Yeast Killer Toxin K28: Biology and Unique Strategy of Host Cell Intoxication and Killing
Total Page:16
File Type:pdf, Size:1020Kb
toxins Review Yeast Killer Toxin K28: Biology and Unique Strategy of Host Cell Intoxication and Killing Björn Becker ID and Manfred J. Schmitt * Molecular and Cell Biology, Department of Biosciences and Center of Human and Molecular Biology (ZHMB), Saarland University, D-66123 Saarbrücken, Germany; [email protected] or [email protected] * Correspondence: [email protected]; Tel.: +49-681-302-4730; Fax: +49-681-302-4710 Academic Editor: Holger Barth Received: 29 September 2017; Accepted: 17 October 2017; Published: 20 October 2017 Abstract: The initial discovery of killer toxin-secreting brewery strains of Saccharomyces cerevisiae (S. cerevisiae) in the mid-sixties of the last century marked the beginning of intensive research in the yeast virology field. So far, four different S. cerevisiae killer toxins (K28, K1, K2, and Klus), encoded by cytoplasmic inherited double-stranded RNA viruses (dsRNA) of the Totiviridae family, have been identified. Among these, K28 represents the unique example of a yeast viral killer toxin that enters a sensitive cell by receptor-mediated endocytosis to reach its intracellular target(s). This review summarizes and discusses the most recent advances and current knowledge on yeast killer toxin K28, with special emphasis on its endocytosis and intracellular trafficking, pointing towards future directions and open questions in this still timely and fascinating field of killer yeast research. Keywords: K28; killer toxin; S. cerevisiae; A/B toxin; cell wall receptor; H/KDEL receptor; retrograde protein transport; retrotranslocation; cell cycle arrest; toxin immunity 1. Introduction A killer phenotype in yeast was originally discovered in certain strains of the wine and brewery yeast, Saccharomyces cerevisiae (S. cerevisiae), and was first described in 1963 [1]. Characteristic of all “killer” yeasts is the production and secretion of a certain type of protein toxin (killer toxin) that is lethal to sensitive strains of the same or related yeast genera. Soon after the initial studies on the killer phenomenon in S. cerevisiae, it became evident that an infection with cytoplasmic inherited double-stranded (ds)RNA viruses is responsible for killer phenotype expression [2]. Interestingly, the occurrence of dsRNA viruses is not restricted to strains of S. cerevisiae, but rather, is widely distributed among various yeast species, including Zygosaccharomyces bailii, Hanseniaspora uvarum and Ustilago maydis [3,4]. Besides non-infectious yeast dsRNA viruses (also designated as virus-like particles, VLPs), the killer phenotype can also be chromosomally encoded (Williopsis californica)[5] or associated with linear dsDNA plasmids (Klyveromyces lactis and Pichia acaciae)[6]. The existence of dsRNA viruses in S. cerevisiae seems to be linked to the absence of RNA interference (RNAi), explaining why killer systems have so far only been found in RNAi-deficient yeast species, while RNAi-proficient yeasts did not develop killer strains during evolution [7]. While relatively little is known about the ecological relevance of killer toxin-producing yeasts, it has been proposed that yeast strains carrying killer viruses presumably possess a competitive advantage in the natural yeast habitat, in the battle for resources, by eliminating sensitive yeasts [8]. Based on the killing properties and the lack of cross-immunity, four different dsRNA-encoded killer types, namely K1, K2, K28, and Klus, have so far been identified in S. cerevisiae [9–12]. Each killer type shows killing activity against non-killer strains as well as killer strains of different killer types, while it is protected and immune against its own toxin. In nature, infected yeast cells only harbor a Toxins 2017, 9, 333; doi:10.3390/toxins9100333 www.mdpi.com/journal/toxins Toxins 2017, 9, 333 2 of 15 single copy of an M-dsRNA genome, whereby the coexistence of multiple M genomes with different killerToxins specificities 2017, 9, 333 is excluded at the replicative level. Artificially, this limitation can be overcome2 of 15 by introducing cDNAs encoding killer toxins, K2 and K28, into a K1 strain, thereby artificially harbor a single copy of an M-dsRNA genome, whereby the coexistence of multiple M genomes with generatingdifferent a triple killer killer specificities strain thatis excl simultaneouslyuded at the replicative expresses level. all three Artificially, killer toxins this limitation and shows can multiple be toxinovercome immunity by [ 13introducing]. To stably cDNAs maintain encoding a virally-encoded killer toxins, K2 killer and phenotypeK28, into a inK1 yeast,strain, twothereby dsRNA genomesartificially must begenerating present ina triple the cytoplasm killer strain of that the infectedsimultaneously host: anexpresses unsegmented all three 4.6 killer kb largetoxins L-dsRNA and genomeshows of the multiple helper toxin virus immunity ScV-L-A [13] and. To one stably of four maintain smaller a virally-enc toxin-encodingoded killer M-dsRNA phenotype satellite in yeast, viruses (ScV-M1,two dsRNA ScV-M2, genomes ScV-M28, must or be ScV-Mlus) present in [the10, 14cytopl]. Thisasm review of the infected will mainly host: an focus unsegmented on ScV-M28 4.6 andkb its encodedlarge killer L-dsRNA toxin, genome K28. of the helper virus ScV-L-A and one of four smaller toxin-encoding M- dsRNA satellite viruses (ScV-M1, ScV-M2, ScV-M28, or ScV-Mlus) [10,14]. This review will mainly 2. K28focus Phenotype: on ScV-M28 Origin, and its Genomic encoded killer Organization toxin, K28. and Viral Replication The2. K28 first Phenotype: detailed Origin, analysis Genomic of the fundamentalOrganization and properties Viral Replication of the K28 killer phenotype in yeast was published in 1990 [11]. The phenotype was initially found in the S. cerevisiae wine strain 28, The first detailed analysis of the fundamental properties of the K28 killer phenotype in yeast whichwas gave published the killer in 1990 toxin [11]. its The designation. phenotype was As alreadyinitially found shown in the for S. other cerevisiae killer wine toxins strain of 28,S. which cerevisiae , K28-producinggave the killer killer toxin strains its designation. harbor two As different already cytoplasmicshown for other persisting killer toxins dsRNA of S. genomes, cerevisiae, which K28- are separatelyproducing encapsidated killer strains into harbor virus-like two particlesdifferent (VLPs)cytoplasmic [15]. persisting The smaller dsRNA 1.8 kb genomes, M-dsRNA which genome are of the ScV-M28separately satellite/killer encapsidated virusinto virus-like encodes particles the unprocessed (VLPs) [15]. K28 The toxin smaller precursor. 1.8 kb M-dsRNA The coding genome (+) strandof of ScV-M28the ScV-M28 contains satellite/killer a single open virus readingencodes the frame unproc (ORF)essed that K28 contains toxin precursor. the genetic The coding information (+) strand for the unprocessedof ScV-M28 K28 contains precursor a single (also open called reading preprotoxin, frame (ORF) pptox) that which contains likewise the genetic confers information toxin immunity for the [14]. As classicalunprocessed satellite, K28 ScV-M28 precursor depends (also called on preprotoxi a second virusn, pptox) (ScV-L-A) which likewise which functionsconfers toxin as aimmunity helper virus [14]. As classical satellite, ScV-M28 depends on a second virus (ScV-L-A) which functions as a helper required for stable maintenance and replication [16]. The linear L-A genome (4.6 kb) possesses two virus required for stable maintenance and replication [16]. The linear L-A genome (4.6 kb) possesses ORFs on its positive strand. The major 76 kDa capsid protein, Gag, is encoded by ORF1 and is two ORFs on its positive strand. The major 76 kDa capsid protein, Gag, is encoded by ORF1 and is requiredrequired for proper for proper dsRNA dsRNA encapsidation encapsidation and VLPand formation.VLP formation. ORF2 ORF2 encodes encodes Pol, an Pol, RNA-dependent an RNA- RNA-polymerasedependent RNA-polymerase (RDRP) which (RDRP) is translated which is as translat a 180ed kDa as Gag-Pola 180 kDa fusionGag-Pol protein fusion protein by a -1 by ribosomal a -1 frameshiftribosomal event frameshift [17–19]. event Figure [17–19].1 summarizes Figure 1 su themmarizes genomic the organization genomic organization and the codingand the capacity coding of the positivecapacity strands of the positive of ScV-M28 strands and of ScV-M28 ScV-L-A. and ScV-L-A. FigureFigure 1. Genomic 1. Genomic organisation organisation of of the the L-A L-A (+) (+) and and M28M28 (+) strands. strands. The The initial initial GAAAAA GAAAAA sequence sequence at at the 5′-end of the M28 (+) ssRNA represents a terminal recognition element (TRE) which is necessary the 50-end of the M28 (+) ssRNA represents a terminal recognition element (TRE) which is necessary for for the initiation of transcription. M28 (+) ssRNA further contains the toxin-encoding K28 open the initiation of transcription. M28 (+) ssRNA further contains the toxin-encoding K28 open reading reading frame (ORF), a poly(A)-rich region and potential 3′ elements which are required for in vivo frame (ORF), a poly(A)-rich region and potential 30 elements which are required for in vivo RNA RNA replication (IRE, internal replication enhancer; TRE, 3′ terminal recognition element) and 0 replicationpackaging (IRE, (VBS, internal viral replicationbinding side). enhancer; Besides the TRE, two 3 ORFsterminal on L-A recognition (+) ssRNA, element)the position and of packagingthe −1 (VBS,frameshift viral binding region, side). the Besidesencapsidation the two signal ORFs and on the L-A replication (+) ssRNA, side the for position (−) strand of the synthesis−1 frameshift are region,indicated the encapsidation in brackets. Reproduced signal and and the modifi replicationed from side[10,20], for 2011 (−) and strand 2013, synthesis Springer. are indicated in brackets. Reproduced and modified from [10,20], 2011 and 2013, Springer. For stable L-A helper and ScV-M28 satellite virus propagation within a killer cell, a subset of at least 28 nuclear MAK genes (MAK1, MAK3-MAK27, PET18 as well as SPE2) is required in S. cerevisiae For stable L-A helper and ScV-M28 satellite virus propagation within a killer cell, a subset [3,21,22].