fig. 4). Sequencing of a K28-coding M28 cDNA revealed a single ORF encoding the 345 amino acid preprotoxin with a predicted size of 37.6 kDa. In vivo expression of this cDNA under transcriptional control of the strong and constitutive PGK promoter resulted in killer toxin secretion and functional (i.e. protective) toxin immunity [13,39], as previously demonstrated for M1 [17]. Mutational analysis further indicated that the toxin precursor (pptox) enters the yeast endoplasmic reticulum via a highly hydrophobic N-terminal secretion signal which precedes the K and L subunits of the mature toxin (10.5 and 11 kDa, respectively) which are separated from each other by a potentially N-glycosylated Q sequence [40]. Screening for mutants of K1 and/or K28 killers that failed to secrete any biologically active toxin while remaining immune resulted in the identi¢cation of two gene loci, KEX1 and KEX2 [41]. Kex2p, the gene product of KEX2, is a subtilisin-like endoprotease which preferentially cleaves proteins after Lys^Arg or Arg^Arg dipeptides [42^44], while Kex1p, the product of KEX1, is a serine carboxypeptidase that removes the C-terminal basic dipeptide exposed by Kex2p action [45]. Both enzymes are membrane-anchored and reside in a late trans-Golgi compartment, having their active sites localized in the organelle lumen [46,47]. Substrates such as prepro-K-factor and the protoxins of K1, K2 and K28 are processed as they pass through the Golgi during secretion in a way that is highly homologous to proinsulin processing and prohormone conversion in higher eukaryotic cells (Fig. 5) [40] : in the case of the K28 toxin precursor, Kex2p/Kex1p-mediated pptox processing removes the N-glycosylated Q sequence, trims the carboxytermini of K and L and ¢nally results in the secretion of a Fig. 5. Analogy in K28 preprotoxin processing in yeast and proinsulin maturation in mammalian cells (h-Kex2p, human homologue to yeast endopeptidase Kex2p). FEMSRE 748 1-8-02 Cyaan Magenta Geel Zwart 264 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 Fig. 6. Receptor-mediated mode of action of the yeast K1 and K28 viral toxins. Killing of a sensitive yeast cell is envisaged in a two-step process involving initial toxin binding to receptors at the level of the cell wall (R1) and the cytoplasmic membrane (R2). After interaction with the plasma membrane, K1 is acting from outside the cell and disrupts cytoplasmic membrane function, while K28 enters the cell by endocytosis in order to reach its ¢nal target, the yeast cell nucleus (note that the cell surface receptors R1 and R2 are di¡erent for both toxins; see text). 21-kDa K/L heterodimeric protein toxin whose K and L subunits are covalently linked through a single disul¢de bond between K-Cys56 and L-Cys340 [40]. Most importantly, the C-terminus of L contains a four amino acid epitope which represents a classical endoplasmic reticulum (ER) retention signal (HDELR). Since this signal is initially masked by a carboxy-terminal arginine residue, ER retention of the toxin precursor is e¡ectively prevented until the protoxin enters a late Golgi compartment in which the Kex1p carboxypeptidase removes this residue and uncovers the toxin’s intracellular targeting signal (see below) [40]. In striking analogy to K28 pptox processing, the K1 viral toxin is likewise processed from a 316 amino acid precursor, resulting in an K/L heterodimer of 21 kDa whose subunits are covalently linked by three disul¢de bonds [48]. In contrast to the well established processing pathways of K1 and K28, it is not yet clear if in vivo processing of K2 pptox results in the secretion of a monomeric or a heterodimeric protein toxin; the K2 sequence is completely unrelated to the sequence of either K1 or K28 and, although its toxic mechanism is similar to that of the ionophoric K1 toxin, K2 immunity seems to be distinct [49]. Since the published data on K2 are insu⁄cient for detailed comparison, and the K1 system has been intensively dealt with in previous reviews [7,50], the following chapter will only emphasize the recently described and unique K28 killer toxin. 2.4. Toxicity of the killer proteins Even though the killer toxins posses di¡erent modes of action, they do have one thing in common: all viral toxins (K1, K2, K28) kill a sensitive yeast cell in a receptormediated two-step process (Fig. 6): the ¢rst step involves a fast and energy-independent binding to a toxin receptor within the cell wall of a sensitive target cell. In case of K1 and K2, this primary receptor has been identi¢ed as L-1,6D-glucan, whereas the cell wall receptor for K28 is a high molecular mass K-1,3-mannoprotein [51,52]. It was specu- FEMSRE 748 1-8-02 lated, however, that toxin binding to the primary cell wall receptor either concentrates the toxin at the level of the cell wall or mediates close contact between the toxin and the target cell membrane [53]. Susceptible strains can become toxin resistant by chromosomal mutations in any of a set of genes that are involved in the structure and/or biosynthesis of yeast cell wall components; in fact, a whole set of toxin resistant yeast kre mutants (for killer resistant) has been identi¢ed for K1, and several yeast mnn mutants (for mannoprotein) have been shown to be highly resistant to K28 at the cell wall level [54,55]. The second, energy-dependent step involves toxin translocation to the cytoplasmic membrane and interaction with a secondary membrane receptor. While the membrane receptor for K28 is still unknown, the K1 membrane receptor was recently identi¢ed as Kre1p, an O-glycosylated yeast cell surface protein which is initially GPI-anchored to the plasma membrane and which is also involved in L-1,6-glucan biosynthesis and K1 cell wall receptor assembly [56]. After having reached the cytoplasmic membrane, the toxin (K1) exerts its lethal e¡ect by ion channel formation and disruption of cytoplasmic membrane function [56^59]. In contrast to the ionophoric mode of action in which K1 acts from outside the cell, K28 represents the ¢rst viral killer toxin for which it was demonstrated that it enters a sensitive target (yeast) cell by endocytosis [37]. After receptor-mediated entry into the cell, the toxin traverses the secretion pathway in reverse (via Golgi and ER), subsequently enters the cytosol, and ¢nally transduces its toxic signal into the yeast cell nucleus where the lethal events occur : inhibition of DNA synthesis and cell cycle arrest at the G1 /S boundary [37,40,60]. Interestingly, this unique retrograde K28 toxin transport critically depends on a short, four amino acid motif (HDEL) at the very C-terminus of L which acts as intracellular targeting signal directing the toxin from an early endosomal compartment back to the trans-Golgi network and thereby preventing toxin degradation in the vacuole (Fig. 7) [37,40]. Just as in higher eukaryotic cells, resident yeast ER proteins like the Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 265 Fig. 7. Retrograde K28 toxin transport in S. cerevisiae. Left panel: Simpli¢ed model of how the K/L heterodimeric K28 virus toxin is entering a sensitive yeast cell in order to reach the cytosol. In vivo toxicity of K28 involves endocytotic uptake, retrograde toxin transport via Golgi apparatus and ER and ER exit into the cytosol through the Sec61 complex (CM, cytoplasmic membrane ; EV, endosomal vesicle). Right panel : Schematic drawing of the subunit structure of selected microbial protein toxins carrying a carboxy-terminal ER targeting signal (modi¢ed according to [37,64]). chaperones Kar2p (BiP) and/or the protein disul¢de-isomerase carry a C-terminal H/KDEL motif which facilitates Golgi-to-ER recycling of the proteins and their ¢nal location within the ER lumen [61]. Interestingly, the K28 virus toxin as well as certain bacterial or plant toxins likewise carry C-terminal H/KDELlike sequences which are responsible for the intracellular targeting and the retrograde transport of the toxins, a strategy that is quite common among microbial protein toxins with enzymatic activity on intracellular targets [37,62]. What all these toxins do have in common is the capacity to ¢rst bind to cell surface receptors, then to enter the cell by endocytosis and, ¢nally, to kill the eukaryotic target cell by modifying some essential cellular components within the cytosol. Some of these protein toxins (like Pseudomonas exotoxin A, cholera toxin and yeast K28 virus toxin) carry a carboxy-terminal H/KDEL (or H/KDEL-like) sequence which can facilitate retrograde toxin transport all the way from the cell surface to the Golgi apparatus and the ER. Consequently, mutations within this C-terminal targeting sequence dramatically reduce cytotoxicity by preventing correct interaction of the toxin’s C-terminus with the K/HDEL receptor of the target cell [40,63]. In this respect it is also interesting to note that some of these toxins contain disul¢de bonds at or near their C-termini whose in vivo function is predicted to ensure correct access of the toxin’s K/HDEL signal to the K/HDEL receptor of the corresponding target cell [64]. FEMSRE 748 1-8-02 This situation is also true for the K28 toxin since it was recently shown that the cysteine residue right next to the L C-terminus (Cys340 ) is part of the disul¢de bond that covalently joins K and L (Fig. 8). It, therefore, has been postulated that correct disul¢de bond formation is required to ensure accessibility of the toxin’s HDEL signal in vivo [40]. However, a noticeable di¡erence between H/KDEL-carrying protein toxins produced and secreted by di¡erent bacteria to that produced by a virus-infected yeast cell is that the K28 toxin itself is expressed in a eukaryotic cell, and therefore the toxin’s C-terminal ER targeting signal has to be masked as long as the toxin resides within the early secretory pathway. Once it has reached a trans-Golgi compartment, the C-terminal arginine residue is removed by Kex1p cleavage, and the toxin’s intracellular targeting signal is uncovered. Because of this ‘intelligent’ strategy, toxin-producing K28 killers are able to prevent toxin retention in the ER lumen [37,40]. Once the toxin has reached the ER lumen of a sensitive yeast cell, it is retranslocated into the cytosol by passing the yeast Sec61 translocon which represents the major ER export channel in yeast [37]. The Sec61 complex itself, which consists of the transmembrane ER protein Sec61p and the two smaller subunits Sbh1p and Sss1p [65], has also been shown to be involved in the export and removal of misfolded ER proteins from the secretory pathway and their subsequent proteasomal degradation in the cytosol [66,67]. However, in contrast to the majority of Cyaan Magenta Geel Zwart 266 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 Fig. 8. Interaction of the toxin’s L-C-terminus with the cellular HDEL receptor Erd2p ensures retrograde toxin transport via endosomes and Golgi to the ER. Subsequent toxin translocation into the cytosol is mediated by the Sec61 complex which represents the major ER export channel in yeast. The indicated ERAD components Cue1p, Ubc7p, Der3p ( = Hrd1p), Ubc6p, and Der1p are not involved in ER-to-cytosol export of the toxin and, therefore, ER exit of K28 should be mechanistically di¡erent from the export of misfolded ER proteins [37]. Due to the reducing environment of the yeast cell cytosol, the toxin dissociates into its two subunits: while L is ubiquitinated and targeted for proteasomal degradation, K is somehow transducing the toxic signal into the nucleus. Previously described data [37,38] have been incorporated into this model. H/KDEL-carrying microbial toxins, ER-to-cytosol translocation of the viral K28 killer toxin does not depend on the ER-associated degradation pathway (ERAD), and yeast mutants defective in ERAD components such as Der3p/Hrd1p and/or Der1p show wild-type sensitivity [37]. It therefore can be predicted that toxin export from the ER should be mechanistically di¡erent from ER-tocytosol export of misfolded proteins. After the heterodimeric K28 toxin has reached the cytosol, it dissociates into its two subunits K and L because of the highly reducing environmental conditions of the yeast cell cytosol. Subsequently after exit from the ER, the toxin’s L subunit is ubiquitinated and targeted for proteasomal degradation, while K is stably transducing the toxic signal into the nucleus where the lethal events ¢nally occur: DNA synthesis is rapidly inhibited, cell viability is lost more slowly and cells arrest in early S phase of the cell cycle with a medium-sized bud, a single nucleus in the mother cell and a pre-replicated 1n DNA content [37,60,68]. So far, it is not known if K is entering the nucleus (either by passive di¡usion due to its small size or by an importin-mediated active transport mechanism) FEMSRE 748 1-8-02 or if it is interacting in the cytosol with one or more yeast proteins to exhibit its lethal e¡ect in the nucleus. Currently, two-hybrid screens are being performed to identify yeast proteins that are potentially interacting with the toxin’s K subunit. 2.5. Toxin immunity Besides the fact that the primary toxin target has yet to be identi¢ed, the question of how immunity occurs in vivo has still not been answered. Although the precise molecular basis for toxin immunity is still unknown, it has been speculated that it might be conferred by the toxin precursor itself. This might act as competitive inhibitor of the mature toxin by saturating or causing elimination of a so far unidenti¢ed plasma membrane receptor that normally mediates toxicity [69^71]. It was also shown that the unprocessed toxin precursor is su⁄cient to confer immunity since in vivo expression of either K1 or K28 cDNAs in a vkex2 null mutant lacking the ability of pptox processing and thus incapable of releasing K and L from the intervening Q sequence results in immune non-killer yeast trans- Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 formants [17,72,73]. Based on all these observations, a model for K1 toxin immunity had been proposed in which either loss or modi¢cation of the toxin’s secondary plasma membrane receptor would cause immunity. When it was demonstrated that expression of a cDNA copy of the preprotoxin was su⁄cient to confer normal immunity, it was postulated that the central Q component of this precursor might not only act as intramolecular chaperone ensuring proper secretory pptox processing, but also by providing some sort of a masking function by protecting membranes of toxin-producing cells against damage by the hydrophobic components within the K subunit of K1 [48]. Later on, a more plausible model was proposed, in which it was speculated that interaction of the secreted protoxin with the receptor, through the same toxin domain involved in lethality, resulted in diversion of the complex to the vacuole and destruction [17]. This model was further strengthened by phenotypic analyses of various mutant pptox derivatives which clearly indicated that the K toxin was the lethal component and that its secretion in the mature form caused severe growth inhibition while secretion of K fused to just an N-terminal fragment of Q was su⁄cient to confer immunity [74]. Dependence of immunity on diversion of the putative membrane receptor to the vacuole is consistent with the defect in immunity observed in many vps mutants (Tipper and Sturley, personal communication). One class of killer resistant kre mutants, therefore, should show similar resistance in spheroplasts. More recently, a chromosomal kre1-12 mutation was described which caused K1 resistance in yeast cell spheroplasts without a¡ecting toxin binding to the cell wall, and the wildtype gene product, Kre1p, was identi¢ed as being the long-sought membrane receptor for the K1 viral toxin [56,75]. Although the precise mechanism for K1 immunity is still obscure, a completely di¡erent mechanism must be postulated for K28 killers, since it has been shown that K28 killer yeasts take up their own toxin after it has been secreted. Furthermore, the toxin subsequently re-enters the secretory pathway of a killer cell and reaches the cytosol, just as in a sensitive target cell. However, in contrast to the latter, the killer cell is not killed but rather protected against the lethal e¡ect of the toxin. This means that K28 immunity is likely to a¡ect some unknown step either within the yeast cell cytosol or eventually downstream from that, possibly within the nucleus. Future experiments will hopefully shed more light onto this interesting and so far unique mode of toxin action and immunity. 2.6. Virus^host cell interactions For maintenance and expression of a stable killer phenotype, chromosomal genes of the host cell play an important, sometimes an essential role. Besides the chromosomal SEC genes required for general secretion of extracellular proteins and glycoproteins, the KEX-encoded FEMSRE 748 1-8-02 267 proteases Kex2p and Kex1p are necessary for preprotoxin processing and precursor maturation of the yeast pheromone K-factor [70,76^79]. Furthermore, there is a whole set of chromosomal yeast genes which either directly or indirectly a¡ect dsRNA virus propagation. As recently reviewed by Wickner [8,22] and Polonelli [7], these genes can be classi¢ed into two major groups, the maintenance of killer genes, MAK, and the superkiller genes, SKI. Among more than 30 chromosomal MAK genes that have been identi¢ed so far, only three genes (MAK3, MAK10, PET18) are required for stable maintenance of both yeast dsRNA viruses, L-A and M; all other MAK genes are only necessary for maintaining either of three known toxin-coding M satellites [73,80]. MAK3 encodes an N-acetyltransferase which is responsible for the N-acetylation of the Gag protein required for the assembly of L-A and M [81,82]. Without this acetylation, the Gag proteins are not able to self-assemble, resulting in a rapid loss of all dsRNA viruses. In addition, Mak3p acetylates at least three additional mitochondrial proteins explaining the slow growth of mak3 mutants on a non-fermentable carbon source. Like Mak3p, Mak10p is also needed for cell growth on a non-fermentable carbon source, and it was further shown to stabilize viral particles, in particular those containing dsRNA, since in mak10 mutants, virions containing single-stranded (+)ssRNA only remain stable while mature viruses containing a complete copy of the viral dsRNA genome are unstable. More recently, physical interaction of Mak3p, Mak10p, and Mak31p was demonstrated and it therefore has been proposed that these Mak proteins might also function as a complex in vivo [83,84]. The gene product of PET18 likewise contributes to the overall virion stability, and Pet18p itself is thought to be particle-associated [85^87]. For stable maintenance of the toxin-coding M killer virions more than 30 MAK genes have been described which, when mutated, all result in an inability to propagate any toxin-coding M satellite virus. While some of the MAK gene products ful¢l completely di¡erent functions within the cell, like Mak1p as DNA topoisomerase I [88], Mak11p as essential membrane-associated protein with L-transducin sequence patterns [89] and the nuclear protein Mak16p which is essential for G1 exit within the cell cycle [90], the majority of the MAK genes are encoding ribosomal proteins and a¡ect 60S ribosomal subunit assembly [91^95]. Correspondingly, most mak mutations either lead to a dramatic decrease in the number of free 60S subunits or weaken ribosomal subunit joining and association [96]. Presumably, L-A is only providing its M satellites with Gag and Gag-Pol when the intracellular concentration of both viral proteins is su⁄cient. Therefore, any diminished Gag to Gag-Pol ratio (either caused by a reduced e⁄ciency of L-A(+)ssRNA translation or due to a limited availability of free 60S ribosomal subunits) would immediately result in a selective loss of the viral M dsRNA genome [97^99]. Mutations in any of six chromosomal SKI genes pheno- Cyaan Magenta Geel Zwart 268 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 Fig. 9. Model of the in vivo functions of the yeast SKI gene system. A: The indicated Ski protein complex Ski2cp (consisting of Ski2p, Ski3p, and Ski8p) is involved in translation repression of poly(A)3 RNA (left panel) and also proposed to act as a RNA helicase and/or cofactor for the exosomal 3P to 5P degradation of mRNA (right panel). This model is based on data mainly derived from the labs of Wickner [107,112,117], Parker [113,120] and Johnson [116]. B: Agar di¡usion assay (left) and Western blot analysis (right) illustrating the phenotypic consequences of a ski8 mutation in meiotic segregants derived from a cross of a K28 wild-type killer (SKI8) to a non-killer ski8 mutant. On sporulation of the heterozygous SKI8/ski8 diploid K28 killer, meiotic segregants from each tetrad show a 2:2 segregation of low and high killer activity (left) as well as low and high levels of secreted K28 toxin (right). (The Western blot shown to the right illustrates the amount of secreted K28 toxin present in the culture supernatant of the indicated ski8 and SKI8 segregants. Samples were separated by SDS^PAGE, electroblotted onto a PVDF membrane, and subsequently probed with a polyclonal antibody against the toxin’s L subunit. Arrows indicate the positions of the K/L heterodimeric protein toxin and of its tetrameric derivative (K/L)2 that forms spontaneously under conditions of a non-reducing SDS^PAGE.) typically suppress mak mutations in dsRNA-harboring killer strains [100]. These so-called superkiller genes (SKI) were initially identi¢ed from mutations that caused overexpression of the killer toxin, indicating that in ski mutants either M dsRNA copy number is increased or translation of the toxin-coding M(+)ssRNA transcript is more e¡ective [101,102]. Subsequent work on the yeast SKI gene system demonstrated that ¢ve out of six genes that have been characterized so far (SKI2, SKI3, SKI6, SKI7 and SKI8) are required in vivo to repress translation of poly- (A)3 RNAs [103,104], whereas SKI1 (which is identical to XRN1) encodes a 5P-exoribonuclease involved FEMSRE 748 1-8-02 in general RNA degradation and turnover [101,105^107]. SKI2 encodes a DEVH box protein of the RNA helicase family, Ski3p is a nucleoprotein, and Ski8p shows L-transducin domains [103,108,109]. All three Ski proteins (Ski2p, Ski3p, and Ski8p) are proposed to form a complex in vivo whose function might be to act as a RNA helicase and/or cofactor for the exosome complex through interaction with Ski7p [110] (reviewed in [111]). Ski6p is the only Ski protein for which it was directly shown that it is part of the eukaryotic exosome complex which ^ in analogy to the Escherichia coli homolog RNase PH ^ is responsible for the 3P-5P degradation of mRNA [112,113]. Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 The precise in vivo function of the SKI genes in yeast and in higher eukaryotes is still controversial. On one hand SKI genes may represent an antiviral defence system whose main function is to down-regulate dsRNA viral copy number and repress expression of non-poly(A) mRNAs, including L-A and M virus RNAs which both lack a 3P poly(A) tract and a 5P cap structure [104,107, 114,115]. Although it has been shown that ski2, ski3 and/or ski8 mutations derepress translation of non-poly(A) mRNAs in vivo, it has also been demonstrated that Ski proteins (like Ski6p/Rrp41p) can be part of the cytoplasmic exosome core complex and, therefore, function in a 3P mRNA degradation pathway [116,117]. This indicates that the yeast SKI genes have at least two important in vivo functions: in addition to their antiviral activity, Ski proteins are also involved in a more basic cell function like 3P mRNA degradation [103,104,113]. Furthermore, in addition to their activity in repressing translation of poly (A) minus mRNAs, the identi¢cation of SKI2 homologous genes in the human genome [118,119] further supports the idea that some of the Ski proteins are indeed part of a more general cellular system which is involved in the overall control of mRNA stability, structure and degradation [113,120,121]. Thus, based on the currently available data it can be concluded that the Ski proteins in yeast (and possibly their homologs in higher eukaryotes as well) have at least two important in vivo functions (Fig. 9): ¢rstly, they block translation of non-poly(A) mRNA possibly by a¡ecting 60S ribosomal subunit joining [104,117] and secondly, as part of the exosome complex they can also be involved in the 3P to 5P degradation of RNA [113,122]. In a superkiller yeast, loss of both functions of the SKI genes leads to a dramatic increase in toxin production and secretion, ¢nally resulting in a superkiller phenotype that originally gave the name to these genes. 2.7. Killer viruses in the non-conventional yeasts Hanseniaspora uvarum and Zygosaccharomyces bailii Besides the viral killer system in S. cerevisiae, toxin-coding mycoviruses have also been identi¢ed in killer strains of the non-conventional yeasts H. uvarum and Z. bailii [9,14,123]. Since the viral killer system in U. maydis has recently been reviewed [7], the following section will focus on the more recently identi¢ed killer viruses ZbV and HuV in Z. bailii and H. uvarum (Table 2). With respect to their structure and intracellular replication cycle, the dsRNA viruses ZbV and HuV are analogous to the S. cerevisiae viruses ScV-L-A and ScV-M, and it was shown that they can be successfully transferred to S. cerevisiae non-killer strains either by VLP transfection or by protoplast fusion, resulting in yeast transfectants and/or heterokaryons that stably express the corresponding killer phenotype [14,123]. However, in contrast to the S. cerevisiae toxins, killer toxins of non-Saccharomycetes ^ and in particular those secreted by virus-infected killer strains of the yeasts Z. bailii FEMSRE 748 1-8-02 269 and H. uvarum ^ show a broad-spectrum antimycotic potential, being not only lethal to a great variety of wood decay basidiomycetes and phytopathogenic fungi (like Heterobasidium, Postia, Serpula, Fusarium and/or Colletotrichum), but also to human pathogenic strains of Candida albicans and Sporothrix schenkii [4,9,124,125]. Lethality of the Z. bailii virus toxin Zygocin resembles that of certain bacteriocins and/or eukaryotic defensins, and involves disruption of cellular integrity by permeabilizing cytoplasmic membrane function. In analogy to the virally encoded preprotoxins in S. cerevisiae killer strains, the toxin-coding dsRNA genome (2.1 kb) in the yeast Z. bailii encodes the unprocessed precursor of the secreted and biologically active killer toxin. Due to its broad killing spectrum, Zygocin e¡ectively kills human pathogenic strains of C. albicans, Candida glabrata and Candida tropicalis, and the toxin is equally e¡ective against the yeast-like and the mycelial stages of the corresponding fungi after germ tube induction through cultivation in the presence of fetal calf serum. Furthermore, on a molar basis, Zygocin is even more e¡ective in killing human pathogenic yeasts than the frequently used antifungals clotrimazole or miconazole [125], indicating that it might be an attractive antifungal candidate, preferably for the treatment of topical mycoses. Undoubtedly, protein toxins secreted by nonconventional killer yeasts are highly promising candidates for use in the therapy of fungal infections, but further pharmacological studies are needed to demonstrate their antifungal potential in the future. 3. Applications During the last two decades, secreted killer toxins and toxin-producing killer yeasts have found several applications. For instance in the food and fermentation industries, killer yeasts have been used to combat contaminating wild-type yeasts which can occur during the production of wine, beer and bread [126^128]. Killer yeasts have also been used as bio-control agents in the preservation of foods [129], in the bio-typing of medically important pathogenic yeasts and yeast-like fungi [130^133], in the development of novel antimycotics for the treatment of human and animal fungal infections [134,135], and ¢nally in the ¢eld of recombinant DNA technology [136,137]. 3.1. Killer yeasts in wine fermentation Soon after the observation that toxin secreting killer yeasts can be the causal agent of stuck and/or protracted wine fermentations by antagonizing toxin-sensitive wine yeasts and thus negatively a¡ecting the sensory quality of the wine [138^141], many e¡orts were undertaken to use natural or ‘constructed’ killer strains as starter culture in beer and wine fermentations (for a recent review see [142]). Thus, in several studies it was demonstrated that Cyaan Magenta Geel Zwart 270 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 genetically modi¢ed wine yeasts producing di¡erent killer toxins simultaneously have an increased antagonistic ability in mixed yeast fermentations, exhibit a signi¢cantly broader killing activity and are thus capable of outcompeting potentially contaminating yeasts like Candida, Hanseniaspora, Kloeckera and Pichia in mixed culture [19,32, 127]. Thus, the use of recombinant DNA technology made it feasible to design multiple killer yeasts which confer a strong selective advantage on their yeast hosts; an interesting observation which might have application in industrial fermentations where contamination by killer yeasts is to be avoided. 3.2. Killer toxins as potential antifungals Antifungal proteins, peptides and their synthetic derivatives possess the potential for being used in the treatment of human fungal infections which dramatically increased during the last two decades, particularly in immunocompromised patients [143,144]. Powerful antifungal proteins are naturally produced by a diverse group of organisms including bacteria, fungi, insects, invertebrates and vertebrates, as well as plants (for a recent review see [145]). Within this group, secreted killer toxins mainly produced by non-Saccharomyces yeasts show a broad spectrum of killing activity against a great number of human and plant pathogens, including the anti-Pneumocystis carinii activity in a recently described killer strain of the yeast Pichia anomala [146]. In the search for novel and more selective antifungals, yeast and fungal cell wall components represent attractive targets, since these structures are usually restricted to yeasts and higher fungi and do not occur in mammalian cells [147,148]. Yeast cell walls predominantly consist of an outer layer of electron-dense mannoproteins, an additional glucan skeleton consisting of linear and branched L-1,3- and L-1,6-D-glucans, and ^ to a much lesser extent ^ of chitin [149]. Since each of these components was shown to act as the primary binding site and cell wall receptor for di¡erent yeast killer toxins [51^53,150], antifungal research is currently focusing on the possible use of yeast killer toxins as novel antifungals [125,145]. For killer toxins secreted by various strains of the yeast genus Hansenula it was demonstrated that these toxins not only bind to yeast cell wall components, but that they also strongly inhibit de novo L-1,3-D-glucan biosynthesis in yeast [151,152]. Therefore, killer toxin-based antifungals that e¡ectively target cell wall components which are restricted to yeasts and fungi should have a comparable selectivity as antibiotics inhibiting bacterial cell wall biosynthesis and/or structure. However, most yeast killer proteins exhibit their cytotoxic activity only within a narrow pH range and at temperatures between 20‡C and 30‡C and, therefore, yeast toxins are probably not suitable for oral and/or intravenous administration [153], but topical applications in the treatment of super¢cial lesions might well be possible [125,134,154]. FEMSRE 748 1-8-02 3.3. Killer toxin expression in transgenic plants As an attempt to engineer pathogen resistance in crop plants, transgenic plants are being constructed which are capable of producing substances such as polypeptides that are toxic to disease-causing pathogens like herbivorous insects [155]. In this respect, killer toxins which are naturally produced and secreted by virus-infected strains of the fungal pathogen U. maydis have been shown to be an attractive and unique model for the introduction of fungal resistance into tobacco plants [156]. Certain strains of the yeast U. maydis, which is also known as smut fungus that infects corn plants and induces tumor-like galls which ¢nally result in a black-colored smut formation, are naturally infected by a dsRNA virus which contains the genetic information for a secreted antifungal protein toxin [7,157^ 159]. From the three di¡erent U. maydis killer toxins that have been identi¢ed so far (KP1, KP4, and KP6), the K/L heterodimeric protein toxin KP6 as well as the monomeric KP4 toxin have been successfully expressed in tobacco plant cells under transcriptional control of the constitutive cauli£ower mosaic virus promoter [156,160]. In both cases it was demonstrated that the transgenic plants are capable of processing the corresponding virus toxin into its active and mature form undistinguishable from the authentic Ustilago toxin. However, the expression level of the KP6 toxin was too low to be useful as biological control of fungal pathogens. In contrast, the expression level of the recombinant KP4 toxin was so high, that a small piece of tobacco leaf secreted enough toxin to e¡ectively kill cells of a KP4-sensitive phytopathogenic fungus [156]. These data indicate that a systemic production of virally encoded yeast toxins in crop plants might provide a novel strategy to engineer biological control of fungal pathogens [161]. 3.4. Heterologous protein secretion via pptox secretion and processing signals Bakers’ yeast (S. cerevisiae) as well as ¢ssion yeast (Schizosaccharomyces pombe) are not only powerful systems for studying eukaryotic cell biology and cell cycle control, they are also becoming increasingly interesting in the fast growing ¢eld of heterologous protein production and secretion. Traditionally, expression and puri¢cation of heterologous proteins is accomplished using prokaryotic systems such as E. coli or Bacillus subtilis. However, when expression of eukaryotic proteins is desired, bacterial systems often turn out to be ine¡ective hosts because of their limited capacity to perform multi-step post-translational modi¢cations such as protein N-glycosylation, phosphorylation and acetylation [162]. Therefore, unicellular eukaryotes such as the yeasts S. cerevisiae, Pichia pastoris, Yarrowia lipolytica, Hansenula polymorpha, K. lactis and S. pombe have become attractive hosts for the expression of heterologous proteins (for review see [163]). In contrast to the great majority of foreign proteins Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 271 Fig. 10. Schematic drawing of a preprotoxin-based DNA construct for the e⁄cient processing and secretion of heterologous proteins in the ¢ssion yeast S. pombe. A: Structure of a chromosomally integrating K28 preprotoxin expression cassette for thiamine-regulated processing and secretion of foreign proteins expressed under transcriptional control of the ¢ssion yeast nmt1 promoter (S/P, secretion and processing signal derived from K28 pptox; P/ Tnmt1 , nmt1 promoter and transcription termination sequence; modi¢ed after [137]). B: Western analysis (left panel) of secreted GFP in culture supernatants of ¢ssion yeast transformants expressing the processing and secretion vector shown in A. Fluorescence microscopy (right panel) of the same cells expressing and secreting recombinant GFP. expressed within the host cell cytosol, only a few proteins have been successfully secreted. Secretory proteins contain a hydrophobic, N-terminal signal sequence that directs entry into the eukaryotic secretion pathway, with the most critical step being protein import into the lumen of the ER and subsequent sorting to the Golgi network. During the past decade, an increasing number of medically and/or pharmaceutically interesting secretory proteins (such as mouse K-amylase, human antithrombin III or placental alkaline phosphatase) have been expressed as extracellular proteins by using homologous secretion signals either derived from yeast invertase, acid phosphatase, pheromone P-factor, or from the plasmid-driven killer toxin of K. lactis [164^166]. More recently it has also been shown that the secretion and processing signal derived from the S. cerevisiae ScV-M28 killer virus is fully functional in ¢ssion yeast (S. pombe) and can be used to target foreign proteins for secretion into the extracellular medium [137]. Moreover, heterologous protein secretion driven by the K28 pptox processing and secretion signal is highly e⁄cient as shown for the green £uorescent protein (GFP; Fig. 10), making ¢ssion yeast an attractive host for the processing and secretion of foreign proteins. There- FEMSRE 748 1-8-02 fore, preprotoxin-based vectors might be an attractive means for e⁄cient processing and high level secretion of heterologous proteins in yeast [137]. 4. Concluding remarks Over the years much has been learned about eukaryotic cell biology by studying killer strains of the yeast S. cerevisiae and their virally encoded protein toxins. The viral killer system in yeast not only developed into a model system to study general eukaryotic cell biology but also to analyze more speci¢c aspects of yeast virology and virus^host cell interactions. Besides that, killer toxins and toxin-producing killer yeasts have also found applications in the food and fermentation industries, where they proved to be e¡ective in combating contaminating wild-type yeasts during wine, beer and bread fermentation. Toxinsecreting killer yeasts have also been used as bio-control agents in the preservation of foods, in the bio-typing of medically important pathogenic yeasts and fungi, in the development of novel antimycotics for the treatment of fungal infections, and in the ¢eld of recombinant DNA Cyaan Magenta Geel Zwart 272 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 technology. All this had been possible by a detailed analysis of at least some of the toxin-secreting killer strains that have been found among around 700 yeast species described to date. Considering that most of the killer yeasts have still not been thoroughly investigated, and that the great majority of yeast and fungi have not yet been discovered, there is a great potential of new yeast species in natural habitats waiting to be characterized ^ let’s do it. [15] [16] [17] [18] [19] Acknowledgements Continuous support of our killer research through grants from the Deutsche Forschungsgemeinschaft is greatly appreciated as well as the many contributions of coworkers and collaborators. In particular, we wish to thank Donald Tipper for critically reading the manuscript, Reed Wickner, Howard Bussey, Roy Parker, Thomas Sommer, Dieter Wolf, Ferdinand Radler, and all past and present members of the Schmitt lab. [20] [21] [22] [23] References [1] Bevan, E.A. and Makower, M. (1963) The physiological basis of the killer-character in yeast. Proc. Int. Congr. Genet. 11203. [2] Young, T.W. and Yagiu, M. (1978) A comparison of the killer character in di¡erent yeasts and its classi¢cation. Antonie Van Leeuwenhoek 44, 59^77. [3] Starmer, W.T., Ganter, P.F., Aberdeen, V., Lachance, M.A. and Pha¡, H.J. (1987) The ecological role of killer yeasts in natural communities of yeasts. Can. J. Microbiol. 33, 783^796. [4] Hodgson, V.J., Button, D. and Walker, G.M. (1995) Anti-Candida activity of a novel killer toxin from the yeast Williopsis mrakii. Microbiology 141, 2003^2012. [5] Bostian, K.A., Sturgeon, J.A. and Tipper, D.J. (1980) Encapsidation of yeast killer double-stranded ribonucleic acids: dependence of M on L. J. Bacteriol. 143, 463^470. [6] Scha¡rath, R. and Breunig, K.D. (2000) Genetics and molecular physiology of the yeast Kluyveromyces lactis. Fungal Genet. Biol. 30, 173^190. [7] Magliani, W., Conti, S., Gerloni, M., Bertolotti, D. and Polonelli, L. (1997) Yeast killer systems. Clin. Microbiol. Rev. 10, 369^400. [8] Wickner, R.B. (1996) Double-stranded RNA viruses of Saccharomyces cerevisiae. Microbiol. Rev. 60, 250^265. [9] Schmitt, M.J., Poravou, O., Trenz, K. and Rehfeldt, K. (1997) Unique double-stranded RNAs responsible for the anti-Candida activity of the yeast Hanseniaspora uvarum. J. Virol. 71, 8852^8855. [10] Wickner, R.B., Bussey, H., Fujimura, T. and Esteban, R. (1995) Viral RNA and the killer phenomenon of Saccharomyces. In: The Mycota, Genetics and Biotechnology (Ku«ck, Ed.), pp. 211^226. Springer Verlag, Berlin. [11] Wickner, R.B. (1989) Yeast virology. FASEB J. 3, 2257^2265. [12] El-Sherbeini, M. and Bostian, K.A. (1987) Viruses in fungi: infection of yeast with the K1 and K2 killer virus. Proc. Natl. Acad. Sci. USA 84, 4293^4297. [13] Schmitt, M.J. and Tipper, D.J. (1990) K28, a unique double-stranded RNA killer virus of Saccharomyces cerevisiae. Mol. Cell. Biol. 10, 4807^4815. [14] Schmitt, M.J. and Neuhausen, F. (1994) Killer toxin-secreting dou- FEMSRE 748 1-8-02 [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] ble-stranded RNA mycoviruses in the yeasts Hanseniaspora uvarum and Zygosaccharomyces bailii. J. Virol. 68, 1765^1772. Esteban, R. and Wickner, R.B. (1986) Three di¡erent M1 RNA-containing viruslike particle types in Saccharomyces cerevisiae : in vitro M1 double-stranded RNA synthesis. Mol. Cell. Biol. 6, 1552^1561. Fujimura, T., Ribas, J.C., Makhov, A.M. and Wickner, R.B. (1992) Pol of gag-pol fusion protein required for encapsidation of viral RNA of yeast L-A virus. Nature 359, 746^749. Sturley, S.L., Elliot, Q., LeVitre, J., Tipper, D.J. and Bostian, K.A. (1986) Mapping of functional domains within the Saccharomyces cerevisiae type 1 killer preprotoxin. EMBO J. 5, 3381^3389. Young, T.W. (1987) In: The Yeasts (Rose, A.H. and Harrison, J.S., Eds.), pp. 131^164. Academic Press, New York. Bussey, H., Vernet, T. and Sdicu, A.M. (1988) Mutual antagonism among killer yeasts: competition between K1 and K2 killers and a novel cDNA-based K1-K2 killer strain of Saccharomyces cerevisiae. Can. J. Microbiol. 34, 38^44. Baker, T.S., Olson, N.H. and Fuller, S.D. (1999) Adding the third dimension to virus life cycles : three-dimensional reconstruction of icosahedral viruses from cryo-electron micrographs. Microbiol. Mol. Biol. Rev. 63, 862^922. Cheng, R.H., Caston, J.R., Wang, G.J., Gu, F., Smith, T.J., Baker, T.S., Bozarth, R.F., Trus, B.L., Cheng, N., Wickner, R.B., et al. (1994) Fungal virus capsids, cytoplasmic compartments for the replication of double-stranded RNA, formed as icosahedral shells of asymmetric Gag dimers. J. Mol. Biol. 244, 255^258. Wickner, R.B. (1996) Prions and RNA viruses of Saccharomyces cerevisiae. Annu. Rev. Genet. 30, 109^139. Caston, J.R., Trus, B.L., Booy, F.P., Wickner, R.B., Wall, J.S. and Steven, A.C. (1997) Structure of L-A virus: a specialized compartment for the transcription and replication of double-stranded RNA. J. Cell Biol. 138, 975^985. Icho, T. and Wickner, R.B. (1989) The double-stranded RNA genome of yeast virus L-A encodes its own putative RNA polymerase by fusing two open reading frames. J. Biol. Chem. 264, 6716^6723. Bruenn, J.A. (1991) Relationships among the positive strand and double-strand RNA viruses as viewed through their RNA-dependent RNA polymerases. Nucleic Acids Res. 19, 217^226. Hayman, G.T. and Bolen, P.L. (1991) Linear DNA plasmids of Pichia inositovora are associated with a novel killer toxin activity. Curr. Genet. 19, 389^393. Huan, B.F., Shen, Y.Q. and Bruenn, J.A. (1991) In vivo mapping of a sequence required for interference with the yeast killer virus. Proc. Natl. Acad. Sci. USA 88, 1271^1275. Dinman, J.D., Icho, T. and Wickner, R.B. (1991) A 31 ribosomal frameshift in a double-stranded RNA virus of yeast forms a gag-pol fusion protein. Proc. Natl. Acad. Sci. USA 88, 174^178. Fujimura, T. and Wickner, R.B. (1988) Gene overlap results in a viral protein having an RNA binding domain and a major coat protein domain. Cell 55, 663^671. Lopinski, J.D., Dinman, J.D. and Bruenn, J.A. (2000) Kinetics of ribosomal pausing during programmed 31 translational frameshifting. Mol. Cell. Biol. 20, 1095^1103. Fujimura, T., Esteban, R., Esteban, L.M. and Wickner, R.B. (1990) Portable encapsidation signal of the L-A double-stranded RNA virus of S. cerevisiae. Cell 62, 819^828. Schmitt, M.J. and Schernikau, G. (1997) Construction of a cDNAbased K1/K2/K28 triple killer strain of Saccharomyces cerevisiae. Food Technol. Biotechnol. 35, 281^285. Dawe, A.L. and Nuss, D.L. (2001) Hypoviruses and chestnut blight : exploiting viruses to understand and modulate fungal pathogenesis. Annu. Rev. Genet. 35, 1^29. Radler, F., Herzberger, S., Schonig, I. and Schwarz, P. (1993) Investigation of a killer strain of Zygosaccharomyces bailii. J. Gen. Microbiol. 139, 495^500. Ribas, J.C., Fujimura, T. and Wickner, R.B. (1994) Essential RNA binding and packaging domains of the Gag-Pol fusion protein of the Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] L-A double-stranded RNA virus of Saccharomyces cerevisiae. J. Biol. Chem. 269, 28420^28428. Ribas, J.C. and Wickner, R.B. (1992) RNA-dependent RNA polymerase consensus sequence of the L-A double-stranded RNA virus: de¢nition of essential domains. Proc. Natl. Acad. Sci. USA 89, 2185^ 2189. Eisfeld, K., Ri¡er, F., Mentges, J. and Schmitt, M.J. (2000) Endocytotic uptake and retrograde transport of a virally encoded killer toxin in yeast. Mol. Microbiol. 37, 926^940. Schmitt, M.J. and Eisfeld, K. (1999) Killer viruses in S. cerevisiae and their general importance in understanding eukaryotic cell biology. Recent Res. Dev. Virol. 1, 525^545. Schmitt, M.J. and Tipper, D.J. (1995) Sequence of the M28 dsRNA: preprotoxin is processed to an alpha/beta heterodimeric protein toxin. Virology 213, 341^351. Ri¡er, F., Eisfeld, K., Breinig, F. and Schmitt, M.J. (2002) Mutational analysis of K28 preprotoxin processing in the yeast Saccharomyces cerevisiae. Microbiology 148, 1317^1328. Wickner, R.B. and Leibowitz, M.J. (1976) Two chromosomal genes required for killing expression in killer strains of Saccharomyces cerevisiae. Genetics 82, 429^442. Fuller, R.S., Brake, A.J. and Thorner, J. (1989) Intracellular targeting and structural conservation of a prohormone-processing endoprotease. Science 246, 482^486. Fuller, R.S., Brake, A. and Thorner, J. (1989) Yeast prohormone processing enzyme (KEX2 gene product) is a Ca2þ -dependent serine protease. Proc. Natl. Acad. Sci. USA 86, 1434^1438. Redding, K., Holcomb, C. and Fuller, R.S. (1991) Immunolocalization of Kex2 protease identi¢es a putative late Golgi compartment in the yeast Saccharomyces cerevisiae. J. Cell Biol. 113, 527^538. Dmochowska, A., Dignard, D., Henning, D., Thomas, D.Y. and Bussey, H. (1987) Yeast KEX1 gene encodes a putative protease with a carboxypeptidase B-like function involved in killer toxin and alpha-factor precursor processing. Cell 50, 573^584. Bryant, N.J. and Boyd, A. (1993) Immunoisolation of Kex2p-containing organelles from yeast demonstrates colocalisation of three processing proteinases to a single Golgi compartment. J. Cell Sci. 106, 815^822. Shilton, B.H., Thomas, D.Y. and Cygler, M. (1997) Crystal structure of Kex1deltap, a prohormone-processing carboxypeptidase from Saccharomyces cerevisiae. Biochemistry 36, 9002^9012. Bostian, K.A., Elliott, Q., Bussey, H., Burn, V., Smith, A. and Tipper, D.J. (1984) Sequence of the preprotoxin dsRNA gene of type I killer yeast: multiple processing events produce a two-component toxin. Cell 36, 741^751. Dignard, D., Whiteway, M., Germain, D., Tessier, D. and Thomas, D.Y. (1991) Expression in yeast of a cDNA copy of the K2 killer toxin gene. Mol. Gen. Genet. 227, 127^136. Bussey, H. (1991) K1 killer toxin, a pore-forming protein from yeast. Mol. Microbiol. 5, 2339^2343. Hutchins, K. and Bussey, H. (1983) Cell wall receptor for yeast killer toxin: involvement of (1 leads to 6)-beta-D-glucan. J. Bacteriol. 154, 161^169. Schmitt, M. and Radler, F. (1988) Molecular structure of the cell wall receptor for killer toxin KT28 in Saccharomyces cerevisiae. J. Bacteriol. 170, 2192^2196. Schmitt, M. and Radler, F. (1987) Mannoprotein of the yeast cell wall as primary receptor for the killer toxin of Saccharomyces cerevisiae strain 28. J. Gen. Microbiol. 133, 3347^3354. Boone, C., Sommer, S.S., Hensel, A. and Bussey, H. (1990) Yeast KRE genes provide evidence for a pathway of cell wall beta-glucan assembly. J. Cell Biol. 110, 1833^1843. Schmitt, M.J. and Radler, F. (1990) Blockage of cell wall receptors for yeast killer toxin KT28 with antimannoprotein antibodies. Antimicrob. Agents Chemother. 34, 1615^1618. Breinig, F., Tipper, D.J. and Schmitt, M.J. (2002) Kre1p, the plasma membrane receptor for the yeast K1 viral toxin. Cell 108, 395^405. FEMSRE 748 1-8-02 273 [57] de la Pena, P., Barros, F., Gascon, S., Lazo, P.S. and Ramos, S. (1981) E¡ect of yeast killer toxin on sensitive cells of Saccharomyces cerevisiae. J. Biol. Chem. 256, 10420^10525. [58] Martinac, B., Zhu, H., Kubalski, A., Zhou, X.L., Culbertson, M., Bussey, H. and Kung, C. (1990) Yeast K1 killer toxin forms ion channels in sensitive yeast spheroplasts and in arti¢cial liposomes. Proc. Natl. Acad. Sci. USA 87, 6228^6232. [59] Ahmed, A., Sesti, F., Ilan, N., Shih, T.M., Sturley, S.L. and Goldstein, S.A. (1999) A molecular target for viral killer toxin: TOK1 potassium channels. Cell 99, 283^291. [60] Schmitt, M.J., Klavehn, P., Wang, J., Schonig, I. and Tipper, D.J. (1996) Cell cycle studies on the mode of action of yeast K28 killer toxin. Microbiology 142, 2655^2662. [61] Schmitt, M.J. (1995) Cloning and expression of a cDNA copy of the viral K28 killer toxin gene in yeast. Mol. Gen. Genet. 246, 236^246. [62] Yoshida, T., Chen, C.C., Zhang, M.S. and Wu, H.C. (1991) Disruption of the Golgi apparatus by brefeldin A inhibits the cytotoxicity of ricin, modeccin, and Pseudomonas toxin. Exp. Cell Res. 192, 389^395. [63] Chaudhary, V.K., Jinno, Y., Gallo, M.G., FitzGerald, D. and Pastan, I. (1990) Mutagenesis of Pseudomonas exotoxin in identi¢cation of sequences responsible for the animal toxicity. J. Biol. Chem. 265, 16306^16310. [64] Pelham, H.R.B., Roberts, L.M. and Lord, J.M. (1992) Toxin entry: how reversible is the secretory pathway ? Trends Cell Biol. 2, 183^185. [65] Beswick, V., Brodsky, J.L., Kepes, F., Neumann, J.M., Sanson, A. and Garrigos, M. (1998) Expression, puri¢cation, and characterization of Sss1p, an essential component of the yeast Sec61p protein translocation complex. Protein Expr. Purif. 13, 423^432. [66] Pilon, M., Schekman, R. and Romisch, K. (1997) Sec61p mediates export of a misfolded secretory protein from the endoplasmic reticulum to the cytosol for degradation. EMBO J. 16, 4540^4548. [67] Plemper, R.K., Bohmler, S., Bordallo, J., Sommer, T. and Wolf, D.H. (1997) Mutant analysis links the translocon and BiP to retrograde protein transport for ER degradation. Nature 388, 891^895. [68] Schmitt, M., Brendel, M., Schwarz, R. and Radler, F. (1989) Inhibition of DNA synthesis in Saccharomyces cerevisiae by yeast toxin K28. J. Gen. Microbiol. 135, 1529^1535. [69] Tipper, D.J. and Schmitt, M.J. (1991) Yeast dsRNA viruses: replication and killer phenotypes. Mol. Microbiol. 5, 2331^2338. [70] Bussey, H., Saville, D., Greene, D., Tipper, D.J. and Bostian, K.A. (1983) Secretion of Saccharomyces cerevisiae killer toxin: processing of the glycosylated precursor. Mol. Cell. Biol. 3, 1362^1370. [71] Bussey, H., Sacks, W., Galley, D. and Saville, D. (1982) Yeast killer plasmid mutations a¡ecting toxin secretion and activity and toxin immunity function. Mol. Cell. Biol. 2, 346^354. [72] Boone, C., Bussey, H., Greene, D., Thomas, D.Y. and Vernet, T. (1986) Yeast killer toxin: site-directed mutations implicate the precursor protein as the immunity component. Cell 46, 105^113. [73] Schmitt, M.J. and Tipper, D.J. (1992) Genetic analysis of maintenance and expression of L and M double-stranded RNAs from yeast killer virus K28. Yeast 8, 373^384. [74] Zhu, Y.S., Kane, J., Zhang, X.Y., Zhang, M. and Tipper, D.J. (1993) Role of the gamma component of preprotoxin in expression of the yeast K1 killer phenotype. Yeast 9, 251^266. [75] Schmitt, M.J. and Compain, P. (1995) Killer-toxin-resistant kre12 mutants of Saccharomyces cerevisiae : genetic and biochemical evidence for a secondary K1 membrane receptor. Arch. Microbiol. 164, 435^443. [76] Wickner, R.B. (1993) Host control of yeast dsRNA virus propagation and expression. Trends Microbiol. 1, 294^299. [77] Douglas, C.M., Sturley, S.L. and Bostian, K.A. (1988) Role of protein processing, intracellular tra⁄cking and endocytosis in production of and immunity to yeast killer toxin. Eur. J. Epidemiol. 4, 400^408. [78] Lolle, S.J. and Bussey, H. (1986) In vivo evidence for posttranslational translocation and signal cleavage of the killer preprotoxin of Saccharomyces cerevisiae. Mol. Cell. Biol. 6, 4274^4280. [79] Zhu, Y.S., Zhang, X.Y., Cartwright, C.P. and Tipper, D.J. (1992) Cyaan Magenta Geel Zwart 274 [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 Kex2-dependent processing of yeast K1 killer preprotoxin includes cleavage at ProArg-44. Mol. Microbiol. 6, 511^520. Sommer, S.S. and Wickner, R.B. (1982) Yeast L dsRNA consists of at least three distinct RNAs; evidence that the non-Mendelian genes [HOK], [NEX] and [EXL] are on one of these dsRNAs. Cell 31, 429^ 441. Tercero, J.C. and Wickner, R.B. (1992) MAK3 encodes an N-acetyltransferase whose modi¢cation of the L-A gag NH2 terminus is necessary for virus particle assembly. J. Biol. Chem. 267, 20277^20281. Tercero, J.C., Dinman, J.D. and Wickner, R.B. (1993) Yeast MAK3 N-acetyltransferase recognizes the N-terminal four amino acids of the major coat protein (gag) of the L-A double-stranded RNA virus. J. Bacteriol. 175, 3192^3194. Rigaut, G., Shevchenko, A., Rutz, B., Wilm, M., Mann, M. and Seraphin, B. (1999) A generic protein puri¢cation method for protein complex characterization and proteome exploration. Nat. Biotechnol. 17, 1030^1032. Polevoda, B. and Sherman, F. (2001) NatC Nalpha-terminal acetyltransferase of yeast contains three subunits, Mak3p, Mak10p, and Mak31p. J. Biol. Chem. 276, 20154^20159. Lee, Y.J. and Wickner, R.B. (1992) MAK10, a glucose-repressible gene necessary for replication of a dsRNA virus of Saccharomyces cerevisiae, has T cell receptor alpha-subunit motifs. Genetics 132, 87^ 96. Fujimura, T. and Wickner, R.B. (1987) L-A double-stranded RNA viruslike particle replication cycle in Saccharomyces cerevisiae : particle maturation in vitro and e¡ects of mak10 and pet18 mutations. Mol. Cell. Biol. 7, 420^426. Fujimura, T., Esteban, R. and Wickner, R.B. (1986) In vitro L-A double-stranded RNA synthesis in virus-like particles from Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 83, 4433^4437. Thrash, C., Voelkel, K., DiNardo, S. and Sternglanz, R. (1984) Identi¢cation of Saccharomyces cerevisiae mutants de¢cient in DNA topoisomerase I activity. J. Biol. Chem. 259, 1375^1377. Icho, T. and Wickner, R.B. (1988) The MAK11 protein is essential for cell growth and replication of M double-stranded RNA and is apparently a membrane-associated protein. J. Biol. Chem. 263, 1467^ 1475. Wickner, R.B. (1987) MKT1, a nonessential Saccharomyces cerevisiae gene with a temperature-dependent e¡ect on replication of M2 double-stranded RNA. J. Bacteriol. 169, 4941^4945. Carroll, K. and Wickner, R.B. (1995) Translation and M1 doublestranded RNA propagation: MAK18 = RPL41B and cycloheximide curing. J. Bacteriol. 177, 2887^2891. Ohtake, Y. and Wickner, R.B. (1995) KRB1, a suppressor of mak7-1 (a mutant RPL4A), is RPL4B, a second ribosomal protein L4 gene, on a fragment of Saccharomyces chromosome XII. Genetics 140, 129^137. Wickner, R.B., Ridley, S.P., Fried, H.M. and Ball, S.G. (1982) Ribosomal protein L3 is involved in replication or maintenance of the killer double-stranded RNA genome of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 79, 4706^4708. Wickner, R.B. (1988) Host function of MAK16: G1 arrest by a mak16 mutant of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 85, 6007^6011. Wickner, R.B. (1992) Double-stranded and single-stranded RNA viruses of Saccharomyces cerevisiae. Annu. Rev. Microbiol. 46, 347^ 375. Ohtake, Y. and Wickner, R.B. (1995) Yeast virus propagation depends critically on free 60S ribosomal subunit concentration. Mol. Cell. Biol. 15, 2772^2781. Valle, R.P. and Wickner, R.B. (1993) Elimination of L-A doublestranded RNA virus of Saccharomyces cerevisiae by expression of gag and gag-pol from an L-A cDNA clone. J. Virol. 67, 2764^2771. Dinman, J.D. and Wickner, R.B. (1992) Ribosomal frameshifting e⁄ciency and gag/gag-pol ratio are critical for yeast M1 doublestranded RNA virus propagation. J. Virol. 66, 3669^3676. FEMSRE 748 1-8-02 [99] Dinman, J.D. and Wickner, R.B. (1994) Translational maintenance of frame: mutants of Saccharomyces cerevisiae with altered 31 ribosomal frameshifting e⁄ciencies. Genetics 136, 75^86. [100] Wickner, R.B. and Toh-e, A. (1982) [HOK], a new yeast non-Mendelian trait, enables a replication-defective killer plasmid to be maintained. Genetics 100, 159^174. [101] Toh, E.A., Guerry, P. and Wickner, R.B. (1978) Chromosomal superkiller mutants of Saccharomyces cerevisiae. J. Bacteriol. 136, 1002^1007. [102] Ridley, S.P., Sommer, S.S. and Wickner, R.B. (1984) Superkiller mutations in Saccharomyces cerevisiae suppress exclusion of M2 double-stranded RNA by L-A-HN and confer cold sensitivity in the presence of M and L-A-HN. Mol. Cell. Biol. 4, 761^770. [103] Widner, W.R. and Wickner, R.B. (1993) Evidence that the SKI antiviral system of Saccharomyces cerevisiae acts by blocking expression of viral mRNA. Mol. Cell. Biol. 13, 4331^4341. [104] Masison, D.C., Blanc, A., Ribas, J.C., Carroll, K., Sonenberg, N. and Wickner, R.B. (1995) Decoying the cap-mRNA degradation system by a double-stranded RNA virus and poly(A)3 mRNA surveillance by a yeast antiviral system. Mol. Cell. Biol. 15, 2763^2771. [105] Toh, E.A. and Wickner, R.B. (1980) ‘Superkiller’ mutations suppress chromosomal mutations a¡ecting double-stranded RNA killer plasmid replication in Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 77, 527^530. [106] Johnson, A.W. and Kolodner, R.D. (1995) Synthetic lethality of sep1 (xrn1) ski2 and sep1 (xrn1) ski3 mutants of Saccharomyces cerevisiae is independent of killer virus and suggests a general role for these genes in translation control. Mol. Cell. Biol. 15, 2719^ 2727. [107] Benard, L., Carroll, K., Valle, R.C., Masison, D.C. and Wickner, R.B. (1999) The ski7 antiviral protein is an EF1-alpha homolog that blocks expression of non-poly(A) mRNA in Saccharomyces cerevisiae. J. Virol. 73, 2893^2900. [108] Rhee, S.K., Icho, T. and Wickner, R.B. (1989) Structure and nuclear localization signal of the SKI3 antiviral protein of Saccharomyces cerevisiae. Yeast 5, 149^158. [109] Matsumoto, Y., Sarkar, G., Sommer, S.S. and Wickner, R.B. (1993) A yeast antiviral protein, SKI8, shares a repeated amino acid sequence pattern with beta-subunits of G proteins and several other proteins. Yeast 9, 43^51. [110] Araki, Y., Takahashi, S., Kobayashi, T., Kajiho, H., Hoshino, S. and Katada, T. (2001) Ski7p G protein interacts with the exosome and the Ski complex for 3P-to-5P mRNA decay in yeast. EMBO J. 20, 4684^4693. [111] Butler, J.S. (2002) The yin and yang of the exosome. Trends Cell Biol. 12, 90^96. [112] Benard, L., Carroll, K., Valle, R.C. and Wickner, R.B. (1998) Ski6p is a homolog of RNA-processing enzymes that a¡ects translation of non-poly(A) mRNAs and 60S ribosomal subunit biogenesis. Mol. Cell. Biol. 18, 2688^2696. [113] Jacobs, J.S., Anderson, A.R. and Parker, R.P. (1998) The 3P to 5P degradation of yeast mRNAs is a general mechanism for mRNA turnover that requires the SKI2 DEVH box protein and 3P to 5P exonucleases of the exosome complex. EMBO J. 17, 1497^1506. [114] Thiele, D.J., Hannig, E.M. and Leibowitz, M.J. (1984) Genome structure and expression of a defective interfering mutant of the killer virus of yeast. Virology 137, 20^31. [115] Searfoss, A.M. and Wickner, R.B. (2000) 3P poly(A) is dispensable for translation. Proc. Natl. Acad. Sci. USA 97, 9133^9137. [116] Brown, J.T., Bai, X. and Johnson, A.W. (2000) The yeast antiviral proteins Ski2p, Ski3p, and Ski8p exist as a complex in vivo. RNA 6, 449^457. [117] Searfoss, A., Dever, T.E. and Wickner, R. (2001) Linking the 3P poly(A) tail to the subunit joining step of translation initiation: relations of Pab1p, eukaryotic translation initiation factor 5b (Fun12p), and Ski2p-Slh1p. Mol. Cell. Biol. 21, 4900^4908. [118] Dangel, A.W., Shen, L., Mendoza, A.R., Wu, L.C. and Yu, C.Y. Cyaan Magenta Geel Zwart M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] (1995) Human helicase gene SKI2W in the HLA class III region exhibits striking structural similarities to the yeast antiviral gene SKI2 and to the human gene KIAA0052: emergence of a new gene family. Nucleic Acids Res. 23, 2120^2126. Lee, S.G., Lee, I., Park, S.H., Kang, C. and Song, K. (1995) Identi¢cation and characterization of a human cDNA homologous to yeast SKI2. Genomics 25, 660^666. van Hoof, A. and Parker, R. (1999) The exosome: a proteasome for RNA? Cell 99, 347^350. Van Hoof, C., Janssens, V., De Baere, I., de Winde, J.H., Winderickx, J., Dumortier, F., Thevelein, J.M., Merlevede, W. and Goris, J. (2000) The Saccharomyces cerevisiae homologue YPA1 of the mammalian phosphotyrosyl phosphatase activator of protein phosphatase 2A controls progression through the G1 phase of the yeast cell cycle. J. Mol. Biol. 302, 103^120. Wang, Z. and Kiledjian, M. (2001) Functional link between the mammalian exosome and mRNA decapping. Cell 107, 751^762. Zorg, J., Kilian, S. and Radler, F. (1988) Killer toxin producing strains of the yeasts Hanseniaspora uvarum and Pichia kluyveri. Arch. Microbiol. 149, 261^267. Radler, F., Schmitt, M.J. and Meyer, B. (1990) Killer toxin of Hanseniaspora uvarum. Arch. Microbiol. 154, 175^178. Theisen, S., Molkenau, E. and Schmitt, M.J. (2000) Wicaltin, a new protein toxin secreted by the yeast Williopsis californica and its broad-spectrum antimycotic potential. J. Microbiol. Biotechnol. 10, 547^550. Young, T.W. (1981) The genetic manipulation of killer character into brewing yeast. J. Inst. Brew. 87, 292^295. Boone, C., Scidu, A.M., Wagner, J., Degre, R., Sanchez, C. and Bussey, H. (1990) Integration of the yeast K1 killer toxin gene into the genome of marked wine yeasts and its e¡ect on vini¢cation. Am. J. Enol. Vitic. 41, 37^42. Bortol, A., Nudel, C., Fraile, E., de Torres, R., Giuletti, A., Spencer, J.F.T. and Spencer, D. (1986) Isolation of the yeast with killer activity and its breeding with an industrial brewing strain by protoplast fusion. Appl. Microbiol. Biotechnol. 24, 414^416. Palpacelli, V., Ciani, M. and Rosini, G. (1991) Activity of di¡erent ‘killer’ yeasts on strains of yeast species undesirable in the food industry. FEMS Microbiol. Lett. 68, 75^78. Morace, G., Manzara, S., Dettori, G., Fanti, F., Conti, S., Campani, L., Polonelli, L. and Chezzi, C. (1989) Biotyping of bacterial isolates using the yeast killer system. Eur. J. Epidemiol. 5, 303^310. Boekhout, T. and Scorzetti, G. (1997) Di¡erential killer toxin sensitivity patterns of varieties of Cryptococcus neoformans. J. Med. Vet. Mycol. 35, 147^149. Buzzini, P. and Martini, A. (2001) Discrimination between Candida albicans and other pathogenic species of the genus Candida by their di¡erential sensitivities to toxins of a panel of killer yeasts. J. Clin. Microbiol. 39, 3362^3364. Buzzini, P. and Martini, A. (2001) Large-scale screening of selected Candida maltosa, Debaryomyces hansenii and Pichia anomala killer toxin activity against pathogenic yeasts. Med. Mycol. 39, 479^482. Polonelli, L., Lorenzini, R., De Bernardis, F. and Morace, G. (1986) Potential therapeutic e¡ect of yeast killer toxin. Mycopathologia 96, 103^107. Polonelli, L., Conti, S., Gerloni, M., Magliani, W., Castagnola, M., Morace, G. and Chezzi, C. (1991) ‘Antibiobodies’ : antibiotic-like anti-idiotypic antibodies. J. Med. Vet. Mycol. 29, 235^242. Meinhardt, F., Kempken, F., Kamper, J. and Esser, K. (1990) Linear plasmids among eukaryotes: fundamentals and application. Curr. Genet. 17, 89^95. Heintel, T., Zagorc, T. and Schmitt, M.J. (2001) Expression, processing and high level secretion of a virus toxin in ¢ssion yeast. Appl. Microbiol. Biotechnol. 56, 165^172. van Vuuren, H.J.J. and Wing¢eld, B.D. (1986) Killer yeasts: cause of stuck fermentations in a wine cellar. S. Afr. J. Enol. Vitic. 7, 113^ 118. FEMSRE 748 1-8-02 275 [139] Longo, E., Vela¤zques, J.B., Cansado, J., Calo, P. and Villa, T.G. (1990) Role of killer e¡ect in fermentations conducted by mixed cultures of Saccharomyces cerevisiae. FEMS Microbiol. Lett. 71, 331^336. [140] Carrau, F.M., Neirotti, E. and Gioia, O. (1993) Stuck wine fermentations: e¡ect of killer/sensitive yeast interactions. J. Ferment. Bioeng. 76, 67^69. [141] Franken, D.B., Ariatti, M., Pretorius, I.S. and Gupthar, A.S. (1998) Genetic and fermentation properties of the K2 killer yeast, Saccharomyces cerevisiae T206. Antonie Van Leeuwenhoek 73, 263^ 269. [142] Vondrejs, V., Janderova, B. and Valasek, L. (1996) Yeast killer toxin K1 and its exploitation in genetic manipulations. Folia Microbiol. 41, 379^393. [143] Fox, J.L. (1993) Fungal infection rates are increasing. ASM News 59, 515^518. [144] Herbrecht, R. (1996) The changing epidemiology of fungal infections: are the lipid-forms of amphotericin B an advance? Eur. J. Haematol. Suppl. 57, 5712^5717. [145] Selitrenniko¡, C.P. (2001) Antifungal proteins. Appl. Environ. Microbiol. 67, 2883^2894. [146] Seguy, N., Polonelli, L., Dei-Cas, E. and Cailliez, J.C. (1998) E¡ect of a killer toxin of Pichia anomala to Pneumocystis. Perspectives in the control of pneumocystosis. FEMS Immunol. Med. Microbiol. 22, 145^149. [147] Hector, R.F. (1993) Compounds active against cell walls of medically important fungi. Clin. Microbiol. Rev. 6, 1^21. [148] Kurtz, M.B. (1998) New antifungal drug targets: A vision for the future. ASM News 64, 31^39. [149] Cabib, E., Drgon, T., Drgonova, J., Ford, R.A. and Kollar, R. (1997) The yeast cell wall, a dynamic structure engaged in growth and morphogenesis. Biochem. Soc. Trans. 25, 200^204. [150] Jablonowski, D., Fichtner, L., Martin, V.J., Klassen, R., Meinhardt, F., Stark, M.J. and Scha¡rath, R. (2001) Saccharomyces cerevisiae cell wall chitin, the Kluyveromyces lactis zymocin receptor. Yeast 18, 1285^1299. [151] Yamamoto, T., Hiratani, T., Hirata, H., Imai, M. and Yamaguchi, H. (1986) Killer toxin from Hansenula mrakii selectively inhibits cell wall synthesis in a sensitive yeast. FEBS Lett. 197, 50^54. [152] Kimura, T., Kitamoto, N., Kito, Y., Iimura, Y., Shirai, T., Komiyama, T., Furuichi, Y., Sakka, K. and Ohmiya, K. (1997) A novel yeast gene, RHK1, is involved in the synthesis of the cell wall receptor for the HM-1 killer toxin that inhibits beta-1,3-glucan synthesis. Mol. Gen. Genet. 254, 139^147. [153] Kandel, J.S. (1988) Killer systems in pathogenic fungi. In: Viruses of Fungi and Simple Eukaryotes (Koltin, Y. and Leibowitz, M., Eds.), pp. 243^263. Marcel Dekker, New York. [154] Walker, G.M., MacLeod, A.M. and Hodgson, V.J. (1995) Interactions between killer yeasts and pathogenic fungi. FEMS Microbiol. Lett. 127, 213^222. [155] Gibbons, A. (1991) Moths take the ¢eld against biopesticide. Science 254, 646. [156] Kinal, H., Park, C.M., Berry, J.O., Koltin, Y. and Bruenn, J.A. (1995) Processing and secretion of a virally encoded antifungal toxin in transgenic tobacco plants: evidence for a Kex2p pathway in plants. Plant Cell 7, 677^688. [157] Koltin, Y. and Day, P.R. (1976) Inheritance of killer phenotypes and double-stranded RNA in Ustilago maydis. Proc. Natl. Acad. Sci. USA 73, 594^598. [158] Koltin, Y. and Day, P.R. (1976) Suppression of the killer phenotype in Ustilago maydis. Genetics 82, 629^637. [159] Park, C.M., Bruenn, J.A., Ganesa, C., Flurkey, W.F., Bozarth, R.F. and Koltin, Y. (1994) Structure and heterologous expression of the Ustilago maydis viral toxin KP4. Mol. Microbiol. 11, 155^164. [160] Kinal, H., Tao, J.S. and Bruenn, J.A. (1991) An expression vector for the phytopathogenic fungus, Ustilago maydis. Gene 98, 129^134. Cyaan Magenta Geel Zwart 276 M.J. Schmitt, F. Breinig / FEMS Microbiology Reviews 26 (2002) 257^276 [161] Park, C.M., Berry, J.O. and Bruenn, J.A. (1996) High-level secretion of a virally encoded anti-fungal toxin in transgenic tobacco plants. Plant Mol. Biol. 30, 359^366. [162] Romanos, M.A., Scorer, C.A. and Clare, J.J. (1992) Foreign gene expression in yeast: a review. Yeast 8, 423^488. [163] Giga-Hama, Y. and Kumagai, H. (1999) Expression system for foreign genes using the ¢ssion yeast Schizosaccharomyces pombe. Biotechnol. Appl. Biochem. 30, 235^244. [164] Bro«ker, M., Ragg, H. and Karges, H.E. (1987) Expression of human anti-thrombin III in Saccharomyces cerevisiae and Schizosaccharomyces pombe. Biochim. Biophys. Acta 908, 203^213. [165] Tokunaga, M., Kawamura, A., Yonekyu, S., Kishida, M. and Hishinuma, F. (1993) Secretion of mouse alpha-amylase from ¢ssion yeast Schizosaccharomyces pombe: presence of chymostatin-sensitive protease activity in the culture medium. Yeast 9, 379^387. [166] Sambamurti, K. (1997) Expression and secretion of mammalian proteins in Schizosaccharomyces pombe. In: Foreign Gene Expression in Fission Yeast: Schizosaccharomyces pombe (Giga-Hama, Y. and Kumagai, H., Eds.), pp. 149^158. Springer, Berlin. [167] Ostergaard, S., Olsson, L. and Nielsen, J. (2000) Metabolic engineering of Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev. 64, 34^50. [168] Pfei¡er, P. and Radler, F. (1982) Puri¢cation and characterization FEMSRE 748 1-8-02 [169] [170] [171] [172] [173] [174] of extracellular and intracellular killer toxin of Saccharomyces cerevisiae strain 28. J. Gen. Microbiol. 128, 2699^2706. Gunge, N., Tamaru, A., Ozawa, F. and Sakaguchi, K. (1981) Isolation and characterization of linear deoxyribonucleic acid plasmids from Kluyveromyces lactis and the plasmid-associated killer character. J. Bacteriol. 145, 382^390. Worsham, P.L. and Bolen, P.L. (1990) Killer toxin production in Pichia acaciae is associated with linear DNA plasmids. Curr. Genet. 18, 77^80. Suzuki, C. and Nikkuni, S. (1994) The primary and subunit structure of a novel type killer toxin produced by a halotolerant yeast, Pichia farinosa. J. Biol. Chem. 269, 3041^3046. Sawant, A.D., Abdelal, A.T. and Ahearn, D.G. (1989) Puri¢cation and characterization of the anti-Candida toxin of Pichia anomala WC65. Antimicrob. Agents Chemother. 33, 48^52. Hanes, S.D., Burn, V.E., Sturley, S.L., Tipper, D.J. and Bostian, K.A. (1986) Expression of a cDNA derived from the yeast killer preprotoxin gene: implications for processing and immunity. Proc. Natl. Acad. Sci. USA 83, 1675^1679. Tao, J., Ginsberg, I., Banerjee, N., Held, W., Koltin, Y. and Bruenn, J. (1990) Ustilago maydis KP6 toxin: structure, expression in Saccharomyces cerevisiae and relationship to other cellular toxins. Mol. Cell. Biol. 10, 1373^1381. Cyaan Magenta Geel Zwart