Investigating the Basis of Trna Editing and Modification Enzyme Coactivation In

Home , Base pair, Trypanosoma brucei, Wybutosine

Investigating the basis of tRNA editing and modification enzyme coactivation in

Trypanosoma brucei.

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Katherine Mary McKenney

Ohio State University Biochemistry Program

The Ohio State University

2018

Dissertation Committee:

Dr. Juan Alfonzo, Adviser

Dr. Anita Hopper

Dr. Jane Jackman

Dr. Paul Herman

Copyrighted by

Katherine Mary McKenney

2018

Abstract

In all domains of life, transfer RNAs (tRNAs) undergo many post-transcriptional

modifications and editing events that fine-tune their structural and functional properties.

These processing events can be indispensable for translation and consequently survival.

In T. brucei, tRNAThr undergoes hydrolytic deamination from adenosine (A) to inosine (I) at position 34, which expands the decoding capacity of tRNAThr by permitting wobble

base pairing with the third position of the codon. tRNAThr in T. brucei also undergoes

cytosine (C) to uridine (U) editing at position 32 of the anticodon loop; both deamination

events are uniquely catalyzed by the same enzyme, TbADAT2/3. In addition to editing at

position 32, tRNAThr is further methylated at the same position, yielding 3-methylcytidine

(m3C) or 3-methyluridine (m3U). Interestingly, in vitro assays have shown that the

3 recombinant m C methyltransferase (TbTrm140) methylates C32 of a synthetic tRNA

substrate only when TbADAT2/3 is present in the reaction. Furthermore, m3C is further

deaminated to form m3U, while also requiring the presence of both enzymes. These

unprecedented results demonstrate that the methyltransferase and deaminase strictly

depend on each other for activity. Additionally, we found that a direct protein-protein

interaction between TbADAT2/3 and TbTrm140 prevents TbADAT2/3, a highly

mutagenic enzyme, from performing rampant mutagenesis of the genome. TbADAT2/3

mutagenicity is analogous to that observed with human activation-induced deaminase,

ii human AID; the interaction between TbTRM140 and TbADAT2/3 provides a foundation for how AID may be regulated.

In the second chapter of this dissertation we show that the TbADAT2/3 and

TbTrm140 act synergistically whereby the deaminase enhances the affinity of the methyltransferase for its substrate and vice versa. We also observe that active site residues in each enzyme have negligible effects on binding synergy. While these enzymes interact directly, these assays suggest that the observed binding synergy results from tRNA binding via domains distal to their active sites. These observations were validated by determination of binding affinity from single-turnover kinetic assays.

Overall, this data provides a rationale for the interaction and functional codependence of

these proteins.

The third focus of this dissertation is the substrate specificity of TbADAT2/3 and

TbTrm140. A series of minimal tRNAThr substrates were synthesized to analyze which

fragments of the tRNA were bound by TbADAT2/3 and TbTrm140. We found that

TbTrm140 binds several small tRNA substrates, particularly those containing the TψC- loop arm, suggesting the importance of the latter for binding. On its own, TbADAT2/3 binds most stably to a substrate composed of the T-arm/anticodon arm/D-arm, but lacking the acceptor stem, with minimal to no affinity for smaller substrates. This likely reflects a requirement of a tertiary folded tRNA for binding. We utilized primer extension analysis from T. brucei Trm5 RNAi cell lines to assess whether the conserved modification m1G at position 37 of tRNAArg affects formation of m3C at position 32. We

6 Thr used similar experiments to analyze the conserved modification, t A37 in tRNA and

iii

tRNASer by RNAi downregulation of the T. brucei Kae1 enzyme, a putative t6A enzyme

subunit. In contrast to what was observed in S. cerevisiae and S. pombe, it appears that

other anticodon loop modifications have minimal influence on methylation at position 32

in T. brucei. Altogether, these findings increase our understanding of the substrate

selectivity of TbTrm140 and TbADAT2/3 and provide insight into the distinct modes of

substrate recognition exhibited by m3C methyltransferase from different organisms.

The final chapter deals with the determination of the subunit orientation of the

TbADAT2/3 heterodimer in collaboration with several other laboratories. We obtained an unpublished crystal structure of the ADAT2 homodimer which was used as a template for two new structural models of the ADAT2/3 heterodimer. We further analyzed ADAT2/3 using hydrogen-deuterium exchange mass spectrometry to map regions of the protein that were solvent-exposed versus buried within the dimer. These results further our knowledge of the tRNA-binding and catalytic properties of TbADAT2/3 and presents the groundwork to further dissect the unique interaction and enzyme coactivation of

TbADAT2/3 and TbTrm140.

Dedication

I would like to dedicate this document to my family for their constant support and encouragement. I am especially grateful for my devoted husband, Ryan McKenney, for everything he does to support me every day.

Acknowledgments

I am exceedingly grateful for the incredible amount of support I received throughout my graduate education. Of course, none of this would be possible without the excellent training and mentorship of my advisor, Dr. Juan Alfonzo, whose passion for science provides a constant source of inspiration for his students. I am deeply appreciative of the extraordinary amount of energy, time, and effort he put into his work to ensure my success throughout my graduate career and beyond. His motivation and guidance has allowed me to grow as a scientist and individual. To my colleague and friend, Dr. Mary Anne Rubio, I will miss our daily conversations from across the lab bench and excursions to the post office. I cannot thank her enough for all of her help throughout my time in the lab, without which I would not have accomplished as much as

I did. I greatly appreciate her unending supply of patience and advice in all matters. I am also very thankful to my thoughtful committee who offered many valuable suggestions to guide my research forward throughout this process.

My graduate career would not have been possible without the influence of my undergraduate advisor Dr. David Lewis. I would like to thank him for encouraging both

Ryan and I to further our education. We both share fond memories of being a part of the

Lewis lab and are forever grateful for his continuous support.

I would like to thank the previous lab members who helped me when I first started in the lab; Dr. Zdenek Paris, for sharing techniques in northern and western vi

blotting and Dr. Paul Sample for his experience in cell culture work. I would like to thank

Dr. Ian Fleming, for the help with protein purifications, polysome analysis, anything

methylase-related, and for always being quick with a joke. I would like to thank Dr.

Raphael Soares, for his assistance with protein purifications, general upbeat attitude, and

constant singing. I am thankful for my good friend, Dr. Alan Kessler, for putting up with

me over the past 6 years. I will always admire his persistent curiosity and exceptional

ability to think outside the box. To my current labmate, Gabriel Silveira d’Almeida, I am

appreciative for all the useful discussions and for being a friendly presence in the lab. I also had the pleasure of working with several outstanding undergraduate researchers:

Scott Hinger, David Salas, Caitlin Moore, Ajith Jaiganesh and Amelia Staats. Thank you all for developing a sound foundation in the lab and being great company to work with.

To my loving parents, Mark and Karen Anderson, thank you for instilling in me a strong drive to learn and investing in my education at a young age. For my entire family, thank you for your patience, enthusiasm, and encouragement throughout this process and my life in general. A special thank you to my dear husband Ryan for always being so considerate and keeping me afloat by working unbelievably hard in all aspects of life.

vii

Vita

May 2008 ...... Stillwater Area High School

May 2012 ...... B.S. Biochemistry/Molecular Biology,

University of Wisconsin-Eau Claire

2012 to present ...... Graduate Research Associate, Department

of Microbiology, The Ohio State University

Publications

McKenney K.M., Rubio, M.A., and Alfonzo, J.D. Binding synergy as an essential step for tRNA editing and modification enzyme co-dependence in Trypanosoma brucei. RNA. 2017. 24(1):56-66.

McKenney K.M., Rubio, M.A., and Alfonzo, J.D. Chapter 2: The evolution of substrate specificity by tRNA modification enzymes. 2017. The Enzymes, 1st Ed., Academic Press, Cambridge MA.

Rubio, M.A., Gaston, K.W., McKenney, K.M., Fleming, I.M., Paris, Z., Limbach P.A., Alfonzo J. D. Editing and methylation at a single site by functionally interdependent activities. 2017. Nature. 542(7642):494-497.

McKenney K.M. and Alfonzo J.D. From prebiotics to probiotics: The evolution and functions of tRNA modifications. Life (Basel). 2016. 6.

Fields of Study

Major Field: Ohio State University Biochemistry Program

viii

Table of Contents

Abstract ...... ii

Acknowledgments...... vi

Vita ...... viii

Publications ...... viii

Fields of Study ...... viii

Table of Contents ...... ix

List of Figures ...... xiii

Chapter 1 : Introduction to tRNA modification and editing in Trypanosoma brucei.... 1

1.1 Global health impact and unusual biochemistry of T. brucei ...... 1

1.1.1 RNA editing in T. brucei mitochondria ...... 4

1.2 tRNA end processing ...... 5

1.3 tRNA splicing ...... 8

1.4 tRNA structure and enzyme recognition...... 9

1.5 Deaminases that target tRNA...... 12

1.5.1 Modularity of tRNA modification enzymes...... 15

1.6 Diversity of RNA methylations, methyltransferases, and their modes of tRNA binding

...... 23

1.6.1 Sequence versus structural determinants for specificity of methyltransferases ... 31

1.6.2 Pre-existing modifications and their requirement for specificity of

methyltransferases ...... 33

1.6.4 Multisubunit methyltransferases ...... 38

Chapter 2 : Binding synergy as an essential step for tRNA editing and modification

enzyme co-dependence in Trypanosoma brucei ...... 42

2.1 Introduction ...... 42

2.2 Results ...... 46

2.2.1 TbADAT2/3 and TbTrm140 bind synergistically to tRNAThr ...... 46

2.2.2 Synergistic binding is independent of enzyme activity ...... 53

2.2.3 Validation of binding constants by single turnover kinetics ...... 60

2.3 Discussion ...... 68

Chapter 3 : Substrate specificity determinants of the Trypanosoma brucei co-

dependent tRNA editing deaminase and m3C methyltransferase ...... 73

3.1 Introduction ...... 73

3.2 Results ...... 76

3.2.1 TbADAT2/3 and TbTrm140 contact distinct sites on the tRNAThr ...... 76

3.2.2 Binding synergism is not limited to substrate tRNA...... 83 x

3.2.3 Methylation at position 32 is not influenced by t6A nor m1G at position 37 in vivo

...... 89

3.3 Discussion ...... 95

Chapter 4 : Analysis of Trypanosoma brucei editing deaminase subunit orientation and consequences for catalysis and tRNA-binding ...... 98

4.1 Introduction ...... 98

4.2 Results ...... 100

4.2.1 Revised structural models of TbADAT2/3 ...... 100

4.2.2 Mapping of TbADAT2/3 buried residues by hydrogen-deuterium exchange ... 103

4.2.3 Deletion of possible flexible regions of ADAT3 for x-ray crystallization ...... 106

4.3 Discussion ...... 110

Chapter 5 : Concluding Remarks and Future Directions ...... 112

5.1 Introduction ...... 112

5.2 Mechanism of enzyme coactivation by TbTrm140 and TbADAT2/3 ...... 113

5.3 Further characterization of substrate specificity of TbTrm140 and TbADAT2/3 .... 114

5.4 Specific role of m3C versus m3U in translation ...... 115

5.5 The effect of compartmentalization on anticodon loop editing ...... 117

5.6 Model of modification interdependence and quality control ...... 118

References ...... 120

Appendix A: Supplementary figures ...... 140

Appendix B: Materials and methods...... 145

B1: Plasmid DNA purification ...... 146

B2: Transformation of E. coli ...... 147

B3: Recombinant protein purification ...... 148

B4: Protein concentration ...... 150

B5: Western blot...... 151

B6: Affinity purification of polyclonal antibody ...... 153

B7: Transfection of T. brucei ...... 155

B8: Co-immunoprecipitation ...... 157

B9: Electrophoretic mobility shift assay ...... 158

B10: A to I deamination and methylation assays ...... 160

B11: Kinetic determination of binding affinity ...... 162

B12: Primer extension analysis of methylation...... 163

xii

List of Figures

Figure 1.1 Trypanosoma brucei life cycle...... 3

Figure 1.2 Post-transcriptional processing of tRNA in eukaryotes...... 6

Figure 1.3 tRNA secondary and tertiary structure...... 11

Figure 1.4 Mechanism of cytosine to uridine and adenosine to inosine editing by hydrolytic deamination...... 14

Figure 1.5 Modularity of deaminase and wybutosine enzymes...... 16

Figure 1.6 A schematic view of the modified nucleotide positions within tRNA...... 19

Figure 1.7 Topology diagram of Class I and Class IV methyltransferases...... 26

Figure 1.8 Topological diagrams of common RNA binding domains...... 28

Figure 1.9 Processing pathway of tRNAPhe in Saccharomyces cerevisiae...... 37

Figure 2.1 TbADAT2/3 and TbTrm140 stably bind tRNA in vitro...... 48

Figure 2.2 TbADAT2/3 and TbTrm140 bind tRNA synergistically...... 50

Figure 2.3 Synergistic binding is exclusively linked to the presence of TbADAT2/3 and

TbTrm140 in the reaction...... 52

Figure 2.4 TbADAT2/3 catalytic residues do not contribute to increased affinity of

TbTrm140 for tRNA...... 54

Figure 2.5 TbTrm140 catalytic residues do not contribute to increased affinity of

TbADAT2/3 for tRNA...... 56

Figure 2.6 N-terminal residues of TbTrm140 are important for tRNA binding...... 58 xiii

Figure 2.7 TbTrm140 N-terminal deletion mutant forms homodimers...... 59

Figure 2.8 C-terminal residues of TbTrm140 are not important for tRNA binding...... 60

Figure 2.9 Kinetic Determination of binding affinities...... 62

Figure 2.10 Kinetic determination of the dissociation constant of TbADAT2/3 for

tRNAThr in the presence of TbTrm140...... 63

Figure 2.11 Kinetic determination of dissociation constant of TbTrm140 to tRNAThr in the presence of TbADAT2/3...... 64

Figure 2.12 Kinetic determination of dissociation constant of TbTrm140 to tRNAThr in the presence of TbADAT2/3 catalytic mutant, E92A...... 66

Figure 2.13 Activity assays with mutant enzymes...... 67

Thr Figure 3.1 RNA minisubstrates derived from T. brucei tRNA CGU...... 77

Figure 3.2 TbADAT2/3 and TbTrm140 do not bind synergistically to the D-

loop/anticodon minisubstrate...... 80

Figure 3.3 TbADAT2/3 and TbTrm140 do not bind synergistically to the TψC-

loop/anticodon minisubstrate...... 81

Figure 3.4 TbADAT2/3 and TbTrm140 do not bind synergistically to the D-

loop/anticodon/TψC-loop minisubstrate...... 82

3 3 Figure 3.5 Cloverleaf secondary structures of m C/m U32 substrates versus non-

substrates...... 85

Val Figure 3.6 TbTrm140 binds nonsubstrate tRNA AAC synergistically...... 86

Glu Figure 3.7 TbADAT2/3 and Trm140 bind stably to nonsubstrate tRNA UUC...... 87

xiv

Glu Figure 3.8 TbADAT2/3 and TbTrm140 bind tRNA nonsubstrate tRNA UUC

synergistically...... 88

Figure 3.9 Putative TbKae1 is important for growth T. brucei...... 91

Figure 3.10 Knockdown of putative TbKae1 does not alter methylation levels at position

32 of tRNAThr...... 92

1 Figure 3.11 m G37 prevents detection of methylation at position 32 by primer extension.

...... 94

Figure 4.1 Revised structural models based on TbADAT2 crystal structure...... 102

Figure 4.2 Exposed and buried residues of TbADAT2/3 identified by hydrogen-

deuterium exchange...... 104

Figure 4.3 Solvent accessibility and HDX values for unswapped and swapped models.

...... 105

Figure 4.4 HDX values mapped onto revised homology models...... 106

Figure 4.5 Western blot detection of TbADAT3 N-terminal deletion variants...... 108

Figure 4.6 tRNA binding and activity of TbADAT3 N-terminal deletion variants...... 109

Figure 5.1 Model of modification interdependence and quality control...... 119

Figure A.1 Complex formation of TbADAT2/3 and TbTrm140…………………...... 141

Figure A.2 Predicted RNA-binding residues of TbTrm140……………………...…….142

Chapter 1 : Introduction to tRNA modification and editing in Trypanosoma

brucei1

1.1 Global health impact and unusual biochemistry of T. brucei

Trypanosoma brucei (T. brucei) is a unicellular eukaryotic parasite belonging to an early-branching group of flagellated protozoans called kinetoplastids. Many kinetoplastids are pathogenic to humans and animals; T. brucei, for instance, is the causative agent for Human African Trypanosomiasis (HAT) also known as African

Sleeping Sickness. HAT is a fatal illness affecting thousands of people in impoverished nations in Sub-Saharan Africa [1,2]. There are currently no good prophylactic drugs or vaccines available and current treatments for the disease are costly, often ineffective, and can cause severe, even deadly, side-effects [2,3]. Certain species of trypanosomes also cause the disease, nagana, which affects livestock [1,2]. Altogether, these diseases yield devastating agricultural and economic ramifications in addition to the human health impact.

The T. brucei parasite is transmitted to a mammalian host via an insect vector, the tsetse fly (genus Glossina). The life cycle of the parasite starts when a tsetse fly takes a bloodmeal from an infected mammal upon which the parasite undergoes drastic

1 The majority of the information presented in this chapter is published as: McKenney K.M. and Alfonzo J.D. From prebiotics to probiotics: The evolution and functions of tRNA modifications. Life (Basel). 2016. 6. and McKenney K.M., Rubio, M.A., and Alfonzo, J.D. Chapter 2: The evolution of substrate specificity by tRNA modification enzymes. 2017. The Enzymes, 1st Ed., Academic Press, Cambridge MA. 1 morphological and metabolic alterations to accommodate the extreme changes of host environments [1,4–7] (Figure 1.1). Three distinct developmental stages occur in the life cycle. The first change takes place once the parasite migrates into the insect mid-gut where it loses its variant-specific surface glycoproteins (VSG), replaces it with a surface protein called procyclin, becomes elongated, and multiplies [1,4]. The parasites then travel from the mid-gut to the salivary gland where they replicate and transform into epimastigotes [1,4]. After proliferation, the epimastigotes recover their VSG coats upon changing into infective metacyclic cells [1,4]. When the insect takes another bloodmeal the metacyclic parasites are injected into a mammalian bloodstream [1,4]. In the mammalian bloodstream, trypanosomes are coated with VSGs which are antigenic determinants for host antibodies [8]. As such, the infectivity of the parasite correlates with the generation of the VSG surface coat [9]. A single trypanosome can give rise to progeny with over a hundred different VSG coats [8]. To add further complexity, the molecular identity of the VSG coat on the parasite changes regularly in an event called

VSG switching. This is carried out through a mechanism called antigenic variation which occurs by transcriptional switching between distinct VSG expression sites, yielding a nearly infinite number of possible unique VSGs [1–4]. This unusual feature of this parasite allows it to evade the host immune system, which cannot keep up with the antigenic variation between generations of parasites, leading to chronic infections.

Figure 1.1 Trypanosoma brucei life cycle.

Trypanosomes multiply in their slender form after injection into the mammalian bloodstream. This cell type produces a bloodstream-specific VSG coat, which upon switching effectively allows them to evade the host immune response. The kinetoplast (mitochondrial genome) is found at the posterior end and the mitochondrion is somewhat repressed. After cell proliferation in the bloodstream, the parasite undergoes a density-dependent morphological change to the stumpy cell form. This form is non-proliferative and prepared for transmission to the tsetse fly. When the fly takes a bloodmeal, the parasite changes to the procyclic form and proliferates in the mid-gut. The procyclic form loses its VSG coat and the kinetoplast returns to the sub-terminal position. Once established in the fly mid-gut, the parasites travel and attach to the salivary gland, transitioning to the proliferative epimastigote form. These cells later convert to the non-proliferative metacyclic form which develop a VSG coat for transmission into another mammalian host.

1.1.1 RNA editing in T. brucei mitochondria

Kinetoplastids are atypical eukaryotes in that they contain a single, complex mitochondrion, originally referred to as a kinetoplast as it is attached to the basal body of the flagellum. In comparison, a classical mammalian cell contains 100 to 10,000 (200 on average) mitochondria with circular genomes of 16 to 17 kilobase pairs [5]. The mitochondrial genome of trypanosomes, often called kDNA, is made up of unusual circular DNA called maxicircles and minicircles. These form a giant network of catenated circles that are arguably the most structurally complex genome found in nature

[5–8]. Maxicircles contain genes characteristic of eukaryotic mitochondria that are necessary for oxidative phosphorylation and encode rRNAs, mRNAs, and guide (gRNAs)

[6,8–10]. Minicircles solely encode gRNAs that direct editing of maxicircle transcripts

[6,11]. RNA editing in the T. brucei mitochondrion involves massive insertions and deletions of uridines, creating initiation codons, stop codons, and entire reading frames, a process often called “pan editing.” RNA editing is defined here as a programmed change of one canonical nucleotide to another or a change in the nucleotide that alters the meaning of the edited RNA beyond what is found in the genome [12]. Insertion editing

(incorporation of new nucleotides) and deletion editing (excision of existing nucleotides) occurs in several mitochondrial transcripts in T. brucei as well as the slime mold

Physarum polycephalum [13]. In T. brucei, the insertion/deletion editing in the mitochondria includes extensive addition or, at a lesser extent, removal of uridines [13].

These activities are catalyzed by a series of enzymes that form a set of three distinct 20S editosomes [14,15] This occurs post-transcriptionally, although editing can also be co-

transcriptional. An example of co-transcriptional editing is seen in paramyxoviruses

where polymerase stuttering results in G insertion leading to expression of alternative

reading frames [16]. In P. polycephalum mainly C, and less often A and U, are inserted

into the mitochondrial mRNA transcripts [13]. Substitutional editing, where one

nucleotide replaces another, has also been observed in chloroplast and mitochondrial

mRNA transcripts of plants [13]. This type of editing will be emphasized later in the

context of tRNA rather than mRNA.

1.2 tRNA end processing

In nearly all organisms, tRNAs are transcribed with 5’ and 3’ extensions that must

be removed for the tRNA to function (Figure 1.2). The organism, Nanoarchaeum

equitans is the only organism known where tRNAs are transcribed without 5’ extensions

[17]. There is no specific 5’-flanking sequence conserved for all tRNA genes of an

organism [18]. The 5’ extensions of tRNAs are removed by the ubiquitous enzyme

RNase P. This process is a key step of tRNA maturation and thus, essential for translation

[19]. T. brucei contains two monomeric forms of RNase P, TbPRORP1 and TbPRORP2,

which localize to the nucleus and mitochondrion respectively. Although TbPRORP2 can

cleave tRNAs in vitro, evidence suggests pre-tRNAs are absent from T. brucei

mitochondria [20–23]. It was first suggested that nuclear pre-tRNAs are imported into the

mitochondrion [24–27]. However, it was later shown that these precursors were permutations of circularized mature tRNAs and that tRNA import is independent of genomic context, demonstrating that only mature tRNAs are substrates for mitochondrial

import [20–23]. This implies TbPRORP2 may function outside the established role in

tRNA maturation.

Figure 1.2 Post-transcriptional processing of tRNA in eukaryotes.

tRNAs are transcribed in the nucleus with 5’ and 3’-extensions which are removed by RNAse P and various endonucleases/exonucleases, respectively. All mature tRNAs contain the trinucleotide CCA at their 3’-end that are added in the nucleus. The tRNA is then exported to the cytosol where it is maintained for translation or is imported into the mitochondrion for translation. Editing and modifications are introduced into the tRNA at any point in the processing pathway and can occur in multiple compartments. Many eukaryotic tRNA precursors also contain an intron located in the anticodon loop, not shown here. These precursors must be spliced to form the mature tRNA. In T. brucei, tRNATyr is the only intron-containing tRNA and splicing occurs in the cytoplasm. Overall, tRNA processing must occur to produce a mature fully functional, properly folded and recognizable tRNA.

The transcription of precursor tRNAs by RNA pol III terminates with a string of

uridines with variable lengths at the 3’ end [28]. Unlike 5’ processing, which is catalyzed

by a single conserved enzyme, these 3’ extensions are processed by a variety

endonucleases and exonucleases including tRNase Z and Rex1p, among others [28]. 3’ processing can occur by alternative pathways depending on the extent of involvement of the poly-U binding protein, La [28]. La-dependent and La-independent pathways can even exist for processing of different subsets of tRNAs in a single cell type [28]. In eukaryotes, as well as many archaea and bacteria, a terminal CCA is then added to the 3’ end of the acceptor stem by tRNA nucleotidyltransferase (Figure 1.2) [18].

Contrastingly, in some bacteria, including E. coli, the tRNA genes are encoded with terminal CCA at the 3’end [18]. Be it encoded in the genome or post-transcriptionally added, the universal CCA sequence is critical for aminoacylation. Other features of the tRNA called “identity elements” are also necessary to ensure the tRNA receives the correct amino acid.

In addition to CCA, further editing at the 5’ or 3’ ends may be needed to produce an intact, functional tRNA. In most organisms tRNAHis, for instance, contains an extra G

at their 5’ end (G-1) which is critical for aminoacylation both in vitro and in vivo [17]. In

some bacteria and archaea, RNase P is able to correctly cleave just after the G-1 position

[17]. However, in other organisms, RNase P removes the G-1, which in turn must be

subsequently re-inserted. This activity is provided by the enzyme Thg1 (tRNAHis

guanylytransferase) which performs 3’-5’ nucleotransferase reaction in a non-templated

7 fashion [17]. Along these lines, some organisms contain mismatches within the first three base pairs of the acceptor stem in a subset of mitochondrial tRNAs. In some eukaryotic organisms, including Distyostelium discoideum and Acanthamoeba castellani, these mismatches are repaired by Thg-1 like enzymes called TLPs (Thg1-like proteins). Unlike

Thg-1, TLPs perform 3’-5’ polymerization in a templated manner in vitro [17]. Similar mismatches are found at the 3’ end of the tRNA which can in theory be repaired by a traditional 5’-3’ polymerase activity [17,29].

1.3 tRNA splicing

Post-transcriptional processing of pre-tRNAs can also occur internally.

Approximately 10 to 20% of eukaryotic tRNA genes contain introns that range from 8 to

60 nucleotides in length [18]. Saccharomyces cerevisiae have about 10 intron-containing tRNA families [18,30]. The intron in these precursor tRNAs have a complementary sequence to the anticodon and a region of the anticodon loop (with exceptions found in

Xenopus laevis, Dictyostelium discoideum, Schizosaccharomyces pombe) [18]. Introns interrupt the tRNA sequence one nucleotide 3’ of the anticodon nucleotides. Although the surrounding sequence differs, the location of the intron within the pre-tRNA in Eukarya is conserved among all sequenced tRNAs to date [18].

In T. brucei, tRNATyr is the only tRNA species with an intron. The 11-nucleotide intron is one of the smallest intervening sequences reported. tRNA splicing in eukaryotes involves the distinct combination of a specific endonuclease, ligase, and 2’ phosphate specific NAD-dependent phosphotransferase [27]. An evolutionarily conserved eukaryotic endonuclease recognizes and cleaves the pre-tRNA intron in the first step of

splicing. The resulting 5’ and 3’ halves are processed and joined by the tRNA ligase

leaving an unusual 2’ phosphate intermediate that is removed by the 2’

phosphotransferase [31]. The S. cerevisiae splicing endonuclease (SEN) complex is made up of four essential subunits, SEN2, 15, 34, and 54. While SEN2 and SEN34 are conserved from Archaea to vertebrates, SEN15 and SEN54 are missing from Archaea and are not highly conserved between budding yeast and vertebrates [32].

1.4 tRNA structure and enzyme recognition

The importance of RNA binding percolates all aspects of RNA processing and

involves an amazing set of different RNA interaction domains; some are readily

recognizable by their degree of sequence conservation, others are more discreet and

escape easy prediction [33]. In general, RNA-binding domains may be divided into two major groups: 1) those belonging to proteins that act alone (“professional RNA-binding proteins”) 2) Those that are appended to other proteins, acting as binding modules that work in cis or trans to anchor other functionalities at the site of action. It is clear that a critical step for the evolution of extant life must have involved the appearance of polypeptides able to interact in some fashion with RNA, facilitating the transition into an

RNA-protein world from an all-RNA world. Such developments had to occur after the emergence of templated peptide synthesis, but not too much later. Although early RNA- binding proteins may have been act alone entities, the necessity for modularity must have been imminent [33]. Modularity implies that discreet protein domains, whose sole purpose was to bind RNA, had to fuse with catalytic domains [34,35]; these must have

been an indispensable invention for the evolution of RNA processing and certainly for

the appearance of RNA modification specificity.

All tRNAs in cells adopt a very similar 3D-folded structure, the characteristic L-

shape, whereby the tRNA can be divided into two functional domains: The codon

recognition domain, formed by the coaxial stacking of the D-arm onto the anticodon arm;

and the amino acid accepting domain, formed by the coaxial stacking of the TΨC-arm on

top of the acceptor arm (Figure 1.3). These two domains come together as an L-shaped molecule whose structure is stabilized by base pair interactions between the D-loop,

variable loop and the aforementioned TΨC loop. Critical to this folding is a stable “elbow

structure”, which is formed by specific base pairs between several nucleotides; residues

15 of the D-loop and 48 of the variable loop, G18 of the D-loop and Ψ55 of the TΨC loop

and finally an additional Watson-Crick base pair between G19 of the D-loop and the

highly conserved C56 (of TΨC). It has become apparent that this conserved structure (the

elbow) has been used repeatedly for recognition by various tRNA-targeting enzymes

such as a number of aminoacyl tRNA synthetases, Ribonuclease P, and some tRNA

transport factors [36]. In terms of modification this conserved structure sets a critical

baseline for how modification enzymes approach the recognition problem. A sure

inference is the fact that if Ψ55 is important for proper tRNA folding, the enzyme that

catalyze its formation cannot be dependent on the L-shape structure for recognition.

Predictably such enzymes learn to target tRNAs before the invention of the L-shape and

inherently must have acted on proto-tRNAs before the elbow structure was adopted.

Figure 1.3 tRNA secondary and tertiary structure. Characteristic secondary cloverleaf structure of tRNA with relevant stem loops indicated and color-coded (left). L-shaped tertiary structure following the same color-coding with the amino acid accepting arm, codon recognition arm, and the elbow indicated (right). The specific base pairs which are needed to form a stable L-shaped structure are numbered and shown as bonds in colors according to their position in the tRNA (base pairs include: nucleotides 15 and 48, 18 and 55, and 19 and 56).

Once the tRNA is folded, then other modifications can be added and the

conserved tertiary features of a tRNA can be exploited for recognition. For example,

many of the modifications that target the anticodon loop demand such degree of

specificity that mistakes, such as modification of neighboring nucleotides that are not the

intended target, could prove detrimental to a cell’s wellbeing. Some modification

enzymes that target the anticodon loop require a fully folded tRNA for activity and it is from the information found in such structured molecule that exquisite specificity is achieved. This is especially true of anticodon deaminases, where deamination of the wrong nucleotide can potentially reassign the tRNA to a different codon leading to a miscoding problem. It has been suggested then, that tRNA modification enzymes can be divided into two major groups, based on how they recognize the substrate they target: 1) architecture-dependent and 2) architecture-independent enzymes; the former require a

fully folded L-shape tRNA for activity, the latter act earlier and are in general a

prerequisite for folding [37]. In turn, many of the architecture-independent enzymes act at

sites away from the anticodon arm and play critical roles as structural modulators, while architecture-dependent ones tend to target the anticodon loop, the modifications that they synthesize affecting only local structure and anticodon function.

1.5 Deaminases that target tRNA

Historically, the modification field has been divided into two groups: the editors and the modifiers. The term modification implies any biological change to a polynucleotide chain that alters its chemical composition beyond that of the canonical nucleotides (G, A, U (T) and C); such changes are programmed rather than stochastic and

different from chemical mutagenesis [12]. Thus, RNA editing can be considered a well- defined category within the vast umbrella term of post-transcriptional modifications.

Editing strictu senso involves a programmed change of one canonical nucleotide to another (e.g. C to U) or a change of a canonical nucleotide that alters the meaning of the edited RNA beyond what is encoded by the genome (e.g. A to I deamination) [12].

Editing may also involve more extensive insertion and deletion of nucleotides as those occurring in Physarum and trypanosomatid mitochondria, but this will not be further discussed here.

Among the tRNA deaminases, some edit the first position of the anticodon

(position 34), while others target additional positions away from the anticodon, for

example A to I deamination at position 37 in eukaryotes, or position 57 or 58 in archaea

[38–40]. The best studied among these enzymes are the adenosine deaminases acting on

tRNA, or ADAT, where ADAT2/3 deaminates adenosine 34 (A34) to inosine 34 (I34). A

to I deamination is the most common form of RNA editing and occurs by hydrolytic

deamination at the C6 position of adenosine (Figure 1.4). The switch from A to I alters

the base-pairing preference from thymine (T) to cytidine (C), expanding the decoding

capacity of tRNAs allowing a single tRNA to decode multiple codons with the same

amino acid. Other ADATs, for example ADAT1, form inosine at position 37 of the

Ala anticodon loop of a single tRNA (tRNA ) [41]. While I37 does not affect decoding

specificity, it is important for translational accuracy. In addition, C to U changes at other

positions have been described in organellar tRNAs from various organisms [42,43], but

the enzymes involved have not been identified. The C to U editing of tRNAthr in T. brucei

13 is the first described non-organellar case of tRNA C to U deamination. The first observed case of C to U editing in vertebrates was in the mRNA encoding apolipoprotein B

(apoB). The editing of this transcript was a nuclear event that changed a Glu codon

(CCA) into a stop codon (UAA) [16]. Conversion of U to C, or “reverse editing,” has also been observed in some plants, amphibians, and mammals, but the mechanism for this editing is unknown [16]. Mechanistically these are inferred to occur by hydrolytic deamination at the C4 position of cytidine (Figure 1.4).

Figure 1.4 Mechanism of cytosine to uridine and adenosine to inosine editing by hydrolytic deamination.

1.5.1 Modularity of tRNA modification enzymes

In general, most enzymes are modular in nature and are known to acquire

domains that become critical for substrate recognition, binding or catalysis. The tRNA

modifying enzymes are not an exception; some enzymes recognize the global structure of

the tRNA and exhibit a broad range of substrate binding, others have recruited RNA

binding domains [44]. These RNA binding domains are commonly fused with catalytic

domains and confer specificity to a reduced set of tRNA substrates. One such example of

the enlistment of RNA binding domains is seen in tRNA editing enzymes that catalyze

the ubiquitous A-to-I deamination. Adenosine deamination is the most common form of

RNA editing and it is not restricted to tRNA. Two major families of editing deaminases

exist, adenosine deaminases acting on RNAs (ADARs) and polynucleotide cytidine

deaminases (CDAs), which are categorized based on their active site and zinc metal-

1 binding sites. The eukaryotic m I37 deaminase enzyme, ADAT1 or Tad1, resembles

classical ADARs except without a double-stranded RNA binding domain (dsRBD)

(Figure 1.5) [41,45]. This close similarity may give insight into the evolution of ADARs

whereby an ADAT1-like enzyme obtained a dsRBD, enabling the switch to mRNA

editing [45]. In fact, protein sequences of ADAT1 from higher eukaryotes are more similar to ADARs than to S. cerevisiae Tad1, further supporting this evolutionary hypothesis [45].

Figure 1.5 Modularity of deaminase and wybutosine enzymes.

Domain organization of (a) Adenosine deaminase enzymes: ADAT1 (human), Tad1/ADAT1 (S. cerevisiae), Tad2/ADAT2 (S. cerevisiae and T. brucei), and Tad3/ADAT3 (S. cerevisiae and T. brucei) and cytidine deaminase enzyme: APOBEC-1 (human). The deaminase domain (purple) contains an active site comprised of characteristic Zn2+-binding residues (denoted by asterisks) and a critical proton-shuttling glutamate (denoted by circle). Z-DNA (dark blue) and double stranded RNA binding domains (green) are unique to ADAR1. Tad2/ADAT2 contain a KR-rich tRNA binding domain (pink) and APOBEC-1 has a pseudoactive site (grey) at the C-terminal end. (b) Enzymes which catalyze 4-demethylwyosine (imG-14), the second step of wyosine formation in Archaea: TAW1 (Methanocaldococcus jannaschii), and wybutosine formation in Eukarya: TYW1 (S. cerevisiae), TYW1L (T. brucei), TYW1S (T. brucei). The wyosine formation domain typical of this family of enzymes is shown in blue. Iron-sulfur cluster (4Fe-4S) coordinating residues (shown by asterisks) are found within and just before the radical SAM domain (orange). The flavin mononucleotide (FMN) domain (red) is found in TYW1 as well as TYW1L, but not TAW1 or TYW1S. The protein length in amino acids is indicated by the numbers at the N-terminus of the protein.

Anticodon A to I deaminases have been described in Bacteria (ADATa or TadA)

and Eukarya (ADAT2/3 or Tad2/3) [46–50]. The bacterial enzyme is homodimeric and

Arg targets a single tRNA, tRNA ACG, the only A34-containing tRNA in bacteria (Figure

1.6). The crystal structure of ADATa was reported by several laboratories, which showed

the general architecture of the active site and confirmed its similarity with other members

of the deaminase superfamily [51–54]. Biochemical analysis demonstrated that this

enzyme could efficiently catalyze deamination of minimal substrates representing only

the anticodon arm of tRNAArg; negligible difference was observed in the catalytic

efficiency of this enzyme with shorter substrates when compared to a full-length tRNA

[51,52]. This observation led to the proposal that ADATa was architecture-independent.

This may have been the result of its specialization towards a single substrate, which then

led to the evolution of a very restrictive RNA binding domain limited to residues in the

proximity of the active site [53]. This proposal was confirmed by the co-crystal structure

of ADATa bound to an anticodon arm mimic, where the nucleotide to be deaminated

(A34) was replaced by the adenosine analog nebularine [52]. This was necessary to

prevent the enzyme from turning over the substrate and enabled trapping of the substrate- bound enzyme. This co-crystal revealed that upon complex formation, the RNA substrate undergoes significant conformational changes. The anticodon nucleotides became splayed out, making them accessible for recognition. The RNA binding cleft is part of a portion of the dimerization domain and makes sequence-specific contacts with the 5 anticodon loop nucleotides, including a direct minor groove interaction with the side chain of a conserved Lys106, which simultaneously forms hydrogen bonds with C32 and

A38. Additionally, a conserved asparagine (Asn138) hydrogen bonds with A38, while a series of positively charged residues (arginines and lysines) make general contacts with backbone phosphates in the anticodon stem. The overall anticodon loop structure poised for catalysis is finally stabilized by specific contacts with G36 and U33 [52]. The take

home lesson is that indeed ADATa has built in its active site all that is needed for the

specific recognition of a single tRNA, while excluding near cognate substrates.

Figure 1.6 A schematic view of the modified nucleotide positions within tRNA.

The cloverleaf secondary structure of tRNA is shown with respective bacterial modification enzymes for methylation (M, left) and A to I deamination (I, right).

Surprisingly, the enzymes responsible for I34 formation in Bacteria and Eukarya are more closely related to CDA family enzymes rather than ADARs [128,129]. The

CDA family contains the sequences H(C)XE and PCXXC (X represents any amino acid) that form the active site of the enzyme with a catalytic glutamate and zinc coordinating residues (Figure 1.6) [130]. Unfortunately, there is no structural information available for the homologous eukaryotic enzyme, ADAT2/3, but biochemical data supports its heterodimeric structure in solution. The heterodimer is formed by two paralogous subunits encoded by the ADAT2 and ADAT3 genes. In eukaryotes, ADAT2 harbors the 19 critical active site glutamate (HAE) while ADAT3 has substituted for a non-catalytic residue (HPV in T. brucei and S. cerevisiae) and was originally thought to play a purely structural role [131]. However, inductively coupled plasma emission spectrometric (ICP) analysis of the enzyme from T. brucei demonstrated that both subunits contribute to zinc binding, suggesting both ADAT2 and ADAT3 are essential for catalysis [132]. Along with the active site, we found that ADAT2 in T. brucei has acquired a lysine and arginine-rich tRNA binding domain, named the KR-domain, at its C-terminal end [130].

Presumably, the density of positive charges interacts favorably with the phosphate backbone of the tRNA. A highly charged domain corresponding to that seen in ADAT2 has also been described in other deaminase enzymes, including ADAR3 [130]. The discovery of the KR-domain supports a previous model whereby the C-terminus of

ADAT2 in eukaryotes evolved a conserved tRNA binding domain expanding its substrate repertoire to include eight different tRNAs [133]. Deletion of the KR-domain leads to lack of deamination activity presumably due to the fact that tRNA binding is abrogated.

However, the active site still contributes to binding and it is the combination of active site residues and of the distal KR-domain that partly explains substrate specificity. Currently, the contribution of the second subunit (TbADAT3) to binding has not been explored.

Regardless, it is clear that unlike bacterial ADATa, eukaryotic ADATs may possess a bipartite binding domain where active-site residues and more distal domains contribute to substrate recognition.

The described dichotomy between a homodimeric deaminase with a single substrate and a heterodimeric enzyme with multiple substrates sets an interesting scenario

that has been suggested for how these enzymes evolved from single-substrate to multi-

substrate specificity. Given their sequence conservation it has been proposed that ADAT3 was the result of gene duplication from ADAT2. This event provided a second copy of an otherwise essential gene, which can now freely mutate by genetic drift. This must have served as the pre-existing condition that allowed the further rearrangement of RNA

binding domains away from the active site, which presumably also led to active-site

relaxation. Once active-site binding becomes less critical, by virtue that binding strength

is provided by a distal domain, the enzyme was probably abled to accommodate a wider

variety of substrates leading to the appearance of the extant eukaryotic tRNA deaminases.

In this realm, gene duplication a purely neutral evolutionary event was a critical step in

the advent of multi-substrate deaminases.

In addition to substrate binding, the presence of critical modular catalytic domains

is also common, such as SAM-binding domains of methyltransferases and FMN-binding

domains of enzymes that form wybutosine and hydroxywybutosine. Wyosine and its

derivatives, including wybutosine (yW) and hydroxywybutosine (OHyW), are critical for

translational fidelity in archaea and eukaryotes, but have not yet been described in

bacteria. These highly complex modifications are found exclusively at position 37 of

tRNAPhe where the bulky hydrophobic side chain stabilizes the anticodon loop structure,

as well as anticodon-codon pairing, preventing potential -1 frameshifting [55,56]. The

biosynthesis pathways for these modifications differ greatly between the two domains.

1 While all wyosine derivatives require m G37 as a precursor, eukaryotes employ a series of

enzymes designated TYW1 through TYW5 whereas the archaeal pathway is comprised

of enzymes TAW1 through TAW3. However, in both Eukarya and Archaea, various

combinations of these enzymes lead to the formation of different derivatives of wyosine depending on the organism. For instance, most eukaryotes carry out sequential reactions by TYW1 through TYW4 with wybutosine as the end product [56–60]. Meanwhile, in other eukaryotic organisms, such as humans, wybutosine can be further modified to hydroxywybutosine by TYW5. The situation in archaea becomes more complex as the enzymes TAW1, TAW2, TAW3 carry out their chemistries either sequentially or in different combinations to generate wyosine derivative ranging from 4-demethylwyosine

(imG-14) to 7-aminocarboxypropylwyosine (yW-72). For instance, Crenoarchaeota lacking TAW2 combine TAW1 and TAW3 to yield wyosine as the final product. The key reaction to form the unique tricyclic core characteristic of wyosine nucleosides is catalyzed by the eukaryotic TYW1 and its archaeal counterpart TAW1. The resulting 4- demethylwyosine is an intermediate in the eukaryotic multienzymatic biosynthesis pathway and can act as a final product in archaea [61]. Once guanosine has been converted to m1G by Trm5, TYW1/TAW1 adds two carbons derived from pyruvate to

create an imidazole ring [61,62]. An N-terminal flavodoxin-like domain and C-terminal

radical SAM catalytic domain, comprised of a characteristic CxxxCxxC iron-sulfur (4Fe-

4S) binding motif, enables the chemistry performed by TYW1 (Figure 1.5) [63,64].

Although it has not been investigated, the FMN prosthetic group undoubtedly participates

in reduction of Fe-S clusters, a function which is required in radical-mediated SAM

reactions [64]. Unlike TYW1, TAW1 has no FMN-binding domain and must obtain

reducing power from an outside source (Figure 1.5) [65].

A phylogenetic analysis of archaeal TAW1 and eukaryotic TYW1 gene sequences

revealed that a gene duplication event likely occurred whereby the eukaryotic lineage

acquired an FMN-binding domain [66]. The possibility of an ancestral bacterial enzyme

is discounted due to the fact that bacteria do not have wyosine. Likewise, horizontal

transfer of the eukaryotic enzyme from Archaea is unlikely as they did not cluster

together in such an analysis [66]. Remarkably, T. brucei has two paralogous TYW1

enzymes, one localized to the cytoplasm and the other to the mitochondrion. It is the only

organism described thus far with wyosine in the mitochondrion. The cytosolic enzyme,

TYW1L, has the FMN binding domain, two 4Fe-4S cluster domains, a radical SAM

domain, and a “wyosine base formation motif” (Figure 1.5). The mitochondrial enzyme,

TYW1S, is reminiscent of the archaeal TAW1 and contains all the previously mentioned

domains except for the FMN binding domain [66]. All in all, it appears as though the

cytosolic TYW1L in T. brucei, akin to other eukaryotes, has more recently acquired the

FMN-binding module. The particular example of the wyosine system in Eukarya and

Archaea yet provides a unique testament to the general modularity of tRNA modification enzymes and how such modularity may have evolved.

1.6 Diversity of RNA methylations, methyltransferases, and their modes of tRNA binding

Methylation is one of the most abundant modifications found ubiquitously in a variety of RNAs including tRNA, rRNA, mRNA, tmRNA, snRNA, snoRNA, miRNA, and viral RNA [67–69]. While methylated nucleotides are involved in affecting functional aspects of certain RNAs, they are particularly important in maintaining

structural elements as methyl groups can stabilize or modulate local RNA structure. In

either case, introducing methylations precisely in the proper location and substrate can be

essential. Consequently, RNA methyltransferases acting on multiple substrates have

evolved different strategies to ensure the correct substrate species and individual target

nucleotide(s) are selected from a pool of potential choices. This portion of the chapter

will focus primarily on tRNA methyltransferases, but will also touch on the methylation

of other RNAs.

The majority of methyltransferases use S-adenosyl-L-methionine (SAM or

AdoMet) as a methyl donor. Methyltransferases which utilize 5,10- methylenetetrahydrofolate or carboxy-S-adenosyl methionine as a methyl donor have been identified, but will not be extensively covered here [70,71]. SAM methyltransferases have been categorized into at least five classes (I-V) based on their conserved structural folds [72–75]. tRNA methyltransferases in particular are predominantly confined to class I and class IV types [72,75]. Despite their highly conserved structural fold, the amino acid sequence as well as the mode of substrate binding can differ, often significantly, between tRNA modification enzymes within the same class. Over time, some methyltransferases have gained distinct auxiliary RNA binding domains, supporting a common theme of modular evolution whereby catalytic domains are combined with different substrate-binding domains to achieve substrate specificity [76–79].

Class I methyltransferases consist of alternating α helices and parallel β sheets characteristic of a typical Rossmann-fold, apart `from the presence of an additional

(seventh) antiparallel β-strand (Figure 1.7) [80]. Within this conserved fold, comparative

sequence analyses have revealed a set of at least six conserved regions (motifs I-VI)

encompassing the SAM-binding domain [72,80]. The first motif harbors a conserved

nucleotide-binding site, GxGxG (x is any residue), which forms the first α-helix as it bends underneath SAM [72,73]. The other known conserved SAM-binding residue is an acidic residue at the end of β2, which forms hydrogen bonds with both hydroxyls of the

SAM ribose [73]. Recently, in the TlyA ribosomal RNA methyltransferase from

Mycobacterium tuberculosis an additional novel SAM-binding motif, RxWV, was discovered [81], which affects the structure of a GxGxG motif in a purposeful way [81], and both motifs are required for SAM-binding [81]. Still, in general the SAM-binding region normally resides within the N-terminus of many class I methyltransferases and is assembled primarily from loops following strands 1, 2, and 3, whereas the substrate- binding region is generally located in the C-terminal portion of the β-sheet [72,82].

Beyond that, substrate binding for class I methyltransferases can vary greatly. There are

several instances described in which a class I methyltransferases is fused to a known

RNA-binding domain as discussed in more detail below [83,84].

Figure 1.7 Topology diagram of Class I and Class IV methyltransferases.

The alpha helices are shown as cylinders and beta sheets as arrows. The alpha helix in green is not always conserved in Class I (Rossmann fold) methyltransferases (left). The SAM binding domains are indicated by brackets. The characteristic trefoil knot motif of Class IV (SPOUT) methyltransferases is highlighted in yellow (right).

Class IV methyltransferases are characterized by the SPOUT domain which was

first discovered by bioinformatics analyses based on the similarity to the SpoU (TrmH)

and TrmD tRNA methyltransferases (Figure 1.6) [85–90]. The SPOUT domain contains

an unusual α/β fold distinguished by a deep trefoil knot formed by the folding of three β- sheets in the C-terminal portion of the SAM-binding motif (Figure 1.7) [75,91,92]. Many

SPOUT proteins have acquired additional N- and/or C-terminal extensions which presumably participate in substrate binding [68,89,93]. Two SPOUT enzymes, TrmL

(tRNA Um34/Cm34) and RlmH (rRNA), are considered “minimalist” proteins where

substrate binding is limited solely to the SPOUT domain and, in the case of TrmL, no

other RNA binding protein assists in binding [68,93–95]. TrmL acts as a homodimer to

26 catalyze 2’-O-methylation at position 34 on two isoacceptors of tRNALeu (anticodons:

CAA and UAA) and like many RNA binding enzymes, it uses positively charged residues for substrate recognition [94,95]. Interestingly, only three other SPOUT tRNA methyltransferases are found in bacteria (TrmH, TrmD, and TrmJ); all methylate many more different substrates than TrmL [94]. Presumably, these enzymes acquired nucleic acid-binding domains as N- and C-terminal extensions that support recognition of multiple substrates, whereas the minimal SPOUT enzyme TrmL binds fewer substrates, making these extensions unnecessary. The nucleic acid-binding domains found within

SPOUT as well as class I methyltransferases include PUA, THUMP, TRAM (OB-fold), and L30e domains (Figure 1.8) [89].

Figure 1.8 Topological diagrams of common RNA binding domains.

RNA binding domains (left) and representative crystal structures of RNA methyltransferases containing these domains (right). In the topological diagrams the alpha helices are shown as cylinders in blue and beta sheets as arrows in red. Alpha helices that are not well-conserved are shown in gray. Representative crystal structures are depicted in cartoon form with the corresponding domain shown with beta sheets in yellow and alpha helices in purple and dark blue. All enzymes are shown here in monomeric form and protein structure outside of the featured domain are shown in gray cartoon. These models were created using the Visual Molecular Dynamics (VMD) program.

The PUA domain (pseudouridine synthase and archeosine transglycosylase domain) is a highly conserved RNA binding motif found in all domains of life. The PUA domain is made up of 64 to 96 amino acids and forms a compact pseudobarrel α/β fold

5 (Figure 1.8) [96–98]. In humans, NSUN6 is a class I methylase which modifies m C72 of tRNACys and tRNAThr and contains a PUA RNA binding motif [96,99]. Interestingly, in

vitro methylation by NSUN6 requires the 3’-CCA sequence which is normally a prerequisite for aminoacylation [99]. A previous crystal structure of another PUA- containing enzyme, Pyrococcus horikoshii archaeosine tRNA-guanine transglycosylase,

revealed that the PUA domain makes direct contact with the 3’-CCA of tRNA indicating

a partially conserved mode of binding for enzymes containing this domain [99].

The putative THUMP RNA binding domain (thiouridine synthase,

methyltransferase and pseudouridine synthase domain) was first discovered by in silico

methods and thought to consist of 100 to 110 amino acids producing an α/β fold [100].

The crystal structure of the Thermatoga maritima Thil 4-thiouridine synthase, in

complex with a truncated tRNA confirmed the role of THUMP domains in RNA binding

and revealed a potential molecular ruler mechanism to define substrate specificity [101].

Interestingly, like Thil, the RNA methyltransferases Trm11, RlmKL, Trm14/TrmN, and

RlmN (a radical SAM enzyme) all contain an N-terminal ferredoxin-like (NFLD) domain

along with the THUMP domain, implying that THUMP and NFLD together could form a

functional binding unit as is seen in Thil [101]. For example, Trm11 from Eukarya and

Archaea are comprised of three domains: NFLD, THUMP, and a class I

methyltransferase catalytic domain (Figure 1.8).

The TRAM domain (Trm2 and MiaB domain) is conserved in Bacteria and

Eukarya, but not widely found in Archaea [78]. Some enzymes have recruited the TRAM

domain for RNA binding. This domain is predicted to form a simple β barrel. In

methyltransferases, TRAM domains have been identified at the N-terminus of Trm2-like

5 tRNA methyltransferases, the N-terminus of RlmD (formerly RumA, m U1939 E. coli 23S

rRNA) and the C-terminus of the 23S RNA-specific uracil O-methyltransferase, FtsJ,

from Halobacterium (Figure 1.8) [76,78]. The EcRlmD enzyme was the first to reveal the structure of the TRAM domain showing a five-stranded antiparallel β-barrel fold with a Greek key topology [102]. This structure adopts an oligonucleotide binding (OB) fold which is commonly found in DNA/RNA enzymes [102]. The putative SPOUT methyltransferase MT1 from Methanobacterium thermoautotrophicum also contains the

OB-fold subdomain unusually located between the center of its β-fold architecture [103].

The L30e domain is an RNA binding domain commonly found in ribosomal proteins (named for the ribosomal L30 protein) and forms a distinctive three-layer α/β/α sandwich. There are three conserved residues within the L30 protein family, Gly26,

Lys28, and Arg52, which are critical for RNA binding. The glycine residue, which lacks a bulky side-chain, provides space for the RNA to bind, while the basic residues form hydrogen bonds with nucleotides in the RNA substrate [91]. The SPOUT 2’-O- methyltransferase, RlmB, is comprised of an N-terminal L30e domain extending from the catalytic subunit (Figure 1.8) [103]. The RlmB family contains residues corresponding to the three RNA binding residues in L30 and indeed RlmB has been shown to methylate

Gm2251 of 23S rRNA in E. coli [103]. A putative SPOUT 2’-O-methyltransferase from

Thermus thermophilus also appears to have an L30e domain at its N-terminus [104].

While the L30e domain is typically found in ribosomal proteins, methylation of 23S rRNA of was not detected in intact ribosomes in T. thermophilus, and the putative

methyltransferase displays no sequence similarity to L30; consequently, the target RNA

is still unknown [104].

1.6.1 Sequence versus structural determinants for specificity of methyltransferases

There are two general strategies used by methyltransferases for identifying the

correct substrate. They either survey structural features or recognize specific sequences

within the RNA substrate. The majority of methyltransferases recognize local secondary

structure or global tertiary folds, but some act based on a specific consensus sequence

where recognition is usually in the context of local structural elements. RNA

methyltransferases commonly detect structural features of RNA, as seen with the

conserved consensus fold of m5U methyltransferases [105]. RlmD has three structural

domains, an N-terminal TRAM domain, a central α/β domain, and a C-terminal class I

methyltransferase domain [105]. RlmD strictly modifies 23S rRNA at a single position

(U1939) to m5U and will not methylate it if the U position is mutated to C [102]. Early

studies of another m5U methyltransferase, TrmA, suggested that it did not recognize a

strict consensus sequence on tRNA and it could efficiently modify short versions of the

T-stem loop [105]. However, a crystal structure of EcTrmA bound to the T-arm revealed a unique colinear base-stacking of G53–A58–G57–C56–U55 in a conformation that is not

normally seen in a typical unbound tRNA [105]. This presents the U54-U55-C56 sequence

(where the first U is the methylation target) into the active site [105]. This interaction has

been proposed to be a conserved specificity determinant for RlmD and TrmA [105].

Outside of this consensus fold, other RNA binding elements allow for discrimination

between the different substrates (rRNA versus tRNA); for example, unlike RlmD, TrmA

does not contain a TRAM domain [76,106].

As previously discussed for deaminases, tRNA methyltransferases are often

categorized into two groups based on substrate recognition: those requiring the entire

tRNA for activity and those that can act on shorter tRNA substrates. For example, the

SPOUT Um34/Cm34 EcTrmL methyltransferase specifically recognizes the stem-loop

structure of the anticodon stem-loop (extended by two nucleotides). EcTrmL also

recognizes the specific nucleotides A36-A37-A38, as well as other determinants emphasized in the next section [94]. On the other hand, the E. coli TrmJ tRNA Xm32 (X is any nucleotide) methyltransferase requires the full-length tRNA for activity [94].

Surprisingly, it detects specific regions within the D-loop rather than the anticodon loop where the nucleotide to be methylated is found [94]. In contrast to EcTrmJ, the archaeal

TrmJ from Sulfolobus acidocaldarius (SaTrmJ) can act on small tRNA substrates,

specifically tRNAs lacking the D- and T-loops [68]. Additionally, SaTrmJ can only

methylate C32 whereas bacterial EcTrmJ can modify all four nucleotides at position 32, although A and G are almost never found at that position [68]. Interestingly, the bacterial

and archaeal TrmJ are SPOUT methyltransferases whereas the eukaryotic tRNA Xm32

enzyme (Trm7 with auxiliary protein Trm732) is a class I methyltransferase [68].

1 Likewise, the tRNA m G37 methyltransferase, TrmD (SPOUT), does not require the L-

shaped tertiary structure of tRNA, but instead recognizes the anticodon arm structure and

also uses G36-G37 as a sequence determinant [94]. Unlike its bacterial counterpart, the

1 eukaryotic m G37 methyltransferase Trm5, an unrelated class I enzyme, surveys the

entirety of the tRNA tertiary structure to identify the correct substrate. It is noteworthy

that both TrmJ and Trm7 as well as TrmD and Trm5 are just a few examples of RNA

modification enzymes that have undergone convergent evolution, whereas two unrelated

enzymes (in this case SPOUT versus Rossmann-fold methyltransferases) have evolved

independently yet perform the same catalytic reaction. This further highlights the

functional significance of tRNA modifications in all domains of life.

1.6.2 Pre-existing modifications and their requirement for specificity of

methyltransferases

Another aspect of specificity in RNA modification enzymes is the order of

processing events and the role played by other modifications. Having a systematic

sequence of processing events or addition of modifications can ensure the proper

structure and function of RNA and allows for quality control. Of course, these events can

be dictated by compartmentalization, perhaps best demonstrated by retrograde transport systems where some modifications require reimport into the nucleus.

Several methylations, in addition to other modifications, require the RNA to be

processed before it can be introduced. This can be due to impediment via steric hindrance in the pre-processed substrate where processing can be necessary to form the proper sequence or structural recognition elements. An interesting case is the splicing of eukaryotic tRNAs introns. The wybutosine (yW) biosynthesis pathway in eukaryotes

1 requires the addition of m G37 by Trm5 as the first of several steps [66,107]. In S.

cerevisiae, the tRNA must be spliced before the methylation reaction can take place;

however, the splicing machinery is located on the cytoplasmic side of the outer

mitochondrial membrane and it is in this context that intron-removal takes place

[30,92,107–110]. The Trm5 enzyme, however, is a nuclear enzyme [107]. Therefore, tRNAPhe must be imported back into the nucleus after splicing in order to be methylated

[30,107]. Afterwards the tRNA is re-exported to the cytoplasm to become further

modified to wybutosine. In contrast, some methylations are actually intron-dependent. In

budding yeast and humans, for example, the multisite m5C tRNA methyltransferase,

Trm4, requires an intron for methylation at positions 34 and 40 [111,112]. Trm4 does not

survey the overall L-shaped tRNA structure, but rather has been suggested to be

dependent on specific nucleotide sequence and base-pairs within the intron. However,

Trm4-mediated m5C modification at positions 48 and 49 is unaffected by the presence or absence of an intron in S. cerevisiae.

The majority of RNA methyltransferases modify canonical nucleotides, but they

can also methylate more complex substrates like nucleotides with preexisting

modifications. RlmH (SPOUT) from E. coli specifically methylates position 1915 of 23S

rRNA exclusively if that position has a preexisting pseudouridine (ψ) [74]. It is the only

methylated pseudouridine found in bacterial RNA to date, although several have been

found in Eukarya (m1Ψ, Ψm, and m1acp3Ψ) [74]. Modifications at positions outside of

the target methylation site also influence methylation events. For example, EcTrmL

methylates pyrimidines but not purines at the wobble position and requires prior

6 formation of i A at position 37 [94]. The sequence A36-A37-A38 is also important for

EcTrmL methylation, but it is unclear whether this sequence is necessary for recognition

6 by the enzyme itself or because of the i A37 modification [94]. In any case, the addition of

6 Leu i A37 in to an otherwise unmodified in vitro transcribed tRNA is sufficient to recruit

6 TrmL to the tRNA [94]. A similar situation occurs in S. cerevisiae where i A37 is a

3 ser prerequisite for m C32 formation in tRNA [113]. These results provide evidence for an

anticodon loop circuitry where modifications within this loop are highly interrelated.

Methylation can also act as a requirement for subsequent modifications at the

same position or elsewhere in the RNA. The methylation of m1A is found universally at

position 58 of tRNA and is additionally present at several other positions within the T- loop depending on the organism [114]. In Archaea, certain tRNAs contain m1A at

1 position 58 and/or 57 as a mandatory step in the biosynthesis of m I57, produced by

adenosine to inosine (A-to-I) deamination. The methylation is catalyzed by the tetrameric

archaeal TrmI enzyme and is strictly required for deamination to occur [40,114]. The

1 orthologous m A58 site-specific methyltransferase in S. cerevisiae also forms a tetramer, but is comprised of two different evolutionary related subunits (further discussed in the next section). A similar, yet unconventional, two-step pathway was recently discovered in

T. brucei involving the TbADAT2/3 deaminase discussed previously. In addition to its role in the conserved A-to-I deamination at position 34, TbADAT2/3 also converts C-to-

3 U at position 32 in trypanosomes [48,115], but it requires a methylated substrate, m C32,

and the corresponding methyltransferase, TbTrm140, for this activity [116,117].

Remarkably, the methyltransferase TbTrm140 also requires the presence of TbADAT2/3

to form the methylated substrate prior to deamination [116,117]. Therefore, these enzymes are mutually dependent on each other for function [116,117].

In an analogous scenario and adding further complexity to the yW biosynthesis in

tRNAphe highlighted above, it has been suggested that one of the enzymes in the pathway,

Tyw1, depends on prior formation of Gm34 and Cm32 by the Trm7 enzyme (along with its

associated partner proteins) [113,118]. Therefore, the entire proposed pathway consists of export of the intron-containing tRNAPhe precursor from the nucleus to the cytoplasm for splicing and 2’-O-methylation by Trm7, followed by import back into the nucleus for

1 m G37 and lastly re-export into the cytoplasm for the final steps to produce yW37 (Figure

1.9) [118].

Figure 1.9 Processing pathway of tRNAPhe in Saccharomyces cerevisiae.

tRNAphe is transcribed in the nucleus with 5’ and 3’ extensions (orange) and an intron (red). The extensions are removed and 3’-CCA is added in the nucleus. The tRNA is exported to the cytoplasm and undergoes splicing by the splicing endonuclease (SEN) protein complex located to the outer mitochondrial membrane facing the cytoplasm. The spliced tRNA is 2’-O-methylated by Trm7/Trm732 for Cm32 and Trm7/Trm734 for Gm34. The tRNA is then imported back into the 1 nucleus where it is methylated to m G37 by Trm5, initiating the first step of the yW pathway. The tRNA is re-exported where it undergoes further biosynthesis steps to form yW when the presence of Cm32 and Gm34 serve as key determinants.

Besides acting as an obligatory intermediate for additional modifications,

methylations can positively or negatively influence other modifications and vice versa.

7 1 The modification m G37 positively impacts Gm18 and m G37 while Ψ55 negatively affects

5 2 1 the levels of Gm18, m S U54, and m A58 in T. thermophilus [113]. In trypanosomes, it has

been proposed that upon mitochondrial import, editing of tRNATrp relies on the interplay

between several modifications; it is first 2’-O-methylated at position 32, thiolated at

2 position 33 to form S U33, edited from C34 to U34, while the unedited C34 is methylated to

Cm34. Interestingly, downregulation of the thiolation activity leads to almost complete

editing at position 34. Therefore, methylation may serve as a positive determinant for C34

to U34 editing, while thiolation negatively affects the editing reaction helping maintain

two functional forms of tRNATrp, edited and unedited [119].

1.6.4 Multisubunit methyltransferases

One noteworthy aspect of RNA methyltransferases is their tendency to recruit additional partner proteins, which often play a part in substrate binding, catalytic efficiency, stability, or localization. Furthermore, oligomerization of the same subunit can also vary between methylation enzymes from different domains of life or even within the same domain affecting substrate specificity or catalysis. It has been proposed that the formation of homodimers could act as a possible step in the evolution of heterodimeric enzymes whereby (as discussed previously), a gene duplication event may arise and lead to functional specialization of each subunit over time [80,120]. In general, SPOUT methyltransferases form homodimers where each monomer partakes in substrate binding and the active site is present at the dimer interface. However, the SPOUT enzyme,

Trm10, acts as a monomeric enzyme, suggesting that catalysis and substrate recognition

for this enzyme may differ from canonical SPOUT methyltransferases [75,121,122].

Unlike the SPOUT methyltransferases, the class I methyltransferases are normally found

as monomers, although there are few exceptions such as Bacillus subtilis TrmB, which

will be discussed later on in this section.

Some enzymes recruit additional proteins to mediate substrate binding. The S.

cerevisiae 2’-O-methyltransferase Trm7 interacts with an associated protein Trm732 to

Phe methylate C32 and with another protein, Trm734 for methylation of X34 on tRNA ,

tRNALeu and tRNATrp [118]. In this case, Trm7 is the catalytic subunit, which uses either

Trm732 or Trm734 to direct substrate specificity. The adaptor protein Trm112 provides

another example in which a methyltransferase requires an unrelated protein factor for

activity. Trm112 interacts with multiple tRNA methyltransferases including Trm9 and

5 2 Trm11 to form mcm U34 and m G10, respectively, in eukaryotic tRNAs [123]. Initial

studies in S. cerevisiae showed that deletion of Trm11 displayed minimal growth defect

whereas Trm112 deletion resulted in a severe growth phenotype, indicating its

importance in other cellular functions [123]. In fact, Trm112 associates with several

different factors involved in other cellular activities; for instance, it provides structural

stability for another methyltransferase, Bud32, which methylates G1575 of 18S rRNA in

eukaryotes [124–126]. Therefore, Trm112 performs multiple functional roles in tRNA modification as well as other activities in the cell.

1 In S. cerevisiae, the m A58 is a common but non-universal modification as it

appears in 21 of 42 sequenced tRNAs [127]. This methylation is catalyzed by a

heterotetrameric enzyme composed of the subunits Trm6 (Gcd10p) and Trm61 (Gcd14p);

mutation of the putative SAM-binding residues of Trm61 abolished activity, while in vitro neither Trm61 nor Trm6 could by themselves bind tRNA [127]. This suggested that

Trm61 was responsible for catalysis whereas Trm6 aided in tight binding to the tRNA

substrate in the context of the other subunit. Mutational studies later demonstrated that

residues in both Trm61 and Trm6 subunits participate in tRNA binding [128]. As

mentioned earlier the orthologous bacterial/archaeal methyltransferase TrmI forms a

tetramer, which may be conserved for its role in recognition, binding, or activity.

However, the binding interface appears to diverge between prokaryotic TrmI and

eukaryotic Trm6/Trm61 indicating that these enzymes evolved different means of

substrate recognition and binding [128].

In eukaryotes, Trm8 and Trm82 form a complex which modifies position 46 of

seven different tRNAs with m7G [129]. While Trm8 contains the putative catalytic site

and has tRNA binding ability, Trm82 does not engage in either catalysis nor tRNA

binding [120]. Instead, the formation of the Trm8-Trm82 complex causes Trm8 to

undergo considerable conformational change forming the correct architecture for

catalysis [75]. Although both Trm8/Trm82 enzymes are conserved among Eukarya,

Bacteria utilize a single subunit enzyme, TrmB, a homolog of Trm8. Interestingly, there

are differences even among bacterial TrmB, where E. coli TrmB is a monomeric enzyme

whereas Bacillus subtilis forms homodimers [80]. The dimerization of BsTrmB is also

unusual for a class I methyltransferase as its structural elements deviate from the

expected Rossmann-fold [80]. This finding supports the evolution of obligatory dimers

observed for Trm8/Trm82 as well as a variety of other dimeric methyltransferases like

those listed above.

Rather than enlisting other proteins, some methyltransferases, lacking intrinsic

substrate specificity, utilize guide RNAs to direct catalysis. These enzymes form

complexes with small C/D box guide RNAs that direct methylation to the target

nucleotide. This mechanism is seen primarily in eukaryotic rRNA and small nuclear

RNA as well as archaeal 2’-O-methylation of rRNA and in some cases tRNA [130]. The need for RNA-dependent methylation of tRNA is not conserved as 2’-O-

methyltransferase of positions 32 and 34 of S. cerevisiae tRNATrp and of position 18 from

Bacteria (TrmH) and Eukarya (Trm3) do not involve guide RNAs [114]. Methylation via

guide RNAs is yet to be seen in Bacteria, suggesting this system likely evolved after the

Bacteria split from Archaea and Eukarya.

Chapter 2 : Binding synergy as an essential step for tRNA editing and

modification enzyme co-dependence in Trypanosoma brucei2

2.1 Introduction

All nucleic acids in cells undergo some type of post-transcriptional chemical modification; these may involve chemically simple modifications such as methylations and thiolations or more complex ones requiring several chemical building blocks.

Regardless, the importance of modifications is highlighted by their prevalence and degree of evolutionary conservation, and further made obvious by the fact that organisms dedicate more than 1% of their genome to encode modification enzymes [131–133].

Functionally, some modifications act as structural modulators increasing flexibility or rigidity as needed, in turn affecting nucleic acid stability, others can directly impact gene function by altering the genetic information of the nucleic acids they target.

By far, tRNAs are the recipients of the largest diversity of modifications. In general, modifications at the anticodon loop play critical roles in translational efficiency and/or fidelity, whereas modifications that are more distal from the anticodon arm ensure proper folding of the tRNA and thus affect tRNA function indirectly. To date over 100 different modifications have been identified in tRNA and in some organisms, for example

2 Chapter 2 is published as: McKenney K.M., Rubio, M.A., and Alfonzo, J.D. Binding synergy as an essential step for tRNA editing and modification enzyme co-dependence in Trypanosoma brucei. RNA. 2017. 24(1):56-66. 42

Saccharomyces cerevisiae, almost a complete set of modification enzymes have been

described. In most cases, however, it is less clear how each enzyme at the molecular level

recognizes and targets specific substrates in a pool of very similar non-substrate tRNAs.

Predictably, their chemical diversity and the variety of tRNA positions targeted dictate

that modification enzymes may have evolved numerous ways of substrate recognition;

some binding to local secondary structural features, some recognizing the global L-

shaped tertiary fold and yet others surveying specific sequences within the tRNA

molecule. Notably, modification enzymes may act on a tRNA molecule at any point in its

folding pathway, with some modifications playing critical roles early, ensuring the

formation of certain structures while avoiding formation of unwanted conformers. These

enzymes lay the structural foundation for subsequent modification enzymes that rely on

tRNA architecture for activity (Helm 2006; Swinehart and Jackman 2015; Grosjean et al.

1996; McKenney et al. 2017). For instance, introduction of m1A at position 9 in human

mitochondrial tRNALys favors formation of its canonical L-shape structure [134,136,137].

2 Further modification of m G10 and pseudouridine (ψ) at positions 27 and 28 strongly

depends on proper formation of tertiary structure, and by extension, upon prior synthesis

1 of m A9 [134]. Many modification enzymes require a fully folded tRNA for activity; among these, S. cerevisiae tRNA adenosine deaminase, ScTad2/3, strictly relies on the

global structure of the tRNA to form inosine at position 34 (I34) [37,138,139]. However,

its bacterial counterpart, ADATa or TadA, in vitro can efficiently deaminate a minimal

substrate composed only of the anticodon arm [46,138]. Likewise, some tRNA

methyltransferases, including bacterial TrmJ and eukaryotic Trm5 methyltransferases

require full-length tRNA, whereas others are less stringent and can modify shorter

substrates in vitro [68].

A growing theme in the RNA modification field is that many modifications do not

occur in isolation and may be part of well-orchestrated cascades, whereby one modification is essential or may influence the synthesis of another; in such cases, modifications are predicted to follow a strictly ordered set of reactions. A model of modification interdependence, where modifications may rely on prior modifications, was first proposed after the observation of potential modification and editing cascades in mitochondrial tRNAs [13,115,116,119,140]. Since then, more examples of modification cascades have surfaced [94,113,118,119,141–144]. In terms of tRNA methylation, there are several cases in which methyltransferases specifically act on a previously modified substrate, or where a methylation serves as a prerequisite for further modification [135].

In E. coli, for instance, the TrmL methyltransferase requires formation of i6A at position

37 [94]. The sequence A36-A37-A38 is important for EcTrmL methylation, but whether

6 this sequence is necessary for recognition by the enzyme itself or for the i A37

6 modification is yet unclear; nonetheless, incorporation of i A37 is sufficient to recruit

EcTrmL to the tRNA in vitro [94]. Likewise, in S. pombe N6-isopentenyl adenosine (i6A)

3 Ser at position 37 is required for m C32 formation on tRNA [113]. Indeed, S. cerevisiae

6 Thr Trm140 recognizes the sequence identity element G35-U36-t A37 in tRNA substrates;

however, the tRNASer substrate lacks this sequence [145]. Instead, tRNASer recognition

was stimulated by seryl-tRNA synthetase (Ses1) and relies on its unique variable loop in

6 6 addition to t A37 and i A37 [145]. In yet another example, deamination of adenosine to

inosine (A-to-I) is required for further methylation at position 37 of tRNAAla in

eukaryotic tRNAs [40,41,146]. In marsupials, deamination editing from cytidine to

uridine (C-to-U) at position 35 acts as a determinant for queuosine formation at position

34 in the anticodon loop of tRNAAsp [13,113]. There is also direct precedence for

methylation as a requirement for deamination; in Archaea, position 58 and/or 57 can be

modified with m1A depending on the organism [40,147]. If m1A occurs at position 57, as

1 in the majority of Archaea, it is subsequently converted to m I57 via adenosine deamination [40,146,147].

We recently described an unexpected and uniquely extreme instance of modification interdependence involving a tRNA deaminase (TbADAT2/3) and a methyltransferase (TbTrm140) from T. brucei. TbADAT2/3 and TbTrm140 act together to edit and modify a single nucleotide position (cytosine 32, C32) in the anticodon loop of

several tRNAs. This position is first methylated to form 3-methylcytosine (m3C) and then deaminated to generate 3-methyluridine (m3U). Remarkably, to form these products both enzymes must be present in the reaction and consequently they form a stable protein complex in vitro and in vivo [116]. This example of strict interdependence by both enzymes then raises questions as to what each enzyme contributes to each other in

3 targeting C32. In the present report, we have taken advantage of the robustness of m C

formation in vitro and performed a series of binding and kinetic studies that show that interdependence may be driven by the synergistic effect that both enzymes have on substrate binding when added together in the same reaction. We also show that active site residues in each enzyme contribute minimally to binding synergy and that the observed

mutual enhancement of substrate binding depends on domains that are more distal to the

active sites. These findings have implications for multi-substrate recognition and are

discussed in the context of how binding synergy may be exploited to ensure high

specificity.

2.2 Results

2.2.1 TbADAT2/3 and TbTrm140 bind synergistically to tRNAThr

To explore the basis for the co-requirement of TbTrm140 and TbADAT2/3 for

methylation and deamination activity at position 32 of tRNAThr, an Electrophoretic

Mobility Shift Assay (EMSA) was established. In these experiments, a slower migrating

band was observed when either TbADAT2/3 or TbTrm140 was incubated with the tRNA,

indicating formation of a stable protein-RNA complex as compared to a tRNA alone

control (Figure 2.1A,B). The resulting data was fitted to a binding isotherm with a single

exponential and an apparent dissociation constant (Kdapp) was calculated. TbADAT2/3

yielded a Kdapp of 0.21 ± 0.03 µM, comparable to that shown by our laboratory with a different tRNA (Figure 2.1A,C) [148]. There are three different isoacceptors of tRNAThr

(anticodons AGU, UGU and CGU); all are substrates for m3C/m3U formation at position

32 [149] and all showed similar Kdapp in these experiments. However, given that position

32 methylation occurs even in tRNAs that lack an adenosine at position 34 (which is

Thr converted to inosine by the same TbADAT2/3 deaminase), we focused on tRNA CGU

for further binding studies, thus uncoupling contributions, however negligible, to binding

from A34 [148,150,151]. No binding parameters have been previously determined for

TbTrm140. Thus, before exploring what the interaction between TbADAT2/3 deaminase

and TbTrm140 contribute to binding, we performed similar binding experiments as above

Thr with recombinant TbTrm140 and tRNA CGU (Figure 2.1B,D). Constant concentrations of substrate tRNA were incubated with increasing concentration of enzymes and, as above, subjected to EMSA. This yielded a binding isotherm with a nominal Kdapp of 0.21

± 0.05 µM (Figure 2.1D), which is similar for the binding of TbADAT2/3 to the same

substrate.

Figure 2.1 TbADAT2/3 and TbTrm140 stably bind tRNA in vitro.

Thr Analysis of protein-tRNA interactions using EMSA where radioactively labeled tRNA CGU (2.5 nM) was incubated with increasing concentrations of enzyme and separated on a native Thr acrylamide gel. (A) Representative EMSA of TbADAT2/3 incubated with tRNA CGU. Lane 1 is a no-enzyme control, lanes 2-6 show tRNA with an increasing concentration of TbADAT2/3 (0.06, 0.12, 0.24, 0.48, and 0.7 µM, respectively). (B) EMSA of TbTrm140 incubated with Thr tRNA CGU. Lane 1 is a no-enzyme control, lanes 2-6 show tRNA with an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). The fraction of total bound tRNA from the EMSA gels was quantified and plotted as a function of protein concentration. The data was fit to a single-ligand binding isotherm and the apparent dissociation constant (Kdapp) was determined as described in the materials and methods. These graphs are Thr shown in the bottom panels, where (C) TbADAT2/3 with tRNA CGU and (D) TbTrm140 with Thr tRNA CGU. Each figure represents at least 5 independent replicates.

Given that TbADAT2/3 and TbTrm140 form stable complexes in vivo and in vitro and both enzymes are required for methylation (Figure A.1) [116], we determined the impact of each protein on substrate binding when incubated together with the tRNA

Thr substrate. Similar EMSAs were performed, but this time, radiolabeled tRNA CGU was incubated with one enzyme at a constant concentration while adding an increasing concentration of the other. First, TbTrm140 was held constant at 0.210 µM and different increasing concentrations of TbADAT2/3 were added, resulting in a stable protein-RNA complex (Figure 2.2A) and yielding a Kdapp of less than 0.03 µM (Figure 2.2C), representing over a 7-fold increase in affinity from what is seen with TbADAT2/3 alone

(compare to Figure 2.1B,D). The Kdapp value determined from the binding isotherm is below the detectable limits of this assay, therefore it is reported here as below 0.03 µM.

An analogous increase in binding affinity was determined when the reciprocal experiment was performed. In this case holding TbADAT2/3 constant while increasing

TbTrm140 again resulted in a Kdapp below 0.03 µM (Figure 2.2B,D). These results show that, when added together, TbTrm140 and TbADAT2/3 bind synergistically exhibiting an improvement in binding that is more than the sum of the individual binding affinities.

Figure 2.2 TbADAT2/3 and TbTrm140 bind tRNA synergistically.

Thr (A) EMSA of TbADAT2/3 incubated with tRNA CGU in the presence of a constant concentration (210 nM) of TbTrm140. Lanes 1 and 2 show a no enzyme control reaction and a control reaction with no TbTrm140 added, respectively. Lanes 3-7 show an increasing concentration of TbADAT2/3 (0.06, 0.12, 0.24, 0.48, and 0.7 µM, respectively). (B) EMSA of Thr TbTrm140 incubated with tRNA CGU in the presence of a constant concentration (210 nM) of TbADAT2/3. Lanes 1 and 2 show a no enzyme control reaction and control reaction with no TbADAT2/3 added, respectively. Lanes 3-7 show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). The data was fit to a single-ligand binding isotherm and the apparent dissociation constant (Kdapp) calculated as in figure 1. These graphs are shown in the bottom panels, (C) and (D) for TbADAT2/3 and TbTrm140 respectively. Each figure represents at least 5 independent replicates.

To rule out the possibility that the observed increase in binding is simply the

result of potential molecular crowding, we performed similar experiments with

TbADAT2 alone without its partner TbADAT3 (Figure 2.3A). This protein can still form a stable homodimer, but by itself is unable to either bind tRNA or catalyze the deamination reaction. To test this, a 3-fold excess of TbADAT2 (600 nM) as compared to the levels of TbADAT2/3 in the previous reactions were added in the presence of increasing concentrations of TbTrm140 as before. This yielded a Kdapp of 0.26 +/- 0.06

µM, which represents binding by TbTrm140 alone. Similar experiments where performed

with bovine serum albumin still at 600 nM constant concentration while increasing either

TbADAT2/3 or TbTrm140 (Figure 2.3B,C). This yielded a Kdapp of 0.18 +/- 0.03 µM

and 0.15 +/- 0.04 µM, respectively. Therefore, no sign of synergy was observed, ruling

out the possibility of non-specific crowding effects as the root cause of the synergy seen

with TbADAT2/3 and TbTrm140. Importantly, in these experiments the order of addition

of the proteins to the tRNA had no impact on binding behavior. Taken together, these

experiments suggest that part of the co-requirements of both enzymes for activity may

rest on their mutual contribution to binding affinity.

Figure 2.3 Synergistic binding is exclusively linked to the presence of TbADAT2/3 and TbTrm140 in the reaction.

Thr (A) EMSA of TbTrm140 to tRNA CGU in the presence of TbADAT2. Lanes 1 and 2 are a no enzyme control reaction and a control reaction with no TbADAT2 added, respectively. Lanes 3-7 show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, Thr respectively). (B) EMSA of TbADAT2/3 to tRNA CGU in the presence of BSA. Lanes 1 and 2 show a no enzyme control reaction and control reaction with no BSA added, respectively. Lanes 3-7 show an increasing concentration of TbADAT2/3 (0.06, 0.12, 0.24, 0.48, and 0.7 µM, Thr respectively). (C) EMSA of TbTrm140 to tRNA CGU in the presence of BSA. Lanes 1 and 2 show a no enzyme control reaction and control reaction with no BSA added, respectively. Lanes 3-7 show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 5 independent replicates.

2.2.2 Synergistic binding is independent of enzyme activity

To further understand how the two enzymes work in complex, we explored whether the synergy observed was simply due to two different active sites converging on a single position. This being the case, it is expected that mutations at the active site of either enzyme that impair activity should abrogate binding synergy. To test this, previously described active-site and binding mutants of TbADAT2/3 were tested in the presence of TbTrm140. Catalytic mutants were previously generated by alanine substitutions of the proton-shuttling glutamate (E92A) of TbADAT2 and a critical zinc binding residue (C291A) of TbADAT3. These catalytically dead mutants can still form heterodimers and bind tRNA with a similar affinity as wild-type TbADAT2/3 alone

[148]. As a control, a tRNA binding-deficient mutant created previously by deletion of the last 10 amino acids of the C-terminal end of TbADAT2 was also tested [148].

Constant concentrations of each mutant were incubated with radiolabeled tRNA in the presence of increasing concentrations of wild-type TbTrm140 as before, and the resulting dissociation constant was estimated. Neither catalytic mutant (ADAT2/3 E92A and

ADAT2/3 C291A) had any effect on binding synergy with Kdapps comparable to wild-type

TbADAT2/3 in the presence of TbTrm140 (Figure 2.4A and 2.4B), whereas the

TbADAT2/3 binding-deficient mutant (ADAT2/3 C-ter∆10) exhibited a Kdapp of 0.12 ±

0.03, reflective of the binding by TbTrm140 with no contribution by TbADAT2/3

(Figure 2.4C). This last observation is expected given that the 10-amino acid C-terminal deletion of TbADAT2 completely abrogates tRNA binding. Taken together this suggests

53 that binding domains, which are not directly involved in catalysis are the major contributors to the observed synergy.

Figure 2.4 TbADAT2/3 catalytic residues do not contribute to increased affinity of TbTrm140 for tRNA.

Thr (A) EMSA of TbTrm140 to assess binding to tRNA CGU in the presence of two TbADAT2/3 Thr catalytic mutants (E92A) and (B) (C291A). (C) EMSA of TbTrm140 to tRNA CGU in the presence of TbADAT2/3 C-terminal deletion binding mutant. In each panel, lanes 1 and 2 show a no enzyme control reaction and a control reaction with no mutant TbADAT2/3 added, respectively. Lanes 3-7 show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 5 independent replicates.

We then generated mutants at conserved SAM-binding residues of TbTrm140.

Methyltransferases which utilize S-adenosylmethionine (SAM) as a methyl donor are categorized into at least five classes (I-V) based on their structural folds. Most SAM methyltransferases, including TbTrm140, are class I which have a conserved and predictable Rossmann-like fold. Class I methyltransferases are made-up of a parallel β-

sheet surrounded by helices, but unlike the Rossmann-fold, they contain an additional antiparallel β-strand. The class I methyltransferases often vary in sequence, but generally contain a characteristic nucleotide binding motif, GXGXG, important for SAM binding.

Analogous conserved residues are found in TbTrm140 and a potential TbTrm140 catalytic mutant was produced by alanine substitution of two of these residues (G124A and G126A). This mutant (Trm140 G124/G126A) had no detectable m3C formation

activity in vitro (data not shown), but despite this, no effect on binding synergy was

observed in the presence of wild-type TbADAT2/3 (Figure 2.5B). Once again

reinforcing the view that analogous to the TbADAT2/3, active-site residues contribute little to binding synergy.

Figure 2.5 TbTrm140 catalytic residues do not contribute to increased affinity of TbADAT2/3 for tRNA.

Thr (A) EMSA of TbTrm140 catalytic mutant (G124/G126A) to tRNA CGU. Lane 1 is a no enzyme control reaction. Lanes 2-6 show an increasing concentration of TbTrm140 catalytic mutant Thr (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). (B) EMSA of TbADAT2/3 to tRNA CGU in the presence of the TbTrm140 catalytic mutant. Lanes 1 and 2 show a no enzyme control reaction and control reaction with no mutant TbTrm140 added, respectively. Lanes 3-7 show an increasing concentration of TbADAT2/3 (0.06, 0.12, 0.24, 0.48, and 0.7 µM, respectively). The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 5 independent replicates.

Some methyltransferases use common specific RNA binding domains to bind

tRNA while others use motifs that are not easily recognizable or that have not been

described yet. No obvious RNA binding motif(s) is present in Trm140 from either S.

cerevisiae (NP_014882.4) or T. brucei. However, the activity of ScTrm140 was

abolished upon deletion of a string of residues at the C-terminus (D602-Q621), notably this domain is distal to the conserved SAM-binding residues and presumably does not form part of the active site, suggesting that this region may be important for tRNA binding [152]. The protein sequence of TbTrm140 was analyzed by the DNA- and RNA- binding prediction tool, BindN (Figure A.2), revealing a region of positively charge

residues found in the N-terminal portion of the protein [153]. A string of positively

charged residues equivalent to those deleted in ScTrm140 was also found at the C-

terminus of TbTrm140. The N-terminal and C-terminal ends of TbTrm140, were respectively deleted and the resulting mutants analyzed by EMSA, only the former impaired binding (Figure 2.6A). Therefore, the N-terminal deletion mutant (TbTrm140

∆S2-G17) was chosen for subsequent studies. Comparable to wild-type TbTrm140, the

N-terminal deletion mutant can still form homodimers (Figure 2.7). The C-terminal deletion mutant (TbTrm140 ∆I320-S340) showed similar binding behavior as the wild- type and was therefore not pursued further (Figure 2.8). Incubation of the binding- defective mutant, TbTrm140 ∆S2-G17, with increasing concentrations of TbADAT2/3 abrogated synergy and a dissociation constant reflective of TbADAT2/3 binding alone was observed (Figure 2.6B). This analysis has thus identified an important tRNA- binding domain in TbTrm140 that is distinct from the analogous domain in S. cerevisiae.

These results also emphasize how tRNA binding domains more distal to the active sites

of TbADAT2/3 and TbTRM140 are important for binding synergy; these must have a

direct bearing in the co-requirement for both enzymes in the reaction.

Figure 2.6 N-terminal residues of TbTrm140 are important for tRNA binding.

Thr (A) EMSA of TbTrm140 N-terminal deletion mutant, ∆S2-G17 to tRNA CGU. Lane 1 is a no enzyme control reaction. Lanes 2-6 show tRNA with an increasing concentration of TbTrm140 N-terminal deletion mutant (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). (B) EMSA of Thr TbADAT2/3 to tRNA CGU in the presence of TbTrm140 binding mutant. Lanes 1 and 2 show a no enzyme control reaction and control reaction with no mutant TbTrm140 added, respectively. Lanes 3-7 show an increasing concentration of TbADAT2/3 (0.06, 0.12, 0.24, 0.48, and 0.7 µM, respectively). The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 5 independent replicates.

Figure 2.7 TbTrm140 N-terminal deletion mutant forms homodimers.

Superdex 200 size exclusion chromatographs of TbTrm140 N-terminal deletion mutant, ∆S2- G17, (MUT, black line) compared to the TbTrm140 wild-type protein (WT, red dashes). The curves represent the absorbance (mAU) at A280 plotted against the elution volumes. Tb Trm140 N-terminal mutant and TbTrm140 wild-type eluted with a peak absorbance at 13.5ml, corresponding to approximately 76 kDa. This size is consistent with the calculated molecular weight of a TbTrm140 homodimer. The locations of protein standards of known sizes are noted above the graph in kilodaltons (kDa).

Figure 2.8 C-terminal residues of TbTrm140 are not important for tRNA binding.

Thr EMSA of TbTrm140 C-terminal deletion mutant, ∆I320-S340 to tRNA CGU. Lane 1 is a no enzyme control reaction. Lanes 2-6 show tRNA with an increasing concentration of TbTrm140 C-terminal deletion mutant (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively).

2.2.3 Validation of binding constants by single turnover kinetics

A limitation of EMSA in our situation is that a precise measurement of dissociation constants cannot be achieved. This is due to the fact that titration of one enzyme while the other is held constant limits the linear range of the assay; eventually the dissociation constant calculation becomes reflective of only the enzyme that was held constant. Thus, single turnover kinetic assays were performed to corroborate and more precisely define the binding affinities established by EMSA. First, to compare affinities between TbADAT2/3 alone and TbADAT2/3 with TbTrm140, we performed A-to-I

deamination assays. Here it is expected that since A-to-I formation is predictably independent of m3C formation, no synergy will be observed when using inosine formation as a reporter. In this experiment a saturating amount of enzyme, to ensure

Thr single-turnover conditions, was incubated with tRNA AGU substrate radiolabeled at

every adenosine and product formation assessed over time by 1D-TLC (Figure 2.9A).

The resulting data were fit to the equation [f = a(1 - e-kt)] as previously described to obtain an observed rate constant (kobs), which was then used to derive an apparent

dissociation constant (Kdapp). TbADAT2/3 alone exhibited a Kdapp of 0.18 ± 0.06 µM,

which is in agreement with the results from EMSA (0.21 ± 0.03 µM) (Figure 2.10A,C)

[148]. As expected, this Kdapp was virtually unchanged when A-to-I activity was tested in the presence of TbTrm140 (Figure 2.10B,D). Similar experiments were performed while

3 monitoring m C formation (Figure 2.9B) to calculate kobs and derive a Kdapp for

Thr methylation, but this time tRNA CGU radiolabeled at every cytosine was used as a

substrate. Here, however, because both enzymes are required for m3C production; the

assay was performed with both present in the reaction. From this experiment the

3 kinetically determine Kdapp for m C was 0.02 ± 0.01 µM, which is in agreement with our

results from the EMSAs (<0.03 µM) and supports the view that when together both

enzymes act synergistically. Importantly, these experiments highlight the fact that while

synergy is important for m3C formation; it plays no role on A-to-I despite both reactions

requiring the same deaminase (Figure 2.11A,B).

Figure 2.9 Kinetic Determination of binding affinities.

Representative 1-Dimensional Thin Layer Chromatography (TLC) activity assay for (A) Adenosine to Inosine deamination assay with TbADAT2/3 in the presence of TbTrm140 and (B) m3C Methylation assay with TbTrm140 in the presence of TbADAT2/3. Samples were taken at increasing timepoints as shown in Figs. 5 and 6.

Figure 2.10 Kinetic determination of the dissociation constant of TbADAT2/3 for tRNAThr in the presence of TbTrm140.

Thr Single turnover assays of (A) TbADAT2/3 with tRNA CGU alone and (B) in the presence of TbTrm140 were performed as described in the materials and methods. The fraction of inosine formed for each TbADAT2/3 protein concentration ranging from 10nM to 2500nM was measured over time as indicated in the graph. Determination of dissociation constants for (C) TbADAT2/3 Thr with tRNA CGU alone and (D) in the presence of TbTrm140. The fraction of inosine produced was plotted as a function of time and fit to a single exponential curve [f = a(1 - e-kt)], where f represents inosine formed, a denotes inosine produced at the end point of the reaction, k signifies kobs and t is time. The resulting kobs values were plotted against the concentration of TbADAT2/3 and fit to a single ligand binding isotherm. The Kdapp was then determined by nonlinear regression using Sigmaplot. Each graph represents at least 5 independent replicates.

Figure 2.11 Kinetic determination of dissociation constant of TbTrm140 to tRNAThr in the presence of TbADAT2/3.

Thr (A) Single turnover assays of TbTrm140 to tRNA CGU in the presence of TbADAT2/3 were performed as described in the materials and methods. The fraction of methylated cytosine 32 was measured for each TbTrm140 protein concentration ranging from 10nM to 2500nM as shown in the graph. The methylated fraction was plotted as a function of time and fit to a single exponential curve [f = a(1 - e-kt)], where f represents methylated cytosine formed, a denotes methylated cytosine produced at the end point of the reaction, k signifies kobs and t is time. (B) The resulting kobs values were plotted against the concentration of TbTrm140 and fit to a single ligand binding isotherm. The Kdapp was then determined by nonlinear regression using Sigmaplot. Each graph represents at least 5 independent replicates.

To validate the observation that the active site residues are not involved in binding synergy we performed methylation assays with different combinations of catalytic and binding mutants of TbTrm140 and TbADAT2/3. TbTrm140 was tested with the catalytically dead mutant TbADAT2/3 E92A; co-incubation of TbTrm140 with this mutant still yielded an active methylase with a kinetically determined dissociation constant of 0.02 ± 0.01 µM. This value is comparable to the wild-type enzymes and again in line with the EMSA results (Figure 2.12). No detectable methylation activity was observed with the TbADAT2/3 C-terminal binding-impaired deletion mutant, even at an upper concentration range for an extended incubation time (Figure 2.13A). Similarly, no methylation activity was detected with the TbTrm140 binding mutant, ∆S2-G17, in the presence of the wild-type TbADAT2/3 (Figure 2.13B). This is expected since the

Trm140 tRNA-binding mutant is unable to bind the tRNA; therefore, it cannot methylate it. These results support our findings by EMSA and underscores the importance of tRNA binding domains as a driving force for synergy, and consequently, for m3C formation.

Figure 2.12 Kinetic determination of dissociation constant of TbTrm140 to tRNAThr in the presence of TbADAT2/3 catalytic mutant, E92A.

Thr (A) Single turnover assays of TbTrm140 to tRNA CGU in the presence of TbADAT2/3 catalytic mutant, E92A were performed as described in the materials and methods. The fraction of methylated cytosine 32 was measured for each TbTrm140 protein concentration ranging from 10nM to 2500nM as shown in the graph. The methylated fraction was plotted as a function of time and fit to a single exponential curve [f = a(1 - e-kt)], where f represents methylated cytosine formed, a denotes methylated cytosine produced at the end point of the reaction, k signifies kobs and t is time. (B) The resulting kobs values were plotted against the concentration of TbTrm140 and fit to a single ligand binding isotherm. The Kdapp was then determined by nonlinear regression using Sigmaplot. Each graph represents at least 5 independent replicates. (C) Representative 1- Dimensional Thin Layer Chromatography (TLC) activity assay for m3C Methylation assay with TbTrm140 in the presence of TbADAT2/3 catalytic mutant, E92A. Samples were taken at increasing timepoints as shown in panel A.

Figure 2.13 Activity assays with mutant enzymes.

Representative 1-Dimensional Thin Layer Chromatography (TLC) activity assay for (A) m3C Methylation assay with TbTrm140 in the presence of TbADAT2/3 C-terminal deletion mutant at 1µM of each enzyme with samples taken at timepoints 1, 2, 6, and 12 hours. The – symbol denotes a no enzyme control while the + signifies a positive control using wild-type TbTrm140 and TbADAT2/3 enzymes. (B) m3C Methylation assay with TbTrm140 binding mutant, ∆S2- G17, in the presence of TbTrm140 at the 1uM of each enzyme with samples taken at timepoints 1, 2, 6, and 12 hours. The – symbol denotes a no enzyme control while the + signifies a positive control using wild-type TbTrm140 and TbADAT2/3 enzymes.

2.3 Discussion

Enzymes interact with partner proteins that may alter their specificity, activity,

and affinity for a given substrate; tRNA modification enzymes are no exception, with

methyltransferases being particularly prone to subunit recruitment [69,75,120,135]. The

S. cerevisiae 2’O-methyltransferase, Trm7 for example, interacts with two different

proteins, Trm732 and Trm734, to modulate its specificity for positions C32 and G34 of the

2 same tRNA anticodon [69,75,118,120,135,154–156]. Similarly, the m G10

5 methyltransferase, Trm11, and the mcm U34 methyltransferase, Trm9, both utilize the

protein Trm112 to direct substrate specificity [69,75,120,123,135,157–161]. Other

enzymes require both subunits for tRNA binding, as is the case with Trm6 and Trm61,

although Trm6 is the catalytic component [69,75,120,127,128,135,162]. In yet another

7 instance, the enzymes Trm8 and Trm82, which methylate tRNA to form m G at position

46, are both required to form an active enzyme complex [69,75,120,129,135,163,164]

We previously demonstrated that two different enzymes, a methyltransferase and

a deaminase, converge on a single anticodon loop nucleotide position to catalyze

formation of m3C and m3U at position 32 of tRNAThr. Unlike many of the aforementioned

multiprotein enzymes, we show here that TbADAT2/3 and TbTrm140 are both capable

of binding tRNA directly; however, binding individually is non-productive, leading to neither methylation nor deamination of position 32 of tRNAs; their intended target. The question then remains as to why these enzymes require each other for function. One possibility could be the increase in binding affinity observed upon addition of both binding partners; this can be attributed to several potential factors. First, it could be a

result of co-activation upon protein complex formation prior to substrate binding,

whereby complex formation leads to a conformational change that then makes the enzymes poised for activity. However, were this the case, it is expected that complex formation alone would be sufficient for activity, even if one of the partners in the complex is unable to bind tRNA, which goes against the observations presented here. The answer instead may lie in how they bind tRNA, where a rearrangement of the enzymes and/or the tRNA could possibly facilitate binding synergy and stimulate enzyme activity.

Most enzymes and enzyme complexes do not bind their substrate as rigid entities, but instead display flexibility upon binding. Binding between protein and RNA is often achieved by an induced-fit mechanism [165–167]. The idea of induced-fit was first used to explain how enzymes in an inactive state become catalytically active upon substrate binding [166,168]. This mode of binding generally utilizes the favorable binding energy to drive entropically unfavorable conformational changes made by either the protein, the

RNA, or both [165–167]. These conformational changes can in turn govern specificity and affinity which often go hand in hand [165,166,169].

There are many cases where recognition of substrate tRNA, particularly by tRNA methyltransferases involves multiple steps, including initial binding and induced-fit processes [69]. The bacterial 2’-O-methyltransferase enzyme TrmH, for instance, methylates G18 on several substrates depending on the organism. In Thermus

thermophilus, the enzyme acts on all tRNAs with G18 whereas it methylates only a subset

of tRNAs in E.coli [68]. It was shown that the regions in the N- and C-terminal ends of

TrmH support initial tRNA binding while the active-site residues participate in substrate

discrimination [170]. Therefore, the enzyme undergoes two steps, the initial tRNA

binding, facilitated by the ends, followed by induced fit, enabled by the active site

[170,171]. Similar to TrmH, the eukaryotic ADAT2/3 deaminase contains a binding domain away from the active site, which permits accommodation of multiple different

Arg substrates, while the bacterial deaminase edits a single A34-containing tRNA substrate

and recognizes a specific sequence in the anticodon loop with high affinity. The evolution

of substrate specificity was proposed whereby the ADAT2/3 enzyme acquired a

“general” tRNA-binding domain away from its active site, which coupled with active site

structural relaxation over time, may have facilitated the accommodation of multiple

different substrates. Thus, together, TbTrm140 and TbADAT2/3 may undergo multiple

binding steps to bind and accommodate their tRNA substrates.

The mode of enzymatic activation described here differs from m3C formation at

position 32 in other systems. The fission yeast, Schizosaccharomyces pombe, relies on

two Trm140-related homologs for m3C methylation, SpTrm140 specific for tRNAThr and

SpTrm141 specific for tRNASer [113,152,172]. It is unclear whether SpTrm140 and

SpTrm141 function as a multi-subunit complex; however, upon deletion of either gene individually, m3C levels decreased in all tRNA substrates, suggesting a connection between these methylation events [113]. Interestingly, T. brucei also harbors two

homologs of TbTrm140 as well. As of yet, only TbTrm140 mentioned here methylates

tRNA whereas the other homolog, TbMTase37, performs an unrelated function in

ribosomal RNA biogenesis [173]. Other metazoans and fungi contain multiple homologs

as well, some with m3C found in an additional substrate, tRNAArg, and present at novel

sites of tRNASer [113,174]. Instead of employing two methyltransferase homologs, the

ScTrm140 enzyme exploits differences in sequence elements, anticodon loop modifications, and utilizes seryl-tRNA synthetase, Ses1, to modulate specificity [145].

These observations are in line with our data showing that TbADAT2/3 is required for

methylation by TbTrm140 and that tRNA-binding plays a major role in T. brucei as an activity determinant [116].

In addition to the functional necessity for the binding strategies described here, binding synergy could also provide a competitive advantage over other tRNA modification enzymes in the nucleus. This would be particularly important for

TbADAT2/3, as the bulk of the enzyme localizes to the cytoplasm and is in low abundance in the nucleus where C-to-U editing takes place. One could also envision the use of binding synergy for regulatory purposes. For instance, we previously showed that complex formation between these two enzymes prevents TbADAT2/3 from mutagenizing the genome [116]. The increase in binding affinity for tRNA offers a possible explanation for how the complex may be sequestered away on the tRNA, precluding interaction and potentially rampant deamination of the genome by TbADAT2/3.

In the current chapter, we have presented a rationale for the observed co-

dependence of TbADAT2/3 and TbTrm140 based on the observed synergistic increase in

substrate binding affinity when combined. Thus, binding synergy is an important

component of enzyme co-activation. Interestingly, the presence of Trm140 in vitro does not affect the A-to-I activity of TbADAT2/3 at position 34. This is not entirely unexpected as the localization of these activities are restricted to separate compartments

71 where A-to-I occurs in the cytoplasm and methylation in the nucleus. The delicate interplay between these seemingly separate editing and methylation events represent an example of the intricacy of tRNA modification pathways and provide a potentially new layer to how modification cascades could be enacted and likely regulated.

Chapter 3 : Substrate specificity determinants of the Trypanosoma brucei co-

dependent tRNA editing deaminase and m3C methyltransferase

3.1 Introduction

Due to the high abundance and diversity of tRNA modifications, modification enzymes have evolved various ways of catalyzing their specific reactions, including different modes of substrate recognition. Selecting the correct substrate often involves recognition of local secondary structural features, the global L-shaped tertiary fold of the tRNA, or specific sequences within the tRNA. While several modification enzymes rely on tRNA structure, many modified nucleotides themselves influence the folding and structure of tRNA by either enhancing structural stability, flexibility, or both. Some modifications modulate structure by increasing flexibility of the ribose sugar favoring a relaxed structure [77]. Other modifications enhance rigidity by creating steric hindrance between the modified base and 2’ hydroxyl of the sugar, restricting the sugar to a primarily C-3’-endo conformation [77]. For instance, tRNA from hyperthermophilic organisms such as Thermus thermophilus and Pyrococcus furiosus contain the modification 2-thiouribothymidine (s2T) within the TψC loop, which provides local structural stability that helps preserve the tRNA tertiary structure at high temperatures

[175–178]. Similarly, methylations commonly affect the folding and stability of tRNA by preventing certain base pairings and/or disfavoring incorrect folding [175]. Several 73

methylations, including m1A, m1G, and m2,2G induce folding of mitochondrial tRNAs

[134,175,179]. In human mt-tRNALys, m1A at position 9 prevents base pairing between

A9 and U64, allowing the tRNA to fold into an appropriate tertiary structure

[134,175,179].

Despite the established role of certain modifications in tRNA structure, in vitro tRNA modification experiments are often performed with in vitro synthesized tRNA substrates lacking modifications found in endogenous tRNA. Although synthetic tRNA folds into the overall tertiary structure sans modifications, subtle aspects of local

structure, which may help distinguish one tRNA from another, are affected in an

unmodified transcript [134,180]. For example, the anticodon loop is dynamic whereby its

structure appears to be dependent on modification status. In the anticodon loop of most

tRNAs, positions 34 and 37 are typically post-transcriptionally modified with well-

studied impacts of modifications on tRNA function covered extensively elsewhere [181–

183]. NMR studies of eukaryotic tRNALys and tRNAPhe and five E.coli tRNAs (Arg, Leu,

Tyr, Pro, and Phe) showed that 1-methylguanosine (m1G), N6-isopentyladenosine (i6A),

and N6-threonylcarbamoyladenosine (t6A) stabilize the 3’ side of the anticodon loop

through stacking interactions and maintain the canonical U-turn with an open loop

structure by abolishing cross-loop interactions [184–189]. It was proposed that these

modifications decrease the conformational space sampled by the anticodon loop shifting

the structure towards the canonical anticodon loop form [186]. The U-turn is a structural

motif first observed in the tRNAPhe crystal structure and is characterized by a 7-

nucleotide loop with a reversed phosphate backbone usually 3’ of a uridine, hence its

name [190,191]. In an unmodified tRNA, the anticodon collapses into a structure which adopts a 3-nucleotide loop rather than the characteristic 7-nucleotide open loop of the U- turn. This was shown in unmodified E. coli tRNAPhe, which formed base pairs between

U32-A38 and U33-A37 leading to a 3-nucleotide loop [186,189]. The defined U-turn

structure of the anticodon is important for mRNA decoding and interactions with the

ribosome [192]. Improper formation of the anticodon loop could also prevent tRNA

modification enzymes which target the anticodon loop from acting on the tRNA [134].

In chapter 2, I showed that binding synergy acts as a critical step for enzyme coactivation involving a dually functional protein complex between the TbADAT2/3 deaminase and the TbTrm140 methylase. In this work, we explored the contribution of

TbADAT2/3 and TbTrm140 to substrate recognition regarding their activities at position

32. Specifically, we utilized synthetic portions of the tRNA to analyze where each enzyme binds, providing insight into where each protein contacts the tRNA and validating synergistic effects seen with full-length tRNA. We examined the in vitro

binding of each enzyme on tRNA species that do not act as substrates in vivo or in vitro.

Additionally, we investigated other anticodon loop modifications for their impacts on

methylation at position 32. The outcomes presented show that the enzymes can bind non-

substrate tRNAs, but cannot modify them, recapitulating what is seen in vivo. However,

the results presented here concerning the influence of other anticodon loop modifications

thus far differ from what was found in S. cerevisiae. Overall, this information contributes

to the knowledge of substrate selectivity of TbTrm140 and TbADAT2/3 and presents

disparities in the approaches taken by the same family of enzymes from different

organisms.

3.2 Results

3.2.1 TbADAT2/3 and TbTrm140 contact distinct sites on the tRNAThr

Thr A series of minimal substrates of tRNA CGU were synthesized to test the ability

of TbADAT2/3 and TbTrm140 enzymes to bind minimal substrates representative of

different domains of a tRNA (Figure 3.1). Notably, mini-substrates were previously used

to study the homologous deaminase from bacteria and a related methylase from archaea

[46,147]. Here, minisubstrates were used to explore whether or not the synergistic effect

of the combined proteins relies solely on binding of a full-length tRNA [193]. We

reasoned that if TbTrm140 can bind to a minisubstrate on its own, while TbADAT2/3

cannot, we can further dissect whether protein-protein interaction alone is sufficient for

binding synergy. An earlier study by Wolf et. al. 2002 showed that E. coli ADATa

deaminates minisubstrate tRNAArg comprised solely of the anticodon stem loop [194].

However, S. cerevisiae Tad2/3 requires the full tRNA substrate for activity, which we

found to be true for TbADAT2/3 as well (data not shown) [68,138,139,146]. For this

reason, we surmised that it was possible that TbADAT2/3 does not bind minisubstrates

and we could use this approach to assess synergism. Although secondary and tertiary

structures have not been directly examined in this case, other studies demonstrated that the anticodon loop and other small tRNA-derived substrates from various tRNA species fold into stem loops in vitro [46,70,94,134,186].

Thr Figure 3.1 RNA minisubstrates derived from T. brucei tRNA CGU.

(A) Tertiary (left) and secondary (right) structures of full length tRNA color-coded according to characteristic structures including the acceptor stem (purple), D-loop (blue), TψC-loop (green), Thr and anticodon (red). (B) Schematic illustrations of the four tRNA CGU minisubstrates that were tested in vitro, color-coded as in panel A. These do not represent actual secondary structures but are used to highlight the portion of the tRNA from which the different mini-substrates were derived.

Electrophoretic Mobility Shift Assays (EMSA) were used to measure binding

affinity as described previously [193]. For these assays, an increasing concentration of

recombinant TbADAT2/3 or TbTrm140 was incubated with a constant concentration of

the minisubstrate and subsequently analyzed by EMSA. Once binding was determined for

the individual enzymes, both TbADAT2/3 and TbTrm140 were incubated together with

the tRNA minisubstrate to evaluate whether binding synergism occurred. TbADAT2/3

and TbTrm140, added individually as well as combined, were unable to bind to the

acceptor stem/TψC-loop minisubstrate (data not shown). By themselves, TbADAT2/3

and TbTrm140 bound to the D-loop/anticodon loop minisubstrate with an apparent

dissociation constant (Kdapp) above 1 µM (Figure 3.2A,B). The Kdapp value determined

from the binding isotherm is above the detectable limits of this assay; therefore, it is

reported here as greater than 1 µM (>1 µM). Theses affinities are at least 5-times lower

Thr than the wild-type enzymes with the full-length tRNA CGU substrate which exhibited

Kdapps of 0.21 ± 0.03 µM for TbADAT2/3 and 0.21 ± 0.05 µM for TbTrm140 [193].

Furthermore, no more than approximately 20% of the D-loop/anticodon loop substrate

was bound by TbADAT2/3. The observed binding is could be due to conformational heterogeneity where only 20% of the substrate population is in a suitable conformation for binding. Notably, the binding isotherm plotted for TbTrm140 appears to display sigmoidal properties indicative of cooperative binding to the D-loop/anticodon loop. As

TbTrm140 is known to form homodimers, this could possibly reflect cooperative binding

of the subunits [193]. When TbADAT2/3 and TbTrm140 were added together to the D-

loop/anticodon loop a Kdapp of 0.57 ± 0.12 µM was observed, indicating at least a 2-fold

increase in affinity of TbTrm140 for tRNA (Figure 3.2C). Since this affinity is

approximately equal to the sum of the individual binding affinities, we consider this additive binding. Binding analysis of the T-loop/anticodon minisubstrate with

TbADAT2/3 again yielded a Kdapp above 1 µM (Figure 3.3A). In contrast, TbTrm140

bound stably to the T-loop/anticodon minisubstrate with a Kdapp of 0.41 ± 0.12 µM

(Figure 3.3B). When TbADAT2/3 and TbTrm140 were combined in the reaction, the

binding affinity of TbTrm140 remained essentially unchanged with a Kdapp of 0.53 ± 0.16

µM (Figure 3.3C). The largest of the tRNA minisubstrates is the combination of the T-

loop/anticodon/D-loop. Incubation of TbADAT2/3 with the T-loop/anticodon/D-loop

minisubstrate resulted in a Kdapp of 0.59 ± 0.14 µM (Figure 3.4A), indicating stable

binding. TbTrm140 also bound stably to the T-loop/anticodon/D-loop mini-substrate with a Kdapp of 0.44 ± 0.14 µM (Fig. 4B). When TbADAT2/3 and TbTrm140 were added

together, the Kdapp decreased by approximately 2-fold to 0.18 ± 0.14 µM, again

suggestive of additive binding, not synergistic binding (Figure 3.4B).

Figure 3.2 TbADAT2/3 and TbTrm140 do not bind synergistically to the D- loop/anticodon minisubstrate.

EMSA of (A) TbADAT2/3, (B) TbTrm140, and (C) TbADAT2/3 and TbTrm140, to assess Thr binding with the tRNA CGU D-loop/anticodon minisubstrate. In each panel, lane 1 shows a no enzyme control reaction and a control reaction. Lanes 2-6 of panels A and B show an increasing concentration of TbADAT2/3 (0.12, 0.24, 0.48, 0.96, and 1.4 µM) and TbTrm140 (0.08, 0.16, 0.32, 0.64, and 1.12 µM), respectively. Lanes 3-7 of panel C show an increasing concentration of TbTrm140 TbTrm140 (0.08, 0.16, 0.32, 0.64, and 1.12 µM) in the presence of a constant concentration (1µM) of TbADAT2/3. The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 3 independent replicates.

Figure 3.3 TbADAT2/3 and TbTrm140 do not bind synergistically to the TψC- loop/anticodon minisubstrate.

EMSA of (A) TbADAT2/3, (B) TbTrm140, and (C) TbADAT2/3 and TbTrm140, to assess Thr binding with the tRNA CGU T-loop/anticodon minisubstrate. In each panel, lane 1 shows a no enzyme control reaction and a control reaction. Lanes 2-6 of panels A and B show an increasing concentration of TbADAT2/3 (0.12, 0.24, 0.48, 0.96, and 1.4 µM) and TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM), respectively. Lanes 3-7 of panel C show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM) in the presence of a constant concentration (1µM) of TbADAT2/3. The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 3 independent replicates.

Figure 3.4 TbADAT2/3 and TbTrm140 do not bind synergistically to the D- loop/anticodon/TψC-loop minisubstrate.

EMSA of (A) TbADAT2/3, (B) TbTrm140, and (C) TbADAT2/3 and TbTrm140, to assess Thr binding with the tRNA CGU D-loop/anticodon/T-loop minisubstrate. In each panel, lane 1 shows a no enzyme control reaction and a control reaction. Lanes 2-6 of panels A and B show an increasing concentration of TbADAT2/3 (0.12, 0.24, 0.48, 0.96, and 1.4 µM) and TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM), respectively. Lanes 3-7 of panel C show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM) in the presence of a constant concentration (590 nM) of TbADAT2/3. The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 3 independent replicates.

From this data, we found that TbTrm140 can bind to several of the small tRNA

substrates albeit with lower affinity than full-length tRNA. While TbTrm140 does not bind the acceptor stem/T-loop, it appears to bind all other substrates that include the T- loop of the tRNA where it possibly makes contacts with the tRNA. On its own,

TbADAT2/3 binds most stably to the T-loop/anticodon/D-loop, but has minimal to no

affinity for the other small tRNA substrates. This may reflect the requirement for the

tertiary structure of tRNA for TbADAT2/3 binding, as mentioned earlier in regard to

catalysis. In the cases when both enzymes are combined, the binding affinity remained

unchanged for the T-loop/anticodon or increased by about 2-fold for the D-

loop/anticodon loop and T-loop/anticodon/D-loop which is likely due to an additive

effect, meaning that the full-length tRNA is required for synergism. Notably, none of the

minisubstrates displayed comparable affinities to the full-length substrate when both

enzymes were added (Kdapp of less than 0.03 µM). These results also support the fact that

direct protein-protein interaction alone may lead to a slight increase in affinity, but it is

not sufficient to yield more robust synergistic effects seen with the full-length substrate.

This corroborates our data showing that tRNA-binding facilitated by distal domains of

TbADAT2/3 and TbTrm140 is critical for binding synergism [193].

3.2.2 Binding synergism is not limited to substrate tRNA

Two different non-substrate tRNA species were tested to determine whether

binding synergism is restricted to substrate tRNA. tRNAGlu was chosen because it

3 3 contains C32 comparable to a substrate tRNA (Figure 3.5), but neither m C nor m U

modifications were detected in tRNAGlu in vivo by mass spectrometry nor in vitro via

reconstitution experiments with TbADAT2/3 and TbTrm140 (data not shown).

Moreover, it is not a known substrate for either methylation or deamination in any

Val organism to date. tRNA , was also tested as it contains a U32 which expectedly

precludes deamination, but not necessarily methylation (Figure 3.5). However, like

tRNAGlu, neither methylation nor deamination was observed at position 32 in tRNAVal in

vivo or in vitro (data not shown). We showed previously that TbADAT2/3 binds tRNAVal

with an affinity similar to the tRNAThr substrate [148,193]. TbTrm140 also binds tRNAVal

Thr with a Kdapp of 0.29 ± 0.09 µM, comparable again to the tRNA (Figure 3.6A). When an

increasing concentration of TbTrm140 was added to a constant concentration of

Val TbADAT2/3 and tRNA , the Kdapp decreased below 0.03 µM, indicative of synergistic

binding (Figure 3.6B). TbADAT2/3 and TbTrm140 individually bound stably to

Glu tRNA with Kdapps of 0.21 ± 0.03 µM and 0.21 ± 0.05 µM, respectively (Figure

3.7A,B). Synergism was observed when both enzymes were added into the reaction with

tRNAGlu regardless of which protein partner was titrated into the sample (Figure

3.8A,B). Altogether, these results demonstrate that binding synergism is not limited to substrate tRNA. Furthermore, solely the presence of C32 in tRNA is not sufficient for

methylation and deamination activity in vitro, indicating there must be additional

specificity determinants or anti-determinants for activity.

3 3 Figure 3.5 Cloverleaf secondary structures of m C/m U32 substrates versus non- substrates.

3 3 Thr Glu Structure of m C/m U32 substrate tRNA CGU (left), non-substrate tRNA UUC (middle), and non- Val substrate tRNA AAC (right).

Val Figure 3.6 TbTrm140 binds nonsubstrate tRNA AAC synergistically.

Val (A) EMSA of TbTrm140 binding to tRNA AAC. Lane 1 is a no enzyme control reaction. Lanes 2-6 show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, Val respectively). (B) EMSA of TbTrm140 with tRNA AAC in the presence of ADAT2/3 (210 nM). Lanes 1 and 2 show a no enzyme control reaction and control reaction with no TbTrm140 added, respectively. Lanes 3-7 show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 3 independent replicates.

Glu Figure 3.7 TbADAT2/3 and Trm140 bind stably to nonsubstrate tRNA UUC.

Glu (A) EMSA of TbADAT2/3 incubated with tRNA UUC. Lane 1 is a no enzyme control reaction in both panels. Lanes 2-6 show an increasing concentration of TbADAT2/3 (0.06, 0.12, 0.24, Glu 0.48, and 0.7 µM, respectively). (B) EMSA of TbTrm140 with tRNA UUC. Lanes 2-6 show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 3 independent replicates.

Glu Figure 3.8 TbADAT2/3 and TbTrm140 bind tRNA nonsubstrate tRNA UUC synergistically.

Glu (A) EMSA of TbADAT2/3 incubated with tRNA UUC in the presence of a constant concentration (210 nM) of TbTrm140. Lanes 1 and 2 show a no enzyme control reaction and a control reaction with no TbTrm140 added, respectively. Lanes 3-7 show an increasing concentration of TbADAT2/3 (0.06, 0.12, 0.24, 0.48, and 0.7 µM, respectively). (B) EMSA of TbTrm140 Glu incubated with tRNA UUC in the presence of a constant concentration (210 nM) of TbADAT2/3. Lanes 1 and 2 show a no enzyme control reaction and control reaction with no TbADAT2/3 added, respectively. Lanes 3-7 show an increasing concentration of TbTrm140 (0.04, 0.08, 0.16, 0.32, and 0.56 µM, respectively). The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 3 independent replicates.

3.2.3 Methylation at position 32 is not influenced by t6A or m1G at position 37 in

vivo

Since the in vitro experiments utilized synthetic tRNAs which lack modifications,

it is possible that another modification impacts the modification levels at position 32 in

some manner. For instance, an open anticodon loop structure could be necessary for

substrate recognition and/or catalysis by TbADAT2/3 or TbTrm140 and this is

6 1 compromised when certain modifications, such as t A37 or m G37, are missing.

TbADAT2/3 acts on eight substrate tRNAs in T. brucei, Val, Ser, Arg, Thr, Ile, Leu, Pro,

and Ala, among these, Thr, Ser, and Arg, are known substrates for TbTrm140

methylation at position 32. Therefore, we examined the anticodon loop modifications

1 6 m G37 and t A37 for their impact on methylation at position 32 of two shared substrates,

tRNAThr and tRNAArg.

6 3 To analyze the influence of t A37 on the m C/U modification at position 32 we

first identified the modification enzyme in T. brucei. Two proteins, Kae1 (kinase

associated endopeptidase) and Sua5 (suppressor of upstream ATG) were identified by

comparative genomic analysis and validated by genetic experiments in S. cerevisiae to be

6 responsible for t A37 modification [195,196]. A putative homolog of the ScKae1 protein

(YKR038C) was identified in T. brucei (Tb927.7.6470) upon a BLAST search of the T.

brucei genome database, TryTrypDB [197]. The putative TbKae1 contains all the

conserved residues important for function in other eukaryotes, such as the ATP-binding and metal ion-binding residues [198]. A portion of the putative TbKae1 coding region was placed into the p2T7-177 RNAi vector containing a tetracycline (TET) inducible

promoter and introduced into procyclic T. brucei strain 29-13 by electroporation [199].

Cell growth was monitored for the RNAi induced (+TET), RNAi uninduced (-TET), and

wild-type cell lines. A growth phenotype was observed after 6 days post-induction

compared to the wild-type and uninduced cells (Figure 3.9). The cells from the

uninduced, induced, and wild-type were collected at day 6 and total RNA was purified.

Thr The presence of methylation at position 32 of tRNA AGU was detected using primer

Thr extension analysis with an oligonucleotide primer specific for the tRNA AGU T-loop.

3 Since the m C/U32 methylation prevents Watson-Crick base pairing, it generates a

“strong” stop one nucleotide 3’ of the modified position. The intensity of the strong stops

observed in the RNAi induced and uninduced samples were comparable, suggesting that

the down-regulation of the putative TbKae1 enzyme by RNAi did not affect the level of

6 methylation at position 32 (Figure 3.10). Therefore, it is unlikely t A37 affects the modification status of position 32 nor acts as a binding or activity determinant for

TbADAT2/3 and TbTrm140.

Figure 3.9 Putative TbKae1 is important for growth T. brucei.

Growth curve of the wild-type, uninduced (-TET), and induced (+TET) cell lines. The uninduced and induced cell lines are T. brucei 29-13 transformed with the p2T7-177 vector containing a 6 fragment of the coding region of the putative TbKae1 subunit of the t A37 modification enzyme. RNAi knockdown was induced by the addition of tetracycline to the media. RNA was collected for analysis at 6 days post-induction as indicated by the arrow.

Figure 3.10 Knockdown of putative TbKae1 does not alter methylation levels at position 32 of tRNAThr.

Schematic of the primer extension assay for detection of methylation at position 32 of Thr tRNA AGU (left). The long black arrow indicates the radioactive probe used for primer extension which anneals to the T-loop 8 nucleotides away from the methylated position. The modification 3 6 sites for m C/U32 and t A37 are labeled along with the primer extension stop. A strong stop at primer +8 is expected if the methylation is present as it blocks canonical base pairing during reverse transcription. Primer extension analysis (right) of RNA collected from TbKae1 RNAi induced (+) and uninduced (-) cell lines. The primer lane refers to a sample where no template RNA was added into the reaction. The percent readthrough displayed under the gel image was determined by quantifying the signal above the strong stop in an entire lane (marked as read through), dividing by the entire lane above the primer, and multiplying by 100.

Previously developed TbTrm5 RNAi lines were used to evaluate the influence of

1 Arg m G37 on methylation at position 32 of tRNA UCG in vivo [182]. Cell growth was

monitored and cells displayed a growth phenotype after 8 days post-induction in

accordance with our previous data [182]. The total RNA was extracted from the uninduced and induced cell lines at 8 days post-induction and similar primer extension

Arg assays were performed with an oligonucleotide primer specific for the tRNA UCG T-

3 1 loop (Figure 3.11). Unfortunately, like m C/U32, the m G37 modification blocks Watson-

Crick base pairing, preventing the extended primer from reaching position 32. Therefore,

1 the effect of m G37 on modification and editing at position 32 remains inconclusive. This

can be resolved in future experiments by applying a different technique to monitor

methylation such as 2D-TLC or northern blotting [116,182].

1 Figure 3.11 m G37 prevents detection of methylation at position 32 by primer extension.

Arg Schematic of the primer extension assay for detection of methylation of tRNA UCG (left). The long black arrow indicates the radioactive probe used for primer extension which anneals to the 3 T-loop 8 nucleotides away from the methylated position 32. The modification sites for m C/U32 1 and m G37 are labeled along with the primer extension stops. A strong stop at primer +8 is expected if the methylation is present as it blocks canonical base pairing during reverse transcription. Primer extension analysis (right) of RNA collected from TbKae1 RNAi induced (+) and uninduced (-) cell lines. The primer lane refers to a sample where no template RNA was 3 1 added into the reaction. Read through for both the strong stop at +8 (m C/U32) and +3 (m G37) are shown. The percent readthrough displayed under the gel image was determined by quantifying the signal above the strong stop in an entire lane, dividing by the entire lane above the primer, and multiplying by 100.

3.3 Discussion

Binding analysis with minisubstrates derived from different portions of a tRNA,

revealed potential binding contacts of the TbADAT2/3 and TbTRM140 enzymes. While

tRNA methyltransferases generally recognize local structure surrounding the target

nucleotide, the ADAT2/3 deaminase requires the full length tRNA, suggesting its dependence on the overall tertiary structure [37,68,69,138,139]. In this study, the T-

loop/acceptor stem mini-substrate did not support binding of TbADAT2/3 and

TbTrm140. This is consistent with the fact that the target position resides in the anticodon

loop, so the enzymes may not necessarily contact the acceptor stem. On the other hand,

we observed that the binding affinity is lower in the T-loop/anticodon/D-loop mini-

substrate compared to the full-length substrate. Since this substrate has all the parts of the

tRNA aside from the acceptor stem, there may still be important contacts on the acceptor

stem that could anchor the enzymes to the tRNA which may allow for docking of the

substrate into the active site. Another explanation could simply be that the tRNA is not

completely folded into its canonical tertiary structure which may lower the affinity of the

enzymes for the tRNA.

The binding synergism observed does not appear to be limited to substrate

tRNAs, bringing up the issue of how the substrate tRNAs are discriminated against non-

substrate tRNA. Unlike ADAT2/3, which is known to deaminate essentially all A34-

containing tRNAs, Trm140 cannot methylate all C32-containing tRNA [152,172]. In a similar case to Trm140, the bacterial TrmD methyltransferase binds all tRNA species including those lacking the target nucleotide G37.; however, it only methylates its

95 substrate tRNAs containing G at position 36 and 37. Therefore, it was suggested that

TrmD requires full-length tRNA for binding and solely forms tight contacts within the active site with cognate tRNA containing both G36 and G37 [86]. Similarly, the methyltransferase RumA and TrmA bind to their tRNA substrates which refold within the active site [105]. In the case of TrmA, the enzyme uses an induced-fit mechanism, making critical contacts with the T-loop, which is re-ordered upon binding. According to

NMR structural analysis, the bound T-loop conformation differs significantly form that seen in the unbound tRNAPhe [105]. Like RumA and TrmA, rather than binding inflexibly to tRNA, many enzyme complexes remodel RNA upon binding in order to increase the affinity or provide access to the target nucleotide [200]. The Trm7 methyltransferase requires the interaction of multiple protein factors to modify its various target

Phe modification sites at C32 and G34 of tRNA [118]. These factors are supposedly necessary to access these positions which are one nucleotide apart, where C32 is buried while G34 is solvent exposed [118]. The ArcTGT-tRNA and RlmD-rRNA complexes are also examples where the RNA is remodeled upon binding [200]. The RNA binding PUA domain of the ArcTGT is responsible for initial binding to tRNA, but is not required for the modification archaeosine at position G15. This result suggests that the PUA domain is not involved in discrimination of the modification site [170]. Perhaps analogous to these enzymes, TbTrm140 and TbADAT2/3 can bind diverse tRNA species by surveying the overall structure and then may undergo a docking mechanism which could disorder the tRNA to allow for catalysis.

Interestingly, ADAT2/3 can deaminate any tRNA substrate with an A34; however,

this appears untrue for the C-to-U deamination activity at C32. As the reaction occurs sequentially, where m3C must be formed prior to deamination, it is possible that the

substrate specificity is forcibly dictated by the methyltransferase activity rather than

selection by the deaminase [116]. This is plausible since there is evidence that some

methyltransferases select substrates from a pool of possible substrates with the same

1 target nucleotide. For instance, the m G9 methyltransferase, Trm10 methylates 13 out of

24 species with G9 without obvious recognition elements [75,145,201]. Evidence for several modes of substrate recognition by Trm140 have been discovered. In S. pombe, two methyltransferase paralogs, Trm140 and Trm141, are required for methylation.

6 3 Ser Within this same organism, i A37 is added before m C32 can be formed in tRNA [113].

6 In S. cerevisiae, Trm140 identifies the sequence identity element G35-U36-t A37 in

tRNAThr substrates. On the other hand, tRNASer substrate specificity relies on an

6 interaction with seryl-tRNA synthetase and its unique variable loop, in addition to t A37

6 6 6 and i A37 [145]. In this work, while the requirement for t A37 was assessed, i A37 was not

tested in this set of experiments, so it could possibly influence these activities and may be

worth testing in the future. It would be interesting to see if the same is true for T. brucei

and if it is enough to discriminate between substrate and non-substrate tRNA in vivo.

Chapter 4 : Analysis of Trypanosoma brucei editing deaminase subunit

orientation and consequences for catalysis and tRNA-binding

4.1 Introduction

Deaminase enzymes that act on DNA and/or RNA are categorized into two major families based on their activity: adenosine deaminases acting on RNA (ADARS) and polynucleotide cytidine deaminases (CDA); all are members of the CDA superfamily.

The active site of the CDA superfamily is characterized by two conserved sequences,

H(C)XE (where X denotes any amino acid), containing the key catalytic glutamate residue for catalysis and two critical cysteines that together with the aforementioned histidine form their characteristic zinc-binding motif. The Zn2+ cofactor is required to coordinate a water molecule, which once activated acts as a nucleophile in the deamination reaction. Meanwhile, the catalytic glutamate acts as a proton shuttle, delivering the hydrogen from water to the ammonia leaving group. In the ADAT2/3 heterodimer, ADAT2 harbors the conserved catalytic site while ADAT3 contains a proposed “pseudo-active site” (HPV in T. brucei and Saccharomyces cerevisiae) and was initially thought to be exclusively structural. However, ICP (Inductively Coupled Plasma-

Mass Spectrometry) analysis of TbADAT3 revealed that it also participates in zinc coordination, indicating that both subunits contribute to catalysis [202]. In addition to the active site, ADAT2 contains a C-terminal extension with string of lysine and arginine 98

residues, termed the KR-domain. Rich with positive basic residues, the KR-domain likely

interacts with the negatively charged tRNA phosphate backbone and is chiefly

responsible for tRNA binding [148]. Consistent with previous models, the KR-domain

presumably allows ADAT2/3 to act on a larger collection of substrates (7 to 8 different

tRNAs) in comparison to the bacterial ADATa which has only one substrate, tRNAArg

[77,119,148,203]. While the bacterial ADATa structure has been solved, high resolution

structural information is currently unavailable for the eukaryotic deaminase enzyme

[204].

Our lab has published three-dimensional structural models of the ADAT2/3 heterodimer [204]. The models were generated using the x-ray crystal structure of the

homologous enzyme, ADATa (TadA), from Staphylococcus aureus (PDB code 2B3J) as

a modeling template [204]. Two versions of the model were created using the

Frankenstein’s Monster method [204,205]. One model is based solely on the SaADATa homodimer and therefore considered the “unswapped” orientation. Like the SaADATa, each subunit coordinates their own zinc (intra-subunit zinc coordination). The alternative

“swapped” model was produced by exchanging the HXE and PCXXC motifs of ADAT2

and ADAT3 [204]. In this model, the enzyme coordinates zinc between the two subunits

(inter-subunit zinc coordination). There are few structural differences between the two

models when they are superimposed and they have nearly identical active sites [202].

Therefore, the models by themselves appear to equally support the possibilities of the

unswapped versus swapped orientation of the subunits. However, taking into account the

biochemical data showing that a single zinc was sufficient to for an active heterodimer,

the inter-subunit model is favored [204].

To validate the subunit arrangement and ultimately address how ADAT2/3 binds and deaminates tRNA we have sought to crystallize the TbADAT2/3 heterodimer with

the help of our collaborators. Therefore, in the following chapter, we have attempted to

create an enzyme variant that could be more easily crystallized by removing the

seemingly disordered N-terminal region of TbADAT3. We have also submitted

TbADAT2/3 for hydrogen-deuterium exchange to gain support for the computational model. We plan to use this structural information in the context of TbTrm140-

TbADAT2/3 complex to analyze the interaction between these two enzymes and produce a complete picture of this interdependent pathway.

4.2 Results

4.2.1 Revised structural models of TbADAT2/3

We previously proposed two opposing models of the TbADAT2/3 heterodimer arrangement with equal probability of supporting the bona fide structure. We obtained an x-ray crystal structure of the TbADAT2 homodimer at 2Å resolution from our collaborators in Dr. Nina Papavasiliou’s laboratory at The Rockefeller University.

Interestingly, TbADAT2 forms homodimers if recombinantly expressed in E. coli.

However, these dimers are bind weakly, if at all, to tRNA and are therefore unable to deaminate tRNA in vitro, even though they share structural features and conserved catalytic residues with the bacterial ADATa homodimeric enzyme. Whether these

TbADAT2 homodimers are biologically relevant within T. brucei is yet to be determined.

100

To resolve the orientation of TbADAT2/3, new structural models were generated in a collaboration with the Dr. Janusz Bujnicki laboratory at the International Institute of

Molecular and Cell Biology (Figure 4.1). The structural information acquired from the

TbADAT2 crystal structure served as a template for the new models. Due to the conserved sequences within their catalytic core, it was suggested that A-to-I deaminases arose from a gene duplication of a precursor CDA. To this end, ADAT3 is thought to have arisen from a gene duplication event of ADAT2 after the divergence of Eukarya and

Bacteria [45,47,206,207]. In the process of evolution, the active site glutamate was replaced by another amino acid in ADAT3. These models were therefore created by homology modeling guided by the presumably conserved core structures of TbADAT2 and TbADAT3 resulting from this proposed gene duplication event [204,208]. These models were generated using computational methods described in the introduction for the previous models to produce an unswapped and swapped orientation [202].

101

Figure 4.1 Revised structural models based on TbADAT2 crystal structure. The TbADAT2 homodimer was used as a template to build two TbADAT2/3 heterodimer models using the Frankenstein’s monster method [202,205]. The Unswapped model (left) represents the heterodimer in which both subunits coordinate one zinc (intra-subunit coordination). The swapped model (right) represents an alternative model where the active sites of ADAT2 and ADAT3 were exchanged, supporting the possibility of inter- subunit zinc coordination. The subunits are indicated by different colors with ADAT2 in blue and ADAT3 in purple. The N-terminus (NH2) and C-terminus (COOH) of each enzyme are indicated in colors corresponding to the subunit and zinc cofactors are shown as teal spheres.

102

4.2.2 Mapping of TbADAT2/3 buried residues by hydrogen-deuterium exchange

With the aim of mapping the orientation and possible inter-subunit contacts of

TbADAT2/3 we performed hydrogen/deuterium exchange mass spectrometry (HDX-

MS). These experiments were performed with the collaboration of the Dr. Otavio

Thiemann laboratory at the University of São Paulo. For this technique, recombinantly expressed TbADAT2/3 was submerged in 90% deuterated water (D2O). The samples

were quenched with acid and by freezing at various timepoints to obtain a range of

hydrogen exchange. The enzyme was digested with pepsin and subject to triple-

quadrupole mass spectrometry followed by fragmentation by electron-spray ionization

(ESI). Regions of the protein exposed to the D2O readily exchanged hydrogen, while

those buried within the protein or were at the subunit interface would not. The percentage

of hydrogen to deuterium exchanged was highlighted in the protein sequences (Figure

4.2). This data resulted in greater than 75% amino acid coverage for each subunit.

103

Figure 4.2 Exposed and buried residues of TbADAT2/3 identified by hydrogen- deuterium exchange. The percent of deuterium exchange was mapped onto the protein sequences for TbADAT2 (top) and TbADAT3 (bottom) after 5 minutes and 30 minutes of exposure to D2O. The color code for percent exchange is indicated in the legend on the right.

To map the HDX data onto the new structural models, properties of solvent

accessibility predicted for the TbADAT2/3 heterodimer were first calculated using the

classical Shrake-Rupley algorithm [209–211]. This was then compared with a plot of the

HDX data (Figure 4.3). From this data, the molecular models are in good agreement

between the predicted solvent-accessible surface area (SASA) values in both models.

However, the current HDX results are too low resolution in that they do not correlate with the solvation results to provide a confident discrimination between the two models; 104 therefore, further structural analysis is necessary to discern the orientation of the heterodimer (Figure 4.4).

Figure 4.3 Solvent accessibility and HDX values for unswapped and swapped models. Solvent-accessible surface area (SASA) values across the protein sequence of TbADAT2 (blue) and TbADAT3 (orange) were calculated using the Shrake-Rupley algorithm method for (A) swapped model and (B) unswapped model. (C) The predicted consensus solvation for TbADAT2/3. (D) The predicted disorder for TbADAT2/3. (D) Solvent accessibility from HDX analysis.

105

Figure 4.4 HDX values mapped onto revised homology models. The homology models as shown in figure 4.1 colored according to the HDX data. The color code for percent exchange is indicated in the legend on the right. The zinc cofactors are shown as teal spheres.

4.2.3 Deletion of possible flexible regions of ADAT3 for x-ray crystallization

Since the bacterial ADATa enzyme is a homodimer, it does not contain all the

residues present in TbADAT2/3. Specifically, the N-terminal domain of TbADAT3 was added to the computational models using an additional program called ROSETTA [204].

Although the molecular models solely predict atomic details, they are accurate at the level of inter-residue contacts within individual domains and can be used to analyze three-dimensional interactions [204]. The N-terminal domain structure of TbADAT3 determined from these models is predicted as a highly flexible, disordered region of the enzyme. Therefore, to create a more compact structure that may facilitate x-ray

crystallography, several N-terminal deletion mutants were created with an increasing

number of residues deleted ranging from 10 amino acids (∆10) to 40 amino acids (∆40).

106

The goal was to create a series of TbADAT3 subunits that were more compact but still

catalytically active and able to bind tRNA when combined with TbADAT2.

Four TbADAT3 deletion mutants were recombinantly co-expressed with

TbADAT2 in E. coli and purified by IMAC (Figure 4.5). The resulting enzymes were

then subject to EMSA and activity assays to determine binding and activity. Of these enzyme variants, TbADAT3∆10, TbADAT3∆20, and TbADAT3∆30 bound to tRNAThr

and formed A-to-I at position 34 in vitro (Figure 4.6). Thus, TbADAT3∆10,

TbADAT3∆20, and TbADAT3∆30 are good candidates for crystallization.

TbADAT3∆40 did not express in E. coli even though there were no errors in the protein sequence. Perhaps the deletion was too great to produce a properly folded protein subunit and the protein was degraded before purification was completed, or the protein was potentially toxic to the cell. We submitted the candidate enzymes, TbADAT3∆10,

TbADAT3∆20, and TbADAT3∆30 to our collaborators to begin the crystallization process.

107

Figure 4.5 Western blot detection of TbADAT3 N-terminal deletion variants.

(A) Western blot detection of His-tagged TbADAT3 N-terminal deletion mutants with anti-His antibodies. The bands corresponding to each TbADAT3 mutant are labeled to the right of the blot. (B) TbADAT2, coexpressed with the TbADAT3 N-terminal deletion mutants, was detected by western blot with anti-TbADAT2 antibodies. In both panels, lanes 1-4 contain wild-type TbADAT2/3, TbADAT2/3∆10, TbADAT2/3∆20, and TbADAT2/3∆30, respectively. The migration of the molecular weight standards is indicated to the left of each panel.

108

Figure 4.6 tRNA binding and activity of TbADAT3 N-terminal deletion variants.

(A) EMSA of TbADAT2/3∆10 (left), TbADAT2/3∆20 (middle), and TbADAT2/3∆30 (right) Thr binding to tRNA CGU. Lane 1 is a no enzyme control reaction in both panels. Lanes 2-6 show an increasing concentration of TbADAT2/3 (0.06, 0.12, 0.24, 0.48, and 0.7 µM, respectively). The bottom panels show the single-ligand binding isotherms used to calculate the individual Kdapp. Each graph represents at least 3 independent replicates. (B) Representative 1-Dimensional Thin Layer Chromatography (TLC) 90-minute end-point activity assay for TbADAT3∆10 (∆10), TbADAT3∆20 (∆20), and TbADAT3∆30 (∆30). The NE and - symbol denote a no enzyme control while the + signifies a positive control using wild-type TbADAT2/3 enzyme.

109

4.3 Discussion

In this work, we obtained an x-ray crystal structure of the TbADAT2 homodimer

which was used to create two new homology models to determine the orientation of the

TbADAT2/3 heterodimer. We performed HDX to assess which residues were exposed to

the solvent versus buried, in order to resolve the dimer orientation. While the coverage

was sufficient across both subunits, the HDX results were not enough to differentiate

between the two molecular models. At present, there is no high resolution structural

information on TbADAT2/3 due to the difficulty crystallizing the enzyme. This is

presumably due to the predicted disordered N-terminal region of ADAT3. The N-

terminal deletions generated in this work will hopefully alleviate this issue in future

crystallization experiments. Ultimately, from this work we hope to further analyze the

interaction between the ADAT2/3 subunits.

Thus far, TbADAT2/3 is the only deaminase known to perform both C-to-U and

A-to-I deamination reactions [48]. This dual function provides support for the evolution from cytidine to adenosine deaminases [48]. Elucidation of the orientation of the

TbADAT2/3 deaminase would provide critical information on the catalytic properties of this unusual enzyme. Previous biochemical analysis favors the swapped orientation and inter-subunit zinc coordination model [202]. Interestingly, while the TbADAT2 homodimer is reminiscent of the active bacterial ADATa enzyme, TbADAT2 is unable to bind tRNA, yielding undetectable deamination activity. This is initially surprising as

TbADAT2 contains the C-terminal KR-domain largely responsible for tRNA-binding.

We could speculate that this “swapped” orientation could be supported if the dimerization

110

between TbADAT2/3 leads to a structural reorganization that is critical to tRNA-binding and catalysis. In this case, the unswapped orientation, which is more strictly representative of the TbADAT2 dimer structural organization, would be predicted to be unable to bind to and efficiently catalyze the tRNA. Alternatively, it could be that

TbADAT3 provides a critical structural element or tRNA-binding residues that are missing in the TbADAT2 homodimer. Additional structural and mutational analysis is necessary to address this.

In T. brucei, TbADAT2/3 is a component of unique interdependent editing and methylation pathway at position 32 of tRNA [116]. If we can generate a clear picture of the structure we can then learn how the enzymes work together to perform their activities at position 32 of tRNA. It was recently proposed that the deaminase and methyltransferase undergo enzyme coactivation to carry out their function [193]. This is facilitated by the binding synergism observed when the enzymes bind their shared tRNA substrate. Structural rearrangements likely stimulate the binding synergism detected when the two enzymes are together versus when they are alone. We hope to dissect these potential structural steps to fully understand the mechanism of enzyme coactivation.

Additionally, given that TbADAT2/3 is an essential enzyme, this structural information could potentially be used in the creation of new treatments for trypanosomiasis.

111

Chapter 5 : Concluding Remarks and Future Directions

5.1 Introduction

Until recently, the intricacy and importance of many tRNA modifications was

largely underappreciated [30,132]. This stems primarily from the fact that most tRNA

modifications are either unessential or their functions were not easily realized, partly due

to a general lack of sensitive techniques to analyze them [30,132]. The development of new approaches in areas such as genetic screens, biochemical assays, mass spectrometry

and next-generation sequencing allowed for the discovery of unsolved functions for numerous modifications [30,132,212]. Consequently, such techniques gave strength to early ideas about modifications forming networks that fine-tune structural and functional aspects of tRNAs. The number of possible interactions between once seemingly unrelated modifications has exposed the potential to uncover pathways more complex than first realized. Related to this work, we discovered an intriguing biochemical pathway involving the interaction between two apparently disparate modification enzymes. The results presented in this thesis address why these enzymes require one another for activity and explores the basis for substrate specificity, which is largely driven by tRNA binding.

These data will be valuable for future studies in tRNA modification and contribute to our knowledge of the biochemistry of tRNA modification enzymes. In all, while the work here resolved several aspects of this editing and modification pathway, there is still much 112 to uncover about these modifications and their corresponding enzymes such as the

3 specific mechanism of enzyme coactivation and the biological significance of of m C32

3 and m U32.

5.2 Mechanism of enzyme coactivation by TbTrm140 and TbADAT2/3

We showed that tRNA binding likely provides a foundation for the interdependent activities of TbTrm140 and TbADAT2/3. It is unclear, however, exactly how this happens mechanistically. With evidence showing that protein complex formation by itself does not result in synergistic binding that would result in coactivation, the likely alternative is that the enzymes undergo a conformational change upon binding the tRNA, or upon binding they distort and reshape the substrate itself. These possibilities can be addressed in several ways. High resolution structural data on TbADAT2/3, TbTrm140, and tRNA obtained from x-ray crystallography or NMR analysis would largely resolve the structural changes that may occur. As mentioned in chapter 4, we are currently working on the crystal structure of TbADAT2/3 and TbTrm140; unfortunately, these enzymes are too large to study by NMR. An alternative method is the use of ion-mobility mass spectrometry, which preserves complexes and can detect conformational changes.

Preliminary data using the latter technique revealed a major limitation in that TbADAT2 readily forms inactive homodimers, and if in extreme excess, even homotetramers. These

TbADAT2 oligomers overwhelm the signal from TbTrm140 and TbADAT3, rendering it difficult to detect the complex by this method. If the enzymes can be purified with a higher yield of TbTrm140 and TbADAT3 while preventing the oligomerization of

TbADAT2, we should be able to get a clearer picture by ion-mobility mass spectrometry.

113

Lastly, experiments using isothermal titration calorimetry could potentially reveal valuable information on the energetics of complex formation and tRNA binding. For example, an energetic penalty may be detected if a conformational change occurs or a positive change in energy could result from the increase in binding affinity observed when the enzymes are added together. The caveat of ITC is that high amounts of enzyme and tRNA are required and the interpretation of thermodynamic parameters is not always clear-cut.

5.3 Further characterization of substrate specificity of TbTrm140 and TbADAT2/3

Another major question is that of substrate specificity and particularly how the substrate tRNAs is chosen from a pool of very similar non-substrate tRNAs. The aspect of substrate specificity is not new, or trivial, as it remains a lingering issue in the field of tRNA modifications and biochemistry in general. There are several major factors that could dictate specificity, including protein-protein recruitment, specific sequences, specific structural features, and other modifications. We attempted to address several of these aspects in this work. In chapter 3, we roughly determined where on the tRNA these enzymes through binding studies with mini-substrate tRNAs. Footprinting analysis would provide further insight into precisely where the proteins are contacting the tRNA at nucleotide resolution. The analysis could be performed with the enzymes individually and together, and results could be compared to see changes in binding contacts between these two states. This would help address how tRNA binding may be altered when the enzymes are added together versus apart.

114

In vitro, the enzymes bind at least two non-substrate tRNAs with comparable affinity to substrate tRNA; however, no activity is detected in vivo or in vitro. This is not entirely unexpected since all tRNAs are structurally very similar and there is the precedent for tRNA modification enzyme binding to non-substrates. In a pool of tRNAs that look structurally identical, the issue may arise that that high affinity binding could result in sequestering the enzymes in an unproductive complex with a non-substrate. This may well be an issue in the cell; however, this may be partially overcome through competition by other tRNA-binding proteins that are specific to those tRNAs, effectively preventing TbADAT2/3 and TbTrm140 from binding.

Swapping certain nucleotides for others can abolish A-to-I activity for

TbADAT2/3, but not the E. coli deaminase, ADATa. We have yet to test these tRNAs for methylation and deamination at position 32 with TbTrm140 and TbADAT2/3. It would be interesting to see if the activities are affected by sequence changes between different substrate tRNAs. However, as they found for Trm140 from S. cerevisiae and S. pombe, it is likely a combination of different factors that culminate in specificity. The answers obtained so far, and in the future, will surely provide insight into the amazing biochemical complexity found in related enzymes from different organisms.

5.4 Specific role of m3C versus m3U in translation

There is still much to learn about the functional implications of the activities of these enzymes. Thus far, we know that the modification at position 32 influences cytoplasmic translation, but its specific function is not yet known [116]. Modifications found at the anticodon stem loop are involved in several aspects of translation including

115

efficiency, fidelity, or reading frame maintenance primarily by altering codon-anticodon

3 pairing. As methylations are often involved in stabilization of tRNA structure, m C32 and

3 m U32 may affect ordering of the anticodon loop, which in turn could influence

translation. Along these lines, 2’O-methylations of the ribose sugar, thiolations, and pseudouridylations are often found at position 32, all of which contribute to the structural integrity of the anticodon loop [134,189,213–215]. Translational fidelity could also be

affected by C32 and U32 due to the differences in base pairing with position 38 in certain

tRNAs, as alluded to in chapter 3. In E. coli, tRNAGly has three isoacceptors for GNN

Gly codons while Mycoplasma mycoides has solely tRNA UCC [216–218]. The difference

between these tRNAs is the nucleotide at position 32, where E. coli contains U32 and M.

mycoides has C32 [216–218]. These variations are thought to influence base pairing of

position 32 with position 38 and ultimately affect codon-anticodon base pairing.

Gly Consistent with this idea, a mutation from U32 to C32 allows E. coli tRNA UCC to read all

four glycine codons [216–218]. Arguably, the methylation of position 32 could have a

similar affect by abolishing base pairing between positions 32-38. Along these lines, the

1 lack of m G37 is thought to abolish the U32-A38 base-pair leading to an increase in +1 frameshifting [219]. The difference between C32 and U32 because of deamination could

also result in differences in affinity for the ribosome which could modulate translational

efficiency or accuracy which has been shown to be true for bacterial tRNAs [220].

Other possibilities outside of translation include resistance or response to external

stresses. Numerous examples of modifications influencing environmental stress response

pathways exist, including preferential transcription of critical response transcripts [221–

116

224]. A recent publication on methylation of the ribose sugar at position 32 of tRNA in

Pseudomonas aeruginosa showed that this modification conferred resistance to the

3 3 oxidative stress response [221]. The involvement of m C32 and m U32 in the stress

response of T. brucei has not been assessed, but could be tested in future work.

5.5 The effect of compartmentalization on anticodon loop editing

We showed that C to U editing occurs in the nucleus while A to I is formed in the

cytoplasm [116,149]. An appealing rationalization is that the TbTrm140-TbADAT2/3

complex prevents TbADAT2/3 from forming A to I at position 34 in the nucleus. Since

TbTrm140 is exclusively nuclear, TbADAT2/3 would then be free to perform A to I in

the cytoplasm [116,149]. This is unlikely, however, since TbADAT2/3 forms A to I in

vitro regardless of whether TbTrm140 is present in the reaction [116,149]. Another

explanation could be that an additional modification in the nucleus acts as a determinant or anti-determinant for TbADAT2/3 activity. For instance, a modification important for tRNA structure could be introduced in the nucleus and be a prerequisite for A to I formation. This would be consistent with observations from S. cerevisiae and data presented in chapter 3 of this thesis showing that TbADAT2/3 is sensitive to tRNA tertiary structure [37,68,69,138,139]. In contrast to this idea, we observe A to I editing using synthetic tRNA in vitro, suggesting that TbADAT2/3 can deaminate A34 without

other modifications. There is also direct evidence for modifications, such as thiolations,

to act as negative determinants for editing [225,226]. Likewise, posttranslational

modifications of TbADAT2/3 could influence its tRNA binding or activity within the

nucleus. RNA binding proteins commonly utilize posttranslational modifications to

117

regulate various features such as RNA-binding, activity, or localization [227]. For

example, a posttranslational modification added to TbADAT2/3 in the nucleus could

prevent it from binding tRNA in a way that is productive for A to I editing. There could

also be other protein factors, potentially other tRNA modification enzymes, that could

prevent TbADAT2/3 from forming A to I at position 34 in the nucleus.

5.6 Model of modification interdependence and quality control

From our collective data we propose a model whereby TbADAT2/3 and

TbTrm140 localize to the nucleus where they act together to form m3C followed by m3U

at position 32 of tRNA. The tRNA is subsequently exported to the cytosol where it

undergoes further A-to-I deamination editing at position 34 carried out by TbADAT2/3.

In the nucleus, the TbADAT2/3 and TbTrm140 complex, likely bound to their tRNA

substrates, prevents TbADAT2/3 from mutagenizing the T. brucei genome (Figure 5.1).

118

Figure 5.1 Model of modification interdependence and quality control.

119

References

1. Pays, E.; Vanhamme, L.; Pérez-Morga, D. Antigenic variation in Trypanosoma brucei: facts, challenges and mysteries. Curr. Opin. Microbiol. 2004, 7, 369–374.

2. Borst, P. Antigenic Variation and Allelic Exclusion. Cell 2002, 109, 5–8.

3. Franco, J. R.; Simarro, P. P.; Diarra, A.; Jannin, J. G. Epidemiology of human African trypanosomiasis. Clin. Epidemiol. 2014, 6, 257–275.

4. Morrison, L. J.; Marcello, L.; McCulloch, R. Antigenic variation in the African trypanosome: molecular mechanisms and phenotypic complexity. Cell. Microbiol. 2009, 11, 1724–1734.

5. van Hellemond, J. J.; Opperdoes, F. R.; Tielens, a G. M. The extraordinary mitochondrion and unusual citric acid cycle in Trypanosoma brucei. Biochem. Soc. Trans. 2005, 33, 967–71.

6. Liu, B.; Liu, Y.; Motyka, S. A.; Agbo, E. E. C.; Englund, P. T. Fellowship of the rings: the replication of kinetoplast DNA. Trends Parasitol. 2005, 21, 363–369.

7. Lukes, J.; Lys Guilbride, D.; Votypka, J.; Zikova, A.; Benne, R.; Englund, P. T. Kinetoplast DNA Network: Evolution of an Improbable Structure. Eukaryot. Cell 2002, 1, 495–502.

8. Koslowsky, D. J. Complex interactions in the regulation of trypanosome mitochondrial gene expression. Trends Parasitol. 2009, 25, 252–255.

9. Shapiro, T. A. Kinetoplast DNA maxicircles: Networks within networks (trypanosomes/topology). Biochemistry 1993, 90, 7809–7813.

10. Koslowsky, D. J.; Yahampath, G. Mitochondrial mRNA 3’ cleavage/polyadenylation and RNA editing in Trypanosoma brucei are independent events. Mol. Biochem. Parasitol. 1997, 90, 81–94.

11. Aphasizhev, R.; Aphasizheva, I. Mitochondrial RNA editing in trypanosomes: small RNAs in control. Biochimie 2014, 100, 125–31.

12. Gray, M. W. Diversity and evolution of mitochondrial RNA editing systems. IUBMB 120

Life 2003, 55, 227–33.

13. Mörl, M.; Dörner, M.; Pääbo, S. C to U editing and modifications during the maturation of the mitochondrial tRNA(Asp) in marsupials. Nucleic Acids Res. 1995, 23, 3380–4.

14. McDermott, S. M.; Luo, J.; Carnes, J.; Ranish, J. A.; Stuart, K. The Architecture of Trypanosoma brucei editosomes. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, E6476– E6485.

15. Carnes, J.; Trotter, J. R.; Peltan, A.; Fleck, M.; Stuart, K. RNA editing in Trypanosoma brucei requires three different editosomes. Mol. Cell. Biol. 2008, 28, 122– 30.

16. Gott, J. M.; Emeson, R. B. Functions and mechanisms of RNA editing. Annu. Rev. Genet. 2000, 34, 499–531.

17. Rammelt, C.; Rossmanith, W. Repairing tRNA termini: News from the 3’ end. RNA Biol. 2016, 1–7.

18. Sharp, S. J.; Schaack, J.; Cooley, L.; Burke, D. J.; Söll, D. Structure and transcription of eukaryotic tRNA genes. CRC Crit. Rev. Biochem. 1985, 19, 107–44.

19. Pinker, F.; Bonnard, G.; Gobert, A.; Gutmann, B.; Hammani, K.; Sauter, C.; Gegenheimer, P. A.; Giegé, P. PPR proteins shed a new light on RNase P biology. RNA Biol. 2013, 10, 1457–1468.

20. Aphasizhev, R.; Karmarkar, U.; Simpson, L. Are tRNAs imported into the mitochondria of kinetoplastid protozoa as 5’-extended precursors? Mol. Biochem. Parasitol. 1998, 93.

21. Alfonzo, J. D.; Söll, D. Mitochondrial tRNA import – the challenge to understand has just begun. Biol. Chem. 2009, 390, 717–722.

22. Hauser, R.; Schneider, A. tRNAs are imported into mitochondria of Trypanosoma brucei independently of their genomic context and genetic origin. EMBO J. 1995, 14, 4212–20.

23. Tan, T. H. P.; Pach, R.; Crausaz, A.; Ivens, A.; Schneider, A. tRNAs in Trypanosoma brucei: genomic organization, expression, and mitochondrial import. Mol. Cell. Biol. 2002, 22, 3707–17.

24. Hancock, K.; LeBlanc, A. J.; Donze, D.; Hajduk, S. L. Identification of nuclear encoded precursor tRNAs within the mitochondrion of Trypanosoma brucei. J. Biol. 121

Chem. 1992, 267, 23963–23971.

25. Salavati, R.; Panigrahi, A. K.; Stuart, K. D. Mitochondrial ribonuclease P activity of Trypanosoma brucei. Mol. Biochem. Parasitol. 2001, 115, 109–117.

26. Hancock, K.; Jahduk, S. L. The mitochondrial tRNAs of Trypanosoma brucei are nuclear encoded. J. Biol. Chem. 1990, 265, 19208–19215.

27. Schneider, A.; Martin, J.; Agabian, N. A nuclear encoded tRNA of Trypanosoma brucei is imported into mitochondria. Mol. Cell. Biol. 1994, 14, 2317–22.

28. Maraia, R. J.; Lamichhane, T. N. 3′ processing of eukaryotic precursor tRNAs. Wiley Interdiscip. Rev. RNA 2011, 2, 362–375.

29. Betat, H.; Long, Y.; Jackman, J.; Mörl, M. From End to End: tRNA Editing at 5’- and 3’-Terminal Positions. Int. J. Mol. Sci. 2014, 15, 23975–23998.

30. Hopper, A. K. Transfer RNA post-transcriptional processing, turnover, and subcellular dynamics in the yeast Saccharomyces cerevisiae. Genetics 2013, 194, 43–67.

31. Paushkin, S. V; Patel, M.; Furia, B. S.; Peltz, S. W.; Trotta, C. R. Identification of a human endonuclease complex reveals a link between tRNA splicing and pre-mRNA 3’ end formation. Cell 2004, 117, 311–21.

32. Dhungel, N.; Hopper, A. K. Beyond tRNA cleavage : novel essential function for yeast tRNA splicing endonuclease unrelated to tRNA processing. 2012, 503–514.

33. Lunde, B. M.; Moore, C.; Varani, G. RNA-binding proteins: modular design for efficient function. Nat. Rev. Mol. Cell Biol. 2007, 8, 479–90.

34. Blanc, V.; Navaratnam, N.; Henderson, J. O.; Anant, S.; Kennedy, S.; Jarmuz, A.; Scott, J.; Davidson, N. O. Identification of GRY-RBP as an apolipoprotein B RNA- binding protein That interacts with both apobec-1 and apobec-1 complementation factor to modulate C to U editing. J. Biol. Chem. 2001, 276, 10272–10283.

35. Mehta, A.; Kinter, M. T.; Sherman, N. E.; Driscoll, D. M. Molecular cloning of apobec-1 complementation factor, a novel RNA-binding protein involved in the editing of apolipoprotein B mRNA. Mol. Cell. Biol. 2000, 20, 1846–54.

36. Zhang, J.; Ferré-D’Amaré, A. R. The tRNA Elbow in Structure, Recognition and Evolution. Life (Basel, Switzerland) 2016, 6, 3.

37. Grosjean, H.; Edqvist, J.; Stråby, K. B.; Giegé, R. Enzymatic formation of modified nucleosides in tRNA: dependence on tRNA architecture. J. Mol. Biol. 1996, 255, 67–85. 122

38. Paris, Z.; Fleming, I. M. C.; Alfonzo, J. D. Determinants of tRNA editing and modification: avoiding conundrums, affecting function. Semin. Cell Dev. Biol. 2012, 23, 269–74.

39. Randau, L.; Stanley, B. J.; Kohlway, A.; Mechta, S.; Xiong, Y.; Söll, D. A cytidine deaminase edits C to U in transfer RNAs in Archaea. Science. 2009, 324, 657–9.

40. Grosjean, H.; Constantinesco, F.; Foiret, D.; Benachenhou, N. A novel enzymatic pathway leading to 1-methylinosine modification in Haloferax volcanii tRNA. Nucleic Acids Res. 1995, 23, 4312–4319.

41. Gerber, A.; Grosjean, H.; Melcher, T.; Keller, W. Tad1p, a yeast tRNA-specific adenosine deaminase, is related to the mammalian pre-mRNA editing enzymes ADAR1 and ADAR2. EMBO J. 1998, 17, 4780–4789.

42. Janke, A.; Pääbo, S. Editing of a tRNA anticodon in marsupial mitochondria changes its codon recognition. Nucleic Acids Res. 1993, 21, 1523–1525.

43. Alfonzo, J. D.; Blanc, V.; Estévez, A. M.; Rubio, M. A. T.; Simpson, L. C to U editing of the anticodon of imported mitochondrial tRNA(Trp) allows decoding of the UGA stop codon in Leishmania tarentolae. EMBO J. 1999, 18, 7056–62.

44. Phizicky, E. M.; Alfonzo, J. D. Do all modifications benefit all tRNAs? FEBS Lett. 2010, 584, 265–271.

45. Gerber, A. P.; Keller, W. RNA editing by base deamination: More enzymes, more targets, new mysteries. Trends Biochem. Sci. 2001, 26, 376–384.

46. Wolf, J.; Gerber, A. P.; Keller, W. tadA, an essential tRNA-specific adenosine deaminase from Escherichia coli. EMBO J. 2002, 21, 3841–51.

47. Gerber, A. P. An Adenosine Deaminase that Generates Inosine at the Wobble Position of tRNAs. Science. 1999, 286, 1146–1149.

48. Rubio, M. A. T.; Pastar, I.; Gaston, K. W.; Ragone, F. L.; Janzen, C. J.; Cross, G. A. M.; Papavasiliou, F. N.; Alfonzo, J. D. An adenosine-to-inosine tRNA-editing enzyme that can perform C-to-U deamination of DNA. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 7821–7826.

49. Torres, A. G.; Piñeyro, D.; Rodríguez-Escribà, M.; Camacho, N.; Reina, O.; Saint- Léger, A.; Filonava, L.; Batlle, E.; Ribas De Pouplana, L. Inosine modifications in human tRNAs are incorporated at the precursor tRNA level. Nucleic Acids Res. 2015, 43, 5145– 5157.

123 50. Zhou, W.; Karcher, D.; Bock, R. Identification of enzymes for adenosine-to-inosine editing and discovery of cytidine-to-uridine editing in nucleus-encoded transfer RNAs of Arabidopsis. Plant Physiol. 2014, 166, 1985–97.

51. Kim, J.; Malashkevich, V.; Roday, S.; Lisbin, M.; Schramm, V. L.; Almo, S. C. Structural and kinetic characterization of Escherichia coli TadA, the wobble-Specific tRNA deaminase. Biochemistry 2006, 45, 6407–6416.

52. Losey, H. C.; Ruthenburg, A. J.; Verdine, G. L. Crystal structure of Staphylococcus aureus tRNA adenosine deaminase TadA in complex with RNA. Nat. Struct. Mol. Biol. 2006, 13, 153–9.

53. Elias, Y.; Huang, R. H. Biochemical and structural studies of A-to-I editing by tRNA:A34 deaminases at the wobble position of transfer RNA. Biochemistry 2005, 44, 12057–12065.

54. Kuratani, M.; Ishii, R.; Bessho, Y.; Fukunaga, R.; Sengoku, T.; Shirouzu, M.; Sekine, S.; Yokoyama, S. Crystal structure of tRNA adenosine deaminase (TadA) from Aquifex aeolicus. J. Biol. Chem. 2005, 280, 16002–16008.

55. Umitsu, M.; Nishimasu, H.; Noma, A.; Suzuki, T.; Ishitani, R.; Nureki, O. Structural basis of AdoMet-dependent aminocarboxypropyl transfer reaction catalyzed by tRNA- wybutosine synthesizing enzyme, TYW2. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 15616–21.

56. Waas, W. F.; de Crécy-Lagard, V.; Schimmel, P. Discovery of a gene family critical to wyosine base formation in a subset of phenylalanine-specific transfer RNAs. J. Biol. Chem. 2005, 280, 37616–37622.

57. Noma, A.; Kirino, Y.; Ikeuchi, Y.; Suzuki, T. Biosynthesis of wybutosine, a hypermodified nucleoside in eukaryotic phenylalanine tRNA. EMBO J. 2006, 25, 2142–54.

58. Kalhor, H. R.; Penjwini, M.; Clarke, S. A novel methyltransferase required for the formation of the hypermodified nucleoside wybutosine in eucaryotic tRNA. Biochem. Biophys. Res. Commun. 2005, 334, 433–40.

59. Suzuki, Y.; Noma, A.; Suzuki, T.; Ishitani, R.; Nureki, O. Structural basis of tRNA modification with CO2 fixation and methylation by wybutosine synthesizing enzyme TYW4. Nucleic Acids Res. 2009, 37, 2910–25.

60. Perche-Letuvée, P.; Molle, T.; Forouhar, F.; Mulliez, E.; Atta, M. Wybutosine biosynthesis: Structural and mechanistic overview. RNA Biol. 2014, 11, 1508–1518.

61. Young, A. P.; Bandarian, V. Mechanistic Studies of the Radical S- Adenosyl- 124

l -methionine Enzyme 4-Demethylwyosine Synthase Reveal the Site of Hydrogen Atom Abstraction. Biochemistry 2015, 54, 3569–3572.

62. Young, A. P.; Bandarian, V. Pyruvate is the source of the two carbons that are required for formation of the imidazoline ring of 4-demethylwyosine. Biochemistry 2011, 50, 10573–5.

63. Suzuki, Y.; Noma, A.; Suzuki, T.; Senda, M.; Senda, T.; Ishitani, R.; Nureki, O. Crystal structure of the radical SAM enzyme catalyzing tricyclic modified base formation in tRNA. J. Mol. Biol. 2007, 372, 1204–14.

64. Young, A. P.; Bandarian, V. Radical mediated ring formation in the biosynthesis of the hypermodified tRNA base wybutosine. Curr. Opin. Chem. Biol. 2013, 17, 613–8.

65. de Crecy-Lagard, V.; Brochier-Armanet, C.; Urbonavicius, J.; Fernandez, B.; Phillips, G.; Lyons, B.; Noma, a.; Alvarez, S.; Droogmans, L.; Armengaud, J.; Grosjean, H. Biosynthesis of Wyosine Derivatives in tRNA: An Ancient and Highly Diverse Pathway in Archaea. Mol. Biol. Evol. 2010, 27, 2062–2077.

66. Sample, P. J.; Kořený L.; Paris, Z.; Gaston, K. W.; Rubio, M. A. T.; Fleming, I. M. C.; Hinger, S.; Horáková, E.; Limbach, P. A.; Lukeš, J.; Alfonzo, J. D. A common tRNA modification at an unusual location: the discovery of wyosine biosynthesis in mitochondria. Nucleic Acids Res. 2015, 43, 4262–4273.

67. Motorin, Y.; Helm, M. RNA nucleotide methylation. Wiley Interdiscip. Rev. RNA 2011, 2, 611–31.

68. Liu, R. J.; Long, T.; Zhou, M.; Zhou, X. L.; Wang, E. D. tRNA recognition by a bacterial tRNA Xm32 modification enzyme from the SPOUT methyltransferase superfamily. Nucleic Acids Res. 2015, 43, 7489–7503.

69. Hori, H. Methylated nucleosides in tRNA and tRNA methyltransferases. Front. Genet. 2014, 5, 144.

70. Yamagami, R.; Yamashita, K.; Nishimasu, H.; Tomikawa, C.; Ochi, A.; Iwashita, C.; Hirata, A.; Ishitani, R.; Nureki, O.; Hori, H. The tRNA recognition mechanism of folate/FAD-dependent tRNA methyltransferase (TrmFO). J. Biol. Chem. 2012, 287, 42480–42494.

71. Kim, J.; Xiao, H.; Bonanno, J. B.; Kalyanaraman, C.; Brown, S.; Tang, X.; Al- Obaidi, N. F.; Patskovsky, Y.; Babbitt, P. C.; Jacobson, M. P.; Lee, Y.-S.; Almo, S. C. Structure-guided discovery of the metabolite carboxy-SAM that modulates tRNA function. Nature 2013, 498, 123–6.

125

72. Kozbial, P. Z.; Mushegian, A. R. Natural history of S-adenosylmethionine-binding proteins. BMC Struct. Biol. 2005, 5, 19.

73. Schubert, H. L.; Blumenthal, R. M.; Cheng, X. Many paths to methyltransfer: a chronicle of convergence. Trends Biochem. Sci. 2003, 28, 329–35.

74. Ero, R.; Peil, L.; Liiv, A.; Remme, J. Identification of pseudouridine methyltransferase in Escherichia coli. RNA 2008, 14, 2223–2233.

75. Swinehart, W. E.; Jackman, J. E. Diversity in mechanism and function of tRNA methyltransferases. RNA Biol. 2015, 12, 398–411.

76. Urbonavicius, J.; Jäger, G.; Björk, G. R. Amino acid residues of the Escherichia coli tRNA(m5U54)methyltransferase (TrmA) critical for stability, covalent binding of tRNA and enzymatic activity. Nucleic Acids Res. 2007, 35, 3297–305.

77. McKenney, K. M.; Alfonzo, J. D. From Prebiotics to Probiotics: The Evolution and Functions of tRNA Modifications. Life (Basel, Switzerland) 2016, 6, 13.

78. Anantharaman, V.; Koonin, E. V.; Aravind, L. TRAM, a predicted RNA-binding domain, common to tRNA uracil methylation and adenine thiolation enzymes. FEMS Microbiol. Lett. 2001, 197, 215–221.

79. Aravind, L.; Koonin, E. V Novel predicted RNA-binding domains associated with the translation machinery. J. Mol. Evol. 1999, 48, 291–302.

80. Zegers, I.; Gigot, D.; van Vliet, F.; Tricot, C.; Aymerich, S.; Bujnicki, J. M.; Kosinski, J.; Droogmans, L. Crystal structure of Bacillus subtilis TrmB, the tRNA (m7G46) methyltransferase. Nucleic Acids Res. 2006, 34, 1925–1934.

81. Witek, M. A.; Kuiper, E. G.; Minten, E.; Crispell, E. K.; Conn, G. L. A Novel Motif for S-Adenosyl-l-methionine Binding by the Ribosomal RNA Methyltransferase TlyA from Mycobacterium tuberculosis. J. Biol. Chem. 2017, 292, 1977–1987.

82. Martin, J. L.; McMillan, F. M. SAM (dependent) I AM: the S-adenosylmethionine- dependent methyltransferase fold. Curr. Opin. Struct. Biol. 2002, 12, 783–93.

83. Pastore, C.; Topalidou, I.; Forouhar, F.; Yan, A. C.; Levy, M.; Hunt, J. F. Crystal structure and RNA binding properties of the RNA recognition motif (RRM) and AlkB domains in human AlkB homolog 8 (ABH8), an enzyme catalyzing tRNA hypermodification. J. Biol. Chem. 2012, 287, 2130–2143.

84. Towns, W. L.; Begley, T. J. Transfer RNA methytransferases and their corresponding modifications in budding yeast and humans: activities, predications, and potential roles in 126

human health. DNA Cell Biol. 2012, 31, 434–54.

85. Anantharaman, V.; Koonin, E. V; Aravind, L. SPOUT: a class of methyltransferases that includes spoU and trmD RNA methylase superfamilies, and novel superfamilies of predicted prokaryotic RNA methylases. J. Mol. Microbiol. Biotechnol. 2002, 4, 71–5.

86. Ahn, H. J.; Kim, H.-W.; Yoon, H.-J.; Lee, B. Il; Suh, S. W.; Yang, J. K. Crystal structure of tRNA(m1G37)methyltransferase: insights into tRNA recognition. EMBO J. 2003, 22, 2593–603.

87. Lim, K.; Zhang, H.; Tempczyk, A.; Krajewski, W.; Bonander, N.; Toedt, J.; Howard, A.; Eisenstein, E.; Herzberg, O. Structure of the YibK methyltransferase from Haemophilus influenzae (HI0766): A cofactor bound at a site formed by a knot. Proteins Struct. Funct. Genet. 2003, 51, 56–67.

88. Elkins, P. A.; Watts, J. M.; Zalacain, M.; Van Thiel, A.; Vitazka, P. R.; Redlak, M.; Andraos-Selim, C.; Rastinejad, F.; Holmes, W. M. Insights into catalysis by a knotted TrmD tRNA methyltransferase. J. Mol. Biol. 2003, 333, 931–949.

89. Tkaczuk, K.; Dunin-Horkawicz, S.; Purta, E.; Bujnicki, J. M. Structural and evolutionary bioinformatics of the SPOUT superfamily of methyltransferases. BMC Bioinformatics 2007, 8, 73.

90. Nureki, O.; Watanabe, K.; Fukai, S.; Ishii, R.; Endo, Y.; Hori, H.; Yokoyama, S. Deep knot structure for construction of active site and cofactor binding site of tRNA modification enzyme. Structure 2004, 12, 593–602.

91. Michel, G.; Sauvé, V.; Larocque, R.; Li, Y.; Matte, A.; Cygler, M. The structure of the RlmB 23S rRNA methyltransferase reveals a new methyltransferase fold with a unique knot. Structure 2002, 10, 1303–1315.

92. Christian, T.; Hou, Y.-M. Distinct determinants of tRNA recognition by the TrmD and Trm5 methyl transferases. J. Mol. Biol. 2007, 373, 623–32.

93. Van Laer, B.; Roovers, M.; Wauters, L.; Kasprzak, J. M.; Dyzma, M.; Deyaert, E.; Singh, R. K.; Feller, A.; Bujnicki, J. M.; Droogmans, L.; Versées, W. Structural and functional insights into tRNA binding and adenosine N1-methylation by an archaeal Trm10 homologue. Nucleic Acids Res. 2016, 44, 940–953.

94. Zhou, M.; Long, T.; Fang, Z.-P.; Zhou, X.-L.; Liu, R.-J.; Wang, E.-D. Identification of determinants for tRNA substrate recognition by Escherichia coli C/U34 2’-O- methyltransferase. RNA Biol. 2015, 12, 900–11.

95. Liu, R. J.; Zhou, M.; Fang, Z. P.; Wang, M.; Zhou, X. L.; Wang, E. D. The tRNA 127

recognition mechanism of the minimalist SPOUT methyltransferase, TrmL. Nucleic Acids Res. 2013, 41, 7828–7842.

96. Cerrudo, C. S.; Ghiringhelli, P. D.; Gomez, D. E. Protein universe containing a PUA RNA-binding domain. FEBS J. 2014, 281, 74–87.

97. Pérez-Arellano, I.; Gallego, J.; Cervera, J. The PUA domain - a structural and functional overview. FEBS J. 2007, 274, 4972–84.

98. Larsen, L. H. G.; Rasmussen, A.; Giessing, A. M. B.; Jogl, G.; Kirpekar, F. Identification and characterization of the Thermus thermophilus 5-methylcytidine (m5C) methyltransferase modifying 23 S ribosomal RNA (rRNA) base C1942. J. Biol. Chem. 2012, 287, 27593–600.

99. Haag, S.; Warda, A. S.; Kretschmer, J.; Günnigmann, M. A.; Höbartner, C.; Bohnsack, M. T. NSUN6 is a human RNA methyltransferase that catalyzes formation of m5C72 in specific tRNAs. RNA 2015, 21, 1532–43.

100. Aravind, L.; Koonin, E. V. THUMP--a predicted RNA-binding domain shared by 4- thiouridine, pseudouridine synthases and RNA methylases. Trends Biochem. Sci. 2001, 26, 215–7.

101. Neumann, P.; Lakomek, K.; Naumann, P. T.; Erwin, W. M.; Lauhon, C. T.; Ficner, R. Crystal structure of a 4-thiouridine synthetase-RNA complex reveals specificity of tRNA U8 modification. Nucleic Acids Res. 2014, 42, 6673–6685.

102. Lee, T. T.; Agarwalla, S.; Stroud, R. M. Crystal structure of RumA, an iron-sulfur cluster containing E. coli ribosomal RNA 5-methyluridine methyltransferase. Structure 2004, 12, 397–407.

103. Leulliot, N.; Bohnsack, M. T.; Graille, M.; Tollervey, D.; Van tilbeurgh, H. The yeast ribosome synthesis factor Emg1 is a novel member of the superfamily of alpha/beta knot fold methyltransferases. Nucleic Acids Res. 2008, 36, 629–639.

104. Nureki, O.; Shirouzu, M.; Hashimoto, K.; Ishitani, R.; Terada, T.; Tamakoshi, M.; Oshima, T.; Chijimatsu, M.; Takio, K.; Vassylyev, D. G.; Shibata, T.; Inoue, Y.; Kuramitsu, S.; Yokoyama, S. An enzyme with a deep trefoil knot for the active-site architecture. Acta Crystallogr. Sect. D Biol. Crystallogr. 2002, 58, 1129–1137.

105. Alian, A.; Lee, T. T.; Griner, S. L.; Stroud, R. M.; Finer-Moore, J. Structure of a TrmA-RNA complex: A consensus RNA fold contributes to substrate selectivity and catalysis in m5U methyltransferases. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 6876–81.

106. Walbott, H.; Leulliot, N.; Grosjean, H.; Golinelli-Pimpaneau, B. The crystal 128

structure of Pyrococcus abyssi tRNA (uracil-54, C5)-methyltransferase provides insights into its tRNA specificity. Nucleic Acids Res. 2008, 36, 4929–4940.

107. Ohira, T.; Suzuki, T. Retrograde nuclear import of tRNA precursors is required for modified base biogenesis in yeast. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 10502–7.

108. Huh, W.-K.; Falvo, J. V; Gerke, L. C.; Carroll, A. S.; Howson, R. W.; Weissman, J. S.; O’Shea, E. K. Global analysis of protein localization in budding yeast. Nature 2003, 425, 686–91.

109. Yoshihisa, T.; Yunoki-Esaki, K.; Ohshima, C.; Tanaka, N.; Endo, T. Possibility of cytoplasmic pre-tRNA splicing: the yeast tRNA splicing endonuclease mainly localizes on the mitochondria. Mol. Biol. Cell 2003, 14, 3266–79.

110. Goto-Ito, S.; Ito, T.; Kuratani, M.; Bessho, Y.; Yokoyama, S. Tertiary structure checkpoint at anticodon loop modification in tRNA functional maturation. Nat. Struct. Mol. Biol. 2009, 16, 1109–15.

111. Motorin, Y.; Grosjean, H. Multisite-specific tRNA:m5C-methyltransferase (Trm4) in yeast Saccharomyces cerevisiae: Identification of the gene and substrate specificity of the enzyme. RNA 1999, 5, 1105–1118.

112. Brzezicha, B.; Schmidt, M.; Makałowska, I.; Jarmołowski, A.; Pieńkowska, J.; Szweykowska-Kulińska, Z. Identification of human tRNA: m5C methyltransferase catalysing intron-dependent m5C formation in the first position of the anticodon of the pre-tRNA(CAA)Leu. Nucleic Acids Res. 2006, 34, 6034–6043.

113. Arimbasseri, A. G.; Iben, J.; Wei, F.; Rijal, K.; Tomizawa, K.; Hafner, M.; Maraia, R. J. Evolving specificity of tRNA 3-methyl-cytidine-32 (m3C32) modification: a subset of tRNAsSer requires N6-isopentenylation of A37. RNA 2016, 22, 1400–1410.

114. Roovers, M.; Wouters, J.; Bujnicki, J. M.; Tricot, C.; Stalon, V.; Grosjean, H.; Droogmans, L. A primordial RNA modification enzyme: the case of tRNA (m1A) methyltransferase. Nucleic Acids Res. 2004, 32, 465–76.

115. Rubio, M. A. T.; Ragone, F. L.; Gaston, K. W.; Ibba, M.; Alfonzo, J. D. C to U editing stimulates A to I editing in the anticodon loop of a cytoplasmic threonyl tRNA in Trypanosoma brucei. J. Biol. Chem. 2006, 281, 115–120.

116. Rubio, M. A. T.; Gaston, K. W.; McKenney, K. M.; Fleming, I. M. C.; Paris, Z.; Limbach, P. A.; Alfonzo, J. D. Editing and methylation at a single site by functionally interdependent activities. Nature 2017, 542, 494–497.

117. Yewdell, W. T.; Chaudhuri, J. Molecular biology: RNA editing packs a one-two 129

punch. Nature 2017, 542, 420–421.

118. Guy, M. P.; Podyma, B. M.; Preston, M. A.; Shaheen, H. H.; Krivos, K. L.; Limbach, P. A.; Hopper, A. K.; Phizicky, E. M. Yeast Trm7 interacts with distinct proteins for critical modifications of the tRNAPhe anticodon loop. RNA 2012, 18, 1921– 1933.

119. Rubio, M.; Alfonzo, J. D. Editing and modification in trypanosomatids: The reshaping of non-coding RNAs. Top. Curr. Genet. 2005, 12, 71–86.

120. Guy, M. P.; Phizicky, E. M. Two-subunit enzymes involved in eukaryotic post- transcriptional tRNA modification. RNA Biol. 2014, 11, 1608–18.

121. Shao, Z.; Yan, W.; Peng, J.; Zuo, X.; Zou, Y.; Li, F.; Gong, D.; Ma, R.; Wu, J.; Shi, Y.; Zhang, Z.; Teng, M.; Li, X.; Gong, Q. Crystal structure of tRNA m1G9 methyltransferase Trm10: Insight into the catalytic mechanism and recognition of tRNA substrate. Nucleic Acids Res. 2014, 42, 509–525.

122. Kurowski, M. A.; Sasin, J. M.; Feder, M.; Debski, J.; Bujnicki, J. M. Characterization of the cofactor-binding site in the SPOUT-fold methyltransferases by computational docking of S-adenosylmethionine to three crystal structures. BMC Bioinformatics 2003, 4, 9.

123. Purushothaman, S. K.; Bujnicki, J. M.; Grosjean, H.; Lapeyre, B. Trm11p and Trm112p are both required for the formation of 2-methylguanosine at position 10 in yeast tRNA. Mol. Cell. Biol. 2005, 25, 4359–4370.

124. Sardana, R.; Johnson, A. W. The methyltransferase adaptor protein Trm112 is involved in biogenesis of both ribosomal subunits. Mol. Biol. Cell 2012, 23, 4313–22.

125. Figaro, S.; Wacheul, L.; Schillewaert, S.; Graille, M.; Huvelle, E.; Mongeard, R.; Zorbas, C.; Lafontaine, D. L. J.; Heurgue-Hamard, V. Trm112 Is Required for Bud23- Mediated Methylation of the 18S rRNA at Position G1575. Mol. Cell. Biol. 2012, 32, 2254–2267.

126. Létoquart, J.; Huvelle, E.; Wacheul, L.; Bourgeois, G.; Zorbas, C.; Graille, M.; Heurgué-Hamard, V.; Lafontaine, D. L. J. Structural and functional studies of Bud23- Trm112 reveal 18S rRNA N7-G1575 methylation occurs on late 40S precursor ribosomes. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, E5518-26.

127. Anderson, J.; Phan, L.; Hinnebusch, A. G. The Gcd10p/Gcd14p complex is the essential two-subunit tRNA(1-methyladenosine) methyltransferase of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 5173–5178.

130

128. Ozanick, S. G.; Bujnicki, J. M.; Sem, D. S.; Anderson, J. T. Conserved amino acids in each subunit of the heteroligomeric tRNA m1A58 Mtase from Saccharomyces cerevisiae contribute to tRNA binding. Nucleic Acids Res. 2007, 35, 6808–6819.

129. Alexandrov, A.; Martzen, M. R.; Phizicky, E. M. Two proteins that form a complex are required for 7-methylguanosine modification of yeast tRNA. RNA 2002, 8, 1253–66.

130. Somme, J.; Van Laer, B.; Roovers, M.; Steyaert, J.; Versées, W.; Droogmans, L. Characterization of two homologous 2’-O-methyltransferases showing different specificities for their tRNA substrates. RNA 2014, 20, 1257–71.

131. Crécy‐Lagard, V. de Identification of Genes Encoding tRNA Modification Enzymes by Comparative Genomics. In Methods in Enzymology; 2007; Vol. 425, pp. 153–183.

132. Hopper, A. K.; Phizicky, E. M. tRNA transfers to the limelight. Genes Dev. 2003, 17, 162–80.

133. Björk, G. R.; Kohli, J. Synthesis and Function of Modified Nucleosides in tRNA. In Chromatography and Modification of Nucleosides. Part B. Biological Roles and Function of Modification.; Gehrke, C.; Kuo, K., Eds.; 1990; pp. B13–B67.

134. Helm, M. Post-transcriptional nucleotide modification and alternative folding of RNA. Nucleic Acids Res. 2006, 34, 721–733.

135. McKenney, K. M.; Rubio, M. A. T.; Alfonzo, J. D. The Evolution of Substrate Specificity by tRNA Modification Enzymes. In The Enzymes; 2017; pp. 51–88.

136. Kobitski, A. Y.; Hengesbach, M.; Seidu-Larry, S.; Dammertz, K.; Chow, C. S.; van Aerschot, A.; Nienhaus, G. U.; Helm, M. Single-molecule FRET reveals a cooperative effect of two methyl group modifications in the folding of human mitochondrial tRNA(Lys). Chem. Biol. 2011, 18, 928–36.

137. Helm, M.; Brulé, H.; Degoul, F.; Cepanec, C.; Leroux, J. P.; Giegé, R.; Florentz, C. The presence of modified nucleotides is required for cloverleaf folding of a human mitochondrial tRNA. Nucleic Acids Res. 1998, 26, 1636–1643.

138. Auxilien, S.; Crain, P. F.; Trewyn, R. W.; Grosjean, H. Mechanism, specificity and general properties of the yeast enzyme catalysing the formation of inosine 34 in the anticodon of transfer RNA. J. Mol. Biol. 1996, 262, 437–58.

139. Haumont, E.; Fournier, M.; Henau, S. De; Grosjean, H. Enzymatic conversion of adenosine to inosine in the wobble position of yeast tRNAAsp: The dependence on the anticodon sequence. Nucleic Acids Res. 1984, 12, 2705–2715.

131

140. Crain, P. F.; Alfonzo, J. D.; Rozenski, J.; Kapushoc, S. T.; McCloskey, J. a; Simpson, L. Modification of the universally unmodified uridine-33 in a mitochondria- imported edited tRNA and the role of the anticodon arm structure on editing efficiency. RNA 2002, 8, 752–61.

141. Tomikawa, C.; Yokogawa, T.; Kanai, T.; Hori, H. N7-Methylguanine at position 46 (m7G46) in tRNA from Thermus thermophilus is required for cell viability at high temperatures through a tRNA modification network. Nucleic Acids Res. 2010, 38, 942– 57.

142. Ishida, K.; Kunibayashi, T.; Tomikawa, C.; Ochi, A.; Kanai, T.; Hirata, A.; Iwashita, C.; Hori, H. Pseudouridine at position 55 in tRNA controls the contents of other modified nucleotides for low-temperature adaptation in the extreme-thermophilic eubacterium Thermus thermophilus. Nucleic Acids Res. 2011, 39, 2304–2318.

143. Shigi, N.; Suzuki, T.; Terada, T.; Shirouzu, M.; Yokoyama, S.; Watanabe, K. Temperature-dependent biosynthesis of 2-thioribothymidine of Thermus thermophilus tRNA. J. Biol. Chem. 2006, 281, 2104–2113.

144. Rider, L. W.; Ottosen, M. B.; Gattis, S. G.; Palfey, B. a Mechanism of dihydrouridine synthase 2 from yeast and the importance of modifications for efficient tRNA reduction. J. Biol. Chem. 2009, 284, 10324–10333.

145. Han, L.; Marcus, E.; D’Silva, S.; Phizicky, E. M. S. cerevisiae Trm140 has two recognition modes for 3-methylcytidine modification of the anticodon loop of tRNA substrates. RNA 2016, rna.059667.116.

146. Grosjean, H.; Auxilien, S.; Constantinesco, F.; Simon, C.; Corda, Y.; Becker, H. F.; Foiret, D.; Morin, A.; Jin, Y. X.; Fournier, M.; Fourrey, J. L. Enzymatic conversion of adenosine to inosine and to N1-methylinosine in transfer RNAs: A review. Biochimie 1996, 78, 488–501.

147. Roovers, M.; Wouters, J.; Bujnicki, J. M.; Tricot, C.; Stalon, V.; Grosjean, H.; Droogmans, L. A primordial RNA modification enzyme: The case of tRNA (m1A) methyltransferase. Nucleic Acids Res. 2004, 32, 465–476.

148. Ragone, F. L.; Spears, J. L.; Wohlgamuth-Benedum, J. M.; Kreel, N.; Papavasiliou, F. N.; Alfonzo, J. D. The C-terminal end of the Trypanosoma brucei editing deaminase plays a critical role in tRNA binding. RNA 2011, 17, 1296–306.

149. Gaston, K. W.; Rubio, M. A. T.; Spears, J. L.; Pastar, I.; Papavasiliou, N. F.; Alfonzo, J. D. C to U editing at position 32 of the anticodon loop precedes tRNA 5′ leader removal in trypanosomatids. Nucleic Acids Res. 2007, 35, 6740–6749.

132

150. Sibert, B. S.; Patton, J. R. Pseudouridine synthase 1: a site-specific synthase without strict sequence recognition requirements. Nucleic Acids Res. 2012, 40, 2107–2118.

151. Arluison, V.; Buckle, M.; Grosjean, H. Pseudouridine synthetase Pus1 of Saccharomyces cerevisiae: kinetic characterisation, tRNA structural requirement and real-time analysis of its complex with tRNA. J. Mol. Biol. 1999, 289, 491–502.

152. Noma, A.; Yi, S.; Katoh, T.; Takai, Y.; Suzuki, T.; Suzuki, T. Actin-binding protein ABP140 is a methyltransferase for 3-methylcytidine at position 32 of tRNAs in Saccharomyces cerevisiae. RNA 2011, 17, 1111–9.

153. Wang, L.; Brown, S. J. BindN: A web-based tool for efficient prediction of DNA and RNA binding sites in amino acid sequences. Nucleic Acids Res. 2006, 34.

154. Purta, E.; van Vliet, F.; Tkaczuk, K. L.; Dunin-Horkawicz, S.; Mori, H.; Droogmans, L.; Bujnicki, J. M. The yfhQ gene of Escherichia coli encodes a tRNA:Cm32/Um32 methyltransferase. BMC Mol. Biol. 2006, 7, 23.

155. Pintard, L.; Lecointe, F.; Bujnicki, J. M.; Bonnerot, C.; Grosjean, H.; Lapeyre, B. Trm7p catalyses the formation of two 2′-O-methylriboses in yeast tRNA anticodon loop. EMBO J. 2002, 21, 1811–1820.

156. Benítez-Páez, A.; Villarroya, M.; Douthwaite, S.; Gabaldón, T.; Armengod, M.-E. YibK is the 2’-O-methyltransferase TrmL that modifies the wobble nucleotide in Escherichia coli tRNA(Leu) isoacceptors. RNA 2010, 16, 2131–43.

157. Mazauric, M. H.; Dirick, L.; Purushothaman, S. K.; Bjö Rk, G. R.; Lapeyre, B. Trm112p is a 15-kDa zinc finger protein essential for the activity of two tRNA and one protein methyltransferases in yeast. J. Biol. Chem. 2010, 285, 18505–18515.

158. Chen, C.; Huang, B.; Anderson, J. T.; Byström, A. S. Unexpected Accumulation of ncm5U and ncm5s2U in a trm9 Mutant Suggests an Additional Step in the Synthesis of mcm5U and mcm5s2U. PLoS One 2011, 6, e20783.

159. Kalhor, H. R.; Clarke, S. Novel methyltransferase for modified uridine residues at the wobble position of tRNA. Mol. Cell. Biol. 2003, 23, 9283–9292.

160. Studte, P.; Zink, S.; Jablonowski, D.; Bär, C.; von der Haar, T.; Tuite, M. F.; Schaffrath, R. tRNA and protein methylase complexes mediate zymocin toxicity in yeast. Mol. Microbiol. 2008, 69, 1266–77.

161. Liger, D.; Mora, L.; Lazar, N.; Figaro, S.; Henri, J.; Scrima, N.; Buckingham, R. H.; Van Tilbeurgh, H.; Heurgué-Hamard, V.; Graille, M. Mechanism of activation of methyltransferases involved in translation by the Trm112 “hub” protein. Nucleic Acids 133

Res. 2011, 39, 6249–6259.

162. Anderson, J.; Phan, L.; Cuesta, R.; Carlson, B. A.; Pak, M.; Asano, K.; Björk, G. R.; Tamame, M.; Hinnebusch, A. G. The essential Gcd10p-Gcd14p nuclear complex 15 required for 1- methyladenosine modification and maturation of initiator methionyl- tRNA. Genes Dev. 1998, 12, 3650–3662.

163. Leulliot, N.; Chaillet, M.; Durand, D.; Ulryck, N.; Blondeau, K.; van Tilbeurgh, H. Structure of the Yeast tRNA m7G Methylation Complex. Structure 2008, 16, 52–61.

164. Matsumoto, K.; Toyooka, T.; Tomikawa, C.; Ochi, A.; Takano, Y.; Takayanagi, N.; Endo, Y.; Hori, H. RNA recognition mechanism of eukaryote tRNA (m7G46) methyltransferase (Trm8-Trm82 complex). FEBS Lett. 2007, 581, 1599–1604.

165. Uter, N. T.; Perona, J. J. Long-range intramolecular signaling in a tRNA synthetase complex revealed by pre-steady-state kinetics. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 14396–401.

166. Leulliot, N.; Varani, G. Current topics in RNA-protein recognition: Control of specificity and biological function through induced fit and conformational capture. Biochemistry 2001, 40, 7947–7956.

167. Williamson, J. R. Induced fit in RNA-protein recognition. Nat. Struct. Biol. 2000, 7, 834–7.

168. Jones, K. A.; Peterlin, B. M. Control of RNA initiation and elongation at the HIV-1 promoter. Annu. Rev. Biochem. 1994, 63, 717–43.

169. Eaton, B. E.; Gold, L.; Zichi, D. A. Let’s get specific: the relationship between specificity and affinity. Chem. Biol. 1995, 2, 633–638.

170. Ochi, A.; Makabe, K.; Yamagami, R.; Hirata, A.; Sakaguchi, R.; Hou, Y. M.; Watanabe, K.; Nureki, O.; Kuwajima, K.; Hori, H. The catalytic domain of topological knot tRNA methyltransferase (TrmH) discriminates between substrate tRNA and nonsubstrate tRNA via an induced-fit process. J. Biol. Chem. 2013, 288, 25562–25574.

171. Ochi, A.; Makabe, K.; Kuwajima, K.; Hori, H. Flexible recognition of the tRNA g18 methylation target site by trmh methyltransferase through first binding and induced fit processes. J. Biol. Chem. 2010, 285, 9018–9029.

172. D’Silva, S.; Haider, S. J.; Phizicky, E. M. A domain of the actin binding protein Abp140 is the yeast methyltransferase responsible for 3-methylcytidine modification in the tRNA anti-codon loop. RNA 2011, 17, 1100–10.

134

173. Fleming, I. M. C.; Paris, Z.; Gaston, K. W.; Balakrishnan, R.; Fredrick, K.; Rubio, M. A. T.; Alfonzo, J. D. A tRNA methyltransferase paralog is important for ribosome stability and cell division in Trypanosoma brucei. Sci. Rep. 2016, 6, 21438.

174. Xu, L.; Liu, X.; Sheng, N.; Oo, K. S.; Liang, J.; Chionh, Y. H.; Xu, J.; Ye, F.; Gao, Y.-G.; Dedon, P. C.; Fu, X.-Y. Three distinct 3-methylcytidine (m3C) methyltransferases modify tRNA and mRNA in mice and humans. J. Biol. Chem. 2017, jbc.M117.798298.

175. Duechler, M.; Leszczyńska, G.; Sochacka, E.; Nawrot, B. Nucleoside modifications in the regulation of gene expression: focus on tRNA. Cell. Mol. Life Sci. 2016, 73, 3075– 3095.

176. Watanabe, K.; Oshima, T.; Saneyoshi, M.; Nishimupa, S. Replacement of ribothymidine by 5-methyl-2-thiouridine in sequence GTψC in tRNA of an extreme thermophile. FEBS Lett. 1974, 43, 59–63.

177. Sprinzl, M.; Horn, C.; Brown, M.; Loudovltch, A.; Steinberg, S. Compilation of tRNA sequences and sequences of tRNA genes. Nucleic Acids Res. 2005, 26, 148–153.

178. Kowalak, J. A.; Dalluge, J. J.; McCloskey, J. A.; Stetter, K. O. The Role of Posttranscriptional Modification in Stabilization of Transfer RNA from Hyperthermophiles. Biochemistry 1994, 33, 7869–7876.

179. Helm, M.; Giegé, R.; Florentz, C. A Watson-Crick base-pair-disrupting methyl group (m1A9) is sufficient for cloverleaf folding of human mitochondrial tRNA(Lys). Biochemistry 1999, 38, 13338–13346.

180. Beuning, P. J.; Musier-Forsyth, K. Transfer RNA recognition by aminoacyl-tRNA synthetases. Biopolymers 1999, 52, 1–28.

181. Agris, P. F. Bringing order to translation: The contributions of transfer RNA anticodon-domain modifications. EMBO Rep. 2008, 9, 629–635.

182. Paris, Z.; Horáková, E.; Rubio, M. A. T.; Sample, P.; Fleming, I. M. C.; Armocida, S.; Lukes, J.; Alfonzo, J. D. The T. brucei TRM5 methyltransferase plays an essential role in mitochondrial protein synthesis and function. RNA 2013, 19, 649–58.

183. Grosjean, H.; de Crécy-Lagard, V.; Marck, C. Deciphering synonymous codons in the three domains of life: Co-evolution with specific tRNA modification enzymes. FEBS Lett. 2010, 584, 252–264.

184. Stuart, J. W.; Gdaniec, Z.; Guenther, R.; Marszalek, M.; Sochacka, E.; Malkiewicz, A.; Agris, P. F. Functional anticodon architecture of human tRNA(Lys3) includes disruption of intraloop hydrogen bonding by the naturally occurring amino acid 135

modification, t6A. Biochemistry 2000, 39, 13396–13404.

185. Fuller, W.; Hodgson, A. Conformation of the anticodon loop in tRNA. Nature 1967, 215, 817–821.

186. Stuart, J. W.; Koshlap, K. M.; Guenther, R.; Agris, P. F. Naturally-occurring Modification Restricts the Anticodon Domain Conformational Space of tRNAPhe. J. Mol. Biol. 2003, 334, 901–918.

187. Ashraf, S. S.; Guenther, R. H.; Ansari, G.; Malkiewicz, a; Sochacka, E.; Agris, P. F. Role of modified nucleosides of yeast tRNA(Phe) in ribosomal binding. Cell Biochem. Biophys. 2000, 33, 241–52.

188. Agris, P. F.; Narendran, A.; Sarachan, K.; Väre, V. Y. P.; Eruysal, E. The Importance of Being Modified. In Enzymes; 2017; pp. 1–50.

189. Lorenz, C.; Lünse, C.; Mörl, M. tRNA Modifications: Impact on Structure and Thermal Adaptation. Biomolecules 2017, 7, 35.

190. Cabello-Villegas, J.; Tworowska, I.; Nikonowicz, E. P. Metal ion stabilization of the U-turn of the A37 N6-dimethylallyl-modified anticodon stem-loop of Escherichia coli tRNAPhe. Biochemistry 2004, 43, 55–66.

191. Ashraf, S. S.; Ansari, G.; Guenther, R.; Sochacka, E.; Malkiewicz, A.; Agris, P. F. The uridine in “U-turn”: contributions to tRNA-ribosomal binding. RNA 1999, 5, 503– 511.

192. Väre, V.; Eruysal, E.; Narendran, A.; Sarachan, K.; Agris, P. Chemical and Conformational Diversity of Modified Nucleosides Affects tRNA Structure and Function. Biomolecules 2017, 7, 29.

193. McKenney, K. M.; Rubio, M. A. T.; Alfonzo, J. D. Binding synergy as an essential step for tRNA editing and modification enzyme co-dependence in Trypanosoma brucei. RNA 2017.

194. Wolf, J.; Gerber, a P.; Keller, W. tadA, an essential tRNA-specific adenosine deaminase from Escherichia coli. Embo J 2002, 21, 3841–3851.

195. Perrochia, L.; Crozat, E.; Hecker, A.; Zhang, W.; Bareille, J.; Collinet, B.; Van Tilbeurgh, H.; Forterre, P.; Basta, T. In vitro biosynthesis of a universal t6A tRNA modification in Archaea and Eukarya. Nucleic Acids Res. 2013, 41, 1953–1964.

196. El Yacoubi, B.; Lyons, B.; Cruz, Y.; Reddy, R.; Nordin, B.; Agnelli, F.; Williamson, J. R.; Schimmel, P.; Swairjo, M. A.; de Crecy-Lagard, V. The universal YrdC/Sua5 136

family is required for the formation of threonylcarbamoyladenosine in tRNA. Nucleic Acids Res. 2009, 37, 2894–2909.

197. Aslett, M.; Aurrecoechea, C.; Berriman, M.; Al., E. TriTrypDB: a functional genomic resource for the Trypanosomatidae. Nucleic Acids Res. 2010, 38, D457-62.

198. Lin, C.; Smibert, P.; Zhao, X.; Hu, J. F.; Ramroop, J.; Kellner, S. M.; Benton, M. A.; Govind, S.; Dedon, P. C.; Sternglanz, R.; Lai, E. C. An extensive allelic series of Drosophila kae1 mutants reveals diverse and tissue-specific requirements for t6A biogenesis. Rna 2015, 1–16.

199. Wickstead, B.; Ersfeld, K.; Gull, K. Targeting of a tetracycline-inducible expression system to the transcriptionally silent minichromosomes of Trypanosoma brucei. Mol. Biochem. Parasitol. 2002, 125, 211–216.

200. Byrne, R. T.; Waterman, D. G.; Antson, A. A. Enzyme-RNA Substrate Recognition in RNA-Modifying Enzymes. In DNA and RNA modification enzymes: Structure, Mechanism, Function and Evolution; 2009; pp. 303–327.

201. Swinehart, W. E.; Henderson, J. C.; Jackman, J. E. Unexpected expansion of tRNA substrate recognition by the yeast m1G9 methyltransferase Trm10. RNA 2013, 19, 1137– 46.

202. Spears, J. L.; Rubio, M. A. T.; Gaston, K. W.; Wywial, E.; Strikoudis, A.; Bujnicki, J. M.; Papavasiliou, F. N.; Alfonzo, J. D. A single zinc ion is sufficient for an active Trypanosoma brucei tRNA editing deaminase. J. Biol. Chem. 2011, 286, 20366–74.

203. Elias, Y.; Huang, R. H. Biochemical and structural studies of A-to-I editing by tRNA:A34 deaminases at the wobble position of transfer RNA. Biochemistry 2005, 44, 12057–65.

204. Spears, J. L.; Rubio, M. A. T.; Gaston, K. W.; Wywial, E.; Strikoudis, A.; Bujnicki, J. M.; Papavasiliou, F. N.; Alfonzo, J. D. A single zinc ion is sufficient for an active Trypanosoma brucei tRNA editing deaminase. J. Biol. Chem. 2011, 286, 20366–74.

205. Kosinski, J.; Cymerman, I. A.; Feder, M.; Kurowski, M. A.; Sasin, J. M.; Bujnicki, J. M. A “FRankenstein’s monster” approach to comparative modeling: Merging the finest fragments of Fold-Recognition models and iterative model refinement aided by 3D structure evaluation. Proteins Struct. Funct. Genet. 2003, 53, 369–379.

206. Kohn, A. B.; Sanford, R. S.; Yoshida, M.; Moroz, L. L. Parallel Evolution and Lineage-Specific Expansion of RNA Editing in Ctenophores. Integr. Comp. Biol. 2015, 55, 1111–1120.

137

207. Keegan, L. P.; Leroy, A.; Sproul, D.; O’Connell, M. A. Adenosine deaminases acting on RNA (ADARs): RNA-editing enzymes. Genome Biol. 2004, 5, 209.

208. Rubio, M. A. T.; Pastar, I.; Gaston, K. W.; Ragone, F. L.; Janzen, C. J.; Cross, G. a M.; Papavasiliou, F. N.; Alfonzo, J. D. An adenosine-to-inosine tRNA-editing enzyme that can perform C-to-U deamination of DNA. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 7821–6.

209. Durham, E.; Dorr, B.; Woetzel, N.; Staritzbichler, R.; Meiler, J. Solvent accessible surface area approximations for rapid and accurate protein structure prediction. J. Mol. Model. 2009, 15, 1093–1108.

210. Lee, B.; Richards, F. M. The interpretation of protein structures: Estimation of static accessibility. J. Mol. Biol. 1971, 55.

211. Shrake, A.; Rupley, J. A. Environment and exposure to solvent of protein atoms. Lysozyme and insulin. J. Mol. Biol. 1973, 79.

212. Jackman, J. E.; Alfonzo, J. D. Transfer RNA modifications: nature’s combinatorial chemistry playground. Wiley Interdiscip. Rev. RNA 2013, 4, 35–48.

213. Ashraf, S. S.; Ansari, G.; Guenther, R.; Sochacka, E.; Malkiewicz, a; Agris, P. F. The uridine in “U-turn”: contributions to tRNA-ribosomal binding. RNA 1999, 5, 503– 511.

214. Motorin, Y.; Helm, M. TRNA stabilization by modified nucleotides. Biochemistry 2010, 49, 4934–4944.

215. Kawai, G.; Yamamoto, Y.; Watanabe, T.; Yokoyama, S.; Kamimura, T.; Masegi, T.; Sekine, M.; Hata, T.; Iimori, T.; Miyazawa, T. Conformational Rigidity of Specific Pyrimidine Residues in tRNA Arises from Posttranscriptional Modifications That Enhance Steric Interaction between the Base and the 2ʹ-Hydroxyl Group. Biochemistry 1992, 31, 1040–1046.

216. Murakami, H.; Ohta, A.; Suga, H. Bases in the anticodon loop of tRNA(Ala)(GGC) prevent misreading. Nat. Struct. Mol. Biol. 2009, 16, 353–358.

217. Lustig, F.; Boren, T.; Guindy, Y. S.; Elias, P.; Samuelsson, T.; Gehrke, C. W.; Kuo, K. C.; Lagerkvist, U. Codon discrimination and anticodon structural context. Proc. Natl. Acad. Sci. U. S. A. 1989, 86, 6873–6877.

218. Lustig, F.; Borén, T.; Claesson, C.; Simonsson, C.; Barciszewska, M.; Lagerkvist, U. The nucleotide in position 32 of the tRNA anticodon loop determines ability of anticodon UCC to discriminate among glycine codons. Proc. Natl. Acad. Sci. U. S. A. 138

1993, 90, 3343–7.

219. Maehigashi, T.; Dunkle, J. a; Miles, S. J.; Dunham, C. M. Structural insights into +1 frameshifting promoted by expanded or modification-deficient anticodon stem loops. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 12740–5.

220. Olejniczak, M.; Uhlenbeck, O. C. tRNA residues that have coevolved with their anticodon to ensure uniform and accurate codon recognition. Biochimie 2006, 88, 943– 50.

221. Jaroensuk, J.; Atichartpongkul, S.; Chionh, Y. H.; Hwa Wong, Y.; Liew, C. W.; McBee, M. E.; Thongdee, N.; Prestwich, E. G.; DeMott, M. S.; Mongkolsuk, S.; Dedon, P. C.; Lescar, J.; Fuangthong, M. Methylation at position 32 of tRNA catalyzed by TrmJ alters oxidative stress response in Pseudomonas aeruginosa. Nucleic Acids Res. 2016, 44, 10834–10848.

222. Chan, C. T. Y.; Dyavaiah, M.; DeMott, M. S.; Taghizadeh, K.; Dedon, P. C.; Begley, T. J. A quantitative systems approach reveals dynamic control of tRNA modifications during cellular stress. PLoS Genet. 2010, 6, 1–9.

223. Chan, C. T. Y.; Pang, Y. L. J.; Deng, W.; Babu, I. R.; Dyavaiah, M.; Begley, T. J.; Dedon, P. C. Reprogramming of tRNA modifications controls the oxidative stress response by codon-biased translation of proteins. Nat. Commun. 2012, 3, 937.

224. Begley, U.; Dyavaiah, M.; Patil, A.; Rooney, J. P.; DiRenzo, D.; Young, C. M.; Conklin, D. S.; Zitomer, R. S.; Begley, T. J. Trm9-Catalyzed tRNA Modifications Link Translation to the DNA Damage Response. Mol. Cell 2007, 28, 860–870.

225. Rubio, M. A. T.; Alfonzo, J. D. Editing and modification in trypanosomatids: the reshaping of non-coding RNAs. In; 2005; pp. 71–86.

226. Wohlgamuth-Benedum, J. M.; Rubio, M. A. F.; Paris, Z.; Long, S.; Poliak, P.; Lukeš, J.; Alfonzo, J. D. Thiolation controls cytoplasmic tRNA stability and acts as a negative determinant for tRNA editing in mitochondria. J. Biol. Chem. 2009, 284, 23947–23953.

227. García-Mauriño, S. M.; Rivero-Rodríguez, F.; Velázquez-Cruz, A.; Hernández- Vellisca, M.; Díaz-Quintana, A.; De la Rosa, M. A.; Díaz-Moreno, I. RNA Binding Protein Regulation and Cross-Talk in the Control of AU-rich mRNA Fate. Front. Mol. Biosci. 2017, 4.

139

Appendix A: Supplementary figures

140

Figure A.1 Complex formation of TbADAT2/3 and TbTrm140.3

(A) Immunoprecipitation with anti-TbADAT2 antibodies and reactions containing no proteins, TbADAT2/3, TbTrm140 or both enzymes mixed together prior to IP, as indicated by (-) and (+). The samples were analyzed by Western blots with anti-TbADAT3 (α-ADAT3), anti-TbADAT2 α-ADAT2) or anti-TbTRM140a (α-Trm140). Rec. refers to recombinant proteins purified from E. coli (B) Similar experiments as in (A) but instead performed with anti-TbTRM140a antibodies, followed by Western analysis as above. (C) Total protein extracts from wild-type T. brucei IP as in (B) with anti-TbTRM140a antibodies, followed by Western analysis as above. The flow- through (FT), washes 1 and 7 (W1 and W7) of the beads and the input (Inp.) are shown in the right panels. (D) Western blots in which all proteins were separated individually by SDS-PAGE and detected with the same antibodies as in (A) and (B), serving as controls for antibody cross- reactivity. All panels are representative of at least 5 independent experiments with identical results.

3 Figure A.1 is published in: Rubio, M.A., Gaston, K.W., McKenney, K.M., Fleming, I.M., Paris, Z., Limbach P.A., Alfonzo J. D. Editing and methylation at a single site by functionally interdependent activities. 2017. Nature. 542(7642):494-497.

141

Figure A.2 Predicted RNA-binding residues of TbTrm140.

(A) Output from BindN predicting RNA-binding residues in the 341 amino acid sequence of Trm140 (Wang and Brown 2006). The + symbols denote likely RNA binding residues. (B) Clustal 2.1 multiple sequence alignment of S. cerevisiae and T. brucei Trm140. Putative motifs (I-VI) characteristic of class 1 Rossmann-fold methyltransferases are indicated above the sequence and key conserved residues are highlighted in black. Gray highlighted regions in both the BindN and alignment correspond to the N-terminal and C-terminal deletion mutants, ∆S2- G17 and ∆I320-S340, respectively.

142

143

144

Appendix B: Materials and methods

145

B1: Plasmid DNA purification

Buffers:

Solution I: 25 mM Tris pH 8.0, 10 mM EDTA

Solution II: 0.2 N NaOH, 1% SDS

Solution III: 14.76 g 5M KOAc, 5.75 ml glacial acetic acid, Fill to 50 ml w/ddH2O

1. Inoculate 2 mL of 2xYT containing appropriate antibiotic with desired E. coli strain. Incubate shaking overnight at 37 ºC.

2. Next day, pour culture in to 2 mL microcentrifuge tube.

3. Spin cells at 5,000 rpm for 10 mins. Pour off supernatant.

4. Suspend pellet in 150 µL solution I and incubate at room temperature for 5 mins.

5. Add 150 µL solution II to sample and invert to mix. Incubate at room temperature for 5 mins.

6. Add 150 µL solution III to sample and invert to mix. Incubate at room temperature for 5 mins.

7. Spin cells at maximum speed for 15 mins to spin down cell debris and bulk proteins.

8. Carefully collect and transfer supernatant to fresh 1.5 mL microcentrifuge tube, avoiding the pellet.

9. Add 100 µL Tris-phenol and invert to mix.

10. Spin cells at maximum speed for 15 mins.

11. Collect top aqueous layer and transfer to new 1.5 mL microcentrifuge tube containing 900 µL EtOH. Mix well by inverting.

12. Pellet DNA at maximum speed for 30 mins.

13. Remove EtOH and dry pellet for 20 mins at room temperature.

14. Suspend DNA pellet in 30 uL ddH2O. Use 10 uL for a 30 uL restriction enzyme digest. 146

B2: Transformation of E. coli

1. Thaw 25 µL E. coli competent cells for each ligation on ice for 15 min.

2. Label competent cell tubes accordingly.

3. Add 5 µL of ligation to the appropriate competent cell tube.

4. Incubate tubes on ice for at least 20 min.

5. Heat shock cells for 2min at 37 °C.

6. Incubate tubes for 2-5 min on ice.

7. Add 50 µl of plain LB kept at room temperature to each tube. Transfer near the flame to prevent contamination.

8. Incubate tubes for 45 min at 37 °C.

9. Plate cells with spreader on appropriate media. Plates should be incubated overnight at 37 °C.

147

B3: Recombinant protein purification

Buffers:

Binding: 100mM Tris pH 8.0, 100mM NaCl, 0.1% NP40, 1x protease inhibitor

Wash: 100mM Tris pH 8.0, 100mM NaCl, 20mM Imidazole (TbADAT2/3)/50mM Imidazole (TbTrm140)

Elution: 100mM Tris pH8.0, 100mM NaCl, 500mM Imidazole

Dialysis: 100mM Tris pH8.0, 100mM NaCl, 0.2mM EDTA, 0.5mM MgCl 2, 2mM DTT

Day 1:

1. Start 10 ml overnight liquid culture from single colony and pre-warm 1.5 L of 2xYT media without antibiotics.

Day 2:

1. Inoculate 1.5 L 2xYT plus the appropriate antibiotic with 10 ml of starter culture.

2. Allow culture to grow to OD600 of 0.6 to 0.9 (this usually takes about 2 hours).

3. Quickly cool culture to room temp by swirling the flask over ice and induce protein expression with 0.5mM IPTG.

4. Induce culture at room temperature shaking overnight (~18hrs).

Day 3:

1. Harvest 1.5 L prep in 2 x 500 ml bottles at 5000 rpm for 10 minutes. You will have to fill and spin bottles multiple times to be able to pellet entire 1.5 L prep . ENTIRE procedure must be done on ICE and gloves must be worn at all times

2. Suspend each pellet in ~160 ml of binding buffer, 1x protease inhibitor (PI), and 1% NP40 and transfer 30 ml each to 4 x 50 ml conical tubes.

3. Sonicate sample in 50 ml conical tubes to lyse the cells.

4. Pool sonicated cell extract into 1 x 250 ml bottle. Pellet cellular debris at 10,000 rpm for 15 min. 148

5. Pour supernatant into 35 ml centrifuge bottles (you will need 5-6 bottles) and spin again at 18,000 rpm for 25 min.

6. Pour supernatant from all 5 bottles into 4 x 50 ml clean conical tubes.

7. Divide 1mL Ni2+ charged beads evenly into all 4 tubes. Let bind for 1.5 hours with rotation in the cold room.

8. Spin down beads at 2000-3000 rpm for 10 minutes. Pipette off supernatant.

9. Add 30 ml of 1x Wash buffer to each tube and gently invert 10 times.

10. Spin down beads at 2000-3000 rpm for 10 minutes. Pipette off supernatant.

11. Pour beads into column and wash with ~200 ml of wash buffer.

12. Elute in 6 x 1ml fractions

13. Pool fractions (usually fractions 1-3 have the highest yield) and dialyze overnight at 4oC in 500 ml dialysis buffer.

149

B4: Protein concentration

1. Rinse out the Amicon® Ultra 4mL or 15 mL filter conical with ddH2O. If it is a brand- new filter conical add 1mL of ddH2O and spin at 2,300 rpm for 10 mins to wet the filter.

2. Rinse the filter conical with 0.1% NP40 and rinse thoroughly 3 times with ddH2O. 3. Add protein sample to top filter piece and spin at 2,300 rpm for 20 mins. This will concentrate 5 mL to 1 mL final volume if no glycerol has been added. If there is glycerol in the protein sample it will filter at a slower rate, usually 5 mL to 2.5 mL in 20 mins. 4. Concentrate protein to desired volume, usually 200 μl to 1 mL. Save sample in 1.5 microcentrifuge tube, rinse out filter with 100 μl of buffer, and add to the sample.

5. Rinse filter with ddH2O and store in 20% EtOH with label indicating which protein was used.

150

B5: Western blot

Gel mix for 10% Tris-glycine acrylamide gel:

Top/Stacking (6 mL) 1 mL 30% acrylamide 0.25 mL Tris pH 6.8 0.06 mL 10% SDS 0.06 mL APS 0.006 mL TEMED

Bottom/Running (10 mL) 3.3 mL 30% acrylamide 2.5 mL Tris pH 8.8 0.1 mL 10% SDS 0.1 mL APS 0.01 mL TEMED

Buffers

10x SDS-PAGE Buffer: 250 mM Tris-base, 25 M glycine, 1% SDS

Western Transfer Buffer: 1x SDS-PAGE, 20% MeOH

Protein separation by SDS-PAGE:

1. Pour Tris-glycine gel containing a concentration acrylamide appropriate for the expected protein size. We typically use 10% acrylamide for 25-100 kDa proteins.

2. Add 5 µL 5x SDS load dye to 20 µL of protein sample (lanes fit ~40 µL maximum).

3. Load samples along with all-blue stained marker. Run gel in 800 mL of 1x SDS-PAGE buffer for 1.5 hrs at 120 V and 45 mA (90 mA for two gels).

Transfer:

4. Remove top plate from gel and remove stacking using a razor blade.

5. Soak transfer components in Western Transfer Buffer before assembling the transfer. Can cut corner of membrane to mark the first lane. Assemble transfer on the cassette in the following order: black side of cassette, sponge, filter paper, gel, membrane, filter paper, sponge, red/white side of cassette. The membrane should always be placed on the side facing the red positive cathode.

151

6. Place ice pack in transfer apparatus and place gel box in an ice bucket. Transfer in 800 mL Western Transfer Buffer for 2 hrs at 75 V and 200 mA.

7. Remove membrane and place face up on paper towel to dry until ready to blot. Mark ladder with colored pens in case they fade later.

Blotting:

8. Make 5% milk mixture by mixing 1 gram of dried milk with 20 mL PBS/Tween. Block the membrane with 5 mL milk for at least 1 hr shaking at room temperature.

9. Add primary antibody to 3 mL milk and incubate at least 1.5 hrs shaking at room temperature. We typically use 1:3000 primary antibody:milk (1 µL in 3 mL) for recombinant proteins, down to 1:200 (15 µL in 3 mL) for native T.brucei extract.

10. Save primary antibody milk in 2 mL microcentrifuge tubes in -20 °C. This may be reused ~5 times.

11. Wash membrane 3x 5 mins with ~5 mL PBS/Tween.

12. Add secondary antibody to 3 mL milk and incubate at least 1 hr shaking at room temperature. We use a ratio of 1:3000-10,000 secondary antibody:milk depending on the antibody.

13. Wash membrane 3x 5 mins with ~5 mL PBS/Tween. Discard last wash and place membrane face up on saran wrap.

14. To visualize mix 0.5 mL of western substrate with 0.5 mL of peroxidase and pipette evenly onto membrane. Incubate at room temperature 5 mins, do not shake. Expose membrane using chemidoc.

152

B6: Affinity purification of polyclonal antibody

Protein separation by SDS-PAGE

1. Run 100 ug protein using large, one-well comb (with small marker lane) and transfer to MeOH activated PVDF membrane

Always activate PVDF membrane in MeOH for at least 10 secs anytime it has been dried before using.

2. Visualize protein with x Ponceau S stain. Incubate membrane shaking at room temperature for 10-15 mins with enough stain to cover the membrane. Remove and keep the stain. This can be re-used ~10x. De-stain with ddH20 until signal is reduced.

3. Cut the membrane as close to the protein band as possible and make sure to indicate which side of the strip is the top (ie. mark with pencil/cut bottom right corner). The procedure can be stopped at this step. Save membrane strip submersed in PBS-tween in a15 mL conical.

Blocking and Incubation

4. Block strip in a 15 mL conical with 5% milk/PBS-tween (enough to cover the strip) shaking at room temperature for 1 hr.

5. Wash the strip 3x with PBS-tween for 5 mins each.

6. After 3rd wash, incubate strip overnight at 4 oC with 1 mL of polyclonal antibody serum. Tip 15mL conical on its side and be sure to pipette serum so it’s covering the topside of membrane. If you need to get rid of any anti-his antibodies, incubate serum with His-tagged protein strip first (overnight). Then, repeat blocking, incubation and elution steps with the protein of interest.

7. The next day, transfer strip to a new conical to wash 3x with PBS-tween. Wash an additional time with 1x PBS (No Tween). Save the leftover bleed at -20 oC until you have tested your antibody.

Elution

8. Place strip face-up on piece of saran wrap.

9. Pipette 300uL 0.1M glycine pH 2.5 covering the top of the strip and let it sit for 3mins. Collect into tube containing 45 uL 1M Tris pH 8 to neutralize.

153

10. Check the pH of neutralized antibody using lithmus paper (should be ~pH8). Repeat step 9 for a total of 2 elutions.

Storage

11. Split up elution 1 into 10uL working aliquots and keep at -20 oC. It should be small enough to avoid freezing/thawing >10x. Otherwise working aliquots can also be kept at 4 oC for several months.

12. Store the leftover strip in PBS-tween. This can be re-used several times.

154

B7: Transfection of T. brucei

CytoMix Buffer:

Components Lab Stock Volume to Add Final Concentration Concentration (M) (mM) HEPES pH 7.6 0.5 2.5 ml 25 KCl 3 2 ml 120 CaCl2 1 7.5 µl 0.15 K2HPO4 pH 7.6 0.1 4.33 ml 10 KH2PO4 pH 7.6 0.1 1.34 ml 10 EDTA 0.5 200 µl 2 Glucose 0.5 600 µl 6 MgCl2 1 250 µl 5

Day 1

1. Mix all the components on a sterile conical tube.

2. Bring volume to 50 ml with autoclaved ddH2O. 3. Filter sterilize inside the biological safety cabinet. 4. Refrigerate on ice for at least 30 min before using (should be ice cold at all times). 5. Harvest 10 ml of mid-log T. brucei culture (10 – 20 x 106 cells/ml) by spinning culture at 2,300 rpm for 10 mins at 4°C. Save supernatant for the conditioned media, keep it on ice. 6. Suspend pellet in 10 ml of ice cold CytoMix buffer. 7. Spin cells at 2300 RPM for 10 min and discard the supernatant by pouring. 8. Suspend the pellet in 1 ml of ice cold CytoMix buffer. 9. Load cuvettes (0.2 cm gap) with 10-20 µg of linearized, sterile DNA, cooled on ice. We usually use two plasmid minipreps (4 ml total) pooled together, resuspended in 20µl of sterile 1 X TE. Make sure the plasmid DNA was digested with Not1 before performing the transfection. 10. Spin the sample down and leave it in ethanol. Remove ethanol and air dry inside the biological safety cabinet. 11. Add 0.5 ml of cell suspension to the cuvettes and mix by pipetting up and down. 12. Electroporate cells one cuvette at a time. BTX settings: 1600 V, 25 Ω, 50 µF. 13. Suspend cells in 5 ml of conditioned media and incubate at 27 °C overnight shaking. Conditioned media is 50/50 old/fresh media without addition of antibiotics. 155

Day 2

14. Add 5 ml of SDM-79 containing the appropriated antibiotic at 2x the concentration to the culture (bringing final volume to 10 ml). 15. Add 1.5 mL of the culture (with antibiotics) to the first row of wells on the 24 well plate. Add 1 mL of the SDM-79 media (with antibiotics) to the second and third rows. Add 0.5 ml of fresh SDM-79 media (with antibiotics) to the fourth row. 16. Perform a serial dilution by transferring 0.5 ml starting from the first well to the second, third, and fourth wells. The fourth well should end up with 1 ml final volume. 17. Write identifying information on the lid of the plate, including date, strain, insert and antibiotics. 18. Wrap plate in plastic wrap and incubate it at 27 °C without shaking for ~ one week. 19. Check cells daily under the microscope. 20. Dilute the fourth row of wells 1:2 after one week of recovery by adding 1 ml of fresh SDM-79 media with antibiotics. Do not remove the old media. 21. Clones that recover can be chosen 10 or 12 days after the plating by transferring the 2 ml from the last row wells into flasks containing 8 ml of fresh SDM-79 with antibiotics.

156

B8: Co-immunoprecipitation

Preparation of total cell extract

1. For binding analysis of whole cell extracts, grow up 50 ml of procyclic T. brucei 29-13 cells to 7x106 cells/ml and pellet at 3,000 rpm for 20 mins at 4oC.

2. Wash the pellet twice in PBS then suspend in 400µl PBS with 1.0% NP40 and protease inhibitor cocktail (Sigma-Aldrich).

3. Sonicate the suspension using a microprobe for three times 30 second pulses with the settings at output 4.2, duty cycle 25.

4. Spin the sample for 10 mins at 5,000 RPM to remove cellular debris.

Preparation of total cell extract

5. Wash 10µl magnetic protein-G Dynabeads (Invitrogen) with 100 µl PBS (0.1 M sodium phosphate, 0.15 M NaCl, pH 7.2). Vortex in short bursts at level six to mix. Apply magnet to the tube to collect beads on the side and easily remove the washes.

6. Incubate the washed beads with antigen-purified polyclonal antibody in 15 µl PBS for 2 hrs on ice.

7. Remove the antibody and wash the beads with 100 µl PBS.

8. Incubate the recombinant proteins or total cell extract with antibody-bound beads for 40 mins on ice.

9. Wash the beads seven times with 100µl PBS with 1.0% NP40.

10. Elute the protein components by boiling beads for 15 mins at 90 oC in SDS loading buffer and analyze by western blot.

157

B9: Electrophoretic mobility shift assay

Buffers and gel mix:

10x TBE buffer: 670 mM Tris- base, 220 mM Boric Acid, 5 mM EDTA

4% acrylamide non-denaturing gel (50 mL): 5 mL of 40% acrylamide 5 mL of 10x TBE 300 μl 10% APS 40 μl TEMED 40 mL of ddH2O

10x HKM assay buffer: 500 mM Hepes 7.5 (or 8.0), 10 mM MgCl2, 50 mM KCl

Standard 1x reaction 2 µl 10x HKM buffer 2 µl tRNA (diluted) µl protein Up to 20 µl ddH2O

Note: This protocol is for a 6 reaction assay. Two can be run at a time on a 12 lane 4% polyacrylamide gel.

1. Slowly pour 30ml of the 4% acrylamide gel mix into the top of two glass plates secured with at least one clamp on either side and one on the bottom.

The size plates we use: front plate is ~20cmx16cm, back plate is ~20cmx20cm. We use 1.5mm spacers and with a 1.5mm 12 lane comb.

2. Allow gel to polymerize for ~1hr.

3. Dilute protein sample to appropriate working concentration. Aim for a range of concentrations that will generate enough points in the lower portion and upper portion of a hyperbolic curve. You will likely need to determine saturating concentration empirically.

158

Example dilutions and concentration ranges:

Reaction Volume of ddH2O Volume of Protein Final [Protein] 1 16 µl - - 2 14.75 µl 1.25 µl 0.044 µM 3 13.5 µl 2.5 µl 0.088 µM 4 11.0 µl 5.0 µl 0.176 µM 5 6.0 µl 10.0 µl 0.352 µM 6 - 16.0 µl 0.563 µM

**Keep protein sample on ice to prevent degradation**

4. Heat 12 µl of 2.5nM P32 labeled tRNA with 24 µl in 70oC heat block for 3 min.

5. Cool RNA at room temp for 1 min.

6. Add equal volume (12 µl) 10x HKM buffer to the tRNA and incubate at 37 oC for 15 mins to refold the tRNA.

7. Add 4 μl of tRNA/HKM buffer mix to each of the 6 reactions containing diluted protein.

8. Incubate on ice for 30 minutes to bind.

9. Add 5 μl of 50% Glycerol to each tube (add dye to at least one lane to monitor progress).

10. Pipette gently 3 times and Load all 25 μl in gel.

11. Remove only the bottom spacer and run gel at 80 volts for 2.5 hrs using 1x TBE running buffer. Make sure the bottom of the gel is submerged in the running buffer and there are no air bubbles where the bottom spacer used to be. You will need to tip the gel slightly to fill the bottom space with buffer.

**Do not exceed 100 volts**

12. Place gel on filter paper and cover with plastic wrap. Dry for 1.5 hrs with heat and 0.5 hrs without heat.

13. Expose to a screen overnight.

14. Calculate percent bound using the equation: Bound = tRNAcomplex/(tRNAfree + tRNAcomplex) 159

B10: A to I deamination and methylation assays

RNA transcription and purification

1. Transcribe RNA using α-32P-labeled CTP and T7 transcription kit (ie. Ambion MEGAshortscriptTM).

2. Add 20 µL urea load dye to 20 µL transcription reaction. Heat for 5 mins at 80 °C then ice 5 mins. For tRNA, separate on an 8% acrylamide urea 1x TBE gel at 150V for 1 hr.

3. Open plates, wrap in plastic, and expose to PhosphorImager screen for 1-5 mins. Cut out the band nearest the Xylene Cyanol dye (should be ~76 nucleotides).

4. Elute in 500 µl 0.3M NaOAc at room temperature overnight (do a 2nd elution next day).

5. Add 1 mL EtOH (200 proof), 50 µl NaOAc, and 2 µl glycogen. Mix well and spin for 30 mins at max speed to precipitate (13.2K rpm). Suspend pellet in 100ul TE and use 0.5-1µl for scintillation counting.

Deamination assay

10x Deaminase Buffer 1x 10µL Deaminase Assay 400 µl 1M Tris pH 7.9 25,000 CPM/µl of RNA (~1 µL) 50 µl 1M MgCl2 1 µl 10x Deaminase buffer 200 µl 0.5M DTT 0.5 µl 10mM SAM 350 µl ddH2O µL protein Up to 10µL with ddH2O

Methylation assay

10x Methylase Buffer 1x 10µL Methylase Assay 250 µl 1M Tris pH 7 25,000 CPM/µl of RNA (~1 µL) 25 µl 1M MgCl2 1 µl 10x Mtase buffer 50 µl 5M NaCl 0.5 µl 10mM SAM 0.5 µl 0.5M EDTA µL protein 725 µl ddH2O Up to 10µL with ddH2O

Start off with a range of protein concentrations such as 0, 0.5µM, 1µM, 5µM over a range of time 1-3 hrs.

160

1. Fold RNA by heating for 3 mins at 75 °C, cool 1 min at room temp, add 10x methylase buffer, then heat for 10 mins at 37 °C.

2. Set up assay and incubate at 27 °C for T. brucei enzymes or 37 °C for E. coli enzymes for chosen times.

3. At each timepoint, quench reaction by adding 50 µL ddH2O and 30 µL saturated phenol:chloroform:isoamyl alcohol (25:24:1). Mix well and spin for 10 mins at max speed to separate phases.

4. Carefully pipette off the top aqueous phase (contains hot tRNA) and add to a labeled microcentrifuge tube containing 250 µL EtOH (200 proof), 6 µL 3M NaOAc, and 2 µL glycogen. Mix well and spin for 30 mins at max speed to precipitate.

5. Carefully remove the EtOH by pipetting. Allow the remaining pellet to air dry (Note: a Geiger counter can be used to ensure no loss of sample).

6. Digest the tRNA to single nucleotides by adding 5 µL of 1x Nuclease P1 buffer and 1 µL of nuclease P1 to the pellet. Incubate at 37 °C overnight.

7. Dry sample using a speed vac (Savant) on high heat. Suspend digested tRNA pellet in 3 µL ddH2O.

8. Spot 1 µL of sample on thin layer chromatography (TLC) plate (7 cm tall) and develop in an appropriate solvent system.

Note: in the case of C to m3C, Solvent C (phosphate buffer, ammonium sulfate, and n- propanol in a 100:60:2 (v:w:v) ratio) is used (see below).

0.1M NaP TLC Buffer C 46.3 mL 1M Na2HPO4 31 mL 0.1M NaP pH 6.8 53.7 mL 1M Na2HPO4 300 µl N-propanol Adjust pH to 6.8 18.5 g (NH4)2SO4 Up to 50 mL ddH2O

9. Allow the solvent front to migrate to the top of the plate (within 1 mm of the top edge of the plate). Remove the plate from the tank and allow it to air dry (or use hairdryer). Wrap the TLC plate in plastic wrap and expose to a PhosphorImager screen. A sample with high specific activity can be visualized after 15–30 minutes but for best results, expose the TLC to the screen overnight

10. Compare the results to cold nucleotide markers and/or published TLC maps.

161

B11: Kinetic determination of binding affinity

1. Perform deamination or methylation assays as detailed in Appendix B10 with a saturating amount of enzyme to the labeled tRNA substrate (2.5nM) and take aliquots at increasing timepoints (0, 2, 5, 10, 20, 30, 40, 60, 90, and 120 mins). 2. Repeat this over a range of concentrations of enzyme from 10 nM to 2.5 µM. 3. Calculate the fraction of inosine or methylated cytosine produced using the equation, pI/(pA + pI) for deamination or pm3C/(pC + pm3C) for methylation. Plot the fraction of product as a function of time and fit to a single exponential curve [f = a(1 - e-kt)], where f represents product formed, a denotes product formed at the end point of the reaction, k signifies kobs and t is time. Plot the resulting kobs values against the protein concentration and fit to a single ligand binding isotherm. Determine the Kdapp by nonlinear regression using Sigmaplot.

162

B12: Primer extension analysis of methylation

Hybridization buffer: 30 mM Tris pH 8.3, 450 mM KCl

1. Add radiolabeled primer at 300,000 cpm (~1 µL) to 25 µg total RNA. Bring up to 50 µL total volume with ddH2O.

2. Precipitate each reaction with 5 µL 3M NaOAC, 140 µL EtOH, 1 µL glycogen spinning at max speed for 30 mins.

3. Wash pellet with 300 µL of 80% EtOH and spin at max speed for 10 mins.

4. Dry pellet at room temp for 20 mins.

5. Suspend pellet in 10 µL ddH2O and add 5 µL hybridization buffer. Heat at 90 °C for 5 mins.

6. Incubate at 37 °C for 90 mins.

7. Equilibrate tubes at 42 °C while preparing the reverse transcription reaction.

1x RT reaction 1.35 µl 1M Tris pH 8.3 3.6 µl 25 mM MgCl2 0.3 µl 200 mM DTT 3.0 µl 10 mM dNTPs 6.25 µl ddH2O 0.5 µl Reverse transcriptase *Always add RT last*

8. Add 15 µl of the RT reaction to each tube and incubate at 42 °C for 3 hrs.

9. Add 1.5 µl 0.5 M EDTA and 3.2 µl 2 M NaOH and incubate at 65 °C for 30 mins.

10. Precipitate reaction with 1 µl glycogen, 10 µl 7.5 M ammonium acetate, and 150 µl EtOH (in that order). This can be left overnight at -20 °C. Spin 30 mins at max speed to pellet.

11. Wash pellet with 300 µL of 80% EtOH and spin at max speed for 10 mins.

12. Suspend pellet in 3 µL TE, add 2 µL urea load dye, and load sample onto standard sequencing gel.

163