Functional aspects of wobble uridine modifications

in yeast tRNA

Anders Esberg

Department of Molecular Biology Umeå University

Umeå, 2007

Department of Molecular Biology Umeå University

SE-90187 Umeå, Sweden

Copyright © 2007 Anders Esberg ISBN: 978-91-7264-302-4

Printed by: Solfjädern Offset, Umeå 2007

2

Table of contents Abstract………………..…………………...………………….. 5 Main references…..…….………………...…………………… 6 1. Introduction

The central dogma.………..……………….....…………….. 7 Transfer RNA……….……..…………………...…………... 8 The genetic code…….………………...………….…….….. 11 Translation………….………………………………….…… 13 Translational fidelity….……….…………….………….….. 17 Translational errors….….…………………….………...….. 18 Wobble uridine modifications.………………..………….… 19 The role of wobble uridine modifications in decoding.……. 21 Proteins required for wobble uridine modifications…..….... 23

2. Results and discussion…….……………….………………. 27 3. Conclusions………….……………………………………… 42 4. Acknowledgements…………………..……………...……… 43 5. References…………….……………………………….……. 44 6. Papers (I-III)…………….….…………….……….……….. I-III

3

4

Abstract Transfer RNAs (tRNA) function as adaptor molecules in the translation of mRNA into protein. These adaptor molecules require modifications of a subset of their nucleosides for optimal function. The most frequently modified nucleoside in tRNA is position 34 (wobble position), and especially uridines present at this position. Modified nucleosides at the wobble position are important in the decoding process of mRNA, i.e., restriction or improvement of codon-anticodon interactions. This thesis addresses the functional aspects of the wobble uridine modifications. The Saccharomyces cerevisiae Elongator complex consisting of the six Elp1-Elp6 proteins has been proposed to participate in three distinct cellular processes; elongation of RNA polymerase II transcription, regulation of polarized exocytosis, and formation of modified wobble nucleosides in tRNA. In Paper I, we show that the phenotypes of Elongator deficient cells linking the complex to transcription and exocytosis are counteracted by increased level of and . These tRNAs requires the Elongator complex for formation of the 5-methoxycarbonylmethyl (mcm

lnGUUGsmcm 25tRNA

LysUUUsmcm 25tRNA

5 ) group of their modified wobble nucleoside 5-methoxycarbonylmethyl-2-thiouridine (mcm 5 s 2 U). Our results therefore indicate that the relevant function of the Elongator complex is in formation of modified nucleosides in tRNAs and the defects observed in exocytosis and transcription are indirectly caused by inefficient translation of mRNAs encoding gene products important for these processes. The lack of defined mutants in eukaryotes has led to limited understanding about the role of the wobble uridine modifications in this domain of life. In Paper II, we utilized recently characterized mutants lacking the 2-thio (s 2 ) or 5-carbamoylmethyl (ncm 5 ) and mcm 5 groups to address the in vivo function of eukaryotic wobble uridine modifications. We show that ncm 5 and mcm 5 side-chains promote reading of G-ending codons, and that presence of a mcm 5 and an s 2 group cooperatively improves reading of both A- and G-ending codons. Previous studies revealed that a S. cerevisiae strain deleted for any of the six Elongator subunit genes shows resistance towards a toxin (zymocin) secreted by the dairy yeast Kluyveromyces lactis. In Paper III, we show that the cytotoxic γ subunit of zymocin is a tRNA endonuclease that target the anticodon of mcm 5 s 2 U 34 containing tRNAs and that the wobble mcm 5 modification is required for efficient cleavage. This explains the γ-toxin resistant phenotype of Elongator mutants which are defective in the synthesis of the mcm 5 group.

5

Main References: This thesis is based on the following papers, which are referred to in the text by their roman numerals.

I Esberg A., Huang B., Johansson M.J.O., and Byström A.S. (2006).

Elevated levels of two tRNA species bypass the requirement for the

Elongator complex in transcription and exocytosis.

Molecular Cell 24, 139–148.

II Johansson M.J.O., Esberg A., Huang B., Björk G., and Byström A.S.

(2007). Eukaryotic wobble uridine modifications promote a functionally

redundant decoding system.

Manuscript.

III Lu J., Huang B., Esberg A., Johansson M.J.O., and Byström A.S.

(2005). The Kluyveromyces lactis γ-toxin targets tRNA anticodons.

RNA 11, 1648-1654.

6

1. Introduction

The central dogma

The central dogma of molecular biology describes the flow of genetic

information from DNA to protein (Figure 1).

mRNA Protein

Transcription TranslationDNA

replication

Figure 1. The central dogma of molecular biology.

DNA is composed of four building blocks: the deoxyribonucleotides

containing the bases Adenine (A), Guanine (G), Cytosine (C), or Thymine

(T). The function of DNA is to provide long-term storage for vital genetic

information, and to function as a blueprint for essential components like

proteins and RNA molecules. The first process in protein synthesis,

transcription, is where an RNA molecule, messenger RNA (mRNA), is

synthesized (transcribed) from a selective DNA region (gene). In the second

process, translation, the mRNA is decoded to generate a specific amino acid

chain (protein) according to the rules defined by the genetic code. The

mRNA nucleotides are decoded in a triplet manner (codons) where each

three-letter code specifies a specific amino acid. However, the mRNA

codons are not directly recognized by amino acids. Translation of mRNA

codons into specific amino acids is dependent on adaptor molecules, which

7

are able to bind a defined amino acid and recognize a specific codon. These

adaptor molecules are small RNA molecules called transfer RNA (tRNA)

that contain a nucleotide-triplet (anticodon) designed to match and decode

the different mRNA codons. The decoding occurs in the ribosome, which is

a large ribonucleoprotein complex capable of binding the mRNA and guide

the tRNA into the correct position on the mRNA. When matching occurs

between codon and anticodon, the amino acid linked to the tRNA will be

incorporated into the growing amino acid chain. This thesis will focus on the

functional aspects of modified uridines present at the first position in the

anticodon.

Transfer RNA

Transfer RNA was discovered by Hoagland et al. 1956 and later described as

a molecule important in translating mRNA into proteins (Hoagland et al.,

1956; For review, RajBhandary and Kohrer, 2006). These small RNA

molecules are composed of four different nucleosides Uridine (U),

Adenosine (A), Guanosine (G), and Cytidine (C). Transfer RNAs are

between 75 and 95 nucleotides in length and usually visualized in the

canonical cloverleaf structure in which the tRNA is divided into the variable

arm, acceptor stem, D-, TΨC-, and anticodon-loop (Figure 2A). All tRNAs

fold into a similar L-shaped three-dimensional structure (Figure 2B)

promoted by base pairs between well-conserved positions within the tRNA

(Dirheimer et al., 1995; Holley, 1965; Holley et al., 1965).

8

In eukaryotes, the RNA polymerase III (Pol III) transcription machinery

transcribes the genes coding for tRNAs. The promoter present in tRNA

genes has internal A- and B-boxes, which correspond to the D- and

TΨC-stem loops of the transcribed tRNA. Transcription by Pol III involves a

multi-step assembly of initiation factors into a pre-initiation complex that

directs recruitment of Pol III. To activate transcription of tRNA genes, the

TFIIIC binds to the A- and B-boxes and acts as an assembly factor to direct

binding of the TFIIIB to an upstream position of the tRNA gene. Assembled

TFIIIB recruits the Pol III enzyme and directs multiple-rounds of

transcription (For review, Schramm and Hernandez, 2002).

3634

CCA-end

TΨC-loopD-loop

Anticodon-loop

Variable arm

Acceptor-stem

ACC

CCA-end

Anticodon-loop

A B

5´

3´

5´3´

3534

3536

Figure 2. Schematic structure of a tRNA. (A) Two-dimensional structure. (B)

Tertiary-structure (Kim et al., 1974). Positions 34, 35, and 36 (anticodon), are shown

in black.

9

All eukaryotic tRNA molecules are transcribed as precursors with extra

sequences at both the 5´ and 3´ ends, which have to be processed to generate

a mature tRNA. Removal of the 5´ leader is accomplished by a

ribonucleoprotein, RNaseP, whereas the 3´ end is processed by the

endoribonuclease RNaseZ. The primary tRNA transcript from eukaryotic

cells does not contain the 3´ CCA sequence, which is essential for

aminoacylation (Figure 2). The 3´ CCA sequence is added by the

tRNA-nucleotidyltransferase. Ten S. cerevisiae tRNA species contain an

intervening sequence in their primary transcript. The introns can vary in size

between 14-60 nucleotides, and are always located one nucleotide 3´ of the

anticodon. Splicing involves multiple-steps, first, the intron is removed by

the tRNA-splicing endonuclease. Secondly, the RNA-ends are ligated

together by the tRNA-ligase, and finally the 2´ phosphate group present after

ligation is removed by a phosphotransferase. For the fidelity of translation,

each tRNA must be charged with its cognate amino acid. The

aminoacyl-tRNA synthetases (aaRS) link the correct amino acid to the 3´

CCA end of the cognate tRNA. In general, there is one aaRS for each of the

20 amino acids used in protein synthesis. Finally, to generate a fully

functional adaptor molecule, specific tRNA modifications have to be

introduced post-transcriptionally. The different modified nucleosides present

in the tRNA are derivatives of the normally present nucleotides A, C, G, or U,

and up till now twenty-five different tRNA modifications have been found in

cytoplasmic S. cerevisiae tRNA (For reviews, Björk, 1995; Hopper and

Phizicky, 2003; Ibba and Söll, 2000; Johansson and Byström, 2005).

10

The genetic code

The mRNA sequence is decoded in a triplet manner into a specific amino

acid sequence by a three-nucleotide complementary region, the anticodon, in

the tRNA (Figure 3).

5´ 3´AI AII AIII

mRNA

5´3´

UU U3436 35

X X X Z Z Z

Figure 3. A schematic drawing of codon-anticodon interaction. The mRNA codon is

composed of three nucleotides A I , A II , and A III . The anticodon of the tRNA is

composed of three nucleotides at position 34, 35, and 36. U 36 will interact with the

first position (A I ), U 35 with the second- (A II ), and U 34 (wobble base) with the

third-position (A III ) of the mRNA codon.

The deciphering of the three-letter codons generated what we today call the

“Genetic Code”. The genetic code is in principle universal, consisting of 64

triplet letters, where 61 codons code for amino acids and three codons

terminate translation . The genetic code is degenerate, i.e., many amino acids

are specified by more than one codon. Two observations provide an

explanation for the degeneracy of the genetic code. First, the existence of

11

isoaccepting tRNA species charged with the same amino acid. Second, the

“wobble hypothesis” proposed by Crick, which predicts that some tRNA

species can decode more than one codon (Crick, 1966; Crick et al., 1961).

The wobble hypothesis stated that the nucleoside at the first position of the

anticodon (wobble position) could form either a canonical or a wobble base

pair with the third nucleoside in the codon, provided that the nucleosides at

position 35 and 36 in the tRNA form Watson-Crick base pairs with the first

two nucleosides of the codon (Crick, 1966) (Figure 3).

The “Codon Table” summarizes all the 64 possible triplets and clarifies

which codon specifies which amino acid (Figure 4). The two first letters of

the three letter codon create 16 different combinations, constructing the 16

codon boxes in the codon table (Figure 4). There are two types of codon

boxes, the family- and the split-codon boxes. In a family codon box, all four

codons code for the same amino acid, whereas a split codon box contains

codons for more than one amino acid.

12

Phe

Leu

Ile

Leu

Val

SerTyr

Stop

His

Gln

Asn

Lys

Asp

Glu

StopTrp

ArgPro

Thr

Ala

Ser

Arg

Gly

UUUUUCUUAUUG

CUUCUCCUACUG

AUUAUCAUAAUG

GUUGUCGUAGUG

UCUUCCUCAUCG

UAUUACUAAUAG

UGUUGCUGAUGG

CCUCCCCCACCG

CAUCACCAACAG

CGUCGCCGACGG

ACUACCACAACG

AAUAACAAAAAG

AGUAGCAGAAGG

GCUGCCGCAGCG

GAUGACGAAGAG

GGUGGCGGAGGG

Cys

Met

Figure 4. The codon table. The genetic code is composed of four different letters U, C,

A, and G, generating in total 64 different three-letter combinations. The codon table

summarizes all 64 codons and clarifies which codon specifies which amino acid.

Translation

Protein translation is one of the most conserved processes in gene expression.

Translating mRNA into protein requires three different RNA species, mRNA,

tRNA, and rRNA. In addition, this process also requires a number of

proteins important for assistance throughout the translation process. The

eukaryotic ribosome is a large multi-subunit complex composed of 4 RNA

molecules and ~76 proteins, which assemble into the small 40S and the large

60S subunit, which together form the mature 80S ribosome. The ribosome is

13

capable of binding tRNA in three separate binding sites: the A-site, P-site,

and E-site. Each of these sites has a distinct function, the A-site binds the

a minoacyl-tRNA, the P-site binds the p eptidyl-tRNA, and the E-site

( E xit-site) binds the deacyl-tRNA. Translation of mRNA into protein can be

divided into three distinct steps, initiation, elongation, and termination.

Initiation of translation is a sequence of events important for

identification of the correct start codon and determination of the correct

reading frame. According to the scanning model of translation initiation, the

pre-initiation complex (40S, initiation factors, and the initiator methionine

) binds to the 5´ end of the mRNA (Figure 5A). The pre-initiation

complex scans the mRNA in the 5´ to 3´ direction searching for the start

codon (Figure 5B). Once the start codon is identified, the initiation factors

are released and the 60S is recruited, assembling the 80S ribosome (Figure

5C). There are three potential reading frames in an mRNA and the reading

frame is fixed by the identification of the start codon. Normally a reading

frame is maintained and the mRNA translated in one continuous motion

from start to finish (Hershey and Merrick, 2000).

MetCAUtRNA

During translation elongation, the ribosome moves stepwise on the

mRNA, codon by codon, guiding the tRNA into the correct position. The

ribosome faithfully selects aminoacyl-tRNA based on how well it fits into

the A-site and the strength of the codon-anticodon interaction. Each

elongation cycle will add one amino acid to a growing amino acid chain. An

elongation cycle starts when a ternary complex, consisting of the

aminoacyl-tRNA, elongation factor 1A (eEF1A), and GTP, enters the

14

ribosomal A-site, (Figure 5D). If matching occurs between the

codon-anticodon, hydrolysis of the GTP promotes eEF1A-GDP release

(Figure 5E). The amino acid chain is transferred from the P-site tRNA to the

A-site tRNA by the enzymatic activity of the ribosome peptidyl transferase

center (Figure 5F). In this process, a new amino acid is added to the growing

amino acid chain. Following peptide transfer, the elongation factor 2 (eEF2)

promotes translocation of the peptidyl-tRNA to the P-site (Figure 5G). In

addition to eEF1A and eEF2, elongation factor 3 (eEF3) promotes release of

the E-site tRNA from the ribosome, and the elongation factor 1B (eEF1B)

recycles the eEF1A-GDP to eEF1A-GTP (Figure 5) (Merrick and Nyborg,

2000).

Of the 64 different codons, three are designated to terminate protein

translation (UAA, UAG, and UGA) (Figure 4). These codons are not

decoded by a tRNA into an amino acid, instead these codons are recognized

by the eukaryotic release factor 1 (eRF1) (Figure 5H). When the eRF1 binds

a stop codon, it will together with release factor 3 (eRF3) promote

translation termination. In this process, the amino acid chain dissociates

from the P-site tRNA, the mRNA is released, and ribosomal subunits

dissociate (Figure 5I) (Welch et al., 2000).

15

C

D

EF

G

APE

APE

APE

APE

H

APE

eRF1

I

Elongation

Initiation Termination

(STOP)

APE

A

APE5´

(AUG)

40S

60S

40S

APE5´

B

Peptidyl transfer

eEF1B

eRF3

Cap

Cap

eEF3

TranslocationeEF2

eEF1

A

GDP

eEF1

A

GTP

Proofreading step 1

Proofreading step 2

eRF3

eRF1

Figure 5. Simple view of eukaryotic protein translation. (A-B) Initiation of translation.

(C-G) Translation elongation. (H-I) Translation termination.

16

Translational fidelity

A vital task in protein synthesis is to incorporate the correct amino acid and

maintain the correct reading frame at the same time sustaining speed. On

average, the translation machinery incorporates ~20 amino acids per second;

however, the machinery is not perfect since an incorrect amino acid is

occasionally incorporated. Cells have in the course of evolution, evolved

numerous proofreading steps to improve translational accuracy.

To assure translational fidelity, each tRNA species has to be charged

with its cognate amino acid. To increase the accuracy of tRNA

aminoacylation, the aaRS accepts only the correct amino acid into the

substrate-binding site. To further reduce misaminacylation of incorrect

substrate tRNAs, a number of identity elements are present within the

different tRNA species. In general, the identity element lies in the two distal

parts of the tRNA, with the major determinants in the anticodon loop and the

acceptor stem. Moreover, modified nucleosides, especially those located in

the anticodon region of the tRNA, have been shown to be important for

charging (For reviews, Giege et al., 1998; Ibba and Söll, 2000).

According to current models, two independent proofreading steps assure

selection of a correct aminoacyl-tRNA into the A-site of the ribosome

(Thompson, 1988). The first proofreading step involves the initial binding of

the ternary complex to the ribosome. Here the primary strength of the

codon-anticodon interaction determines if the tRNA should be accepted into

the ribosome (Figure 5D). The second proofreading step is after GTP

hydrolysis and release of the eEF1A-GDP, here the ribosome senses the

17

codon-anticodon strength and the A-site fitness, and together these features

determine whether the aminoacyl-tRNA should be rejected or not (Figure 5E)

(For review, Ogle and Ramakrishnan, 2005). In the selection of the cognate

aminoacyl-tRNA into the A-site of the ribosome, modifications present at

the wobble position have been shown to play a vital role, primarily by

influencing the codon-anticodon interaction (Agris, 1991; Lim, 1994; Takai

and Yokoyama, 2003; Yokoyama and Nishimura, 1995; Yokoyama et al.,

1985). In addition to these proofreading steps, the E-site tRNA has also been

shown to be important for the selection of correct A-site tRNA, but the

mechanism for this proofreading step is not clear (Nierhaus, 2006).

Translational errors

Even though the translational machinery is capable of translating long

mRNA faithfully, errors now and then occur. During the translation

elongation process, three classes of translational errors can be distinguished.

(I) Progressive errors, the release of either a shorter or a longer form of

the amino acid chain. At least three different events can cause this kind of

translational error, (1) read through of stop codons, (2) incorrect

translational termination, or (3) drop-off of peptidyl-tRNA.

(II) Missense errors, an incorrect amino acid incorporated at sense

codons. This kind of error is usually caused by misreading of the third

nucleotide of the mRNA codon by the wobble base of the tRNA. In addition,

tRNA can also be misaminoacylated and thereby incorporate an incorrect

amino acid even though it pairs with the matching codon.

18

(III) Frameshifting errors, a shift in the reading frame. Hypothetically,

there are four different ways to induce frameshifting: (1) Transcriptional

errors expanding or contracting the codon size, (2) incoming

aminoacyl-tRNA binds out of frame, (3) translocation error, or (4) slippage

of the peptidyl-tRNA. After a frameshifting error, the original genetic

message is completely altered and can produce either longer or shorter form

of proteins.

It is broadly accepted that a single mRNA that encodes for a defined

amino acid chain will give rise to identical protein products, assuming that

these proteins go through the same post-translational modifications. This

notion was recently challenged by the findings that altered translational rate

could change protein folding and modify the protein function

(Kimchi-Sarfaty et al., 2007). It was hypothesized that temporal separation

of folding events throughout synthesis of a protein molecule might be vital

to ensure correct protein function. This implies that cells have a high

demand for a fully functionally translational machinery to maintain correct

translational speed to assure accurate protein folding.

Wobble uridine modifications

Transfer RNA modifications are found in all organisms and all over the

tRNA molecule. These modifications are introduced post-transcriptionally

and are derivatives of the normal nucleotides A, C, U, and G (For reviews,

Björk, 1995; Johansson and Byström, 2005). So far, 91 different tRNA

modifications have been found in the three kingdoms of life. Some tRNA

19

modifications are present at the same position, in the same population of

tRNA, and within all phylogenic domains, suggesting a conserved function

(Björk et al., 2001; RNAmods@lib.med.utah.edu).

The modifications present in tRNA have been shown to play a vital role

in numerous processes e.g., folding and stability of tRNA (Anderson et al.,

1998; Helm, 2006), recognition and anti-determinants for proteins

interacting with tRNAs (Åström and Byström, 1994; For review, Giege et al.,

1998), and intracellular localization of tRNA (Kaneko et al., 2003). The

most frequently modified base in tRNA is positions 34, and especially tRNA

carrying a uridine at the wobble position. Modifications at wobble position

are primarily thought to influence the decoding properties of the tRNA.

In total, there are 42 cytoplasmic S. cerevisiae tRNA species, and out of

these, thirteen contain a uridine at the wobble position in the primary

transcript. Of the thirteen U 34 containing tRNA species, one carries an

unmodified U ( ) and one has a pseudouridine (Ψ) ( ). The

remaining eleven U

LeuUAGtRNA Ile

AtRNA ΨΨ

34 containing tRNAs have either a 5-carbamoylmethyl

(ncm 5 ) or a 5-methoxycarbonylmethyl (mcm 5 ) group at the position 5

(Figure 6) (in many cases the ncm 5 and the mcm 5 modifications will be

referred to as xm 5 ) (Huang et al., 2005; For review, Johansson and Byström,

2005) (Paper II). Three of the mcm 5 containing tRNAs are thiolated at

position 2, generating the hyper modified

5-methoxycarbonylmethyl-2-thiouridine nucleoside (mcm 5 s 2 U) (Kobayashi

et al., 1974; Smith et al., 1973) (Paper III) (Figure 6).

20

N

OOH

OH

NH

O

O

OH

CH2

N

NH

O

O

OOH

OHOH

COCH 3

O

OOH

OHOH

CH2CNH 2

N

NH

O

O

O

CH2COCH 3

OOH

OHOH

N

NH

O

S

O

U ncm5U mcm5U mcm5s2U

Figure 6. Examples of uridine or uridine derivatives present at position 34 (wobble

position). From the left; uridine (U), 5-carbamoylmethyl-uridine (ncm 5 U),

5-methoxycarbonylmethyl-uridine (mcm 5 U), and 5-methoxycarbonylmethyl-2-thio

-uridine (mcm 5 s 2 U).

The role of wobble uridine modifications in decoding

In the decoding process, an aminoacyl-tRNA can potentially direct decoding

of three different types of codons. Cognate codon: the tRNA is able to form

three Watson-Crick base pairs, Near-cognate codons: two Watson-Crick base

pairs and one wobble base pair at the third position, and Non-cognate codon:

one or less Watson-Crick base pair. In the translational process, a

unmodified U 34 will according to Crick’s original wobble hypothesis

preferentially decode cognate A- and near cognate G-ending codons,

assuming that the nucleotides at position 35 and 36 in the tRNA form

Watson-Crick base pairs with the two first nucleotides of the mRNA codon

(Crick, 1966). At the time when Crick presented his hypothesis, it was not

known that cytoplasmic tRNA almost never contains an unmodified U 34 .

Since then, Crick’s hypothesis has been revised and it was suggested that an

21

unmodified U 34 recognizes all four nucleotides U, C, A, and G, whereas a

modified uridine can alter the decoding capacity (Agris, 1991; Björk, 1995;

Lim, 1994; Takai and Yokoyama, 2003; Yokoyama and Nishimura, 1995;

Yokoyama et al., 1985). Based on results from in vitro studies, as well as

structural considerations, it has been proposed that the presence of an xm 5

group at a wobble uridine would prevent pairing with U- and C-ending

codons (Lim, 1994; Yokoyama and Nishimura, 1995; Yokoyama et al., 1985).

For the decoding of A- and G-ending codons, two models are proposed. The

first and most accepted model states that presence of an xm 5 side-chain at

U 34 improves reading of A- and causes inefficient decoding of G-ending

codons (Yokoyama and Nishimura, 1995; Yokoyama et al., 1985). The

second model suggests that an xm 5 U nucleoside efficiently reads both A-

and G-ending codons (Kruger et al., 1998; Lim, 1994; Murphy et al., 2004;

Takai and Yokoyama, 2003; Yarian et al., 2002).

Three mcm 5 U 34 containing tRNAs are thiolated at position 2 of the

base. Interestingly, the 2-thio group is only found in tRNAs decoding in split

codon boxes. Similar to an xm 5 U 34 nucleoside, presence of an mcm 5 s 2 U 34

residue was suggested to prevent pairing with U- and C-ending codons,

allow efficient reading of A-, and simultaneously reduce the ability to pair

with G-ending codon (Lustig et al., 1981; Sekiya et al., 1969; Yokoyama et

al., 1985). More recent data have indicated that a mcm 5 s 2 U 34 residue would

allow reading of both A- and G-ending codons (Murphy et al., 2004).

22

Proteins required for wobble uridine modifications

Today over 50 gene products are dedicated for synthesis of tRNA

modifications in S. cerevisiae (Huang et al., 2005; For review, Johansson

and Byström, 2005; Nakai et al., 2007) (Lu J. personal communication).

Surprisingly, at least twenty-four of these are required for the synthesis of

wobble uridine modifications.

At this moment, fourteen gene products have been identified that are

required for synthesis of the xm 5 modification at U 34 . The Elongator

complex, consisting of six proteins Elp1-Elp6, is required for the formation

of ncm 5 and mcm 5 side-chains at wobble uridines (Figure 6) (Huang et al.,

2005). In addition to a role in tRNA modification, the Elongator complex

was proposed to participate in two other distinct cellular processes.

Originally the Elongator complex was identified based on its association

with the hyper-phosphorylated elongating form of RNA polymerase II (Pol

II) (Otero et al., 1999). Purified Elongator complex from yeast cells is able

to transfer acetyl groups from acetyl-CoA to histones H3 and H4 in vitro

(Hawkes et al., 2002; Kim et al., 2002; Winkler et al., 2002; Wittschieben et

al., 1999). Moreover, an elp3Δ strain shows a decreased acetylation level of

primarily histone H3 in vivo (Winkler et al., 2002). Together these data

suggested that the Elongator complex mediates Pol II transcription

elongation through chromatin structure. Another distinct function described

for the Elongator complex is in polarized transport of secretory vesicles to

the bud tip (Rahl et al., 2005). This transport event requires activation of the

vesicle-associated GTPase Sec4p by its guanine nucleotide exchange factor

23

Sec2p (Salminen and Novick, 1987; Walch-Solimena et al., 1997). It was

found that Sec2p associated with Elp1p and that polarized localization of

Sec2p was dependent on the presence of Elp1p, suggesting that the

Elongator complex regulates exocytosis by influencing the localization of

Sec2p (Rahl et al., 2005). In this thesis, we address whether the Elongator

complex has three distinct functions or whether it regulates one key process

that sequentially leads to downstream effects in the two other processes

(Paper I). We also utilized an elp3 null mutant to address the function of the

ncm 5 and mcm 5 side-chains at wobble uridines (Paper II). In addition to the

elp1–elp6 mutants, the formation of ncm 5 and mcm 5 modified wobble

uridines is abolished in the kti11, kti12, sit4, kti14, and the double sap185

sap190 mutants, whereas a kti13 mutant show significantly reduced levels

(Huang et al., 2005) (Lu J. personal communication).

The esterified methyl constituent of the mcm 5 group is synthesized by

the tRNA carboxyl methyltransferase, Trm9 protein (Kalhor and Clarke,

2003). The TRM9 gene was identified in a search for S. cerevisiae proteins

that contained putative AdoMet binding motifs, and later it was shown that

Trm9p used AdoMet as the donor for the tRNA methylation reaction (Kalhor

and Clarke, 2003; Niewmierzycka and Clarke, 1999). A strain deleted for the

TRM9 gene did not contain mcm 5 U or mcm 5 s 2 U in total tRNA (Figure 6),

and lacked methyl-esterified nucleosides in purified tRNA Arg and tRNA Glu .

Extracts from wild-type but not the trm9 mutant catalyzed incorporation of

methyl-esters into a substrate tRNA. This suggested that the Trm9 protein

catalyzed formation of methyl-esters in both mcm 5 U and mcm 5 s 2 U in

24

tRNA (Kalhor and Clarke, 2003). In this thesis, we used a trm9 null mutant

to address the role of the esterified methyl group in decoding (Paper II).

Three tRNAs, , , and carry

the hyper-modified nucleoside mcm

GlnUUGsmcm 25tRNA Lys

UUUsmcm 25tRNA GluUUCsmcm 25tRNA

5 s 2 U at the wobble position (Figure 6).

At the moment, there are ten gene products known to be required for the

synthesis of the 2-thio group at U 34 . The Tuc1 protein was identified by

sequence homology to the Escherichia coli TtcA protein, which is required

for the synthesis of s 2 C in position 32 of a subset of tRNAs (Björk G.

personal communication). However, no s 2 C modification has been identified

in S. cerevisiae tRNA, so it was speculated if this protein could be required

for the formation of the 2-thio group present in the mcm 5 s 2 U 34 nucleoside.

Accordingly, total tRNA isolated from a tuc1 null strain lack the s 2 group of

the mcm 5 s 2 U wobble nucleoside (Björk G. personal communication). The

Kluyveromyces lactis zymocin is a heterotrimeric toxin consisting of a α, β,

and γ subunit (Schaffrath and Meinhardt, 2005). The γ-toxin subunit is a

tRNA endonuclease cleaving , , and

at the 3’ side of the wobble nucleoside mcm

GlnUUGsmcm 25tRNA Lys

UUUsmcm 25tRNA

GluUUCsmcm 25tRNA 5 s 2 U (Paper III).

Presence of the mcm 5 s 2 U 34 nucleoside is important for efficient cleavage of

substrate tRNA and mutants defective in formation of either mcm 5 or s 2

side-chains are resistant to zymocin (Paper III) (Lu J. personal

communication). In a screen for zymocin resistant mutants, we identified

five mutants defective in formation of the s 2 group of the mcm 5 s 2 U 34 ,

among these the tuc1 and the tuc2 mutants (Lu J. personal communication).

In addition to Tuc1p and Tuc2p, the Nfs1p, Yor251Cp, Nbp35p, Cia1p,

25

Urm1p, Uba4p, Isu1p, and Isu2p are also required for the formation of s 2 U

at wobble position (Nakai et al., 2007) (Lu J. personal communication). In

this thesis, we use the tuc1 and tuc2 mutant (also called NCS6 and NCS2,

respectively) to address the function of the 2-thio group in decoding (Paper

II).

26

2. Results and discussion

Paper I: Elevated levels of two tRNA species bypass the

requirement for the Elongator complex in transcription and

exocytosis.

The S. cerevisiae Elongator complex, consisting of the Elp1-Elp6 proteins,

has been proposed to participate in three distinct cellular processes:

transcriptional elongation, exocytosis, and formation of modified wobble

uridines in tRNA. It was important to clarify whether the Elongator complex

has three distinct functions or whether it regulates one key process that

sequentially leads to downstream effects. A strain lacking Elongator function

displays a multitude of phenotypes, including temperature sensitive (Ts)

growth at 38°C (Frohloff et al., 2001; Jablonowski et al., 2001a; Krogan and

Greenblatt, 2001).

We investigated if there were genes in high dosage that could bypass the

requirement for an elp3 null mutant. Except for plasmids carrying the ELP3

gene, we identified five plasmids that partially suppressed the Ts phenotype.

All these plasmids contained inserts carrying a tK(UUU) gene, which codes

for . In a strain lacking Elongator function, this tRNA species

contains s

LysUUUsmcm 25tRNA

2 U rather than mcm 5 s 2 U at the wobble position. Since the

ELP1-ELP6 genes are required for formation of wobble ncm 5 and mcm 5

side-chains in 11 tRNA species (Huang et al., 2005) (Paper II) (Paper III),

we investigated whether elevated levels of any of the remaining 10 tRNAs

27

species would suppress the Ts phenotype of the elp3Δ strain. Weak

suppression of the Ts phenotype of the elp3Δ strain was obtained by

increased dosage of a tQ(UUG) gene which codes for mcm 5 s 2 U containing

. Simultaneously increased expression of the tK(UUU) and

tQ(UUG) genes in an elp3Δ strain resulted in a cooperative growth

improvement at both 30°C and 38°C. The species contain the

same wobble nucleoside (mcm

GlnUUGsmcm 25tRNA

GlnUUGsmcm 25tRNA

5 s 2 U) as and contain sLysUUUsmcm 25tRNA 2 U in

Elongator deficient cells (Paper III). Because strains with a mutation in any

of the ELP1-ELP6 genes display similar phenotypes, we tested if the

suppression generated by increased dosage of the tK(UUU) and tQ(UUG)

genes is also true in other Elongator mutants. Increased dosage of the

tK(UUU) and tQ(UUG) genes suppressed the Ts phenotype of all the

elp1-elp6 mutants. These results indicate with respect to growth defects,

elevated levels of the hypomodified and species

bypass the requirement for the Elongator complex.

GlnUUGs2tRNA Lys

UUUs2tRNA

The second function proposed for the Elongator complex was

acetylation of histones, and the histone acetyl transferase (HAT) activity was

primarily associated with the Elp3 protein. HATs transfer acetyl groups from

acetyl-CoA to the amino group of certain lysine residues within histone

N-terminal tails (Sterner and Berger, 2000). Lysine 14 in histone H3 has

been shown to be the major target for the HAT activity of Elp3p (Kuo et al.,

1996; Winkler et al., 2002). Histones isolated from elp3Δ cells with or

without elevated levels of and were analyzed with

anti-histone H3 and anti-acetyl-lys14-histone H3 antiserum. Increased levels

LysUUUs2tRNA Gln

UUGs2tRNA

28

of the hypomodified and restored the Lysine 14

acetylation levels of histone H3 in elp3Δ cells. Thus, the low acetylation

level of Lysine 14 in histone H3 isolated from elp3Δ cells is most likely

caused by a reduced function of the hypomodified and

, strongly questioning a role of Elp3p as a bona fide HAT.

LysUUUs2tRNA Gln

UUGs2tRNA

LysUUUs2tRNA

GlnUUGs2tRNA

The third distinct function proposed for the Elongator complex was the

delivery of transport vesicles to the bud-site (Rahl et al., 2005). Sec2p is a

guanine nucleotide exchange factor that triggers GDP to GTP exchange of

Sec4p (Walch-Solimena et al., 1997). Activated GTP-bound Sec4p

participates in delivery of vesicles to a specific region of the plasma

membrane (Salminen and Novick, 1987) (Walch-Solimena et al., 1997). A

sec2-59 mutant displays a Ts phenotype (Nair et al., 1990) and introduction

of an elp1 null allele into the sec2-59 mutant strain suppress the Ts

phenotype (Rahl et al., 2005). Increased expression of the hypomodified

and in the elp1Δ sec2-59 double mutant generated a

Ts phenotype, as observed in the sec2-59 single mutant. Thus, increased

levels of these tRNAs mimic the presence of Elp1p. Sec2p is a cytosolic

protein that normally concentrates in a distinct polarized manner to the site

of exocytosis (Elkind et al., 2000). In an elp1Δ strain, Sec2p is mislocalized

and it was suggested that a physical interaction between Elp1p and Sec2p

was important for the polarized localization of Sec2p (Rahl et al., 2005).

Elevated levels of the hypomodified and in the

elp1Δ strain restored polarized localization of Sec2p. This shows that an

interaction to Elp1p is not a requirement for a polarized localization of

LysUUUs2tRNA Gln

UUGs2tRNA

LysUUUs2tRNA Gln

UUGs2tRNA

29

Sec2p, rather it reflects the importance of the mcm 5 side-chain in

and for post-transcriptional expression of

gene products crucial in the exocytosis process.

LysUUUsmcm 25tRNA Gln

UUGsmcm 25tRNA

The mcm 5 or s 2 groups of the wobble mcm 5 s 2 U residue are both

believed to affect decoding in a similar manner (see above). Phenotypes

caused by removal of the s 2 or the mcm 5 group in and

are expected to be a consequence of less efficient decoding

Lys and Gln codons. We recently found that a strain with a tuc2 (also called

ncs2) null allele lacks the s

LysUUUsmcm 25tRNA

GlnUUGsmcm 25tRNA

2 group in mcm 5 s 2 U containing tRNA species

(see above). A tuc2 strain displays similar phenotypes as elp1 and elp3

mutant strains e.g., delayed transcriptional activation, chromatin

remodeling-, and exocytosis-defects. Increased levels of the mcm 5

containing and suppressed the phenotypes of a

tuc2 null strain, suggesting that the defects were caused by lack of the 2-thio

group at wobble position. We conclude that phenotypes proposed to be

caused by a direct effect of Elongator on exocytosis or transcription are

observed in a tuc2 mutant where the functionality of and

has been reduced by an Elongator-independent mechanism.

LysUUUmcm5tRNA Gln

UUGmcm5tRNA

LysUUUsmcm 25tRNA

GlnUUGsmcm 25tRNA

Summary:

In Paper I, we show that increased levels of the hypomodified

and could bypass the requirement for the Elongator complex in

transcription and exocytosis. As the Elongator complex is required for the

formation of the mcm

LysUUUs2tRNA

GlnUUGs2tRNA

5 group at position 34 in these tRNAs, we suggest that

30

the relevant function of the Elongator complex is to modify wobble uridine

in tRNA, and that transcription and exocytosis defects in Elongator mutant

are indirect effects resulting from a primary defect in translation (Figure 7).

mcm5s2

tRNA modificationExocytosis Transcription

Elongator complex

Sec2

Pol II

mRNA

Figure 7. Model: The Elongator complex is required to modify uridines at the wobble

position. The defects observed in exocytosis and transcription are indirectly caused

by inefficient translation of mRNAs encoding gene products important for these

processes.

31

Paper II: Eukaryotic wobble uridine modifications promote a

functionally redundant decoding system

Modified nucleosides in the anticodon region of tRNA are important in the

decoding process. Since defined mutants defective in formation of the s 2 or

ncm 5 and mcm 5 groups at wobble uridines have not been available, the in

vivo roles of these modifications have not been investigated. The Elongator

complex was recently shown to be required for formation of ncm 5 and mcm 5

side-chains at position 34 (Huang et al., 2005). In addition, the last methyl

group in the mcm 5 side-chain is synthesized by the Trm9p (Kalhor and

Clarke, 2003). Furthermore, the Tuc2p is required for the formation of the

2-thio group in mcm 5 s 2 U 34 containing tRNA species (Paper I) (Björk G.

personal communication). Strains with mutations in any of the ELP1-ELP6,

TRM9, or TUC2 genes are viable, making it possible to investigate the in vivo

role of the wobble ncm 5 , mcm 5 , and s 2 groups.

According to current models based on mainly physicochemical studies, a

tRNA with an unmodified U 34 may decode codons ending with any of the

four nucleotides (Lim, 1994; Yokoyama and Nishimura, 1995). To

investigate if an unmodified wobble uridine is able to read codons ending

with any of the nucleotides in vivo, we utilized the distribution of tRNA

species in the Leucine and Proline family codon boxes. In the Leucine box,

there are only two tRNA species present, the containing an

unmodified U

LeuUAGtRNA

34 (Randerath et al., 1979), and the harboring a GLeuGAGtRNA 34 .

Interestingly, a strain lacking the G 34 containing species is viable, LeuGAGtRNA

32

suggesting that an unmodified wobble uridine can read all four nucleotides.

In the Proline box, there are two tRNA species present, one contains

ncm 5 U 34 ( ) and the other one contains an A at the wobble

position (most likely converted to I

oPrUGGncm5tRNA

34 ). A strain lacking the A 34 containing

species is viable with no apparent growth defect. Introduction of

an elp3Δ allele into this strain did not generate a synthetic growth defect,

suggesting that the unmodified U

oPrAGGtRNA

34 in is also able to decode

codons ending with any of the four nucleotides. The ability of

to decode U- and C-ending codons is independent of the ncm

oPrUGGtRNA

oPrUGGncm5tRNA

5 group, which

could imply that reading of these codons involves a two out of three

interaction (Lagerkvist, 1978).

To investigate if the ncm 5 group in is important for

decoding A-ending codons, we reduced the copy number of the genes coding

for ncm

ValUACncm5tRNA

5 containing in yeast cells lacking the elp3 allele. In a

strain deleted for one of the two genes coding for , introduction

of the elp3Δ allele did not generate a synthetic growth defect. Since a tRNA

species with an I

ValUACncm5tRNA

ValUACncm5tRNA

34 residue is present in the Valine family codon box

( ), it is possible that lack of a synergistic phenotype is due to the

ability of the I

ValIACtRNA

34 containing tRNA to decode GUA codons. However, a strain

lacking the ncm 5 containing species is inviable, showing that

the I

ValUACncm5tRNA

34 containing cannot decode the GUA codons. We conclude

that the absence of an ncm

ValIACtRNA

5 side-chain at U 34 does not reduce the ability of

the to decode the GUA codons. ValUACncm5tRNA

To investigate if the ncm 5 group in is important for ValUACncm5tRNA

33

decoding G-ending codons, we deleted the two genes coding for the C 34

containing species. As this strain was viable, the ncmValCACtRNA 5 U 34

containing must be able to decode GUG codons. However,

introduction of the elp3 allele into a strain lacking the C

ValUACncm5tRNA

34 containing

generated lethality, showing that the ncmValCACtRNA 5 group in

is required for reading the GUG codons. We obtained similar results using

comparable genetic strategies for the Threonine and the Serine codon boxes.

Together these data supports a model where the ncm

ValUACncm5tRNA

5 U 34 nucleoside does

not enhance the ability to decode A-, but improves reading of G-ending

codons.

To test if the mcm 5 group at U 34 is important for decoding A-ending

codons, we deleted two of the three genes coding for the mcm 5 U 34

containing . Introduction of the elp3Δ allele into this strain did

not generate a synthetic growth defect, suggesting that the mcm

GlyUCCmcm5tRNA

5 group does

not notably affect the ability of to decode GGA codons. To

address the role of the mcm

GlyUCCmcm5tRNA

5 group in decoding G-ending codons, we

introduced an elp3Δ allele into a strain lacking the C 34 containing

species. A synthetic growth defect was observed, suggesting that the

presence of a mcm

GlyCCCtRNA

5 side-chain improves decoding of G-ending codons. We

obtained similar results using a comparable genetic approach in the Arginine

split codon box (AGN) to address the role of the mcm 5 group in decoding

G-ending codons. Taken together, this suggests that the mcm 5 residue

improves reading of G-, but has no significant influence on decoding

A-ending codons.

34

To investigate the role of the mcm 5 s 2 U 34 residue in the decoding of

G-ending codons, a strain with a deletion of the single copy C 34 containing

gene was constructed. This strain is inviable, suggesting that

is unable to read the G-ending codons. The presence of either

a mcm

lnGCUGtRNA

lnGUUGsmcm 25tRNA

5 or an s 2 group could potentially prevent reading of G-ending

codons. To test this hypothesis we deleted the single copy C 34 containing

gene in either an elp3Δ or a tuc1Δ null strain. Lack of neither the

mcm

lnGCUGtRNA

5 nor the s 2 group suppressed the inviability of a strain lacking the C 34

containing species. Thus, normal levels of the

carrying either the mcm

lnGCUGtRNA lnG

UUGsmcm 25tRNA5 s 2 U 34 , mcm 5 U 34 or the s 2 U 34 at wobble position

are not able to decode G-ending codons. Interestingly, elevated levels of the

suppressed the need for ClnGUUGsmcm 25tRNA 34 containing in a

wild-type background. This suppression was not observed in an elp3Δ or

tuc1Δ background, suggesting that the mcm

lnGCUGtRNA

5 and the s 2 groups

cooperatively improve decoding of G-ending codons.

Summary

According to current models, an unmodified U 34 should be able to decode all

four nucleotides U, C, A, and G (Lim, 1994; Yokoyama and Nishimura,

1995). This is true for two of the tRNA species we have investigated. We

show that the and the hypomodified , both containing

an unmodified U

LeuUAGtRNA oPr

UGGtRNA

34 , are able to decode codons ending with any of the four

nucleosides. However, the decoding ability of U 34 is highly specific as an

unmodified U 34 in or , is incapable of reading their ThrUGUtRNA Ser

UGAtRNA

35

respective G-ending codons. Therefore, we suggest that it is rather an

exception than a role for an unmodified U 34 to be able to decode all four

nucleotides and we do not exclude the possibility that reading of these

codons could involve a two out of three interaction (Lagerkvist, 1978).

According to current models, presence of an xm 5 side-chain is

predicted to either restrict reading to A- or allow efficient decoding of both

A- and G-ending codons (Kruger et al., 1998; Lim, 1994; Murphy et al.,

2004; Takai and Yokoyama, 2003; Yarian et al., 2002; Yokoyama and

Nishimura, 1995; Yokoyama et al., 1985). We found no evidence for a role

of the xm 5 side-chain in , , and in

the decoding of their respective A-ending codons. However, we could show

that presence of an xm

ValUACncm5tRNA Ser

UGAncm5tRNA GlyUCCmcm5tRNA

5 side-chain in ,

, and improves reading of G-ending codons.

ValUACncm5tRNA Ser

UGAncm5tRNAArg

UCUmcm5tRNA GlyUCCmcm5tRNA

The presence of a mcm 5 s 2 U 34 residue was originally proposed to allow

the tRNA to efficiently read the cognate A-ending codon and simultaneously

reduce the ability to decode G-ending codons (Lustig et al., 1981; Sekiya et

al., 1969; Yokoyama et al., 1985). Together the data presented in Paper I and

Paper II, suggests that the presence of the mcm 5 and the s 2 group

cooperatively improves decoding of both A- and G-ending codons.

36

(9)mcm5s2U

U

C

A

G

(7)ncm5Um

(3) U

(2)b ncm5U

(2) Ψ (7)mcm5s2U

(14)mcm5s2U

Phe

Leu

Ile

Leu

Met

Val

Ser Tyr

His

Gln

Asn

Lys

Asp

Glu

Cys

Trp

Arg

Stop

Stop

(10)ncm5U

(5) ncm5U

(4) ncm5U

(3)a ncm5U

UCAGUCAGUCAGUCAG

UCAGUCAGUCAGUCAG

U AC GU AC G

Stop

(1)b

(1)a

(10)

(13)

(14)

Pro

Thr

Ala

(10)

(4+5)c

(2)a

(1)b

(11)

(2)a

(11)

(11)

(8)

(7)

(10)

(15)

(4)

(6)

Ser

Arg

Gly

(11)mcm5U

(3)mcm5U

(1)b

(14)

(2)b

(4)

(16)

(6)

(1)b

(1)a

(2)a

Third position

Second position

Fris

t pos

ition

Figure 8. Summary of the result presented in Paper II. Circles connected with a line

indicate one tRNA species. Transfer RNA species addressed in Paper II are labeled

with black or gray circles. Black circles connected with a line shows that a tRNA

species efficiently reads the indicated codons. Gray circles linked with a dashed line

indicate that a tRNA species reads the specified codons only when over-expressed.

The nucleoside at the wobble position is given for the U 34 containing tRNA species.

The number of genes coding for a tRNA species is indicated next to the circle for the

cognate codon, which also means that the primary anticodon sequence can be

deduced from the figure. a The gene(s) encoding the tRNA is nonessential. b The

gene(s) encoding the tRNA is essential.

37

Paper III: The Kluyveromyces lactis γ-toxin targets tRNA

anticodons.

To gain competitive growth advantage, various microorganisms produce and

secrete toxins. One example is the dairy yeast K. lactis, which secretes a

heterotrimeric toxin (zymocin), that causes growth arrest of sensitive yeast

cells (Gunge and Sakaguchi, 1981; For review, Schaffrath and Meinhardt,

2005; Sugisaki et al., 1983; White et al., 1989). Zymocin is composed of a α,

β, and γ subunit. The α and β subunits assist transfer of the cytotoxic

γ subunit into cells (Butler et al., 1991a; Gunge et al., 1981; For reviews,

Schaffrath and Meinhardt, 2005; Stark et al., 1990; Tokunaga et al., 1989).

Two classes of S. cerevisiae mutants resistant to zymocin have been

isolated (Butler et al., 1991a; Butler et al., 1994; For review, Schaffrath and

Meinhardt, 2005; Sugisaki et al., 1983; White et al., 1989). Class I mutants

are defective in binding and/or uptake of zymocin, as these mutants are

resistant to extra genius zymocin, but sensitive to intracellular γ-toxin. Class

II mutants are resistant to both extra genius zymocin and intracellular γ-toxin,

and therefore considered target mutants. Interestingly, strains deleted for any

of the ELP1-ELP6 or KTI11–KTI13 genes have been shown to be class II

mutants (Butler et al., 1994; Fichtner and Schaffrath, 2002; Frohloff et al.,

2001; Jablonowski et al., 2001b). As the Elongator complex was primarily

thought to function in elongation of Pol II transcription, it was assumed that

the γ-toxin was targeted to the Pol II transcription machinery in an

Elongator-dependent manner, and that the growth arrest was caused by

transcriptional inactivation (For review, Schaffrath and Meinhardt, 2005).

38

Previously, it was shown that elevated levels of the mcm 5 s 2 U 34 containing

suppressed the zymocin sensitivity of a wild-type S.

cerevisiae strain (Butler et al., 1994). As all the class II mutants (elp1-elp6

and kti11-kti13) affect the synthesis of the mcm

GluUUCsmcm 25tRNA

5 group present in

, we hypothesized that the killer toxin-resistant phenotype

induced by these mutations could be a consequence of the inability to modify

wobble uridine in .

GluUUCsmcm 25tRNA

GluUUCsmcm 25tRNA

To test this hypothesis, we investigated the levels of in

wild-type, elp3Δ, or trm9Δ cells after induction of intracellular γ-toxin. The

amount of was reduced with increased induction time in

wild-type cells, whereas no reduction could be detected in elp3Δ or trm9Δ

cells. This supported the idea that the entire mcm

GluUUCsmcm 25tRNA

GluUUCsmcm 25tRNA

5 side-chain in

is required for the cytotoxicity of the γ-toxin. In addition to

the , the mcm

GluUUCsmcm 25tRNA

GluUUCsmcm 25tRNA 5 side-chain is present in ,

, , and , but no reduction of these

tRNAs could be observed in wild-type cells after intracellular γ-toxin

expression, suggesting that these tRNAs are not substrates in vivo.

lnGUUGsmcm 25tRNA

LysUUUsmcm 25tRNA Arg

UCUmcm5tRNA GlyUCCmcm5tRNA

To investigate if γ-toxin acts directly on tRNA, we incubated total tRNA

isolated from wild-type, elp3, or trm9 cells with purified γ−toxin. Cleavage

of , , and could be detected of

tRNA isolated from wild-type cells however, and

were cleaved with lower efficiency. The mcm

GluUUCsmcm 25tRNA Lys

UUUsmcm 25tRNA lnGUUGsmcm 25tRNA

LysUUUsmcm 25tRNA

lnGUUGsmcm 25tRNA 5 group is

important for efficient cleavage in vitro, as reduced γ−toxin cleavage

efficiency was observed for tRNA isolated from elp3 or trm9 null mutants.

39

Analysis of the cleavage products revealed that the γ−toxin cleaves

, , and between position 34 and

35, generating a 2´, 3´ cyclic-phosphate and a 5´ hydroxyl-group. This

confirms that the γ-toxin acts directly on tRNA and suggests that

, and might be γ-toxin targets

in vivo.

GluUUCsmcm 25tRNA Lys

UUUsmcm 25tRNA lnGUUGsmcm 25tRNA

GluUUCsmcm 25tRNA Lys

UUUsmcm 25tRNA lnGUUGsmcm 25tRNA

Consistent with the γ-toxin reactivity of the substrate tRNA in vivo and

in vitro, we showed that elevated levels of , but not

or , made a wild-type strain resistant to

exogenous zymocin. However, over-expression of in

combination with or or all three mcm

GluUUCsmcm 25tRNA

LysUUUsmcm 25tRNA lnG

UUGsmcm 25tRNA

GluUUCsmcm 25tRNA

LysUUUsmcm 25tRNA lnG

UUGsmcm 25tRNA 5 s 2 U 34

containing tRNA species synergistically increased zymocin resistance. These

data suggests that all three mcm 5 s 2 U 34 containing tRNAs, ,

, and , are γ-toxin substrates in vivo.

GluUUCsmcm 25tRNA

LysUUUsmcm 25tRNA lnG

UUGsmcm 25tRNA

Inhibition of protein synthesis arrests yeast cells in the G 1 phase of the

cell cycle (Hohmann and Thevelein, 1992; Johnston et al., 1977; Pyronnet

and Sonenberg, 2001; Unger and Hartwell, 1976; Wrobel et al., 1999).

Interestingly, zymocin also arrests sensitive yeast cells in the G 1 phase of the

cell cycle (Butler et al., 1991b). Therefore, we propose that the zymocin

induced G 1 arrest is most likely a consequence of a translational defect

caused by depletion or reduction of , , and

.

GluUUCsmcm 25tRNA Lys

UUUsmcm 25tRNA

lnGUUGsmcm 25tRNA

40

Summary:

In Paper III, we characterize the first eukaryotic toxin shown to target tRNAs.

We show that all three mcm 5 s 2 U 34 containing tRNA species, ,

, and are cleaved by the γ-toxin between

position 34 and 35. The γ-toxin resistance of elp1-elp6 and kti11-kti13

mutants can be explained by the fact that the these mutants are defective in

formation of the mcm

lnGUUGsmcm 25tRNA

LysUUUsmcm 25tRNA Glu

UUCsmcm 25tRNA

5 group present at position 34 which is required for

efficient cleavage (Figure 9).

34 3635

γ-toxin cleaves the three mcm s U containing tRNA species between nucleoside 34 and 35

5 2

mcm5s2

Figure 9. Model: K. lactis γ-toxin cleaves , , and

between positions 34 and 35; the mcm

Glu

UUC2s5mcmtRNA Lys

UUU2s5mcmtRNA lnG

UUG2s5mcmtRNA

5 group at position 34 is important for efficient

cleavage.

41

3. Conclusions

Elevated levels of two tRNA species bypass the requirement for

the Elongator complex in transcription and exocytosis.

Increased expression of and bypass the

requirement for the Elongator complex in transcription and exocytosis. This

observation suggests that the relevant function for the Elongator complex is

in tRNA modification (Paper I) (Figure 7).

GlnUUGsmcm 25tRNA Lys

UUUsmcm 25tRNA

Eukaryotic wobble uridine modifications promote a functionally

redundant decoding system.

The xm 5 side-chain present at the wobble uridine improves reading of

G-ending codons. The mcm 5 and s 2 groups cooperatively improve reading

of A- and G-ending codons (Paper II) (Figure 8).

The Kluyveromyces lactis γ-toxin targets tRNA anticodons.

The γ-toxin cleaves the three mcm 5 s 2 U 34 containing tRNA species,

, , and between position 34 and

35. The γ-toxin resistance of mutants defective in formation the mcm

GlnUUGsmcm 25tRNA Glu

UUCsmcm 25tRNA LysUUUsmcm 25tRNA

5 group

can be explained by the fact that this group is important for efficient cleavage

(Paper III) (Figure 9).

42

4. Acknowledgements

First, I would like to thank my supervisor Anders Byström for your passion in research, and for your ability to create an educating environment. Special thanks for all the wild ideas and discussions, and clearly showing me that the “crazy scientist” still exists. Not to be forgotten, behind a wild scientist there is a hardworking and helping wife, many thanks Mona for keeping him in line when needed and for the many great dinners during the years. Thereafter I would like to acknowledge past and present members of the AB group: Jian, Bo, Chen, Monica, and Marcus for everything during these years. Moreover, I would like to thank Glenn, Mikael, Tord, and their group members for the scientific discussions. I also would like to thank P-A, Johnny, the dishwashing-, media-, and secretarial-personnel for valuable services. Special thanks to the “Innebandy people” for many excellent moments on and off the court. It was great to know that Wednesdays was linked to entertainment, laughter, and now and then some pain. Promise me to at least try to keep up the good work even though, I know, it will be hard without me!

Outside the research field, I would like to acknowledge all my close friends: Ante, Blomman, Mike, Nisse, Niklas, Erik and Carola, Tobbe, Maria, and Linea, Magnus, Jessika and Petter, Per, Sussi, and Anna, Anna and Andreas, Niklas and Soffan, Leif and Kristina, and many, many, more. A special thanks to “The brothers” for not trying to understand my research. Nevertheless, all of you have provided with everything close friends can and much more, thanks again.

Special thanks to my mother Birgitta, father Folke, brother Stefan, his sambo

Nina, and their lovely children Viktor and Emilia for being there and supporting me when needed. I know, you know, how much you mean to me!

Finally, I would like to thank my life friend Maria for supporting me through

both the good and bad times. I hope that we have at least 70 more years to share and enjoy together, love U.

Thank you all /Anders

43

5. Reference

Agris, P.F. (1991) Wobble position modified nucleosides evolved to select transfer

RNA codon recognition: a modified-wobble hypothesis. Biochimie, 73,

1345-1349.

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.

Åström, S.U. and Byström, A.S. (1994) Rit1, a tRNA backbone-modifying enzyme

that mediates initiator and elongator tRNA discrimination. Cell, 79,

535-546.

Björk, G.R. (1995) Biosynthesis and Function of Modified Nucleosides. In tRNA:

Structure, Biosynthesis, and Function. ASM Press, Washington, DC, pp.

165-205.

Björk, G.R., Jacobsson, K., Nilsson, K., Johansson, M.J., Byström, A.S. and Persson,

O.P. (2001) A primordial tRNA modification required for the evolution of

life? Embo J, 20, 231-239.

Butler, A.R., Porter, M. and Stark, M.J. (1991a) 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

44

G1 arrest induced by Kluyveromyces lactis toxin. Mol Cell Biol, 14,

6306-6316.

Butler, A.R., White, J.H. and Stark, M.J. (1991b) Analysis of the response of

Saccharomyces cerevisiae cells to Kluyveromyces lactis toxin. J Gen

Microbiol, 137 ( Pt 7), 1749-1757.

Crick, F.H. (1966) Codon--anticodon pairing: the wobble hypothesis. J Mol Biol, 19,

548-555.

Crick, F.H., Barnett, L., Brenner, S. and Watts-Tobin, R.J. (1961) General nature of

the genetic code for proteins. Nature, 192, 1227-1232.

Dirheimer, D., Keith, G., Dumas, P. and Westhof, E. (1995) Primary, Secondary, and

Tertiary Structures of tRNAs. In Söll, D. and RajBhandary, U.L. (eds.),

tRNA: Structure, Biosynthesis and Function. ASM Press, Washington, DC,

pp. 93-126.

Elkind, N.B., Walch-Solimena, C. and Novick, P.J. (2000) The role of the COOH

terminus of Sec2p in the transport of post-Golgi vesicles. J Cell Biol, 149,

95-110.

Fichtner, L. and Schaffrath, R. (2002) KTI11 and KTI13, Saccharomyces cerevisiae

genes controlling sensitivity to G1 arrest induced by Kluyveromyces lactis

zymocin. Mol Microbiol, 44, 865-875.

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.

Giege, R., Sissler, M. and Florentz, C. (1998) Universal rules and idiosyncratic

features in tRNA identity. Nucleic Acids Res, 26, 5017-5035.

45

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.

Hawkes, N.A., Otero, G., Winkler, G.S., Marshall, N., Dahmus, M.E., Krappmann,

D., Scheidereit, C., Thomas, C.L., Schiavo, G., Erdjument-Bromage, H.,

Tempst, P. and Svejstrup, J.Q. (2002) Purification and characterization of

the human elongator complex. J Biol Chem, 277, 3047-3052.

Helm, M. (2006) Post-transcriptional nucleotide modification and alternative

folding of RNA. Nucleic Acids Res, 34, 721-733.

Hershey, J.W.B. and Merrick, W.C. (2000) The pathway and mechanism of

initiation of protein synthesis. In Translational Control of Gene Expression.

Cold spring harbor laboarotory press, New York, Vol. 2, pp. pp. 33-88.

Hoagland, M.B., Keller, E.B. and Zamecnik, P.C. (1956) Enzymatic carboxyl

activation of amino acids. J Biol Chem, 218, 345-358.

Hohmann, S. and Thevelein, J.M. (1992) The cell division cycle gene CDC60

encodes cytosolic leucyl-tRNA synthetase in Saccharomyces cerevisiae.

Gene, 120, 43-49.

Holley, R.W. (1965) Structure of an alanine transfer ribonucleic acid. Jama, 194,

868-871.

Holley, R.W., Apgar, J., Everett, G.A., Madison, J.T., Marquisee, M., Merrill, S.H.,

46

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.

Ibba, M. and Söll, D. (2000) Aminoacyl-tRNA synthesis. Annu Rev Biochem, 69,

617-650.

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., Frohloff, F., Fichtner, L., Stark, M.J. and Schaffrath, R. (2001b)

Kluyveromyces lactis zymocin mode of action is linked to RNA

polymerase II function via Elongator. Mol Microbiol, 42, 1095-1105.

Johansson, M.J.O. 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-Verlag,

New-York, Vol. 12, pp. 87-120.

Johnston, G.C., Pringle, J.R. and Hartwell, L.H. (1977) Coordination of growth with

cell division in the yeast Saccharomyces cerevisiae. Exp Cell Res, 105,

79-98.

Kalhor, H.R. and Clarke, S. (2003) Novel methyltransferase for modified uridine

47

residues at the wobble position of tRNA. Mol Cell Biol, 23, 9283-9292.

Kaneko, T., Suzuki, T., Kapushoc, S.T., Rubio, M.A., Ghazvini, J., Watanabe, K.

and Simpson, L. (2003) Wobble modification differences and subcellular

localization of tRNAs in Leishmania tarentolae: implication for tRNA

sorting mechanism. Embo J, 22, 657-667.

Kim, J.H., Lane, W.S. and Reinberg, D. (2002) Human Elongator facilitates RNA

polymerase II transcription through chromatin. Proc Natl Acad Sci U S A,

99, 1241-1246.

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.

Kimchi-Sarfaty, C., Oh, J.M., Kim, I.W., Sauna, Z.E., Calcagno, A.M., Ambudkar,

S.V. and Gottesman, M.M. (2007) A "silent" polymorphism in the MDR1

gene changes substrate specificity. Science, 315, 525-528.

Kobayashi, T., Irie, T., Yoshida, M., Takeishi, K. and Ukita, T. (1974) The primary

structure of yeast glutamic acid tRNA specific to the GAA codon. Biochim

Biophys Acta, 366, 168-181.

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.

Kruger, M.K., Pedersen, S., Hagervall, T.G. and Sörensen, M.A. (1998) The

modification of the wobble base of tRNAGlu modulates the translation rate

of glutamic acid codons in vivo. J Mol Biol, 284, 621-631.

Kuo, M.H., Brownell, J.E., Sobel, R.E., Ranalli, T.A., Cook, R.G., Edmondson, D.G.,

48

Roth, S.Y. and Allis, C.D. (1996) Transcription-linked acetylation by

Gcn5p of histones H3 and H4 at specific lysines. Nature, 383, 269-272.

Lagerkvist, U. (1978) "Two out of three": an alternative method for codon reading.

Proc Natl Acad Sci U S A, 75, 1759-1762.

Lim, V.I. (1994) Analysis of action of wobble nucleoside modifications on

codon-anticodon pairing within the ribosome. J Mol Biol, 240, 8-19.

Lustig, F., Elias, P., Axberg, T., Samuelsson, T., Tittawella, I. and Lagerkvist, U.

(1981) Codon reading and translational error. Reading of the glutamine and

lysine codons during protein synthesis in vitro. J Biol Chem, 256,

2635-2643.

Merrick, W.C. and Nyborg, J. (2000) The protein biosynthesis elongation cycle. In

Translational Control of Gene Expression. Cold spring harbor laboarotory

press, New York, Vol. 2, pp. pp. 89-125.

Murphy, F.V.t., Ramakrishnan, V., Malkiewicz, A. and Agris, P.F. (2004) The role of

modifications in codon discrimination by tRNA(Lys)UUU. Nat Struct Mol

Biol, 11, 1186-1191.

Nair, J., Muller, H., Peterson, M. and Novick, P. (1990) Sec2 protein contains a

coiled-coil domain essential for vesicular transport and a dispensable

carboxy terminal domain. J Cell Biol, 110, 1897-1909.

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.

Nierhaus, K.H. (2006) Decoding errors and the involvement of the E-site. Biochimie,

88, 1013-1019.

49

Niewmierzycka, A. and Clarke, S. (1999) S-Adenosylmethionine-dependent

methylation in Saccharomyces cerevisiae. Identification of a novel protein

arginine methyltransferase. J Biol Chem, 274, 814-824.

Ogle, J.M. and Ramakrishnan, V. (2005) Structural insights into translational

fidelity. Annu Rev Biochem, 74, 129-177.

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.

Pyronnet, S. and Sonenberg, N. (2001) Cell-cycle-dependent translational control.

Curr Opin Genet Dev, 11, 13-18.

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.

RajBhandary, U.L. and Kohrer, C. (2006) Early days of tRNA research: discovery,

function, purification and sequence analysis. J Biosci, 31, 439-451.

Randerath, E., Gupta, R.C., Chia, L.L., Chang, S.H. and Randerath, K. (1979) Yeast

tRNA Leu UAG. Purification, properties and determination of the

nucleotide sequence by radioactive derivative methods. Eur J Biochem, 93,

79-94.

RNAmods@lib.med.utah.edu. The RNA modification database.

Salminen, A. and Novick, P.J. (1987) A ras-like protein is required for a post-Golgi

event in yeast secretion. Cell, 49, 527-538.

Schaffrath, R. and Meinhardt, F. (2005) Kluveromyces lactis zymocin and other

50

plasmid-encoded yeast killer toxins. In Schmitt, M.J. and Schaffrath, R.

(eds.), Microbial protein toxins. Springer-Verlag, Berlin, Vol. 11, pp.

133-155.

Schramm, L. and Hernandez, N. (2002) Recruitment of RNA polymerase III to its

target promoters. Genes Dev, 16, 2593-2620.

Sekiya, T., Takeishi, K. and Ukita, T. (1969) Specificity of yeast glutamic acid

transfer RNA for codon recognition. Biochim Biophys Acta, 182, 411-426.

Smith, C.J., Teh, H.S., Ley, A.N. and D'Obrenan, P. (1973) The nucleotide

sequences and coding properties of the major and minor lysine transfer

ribonucleic acids from the haploid yeast Saccharomyces cerevisiae S288C.

J Biol Chem, 248, 4475-4485.

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.

Sterner, D.E. and Berger, S.L. (2000) Acetylation of histones and

transcription-related factors. Microbiol Mol Biol Rev, 64, 435-459.

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.

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,

6383-6391.

Thompson, R.C. (1988) EFTu provides an internal kinetic standard for translational

accuracy. Trends Biochem Sci, 13, 91-93.

51

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.

Unger, M.W. and Hartwell, L.H. (1976) Control of cell division in Saccharomyces

cerevisiae by methionyl-tRNA. Proc Natl Acad Sci U S A, 73, 1664-1668.

Walch-Solimena, C., Collins, R.N. and Novick, P.J. (1997) Sec2p mediates

nucleotide exchange on Sec4p and is involved in polarized delivery of

post-Golgi vesicles. J Cell Biol, 137, 1495-1509.

Welch, E.M., Wang, W. and Peltz, S.W. (2000) Translation termination: it’s not the

end of the story. In Translational Control of Gene Expression. Cold spring

harbor laboarotory press, New York, Vol. 2, pp. pp. 467-485.

White, J.H., Butler, A.R. and Stark, M.J.R. (1989) Kluyveromyces lactis toxin does

not inhibit yeast adenylyl cyclase. Nature, 341, 666-668.

Winkler, G.S., Kristjuhan, A., Erdjument-Bromage, H., Tempst, P. and Svejstrup,

J.Q. (2002) Elongator is a histone H3 and H4 acetyltransferase important

for normal histone acetylation levels in vivo. Proc Natl Acad Sci U S A, 99,

3517-3522.

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.

Wrobel, C., Schmidt, E.V. and Polymenis, M. (1999) CDC64 encodes cytoplasmic

alanyl-tRNA synthetase, Ala1p, of Saccharomyces cerevisiae. J Bacteriol,

181, 7618-7620.

52

Yarian, C., Townsend, H., Czestkowski, W., Sochacka, E., Malkiewicz, A.J.,

Guenther, R., Miskiewicz, A. and Agris, P.F. (2002) Accurate translation of

the genetic code depends on tRNA modified nucleosides. J Biol Chem, 277,

16391-16395.

Yokoyama, S. and Nishimura, S. (1995) Modified nucleosides and codon

recognition. In tRNA: Structure, Biosynthesis, and Function. ASM Press,

Washington, DC, 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.

53