Dissertations / Theses on the topic 'Saccharomyces cerevisiae – Genetics'

For organisms undergoing a developmental process it is ideal that specific genes are induced and repressed at the correct time and to the correct level in a coordinated manner. The process of meiosis and spore formation (collectively known as sporulation) in Saccharomyces cerevisiae provides a convenient system to elucidate transcriptional mechanisms of gene repression and the contribution such repression mechanisms offer to cells capable of undergoing a developmental process. This thesis focuses on transcriptional repression of sporulation-specific genes during both vegetative/mitotic conditions and sporulation. The fitness contribution of transcriptional repressors that regulate sporulationspecific genes during vegetative growth were investigated considering the similarities between meiosis and mitosis such as DNA replication, chromosome segregation and cytokinesis. Well-characterised sporulation genes of different functions were expressed in vegetative cells and ectopic expression of these genes was found not to be lethal. It was ascertained through strain competition studies that ectopic expression of the genes IME1, SMK1, SPR3 and DIT1 during mitotic growth did not affect cellular fitness. The expression of NDT80 in vegetative cells, however, caused a marked reduction in fitness and cells were also further compromised in the absence of the Sum1p repressor that regulates NDT80 transcription. The role of NDT80 as a transcriptional activator of middle sporulation genes, rather than the over-expression of NDT80 as a protein, caused the reduction of cell viability. Transcriptional regulation of the middle sporulation-specific gene SPR3 by the meiosis-specific Set3p repressor complex was investigated using synchronous sporulation cultures of the W303a/?? strain commonly used for sporulation studies. In a mutant W303a/?? ??set3/??set3 strain, lacking a key component of the Set3p repression complex, the transcription of SPR3 was uncharacteristically expressed at higher levels and derepressed during late sporulation. This SPR3 expression was consistent for both SPR3 transcript and SPR3::lacZ reporter protein studies. This preliminary work will enable future studies, using SPR3 promoter deletions fused to a lacZ reporter, aimed at determining the region of the SPR3 promoter that the Set3p complex may interact with to transcriptionally repress the gene during sporulation.

The yeast FBP1 and PCK1 genes and the gluconeogenic enzymes that they encode, fructose-1,6-bisphosphatase and phosphoenolpyruvate carboxykinase, are subject to multiple levels of regulation by glucose. It has been reported that transcriptional repression of these genes is exceptionally sensitive to glucose, being triggered by glucose concentrations of less than 0.005% (0.25 mM). It was shown here that in addition at transcriptional repression, the FBP1 and PCK1 and mRNAs are destabilised about 2-fold upon addition of the same low levels of glucose. Low levels of the fermentable sugars fructose or sucrose also stimulated this effect but galactose did not. This destabilisation was lost in a triple hxk1, hxk2, glk1 mutant, but was not triggered by addition of 2-deoxyglucose. The data suggests that sugar phosphorylation and further metabolism of glucose is required to trigger this response. Analysis of metabolic mutants showed that mutations in the upper part of the glycolytic pathway abolish the destabilisation of the FBP1 mRNA. Differences were shown to exist between the regulatory pathways that mediate glucose-stimulated mRNA decay and transcriptional repression. Models which might account for the mechanisms by which rapid decay of the gluconeogenic mRNAs is triggered are discussed. A strategy based on gene fusions with the stable PGK1 mRNA was designed in order to map cis-acting regions which influence PCK1 mRNA stability. A fusion mRNA containing the PCK1 mRNA protein coding region was not destabilised upon addition of low levels of glucose. It was therefore suggested that glucose-stimulated mRNA decay might in some way be dependent upon translation initiation via an interaction with the 5'-leader.

Studies were made of 2 mum based chimaeric plasmid copy number in Saccharomyces cerevisiae. A plasmid (pAYE56) containing three selectable genes in yeast (yeast LEU2, bacterial CAT and HSV-1 - TK) was constructed to reflect changes in copy number. Yeast transformants could be grown under three selection regimes and plasmid copy number estimated. During selective growth for the LEU2 gene there are about 20 plasmids per cell. This increases to about 100 during selective growth for the TK gene and furthermore the copy number can be controlled by the stringency of selection. Simultaneous selection for the TK and CAT genes may lead to a further increase (160 copies). Two models are proposed to account for these increases. The amplification model proposes plasmid replication without cell growth whilst the selection model suggests that plasmid copy number varies greatly in a population of transformants and cells with a high copy number are selected for growth under the TK/CAT selection conditions. Whilst the mechanism of copy number increase is unclear, an attempt was made to relate the expression of a heterologous gene (Human alpha2-IFN) to gene dosage using the promotion and secretion signals of the alpha-factor gene. Production of intracellular alpha2-IFN was unaffected by copy number whilst secreted material showed a 100 fold increase over a ten fold increase in gene dosage. Attempts were made to isolate plasmid copy number mutants. After mutagenesis (of cells or plasmid) transformants were selected under conditions for simultaneous over-expression of the TK and CAT genes. Mutants capable of growth under these conditions were obtained. In one group the mutant phenotype was lost upon curing but did not return upon retransformation. In a second group a chromosomal mutation was isolated. Plasmid copy number estimates indicated that this was unchanged however. Alternative strategies are discussed for the isolation of mutants.

Thesis (PhD)--University of Stellenbosch, 2002
ENGLISH ABSTRACT: Upon nutrient limitation, normal cells of the budding yeast, Saccharomyces cerevisiae, undergo a transition from ovoid cells that bud in an axial (haploid) or bipolar (diploid) fashion to elongated cells that bud in a unipolar fashion. The daughter cells stay attached to the mother cells, resulting in chains of cells referred to as pseudohyphae. These filaments can grow invasively into the growth substrate (haploid), or away from the colony (diploid), and are hypothesised to be an adaptation of yeast cells that enables them to search for nutrientrich substrates. This filamentous growth response to nutrient limitation was shown to be dependent on the expression of, amongst others, the MUC1 gene. MUC1 (also known as FL011) encodes a large, cell wall-associated, GPI-anchored threonine/serine-rich protein that bears structural resemblance to mammalian mucins and to the yeast flocculins. Deletion and overexpression studies demonstrated that it is critical for pseudohyphal differentiation and invasive growth, and that overexpression of the gene also results in strongly flocculating yeast strains. The upstream regulatory region of MUC1 comprises the largest yeast promoter identified to date and areas as far as 2.4 kb upstream of the translational start site have been shown to confer regulation on MUC1 expression. The large promoter region is not unique to MUC1, however, since it is almost identical to that of the functionally unrelated STA2 gene. The STA2 gene, as well as the identical STA1 and STA3 genes, encodes extracellular glucoamylase isozymes that enable the yeast cell to utilise starch as a carbon source. Glucoamylases liberate glucose residues from the non-reducing end of the starch molecule, thereby making it accessible to yeast cells. The high identity between the promoters of MUC1 and STA1-3 suggests that the two genes are co-regulated. In addition, several transcription factors that regulate the transcriptional levels of both MUC1 and STA2 have been identified and include Msn1p and the previously uncharacterised Mss11p. Overexpression of either Msn1p or Mss11p results in elevated levels of MUC1 and STA2 transcription and a dramatic increase in flocculation, invasive growth, pseudohyphal differentiation and the ability to utilise starch, suggesting that the two genes are indeed co-regulated. The main objective of this study was to characterise Mss11p and its role in the co-regulation of MUC1 and STA2 (as a representative member of the STA gene family). A detailed expression analysis, using Northern blots and Lacl reporter gene expression studies in different media, confirmed that these genes are indeed co-regulated to a large extent. MUC1 and STA2 are also regulated by the same transcriptional regulators, which include not only Msn1pand Mss11p, but also Ste12p, the transcription factor of the mating pheromone/filamentous growth signalling cascade, and Flo8p, a transcriptional activator of the flocculation genes. Overexpression of the genes encoding these factors results in elevated expression levels of both MUC1 and STA2 in most nutritional conditions and enhances the filamentous growth phenotypes of the strain, as well as the ability to degrade starch. On the other hand, the deletion thereof results in severe reductions in the transcription levels of MUC1 and STA2, with equally severe reductions in filamentous growth and the ability to hydrolyse starch. These expression studies also showed that the repressive effect of STA10, a previously uncharacterised negative regulator of STA2, is actually a phenotype conferred by a FLOB mutation in some laboratory strains of S. cerevisiae. The upstream regulatory regions of MUC1 and STA2 are the largest promoters in the yeast genome. By sequencing the upstream areas of STA2 and STA3 and comparing them to the sequence of MUC 1, it was shown that these upstream areas are 99.7%identical over more than 3 900 base pairs (bp) upstream of the translational start. With the exception of a few minor substitutions, the only significant difference between the MUC1 and STA2 promoters is the presence of a 20-bp and a 64-bp sequence found in the MUC1 promoter, but not in the promoters of any of the STA1-3 genes. Through a promoter-deletion analysis, it was shown that Mss11p, Msn1pand Flo8p exert their control over the transcription of MUC1 and STA2 from an 90-bp sequence located at -1 160 to -1 070 in the STA2 and -1 210 to -1 130 in the MUC1 promoters. This sequence also mediates the effect of carbon catabolite repression on the transcription of STA2 and MUC1. Despite the similarities in the expression patterns of MUC1 and STA2, some discrepancies also exist. The most significant difference is that, in wild-type cells and under all nutritional conditions tested, MUC1 transcription is reduced significantly if compared to the transcription levels of STA2. This was attributed to the presence of the 20- and 64-bp sequences, that are present in the promoter region of MUC1, but absent from that of STA2. To place the transcriptional regulators of MUC1 and STA2 in the context of known signal transduction pathways, an epistasis analysis was conducted between MSN1, MSS11 and components of the mating pheromone/filamentous response MAPkinase cascade and cAMPPKA pathway that were shown to be required for the filamentous growth response. This analysis revealed that Msn1p functions in a third, as yet uncharacterised, signal transduction pathway, also downstream of Ras2p,but independent of the two identified pathways, i.e. the cAMP-PKA and pheromone response/filamentous growth response MAP kinase pathways. However, Mss11p seems to function downstream of all three the identified pathways. This suggestsa critical and central role for Mss11p in determining the transcription levels of MUC1 and STA2. To further characterise Mss11p and its role in the transcriptional regulation of MUC1 and STA2, it was also subjected to a detailed deletion and mutation analysis. Mss11p was shown to harbour two distinct activation domains required for the activation of MUC1 and STA2, but also able to activate a reporter gene expressed from under the GALl promoter. The more prominent of the activation domains of Mss11p was shown to be one of the domains with homology to Flo8p, designated H2. The H2 domain has significant homology to a number of proteins of unknown function from a range of different organisms. A multi-sequence alignment allowed the identification of conserved amino acids in this domain. Mutations in two of the four conserved amino acid pairs in the H2 domain completely eliminated the activation function of Mss11p. The poly-glutamine and poly-asparagine domains of Mss11p are not required for its activation function. The deletion of these domains has no impact on the ability of Mss11p to activate MUC1 or STA2 or of the Gal4p-Mss11p fusion to activate the lacl reporter gene expressed from under the GAL7 promoter. Gal4p fusions of either of these domains were also unable to trans-activate the PGAL7-lacl reporter gene. As such, it was concluded that neither of these domains performs a function in the role of Mss11p as a transcriptional activator. We also demonstrated that the putative ATP/GTP-binding domain (P-loop) is not required for the transcriptional activation function of Mss11p. In an attempt to identify other target genes of Mss11p, the use of micro-arrays was employed to assessthe impact of the overexpression and deletion of MSS11 on the total yeast transcriptome. These results showed that MUC1 and STA2 are the only two genes in the ISP15 genetic background that are significantly (more than 15-fold) enhanced by the overexpression of MSS11. Mss11p therefore seemsto playa very specific or dedicated role in MUC1 and STA2 transcription. This analysis also identified several genes (DBP2, ROM2, YPLOBOC, YGR053C, YNL179C, YGR066C) that are repressed by overexpression of MSS11 and activated when MSS11 is deleted. To integrate all the results, three possible models for the activation of MUC1 and STA2 transcription by Mss11p are proposed: (i) Mss11p performs the role of a transcriptional mediator, possibly in a protein complex, to convey information from upstream regulatory elements to the transcription machinery assembledat the core promoters of MUC1 and STA2; (ii) Mss11p plays a more direct role in transcriptional activation, possibly as a transcription factor itself; and (iii) Mss11p facilitates transcription of the MUC1 and STA2 promoters as part of a larger complex that removes or releases the chromatin barrier over the MUC1 and STA2 promoters in responseto specific nutritional signals.
AFRIKAANSE OPSOMMING: Wanneer voedingstowwe beperkend raak, ondergaan selle van die botselvormende gis, Saccharomyces cerevisiae, fn transformasie vanaf ronde selle, wat in fn aksiale (haploïede) of bipolêre (diploïede) patroon bot, tot verlengde selle, wat slegs op een punt bot. Die dogterselle blyaan die moederselle geheg, sodat kettings van selle, wat as pseudohifes bekend staan, gevorm word. Hierdie filamente kan fn groeisubstraat binnedring (haploïede) of vanaf die kolonie weggroei (diptoïede), en is moontlik fn aanpassing van die gisselle wat hulle in staat stelom na meer voedingstofryke substrate te groei. Die vermoë om filamente in respons tot voedingstoftekorte te vorm, is onderhewig aan die uitdrukking van, onder meer, die MUC1-geen. MUC1 (ook bekend as FL011) kodeer vir fn selwand-geassosieerde treonien/serien-ryke proteten met fn GPI-anker wat strukturele verwantskappe met die mukiene van soogdiere en die flokkuliene van giste toon. Delesie- en ooruitdrukkingstudies het bewys dat dit krities is vir die ontwikkeling van pseudohifes en penetrerende groei, terwyl die ooruitdrukking daarvan ook tot sterk flokkulerende gisrasse lei. Die stroom-op regulatoriese area van MUC1 vorm die grootste promotor wat tot dusver in gis geïdentifiseer is, en daar is bewys dat areas so ver as 2.4 kb stroom-op van die translasie-inisiëringsetel die regulering van MUC1 beïnvloed. Hierdie groot promotor is egter nie uniek tot MUC1 nie, aangesien fn amper identiese promotor die regulering van die funksioneelonverwante STA2-geen beheer. Die STA2-geen, asook die identiese STA1- en STA3-gene, kodeer vir ekstrasellulêre glukoamilase isosieme wat die gis in staat stelom stysel as koolstofbron te benut. Dit bevry glukosemolekules vanaf die nie-reduserende punt van die styselmolekuul en stel dit sodoende aan gisselle beskikbaar. Die hoë vlak van eendersheid tussen dié twee promotors veronderstel dat die twee gene op soortgelyke wyse gereguleer word. Verskeie transkripsiefaktore wat die transkripsievlakke van beide MUC1 en STA2 beheer, is ook geïdentifiseer, Dit sluit Msn1p en die tot dusver ongekarakteriseerde Mss11p in. Ooruitdrukking van Msn1p of Mss11p lei tot verhoogde vlakke van MUC1 en STA2 se transkripsie en fn dramatiese toename in flokkulasie, asook die vermoë om penetrerend te groei, pseudohifes te vorm en stysel te benut. Dit bevestig dat die twee gene wel tot fn groot mate op dieselfde wyse gereguleer word. Die hoofdoel van hierdie studie was om Mss11p en die rol daarvan in die regulering van MUC1 en STA2 te karakteriseer. Gedetailleerde uitdrukkingsanalises met behulp van die Northern-kladtegniek en facZverklikkergeeneksperimente in verskillende media het bevestig dat die gene wel tot fn groot mate op dieselfde wyse gereguleer word. Transkripsie van MUC1 en STA2 word ook deur dieselfde transkripsionele reguleerders beheer, wat nie net Msn1pen Mss11p insluit nie, maar ook Ste12p, die transkripsiefaktor van die paringsferomoon/filamentagtige groei seintransduksiekaskade, en Fl08p, fn transkripsionele aktiveerder van die flokkulasiegene. Ooruitdrukking van die gene wat vir hierdie faktore kodeer, veroorsaak verhoogde uitdrukkingsvlakke van beide MUC1 en STA2 onder die meeste groeitoestande en verbeter die vermoë van die gisras om filamentagtig te groei en om stysel te benut. Andersyds veroorsaak delesies van die gene 'n dramatiese afname in die transkripsievlakke van MUC1 en STA2, met vergelykbare afnames in die vermoë van die gisras om filamentagtig te groei en om stysel te benut. Hierdie uitdrukkingstudies het ook bewys dat die onderdrukkingseffek van STA10, 'n tot dusver ongekarakteriseerde, negatiewe reguleerder van STA2, aan 'n mutasie in FLOB in sekere laboratoriumrasse van S. cerevisiae toegeskryf kan word. Die stroom-op regulatoriese areas van MUC1 en STA2 is die grootste promotors in die gis se genoom. Deur die nukleotiedvolgordes van die ver stroom-op areas van STA2 en STA3 te bepaal en hulle met dié van MUC1 te vergelyk, is daar vasgestel dat die stroom-op areas van die gene 99.7% identies is oor meer as 3 900 basispare (bp) stroom-op van die beginsetel van translasie. Met die uitsondering van enkele basispaarverskille, is die enigste noemenswaardige verskil tussen die promotors van MUC1 en STA2 die teenwoordigheid van 'n 20 bp- en 'n 64 hp-fragment wat in die MUC1-promotor aangetref word, maar nie in die promotors van die STA1-3 gene nie. Deur 'n promotordelesie-analise kon daar bewys word dat Mss11p, Msn1p en Flo8p beheer uitoefen oor die transkripsie van MUC1 en STA2 vanaf 'n 90-bp-fragment, wat by posisie -1 160 tot -1 070 in die STA2-promotor en posisie -1 210 tot -1 130 in die MUC1-promotor aangetref word. Koolstofkatabolietonderdrukking van MUC1 en STA2 se transkripsie geskied ook deur middel van hierdie fragment. Ten spyte van die ooreenkomste in die uitdrukkingspatrone van MUC1 en STA2, kom daar tog ook verskille voor. Die mees opvallende verskil is dat, in wilde-tipe selle en onder alle toestande tot dusver getoets, die transkripsievlakke van MUC1 aansienlik laer is as dié van STA2. Dit word toegeskryf aan die teenwoordigheid van die 20 bp- en 64 bp-fragmente, wat in die promotor van MUC1 teenwoordig is, maar in die promotor van STA2 afwesig is. Om die transkripsionele reguleerders van MUC1 en STA2 in die konteks van bekende seintransduksieweë te plaas, is 'n epistase-analise gedoen tussen MSN1, MSS11 en komponente van die paringsferomoon/filamentagtige groei MAP-kinasekaskade en die cAMPPKA- weg wat uitgewys het dat dit 'n rol in die filamentagtige groeirespons speel. Hierdie analise het onthul dat Msn1p in 'n derde, tot dusver onbeskryfde, seintransduksieweg funksioneer, wat ook stroom-af van Ras2p is, maar wat onafhanklik funksioneer van die twee bekende weë, die cAMP-PKA-weg en die paringsferomoon/filamentagtige groei MAPkinasekaskade. Mss11p blyk egter stroom-af van al drie dié weë te funksioneer. Dit wys dat Mss11p 'n kritiese en sentrale rol in die bepaling van MUC1 en STA2 se transkripsievlakke speel. Om Mss11p en die rol daarvan in die regulering van MUC1 en STA2 se transkripsie verder te karakteriseer, is dit aan 'n volledige delesie- en mutasie-analise onderwerp. Dit het gewys dat Mss11p twee verskillende aktiveringsdomeine bevat wat vir die transkripsionele aktivering van STA2 en MUC1 benodig word, maar wat ook 'n verklikkergeen kon aktiveer wat onder die GAL7-promotor uitgedruk word. Die prominentste van die twee aktiveringsdomeine van Mss11p is een van die domeine wat homologie toon met 'n soortgelyke domein van Flo8p, die sogenaamde H2-domein. Die H2-domein toon hornologie met 'n verskeidenheid van organismesse proteïene, waarvan die funksie onbekend is. 'n Vergelyking van al die relevante aminosuurvolgordes uit dié proteïene het gehelp om 'n aantal gekonserveerde aminosure te identifiseer. Mutasies van twee van die vier gekonserveerde aminosuurpare het die vermoë van Mss11p om transkripsie te aktiveer, heeltemal geëlimineer. Die poliglutamien- en poliasparagiendomeine van Mss11p word nie vir die aktiveringsfunksie benodig nie. Die delesie van die domeine het geen impak gehad op die vermoë van Mss11p om die transkripsie van MUC1 en STA2 te aktiveer nie, of op die vermoë van die Gal4p-Mss11p fusie om die lacZ-verklikkergeen onder regulering van die GAL7-promotor te aktiveer nie. Gal4p-fusies met enige van die domeine was ook nie in staat om die PGAL7-lacZverklikkergeen te aktiveer nie. Daar kan dus afgelei word dat nie een van die twee domeine 'n funksie in die rol van Mss11p as transkripsionele aktiveerder het nie. Soortgelyke eksperimente het bewys dat die moontlike ATP/GTP-bindingsdomein (P-lus) nie vir die transkripsionele aktiveringsfunksie van Mss11p benodig word nie. In 'n poging om ander teikengene van Mss11p te identifiseer, is mikro-ekspressieroosters gebruik om die impak van die ooruitdrukking en delesie van MSS11 op die totale transkriptoom van die gis te bepaal. Dié resultate het gewys dat MUC1 en STA2 die enigste twee gene in die ISP15genetiese agtergrond is waarvan transkripsie noemenswaardig (meer as 15-voudig) deur die ooruitdrukking van MSS11 verhoog word. Dit wil dus voorkom asof Mss11p 'n baie spesifieke rol in die transkripsie van MUC1 en STA2 speel. Hierdie analise het ook verskeie gene (DBP2, ROM2, YPLOBOC,YGR053C, YNL179C, YGR066C) geïdentifiseer wat deur die ooruitdrukking van MSS11 onderdruk word en deur die delesie van MSS11 geaktiveer word. Ten einde al die resultate te integreer, word drie moontlike modelle vir die aktivering van MUC1- en STA2-transkripsie deur Mss11p voorgestel: (i) Mss11p vervul die rol van 'n transkripsionele tussenganger, moontlik as deel van 'n proteïenkompleks, om die inligting van die stroom-op regulatoriese elemente aan die transkripsiemasjinerie wat oor die kernpromotor van MUC1 en STA2 gebind is, oor te dra; (ii) Mss11p speel 'n meer direkte rol in transkripsionele aktivering, moontlik as 'n transkripsiefaktor self; en (iii) Mss11p maak die transkripsie van MUC1 en STA2 moontlik as deel van 'n groter kompleks wat die chromatienblokkade oor die promotors van STA2 en MUC1 in respons tot spesifieke seine verslap of verwyder.

The MAT∝ locus of the yeast Saccharomyces cerevisiae encodes two regulatory proteins responsible for determining the ∝cell type. The MAT∝1 gene encodes ∝1, a positive regulator of ∝cell-specific genes, whereas the MAT∝2 gene encodes a negative regulator of a cell-specific genes (∝2). MAT∝2. (in conjunction with the MATα1 gene) also determines the α/∝ diploid cell type by repressing haploid-specific genes. ∝2 exerts its effect at the transcriptional level in the ∝ cell by binding to a sequence located upstream of α cell-specific genes. The present study undertook to examine, through in vitro genetic manipulation, the structure/function relationship of the MAT∝ regulatory proteins, particularly∝2, in their role as gene regulators. The construction of mutant MAT∝2 genes containing termination codons at various points within the gene, and subsequent transformation of the mutant genes into mat∝2 yeast, indicated that the carboxy-terminal one-third of the gene product was necessary for full repressor activity in the haploid as well as in the diploid. A segment within the carboxy-terminal one-third of ∝2 displays some homology to the higher eukaryote homeo domain as well as to a prokaryotic bihelical DNA-binding structural motif. This region of the gene was subjected to semi-random missense mutagenesis in vitro and the mutant genes were analyzed by transformation into strains containing chimaeric genes that encode β-galactosidase from ∝2 and a1/∝2. repressible promoters. In this manner it was demonstrated that most of those residues in ∝2. which correspond to conserved amino acids in the prokaryotic DNA-binding structure and in the homeo domain are essential for the two repressor activities of ∝2. Several mutations more severely affected the ability of ∝2 to repress α-specific genes than haploid-specific genes. Analysis of the temperature dependence of the activities of some of the mutants was consistent with the existence of a helix-turn-helix structure at this region of the protein. Finally, further analysis of some of these mutants in vitro confirmed that the observed defect correlated with a loss of DNA-binding activity.
Medicine, Faculty of
Biochemistry and Molecular Biology, Department of
Graduate

The kre2-1 mutation is one of a number of K1 killer resistant complementation groups in Saccharomyces cerevisiae. kre2-1 cells bind less K1 killer toxin than wild type cells due to a reduced affinity for the toxin molecule and yet contain wild type levels of (1$ to$6)-$ beta$-D-glucan, a component of the toxin receptor molecule. The KRE2 gene was cloned by complementation of the kre2-1 mutation and is predicted to encode a 433 amino acid protein directed into the secretory pathway. Haploid strains which carried a disruption at the KRE2 locus, grew more slowly than wild type cells, and showed a defect in the N-linked glycosylation of secreted proteins analysed. Genetic and protein analyses showed the defect to be in the assembly of the core oligosaccharide, most likely in the attachment of one or perhaps two mannose residues to the core structure. The mutant core in kre2 cells appeared to be a less efficient substrate for outer chain elaboration.
A plausible interpretation of these results is that those mannose residues added to the core oligosaccharide in a KRE2 dependent manner are involved in the cross-linking of (1$ to$6)-$ beta$-D-glucan to mannoprotein in the cell wall to complete the K1 killer toxin receptor and failure to make such attachments is the basis of resistance to toxin in kre2 cells.

Dissertation (PhD)--University of Stellenbosch, 2005.
ENGLISH ABSTRACT: To manipulate recombinant microorganisms for industrial processes, controllable genetic systems are needed that can coordinate expression of recombinant metabolic pathways. All components are sensitive to change and thus putative targets for modification and genetic elements and regulatory systems need to be understood and determined. Central in gene regulation is the transcription activators that mediate gene transcription mechanisms by binding to promoters in response to environmental signals. Promoter engineering entails the modification of transcription factors and their target promoters. In this study, a metabolic control system in Saccharomyces cerevisiae was constructed that would allow induction in response to physiological environment, specifically hypoxia and low temperature conditions. Two approaches were undertaken to find such a system. Firstly, a bi-directional reporter gene cloning vector was designed to search for novel hypoxiainducible promoters. Secondly, a transcription regulatory circuit was built, consisting of an inducible transcription regulator and promoter with a reporter gene through which it mediates transcription. Advantage was taken of the modular nature of proteins and functional domains originating from different transcriptional proteins were combined. A search for promoter elements sensitive to hypoxia from a S. cerevisiae genomic DNA (gDNA) library, using a bi-directional cloning vector, did not yield highly inducible promoters. It was concluded that a multitude of signals overlap, rendering genetic induction difficult to control. A synthetic regulatory system would minimize the impact of these multiple interactions. Such a genetic circuit was constructed, consisting of a chimeric transcription activator and a target fusion promoter. The chimeric transcription activator consisted of the GAL4 DNA binding domain, ADR1 TADIII transactivation domain and three domains of the MGA2 regulatory protein. The functional domains of Mga2p responsible for unregulated expression (at high basal levels) under both aerobic and hypoxia conditions were located, as well as a further upregulation under low temperature, and were mapped to the Nterminal and mid-Mga2p regions. A target fusion promoter consisting of a partial GAL10/1 promoter sequence and a Trichoderma reesei core xyn2 promoter were constructed as target for this chimeric transactivator. This synthetic promoter was fused to the T. reesei xyn2 open reading frame encoding for a readily assayable β-xylanase activity. Both the chimeric transactivator and fusion promoter-reporter gene cassettes were expressed from the same episomal plasmid, named pAR. Transformed into S. cerevisiae Y294, this regulatory system induced transcription under aerobic and hypoxia conditions. Furthermore, the reporter gene expression was upregulated by the chimeric transactivator at low temperatures. The chimeric transactivator mediated a seven-fold induction of the reporter gene under aerobic conditions in S. cerevisiae Y294 when transformed with plasmid AR. A two- to three-fold induction at 23ºC was reported under anaerobic conditions, relative to a reference strain expressing a transcription activator without the Mga2p domains. At 30ºC, a two- to three-fold induction under aerobic conditions and similar induction under oxygen-limited conditions were observed. Replacing the reporter gene with your favorite gene (for example a recombinant enzyme) and incorporating such a pAR system into a recombinant yeast should induce expression of the chosen gene under low temperatures, both aerobic and anaerobically (thus creating a controllable system). The system also has wider application in identifying other transcription factors’ signal-sensitive domains. The design of this system provides the ability to add a linker to a transactivator and to either create specific signal sensitivity or relieve the regulator of its signal dependence. It creates an easy system for assessing other transactivators and their domains with unknown functions and thus provides a ”workhorse and prospector in one”.
AFRIKAANSE OPSOMMING: Vir die manipulering van rekombinante mikroörganismes vir industriële prosesse word beheerbare genetiese stelsels benodig om gekoördineerde uitdrukking van rekombinante metaboliese weë teweeg te bring. Alle komponente van sulke stelsels is sensitief vir verandering en genetiese elemente en reguleerbare sisteme moet dus deeglik verstaan of bepaal word. Sentraal tot geenregulering is die transkripsie-aktiveerders wat geentranskripsie beheer deur aan promoters te bind in reaksie op eksterne omgewingsfaktore. Promotoringenieurswese behels wysigings van transkripsiefaktore en hul teikenpromotors. In hierdie studie is 'n genetiese beheerstelsel vir Saccaromyces cerevisiae ontwikkel wat induksie in reaksie tot spesifieke fisiologiese omgewingreaksies, naamlik hipoksie- en lae temperatuur, toelaat. Twee benaderings is gevolg: eerstens is ‘n tweerigting verklikker-geen vektor ontwikkel en gebruik om vir unieke induseerbare hipoksie-promoters te soek. Tweedens is ‘n transkripsie reguleringstelsel gebou wat uit ‘n induseerbare transkripsiereguleerder and promotor met ‘n verklikkergeen bestaan, waardeur transkripsie bemiddel kan word. Hierdie benadering benut die modulêre onderbou van proteïene en funksionele domeine afkomstig vanaf verskillende transkripsiefaktore is gekombineer. 'n Soektog na hipoksie-sensitiewe promotors vanuit 'n Saccharomyces cerevisiae-genoom- DNA (gDNA), deur van ‘n tweerigting verklikker-vektor gebruik te maak, het ongelukkig nie hoogs-induseerbare promotors opgelewer nie. Die gevolgtrekking was dat ‘n veelvoud van seine met mekaar oorvleuel en die beheer van genetiese induksie dus bemoeilik. Die ontwikkeling van ‘n sintetiese regulering-sisteem kan die impak van die veelvuldige interaksies verminder. Vir dié doel is ‘n sintetiese reguleringstelsel ontwerp, bestaande uit ‘n chimeriese transkripsie-aktiveerder met ‘n teiken fusie-promotor. Die chimeriese transaktiveerder bestaan uit die GAL4 DNA bindingsdomein, die ADR1 TAD III transaktiveringsdomein en drie domeine van die Mga2 reguleringsproteïen. In die studie is die funksionele domeins van Mga2p betrokke by lae temperatuur-respons en ongereguleerde uitdrukking (teen hoë basale vlakke) onder beide aërobiese en anaërobiese toestande aangedui en is tot die N-terminaal en middel-Mga2p areas gekarteer. ‘n Teiken-fusie-promoter, bestaande uit 'n gedeeltelike GAL1/10 DNA promotoropeenvolging en ‘n Trichoderma reesei kern xyn2-promoter, is as teiken vir hierdie chimeriese transaktiveerder saamgestel. Hierdie sintetiese promotor is aan die T. reesei xyn2 oopleesraam, wat vir ‘n maklik meetbare β-xylanase aktiwiteit kodeer, gekoppel. Beide die chimeriese transaktiveerder and fusie-promoter-verklikker-geenkaset word vanaf dieselfde episomale plasmied, bekend as pAR, uitgedruk. Hierdie reguleringsisteem induseer transkripsie onder aërobiese en hipoksie toestande in S. cerevisiae Y294. Verder word die verklikkergeen se uitdrukking deur die chimeriese transaktiveerder by lae temperature verhoog. Die chimeriese transaktiveerder induseer ‘n sewe-voudige induksie van die verklikkergeen onder aërobiese toestande by 23ºC vanaf die pAR-stelsel in S. cerevisiae Y294. ‘n Twee- tot drie-voudige induksie teen 23ºC is onder hipoksie toestande gevind, relatief tot induksievlakke van ‘n verwysingstam met ‘n transaktiveerder sonder die Mga2 domeine. By 30ºC is ‘n twee- tot drie-voudige induksie onder aërobiese en lae suurstofvlakke waargeneem. Deur die verklikker geen met ‘n jou-gunsteling-geen te vervang (bv. ‘n rekombinante ensiem) en so 'n pAR-sisteem in ‘n rekombinante gis te inkorporeer, word uitdrukking onder lae temperature onder beide aërobiese- en anaërobiese toestande geïnduseer (en sodoende word ‘n reguleerbare sisteem geskep). Die sisteem het wyer toepassing om sein-sensitiewe domeine van ander transkripsiefaktore te identifiseer. Die ontwerp van die stelsel maak dit moontlik om 'n skakel tot die transaktiveerder by te voeg wat óf sensitiwiteit tot 'n spesifieke sein skep, óf die reguleerder vanaf seinafhanklikheid verlos. So word ‘n bruikbare stelsel vir die bestudering van ander transaktivators en hul domeine met onbekende funksie geskep – ‘n “werksesel en prospekteerder in een”.

This thesis reports on some sequencing and functional studies on chromosome I of Saccharomyces cerevisiae.
Eight open reading frames (ORFs) have been identified in S. cerevisiae which have similarity to the canine 24kD glycoprotein, gp25L (Wada et al., 1991). In chapter 2 of this thesis, I report the gene disruption and functional characterization of three of these ORFs: YAR002Ac, YAL007c and YGL002w. Disruption of YAR002Ac resulted in calcofluor white resistance, disruption of YGL002w increased sensitivity to this compound, while yal007c$ Delta$ mutants had no calcofluor white phenotype. The expression of Kre9p was partially increased in the ygl002w$ Delta$ mutant. All single, double and triple mutants grew, mated and sporulated normally.
The sequence of each S. cerevisiae chromosome, released in GenBank database, lacked most of the telomere sequence. This was due to the fact that a telomeric fragment with one clonable end cannot be integrated into a vector by the classical cloning method. Recently, Louis and Borts cloned all telomeres of S. cerevisiae successfully using an alternative cloning method (Louis and Borts, 1995). In chapter 3 of this thesis, I report the DNA sequence of the right telomeric region of chromosome I. The sequence indicates that this region represents a typical yeast telomeric region and contains 98bp of TG$ sb{1-3}$ repeats and an X subtelomeric element. Another subtelomeric element, the Y$ sp prime$ element, is absent from the right telomeric region of chromosome I. (Abstract shortened by UMI.)

A mutational analysis of genes which confer resistance to the yeast K1 killer toxin has identified a number of components involved in the synthesis of a $ beta(1 to 6)$-linked glucan polymer found in the Saccharomyces cerevisiae extracellular matrix. The KRE9 gene is predicted to encode a 30 kDa serine/threonine rich cell wall protein, which is modified by O-glycosylation before being secreted at the cell surface where it likely functions in $ beta(1 to 6)$-glucan assembly. Null mutations in KRE9 lead to killer resistance, slow vegetative growth, and reduced cell wall $ beta(1 to 6)$-glucan levels which are 10 to 20% of wild type. A study of the KRE11 gene has revealed that its product, Kre11p, is a 63 kDa cytoplasmic protein which appears to be involved in the regulation of glucan assembly. Through genetic interactions and epistasis assignments with these and other KRE genes, the basis of a biosynthetic pathway for the synthesis of this extracellular matrix polymer has emerged.
A search for genes involved in the regulation of cell surface assembly has led to the identification of SKN7. Sequence analysis of Skn7p revealed a region of homology to the DNA-binding domains found on heat-shock transcription factors, and a distinct region of similarity to a large family of bacterial "two-component" signal-transduction proteins. Two-component systems have historically been confined to prokaryotic organisms, and the identification of SKN7 has raised the possibility that two-component signaling pathways involving phospho-histidine and phospho-aspartate transfer reactions may exist in higher eukaryotes. Skn7p appears to function in yeast as a nuclear localized two-component response regulator, whose transcriptional B activity is regulated through aspartic acid phosphorylation. The Skn7p signal transduction pathway may act in concert with the yeast PKC1-mediated MAP-kinase cascade, to regulate cellular growth events at the cell surface.

The putative role of the S. cerevisiae vacuole in osmohomeostasis, as well as its biogenesis was analysed by taking a mutational approach. 97 mutants unable to tolerate high concentrations of salt were isolated and examined for aberrant vacuolar phenotypes. A comprehensive phenotypic analysis was able to demonstrate that apart from osmosensitivity most mutations conferred other properties such as altered vacuolar morphology, the inability to perform gluconeogenesis and/or the mislocalization of vacuolar proteins to the cell surface. The mutants fall into at least 20 complementation groups, termed ssv for salt sensitive vacuolar mutants, of which 3 genetically overlap with complementation groups isolated by others. This analysis provides evidence that in 5. cerevisiae correct vacuolar biogenesis is required for osmotolerance as well as other important cellular processes. To elucidate vacuolar osmohomeostasis at the molecular level, one gene, SSV7, was cloned from a genomic DNA library by complementation of a ssv7-l mutation and its sequence determined. It encodes a novel 927 amino acid protein with limited structural homology to the functional domains of two nucleotide exchange factors from 5. cerevisiae, namely CDC25 and BUD5. A mutation in the SSV7 gene confers a pleiotropic phenotype including fragmented vacuoles and the absence of a nucleus. The mutation apparently uncouples mitosis from cell growth, which results in an increased cell size of a ssv7-l mutant. The role of SSV7 in the general concept of vacuole biogenesis is discussed, as is the observed fact that a number of mutants in intermediate biosynthetic pathways confer a vacuolar protein sorting defect.

This thesis examines the post-transcriptional regulation of the gluconeogenic mRNAs PCK1 and FBP1 in Saccharomyces cerevisiae. By construction of a set of chimaeric PCK1-PGK1 reporter genes, it was possible to show that sequences from the 5' end of the PCK1 mRNA were capable of conferring post-transcriptional glucose regulation on the reporter. These regions were presumably the ultimate target of the glucose signal on the wild-type PCK1 mRNA. Sequences from the 5' end of the SDH2 mRNA had previously been shown to be required for the rapid degradation of that transcript observed in the presence of glucose. In addition, a number of other elements within the PCK1 gene proved capable of constitutively destabilising the reporter mRNA. One of these bore similarity to a 70 bp mRNA destabilising element from the highly unstable MATα1 mRNA. Overall, the decay patterns of the chimaeric PCK1-PGK1 and FBP1-PGK1 mRNAs were consistent with other studies of the decay of chimaeric PGK1 mRNAs.

A search of the Saccharomyces genome data base revealed an open reading frame of 2039 amino acids with homology to L-type calcium channels. Northern blots probed with a 540 bp PCR product of the ORF showed a transcript of 6.1 kb. Two procedures were used to disrupt the gene: the ORF was truncated by an integrative disruption after the third pore motif, or deleted in the first three pore domains using a one-step disruption construct. In most strains tested, the disruptions gave no apparent phenotype when tested under a variety of conditions. However, conspicuous phenotypes were seen in the strain YEL161-2A, a strain super-sensitive to alpha-mating factor (sst1). In most respects, truncation gave a less severe phenotype than deletion, suggesting that the truncated gene retains partial function. Calcium uptake during normal growth, as well as the increased calcium uptake in response to mating factor, were reduced progressively by the truncation and deletion respectively. Growth rate and cell viability were reduced, cell size heterogeneity increased, and recovery from mating factor arrest was delayed and abnormal. The cells became sensitive to MnCl$ sb2.$ The phenotype resulting from gene truncation was alleviated by a high-calcium medium, and exacerbated by low calcium. Complementation of the deleted strain by a Yep13 plasmid containing BAR1 (SST1) restored normal growth and viability. However, somewhat paradoxically, deletion of the putative calcium channel gene in another sst1 strain (SY1159) showed no phenotype.

Dissertation (PhD)--University of Stellenbosch, 2004.
ENGLISH ABSTRACT: Cells of the yeast Saccharomyces cerevisiae are able to change their morphological appearance in response to a variety of extracellular and intracellular signals. The processes involved in morphogenesis are well characterised in this organism, but the exact mechanism by which information emanating from the environment is integrated into the regulation of the actin cytoskeleton and the yeast cell cycle, is still not clearly understood. Considerable progress has, however, been made. The processes are investigated on various levels including: (i) the nature of the signals required to elicit a morphological adaptation, (ii) the mechanism by which these signals are perceived and transmitted to the nucleus for gene transcription regulation (signal transduction pathways), (iii) the role of the cytoskeleton, particularly actin, in morphogenesis, and (iv) the relationship between cell cycle regulators and factors required for alterations in cellular shape. The focus of this study was on elements involved in the regulation of one of these morphological processes, pseudohyphal formation, in S. cerevisiae. During pseudohyphal differentiation normal oval yeast cells become elongated and mother and daughter cells stay attached after cytokinesis to give rise to filaments. These filaments are able to penetrate the growth substrate, a phenomenon referred to as invasive growth. Actin remodelling is a prerequisite for the formation of elongated cells during pseudohyphal development and invasive growth. Its main contribution to this event is the directing of vesicles, containing cell wall constituents and enzymes, to specific sites of cell wall growth at the cell periphery. In order to fulfil this cellular function, actin is regulated on several levels. Signal transduction pathways that are activated in response to external nutritional signals play important roles in the regulation of the actin cytoskeleton during pseudohyphal differentiation. For this reason a literature review was compiled to introduce various aspects of actin-structure, the regulation of this structure and the functions actin performs during morphogenesis. The connection between signal transduction elements involved in morphological processes and actin remodelling is also reviewed. This study entailed the genetic analysis of numerous factors involved in the regulation of pseudohyphal differentiation, invasive growth and starch metabolism. Several transcriptional regulators playing a role in these phenomena were investigated. Apart from the transcription factors, which include Mss11p, Msn1p, Ste12p, F108p,Phd1p and Tec1p, additional elements ranging from transporters to G-proteins, were also investigated. Mutant strains deleted for one or more of these factors were constructed and tested to assess their abilities to form filaments that penetrate the growth substrate, and to utilise starch as a carbon source. Complex genetic relationships were observed for various combinations of these factors. Specifically, F108p,Msn1p and Ste12p were shown to act independently in controlling invasive growth and starch metabolism, suggesting that these factors are regulated by different signal transduction pathways. Mss11p, on the other hand, was found to play an indispensable role and seems to act as a downstream factor of Msn1 p, Fl08p, Ste12p and Tec1 p. The exception to this is Phd1 p, since multiple copies of PHD1 partially suppress the effect of a MSS11 deletion. The data suggests that Mss11 p functions at the confluence of several signalling pathways controlling the transcriptional regulation of genes required for invasive growth and starch degradation. Different nutritional signals were also found to differentially regulate specific signalling elements during the invasive growth response. For example, Tec1 p requires Msn1 p activity in response to growth on media containing a limited nitrogen source. This dependency, however, was absent when invasive growth was tested on glucose and starch media. Evidence was also obtained that confirmed the transcriptional co-regulation of MUC1 and STA2. MUC1 encodes a mucin-like protein that is required for invasive growth and pseudohyphal differentiation, whereas STA2 encodes a glucoamylase required for starch degradation. Unpublished results indicated that several transcriptional regulators of invasive growth also exert an effect on starch metabolism. The data generated during this study complemented and confirmed published results. It also contributed to the compilation of a more detailed model, integrating the numerous factors involved in these signalling processes.
AFRIKAANSE OPSOMMING: Saccharomyces cerevisiae gisselle beskik oor die vermoë om hul morfologiese voorkoms in responstot 'n verskeidenheid van ekstrasellulêre en intrasellulêre seine te verander. Die prosesse betrokke by morfogenese is goed gekarakteriseerd in hierdie organisme, maar die presiese meganisme waardeur inligting vanuit die omgewing geïntegreer word in die reguleringvan die aktien-sitoskelet en die gisselsiklus, word nog nie ten volle verstaan nie. Aansienlike vordering in die verband is egter gemaak. Die prosesse word op verskeie vlakke ondersoek, insluitende: (i) die aard van die seine wat benodig word om 'n morfologiese aanpassing te inisïeer; (ii) die meganisme waardeur hierdie seine waargeneem en herlei word na die selkern vir die regulering van geen-transkripsie (seintransduksie paaie); (iii) die rol van die sitoskelet, spesifiek aktien, in morfogenese en (iv) die verhouding tussen selsiklusreguleerders en faktore wat benodig word vir verandering in selvorm. Hierdie navorsing fokus op elemente betrokke by die regulering van een van hierdie morfologiese prosesse in S. cerevisiae, naamlik pseudohife-vorming. Gedurende pseudohife-differensiëring neem tipiese ovaalvormige selle 'n verlengde voorkoms aan wat tot die vorming van filamente lei. Hierdie filamente is in staat om die groeisubstraat te penetreer, 'n verskynsel bekend as penetrasie-groei. Aktienherrangskikking is 'n voorvereiste vir die vorming van verlengde selle tydens pseudohife-ontwikkeling. Die hoofbydrae van aktien tot hierdie verskynsel is die oriëntering van uitskeidingsvesikels, wat selwandkomponente en ensieme bevat, na spesifieke areas van selwandgroei op die seloppervlak. Aktien word op verskeie vlakke gereguleer om hierdie sellulêre funksie te vervul. Seintransduksiepaaie wat geaktiveer word in respons tot ekstrasellulêre voedingsseine speel 'n belangrike rol in die regulering van die aktien-sitoskelet tydens pseudohife-differensiëring. Op grond hiervan is 'n literatuuroorsig saamgestel vir die bekendstelling van verskeie aspekte van aktienstruktuur, die regulering van hierdie strukture en die funksies wat deur aktien gedurende morfogenese vervul word. Die verband tussen seintransduksie-elemente betrokke by morfologiese prosesse en aktien herrangskikkingword ook behandel. Hierdie studie het die genetiese analisering van verskeie faktore betrokke by pseudohife-differensiëring, penetrasie-groei en styselmetabolisme, behels. Verskeie transkripsionele reguleerders wat In rol speel in hierdie prosesse was bestudeer. Buiten die transkripsiefaktore Mss11p, Msn1p, Ste12p, F108p,Phd1P en Tec1p, was addisionele faktore, wat gewissel het van transporters tot G-proteïene, ook ondersoek. Mutante-rasse met geendelesies vir een of meer van hierdie faktore is gekonstrueer en getoets om vas te stel hoe dit hul vermoë raak om penetrerende filamente te vorm, asook om te bepaal of stysel as koolstofbron gebruik kan word. Komplekse genetiese interaksies vir verskeie kombinasies van hierdie faktore is waargeneem. Dit was waargeneem dat F108p,Msn1p en Ste12p onafhanklik funksioneer tydens die regulering van penetrasie-groei en styselmetabolisme, wat impliseer dat hierdie faktore deur verskillende seintransduksiepaaie gereguleer word. Mss11 p word beskou as In onmisbare rolspeler in hierdie prosesse en dit kom voor asof hierdie protein as 'n stroom-af faktor is en vereis word vir die funksionering van Msn1p, F108p, Ste12p en Tec1p. Phd1p is egter 'n uitsondering, aangesien veelvuldige kopieë van PHD1 die effek van 'n MSS11-delesie gedeeltelik oorkom. Die data impliseer dat Mss11 p by die samevloei van verskeie seintransduksiepaaie, benodig vir die transkripsionele regulering van gene betrokke by penetrasie-groei en styselmetabolisme, funksioneer. Dit was ook waargeneem dat verskillende voedingsseine die faktore betrokke by die penetrasie-groeirespons differensieel reguleer. Tec1 p byvoorbeeld benodig Msn1paktiwitieit in respons tot groei op media met 'n beperkte stikstofbron. Hierdie afhanklike interaksie is egter afwesig wanneer penetrasie-groei bestudeer word op glukose- en styselmedia. Resultate wat die gesamentlike transkripsionele regulering van MUC1 en STA2 bevestig, is ook verkry. MUC1 kodeer vir 'n mukienagtige proteïen wat benodig word vir pseudohife-vorming en penetrasie-groei, terwyl STA2 kodeer vir 'n glukoamilase essensieël vir styselafbraak. Ongepubliseerde resultate dui daarop dat verskeie transkripsionele reguleerders van penetrasie-groei ook In effek uitoefen op styselmetabolisme. Die data wat gegenereer is tydens hierdie studie komplementeer en bevestig reeds gepubliseerde resultate. Dit het ook bygedra tot die samestelling van 'n gedetaileerde model wat die verskillende faktore, betrokke by hierdie seintransduksieprosesse, integreer.

The N-terminal tails of histone proteins protrude from the nucleosome core and are extensively post-translationally modified. These modifications are proposed to affect many DNA-based processes such as transcription, DNA replication and repair. Post-translational modifications on histone tails do not act independently but are subject to crosstalk. One example of crosstalk is on histone H3 between lysine 14 (H3K14) and trimethylated lysine 4 (H3K4me3), a modification found at the 5’ end of most active or poised genes. In this work, Western blots and chromatin immunoprecipitation (ChIP) experiments show that different amino acid substitutions at histone H3 position 14 cause varying degrees of H3K4me3 loss, indicating that H3K14 is not essential for H3K4me3 but acts as a modulator of H3K4me3 levels. A neighbouring residue, H3P16 is also important for H3K4me3 and may operate in concert with H3K14 to control H3K4me3. These crosstalk pathways have gene-specific effects and the levels of H3K4me3 are influenced to different extents on genes that fall into functionally distinct classes. A model is proposed to explain how H3K14/H3P16 may exert these varying effects on H3K4me3 at individual genes. In addition to its ability to regulate H3K4me3, H3K14 also influences the levels of two modifications on H3K18, acetylation and monomethylation. A ChIP-sequencing experiment has shown that H3K18me1, a previously uncharacterised modification in S. cerevisiae, is widely distributed throughout the genome and correlates strongly with histone H3 levels. The potential for a functional acetyl/methyl switch at H3K18 is explored. Together, these data indicate that, with gene-specific effects, crosstalk between histone modifications may be even more complex than originally thought.

Expression of cDNA clones of the M1 double-stranded RNA killer preprotoxin coding region in Saccharomyces cerevisiae successfully directed the synthesis of secreted active toxin. Transformants harbouring these expression plasmids also displayed a K1 specific immunity phenotype. Immunoprecipitation of intracellular proteins with antitoxin antiserum showed that these transformants synthesize a 42kd glycosylated preprotoxin precursor. Two smaller unglycosylated immunoreactive species could also be resolved. These toxin precursor species were characterized by using secretory-defective hosts, by comparative electrophoretic mobilities, and by tunicamycin susceptibility. Such studies indicate that these protein species represent intermediates generated by signal cleavage of the preprotoxin and its subsequent glycosylation and provide evidence that these events occur post-translationally. Mutational analysis of the 44 amino acid preprotoxin N-terminal leader indicated that it is functionally bipartite, consisting of an N-terminal signal sequence and a C-terminal pro-sequence. Deletion of the leader perturbed but did not eliminate secretion of toxin.

The Saccharomyces cerevisiae KEX1 gene product, Kex1p, has been identified and partially characterised to assess its role in processing secreted protein precursors and to define its intracellular location. Kex1p antiserum identified a 113 kDa protein that was absent in kex1-$ Delta$ cells and more abundant in cells overexpressing KEX1. Kex1p was found to be a type I membrane associated glycoprotein with N-linked carbohydrate. The N-linked oligosaccharide was modified in a progressive manner after synthesis, causing the glycoprotein to slowly increase in mass to 115 kDa.
After a Kex2p-mediated cleavage event at specific pairs of basic amino acids, $ alpha$-factor and K1 killer toxin precursors have COOH-terminal dibasic residue extensions and require a carboxypeptidase B-like activity to process the precursors to maturity. A carboxypeptidase activity, with apparent specificity for basic amino acids, was detected in KEX1 cells. Disruption of the KEX1 gene abolished this activity, while overexpression of KEX1 increased it. These results provide biochemical evidence, consistent with earlier genetic work, that KEX1 encodes a serine carboxypeptidase involved in the processing of precursors to secreted mature proteins.
Immunological and activity studies indicate that most Kex1p is intracellular and suggests that the enzyme is retained within the secretory pathway. COOH-terminal truncations of the protein indicate that the cytoplasmically exposed domain of Kex1p is responsible for correct localisation of the protein, probably in the late Golgi.
When KEX1 was expressed in Schizosaccharomyces pombe, Kex1p was localised in structures consistent with components of the Golgi. Mammalian cells expressing KEX1 produce a membrane associated activity that is not detected in the medium. In immunofluorescence studies on mammalian cells, Kex1p was localised to the ER and Golgi but not to the plasma membrane. Kex1p in such cells was responsible for completing the processing of the neuropeptide, $ gamma$-lipotropin. This in vivo processing of $ gamma$-lipotropin by Kex1p demonstrates a significant functional homology of the basic prohormone processing machinery in yeast and neuroendocrine cells.

Isoleucyl-tRNA synthetase genes from the yeast Saccharomyces cerevisiae and the ciliated protozoan Tetrahymena thermophila were sequenced. The intronless S. cerevisiae gene (ILS1) encodes a putative polypeptide of 1072 amino acids. Two putative promoter elements were identified, one for general amino acid control and one for constitutive transcription. A heat shock protein gene lies upstream of ILS1. The T. thermophila isoleucyl-tRNA synthetase gene (ilsA: formerly cupC) has eight introns, four transcription start sites, and codes for a putative polypeptide of 1081 amino acids with two leucine-zippers. These eukaryotic isoleucyl-tRNA synthetases are 47% identical. They are compared to homologous enzymes from Escherichia coli and an archaebacterium, and to other aminoacyl-tRNA synthetases.
Intron sequences and junctions from T. thermophila and other eukaryotes were analyzed and all but yeast and mammalian introns were found to be A + T enriched. T. thermophila transcription start sites were analyzed and occur at a T or an A within the consensus sequence (A/T)$ sb{ rm n}$ T A A (A)$ sb{ rm n}.$

Transcription by RNA polymerase II is a highly regulated process requiring a number of general and promoter specific transcription factors. Although many of the factors involved in the transcription reaction are known, exactly how they function to stimulate or repress transcription is not well understood. Central to understanding gene regulation is understanding the mechanism by which promoter specific transcription activators (activators) stimulate transcription. A group of factors called coactivators have been shown to be required for activator function in vitro. The best characterized coactivators to date are members of the TFIID complex. TFIID is a multisubunit complex composed of the TATA box binding protein (TBP) and 8-12 TBP associated factors (TAFIIs). Results from numerous in vitro experiments indicate that TAFIIs function by binding to activators and forming a bridge between the activator and the basal transcription machinery. In order to gain insight into the mechanism by which activators stimulate transcription, we chose to analyze the in vivo function of TAFIIs, their proposed targets. Results from the genetic disruption of a number of TAFIIs in the yeast Saccharomyces cerevisiae showed that most are encoded by essential genes. In order to study their function, temperature-sensitive and conditional alleles were constructed. Cells depleted of individual TAFIIs by either of these two methods displayed no defect in global transcription activation. Inactivation of yTAFII17, however, resulted in a promoter specific defect. In addition, inactivation of yTAFII145, yTAFII90, or TSM1, resulted in an inability of cells to progress through the cell-cycle. In an attempt to identify genes whose expression required yTAFII90, we performed subtractive hybridization on strains containing wild-type and temperature-sensitive alleles. Although this technique successfully identified genes differentially expressed in the two strains, it failed to identify genes whose expression required yTAFII90. These results indicate that TAFIIs are not the obligatory targets of activators, and that other factors must provide this role in vivo. Furthermore, that many of TAFIIs are required for cell-cycle progression.

Much of the eukaryotic transcriptional machinery is conserved from yeast to human. However, the distance over which transcriptional activation can occur differs between Saccharomyces cerevisiae and metazoans. In S. cerevisiae, the upstream activating sequence (UAS) is generally found within 300 base pairs of the transcription start site; when the UAS is moved too far away, activation no longer occurs. In contrast, metazoan enhancers can activate from as far as 100 kilobases from the start site. In past work, our lab identified five genes that, when mutant, allow transcription activation to occur at a greater-than-normal distance from the GAL1 UAS. As this long-distance activation phenotype was weak, we have now studied long-distance activation as a polygenic trait, isolating strains with multiple mutations that together confer a strong phenotype. To do this, we constructed strains containing two reporters, HIS3 and URA3. For each reporter, the GAL1 UAS was placed approximately 800 base pairs upstream of the transcription start sites. By iterative selection for stronger and stronger expression of HIS3, followed by screening for stronger expression of URA3, we isolated three strains, each containing multiple mutations that contribute to the strength of the long distance activation phenotype. Causative mutations were identified in MOT3, GRR1, MIT1, PTR3, YOR019W, and MSN2 that contribute to the long distance activation phenotype. Strains containing multiple mutations were found to activate the reporter construct at distances up to 2 kilobases. Microarray analysis revealed genome wide transcriptional changes in the mutant strains. Statistical analysis of the microarray results suggests other potential sites of long distance activation throughout out the genome. These results have extended our understanding of mutations that allow long distance activation and have demonstrated the value of studying a phenotype as a polygenic trait.

The bipolar budding pattern of a/a Saccharomyces cerevisiae cells appears to depend on persistent spatial markers. Genetic analysis reported here indicates that BUD8 and BUD9 potentially encode components of the markers at the distal and proximal poles, respectively. Mutants deleted for BUD8 or BUD9 bud exclusively from the proximal and distal poles, respectively, and the double-mutant phenotype suggests that the bipolar budding pathway has been totally disabled. Moreover, overexpression of these genes can cause either an increased bias for budding at the distal (BUD8) or proximal (BUD9) pole or a randomization of bud position, depending on the level of expression. Both molecules are related plasma membrane glycoproteins that are both N- and O-glycosylated. Each protein was localized predominantly in the expected location, with Bud8p delivered to the presumptive bud site just before bud emergence, and Bud9p delivered to the bud side of the mother-bud neck just before cytokinesisis. Promoter-swap experiments revealed the importance of time of transcription in localization: expression of Bud8p from the BUD9 promoter leads to its localization predominantly in the sites typical for Bud9p, and vice versa. Moreover, expression of Bud8p from the BUD9 promoter fails to rescue the budding-pattern defect of a bud8 mutant but fully rescues that of a bud9 mutant. However, although expression of Bud9p from the BUD8 promoter fails to rescue a bud9 mutant, it also rescues only partially the budding-pattern defect of a bud8 mutant.
Using a collection of mutants individually deleted for almost every yeast gene, I undertook a genome-wide phenotype analysis for altered sensitivity to a yeast antifungal protein, the K1 killer toxin. Mutations in most genes have no effect on toxin sensitivity, with less than 10% having a phenotype. Only 4% of these were previously known to have a toxin phenotype. There is a markedly non-random functional distribution of mutants with a toxin phenotype. Many genes fall into a limited set of functional classes or modules, which define specific areas of cellular function. These include known pathways of cell wall synthesis and signal transduction, and offer new insights into these processes and into cell wall morphogenesis.

We have previously demonstrated that Mid2p is required for the activation of the PKC1-MPK1 cell integrity pathway during cell exposure to mating pheromone, calcofluor white (CFW), and heat. Accumulating evidence indicates that Mid2p might regulate this pathway via the small GTPase, Rho1p. To understand the mechanism by which Mid2p signals, we initiated a two hybrid screen using the essential cytoplasmic tail of Mid2p as bait. ZEO1 (YOL109w), a previously uncharacterized open reading frame, was identified. ZEO1 encodes a 12kDa protein that co-localizes to the plasma membrane and interacts with the cytoplasmic tail of Mtl1p, a Mid2p functional homologue. Like mid2Delta mutants, cells deleted for ZEO1 are resistant to calcofluor white. In addition, ZEO1 null strains are no longer hypersensitive to calcofluor white caused by high copy expression of MID2. A role for Zeo1p in the cell integrity pathway is supported by the finding that disruption of ZEO1 leads to a Mid2p-dependent constitutive phosphorylation of Mpk1p. (Abstract shortened by UMI.)

The PMT (Protein O-mannosyl transferase) family members are involved in the initial step of protein-O-mannosylation. A large scale procedure (Systematic Genetic Array) was performed using pmt1, pmt2, pmt3, pmt5, and pmt6 as query deletions and a set of 4700 non-essential array gene deletions, to screen for query/array double deletion mutant combinations affecting growth of Saccharomyces cerevisiae. This procedure revealed a genetic interaction network consisting of 53 interacting genes. Functional grouping of these 53 genes revealed 9 functional categories that were analyzed according to gene function to elucidate how they might buffer defects in protein-O-mannosylation. Synthetic genetic interactions were also identified between PMT family members demonstrating redundancies among them. Protein-O-mannosylation is a protein modification conserved from yeast to human. POMT1 and POMT2 (Human PMTs counterparts) catalyze mannosyl residue transfer in mammals, with mutations identified to be involved in Walker-Warburg Syndrome (WWS). ∼72 % of the genes in the yeast PMT genetic network have human homologs, and ∼55 % of these are associated with human disease. Using the yeast genetic interaction network as a model for human genetic interactions may help in the understanding of complex inherited human disease.

Mutations in the yeast gene KRE1 lead to resistance to the K1 killer toxin of S. cerevisiae. The resistant phenotype is associated with a 40% reduction of the cell wall (1 $ to$ 6)-$ beta$-glucan fraction. Yeast cell wall (1 $ to$ 6)-$ beta$-glucan is a highly branched glucose polymer composed predominantly of linear (1 $ to$ 6)-$ beta$- sc D-linked glucopyranosyl residues. This glucan acts as a receptor for the killer toxin, leading to a concentration of active toxin on the yeast cell surface. The KRE1 gene was cloned by complementation of the kre1-1 mutant allele. The KRE1 gene encodes a serine and threonine rich protein, that is directed into the yeast secretory pathway, where it is highly modified, probably through O-linked glycosylation. Haploid strains with a kre1::HIS3 disruption appear to grow somewhat more slowly than wild type, and show an unusual wall structure when examined using electron microscopy. As with strains that carry a mutant kre1-1 allele those with a kre1::HIS3 disruption show a 40% reduced level of cell wall (1 $ to$ 6)-$ beta$-glucan. Structural comparison of the (1 $ to$ 6)-$ beta$-glucan fraction isolated from a wild type strain and a kre1 mutant, showed that the glucan obtained from the mutant had fewer (1 $ to$ 6)-linked residues and displayed a smaller average polymer size. Therefore, the KRE1 gene product appears to be required for the stepwise synthesis of cell wall (1 $ to$ 6)-$ beta$-glucan.

Analyses of genes involved in yeast cell wall $ beta$1,6-glucan assembly have identified CWH41, PTC1/CWH47, EXG1, PBS2 and a family of genes related to the human oxysterol binding protein, OSBP. CWH41 encodes a novel membrane N-glycoprotein located in the ER. Disruption of CWH41 leads to a K1 killer toxin resistant phenotype, and a 50% reduction in the $ beta$1,6-glucan level The $cwh{ it 41 /} Delta$ mutant displayed strong synergistic defects with $kre{ it 1 /} Delta$ or $kre{ it 1 /} Delta$ null mutations: the $cwh{ it 41 /} Delta kre{ it 6 /} Delta$ double mutant showed a slow-growth phenotype and a 75% reduction in $ beta$1,6-glucan level, and cells containing $cwh{ it 41 /} Delta kre{ it 6 /} Delta$ double mutations were nonviable. These results indicate that CWH41 is involved in the assembly of $ beta$1,6-glucan.
PTC1/CWH47 encodes a serine/threonine phosphatase, PBS2 is the structural gene for a MAPK kinase, and EXG1 codes for an exo-$ beta$-glucanase. Overexpression of EXG1 led to a killer resistant phenotype and a reduction in ($ beta$1,6-glucan level; whereas the $exg{ it 1 /} Delta$ mutant displayed modest increases in killer sensitivity and $ beta$1,6-glucan levels. Disruption of PTC1/CWH47 and overexpression of PBS2 resulted in similar $ beta$-glucan related phenotypes, with elevated EXG1 transcription, increased Exg1p activity, reduced $ beta$1,6-glucan levels, and resistance to killer toxin. The killer resistant phenotype caused by PTC1/CWH47 disruption or PBS2 overproduction were partially suppressed by the $exg{ it 1 /} Delta$ null mutation. These results suggest that Ptc1p/Cwh47p and Pbs2p play opposing regulatory roles in $ beta$-glucan assembly, and this is effected in part by modulating Exg1p activity.
Three yeast genes, KES1, HES1 and OSH1, whose products show homology to the human oxysterol binding protein, were also identified. Mutations in these genes resulted in sterol-related phenotypes, which include tryptophan-transport defects and nystatin resistance. In addition, mutant combinations showed small but cumulative reductions in membrane ergosterol levels. The three genes are also functionally related; since overexpression of HES1 or KES1 alleviated the tryptophan-transport defect in $kes{ it 1 /} Delta$ or $osh{ it 1 /} Delta$ mutants, respectively. These observations implicate the KES1-related gene family in ergosterol synthesis and provide comparative evidence of a role for human OSBP in cholesterol synthesis.

The Saccharomyces cerevisiae genes KRE6 and SKN1 encode a novel pair of highly homologous proteins involved in cell wall (1$ rightarrow$6)-$ beta$-glucan assembly. Disruption of KRE6 results in a slow growing, killer-toxin resistant mutant possessing reduced levels of structurally wild type (1$ rightarrow$6)-$ beta$-glucan. Although deletion of SKN1 has no effect on killer sensitivity, growth, or (1$ rightarrow$6)-$ beta$-glucan levels, SKN1 appears to overlap in function with KRE6, suppressing kre6 null alleles in a dosage-dependent manner. Strains deleted of both KRE6 and SKN1 possess an exaggerated growth phenotype, enhanced cell wall ultrastructure defects, and more severe (1$ rightarrow$6)-$ beta$-glucan reductions compared with either single disruptant. Moreover, the residual (1$ rightarrow$6)-$ beta$-glucan polymer in kre6 skn1 double mutants is smaller in size and altered in structure. Since single disruptions of either gene lead to structurally wild type (1$ rightarrow$6)-$ beta$-glucan, KRE6 and SKN1 appear to function independently and to act early in the assembly of the polymer, possibly as glucan synthases. Consistent with their direct role in the assembly of this polymer, both Kre6p and Skn1p possess C-terminal domains with significant sequence similarity to two recently identified glucan-binding proteins.
An initial characterization of Kre6p and Skn1p reveal both to be phosphorylated integral-membrane glycoproteins, with Kre6p likely localized to the Golgi apparatus. The topology implied by the post-translational modifications of Kre6p and Skn1p, offers the potential for both proteins to link cytoplasmic regulation with a secretory pathway-based assembly of the (1$ rightarrow$6)-$ beta$-glucan polymer. The observed phosphorylation of both Kre6p and Skn1p prompted an examination for genetic interactions with suspected cell wall regulating kinases. KRE6-dependent suppression of the pkc1 lysis defect, as well as synthetic lethal interactions between several KRE genes and members of the PKC1-mediated MAP kinase pathway, supports a role for the PKC1 pathway in regulating synthesis of cell wall components, including (1$ rightarrow$6)-$ beta$-glucan.

The Saccharomyces cerevisiae SKN7 gene has been identified through a search for genes which, at a high copy number, could restore the growth of the $kre9 Delta$ disrupted strain showing a cell wall $ beta$-glucan defect. SKN7 was mapped to the right arm of chromosome VIII, and is predicted to encode a 70 kDa protein, Skn7p, with a region of homology to the DNA binding domain of the yeast heat shock transcription factor, Hsf1p. Skn7p also has a domain which shows similarity to the prokaryotic receiver modules found on an extensive family of two-component response regulators, including the product of the rcsC gene. While restoring the growth rate to near wild type levels, SKN7 does not appear to restore the $ beta$-glucan levels of the $kre9 Delta$ mutant. However, SKN7-suppressed cells show a partially restored cellular morphology, and a restored cell wall resistance to mechanical stress. SKN7 does not suppress other mutations in the ($1 rightarrow6$)-$ beta$-glucan biosynthetic pathway, suggesting that it does not act as a general bypass suppressor of this glucan.

Saccharomyces cerevisiae Kre5p is important for the biosynthesis of beta-1,6-glucan, which is required for proper cell wall assembly and architecture. A functional and cell biological analysis of Kre5p was conducted to further elucidate its role in beta-1,6-glucan synthesis. Kre5p was found to be a primarily soluble N-glycoprotein of ∼200 kD that localizes to the endoplasmic reticulum. Observation of Kre5p-deficient cells reveals a severe cell wall morphological defect, and kre5Delta cells were shown to have only residual levels of beta-1,6-glucan. KRE6 was identified as a multicopy suppressor of a temperature-sensitive kre5 allele, suggesting these proteins participate in a common pathway. An analysis of truncated versions of Kre5p indicates that it may have two independent, essential activities, or that it functions in a homodimeric state. Finally, Candida albicans KRE5 was shown to partially restore growth to kre5Delta cells, suggesting it has a function similar to that of the S. cerevisiae protein.

ChIP-chip is a technology originally developed to determine the binding sites of proteins in chromatin on a genome wide scale. Its uses have since been expanded to analyse other genome features, such as epigenetic modifications and, in our laboratory, DNA damage. Datasets comprise many thousands of data points and therefore require bioinformatic tools for their analysis. Currently available tools are limited in their applications and lack the ability to normalise data so as to allow relative comparisons between different datasets. This has limited the analyses of multiple ChIP-chip datasets from different experimental conditions. The first part of the study presented here is bioinformatic, presenting a selection of tools written in R for ChIP-chip data analysis, including a novel normalisation procedure which allows datasets from different conditions to be analysed together, permitting comparisons of values between different experiments and opening up a new dimension of analysis of these datasets. A novel enrichment detection procedure is presented, suited to many formats of data, including protein binding (which forms peaks) and epigenetic modifications (which can form extended regions of enrichment). Graphical tools are also presented, to facilitate the analysis of these large datasets. A method of predicting the output of a ChIP-chip dataset is presented, which has been used to show that ChIP-chip is capable of detecting sequence dependent damage events. All functions work together, using a common data format, and are effcient and easy to use. The second part of this study applies these bioinformatic tools in a biological context. An analysis of Abf1 protein binding datasets has been undertaken, revealing many more binding sites than had previously been identified. Analysis of the sequences at these binding sites identifed the previously determined consensus binding motif in only a subset, with no novel motif identifiable in the remainder, suggesting binding may be in uenced by factors other than sequence.

CWH41 is required for beta-1,6-glucan biosynthesis and encodes glucosidase I, an enzyme involved in protein N-chain glucose processing. Therefore, the effects of N-chain glucosylation and processing on beta-1,6-glucan biosynthesis were examined, and it was shown that incomplete N-chain glucose processing results in loss of beta-1,6-glucan. To explore the involvement of other N-chain-dependent events with beta-1,6-glucan synthesis, the S. cerevisiae KRE5 and CNE1 genes were investigated, which encode homologs of the 'quality control' components UDP-Glc:glycoprotein glucosyltransferase and calnexin, respectively. The essential activity of Kre5p was found to be separate from its possible role as a UDP-Glc:glycoprotein glucosyltransferase. A ∼30% decrease in beta-1,6-glucan was observed upon disruption of CNE1, a phenotype which is additive with other beta-1,6-glucan synthetic mutants. Analysis of the cell wall anchorage of alpha-agglutinin suggests the existence of two beta-1,6-glucan biosynthetic pathways, one N-chain dependent, the other involving protein glycosylphosphatidylinositol modification.
Fks1p and Fks2p are related proteins thought to be catalytic subunits of the beta-1,3-glucan synthase. The fks1Delta mutant was partial K1 killer toxin resistant and showed a 30% reduction in alkali-soluble beta-1,3-glucan that was accompanied by a modest reduction in beta-1,6-glucan. The gas1Delta mutant lacking a 1,3-beta-glucanosyltransferase displayed a similar reduction in alkali-soluble beta-1,3-glucan but did not share the beta-1,6-glucan defect, indicating that beta-1,6-glucan reduction is not a general phenotype among beta-1,3-glucan biosynthetic mutants. FKS2 overexpression suppressed the killer toxin phenotype of fks1Delta mutants, implicating Fks2p in the biosynthesis of the residual beta-1,6-glucan present in fks1Delta cells. Eight out of twelve fks1tsfks2Delta mutants had altered beta-glucan levels at the permissive temperature: the FKS1F1258Y N1520D allele was severely affected in both polymers and displayed a 55% reduction in beta-1,6-glucan, while the in vitro hyperactive FKS1T6051 M761T allele increased both beta-glucan levels. These beta-1,6-glucan phenotypes may be due to altered availability of, and structural changes in, the beta-1,3-glucan polymer, which might serve as a beta-1,6-glucan acceptor at the cell surface. Alternatively, Fks1p and Fks2p could actively participate in the biosynthesis of both polymers as beta-glucan transporters. beta-1,6-Glucan deficient mutants had reduced in vitro glucan synthase activity and mislocalized Fks1p and Fks2p, possibly contributing to the observed beta-1,6-glucan defects.

There are two copies of each core histone in a nucleosome, however, it is unclear whether post-translational modifications on each molecule function redundantly or if symmetrical modifications are required to properly regulate gene expression. We tried to address this question by breaking nucleosomal symmetry and measuring its impact on gene expression. Our strategy includes re-engineering specific residues at the H3-H3 interface, generating pairs of mutant proteins, which were predicted by computational methods to form obligate heterodimers. Using S. cerevisiae as a model system, we tested the viability of strains with mutant histones, and analyzed the interaction between by co-immunoprecipitation from mononucleosome preparations. We also measured the changes of gene expression in the strains bearing single-tailed or tailless H3 heterodimers. The data suggested that the best computationally-derived H3 pair was frequently, but not exclusively heterodimeric in vivo. In order to obtain a more stringent H3 heterodimer, random mutagenesis was performed on four codons in the original computational design, and then genetic screening of the mutant libraries was performed.