K. Lactis Toxin Is a Trna Endonuclease

The Kluyveromyces lactis killer toxin is a transfer RNA endonuclease

Jian Lu

Department of Molecular Biology Umeå University Umeå, Sweden 2007

Doctoral Dissertation 2007 Department of Molecular Biology Umeå University SE-901 87, Umeå, Sweden

ISBN: 978-91-7264-303-1

Printed by Solfjädern Offset AB Umeå, Sweden

2 TABLE OF CONTENTS

ABSTRACT...... 4 PAPERS IN THIS THESIS...... 5 1 INTRODUCTION 1.1 Kluyveromyces lactis killer toxin (zymocin) 1.1.1 Zymocin-encoding linear plasmids...... 6 1.1.2 Biogenesis of zymocin…...... 8 1.1.3 Early events in zymocin toxicity…………………………………….…...9 1.1.4 Mechanism of growth inhibition caused by zymocin………………..…10 1.1.5 Immunity ……………………………………………………………...... 13 1.1.6 Other linear plasmids encoding killer toxin ……………………...... …13 1.2 Transfer RNA modification 1.2.1 Transfer RNA (tRNA)……………………………………………..…….14 1.2.2 Biogenesis and functions of modified nucleosides in tRNA………...…17 1.3 Transfer RNA anticodon nucleases 1.3.1 Reaction mechanism of ribonucleases……………………………….....19 1.3.2 PrrC……………………………………………………………………....21 1.3.3 Colicin E5………………………………………………………………..22 1.3.4 Colicin D…………………………………………………………………23 2 RESULTS AND DISCUSSION 2.1 The K. lactis γ-toxin is a tRNA anticodon nuclease (Paper I) ……….……....25 2.2 Determinants in tRNA required for γ-toxin cleavage (Paper II)………...... 28 2.3 γ-toxin as a tool to identify mutants defective in the formation of modified wobble uridines (Paper III)…………..…………...31 3 CONCLUSIONS………………………………………………..……..…….35 ACKNOWLEDGEMENTS………………………………………..…….……..36 REFERENCES………………………………………………………….…..…....37 PAPERS I-III

3 ABSTRACT

Killer strains of the yeast Kluyveromyces lactis secrete a heterotrimeric protein toxin (zymocin) to inhibit the growth of sensitive yeasts. The cytotoxicity of zymocin resides in the γ subunit (γ-toxin), however the mechanism of cytotoxicity caused by γ-toxin was previously unknown. This thesis aimed to unravel the mode of γ-toxin action and characterize the interaction between γ-toxin and its substrates. Previous studies suggested a link between the action of γ-toxin and a distinct set of transfer RNAs. In paper I, we show that γ-toxin is a tRNA anticodon endonuclease which cleaves tRNA carrying modified nucleoside 5-methoxycarbonylmethyl-2-thiouridine (mcm 5 s 2 U) at position 34 (wobble position). The cleavage occurs 3’ to the wobble uridine and yields 2’, 3’-cyclic phosphate and 5´-hydroxyl termini. In paper II, we identified the determinants in tRNA important for efficient γ-toxin cleavage. The modifications present on the wobble uridines have different effects on tRNA cleavage by γ-toxin. The Saccharomyces cerevisiae wobble modification mcm 5 group has a strong positive effect, whereas the Escherichia coli wobble modification 5-methylaminomethyl group has a strong inhibitory effect on tRNA cleavage. The s 2 group present in both S. cerevisiae and E. coli tRNAs has a weaker positive effect on the cleavage. The anticodon stem loop (ASL) of tRNA represents the minimal structural requirement for γ-toxin action.

Nucleotides U 34 U 35 C 36 A 37 C 38 in the ASL are required for optimal cleavage by γ-toxin, whereas a purine at position 32 or a G at position 33 dramatically reduces the reactivity of ASL. Screening for S. cerevisiae mutants resistant to zymocin led to the identification of novel proteins important for mcm 5 s 2 U formation (paper III). Sit4p (a protein phosphatase), Sap185p and Sap190p (two of the Sit4 associated proteins), and Kti14p (a protein kinase) are required for the formation of mcm 5 side chain. Ncs2p, Ncs6p, Urm1p, and Uba4p, the latter two function in a protein modification (urmylation) pathway, are required for the formation of s 2 group. The gene product of YOR251C is also involved in the formation of s 2 group. The involvement of multiple proteins suggests that the biogenesis of mcm 5 s 2 U is very complex.

Key words: K. lactis γ-toxin, tRNA endonuclease, modified nucleosides, anticodon stem loop, 5-methoxycarbonylmethyl-2-thiouridine, 5-methylaminomethyl-2-thiouridine.

4 PAPERS IN THIS THESIS

This thesis is based on the following publications referred to in the text by their roman numbers (I-III)

I The Kluyveromyces lactis γ-toxin targets tRNA anticodons.

Lu J, HUANG B, ESBERG A, JOHANSSON M.J.O, and BYSTRÖM A.S.

(2005)

RNA 11:1648-1654.

Reprinted with the permission from RNA society.

II Identification of determinants in tRNA required for Kluyveromyces lactis

γ-toxin cleavage.

Lu J, ESBERG A, HUANG B, and BYSTRÖM A.S. (2007)

Manuscript

III A genome-wide screen in Saccharomyces cerevisiae for mutants defective

in the formation of wobble nucleoside 5-methoxycarbonylmethyl-2-

thiouridine.

Lu J*, HUANG B*, and BYSTRÖM A.S. (2007)

These authors contributed equally to this work

Manuscript

5 INTRODUCTION

1.1. Kluyveromyces lactis killer toxin (zymocin) Secreting toxins to inhibit the growth of other microorganisms is a common phenomenon in microbial competition. Protein toxins secreted by yeast that are able to kill other yeast cells are termed killer toxins. Killer yeasts are toxin-producing fungi that are immune to their own killer toxins. The first yeast killer toxin was described in the yeast Saccharomyces cerevisiae in the early 1970s. This toxin was found to be associated with double-stranded RNA (dsRNA) viruses (Bevan et al., 1973). Subsequently, killer toxins have been discovered in many yeast species and are found to be associated with, in addition to dsRNA virues, linear dsDNA plasmids, and chromosomal genes (for review, Magliani et al., 1997). Killer strains of the dairy yeast Kluyveromyces lactis secrete a hetrotrimeric protein toxin (zymocin) to inhibit the growth of other yeast strains. In addition to non-killer K. lactis strains, zymocin sensitive yeasts include Kluyveromyces thermotolerans, Kluyveromyces vanudenii, Candida intermedia, Candida utilis, Saccharomyces italicus, Saccharomyces rouxii, S. cerevisiae, and Torulopsis glabrata (Gunge and Sakaguchi, 1981; Gunge et al., 1981)

1.1.1 Zymocin-encoding linear plasmids

The production of zymocin is dependent on two linear, double-stranded DNA plasmids (pGKL1 and pGKL2) which reside in the cytosol. This is distinct from other yeast toxins that are encoded by double-stranded RNA plasmids or chromosomal genes (Magliani et al., 1997). The plasmid pGKL1 is 8,874 base pairs in length and contains four open reading frames (ORFs). As shown in Figure 1, there are genes encoding the three zymocin subunits (α and β: ORF2; γ: ORF4) (Stark and Boyd, 1986) and an immunity determinant (ORF3) (Tokunaga et al., 1987a). The remaining gene codes for a

6 pGKL1-specific DNA polymerase (ORF1) (Stark and Boyd, 1986). The larger plasmid pGKL2 is 13,457 base pairs in length and contains eleven ORFs. Many of them encode proteins required for maintenance of the linear plasmids: ORF2 for a DNA polymerase and a terminal protein specific for pGKL2 (Takeda et al., 1996; Tommasino et al., 1988), ORF6 and ORF7 for plasmid specific RNA polymerases (Schaffrath et al., 1997; Wilson and Meacock, 1988), ORF3 for a mRNA capping protein (Larsen et al., 1998). Several ORFs encode putative maintenance factors: ORF5 for a single-stranded DNA binding protein (Schaffrath and Meacock, 2001), ORF10 for a replication initiator (McNeel and Tamanoi, 1991), and ORF4 for a helicase (Tommasino et al., 1988). There are four ORFs with undefined functions.

ORF1 is known to be dispensable for maintenance of the linear plasmids and immunity (Schaffrath et al., 1992), whereas ORF8 and ORF9 are essential for maintenance of the plasmids (Schaffrath et al., 1999). The function of ORF11 has not been studied (Larsen and Meinhardt, 2000).

DNP Zymocin(αβ) Zymocin(γ) pGKL1 1 2 4 3 TIR Immunity

Helicase?SSB? RNP TRF1 pGKL2 1 11 4 5 6 9 10 2 3 7 8 TIR TP/DNP Cap-enzyme RNP

Figure 1. Genetic organization of the killer plasmids pGKL1 and pGKL2. DNP: DNA polymerase, RNP: RNA polymerase, SSB: single-stranded DNA binding protein, TIR: terminal inverted repeats, TP: terminal proteins, TRF1: terminal recognition factor 1. The ORFs in black denote that they are functionally unassigned. Circles of different colors denote pGKL1 or pGKL2 specific terminal proteins (adapted from Schaffrath and Meinhardt, 2005).

7 Both linear plasmids contain unique terminal inverted repeats and terminal proteins. The presence of terminal proteins and plasmid specific DNA polymerases is reminiscent of the replication of adenovirus and some bacteriophages, suggesting a virus-like protein-primed plasmids replication model (Meijer et al., 2001; Stark et al., 1990). The genes on the linear plasmids are transcribed from plasmid specific promoters containing u pstream c onserved s equence (UCS) that are incompatible with the host RNA polymerase (Kamper et al., 1991; Romanos and Boyd, 1988;

Schaffrath et al., 1996; Schickel et al., 1996; Wilson and Meacock, 1988). Taken together, the linear plasmids are replicated and transcribed independent of the host replication and transcription machinery.

1.1.2 Biogenesis of zymocin

The comparison of polypeptides secreted by killer strain and non-killer strain of K. lactis revealed that zymocin is composed of three subunits. These subunits are termed α, β, and γ according to their molecular weights in decreasing order (99, 30, and 28 kD, respectively) (Stark and Boyd, 1986). Determination of the amino-termini of purified zymocin subunits showed that the

α and β subunits are encoded by ORF2, and the γ subunit is encoded by ORF4 in pGKL1. The Kex1 endopeptidase was suggested to cleave the primary polypeptide translated from ORF2 to generate the matured amino-termini of α and β subunits. On the other hand, the γ subunit is likely to be processed by a signal peptidase (Julius et al., 1984; Stark and Boyd, 1986; von Heijne, 1983). The processing of α and β subunits is required for zymocin secretion, since a mutation in the KEX1 gene abolished the secretion of zymocin (Wesolowski-Louvel et al., 1988). The affinity of zymocin to the glycoprotein-binding resin (Sugisaki et al., 1984) and the inhibitory effect of glycosylation inhibitor on the synthesis of zymocin (Sugisaki et al., 1985) suggested that zymocin is a glycoprotein. The different

8 sensitivities of zymocin subunits to endo-β-N-acetylglucosaminidase H (endo H) indicated that the α subunit contains a single, N-linked oligosaccharide chain, whereas the β and γ subunits are not glycosylated (Stark and Boyd, 1986). The disulfide bonds are important for the structural integrity of zymocin, since disruption of disulfide bonds in zymocin resulted in dissociation of the β and γ subunits and abolished toxin activity (Stark and Boyd, 1986; Tokunaga et al., 1987a). The secretion of zymocin is coupled with its assembly, since the γ subunit can not be secreted when expressed alone in yeast (Tokunaga et al., 1989; Tokunaga et al., 1990). This is surprising because the γ subunit contains a functional secretion signal that when fused to other proteins can target their secretion (Baldari et al., 1987;

Tokunaga et al., 1987b). One possible explanation for this may be that the γ subunit requires the other subunits for optimal folding. Alternatively, the α subunit might aid the formation of disulfide bonds between the β and γ subunits, and/or provide the glycosylated entity that is typical for secreted proteins (Stark et al., 1990).

1.1.3 Early events in zymocin toxicity

S. cerevisiae mutants resistant to zymocin have been a crucial tool to understand how zymocin enters cells and inhibits growth. Mutants termed skt (s ensitivity to K . lactis t oxin) (Kawamoto et al., 1993), iki (i nsensitive to killer) (Kishida et al., 1996), kti (K . lactis t oxin i nsensitive) (Butler et al., 1991b; Butler et al., 1994), tot ( toxin-t arget) (Frohloff et al., 2001) have been isolated in different studies. The γ subunit is the cytotoxic subunit of zymocin (hence the name γ-toxin), since its expression in toxin sensitive cells causes similar inhibitory effect as zymocin treatment (Butler et al., 1991b; Tokunaga et al., 1989). However, exogenously applied γ-toxin is not toxic, suggesting that the α and β subunits are required for γ-toxin entry. The α subunit contains a chitin binding domain and a chitinase motif. Consistently, it is able to bind chitin in vitro and displays exo-chitinase activity (Butler

9 et al., 1991a; Jablonowski et al., 2001b). This suggests that the initial step of zymocin docking could involve chitin binding and hydrolysis. In support to this notion, mutants affected in the chitin synthesis (Chs3, Chs7, and Chs4) are resistant to exogenous zymocin (Jablonowski et al., 2001b), because zymocin entry is blocked in these mutants (Mehlgarten and Schaffrath, 2004). How the γ-toxin is imported into the cell after zymocin docking is poorly understood. Two components were found to be involved in the events after zymocin docking but before γ-toxin acting on the target. Mutations in PMA1/KTI10 gene, which encodes the major plasma membrane H + pump, confer resistance to exogenous zymocin, but not to intracellularly expressed γ-toxin. It is likely that the Pma1 protein is required to promote γ-toxin trafficking and/or cytosolic release after entry, because γ-toxin is able to enter pma1 cells treated with exogenous zymocin (Mehlgarten and Schaffrath, 2004). The mannosyl-diinositolphospho-ceramide

(M(IP) 2 C), the major plasma membrane sphingolipid, was shown to be important for

γ-toxin entry. Mutations in M(IP) 2 C synthase (encoded by IPT1/KTI6) and other proteins involved in M(IP) 2 C synthesis all cause resistance to exogenous zymocin but not to intracellularly expressed γ−toxin (Zink et al., 2005). In contrast to pma1 cells, the entry of γ-toxin is blocked in the ipt1 cells.

1.1.4 Mechanism of growth inhibition caused by zymocin

According to the reaction to intracellular γ-toxin expression, zymocin resistant mutants can be divided into two classes (Butler et al., 1991b). The class I mutants are resistant to extracellular zymocin, but sensitive to intracellular expression of γ-toxin. These mutants are affected in early events before the γ-toxin acts on the target, including those with altered chitin synthesis, plasma membrane H + pump, or sphingolipid synthesis. Class II mutants are resistant to both extracellular zymocin and intracellularly expressed γ-toxin, and are thus target mutants. Numerous class II

10 mutants have been identified (Table 1), however, the target of γ-toxin remained unclear before this study.

Table 1. Class II zymocin resistant mutants

Gene(s) mutated Description References

ELP1 Subunit of Elongator complex (Frohloff et al., 2001)

ELP2 Subunit of Elongator complex (Frohloff et al., 2001)

ELP3 Subunit of Elongator complex (Frohloff et al., 2001)

ELP4 Subunit of Elongator complex (Jablonowski et al., 2001c)

ELP5 Subunit of Elongator complex (Frohloff et al., 2001)

ELP6 Subunit of Elongator complex (Jablonowski et al., 2001c)

KTI11 Protein associated with Elongator complex, required for synthesis of diphthamide (Butler et al., 1991a)

KTI12 Protein associated with Elongator complex (Butler et al., 1991a)

KTI13 Function unknown (Butler et al., 1991a)

(Mehlgarten and Schaffrath, KTI14 Protein kinase 2003)

SIT4 Protein phosphatase (Jablonowski et al., 2001a)

SAP185 SAP190 Sit4 associated proteins (Jablonowski et al., 2001a)

URM1 Ubiquitin-like protein conjugation system (Fichtner et al., 2003)

Intracellular γ-toxin expression causes sensitive budding yeast cells to arrest at

G 1 stage of the cell cycle, evidenced by the unbudded morphology and an unreplicated (1n) DNA content (Butler et al., 1991b). It was first proposed that the cell cycle effect of zymocin was caused by inhibiting the adenylate cyclase and depleting cAMP which is essential for mitotic growth and cell division (Sugisaki et al., 1983). However, this hypothesis was later disproved (White et al., 1989).

11 Characterization of class II zymocin resistant mutants has led to the hypothesis that the Elongator is the target of γ-toxin (Frohloff et al., 2001). The Elongator complex is composed of Elongator p roteins 1 to 6 (Elp1 to Elp6) and has been proposed to participate in various cellular processes; elongation of RNA polymerase II transcription (Krogan and Greenblatt, 2001; Otero et al., 1999; Winkler et al., 2001), regulation of exocytosis (Rahl et al., 2005), and formation of modified nucleosides in tRNA (Huang et al., 2005). In addition to other phenotypes such as G1 cell cycle delay, slow growth, and delayed transcriptional activation (Frohloff et al., 2001; Wittschieben et al., 1999), defects in the Elongator result in class II zymocin resistance (Frohloff et al., 2001; Jablonowski et al., 2001c). However, in contrast to zymocin treatment, deletion of elongator is not lethal to the cell, suggesting the Elongator itself might not be the target of γ toxin. Defects in some of the class II resistant mutants are believed to be related to the

Elongator function. For example, Kti11p and Kti12p are associated with the Elongator and are suggested to regulate functions of Elongator (Fichtner et al., 2002a; Fichtner et al., 2002b; Fichtner et al., 2003; Frohloff et al., 2001; Frohloff et al., 2003; Petrakis et al., 2005); Sit4p and its associated proteins (Sap185p and Sap190p), together with Kti12p, are proposed to regulate the phosphorylation of Elp1p (Jablonowski et al., 2004). The mechanisms of resistance in the remaining class II mutants, kti13, kti14, and urm1 are less clear. The linkage between transfer RNA and γ-toxin toxicity was first suggested based

Glu on the finding that elevated level of tRNA 25 UUCsmcm suppresses the zymocicity (Butler

Glu et al., 1994). The tRNA 25 UUCsmcm has an anticodon sequence U 34 U 35 C 36 in which U 34 is modified to 5-methoxycarbonylmethyl-2-thiouridine (mcm 5 s 2 U). Interestingly, the Elongator mutants, which are class II zymocin resistant, are all defective in the formation of mcm 5 s 2 U in tRNA (Huang et al., 2005). Furthermore, phenotypes in transcription and exocytosis observed in the Elongator mutants can be suppressed by

12 overexpression of tRNAs that normally contain mcm 5 s 2 U. This suggests that these phenotypes are secondary and are caused by inefficient translation due to the tRNA modification defect (Esberg et al., 2006). Taken together, class II zymocin resistance is linked to a defect in mcm 5 s 2 U formation, suggesting that tRNAs containing this modified nucleoside could be the target of γ-toxin.

1.1.5 Immunity

The K. lactis killer strain is immune to zymocin (Gunge et al., 1981; Niwa et al.,

1981). The protein coded by ORF3 in pGKL1 has been suggested to be responsible for the immunity. The K. lactis strain harboring a mutant form of pGKL1 plasmid, in which ORF2 is deleted but ORF3 intact, loses the killer activity but retains the immunity (Niwa et al., 1981). Furthermore, intracellular expression of ORF3 renders non-killer strain partially resistant to intracellular γ-toxin expression, providing the pGKL2 is present in the strain (Tokunaga et al., 1987a). Given that all genes in linear plasmids require plasmid-specific promoter motifs that are incompatible with the K. lactis host-encoded RNA polymerase (Romanos and Boyd, 1988), the pGKL2 is likely to be essential for the expression of cloned ORF3, but not for the resistance per se

(Stark et al., 1990). The mechanism of immunity is unclear. If or not the protein encoded by ORF3 functions by binding and inhibiting the activity of γ-toxin remains to be established.

1.1.6 Other yeast killer-encoding linear plasmids

Linear double-stranded (ds) DNA plasmid systems are observed in many other yeast genera (Fukuhara, 1995). In addition to K. lactis, linear plasmids isolated from

Pichia acaciae (Worsham and Bolen, 1990), Pichia inositovora (Hayman and Bolen, 1991), and Wingea robertsiae (Klassen and Meinhardt, 2002) are also associated with

13 killer phenotypes. According to the genetic organization of the plasmids and mechanisms of toxicity, these killer systems can be divided into two groups. The first group consists of killer toxins from K. lactis and P. inositovora. Plasmids from these systems encode the plasmid specific DNA polymerases. Even though the potential cytotoxic subunit of P. inositovora killer toxin shows only limited similarity in primary sequence with the γ subunit of K. lactis zymocin, these two toxins may share common target(s) (Klassen and Meinhardt, 2003), since an elp3 mutant that is resistant to zymocin also displays resistance against P. inositovora killer toxin. The second group includes killer toxins from P. acaciae and W. robertsiae. Plasmids from these systems do not encode a plasmid specific DNA polymerase. In addition, these toxins elicit inhibitory effect through distinct mechanism than K. lactis zymocin. Killer toxins from P. acaciae and W. robertsiae interfere with S-phase progression (Klassen et al., 2004), whereas K. lactis zymocin causes G1 arrest. Interestingly, all these toxins are able to bind chitin in vitro via chitin binding domains (Klassen and Meinhardt, 2002; Klassen and Meinhardt, 2003; Klassen et al., 2004). In addition, the mutants defective in chitin synthesis are protected from these toxins applied exogenously

(Klassen and Meinhardt, 2003; Klassen et al., 2004), indicating that all these killer toxins utilize chitin as receptor to enter the cell.

1.2 Transfer RNA modification

1.2.1 Transfer RNA (tRNA)

Transfer RNAs are adaptor molecules that deliver amino acids to the ribosome where messenger RNA (mRNA) is translated into protein. Transfer RNAs are typically between 75 and 95 nucleotides in length. The secondary structure of tRNA is commonly represented by a cloverleaf that is composed of four base-paired stems and three non-base-paired loops: D, anticodon,

14 TΨC loops (Holley et al., 1965) (Figure 2). The D and TΨC loops are conserved in sequence and interact with each other in the tertiary structure. All tRNA molecules form similar L-shaped tertiary structures in which the anticodon loop and the amino acid accepting stem (with its 3’ CCA end) are located at the two distal ends (Kim et al., 1974; Robertus et al., 1974) (Figure 2). The CCA end of the amino acid accepting stem is where the amino acid is attached. The anticodon is a triplet of nucleotides at position 34-36 in tRNA that basepair with the triplet of nucleotides of a codon in the mRNA (Figure 3). A C C A C C 1 72

Acceptor stem

TΨC-loop D-loop

Variable loop

Anticodon-loop 34 35 36

Figure 2. Schematic drawing of the secondary (left) and tertiary structure (right) of tRNA. The anticodon- loop is labeled with , the D-loop is labeled with , and the TΨC-loop is labeled with . Note the interaction between the D-loop and TΨC-loop in the tertiary structure.

The biogenesis of eukaryotic tRNA is a multi-step process. Before mature tRNAs are able to function in translation, the primary tRNA transcripts have to undergo extensive maturation steps, including removal of 5’ leader and 3’ trailer sequences, addition of CCA to the 3’ end, splicing of introns in some tRNAs, and modified nucleoside formation at multiple positions. In addition, tRNAs are transported into

15 different cellular compartments through complicated transporting pathways (reviewed by Hopper and Phizicky, 2003). The number of tRNA species is always less than that of codons, showing that some tRNAs have to read more than one codon. To explain this, Crick proposed the Wobble Hypothesis (Crick, 1966). It was suggested that the base-pairs between the first two nucleotides of the codon and the two last two nucleotides of the anticodon (position 35 and 36) must be canonical (Watson-Crick) base parings, but the base-pair between the third nucleotide of the codon and the first nucleotide (wobble nucleotide) of the anticodon can be either canonical or non-canonical (wobble) base pairing. For

Arg 5 example, tRNA 5UCUmcm that carries a modified wobble uridine mcm U is able to decode both AGA and AGG codon, because mcm 5 U can form Watson-Crick base pairing with A and wobble base paring with G (Johansson M.J.O., personal communication)(Figure 3). It was estimated that depending on the codon usage in an organism, about 30% - 40% of the codons are decoded by tRNA wobble recognition (reviewed by Agris et al., 2007).

3’ 3’ 5’ 5’

tRNA

U36 C35 U34* U36 C35 U34*

A G A A G G 5’ 3’ mRNA 5’ 3’

Figure 3. Schematic drawing of the interaction between the codon and the

Arg anticodon. tRNA 5UCUmcm that can decode both AGA and AGG codons is used as an example. The wobble nucleotide mcm 5 U (denoted with U*) can form Watson-Crick base pairing with A, or wobble base pairing with G.

16 1.2.2 Biogenesis and functions of modified nucleosides in tRNA

The tRNA molecules are the most extensively modified RNA in the cell. Currently, 91 different modified nucleosides have been found in tRNA from organisms within the three domains of life, Archaea, Bacteria, and Eukarya (Rozenski et al., 1999). Figure 4 shows the structures of some modified nucleosides which range from a simple methylation on the base (e.g. 5-methyluridine, m 5 U) to addition of a bulky complex

(e.g. N 6 -threonylcarbamoyladenosine, t 6 A). In addition to the base, modifications can occur on the ribose moiety (e.g. 2’-O-ribosyladenosine (phosphate), Ar(p)).

m5Ut6A Ar(p)

m cm5s2U mnm5s2U ncm5U

Figure 4. Structures of some modified nucleosides found in tRNA.

17 All the modified nucleosides in tRNAs are introduced post-transcriptionally by various tRNA modifying enzymes. So far, over 50 gene products have been found to be involved in formation of modified nucleosides in S. cerevisiae (reviewed by Johansson and Byström, 2005; paper III). Some of the tRNA modifying enzymes only modify cytosolic tRNAs, whereas certain tRNA modifying enzymes only modify mitochondrial tRNAs (Umeda et al., 2005). There are also tRNA modifying enzymes that could act on both cytosolic and mitochondrial tRNAs. For example, Pus4p is responsible for the pseudouridylation of tRNAs in both compartments (Becker et al., 1997). For modified nucleosides present in more than one position in the tRNA, the corresponding modifying enzymes may function in either a site-specific, region-specific, or multi-site way (Becker et al., 1997; Lecointe et al., 1998; Motorin et al., 1998). The formation of modified nucleosides occurs in concurrence with other tRNA processing events. For example, some modifications are made prior to, while others are made after the removal of intron (Grosjean et al., 1997). However, a specific order of modification seems not obligatory, since lack of earlier modifications does not affect the formation of other modifications.

Formation of certain modifications requires more than one protein. For example, numerous proteins are involved in the formation of the wobble modified nucleoside mcm 5 s 2 U. Subunits of the Elongator complex (Elp1p-Elp6p), Kti11p, Kti12p, and Kti13p are all required for the formation of the mcm 5 side chain (Huang et al., 2005). Trm9p catalyzes the last step of mcm 5 formation (Kalhor and Clarke, 2003). Recently it was shown that the biosynthesis of 2-thio (s 2 ) group on cytosolic tRNAs is iron-sulfur (Fe/S) protein dependent and requires mitochondrial and cytosolic Fe/S cluster assembly machineries termed ISC and CIA. Three components of CIA (Cfd1, Nbp35, and Cia1) and two mitochondria scaffold proteins (Isu1 and Isu2) are required for the s 2 modification of the cytosolic tRNAs (Nakai et al., 2007). In addition, the

18 Nfs1p, a cysteine desulfurase that might function as a sulfur donor, is also required for the synthesis of s 2 group in tRNA (Nakai et al., 2004). Certain modified nucleosides are found in all organisms and some are even found in identical positions of the tRNA, suggesting they have conserved functions (for review, Björk, 1995). The modifications present in the anticodon region are generally believed to be important for the decoding ability of the tRNA. It was suggested that modifications in the anticodon region selectively restrict some anticodon–codon interactions and expand others (Agris, 1991; Lim, 1994; Takai and Yokoyama, 2003; Yokoyama and Nishimura, 1995; Yokoyama et al., 1985). For example, mcm 5 , 5-carbamoylmethyl (ncm 5 ), and s 2 groups present on the wobble uridine in certain tRNAs are important for decoding A and G-ending codons (Björk G, personal communication; Johansson M.J.O, personal communication; Esberg A, personal communication). In some tRNAs, the modified nucleosides in the anticodon loop are identity elements for tRNA aminoacylation (for review, Giege et al., 1998). The modification on the wobble nucleoside can be used to sort tRNA in different cellular compartments in Leishmania tarentolae (Kaneko et al., 2003). The modifications present in other regions in tRNA might be important for the structure and/or stability of the tRNA (Anderson et al., 1998; Helm, 2006).

1.3 tRNA anticodon nucleases

1.3.1 Reaction mechanism of ribonucleases

Ribonucleases (RNases) cleave phosphodiester bonds in RNA. Most of the known RNases are protein enzymes (Aravind and Koonin, 2001), but some of the catalytic RNAs (ribozymes) are also capable of cleaving RNAs (for review, Fedor and Williamson, 2005). Notably, the RNA component of ribonuclease P (RNase P) in bacteria and some archaea can catalyze the maturation of the 5’ end of tRNA in vitro in

19 the absence of proteins (for reviews, Kazantsev and Pace, 2006; Kirsebom, 2002). Except for colicin E5 (see below), all protein RNases utilize at least one histidine residue as catalytic residues (Gerlt, 1993). The best studied RNase is RNase A which catalyzes the cleavage of the P-O 5’ bond of an RNA strand (transphosphorylation reaction) and the hydrolysis of the P-O 2’ bond of a nucleoside 2’,3’-cyclic phosphodiester (hydrolysis reaction). In both reactions, two of the histidine residues

(His 12 and His 119 ) play important roles in the catalysis (reviewed by Raines, 1998).

For example, in the transphosphorylation reaction, the His 12 acts as a base to abstract a proton from the 2’-oxygen and facilitates its attack on the phosphodiester bond, and

His 119 acts as an acid to protonate 5’-oxygen to facilitate its displacement. The amino group of Lys 41 is proposed to form hydrogen-bond with the nonbridging phophoryl oxygen, thus stabilizing the pentacovalent transition state. In this step, a 2’,3’-cyclic nucleotide intermediate is generated (Figure 5). For RNase T1 which also hydrolyzes

RNA via the formation of 2’,3’-cyclic nucleotide intermediates, it was suggested that

Glu 58 and His 92 function as the base and acid catalysts, respectively. In addition,

His 40 and Arg 77 might stabilize the transition state (Arni et al., 1988; Steyaert et al., 1990)

Figure 5. Putative mechanism of the transphosphorylation reaction catalyzed by

RNase A . His 12 and His 119 function as base and acid catalysts, respectively. The transition state is stabilized by Lys 41 (adapted from Fedor and Williamson, 2005).

20 1.3.2 PrrC

Enzymes that cleave tRNA in the anticodon region are termed A ntic odon N uclease (ACN) (Kaufmann, 2000). Presently, ACN family includes PrrC, colicin E5, and colicin D. All these bacterial ACNs generate 2’,3’-cyclic phosphate and 5’-hydroxyl termini, suggesting a similar cleavage mechanism as RNase A and RNase T1. It is noteworthy that eukaryotic and archaeal tRNA splicing endonucleases also utilize the same mechanism to cleave pre-tRNA, since they generate 2’,3’-cyclic phosphate and 5’-hydroxyl termini and use a histidine residue as acid catalyst (reviewed by Abelson et al., 1998). PrrC is encoded by the prr locus that is present in some Escherichia coli strains. This locus also encodes another protein, EcoprrI. The EcoprrI is a type I DNA restriction- modification enzyme that binds to PrrC and masks its ACN activity (Levitz et al., 1990; Linder et al., 1990; Tyndall et al., 1994). When E. coli strain carrying prr locus is infected with phage T4, phage-encoded Stp inhibits EcoprrI and activates PrrC

Lys (Penner et al., 1995). Activated PrrC cleaves tRNAUUU in the anticodon region, but the

Lys phage is able to repair the cleaved tRNAUUU by using T4 polynucleotide kinase and

Lys RNA ligase (Amitsur et al., 1987). The depletion of tRNAUUU by PrrC is considered a suicidal action operated in individual cells to favor the survival of the population.

Lys PrrC cleaves tRNAUUU between position 33 and 34, yielding 2’,3’-cyclic phosphate and 5’-OH termini (Amitsur et al., 1987). The catalytic residues of PrrC are proposed to be Arg 320 , Glu 324 , and possibly His 356 based on the observations that: 1)

Arg 320 , Glu 324 , and His 356 are conserved among bacterial PrrC homologs. 2) Mutation of these residues abolished ACN activity. 3) PrrC proteins carrying either R320A or E324A mutation could be activated by guanidine or acetate, respectively, however, H356A mutant allele could not be rescued by imidazole or its derivatives (Blanga-Kanfi et al., 2006).

21 The major recognition element for PrrC is the a nticodon s tem loop (ASL)

Lys structure, since synthetic RNA oligonucleotide corresponding to the ASL of tRNAUUU is a comparable substrate as the full length tRNA (Jiang et al., 2002). It was found that modifications at the wobble position play important roles in the interaction between PrrC and tRNA. PrrC cleaves tRNA carrying different wobble nucleoside with the following efficiency: mnm 5 s 2 U>U>C> mcm 5 s 2 U (Jiang et al., 2001). Further study using ASL carrying different modified nucleosides showed that t 6 A has the strongest positive effect, mnm 5 s 2 U has a weak positive effect, and mcm 5 s 2 U has a negative effect on the cleavage of ASL (Jiang et al., 2002).

1.3.3 Colicin E5

Toxins produced by some strains of E. coli that inhibit or kill closely related species are termed colicins. Colicins are encoded by plasmids which also carry a gene encoding corresponding immunity protein which protects the host (for review,

Cascales et al., 2007). Colicins could kill sensitive bacteria by forming pores on the cytoplasmic membrane (Lakey and Slatin, 2001), or working as nucleases which attack DNA, rRNA, or tRNA (Masaki and Ogawa, 2002).

Colicin E5 inhibits protein synthesis by cleaving tRNATyr , tRNA His , tRNA Asn , and tRNAAsp between position 34 and 35, yielding 2’,3’-cyclic phosphate and 5’-OH termini (Ogawa et al., 1999). All four substrate tRNAs of colicin E5 carry a hypermodified nucleotide queuosine (Q) at the wobble position, however, the Q is not a recognition element for colicin E5 (Lin et al., 2005; Ogawa et al., 2006; Ogawa et al.,

1999). Synthetic RNA oligonucleotide corresponding to the ASL of tRNATyr is cleaved in a similar way as tRNATyr . It was proposed that colicin E5 cleaves tRNA in a sequence-dependent way, with Y 32 G 33 U 34 N 35 (Y is a pyrimidine and N is any nucleotide) in the tRNA as the specificity determinant. In addition, the stem-loop structure of ASL is important for colicin E5 cleavage. Dinucleotide GpUp that is

22 cleaved by colicin E5 was proposed to represent a minimal substrate, however, it is not known if GpUp is cleaved with same efficiency as the RNA oligonucleotide (Ogawa et al., 2006). The three-dimensional structure of the C -terminal r ibonuclease d omain of Colicin E5 (E5-CRD) has been solved using crystallography (Lin et al., 2005; Yajima et al., 2006). The tertiary structure of E5-CRD has a weak homology with several other ribonucleases, including archaeal toxin RelE (Takagi et al., 2005), the catalytic domain of colicin D (Graille et al., 2004)(see below), and nuclease T1 (Martinez-Oyanedel et al., 1991). According to the structural model in which the

Phe ASL of tRNA was manually docked to the E5-CRD, Asp 54 and Arg 56 are proposed to be the putative catalytic residues (Lin et al., 2005). However, these residues can not reach enough proximity to the target site of the tRNA in the structural model where E5-CRD is complexed with substrate analog dGpdUp (Yajima et al., 2006).

Interestingly, colicin E5 does not contain histidine residue which is utilized as catalytic residue by other ribonucleases. The cleavage mechanism of colicin E5 thus requires further study.

1.3.4 Colicin D

Colicin D is another colicin that inhibits growth by cleaving tRNAs. It cleaves four isoaccepting arginine tRNAs between nucleotides 38 and 39, yielding 2’,3’- cyclic phosphate and 5’-OH termini (Tomita et al., 2000). The cytotoxicity of colicin D requires an endoproteolytic processing that releases a smaller C-terminal fragment into cytoplasm (de Zamaroczy et al., 2001). Interestingly, colicins acting as nuclease share a consensus sequence at their putative processing site, suggesting that proteolytic processing might be a common requirement for entry of nuclease type colicins into the cytoplasm (de Zamaroczy et al., 2001).

The crystal structure of the complex between the catalytic domain of Colicin D

23 and its immunity protein ImmD has been reported (Graille et al., 2004; Yajima et al., 2004). As mentioned above, the tertiary structure of the catalytic domain of colicin D shows weak similarity with that of the catalytic domain of colicin E5. It was proposed that the central groove formed by the C-terminal part of the α1 helix and neighboring positive charged cluster may form the active site of colicin D. The His 611 residue located in the central groove might be the catalytic residue, since its mutation abolishes both nuclease activity and cytotoxicity (Graille et al., 2004; Tomita et al.,

2000). A row of positively charged residues surrounding His 611 was suggested to form a tRNA binding site (Yajima et al., 2004). The determinants in the tRNA for colicin D cleavage remain to be identified.

24 RESULTS AND DISCUSSION

3.1 γ subunit of K. lactis zymocin (γ-toxin) is a tRNA anticodon nuclease (Paper I)

The linkage between tRNA and γ-toxin was suggested by the following

Glu 5 2 observations; First, elevated level of tRNA 25 UUCsmcm that carries mcm s U causes zymocin resistance in wild-type strain (Butler et al., 1994). Second, the class II zymocin resistant elp1-elp6 and kti11-kti13 mutants are deficient in the formation of mcm 5 and ncm 5 side chains on wobble uridines in tRNA (Huang et al., 2005). We found that a strain deleted for the TRM9 gene is also a class II resistant mutant (After this paper was published, it was confirmed by others that trm9 mutant displays a type II zymocin resistance, see Jablonowski et al., 2006). Trm9 protein is a methyltransferase responsible for the final step in the synthesis of the wobble mcm 5 side chain (Kalhor and Clarke, 2003). Thus, the lack of either part of, or the whole mcm 5 side chain in tRNA correlates with class II zymocin resistance, indicating that tRNA(s) containing this side chain could be the target(s) of γ-toxin.

Glu To investigate if tRNA 25 UUCsmcm is a target of the γ-toxin, we analyzed the levels of this tRNA in wild-type cells after either intracellular γ-toxin expression or

Glu exogenous zymocin treatment. In both cases, a reduction in the level of tRNA 25 UUCsmcm together with a reduction of cell viability was observed. In contrast, no reduction in cell viability or tRNA Glu levels was observed upon γ-toxin induction in the elp3Δ and

5 Glu trm9Δ mutants, suggesting that the mcm side chain in tRNA 25 UUCsmcm is required for

Lys the cytotoxicity of γ-toxin. No obvious reduction in the levels of tRNA 25 UUUsmcm ,

Gln Gly Arg 5 tRNA 25 UUGsmcm , tRNA 5UCCmcm , or tRNA 5UCUmcm , which all contain a mcm side chain on the wobble uridine, was observed after γ-toxin induction in wild-type cells.

Glu To examine if the reduction of tRNA 25 UUCsmcm is a direct effect of γ-toxin, we

25 Glu assayed the level of tRNA 25 UUCsmcm after incubation with recombinant γ-toxin purified from E. coli. Upon treatment with recombinant γ-toxin, the amount of full length

Glu Glu tRNA 25 UUCsmcm decreased. This reduction in tRNA 25 UUCsmcm level is caused by the cleavage, rather than degradation of the tRNA, because γ-toxin treatment yielded tRNA halves with approximate sizes of 34 and 41 nucleotides. The mcm 5 side chain on wobble uridine is important for the in vitro cleavage by γ-toxin, since hypomodified tRNA Glu prepared from elp3Δ or trm9Δ cells was cleaved with a lower efficiency than the fully modified counterpart.

Lys The in vitro cleavage was also observed in the fully modified tRNA 25 UUUsmcm

Gln Glu and tRNA 25 UUGsmcm at higher γ-toxin concentrations. However, tRNA 25 UUCsmcm appears to be the best substrate among these tRNAs, since its cleavage was detected at

Gly Arg significantly lower γ-toxin concentrations. tRNA 5UCCmcm and tRNA 5UCUmcm that also contain mcm 5 side chain on wobble uridine were not cleaved by purified γ-toxin, suggesting that in addition to mcm 5 group, there are other determinants for γ-toxin cleavage.

Glu Elevated levels of tRNA 25 UUCsmcm caused a wild-type strain to become resistant

Lys to exogenous zymocin. Similarly, simultaneous over expression of tRNA 25 UUUsmcm and

Gln tRNA 25 UUGsmcm generated a comparable resistance. A higher resistance to zymocin was

Glu Lys achieved by over-expressing tRNA 25 UUCsmcm in combination with tRNA 25 UUUsmcm or

Gln tRNA 25 UUGsmcm , or over-expressing all three tRNA species. In line with the difference in γ-toxin reactivity of the substrate tRNAs, these data suggest that not only

Glu Lys Gln tRNA 25 UUCsmcm , but also tRNA 25 UUUsmcm and tRNA 25 UUGsmcm are γ-toxin substrates in vivo. The position of the cleavage by γ-toxin was determined by primer extension analysis. It was shown that the 5’ nucleotide of the 3’-halves corresponds to position

35 in substrate tRNAs. The 3’-tRNA fragments could be directly labeled with [γ- 32 P] ATP using T4 polynucleotide kinase (T4 PNK), suggesting the presence of a

26 5’-hydroxyl terminus. The 3’ end of the 5’-tRNA fragments was accessible for ligation only after treatment with T4 PNK which has a 2’,3’-cyclic phosphodiesterase activity, suggesting the presence of a 2’,3’-cyclic phosphate. Taken together, these data show

Glu Lys Gln that γ-toxin cleaves tRNA 25 UUCsmnm , tRNA 25 UUUsmcm , and tRNA 25 UUGsmcm between position 34 and 35, generating a 2’,3’-cyclic phosphate and a 5’-hydroxyl termini. The zymocin induced G1 arrest is most likely a consequence of a translational defect

Glu caused by depletion or reduction of tRNA 25 UUCsmnm . The γ-toxin is the first eukaryotic toxin that targets tRNA. Secreting toxins to deplete competing fungi of tRNA could be a strategy to gain a growth advantage. For example, the toxin secreted by yeast Pichia inositovora might also target U34 containing tRNA species, as one subunit of the P. inositovora toxin shares a weak sequence similarity with the γ-toxin. In addition, an elp3 mutant that is resistant to zymocin, is also less sensitive to the P. inositovora toxin (Klassen and Meinhardt,

2003). All known anticodon nucleases, i.e., γ-toxin, PrrC (Amitsur et al., 1987), colicin E5 (Ogawa et al., 1999), and colicin D (Tomita et al., 2000), generate a 2’,3’-cyclic phosphate and a 5’-hydroxyl group at the cleavage site. Even though there is no obvious sequence homology among these proteins, recent studies revealed that colicin E5 and colicin D share similar core folding with known ribonucleases (Graille et al., 2004; Lin et al., 2005; Yajima et al., 2006). Preliminary data showed that the only histidine residue in γ-toxin is required for both nuclease activity and cytotoxicity of γ-toxin (data not shown), suggesting its potential role in the catalytic mechanism. The crystal structure of γ-toxin complexed with its substrate tRNA would provide the detailed pictures of the interaction between tRNA and γ-toxin.

27 3.2 Determinants in tRNA required for K. lactis γ-toxin cleavage

(Paper II)

Glu Lys Even though the substrate tRNAs, i.e., tRNA 25 UUCsmnm , tRNA 25 UUUsmcm , and

Gln 5 2 tRNA 25 UUGsmcm , all contain mcm s U at the wobble position, the latter two tRNA species are cleaved with much lower efficiency. This suggests that in addition to the

Glu wobble uridine modification, there are other features in the tRNA 25 UUCsmnm improving the cleavage by γ-toxin.

Lys Gln Interestingly, unmodified tRNAUUU and tRNAUUG were cleaved with much

Glu lower efficiency than unmodified tRNAUUC , suggesting that the sequence difference between the tRNAs is the cause of the difference in their reactivities. Notably, these unmodified tRNAs were much less sensitive to γ-toxin cleavage than their native counterparts, confirming the importance of tRNA modification. The importance of the anticodon stem loop (ASL) for γ-toxin cleavage was investigated by comparing the reactivities of various unmodified chimeric tRNAs.

Lys Glu Replacing ASL of tRNAUUU by the counterpart of tRNAUUC increased the reactivity

Lys Glu of tRNAUUU drastically, making it almost as good substrate as tRNAUUC . On the

Glu Lys contrary, exchange of ASL in tRNAUUC by the counterpart of tRNAUUU decreased its

Lys reactivity, rendering it as sensitive to γ-toxin as tRNAUUU . In addition, replacement of

Glu C 34 in the anticodon loop of tRNACUC to U 34 was sufficient to change it from being

Glu resistant to γ-toxin to becoming as sensitive as tRNAUUC . Taken together, these data

Glu showed that the ASL in tRNAUUC is important for the cleavage by γ-toxin. To investigate if an ASL by itself is sufficient for γ-toxin cleavage, we utilized a

Glu 17mer RNA oligonucleotide with identical sequence to the ASL of tRNAUUC (designated ASLGlu ). The RNA oligonucleotide was chemically synthesized, and lacks modified nucleosides that are present in native tRNA. The ASLGlu displayed similar

Glu cleavage kinetics upon γ-toxin treatment as tRNAUUC , supporting the notion that the ASLGlu contains all the sequence information required for efficient γ-toxin cleavage.

28 This enabled us to pinpoint the sequence identity elements in the ASL by using a series of RNA oligonucleotides corresponding to the wild-type and mutant ASLs. Disruption of the stem structure or alteration of stem and loop size in the ASLGlu caused a significant decrease in the reactivity. Changing the stem sequence to that of E. coli ASLGlu (5 G-C base-pairs) did not alter the reactivity, while a stem with 5 A-U base-pairs made ASL five fold less sensitive to γ-toxin. These data suggested that γ-toxin requires the canonical anticodon stem loop structure for efficient cleavage, and that ASL with a more stable stem structure may be preferred by γ-toxin. The bacterial anticodon nuclease PrrC also recognizes ASL as the minimal substrate, however, it prefers ASL with a partially destabilized stem (Jiang et al., 2002). For another bacterial anticodon nuclease colicin E5, dinucleotide GpUp that is corresponding to

G 34 U 35 shared between substrate tRNAs was proposed to be the minimal recognition sequence (Ogawa et al., 2006). However, it remains to be determined whether this dinucleotide is cleaved with same efficiency as the full length substrate tRNA. Interestingly, disruption of the stem loop structure greatly decreases the reactivity of the ASL (Ogawa et al., 2006), implying that the canonical anticodon stem loop structure is important for efficient cleavage by colicin E5. In the case of bacterial anticodon nuclease colicin D, the contacting region of substrate tRNA to colicin D was proposed to encompass the anticodon stem loop and D stem loop (Yajima et al., 2004).

To pinpoint which nucleotides in the ASL are required for γ-toxin cleavage, a series of ASLs, each with one of the loop nucleotides changed, were synthesized. Comparing the reactivities of these ASLs with that of wild-type ASLGlu revealed that nucleotides U 34 U 35 C 36 A 37 C 38 in the ASL were important for efficient cleavage by γ-toxin, whereas a purine at position 32 or a G at position 33 remarkably reduced the

Glu cleavage. The unaffected reactivity of mutant ASL which has a C 33 instead of U 33 is seemingly surprising, since U at position 33 is indispensable for the U-turn formation which is crucial for the canonical anticodon loop structure (Kim et al., 1974;

29 Quigley and Rich, 1976; Robertus et al., 1974). However, this does not rule out the possibility that canonical U-turn structure is required for efficient γ-toxin cleavage, because unmodified ASL might lack a U-turn (Ashraf et al., 1999; Cabello-Villegas et al., 2002; Durant and Davis, 1999).

Glu Glu In contrast to S. cerevisiae tRNA 25 UUCsmcm , the E. coli tRNA 25 UUCsmnm contains 5-methylaminomethyl-2-thiouridine (mnm 5 s 2 U) at the wobble position (Sprinzl et al.,

Glu 2 1987). In addition, the E. coli tRNA 25 UUCsmnm carries a 2-methyladenosine (m A) at

Glu position 37, whereas S. cerevisiae tRNA 25 UUCsmcm contains unmodified A 37 (Sprinzl et al., 1987). The other nucleotides in the anticodon loop are identical between the two tRNAs. It was shown that mnm 5 s 2 U has a positive whereas mcm 5 s 2 U has a negative effect on the cleavage of E. coli tRNA Lys by PrrC (Jiang et al., 2002; Jiang et al., 2001).

5 5 2 Glu We investigated the effect of mcm , mnm , or s group on the reactivity of tRNAUUC . A yeast elp3 mutant and an E. coli mnmE mutant lack the mcm 5 and mnm 5 side chains, respectively, but both retain the s 2 group in tRNA (Hagervall et al., 1998; Huang et al., 2005). A yeast tuc1 mutant and an E. coli mnmA mutant lack the s 2 group, but retain mcm 5 and mnm 5 groups in tRNA (Björk G, personal communication) (Hagervall et al.,

1998). Total RNA prepared from these strains and corresponding wild-type strains was

Glu treated with γ-toxin and the cleavage of tRNAUUC was compared. Our results showed that: 1) The mcm 5 group has a positive effect, 2) The mnm 5 group has a negative effect

Glu 2 on the reactivity of tRNAUUC , 3) The s group has a weaker positive effect on the

2 cleavage by γ-toxin, 4) The m A 37 is likely not a negative element in E. coli

Glu tRNA 25 UUCsmnm for γ-toxin cleavage.

Figure 6. Summary of Paper I and II. γ-toxin cleaves tRNAs containing mcm 5 s 2 U (indicated by U*) at the wobble position.

Y C The sequence identity elements for efficient X A U* C γ-toxin cleavage are shown (Y is a pyrimidine, U ✄ X is A, C, or U).

3.3 γ-toxin as a tool to identify mutants defective in the formation of modified wobble uridines (Paper III) The correlation between class II zymocin resistance and mcm 5 modification deficiency in tRNA should enable the isolation of mutants affecting this modification by screening for zymocin resistance. A systematic screen for mutants resistant to zymocin was performed using the yeast homozygous diploid deletion collection which consists of 4826 strains each with one non-essential gene deleted. Totally 63 mutants were found to be resistant to zymocin. Transfer RNA from these mutants was further analyzed by HPLC to identify mutants defective in mcm 5 s 2 U formation. The validity of the screen was confirmed by the identification of deletion strains known to affect synthesis of mcm 5 , i.e., elp1-elp4, elp6, kti12, kti13, and trm9. Null strains of elp5 or kti11 are not present in the deletion collection. Interestingly, we also identified novel genes involved in the formation of mcm 5 and ncm 5 (SIT4) and s 2 group (URM1, UBA4, NCS2, NCS6, and YOR251C), demonstrating that this screen is a powerful tool to identify mutants defective in wobble uridine modification (Figure 7).

Sit4·Sap185·Sap190 Nfs·Isu1 Isu2 Kti14 Cfd1 Kti12 Nbp35 Kti11·Kti13 Cia1 Elp1·Elp2·Elp3·Elp4·Elp5·Elp6 Urm1·Uba4 Ncs2 Ncs6 Yor251c

Trm9

Figure 7. Proteins involved in the formation of mcm 5 s 2 U at wobble position. Proteins identified in this study are circled. For details and references, see Paper III.

31 The Sit4 protein is a protein phosphatase that plays a critical role in cell growth and proliferation (Sutton et al., 1991). Consistent with its class II zymocin resistance (Jablonowski et al., 2001a), the sit4 null mutant was unable to form mcm 5 and ncm 5 side chain. The presence of 2-thiouridine (s 2 U) in tRNA from a sit4 null mutant indicates that thiolation at position 2 is not affected. The function of the Sit4 protein is regulated by various Sit4 associated proteins (Sap185, Sap190, Sap155, and Sap4) (Luke et al., 1996). Deletion of SAP185 or SAP190 gene individually does not affect zymocin resistance, whereas deletion of both SAP185 and SAP190 genes causes a class II zymocin resistance (Jablonowski et al., 2001a; Jablonowski et al., 2004). Consistently, we found that a single sap185Δ or sap190Δ mutant was not affected in wobble uridine modification whereas a sap185Δ sap190Δ double mutant was impaired in the formation of mcm 5 and ncm 5 side chain. On the contrary, a strain deleted for genes encoding the other two Sit4 associated proteins (sap4Δ sap155Δ), which is not resistant to zymocin (Jablonowski et al., 2001a), did not show any defect in wobble uridine modification. This suggested that the role of Sit4p in tRNA modification is modulated through the association with Sap185p and Sap190p.

The Sit4 protein, together with its associated proteins Sap185p and Sap190p, plays an important role in nutrient response mediated by TOR pathway (Di Como and Arndt, 1996; Luke et al., 1996; Rohde et al., 2004). We asked if TOR pathway, through Sit4p-Sap185p-Sap190p, regulates tRNA modification. Deletion of TOR1 gene downregulates the TOR pathway (Kaeberlein et al., 2005; Powers et al., 2006; Reinke et al., 2004). However, a tor1Δ strain was not resistant to zymocin and the level of mcm 5 s 2 U was not affected in total tRNA prepared from a tor1Δ strain, suggesting it is unlikely that TOR pathway is required for the synthesis of mcm 5 side chain.

In addition to the Sit4 phosphatase, we also identified a kinase Kti14p in the present study. A kti14 mutant that is resistant to intracellular γ-toxin (Mehlgarten and

32 Schaffrath, 2003) is impaired in the formation of mcm 5 and ncm 5 side chains. Kti14p is a homolog of mammalian casein kinase 1δ which is involved in various cellular processes, e.g., the secretion pathway, DNA repair, calcineurin signalling, maturation of 40S ribosomal subunits, and chromosome segregation (DeMaggio et al., 1992; Ho et al., 1997; Hoekstra et al., 1991; Kafadar et al., 2003; Murakami et al., 1999; Schafer et al., 2006). Interestingly, Kti14 protein displays physical interaction with Elongator complex (Elp1, Elp2, and Elp3 proteins) and Sit4p-Sap185p-Sap190p complex (Gavin et al., 2006; Ho et al., 2002; Schafer et al., 2003). We hypothesized that Sit4 and Kti14 proteins may cooperate to regulate the phosphorylation status of proteins required for mcm 5 and ncm 5 side chain formation. One candidate substrate of Sit4p/Kti14p regulation is the Elp1 protein whose phosphorylation status is affected by several proteins including Sit4p (Jablonowski et al., 2004), however, it remains to be established that the changes in the phosphorylation status of Elp1p are the direct effect of Sit4p.

Glu Lys Αll three γ-toxin target tRNAs, tRNA 25 UUCsmnm , tRNA 25 UUUsmcm , and

Gln 2 tRNA 25 UUGsmcm , contain a 2-thio (s ) modification at the wobble uridine. In the present screen, we identified 4 mutants, urm1, uba4, ncs2, and ncs6, which lacked the 2-thio modification on wobble uridines. Another mutant, yor251c, displayed a reduced level

2 Glu of s -containing modified wobble uridine. The tRNA 25 UUCsmnm from mutants lacking s 2 group was more resistant to γ-toxin cleavage than that from wild-type strain, reinforcing the previous finding (see section 3.2) that 2-thio modification on tRNAs is also required for effective γ-toxin cleavage. Based on its homology to the bacterial TtcA protein, which is required for 2-thiocytidine synthesis (Jager et al., 2004), Ncs6 has recently been proposed to be responsible for making 2-thiouridine in yeast (Björk G, personal communication).

Urm1 and Uba4 protein function in the urmylation pathway that is one of ubiquitin-like protein conjugation systems in S. cerevisiae (Furukawa et al., 2000). It

33 is possible that s 2 -forming enzymes, e.g., Ncs6p, need to be urmylated to become functional. Recently, it was reported that thiolation of cytosolic tRNA is iron-sulfur (Fe/S) protein dependent. Three components of cytosolic Fe/S cluster assembly machinery (Cfd1, Nbp35, and Cia1) and two mitochondria scaffold proteins (Isu1 and Isu2) are required for the s 2 modification of the cytosolic tRNAs but are dispensable for the thiolation of the mitochondrial tRNAs (Nakai et al., 2007). On the other hand, the

NFS1 gene which encodes a cysteine desulfurase that might function as a sulphur donor is required for the thiolation of both mitochondrial and cytoplasmic tRNA (Nakai et al., 2004). Notably, these genes (NFS1, CFD1, NBP35, CIA1, ISU1, and

ISU2) are essential and thus were not detected in our screen using yeast homozygous diploid deletion collection.

34 CONCLUSIONS

Glu 1. The γ-toxin of K. lactis is a tRNA endonuclease. It cleaves tRNA 25 UUCsmnm ,

Lys Gln tRNA 25 UUUsmcm , and tRNA 25 UUGsmcm between position 34 (wobble position) and position 35, generating tRNA fragments with 2’,3’-cyclic phosphate and

5’-hydroxyl ends.

2. The mcm 5 group on the wobble uridine has a positive effect, whereas the

mnm 5 group has a inhibitory effect on tRNA cleavage. The s 2 group has a

weaker positive effect on tRNA cleavage.

3. The anticodon stem loop is important and sufficient for γ-toxin cleavage.

4. Sit4p-Sap185p-Sap190p, and Kti14p are required for the formation of mcm 5

side chain on the wobble uridine.

5. Urm1p, Uba4p, Ncs2p, and Ncs6p are required for the s 2 formation on the

wobble uridine.

35 ACKNOWLEDGEMENTS

I would like to thank my supervisor Anders Byström for his guidance and support. His enthusiasm and creativity are amazing and will keep on inspiring me. I am grateful to all members in group AB for making an enjoyable environment. Thanks are due to: Marcus, for always being helpful, from my early days to final stage; Anders E, for sharing knowledge and painfulness during these years; Bo, for opening new projects and wonderful collaboration; Changchun, for fresh ideas, taking over burdens in time, and fun in worms and football. I also would like to thank: Glenn Björk, Mikael Wikstrom, Tord Hagervall, and their groups for group meetings, journal clubs, and helps in research and manuscript preparation. Sincere thanks to Edmund Loh for proofreading this thesis. Dr. Bengt Hallberg and his group members, especially Dr. Hailing Yang, are acknowledged for the great help with cell culture. Thanks are also due to Prof. Elisabeth Sauer-Eriksson and her group members, Anders Karlsson, Dr. Anders Olofsson, Carlos Cabrera, for valuable help with protein crystallization. Dishwashing, media, Marita, and Berit, as well as P-A/Johnny have provided valuable services. Living far away from China became easier with all Chinese friends in Umeå. Special thanks to Teacher Gu, families of Chen Peng’s, Chen Sa’s, Gong’s, Jin Weiqiang’s, Lai’s, Liu Jianming’s, Wang Daping’s, Xiong’s, Zhang Jiexie’s. Mei Yafang and Zhao Yunsheng are acknowledged for help with the cars. I am especially grateful to Prof. Jin Taiyi for never-failing support, Dr. Zhou Guangqian for introducing me to Umeå. Yun Jun is acknowledged for all the help. Poker club and Fishing club members: Bo, Jufang, Gangwei - It is a pity that banquet will not last for ever! Badminton fans: Bo (you are a very good instructor), Jun, Xinhe, Nicklas – hope you stay in B! Prof. Gunnar and Monica Nordberg, thanks for the cooperation and care. Mom and big brother, I love you. Don’t worry, I will be fine. OK? Ming and Yuanyang, you are the fortune that heaven has given to me - Everything I do, I do it for you.

36 REFERENCES

Abelson, J., Trotta, C.R. and Li, H. (1998) tRNA splicing. J Biol Chem, 273, 12685-12688. Agris, P.F. (1991) Wobble position modified nucleosides evolved to select transfer RNA codon recognition: a modified-wobble hypothesis. Biochimie, 73, 1345-1349. Agris, P.F., Vendeix, F.A. and Graham, W.D. (2007) tRNA's wobble decoding of the genome: 40 years of modification. J Mol Biol, 366, 1-13. Amitsur, M., Levitz, R. and Kaufmann, G. (1987) Bacteriophage T4 anticodon nuclease, polynucleotide kinase and RNA ligase reprocess the host lysine tRNA. Embo J, 6, 2499-2503. Anderson, J., Phan, L., Cuesta, R., Carlson, B.A., Pak, M., Asano, K., Björk, G.R., Tamame, M. and Hinnebusch, A.G. (1998) The essential Gcd10p-Gcd14p nuclear complex is required for 1- methyladenosine modification and maturation of initiator methionyl-tRNA. Genes Dev, 12, 3650-3662. Aravind, L. and Koonin, E.V. (2001) A natural classification of ribonucleases. Methods Enzymol, 341, 3-28. Arni, R., Heinemann, U., Tokuoka, R. and Saenger, W. (1988) Three-dimensional structure of the ribonuclease T1 2'-GMP complex at 1.9-A resolution. J Biol Chem, 263, 15358-15368. Ashraf, S.S., Ansari, G., Guenther, R., Sochacka, E., Malkiewicz, A. and Agris, P.F. (1999) The uridine in "U-turn": contributions to tRNA-ribosomal binding. Rna, 5, 503-511. Baldari, C., Murray, J.A., Ghiara, P., Cesareni, G. and Galeotti, C.L. (1987) A novel leader peptide which allows efficient secretion of a fragment of human interleukin 1 beta in Saccharomyces cerevisiae. Embo J, 6, 229-234. Becker, H.F., Motorin, Y., Planta, R.J. and Grosjean, H. (1997) The yeast gene YNL292w encodes a pseudouridine synthase (Pus4) catalyzing the formation of psi55 in both mitochondrial and cytoplasmic tRNAs. Nucleic Acids Res, 25, 4493-4499.

37 Bevan, E.A., Herring, A.J. and Mitchell, D.J. (1973) Preliminary characterization of two species of dsRNA in yeast and their relationship to the "killer" character. Nature, 245, 81-86. Björk, G. (1995) Biosynthesis and Function of Modified Nucleosides. In Söll, D. and RajBhandary, U. (eds.), tRNA : structure, biosynthesis, and function. ASM Press, Washington, D.C., pp. 165-205. Blanga-Kanfi, S., Amitsur, M., Azem, A. and Kaufmann, G. (2006) PrrC-anticodon nuclease: functional organization of a prototypical bacterial restriction RNase. Nucleic Acids Res, 34, 3209-3219. Butler, A.R., O'Donnell, R.W., Martin, V.J., Gooday, G.W. and Stark, M.J. (1991a) Kluyveromyces lactis toxin has an essential chitinase activity. Eur J Biochem, 199, 483-488. Butler, A.R., Porter, M. and Stark, M.J. (1991b) Intracellular expression of Kluyveromyces lactis toxin gamma subunit mimics treatment with exogenous toxin and distinguishes two classes of toxin-resistant mutant. Yeast, 7, 617-625. Butler, A.R., White, J.H., Folawiyo, Y., Edlin, A., Gardiner, D. and Stark, M.J. (1994) Two Saccharomyces cerevisiae genes which control sensitivity to G1 arrest induced by Kluyveromyces lactis toxin. Mol Cell Biol, 14, 6306-6316. Cabello-Villegas, J., Winkler, M.E. and Nikonowicz, E.P. (2002) Solution conformations of unmodified and A(37)N(6)-dimethylallyl modified anticodon stem-loops of Escherichia coli tRNA(Phe). J Mol Biol, 319, 1015-1034. Cascales, E., Buchanan, S.K., Duche, D., Kleanthous, C., Lloubes, R., Postle, K., Riley, M., Slatin, S. and Cavard, D. (2007) Colicin biology. Microbiol Mol Biol Rev, 71, 158-229. Crick, F.H. (1966) Codon--anticodon pairing: the wobble hypothesis. J Mol Biol, 19, 548-555. de Zamaroczy, M., Mora, L., Lecuyer, A., Geli, V. and Buckingham, R.H. (2001) Cleavage of colicin D is necessary for cell killing and requires the inner membrane peptidase LepB. Mol Cell, 8, 159-168. DeMaggio, A.J., Lindberg, R.A., Hunter, T. and Hoekstra, M.F. (1992) The budding

38 yeast HRR25 gene product is a casein kinase I isoform. Proc Natl Acad Sci U S A, 89, 7008-7012. Di Como, C.J. and Arndt, K.T. (1996) Nutrients, via the Tor proteins, stimulate the association of Tap42 with type 2A phosphatases. Genes Dev, 10, 1904-1916. Durant, P.C. and Davis, D.R. (1999) Stabilization of the anticodon stem-loop of tRNALys,3 by an A+-C base-pair and by pseudouridine. J Mol Biol, 285, 115-131. Esberg, A., Huang, B., Johansson, M.J. and Byström, A.S. (2006) Elevated levels of two tRNA species bypass the requirement for elongator complex in transcription and exocytosis. Mol Cell, 24, 139-148. Fedor, M.J. and Williamson, J.R. (2005) The catalytic diversity of RNAs. Nat Rev Mol Cell Biol, 6, 399-412. Fichtner, L., Frohloff, F., Burkner, K., Larsen, M., Breunig, K.D. and Schaffrath, R. (2002a) Molecular analysis of KTI12/TOT4, a Saccharomyces cerevisiae gene required for Kluyveromyces lactis zymocin action. Mol Microbiol, 43, 783-791. Fichtner, L., Frohloff, F., Jablonowski, D., Stark, M.J. and Schaffrath, R. (2002b) Protein interactions within Saccharomyces cerevisiae Elongator, a complex essential for Kluyveromyces lactis zymocicity. Mol Microbiol, 45, 817-826. Fichtner, L., Jablonowski, D., Schierhorn, A., Kitamoto, H.K., Stark, M.J. and Schaffrath, R. (2003) Elongator's toxin-target (TOT) function is nuclear localization sequence dependent and suppressed by post-translational modification. Mol Microbiol, 49, 1297-1307. Frohloff, F., Fichtner, L., Jablonowski, D., Breunig, K.D. and Schaffrath, R. (2001) Saccharomyces cerevisiae Elongator mutations confer resistance to the Kluyveromyces lactis zymocin. Embo J, 20, 1993-2003. Frohloff, F., Jablonowski, D., Fichtner, L. and Schaffrath, R. (2003) Subunit communications crucial for the functional integrity of the yeast RNA polymerase II elongator (gamma-toxin target (TOT)) complex. J Biol Chem, 278, 956-961. Fukuhara, H. (1995) Linear DNA plasmids of yeasts. FEMS Microbiol Lett, 131, 1-9. Furukawa, K., Mizushima, N., Noda, T. and Ohsumi, Y. (2000) A protein conjugation

39 system in yeast with homology to biosynthetic enzyme reaction of prokaryotes. J Biol Chem, 275, 7462-7465. Gavin, A.C., Aloy, P., Grandi, P., Krause, R., Boesche, M., Marzioch, M., Rau, C., Jensen, L.J., Bastuck, S., Dumpelfeld, B., Edelmann, A., Heurtier, M.A., Hoffman, V., Hoefert, C., Klein, K., Hudak, M., Michon, A.M., Schelder, M., Schirle, M., Remor, M., Rudi, T., Hooper, S., Bauer, A., Bouwmeester, T., Casari, G., Drewes, G., Neubauer, G., Rick, J.M., Kuster, B., Bork, P., Russell, R.B. and Superti-Furga, G. (2006) Proteome survey reveals modularity of the yeast cell machinery. Nature, 440, 631-636. Gerlt, J.A. (1993) Nucleases. In Linn, S.M., Lloyd, R.S. and Roberts, R.J. (eds.), Cold Spring Harbor monograph series ; 25. Cold Spring Harbor, Cold Spring Harbor, N.Y., pp. 1-34. Giege, R., Sissler, M. and Florentz, C. (1998) Universal rules and idiosyncratic features in tRNA identity. Nucleic Acids Res, 26, 5017-5035. Graille, M., Mora, L., Buckingham, R.H., van Tilbeurgh, H. and de Zamaroczy, M. (2004) Structural inhibition of the colicin D tRNase by the tRNA-mimicking immunity protein. Embo J, 23, 1474-1482. Grosjean, H., Szweykowska-Kulinska, Z., Motorin, Y., Fasiolo, F. and Simos, G. (1997) Intron-dependent enzymatic formation of modified nucleosides in eukaryotic tRNAs: a review. Biochimie, 79, 293-302. Gunge, N. and Sakaguchi, K. (1981) Intergeneric transfer of deoxyribonucleic acid killer plasmids, pGKl1 and pGKl2, from Kluyveromyces lactis into Saccharomyces cerevisiae by cell fusion. J Bacteriol, 147, 155-160. Gunge, N., Tamaru, A., Ozawa, F. and Sakaguchi, K. (1981) Isolation and characterization of linear deoxyribonucleic acid plasmids from Kluyveromyces lactis and the plasmid-associated killer character. J Bacteriol, 145, 382-390. Hagervall, T.G., Pomerantz, S.C. and McCloskey, J.A. (1998) Reduced misreading of asparagine codons by Escherichia coli tRNALys with hypomodified derivatives of 5-methylaminomethyl-2-thiouridine in the wobble position. J Mol Biol, 284, 33-42. Hayman, G.T. and Bolen, P.L. (1991) Linear DNA plasmids of Pichia inositovora are

40 associated with a novel killer toxin activity. Curr Genet, 19, 389-393. Helm, M. (2006) Post-transcriptional nucleotide modification and alternative folding of RNA. Nucleic Acids Res, 34, 721-733. Ho, Y., Gruhler, A., Heilbut, A., Bader, G.D., Moore, L., Adams, S.L., Millar, A., Taylor, P., Bennett, K., Boutilier, K., Yang, L., Wolting, C., Donaldson, I., Schandorff, S., Shewnarane, J., Vo, M., Taggart, J., Goudreault, M., Muskat, B., Alfarano, C., Dewar, D., Lin, Z., Michalickova, K., Willems, A.R., Sassi, H., Nielsen, P.A., Rasmussen, K.J., Andersen, J.R., Johansen, L.E., Hansen, L.H., Jespersen, H., Podtelejnikov, A., Nielsen, E., Crawford, J., Poulsen, V., Sorensen, B.D., Matthiesen, J., Hendrickson, R.C., Gleeson, F., Pawson, T., Moran, M.F., Durocher, D., Mann, M., Hogue, C.W., Figeys, D. and Tyers, M. (2002) Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature, 415, 180-183. Ho, Y., Mason, S., Kobayashi, R., Hoekstra, M. and Andrews, B. (1997) Role of the casein kinase I isoform, Hrr25, and the cell cycle-regulatory transcription factor, SBF, in the transcriptional response to DNA damage in Saccharomyces cerevisiae. Proc Natl Acad Sci U S A, 94, 581-586. Hoekstra, M.F., Liskay, R.M., Ou, A.C., DeMaggio, A.J., Burbee, D.G. and Heffron, F. (1991) HRR25, a putative protein kinase from budding yeast: association with repair of damaged DNA. Science, 253, 1031-1034. Holley, R.W., Apgar, J., Everett, G.A., Madison, J.T., Marquisee, M., Merrill, S.H., Penswick, J.R. and Zamir, A. (1965) Structure of a Ribonucleic Acid. Science, 147, 1462-1465. Hopper, A.K. and Phizicky, E.M. (2003) tRNA transfers to the limelight. Genes Dev, 17, 162-180. Huang, B., Johansson, M.J. and Byström, A.S. (2005) An early step in wobble uridine tRNA modification requires the Elongator complex. Rna, 11, 424-436. Jablonowski, D., Butler, A.R., Fichtner, L., Gardiner, D., Schaffrath, R. and Stark, M.J. (2001a) Sit4p protein phosphatase is required for sensitivity of Saccharomyces cerevisiae to Kluyveromyces lactis zymocin. Genetics, 159, 1479-1489. Jablonowski, D., Fichtner, L., Martin, V.J., Klassen, R., Meinhardt, F., Stark, M.J. and

41 Schaffrath, R. (2001b) Saccharomyces cerevisiae cell wall chitin, the Kluyveromyces lactis zymocin receptor. Yeast, 18, 1285-1299. Jablonowski, D., Fichtner, L., Stark, M.J. and Schaffrath, R. (2004) The Yeast Elongator Histone Acetylase Requires Sit4-Dependent Dephosphorylation for Toxin-Target Capacity. Mol Biol Cell. Jablonowski, D., Frohloff, F., Fichtner, L., Stark, M.J. and Schaffrath, R. (2001c) Kluyveromyces lactis zymocin mode of action is linked to RNA polymerase II function via Elongator. Mol Microbiol, 42, 1095-1105. Jablonowski, D., Zink, S., Mehlgarten, C., Daum, G. and Schaffrath, R. (2006) tRNAGlu wobble uridine methylation by Trm9 identifies Elongator's key role for zymocin-induced cell death in yeast. Mol Microbiol, 59, 677-688. Jager, G., Leipuviene, R., Pollard, M.G., Qian, Q. and Björk, G.R. (2004) The conserved Cys-X1-X2-Cys motif present in the TtcA protein is required for the thiolation of cytidine in position 32 of tRNA from Salmonella enterica serovar Typhimurium. J Bacteriol, 186, 750-757. Jiang, Y., Blanga, S., Amitsur, M., Meidler, R., Krivosheyev, E., Sundaram, M., Bajji, A.C., Davis, D.R. and Kaufmann, G. (2002) Structural features of tRNALys favored by anticodon nuclease as inferred from reactivities of anticodon stem and loop substrate analogs. J Biol Chem, 277, 3836-3841. Jiang, Y., Meidler, R., Amitsur, M. and Kaufmann, G. (2001) Specific interaction between anticodon nuclease and the tRNA(Lys) wobble base. J Mol Biol, 305, 377-388. Johansson, M.J. and Byström, A.S. (2005) Transfer RNA modifications and modifying enzymes in Saccharomyces cerevisiae. In Grosjean, H. (ed.), Fine-Tuning of RNA Functions by Modification and Editing. Springer Berlin / Heidelberg, Vol. 12, pp. 87-120. Julius, D., Brake, A., Blair, L., Kunisawa, R. and Thorner, J. (1984) Isolation of the putative structural gene for the lysine-arginine-cleaving endopeptidase required for processing of yeast prepro-alpha-factor. Cell, 37, 1075-1089. Kaeberlein, M., Powers, R.W., 3rd, Steffen, K.K., Westman, E.A., Hu, D., Dang, N., Kerr, E.O., Kirkland, K.T., Fields, S. and Kennedy, B.K. (2005) Regulation of yeast replicative life span by TOR and Sch9 in response to nutrients. Science,

42 310, 1193-1196. Kafadar, K.A., Zhu, H., Snyder, M. and Cyert, M.S. (2003) Negative regulation of calcineurin signaling by Hrr25p, a yeast homolog of casein kinase I. Genes Dev, 17, 2698-2708. Kalhor, H.R. and Clarke, S. (2003) Novel methyltransferase for modified uridine residues at the wobble position of tRNA. Mol Cell Biol, 23, 9283-9292. Kamper, J., Esser, K., Gunge, N. and Meinhardt, F. (1991) Heterologous gene expression on the linear DNA killer plasmid from Kluyveromyces lactis. Curr Genet, 19, 109-118. Kaneko, T., Suzuki, T., Kapushoc, S.T., Rubio, M.A., Ghazvini, J., Watanabe, K., Simpson, L. and Suzuki, T. (2003) Wobble modification differences and subcellular localization of tRNAs in Leishmania tarentolae: implication for tRNA sorting mechanism. Embo J, 22, 657-667. Kaufmann, G. (2000) Anticodon nucleases. Trends Biochem Sci, 25, 70-74. Kawamoto, S., Sasaki, T., Itahashi, S., Hatsuyama, Y. and Ohno, T. (1993) A mutant allele skt5 affecting protoplast regeneration and killer toxin resistance has double mutations in its wild-type structural gene in Saccharomyces cerevisiae. Biosci Biotechnol Biochem, 57, 1391-1393. Kazantsev, A.V. and Pace, N.R. (2006) Bacterial RNase P: a new view of an ancient enzyme. Nat Rev Microbiol, 4, 729-740. Kim, S.H., Suddath, F.L., Quigley, G.J., McPherson, A., Sussman, J.L., Wang, A.H., Seeman, N.C. and Rich, A. (1974) Three-dimensional tertiary structure of yeast phenylalanine transfer RNA. Science, 185, 435-440. Kirsebom, L.A. (2002) RNase P RNA-mediated catalysis. Biochem Soc Trans, 30, 1153-1158. Kishida, M., Tokunaga, M., Katayose, Y., Yajima, H., Kawamura-Watabe, A. and Hishinuma, F. (1996) Isolation and genetic characterization of pGKL killer-insensitive mutants (iki) from Saccharomyces cerevisiae. Biosci Biotechnol Biochem, 60, 798-801. Klassen, R. and Meinhardt, F. (2002) Linear plasmids pWR1A and pWR1B of the yeast Wingea robertsiae are associated with a killer phenotype. Plasmid, 48, 142-148.

43 Klassen, R. and Meinhardt, F. (2003) Structural and functional analysis of the killer element pPin1-3 from Pichia inositovora. Mol Genet Genomics, 270, 190-199. Klassen, R., Teichert, S. and Meinhardt, F. (2004) Novel yeast killer toxins provoke S-phase arrest and DNA damage checkpoint activation. Mol Microbiol, 53, 263-273. Krogan, N.J. and Greenblatt, J.F. (2001) Characterization of a six-subunit holo-elongator complex required for the regulated expression of a group of genes in Saccharomyces cerevisiae. Mol Cell Biol, 21, 8203-8212. Lakey, J.H. and Slatin, S.L. (2001) Pore-forming colicins and their relatives. Curr Top Microbiol Immunol, 257, 131-161. Larsen, M., Gunge, N. and Meinhardt, F. (1998) Kluyveromyces lactis killer plasmid pGKL2: evidence for a viral-like capping enzyme encoded by ORF3. Plasmid, 40, 243-246. Larsen, M. and Meinhardt, F. (2000) Kluyveromyces lactis killer system: identification of a new gene encoded by pGKL2. Curr Genet, 38, 271-275. Lecointe, F., Simos, G., Sauer, A., Hurt, E.C., Motorin, Y. and Grosjean, H. (1998) Characterization of yeast protein Deg1 as pseudouridine synthase (Pus3) catalyzing the formation of psi 38 and psi 39 in tRNA anticodon loop. J Biol Chem, 273, 1316-1323. Levitz, R., Chapman, D., Amitsur, M., Green, R., Snyder, L. and Kaufmann, G. (1990) The optional E. coli prr locus encodes a latent form of phage T4-induced anticodon nuclease. Embo J, 9, 1383-1389. Lim, V.I. (1994) Analysis of action of wobble nucleoside modifications on codon-anticodon pairing within the ribosome. J Mol Biol, 240, 8-19. Lin, Y.L., Elias, Y. and Huang, R.H. (2005) Structural and mutational studies of the catalytic domain of colicin E5: a tRNA-specific ribonuclease. Biochemistry, 44, 10494-10500. Linder, P., Doelz, R., Gubler, M. and Bickle, T.A. (1990) An anticodon nuclease gene inserted into a hsd region encoding a type I DNA restriction system. Nucleic Acids Res, 18, 7170. Luke, M.M., Della Seta, F., Di Como, C.J., Sugimoto, H., Kobayashi, R. and Arndt, K.T. (1996) The SAP, a new family of proteins, associate and function

44 positively with the SIT4 phosphatase. Mol Cell Biol, 16, 2744-2755. Magliani, W., Conti, S., Gerloni, M., Bertolotti, D. and Polonelli, L. (1997) Yeast killer systems. Clin Microbiol Rev, 10, 369-400. Martinez-Oyanedel, J., Choe, H.W., Heinemann, U. and Saenger, W. (1991) Ribonuclease T1 with free recognition and catalytic site: crystal structure analysis at 1.5 A resolution. J Mol Biol, 222, 335-352. Masaki, H. and Ogawa, T. (2002) The modes of action of colicins E5 and D, and related cytotoxic tRNases. Biochimie, 84, 433-438. McNeel, D.G. and Tamanoi, F. (1991) Terminal region recognition factor 1, a DNA-binding protein recognizing the inverted terminal repeats of the pGKl linear DNA plasmids. Proc Natl Acad Sci U S A, 88, 11398-11402. Mehlgarten, C. and Schaffrath, R. (2003) Mutant casein kinase I (Hrr25p/Kti14p) abrogates the G1 cell cycle arrest induced by Kluyveromyces lactiszymocin in budding yeast. Mol Genet Genomics, 269, 188-196. Mehlgarten, C. and Schaffrath, R. (2004) After chitin docking, toxicity of Kluyveromyces lactis zymocin requires Saccharomyces cerevisiae plasma membrane H+-ATPase. Cell Microbiol, 6, 569-580. Meijer, W.J., Horcajadas, J.A. and Salas, M. (2001) Phi29 family of phages. Microbiol Mol Biol Rev, 65, 261-287 ; second page, table of contents. Motorin, Y., Keith, G., Simon, C., Foiret, D., Simos, G., Hurt, E. and Grosjean, H. (1998) The yeast tRNA:pseudouridine synthase Pus1p displays a multisite substrate specificity. Rna, 4, 856-869. Murakami, A., Kimura, K. and Nakano, A. (1999) The inactive form of a yeast casein kinase I suppresses the secretory defect of the sec12 mutant. Implication of negative regulation by the Hrr25 kinase in the vesicle budding from the endoplasmic reticulum. J Biol Chem, 274, 3804-3810. Nakai, Y., Nakai, M., Lill, R., Suzuki, T. and Hayashi, H. (2007) Thio modification of yeast cytosolic tRNA is an iron-sulfur protein-dependent pathway. Mol Cell Biol. Nakai, Y., Umeda, N., Suzuki, T., Nakai, M., Hayashi, H., Watanabe, K. and Kagamiyama, H. (2004) Yeast Nfs1p is involved in thio-modification of both mitochondrial and cytoplasmic tRNAs. J Biol Chem, 279, 12363-12368.

45 Niwa, O., Sakaguchi, K. and Gunge, N. (1981) Curing of the killer deoxyribonucleic acid plasmids of Kluyveromyces lactis. J Bacteriol, 148, 988-990. Ogawa, T., Inoue, S., Yajima, S., Hidaka, M. and Masaki, H. (2006) Sequence-specific recognition of colicin E5, a tRNA-targeting ribonuclease. Nucleic Acids Res. Ogawa, T., Tomita, K., Ueda, T., Watanabe, K., Uozumi, T. and Masaki, H. (1999) A cytotoxic ribonuclease targeting specific transfer RNA anticodons. Science, 283, 2097-2100. Otero, G., Fellows, J., Li, Y., de Bizemont, T., Dirac, A.M., Gustafsson, C.M., Erdjument-Bromage, H., Tempst, P. and Svejstrup, J.Q. (1999) Elongator, a multisubunit component of a novel RNA polymerase II holoenzyme for transcriptional elongation. Mol Cell, 3, 109-118. Penner, M., Morad, I., Snyder, L. and Kaufmann, G. (1995) Phage T4-coded Stp: double-edged effector of coupled DNA and tRNA-restriction systems. J Mol Biol, 249, 857-868. Petrakis, T.G., Sogaard, T.M., Erdjument-Bromage, H., Tempst, P. and Svejstrup, J.Q. (2005) Physical and functional interaction between Elongator and the chromatin-associated Kti12 protein. J Biol Chem, 280, 19454-19460. Powers, R.W., 3rd, Kaeberlein, M., Caldwell, S.D., Kennedy, B.K. and Fields, S. (2006) Extension of chronological life span in yeast by decreased TOR pathway signaling. Genes Dev, 20, 174-184. Quigley, G.J. and Rich, A. (1976) Structural domains of transfer RNA molecules. Science, 194, 796-806. Rahl, P.B., Chen, C.Z. and Collins, R.N. (2005) Elp1p, the yeast homolog of the FD disease syndrome protein, negatively regulates exocytosis independently of transcriptional elongation. Mol Cell, 17, 841-853. Raines, R.T. (1998) Ribonuclease A. Chem Rev, 98, 1045-1066. Reinke, A., Anderson, S., McCaffery, J.M., Yates, J., 3rd, Aronova, S., Chu, S., Fairclough, S., Iverson, C., Wedaman, K.P. and Powers, T. (2004) TOR complex 1 includes a novel component, Tco89p (YPL180w), and cooperates with Ssd1p to maintain cellular integrity in Saccharomyces cerevisiae. J Biol Chem, 279, 14752-14762. Robertus, J.D., Ladner, J.E., Finch, J.T., Rhodes, D., Brown, R.S., Clark, B.F. and

46 Klug, A. (1974) Structure of yeast phenylalanine tRNA at 3 A resolution. Nature, 250, 546-551. Rohde, J.R., Campbell, S., Zurita-Martinez, S.A., Cutler, N.S., Ashe, M. and Cardenas, M.E. (2004) TOR controls transcriptional and translational programs via Sap-Sit4 protein phosphatase signaling effectors. Mol Cell Biol, 24, 8332-8341. Romanos, M.A. and Boyd, A. (1988) A transcriptional barrier to expression of cloned toxin genes of the linear plasmid k1 of Kluyveromyces lactis: evidence that native k1 has novel promoters. Nucleic Acids Res, 16, 7333-7350. Rozenski, J., Crain, P.F. and McCloskey, J.A. (1999) The RNA Modification Database: 1999 update. Nucleic Acids Res, 27, 196-197. Schafer, T., Maco, B., Petfalski, E., Tollervey, D., Bottcher, B., Aebi, U. and Hurt, E. (2006) Hrr25-dependent phosphorylation state regulates organization of the pre-40S subunit. Nature, 441, 651-655. Schafer, T., Strauss, D., Petfalski, E., Tollervey, D. and Hurt, E. (2003) The path from nucleolar 90S to cytoplasmic 40S pre-ribosomes. Embo J, 22, 1370-1380. Schaffrath, R. and Meacock, P.A. (2001) An SSB encoded by and operating on linear killer plasmids from Kluyveromyces lactis. Yeast, 18, 1239-1247. Schaffrath, R. and Meinhardt, F. (2005) Kluyveromyces lactis zymocin and other plasmid-encoded yeast killer toxins. In Schmitt, M. and Schaffrath, R. (eds.), Microbial protein toxins. Springer, New York, NY, Vol. 11, pp. 133-155. Schaffrath, R., Meinhardt, F. and Meacock, P.A. (1996) Yeast killer plasmid pGKL2: molecular analysis of UCS5, a cytoplasmic promoter element essential for ORF5 gene function. Mol Gen Genet, 250, 286-294. Schaffrath, R., Meinhardt, F. and Meacock, P.A. (1997) ORF7 of yeast plasmid pGKL2: analysis of gene expression in vivo. Curr Genet, 31, 190-192. Schaffrath, R., Meinhardt, F. and Meacock, P.A. (1999) Genetic manipulation of Kluyveromyces lactis linear DNA plasmids: gene targeting and plasmid shuffles. FEMS Microbiol Lett, 178, 201-210. Schaffrath, R., Stark, M.J., Gunge, N. and Meinhardt, F. (1992) Kluyveromyces lactis killer system: ORF1 of pGKL2 has no function in immunity expression and is dispensable for killer plasmid replication and maintenance. Curr Genet, 21,

47 357-363. Schickel, J., Helmig, C. and Meinhardt, F. (1996) Kluyveromyces lactis killer system: analysis of cytoplasmic promoters of the linear plasmids. Nucleic Acids Res, 24, 1879-1886. Sprinzl, M., Hartmann, T., Meissner, F., Moll, J. and Vorderwulbecke, T. (1987) Compilation of tRNA sequences and sequences of tRNA genes. Nucleic Acids Res, 15 Suppl, r53-188. Stark, M.J. and Boyd, A. (1986) The killer toxin of Kluyveromyces lactis: characterization of the toxin subunits and identification of the genes which encode them. Embo J, 5, 1995-2002. Stark, M.J., Boyd, A., Mileham, A.J. and Romanos, M.A. (1990) The plasmid-encoded killer system of Kluyveromyces lactis: a review. Yeast, 6, 1-29. Steyaert, J., Hallenga, K., Wyns, L. and Stanssens, P. (1990) Histidine-40 of ribonuclease T1 acts as base catalyst when the true catalytic base, glutamic acid-58, is replaced by alanine. Biochemistry, 29, 9064-9072. Sugisaki, Y., Gunge, N., Sakaguchi, K., Yamasaki, M. and Tamura, G. (1983) Kluyveromyces lactis killer toxin inhibits adenylate cyclase of sensitive yeast cells. Nature, 304, 464-466. Sugisaki, Y., Gunge, N., Sakaguchi, K., Yamasaki, M. and Tamura, G. (1984) Characterization of a novel killer toxin encoded by a double-stranded linear DNA plasmid of Kluyveromyces lactis. Eur J Biochem, 141, 241-245. Sugisaki, Y., Gunge, N., Sakaguchi, K., Yamasaki, M. and Tamura, G. (1985) Transfer of DNA killer plasmids from Kluyveromyces lactis to Kluyveromyces fragilis and Candida pseudotropicalis. J Bacteriol, 164, 1373-1375. Sutton, A., Immanuel, D. and Arndt, K.T. (1991) The SIT4 protein phosphatase functions in late G1 for progression into S phase. Mol Cell Biol, 11, 2133-2148. Takagi, H., Kakuta, Y., Okada, T., Yao, M., Tanaka, I. and Kimura, M. (2005) Crystal structure of archaeal toxin-antitoxin RelE-RelB complex with implications for toxin activity and antitoxin effects. Nat Struct Mol Biol, 12, 327-331. Takai, K. and Yokoyama, S. (2003) Roles of 5-substituents of tRNA wobble uridines in the recognition of purine-ending codons. Nucleic Acids Res, 31,

48 6383-6391. Takeda, M., Hiraishi, H., Takesako, T., Tanase, S. and Gunge, N. (1996) The terminal protein of the linear DNA plasmid pGKL2 shares an N-terminal domain of the plasmid-encoded DNA polymerase. Yeast, 12, 241-246. Tokunaga, M., Kawamura, A. and Hishinuma, F. (1989) Expression of pGKL killer 28K subunit in Saccharomyces cerevisiae: identification of 28K subunit as a killer protein. Nucleic Acids Res, 17, 3435-3446. Tokunaga, M., Kawamura, A., Kitada, K. and Hishinuma, F. (1990) Secretion of killer toxin encoded on the linear DNA plasmid pGKL1 from Saccharomyces cerevisiae. J Biol Chem, 265, 17274-17280. Tokunaga, M., Wada, N. and Hishinuma, F. (1987a) Expression and identification of immunity determinants on linear DNA killer plasmids pGKL1 and pGKL2 in Kluyveromyces lactis. Nucleic Acids Res, 15, 1031-1046. Tokunaga, M., Wada, N. and Hishinuma, F. (1987b) A novel yeast secretion vector utilizing secretion signal of killer toxin encoded on the yeast linear DNA plasmid pGKL1. Biochem Biophys Res Commun, 144, 613-619. Tomita, K., Ogawa, T., Uozumi, T., Watanabe, K. and Masaki, H. (2000) A cytotoxic ribonuclease which specifically cleaves four isoaccepting arginine tRNAs at their anticodon loops. Proc Natl Acad Sci U S A, 97, 8278-8283. Tommasino, M., Ricci, S. and Galeotti, C.L. (1988) Genome organization of the killer plasmid pGK12 from Kluyveromyces lactis. Nucleic Acids Res, 16, 5863-5878. Tyndall, C., Meister, J. and Bickle, T.A. (1994) The Escherichia coli prr region encodes a functional type IC DNA restriction system closely integrated with an anticodon nuclease gene. J Mol Biol, 237, 266-274. Umeda, N., Suzuki, T., Yukawa, M., Ohya, Y., Shindo, H., Watanabe, K. and Suzuki, T. (2005) Mitochondria-specific RNA-modifying enzymes responsible for the biosynthesis of the wobble base in mitochondrial tRNAs. Implications for the molecular pathogenesis of human mitochondrial diseases. J Biol Chem, 280, 1613-1624. von Heijne, G. (1983) Patterns of amino acids near signal-sequence cleavage sites. Eur J Biochem, 133, 17-21.

49 Wesolowski-Louvel, M., Tanguy-Rougeau, C. and Fukuhara, H. (1988) A nuclear gene required for the expression of the linear DNA-associated killer system in the yeast Kluyveromyces lactis. Yeast, 4, 71-81. White, J.H., Butler, A.R. and Stark, M.J. (1989) Kluyveromyces lactis toxin does not inhibit yeast adenylyl cyclase. Nature, 341, 666-668. Wilson, D.W. and Meacock, P.A. (1988) Extranuclear gene expression in yeast: evidence for a plasmid-encoded RNA polymerase of unique structure. Nucleic Acids Res, 16, 8097-8112. Winkler, G.S., Petrakis, T.G., Ethelberg, S., Tokunaga, M., Erdjument-Bromage, H., Tempst, P. and Svejstrup, J.Q. (2001) RNA polymerase II elongator holoenzyme is composed of two discrete subcomplexes. J Biol Chem, 276, 32743-32749. Wittschieben, B.O., Otero, G., de Bizemont, T., Fellows, J., Erdjument-Bromage, H., Ohba, R., Li, Y., Allis, C.D., Tempst, P. and Svejstrup, J.Q. (1999) A novel histone acetyltransferase is an integral subunit of elongating RNA polymerase II holoenzyme. Mol Cell, 4, 123-128. Worsham, P.L. and Bolen, P.L. (1990) Killer toxin production in Pichia acaciae is associated with linear DNA plasmids. Curr Genet, 18, 77-80. Yajima, S., Inoue, S., Ogawa, T., Nonaka, T., Ohsawa, K. and Masaki, H. (2006) Structural basis for sequence-dependent recognition of colicin E5 tRNase by mimicking the mRNA-tRNA interaction. Nucleic Acids Res, 34, 6074-6082. Yajima, S., Nakanishi, K., Takahashi, K., Ogawa, T., Hidaka, M., Kezuka, Y., Nonaka, T., Ohsawa, K. and Masaki, H. (2004) Relation between tRNase activity and the structure of colicin D according to X-ray crystallography. Biochem Biophys Res Commun, 322, 966-973. Yokoyama, S. and Nishimura, S. (1995) Modified nucleosides and codon recognition. In Söll, D. and RajBhandary, U. (eds.), tRNA : structure, biosynthesis, and function. ASM Press, Washington, D.C., pp. 207-223. Yokoyama, S., Watanabe, T., Murao, K., Ishikura, H., Yamaizumi, Z., Nishimura, S. and Miyazawa, T. (1985) Molecular mechanism of codon recognition by tRNA species with modified uridine in the first position of the anticodon. Proc Natl Acad Sci U S A, 82, 4905-4909.

50 Zink, S., Mehlgarten, C., Kitamoto, H.K., Nagase, J., Jablonowski, D., Dickson, R.C., Stark, M.J. and Schaffrath, R. (2005) Mannosyl-diinositolphospho-ceramide, the major yeast plasma membrane sphingolipid, governs toxicity of Kluyveromyces lactis zymocin. Eukaryot Cell, 4, 879-889.