UvA-DARE (Digital Academic Repository)

1,6-beta-glucan synthesis in Saccharomyces cerevisiae

Vink, E.

Publication date 2003 Document Version Final published version

Link to publication

Citation for published version (APA): Vink, E. (2003). 1,6-beta-glucan synthesis in Saccharomyces cerevisiae.

General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

Download date:10 Oct 2021 1,6-p-glucann synthesis in SaccharomycesSaccharomyces cerevisiae l96-/?-glucann synthesis in Saccharomyces cerevisiae l,6-/?-glucann synthesis in Saccharomyces cerevisiae

ACADEMISCHH PROEFSCHRIFT

terr verkrijging van de graad van doctor aann de Universiteit van Amsterdam opp gezag van de Rector Magnificus Prof.. mr. P.F. van der Heijden tenn overstaan van een door het college voor promoties ingestelde commissie,, in het openbaar te verdedigen in de Aula der Universiteit

opp woensdag 3 september 2003, te 10.00 uur

doorr Edwin Vink

geborenn te Zaandam Promotiecommissie: :

Promotor r Prof.. Dr. H. van den Ende

Co-promotores s Dr.. F.M. Klis

Dr.. J.G. de Nobel

Overigee leden Prof.. Dr. S. Brul

Prof.. Dr. M.A. Haring

Prof.. Dr. K.J. Hellingwerf

Prof.. Dr. H.A.B. Wösten

Faculteitt der Natuurwetenschappen, Wiskunde en Informatica

Thee research described in this thesis was carried out in the Swammerdam Institute for Life Sciences,, Section Plant Physiology, University of Amsterdam, Kruislaan 318, NL-1098 SM, Amsterdam,, The Netherlands. "Iederr nadeel heb z'n voordeel!" JohanJohan Cruijf

AanAan mijn ouders

TableTable of contents

Abbreviations s

Chapterr 1 Generall Introduction: 1,6-/?-glucann synthesis in Saccharomyces cerevisiae

Chapterr 2 Thee protein kinase Kiel affects 1,6-/?-glucan levels in the 41 1 celll wall of Saccharomyces cerevisiae

Chapterr 3 Localizationn of synthesis of 1,6-/?-glucan in 71 1 SaccharomycesSaccharomyces cerevisiae

Chapterr 4 InIn vitro measurement of Saccharomyces cerevisiae 1,6-/?- 93 3 glucann synthase activity

Chapterr 5 Generall Discussion 111 1

Summary y 129 9

Samenvatting g 131 1

Dankwoord d 133 3 Abbreviations Abbreviations

BSS A bovine serum albumin CSS chitin synthase CWHCWH calcofluor white hypersensitive CWPP cell wall protein ERR endoplasmic reticulum GCC gas chromatography GFPP green fluorescent protein GPII glycosylphosphatidylinositol GTT UDP-glucose: glycoprotein glucosy transferase HAA hemagglutinin HCAA hydrophobic cluster analysis KREKRE killer resistant MAPP mitogen activated protein MSS mass spectrometry PBSS phosphate buffered saline PIRR proteins with internal repeats PVDFF polyvinylidene fluoride

8 8 ChapterChapter 1

Generall Introduction

l,6-/7-glucann synthesis in Saccharomyces cerevisiae ChapterChapter I

1.. Introduction 11 2.. Cell wall proteins 11 2.1.. Cell wall proteins are heavily glycosylated 11 2.2.. GPI-dependent cell wall proteins 12 2.3.. Pir cell wall proteins 14 3.. Chitin 15 3.1.. Chitin in the cell wall 15 3.2.. The three chitin synthases 16 3.3.. Zymogenic nature of chitin synthases 16 3.4.. Localization of chitin synthases 16 4.1,3-A-Glucann 17 4.1.1,3-/?-Glucann in the cell wall 17 4.2.. Enzymatic properties of the l,3-/?-ghican synthase 17 4.3.. GTP is required for l,3-/?-glucan synthase 17 4.4.. Rhol is the GTP-binding protein 18 4.5.. Fksl and Gsc2 are essential components of 1,3-/?-glucan synthase 18 4.6.. More on the regulation of l,3-/?-ghican synthase 20 4.7.. Localization of l,3-/?-glucan synthase 20 5.. l,6-/?-Glucan 21 5.1.1,6-y?-Glucann in the cell wall 21 5.2.. Identification of genes involved in 1,6-p-glucan synthesis 21 5.3.. Endoplasmic Reticulum (ER) 22 5.4.. Golgi 23 5.5.. Cytoplasm 24 5.6.. Plasma membrane 24 5.7.. Extracellular space 24 5.8.. Localization of l,6-/?-glucan synthesis 25 5.9.. Regulation of l,6-/?-glucan 25 5.10.. Biochemical assay for l,6-/?-glucan 26 6.. References 26

10 0 1,6-fi-glucan1,6-fi-glucan synthesis in Saccharomyces cerevisiae

1.1. Introduction Thee baker's yeast Saccharomyces cerevisiae is protected from its environment by its celll wall. The cell wall - which can be found throughout all fungal species - provides the cell withh both physical protection and osmotic support. Although the term "wall" suggests this structuree to be very solid and rigid, for it to allow growth and mating it must also have dynamicc properties. It is a layer composed of about equal amounts of glucan, and mannoproteinss and a small amount of chitin (1-2%). The glucan consists mainly of 1,3-/?- linkedd residues, and this so-called 1,3-/?-glucan - together with the chitin - can be visualized ass an electron-transparent layer by electron microscopy. This layer makes up the inner part of thee cell wall, while the mannoproteins appear as an electron-dense layer on the outer side. Thee proteins which are covalently linked to the cell wall can be divided into two groups: (I) thee glycosylphosphatidylinositol-dependent cell wall proteins (GPI-CWPs), which are linked too the 1,3-/?-glucan through a 1,6-/?-glucan moiety, and (II) the Pir cell wall proteins (Pir- CWPs),, which are presumably directly linked to l,3-/?-glucan via an alkali-sensitive binding (forr reviews, see Orlean, 1997; Klis et al, 2002). The cell wall proteins are among others involvedd in interactions of the cell with its surroundings, and processes like mating and flocculationn depend on them (Lipke and Kurjan, 1991; Teunissen and Steensma, 1995). Furthermore,, they limit the permeability of the wall to macromolecules (Zlotnik et al, 1984; Dee Nobel et al, 1991) and protect the glucan and the cell itself against external perils. 1,6-/7-glucann is an essential component of the cell wall, which interconnects the GPI- CWPss with the l,3-/?-glucan network as mentioned above (Kapteyn et al, 1996; Kollar et al, 1997;; Kapteyn et al, 1999). Mutations which result in lowered 1,6-/?-glucan levels often lead too decreased viability (Meaden et al, 1990; Roemer et al, 1993; Brown and Bussey, 1993; Vinkk et al, 2002, Chapter I). Also in Candida albicans, genes involved in 1,6-^-glucan were foundd to be essential (Mio et al, 1997; Lussier et al, 1998). This stresses the importance of thiss minor component of the yeast cell wall. However, little is known about its biosynthesis andd regulation.

2.2. Cell wall proteins 2.1.2.1. Cell wall proteins are heavily glycosylated Mannoproteinss destined for the cell wall travel through the secretory pathway before theyy are incorporated into the cell wall. Along this route, they become heavily glycosylated onn both N- and O-glycosylation sites. O-chains are short oligomannosyl side chains, which cann be added onto either serine or threonine residues. However, the asparagine-linked side chainss may contain up to 200 mannose residues, and form large, branched structures. They cann make up 90% of the mannoprotein weight (Orlean, 1997). Both N- and O-chains are initiatedd in the ER, where core structures of the respective chains are added to the protein.

11 1 ChapterChapter 1

Furtherr extension of these core chains is accomplished in the Golgi, which solely concerns mannosyll additions. Although there is evidence for the presence of an UDP-galactose antiporterr in S. cerevisiae, no galactosyl modifications have ever been described (Gemmill andd Trimble, 1999). Besides the addition of mannose residues, both N- and O-chains may alsoo be modified by the addition of mannosyl phosphate. The resulting mannose phospodiesterr bridges give a negative charge to the neutral mannan, the exact function of whichh remains to be clarified. It has been reported, however, that mannosyl phosphorylation iss increased by the addition of KC1 to the growth medium and thus might be involved in stresss response (Jigami and Odani, 1999). 2.2.2.2. GPI-dependent cell wall proteins GPI-anchorss are found in numerous species ranging from fungi to mammals (Ikezawa,, 2002). GPI-anchor addition occurs in thee ER to newly synthesized proteins, and for thiss a carboxy-terminal signal peptide is required. Generally, the signal peptide consists of a cleavage/attachmentt site, a spacer domain of 6 - 14 hydrophilic amino acid residues, and a hydrophobicc carboxy-terminal generally ranging from 15-30 residues. The attachment site, alsoo named the co-site, preferentially is a small amino acid such as glycine or asparagine. At thee co+1 and co+2 positions also small amino acids are found (Udenfriend & Kodukula, 1995; Ikezawa,, 2002; De Groot et al, 2003). In general, GPI-anchors result in a permanent attachmentt of specific proteins to the plasma membrane (Figure 1), but in yeast GPI-anchors cann also function in binding cell wall proteins. Then, the GPI-anchor is cleaved to release the celll wall protein from the plasma membrane, after which the GPI-remnant on the protein is cross-linkedd to 1,6-/?-glucan, which in turn is connected to 1,3-/ï-glucan (Kapteyn et al, 1996;; Kollar et al, 1997; Klis et al, 2002). Thee above raises the question how GPI-anchored proteins that remain bound to the plasmaa membrane are distinguished from those destined for incorporation in the cell wall. A dibasicc motif upstream of the GPI-attachment site might serve to prevent cell wall incorporationn for GPI-anchored proteins (Vossen et al, 1997; Caro et al, 1997; Hamada et al,al, 1999). In contrast, the amino acid at sites 4 or 5 upstream of the GPI-attachment site (the coo site) and the amino acid at site 2 upstream of the co site seem to correlate with cell wall incorporation.. This region is also called the co-minus 5-1 sequence. The presence of valine, isoleucine,, or leucine at the co-5 or co-4 position and tyrosine or asparagine at the co-2 position tendss to direct a GPI-anchored protein to the cell wall (Hamada et al, 1998; Hamada et al, 1999).. The exact mechanism of both negative and positive signals has not yet been resolved. Interestingly,, mutations in the co-minus 5-1 sequence that reduce cell wall incorporation, do nott result in the increased release into the medium of fusion proteins (Hamada et al, 1998). Thiss suggests that the co-minus 5-1 sequence serves as a recognition/target site for the release fromm the plasma membrane, and thereby facilitates cell wall incorporation. Another mechanismm that might help the cell to distinguish between plasma membrane GPI-anchored

12 2 /,, 6-fi-glucan synthesis in Saccharomyces cerevisiae

proteinss and cell wall GPI-anchored proteins, is the clustering of GPI-anchored proteins in sphingolipid-cholesteroll microdomains or rafts. These rafts might function as selective platformss to target proteins to specific locations, possibly driven by differential remodelling off the lipid moiety of the GPI-anchors (Sipos et ai, 1997; Reggiori et ai, 1997; Muniz and Riezman,, 2000).

1,6-p-glucann attachment

r~~ (Man)4 )

myo-lno o

Inositol l phospholipid d

lipidd moiety

Figuree 1. Schematic representation of a GPI-anchored protein. EtN = ethanolamine, P = phosphate, Man mannose,, GlcN = glucosamine, mw-Ino = mvo-inositol, PM = plasma membrane.

Detailedd studies have been dedicated to address the question of how the cell wall proteinss are linked to the 1,6-/?-glucan. The GPI-CWP a-agglutinin has been studied most extensively.. Upon release from the plasma membrane, part of the GPI-anchor is removed fromm the protein, including the inositol residue (Lu et ai, 1994). However, no l,6-/?-glucan wass found to be associated with the periplasmic intermediate form of a-agglutinin. Instead, onlyy a-agglutinin released from the cell wall with laminarinase contained 1,6-/?-glucan (Lu et ai,ai, 1995). This suggests that cross-linking of a-agglutinin to 1,6-/?-glucan results in cell wall attachment.. Close examination of another GPI-CWP, Tipl, revealed that the C-terminus of thiss protein was linked via ethanolamine, a phophodiester bridge, and mannan to 1,6-/?-glucan

13 3 ChapterChapter 1

(Fujiii et al, 1999). Further insight in the linkage between the GPI-remnant and 1,6-/?-glucan camee from a study by Kollar and co-workers (1997). They digested cell walls with 1,3-/?- glucanase,, labeled them by reduction with tritiated borohydride, and then further digested themm with chitinase. After chromatography on a Bio-Gel P2 column, the void volume was subjectedd to ConA-chromatography. The ConA+ fraction was analyzed by NMR, gel chromatography,, immunoblotting, chemical and enzymatic degradation. This study confirmedd that 1,6-/?-glucan is attached to the remnant of the GPI-anchor, through a yet unknownn linkage to the mannan core of the GPI-remnant. Furthermore it was established that thee terminal glucose residue at the reducing end of l,6-/?-glucan is directly attached to a terminall glucose residue at the non-reducing end of a l,3-/?-glucan polymer (Kollar et al, 1997).. Taken together, this suggests that the original GPI-anchor is cleaved between the mannann core and the glucosamine (Figure 1), or at least that glucosamine is lost upon cross- linkagee to 1,6-/?-glucan. It has been proposed that the non-reducing end of 1,6-/?-glucan acts ass the accepting group in a transglycosylation reaction linking it to the remnant of the GPI- anchorr (Kollar et al, 1997). Although the GPI-CWPs are generally connected via 1,6-/?- glucann to l,3-/?-glucan, which in turn is connected to chitin, there are also examples of 1,6-/?- glucann directly linked to chitin, which are predominantly seen under cell wall stress conditionss (Kollar et al, 1997; Kapteyn et al, 1997; Figure 2). 2.3.2.3. Pir cell wall proteins Anotherr class of covalently linked CWPs, Pir-CWPs, does not contain C-terminal GPI-anchorr addition signals, but does get covalently linked to the glucan network of the wall. Fourr PIR (proteins with internal repeats) proteins have been found in S. cerevisiae: Pirl, Pir2/Hspl50,, Pir3, and Pir4/Cis3 (Toh-e et al, 1993; Yun et al, 1997; Mrsa and Tanner, 1999;; Moukadiri et al, 1999). Two PIR genes were initially identified as genes involved in heatt shock tolerance. In addition, a third homolog was identified (Toh-e et al, 1993). Later, it wass found that overexpression of PIR J - PIR3 conferred resistance to tobacco osmotin, and thatt the gene products were localized to the cell wall (Yun et al, 1997). The PIR proteins are covalentlyy linked to the cell wall, probably by direct attachment to l,3-/?-glucan, through a linkagee that is sensitive to mild alkali-treatment (Mrsa et al, 1997; Kapteyn et al, 1999; Mr§aa and Tanner, 1999; Figure 2). All PIR proteins have a similar structure: they consist of ann N-terminal signal peptide, a Kex2 protease cleavage site, and a region of one or more repeatss with the conserved core sequence Q[IV][STGNH]DGQ[LIV]Q. In addition, the C- terminall domain of the members of the PIR family is highly homologous and contains a conservedd four-cysteine motif (Klis et al, 2002). Except for the presence of a Kex2-site in thee putative gene product of YJL160c, this gene also appears to encode a protein with an N- terminall signal sequence and 4 PIR-repeats (Apweiler et al, 2001), and thus may represent a fifthfifth member of the PIR-family.

14 4 1,6-fS-glucan1,6-fS-glucan synthesis in Saccharomyces cerevisiae

Interestingly,, PIR-repeats can also be found in the GPI-CWPs Cwpl, Cwp2, Tirl, and Tir2,, where they seem to be present as a single "repeat". It has been shown that at low pH Cwpll is both linked via its GPI-remnant to a 1,6-yS-glucan - l,3-/?-glucan heteropolymer and viaa an alkali-sensitive linkage directly to l,3-/?-glucan. In addition, at low pH, Pir2 was incorporatedd in the cell wall with higher efficiency. This suggests that the incorporation mechanismm for Pir-CWPs is functioning more efficiently at low pH, and therefore the alkali- sensitivee linkage of Cwpl to the cell wall might occur more frequently under those conditions (Kapteynn et al, 2001). It is tempting to speculate that the PIR-repeat is necessary and sufficientt for this linkage, which could also imply multiple linkages for proteins with multiple PIR-repeats.. Furthermore, it would be interesting to study the cell wall linkages of Cwp2, Tirl,, and Tir2 at low pH.

(( GPI-CWP J ( GPI-CWP J © \.. J V. J 1.6-[i-(]lucan 1,3-pi-glucann 1,3-p-glucan 1,3-p-glucan 1,6-p-glucan 1,3-p-glucan 1,3-|3-glucan

chitin n

Figuree 2. Overview of existing CWP-polysaccharide interconnections in S. cerevisiae. For details, see text.

3.3. Chitin 3.1.3.1. Chitin in the cell wall Chitin,, which consists of 1,4-/?-linked ./V-acetylglucosamine (GlcNAc) residues, is predominantlyy present in the septal region of vegetatively growing cells of S. cerevisiae (Orlean,, 1997; Cabib et al, 2001). It is routinely visualized with the fluorescent compound calcofluorr white. There is also some chitin dispersed throughout the lateral wall, but this is a minorr part compared to the septal region. However, under stress conditions this can be considerablyy increased, and can account for up to 20% of the cell wall dry weight (Popolo et al.,al., 1997; Dallies et al, 1998; Kapteyn et al, 1998; Ram et al, 1998). The increase in chitin levelss in the lateral wall upon activation of the cell wall salvage mechanism is accomplished byy Chitin Synthase 3 (CSIII), one of the three chitin synthases in S. cerevisiae (Valdivieso et al,al, 2000).

15 5 ChapterChapter 1

3.2.3.2. The three chit in synthases Threee chitin synthase activities have been identified in S. cerevisiae, named CSI, CSII,, and CSIII. The catalytic part of these synthases is supposed to be encoded by CHS1, CHS2,CHS2, and CHS3, respectively (Cabib et al, 2001). All three synthases share the QRRRW motif,, which is essential for activity (Cos et al, 1998). Chitin synthases are processive enzymes,, which catalyze the transfer of GlcNAc from UDP-GlcNAc to a growing chitin chain.. In addition, a number of genes {i.e. CHS4 - CHS7) have been identified whose gene productss regulate chitin synthesis (Cabib et al, 2001). Althoughh sharing a common biochemical function, the three chitin synthases play a distinctt physiological function. CSIII is responsible for the synthesis of the chitin ring at the basee of emerging buds and the chitin in the lateral wall, which together represent the bulk of thee chitin in S. cerevisiae (Shaw et al., 1991; Bulawa, 1992). The synthesis of this chitin ring att the bud neck is the first step in the synthesis of the septum, a cross-wall separating the motherr and daughter cell. The second step encompasses the synthesis of the primary septum, aa chitin disc perpendicular to the mother-daughter axis, which is accomplished by CSII (Silvermann et al, 1988; Shaw et al, 1991). The third and last step in septum formation comprisess the deposition of secondary septa on both mother and daughter sides of the primaryy septum. Cell separation is finally completed by means of a chitinase that partially hydrolyzess the primary septum (Kuranda and Robbins, 1991). During the separation, CSI seemss to function as a repair mechanism to protect cell wall integrity from excessive chitinasee activity (Cabib et al, 1989; Cabib et al, 1992). 3.3.3.3. Zymogenic nature of chitin synthases Thee activity of all three chitin synthases is significantly increased after partial proteolysiss by trypsin. CSI and CSII are predominantly found in an inactive state. This state hass been designated as zymogenic (Cabib et al, 1996). In contrast, CSIII can be isolated in ann active state, but in chs4 mutants or after extractions with detergents, CSIII also appeared too have a zymogenic state (Choi et al, 1994a; Trilla et al, 1997). For Chsl and Chs3 activity, posttranslationall regulation seems to be the major mechanism during vegetative growth in richrich medium. Chs2 also appears to be regulated on the posttranslational level, yet Chs2 levels oscillatee during the cell cycle whereas Chsl and Chs3 levels remain constant (Choi et al, 1994b;; Chuang and Schekman, 1996). 3.4.3.4. Localization of chitin synthases Thee chitin synthases must be located in the plasma membrane to perform their biologicall activities. Best documented is the migration of Chs3, which requires Chs7 to exit thee endoplasmic reticulum (ER) (Trilla et al, 1999). In the Golgi, Chs3 is transferred to so- calledd chitosomes, a process for which Chs5 and Chs6 are necessary (Santos and Snyder, 1997;; Ziman et al, 1998). Finally, the chitosomes transport Chs3 to the plasma membrane (Chuangg and Schekman, 1996; Ziman et al, 1998). The translocation of Chsl and Chs2 to

16 6 1,6-fï-glucan1,6-fï-glucan synthesis in Saccharomyces cerevisiae

thee plasma membrane is poorly documented. The bulk of these proteins is located at the plasmaa membrane (Cabib et ai, 1996), but also a small amount can be detected in chitosomes (Chuangg and Schekman, 1996; Ziman et ai, 1996).

4.4. 1,3-p-Glucan 4.1.4.1. 1,3-fi-Glucan in the cell wall 1,3-/?-Glucann represents the major class of/?-glucans in the cell wall of S. cerevisiae, andd is responsible for the mechanical strength of the cell wall. The elasticity is exemplified byy the observation that cells exposed to hypertonic pressure show dramatic shrinkage, a processs that is reversed when they are placed back in an isotonic environment (Morris et ai, 1986).. This suggests that the cell wall is in an extended state under normal growth conditions, whichh is supported by the observation that isolated walls are far less permeable to macromoleculess than the stretched walls of living cells (Scherrer et ai, 1974; De Nobel et ai, 1990a;; De Nobel et ai, 1990b; De Nobel et ai, 1991). The elastic nature of the cell wall can bee explained by the 3-D structure of l,3-/?-glucan chains, which can assume the shape of a springg (Klis et ai, 2002). Of the carbohydrate polymers in the yeast cell wall, l,3-/?-glucan hass the highest degree of polymerization with an average of 1,500 glucose residues per chain inn stationary phase cells (Fleet, 1991). 4.2.4.2. Enzymatic properties of the 1,3-fi-glucan synthase Thee development of an in vitro assay for l,3-/?-glucan synthesis has been of great valuee to cell wall research. After important groundwork (Sentandreu et ai, 1975; Balint et ai,ai, 1976; Lopez-Romero and Ruiz-Herrera, 1977), an efficient assay for in vitro biosynthesis off this polymer was devised (Shematek et ai, 1980). The activity is associated with the plasmaa membrane, and requires the presence of GTP or ATP (Shematek et ai, 1980; Shematekk and Cabib, 1980). UDP-glucose is used as sugar donor for this assay, and the in vitrovitro reaction product is a linear l,3-/?-glucan of 60 - 80 sugar residues long. This polysaccharidee is synthesized de novo, since the terminal reducing group was found to containn 14C from the UDP-[,4C]glucose (Shematek et ai, 1980). However, l,3-/?-glucan occurss in the cell wall as a large linear polymer, branched with some 1,6-/9 residues (Fleet, 1991).. It is unlikely that in vivo l,3-/?-glucan is directly synthesized into such a large polymer,, and there are candidate genes for extracellular modification (Popolo and Vai, 1999). 4.3.4.3. GTP is required for l,3-fi-glucan synthase Thee activation of l,3-/?-glucan synthesis by nucleotide triphosphates, in particular GTP,, suggested that these nucleotides are involved in a regulatory mechanism. Evidence camee from reconstitution experiments, in which l,3-/?-glucan synthase was dissociated by detergentt and NaCl into a soluble- and a membrane fraction. Either fraction was practically inactive,, while activity was reconstituted by mixing both fractions. The catalytic activity appearedd to reside in the membrane fraction, whereas the soluble fraction contained a

17 7 ChapterChapter 1

putativee regulatory GTP-binding component (Kang and Cabib, 1986). A small GTP-binding proteinn was purified from the soluble fraction, which no longer required the addition of GTP forr activity. Its ability to stimulate l,3-/?-glucan synthesis depended on the association with GTP,, as hydrolysis of the bound GTP to GDP lead to the recurring need for GTP addition (Mole/a/.,, 1994). 4.4.4.4. Rhol is the GTP-binding protein Inn search for the identity of the small GTP-binding protein that regulates l,3-/?-glucan synthase,, the known small GTP-binding proteins were explored. Conditional mutations in RHOlRHOl - RH04 result in the lysis of cells with a small bud (Madaule et al, 1987; Matsui and Toh-e,, 1992; Yamochi et al, 1994), which is a stage in the cell cycle that clearly requires 1,3- /?-glucann synthesis. Thus, these genes were candidates for the regulatory G-protein involved inn l,3-/?-glucan synthesis. Rhol turned out to be the GTP-binding regulatory component of 1,3-yS-glucann synthase, as extracts of rhol mutants were defective in l,3-/?-glucan synthase activity.. Also, the addition of recombinant Rho 1 could reconstitute the l,3-/?-glucan synthase activityy of the membrane fraction of rhol mutants (Drgonova et al, 1996; Qadota et al, 1996).. Furthermore, Rhol copurified with the l,3-/7-glucan synthase complex, and co- precipitatedd when Fksl - another component of the l,3-/?-glucan synthase - was "pulled down"" with antibodies (Qadota et al, 1996; Mazur et al, 1996). 4.5.4.5. Fksl and Gsc2 are essential components ofl,3-(5-glucan synthase FKS1FKS1 was first identified in a screen for mutants hypersensitive for the immunosuppressantss FK506 and Cyclosporin A (Parent et al, 1993). Also in other studies thiss gene was discovered, which, when mutated caused resistance to l,3-/7-glucan synthase inhibitorss ietgl; Douglas et al, 1994a), or displayed hypersensitivity to the cell wall perturbingg compound calcofluor white (cwh53; Ram et al, 1994). Mutants of fksl showed decreasedd in vitro l,3-/?-glucan synthase activity (Douglas et al, 1994b). The Fksl protein wass highly enriched in l,3-/?-glucan synthase purified by product entrapment (Inoue et al, 1995),, and when immunoprecipitated, the synthase activity was copurified (Mazur et al, 1996).. In S. cerevisiae, also a close homolog of FKS1 is present, named GSC2 or FKS2 (Inouee et al, 1995; Mazur et al, 1995). Since GSC2 was used in the first publication that describedd this gene, this name will be used here. According to their sequence, both FKS1 and GSC2GSC2 encode large proteins with 16 putative membrane-spanning domains (Douglas et al, 1994;; Mazur et al, 1995). Althoughh deletion of either FKS1 or GSC2 is not lethal, cells in which both genes are deletedd simultaneously are in viable (Inoue et al, 1995; Mazur et al, 1995). FKS1 and GSC2 aree considered to encode alternative subunits of the l,3-/?-glucan synthase complex, since theyy are differentially expressed. FKS1 is expressed in vegetatively growing cells, cultured in richh glucose-containing medium, whereas GSC2 is expressed in medium without glucose. Expressionn of GSC2 in glucose-containing medium is induced by calcium or mating

18 8 1,6-fS-glucan1,6-fS-glucan synthesis in Saccharomyces cerevisiae

pheromones,, in a calcineurin-dependent manner. FK506 inhibits the expression of GSC2, explainingg the hypersensitivity of 'fksl mutants for this compound (Mazur et ai, 1995). Both Fksll and Gsc2 are clearly essential components of l,3-/?-glucan synthase, and they are generallyy proposed to be the catalytic subunit of this complex. However, no unequivocal evidencee has been presented yet to confirm this. Also, these proteins lack the nucleotide- glucosee binding consensus site (K/R-X-G-G), as present in Escherichia coli glycogen synthasee (Furukawa et ah, 1993), so it cannot be excluded that there are more - yet unknown -- components of the 1,3-/?-glucan synthase.

1,3-(i-glucan n

inactive e

PM M

UDP-GIc c

(( Rho1 (GTP Rom2 2 Lrg1 1 GDI-- (( Rho1 (GDP

Figuree 3. Schematic representation of Rhol activating the l,3-/?-glucan synthase. Rhol can be converted to the GTP-boundd form, facilitated by Rom2. This active form can activate the 1,3-/?-glucan synthase. The activity of thee synthase can be down regulated by Lrgl.

19 9 ChapterChapter I

4.6.4.6. More on the regulation of 1,3-fl-glucan synthase Smalll G-proteins have two forms: (1) an inactive form that binds GDP, and (2) a GTP-boundd active form (Figure 3). Inactive GDP-bound G-proteins, upon receiving the correctt stimulation, can be converted into the GTP-bound form. This process is facilitated by so-calledd guanine nucleotide exchange factors (GEFs), and is inhibited by GDP dissociation inhibitorss (GDIs). Conversion of GTP to GDP occurs by the intrinsic GTPase activity of the G-protein,, which is promoted by GTPase activating proteins (GAPs), because the default GTPasee activity is very low (Matozaki et al, 2000). Inn S. cerevisiae the activity of Rhol is controlled by several GEFs and GAPs. With respectt to the function of Rhol in regulating 1,3-/?-glucan synthesis, the Rhol GEF encoded byy ROM2 (Ozaki et al, 1996) activates Rhol upon cell wall alterations (Bickle et al, 1998) andd stimulates l,3-/7-glucan synthase in the fksl-1154 fks2A mutant when overexpressed. Thiss mutant has a temperature-sensitive defect in l,3-/?-glucan synthase activity (Sekiya- Kawasakii et al, 2002). Furthermore, cells deleted for ROM2 show decreased incorporation of glucosee into l,3-/?-glucan, a phenomenon which is also seen in cells deleted for WSC1 (also knownn as SLG1/HCS77). The WSC family of genes is thought to encode membrane-located sensorss to control cell integrity (Gray et al, 1997; Verna et al, 1997; Jacoby et al, 1998). Overexpressionn of WSC1 or WSC3 was found to stimulate 1,3-/?-glucan synthase in the fksl- 11541154 fks2A mutant (Sekiya-Kawasaki et al, 2002). The Wscl protein is known to directly interactt with Rom2 to convert Rhol to the active GTP-bound form (Philip and Levin, 2001), soo this signal cascade seems to provide a mechanism for regulation of l,3-/?-glucan synthesis. Thee GAP Lrgl is a negative regulator of Rhol activity in l,3-/?-glucan synthesis. When deleted,, l,3-/7-glucan synthesis is restored in the fksl-1154 fks2A mutant. This is also seen in aa rhol-2 mutant in which LRG1 has been deleted, indicating an inhibitory function for Lrgl inn 1,3-/?-glucan synthase activity (Watanabe et al, 2001). Altogether, based on the currently availablee knowledge, 1,3-/?-glucan synthase activity seems to be tightly regulated. 4.7.4.7. Localization of 1,3-(3-glucan synthase Ass a complex that manufactures a major cell wall component, one would expect the 1,3-/?-glucann synthase to localize to sites of growth. Indeed, the synthase is present at the plasmaa membrane (Shematek et al., 1980). For Rhol to activate the l,3-/?-glucan synthase, it mustt also localize to the plasma membrane. Small G-proteins are known to be prenylated, i.e. theyy are modified with a lipid moiety that promotes binding to the membrane (Zhang and Casey,, 1996). Rhol is also known to be prenylated, and this modification is critical for its capabilityy to activate 1,3-/?-glucan synthase (Inoue et al, 1999). Both Rhol and Fksl localize too the bud site at bud emergence, and to the bud tip of small buds. Subsequently, Rhol and Fksll disperse throughout the bud during isotropic growth of the bud, in order to finally repolarizee to the bud-neck region before cell separation (Yamochi et al, 1994; Qadota et al,

20 0 /,, 6-fi-glucan synthesis in Saccharomyces cerevisiae

1996).. In conclusion, the dispersion of the Rhol - Fksl complex during the cell cycle coincidess perfectly with the occurrence of cell wall synthesis in the growing bud.

5.. 1,6-p-Glucan 5.1.5.1. 1,6-ft-Glucan in the cell wall 1,6-/?-Glucann isolated from S. cerevisiae cell walls is a highly branched, water-soluble polymer,, which is reported to consist on average of 130 - 350 glucose residues (Manners et al,al, 1973; Kollar et al, 1997). It only accounts for 5 - 10 % of the cell wall dry weight (Mannerss et al, 1973), yet it plays a crucial role in the cell wall of S. cerevisiae and C. albicans.albicans. Mutations in genes involved in 1,6-/?-glucan synthesis often result in cells with severelyy decreased viability (Meaden et al., 1990; Roemer et al., 1993; Brown and Bussey, 1993;; Mio et al., 1997; Lussier et al, 1998; Sullivan et al, 1998; Vink et al., 2002, Chapter 2).2). At the molecular level, 1,6-/?-glucan functions as a linker between cell wall proteins and 1,3-yS-glucann (Kapteyn et al, 1996; Kollar et al, 1997). Under stress conditions, it might also bee connected directly to chitin (Kollar et al, 1997; Kapteyn et al, 1997). 5.2.5.2. Identification of genes involved in 1,6-fi-glucan synthesis Geneticc screens have been most valuable in identifying genes involved in 1,6-/?- glucann synthesis. One approach exploited the cell wall perturbing compound calcofluor white too identify genes involved in cell wall biogenesis. Mutants with increased sensitivity to this compoundd were expected to have a weakened cell wall, and the corresponding mutations mightt therefore identify genes involved in cell wall synthesis. This approach resulted in the identificationn of a large number of genes involved in many aspects of cell wall synthesis (Ramm et al, 1994; Lussier et al, 1997). A method for identifying genes more specifically involvedd in 1,6-/?-glucan synthesis was developed by Bussey and co-workers, taking advantagee of the fact that mutants with reduced levels of 1,6-/?-glucan display a resistance to Kll killer toxin, which is a pore-forming protein that binds 1,6-/?-glucan and kills cells (Al- Aidrooss and Bussey, 1978; Boone et al, 1990; Bussey, 1991; Brown et al, 1993a; Page et al,al, 2003). These screens have resulted in a large number of genes either involved in cell wall biogenesiss in general, or, in 1,6-/?-glucan biogenesis in particular. These genes are named KREKRE (for Killer REsistant), CWH (for Calcofluor White Hypersensitive), or ECM (for ExtraCellularr Matrix). Sincee the synthesis of both chitin and l,3-/?-glucan takes place at the plasma membrane,, it came as a surprise that the gene products involved in l,6-/?-glucan synthesis, identifiedd in the genetic screens mentioned above, were localized along the secretory pathway.. Although this might suggest that l,6-/?-glucan is synthesized throughout the secretoryy pathway, there are at least two other explanations for this observation: l,6-/?-glucan synthesiss may take place at the plasma membrane, but it requires some intracellular key

21 1 ChapterChapter 1

eventss such as the synthesis of a primer, or, alternatively, the synthesis takes place at the plasmaa membrane, but this process is very sensitive to certain intracellular defects. 5.3.5.3. Endoplasmic Reticulum (ER) Thee KRE5 gene encodes a large, soluble secretory glycoprotein with a carboxy- terminall HDEL sequence (Meaden et al, 1990), which is an ER-retention signal for soluble proteins.. Deletion of KRE5 results in extremely compromised growth, and an aberrant cell walll (Meaden et al., 1990) which is entirely void of the electron dense outer layer (Simons et al,al, 1998; Levinson et al, 2002). Furthermore, kre5 mutants have a substantial defect in the incorporationn of the GPI-CWP a-agglutinin (Lu et al, 1995), which might be a consequence off their lack of 1,6-/?-glucan (Meaden et al, 1990). Also, the percentage of Pir-CWPs which aree linked directly to l,3-/?-glucan is dramatically increased in the kre5A mutant. The transcriptt levels of PIR2 and other Pir-family members are upregulated, which probably is the resultt of a compensatory mechanism to ensure cell wall integrity under compromised conditionss (Kapteyn et al, 1999). The Kre5 protein has weak, though significant similarity to UDP-glucose:glycoproteinn glucosyltransferase (GT) of Drosophila (Parker et al, 1995) and S.S. pombe (Fernandez et al, 1996). GT is part of the ER quality control mechanism for protein folding.. Glycoproteins that are ^-glycosylated receive a core oligosaccharide

(Glc3Man9GlcNAc2)) in the ER, of which the glucose residues subsequently are removed by glucosidasee I and II. However, if a protein is not yet properly folded, GT attaches one glucose too the core which allows recognition of the protein by calnexin/calreticulin. This facilitates correctt folding of the protein. The glucose can be removed again by glucosidase II, and the cyclee may repeat itself until the protein is folded correctly (Parodi, 1999). However, there is evidencee speaking against the idea that Kre5 has a GT-like function: (1) there is no detectable GTT activity in S. cerevisiae (Fernandez et al, 1994; Jakob et al, 1998), and (2) the essential functionn of Kre5 is not that of a GT, since KRE5 is essential in a mutant in which the requirementt of a GT is bypassed (Shahinian et al, 1998). Although this can hardly be seen as conclusivee evidence, an alternative possibility is that the function of Kre5 is related to that of aa GT, but that it has diverged and is now involved in 1,6-/?-glucan synthesis directly or indirectly.. Nonetheless, it has recently been found that the C. albicans Kre5 homolog (CaKre5)) can partially restore the kre5 mutant. Although it has not been tested for GT activity,, it has a higher similarity to GT than Kre5 (Levinson et al, 2002). Alsoo other mutations in genes encoding components of the quality control mechanism weree found to affect l,6-y?-glucan. These include CWH41/GLS1, ROT2/GLS2, and CNE1, whichh encode glucosidase I (Romero et al, 1997), glucosidase II (Trombetta et al, 1996), andd the yeast homolog of mammalian calnexin/calreticulin (De Virgilio et al, 1993), respectively.. The cwh41 mutant has lowered levels of l,6-/?-glucan, but this is not accompaniedd by a growth defect (Jiang et al., 1996; Shahinian et al, 1998). Also, the rotl mutantt shows a decrease in 1,6-/?-glucan levels, and these cells display killer resistance

22 2 1,6-fi-glucan1,6-fi-glucan synthesis in Saccharomyces cerevisiae

(Shahiniann et al, 1998) just like the cwh41 mutant (Jiang et al, 1996; Shahinian etal, 1998). Thee CWH41 gene displays strong genetic interaction with KRE6 and KRE1, two genes that aree involved in the biogenesis of l,6-/?-glucan (discussed below), as double mutants are inviablee or show exacerbated phenotypes, respectively (Jiang et al, 1996). Interestingly, the strongg genetic interactions with both genes disappeared upon deletion of ALG5 - encoding the dolichol-P-glucosee synthase - which results in unglucosylated core TV-chains. As this bypasses thee requirement for glucosidase I activity, this indicates that indeed this function of Cwh41 is essentiall in the kre6 mutant. It was found that Kre6 is unstable in the absence of CWH41, suggestingg that the effect of Cwh41 on l,6-/?-glucan synthesis is indirect (Abeijon and Chen, 1998). . Cellss deleted for CNE1 show a moderate defect in l,6-/?-glucan synthesis (Shahinian etet al, 1998), although this conclusion is contradicted by a study of Simons and co-workers, whoo saw no effect (1998). However, the outer electron dense layer of the cnelA mutant was reducedd and the cell wall surface had an irregular shape (Simons et al, 1998). Cnel appears too function in the quality control mechanism, as its deletion results in a marked protein foldingg defect, yet its function is not essential in S. cerevisiae (Parlati et al, 1995). Deletion off CNE1 enhanced the l,6-/?-glucan defects of kreóA, cwh41A, and rot2A (Shahinian et al, 1998).. Another component of the quality control mechanism, BiP/Kar2, also exacerbated the l,6-/?-glucann defect of cwh41A and rot2A when deleted (Simons et al, 1998). This further demonstratess that defects in the quality control mechanism may result in general cell wall defects,, amongst others by indirectly affecting l,6-/?-glucan. 5.4.5.4. Golgi Thee KRE6 gene encodes a type II transmembrane glycoprotein located in the Golgi (Boonee et al, 1990; Roemer and Bussey, 1991; Roemer et al, 1994). Deletion of this gene resultss in a growth defect, a significant reduction of 1,6-/?-glucan content, and cell walls with ann aberrant structure (Roemer and Bussey, 1991; Roemer et al, 1994). Furthermore, these cellss are defective in the incorporation of a-agglutinin (Lu et al, 1995) and Cwpl into the celll wall, yet they incorporate more Pir2 protein (Kapteyn et al, 1999). SKN1 was identified ass a multi-copy suppressor of the kre6A mutant, and encodes a homolog of KRE6. Deletion of SKN1SKN1 did not result in a noticeable phenotype, but a kreóAsknlA double mutant is hardly viablee and displays a dramatic defect in 1,6-/?-glucan synthesis (Roemer et al, 1993). Hydrophobicc cluster analysis showed that Kre6 and Sknl have similarities to family 166 glycosylhydrolases, indicating that they function as glycosylases or transglycosylases. Thiss seems to exclude a function as nucleotide sugar glucosyltransferase (Montijn et al, 1999,, Chapter 3). The cytoplasmic tail of Kre6 can be phosphorylated, which might be a potentiall mode of regulation. The kre6A mutant is synthetic lethal with several components off the cell wall integrity Pkc 1 MAP kinase cascade, and overexpression of the KRE6 gene

23 3 ChapterChapter 1

cann suppress the lysis defect of the/?A;cVA mutant (Roemer et al, 1994). This implicates a rolee for PKC1 in whatever function Kre6/Skn 1 may have. Onee of the genes identified in the screen for Kl killer toxin resistant mutants, KRE2/MNT1,KRE2/MNT1, appeared to encode a Golgi mannosyltransferase involved in O- and jV-linked glycosylationn (Hill et al, 1992; Hausler etai, 1992). It would be interesting to see what the underlyingg mechanism is of the KRE phenotype of the kre2A mutant. One possibility might bee that in this mutant Kiel (discussed below) is not properly O-glycosylated, but there may bee more proteins suffering from this defect. 5.5.5.5. Cytoplasm Alsoo KRE 11 was identified in the screen for Kl toxin resistant genes. It encodes a putativee cytoplasmic protein. Deletion of this gene results in lowered 1,6-/?-glucan levels (Brownn et al, 1993). Krel 1 was identified as a component of TRAPP (Sacher et al, 2000), a complexx on the ds-Golgi that functions in docking and/or fusion of ER-to-Golgi transport vesicless (Sacher et al, 1998). The synthetic lethality of krel 1A with kre6A (Brown et al, 1993)) might thus be explained by the incorrect delivery of Kre6 and Sknl to the Golgi in a krelkrel 1A background. 5.6.5.6. Plasma membrane Thee KRE1 gene is predicted to encode a heavily ^-glycosylated GPI-anchored plasma membranee protein (Boone et al, 1991; Roemer and Bussey, 1995), and it was confirmed by immunofluorescencee that Krel indeed localizes to the cell surface (Roemer and Bussey, 1995).. Deletants for krel A show an aberrant morphology of the cell wall, demonstrated by a dispersee electron dense outer layer. These cells display a decrease in 1,6-/?-glucan levels, mostlyy due to much shorter chains of 1,6-/?-glucan (Boone et al, 1991). Recently, it was foundd that Krel is the receptor for the Kl killer toxin. The Kl toxin directly binds to the Krel proteinn and probably allows the subsequent and lethal formation of the ion channel. It was proposedd that the other receptor for the Kl toxin, 1,6-/?-glucan, functions to concentrate the toxinn (Breinig et al, 2002). 5.5.7.7. Extracellular space KRE9KRE9 encodes a 276-amino acid secretory protein, with a predicted molecular weight off 30 kDa. It has many potential O-glycosylation sites, many of which presumably are occupiedd since the protein has an apparent molecular weight of about 55-60 kDa on SDS- PAGEE (Brown and Bussey, 1993). This might be another important target of the Kre2 mannosyltransferase.. The kre9A deletants display severely reduced l,6-/?-glucan levels, and thee electron dense layer of their cell wall appears to be absent (Brown et al, 1991; Brown andd Bussey, 1993). In addition, these mutants fail to form a mating projection upon a-factor treatmentt and they are defective in mating (Brown and Bussey, 1993). The KNH1 gene is a functionall homolog of KRE9, which when overexpressed can suppress the kre9A mutation. Deletionn of KNH1 does not result in a clear phenotype, but in a kre9A background this

24 4 1,6-P'glucan1,6-P'glucan synthesis in Saccharomyces cerevisiae

deletionn is lethal. Overexpression of KNH1 is able to suppress the kre9A mutant. The expressionn of KNH1 is upregulated when the cells are grown on galactose, explaining the partiall suppression of the kre9A phenotypes by growth on galactose. 5.8.5.8. Localization of 1,6-p-glucan synthesis Basedd on the above described genes - identified on the basis of their resistance to Kl killerr toxin - it was thought for a long time that the synthesis of 1,6-/ï-glucan was a process thatt started intracellularly, and that the resulting polymer was extended while traveling throughh the secretory pathway. However, this was all based on genetic evidence, and no biochemicall data is available to support this idea. Using immunogold labeling, Montijn and co-workerss (1999, Chapter 3) showed that there was no detectable intracellular 1,6-/?-glucan inn a secl-1 mutant grown at the restrictive temperature. This mutant accumulates post-Golgi secretoryy vesicles, and after 2 hours of growth at the restrictive temperature one would expect att least some 1,6-/?-glucan to accumulate intracellularly if the hypothesis above was correct. Microsomess isolated from these cells were separated by gel filtration and analyzed for the presencee of 1,6-/?-glucan in different fractions. The l,6-/7-glucan co-eluted with the plasma membranee fraction, supporting the idea that the bulk of l,6-/?-glucan synthesis occurs at the plasmaa membrane (Montijn et al., 1999, Chapter 3). 5.9.5.9. Regulation of 1,6-fi-glucan Littlee is known about the regulation of 1,6-/?-glucan synthesis. The original screens forr Kl killer toxin resistant mutants did not result in the isolation of genes with potential regulatoryy functions. In contrast, a more general screen for genes involved in cell wall biogenesiss - using calcofluor white - did yield potential regulatory genes with defects in 1,6- /?-glucann synthesis (Ram et al, 1994; Jiang et al, 1995; Vink et al, 2002, Chapter 2). The PTC1/CWH47PTC1/CWH47 gene encodes a protein phosphatase type 2C, and mutations in this gene resultedd in resistance to the Kl killer toxin. Ptcl appeared to have a negative regulatory functionn on the Pbs2-Hogl pathway (Jiang et al, 1995), which later proved to be a direct dephosphorylationn of Hog 1 (Warmka et al, 2001). Accordingly, overexpression of the kinase off Hogl, PBS2, also results in Kl toxin resistance (Jiang et al, 1995), whereas the pbs2A mutantt is hypersensitive (Jiang et al, 1995; Vink et al, 2002, Chapter 2). This clearly indicatess that 1,6-/?-glucan synthesis is inhibited by the Pbs2-Hogl MAP kinase pathway whichh is otherwise known to control the response to high osmolarity (reviewed in Hohmann, 2002).. The mechanism of how the Pbs2-Hogl pathway affects l,6-/?-glucan synthesis is not clear. . KIC1KIC1 encodes a protein kinase that was identified as a two-hybrid interactor with Cdc31,, revealing a novel function for this yeast centrin homolog in the control of cell wall integrityy (Sullivan et al, 1998). A mutation in this gene was identified in a screen for calcofluorr white hypersensitivity (Ram et al, 1994; Vink et al, 2002, Chapter 2). Recently, Vinkk and co-workers (2002, Chapter 2) found that kicl mutant cells displayed Kl killer toxin

25 5 ChapterChapter I

resistance,, which was accompanied by a defect in 1,6-/?-glucan. Reciprocally, overexpression off K1C1 resulted in hypersensitivity to the Kl toxin and a slight increase in 1,6-^-glucan levels.. The Kiel gene was therefore proposed to participate in the regulation of 1,6-/?-glucan synthesis.. Interestingly, a multicopy suppressor of the kicl calcofluor white hypersensitivity wass isolated that shared some of the kicl phenotypes. This was the RH03 gene, that, when deleted,, also resulted in a higher degree of resistance to the Kl toxin, concomitant with reducedd l,6-/?-glucan levels. Overexpression of RH03 resulted in Kl toxin hypersensitivity (Vinkk et al, 2002, Chapter 2). The RH03 gene encodes a Rho-type small G-protein that also iss involved in bud growth, in the actin cytoskeleton, and in exocytosis (Matsui and Toh-E, 1992b;; Imai et al, 1996; Robinson et al, 1999; Adamo et al, 1999). In the absence of RHÖ3,RHÖ3, overexpression of RHÖ4 can suppress the mutant, and the rho3Arho4A double mutant iss inviable. This suggests that RH04 is functionally related to RH03 (Matsui and Toh-E, 1992a).. However, RH04 does not share the regulating function of RHÖ3 in cell wall synthesiss (Vink et al, 2002, Chapter 2). It is tempting to speculate that the function of Rho3 inn 1,6-/?-glucan synthesis is analogous to that of Rhol in l,3-/?-glucan synthesis, i.e. in the directt regulation of the 1,6-/?-glucan synthase itself. However, evidence for this is lacking sincee l,6-/?-glucan synthase has not yet been identified. 5.10.5.10. Biochemical assay for 1,6-B-glucan Ass yet, no genes have been identified as potential candidates encoding for proteins thatt are (part of) the putative plasma membrane 1,6-/?-glucan synthase complex. How can this importantt enzyme have been overlooked, whereas (at least some of the) components of both chitin-- and 1,3-/J-ghican synthase have been isolated? One major help in the identification of chitin-- and l,3-/?-glucan synthase components has been the availability of biochemical assays too measure these activities (Shematek et al, 1980; Kang et al, 1984). In this thesis a method iss described for measuring the activity of 1,6-/?-glucan synthesis, making use of antibodies for thee detection of 1,6-/?-glucan rather than expecting it to behave as l,3-/?-glucan {Chapter 4). Thiss might be key in the development of an assay for l,6-/?-glucan synthesis activity, since thee reaction product appeared to be water-soluble in contrast to the products of chitin- and 1,3-/?-glucann assays. As this method uses a membrane extract as starting material, this implicatess that there is at least some activity associated with membranes. Further work is neededd to optimize this approach, but it is expected to become an important tool to identify componentss of the 1,6-/?-glucan synthase complex.

6.6. References

Adamoo J.E., Rossi G., and Brennwald P. (1999). The Rho GTPase Rho3 has a direct role inn exocytosis that is distinct from its role in actin polarity. Mol Biol Cell 10: 4121-4133

26 6 /,, 6-p-glucan synthesis in Saccharomyces cerevisiae

Al-Aidrooss K., and Bussey H. (1978). Chromosomal mutants of Saccharomyces cerevisiae affectingg the cell wall binding site for killer factor. Can. J. Microbiol. 24: 228-37

Apweilerr R., Attwood T.K., Bairoch A., Bateman A., Birney E., Biswas M., Bucher P., Ceruttii L., Corpet F., Croning M.D., Durbin R., Falquet L., Fleischmann W., Gouzy J., Hermjakobb H., Hulo N., Jonassen I., Kahn D., Kanapin A., Karavidopoulou Y., Lopez R.,, Marx B., Mulder N.J., Oinn T.M., Pagni M., Servant F., Sigrist C.J., and Zdobnov E.M.. (2001). The InterPro database, an integrated documentation resource for protein families,, domains and functional sites. Nucleic Acids Res. 29: 37-40

Balintt S., Farkas V., and Bauer S. (1976). Biosynthesis of beta-glucans catalyzed by a particulatee enzyme preparation from yeast. FEBS Lett. 64: 44-47

Bicklee M., Delley P.A., Schmidt A., and Hall M.N. (1998). Cell wall integrity modulates RHOlRHOl activity via the exchange factor ROM2. EMBOJ. 17: 2235-2245

Boonee C, Sommer S.S., Hensel A., and Bussey H. (1990). Yeast KRE genes provide evidencee for a pathway of cell wall beta-glucan assembly. J. Cell Biol. 110: 1833-1843

Breinigg F., Tipper D.J., and Schmitt M.J. (2002). Krelp, the plasma membrane receptor forr the yeast Kl viral toxin. Cell 108: 395-405

Brownn J.L., Kossaczka Z., Jiang B., and Bussey H. (1993). A mutational analysis of killer toxinn resistance in Saccharomyces cerevisiae identifies new genes involved in cell wall (1 — >6)-beta-glucann synthesis. Genetics 133: 837-49

Brownn J.L., and Bussey H. (1993). The yeast KRE9 gene encodes an O-glycoprotein involvedd in cell surface beta-glucan assembly. Mol. Cell Biol. 13: 6346-6356

Bulawaa C.E. (1992). CSD2, CSD3, and CSD4, genes required for chitin synthesis in SaccharomycesSaccharomyces cerevisiae: the CSD2 gene product is related to chitin synthases and to developmentallyy regulated proteins in Rhizobium species and Xenopus laevis. Mol. Cell Biol. 12:: 1764-1776

Cabibb E., Silverman S.J., and Shaw J.A. (1992). Chitinase and chitin synthase 1: counterbalancingg activities in cell separation of Saccharomyces cerevisiae. J. Gen. Microbiol. 138:: 97-102

27 7 ChapterChapter 1

Cabibb E., Sburlati A., Bowers B., and Silverman S.J. (1989). Chitin synthase 1, an auxiliaryy enzyme for chitin synthesis in Saccharomyces cerevisiae. J. Cell Biol. 108: 1665- 1672 2

Cabibb E., Shaw J. A., Mol P. C, Bowers B., and Choi, W.-J. (1996). In: The Mycota (Bramblee R., and Marzluf G. A., Eds.) Vol. Ill, pp. 243-267, Springer-Verlag, Berlin

Cabibb E., Roh D.H., Schmidt M., Crotti L.B., and Varma A. (2001). The yeast cell wall andd septum as paradigms of cell growth and morphogenesis. J. Biol. Chem. 276: 19679- 19682 2

Caroo L.H.P., Tertelin H., Vossen J.H., Ram A.F.J., van den Ende H., and Klis F.M. (1997).. In silicio identification of glycosyl-phosphatidylinositol-anchored plasma-membrane andd cell wall proteins of Saccharomyces cerevisiae. Yeast 13: 1477-1489

Choii W.J., Sburlati A., and Cabib E. (1994a). Chitin synthase 3 from yeast has zymogenic propertiess that depend on both the CAL1 and the CAL3 genes. Proc. Natl. Acad. Sci U.S.A. 91:4727-4730 0

Choii W.J., Santos B., Duran A., and Cabib E. (1994b). Are yeast chitin synthases regulatedd at the transcriptional or the posttranslational level? Mol. Cell Biol 14: 7685-7694

Chuangg J.S., and Schekman R.W. (1996). Differential trafficking and timed localization of twoo chitin synthase proteins, Chs2p and Chs3p. J. Cell Biol. 135: 597-610

Coss T., Ford R.A., Trilla J.A., Duran A., Cabib E., and Roncero C. (1998). Molecular analysiss of Chs3p participation in chitin synthase III activity. Eur. J. Biochem. 256: 419-426

Dalliess N., Francois J., and Paquet V. (1998). A new method for quantitative determination off polysaccharides in the yeast cell wall. Application to the cell wall defective mutants of SaccharomycesSaccharomyces cerevisiae. Yeast 14:1297-1306

Dee Groot P.W.J., Hellingwerf K.J., and Klis F.M. (2003). Genome-wide identification of fungall GPI-proteins. submitted submitted

Dee Nobel J.G., Klis F.M., Munnik T., Priem J., and van den Ende H. (1990a). An assay off relative cell wall porosity in Saccharomyces cerevisiae, Kluyveromyces lactis and SchizosaccharomycesSchizosaccharomyces pombe. Yeast 6: 483-490

28 8 1,6-fi-glucan1,6-fi-glucan synthesis in Saccharomyces cerevisiae

Dee Nobel J.G., Klis F.M., Priem J, Munnik T, and van den Ende H. (1990b). The glucanase-solublee mannoproteins limit cell wall porosity in Saccharomyces cerevisiae. Yeast 6:491-499 9

Dee Nobel J.G., and Barnett J.A. (1991). Passage of molecules through yeast cell walls: a brieff essay-review. Yeast7: 313-323

Dee Virgilio C, Burckert N., Neuhaus J.M., Boiler T., and Wiemken A. (1993). CNE1, a SaccharomycesSaccharomyces cerevisiae homologue of the genes encoding mammalian calnexin and calreticulin.. Yeast 9: 185-188

Douglass CM., Marrinan J.A., Li W., and Kurtz M.B. (1994a). Saccharomyces cerevisiae mutantt with echinocandin-resistant 1,3-beta-D-glucan synthase. J. Bacteriol. 176: 5686-5696

Douglass C.M., Foor F., Marrinan J.A., Morin N., Nielsen J.B., Dahl A.M., Mazur P., Baginskyy W., Li W., el-Sherbeini M., Clemas J.A., Mandala, S.M., Frommer B.R., and Kurtzz M.B. (1994b). The Saccharomyces cerevisiae FKS1 (ETG1) gene encodes an integral membranee protein which is a subunit of 1,3-beta-D-glucan synthase. Proc. Natl. Acad. Sci. USA9\:USA9\: 12907-12911

Drgonovaa J., Drgon T., Tanaka K., Kollar R., Chen G.C., Ford R.A., Chan C.S., Takai Y.,, and Cabib E. (1996). Rholp, a yeast protein at the interface between cell polarization andd morphogenesis. Science 111: 277-279

Fernandezz F.S., Trombetta S.E., Hellman U., and Parodi A.J. (1994). Purification to homogeneityy of UDP-glucose:glycoprotein glucosyltransferase from Schizosaccharomyces pombepombe and apparent absence of the enzyme from Saccharomyces cerevisiae. J. Biol. Chem. 269:: 30701-30706

Fernandezz F., Jannatipour M., Hellman U., Rokeach L.A., and Parodi A.J. (1996). A neww stress protein: synthesis of Schizosaccharomyces pombe UDP—Glc: glycoprotein glucosyltransferasee mRNA is induced by stress conditions but the enzyme is not essential for celll viability. EMBOJ. 15: 705-713

Fleet,, G.H. (1991). Cell walls. In: The Yeasts, 2nd edition., Vol. 4 (Rose A.H. and Harrison, J.S.,, Eds.), pp. 199-277. Academic Press, New York.

29 9 ChapterChapter 1

Fujiii T., Shimoi H., and Iimura Y. (1999). Structure of the glucan-binding sugar chain of Tiplp,, a cell wall protein of Saccharomyces cerevisiae. Biochim. Biophys. Acta 1427: 133- 144 4

Furukawaa K., Tagaya M., Tanizawa K., and Fukui T. (1993). Role of the conserved Lys- X-Gly-Glyy sequence at the ADP-glucose-binding site in Escherichia coli glycogen synthase. J.J. Biol. Chem. 268: 23837-23842

Gemmilll T.R., and Trimble R.B. (1999). Overview of N- and O-linked oligosaccharide structuress found in various yeast species. Biochim. Biophys. Acta 1426: 227-237

Grayy J.V., Ogas J.P., Kamada Y., Stone M., Levin D.E., and Herskowitz I. (1997). A rolee for the Pkcl MAP kinase pathway of Saccharomyces cerevisiae in bud emergence and identificationn of a putative upstream regulator. EMBO J. 16: 4924-4937

Hamadaa K., Terashima H., Arisawa M., and Kitada K. (1998). Amino acid sequence requirementt for efficient incorporation of glycosylphosphatidylinositol-associated proteins intoo the cell wall of Saccharomyces cerevisiae. J. Biol. Chem. 273: 26946-24953

Hamadaa K., Terashima H., Arisawa M., Yabuki N., and Kitada K. (1999). Amino acid residuess in the omega-minus region participate in cellular localization of yeast glycosylphosphatidylinositol-attachedd proteins. J. Bacteriol. 181: 3886-3889

Hauslerr A., Ballou L., Ballou C.E., and Robbins P.W. (1992). Yeast glycoprotein biosynthesis:: MNT1 encodes an alpha-1,2-mannosyltransferase involved in O-glycosylation. Proc.Proc. Natl Acad. Sci. USA 89: 6846-6850

Hilll K., Boone C, Goebl M., Puccia R., Sdicu A.M., and Bussey H. (1992). Yeast KRE2 definess a new gene family encoding probable secretory proteins, and is required for the correctt N-glycosylation of proteins. Genetics 130: 273-283

Hohmannn S. (2002). Osmotic stress signaling and osmoadaptation in yeasts.. Microbiol. Mol. Biol.Biol. Rev. 66: 300-372

Ikezawaa H. (2002). Glycosylphosphatidylinositol (GPI)-anchored proteins. Biol. Pharm. Bull.Bull. 25: 409-417

30 0 1,6-fi-gIucan1,6-fi-gIucan synthesis in Saccharomyces cerevisiae

Imaii J., Toh-e A., and Matsui Y. (1996). Genetic analysis of the Saccharomyces cerevisiae RH03RH03 gene, encoding a rho-type small GTPase, provides evidence for a role in bud formation.. Genetics 142: 359-369

Inouee S.B., Takewaki N., Takasuka T., Mio T., Adachi M., Fujii Y., Miyamoto C, Arisawaa M., Furuichi Y., and Watanabe T. (1995). Characterization and gene cloning of 1,3-beta-D-glucann synthase from Saccharomyces cerevisiae. Eur. J. Biochem. 231: 845-854

Inouee S.B., Qadota H., Arisawa M., Watanabe T., and Ohya Y. (1999). Prenylation of Rholpp is required for activation of yeast 1, 3-beta-glucan synthase. J. Biol. Chem. 274: 38119-38124 4

Jacobyy J.J., Nilius S.M., and Heinisch J.J. (1998). A screen for upstream components of thee yeast protein kinase C signal transduction pathway identifies the product of the SLG1 gene.. Mol. Gen. Genet. 258: 148-155

Jakobb C.A., Burda P., te Heesen S., Aebi M., and Roth J. (1998). Genetic tailoring of N- linkedd oligosaccharides: the role of glucose residues in glycoprotein processing of SaccharomycesSaccharomyces cerevisiae in vivo. Glycobiology 8: 155-164

Jiangg B., Ram A.F.J., Sheraton J., Klis F.M., and Bussey H. (1995). Regulation of cell walll beta-glucan assembly: PTC1 negatively affects PBS2 action in a pathway that includes modulationn of EXG1 transcription. Mol. Gen. Genet. 248: 260-269

Jiangg B., Sheraton J., Ram A.F.J., Dijkgraaf G.J., Klis F.M., and Bussey H. (1996). CWH41CWH41 encodes a novel endoplasmic reticulum membrane N-glycoprotein involved in beta 1,6-glucann assembly. J. Bacteriol. 178: 1162-1171

Jigamii Y., and Odani T. (1999). Mannosylphosphate transfer to yeast mannan. Biochim. Biophys.Biophys. Acta 1426: 335-345

Kangg M.S., Elango N., Mattia E., Au-Young J., Robbins P.W., and Cabib E. (1984). Isolationn of chitin synthetase from Saccharomyces cerevisiae. Purification of an enzyme by entrapmentt in the reaction product. J. Biol. Chem. 259: 14966-14972

Kangg M.S., and Cabib E. (1986). Regulation of fungal cell wall growth: a guanine nucleotide-binding,, proteinaceous component required for activity of (1—3)-beta-D-glucan synthase.. Proc. Natl. Acad. Sci. USA 83: 5808-5812

31 1 ChapterChapter 1

Kapteynn J.C., Montijn R.C., Vink E., de la Cruz J., Llobell A., Douwes J.E., Shimoi H., Lipkee P.N., and Klis F.M. (1996). Retention of Saccharomyces cerevisiae cell wall proteins throughh a phosphodiester-linked beta-l,3-/beta-l,6-glucan heteropolymer. Glycobiology 6: 337-345 5

Kapteynn J.C., Ram A.F.J., Groos E.M., Kollar R., Montijn R.C., Van Den Ende H., Llobelll A., Cabib E., and Klis F.M. (1997). Altered extent of cross-linking of betal,6- glucosylatedglucosylated mannoproteins to chitin in Saccharomyces cerevisiae mutants with reduced cell walll betal,3-glucan content. J. Bacteriol 179: 6279-6284

Kapteynn J.C., Van Egmond P., Sievi E., Van Den Ende H., Makarow M., and Klis F.M. (1999).. The contribution of the O-glycosylated protein Pir2p/Hspl50 to the construction of thee yeast cell wall in wild-type cells and beta 1,6-glucan-deficient mutants. Mol. Microbiol. 31:: 1835-1844

Kliss F.M., Mol P., Hellingwerf K., and Brul S. (2002). Dynamics of cell wall structure in Saccharomycess cerevisiae. FEMS Microbiol. Rev. 26: 239-256

Kollarr R., Reinhold B.B., Petrakova E., Yeh H.J., Ashwell G., Drgonova J., Kapteyn J.C.,, Klis F.M., and Cabib E. (1997). Architecture of the yeast cell wall. Beta(l~>6)-glucan interconnectss mannoprotein, beta(l-->)3-glucan, and chitin. J. Biol. Chem. 272: 17762-17775

Kurandaa M.J., and Robbins P.W. (1991). Chitinase is required for cell separation during growthh of Saccharomyces cerevisiae. J. Biol. Chem. 266: 19758-19767

Levinsonn J.N., Shahinian S., Sdicu A.M., Tessier D.C., and Bussey H. (2002). Functional, comparativee and cell biological analysis of Saccharomyces cerevisiae Kre5p. Yeast 19: 1243- 1259 9

Lopez-Romeroo E., and Ruiz-Herrera J. (1977). Biosynthesis of beta-glucans by cell-free extractss from Saccharomyces cerevisiae. Biochim. Biophys. Acta 500: 372-384

Luu C.F., Kurjan J., and Lipke P.N. (1994). A pathway for cell wall anchorage of SaccharomycesSaccharomyces cerevisiae alpha-agglutinin. Mol. Cell Biol. 14: 4825-4833

Luu C.F,, Montijn R.C., Brown J.L., Klis F.M., Kurjan J., Bussey H., and Lipke P.N. (1995).. Glycosyl phosphatidylinositol-dependent cross-linking of alpha-agglutinin and beta 1,6-glucann in the Saccharomyces cerevisiae cell wall. J. Cell Biol. 128: 333-340

32 2 1,6-p-glucan1,6-p-glucan synthesis in Saccharomyces cerevisiae

Lussierr M., White A.M., Sheraton J., di Paolo T., Treadwell J., Southard S.B., Horensteinn C.I., Chen-Weiner J., Ram A.F.J., Kapteyn J.C., Roemer T.W., Vo D.H., Bondocc D.C., Hall J., Zhong W.W., Sdicu A.M., Davies J., Klis F.M., Robbins P.W., and Busseyy H. (1997). Large scale identification of genes involved in cell surface biosynthesis andd architecture in Saccharomyces cerevisiae. Genetics 147: 435-450

Lussierr M., Sdicu A.M., Shahinian S., and Bussey H. (1998). The Candida albicans KRE9 genee is required for cell wall beta-1, 6-glucan synthesis and is essential for growth on glucose.. Proc. Natl. Acad. Sci. USA 95: 9825-9830

Madaulee P., Axel R., and Myers A.M. (1987). Characterization of two members of the rho genee family from the yeast Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 84: 779- 783 3

Mannerss D.J., Masson A.J., Patterson J.C., Bjorndal H., and Lindberg B. (1973). The structuree of a beta-(l—6)-D-glucan from yeast cell walls. Biochem. J. 135: 31-36

Matozakii T., Nakanishi H., and Takai Y. (2000). Small G-protein networks: their crosstalk andd signal cascades. Cell Signal. 12: 515-524

Matsuii Y., and Toh-e A. (1992a). Isolation and characterization of two novel ras superfamilyy genes in Saccharomyces cerevisiae. Gene 114: 43-49

Matsuii Y., and Toh-E A. (1992b). Yeast RH03 and RH04 ras superfamily genes are necessaryy for bud growth, and their defect is suppressed by a high dose of bud formation geness CDC42 and BEM1. Mol. Cell. Biol. 12: 5690-5699

Mazurr P., Morin N., Baginsky W., el-Sherbeini M., Clemas J.A., Nielsen J.B., and Foor F.. (1995). Differential expression and function of two homologous subunits of yeast 1,3-beta- D-glucann synthase. Mol. Cell Biol. 15: 5671-5681

Mazurr P., and Baginsky W. (1996). In vitro activity of 1,3-beta-D-glucan synthase requires thee GTP-binding protein Rhol. J. Biol. Chem. 271: 14604-14609

Meadenn P., Hill K., Wagner J., Slipetz D., Sommer S.S., and Bussey H. (1990). The yeast KRE5KRE5 gene encodes a probable endoplasmic reticulum protein required for (1—6)-beta-D- glucann synthesis and normal cell growth. Mol. Cell Biol. 10: 3013-3019

33 3 ChapterChapter 1

Mioo T., Yamada-Okabe T., Yabe T., Nakajima T., Arisawa M., and Yamada-Okabe H. (1997).. Isolation of the Candida albicans homologs of Saccharomyces cerevisiae KRE6 and SKN1:SKN1: expression and physiological function. J. Bacteriol. 179: 2363-2372

Moll P.C., Park H.M., Mullins J.T., and Cabib E. (1994). GTP-binding protein regulates thee activity of (l~>3)-beta-glucan synthase, an enzyme directly involved in yeast cell wall morphogenesis.. J. Biol. Chem. 269: 31267-31274

Montijnn R.C., Vink E., Muller W.H., Verkleij A.J., Van Den Ende H., Henrissat B., and Kliss F.M. (1999). Localization of synthesis of betal,6-glucan in Saccharomyces cerevisiae. J.J. Bacteriol. 181: 7414-7420

Morriss G.J., Winters L., Coulson G.E., and Clarke K.J. (1986). Effect of osmotic stress onn the ultrastructure and viability of the yeast Saccharomyces cerevisiae. J. Gen. Microbiol. 132:: 2023-2034

Moukadirii I., Jaafar L., and Zueco J. (1999). Identification of two mannoproteins released fromm cell walls of a Saccharomyces cerevisiae mnnl mnn9 double mutant by reducing agents. J.J. Bacteriol. 181: 4741-4745

Mr§aa V., Seidl T., Gentzsch M., and Tanner W. (1997). Specific labelling of cell wall proteinss by biotinylation. Identification of four covalently linked O-mannosylated proteins of SaccharomycesSaccharomyces cerevisiae. Yeast 13: 1145-1154

MrSaa V., and Tanner W. (1999). Role of NaOH-extractable cell wall proteins Ccw5p, Ccw6p,, Ccw7p and Ccw8p (members of the Pir protein family) in stability of the SaccharomycesSaccharomyces cerevisiae cell wall. Yeast 15: 813-820

Munizz M., and Riezman H. (2000). Intracellular transport of GPI-anchored proteins. EMBO J.J. 19: 10-15

Nagahashii S., Lussier M., and Bussey H. (1998). Isolation of Candida glabrata homologs off the Saccharomyces cerevisiae KRE9 and KNH1 genes and their involvement in cell wall beta-l,6-glucann synthesis. J Bacteriol. 180: 5020-5029

Orlean,, P. (1997). Biogenesis of yeast wall and surface components. In: Molecular and Cellularr Biology of the Yeast Saccharomyces, Vol. 3, Cell Cycle and Cell Biology (Pringle,

34 4 1,6-fi-glucan1,6-fi-glucan synthesis in Saccharomyces cerevisiae

J.R.,, Broach, J.R., and Jones, E.W., Eds.) Cold Spring Harbor, NY: Cold Spring Harbor Laboratoryy Press, pp. 229-362

Ozakii K., Tanaka K., Imamura H., Hihara T., Kameyama T., Nonaka H., Hirano H., Matsuuraa Y., and Takai Y. (1996). Romlp and Rom2p are GDP/GTP exchange proteins (GEPs)) for the Rholp small GTP binding protein in Saccharomyces cerevisiae. EMBO J. 15: 2196-2207 7

Pagee N., Gérard-Vincent M., Menard P., Beaulieu M., Azuma M., Dijkgraaf G.J., Li H., Marcouxx J., Nguyen T., Dowse T., Sdicu A.M., and Bussey H. (2003). A Saccharomyces cerevisiaecerevisiae Genome-Wide Mutant Screen for Altered Sensitivity to Kl Killer Toxin. Genetics 163:: 875-894

Parentt S.A., Nielsen J.B., Morin N., Chrebet G., Ramadan N., Dahl A.M., Hsu M.J., Bostiann K.A., and Foor F. (1993). Calcineurin-dependent growth of an FK506- and CsA- hypersensitivee mutant of Saccharomyces cerevisiae. J. Gen. Microbiol. 139: 2973-2984

Parkerr C.G., Fessier L.I., Nelson R.E., and Fessler J.H. (1995). Drosophila UDP- glucose:glycoproteinn glucosyltransferase: sequence and characterization of an enzyme that distinguishess between denatured and native proteins. EMBO J. 14: 1294-1303

Parlatii F., Dominguez M., Bergeron J.J., and Thomas D.Y. (1995). Saccharomyces cerevisiaecerevisiae CNE1 encodes an endoplasmic reticulum (ER) membrane protein with sequence similarityy to calnexin and calreticulin and functions as a constituent of the ER quality control apparatus.. J. Biol. Chem. 270: 244-253

Parodii A.J. (1999). Reglucosylation of glycoproteins and quality control of glycoprotein foldingg in the endoplasmic reticulum of yeast cells. Biochim. Biophys. Acta 1426: 287-295

Philipp B., and Levin D.E. (2001). Wscl and Mid2 are cell surface sensors for cell wall integrityy signaling that act through Rom2, a guanine nucleotide exchange factor for Rhol. Mol.Mol. Cell Biol. 21: 27'1-280

Popoloo L., Gilardelli D., Bonfante P., and Vai M. (1997). Increase in chitin as an essential responsee to defects in assembly of cell wall polymers in the ggpl delta mutant of SaccharomycesSaccharomyces cerevisiae. J. Bacteriol. 179: 463-469

35 5 ChapterChapter 1

Popoloo L., and Vai M. (1999). The Gasl glycoprotein, a putative wall polymer cross-linker. Biochim.Biochim. Biophys. Acta 1426: 385-400

Qadotaa H., Python C.P., Inoue S.B., Arisawa M., Anraku Y., Zheng Y., Watanabe T., Levinn D.E., and Ohya Y. (1996). Identification of yeast Rholp GTPase as a regulatory subunitt of 1,3-beta-glucan synthase. Science 111: 279-281

Ramm A.F.J., Wolters A., Ten Hoopen R., and Klis F.M. (1994). A new approach for isolatingg cell wall mutants in Saccharomyces cerevisiae by screening for hypersensitivity to calcofluorr white. Yeast 10: 1019-1030

Ramm A.F.J., Kapteyn J.C., Montijn R.C., Caro L.H.P., Douwes J.E., Baginsky W., Mazurr P., van den Ende H., and Klis F.M. (1998). Loss of the plasma membrane-bound proteinn Gaslp in Saccharomyces cerevisiae results in the release of betal,3-gliican into the mediumm and induces a compensation mechanism to ensure cell wall integrity. J. Bacteriol. 180:: 1418-1424

Reggiorii F., Canivenc-Gansel E., and Conzelmann A. (1997). Lipid remodeling leads to thee introduction and exchange of defined ceramides on GPI proteins in the ER and Golgi of SaccharomycesSaccharomyces cerevisiae. EMBOJ. 16: 3506-3518

Robinsonn N.G., Guo L., Imai J., Toh-E A., Matsui Y., and Tamanoi F. (1999). Rho3 of SaccharomycesSaccharomyces cerevisiae, which regulates the actin cytoskeleton and exocytosis, is a GTPasee which interacts with Myo2 and Exo70. Mol. Cell. Biol. 19: 3580-3587

Roemerr T., and Bussey H. (1991). Yeast beta-glucan synthesis: KRE6 encodes a predicted typee II membrane protein required for glucan synthesis in vivo and for glucan synthase activityy in vitro. Proc. Natl. Acad Sci. USA 88: 11295-11299

Roemerr T., Delaney S., and Bussey H. (1993). SKN1 and KRE6 define a pair of functional homologss encoding putative membrane proteins involved in beta-glucan synthesis. Mol. Cell. Biol.Biol. 13: 4039-4048

Roemerr T., Paravicini G., Payton M.A., and Bussey H. (1994). Characterization of the yeastt (l~>6)-beta-glucan biosynthetic components, Kre6p and Sknlp, and genetic interactionss between the PKC1 pathway and extracellular matrix assembly. J. Cell Biol. 127: 567-579 9

36 6 1,6'p-glucan1,6'p-glucan synthesis in Saccharomyces cerevisiae

Romeroo P.A., Dijkgraaf G.J., Shahinian S., Herscovics A., and Bussey H. (1997). The yeastt CWH41 gene encodes glucosidase I. Glycobiology 7: 997-1004

Sacherr M., Jiang Y., Barrow man J., Scarpa A., Burston J., Zhang L., Schieltz D., Yates J.R.. HI, Abeliovich H., and Ferro-Novick S. (1998). TRAPP, a highly conserved novel complexx on the cis-Golgi that mediates vesicle docking and fusion. EMBOJ. 17: 2494-2503

Sacherr M., Barrowman J., Schieltz D., Yates J.R. Ill, and Ferro-Novick S. (2000). Identificationn and characterization of five new subunits of TRAPP. Eur. J. Cell Biol. 79: 71- 80 0

Santoss B., and Snyder M. (1997). Targeting of chitin synthase 3 to polarized growth sites in yeastt requires Chs5p and Myo2p. J. Cell Biol. 136: 95-110

Sekiya-Kawasakii M., Abe M., Saka .A, Watanabe D., Kono K., Minemura-Asakawa M., Ishiharaa S., Watanabe T., and Ohya Y. (2002). Dissection of upstream regulatory componentss of the Rholp effector, 1,3-beta-glucan synthase, in Saccharomyces cerevisiae. GeneticsGenetics 162: 663-676

Sentandreuu R., Elorza M.V., and Villanueva J.R. (1975). Synthesis of yeast wall glucan. J. Gen.Gen. Microbiol. 90: 13-20

Shahiniann S., and Bussey H. (2000). |3-1,6-Glucan synthesis in Saccharomyces cerevisiae. Mol.Mol. Microbiol. 35: 477-489

Shaww J.A., Mol P.C., Bowers B., Silverman S.J., Valdivieso M.H., Duran A., and Cabib E.. (1991). The function of chitin synthases 2 and 3 in the Saccharomyces cerevisiae cell cycle.. J. Cell Biol. 114: 111-123

Shematekk E.M., Braatz J.A., and Cabib E. (1980). Biosynthesis of the yeast cell wall. I. Preparationn and properties of beta-(l leads to 3)glucan synthetase. J. Biol. Chem. 255: 888- 894 4

Shematekk E.M., and Cabib E. (1980). Biosynthesis of the yeast cell wall. II. Regulation of beta-(ll leads to 3)glucan synthetase by ATP and GTP. J. Biol. Chem. 255: 895-902

37 7 ChapterChapter 1

Silvermann S.J., Sburlati A., Slater M.L., and Cabib E. (1988). Chitin synthase 2 is essentiall for septum formation and cell division in Saccharomyces cerevisiae. Proc. Natl. Acad.Acad. Sci. USA 85: 4735-4739

Simonss J.F., Ebersold M., and Helenius A. (1998). Cell wall 1,6-beta-glucan synthesis in SaccharomycesSaccharomyces cerevisiae depends on ER glucosidases I and II, and the molecular chaperone BiP/Kar2p.. EMBOJ. 17: 396-405

Siposs G., Reggiori F., Vionnet C, and Conzelmann A. (1997). Alternative lipid remodellingg pathways for glycosylphosphatidylinositol membrane anchors in Saccharomyces cerevisiae.cerevisiae. EMBO J. 16: 3494-3505

Sullivann D.S., Biggins S., and Rose M.D. (1998). The yeast centrin, cdc31p, and the interactingg protein kinase, Kiclp, are required for cell integrity. J. Cell Biol. 143: 751-765

Terashimaa H., Fukuchi S., Nakai K., Arisawa M., Hamada K., Vabuki N., and Kitada K.. (2002). Sequence-based approach for identification of cell wall proteins in Saccharomyces cerevisiae.cerevisiae. Curr. Genet. 40: 311-316

Toh-ee A., Yasunaga S., Nisogi H., Tanaka K., Oguchi T., and Matsui Y. (1993). Three yeastt genes, PIR1, PIR2 and PIR3, containing internal tandem repeats, are related to each other,, and PIR1 and PIR2 are required for tolerance to heat shock. Yeast 9: 481-494

Trillaa J.A., Cos T., Duran A., and Roncero C. (1997). Characterization of CHS4 (CAL2), a genee of Saccharomyces cerevisiae involved in chitin biosynthesis and allelic to SKT5 and CSD4.CSD4. Yeast 13: 795-807

Trillaa J.A., Duren A., and Roncero C. (1999). Chs7p, a new protein involved in the control off protein export from the endoplasmic reticulum that is specifically engaged in the regulationn of chitin synthesis in Saccharomyces cerevisiae. J. Cell Biol. 145: 1153-1163

Trombettaa E.S., Simons J.F., and Helenius A. (1996). Endoplasmic reticulum glucosidase III is composed of a catalytic subunit, conserved from yeast to mammals, and a tightly bound noncatalyticc HDEL-containing subunit. J. Biol. Chem. 211: 27509-27516

Udenfriendd S., and Kodukula K. (1995). How Glycosylphosphatidylinositol-Anchored Membranee Proteins Are Made. Annu. Rev. Biochem. 64: 563-591

38 8 1,6-fi-glucan1,6-fi-glucan synthesis in Saccharomyces cerevisiae

Valdiviesoo M.H., Ferrario L., Vai M., Duran A., and Popoio L. (2000). Chitin synthesis in aa gasl mutant of Saccharomyces cerevisiae. J. Bacteriol. 182: 4752-4757

Vinkk E., Vossen J.H., Ram A.F.J., van den Ende H., Brekelmans S.S.C., de Nobel H., andd Klis F.M. (2002). The protein kinase Kiel affects 1,6-beta-glucan levels in the cell wall off Saccharomyces cerevisiae. Microbiology 148:4035-4048

Vossenn J.H., Muller W.H., Lipke P.N., and Klis F.M. (1997). Restrictive glycosylphosphatidylinositoll anchor synthesis in cwh6/gpi3 yeast cells causes aberrant biogenesiss of cell wall proteins. J. Bacteriol. 179: 2202-2209

Warmkaa J., Hanneman J., Lee J., Amin D., and Ota I. (2001). Ptcl, a type 2C Ser/Thr phosphatase,, inactivates the HOG pathway by dephosphorylating the mitogen-activated proteinn kinase Hogl. Mol. Cell. Biol. 21: 51-60

Watanabee D., Abe M., and Ohya Y. (2001). Yeast Lrglp acts as a specialized RhoGAP regulatingg 1,3-beta-glucan synthesis. Yeast 18: 943-951

Yamochii W., Tanaka K., Nonaka H., Maeda A., Musha T., and Takai Y. (1994). Growth sitee localization of Rho 1 small GTP-binding protein and its involvement in bud formation in SaccharomycesSaccharomyces cerevisiae. J. Cell Biol. 125: 1077-1093

Yunn D.J., Zhao Y., Pardo J.M., Narasimhan M.L., Damsz B., Lee H., Abad L.R., D'Urzoo M.P., Hasegawa P.M., and Bressan R.A. (1997). Stress proteins on the yeast cell surfacee determine resistance to osmotin, a plant antifungal protein. Proc. Natl. Acad. Sci. USAUSA 94: 7082-7087

Zhangg F.L., and Casey P.J. (1996). Protein prenylation: molecular mechanisms and functionall consequences. Annu. Rev. Biochem. 65: 241-69

Zimann M., Chuang J.S., and Schekman R.W. (1996). Chslp and Chs3p, two proteins involvedd in chitin synthesis, populate a compartment of the Saccharomyces cerevisiae endocyticc pathway. Mol. Biol. Cell. 7: 1909-1919

Zimann M., Chuang J.S., Tsung M., Hamamoto S., and Schekman R. (1998). Chs6p- dependentt anterograde transport of Chs3p from the chitosome to the plasma membrane in SaccharomycesSaccharomyces cerevisiae. Mol. Biol. Cell 9'. 1565-1576

39 9

ChapterChapter 2

Thee protein kinase Kiel affects l,6-/7-glucan levels in the celll wall of Saccharomyces cerevisiae

Edwinn Vink Jackk H. Vossen Arthurr F J. Ram Hermann van den Ende Stephann S.C. Brekelmans Hanss de Nobel Franss M. Klis

Thiss chapter was published with minor modifications in Microbiology (2002) 148:: 4035 - 4048 ChapterChapter 2

SUMMARY Y

K1C1K1C1 encodes a PAK kinase that is involved in morphogenesis and cell integrity. Both over- andd underexpressing conditions of KICJ affected cell wall composition. Kiel-deficient cells weree hypersensitive to the cell wall perturbing agent Calcofluor white and had less 1,6-/?- glucan.. When Kiel-deficient cells were crossed with various kre mutants, which also have lesss l,6-/?-glucan in their wall, the double mutants displayed synthetic growth defects. However,, when crossed with the 1,3-/?-glucan-deficient strain JkslA, no synthetic growth defectt was observed, supporting a specific role for KIC1 in regulating 1,6-/?-glucan levels. Kicc 1 -deficient cells also became highly resistant to the cell wall-degrading enzyme mixture Zymolyase,, and exhibited higher transcript levels of the cell wall protein-encoding genes CWP2CWP2 and SED1. Conversely, overexpression of K1C1 resulted in increased sensitivity to Zymolyasee and in a higher level of l,6-/?-glucan. Multicopy suppressor analysis of a Kicl- deficientt strain identified RH03. Consistent with this, expression levels of RH03 correlated withh 1,6-/?-glucan levels in the cell wall. Interestingly, expression levels of KIC1 and the MAPP kinase kinase PBS2 had opposite effects on Zymolyase sensitivity of the cells and on celll wall l,6-/?-glucan levels in the wall. We propose that Kiel affects cell wall construction inn multiple ways and in particular in regulating l,6-/?-glucan levels in the wall.

42 2 KIClKICl affects 1,6-fi-glucan synthesis

INTRODUCTION N

SaccharomycesSaccharomyces cerevisiae is protected from extracellular challenges by its cell wall. Thesee challenges can vary from hypo-osmotic stress to mechanical damage and toxic compoundss from other organisms. The yeast cell wall consists of l,3-/?-glucan, 1,6-/7-glucan, chitinn and mannoproteins, which are interconnected in an ordered manner (Klis et ah, 2002). Celll wall construction and composition are highly dynamic: the composition and structure of thee newly formed cell wall are continuously adjusted in response to extracellular conditions, andd even to progress in the cell cycle. This indicates that cell wall construction is highly regulated. . Thee cell wall perturbing agent Calcofluor white has been a valuable tool in identifying mutantss with a defective cell wall (Roncero et ah, 1988; Ram et ah, 1994; Lussier et ah, 1997;; De Groot et ah, 2001). Analysis of mutants hypersensitive to Calcofluor white has resultedd in the identification of numerous genes involved in different aspects of cell wall biogenesiss (Ram et ah, 1995, Jiang et ah, 1995; Vossen et ah, 1995; Jiang et ah, 1996; Van Berkell et ah, 1999). In addition, screening for mutants resistant to this compound, has led to thee identification of several genes involved in chitin biosynthesis (Roncero et ah, 1988). Thee KICl gene encodes an essential protein kinase, which is involved in cell integrity andd morphogenesis. This kinase was identified in a two-hybrid screen with the yeast centrin CDC31.CDC31. The in vitro kinase activity of Kiel was found to be dependent on CDC31. However, KIClKICl did not share the CDC31 functions in spindle pole body (SPB) duplication, but rather revealedd a novel function for CDC31 (Sullivan et ah, 1998). This was further supported by a mutationall analysis of CDC31, which resulted in dissection of the SPB-related functions and A7C7-relatedd functions (Ivanovska and Rose, 2001). Thee PKC mitogen activated protein kinase (MAPK) pathway is commonly known as thee cell (wall) integrity pathway, although this is certainly not the only MAPK pathway that hass an effect on cell wall biosynthesis and composition (Klis et ah, 2002). For instance, activationn of the pheromone response pathway results in the formation of a mating projection andd considerable alterations in the cell wall. In this case, at least part of this may be coordinatedd through the PKC MAPK pathway (Buehrer and Errede, 1997). Evidence is also accumulatingg that the HOG MAPK pathway plays a role in cell wall assembly. Overexpressionn of PBS2 - the MAP kinase kinase of this pathway - results in resistance to the Kll killer toxin (Jiang et ah, 1995) and to laminarinase (Lai et ah, 1997), an enzyme preparationn that contains l,3-/?-glucanase as its main activity. In addition, mutations in either PBS2PBS2 or HOG1 results in Calcofluor white resistance (Garcia-Rodriquez et ah, 2000). Kll killer toxin is a powerful tool for identifying mutants with defects in 1,6-/?-glucan synthesiss and/or assembly (Al-Aidroos and Bussey, 1978; Boone et ah, 1990; Brown et ah, 1993).. This pore-forming protein needs to bind its acceptors (l,6-/?-glucan and Krel) in order

43 3 ChapterChapter 2

too perform its lethal action (Bussey, 1991; Breinig et al, 2002). Mutants defective in the biogenesiss of 1,6-/?-glucan hold less of this acceptor, and therefore are resistant to this killer toxin.. Several genes have been identified, whose gene products mainly localize throughout thee secretory pathway (reviewed in Shahinian and Bussey, 2000). How the synthesis of 1,6-/?- glucann is regulated is largely unknown. Heree we describe the identification and characterization of CWH30, which is allelic to thee previously described KIC1 gene. A Kiel-deficient strain is not only hypersensitive to Calcofluorr white, but is also resistant to Zymolyase, a cell wall degrading enzyme mixture, indicatingg that its cell wall is affected. We show that mutation of K1C1 results in Kl killer toxinn resistance and decreased levels of l,6-/?-glucan. Furthermore, KIC1 expression levels weree found to correlate with 1,6-/?-glucan levels in the cell wall. Multicopy suppressor analysiss of a Kic 1-deficient strain identified RHÖ3 which itself was found to strongly affect l,6-/ï-glucann levels. We propose that KIC J is involved in regulating the 1,6-/?-glucan levels in thee cell.

44 4 K1C1K1C1 affects 1,6-fi-gIucan synthesis

MATERIALSS AND METHODS

Strainss and media. The yeast strains used in this study are listed in Table 1. The strains were grownn in YPD (1% [w/v] yeast extract, 1% [w/v] Bacto Peptone, 3% [w/v] glucose), YPGal (1%% [w/v] yeast extract, 1% [w/v] Bacto Peptone, 3% [w/v] galactose), or in SD (0.17% [w/v]] yeast nitrogen base without amino acids and ammonium sulfate, 2% [w/v] glucose, 0.5%% [w/v] ammonium sulfate, buffered at pH 6.0 with 1% [w/v] MES [morpholinoethanesulfonicc acid]) supported with the necessary amino acids, at 28°C or 37°C. Forr solid media, 2% (w/v) Bacto Agar was added. The Kl killer assays were performed in eitherr YPD or SD media, which for this purpose were buffered at pH 4.7 using 3% (w/v) sodiumm citrate and supplemented with 0.003% methylene blue. Yeast genetics, sporulation, andd transformation followed established procedures (Sherman and Hicks, 1991). Escherichia coltcolt strain DH5cc was used for propagation of all plasmids, and was grown in LB medium (1%% [w/v] Bacto Tryptone, 1% [w/v] NaCl, 0.5% [w/v] yeast extract). Yeast extract, Bacto Peptone,, Bacto Tryptone, yeast nitrogen base, and Bacto Agar were all from Difco Laboratories,, Detroit, Michigan (USA).

Strainn construction. Strains JV67 and JV68 were constructed by transforming strains HAB251-15BB and AR835 with KIC1 disruption constructs, using the HIS3 and TRP1 markers,, respectively. The disruption constructs were created according to Berben et al, (1991).. Primers used for this purpose are listed in Table 2. Correct integration of the disruptionn constructs was confirmed by Southern analysis, using the 4 kb Hindlll fragment fromm plasmid pl4 as a probe (see below). JV80 and JV83 were haploid offspring of JV68 and JV67,, respectively. JV142 and JV143 were haploid offspring of JV67 transformed with the PGALI-KICIPGALI-KICI plasmid (see below). JV144 and JV145 were offspring of JV68 transformed with thee PGALI'KICI plasmid. These four haploid strains propagated the PGAU:KIC1 plasmid even withoutt selective pressure. JV2022 was the progeny of the JV143 x HAB813 diploid. JV215 was a haploid offspringg of the JV143 x TR95 diploid. JV220 was a haploid descendant from the JV143 x HAB637-1AA diploid. The JV168 strain was a haploid descendant from the JV145 x AR100 diploid.. JV264 was constructed in the FY833 background, using a PBS2 disruption construct withh a HIS3 marker. This construct was created as described by Berben et al, (1991), and correctt integration was confirmed by PCR. Primers are all listed in Table 2. JV268 resulted fromm the JV144 * JV264 diploid. Strainss EV116 and EV077 were constructed in the FY834 background, using PCR- generatedd disruption constructs with the HIS3 marker (Berben et al, 1991). Primers are listed inn Table 2. Correct integration was confirmed by PCR.

45 5 ChapterChapter 2

Tablee 1. Yeast strains used in this study Strain n Genotype e Source/reference e

SEY6210 0 MATaa leu2-3,112 ura3-52 his3A200 Iys2-80I trpl-A901 suc2A9 S.D.. Emr

HAB251-15B B MATa/aa SEY6210 autodiploid Roemerr and Bussey (1991) ) FY833 3 MATaa his3A200 ura3-52 leu2Al fys2A202 trplA63 Winstonn et al., (1995) )

FY834 4 MATaa his3A200 ura3-52 leu2Al lys2A202 trplA63 Winstonn et al., (1995) )

AR835 5 MATa/aa FY833 * FY834 A.F.J.. Ram

T158C/S14a a killerr toxin producing strain Busseyy et al, (1979) ) cwh30-l cwh30-l MATaa ura3-52 cwh30-l Ramm et al., (1994)

JV67 7 HAB251-15BB kicL:HIS3 Thiss study

JV68 8 AR8355 kicl.TRPl Thiss study

JV80 0 MATaa his3A200ura3-52!eu2AItys2A202trplA63 kk!::HIS3 Thiss study

JV83 3 MATaa Ieu2-3,1I2 ura3-52 his3A200 lys2-801 trpl-A901 suc2A9 Thiss study kicl::HIS3 kicl::HIS3 JV141 1 MATaa leu2-3,112 ura3-52 his3A200 lys2-801 trpl-A901 suc2A9 Thiss study

kicl::TRPlkicl::TRPl + PGALI:KIC1 JV142 2 MATaa Ieu2-3,U2 ura3-52 his3A200 lys2-801 trpl-A901 suc2A9 Thiss study

kicl::HIS3kicl::HIS3 + PCAL1:KIC1 JV143 3 MATaa leu2-3,112 ura3-52 his3A20O Iys2-80I trpl-A901 suc2A9 Thiss study

kicl::HIS3kicl::HIS3 + PGALI:K1C1 JV144 4 MATaa his3A200 ura3-52 leu2Al lys2A202 trplA63 kicl::TRPI + Thiss study

PPGAllGAll:KIC! :KIC! JV145 5 MATaa his3A200 ura3-52 leu2Al \ys2A202 trplA63 kicl::TRPl + Thiss study pQAU'^lCl pQAU'^lCl

AR100 0 MATaa his3A200 ura3-52 leu2Al \ys2A202 trplA63 jksl::HIS3 Ramm etal, (1998)

HAB637-1A A MATaa leu2-3,112 ura3-52 his3A200 fys2-801 trpl-A90I suc2A9 Boonee et al, (1990) krel::HIS3 krel::HIS3

46 6 KIClKICl affects 1,6-fi-glucan synthesis

HAB813 3 MATaa Ieu2-3,U2 ura3-52 his3A200 Iys2-801 trpl-A901 suc2A9 Brownn and Bussey kre9::H!S3 kre9::H!S3 (1993) ) TR95 5 MATaa Ieu2-3,U2 ura3-52 his3A200 Iys2-80J trpl-A901 suc2A9 Roemerr and Bussey kre6::HIS3 kre6::HIS3 (1991) ) JV168 8 MATaa his3A200 ura3-52 leu2Al lys2A202 trp!A63 flcsl::HIS3 Thiss study

kicl::TRPl+Pkicl::TRPl+PGAL1GAL1:KICl :KICl JV202 2 MATaa leu2-3,112 ura3-52 his3A200 Iys2-80I trpI-A901 suc2A9 Thiss study

kre9::HIS3kre9::HIS3 kicl::HIS3 + PGAL1:KIC1 JV215 5 MATaa leu2-3,112 ura3-52 his3A200 lys2-801 trpl-A901 suc2A9 Thiss study

kre6::HIS3kre6::HIS3 kicl::HIS3+PGALI:KIC1 JV220 0 MATaa Ieu2-3,112 ura3-52 his3A200 lys2-801 trpl-A901 suc2A9 Thiss study

krel::HIS3krel::HIS3 kicl::HIS3 + PGAL,:KIC1 JV264 4 MATaa his3A200 ura3-52 leu2Al fys2A202 trplA63 pbs2::H!S3 Thiss study

JV268 8 MATaa his3A200 ura3-52 leu2Al lys2A202 trplA63 kicl::TRPl Thiss study

pbs2::HIS3pbs2::HIS3 + PGAL1:KIC1 EV116 6 MATaa his3A200 ura3-52 leu2Al lys2A202 trplA63 rho3::HIS3 thiss study

EV077 7 MATaa his3A200 ura3-52 leu2Al lys2A202 trplA63 rho4::HIS3 thiss study

Plasmids,, oligonucleotides and recombinant DNA techniques. For the cloning of the KIC1/CWH30KIC1/CWH30 gene, a YCp50-based genomic library was used (Rose et al, 1987). The plasmidd that could complement the cwh30-l mutant was named pi4. The 4 kb Hin&lll fragment,, containing the complete KICl open reading frame, was subcloned into YEplacl95 (Gietzz and Sugino, 1988) and named p61. Plasmid p62 was isolated from a YEpl3 based genomicc library based on its ability to complement the cwh30-l mutant. DNAA handling and manipulation were carried out according to Sambrook et al., (1989).. DNA sequencing was performed as described by Sanger et al, (1977), using T7 DNAA polymerase (Pharmacia). Restriction enzymes, nucleotides, Klenow fragment, and alkalinee phosphatase were all from Pharmacia. DNA ligase was purchased from Gibco-BRL, SuperTaqq polymerase was from HT Biotechnology (Cambridge, England), Expand high fidelityy polymerase was from Boehringer-Mannheim, and oligonucleotides were from Eurogentecc (Seraing, Belgium).

Cloningg of KICl behind the GAL1 promoter. The Expand high fidelity polymerase was usedd to create a KICl fragment with a 5' Xba\ restriction site, followed by three bases of the 5'' untranslated region (UTR) of KICl and then the KICl open reading frame. The first 500

47 7 ChapterChapter 2

basess of the 3' UTR were included in this fragment, which was followed by a Xhol restriction site.. The primers used for this are listed in Table 2. This PCR generated fragment was digestedd with Xba\ and Xhol, and subsequently cloned into the corresponding sites of the pYEUra33 plasmid (Clontech), resulting in the PGAL1:KIC1 plasmid.

Tablee 2. Oligonucleotide primers used in this study Name e Sequencee (5'-3')

KIC1KIC1 forward disruption CAGTAATGAC GACGAAGCCA CAAAATAGTA AGCAGGGTTT AGCCGAAGGAA GAATTCCCGG GGATCCG K1C1K1C1 reverse disruption CAGAACCTGG GGTTTCCTGT CTACCCGCCT CTGTTTTCCT GGCAGAAACGG AAGCTAGCTT GGCTGCAG

PPGALIGALI:KIC1:KIC1 forwarACTCTAGATd AA ATGACGACGA AGCCAC

PPGALIGALI:KIC1:KIC1 reversTTCTCGAGTe CC TAGGCGCGTT TATAAG

PBS2PBS2 forward disruption AAGATGGAAG ACAAGTTTGC TAACCTCAGT CTCCATGAGA AAACTGGTAAA GAATTCCCGG GGATCCG PBS2PBS2 reverse disruption ACGCTATAAA CCACCCATAT GTAATGCCGG TACATTTTTA GATAAACCATT AAGCTAGCTT GGCTGCAG

PBS2PBS2 forward control CAGATCGAGA CGTTAATTTC TCAAA

PBS2PBS2 reverse control TCACGTGCCTT GTTTGCTTTT

RHÖ3RHÖ3 forward disruption AACATGTCAT TTCTATGTGG GTCAGCTTCA ACGTCAAATA AACCGATCGAA GAATTCCCGG GGATCCG RHÖ3RHÖ3 reverse disruption ATATTACATA ATGGTACAGC TGGATCCACT GTCACTTTTC ACTTCGGTTGG AAGCTAGCTT GGCTGCAG RH04RH04 forward disruption TTCATGAATA CACTATTATT TAAGCGAAAA GGTGGCAATT GTGGGAACGAA GAATTCCCGG GGATCCG RH04RH04 reverse disruption TTCTTACATT ATAATACACT TGTTTTTTCT TAATCTTTTC GTTCCGGAAA AAGCTAGCTTT GGCTGCAG

Multicopyy suppressor screen. A high-copy pMA3a-based genomic library (kindly provided byy M. Crouzet and M. F. Tuite) with an average insert size of 5 - 10 kb was transformed to

strainn JV141, which contains kiclr.TRPl and carries the PGALJ-KIC1 plasmid. A total of 3 5000 transformants were replica-plated on selective SD medium containing 100 ug/ml Calcofluorr white. Viable colonies were isolated, library plasmids were recovered and

48 8 KIC1KIC1 affects 1,6-fi-glucan synthesis

retransformedd into the JV141 strain. Serial dilutions of transformants of these strains were spottedd on selective SD medium containing 50 jag/ml Calcofluor white. The seven plasmids thatt showed the best suppression were selected for further analysis, and were found to contain fivee unique inserts (Table 4). Clone 2 contained amongst others the RH03 gene, which provedd to be the gene responsible for the suppression. A 2.2 kb SaWXhol fragment containingg the RHÖ3 open reading frame (ORF) was removed from clone 2 and cloned into a YEplacl811 vector (Gietz and Sugino, 1988) resulting in the pEV021 plasmid. The remaining partt of clone 2 was religated, and the resulting plasmid lost the ability to suppress the kicl mutant,, whereas plasmid pEV021 retained the ability to suppress the kicl mutant. Clone 11 andd 13 contained an identical insert, and clone 11 was subjected to further analysis. A 2.4 kb XhoVSaRXhoVSaR fragment (the Sail restriction site was located in the pMA3a plasmid, 0.3 kb from thee BamHl site which was used for the insertion of the genomic fragments) was cloned into YEplacl811 resulting in plasmid pEV017, which contained MSG5 ORF and could suppress thee kicl mutant. Clone 23 and 44 also showed an identical insert, in which the STB3 gene was thee only complete ORF. In addition, 2.3 kb of the 3' end of the SEC7 coding sequence was presentt in this insert. This 2.3 kb fragment was removed by a XhollSall digest of clone 44, usingg the Xho\-s\te in the insert and the Safl-site in the pMA3a vector, 0.3 kb from the insert. Thee resulting plasmid (pEV020) retained the ability to suppress the kicl mutant.

Phenotypicc screens. Calcofluor white sensitivity was analyzed as described previously (Ram etet al, 1998). Precultures were concentrated or diluted to OD530 10. Subsequently, ten-fold dilutionn series were made of which 4 ul of each dilution was spotted onto YPD or SD plates containingg 0, 10, and 50 ng/ml Calcofluor white. Plates were incubated for 3 days at 28°C. Kll killer toxin sensitivity was measured using the halo assay (Brown et al, 1994), withh some modifications. Precultures were concentrated or diluted to OD530 10, and 45 \x\ cellss were seeded in 13 ml of killer agar medium. On the surface, 5 ul of a dilution series of 10°,, 5 * 10"', 10"1, 5 x 10"2, and 10"2 of isolated killer toxin was spotted. Killer toxin was isolatedd according to Brown et al, (1994). In short, the medium of the Kl killer toxin- producingg strain was concentrated a 1000-fold by ultrafiltration using a 10-kDa Amicon filter,, and was used as such in the halo assay. In all experiments, samples and controls were treatedd with toxin from the same isolation. Plates were incubated for 4 - 6 days at 20°C. The diameterss of the halos were measured for each toxin dilution. Relative apparent sensitivities weree calculated according to Reneke et al, (1988). In short, the diameter of the growth inhibitionn zone is proportional to the logarithm of the Kl killer toxin dose applied in the centerr of the zone. Plotting these parameters against each other allows estimation of the dose requiredd to produce an inhibition zone of a given diameter. The ratio of this dose estimated forr wild type cells divided by the dose found for mutant cells is termed the 'relative apparent sensitivity'. .

49 9 ChapterChapter 2

Zymolyasee sensitivities were measured essentially as described by De Nobel et al.,

(1990).. Yeast strains were grown to equivalent optical densities, and 1 OD530 unit was taken forr analysis. Cells were washed once and resuspended in 900 jxl 10 mM Tris/HCl, pH 7.5. Thee OD530 was followed for 1 hour after the addition of 100 ul Zymolyase 20T (10 mg/ml in 100 mM Tris/HCl, pH7.5; Zymolyase 20T was from Kirin Brewery, Japan).

Isolationn of cell walls. Cell walls from cells grown to early logarithmic phase were isolated accordingaccording to Van Rinsum and co-workers (1991). Walls were extracted twice in 50 mM Tris/HCl,, pH 7.4, 150 mM NaCl, 5 mM EDTA, 2% [w/v] SDS, 0.3% [v/v] B- mercaptoethanoll for 5 min at 100°C. Walls were extensively washed in distilled water, and subsequentlyy freeze dried.

Determinationn of l,6-/?-glucan levels. The levels of 1,6-/?-glucan in die alkali-insoluble cell walll fraction were basically determined in accordance to Brown and co-workers (1994). Essentially,, cells were grown for 24 h at 28°C to stationary phase, harvested, and washed twicee in distilled water. Samples were split up into four aliquots, and three fractions were eachh three times extracted in 3% [w/v] NaOH at 75°C for 1 h. The remaining fraction was freeze-driedd and used for the determination of the cell dry weight. The alkali-insoluble materiall was washed in 100 mM Tris/HCl, pH 7.5, twice, and subsequently washed in distilledd water. The pellet was resuspended in 10 mM Tris/HCl, pH 7.5, 10% [v/v] glycerol, 1 mg/mlmg/ml Zymolyase 100T (Kirin Brewery, Japan), and incubated at 37°C overnight. Following incubation,, samples were centrifuged at 14,000 g for 5 min, and the supernatant was dialyzed againstt distilled water using a Spectra/POR® 3 (6,000-8,000 MWCO) dialysis membrane. Thee glucose content of the residue - l,6-/?-glucan and chitin - was determined by the phenol- sulfuricc acid method (Dubois et al, 1956). In later experiments, the procedure was as follows.. Cells were grown for 24 h in YPD to early stationary phase, washed twice in 30 mM Tris-HCl,, 1 mM EDTA, pH 7.4, and collected in five aliquots. Two were freeze-dried in orderr to quantify the cell dry weight, whereas the three other samples were resuspended in 6000 ul 50 mM Tris-HCl, 150 mM NaCl, 5 mM EDTA, pH 7.4, and broken with glass beads. Wallss were collected and extracted twice in 50 mM Tris-HCl, 150 mM NaCl, 5 mM EDTA, 2%% SDS, 40 mM p-mercaptoethanol, pH 7.4, at 100°C for 5 min. Walls were extensively washedd in distilled water, and suspended in 10 mM Tris-HCl, pH 7.5, with 1 mg/ml (w/v) Zymolyasee 100T (Kirin Brewery Co., Ltd.) and incubated at 37°C for 16 h. Following incubation,, solutions were centrifuged for 5 min at 15 000 g and the supernatant was dialyzed againstt distilled water using a Spectra/POR® 3 (3,500 MWCO) dialysis membrane. The residuee was hydrolyzed in 2 M trifluoroacetic acid at 100°C for 4 h, freeze-dried, and subsequentlyy glucose levels were determined using the D-glucose oxidase assay.

50 0 KIC1KIC1 affects 1,6-fi-glucan synthesis

Northernn analysis. RNA was isolated from early log phase cells with hot acidic phenol (Currentt Protocols in Molecular Biology ; Greene Publishing Associates, Inc. and John Wiley && Sons, Inc., 1998). Fifteen u£ RNA was loaded on a 1% agarose gel containing 2.4% formaldehyde.. Following electrophoresis, the RNA was blotted onto Hybond-N+ (Amersham)) through capillary paper transfer and UV-crosslinked to the membrane.

51 1 ChapterChapter 2

RESULTS S

CWH30CWH30 is allelic to KIC1 Thee cwh30-l mutant was isolated in a general screen for cell wall mutants (Ram et al., 1994), basedd on their hypersensitivity to the cell wall perturbing agent Calcofluor white. The gene mutatedd in cwh30-l was identified by functional complementation of the hypersensitivity to Calcofluorr white, using a YCp50-based genomic library. Out of 40,000 transformants, only 155 were able to grow in the presence of 50 u.g/ml Calcofluor white. The plasmids from these transformantss were recovered, and upon retransformation only five were able to complement thee Calcofluor white hypersensitivity of the cwh30-l mutant. All five complementing plasmidss contained the same genomic insert (data not shown). The complementing activity of thiss insert resided in a 4.4 kb fragment that remained after HindUl truncation of the fragment, butt was lost after further truncation to 4.2 kb with Xba\. The 4.4 kb fragment contained the YHR102wYHR102w open reading frame, which previously has been named KIC1 for kinase interacting withh Cdc31 (Sullivan et al, 1998). Thecal truncation removed part of the 5' upstream regionn of the KIC1 gene, which seemed essential for complementation of the mutant. Additionall evidence that CWH30 was allelic to KIC1 came from the resulting diploid of the cwh30-lcwh30-l mutant and the kicl disruption strain JV143. The JV143 strain has been disrupted forr kicl but is supported by a plasmid-borne PGALI'KICI fusion. When the diploid was culturedd on glucose-containing medium, the Calcofluor white hypersensitivity caused by the recessivee cwh30-l mutation was not complemented (not shown), indicating that CWH30 is allelicc to KICL

Calcofluorr white YEPD (500 jug ml-1) www type i M T m a^ i i j ¥ mm

kidkid A rr T 11 ia

Figuree 1. Both kicl A and PGAU:K1C1 strains are Calcofluor white hypersensitive.

Tenfoldd serial dilutions of precultures of wild type (FY834), kicl A (JV80), and PGAU:KIC1 (JV144) were spottedd onto YEPD and YEPD containing 50 ug/ml Calcofluor white. Cells were grown for 2 days at 28'C. JV800 grows faster than the original kicl A strain, probably because it has acquired a second-site suppressor mutation. .

52 2 KICIKICI affects 1,6-fi-glucan synthesis

Disruptionn of KICI Inn order to generate a KICI knock-out mutant, the KICI gene was replaced by the TRP1 markerr in the diploid AR835 wild type strain, and by the HIS3 marker in the diploid HAB251-15BB wild type strain. For both heterozygous diploids (JV68 and JV67) tetrad analysiss resulted in two wild type colonies and two very poorly growing mutant colonies (dataa not shown). The poor growth of the kiclA strain could not be suppressed by osmotic supportt of the medium (data not shown). In addition our results show that mutant spores failedd to germinate on medium with galactose as the sole carbon source (data not shown). Microscopicc analysis showed that kiclA cells were enlarged and round, forming large clumps indicatingg a cell separation defect which is in accordance with previously described phenotypess of kicl mutants (data riot shown). In addition, the cells were very sensitive to pipettingg and centrifugation. Fig. 1 shows that the kicl A mutant displayed Calcofluor white hypersensitivity.. This was similar to that of the cwh30-l mutant (Ram et al, 1994; data not shown).. The morphological defects, the hypersensitivity to Calcofluor white, and the fragility off the cells are all in agreement with a role in cell wall integrity for KICI.

Constructionn of PGALI-'KICI Thee kicl A cells were not only very fragile, but they also had low transformation efficiencies, andd even heterozygous diploids sporulated very poorly. In addition, kicl A cells occasionally developedd second-site suppressor mutations. To circumvent the technical problems of workingg with a kicl A strain, a repressible KICI allele was constructed by placing it under the

controll of the GAL1 promoter in the pYEUra3 plasmid. The activity of the PGALI'KIC1 constructt was confirmed by its ability to complement the kicl A growth defect (data not shown).. This plasmid was transformed into a diploid heterozygous for kicl, and this strain wass sporulated. When germinated on YPGal medium, spores lacking the endogenous KICI genee grew indistinguishably from spores with the endogenous KICI gene (data not shown). However,, when germinated on YEPD medium, the spores lacking the endogenous KICI gene hadd a noticeable growth defect, albeit not as severe as a kicl A strain without the PGALI'KICI plasmidd (data not shown). This suggests that under glucose repression conditions there still is somee expression of the KICI gene. Under these conditions however, cells still display Calcofluorr white hypersensitivity and other defects in cell wall integrity (Fig. 1, 2, and 4).

Thee kicl A mutant carrying the PGALI'KICI plasmid will be referred to as the PCAu:KICl mutant.. Interestingly, when grown on nonselective media (e.g. YEPD), the kicl A strain was nott cured from the PGAU'-KICI plasmid. In addition, no viable colonies were found when PGALI'KICIPGALI'KICI cells were put on media containing 5'-FOA (data not shown).

53 3 ChapterChapter 2

(a) ) (d) ) \ \ £vO £ O i i K K i i S S

SED1 SED1

CWP2 CWP2 (b) ) i i

CWP1 CWP1

ACT1 ACT1

(c) )

Figuree 2. K1C1 gene dosage affects cell wall composition.

(a)) Cells from wild type (squares), kicJA (diamonds), PGALI:K1C1 (circles), and wild type with 2u KJC1 (triangles)) were precultured, washed and incubated in the presence of Zymolyase 20T. (b) Isolated cell walls of wildd type (squares), kiclA (diamonds), PGAU:KIC1 (circles), and wild type with 2u KJC1 (triangles) were incubatedd in the presence of Zymolyase 20T. The decrease in ODssonm was followed in time and is expressed in

54 4 KIC1KIC1 affects 1,6-fi-glucan synthesis

percentagee of the starting ODsaonn,. (c) KIC1 and PBS2 have opposing effects on Zymolyase sensitivity. Wild typee (squares), PGALI'KICI (diamonds), pbs2A (circles), and PGALI-KICI pbs2A (triangles) cells were precultured,, washed and incubated in the presence of Zymolyase 20T. The decrease in OD530nm was followed in timee and is expressed in percentage of the starting ODïsonm. (d) mRNA levels of cell wall protein encoding geness SED1, CWP2, and CWPl in wild type and the PGAL!:KIC1 mutant. Actin mRNA levels are shown as loadingg reference.

PPGAL1GAL1:KIC1:KIC1 mutant ceils are resistant to Zymolyase Zymolyasee is a commercial enzyme preparation with both l,3-/?-glucanase and protease activities,, which can be used to assay differences in cell wall structure and composition (De Nobell et al, 1990; Ram et al, 1994; De Groot et al, 2001). Wild type cells were sensitive to treatmentt with Zymolyase 20T, whereas Kiel-deficient cells were resistant (Fig. 2a). Conversely,, KIC1 expressed from a high-copy plasmid resulted in hypersensitivity to Zymolyasee (Fig. 2a). The major substrate for Zymolyase, i.e. l,3-/ï-glucan, forms the inner layerr of the cell wall. The outer layer consists of mannoproteins, which in intact cells limits thee permeability to macromolecules (Zlotnik et al, 1984; De Nobel et al, 1990). It is conceivablee that (at least part) of the Zymolyase sensitivity of intact cells can be attributed to changess in the protein outer layer and thus in cell wall permeability. One way to investigate thiss is to compare the Zymolyase sensitivity of intact cells and isolated walls, in which the innerr layer is now exposed to Zymolyase. The increase in Zymolyase resistance of cell walls fromm the PGALI:KIC1 mutant was much less dramatic compared to intact cells (Fig. 2b), but wass still significant. Cell walls derived from a strain with high copy numbers of KIC1 now weree more resistant to Zymolyase than wild type (Fig. 2b). Taken together, these data suggest thatt KIC1 affects cell wall permeability. This might be caused by altered mannoprotein levels (seee below) and possibly to some extent altered glucan levels.

Expressionn of some cell wall proteins is altered in the kicl mutant Thee mRNA expression levels of some known cell wall proteins in the PGAII-'KICI mutant showedd increased levels of both SED1 and CWP2 in comparison to wild type levels (Fig. 2d). CWPlCWPl mRNA levels, however, remained unaffected (Fig. 2d). Interestingly, both sedlA and cwplAcwplA mutants are more sensitive to Zymolyase than wild type (Van der Vaart et al, 1995; Shimoii et al, 1998). Overexpression of SED1 resulted in Zymolyase resistance (Shimoi et al,al, 1998). This strongly suggests that at least part of the Zymolyase resistance of the

PPGAUGAU:K1CI:K1CI mutant is caused by increased levels ofSEDl and CWP2.

Thee PcALiiKICl is more resistant to Kl killer toxin Kll killer toxin has been a powerful tool for identifying genes involved in l,6-/?-glucan synthesis.. Lower sensitivity to this toxin is generally associated with decreased levels of 1,6- /?-glucann in the cell wall, which is a receptor for the toxin (Boone et al, 1990; Brown et al,

55 5 ChapterChapter 2

1993).. KI killer toxin sensitivities were compared using the halo assay. Fig. 3(a) shows the haloo assays of some strains tested, to exemplify the difference in sensitivity. The Kl killer toxinn sensitivity of various strains compared to their corresponding wild type, is depicted graphicallyy in Fig. 3(b). Note that the overexpression studies were performed on supplementedd SD-based media, as opposed to the other strains which were tested on YEPD media.. The halo assays performed on SD-based media commonly showed larger halos than onn YEPD media, but the sensitivities compared to wild type remained consistent. Thee sensitivity of the kiclA mutant (JV83) to the Kl killer toxin was very low (not shown).. The PGAL/'KJCI mutant (JV142)retained a low sensitivity to the toxin, although not too the same extent as the deletion mutant. Conversely, overexpression of KIC1 in wild type (JV399 with plasmid p62) resulted in an increased sensitivity to the Kl killer toxin (Fig. 3a). Thee KIC1 gene dosage effect on Kl killer toxin sensitivity might reflect altered 1,6-/?-glucan levels. .

KIC1KIC1 affects cell wall l,6-/?-glucan levels Inn order to determine if the changes in Kl killer toxin sensitivity can be attributed to changes inn l,6-/?-glucan levels, the alkali-insoluble cell wall fraction were analyzed. As expected, the kiclAkiclA strain showed a marked decrease in l,6-/?-glucan levels (Table 3). Consistently, overexpressionn of KIC1 resulted in a slight increase in l,6-/?-glucan levels (Table 3). These dataa are in agreement with the data from the Kl killer toxin assay. The gene dosage effect of KIC1KIC1 on 1,6-/?-glucan levels suggests a role for KIC1 in 1,6-/7-glucan deposition.

Thee PGALÏ-KICI mutant displays synthetic growth defects with kre mutants Thee proposed role of KIC1 in 1,6-/7-glucan deposition in the cell wall is further supported by thee strong genetic interaction occurring between a PGALI'KICI mutant and kre mutants, which havee lowered 1,6-/?-glucan levels in the cell wall. The PGALI:KIC1 mutant was crossed with kre6A,kre6A, kre9A, and kre J A mutants, and the resulting diploids were sporulated. Double mutants weree selected on galactose-containing media, and the growth phenotypes were analyzed on glucose-containingg media. KIC1 showed a strongly enhanced growth defect with both KRE6 andd KRE9. Interestingly, when the PGALI'KICI mutant was crossed with krelA, enhancement off the growth defect was minor (Fig. 4). In contrast, KIC1 did not show an enhanced growth defectt with FKS1, a mutant impaired in 1,3-/?-glucan synthesis (Fig. 4). This suggests that the growthh defects of Kiel-deficient cells are to a large extent related to the biogenesis of 1,6-/?- glucan,, and offers further support for the notion that K1C1 is involved in the regulation of l,6-/?-glucann biogenesis.

56 6 KIClKICl affects 1,6-fS-glucan synthesis

(a) )

LL,:KIC1 ,:KIC1

(b) )

pbs2A pbs2A

PPGAUGAU:KIC1:KIC1 pbs2A

1000 1000 10000 relativee apparent sensitivity (%)

Figuree 3. KICl expression levels affect Kl killer toxin sensitivity. Relativee apparent sensitivity to the Kl killer toxin was determined as described in Materials and Methods, (a) Somee examples of the halo plate assay. Tenfold dilution series of isolated Kl killer toxin was spotted onto seededd plates. After 4 days at 20°C, halo diameters were measured and relative apparent sensitivities were calculated,, (b) Relative apparent sensitivities of several strains. Strains displayed in the upper panel were grown

57 7 ChapterChapter 2

onn SD-based medium with selective amino acid mix. Strains displayed in the lower panel were grown in YEPD- basedd medium.

Multicopyy suppressor screen of the Calcofluor white sensitivity of PGALI'KICI mutant Inn order to further elucidate the regulatory function of KIC1 in cell wall biosynthesis, we introducedd a high-copy pMA3a-based genomic library into the PGALI:KIC1 mutant strain JV141.. Thirty-five hundred transformants were replica-plated on selective SD medium containingg 100 jig/ml Calcofluor white. Plasmids isolated from the surviving colonies were retransformedd into JV141 and the original cwh30-I point mutant to ensure plasmid-dependent suppressionn of the Calcofluor white phenotype. Serial dilutions of cells were spotted onto selectivee SD containing 50 p.g/ml Calcofluor white. Partial sequence analysis of the inserts of thee seven strongest suppressors, revealed five separate genomic regions (Table 4). Several cloness were subjected to subcloning and deletion experiments, identifying the genes responsiblee for the (partial) suppression of the Calcofluor white hypersensitivity of the PGALI'KICIPGALI'KICI mutant (Fig. 5a). These consisted of (1) STB3, a gene encoding a 2-hybrid interactorr with the Sin3p protein (Kasten and Stillman, 1997), (2) MSG5, encoding a dual- specificityy protein phosphatase involved in the pheromone adaptation response (Doi et ai, 1994;; Zhan et ai, 1997) and, in addition, capable of influencing the phosphorylation state of Slt2pp (Watanabe et ai, 1995; Martin et ai, 2000), and (3) RH03. RH03 encodes a small G- protein,, which is known to be involved in bud formation and growth, organization of the actinn cytoskeleton and exocytosis (Matsui and Toh-e, 1992b; Imai et ai, 1996; Robinson et ai,ai, 1999; Adamo et ai, 1999). JV141 cells were transformed with either of the before mentionedd plasmids, and grown on selective medium containing 5'-FOA to determine if the presencee of these multicopy suppressor plasmids allowed the loss of the PGALI'KICI plasmid. Noo viable colonies could be found, indicating that none of these plasmids could restore (all of)) the essential function(s) of KIC1 (data not shown). RHÖ3 was chosen for further analysis.

Tablee 3. KIC1 gene dosage affects 1,6-/?-glucan levels Strain n Alkali-insolublee 1,6-/?-glucan (%)* * wildd type 100 0 kicld kicld 722 7 wildd type" 100 0 kiclA^ kiclA^ 8 8 wildd type + 2\iKlCl* 1122 1 ** Measured as ug alkali-insoluble glucan per mg dry weight cell wall, and expressed as percentages relative to wildd type as described in materials and methods. Values represent mean SEM (n=3). ## These strains were grown on SD medium ff Strain with improved growth rate, probably resulting from a second-site suppressor mutation

58 8 KICIKICI affects 1,6-fi-gIucan synthesis

Thee rho3 deletant is impaired in cell wall biogenesis RH03RH03 was disrupted, and the cells were tested for their sensitivity to Calcofluor white. Althoughh not to the same extent as Kiel-deficient cells, rho3A cells showed increased sensitivityy to Calcofluor white, indicating a defect in cell wall biogenesis (Fig. 5b). In the absencee of RH03, the functionally related RH04 can suppress the growth defect when overexpressedd (Matsui and Toh-e, 1992a), whereas the rho3 rho4 double mutant is inviable. Thee double mutant also proved inviable in our genetic background (data not shown). Cells disruptedd for RH04 did not show an increased sensitivity to Calcofluor white (Fig. 5b). In addition,, overexpression of RH04 did not restore the Calcofluor white hypersensitivity of the PGALI'KICPGALI'KIC mutant (data not shown). This suggests that the function that is suppressed by overexpressionn OÏRHÖ3 in the PGALI'KIC I mutant, is not shared by RH04. Overexpressionn of the RH03 gene in the PGALI'KICI mutant showed an increase in sensitivityy to the Kl killer toxin, in contrast to overexpression of the RH04 gene (Fig. 6a). Deletionn of RH03 resulted in a decrease in Kl killer toxin sensitivity (Fig. 6b), and a decreasee of about 40% in cell wall 1,6-/?-glucan (3.3% of cell dry weight in wild-type cells comparedd to 1.9). In contrast, deletion of RHÖ4 had no effect on Kl killer toxin sensitivity (Fig.. 6b). Overexpression of the RHÖ3 gene in wild type also resulted in an increase in Kl killerr toxin sensitivity (Fig. 6c). However, when RH04 was overexpressed in wild type, cells displayedd a decrease in killer sensitivity (Fig. 6c). A possible explanation for this is that high levelss of Rho4 might compete with Rho3. Taken together, the observations suggest that the levelss of RHÖ3 influence the level of cell wall l,6-/?-glucan and thus the sensitivity of the cellss to the Kl killer toxin. This further implicates KICI in 1,6-/?-glucan biogenesis, evidently forr a part through RH03.

Tablee 4. High copy suppressors of KICI Clonee number Chromosomee . Coordinates s Completee ORFs 2 2 9 9 135225-144766 6 YIL120w,, RPI1, RH03, YIL117c, HIS5, NUP159 11/13* * 14 4 523677-531475 5 VAC7,MSG5 VAC7,MSG5 23/44* * 4 4 791592-798528 8 STB3 STB3 24 4 11 1 487450-497450t t GCN3,GCN3, YKR027w, SAP 190 43 3 2 2 243957-253522 2 YBR004c,, YBR005w, UGA2, YBR007c —— —' - _ -— Boldfacee type is used in cases where the ORF responsible for the suppression has been determined by subcloning. . *Twoo clones with identical inserts were identified. tThiss clone was only sequenced from the left flank. The right flank was estimated based on average insert size andd restriction analysis.

KICIKICI antagonizes the PBS2-HOG1 pathway in cell wall biogenesis Celll wall phenotypes caused by overexpression of the MAP kinase kinase PBS2 from the HOGG pathway show a remarkable resemblance with some of the phenotypes shown by

59 9 ChapterChapter 2

PGALI'KICIPGALI'KICI mutant cells. Similar to cells that overexpress PBS2 (Jiang et ah, 1995; Lai et ah, 1997),, PGALI'KICI mutant cells were less sensitive to Kl killer toxin, and showed a modest decreasee in 1,6-yS-glucan levels (Fig. 3b; Table 4). The reverse phenotypes were found in the pbs2Apbs2A mutant and cells overexpressing KIC1, i.e., hypersensitivity to the Kl killer toxin (Jiangg et ah, 1995; Lai et ah, 1997; Fig. 3b) and an increase in 1,6-/?-glucan levels (Jiang et ah,ah, 1995; Table 3). A PGALI'KICI pbs2A double mutant was generated, and this strain displayedd an intermediate sensitivity to Kl killer toxin (Fig. 3b). KIC1 deficiency and deletionn of PBS2 also reversely affected the sensitivity of the cells to cell wall degrading enzymess (Fig. 2c), whereas the double mutant displayed an intermediate phenotype. In summary,, our observations in combination with data from the literature indicate that K1C1 andd PBS2 have opposing roles in cell wall biogenesis.

#

Figuree 4. K1C1 displays synthetic growth phenotypes with several kre mutants.

Thee PGAU:K1C1 mutant strain was crossed with kre6A , kre9A, krelA, andjkslA. Tetrads were dissected on YPGal,, and the double and single mutants were tested for growth on YEPD in tenfold serial dilutions.

60 0 KIClKICl affects 1,6-p-glucan synthesis

DISCUSSION N

Thee KICl gene was originally identified in a two-hybrid screen interacting with CDC31,CDC31, which encodes yeast centrin. KIC1, member of the PAK1/Ste20 kinase family, encodess a 116 kDa protein which interacts in vivo with CDC31 and has in vitro kinase activityy dependent on CDC3L However, it was shown that KICl did not play a role in the SPBB duplication function of CDC31, KICl rather contributes to its function in cell integrity andd morphogenesis, since several kicl mutants showed aberrant cell wall morphology, wide budd necks, failure in cell separation, and cell lysis (Sullivan et al, 1998). Previously, cells deletedd for KICl were found to be inviable (Sullivan et al, 1998). However, differences in geneticc backgrounds might explain the (albeit very poor) viability of the kicl A strain in our backgrounds. . Thee cwh30/kicl mutant was originally discovered because it was hypersensitive to the celll wall perturbing agent Calcofluor White (Ram et al., 1994). Interestingly, mutant cells weree also resistant to the Kl killer toxin, indicating that its walls contained less l,6-/7-glucan. Ass KIC1 encodes a protein kinase, this marked it as a potential regulator of 1,6-/?-glucan biogenesis.. The following evidence supports this. KICl -deficiency resulted in decreased sensitivityy to the Kl killer toxin, and lower levels of 1,6-/?-glucan in its walls. Conversely, overexpressionn of KICl resulted in increased sensitivity to Kl killer toxin and elevated levels

off l,6-/?-glucan. In addition, the PGAU:KIC1 mutant crossed with various kre mutants resulted

inn double mutants with a synthetic growth defect, whereas a combination of the PGAU:KIC1 mutantt with the l,3-/?-glucan impaired mutant fkslA did not result in a synthetic growth defect.. Taken together, these results imply a role for KICl in the regulation of 1,6-/?-glucan biogenesis.. Finally, the expression levels of RH03 correlated with the sensitivity to Kl killer toxinn and thus probably with 1,6-/?-glucan levels. In addition, the mutant phenotypes of Kiel - deficientt cells were partially suppressed by overexpression of RH03. This is consistent with thee postulated role for KICl in regulating 1,6-/?-glucan biogenesis. Besidess the defects in 1,6-/?-glucan deposition, the Kicl-deficient cells also displayed resistancee to the cell wall-degrading enzyme mixture Zymolyase, whereas overexpression of KIClKICl resulted in hypersensitivity to Zymolyase. This effect may partly be caused by changes inn the cell wall mannoprotein composition, since the external layer of mannoproteins in the celll wall determines the porosity and needs to be removed for efficient cell wall degradation (Zlotnikk et al, 1984; De Nobel et al, 1990). There are two lines of evidence that confirm this notion.. First, when cell walls were isolated prior to Zymolyase treatment, walls from Kicl- deficientt cells showed a much less pronounced resistance to Zymolyase compared to wild typee cell wall. Second, two known cell wall protein encoding genes, SED1 and CWP2, were

foundd to be upregulated in the PGAU:KIC1 mutant. Interestingly, overexpression of SED1 leadss to Zymolyase resistance (Shimoi et al, 1998), and, reversely, deletion of both SED1

61 1 ChapterChapter 2

andd CWP2 results in increased sensitivity to Zymolyase (Van der Vaart et al, 1995; Shimoi etet al, 1998). Increased expression of these genes in the Kiel-deficient mutants might thus at leastt in part explain the resistance to Zymolyase. By which mechanism the expression of thesee cell wall proteins is induced is unknown. It might reflect the induction of a cell wall repairr mechanism as the result of the decrease in 1,6-yS-glucan. (Popoio et ah, 1999; Kapteyn etet al, 1999; Klis et al, 2002), but this normally includes induction of CWP1 expression (Terashimaa et al, 2000; Kapteyn et al, 2001). However, CWP1 expression was not induced inn the PGAL,:KIC1 mutant. Alternatively, Kiel might have a regulatory role in multiple cell walll biosynthetic steps and not only in 1,6-^-glucan biogenesis.

(a) (a) Calcofluorr white YEPD D (500 |^g ml-1)

wildd type + 2M

PPGALtGALt:KIC1:KIC1 + 2M

PPGAUGAU:KIC1:KIC1 + 2|J RH03 # # *

PPGAL1GAL1:KIC1:KIC1 + 2M MSG5 ' t m «

PPGAL1GAL1:KIC1:KIC1 + pMA3a

PPrr.,,:KIC1.,,:KIC1 + STB3IB»» .t . ^^ ^ # €1 ^ ^

(b) ) Calcofluorr white YEPD D (50p.gg ml"1)

WcM M

Figuree 5. Multicopy suppressors of the Calcofluor white hypersensitivity of the PGAL1:KIC1 mutant,

(a)) Tenfold serial dilutions of wild type (FY69) + YEplacl81, PGALI:K1C1 (JV141) + YEplacl81, PGAL1:KIC1

(JV141)) + 2p. RH03 (pEV021), PGAL,:KIC1 (JV141) + 2M MSG5 (pEV017), PGAL,:KIC1 (JV141) + pMA3a, and PGALI-KICIPGALI-KICI (JV141) + pMA3a-SrS5 (pEV020) were spotted onto selective SD media with or without 50 ug/ml Calcofluorr white, (b) Tenfold dilution series of wild type (FY834), rho3A (EV116), rho4A (EV077), kid A +

62 2 KIC1KIC1 affects 1,6-fi-glucan synthesis

PGALI-KICIPGALI-KICI (JV144), and kiel A (JV80) were spotted onto YPD with or without 50 ug/ml Calcofluor white. Cellss were grown for 2-3 days at 30°C.

Thee identification of RH03 as a multicopy suppressor of the PGALI'KICI mutant suggestss that RH03 might be a downstream target of KIC1 in cell wall biogenesis. This is supportedd by the RH03 gene dosage relationship with Kl killer toxin sensitivity, and the reductionn of 1,6-/7-levels in the rho3A mutant. Whereas RH04 contributes to some of the knownn functions of RH03 (Matsui and Toh-e, 1992; Imai et al, 1996), the effects on cell walll biogenesis are not shared by RH04. Evidencee is accumulating that suggests a role for the PBS2-HOG1 pathway in cell walll construction. Overexpression of PBS2 causes resistance to laminarinase, a cell wall degradingg enzyme mixture (Lai et al, 1997), and deletion results in hypersensitivity (Fig. 2b; Alonso-Mongee et al, 2001). In addition, PBS2 overexpression results in resistance to the Kl killerr toxin and a decrease in cell wall 1,6-/?-glucan levels (Jiang et al, 1995). Also, under noninducingg conditions the HOG pathway contributes to the maintenance of cell wall architecturee (Garcia-Rodriquez et al, 2000). Furthermore, overexpression of some cell wall relatedd genes suppress the hyperosmosensitive phenotype of a stel Issk2ssk22 mutant. These includee LRE1 and HLR1, which can also suppress the osmosensitivity and the glucanase sensitivityy of both pbs2A and hoglA mutants (Alonso-Monge et al, 2001). Our report further supportss a role for PBS2 in cell wall biogenesis. The PGALI:KIC1 mutant and the pbs2A single mutantt cells had reverse phenotypes in both Kl killer toxin and Zymolyase sensitivities. In thee PGALI:K1C1I pbs2A double mutant an intermediate phenotype was observed (Fig. 3b). Thesee results suggest that K1C1 and PBS2 play opposing roles in cell wall biogenesis. The mechanismm by which KIC1 and PBS2 counteract each other remains obscure. In summary, thee protein kinase Kiel is involved in regulating cell wall construction in multiple ways and seemss to have a specific role in controlling 1,6-/?-glucan levels.

63 3 ChapterChapter 2

(a) )

PPGAL1GAL1:KIC1:KIC1 + 2M

PPGAL1GAL1:KIC1:KIC1 + 2M RH03

PPGAL1GAL1:KIC1:KIC1 + 2M RH04

—ii 1 1 10 0 155 20 25 (b) )

wildd type

rho3& rho3&

rho4& rho4&

PQ^,:/C/C1 1

ii i i ii i ()) ( 25 50 75 1000 125 (c) )

wildd type + 2M

wildd type + 2M RH03

wildd type + 2p RH04

II I I I I 00 50 100 150 200

relativee apparent sensitivity (%)

Figuree 6. RH03 gene dosage affects K.1 killer toxin sensitivity. Relativee apparent sensitivity to the Kl killer toxin was determined as described in Materials and Methods, (a) kidkid A + PGAU:KICI (JV144) + YEplacl81, kid A + PGAL,:K1C1 (JV144) + 2u RH03, and kid A + PGAU:KIC1 (JV144)) + RH04 are compared, (b) Wild type (FY834), rhoSA (EV116), rho4A (EV077), and kid A +

PPGALIGALI:KIC1:KIC1 (JV144) were compared, (c) Wild type (FY834) + YEplac 181, wild type (FY834) + 2p RH03, and wildd type (FY834) + 2p RH04 were compared.

ACKNOWLEDGMENTS S Thee authors thank all members of the Klis lab for helpful comments and suggestions. Dr. Marcoo Siderius is thanked for many stimulating discussions. We are indebted to Dr. Howard Busseyy for sharing strains and plasmids, and to Drs Crouzet and Tuite for sharing their pMA3aa genomic library.

64 4 KIC1KIC1 affects 1,6-fi-glucan synthesis

REFERENCES S

Al-Aidrooss K., and Bussey H. (1978). Chromosomal mutants of Saccharomyces cerevisiae affectingg the cell wall binding site for killer factor. Can J Microbiol 24:228-237

Alonso-Mongee R., Real E., Wojda I., Bebelman J.P., Mager W.H., and Siderius M. (2001).. Hyperosmotic stress response and regulation of cell wall integrity in Saccharomyces cerevisiaecerevisiae share common functional aspects. Mol Microbiol 41:717-730

Berbenn G., Dumont J., Gilliquet V., Bolle P.A., and Hilger F. (1991). The YDp plasmids: aa uniform set of vectors bearing versatile gene disruption cassettes for Saccharomyces cerevisiae.cerevisiae. Yeast 1:415-477

Boonee C, Sommer S.S., Hensel A., and Bussey H.(1990). Yeast KRE genes provide evidencee for a pathway of cell wall beta-glucan assembly. J Cell Biol 110:1833-1843

Brownn J.L., Kossaczka Z., Jiang B., and Bussey H. (1993). A mutational analysis of killer toxinn resistance in Saccharomyces cerevisiae identifies new genes involved in cell wall (1~ >6)-beta-glucann synthesis. Genetics 133:837-849

Brownn J.L., Roemer T., Lussier M., Sdicu A.M., and Bussey H. (1994). The Kl killer toxin:: molecular and genetic applications to secretion and cell surface assembly. In MolecularMolecular Genetics of Yeast: A Practical Approach, pp. 217-232. Edited by J.R. Johnston. Oxfordd University Press, U.K.: IRL Press.

Buehrerr B.M., and Errede B. (1997). Coordination of the mating and cell integrity mitogen-activatedd protein kinase pathways in Saccharomyces cerevisiae. Mol Cell Biol 17:6517-6525 5

Busseyy H. (1991). Kl killer toxin, a pore forming protein from yeast. Mol Microbiol 5:2339- 2343 3

Dee Nobel J.G., Klis F.M., Priem J., Munnik T., and Van Den Ende H. (1990). The glucanase-solublee mannoproteins limit cell wall porosity in Saccharomyces cerevisiae. Yeast 6:491-499 9

Dee Groot P.W.J., Ruiz C, Vazquez de Aldana C.R., Duenas E., Cid V.J., Del Rey F., Rodriquez-Penaa J.M., Perez P., Andel A., Caubin J., Arroyo J., Garcia J.C., Gil C,

65 5 ChapterChapter 2

Molinaa M., Garcia L.J., Nombela C, and Klis F.M. (2001). A genomic approach for the identificationn and classification of genes involved in cell wall formation and its regulation in SaccharomycesSaccharomyces cerevisiae. Comp Fund Genomics 2:124-142

Duboiss M., Gilles K.A., Hamilton J.K., Rebers P.A., and Smith F. (1956). Colometric methodd for determination of sugars and related substances. Anal Chem 28:350-356

Garcia-Rodriguezz L.J., Duran A., and Roncero C. (2000). Calcofluor antifungal action dependss on chitin and a functional high-osmolarity glycerol response (HOG) pathway: evidencee for a physiological role of the Saccharomyces cerevisiae HOG pathway under noninducingg conditions. J Bacteriol 182:2428-2437

Gietzz R.D., and Sugino A. (1988). New yeast-Escherichia coli shuttle vectors constructed withh in vitro mutagenized yeast genes lacking six-base pair restriction sites. Gene 74:527-534

IvanovskaIvanovska I., and Rose M.D. (2001). Fine structure analysis of the yeast centrin, Cdc31p, identifiess residues specific for cell morphology and spindle pole body duplication. Genetics 157:503-518 8

Jiangg B., Ram A.F.J., Sheraton J., Klis F.M., and Bussey H. (1995). Regulation of cell walll beta-glucan assembly: PTC J negatively affects PBS2 action in a pathway that includes modulationn ofEXGI transcription. Mol Gen Genet 248:260-269

Jiangg B., Sheraton J., Ram A.F.J., Dijkgraaf G.J., Klis F.M., and Bussey H. (1996). CWH41CWH41 encodes a novel endoplasmic reticulum membrane N-glycoprotein involved in beta 1,6-glucann assembly. J Bacteriol 178:1162-1171

Kapteynn J.C., Van Egmond P., Sievi E., Van Den Ende H., Makarow M., and Klis F.M. (1999).. The contribution of the O-glycosylated protein Pir2p/Hspl50 to the construction of thee yeast cell wall in wild-type cells and beta-1,6-glucan-deficient mutants. Mol Microbiol 31:1835-1844 4

Kapteynn J.C., ter Riet B., Vink E., Blad S., De Nobel H., Van Den Ende H., and Klis F.M.. (2001). Low external pH induces HOG1 -dependent changes in the organization of the SaccharomycesSaccharomyces cerevisiae cell wall. Mol Microbiol 39:469-479

Kliss F.M., Mol P.C., Hellingwerf K., and Brul S. (2002). Dynamics of cell wall structure in SaccharomycesSaccharomyces cerevisiae. FEMS Microbiol Rev, in press

66 6 KIC1KIC1 affects 1,6-fJ-glucan synthesis

Laii M.H., Silverman S.J., Gaughran J.P., and Kirsch D.R. (1997). Multiple copies of PBS2,PBS2, MHP1 or LRE1 produce glucanase resistance and other cell wall effects in SaccharomycesSaccharomyces cerevisiae. Yeast 13:199-213

Lussierr M., White A.M., Sheraton J., di Paolo T., Treadwell J., Southard S.B., Horensteinn C.I., Chen-Weiner J., Ram A.F.J., Kapteyn J.C., Roemer T.W., Vo D.H., Bondocc D.C., Hall J., Zhong W.W., Sdicu A.M., Davies J., Klis F.M., Robbins P.W., and Busseyy H. (1997). Large scale identification of genes involved in cell surface biosynthesis andd architecture in Saccharomyces cerevisiae. Genetics 147: 435-450

Popoloo L., and Vai M. (1999). The Gasl glycoprotein, a putative wall polymer cross-linker. Biochim.Biochim. Biophys. Acta 1426:385-400

Posass F., Chambers J.R., Heyman J.A., Hoeffler J.P., de Nadal E„ and Arino J. (2000). Thee transcriptional response of yeast to saline stress. J Biol Chem 275:17249-17255

Ramm A.F.J., Wolters A., Ten Hoopen R., and Klis F.M. (1994). A new approach for isolatingg cell wall mutants in Saccharomyces cerevisiae by screening for hypersensitivity to calcofluorr white. Yeast 10:1019-1030

Ramm A.F.J., Brekelmans S.S.C., Oehlen L.J.W.M., and Klis F.M. (1995). Identification of twoo cell cycle regulated genes affecting the beta-1,3-ghican content of cell walls in SaccharomycesSaccharomyces cerevisiae. FEBS Lett 358:165-170

Ramm A.F.J., Kapteyn J.C., Montijn R.C., Caro L.H.P., Douwes J.E., Baginsky W., Mazurr P., Van Den Ende H., and Klis F.M. (1998). Loss of the plasma membrane-bound proteinn Gaslp in Saccharomyces cerevisiae results in the release of beta-l,3-ghican into the mediumm and induces a compensation mechanism to ensure cell wall integrity. J Bacteriol 180:: 1418-1424

Renekee J.E., Blumer K.J., Courchesne W.E., and Thorner J. (1988). The carboxy- terminall segment of the yeast alpha-factor receptor is a regulatory domain. Cell 55:221-234

Repp M., Krantz M., Thevelein J.M., and Hohmann S. (2000). The transcriptional response off Saccharomyces cerevisiae to osmotic shock. Hotlp and Msn2p/Msn4p are required for the inductionn of subsets of high osmolarity glycerol pathway-dependent genes. J Biol Chem 275:8290-8300 0

67 7 ChapterChapter 2

Reynoldss T.B., Hopkins B.D., Lyons M.R., and Graham T.R. (1998). The high osmolality glyceroll response (HOG) MAP kinase pathway controls localization of a yeast golgi glycosyltransferase.. J Cell Biol 143:935-946

Ronceroo C, Valdivieso M.H., Rib as J.C., and Duran A. (1988). Isolation and characterizationn of Saccharomyces cerevisiae mutants resistant to Calcofluor white. J BacteriolBacteriol 170:1950-1954

Rosee M.D., Novick P., Thomas J.H., Bothstein D., and Fink G.R. (1987). A SaccharomycesSaccharomyces cerevisiae genomic plasmid bank based on a centromeric-containing shuttle vector.. Gene 60:237-243

Sambrookk J., Fritsch E. F., and Maniatis T. (1989). Molecular Cloning: A Laboratory Manual,Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sangerr F., Nicklen S., and Coulson A. R. (1977). DNA sequence with chain-terminating inhibitors.. Proc Natl Acad Sci 74, 5463-5467

Shahiniann S., and Bussey H. (2000). P-1,6-Glucan synthesis in Saccharomyces cerevisiae. MolMol Microbiol 35:477-489

Shermann F., and Hicks J. (1991). Micromanipulation and dissection of asci. Methods EnzymolEnzymol 194:21-37

Shimoii H., Kitagaki H., Ohmori H., Iimura Y., and Ito K. (1998). Sedlp is a major cell walll protein of Saccharomyces cerevisiae in the stationary phase and is involved in lytic enzymee resistance. J Bacteriol 180:3381-3387

Smitss G.J., Van Den Ende H., and Klis F.M. (2001). Differential regulation of cell wall biogenesiss during growth and development in yeast. Microbiology 147:781-794

Sullivann D.S., Biggins S., and Rose M.D. (1998). The yeast centrin, Cdc31p, and the interactingg protein kinase, Kic 1, are required for cell integrity. J Cell Biol 143:751 -765

Terashimaa H., Yabuki N., Arisawa M., Hamada K., and Kitada K. (2000). Up-regulation off genes encoding glycosylphosphatidylinositol (GPI)-attached proteins in response to cell walll damage caused by disruption of FKS1 in Saccharomyces cerevisiae. Mol Gen Genet 264:64-74 4

68 8 KICJKICJ affects 1,6-p-glucan synthesis

vann Berkel M.A., Rieger M., te Heesen S., Ram A.F.J., Van Den Ende H., Aebi JV1., and Kliss F.M. (1999). The Saccharomyces cerevisiae CWH8 gene is required for full levels of dolichol-linkedd oligosaccharides in the endoplasmic reticulum and for efficient N- glycosylation.. Glycobiology 9:243-253

Vann Der Vaart J.M., Caro L.H.P., Chapman J.W., Klis F.M., and Verrips C.T. (1995). Identificationn of three mannoproteins in the cell wall of Saccharomyces cerevisiae. J BacteridBacterid 177:3104-3110

Vann Rinsum J., Klis F.M., and Van Den Ende H. (1991). Cell wall glucomannoproteins of SaccharomycesSaccharomyces cerevisiae mnn9. Yeast 7:717-26

Vossenn J.H., Muller W.H., Lipke P.N., and Klis F.M. (1997). Restrictive glycosylphosphatidylinositoll anchor synthesis in cwh6/gpi3 yeast cells causes aberrant biogenesiss of cell wall proteins. J Bacteriol 179:2202-2209

Vossenn J.H., Ram A.F.J., and Klis F.M. (1995). Identification of SPT14/CWH6 as the yeast homologuee of hPIG-A, a gene involved in the biosynthesis of GPI anchors. Biochim Biophys ActaActa 1243:549-551

Zlotnikk H., Fernandez M.P., Bowers B., and Cabib E. (1984). Saccharomyces cerevisiae mannoproteinss form an external cell wall layer that determines wall porosity. J Bacteriol 159:1018-1026 6

69 9

ChapterChapter 3

Localizationn of synthesis of l,6-/?-glucan in Saccharomyces cerevisiae cerevisiae

Royy C. Montijn Edwinn Vink Wallyy H. Muller Ariee J. Verkleij Hermann van den Ende Bernardd Henrissat Franss M. Klis

Thiss chapter was published with minor modifications in J. Bacteriology (1999) 181:7414-7420 0 ChapterChapter 3

SUMMARY Y

1,6-^-Glucann is a key component of the yeast cell wall, interconnecting cell wall proteins, 1,3-/?-glucan,, and chitin. It has been postulated that the synthesis of 1,6-/?-glucan begins in thee endoplasmic reticulum with the formation of protein-bound primer structures and that thesee primer structures are extended in the Golgi complex by two putative glucosyltransferasess that are functionally redundant, Kre6 and Sknl. This is followed by maturationn steps at the cell surface and by coupling to other cell wall macromolecules. We havee reinvestigated the role of Kre6 and Sknl in the biogenesis of l,6-/?-glucan. Using hydrophobicc cluster analysis, we found that Kre6 and Sknl show significant similarities to familyy 16 glycoside hydrolases but not to nucleotide diphospho-sugar glycosyltransferases, indicatingg that they are glucosyl hydrolases or transglucosylases instead of genuine glucosyltransferases.. Next, using immunogold labeling, we tried to visualize intracellular 1,6- /?-glucann in cryofixed sec 1-1 cells which had accumulated secretory vesicles at the restrictive temperature.. No intracellular labeling was observed, but the cell surface was heavily labeled. Consistentt with this, we could detect substantial amounts of 1,6-/?-glucan in isolated plasma membrane-derivedd microsomes but not in post-Golgi secretory vesicles. Taken together, our dataa indicate that the synthesis of 1,6-^-glucan takes place largely at the cell surface. An alternativee function for Kre6 and Sknl is discussed.

72 2 LocalizationLocalization of 1,6-fi-glucan synthesis

INTRODUCTION N

Thee cell wall of the yeast Saccharomyces cerevisiae consists of four classes of macromoleculess organized in the form of supramolecular complexes (Fujii et al, 1999; Kapteynn et al, 1997; Kapteyn et al, 1999a; Kapteyn et al, 1999b; Kollar et al, 1995; Kollar etet al, 1997; Lipke and Ovalle, 1998). About 40 to 50% of the cell wall is accounted for by mannoproteins,, and the other 50 to 60% is composed of 1,3-/?-glucan and l,6-/?-glucan, with aa small amount of chitin (Hartland et al, 1994; Dallies et al, 1998). Chitin and l,3-/?-glucan aree synthesized by separate enzyme complexes located in the plasma membrane (Shematek et al,al, 1980; Cabib et al, 1983; Orlean, 1997; Cabib et al, 1998), and cell wall proteins are processedd and transported to the cell surface in a stepwise process by the secretory pathway (Orlean,, 1997). The biogenesis of l,6-/?-glucan, which in its mature form consists of about 1400 glucose residues (Manners et al, 1973), is less well known. By screening for increased resistancee to Kl killer toxin, several genes (often designated KRE genes) that are required for normall cell wall levels of 1,6-/?-glucan, have been identified (Boone et al, 1990; Meaden et al,al, 1990; Roemer and Bussey, 1991; Brown and Bussey, 1993; Roemer et al, 1993; Roemer etet al, 1994; Jiang et al, 1995; Dijkgraaf et al, 1996; Simons et al, 1998; Shahinian et al, 1998;; Bickle et al, 1998). The corresponding gene products operate in the endoplasmic reticulumm (ER) {CWH41, GLS2, KRE5, and CNE1), in the Golgi apparatus (KRE6 and probablyy SKN1), or at the cell surface (KREJ, KRE9, and KNH1). This largely genetic evidencee has led to the proposal that the synthesis of 1,6-/?-glucan is a stepwise process that beginss in the ER with the synthesis of protein-bound primer structures. These primer structuress are believed to be extended by Golgi-located 1,6-/?-glucosyltransferases, encoded byy KRE6 and its functional counterpart SKN1, whereas remodeling and maturation of 1,6-/?- glucann is believed to take place at the cell surface (Boone et al, 1990; Klis, 1994; Montijn et al,al, 1994; Orlean, 1997). Wee have reinvestigated the role of Kre6 and Sknl in the biogenesis of l,6-/7-glucan. Usingg hydrophobic cluster analysis (HCA), we found unexpectedly that Kre6 and Skn 1 show significantt similarities to family 16 glycoside hydrolases (Henrissat, 1991; Coutinho and Henrissat,, 1999). This called into doubt the putative function of Kre6 and Sknl. Using advancedd immunocytochemical techniques and affinity purified antibodies raised against protein-boundd 1,6-/?-glucan oligosaccharides, we could detect only 1,6-/7-glucan at the cell surface.. In addition, post-Golgi secretory vesicles isolated by gel filtration did not contain detectablee amounts of 1,6-/?-glucan, whereas plasma membrane-derived microsomes did containn substantial amounts of l,6-/?-glucan. Together, these findings seem to exclude Kre6 andd Sknl as genuine glucosyltransferases involved in the extension in the Golgi apparatus of 1,6-/?-glucann chains. The results also indicate that the majority of 1,6-/?-glucan is synthesized att the plasma membrane.

73 3 ChapterChapter 3

MATERIALSS AND METHODS

Strainss and media. The strains used in this study were X2180A (M47a) and HMSF1 (MATu secl-1).secl-1). Cells were grown in YPD medium (1% [wt/vol] yeast extract, 1% [wt/vol] Bacto Peptonee [Difco Laboratories, Detroit Mich.], 3% [wt/vol] glucose) at 28°C.

HCA.. HCA (recently reviewed by Callebaut et al. [1997]) uses a two-dimensional (2-D) plot inn which the amino acid sequence of a protein is displayed as an unrolled and duplicated longitudinall cut of a cylinder, where the amino acids follow an a-helical pattern. The duplicationn of the helical net allows the full sequence environment of each amino acid to be represented.. On this representation, the clusters of contiguous hydrophobic residues (V, I, L, F,, M, W, Y) correspond significantly to secondary structure elements in globular proteins. Thee segmentation of a protein into successive secondary structure elements becomes visible alongg the horizontal axis of the diagram, whereas the sequence itself can be read on an almost verticall axis. The analysis then involves the comparison of cluster shape (for instance, large horizontall clusters correspond predominantly to a helices and short vertical ones correspond too (3 strands) and cluster distribution between several plots in order to find correspondences.

Cryofixation,, freeze-substitution, low-temperature embedding, and immunogold labelingg of l,6-/ï-glucan. Samples of secl-1 mutant cells were cryofixed in liquid propane by meansmeans of a Reichert-Jung KF80 apparatus and were freeze-substituted as described by Schwarzz and Humbel (1989) by placing the cryofixed samples in 0.3% uranyl acetate and 0.01%% glutaraldehyde in methanol at -90°C for 2 days. The samples were subsequently warmedd to -45°C at a rate of 5°C/h, rinsed with methanol, and infiltrated with Lowicryl HM20.. After 16 h, the specimens were transferred to an embedding mold filled with Lowicryll HM20 at -45°C. Polymerization at -45°C for 48 h was carried out in a CS auto apparatuss using a UV light source attachment (360 nm); this was followed by a 2-day curing periodd using UV light at room temperature (Sitte et al., 1985). Ultrathin HM20 sections of the yeastt cells were mounted on nickel grids and incubated with affinity-purified anti-1,6-/?- glucann polyclonal antibodies (1:300) (Klis et al, 1998). The antigen-antibody complex was visualizedd with secondary goat anti-rabbit antibodies (1:20) conjugated with 10-nm gold particless (Aurion, Wageningen, The Netherlands) (Muller et al, 1998). The labeled ultrathin sectionss were viewed in a Philips EM420 electron microscope, and micrographs were taken att an acceleration voltage of 80 kV.

Cross-linkingg and staining of carbohydrates. Samples of secl-1 mutant cells were fixed in 2%% glutaraldehyde in 0.1 M phosphate-buffered saline (PBS; pH 7.4) for 16 h at 4°C. After beingg rinsed with PBS, the yeast cells were put for 1.5 h into a mixture of 0.1 M lysine-HCl,

74 4 LocalizationLocalization of1,6-fi-glucan synthesis

0.055 M PBS, and 0.02 M sodium /w-periodate to cross-link the carbohydrates (McLean and Nakane,, 1974). After a rinse with PBS, the yeast cells were postfixed with 1% OsC>4 in 50 mMM PBS for 2 h. After dehydration in graded acetone solutions, the yeast cells were embeddedd in Epon. Ultrathin Epon sections of the yeast cells were mounted on nickel grids, immunogoldd labeled as described above, and subsequently stained with alkaline bismuth for 300 min at 37°C as described by Shinji et al. (1975). Sections were viewed in a Philips EM420 electronn microscope, and micrographs were taken at an acceleration voltage of 80 kV.

Fractionationn and characterization of microsomes. The protocols of Walworth and coworkerss (Walworth and Novick, 1987; Walworth et al, 1989) were used, secl-1 mutant cellss were grown at 25°C in rich medium containing 2% glucose and then transferred to a mediumm kept at the restrictive temperature (37°C) and containing only 0.2% glucose. This mediumm shift simultaneously imposes the secretory block, resulting in the accumulation of post-Golgii secretory vesicles within the cytosol, and derepresses the synthesis of invertase. Thee cells were converted to spheroplasts in 1.4 M sorbitol and lysed osmotically in 0.8 M sorbitol.. The latter sorbitol concentration was maintained throughout the fractionation proceduree to preserve the integrity of the organelles. The lysate was subjected to differential centrifugationn to remove unlysed cells, cell wall debris, nuclei, and mitochondria, and the microsomall fraction was fractionated by passage through a Sephacryl S-1000 gel filtration column.. Aliquots from each column fraction were assayed for protein content, plasma membranee ATPase activity, invertase activity, and 1,6-/?-glucan content. Protein was measuredd by bicinchonic acid protein assay (Pierce, Rockford, 111.) with bovine serum albuminn (BSA) as a reference protein. Vanadate-sensitive plasma membrane ATPase activity wass assayed as described elsewhere (Bowman and Slayman, 1979). Invertase activity was assayedd with sucrose as substrate, as described previously (Goldstein and Lampen, 1975), exceptt that the resulting reducing sugars were determined by the Nelson-Somogyi method (Spiro,, 1966). Distribution of l,6-/?-glucan across the eluate was determined by dot blot analysiss using affinity-purified l,6-/?-glucan antibodies (Klis et al, 1998). An aliquot of 1 u.1 fromm each column fraction was spotted on a polyvinylidene difluoride membrane and left for 300 min in a closed container. The membrane was incubated for 1 h with a blocking buffer containingg 5% nonfat milk powder in PBS. For immunodetection, the membrane was treated ass described below, and the staining intensities of the spots were measured by densitometric scanning. .

Conjugationn of gentiobiose to BSA. Gentiobiose was covalently linked to lysine residues of BSAA by reductive amination (Roy et al, 1984; Klis et al, 1998). Sodium dodecyl sulfate- polyacrylamidee gel electrophoresis was performed on a linear 2.2 - 20% gradient of polyacrylamidee (Montijn et al, 1994). For Western analysis, proteins were

75 5 ChapterChapter 3

electrophoreticallyy transferred to polyvinylidene difluoride membranes overnight at 45 V. Thee membranes were washed three times with PBS and incubated for 1 h with a blocking bufferr containing 5% nonfat milk powder in PBS. The membranes were washed three times withh PBS and then incubated for 1 h with affinity-purified anti-l,6-/?-glucan antibodies (1:5,000)) in 3% BSA in PBS. The membranes were washed five times with PBS and incubatedd with peroxidase-conjugated goat anti-rabbit antibodies (Bio-Rad Laboratories, Hercules,, Calif.). The immunoblots were developed with ECL (enhanced chemiluminescence)) Western blotting detection reagents (Amersham, Arlington Heights, 111.).

76 6 LocalizationLocalization of 1,6-ft-glucan synthesis

RESULTS S

Structurall relationships of Kre6 and Sknl with family 16 glycoside hydrolases and transglycosidases. . HCAA is based on a 2-D helical representation of protein sequences (Gaboriaud et al., 1987;; Callebaut et al, 1997). It is a powerful tool for sequence comparison at low sequence identityy levels and for the detection of secondary structure elements (a helixes, P strands, and loops).. Using this method, we could not find sequence similarities with nucleotide diphospho-sugarr glycosyltransferases (Campbell et al, 1997) but instead detected significant sequencee similarities with family 16 glycoside hydrolases and transglycosidases (Henrissat, 1991;; Coutinho and Henrissat, 1999). Figure 1 shows that throughout most of their lumenal portions,, Kre6 (and Sknl [not shown]) display significant HCA similarities with well- characterizedd family 16 members. In particular, they share a conserved motif with two glutamicc acid residues separated by either three or four amino acids (Henrissat et al., 1995). Thesee two residues are the catalytic amino acids in family 16 glycoside hydrolases. The 3-D structuree of the l,3-l,4-/?-glucanase of Bacillus macerans (Protein Data Base [PDB] entry 1BYH),, which belongs to the same family, has been resolved (Keitel et al, 1993). An interestingg feature of Kre6 (and Sknl [not shown]) is the insertion of two segments (A and B inn Fig. 2). The inserted elements are localized on two adjacent loops of the structure where theyy might form a protuberance in Kre6 without affecting the catalytic machinery. Pairwise BLASTT analysis supported the results obtained by HCA analysis. Comparison of the full- lengthh putative catalytic region of Kre6 (amino acids 321 to 660) with the clotting factor G alphaa subunit resulted in the identification of two sequences with P = 8e-04 (significant) for thee first sequence (amino acids 573 to 660) and P = 0.94 (not significant) for the other sequencee (amino acids 325 to 433). When regions A and B (Fig. 1) were left out, a single sequencee (amino acids 325 to 660) was identified with a probability of le-12 (significant). In otherr words, removal of regions A and B considerably improved the significance of the resemblance.. Similarly, comparison of the full-length catalytic region of Kre6 with the full- lengthh catalytic region of the l,3-/?-glucanase II of Oerskovia again resulted in the identificationn of two sequences with probabilities of 0.43 for the first sequence and 0.63 for thee second. Again, when regions A and B were left out of consideration, a single sequence (aminoo acids 329 to 659) was identified with a probability of 2e-9 (significant). Comparison off the full-length catalytic regions of Kre6 and l,3-l,4-/?-glucanase from B. macerans resultedd in a nonsignificant P value even when both regions A and B were left out. However, comparisonn between the full catalytic region of l,3-l,4-/?-glucanase from B. macerans with thee full catalytic region of l,3-/?-glucanase II of Oerskovia, which as discussed above shows significantt similarity with the Kre6 catalytic domain, resulted in a P value of 8e-4 (significant). .

77 7 ChapterChapter 3

„-.-w^po-- Figuree 1. HCA plots of selected members of family 16 glycoside hydrolases and transglycosidases (Henrissat,

78 8 LocalizationLocalization of 1,6-B-glucan synthesis

1991;; Henrissat et ai, 1995). From top to bottom: Kre6 of Saccharomyces cerevisiae (SwissProt P32486), clottingg factor G alpha subunit of Tachypleus tridentatus (GenBank D16622), l,3-/?-glucanase II of Oerskovia xanthineolyticaxanthineolytica (GenBank AF052745), and l,3-l,4-/?-glucanase of Bacillus macerans (SwissProt P23904; PDB 1BYH).. The HCA plots were made, edited, and analyzed as described elsewhere (Gaboriaud et ai, 1987; Henrissat,, 1991). To facilitate visual inspection of the plots, the symbols *, , 0, and D are used for proline, glycine,, serine, and threonine, respectively. Vertical lines show correspondences between proteins. The two catalyticc glutamate residues are shown in white on black circles. The secondary structure elements of 1BYH are shownn as open (p strand) and grey (a helix) boxes under the corresponding plot. The two insertions found in Kre66 are marked A and B.

Inn summary, HCA and pairwise BLAST analyses point to a clear resemblance of Kre6 andd Sknl with glycoside hydrolases of family 16. This is difficult to reconcile with their postulatedd function as nucleotide sugar glucosyltransferases, which use activated sugars as a substrate,, and suggests instead that Kre6 and Sknl have either glucosidase or transglucosidasee activity.

Figuree 2. Schematic 3-D structure of the l,3-l,4-/ï-glucanase of B. macerans (PDB 1BYH). The two glutamate residuess that belong to the catalytic site are shown in ball-and-stick representation. The two loops which carry thee insertions in Kre6 are labeled A and B. The location of the resulting putative protuberance in Kre6 is shown byy dashed lines. The figure was prepared with the program MOLSCRIPT (Kraulis, 1991).

Detectionn of l,6-/?-glucan at the cell surface by immunogold labeling. Ass the results of the HCA analysis seemed to be inconsistent with the proposed functionss of Kre6 and Sknl as nucleotide sugar glucosyltransferases responsible for elongatingg l,6-/?-glucan chains in the Golgi apparatus, we decided to look for intracellular 1,6-/?-glucann immunocytochemically. The 1,6-/?-glucan antiserum that we used was raised

79 9 ChapterChapter 3

againstt 1,6-/?- glucan oligosaccharides with an average chain length of 15 glucose residues coupledd to BSA (Montijn et ai, 1994). The specificity of the 1,6-/?-glucan antiserum has been confirmedd in various ways. Recognition of the epitope is competitively inhibited by pustulan (l,6-/?-glucan)) but not by laminarin (1,3-/?-glucan) or mannan (Montijn et ah, 1994). In addition,, periodate treatment of the 1,6-/?-glucan epitope completely abolishes the signal (Montijnn et ai, 1994). The anti-1,6-/?- glucan antiserum has been further purified by affinity chromatographyy using 1,6-/?-glucan oligosaccharides immobilized on an epoxy-Sepharose columnn (Klis et ai, 1998). This raises the question of whether the affinity-purified anti-1,6-/?- glucann antiserum also recognizes short 1,6-/?-glucan oligosaccharides. To answer this question,, we analyzed the effectiveness of the antiserum toward protein-bound gentiobiose (Glc-l,6-/?-Glc).. Figure 3 shows that the antiserum efficiently bound to gentiobiose coupled too BSA, whereas it had no activity toward BSA itself (Fig. 3). This shows that our affinity- purifiedd antiserum is capable of recognizing short 1,6-/?-glucan oligosaccharides.

11 2 3

2066 — 1255 —

488 — 355 —

Figuree 3. Pustulan affinity-purified antibodies recognize protein-bound gentiobiose. Lane 1, 1 |ag of BSA; lane 2,, 1 ng of gentiobiose-BSA; lane 3, 5 ng of gentiobiose-BSA. The blot was developed with ECL for 1 min. Sizess are indicated in kilodaltons.

Too study subcellular localization, yeast cells were first cryofixed and freeze- substituted.. Ultrathin Lowicryl sections were immunogold labeled with affinity-purified 1,6- /?-glucann antibodies. In wild-type cells, the cell surface became clearly labeled, but no intracellularr labeling was observed (data not shown). Since in wild-type cells only a few secretoryy vesicles are seen and intracellular 1,6-/?-glucan might for that reason have been overlooked,, labeling experiments were also performed on sec 1-1 cells (Novick et ai, 1980), keptt at the restrictive temperature for 2 h. This is roughly equivalent to one generation time, implyingg that collectively the secretory vesicles are expected to contain sufficient cell wall precursorr material to build an entire cell wall. In the section shown, about 360 gold particles aree visible at the cell surface, whereas intracellular labeling is negligible (Fig. 4A). This is

80 0 LocalizationLocalization of 1,6-fS-glucan synthesis

difficultt to reconcile with the notion that the bulk of 1,6-yS-glucan is synthesized intracellularly.. To exclude the possibility that 1,6-/?- glucan had leaked out of the secretory vesicless during the processing steps prior to electron microscopy, in the next experiment we introducedd a cross-linking step, which results in the formation of aggregates of indiscriminatelyy cross-linked carbohydrates and proteins (McLean and Nakane, 1974). This wass followed by specific staining of the carbohydrate cargo of the vesicles (Shinji et al, 1975)) (Fig. 4B). Although the vesicles were heavily stained under these conditions, indicatingg that possible losses of their contents were limited, still no immunogold labeling of thee vesicles was observed. In summary, our data are consistent with the notion that the bulk off 1,6-yS-glucan synthesis takes place at the plasma membrane.

Figuree 4. Immunogold labeling of 1,6-/i-glucan in secl-1 cells. (A) To induce accumulation of post-Golgi secretorysecretory vesicles, secl-1 cells were kept at the restrictive temperature for 2 h. A representative cell is shown. Aboutt 360 gold particles are visible at the cell surface, whereas intracellular labeling is negligible. (B) The vesicless were visualized by cross-linking and staining of the carbohydrate cargo by the methods of McLean and

81 1 ChapterChapter 3

Nakanee (1974) and Shinji et al. (1975), respectively. Bar = 250 nm. Buds at various stages of the cell cycle are shownn (a), (b), and (c).

1,6-/F-Glucann is absent from post-Golgi secretory vesicles. Walworthh and coworkers have developed an efficient and well-documented method to isolatee intact post-Golgi secretory vesicles that are largely free from contaminating organelles includingg ER-, plasma membrane-, and vacuolar membrane-derived microsomes. An additionall advantage of their method is that it results in a considerable purification of plasma membrane-derivedd microsomes which are relatively free from post-Golgi secretory vesicles andd vacuolar membrane-derived microsomes (Walworth and Novick, 1987; Walworth et al, 1989).. Their method makes use of a late secretory mutant which is allowed to accumulate secretoryy vesicles at the restrictive temperature. The microsomal fraction obtained by differentiall centrifugation of osmotically lysed spheroplasts is further fractionated by gel filtrationn in the presence of stabilizing concentrations of sorbitol, resulting in a clear separationn of plasma membrane-derived microsomes and post-Golgi secretory vesicles. Using thiss approach, we analyzed the eluate for the presence of protein, invertase activity, plasma membranee ATPase activity, and 1,6-/?-glucan. As shown in Fig. 5A, protein eluted as three majorr peaks. The first protein peak coeluted with the first peak of the plasma membrane marker,, vanadate-sensitive ATPase activity, and represented the plasma membrane-derived microsomess (Fig. 5C) (Walworth and Novick, 1987; Walworth et al, 1989). The second proteinn peak cofractionated with the major peak of invertase activity, which marks post-Golgi secretoryy vesicles (Fig. 5B). The second ATPase peak, which coeluted with the major peak of invertasee activity, probably represents ATPase transported by secretory vesicles (Walworth andd Novick, 1987; Walworth et al, 1989). The last protein peak cofractionated with an invertasee peak. This probably represents material escaped from leaky secretory vesicles. Finally,, the column fractions were analyzed by a dot blot assay using affinity-purified 1,6-/?- glucann antibodies (Fig. 5D). Although some signal (23%) was detected in the fractions correspondingg to the post-Golgi secretory vesicles, most of the signal (77%) was found in the fractionsfractions containing plasma membrane-derived vesicles, suggesting the existence of a 1,6-0- glucan-synthesizingg protein (complex) associated with the plasma membrane. The presence off a weak positive signal in the fractions containing post-Golgi secretory vesicles might be duee to contamination with plasma membrane-derived vesicles. However, we cannot exclude thee possibility that secretory vesicles contain some l,6-/?-glucan that cannot be detected by immunogoldd labeling.

82 2 LocalizationLocalization of 1,6-fS-glucan synthesis

30 0 fractionn number

Figuree 5. 1,6-/7-Glucan colocalizes with plasma membrane-derived vesicles. To induce accumulation of post- Golgii secretory vesicles, secl-1 cells were kept at the restrictive temperature for 2 h. The cells were spheroplastedd and gently lysed. The resulting homogenate was fractionated by differential centrifugation, and thee microsomal fraction was separated by gel filtration on a Sephacry 1 S-1000 column (Walworth and Novick, 1987;; Walworth et al., 1989). Aliquots of each column fraction (x axis) were assayed for protein content, invertase,, plasma membrane ATPase, and 1,6-/?-glucan content, v axes represent, from top to bottom, protein (micrograms),, invertase (micromoles of Glc per minute), plasma membrane ATPase (micromoles of P, per minute),, and 1,6-/?-glucan (arbitrary densitometric units).

83 3 ChapterChapter 3

DISCUSSION N

KREKRE and A7?£-related genes have been isolated as genes that are required for full levelss of cell wall 1,6-/?-glucan and which, when nonfunctional, confer resistance to Kl killer toxinn (reviewed in Orlean, 1997). As several of the corresponding proteins have been localizedd in the ER (Kre5 and Cwh41), the Golgi complex (Kre6 and possibly also its homologg Sknl), and at the cell surface (Krel and Kre9), it has been proposed that they might bee involved in sequential steps of the biosynthesis of 1,6-/7-glucan (Boone et al, 1990; Klis et al,al, 1994; Roemer et al, 1994; Orlean, 1997). Kre6 and Sknl are predicted to have a single amino-terminall transmembrane domain and a long carboxy-terminal lumenal domain. It has beenn proposed that Kre6 and Sknl are Golgi-located glucosyltransferases that elongate a protein-boundd 1,6-/?-glucan primer structure formed in the ER (Roemer et al, 1994; Orlean, 1997).. Here we present evidence that Kre6 and Sknl are not genuine glucosyltransferases and thatt the synthesis of 1,6-/?-glucan takes largely place at the plasma membrane. First, homologyy searches based on hydrophobic cluster analysis show that Kre6 and Sknl have the hallmarkss of glycoside hydrolases or transglycosidases but not of nucleotide diphospho-sugar glycosyltransferasess (Henrissat, 1991; Henrissat et al, 1995; Barbeyron et al, 1998). Second, wee were unable to detect any intracellular 1,6-/?-glucosylated proteins, neither in wild-type cellss nor in seclH, sec7, and seel cells kept at the restrictive temperature to accumulate ER, Golgi-likee structures, and post-Golgi secretory vesicles, respectively (R.C. Montijn, E. Vink, andd F.M. Klis, unpublished results). As our antibodies efficiently recognize 1,6-/?- glucosylatedd cell wall proteins in yeast (Montijn et al, 1994), intracellular l,6-/?-glucan in the myceliall fungus Trichosporum sporotrichoides (Muller et al, 1998), and even protein-bound gentiobiosee (this report), extensive 1,6-/?-glucosylation of intracellular proteins in yeast seems unlikely.. Third, immunogold labeling of post-Golgi secretory vesicles in cryofixed cells gave negativee results even after an additional cross-linking step to avoid potential losses of the vesiclee contents during the processing steps prior to electron microscopy, whereas a strong signall was seen all over the cell surface, showing that our antiserum efficiently recognizes 1,6-/?-glucan.. An alternative explanation of this result is that in contrast to wild-type cells, seel-Iseel-I cells immediately halt the production of 1,6-/?-glucan when transferred to 37°C. However,, it is known that other cell surface components like plasma membrane ATPase, invertase,, and acid phosphatase continue to be synthesized at this temperature (Walworth and Novick,, 1987; Walworth et al, 1989). Fourth, dot blot analysis of membrane vesicles fractionatedd by gel filtration revealed only a small amount of l,6-/?-glucan in the fractions containingg post-Golgi secretory vesicles, possibly due to contamination with plasma membrane-derivedd vesicles. However, in the fractions that contained plasma membrane- derivedd vesicles, substantial amounts of l,6-/?-glucan were present. As post-Golgi secretory vesicless are destined to become part of the plasma membrane, these data also suggest that the

84 4 LocalizationLocalization of 1,6-p-glucan synthesis

plasmaa membrane contains not only an activatable l,3-/?-glucan synthase but also an activatablee 1,6-/?-glucan synthase. Stilll unanswered is the question of how the loss of function of Kre proteins in the secretoryy pathway, including Kre6 and Sknl, could lead to a reduction in cell wall 1,6-/?- glucan.. One possibility is that the postulated plasma membrane-associated 1,6-/?-glucan synthasee complex is for unknown reasons extremely sensitive to defects in glycosylation. Thiss seems less likely because severe defects in TV-glycosylation as observed in mnn9A and ochlAochlA cells do not result in decreased levels of 1,6-/?-glucan in the cell wall (Shahinian et al, 1998).. Alternatively, Kre proteins in the secretory pathway may contribute to the construction off glucose-containing protein-bound carbohydrate structures, which may act as acceptor sites forr the addition of 1,6-/?-glucan at the cell surface. For example, Kre6 and Sknl could act as transglucosidasess on a protein-bound glucan structure formed in the ER by Kre5. The nature off the postulated acceptor structures is unknown and could include modified glycosylphosphatidylinositoll (GPI) anchors, TV-chains, and O-chains, but not necessarily on thee same proteins. This is consistent with earlier observations by Van Rinsum and coworkers (1991)) (Fig. 4), who provided evidence for the presence of three different types of glucose- containingg carbohydrate side chains in cell wall proteins, possibly corresponding with extendedd TV-chains, O-chains, and GPI anchors. Indeed, chemical analysis has revealed a directt linkage between a processed GPI anchor and 1,6-/?-glucan (Kollar et al, 1997; Fujii et al,al, 1999). Recently, a genetic analysis of ER-located Kre proteins has provided evidence that TV-chainss may also be involved as an alternative attachment site for 1,6-/?-glucan (Shahinian etet al, 1998). Finally, mutants defective in the first steps of O-glycosylation show partial resistancee to Kl killer toxin (Strahl-Bolsinger et al, 1993; Lussier et al, 1996; Gentzsch and Tanner,, 1996), consistent with the notion that in some cases also O-chains may function as attachmentt sites for l,6-/?-glucan. In summary, we propose that the incorporation of 1,6-/?- glucann into the cell wall requires three critical steps: (i) the construction of glucose- containingg protein-bound acceptor sites by Kre proteins in the early compartments of the secretoryy pathway for the later addition of 1,6-/?-glucan; (ii) the extension of these primer structuress with 1,6-/?-glucan at the plasma membrane; (iii) the addition of 1,6-/?-glucan to cell walll proteins that have newly arrived at the cell surface, as has been described for a- agglutininn and Tipl (Lu et al, 1995; Fujii et al., 1999).

ACKNOWLEDGMENTS S Wee thank Pedro M. Coutinho for help with the preparation of Fig. 2, Howard Bussey and Terryy Roemer for supplying the kre mutants, and Alfred Van Kuik for helpful discussions. Wee also thank Annemiek Andel, Sylvia Blad, and Piet De Groot for preparing and analyzing thee gentiobiose-BSA neoglycoprotein and Hans de Nobel for critical reading of the manuscript.. This work received support from the EU program EUROFAN II.

85 5 ChapterChapter 3

REFERENCES S

Barbeyronn T., Gerard A., Potin P. , Henrissat B., and Kloareg B. (1998). The kappa- carrageenasee of the marine bacterium Cytophaga drobachiensis. Structural and phylogenetic relationshipss within family-16 glycoside hydrolases. Mol. Biol. Evol. 15:528-537

Bicklee M., Delley P.A., Schmidt A., and Hall M.N. (1998). Cell wall integrity modulates RHOlRHOl activity via the exchange factor R0M2. EMBOJ. 17:2235-2245

Boonee C, Sommer S.S., Hensel A., and Bussey H. (1990). Yeast KRE genes provide evidencee for a pathway of cell wall p-glucan assembly. J. Cell Biol. 110:1833-1843

Bowmann B.J., and Slayman C.W. (1979). The effect of vanadate on the plasma membrane ATPasee of Neurospora crassa. J. Biol. Chem. 254:2928-2934

Brownn J.L., and Bussey H. (1993). The yeast KRE9 gene encodes an O-glycoprotein involvedd in cell surface P-glucan assembly. Mol. Cell. Biol. 13:6346-6356

Brownn J.L., Kossaczka Z., Jiang B., and Bussey H. (1993). A mutational analysis of killer toxinn resistance in Saccharomyces cerevisiae identifies new genes involved in cell wall (1—^6)-beta-glucann synthesis. Genetics 133:837-849

Cabibb E., Bowers B., and Roberts R.L. (1983). Vectorial synthesis of a polysaccharide by isolatedd plasma membranes. Proc. Proc. Natl. Acad. Sci. USA 80:3318-3321

Cabibb E., Drgonova J., and Drgon T. (1998). Role of small G proteins in yeast cell polarizationn and wall biosynthesis. Annu. Rev. Biochem. 67:307-333

Callebautt I., Labesse G., Durand P., Poupon A., Canard L., Chomilier J., Henrissat B., andd Mornon J.-P. (1997). Deciphering protein sequence formation through hydrophobic clusterr analysis (HCA): current status and perspectives. Cell. Mol. Life Sci. 53:621-645

CampbellCampbell J.A., Davies G.J., Bulone V., and Henrissat B. (1997). A classification of nucleotide-diphospho-sugarr glycosyltransferases based on amino acid sequence similarities. Biochem.Biochem. J. 326:929-939

86 6 LocalizationLocalization of 1,6-fi-glucan synthesis

Coutinhoo P.M. and Henrissat B. 27 August 1999, revision date. [Online.] Carbohydrate- activee enzymes server. http://afmb.cnrs-mrs.fr/~pedro/CAZY/db.html. [22 October 1999, last datee accessed]

Dalliess N., Francois J., and Paquet V. (1998). A new method for quantitative determination off polysaccharides in the yeast cell wall. Application to the cell wall defective mutants of SaccharomycesSaccharomyces cerevisiae. Yeast 14:1297-1306

Dijkgraaff G.J., Brown J.L., and Bussey H. (1996). The KNH1 gene of Saccharomyces cerevisiaecerevisiae is a functional homolog OÏKRE9. Yeast 15:683-692

Fujiii T., Shimoi H., and Iimura Y. (1999). Structure of the glucan-binding sugar chain of Tiplp,, a cell wall protein of Saccharomyces cerevisiae. Biochem. Biophys. Acta 1427:133— 144 4

Gaboriaudd C, Bissery V., Benchetrit T., and Mornon J.-P. (1987). Hydrophobic cluster analysis:: an efficient new way to compare and analyse amino acid sequences. FEBS Lett. 224:149-155 5

Gentzschh M., and Tanner W. (1996). The PMT gene family: protein O-glycosylation in SaccharomycesSaccharomyces cerevisiae is vital. EMBOJ. 15:5752-5759

Goldsteinn A., and Lampen J.O. (1975). P-D-Fructofuranoside fructohydrolasefrom yeast. MethodsMethods Enzymol. 42:504-511

Hartlandd R.P., Vermeulen CA., Klis F.M., Sietsma J.H, and Wessels J.G. (1994). The linkagee of (l-3)-beta-glucan to chitin during cell wall assembly in Saccharomyces cerevisiae. YeastYeast W.\59\-\599

Henrissatt B. (1991). A classification of glycosyl hydrolases based on amino acid sequence similarities.. Biochem. J. 280:309-316

Henrissatt B., Callebaut I., Fabrega S., Lehn P., Mornon J.-P., and Davies G. (1995). Conservedd catalytic machinery and prediction of a common fold for several families of glycosyll hydrolases. Proc. Natl. Acad. Sci. USA 92:7090-7094

87 7 ChapterChapter 3

Jiangg B., Sheraton J., Ram A.F.J., Dijkgraaf G.J.P., Klis F.M., and Bussey H. (1995). CHW41CHW41 encodes a novel endoplasmic reticulum membrane N-giycoprotein involved in 01,6- glucann assembly. J. Bacteriol. 178:1162-1171

Kapteynn J.C., Ram A.F.J., Groos E.M., Kollar R., Montijn R.C., Van Den Ende H., Llobelll A., Cabib E., and Klis F.M. (1997). Altered extent of crosslinking of bl,6- glucosylatedd mannoproteins to chitin in Saccharomyces cerevisiae mutants with reduced cell walll pl,3-glucan content. J. Bacteriol. 179:6279-6284

Kapteynn J.C., De Nobel J.G., and Klis F.M. (1999). The contribution of cell wall proteins too the organization of the yeast cell wall. Biochim. Biophys. Acta 1426:373-383

Kapteynn J.C., Van Egmond P., Sievi E., Van den Ende H., Makarow M., and Klis F.M. (1999).. The contribution of the O-glycosylated protein Pir2p/Hspl50 to the construction of thee yeast cell wall in wild-type cells and pl,6-glucan-deficient mutants. Mol. Microbiol. 31:1835-1844 4

Keitell T., Simon O., Borriss R., and Heinemann U. (1993). Molecular and active-site structuree of a Bacillus 1,3-1,4-P-glucanase. Proc. Natl. Acad. Sci. USA 90:5287-5291

Kliss F.M. (1994). Review: cell wall assembly in yeast. Yeast 10:851-869

Kliss F.M., Ram A.F.J., Montijn R.C., Kapteyn J.C., Caro L.H.P., Vossen J.H., Van Berkell M.A.A., Brekelmans S.S.C., and Van Den Ende H. (1998). Posttranslational modificationss of secretory proteins. Methods Microbiol. 26:233-238

Kollarr R., Petrakova E., Ashwell G., Robbins P.W., and Cabib E. (1995). Architecture of thee yeast cell wall. The linkage between chitin and beta(l—»3)glucan. J. Biol. Chem. 270:1-9

Kollarr R., Reinhold B.B., Petrakova E., Yeh H.J., Ashwell G., Drgonova J., Kapteyn J.C.,, Klis F.M., and Cabib E. (1997). Architecture of the yeast cell wall - beta(l-^6)-glucan interconnectss mannoprotein, beta(l—>-3)-glucan, and chitin. J. Biol. Chem. 272:17762-17775

Kraulis,, P. J. (1991). MOLSCRIPT: a program to produce both detailed and schematic plots off protein structures. J. Appl Crystallogr. 24:946-950

Lipkee P.N., and Ovalle R. (1998). Cell wall architecture in yeast: new structure and new challenges.. J. Bacteriol. 180:3735-3740

88 8 LocalizationLocalization of 1,6~fi-glucan synthesis

Luu C.F., Montijn R.C., Brown J.L., Klis F.M., Kurjan J., Bussey H., and Lipke P.N. (1995).. Glycosyl phosphatidylinositol-dependent cross-linking of a-agglutinin and pi,6- glucann in the Saccharomyces cerevisiae cell wall. J. Cell Biol. 128:333-340

Lussierr M., Sdicu A.M., Camirand A., and Bussey H. (1996). Functional characterization off the YURI, KTR1, and KTR2 genes as members of the yeast KRE2/MNT1 mannosyltransferasee gene family. J. Biol. Chem. 271:11001-11008

Mannerss D.J., Masson A.J., Patterson J.C., Bjorndal H., and Lindberg B. (1973). The structuree of a (3-(l-6)-D-glucan from yeast cell walls. Biochem. J. 135:31-36

McLeann I.W., and Nakane P.K. (1974). Periodate-lysine-paraformaldehyde fixative. A new fixativee for immunoelectron microscopy. J. Histochem. Cytochem. 22:1077-1083

Meadenn P., Hill K., Wagner J., Slipetz D., Sommer S.S., and Bussey H. (1990). The yeast KRE5KRE5 gene encodes a probable endoplasmic reticulum protein required for 1,6-p-D-glucan synthesiss and normal cell growth. Mol. Cell. Biol. 10:3013-3019

Montijnn R.C., Van Rinsum J., Van Schagen F.A., and Klis F.M. (1994). Glucomannoproteinss in the cell wall of Saccharomyces cerevisiae contain a novel type of carbohydratee side chain. J. Biol. Chem. 269:19338-19342

Mullerr W.H., Van der Krift T.P., Knoll G., Smaal E.B., and Verkleij A.J. (1991). A preparationn method of specimens of the fungus Penicillium chrysogenum for ultrastructural andd immuno-electron microscopical studies. J. Microsc. 169:29^1

Mullerr W.H., Montijn R.C., Humbel B.M., Van Aelst A.C., Boon E.J., Van der Krift T.P.,, and Boekhout T. (1998). Structural differences between two types of basidiomycete septall pore caps. Microbiology 144:1721-1730

Novickk P., Field C, and Schekman R. (1980). Identification of 23 complementation groups requiredd for post-translational events in the yeast secretory pathway. Cell 21:205-215

Orleann P. (1997). Biogenesis of yeast wall and surface components. In: Molecular and Cellularr Biology of the Yeast Saccharomyces, Vol. 3, Cell Cycle and Cell Biology. Pringle J.R.,, Broach J.R., and Jones E.W. (eds). Cold Spring Harbor, NY: Cold Spring Harbor Laboratoryy Press, pp. 229-362

89 9 ChapterChapter 3

Roemerr T., and Bussey H. (1991). Yeast |3-glucan synthesis: Kre6p encodes a predicted typee II membrane protein required for glucan synthesis in vivo and for glucan synthase activityy in vitro. Proc. Natl Acad. Sci. USA 88:11295-11299

Roemerr T., Delaney S., and Bussey H. (1993). SKNl and KRE6 define a pair of functional homologuess encoding putative membrane proteins involved in (5-glucan synthesis. Mol Cell Biol.Biol. 13:4039^*048

Roemerr T., Paravicini G., Payton M.A., and Bussey H. (1994). Characterization of the yeastt pi,6-glucan biosynthetic components, Kre6p and Sknlp, and genetic interaction betweenn the PKC1 pathway and the extracellular matrix. J. Cell Biol. 127:567-579

Royy R., Katzenellenbogen E., and Jennings H.J. (1984). Improved procedures for the conjugationn of oligosaccharides to protein by reductive amination. Can. J. Biochem. Cell Biol.Biol. 62:270-275

Schwarzz H., and Humbel B.M. (1989). Influence of fixatives and embedding media on immunolabellingg of freeze-substituted cells. Scanning Microsc. Suppl. 3:57-63

Shahiniann S., Dijkgraaf G.J., Sdicu A.M., Thomas D.Y., Jakob C.A., Aebi M., and Busseyy H. (1998). Involvement of protein N-glycosyl chain glucosylation and processing in thee biosynthesis of cell wall P-l-6-glucan of Saccharomyces cerevisiae. Genetics 149:843- 856 6

Shematekk E.M., Braatz J.A., and Cabib E. (1980). Biosynthesis of the yeast cell wall. I. Preparationn and properties of P-(l-3)glucan synthetase. J. Biol Chem. 255:888-894

Shinjii Y., Shinji E., and Mizuhira V. (1975). A new electron microscopic histo- cytochemicall staining method: demonstration of glycogen particles. Acta Histochem. Cytochem.Cytochem. 8:139-146

Simonss J.F., Ebersold M., and Helenius A. (1998). Cell wall 1,6-beta-glucan synthesis in SaccharomycesSaccharomyces cerevisiae depends on ER glucosidases I and II, and the molecular chaperone BiP/Kar2p.. EMBOJ. 17:396^05

Sittee H., Neumann K., and Edelmann L. (1985). Cryofixation and cryosubstitution for routinee work in transmission electron microscopy. In: Science of biological specimen

90 0 LocalizationLocalization of 1,6-fi-glucan synthesis

preparation.. Muller M., Becker R.P., Boyde A., and Wolosewick J.J. (eds.). SEM Inc., AMF O'Hare,, Chicago, 111., pp. 103-118

Spiroo R.G. (1966). Analysis of sugars found in glycoproteins. Methods Enzymol. 8:3-26

Strahl-Bolsingerr S., Immervoll T., Deutzmann R., and Tanner W. (1993). PMT1, the genee for a key enzyme of protein 0-glycosylation in Saccharomyces cerevisiae. Proc. Natl. Acad.Acad. Sci. USA 90:8164-8168

Vann Rinsum J., Klis F.M., and Van den Ende H. (1991). Cell wall glucomannoproteins of SaccharomycesSaccharomyces cerevisiae mmn9. Yeast 7:717-726

Walworthh N.C., and Novick P.J. (1987). Purification and characterization of constitutive secretoryy vesicles from yeast. J. Cell Biol. 105:163-174

Walworthh N.C., Goud B., Ruohola H., and Novick P.J. (1989). Fractionation of yeast organelles.. Methods Cell Biol. 31:335-356

91 1

ChapterChapter 4

InIn vitro measurement of Saccharomyces cerevisiae l,6-/?-glucann synthase activity

Edwinn Vink Juann Carlos Ribas Hanss de Nobel Hermann van den Ende Angell Duran Franss M. Klis ChapterChapter 4

SUMMARY Y

1,6-/7-Glucann is a key component of the cell wall of Saccharomyces cerevisiae and Candida albicans.albicans. Several genes have been cloned which affect the levels or structure of l,6-/7-glucan inn the cell wall, but their precise role in the formation of mature 1,6-/?-glucan is ill understood.. Research in this area has been particularly hampered by the lack of a suitable methodd to measure 1,6-/?-glucan synthase activity in vitro. Here, non-radioactive conditions forr the detection of in vitro synthesis of l,6-/?-glucan are described. Crude membrane preparationss from S. cerevisiae were isolated, and incubated in the presence of UDP-glucose andd GTP. Using antibodies directed against 1,6-/?-glucan, the increase of the amount 1,6-/?- glucann in time could be visualized either qualitatively by using a dot blot assay, or quantitativelyy by using an inhibition enzyme immunoassay. The specificity of the antibodies wass validated by competition experiments, using pustulan (a 1,6-/?-glucan) and other polysaccharidess such as laminarin (a l,3-/?-glucan) and yeast mannan. The identity of the reactionn product was also confirmed by its sensitivity to a recombinant 1,6-/?-glucanase. The optimall pH in MES buffer for the enzyme was 6.5. This approach may allow the identificationn of the gene product(s) that is/are responsible for l,6-/?-glucan synthesis and its regulation,, and the development of new antimycotics based on specific inhibition of 1,6-/?- glucann synthesis.

94 4 InIn vitro I,6-fi-gIucan synthase activity

INTRODUCTION N

Thee cell wall of S. cerevisiae and C. albicans consists of four different components, whichh are kept together by covalent linkages (Kollar et al, 1997; Kapteyn et ai, 1999; Klis et al,al, 2001). The outer layer of the cell wall consists of heavily glycosylated mannoproteins, mostt of which are interconnected to the l,3-/?-glucan network through a 1,6-^-glucan moiety. 1,6-/?-Glucann thus plays a central role in the molecular organization of the yeast cell wall. In S.S. cerevisiae, the mature form of l,6-/?-glucan is heavily branched and consists of about 130 glucosee residues (Orlean, 1997). In C. albicans and some other pathogenic fungi l,6-/?-glucan synthesiss has been shown critical for survival (James et al, 1990; Chaffin et al, 1998). In addition,, the cell wall contains minor amounts of chitin (Orlean, 1997; Klis et al, 2002). Thee synthesis of the various cell wall components in S. cerevisiae takes place at differentt cellular locations. Mannoproteins are synthesized and processed throughout the secretoryy pathway (Orlean, 1997). l,3-/?-Glucan and chitin, on the other hand, are synthesized att the cell surface. The putative l,3-/?-glucan synthase catalytic subunits, the homologues Fkslpp and Fks2p, are multispanning transmembrane proteins located at the plasma membrane.. The activity of the l,3-/?-glucan synthase complex is regulated by the intracellular GTP-bindingg protein Rholp (Cabib et al, 1998). The three chitin synthases are also predominantlyy found at the plasma membrane, while minor amounts are present in specializedd intracellular vesicles known as chitosomes (Cabib et al, 1996; Chuang and Schekman,, 1996). Biochemical assays have been devised for measuring the in vitro activity off both l,3-y?-glucan synthase (Cabib and Kang, 1987; Shedletzky et al, 1997) and chitin synthasee (Cabib et al, 1987). Muchh less is known about the synthesis of 1,6-/?-glucan. Bussey and co-workers used aa genetic approach to study the synthesis of 1,6-/?-glucan by exploiting the Kl killer toxin. Sincee 1,6-/?-glucan serves as a receptor for this toxin, several gene products that affect 1,6-fi- glucann levels were identified by their effect on the sensitivity to the toxin (reviewed in Shahiniann and Bussey, 2000). Interestingly,, many of the corresponding gene products were found to be localized throughoutt the secretory pathway. This led to the hypothesis that the synthesis of 1,6-/?- glucann begins in the endoplasmic reticulum, that the product is extended in the Golgi, and thatt the final processing steps take place at the cell surface (reviewed in Shahinian and Bussey,, 2000). However, we found no evidence for the presence of intracellular l,6-/?-glucan, andd concluded that its synthesis predominantly takes place at the plasma membrane (Montijn etet a/., 1999, Chapter 3), Further progress in this area would strongly benefit from the developmentt of suitable in vitro assays for the various proteins that have been shown to be directlyy or indirectly involved in the biogenesis of 1,6-/?-glucan. In particular, an in vitro methodd for measuring 1,6-/?-glucan synthase activity directly is urgently needed. Here we

95 5 ChapterChapter 4

describee two non-radioactive, immunological methods to meet this requirement. Using 1,6-fi- glucann antibodies, we developed (I) a simple, qualitative assay based on dot blot analysis, and (II)) an inhibition enzyme immunoassay that allows quantification of the amount of product formed. .

96 6 InIn vitro 1,6-fi-glucan synthase activity

MATERIALSS AND METHODS

Yeastt strain and medium. For this study, yeast strain FY834 (MATa his3A200 ura3-52 leu2Alleu2Al lys2A202 trp!A63) was used (Winston et al, 1995). This strain was grown in YEPD (1%% [w/v] yeast extract, 2% [w/v] Bacto Peptone, and 2% [w/v] glucose) at 30°C. For solid media,, 2% [w/v] Bacto Agar was added. Yeast extract, Bacto Peptone, and Bacto Agar were purchasedd from Difco Laboratories, Detroit Mich. (USA)

Reagents.. Preparation and characterization of anti- 1,6-/?-glucan antibodies used in this study wass described by Montijn et al. (1994). Pustulan was purchased from Calbiochem, San Diego California.. Laminarin was purchased from Fluka. GTP, UDP-glucose, MES, yeast mannan, andd BSA were from Sigma. Zymolyase 100T was purchased from Seikagaku, Tokyo.

Membranee preparation and /?-glucan synthase conditions. The isolation of membranes andd subsequent l,3-/?-glucan synthase assays were carried out as described before (Ishiguro et al,al, 1997). For 1,6-/?-glucan synthase assays, membrane extracts were prepared as described beforee (Ishiguro et al., 1997), with minor modifications. In short, yeast cells were grown to earlyy logarithmic phase, collected, and washed in 50 mM Tris-Cl, pH 7.5. Cells were resuspendedd in the same buffer, and subsequently broken with glass beads in a FastPrep FP1200 instrument (Qbiogene). After collection of the broken material, cell debris was removedd by low speed centrifugation (3,000 xg, 5 min at 4°C). The supernatant was centrifugedd at 36,600 xg for 30 min at 4°C, and the pellet was resuspended in 50 mM Tris-Cl, pHH 7.5, with 33% (v/v) glycerol and stored at -70°C. Protein content of the membrane extract wass measured using the Bradford protein assay (Bio-Rad), using BSA as a standard (Bradford,, 1976). The 40 ul assay mixture consisted of 25 mM UDP-D-glucose, 150 \\M GTP,, 2.1 mM EDTA, 0.75% (w/v) BSA, 4.1% (v/v) glycerol, and 100 mM MES at pH 6.5 unlesss stated otherwise. For pH optimization, 75 mM NaAc at pH 4.0, 5.0, and 6.0, 25 mM sodiumm phosphate at pH 6.0, 7.0, and 8.0, 25 mM MES pH 5.5, 5.75, 6.0, 6.25, and 6.5, 50 mMM MES at pH 6.2, 6.3, 6.4, 6.5, 6.6, and 6.7 were tested. To terminate the reaction, samples weree heated for 15 min at 75°C for detection with dot blot assay. In case of detection by inhibitionn Enzyme ImmunoAssay (EIA), ethanol was added to a final concentration of > 90% (v/v).. Samples were incubated on ice for at least 2h, and subsequently centrifuged at 15,000 x gg for 15 min at 4°C. Pellets were suspended in 220 ul PBS, and then 110 ul was taken and dilutedd 1:1 in PBS. One hundred microliters of either dilution was applied per well.

Enzymaticc treatments. Assay reaction mixtures were stopped by heating for 15 min at 75°C. Sampless were digested with 150 mU/ml Laminarinase (Sigma; L-5272, lot nr. 50H0078) in 1000 mM MES pH 5.5 o/n at 30°C. Alternatively, samples were treated with 60 U/ml

97 7 ChapterChapter 4

recombinantt 1,6-/?-glucanase (Bom et al. 1998) in 50 mM NaAc, pH 5.0, at 30°C. For Zymolyasee 100T treatment, samples were digested in 8.7 U/ml Zymolyase 100T in 50 mM sodiumm phosphate, pH 7.0, o/n at 30°C.

Dott blot assay. Unless stated otherwise, for each assay two microliters of reaction mixture wass spotted onto nitrocellulose membranes, and allowed to dry for at least 30 min. Since this materiall was only loosely associated with the membranes, the dot blots required gentle handling.. Membranes were blocked in 5% (w/v) non-fat milk powder in PBS, and probed withh anti 1,6-yS-glucan antibodies at a dilution of 1:50,000 in PBS with 3% (w/v) BSA. For thee detection of l,3-/?-glucan, PVDF membranes were used to limit background staining. Thesee membranes were blocked in 3% (w/v) BSA in PBS, and probed with anti 1,3-/?-glucan antibodiess at a 1:25,000 dilution in PBS with 3% (w/v) BSA. Subsequently, membranes were probedd with anti-rabbit IgG peroxidase at a dilution of 1:10,000 PBS with 3% (w/v) BSA. Thee blots were visualized with ECL Western blotting detection reagent (Amersham). Competitionn experiments were carried out with pustulan, laminarin, yeast mannan, and starch inn a final concentration of 0.5 mg/ml. l,6-/?-Glucann detection by inhibition EIA. Detection of 1,6-/?-glucan was based on a methodd developed by Douwes et al. (1996). Wells of a 96-wells microtiter plate (Greiner; 6555 092; batch number 99220103) were coated with 200 ul of a 2 ug/ml pustulan solution in PBS,, pH 7.0, overnight at 4°C. Prior to use, the pustulan solution was extensively boiled to dissolvee the material completely. In order to avoid problems caused by static electricity, platess were stored in the refrigerator and all handling was performed on pre-wetted towels. Platess were washed three times with 0.05% (v/v) Tween 20 in PBS (PBT), and subsequently blockedd with 300 ul of 0.5% (w/v) gelatin in PBT (PBTG). Test samples or pustulan standardss were applied in a volume of 100 ul per well, and mixed with an equal volume of antii 1,6-/?-glucan antibodies in a dilution of 1:100,000 in PBTG. After incubation for lh, the wellss were thoroughly washed and incubated with 200 ul of goat anti-rabbit peroxidase antibodiess diluted 1:5,000 in PBTG. After thorough washing, the wells were incubated with 2000 ul of o-phenylenediamine (2 mg/ml) in a O.05 M citrate-phosphate buffer, pH 5.5, with 0.015%% (v/v) hydrogen peroxide. After 10 min, the reaction was terminated by adding 50 ul off 2 M HCl. The absorbance at 490 nm was measured in a SpectraCount microplate reader (Packardd BioScience), and data were processed with I-Smart (Packard BioScience) using the 55 PL curve fitting preset.

98 8 InIn vitro 1,6-fi-glucan synthase activity

RESULTS S l,6-/7-Glucann is not synthesized under l,3-/?-glucan synthase conditions Inn the usual assay for in vitro l,3-/?-glucan synthase activity in fungi a crude membranee preparation is used to which radioactively labeled UDP-glucose is added as substratee donor and GTP as activator. This assay is carried out at alkaline pH (Shematek et al,al, 1980; Shematek and Cabib, 1980). The reaction is terminated by the addition of TCA to a finall concentration of 10%. To remove the unreacted substrate, the reaction mixture is filtered overr a glass fiber filter, which retains the reaction product. As a first step, we investigated whetherr 1,6-/?-glucan was formed during the in vitro synthesis of l,3-/?-glucan. Previously, whenn using C. albicans extracts, Frost et al. (1994) found that about 20% of the reaction productt was resistant to Zymolyase 100T digestion, an enzyme preparation with 1,3-/?- glucanasee and protease activity. In contrast, laminarinase, a l,3-/?-glucanase preparation with somee 1,6-/?-glucanase activity, could fully degrade the reaction product (Frost et al., 1994). Wee verified these results using Saccharomyces cerevisiae extracts. Again, about 20% of the reactionn products formed in the l,3-/?-glucan synthase assay resisted digestion by Zymolyase 100T.. However, both the total and the l,3-/?-glucanase resistant fraction of the 1,3-/?-glucan synthasee reaction product also proved insensitive to a recombinant l,6-/?-glucanase (Figure 1A). . Subsequently,, we analyzed the product formed in the l,3-/?-glucan synthase assay usingg antibodies raised against 1,3-/?-glucan and l,6-/?-glucan. A dot blot assay was used, in whichh the reaction mixture is spotted onto a nitrocellulose membrane and allowed to dry beforee probing the spots with antibodies. The advantage of this method compared to the previouss one is that loss of reaction products is minimized. Using antibodies directed against l,3-/?-glucan,, an increase of l,3-/?-glucan in time could clearly be detected. However, when usingg antibodies directed against l,6-/?-glucan, no increase in l,6-/?-glucan in time was seen underr these conditions (Figure IB). Therefore we conclude that under the conditions used for thee in vitro synthesis of l,3-/?-glucan, no detectable amount of l,6-/?-glucan is formed.

Inn vitro synthesis of l,6-/?-glucan Thee results of the described dot blot experiments prompted us to further exploit this methodd for the detection of in vitro 1,6-/?-glucan synthesis. A possible reason why this approachh did not yield the production of 1,6-/?-glucan under the given conditions was the use off a suboptimal buffer or pH or the presence of EDTA. We tested various buffers such as sodiumm acetate (pH 4 - 6), sodium phosphate (pH 6 - 8), and MES-NaOH (pH 5.5 - 6.7) in the absencee of EDTA and found that the use of MES buffer resulted in optimal activity at about pHH 6.5 (Figure 2A). The reaction was dependent on the presence of UDP-glucose and GTP

99 9 ChapterChapter 4

(Figuree 2B). It was also stimulated by the addition of EDTA. This chelator was therefore includedd in all further experiments.

100 0

B B timee (min) 0 15 30 60 90 120

1,6-p-glucann pAbs

1,3-p-glucann pAbs »» * «

Figuree 1 Under l,3-/?-glucan synthase conditions 1,6-/?-glucan is not synthesized. (A) l,3-/?-Glucan synthase assayss were performed using UDP-[14C]-glucose for 1 hr at 30°C, heat inactivated (15 min at 75°C), and next thee reaction products were subjected to various enzymatic treatments. The residual material after enzymatic treatmentt is expressed as percentage of the untreated sample. (1) Untreated sample, (2) Zymolyase 100T treated sample,, (3) sample treated by Zymolyase 100T followed by 1,6-/?-glucanase, (4) sample treated by 1,6-/?- glucanase.. Mock incubations showed no decrease in residual material. Error bars indicate standard deviation (n=2).. (B) l,3-/?-Glucan synthase assays were carried out and stopped at different time points by heat inactivation.. Samples (5 ul) were spotted onto a nitrocellulose membrane for probing with anti 1,6-/?-glucan antibodies,, and onto a polyvinylidene fluoride (PVDF) membrane for probing with anti l,3-/?-glucan antibodies. Thee difference of membranes explains the dissimilarity in the spot shapes.

Next,, we characterized the reaction product in two ways. First, we did competition experimentss and found that the signal was only sensitive to the presence of pustulan (a commerciallyy available 1,6-/?-glucan), but not to the presence of laminarin (a commercially

100 0 InIn vitro 1,6-fi-glucan synthase activity

availablee l,3-/?-glucan), mannan, or starch (Figure 3A). Second, the reaction product was sensitivee to treatment with 1,6-/?-glucanase (Figure 3B). This indicates that in our assay 1,6- /?-glucann was a reaction product. The presence of Zymolyase in a dot blot assay resulted in a veryy high background signal, which rendered it impractical in these experiments (Figure 3C).

B B

imee (min) 0 0 30 0 60 0 120 0 180 0 timee (min) 0 180

pH5.5 5 s s « « controll

pHH 5.75 0 0 w w w/oo UDP-GIc

pH6.0 0 » » t t w/oo GTP

pH6.25 5 1 1 withh EDTA I

pH66 5

pHH 6.75 , , # # I I

Figuree 2 In vitro synthesis of 1,6-/?-glucan at various pHs. Two microliters of a reaction mixture were spotted ontoo a nitrocellulose membrane. After drying, membranes were probed with anti l,6-/?-glucan antibodies and subsequentlyy with peroxidase-labeled secondary antibodies. The signal was visualized by ECL. (A) pH dependencyy of 1,6-/?-glucan synthase activity in MES-NaOH. (B) l,6-/?-Glucan synthase activity in MES-NaOH att pH 6.0 without UDP-glucose or GTP, or in the presence of EDTA.

AA quantitative method for the detection of in vitro synthesized l,6-/?-glucan Ass the dot blot assay was difficult to quantify, we developed an inhibition enzyme immunoassayy based on the method developed by Douwes and co-workers (1996) for the detectionn of l,3-/?-glucan. Figure 4A depicts a schematic representation of the inhibition enzymee immunoassay. Pustulan was used to coat the wells of a 96-well microtiter plate. The wellss were blocked with gelatin, and the samples were applied in the presence of the anti 1,6- /?-glucann antibodies. For calibration, a two-fold dilution series of pustulan ranging from 1,000 too 1.9 ng/ml was used in each microtiter plate. Enzyme-linked secondary antibodies were usedd to visualize the primary antibodies bound to the coated l,6-/?-glucan. A typical inhibitionn dilution curve is shown in Figure 4B; the absolute absorption values, however, variedd between experiments. The working range of the assay was arbitrarily set between 15 to 85%% inhibition (dashed lines), corresponding to approximately 5 to 400 ng/ml pustulan. Laminarinn did not compete in this assay up to concentrations of 1 mg/ml (data not shown),

101 1 ChapterChapter 4

indicatingg that the assay is specific for l,6-/?-glucan. Using this method for the quantification off 1,6-/?-glucan under our assay conditions, we found that already at t = 0 some 1,6-/?-glucan wass present, varying between 100 and 200 pg, in agreement with the results of the dot blot assay.. This could be explained by the fact that the membrane extract used, contained some celll wall debris as could be shown by staining with calcofluor white (data not shown). The amountt of 1,6-//-glucan present at t = 0 was subtracted from the levels found at other time points.. Without the presence of UDP-glucose as sugar donor, no increase in 1,6-/?-glucan levelss was seen (Figure AC). However, in the presence of UDP-glucose, a clear increase couldd be noticed. From 15 minutes onwards, the signal appeared to increase linearly. In this particularr experiment, the specific enzyme activity in the linear range was approximately 26 pg/ugg protein/min pustulan equivalents.

B B

timee (min) 0 15 30 70 97 timee (min) 0 15 30

pustulan n mock k

laminarmm 0 1,6-P-glucanase e

mannann O *

starchh « c m

timee (min) 0 15 30

Zymolyase e ^^^^ ^KF ^^^

Figuree 3. Characterization of the in vitro synthesized 1,6-/?-glucan. (A) Reactions were stopped at different timee points, and spotted onto nitrocellulose membranes. The anti 1,6-/?-glucan antibodies were probed in the presencee of either pustulan, laminarin, yeast mannan, or starch. (B) Reactions were stopped at different time points,, and subsequently either mock incubated, incubated with recombinant 1,6-/?-glucanase, or precipitated withh 10% TCA. (C) Reactions were stopped at different time points, and incubated with Zymolyase 100T.

102 2 InIn vitro 1,6-[i-glucan synthase activity

1,6-p-glucann in sample secondaryy Ab ++ primary Ab

ighh 1,6-)J-glucan concentration

VV o^

OO 1,6-P-glucan —<< primary antibody 4—'' secondary antibody

B B

100 100 1000 10000 0 ng/mll pustulan equivalents

++ UDP-GIc -- UDP-GIc

200 30 timee (min)

Figuree 4. In vitro synthesis of 1,6-/?-glucan measured by an inhibition enzyme immunoassay. (EIA). (A) Schematicc representation of an inhibition EIA: wells of a microtiter plate are coated with 1,6-/9-glucan (pustulan),, and blocked afterwards. Samples or standards are applied in the presence of anti 1,6-/?-glucan

103 3 ChapterChapter 4

antibodies.. The number of antibodies that bind to coated 1,6-/f-glucan is inversely related to the concentration of addedd l,6-/?-glucan. Antibodies bound to 1,6-/J-gIucan are visualized by enzyme-labeled secondary antibodies. (B)) A typical pustutan standard curve. The dashed horizontal lines indicate 15% and 85% inhibition, respectively.. (C) Time course of 1,6-/?-glucan synthesis. Closed circles represent time points in the presence of UDP-glucosee and closed squares represent time points in the absence of UDP-glucose.

104 4 InIn vitro 1,6-p-glucan synthase activity

DISCUSSION N

Becausee in vitro synthesis of l,6-/?-glucan could not be detected when a conventional l,3-/?-glucann synthase assay was used, a dot blot technique was exploited. Using UDP- glucosee as sugar donor, we found maximal 1,6-/?-glucan synthesis at pH 6.5 in MES-NaOH bufferr in the presence of EDTA. When during the blotting step the antibodies against 1,6-fi- glucann were incubated in the presence of pustulan, a commercially available 1,6-/7-glucan, the signall was lost, whereas addition of laminarin (l,3-/?-glucan), yeast mannan, or starch had no effect.. This confirmed that the antibodies specifically recognized l,6-/?-glucan. This was furtherr supported by the observation, that the signal was lost upon preincubation of the reactionn products with 1,6-/?-glucanase. However, the dot blots required gentle handling, as vigorouss washing resulted in the dissociation of the epitope from the blot. The signal also appearedd to be rapidly saturated, suggesting that the linear range of the signal was limited. Thiss might be the result of the limited binding capacity of the nitrocellulose membrane for thee 1,6-/?-glucan produced in vitro. These limitations prompted us to explore another method forr the quantification of in vitro synthesized 1,6-/?-glucan. Douwess et al. (1996) developed an inhibition enzyme immunoassay for the quantificationn of 1,3-/?-glucan. We adapted this procedure, and used it for the quantification off l,6-/?-glucan. In this method, the 1,6-/?-glucan content of a sample is expressed in pustulan equivalents,, i.e. the level of inhibition produced by a sample is expressed in the amount of pustulann that results in the same level of inhibition. To roughly estimate how much glucose is incorporatedd into 1,6-/?-glucan in our assay, the following calculation was carried out. Figure 44 shows that in the linear range of the assay about 26 pg of pustulan equivalents per minute aree produced per ug protein, corresponding to about 0.16 pmol of glucose per min. If we definee one unit of 1,6-/?-glucan synthase activity equal to the incorporation of 1 umol glucose perr minute at 30°C, the extract used in our experiments had a specific activity of 0.16 mU/mg protein.. The specific activity of crude l,3-/?-glucan synthase extracts has been reported to rangee from about 15 to 35 mU/mg protein (Shematek et al., 1980). Although it is not to be expectedd that both enzymes have the same specific activity, this large difference suggest that thee current conditions for in vitro 1,6-/?-glucan synthase activity may be further optimized. Onee possible modification would be the application of other sugar-donors, since it has been suggestedd that in a crude (3-glucan synthase assay the addition of GDP-glucose results in a productt consisting of 1,6-/?-gIucan with some l,3-/?-linkages in the side chains (Balint et al., 1976). . Inn S. cerevisiae and S. pombe, GTP-binding proteins are known to regulate in vivo synthesiss of 1,3-/?-glucan and a-glucan, respectively. In Schizosaccaromyces pombe it has beenn found that Rho2p regulates the synthesis of a-glucan (Calonge et al, 2000). Accordingly,, the GTP-binding protein Rholp is known to regulate l,3-/?-glucan synthesis in

105 5 ChapterChapter 4

S.S. cerevisiae (Drgonova et al, 1996; Qadota et al, 1996). Interestingly, we find in vitro 1,6- /?-glucann synthase activity also depends on the addition of GTP, which might suggest a role forr a GTP-binding protein in the regulation of 1,6-/?-glucan synthesis. A possible candidate forr the regulation l,6-/?-glucan synthesis is Rho3p, since we have found that RHÖ3 is involvedd in maintaining 1,6-/?-glucan levels in the cell wall (Vink et al, 2002, Chapter 2). Thee RH03 gene is involved in actin polarization and exocytosis (Matsui and Toh-e, 1992; Imaii et al, 1996; Robinson et al, 1999; Adamo et al, 1999), and affects integrity of the emergingg bud in combination with its functionally related homolog RH04 (Matsui and Toh- e,, 1992). When the RH03 gene is deleted, cells become resistant to the Kl killer toxin, and havee less l,6-/?-glucan in their walls. Conversely, when the RH03 gene is overexpressed, cellss become more sensitive to the Kl killer toxin. However, the functionally related RH04 doess not appear to contribute to this specific function OÏRH03 (Vink et al, 2002, Chapter 2). Ourr in vitro assay for l,6-/?-glucan synthase provides a tool for addressing several importantt questions pertinent to the synthesis of 1,6-/?-glucan. First of all, it might be instrumentall in the identification of the gene or genes that encode l,6-/?-glucan synthase. In addition,, it should provide the opportunity to definitely solve the question where 1,6-/?-glucan synthesiss takes place. Finally, as l,6-/?-glucan has been found as a crucial cell wall componentt in many pathogenic fungi, such as Candida species and Cryptococcus neoformam (Chaffinn et al, 1998; James et al, 1990), this method provides a tool for screening for new antifungals,, specifically directed against l,6-/?-glucan synthesis.

ACKNOWLEDGMENTS S Thee authors thank all members of the Klis and Duran labs for stimulating discussions. Also, Drss G. Doekes, R.C. Montijn, T. Munnik and E. Cabib are thanked for helpful suggestions andd comments. EV acknowledges a traveling grant from the Netherlands Organization for Scientificc Research (NWO).

106 6 InIn vitro 1,6-{}-gIucan synthase activity

REFERENCES S

Adamoo J.E., Rossi G., and Brennwald P. (1999). The Rho GTPase Rho3 has a direct role inn exocytosis that is distinct from its role in actin polarity. Mol. Biol. Cell 10:4121-4133

Balintt S., Farkas V., and Bauer S. (1976). Biosynthesis of P-glucans catalyzed by a particulatee enzyme preparation from yeast. FEBS Lett. M:A4-A1

Bomm I.J., Dielbandhoesing S.K., Harvey K.N., Oomes S.J., Klis F.M., and Brul S. (1998). AA new tool for studying the molecular architecture of the fungal cell wall: one-step purificationn of recombinant Trichoderma P-(l-6)-glucanase expressed in Pichia pastoris. Biochim.Biochim. Biophys. Acta. 1425:419-424

Bradfordd M.M. (1976). A rapid and sensitive method for the quantitation of microgram quantitiess of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248- 254 4

Cabibb E., and Kang M.S. (1987). Fungal 1,3-P-glucan synthase. Methods Enzymol. 138:637-642 2

Cabibb E., Kang M.S., and Au-Young J. (1987). Chitin synthase from Saccharomyces cerevisiae.cerevisiae. Methods Enzymol. 138:643-649

Cabib,, E., Shaw, J. A., Mol, P. C, Bowers, B., and Choi, W.-J. (1996). In The Mycota, vol.. Ill, pp. 243-267. Edited by R. Bramble and G.A. Marzluf. Springer-Verlag, Berlin

Cabibb E., Drgonova J., and Drgon T. (1998). Role of small G proteins in yeast cell polarizationn and wall biosynthesis. Annu. Rev. Biochem. 67:307-333

Calongee T.M., Nakano K., Arellano M., Arai R., Katayama S., Toda T., Mabuchi I., and Perezz P. (2000). Schizosaccharomyces pombe Rho2p GTPase regulates cell wall a-glucan biosynthesiss through the protein kinase Pck2p. Mol. Biol. Cell 11:4393-4401

Chaffinn W.L., Lopez-Ribot J.L., Casanova M., Gozalbo D., and Martinez J.P. (1998). Celll wall and secreted proteins of Candida albicans: identification, function, and expression. Microbiol.Microbiol. Mol. Biol. Rev. 62:130-180

107 7 ChapterChapter 4

Chuang,, J.S., and Schekman, R.W. (1996). Differential trafficking and timed localization off two chitin synthase proteins, Chs2p and Chs3p. J. Cell Biol. 135:597-610

Douwess J., Doekes G., Montijn R., Heederik D., and Brunekreef B. (1996). Measurement off (3(1—>-3)-glucans in occupational and home environments with an inhibition enzyme immunoassay.. Appl. Environ. Microbiol. 62:3176-82

Drgonovaa J., Drgon T., Tanaka K., Kollar R., Chen G.C., Ford R.A., Chan C.S., Takai Y.,, and Cabib E. (1996). Rholp, a yeast protein at the interface between cell polarization andd morphogenesis. Science 211:211-21$

Frostt D.J., Brandt K., Capobianco J., and Goldman R. (1994). Characterization of (1,3)- (3-glucann synthase in Candida albicans: microsomal assay from the yeast or mycelial morphologicall forms and a permeabilized whole-cell assay. Microbiology 140:2239-2246

Imaii J., Toh-e A., and Matsui Y. (1996). Genetic analysis of the Saccharomyces cerevisiae RH03RH03 gene, encoding a rho-type small GTPase, provides evidence for a role in bud formation.. Genetics 142:359-369

Ishiguroo J., Saitou A., Duran A., and Ribas J.C. (1997). cpsJ+, a Schizosaccharomyces pombepombe gene homolog of Saccharomyces cerevisiae FKS genes whose mutation confers hypersensitivityy to cyclosporin A and papulacandin B. J. Bacteriol. 179:7653-7662

Jamess P.G., Cherniak R., Jones R.G., Stortz C.A., and Reiss E. (1990). Cell-wall glucans off Cryptococcus neoformans Cap 67. Carbohydr. Res. 198:23-38

Kapteynn J.C, Van Den Ende H., and Klis F.M. (1999). The contribution of cell wall proteinss to the organization of the yeast cell wall. Biochim. Biophys. Acta 1426:373-383

Kliss F.M., de Groot P., and Hellingwerf K. (2001). Molecular organization of the cell wall off Candida albicans. Med. Mycol. 39, Suppl. 1:1-8

Kliss F.M., Mol P., Hellingwerf K., and Brul S. (2002). Dynamics of cell wall structure in Saccharomycess cerevisiae. F E MS Microbiol. Rev. 26: 239-256

Kollarr R., Reinhold B.B., Petrakova E., Yeh H.J., Ashwell G., Drgonova J., Kapteyn J.C,, Klis F.M., and Cabib E. (1997). Architecture of the yeast cell wall. Beta( 1->6)-glucan interconnectss mannoprotein, beta(l~>)3-glucan, and chitin. J. Biol. Chem. 272:17762-17775

108 8 InIn vitro 1,6-fi-glucan synthase activity

Matsuii Y., and Toh-e A. (1992). Yeast RH03 and RH04 ras superfamily genes are necessaryy for bud growth, and their defect is suppressed by a high dose of bud formation geness CDC42 and BEM1. Mol. Cell. Biol. 12:5690-5699

Montijnn R.C., van Rinsum J., van Schagen F.A., and Klis F.M. (1994). Glucomannoproteinss in the cell wall of Saccharomyces cerevisiae contain a novel type of carbohydratee side chain. J. Biol. Chem. 269:19338-19342

Montijnn R.C., Vink E., Muller W.H., Verkleij A.J., Van Den Ende H., Henrissat B., and Kliss F.M. (1999). Localization of synthesis of J31,6-glucan in Saccharomyces cerevisiae. J. Bacteriol.Bacteriol. 181:7414-7420

Orlean,, P. (1997). Biogenesis of yeast wall and surface components. In The Molecular BiologyBiology of the Yeast Saccharomyces, vol. 3, pp. 229-362. Edited by J. R. Pringle, J. R. Broach andd E. W. Jones. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Qadotaa H., Python C.P., Inoue S.B., Arisawa M., Anraku Y., Zheng Y., Watanabe T., Levinn D.E., and Ohya Y. (1996). Identification of yeast Rholp GTPase as a regulatory subunitt of l,3-(3-glucan synthase. Science 272:279-281

Robinsonn N.G., Guo L., Imai J., Toh-e A., Matsui Y., and Tamanoi F. (1999). Rho3 of SaccharomycesSaccharomyces cerevisiae, which regulates the actin cytoskeleton and exocytosis, is a GTPasee which interacts with Myo2 and Exo70. Mol. Cell. Biol. 19:3580-3587

Shahiniann S., and Bussey H. (2000). (3-1,6-Glucan synthesis in Saccharomyces cerevisiae. Mol.Mol. Microbiol. 35:477-489

Shedletzkyy E., Unger C, and Delmer D.P. (1997). A microtiter-based fluorescence assay forr (l,3)-(3-glucan synthases. AnalBiochem. 249:88-93

Shematekk E.M., Braatz J.A., and Cabib E. (1980). Biosynthesis of the yeast cell wall. I. Preparationn and properties of p-( 1 —>3)glucan synthetase. J. Biol. Chem. 255:888-894

Shematekk E.M., and Cabib E. (1980). Biosynthesis of the yeast cell wall. II. Regulation of (3-(1^3)glucann synthetase by ATP and GTP. J. Biol. Chem. 255:895-902

109 9 ChapterChapter 4

Vinkk E., Vossen J.H., Ram A.F.J., van den Ende H., Brekelmans S.S.C., de Nobel H., andd Klis F.M. (2002). The protein kinase Kiel affects 1,6-beta-glucan levels in the cell wall off Saccharomyces cerevisiae. Microbiology 148:4035-4048

Winstonn F., Dollard C, and Ricupero-Hovasse S.L. (1995). Construction of a set of convenientt Saccharomyces cerevisiae strains that are isogenic to S288C. Yeast 11:53-55

110 0 ChapterChapter 5

Generall Discussion ChapterChapter 5

1.1. Introduction Thee cell wall of baker's yeast has a bilayered structure and is composed of four components: l,3-/7-glucann and chitin, which form the inner layer of the cell wall and represent the skeletal network,, and mannoproteins which are mainly linked through l,6-/?-glucan to the skeletal network,, and which form the outer layer of the cell wall (reviewed by Klis, 1994, Orlean, 1997,, and Klis et al, 2002). The mannoproteins and glucans are present in about equal amounts,, whereas chitin represents only a minor percentage of the cell wall (1 - 2%). About 80%% of the glucan is represented by 1,3-/?-glucan, and the remaining 20% is 1,6-^-glucan. In thee cell wall of Saccharomyces cerevisiae, 1,6-/?-glucan occurs as a highly branched, water- solublee polymer consisting of about 130 residues (Manners et al., 1973). Roughlyy six major research questions can be posed with regard to the biogenesis of 1,6-/?-glucan:: (i) is 1,6-/?-glucan synthesis a single- or a multistep process? (ii) at which locationn are these steps performed? (iii) which enzymes are involved? (iv) does the primary syntheticc step involve a single protein or multiple proteins? (v) how is l,6-/?-glucan assembledd into the cell wall? (vi) how is l,6-/?-glucan synthesis regulated? It is remarkable howw little is known about the biosynthesis of 1,6-/?-gIucan and its regulation in comparison withh the other cell wall components. When this work was started, our knowledge on this subjectt was mainly based on a genetic screen developed by Bussey and co-workers that exploitss the Kl killer toxin (reviewed by Shahinian and Bussey, 2000). As a rule, the sensitivityy of cells for this toxin is correlated with their cell wall 1,6-/?-glucan levels: cells withh lowered levels of 1,6-/?-glucan are more resistant to this toxin, and vice versa. Based on thee identification of a relatively small number of genes by this screen, it was hypothesized thatt 1,6-y6-glucan synthesis is a stepwise process that starts in the ER, continues along the secretoryy pathway, and is completed at the cell surface (reviewed by Klis, 1994, and Orlean, 1997).. As depicted in figure 1, some of the KRE genes identified in the killer toxin screen allegedlyy encoded glucan synthases. Kre5 was believed to initiate the synthesis of 1,6-/?- glucann in the endoplasmic reticulum (ER), either as a free molecule or attached to precursor formss of cell wall proteins. Extension of l,6-/?-glucan was believed to occur by a related pair off Golgi-located enzymes (Kre6 and Sknl). Finally, maturation steps were believed to take placee at the cell surface, and involved Krel and Kre9. As the original screen appeared to be exhaustive,, it was assumed that most of the important genes in 1,6-/?-glucan biogenesis had beenn identified. Recently, a genome-wide screen was carried out based on Kl killer toxin sensitivityy (Page et al, 2003). In this study 268 genes were identified with a toxin phenotype, off which 186 displayed increased resistance and 82 hypersensitivity. The genes that were identifiedd in this screen coded for proteins from many different functional classes, such as signaling,, transcription, secretion, protein modification, and ribosomal subunits. Unfortunately,, no obvious candidate genes were found and/or proposed to encode subunits of thee 1,6-/7-glucan synthase, although it might be difficult to identify such a gene without

112 2 GeneralGeneral Discussion

preciselyy knowing the diagnostic features of the predicted protein. Moreover, although heterozygouss mutant diploids for these genes were used for analysis, essential genes are easilyy missed in this screen, because not all essential genes display a so-called haploinsufficiencyy phenotype when one copy is disrupted in a diploid. Inn the following paragraphs, the contributions of this thesis to solving some of the questionss on l,6-/?-glucan synthesis are discussed.

2.2. Localization ofl,6-fi-glucan synthesis 1,6-/?-Glucann synthesis has long been presumed to start in the ER, and to be continued alongg the rest of the secretory pathway (reviewed in Klis, 1994, and Orlean, 1997). However, ass extensively discussed in chapter 3 thus far there is no significant evidence for the presence off intracellular l,6-/?-glucan. Rather, the l,6-/?-glucan seems to be associated with the plasma membranee and the cell wall, suggesting that the major synthase activity is located in the plasmaa membrane (Figure 2). This was shown by using a temperature-sensitive secretory mutant,, which was blocked in exocytosis when grown at the restrictive temperature. Secretoryy vesicles were allowed to accumulate for the duration of roughly one cell cycle. Afterr fractionation of the cell organelles, the bulk of l,6-/?-glucan appeared to co-localize withh the plasma membrane-derived fractions. If its synthesis was to take place intracellularly, onee would expect that the intracellular l,6-/?-glucan content would be about equivalent to the amountt of 1,6-/?-glucan produced by the cells during a generation time. The same secretory mutantt grown at the restrictive temperature was also used for immunogold labeling, showing thatt 1,6-/?-glucan was only present at the cell surface and in the cell wall. In addition, using otherr secretory mutants blocked at various stages of the secretory pathway to allow accumulationn of precursor forms of the cell wall protein a-agglutinin, the various precursor formss were easily identifiable, but they never became attached to l,6-/?-glucan, although the celll wall form of a-agglutinin is known to be 1,6-/?-glucan-bound (Lu et al.t 1995). Together thesee observations strongly suggest that the bulk of 1,6-/?-glucan is synthesized at the plasma membranee (Figure 2). These data contradict the earlier ideas about the biogenesis of 1,6-/?- glucann as presented in figure 1. In particular, the idea that Kre5 and Kre6/Sknl are involved inn the initiation and elongation of 1,6-/?-glucan chains has become untenable. In the next section,, we will discuss alternative functions for these proteins in the biogenesis of 1,6-/?- glucan. .

3.3. Role ofER and Golgi proteins in 1,6-ft-glucan synthesis Ass stated before, a number of genes have been proposed mainly based on genetic evidencee to function in 1,6-/7-glucan synthesis. As intracellular synthesis of 1,6-/7-glucan seemss unlikely, the roles of Kre5 and Kre6/Sknl have to be reinterpreted. The KRE5 gene encodess an ER-localized soluble glycoprotein with a signal sequence and the C-terminal ER-

113 3 ChapterChapter 5

retentionn signal HDEL (Meaden et ai, 1990; Levinson et ai, 2002). Its deletion severely compromisess growth in S. cerevisiae (Meaden et ai, 1990; Shahinian et ai, 1998; Levinson etet ai, 2002). The Kre5 protein has significant homology with UDP-glucose: glycoprotein glucosyltransferasee (GT), an enzyme that is involved in quality control of proteins in the ER. However,, this enzyme activity has not been detected in S. cerevisiae (Fernandez et ai, 1994; Jakobb et ai, 1998). The Kre5 function might thus be related to the GT function, yet might havee diverged from the quality control of protein folding to a function in 1,6-/?-glucan synthesis.. For example, Kre5 may be involved in the production of an acceptor structure or a primerr structure that is extended by the l,6-/?-glucan synthase (Figure 2). This primer might bee a glucosylated protein, as is the case for glycogen synthesis where the self-glucosylating proteinn glycogenin functions as a primer (Alonso et ai, 1995). Another option is that this primerr is a glucosylated sterol. It was recently found that cellulose synthesis in plants requiress sitosterol-/?-glucoside as a primer (Peng et ai, 2002). Sterol glucosides have also beenn identified in S. cerevisiae, which are produced by the UDP-glucose: sterol glucosyltransferasee encoded by UGT51. However, cells deleted for the sterol glucosyltranferasee lack sterol glucosides yet exhibit normal growth under several conditions (Warneckee et ai, 1999), and therefore it is unlikely that a glucosylated sterol functions as a primerr for l,6-/?-glucan synthesis. It has also been proposed that Kre5 may glucosylate the GPI-anchorr of precursor forms of GPI-CWPs, a process that might be key in the later attachmentt of l,6-/7-glucan (Shahinian and Bussey, 2000). However, it is unlikely that these modificationss occur on precursor forms of GPI-CWPs, as the assembly of the l,3-/?-glucan- 1,6-/?-glucann complex in the cell wall does not seem to depend on GPI-CWP incorporation (Rohh et ai, 2002). Interestingly, there is import of UDP-glucose in the ER of S. cerevisiae (Castroo et ai, 1999). However, up to this point no UDP-glucose dependent reaction in the ER lumenn of S. cerevisiae has been identified. This nucleotide sugar might thus be a candidate sugarr donor for Kre5. Cells deleted for BIG I, a gene that encodes an ER transmembrane protein,, also show a dramatic reduction in 1,6-/?-glucan levels. Although its function is unclear,, it is supposed to play a role together with Kre5 in a common pathway, yet with partiallyy distinct functions. This could be the modification of a precursor, or the assembly of aa functional synthase complex (Azuma et ai, 2002). If indeed Kre5 is involved in synthesizingg an acceptor structure, it seems possible that cells deleted for KRE5 would still be ablee to produce 1,6-jö-glucan, yet the incorporation of 1,6-/?-glucan in the cell wall would thenn be disturbed. It would thus be interesting to test the culture medium of these cells for the presencee of l,6-/?-glucan.

114 4 GeneralGeneral Discussion

PM M

latee Golgi/ secretory y compartment? ?

Golgi i

ER R

AA 1,3-p-linked glucose

Figuree 1. Previous model for 1,6-/?-glucan synthesis in S. cerevisiae (adapted from Orlean, 1997). The order of eventss was based on localization of the gene products, and on epistatic analysis of the genes. In the ER, 1,6-/?- glucann synthesis was presumed to be initiated by Kre5. In the Golgi, Kre6/Sknl were believed to be responsible

115 5 ChapterChapter 5

forr the extension of 1,6-/?-glucan. Finally, at the cell surface the maturation of l,6-/?-glucan was presumably establishedd by Krel and Kre9, although during their passage through the secretory pathway these proteins might alreadyy have carried out (part of) their function.

Usingg hydrophobic cluster analysis (HCA), we found that the Golgi-localized Kre6 proteinn and its close homolog Sknl share significant homology to family 16 glycoside hydrolases,, which indicates that they function as glycoside hydrolases or transglycosidases (Chapterr 3). This excludes that, in contrast to what was postulated before (Figure 1), Kre6 andd Sknl function as glucosyltransferases, which, unlike transglycosidases, utilize activated sugars.. The action of Kre6 might be to modify a glucose-containing acceptor structure associatedd with GPI-anchors in the ER or to other such structures (Figure 2).

4.4. Development of an assay for 1,6-fi-glucan synthase activity Althoughh the genetic screens using the Kl killer toxin have been of great value to the studyy of cell wall biogenesis, this genetic approach by itself seems to be exhausted. Up to now,, no potential candidate genes that encode (components of) the 1,6-/?-glucan synthase complexx have been identified. In particular, the lack of a biochemical assay to measure 1,6-/?- glucann synthase activity has hampered the use of reverse genetics to identify components of thiss enzyme activity. For this reason, we decided to develop a biochemical assay for in vitro 1,6-/?-glucann synthase activity. In principle, purification of the activity in combination with currentlyy available techniques such as mass spectrometry could provide a simple means to identifyy the components of the synthase. A similar approach has helped to identify the catalyticc subunit of the 1,3-/?-glucan synthase complex (Inoue et ai, 1995). Chapter 4 describess the development of a method for measuring l,6-/?-glucan synthase activity in vitro. Thee reaction products of a l,6-/?-glucan synthase assay were analyzed in two ways: by a qualitativee dot blot assay, and by a quantitative enzyme immunoassay, both using affinity- purifiedd antibodies raised against l,6-/?-glucan. The optimal pH was 6.5 for the in vitro 1,6-/?- glucann synthase activity. The structure of the reaction products is unknown, so the question if thee 1,6-/?-glucan synthase produces linear polymers of 1,6-/?-glucan or perhaps a more branchedd product remains unanswered. If sufficient amounts of reaction product could be producedd in vitro, the structure could be resolved using NMR. For example, the l,3-/?-glucan synthasee produces linear l,3-/?-glucan polymers of 60 - 80 residues long (Shematek et al, 1980).. It would also be interesting to know what the chain lengths are of the 1,6-/?-glucan synthasee products. These reaction products appear to be water-soluble, in contrast to e.g. the l,3-/?-glucann synthase products and the commercially available 1,6-/?-glucan pustulan (with ann average chain length of about 110 residues). This suggests that the 1,6-/?-glucan synthase producess either relatively short oligomers which are subsequently remodeled, or highly branchedd polymers. Both hypotheses are in agreement with the fact that 1,6-/?-glucan in the

116 6 GeneralGeneral Discussion

celll wall of S. cerevisiae occurs as a highly branched, water-soluble polymer consisting of aboutt 130 residues (Manners et ah, 1973). Anotherr important question is if the 1,6-/?-glucan synthase can produce 1,6-/?-glucan polymerss de novo, or if it needs a primer structure to which the synthase can transfer 1,6-/?- linkedd glucose residues. This could be investigated by analyzing the terminal reducing glucose,, which, if UDP-[l4C]-gIucose would be used as substrate, would contain this Re- label.. This can be done by reduction by sodium borohydride followed by acid hydrolysis. The liberatedd sorbitol (derived from the terminal reducing glucose) can subsequently be analyzed forr the presence of the 14C-label. Thee in vitro assay for 1,6-/?-glucan synthase activity, as described in Chapter 4, displayss a relatively low activity compared to for example the in vitro t,3-/?-glucan synthase activity.. This might very well be due to non-optimal assay conditions. An interesting possibilityy is that the putative primer structure is present at a limiting concentration and does nott allow the synthase to function at maximal speed. Currently, our in vitro 1,6-/?-glucan synthasee assay is being used and further developed by Drs Manon Gerard-Vincent and Howardd Bussey from the McGill University in Montreal, Canada. Sincee the l,6-/?-glucan synthase itself has not yet been identified, it is also unknown if thee synthase is made up of one or more subunits. The l,3-/?-glucan synthase consists of at leastt two components: the putative catalytic subunit encoded by FKS1, which is a multipass transmembranee protein, and the cytoplasmic regulatory subunit, a small G-protein encoded byy RHOl (reviewed in Cabib et ah, 1998). Although Fksl lacks the conserved UDP-glucose bindingg site (reviewed in Klis et ah, 2002), it was recently shown that the UDP-glucose analogg 5-azido-[beta-32P]-UDP-glucose binds to the Neurospora crassa Fksl homolog (Schimoler-O'Rourkee et ah, 2003). This strongly suggests that Fksl is indeed the substrate- bindingg subunit of the 1,3-/?-glucan synthase. One may presume that the l,6-/?-glucan synthasee also requires a multipass transmembrane domain, since such a domain probably facilitatess transmembrane transport of the product as for example postulated in the case of the putativee 1,3-cc-glucan synthase Agsl in S. pombe pombe (Hochstenbach et ah, 1998). On the other hand,, the l,6-/?-glucan synthase might well consist of multiple proteins, which means that the variouss functions are not necessarily restricted to one protein. If the l,6-/?-glucan synthase hass a function analogous to that of l,3-/?-glucan synthase, it would probably also use UDP- glucosee as a substrate. The UDP-glucose analog 5-azido-[beta- P]-UDP-glucose may then be exploitedd for the identification of components of the 1,6-/?-glucan synthase as well. When the inin vitro assay for 1,6-/?-glucan synthesis is further refined so that it can be routinely used, a purificationn strategy might be used after which UDP-glucose binding proteins can be labeled withh this substrate analog. Subsequently, mass spectrometry could be used to identify these proteins.. Interestingly, Frost and co-workers labeled yeast microsomal fractions with 5-azido-

117 7 ChapterChapter 5

UDP-glucosee and identified several potential UDP-glucose binding proteins (1992). These havee not been further investigated, but this observation merits re-investigation.

5.5. Assembly of 1,6-fi-glucan in the cell wall Thee speculative assembly of 1,6-/?-glucan in the cell wall presumably consists of three enzymaticc steps: (1) elongation and remodeling of linear 1,6-/?-glucan chains to form mature celll wall 1,6-/?-glucan polymers, (2) the attachment of the 1,6-/?-glucan polymers to 1,3-/?- glucan,, and (3) the attachment of GPI-CWPs to the 1,6-/?-glucan. It was proposed by Roh and co-workerss (2002) that the order of assembly of the cell wall components is as follows: 1,3-/?- glucann is produced first, followed by the covalent addition of 1,6-/?-glucan to 1,3-/?-glucan, andd finally the mannoproteins are linked to 1,6-/?-gIucan. Chitin then is the last component to bee added to die complex. Interestingly, mutations in Fksl, putatively encoding die catalytic subunitt of 1,3-/?-glucan synthase, also affect l,6-/?-glucan levels. This might be caused by a decreasee in l,3-/?-glucan levels, and thus decreased acceptor levels for attachment of 1,6-/?- glucann at the celll surface (Dijkgraaf et al, 2002). Nonee of the enzymatic activities for covalently connecting 1,6-/?-glucan to other macromoleculess in the wall has been identified, yet some known genes might play a role in thesee steps. The Krel O-glycoprotein is probably associated with the plasma membrane throughh a GPI-anchor, and could play a role in the elongation and remodeling of 1,6-/?-glucan chainss since the average length of the 1,6-/?-glucan polymers in thee mutant strain was reduced too about 50% of the wild type. The total amount of 1,6-/?-glucan was reduced by 40% (Boone etet al, 1990; Roemer and Bussey, 1995). Krel might thus be involved in the remodeling of freshlyy synthesized l,6-/?-glucan polymers resulting in more branched chains (Figure 2). Thee Kre9 O-glycoprotein and its homolog Knhl are soluble secretory proteins that alsoo influence l,6-/?-glucan levels (Brown and Bussey, 1993; Dijkgraaf et al, 1996). The 1,6- /?-glucann in the cell wall of kre9ts cells had an altered structure, and the levels were reduced too 10 - - 20% of the wild type level (Brown and Bussey, 1993). While knhl IS. mutants did not showw defects in 1,6-/?-glucan synthesis, overexpression of KNH1 in kre9A mutants almost completelyy restored the l,6-/?-glucan levels. Moreover, deletion of both KRE9 and KNH1 was lethall (Dijkgraaf et al, 1996). This suggests a role for Kre9 and Knhl in remodeling and incorporatingg the l,6-/?-glucan (Figure 2).

118 8 GeneralGeneral Discussion

1,3-p-glucann - 1,6-P-glucan

incorporationn ..-- 1,6-p-glucan n

:: -e9/Knh1^_

remodeling g

Figuree 2. Current model for 1,6-/?-glucan synthesis. Kre5 initiates the synthesis of a primer structure in the ER. Kre6/Sknll are responsible for the maturation of this primer, indicated by an asterisk. At the cell surface, the actuall l,6-/?-glucan is synthesized using the primer. Subsequently, 1,6-/?-glucan is remodeled by Krel, and Kre9/Knhll are involved in the coupling of 1,6-/?-glucan to other cell wall macromolecules. Rho3 may directly regulatee the activity of the synthase. Kiel and Pbs2 - Hogl play opposite roles in controlling 1,6-/?-glucan levels,, probably in an indirect way.

6.6. Regulation of 1,6-fi-glucan synthesis Thee original screen for mutants that are resistant to the Kl killer toxin, did not identifyy potential regulatory proteins involved in controlling the levels of l,6-/?-glucan. However,, another genetic screen which was devised to identify genes involved in cell wall biogenesiss in general and which exploits the observation that cell wall mutants are often

119 9 ChapterChapter 5

hypersensitivee to calcofluor white, did result in the identification of genes encoding proteins involvedd in regulating 1,6-/?-glucan levels. The corresponding mutants were not only hypersensitivee to calcofluor white, but also displayed resistance to the K.1 killer toxin (Ram et ah,ah, 1994). Among these genes were PTC J, a protein phosphatase type 2C that is directly involvedd in the regulation of the Pbs2-Hogl pathway (see below) (Jiang et ah, 1995; Warmka etet ah, 2001), and KIC1, encoding a serine/threonine protein kinase that is involved in cell walll integrity (Sullivan et ah, 1998; Chapter 2). Studies on both these genes have shown that thee Pbs2-Hogl pathway, a MAP kinase pathway that regulates adaptation to high osmolarity (reviewedd in Hohmann, 2002), also appears to play a role in the regulation of 1,6-/?-glucan synthesiss (Jiang et ah, 1995; Chapter 2). The precise mechanism of how the Pbs2-Hogl pathwayy influences l,6-/7-glucan levels is unclear. The Pbs2-Hogl pathway possibly plays an indirectt role in l,6-/?-glucan synthesis by influencing the expression of genes that modulate 1,6-/?-glucann levels in the wall. Yet, it is also conceivable that the Pbs2-Hogl pathway is moree directly involved in 1,6-/7-glucan synthesis, for example by regulating the expression of geness encoding subunits of the synthase. Thee protein kinase Kiel also appears to play a role in regulating 1,6-/?-glucan synthesis.. Deletion of the KIC1 gene results in Kl killer toxin resistance and a decrease in 1,6-/?-glucann levels in the wall. Reciprocally, overexpression of the KIC1 gene results in hypersensitivityy to the killer toxin and an increase in 1,6-/?-glucan levels (Chapter 2). Kiel interactss with and is activated by Cdc31, the yeast centrin, demonstrating a novel function of Cdc311 in cell wall integrity (Sullivan et ah, 1998; Ivanovska and Rose, 2001). As there are no knownn targets of the Kiel kinase, it is unclear by which mechanism Kiel is able regulate 1,6- /?-glucann levels. One way to identify downstream targets of a kinase is by using a multicopy suppressorr screen, where cells mutated in the gene of interest are transformed with a genomic libraryy and screened for suppression of an associated phenotype of the mutation. Using this technique,, the small G-protein encoding RH03 was identified. Small G-proteins are regulated byy a set of specific proteins, such as GTPase activating proteins (GAPs), guanine nucleotide exchangee factors (GEFs), and GDP dissociating inhibitors (GDIs) (reviewed in Matozaki, 2000).. As Kiel does not belong to any of these classes of proteins, it is unlikely that Kiel directlyy interacts with Rho3. It might be possible that Kiel modulates the activity of the above-mentionedd G-protein regulatory proteins, but there is no evidence for this as a yeast two-hybridd screen using Kiel as bait did not result in the identification of such G-protein regulatorss (Ito et ah, 2001). This two-hybrid screen did not yield interactors that clarify the rolee of Kiel in cell wall biogenesis, but recently it was established in a large scale analysis of proteinn complexes by mass spectrometry that Kiel complexes with Slt2 (Ho et ah, 2002). Thee Slt2 protein is the MAP kinase of the PKC1 cell wall integrity pathway (reviewed in Heinischh et ah, 1999), and therefore represents a link to cell wall biogenesis. However, it mustt be noted that mutants of slt2 do not share the kicl defects in 1,6-/?-glucan synthesis

120 0 GeneralGeneral Discussion

(Pagee et al, 2003; Chapter 2). As pkcl mutants display Kl killer toxin hypersensitivity, whichh is not shared by components of the downstream MAPK pathway (Page et al, 2003), thiss suggests that the interaction of Kiel with Slt2 is either unrelated to 1,6-/?-glucan synthesiss or causes feedback to Pkcl. Perhaps Kiel regulates l,6-/?-glucan synthesis through onee (or more) of its two-hybrid interactors with yet unknown functions (Ito et al, 2001). In thiss model, Kiel would probably function in parallel of Rho3. Interestingly, Kiel and the Pbs2-Hogll pathway seem to have counteracting effects on 1,6-/?-glucan levels. Thee exact role of the small G-protein Rho3 in 1,6-/?-glucan biogenesis is also unknown,, but one may predict that Rho3 has a role in 1,6-/?-glucan synthesis, which is similar too the activating role of Rhol in l,3-/?-glucan synthesis (Figure 2). Although this is an appealingg scheme, another possibility is that the stimulatory role of Rho3 in exocytosis (Adamoo et al, 1999; Robinson et al, 1999) contributes to 1,6-/?-glucan synthesis, perhaps throughh the correct delivery of 1,6-/?-glucan synthetic components. The most obvious locationn for the l,6-/?-glucan synthase is at the sites of growth, as is the case for the 1,3-/?- glucann synthase (Yamochi et al, 1994; Qadota et al, 1996). Interestingly, Rho3 localizes to thee bud tip of cells with a small bud (Robinson et al, 1999), a place where active cell growth occurs.. However, this does not distinguish between a direct and an indirect role in 1,6-/?- glucann synthesis. Finally, one could speculate that the 1,6-/?-glucan synthase is co-localized withh the l,3-/?-glucan synthase, which probably would facilitate the coordination of the synthesiss of both cell wall components. Also, this would simplify the delivery of the different biosyntheticc components to the site(s) of growth.

7.. Proposed model for 1,6-fl-glucan synthesis Consideringg the data presented in this thesis and those described in the literature, we wouldd like to propose the following model for l,6-/?-glucan synthesis (Figure 2): Kre5 is involvedd in the production of a primer structure in the ER. Although more proteins are probablyy involved (discussed above), these are omitted for clarity. The proposed primer structuree is further processed by Kre6/Sknl in the Golgi. The mature primer can then functionn as an acceptor for the 1,6-/7-glucan synthase in the plasma membrane. We propose thatt the small G-protein Rho3 is a regulatory subunit of the l,6-/?-glucan synthase complex. Thee protein kinase Kiel controls the levels of 1,6-/?-glucan, counteracting the Pbs2-Hogl pathway.. Its influence may be indirect, for example through transcriptional regulation. The 1,6-/?-glucann synthase then produces l,6-/?-glucan precursor chains for cell wall assembly, possiblyy followed by the removal of the primer structure. The assembly of l,6-/7-glucan in the celll wall may involve enzymatic activities for elongation and remodeling, and the attachment too 1,3-/?-glucan. At the cell surface, Krel and Kre9/Knhl are probably involved in these processes.. The KRE1 gene is not essential, and the average chain length of 1,6-/?-glucan is reducedd by 50% in krel A mutants. This suggests that Krel is involved in the elongation and

121 1 ChapterChapter 5

remodelingg of l,6-/?-glucan. On the other hand, the KRE9/KNH1 gene pair is essential suggestingg a role in the incorporation of l,6-/?-glucan in the cell wall.

8.8. Perspectives Thee study of 1,6-/?-glucan synthesis was initiated by using a genetic approach, i.e. the K.11 killer toxin resistance screen (reviewed by Shahinian and Bussey, 2000). This screen was furtherr extended by a genome-wide screen for altered killer toxin sensitivity (Page et al, 2003).. Although many genes were identified that influence l,6-/?-glucan levels, this approach farr from solved the many questions about l,6-/?-glucan synthesis. Probably, this has to do withh the limitations of using a single genetic approach. In this thesis, l,6-/?-glucan synthesis wass shown to take place mainly at the cell surface, excluding the existence of an intracellular biosyntheticc machinery for 1,6-/?-glucan. We have further shown that Kiel and Pbs2-Hogl havee an opposite role in controlling l,6-/?-glucan levels. Also, we propose that the small G- proteinn Rho3 might play a direct role in activating the l,6-/?-glucan synthase. Finally, an in vitrovitro biochemical assay for the detection of 1,6-/?-glucan synthase activity has been developed.. This assay will contribute to finding an answer to many of the questions posed in thee beginning of this chapter, such as (i) if l,6-/?-glucan synthesis is a single- or a multistep process.. This can be done by analysis of the terminal reducing glucose to determine if it is derivedd from UDP-[14C]-glucose. The identification of the catalytic subunit of the 1,6-/?- glucann synthase is crucial for the further study of the various aspects of 1,6-/?-glucan synthesis,, as has been the case for l,3-/?-glucan synthesis. Using the l,6-/?-glucan synthase activityy assay, the purification of the activity could lead to the identification of the synthase componentss (questions iii & iv). When these components are known, it is relatively simple to studyy their exact subcellular localizations (question ii), for example by the addition of a GFP- tagg or by immunofluorescence. The assay itself might also give further insight in how 1,6-/7- glucann is synthesized, by detailed analysis of the structures of the reaction products. In addition,, knowing the structures of the l,6-/?-glucan synthase reaction products could be of valuee in the understanding of the remodeling and incorporation reactions, since the substrates forr these enzymes would then be identified (question v). Finally, the identification of the catalyticc subunit of the 1,6-/?-glucan synthase will allow the analysis of its regulation on the transcriptionall level, and the discovery of proteins interacting with this subunit. For example, interactingg proteins might be involved in the regulation of the synthase or its correct localizationn (question vi).

122 2 GeneralGeneral Discussion

References References

Adamoo J.E., Rossi G., and Brennwald P. (1999). The Rho GTPase Rho3 has a direct role inn exocytosis that is distinct from its role in actin polarity. Mol. Biol. Cell 10: 4121-4133

Alonsoo M.D., Lomako J., Lomako W.M., and Whelan W.J. (1995). A new look at the biogenesiss of glycogen. FASEBJ. 9:1126-1137

Azumaa M., Levinson J.N., Page N., and Bussey H. (2002). Saccharomyces cerevisiae Biglp,, a putative endoplasmic reticulum membrane protein required for normal levels of cell walll beta-1,6-glucan. Yeast 19:783-793

Boonee C, Sommer S.S., Hensel A., and Bussey H. (1990). Yeast KRE genes provide evidencee for a pathway of cell wall beta-glucan assembly. J. Cell Biol. 110: 1833-1843

Brownn J.L., and Bussey H. (1993). The yeast KRE9 gene encodes an O-glycoprotein involvedd in cell surface beta-glucan assembly. Mol. Cell. Biol. 13: 6346-6356

Cabibb E., Drgonova J., and Drgon T. (1998). Role of small G proteins in yeast cell polarizationn and wall biosynthesis. Annu. Rev. Biochem. 67:307-333

Castroo O., Chen L.Y., Parodi A.J., and Abeijon C. (1998). Uridine diphosphate-glucose transportt into the endoplasmic reticulum of Saccharomyces cerevisiae: in vivo and in vitro evidence.. Mol. Biol. Cell 10:1019-1030

Dijkgraaff G.J., Brown J.L., and Bussey H. (1996). The KNH1 gene of Saccharomyces cerevisiaecerevisiae is a functional homolog of KRE9. Yeast 12:683-692

Dijkgraaff G.J., Abe M., Ohya Y., and Bussey H. (2002). Mutations in Fkslp affect the cell walll content of beta-1,3- and beta-1,6-glucan in Saccharomyces cerevisiae. Yeast 19:671-690

Fernandezz F.S., Trombetta S.E., Hell man U., and Parodi A.J. (1994). Purification to homogeneityy of UDP-glucose: glycoprotein glucosyltransferase from Schizosaccharomyces pombepombe and apparent absence of the enzyme from Saccharomyces cerevisiae. J. Biol. Chem. 269:: 30701-30706

123 3 ChapterChapter 5

Heinischh J.J., Lorberg A., Schmitz H.P., and Jacoby J.J. (1999). The protein kinase C- mediatedd MAP kinase pathway involved in the maintenance of cellular integrity in SaccharomycesSaccharomyces cerevisiae. Mol. Microbiol. 32:671-680

Hoo Y., Gruhler A., Heilbut A., Bader G.D., Moore L., Adams S.L., Millar A., Taylor P., Bennettt K., Boutilier K., et at (2002). Systematic identification of protein complexes in SaccharomycesSaccharomyces cerevisiae by mass spectrometry. Nature 415:180-183

Hochstenbachh F., Klis F.M., van den Ende H., van Donselaar E., Peters P.J., and Klausnerr RD. (1998). Identification of a putative alpha-glucan synthase essential for cell walll construction and morphogenesis in fission yeast. Proc. Natl. Acad. Sci. USA 95:9161- 9166 6

Hohmannn S. (2002). Osmotic stress signaling and osmoadaptation in yeasts. Microbiol. Mol. Biol.Biol. Rev. 66: 300-372

Inouee S.B., Takewaki N., Takasuka T., Mio T., Adachi M., Fujii Y., Miyamoto C, Arisawaa M., Furuichi Y., and Watanabe T. (1995). Characterization and gene cloning of 1,3-beta-D-glucann synthase from Saccharomyces cerevisiae. Eur. J. Biochem. 231: 845-854

Itoo T„ Chiba T., Ozawa R., Yoshida M., Hattori M., and Sakaki Y. (2001). A comprehensivee two-hybrid analysis to explore the yeast protein interactome. Proc. Natl. Acad.Acad. Sci. USA 98:4569-4574

Ivanovskaa I., and Rose M.D. (2001). Fine structure analysis of the yeast centrin, Cdc31p, identifiess residues specific for cell morphology and spindle pole body duplication. Genetics 157:503-518 8

Jakobb C.A., Burda P., te Heesen S., Aebi M., and Roth J. (1998). Genetic tailoring of N- linkedd oligosaccharides: the role of glucose residues in glycoprotein processing of SaccharomycesSaccharomyces cerevisiae in vivo. Gtycobiology 8: 155-164

Jiangg B., Ram A.F.J., Sheraton J., Klis F.M., and Bussey H. (1995). Regulation of cell walll beta-glucan assembly: PTC I negatively affects PBS2 action in a pathway that includes modulationn of EXGl transcription. Mol. Gen. Genet. 248: 260-269

Kliss F.M. (1994). Cell wall assembly in yeast. Yeast 10:851-869

124 4 GeneralGeneral Discussion

Kliss F.M., Mol P., Hellingwerf K., and Brul S. (2002). Dynamics of cell wall structure in Saccharomycess cerevisiae. FEMS Microbiol. Rev. 26: 239-256

Levinsonn J.N., Shahinian S., Sdicu A.M., Tessier D.C., and Bussey H. (2002). Functional, comparativee and cell biological analysis of Saccharomyces cerevisiae Kre5p. Yeast 19: 1243- 1259 9

Luu C.F., Montijn R.C., Brown J.L., Klis F.M., Kurjan J., Bussey H., and Lipke P.N. (1995).. Glycosyl phosphatidylinositol-dependent cross-linking of alpha-agglutinin and beta 1,6-glucann in the Saccharomyces cerevisiae cell wall. J. Ceil Biol. 128: 333-340

Mannerss D.J., Masson A.J., Patterson J.C., Bjorndal H., and Lindberg B. (1973). The structuree of a beta-(l~6)-D-glucan from yeast cell walls. Biochem. J. 135: 31-36

Matozakii T., Nakanishi H., and Takai Y. (2000). Small G-protein networks: their crosstalk andd signal cascades. Cell. Signal. 12: 515-524

Meadenn P., Hill K., Wagner J., Slipetz D., Sommer S.S., and Bussey H. (1990). The yeast KRE5KRE5 gene encodes a probable endoplasmic reticulum protein required for (1—6)-beta-D- glucann synthesis and normal cell growth. Mol. Cell Biol. 10: 3013-3019

Orleann P. (1997). Biogenesis of yeast wall and surface components. In: Molecular and Cellularr Biology of the Yeast Saccharomyces, Vol. 3, Cell Cycle and Cell Biology (Pringle, J.R.,, Broach, J.R., and Jones, E.W., Eds.) Cold Spring Harbor, NY: Cold Spring Harbor Laboratoryy Press, pp. 229-362

Pagee N., Gerard-Vincent M., Menard P., Beaulieu M., Azuma M., Dijkgraaf G.J., Li H., Marcouxx J., Nguyen T., Dowse T., Sdicu A.M., and Bussey H. (2003). A Saccharomyces cerevisiaecerevisiae Genome-Wide Mutant Screen for Altered Sensitivity to Kl Killer Toxin. Genetics 163:: 875-894

Pengg L., Kawagoe Y., Hogan P., and Delmer D. (2002). Sitosterol-beta-glucoside as primer forr cellulose synthesis in plants. Science 295:147-150

Qadotaa H., Python C.P., Inoue S.B., Arisawa M., Anraku Y., Zheng Y., Watanabe T., Levinn D.E., and Ohya Y. (1996). Identification of yeast Rholp GTPase as a regulatory subunitt of 1,3-beta-glucan synthase. Science 272: 279-281

125 5 ChapterChapter 5

Ramm A.F.J., Wolters A., Ten Hoopen R., and Klis F.M. (1994). A new approach for isolatingg cell wall mutants in Saccharomyces cerevisiae by screening for hypersensitivity to calcofluorr white. Yeast 10: 1019-1030

Robinsonn N.G., Guo L., Imai J., Toh-E A., Matsui Y., and Tamanoi F. (1999). Rho3 of SaccharomycesSaccharomyces cerevisiae, which regulates the actin cytoskeleton and exocytosis, is a GTPasee which interacts with Myo2 and Exo70. Mol Cell. Biol 19: 3580-3587

Roemerr T., and Bussey H. (1995). Yeast Krelp is a cell surface 0-glycoprotein. Mol. Gen. Genet.Genet. 249:209-216

Rohh D.H., Bowers B., Riezman H., and Cabib E. (2002). Rholp mutations specific for regulationn of beta(l->3)glucan synthesis and the order of assembly of the yeast cell wall. Mol.Mol. Microbiol. 44:1167-1183

Schimoler-O'Rourkee R., Renault S., Mo W., and Selitrennikoff C.P. (2003). Neurospora crassacrassa FKS Protein Binds to the (l,3)beta-Glucan Synthase Substrate, UDP-Glucose. Curr. Microbiol.Microbiol. 46:408-412

Shahiniann S., Dijkgraaf G.J., Sdicu A.M., Thomas D.Y., Jakob C.A., Aebi M., and Busseyy H. (1998). Involvement of protein jV-glycosyl chain glucosylation and processing in thee biosynthesis of cell wall beta-l,6-glucan of Saccharomyces cerevisiae.Genetics 149:843- 856 6

Shahiniann S., and Bussey H. (2000). P-1,6-Glucan synthesis in Saccharomyces cerevisiae. MolMol Microbiol 35: 477-489

Shematekk E.M., Braatz J.A., and Cabib E. (1980). Biosynthesis of the yeast cell wall. I. Preparationn and properties of beta-(l leads to 3)glucan synthetase. J. Biol. Chem. 255: 888- 894 4

Sullivann D.S., Biggins S., and Rose M.D. (1998). The yeast centrin, cdc31p, and the interactingg protein kinase, Kiclp, are required for cell integrity. J. Cell Biol 143: 751-765

Warmkaa J., Hanneman J., Lee J., Amin D., and Ota I. (2001). Ptcl, a type 2C Ser/Thr phosphatase,, inactivates the HOG pathway by dephosphorylating the mitogen-activated proteinn kinase Hog 1. Mol Cell Biol. 21: 51-60

126 6 GeneralGeneral Discussion

Warneckee D., Erdmann R., Fahl A., Hube B., Muller F., Zank T., Zahringer U., and Heinzz E. (1999). Cloning and functional expression of UGT genes encoding sterol glucosyltransferasess from Saccharomyces cerevisiae, Candida albicans, Pichia pastoris, and DictyosteliumDictyostelium discoideum, J. Biol. Chem. 274:13048-13059

Yamochii W., Tanaka K., Nonaka H., Maeda A., Musha T., and Takai Y. (1994). Growth sitee localization of Rhol small GTP-binding protein and its involvement in bud formation in SaccharomycesSaccharomyces cerevisiae. J. Cell Biol. 125: 1077-1093

127 7

Summary Summary

Yeastss and fungi are surrounded by a protective cell wall, ensuring the integrity of the cell. Sincee these cells are constantly challenged by a variable environment, their cell walls must be highlyy adaptable to match the various forms of stress they encounter. The synthesis of cell walll components is a tightly regulated process, facilitating the adaptational capacity of the celll wall. The cell wall of baker's yeast - Saccharomyces cerevisiae - consists of about equal amountss of mannoproteins and carbohydrates. As described in Chapter 1, the mannoproteins aree synthesized and modified throughout the secretory pathway. They emerge at the cell surfacee as separate moieties, which are subsequently incorporated into the cell wall. The enzymess responsible for this incorporation mechanism, however, are still unknown. The carbohydratess can be subdivided into glucans and chitin, the latter only representing a minor percentagee of the wall. The glucans can be further subdivided into 1,3- and 1,6-/?-glucan. Aboutt 80% of the glucan consists of 1,3-/Minked glucose, which is produced by the 1,3-/?- glucann synthase complex. The activity of this complex can be studied in vitro, and the complexx is known to have a plasma membrane bound catalytic subunit ,thought to be encodedd by FKS1. Rhol, a small GTP-binding protein that binds to Fksl, is the regulatory subunit.. This facilitates tight regulation of the synthesis of l,3-/?-glucan. Thee other 20% of the total cell wall glucan is composed of l,6-/?-glucan. This polymer tetherss cell wall proteins to 1,3-/?-glucan and therefore plays a pivotal role in cell wall integrity.. The observation that the protein kinase Kiel and the small G-protein Rho3 are involvedd in 1,6-/?-glucan synthesis (Chapter 2), introduces two candidates for a regulatory functionn in this process. It is unknown how these two proteins influence 1,6-/?-glucan synthesis.. Kiel may modulate transcription of genes involved in l,6-/?-glucan synthesis. In thee conditional ?CAU'KIC1 mutant the expression of two cell wall proteins is upregulated, however,, a more general approach based on DNA microarrays may yield more insight on this.. Using the Kl killer toxin, which binds l,6-/?-glucan and then kills cells, many mutants havee been isolated with decreased levels of 1,6-/?-glucan in the cell wall. As the correspondingg (KRE) genes were localized throughout the secretory pathway, it was long thoughtt that 1,6-yS-glucan was synthesized intracellularly via a multi-step process. However, somee of these genes were found to be involved in jV-glycosylation. The gene products of KRE6KRE6 and SKN1 were shown to have similarities to "family 16 glycoside hydrolases" (Chapterr 3), which basically excludes their direct function in 1,6-/?-glucan synthesis. Furthermore,, it is demonstrated in Chapter 3 that the bulk of 1,6-/?-glucan synthesis seems to takee place at the cell surface. This evidence was provided in two ways: (1) immunogold labelingg of cells that accumulate post-Golgi secretory vesicles, using antibodies directed

129 9 Summary Summary

againstt 1,6-/?-glucan, and (2) fractionation of microsomes by gel filtration, followed by detectionn of l,6-/?-glucan. In both cases, the bulk of 1,6-/?-glucan was detected at the cell surfacee or in cell surface-derived vesicles. Inn order to get conclusive data on the localization and regulation of 1,6-/?-glucan synthesis,, the development of an assay for 1,6-/?-glucan synthase activity is crucial. In Chapterr 4 the first steps are described of the development of such an assay, including both a qualitativee and a quantitative approach, using a dot blot and an enzyme immunoassay, respectively.. In collaboration with Manon Gerard-Vincent and Howard Bussey, McGill University,, Canada, these methods are currently further being optimized.

130 0 Samenvatting Samenvatting

Gistenn en schimmels zijn omgeven door een beschermende celwand, die zorg draagt voor de integriteitt van de cel. Aangezien gisten en schimmels continu door een veranderlijke omgevingg worden belaagd, moet hun celwand een zeer hoog aanpassingsvermogen hebben omm het hoofd te kunnen bieden aan de verschillende vormen van stress. De synthese van de celwandd componenten is een proces dat nauwgezet wordt gereguleerd, wat het aanpassingsvermogenn van de cel mogelijk maakt. De celwand van bakkersgist - SaccharomycesSaccharomyces cerevisiae - bestaat ongeveer uit gelijke hoeveelheden mannoproteïnen en koolhydraten.. Zoals beschreven in Hoofdstuk 1, worden de mannoproteïnen gesynthetiseerd enn gemodificeerd tijdens de passage door de secretie route en komen als afzonderlijke eenhedenn te voorschijn aan het cel oppervlak, die vervolgens in de celwand geïncorporeerd worden.. De enzymen die verantwoordelijk zijn voor de incorporatie in de celwand zijn nog steedss onbekend. De koolhydraten kunnen onderverdeeld worden in glucanen en chitine, waarvann het laatste slechts een gering percentage van de celwand beslaat. De glucanen kunnenn verder onderverdeeld worden in 1,3- en 1,6-/?-glucanen. Ongeveer 80% van het glucaann bestaat uit l,3-/?-gekoppelde glucose eenheden, hetgeen wordt geproduceerd door het 1,3-/?-glucaann synthase complex. De activiteit van dit complex kan in vitro bestudeerd worden.. Het wordt aangenomen dat de plasma membraan gebonden katalytische eenheid wordtt gecodeerd door het FKS1 gen. Het kleine G-eiwit Rhol dat aan Fksl bindt is de regulatoiree eenheid van dit complex. Hierdoor kan de synthese van 1,3-/?-glucaan zeer nauwkeurigg worden gereguleerd. Dee andere 20% van de totale hoeveelheid celwand glucaan bestaat uit 1,6-/?-glucaan. Ditt polymeer verbindt de celwand eiwitten met het 1,3 -/?-glucaan, en speelt daarom een essentiëlee rol in de integriteit van de celwand. De observatie dat het proteïne kinase Kiel en hett kleine G-eiwit Rho3 betrokken zijn bij de synthese van 1,6-/?-glucaan (Hoofdstuk 2), introduceertt twee nieuwe kandidaten die een regulerende functie zouden kunnen hebben bij ditt proces. Het is echter onbekend hoe deze eiwitten de synthese van 1,6-/?-glucaan kunnen beïnvloeden.. Een mogelijkheid is dat Kiel de transcriptie beïnvloedt van andere genen die betrokkenn zijn bij l,6-/?-glucaan synthese. In het geval van de conditionele PGALI'KICI mutantt is de expressie van twee celwand eiwitten verhoogd, maar wellicht kan een meer algemenee benadering zoals de DNA microarray hier meer inzicht in geven. Met behulp van hett KI killer toxine, dat aan l,6-/?-glucaan bindt en vervolgens cellen doodt, zijn er veel mutantenn geïsoleerd met verlaagde hoeveelheden l,6-/?-glucaan in de celwand. Er werd lang verondersteldd dat de synthese van 1,6-/7-glucaan een stapsgewijs proces is dat intracellulair plaatss vindt. De reden hiervan was dat de genproducten van de KRE genen zich bevinden op

131 1 Samenvatting Samenvatting

verschillendee plaatsen van de secretie route. Er werd echter ontdekt dat een aantal van deze genenn betrokken zijn bij TV-glycosylering. Weer andere genen, KRE6 en SKN1, bleken voor genproductenn te encoderen die gelijkenis vertonen met glycoside hydrolases (Hoofdstuk 3), hetgeenn min of meer uitsluit dat ze direct betrokken zijn bij l,6-/?-glucaan synthese. Bovendienn wordt in Hoofdstuk 3 beschreven dat het grootste deel van 1,6-/?-glucaan synthese aann het cel oppervlak plaats lijkt te vinden. Het bewijs hiervoor is op twee manieren verkregen:: (1) door immunogoud labeling van cellen die post-Golgi secretieblaasjes ophopen, mett behulp van antilichamen tegen 1,6-/?-glucaan, en (2) door het fractioneren van microsomenn door middel van gel filtratie gevolgd door immunologische detectie van 1,6-/?- glucaan.. In beide gevallen kon 1,6-/?-glucaan slechts aan het cel oppervlak aangetoond wordenn of in blaasjes afkomstig van de plasma membraan. Tenn einde een definitief beeld te krijgen van de lokalisatie en regulatie van 1,6-/7- glucaan,, is een manier om de activiteit van 1,6-/?-glucaan synthase te meten essentieel. Hoofdstukk 4 beschrijft de eerste stappen van de ontwikkeling van zowel een kwalitatieve als eenn kwantitatieve methode om 1,6-/?-glucaan synthese te meten. De eerstgenoemde maakt gebruikk van een dot blot, de tweede van een enzym immunoassay. In samenwerking met Manonn Gérard-Vincent en Howard Bussey (McGill University, Canada) worden deze techniekenn verder ontwikkeld.

132 2 Dankwoord Dankwoord

Watt heb ik hier lang naar uitgekeken! Eerlijk gezegd heb ik me vanaf het begin van m'n AIO-tijdd al verheugd op het schrijven van een dankwoord, niet omdat ik er zo snel mogelijk vanaff wilde zijn maar omdat me dat gewoon leuk leek. Bovendien is dit vaak het meest gelezenn deel van een proefschrift, en verdient dus een weloverwogen stukje tekst. Zoals waarschijnlijkk gebruikelijk, is ook dit dankwoord een "sluitpost op de begroting" en schrijf ik ditt met de hete adem van de drukker in mijn nek. Ik hoop dan ook van harte dat ik niemand vergeet. .

Omm maar eens een open deur in te trappen: het maken van een proefschrift doe je niet alleen. Gedurendee mijn promotieonderzoek ben ook ik er achter gekomen dat onderzoek doen een teamsportt is, en dat alhoewel de bijdrage van sommige teamleden soms schijnbaar onzichtbaarr is, deze toch van wezenlijk belang kan zijn.

Frans,, na een kijkje in de keuken tijdens één van mijn stages werd ook ik deel van je "gistgroep".. Ik wil je dan ook bedanken voor het vertrouwen in mij, en de onderzoeksvrijheid diee je me geschonken hebt. Ook je nimmer aflatend enthousiasme heb ik enorm gewaardeerd. Jee bent altijd bereikbaar geweest en de snelheid waarmee je manuscripten van commentaar kann voorzien is onovertroffen. Ik verbaas me er nog steeds over hoe je door een paar woorden tee veranderen een alinea veel krachtiger kan maken. De afgelopen jaren heb ik dan ook met zeerr veel plezier doorgebracht. Herman,, jij was in het begin wat meer op de achtergrond aanwezig, maar door de legendarischee muziekavonden maakte je toch een onvergetelijke indruk op mij. In een later stadiumm werd jouw rol steeds duidelijker, en ik wil je dan ook bedanken voor de snelheid waarmeee je manuscripten hebt nagekeken en gecorrigeerd, alswel voor de steun op 't laatst. Hans,, hoewel je pas halverwege mijn AIO-tijd ten tonele verscheen, heb ik toch enorm veel vann je geleerd. Je was een steun en toeverlaat (zowel wetenschappelijk als niet- wetenschappelijk),, en een ideale "sparring-partner" voor een partijtje brainstormen. Enorm bedanktt daarvoor. Ik denk nog altijd met plezier terug aan de tijd tegenover elkaar op 't lab enn samen op de zitkamer.

Mijnn eerste echte kennismaking met de wetenschap was tijdens m'n stages. Roy, na het schrijvenn van een scriptie, mocht ik onder jouw bezielende begeleiding het lab op om het allemaall zelf te doen. Daar heb ik geweldig van genoten, en mijn keuze om "verder te gaan" inn de wetenschap is hierdoor bepaald. Onze gezamenlijke voorliefde voor de muziek van Jimi

133 3 Dankwoord Dankwoord

Hendrixx uitte zich soms door scheurende gitaarsolo's van hem op het isotopenlab. Ook heb ik genotenn van het samen in een band spelen, waar ik ook op het gebied van gitaarspelen veel vann je heb geleerd. Het wordt tijd om weer eens een biertje te drinken in 't Sluisje. John,, when I started my research training at the NKI, you were still in Oz for your holiday. I wass told that it was a risk not seeing your supervisor before starting, but it turned out to be allright.. I had a great time there and you introduced me to the wonders of molecular biology. Manyy thanks for that, as well as for introducing me to Led Zeppelin. By now I've collected alll their albums. I hope you'll be around, but if not I hope to see you at another occasion.

Nuu komen natuurlijk mijn collega's uit de gistgroep. Jack, ook al was je al snel weer weg uit dee gistgroep toen ik als AIO (of eigenlijk bursaal) begon, toch heb je mij in die korte tijd op dee rit geholpen. Gelukkig was je na je transfer naar de Fyto's toch nog bereikbaar en bleef je altijdd geïnteresseerd. Verder ook de "oude" gistgroep, bestaande uit Arthur, Heleen, Kappie, Marianne,, Piet van E., Stephan en natuurlijk Roy. Jullie heb ik al tijdens m'n stageperiode lerenn kennen en vormden een goede reden om bij de gistgroep te willen horen. Om Alleen-In- Opleidingg te zijn is niet zo leuk. Gertien en Sylvia, ik ben blij dat jullie er ook waren. Behalvee dat we een enorm gezellige tijd hadden, kon ik bij jullie ook m'n hart luchten wanneerr het me zo nu en dan allemaal eventjes te veel werd. Dan hebben we natuurlijk ook nogg Frans H., Piet de G., Sandra (wat zal jij een lieve juf zijn), Miekje (het vuurspuwende begintt er inmiddels wel wat af te raken), Els (bedankt voor de "famous candle-light suppers"),, Michiel (ouwe brombeer), Bas (ouwe Goorlap) en natuurlijk Hans. Also I really enjoyedd that you, Neta and Jay, have spent about a year in the group. Blijven natuurlijk nog overr de studenten die me hebben bijgestaan, Jacco en Merijn. Jacco, jongen, je hebt me inmiddelss bijna ingehaald! Veel succes nog met de zware laatste loodjes. Merijn, jij kreeg mij minn of meer cadeau toen Hans de gistgroep verliet, maar het is allemaal toch nog goed gekomen.. Ook jij veel succes met je promotieonderzoek!

Duringg my stay in Salamanca, Spain, the basis for the 1,6-/?-glucan synthase assay was produced.. I would like to thank everybody overthere for their warm welcome. Especially I wantt to thank Juan Carlos and Angel for their help, advice and hospitality. Also I would like too say to Hugo, Juan Carlos Jr., Victoria, Elena, and Teresa: "Muchas gracias por el buen tiempo!".. Manon and Howard, I'm very glad you are willing to continue to work on the 1,6- /?-glucann synthase assay, and I wish you the best of luck.

Niett te vergeten wil ik de overige Plantenfysiologen bedanken die er gedurende mijn tijd waren:: Teun, Alan, Steven, Harold, John, Ana,, Appie, Hans H., Hilda, Michel, Wies, Jet en Eddy.. Ook de huidige generatie wil ik niet overslaan, met name Bastiaan en Juul: wanneer doenn we weer eens een "Scientific Jam"? En natuurlijk ook de Fyto's die ook altijd van de

134 4 Dankwoord Dankwoord

partijj waren. Last but not least wil ik Kitty bedanken (en natuurlijk "broertje" Jasper) voor de gezelligheidd en alle wijze raad. Allemaal bedankt voor de BBQ's, borrels, en natuurlijk ons verblijff in 't Kanohoes in Groningen. Het was een fantastische tijd!

Ikk mag natuurlijk niet m'n nieuwe collega's van het AMC Lever Centrum vergeten. Gelukkig blijktt maar weer eens dat wetenschappers vaak gezellige mensen zijn. Ik heb het gevoel dat dee tijd vliegt, en dat gebeurt niet wanneer je je verveelt. Allemaal bedankt!

Dann blijven vrienden, bandleden, motormuizen en familieleden over, die misschien niet allemaall zo direct betrokken zijn geweest bij het maken van een proefschrift, maar wel bij alless eromheen. Dus jullie zijn minstens net zo belangrijk geweest, en dat zijn jullie nog steeds.. Bedankt daarvoor! Paa en ma, jullie hebben mij altijd gestimuleerd om door te leren en het ook mogelijk gemaakt omm dat te doen. Ook al was het soms niet te volgen wat ik aan het doen was, jullie zijn altijd geïnteresseerdd gebleven en hebben mij onvoorwaardelijk gesteund. Ik ben blij dat jullie dat zoo gedaan hebben, want anders zou dit boekje er nooit zijn geweest. Tsja,, dan blijft er natuurlijk nog één iemand over. Die iemand had ik al eerder bij de Plantenfysiologenn kunnen noemen, maar ik heb er toch maar mee gewacht tot nu. Eigenlijk is zijj voor mij nog een belangrijker eindresultaat dan dit boekje, al zou ik zonder de aanzet tot ditt boekje haar waarschijnlijk nooit ontmoet hebben. Dus ook vanuit dat oogpunt had ik dit nooitt willen missen. Martine, je bent een schat!

135 5