Trypanosoma Brucei Trna Editing Deaminase: Conserved Deaminase Core, Unique Deaminase Features

Trypanosoma brucei tRNA Editing Deaminase: Conserved Deaminase Core, Unique Deaminase Features

DISSERTATION

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

Jessica Lynn Spears

Graduate Program in Microbiology

The Ohio State University

2011

Dissertation Committee:

Dr. Juan Alfonzo, Advisor

Dr. Michael Ibba

Dr. Chad Rappleye

Dr. Venkat Gopalan

Jessica Lynn Spears

2011 Abstract

Inosine, a guanosine analog, has been known to function in transfer RNAs

(tRNAs) for decades. When inosine occurs at the wobble position of the tRNA, it functionally expands the decoding capability of a single tRNA because inosine can base pair with cytosine, adenosine, and uridine. Because inosine is not genomically encoded, essential enzyme(s) are responsible for deaminating adenosine to inosine by a conserved zinc-mediated hydrolytic deamination mechanism. Collectively called ADATs

(Adenosine deaminase acting on tRNA), these enzymes are heterodimeric in eukaryotes and are comprised of subunits called ADAT2 and ADAT3. ADAT2 is presumed to be the catalytic subunit while ADAT3 is thought to be just a structural component. Although these enzymes are essential for cell viability and their products (inosine-containing tRNAs) have a direct effect on translation, little is known about ADAT2/3. Questions such as what is ADAT3’s role in enzyme activity, how many zinc ions are coordinated, how many tRNAs are bound per heterodimer per catalytic cycle and what is the nature of the tRNA binding domain were all open questions until this work.

The focus of chapter two is on the specific contributions of each subunit to catalysis. Steady state kinetic measurements with a series of ADAT2/3 mutants show that ADAT3 contributes directly to catalysis via participation in inter-subunit zinc coordination. A molecular model is presented that is corroborated by ICP (inductively coupled plasma) studies which further show that unlike other multimeric deaminase, one

ii zinc ion is necessary and sufficient for deaminase activity. This effectively means that

ADAT2/3 has one complete active site and a pseudo-active site (which is not complete because of a naturally missing catalytic glutamate). Furthermore, electrophoresis mobility shift assays (EMSAs) show a predicted 1:1 stoichiometry of tRNA: ADAT2/3 heterodimer.

In chapter three, the focus shifts from the catalytic site and moves to the tRNA binding domain(s). In silico work predicted an RNA binding domain away from the active site and at the C-terminus of ADAT2 (KR-domain). A combination of enzyme kinetics and EMSAs show that not only are the positively charged arginines and lysines critical for substrate binding but also the pseudo-active site is critical for binding. Moreover, these two binding sites work cooperatively to bind and position a single tRNA for catalysis.

The final study presented in chapter four examines inosine formation away from

1 the anticodon. In archaeal tRNAs, A57 in the TΨC loop is first methylated (m A57) by the

1 SAM-dependent TrmI and then deaminated (m I57) by an unknown enzyme(s). Using a combination of bioinformatics and protein purification via column chromatography, progress was made towards the final goal of identifying the enzyme responsible for m1A to m1I activity.

In summary this dissertation presents interesting findings with respect to tRNA editing deaminases and fills important data gaps including new idea about the active site works and how protein binds its tRNA substrate.

iii

Acknowledgments

I would like to acknowledge and thank first and foremost my advisor, Juan “Sir”

Alfonzo, without whom most of the work presented herein would not have been possible.

His constant guidance, life lessons, and invaluable discussions are appreciated and will be greatly missed. He demands nothing less than excellence from himself and those around him and I am a stronger person and scientist because of it. And of course there are no words to express the deepest of gratitude to the wonderful Mary Anne Rubio, to whom I will be forever indebted for becoming my Columbus mom, assisting in technical challenges and keeping the lab a functional and (mostly) sane place to be. I am also very thankful to have had a very thoughtful dissertation committee. Their suggestions, advice, and willingness to help me succeed will not soon be forgotten.

To all of my labmates (especially Zdenek Paris, Kirk Gaston, Ashlie Tseng, and

Jessica Wohlgamuth-Benedum) who made working long hours in a windowless building enjoyable, I thank you. To all of my labmates (namely Zdenek, Paul Sample, and Ian

Fleming) who have and will continue to take the lab to higher heights, it has been a pleasure working with all of you. I also owe a huge thank you to one of my dearest friends, Michael Carter. He has been a very valuable sounding board for many ideas, scientific and otherwise.

I would also like to acknowledge my parents Donna and Steve Spears (Dee Dee and Sparky) and my sister, Nicole Brei, who have been supportive not just during my graduate studies but in all my life endeavors. They perhaps still don’t quite understand

iv what a Ph.D. is or why tRNA editing is important but they never stopped believing in me or reminding me that I could do it when I wanted to just give up. And last but certainly not least I would like to acknowledge my very loving and supportive partner, Adriane Brown.

She not only had to bear the brunt of my moodiness and stress during my dissertation process but also had to do it while teaching and writing her own dissertation. She has been a constant source of love and support through it all and to her I will be forever grateful.

Vita

2005 ...... B.S. Biology, University of Wisconsin-Eau Claire

2005 to present ...... Graduate Teaching and Research Associate,

Department of Microbiology, The Ohio State

University

Publications

Peer-reviewed journal articles

Spears, JL, Rubio, MA, Gaston, KW, Wywial, E, Strikoudis, A, Papavasiliou, FN, Bujnicki, JM, and Alfonzo, JD. (2011) One Zinc ion is sufficient to create an active Trypanosoma brucei heterodimeric tRNA editing deaminase. JBC (Apr 20 Epub ahead of print).

Ragone, F*, Spears, JL*, Wohlgamuth, JM, Kreel, N, Rubio, MA, Papavasiliou, FN and Alfonzo, JD. (2011) The C-terminal end of T. brucei adenosine deaminase acting on tRNA plays a critical role in tRNA binding and editing. RNA (In Press).

*These authors contributed equally to the work.

Gaston, KW, Rubio, MA, Spears, JL, Pastar, I, Papavasiliou, FN, and Alfonzo, JD. (2007) C to U editing at position 32 of the anticodon loop precedes tRNA 5’ leader removal in trypanosomatids. Nucleic Acid Res. 35(20): 6740-6749.

Book Chapters

Spears, JL, Rubio, MA, Sample, PJ, and Alfonzo JD. (2011) Working title: tRNA biogenesis and processing in Trypanosome review (In press).

Spears, JL, Gaston, KW, and Alfonzo JD. (2011) Analysis of tRNA Editing in Native and Synthetic Substrates. Methods Mol Biol. 2011; 718: 209-26.

Fields of Study

Major Field: Microbiology

vii

Table of Contents

Page

Abstract ...... ii Acknowledgements ...... iv Vita ...... vi Publications ...... vi Field of Study ...... vii List of Tables ...... xi List of Figures ...... xii

Chapters:

1. Introduction

1.1 Brief introduction to tyrpanosomes ...... 1 1.2 Medical significance of trypanosomes ...... 2 1.3 tRNA processing in trypansomes ...... 3 1.3.1 Trimming of 5’ and 3’ ends to generate full length tRNA ...... 5 1.3.2 Intron removal ...... 6 1.3.3 tRNA export to the cytoplasm and mitochondrial import ...... 6 1.4 RNA editing ...... 7 1.4.1 Insertion/deletion editing ...... 8 1.4.2 Nucleotide substitution editing by deamination ...... 9 1.4.3 mRNA editing: APOBEC-1 ...... 11 1.4.4 mRNA editing: ADAR1 and ADAR2 ...... 13 1.4.5 tRNA editing ...... 13 1.4.5.1 C to U tRNA editing ...... 14 1.4.5.2 A to I tRNA editing...... 17

2. A Single Zinc Ion is Sufficient for an Active Trypanosoma brucei tRNA Editing Enzyme: Insights into the Function of Multimeric Deaminases

2.1 Introduction ...... 21 2.2 Results …...... 26 2.2.1 Mutation of a key Zn2+-coordinating cysteine in TbADAT2 still yields an active enzyme ...... 26

viii

2.2.2 A single Zn2+ is necessary and sufficient for ADAT2/3 activity ...... 34 2.2.3 TbADAT2 forms inactive homodimers in solution despite coordinating Zn2+ ...... 38 2.2.4 TbADAT2/3 binds one tRNA per heterodimer ...... 41 2.3 Discussion ...... 41

3. TbADAT2/3 Has Two Distinct tRNA Binding Domains that Act Cooperatively

3.1 Introduction……… ...... 54 3.2 Results…………… ...... 58 3.2.1 Recombinant TbADAT2/3 stably binds tRNA ...... 58 3.2.2 Single-turnover kinetics validates G34-containing tRNA substrate for binding studies ...... 60 3.2.3 TbADAT2 has a predicted RNA binding motif at its C-terminal end important for tRNA binding ...... 63 3.2.4 The C-terminus of TbADAT2 is crucial for enzyme activity ...... 75 3.2.5 Pseudo-active site also contributes to substrate binding ...... 78 3.2.6 Pseudo-active site and KR-domain act cooperatively to bind a single tRNA ...... 83 3.3 Discussion………...... 86

4. Partial Purification of the Archaeal m1A to m1I tRNA Deaminase Enzyme

4.1 Introduction ...... 93 4.2 Results…...... 94 4.2.1 Thermococcus kodakaraensis can deaminate m1A to m1I ...... 94 4.2.2 Bioinformatic searches to identify candidate deaminases ...... 95 4.2.3 Native protein purification via column chromatography (pilots) ... 98 4.2.4 Native protein purification scheme ...... 104 4.2.5 Mass spectrometry analysis ...... 112 4.3 Discussion ...... 115

5. Concluding remarks ...... 118

Appendices:

A. Materials and Methods ...... 120

A.1 Plasmid DNA Mini-Prep ...... 121 A.2 Site directed mutagenesis...... 122 A.3 Transformation of E. coli ...... 123 A.4 Protein Over-expression in E. coli ...... 124 A.5 Native Protein Purification...... 125 A.6 Labeling and purifying protein with 35S-Met/Cys ...... 128 A.7 10% Tricine-SDS PAGE ...... 130 A.8 Protein concentration via Bradford reaction ...... 131 A.9 Circular dichroism ...... 132 A.10 Adenosine to inosine activity assay ...... 133

A. 11 Kinetic adenosine to inosine assay ...... 134 A. 12 Single-turnover kinetic assay to determine binding dissociation constant (Kd) ...... 136 A. 13 1-methyl adenosine (m1A) to 1-methyl inosine (m1I) activity assay...... 137 A. 14 Electrophoresis mobility shift assay ...... 139

References ...... 141

List of Tables

Table Page

Table 2.1 The effect of amino acid substitutions at evolutionarily conserved residues in TbADAT2 and TbADAT3 ...... 28

Table 2.2 The effect of amino acid substitutions at evolutionarily conserved residues in TbADAT2 and TbADAT3 ...... 33

Table 2.3 Metal content in the ADAT2 complex ...... 40

Table 3.1 Binding parameters for wild type and C-terminus mutants of TbADAT2 ...... 66

Table 3.2 Kinetic parameters for wild type and C-terminus mutants of TbADAT2 ...... 76

Table 3.3 The pseudo-active site cannot rescue mutations of the active site ...... 78

Table 3.4 The pseudo-active site and the KR-domain act cooperatively to bind single tRNA ...... 85

Table 4.1 m1A to m1I native protein fold purification ...... 107

List of Figures

Page

Figure 1.1 tRNA processing in Trypanosoma brucei ...... 4

Figure 1.2 Hydrolytic deamination reaction ...... 10

Figure 1.3 Adenosine to inosine editing in tRNAs ...... 19

Figure 2.1 Evolutionarily conserved residues in the active sites of Sc and TbADAT2, Sc and TbADAT3, and ScADAT1 ...... 23

Figure 2.2 Steady state kinetic analysis of recombinant wild type TbADAT2/3 ...... 27

Figure 2.3 TbADAT2/3 and TbADAT2 are hetero and homodimers, respectively (Chromatographs) ...... 31

Figure 2.4 TbADAT2/3 and TbADAT2 are hetero and homodimers, respectively (Standard curve) ...... 32

Figure 2.5 Amino acid substitution of one of the key Zn-coordinating cysteines in TbADAT2 yields and active enzyme...... 36

Figure 2.6 A single tightly bound zinc ion is sufficient for enzyme activity ...... 37

Figure 2.7 TbADAT2 in the absence of TbADAT3 has no detectable A to I activity..... 39

Figure 2.8 Wild type TbADAT2/3 binds 1 tRNA per heterodimer...... 42

Figure 2.9 TbADAT2 binds tRNA substrate with lower affinity than wild type ...... 43

Figure 2.10 Structural modeling supports the possibility of alternative inter-subunit zinc coordination in TbADAT2/3 ...... 47

Figure 2.11 Structural modeling supports the possibility of alternative inter-subunit zinc coordination in TbADAT2/3 (colored by conservation) ...... 48

Figure 2.12 Evolutionary relationships among ADARs, ADATs, and Cytidine deaminases ...... 53

Figure 3.1 TbADAT2/3 herterodimer has one active site and one pseudo-active site . 56

Figure 3.2 Wild type TbADAT2/3 stably binds tRNA in vitro ...... 59

Figure 3.3 Wild type TbADAT2/3 specifically binds tRNA in vitro ...... 61

xii

Figure 3.4 Kinetic determination of the dissociation constant for TbADAT2/3 ...... 62

Figure 3.5 Alignment of SaADATa and TbADAT2 suggests differences in tRNA binding ...... 64

Figure 3.6 KR-domain is a predicted tRNA binding domain ...... 68

Figure 3.7 The C-terminus of TbADTA2 is critical for tRNA binding ...... 70

Figure 3.8 The KR-domain of TbADAT2 is critical for tRNA binding ...... 71

Figure 3.9 None of the protein variants had drastic effects on the overall structure of TbADAT2/3 ...... 73

Figure 3.10 TbADAT2 C-terminal mutants still heterodimerize with ADAT3 ...... 74

Figure 3.11 Steady-state kinetic analysis of TbADAT2/3 KR-domain mutants ...... 77

Figure 3.12 The pseudo-active site contributes to substrate binding ...... 81

Figure 3.13 The pseudo-active site proline contributes to substrate binding ...... 82

Figure 3.14 The pseudo-active site and the KR-domain act cooperatively to bind a single tRNA ...... 84

Figure 3.15 ADAT2/3-tRNA binding model ...... 92

Figure 4.1 m1A to m1I activity assay ...... 96

Figure 4.2 Thermococcus kodakaraensis has m1A to m1I activity ...... 97

Figure 4.3 Native protein purification column pilots (Q Seph and DEAE) ...... 100

Figure 4.4 Native protein purification secondary column pilots (Mono S) ...... 102

Figure 4.5 Native protein purification column pilots (HIC screen) ...... 103

Figure 4.6 Heparin chromatograph ...... 105

Figure 4.7 Heparin fraction analysis via TLC and SDS-PAGE ...... 106

Figure 4.8 Q sepharose chromatograph ...... 108

Figure 4.9 Q sepharose fraction analysis via TLC and SDS-PAGE ...... 109

Figure 4.10 HIC-Phenyl chromatograph ...... 110

Figure 4.11 HIC-Phenyl fraction analysis via TLC and SDS-PAGE ...... 111

Figure 4.12 Mono Q chromatograph ...... 113

Figure 4.13 Mono Q fraction analysis via TLC and SDS-PAGE ...... 114

xiii

Figure 4.14 Mass spectrometry analysis ...... 117

xiv

Chapter 1:

Introduction

1.1 Brief introduction to trypanosomes

Trypanosomes are a group of single-celled flagellated protozoa that belong to the order Kinetoplastida. Named for one of their most distinguishing features—the kinetoplast, a DNA containing granule found within the single mitochondrion—these organisms are among the earliest diverging eukaryotes. The mitochondrial genome

(kinetoplast DNA, kDNA) is unique to this order of organisms and is composed of about

40 maxicircles (22 kb each) and thousands of minicircles (1 kb each); maxicircles are most similar to the mitochondrial DNA of other organsims in that they encode components of the mitochondrial oxidative phosphorylation system and mitochondrial ribosomal RNAs (rRNA), while minicircles encode “guide” RNAs (gRNAs) which together with maxicircle-encoded gRNAs provide the genetic information needed for mRNA editing as will be discussed in more detail in subsequent sections of this chapter. The trypanosomes have two distinct life cycles, one within its insect vector (procyclic stage) and another in its mammalian host (bloostream stage). These two stages encounter very distinct environmental conditions and a number of unique biochemical and physiological mechanisms exist in trypanosomes not only to ensure survival under such diverse conditions but

also to evade host immunity (reviewed in Campbell, Thomas, Sturm 2003; Clayton 2002;

Das and Bellofatto 2004; Martinez-Calvillo et. al. 2010; Palenchar and Bellofatto 2006).

1.2 Medical significance of trypanosomes

Trypanosomes are medically important because particular species can be parasitic to humans and cattle. Trypanosoma brucei gambiense, for example causes

African trypnosomiasis or sleeping sickness, a neurological disease that if left untreated can lead to death. Primarily endemic to sub-Saharan Africa, the World Health

Organization (WHO) recently reported that due to disease control efforts the number of new incidents of African sleeping sickness fell below 10,000 for the first time in 50 years.

On the other hand, over 10 million people worldwide are reportedly infected with

American trypanosomiasis or Chagas disease, a disease caused by T. cruzi. In the chronic phase of this disease, trypanosomes reside in heart and digestive muscle, which can cause gradual muscle deterioration and death. While there are relatively effective treatments for trypanosome infections, no vaccines against these organisms exist; even effective treatments are not without side effects. Thus, there is much hope that exploiting the very unique biochemistry and physiology of these parasites will prove fruitful for developing more effective drugs and vaccines against trypanosomes.

To this end, the Alfonzo laboratory studies unique features of the trypanosomes with respect to tRNA editing and modifications and tRNA import into the mitochondrion.

Finding differences among trypanosomes and the human host in essential editing and modification enzymes could help identify new drug targets. Along these lines the Alfonzo laboratory recently showed, in collaboration with the Papavasiliou laboratory, that the T. brucei tRNA editing deaminase is unlike any other deaminase in that it can perform two

types of deamination reactions: adenosine to inosine and cytosine to uridine (Rubio

2007). This suggests that perhaps the active site or substrate binding site(s) are unique to the trypanosome enzyme thus providing targets for new drug development. These types of deaminase reactions in transfer RNAs (tRNAs) has been the focus of my work.

1.3 tRNA processing in trypanosomes

In most eukaryotes, transfer RNAs (tRNAs) are encoded in either of two DNA- containing compartments: the nucleus or the genome-containing organelles (chloroplast and mitochondria). However, in trypanosomatids, mitochondrial genomes do not appear to contain any tRNA genes. Therefore, the complete set of tRNAs used in both cytoplasmic and mitochondrial protein synthesis is encoded solely in the nuclear genome. In these organisms tRNAs are thus transcribed in the nucleus, exported into the cytoplasm and later a subset of cytoplasmic tRNAs are actively imported into the mitochondria. However, before tRNAs can be rendered functional in any cellular compartment, pre-tRNA transcripts face a number of enzymatic reactions including end trimming, intron splicing, and tRNA editing and modifications. Some of these processes, for example those involved in trimming of extraneous sequences found at tRNA ends, occur in the nucleus and usually precede cytoplasmic export. Other processes, like editing and modification, may occur at any point in the tRNA maturation pathway and in any of the tRNA-containing compartments (Fig. 1.1).

tRNAThr tRNATyr 5’ trimming 3’ cleavage Nucleus

C C splicing editing

CCA CCA

3’ cleavage U + CCA 5’ cleavage U + CCA

CCA A portion of tRNAs CCA CCA are kept for cytoplasmic A to I editing translation… U I U A

tRNATrp CCA …Another portion CCA CCA is imported into Mitochondrion the mitochondria 2 s U33 C U U Thiolation

Figure 1.1 tRNA processing in Trypanosoma brucei

After being transcribed in the nucleus tRNAs are not fully functional until after further processing which takes place primarily in, but is not limited to, the nucleus. Shown on the left is the fate of a newly transcribed tRNA, specifically highlighting tRNATyr, the only intron containing tRNA. The 5'- leader sequence is removed followed by intron splicing and 3' trailer cleavage. CCA is added to the 3'-end of the tRNA which is then exported into the cytoplasm where a portion of the tRNAs are kept for cytoplasmic translation and another portion is imported into the mitochondrion for mitochondrial translation. Editing as exemplified by C to U and A to I can occur in all 3 compartments. Highlighted in the pathway on the right is C to U editing of tRNAThr that occurs in the nucleus before the cleavage of the 5' leader and the addition of the CCA end. This tRNA is further edited (A to I) after being exported to the cytoplasm. Mitochondrial C to U editing in tRNA tryptophan is also shown. tRNA modifications such as thiolation (s2U) have been observed in the cytoplasm (ie. tRNAGln, tRNAGlu, tRNALys) as well as in the mitochondrion (ie. tRNATrp). Arrows indicate the sequential order of reactions and importantly those processes highlighted by gray circles have been observed in trypanosomatids however, the enzymes responsible for those reactions remain unknown. (Figure modified from review chapter Spears, Rubio, Sample and Alfonzo, in press).

1.3.1 Trimming of 5' and 3' ends to generate a full-length tRNA

Trypanosome tRNAs are transcribed by RNA polymerase III in the nucleus but like in most organisms, tRNAs are not transcribed as unit-length molecules; they contain extra sequences at their 5'- and 3' ends (5'-leader and 3'-trailer, respectively) that have to be removed as part of the standard maturation process. The exact order of end trimming events depends on the tRNA species, but most commonly in eukaryotic organisms, processing begins in the nucleus with 5'-leader cleavage followed by 3'- trailer removal. There are some of examples from T. brucei where this is not the case

(i.e. 3'-trailer is removed before the 5'-leader, tRNAThr) (Gaston et. al. 2007) but typically the 5' end is processed first.

In eukaryotic pre-tRNAs, 5'-leader removal is mediated by the highly conserved endonuclease ribonuclease P (RNase P) (most recently reviewed in McClain, Lai,

Gopalan 2010). While the 5'- leader sequence of pre-tRNAs from trypanosomatids is removed by RNase P, the exact nature of the enzyme(s) responsible for this cleavage event is still elusive. Maturation of the 3'-end of eukaryotic pre-tRNA involves two sequential events: 3' end removal followed by CCA-addition. Cleavage of the 3' trailer of pre-tRNAs is accomplished by the highly conserved tRNase Z (Schiffer, Rosch,

Marchfelder 2002; Vogel et. al. 2005) and in some cases enhanced by the multi- substrate RNA binding La protein (Van Horn et. al. 1997; Yoo and Wolin 1997).

However, not much is known about the 3' end maturation enzyme(s) in trypanosomes.

Bioinformatic analysis in T. brucei reveals the presence of a potential homolog only in the nucleus. tRNA import to the mitochondrion for all tRNAs in trypanosomes could help explain why a homolog to this highly conserved enzyme is not found in the mitochondrion; its function would be unnecessary in that there is no need in the

mitochondria for 3'-end processing of an already mature tRNA. Before the end-matured tRNA can be aminoacylated and used in translation, a three-nucleotide sequence, 5'-

CCA-3', must be added post-transcriptionally by the 3'-specific CCA nucleotidyltransferase. Remarkably, these enzymes add nucleotides to the 3'-end of the tRNA in a template-independent manner (Betat, Rammelt, Morl 2010; Xiong and Steitz

2006).

1.3.2 Intron removal

In T. brucei, the only intron-containing tRNA is tRNATyr, which has an 11- nucleotide intervening sequence, one of the smallest known introns. Intron-containing tRNAs can be found in all of the phylogenic kingdoms and usually consist of a short sequence of nucleotides that occur immediately 3' of the anticodon loop in eukaryotes.

The function of introns within tRNAs is largely unknown; however, in yeast tRNATyr, the intron is required for conversion of the central uridine of the anticodon to pseudouridine

(Johnson and Abelson 1983). The mechanism of tRNA intron splicing in trypanosomes likely proceeds in two steps. First, the intron is removed by a specific endonuclease that recognizes conserved elements of the anticodon stem; recent data suggest that the actual sequence of the intron might also be an important recognition element (Rubio and

Alfonzo, unpublished data). The two halves generated by the endonuclease must then be joined together to complete the full length tRNA; this is likely catalyzed by a ligase activity although the ligase has yet to be characterized.

1.3.3 tRNA export to the cytoplasm and mitochondrial import

Once the pre-tRNA ends have been cleaved and in the case of tRNATyr, once the intron has been removed, the mature tRNA must be exported from the nucleus to the

cytoplasm. Again, little is known about this process in T. brucei, but in other eukaryotic systems a Ran-GTP dependent protein, exportin-t, carries tRNA across the nuclear membrane and releases it after Ran-GTP hydrolysis to form Ran-GDP. The maturation status of tRNAs destined for mitochondrial import is currently a point of contention, so it will be interesting to see how exportin-t discriminates substrates for different localization or if other factors are also involved in tRNA export, as has been seen in fungi (reviewed in (Hopper, Pai, Engelke 2010). Once in the cytoplasm, the tRNA can be edited/modified, used for cytoplasmic translation or imported into the mitochondrion where the tRNA can be further edited and/or modified before use in mitochondrial translation.

1.4 RNA Editing

RNA editing, a term coined after the discovery of post-transcriptional uridine insertion and deletions in the mitochondrial messenger RNAs (mRNAs) of trypanosomes, is now a widely accepted process that encompasses many different nucleotide additions, deletions, and substitutions of RNA to result in final products which differ from what was originally encoded by the gene. RNA editing has been previously been described to fall into two general types: Insertion/deletion editing and nucleotide substitution editing (Gray 2003). Insertion/deletion editing involves the addition (or removal) of nucleotides to the DNA-encoded RNA transcript such as seen in kinetoplastids and Physarum. Substitution editing includes post-transcriptional base changes such as cytidine (C) conversion to uridine (U) or adenosine (A) conversion to inosine (I). In mRNAs, both nucleotide insertion/deletion and substitution can change codons in the transcript and thus create protein isoforms; in tRNAs, substitution editing at the wobble position can change the decoding capacity of the tRNA. Both

insertion/deletion and nucleotide substitution types of RNA editing will be discussed in more detail below and while editing is not limited to U insertion/deletion and C to U and

A to I deamination, these types of editing are most prevalent and will be the focus of review.

1.4.1 Insertion/deletion editing

Insertion/deletion editing was first discovered in the late 1980s by Rob Benne and colleagues (Benne et. al. 1986). Setting out to initially examine the nature of a presumed -1 frameshift within the trypanosome mitochondrial cytochrome oxidase subunit II (coxII) gene, Benne came across a most unique finding: the mature messenger RNA (mRNA) transcript contained 4 uridine (U) nucleotides not encoded by its DNA template (Benne et. al. 1986). Remarkably, translation of the functional coxII requires the transcript with 4 additional uridine nucleotides; without the additional uridines the protein is truncated because of the presence of an in frame stop codon.

Now it is well accepted that kinetoplastid DNA is extensively edited by insertion and deletion of multiple uridines mediated by guide RNAs (gRNAs) (Blum, Bakalara,

Simpson 1990) encoded by maxicircles as well as minicircles (Koslowsky et. al. 1990;

Pollard et. al. 1990). Guide RNAs direct, via complementary base pairing, where cleavage and subsequent U insertions and deletions will in a processive 3'-5' fashion

(Koslowsky et. al. 1991).

Just five years after the initial discovery a similar insertion editing was identified in the slime mold, Physarum polycephalum (Mahendran, Spottswood, Miller 1991). In this system, single cytidine insertions occurred at 54 different sites. Like in the trypanosome system, this extensive editing is required for the synthesis of full length,

functional proteins; however, this process does not seem to be mediated by gRNAs.

Since the original discovery of insertion/deletion editing other examples of insertion editing, albeit less extensive, have been shown in rRNAs in Physarum (Mahendran et. al. 1994) and mitochondrial tRNAs in Physarum as well as Didymium nigripes (Antes et. al. 1998) and in mRNAs in viruses including vaccinia (Schwer and Stunnenberg 1988) and measles virus (Cattaneo et. al. 1989).

1.4.2 Nucleotide substitution editing by deamination

As mentioned above, nucleotide substitution editing will be discussed in the context of editing by deamination although other types of base substitution exist (see section 1.4.5). Specifically, mRNA editing by APOBEC, ADAR1 and ADAR2 and tRNA editing by ADATs will be discussed in more detail. Regardless of the enzyme or the substrate being discussed, the deamination mechanism is conserved throughout all domains of life. The reaction proceeds by a zinc-mediated hydrolytic deamination whereby a proton from an activated water molecule is shuttled via a proton shuttling glutamate to the exocyclc nitrogen at C6 of the purine ring creating ammonia as the leaving group (Fig. 1.2; Betts et. al. 1994; Carter 1995; Wolfenden 1969). Within the respective deaminases, there is a conserved catalytic core with two highly conserved active site domains, the H(C)xE and the PCxxC domains (where x represents any amino acid). The histidine and two cysteines are responsible for the zinc coordination while the glutamate of H(C)xE is the proton shuttle (Fig. 1.2; Carter 1995). Although the reaction mechanism is conserved from nucleotide to polynucleotide deaminases, as will be seen in subsequent subsection and chapters, the substrate specificity varies greatly.

Cys Glu Cys Zn His

NH2 O

N N N HN

N N N N HO HO O H2O O

NH3 OH OH OH OH Inosine Adenosine

Figure 1.2 Hydrolytic deamination reaction

This schematic shows the hydrolytic deamination reaction for adenosine to inosine conversion. While the example is specific the mechanism is common among deaminases including those that catalyze A to I and C to U. The enzyme coordinates zinc via three conserved amino acids (histidine, cysteine and cysteine). The fourth zinc coordination comes from water which actives the water molecule for nucleophilic attack on the pyrimidine ring at C6. An essential glutamate shuttles acts as a proton shuttle creating an ammonia leaving group (Carter 1995).

1.4.3 mRNA editing: APOBEC-1

Apolipoprotein B (ApoB), a major protein component of the low-density lipoprotein (LDL) that carries cholesterol to various tissues in mammals (Kane 1983), is site-specifically edited at the mRNA level at position C6666 (Powell et. al. 1987).

APOBEC-1 (Apolipoprotein B mRNA editing catalytic polypeptide) is the catalytic component of the enzyme complex responsible for this tissue specific C to U deamination which introduces an in frame stop codon. In this system, RNA editing yields a truncated protein (Apo-B48 vs full length Apo-B100) in the small intestines (or liver in some mammals such as rodents) (Greeve, Navaratnam, Scott 1991; Navaratnam et. al.

1993; Powell et. al. 1987; Teng, Burant, Davidson 1993). Because of its association with

LDL, ApoB100 in high levels correlate to atherosclerosis (Scott et. al. 1987). ApoB48 is not associated with VLDL (very low densitiy lipoproteins) or LDL but is instead the primary component of chylomicrons and is responsible for fat absorption in the small intestines (Kane 1983).

While the exact size and composition of the ApoB mRNA editing enzyme complex have yet to be definitively determined, numerous studies have shed light on functional and structural aspects of APOBEC-1. Notably, the minimal requirements to reconstitute ApoB mRNA deamination activity in vitro are the 27-kDa catalytic protein

APOBEC-1, 65-kDa RNA binding protein, APOBEC-1 complementation factor (ACF)

(Lellek et. al. 2000; Mehta et. al. 2000), and a mRNA substrate at least 26 nucleotides

(nt) in length that contains the 11-nt sequence 5'-TGATCAGGATA-3' five nt downstream of C6666 (so called mooring sequence) (Shah et. al. 1991). A combination of UV- crosslinking and immunoprecipitation experiments showed that the ACF is able to bind

to ApoB mRNA at the mooring sequence and interact with APOBEC-1 for activity (Mehta et. al. 2000).

As observed by viable APOBEC-1-/- mice, APOBEC-1 is required for ApoB editing but is not essential for viability as knockout mice were phenotypically nearly indistinguishable from their respective control group (Hirano et. al. 1996; Morrison et. al.

1996). To the contrary, ACF-/- mice are embryonic lethal suggesting an important function in addition to ApoB editing (Blanc et. al. 2005). Indeed ACF is able to bind other

AU-rich RNAs, and Blanc et. al.suggest that binding to other RNA targets (i.e. Cox2 and interferon gamma) early in embryonic development could explain the lethality of the transgenic mouse (Blanc et. al. 2005; Blanc and Davidson 2010).

Strong evidence supports ApoB mRNA as APOBEC-1’s primary target in vivo; however, recent studies via transcriptome-wide searches reveal that at least 30 more mRNA sequences undergo C to U editing. Interestingly these sequences were all located in AU-rich 3'-UTRs (3'-untranslated regions), the first reported example of C to U editing in 3'-UTRs (Rosenberg et. al. 2011). Papavasiliou’s group concluded the work by suggesting additional functions for APOBEC-1 beyond a role in ApoB editing and lipid transport. This idea is further supported by work from the Conticello group (Severi,

Chicca, Conticello 2011). Additionally, studies showed that human and mouse APOBEC-

1 can also mutate C to U in ssDNA in vitro and in vivo (Harris, Petersen-Mahrt,

Neuberger 2002; Petersen-Mahrt and Neuberger 2003; Petit et. al. 2009). Furthermore,

Conticello’s group has demonstarted, for the first time, the presence of APOBEC-1 homologues (Blanc et. al. 2005) in non-mammalian organisms. Closer examination of the recombinant lizard APOBEC-1 reveals that not only can this protein edit ssDNA but the C to U patterning and distribution is similar to the mammalian APOBEC-1 (Severi,

Chicca, Conticello 2011). Taken together, these studies are interesting from an evolutionary standpoint and suggest that DNA mutation could be the more ancestral

APOBEC-1 function (Conticello et. al. 2005; Severi, Chicca, Conticello 2011). This could have functional relevance in cells with respect to DNA mutations that lead to tumor formation (Harris, Petersen-Mahrt, Neuberger 2002; Severi, Chicca, Conticello 2011).

1.4.4 mRNA editing: ADAR1 and ADAR2

Editing also leads to protein diversity as a result of adenosine to inosine deaminations of mRNAs catalyzed by ADARs (adenosine deaminase acting on RNA).

These enzymes contain a variable number of double stranded RNA (dsRNA)-binding sites (1 to 3) and a core deaminase domain consisting of the active site residues histidine, cytidine, cytidine, and glutamate, and target dsRNA. First discovered in the late

1980s as an activity that unwinds dsRNA by introducing less stable I-U base pairs (Bass and Weintraub 1988), adenosine to inosine deamination of RNA is now widely accepted as an important cellular event for which some of its functions still remain unclear

(Hundley and Bass 2010). Direct coding alterations of edited mRNAs can of course affect the final protein product but often ADARs act on noncoding RNAs such as the 3' and 5' UTR (untranslated region) and introns (reviewed in Nishikura 2010). With inosine- participating base pairs affecting dsRNA structure/stability, many studies in Drosophila and C. elegans have pointed to functions in gene regulation by, for example, inhibiting steps of RNA silencing and perhaps affecting dicer’s ability to cleave a given RNA

(Knight and Bass 2002; Scadden and Smith 2001).

ADAR1 and ADAR2 are two ADARs that catalyze the deamination of mRNAs in a site-specific manner. Each enzyme functions as a homodimer to deaminate specific

amino acids on mRNA substrates such as GluR-B, a subunit of the AMPA (α-Amino-3- hydroxy-5 methyl 4 isoxazolepropionate) receptor. This receptor is a class of L- glutamate-activated cation channels and mediates the majority of fast excitatory neurotransmission in the central synapses (Higuchi et. al. 1993). In the subunits of

AMPA, a critical arginine residue is responsible for differences seen in calcium permeability and gated ion-channel behavior. This site is termed the Q/R site for the presence of either glutamine (CAG) or arginine (CGG). Not surprisingly, this site is the location of adenosine deamination where the result is a change in the glutamine codon

CAG to CIG which is read by the translation machinery as CGG (I is read as G), an arginine codon (Higuchi et. al. 1993; Sommer et. al. 1991). This editing event occurs in nearly 100% GluR-B mRNA with a genomically encoded glutamine and is critical for proper brain function as hypo-edited messages lead to epilepsy (Brusa et. al. 1995) and were also shown to be present in malignant tumors (Maas et. al. 2001).

1.4.5 tRNA editing

tRNA editing most commonly involves a base substitution whereby post- transcriptionally an adenosine (A) or cytidine (C) is deaminated to inosine (I) or uridine

(U), respectively. These nucleotide conversions have been observed in the acceptor stem (Lonergan and Gray 1993) as well as in the anticodon loop (Morl, Dorner, Paabo

1995; Gerber and Keller 1999; Alfonzo et. al. 1999). The first example of tRNA editing shown occurs in the mitochondrial tRNAs of Acanthamoeba. In this system, tRNAs undergo U to A, U to G and A to G nucleotide substitutions in the first three nucleotides of the acceptor stem to repair base pair mismatches (Lonergan and Gray 1993). A similar mismatch base pair repair editing has most recently been found in Physarum tRNA where a C to G or U to G change is seen in tRNAMet (Gott, Somerlot, Gray 2010).

Being the most well studied and supported examples of tRNA editing in trypanosomes,

C to U and A to I editing will be elaborated in the following subsections.

1.4.5.1 C to U tRNA editing

C to U editing in tRNAs was first described in rat cytoplasmic tRNAAsp (Beier et. al. 1992) and marsupial mitochondrial tRNAAsp (Morl, Dorner, Paabo 1995). In both cases, the anticodon loop is edited; however, in the editing of position 35 (2nd position of the anticodon) in tRNAAsp from marsupials, the edited nucleotide has a direct bearing on translation as U35 is necessary for decoding aspartate (Morl, Dorner, Paabo 1995).

In trypanosomes, all the tRNAs used in mitochondrial translation are nucleus- encoded and subsequently imported into the mitochondrion. This poses a problem when considering the translation of tryptophan codons because, like with many eukaryotic organisms, in the trypanosomatid mitochondrial genome the canonical 5'-UGG-3' tryptophan codon is often replaced by 5'-UGA-3', a stop codon in cytoplasmic translation. This led to the interesting question of how organisms with a single tRNATrp with anticodon 5'-CCA-3' could decode 5'-UGG-3' as tryptophan and 5'-UGA-3' as stop in the cytoplasm and UGA as tryptophan once imported into the mitochondrion. First shown in Leishmania tarentolae more than a decade ago (Alfonzo et. al. 1999) and later in Trypanosoma brucei (Charriere et. al. 2006), these organisms have solved this decoding conundrum in a simple yet rather elegant way. The tRNA undergoes cytidine

(C) to uridine (U) editing at the first position of the anticodon thus changing the 5'-CCA-3' anticodon to 5'-UCA-3' which can now decode 5'-UGA-3' by canonical base pairing and even 5'-UGG-3' by wobbling. Clearly, compartmentalization of such an activity avoids the

potential generation of a high-copy suppressor tRNA (with anticodon 5'-UCA-3') in the cytoplasm (Alfonzo et. al. 1999).

While the reason for this type of editing is clear, its mechanism remains uncertain. The simplest activity accounting for the observed C to U editing would involve a tRNA-specific cytidine deaminase. A precedent for a polynucleotide specific cytidine deaminase already exists in the editing of mRNA in mammals (ie APOBEC-1 as discussed above). The possibility of such a deaminase enzyme in trypanosomatid mitochondrial tRNA has been most recently reinforced with the discovery of an analogous activity in archaea, where the first cytidine deaminase acting on tRNAs

(CDAT8) has been described (Randau et. al. 2009). This enzyme, a member of the larger cytidine deaminase family, catalyzes C8 to U8 editing via the conserved hydrolytic deamination reaction in a zinc-dependent manner (Fig. 1.2). Unfortunately, no protein similar to CDAT8 in Trypanosoma or Leishmania could be identified by bioinformatic analysis. Therefore, the enzyme responsible for this mitochondrial editing activity remains at large. However, an in vivo approach still allowed for defining the last base

Trp pair of the anticodon stem (A31-U39) as important for tRNA editing (Crain et. al. 2002).

Beyond this position, it is not clear what other determinants of editing exist in the natural

Trp 2 tRNA substrate in Leishmania, but modifications such as S U33 in T. brucei may serve as an anti-determinant for editing (Wohlgamuth-Benedum et. al. 2009). C to U editing has a direct bearing on translation and because an estimated 88% of the conserved mitochondrial tryptophan codons are UGA, this editing is essential for mitochondrial translation and cell survival (Alfonzo et. al. 1999).

The only other example of C to U editing in trypanosomatids occurs outside the mitochondrion and is, in fact, the first example of C to U editing outside organelles in

eukarya. Occurring just 5' of the tRNA anticodon (position 32) in all three isoacceptors of tRNAThr, this editing event takes place while the tRNA is still in the nucleus and before the removal of the 5' leader from the tRNA (Gaston et. al. 2007). However, unlike the

Trp mitochondrial editing of tRNA , C32 to U32 editing has no direct bearing on decoding; interestingly, in vitro, C32 to U32 stimulates, although is not required for, the essential

Thr adenosine to inosine formation at the wobble position of tRNA AGU (Rubio et. al. 2006).

Because A to I has a direct effect on translation and is essential for cell viability, it is possible that C32 to U32 may indirectly affect translational efficiency of some, if not all, threonine codons.

1.4.5.2 A to I tRNA editing

Although not originally recognized as editing, the discovery of inosine in

Ala tRNA AUA by Holley and co-workers over 45 years ago (Holley et. al. 1965) led Francis

Crick to propose that the base-pair capabilities of this tRNA were not limited to canonical nucleotides and inosine could pair with three different codons (ending in A, C, or U) to specify the same amino acid. This notion was further elaborated in Crick's famous

"wobble hypothesis" (Crick 1966), which provides explanation for (1) the presence of multiple codes for one amino acid, (2) the existence of more than one tRNA to specify a single amino acid, and (3) the implicit ability of a single tRNA to decode more than one codon (as in the case of inosine). It is now well established that tRNAs encoded with an adenosine at position 34 are almost universally edited to inosine, expanding the decoding capacity of that tRNA.

Adenosine to inosine is not only found in the anticodon but also 3'-adjacent to the anticodon (position 37) in eukaryotes and in the TΨC loop (position 57) in archaea. In

both these locations, the inosine also carries a methyl group (1-methylinosine or m1I) however the order of events (methylation and deamination) are different at the two sites.

In eukaryotes, A37 is first deaminated by ADAT1 (adenosine deaminase acting on tRNAs

1) and then methylated by the SAM-dependent methyltransferase (Grosjean et. al.

1996). However, in archaeal tRNAs, A57 is first methylated by the SAM dependent methylatransferase TrmI, and is further deaminated by an unknown m1A to m1I enzyme

(Fig 1.3). This enzyme, presumably a deaminase is the focus of chapter 4.

Mechanistically all these deamination reactions are believed to occur via a conserved hydrolytic deamination reaction mentioned above (Fig 1.2). By far the most studied adenosine deamination in tRNA is that occurring at the wobble position because inosine has a direct bearing on translation. This reaction is conserved and essential in bacteria as well as in eukaryotes. The enzymes responsible for A to I, ADATa in bacteria and ADAT2/3 in eukaryotes, are essential in their respective organisms. Because crystal structures of the bacterial but not the eukaryotic ADAT exist (Elias and Huang 2005; Kim et. al. 2006; Losey, Ruthenburg, Verdine 2006), ADATa is often used as a model for studying ADAT2/3 and thus the bacterial enzyme will be discussed first. ADATa functions as a homodimer (mol. wt. ~40 kDa; ~20 kDa per monomer) to deaminate a single tRNA, tRNAArg. Importantly, the enzyme can also deaminate just an anticodon stem loop in vitro. With two complete active sites (one per subunit) each subunit

(monomer) coordinates a single zinc ion and binds a single tRNA per catalytic cycle

(Losey, Ruthenburg, Verdine 2006; Wolf, Gerber, Keller 2002). The co-crystal structure of ADATa bound to an anticodon stem loop revealed that all the important and necessary contacts for substrate binding were contained within the active site of the

Tb tRNAVal CCA-3’ 5’ A G U C G G C C G 1 1 U G A57 m A m I57 G C A U U UG A57 G G G AGCCC A U AUCUG G G UCGGG C U U G UGAC C U U A G U U A AGG C G G C C G U G U C U A A34 A C A37

I 1 34 A37 I37 m I37

Figure 1.3 Adenosine to inosine editing in tRNAs

Val Shown on the left is Trypanosoma brucei tRNA ACC transcript that is used throughout the studies presented in this dissertation. Highlighted is the A to I editing by ADAT2/3 that occurs at the 1 wobble position (A34). Shown on the right is a generic tRNA that highlights the A to m I37 that 1 occurs in eukaryotes and the A57 to m I57 that occurs in archaea.

enzyme and within the anticodon stem loop of tRNAArg, as predicted by in vitro studies

(Losey, Ruthenburg, Verdine 2006).

On the other hand, ADAT2/3 recognizes seven or eight different tRNA substrates

(Leu, Ile, Val, Ser, Pro, Thr, Ala, and Gly) depending on the organism. The enzyme functions as a heterodimer comprised of subunits ADAT2 and ADAT3. It has been proposed that ADAT2 is the catalytic component of the enzyme while ADAT3 plays a more structural role because it is naturally missing the essential glutamate needed for catalysis (Gerber and Keller 1999). Here it is worth noting distinctive features from the

Trypanosoma brucei enzyme (TbADAT2/3). First, this enzyme is the only known RNA deaminase that can perform both types of deaminations, A to I and C to U, albeit in different substrates (tRNA and DNA, respectively) (Rubio et. al. 2007). Also interesting is the fact that ADAT2 knock-down in T. brucei led to not only a decrease in A34 to I34 but

Thr also an unexpected decrease in C32 to U32 tRNA editing in tRNA in vivo. However,

ADAT2/3 does not produce C to U in vitro which raises the possibility that the substrate specificity can change by virtue of which other proteins associate with ADAT2 within a given compartment (Rubio et. al. 2007). Despite being an essential enzyme with products important in translation, there are many gaps in the knowledge about ADAT2/3.

For instance how many zinc ions are coordinated, how the tRNA is recognized, and how many tRNAs are bound per catalytic cycle are still open questions.

The focus of the work presented in the following chapters is on answering these unresolved questions with respect to the T. brucei ADAT2/3 enzyme (Chap. 2 and 3).

1 Chapter 4 will focus on tRNA editing in the TΨC loop and the purification of the m A57 to

1 m I57 enzyme.

CHAPTER 2:

A Single Zinc Ion is Sufficient for an Active Trypanosoma brucei tRNA Editing Enzyme: Insights into the Function of Multimeric Deaminases

2.1 Introduction

All genomes encode far fewer tRNAs than would be required to decode all codons used in translation (Agris, Vendeix, Graham 2007). This apparent decoding conundrum is resolved by the inherent base pairing flexibility between the third codon position in mRNA and the first position of the anticodon in tRNA, as originally proposed in Crick’s wobble rules (Crick 1966). Wobbling permits some nucleotides (and modified nucleotides (Yokoyama and Nishimura 1995)) to form sufficiently stable non-Watson and

Crick base pairs providing critical flexibility without sacrificing translational fidelity.

Among the most extreme cases of base-pairing flexibility in protein synthesis is the use of the nucleotide inosine (I), a guanosine analog that can base pair with adenosine (A), cytosine (C) or uridine (U) (Murphy and Ramakrishnan 2004). Thus, in the context of a codon-anticodon interaction, inosine at the first position of the anticodon (I34) effectively allows a single tRNA to faithfully decode three different codons for the same amino acid obviating the need for additional tRNAs to be encoded in genomes. Given the critical role of I34 in decoding in bacteria and eukarya (including trypanosomes), it is not surprising that the enzymes responsible for inosine formation are essential for viability

(Gerber and Keller 1999; Rubio et. al. 2007; Wolf, Gerber, Keller 2002).

Inosine formation in tRNAs, however, is not relegated to the anticodon and has been described at various positions in tRNAs from a number of different organisms from all three domains of life. For example, in most archaeal tRNAs inosine exists in the TΨC

1 1 loop as a methylated species (N1-methylinosine, m I57). The presence of m I in the TΨC loop is so far exclusive to archaea (Czerwoniec et. al. 2009; Grosjean et. al. 1996) and will be discussed in more details in Chapter 4. In eukaryotic tRNA, m1I is also observed but it occurs 3’-adjacent to the anticodon at position 37 where it may affect anticodon loop structure (Gerber et. al. 1998; Grosjean et. al. 1996). Interestingly, the sequence of the enzyme responsible for synthesis of I37 (Adenosine Deaminases Acting on tRNA

ADAT1) resembles classic adenosine deaminases while the enzyme for I34 has all the sequence signature motifs of cytidine deaminases (CDA) (Fig. 2.1), perhaps reflecting their different evolutionary paths. Regardless of nucleotide position, all inosines in RNA are presumably formed by the Zn2+-dependent hydrolytic deamination of adenosine, catalyzed by Adenosine Deaminases Acting on tRNAs (ADATs) or Adenosine

Deaminases Acting on RNA (ADARs) for all other RNAs. These enzymes can specifically target a variety of RNA substrates, including tRNAs, mRNAs, miRNAs and even rRNA, while excluding free nucleotides (Gray 1976; Habig, Dale, Bass 2007; Hurst et. al. 1995; Kawahara et. al. 2007).

Currently, the best-studied example of inosine in tRNA occurs at the first position of the anticodon (I34) where the essential inosine has, as discussed above, a direct

Arg bearing on translational decoding. In bacteria, I34 is generated only in tRNA ACG by the homodimeric ADATa (Wolf, Gerber, Keller 2002). Each homodimer contains two complete active sites, coordinates two Zn2+ ions (1 per monomer), and binds 2 tRNAs (1 per monomer) per catalytic cycle

Proton-shuttling glutamate Residue # 90 92 136 139 TbADAT2 LAHAEFVAVEELLRQATAGTS----VLYVVVEPCIMCAAML

ScADAT2 VAHAEFMGIDQIKAMLGSRGE----TLYVTVEPCIMCASAL

No Proton-shuttling glutamate Residue # 252 254 291 294 TbADAT3 LDHPVTFVLKEVTRKQCKDR---DMFVSHEPCVMCSMAL

ScADAT3 IDHSVMVGIRAVGERLREGV-----DVYLTHEPCSMCSMAL

ScADAT1 DCHAEILALRGANTVLLNRIG-----LALYISRLPC---~60 aa---CSDKL

motif 3

Figure 2.1: Evolutionarily conserved residues in the active sites of Sc and TbADAT2, Sc and TbADAT3, and ScADAT1.

This figure shows a schematic highlighting conserved amino acids involved in Zn2+ coordination in TbADAT2, TbADAT3, ScADAT2, and ScADAT3 (denoted by a star). Conserved deaminase domains are highlighted in black. The key glutamate required for catalysis is also indicated. This glutamate is involved in shuttling a proton from an activated water molecule to the leaving amino group which is released as ammonia during the reaction. This glutamate, however, has been naturally replaced during evolution by a valine in Sc and TbADAT3. Numbers above highlighted residues indicate amino acid positions in each respective sequence relative to the first methionine in T. brucei. “Motif 3” refers to a motif characteristic of adenosine deaminases (ADAT1 for example) where the second Zn-coordinating cysteine is separated from the first by an approximately 60 amino acid spacer. This motif is not found in cytidine deaminases.

(Elias and Huang 2005; Kim et. al. 2006; Kuratani et. al. 2005; Losey, Ruthenburg,

Verdine 2006). The Verdine laboratory made significant contributions to ADATa studies with their co-crystal structure of S. aureus ADATa with a bound anticodon-stem loop

(ASL). The structure highlighted, for the first time, the important RNA recognition and substrate-protein contacts necessary for catalysis (Losey, Ruthenburg, Verdine 2006).

While the natural substrate for ADATa is presumably the full length tRNAArg, in vivo, the enzyme can also efficiently catalyze the deamination of a minimal tRNA substrate consisting of only the ASL in vitro (Wolf, Gerber, Keller 2002). Importantly, all the key contacts highlighted in the crystal structure are maintained within this shortened substrate (Losey, Ruthenburg, Verdine 2006).

In contrast, the eukaryotic enzymes target 7 or 8 different tRNAs depending on the organism, raising questions about the evolutionary transition to this broader, multi- substrate specificity. The eukaryotic enzyme functions as a heterodimer, comprised of 2 subunits, ADAT2 and ADAT3, and unlike the bacterial enzyme, it needs a full-length tRNA for activity (Auxilien et. al. 1996; Gerber and Keller 1999). Currently, however, due to the lack of a crystal structure, little is known about what governs substrate recognition among eukaryotic tRNA anticodon deaminases or how zinc is coordinated in their active site. Because of its sequence similarity with other members of the cytidine deaminase superfamily, including bacterial ADATa, it has been suggested that the eukaryotic enzymes use similar metal coordinating motifs, H(C) xE and PCxxC (where x represents any amino acid). Presumably, the conserved histidine and two cysteines coordinate a zinc ion; the fourth coordination is with an activated water molecule. A conserved glutamate in ADAT2 then acts as a proton shuttle between the activated water and the exocyclic nitrogen at C6 of the purine ring. However in ADAT3 the catalytic glutamate

has been replaced by a non-catalytic amino acid. In about 60% of all ADAT3s, including

T. brucei ADAT3, a valine is found in place of the conserved catalytic glutamate seen in

ADAT2 and ADATa. (Fig. 2.1) (Rubio et. al. 2007). Since ADAT3 still has all the Zn2+- coordinating residues (His, Cys, and Cys), it has been suggested that this subunit might only play a structural role (Gerber and Keller 1999; Rubio et. al. 2007). In the absence of evidence to suggest otherwise, the eukaryotic A34-specific tRNA deaminases are thus presumed to bind two zinc ions (one catalytic and one structural), both essential for activity (Gerber and Keller 1999).

The work in this chapter focuses on the exploration of the general contributions of each subunit to deaminase activity. Specifically the intrinsic asymmetry of the heterodimeric ADAT2/3 was exploited to determine the role of each conserved Zn2+- coordinating residue to activity. Earlier work from the Alfonzo lab showed that ADAT2/3 from T. brucei (TbADAT2/3) is unusual in that it performs both C to U and A to I deamination reactions, albeit in different substrates (C to U in DNA and A to I in RNA)

(Rubio et. al. 2007). Here, another unique feature of TbADAT2/3 is highlighted. Using a combination of molecular and biochemical approaches, the steady-state kinetic behavior of TbADAT2/3 was studied. Our data show that this enzyme is mechanistically distinct from other multimeric RNA deaminases in its ability to catalyze the deamination reaction with only a single bound Zn2+. We also present a molecular model which supports the experimental data and highlights the possible tight interplay between subunits to generate a functional active site. These new insights are discussed with respect to implications for the function and evolution of multimeric editing deaminases.

2.2 Results

2.2.1 Mutation of a key Zn2+-coordinating cysteine in ADAT2 still yields an active enzyme

Shown previously, the heterodimeric TbADAT2/3 had sufficient catalytic flexibility to perform both adenosine and cytidine deamination reactions, albeit in two different substrates, tRNA and DNA respectively (Rubio et. al. 2007). In light of these observations, the general contribution of each subunit (TbADAT2 and TbADAT3) to enzymatic activity was further probed. Previous studies reported the identification and partial characterization of the S. cerevisiae ADAT2/3 (ScADAT2/3) enzyme (Gerber and

Keller 1999). This S. cerevisiae enzyme, like T. brucei, is also heterodimeric, and while the ScADAT2 subunit harbors all the key conserved amino acids required for deamination, the sequence of ScADAT3 is also naturally missing the conserved active site glutamate essential for deaminase activity. ADAT2 was thus proposed to play a catalytic role while ADAT3 was proposed to play only a structural role in activity (Gerber and Keller 1999). However, this remains an untested premise, and the general contribution of each subunit to catalysis has not been formally established.

Before establishing the contribution of each subunit to catalysis, existing kinetic parameters for the recombinant enzyme were confirmed. Recombinant protein was incubated with increasing saturating levels of radio-labeled substrate ([tRNAVal] = 0.1 M to 1.6 M). Plots of the initial reaction velocity (pmol/min) against substrate concentration

(M) yielded a Vmax of 0.25 ± 0.06 pmol/min, a Km value of 0.75 ± 0.11 M and a kcat of

-1 0.19 ± 0.07 min (Fig. 2.2 and Table 2.1). A similar Km value of 0.72 ± 0.11 M was observed for the partially purified (600-fold pure) native enzyme, suggesting that the recombinant enzyme is a fair representation of that natively expressed in T. brucei

A B 0.20 [tRNA] 0.18 0.16 0.14 0.12 pI /min) 0.10

0.08 pmoles

( 0.06

pA o v 0.04 KM = 0.75 ± 0.11 M -1 0.02 kcat = 0.19 ± 0.07 min 0.00 1 2 3 4 5 6 7 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6

[tRNA] (M)

Figure 2.2: Steady-state kinetic analysis of recombinant wild type TbADAT2/3.

A) A representative one-dimensional thin-layer chromatography (TLC) analysis of the reaction products. pA and pI denote the migration of unlabeled 5'-AMP (pA) and 5'-IMP (pI) used as TLC markers and visualized by UV shadowing (not shown). The fraction of pA converted into pI during each reaction was calculated by dividing the amount of radioactive pI produced by the total (pA+pI); this value was then used to calculate the picomoles of 5'-IMP produced. A no enzyme control was routinely used for background subtraction. B) The initial velocity (vo) was plotted as a function of substrate concentration in M. The data were fitted by non-linear regression to the Michaelis-Menten equation using SigmaPlot kinetic software.

Table 2.1: The effect of amino acid substitutions at evolutionarily conserved residues in TbADAT2 and TbADAT3.

a a,b Enzyme Km kcat kcat/Km M min-1 min-1/M

Wild type 0.75 ± 0.11 0.19 ± 0.07 0.25 ADAT2 E92A/ADAT3 V254E ------ADAT3 V254E ------ADAT3 V254D ------ADAT3 V254L 2.77 ± 0.58 0.05 ± 0.03 0.02 ADAT3 V254T 2.14 ± 0.79 0.11 ± 0.01 0.05 ADAT3 V254S 2.67 ± 0.81 0.08 ± 0.03 0.03

a Values are obtained from at least three independent trials. “---” denotes no activity detected after 24 hours. b kcat values are calculated using experimental Vmax values, protein concentrations via Bradford assay, and fraction of active protein.

(Ragone and Alfonzo, unpublished results). The observed Km is also within the range of the E. coli ADATa enzyme (Km= 0.83 M). However, the E. coli enzyme shows a 10-fold faster kcat (Kim et. al. 2006).

To investigate the specific contributions of ADAT2 and ADAT3 to catalysis, a number of single-amino acid substitutions were systematically made in each subunit of the homologous TbADAT2/3 at conserved residues. In deaminases of the cytidine deaminase family, Zn2+-coordination occurs via an amino acid triad composed of one histidine (or cysteine) separated by a spacer region (with a variable number of amino acids) followed by two additional cysteines (the so-called H(C)…CxxC motif) (Gerber et. al. 1998; Xie et. al. 2004) (Fig. 2.1). Attention was focused on amino acid substitutions in these regions.

In the ADAT3 sequence from both Trypanosoma brucei and yeast, a valine is found in place of the key catalytic glutamate (V254, HPV instead of HAE in Tb; V218,

HSV instead of HAE in Sc, Fig. 2.1). Within the context of the heterodimer, previous studies showed that replacing the naturally occurring valine in ScADAT3 with glutamate

(V218E) could not rescue the activity lost by substituting the catalytic glutamate for alanine (ScADAT2 E56A) suggesting that the position of the glutamate in ScADAT2 tertiary structure is crucial for catalysis (Gerber and Keller 1999). Analogous substitutions in TbADAT2/3 (TbADAT2 E92A/TbADAT3 V254E) caused similar effects with one striking difference; replacing V254 of TbADAT3 with glutamate (V254E) in the context of a wild type ADAT2 also completely inactivated the enzyme (Table 2.1). It should be noted that substitutions of V254 in TbADAT3 by aspartate also inactivated the enzyme while amino acids such as leucine, threonine, and serine were tolerated, albeit with a 3- to 7-fold increase in Km and smaller or insignificant changes in kcat (compared to 29

the wild type, Table 2.1). These data reinforce the view that the Zn2+-coordination by the

HPV region of TbADAT3 serves either a structural and/or binding role but does not directly affect catalysis. However, the fact that changes of nearby residues do affect activity (Table 2.1, i.e. ADAT3 V254E), it is concluded that even minor perturbations of residues proximal to H252 of TbADAT3 can have drastic effects on enzyme function.

Similarly, substitutions in the conserved HAE region of TbADAT2 showed that the key catalytic glutamate (E92) in TbADAT2 was essential for activity, but in contrast to V254, other amino acid substitutions at E92 could not be tolerated highlighting the essential role of the conserved glutamate in catalysis.

To further assess the specific contributions of TbADAT2 and TbADAT3 to catalysis, amino acids in the conserved PCxxC domain (Fig. 2.1) of each subunit were substituted for alanine; individual alanine substitutions were made in TbADAT2 and

TbADAT3 so that one subunit was bearing the mutation in the context of the second subunit remaining wild type (unless otherwise noted), yielding a heterodimer (Fig. 2.3 and Fig. 2.4). In TbADAT3, alanine substitutions of either of the conserved cysteines

(C291 and C294) yielded an inactive enzyme (Table 2.2). Notably, these ADAT3 mutants could still heterodimerize with TbADAT2 as shown by size exclusion chromatography. Two inferences can be drawn from the lack of activity: first, these

TbADAT3 mutants are unable to coordinate Zn2+; second, Zn2+-coordination by

TbADAT3 is not a pre-requisite for heterodimerization. As expected, TbADAT2 C139A abrogated enzymatic activity (Table 2.2), indicating the essential role of this cysteine for activity, likely due to its ability to participate in Zn2+ coordination. Quite unexpectedly, however, replacing the first cysteine of the TbADAT2 CxxC domain (C136A) generated

Mol. wt (kDa) Void 670 158 69 44 17 3

A 2.0 Wild type TbADAT2/3 heterodimer 1.5

1.0

0.5

0.0 0 2 4 6 8 10 12 14 16 18 20 22 24

B ) 2.0

mAU 1.5 TbADAT2 C136A/3 heterodimer

1.0

0.5 Abs 280nm ( 280nmAbs

0.0 0 2 4 6 8 10 12 14 16 18 20 22 24 C 1.0

TbADAT2 homoodimer

0.5

0.0 0 2 4 6 8 10 12 14 16 18 20 22 24

Retention Volume (mL)

Figure 2.3: TbADAT2/3 and TbADAT2 are hetero and homodimers, respectively (Chromatographs).

6xHis-tagged Wild type, ADAT2 C136A (in the context of wild type ADAT3) and ADAT2 proteins were over-expressed in E. coli and purified via Ni2+ affinity. Peak fractions containing the respective protein eluted from Ni2+ beads were pooled and concentrated. The elution buffer (20 mM Tris pH 8, 500 mM NaCl, and 1 M imidazole) was exchanged for the Superdex 200 size exclusion column running buffer (50 mM Hepes pH 8, 100 mM KCl). Shown are the size exclusion chromatographs for wild type TbADAT2/3 (A), variant ADAT2 C136A/3 (B), and TbADAT2 (C) proteins. Shown across the top of the elution profiles is molecular weights based on protein standards (from BioRad) separated using the Superdex 200 10/300GL with the same running conditions (0.5 ml/min) as used for the experimental trials. As seen in the elution profile the wild type and mutant proteins elute as a 68 kDa heterodimer. ADAT2 expressed in the absence of ADAT3 is able to homodimerize as seen by the peak at 57 kDa.

0.9

0.8 Vitamin B12 (1.35 kDa)

0.7

0.6 Myoglobin (17 kDa)

0.5

av Ovalbumin (44 kDa) K 0.4 TbADAT2 (57 kDa) TbADAT2/3 (69 kDa) 0.3 TbADAT2 C136A/3 (68 kDa) Gamma globulin (158 kDa) 0.2

0.1 Thyroglobulin (670 kDa) 0.0 1 10 100 1000

Molecular weight (kDa)

Figure 2.4: TbADAT2/3 and TbADAT2 are heterodimers and homodimers, respectively (Standard curve).

A molecular weight standard curve was generated using protein standards (from BioRad) separated by Superdex 200 10/300GL size exclusion column. Kav values were calculated using the formula Kav = [Velution-Vvoid]/[Vcolumn-Vvoid] where V is volume (ml). The void volume, as based on the elution of blue dextran, is 8.1 ml and the column volume is 23.6 ml. The elution volume of each protein sample was converted to Kav and those values were used to calculate molecular weight using the standard curve. The standard proteins and their respective sizes are shown as labeled gray diamonds; the experimental data (wild type and mutant TbADAT2/3 and TbADAT2 alone) are shown as black circles.

Table 2.2: The effect of amino acid substitutions at evolutionarily conserved residues in TbADAT2 and TbADAT3.

a a,b Enzyme Km kcat kcat/Km M min-1 min-1/M

Wild type 0.75 ± 0.11 0.19 ± 0.07 0.25 ADAT3 H252A ------ADAT3 C291A ------ADAT3 C294A ------ADAT2 E92A ------ADAT2 H90A ------ADAT2 C136A/ADAT3 C291A ------ADAT2 C136A 1.48 ± 0.40 0.05 ± 0.01 0.02 ADAT2 C139A ------

an active enzyme (Table 2.2 and Fig. 2.5). This recombinant variant showed only a 4- fold reduction in kcat and a 12.5-fold reduction in kcat/Km compared to the wild type. The larger decrease in catalytic efficiency was partly due to a 2-fold increase in the Km, suggesting that the modest change in kcat was either partly structural or due to small effects on substrate binding. This is in stark contrast with similar mutations in the cytidine deaminase from Escherichia coli, which led to either no detectable activity or a

>100,000-fold decrease in specific activity (Smith et. al. 1994). This observation suggests that although both cysteines in the CxxC motif of TbADAT2 play a role in optimal enzyme function, there is a first among equals: Zn2+ coordination via the first cysteine is not essential for catalysis. Taken together, the results in table 2.2 suggest marked differences in the functional significance of residues within conserved motifs between TbADAT2/3 and other nucleotide deaminases. These observations also reveal that all of the key Zn2+-coordinating residues of TbADAT3 are essential, but at least one of the three Zn2+-coordinating residues in TbADAT2 is dispensable for activity.

2.2.2 A single Zn2+ is necessary and sufficient for ADAT2/3 activity

Two possibilities could account for the above results: 1) the TbADAT2 C136A mutant might somehow coordinate a loosely bound catalytic Zn2+, or 2) a single Zn2+ is sufficient for catalysis. To rule out the first possibility and test the idea that loose chelation of a zinc ion would effectively enable the enzyme to remain active, deamination reactions were performed with the wild type and TbADAT2 C136A mutant in the presence of increasing concentrations of EDTA (ethylenediaminetetraacetic acid), a carboxylic acid able to chelate metal ions. However, no significant effects on the enzyme activity of the mutant or the wild type were observed even in the presence a

>650-fold molar excess of EDTA compared to the enzyme (1.5 M in these reactions)

(Fig. 2.6 A). Similar mutations in a number of deaminases can be partially rescued by increasing exogenous Zn2+ concentration (Smith et. al. 1994). In the T. brucei enzyme,

Zn2+ addition failed to rescue any of the inactive mutants; importantly, the C136A mutant had comparable activity in the absence of added Zn2+ (Fig. 2.6 B). In fact, as in the case of wild type, this variant was inhibited to varying degrees by ZnCl2 concentrations exceeding 50 M, which is not unprecedented (Smith et. al. 1994). The lack of rescue by addition of Zn2+ may be related to the observation that only after co-expression of both subunits in E. coli was an active enzyme generated and perhaps suggests that the catalytic Zn2+ has to be incorporated into the enzyme during de novo assembly of the recombinant heterodimer.

To explore the second possiblity the Zn2+ content of the wild type and TbADAT2

C136A mutant heterodimers was measured by Inductively Coupled Plasma (ICP)- emission spectrometry (Fassel 1978). Recombinant wild type or mutant enzyme was purified by Ni2+-chelate chromatography and followed by extensive dialysis in the presence of EDTA to remove loosely and non-bound zinc. This extensive washing was also to insure that no metals remained bound to the 6xHis-tag. These samples were then subjected to ICP, which revealed that, as expected, the wild type enzyme contained an average of 1.99 ± 0.12 zinc ions per heterodimer (Fig. 2.6 C). Remarkably, the

TbADAT2 C136A mutant contained an average of 0.78 ± 0.22 picomole of zinc bound per picomole of heterodimeric enzyme (Fig. 2.6 C). It is possible that a metal other than zinc remains bound in the mutant, coordinated by residues other than the conserved

H…CxxC triad, and this alternate metal is the one catalyzing the deamination reaction in

TbADAT2 C136A. Replacement of zinc by cobalt, for example, in other deaminases can still yield an active enzyme, albeit with a reduced catalytic rate (Betts et. al. 1994).

Time Time Time

no enzyme no no enzyme no enzyme no

pI pI pI

pA pA pA

1 2 3 4 1 2 3 4 1 2 3 4 Wild type AD2 C136A AD3 C291A

B C 4 0.020

) AD2 C136A

0.016

pmol 3 /min) 0.012 0.25

0.20 pmol 2 0.15 0.008 0.10

Wild type Velocity 0.05 1 AD2 C136A Wild type 0.004 0.00 AD3 C291A ( Velocity 0.0 0.5 1.0 1.5 2.0 2.5 3.0

[tRNA] M Inosine produced ( produced Inosine 0 0.000 0 50 100 150 200 250 300 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Time (min) [tRNA] (M)

Figure 2.5: Amino acid substitution of one of the key Zn-coordinating cysteines in TbADAT2 yields an active enzyme.

32 A) A site-specifically P-labeled (at A34) tRNA substrate was incubated with constant concentrations of enzyme for increasing time. Reaction products were then digested to 5’- nucleotide monophosphates and separated by thin-layer chromatography. Arrows indicate the position of 5’-adenosine monophosphate (pA) and the product 5’-inosine monophosphate (pI). Non-radioactive nucleotides were used as markers (not shown). In all panels, the no enzyme lane refers to a negative control in which tRNA was incubated with buffer in the absence of enzyme; this also served as a control for background during quantitation. Samples were incubated with wild type (left panel) or mutant enzymes (middle and right panel) for 60, 90, 120 and 180 minutes (lanes 1-4, respectively). B) This panel shows a graph of a time course of similar reactions as above but for extended periods of time. AD2 C136A denotes a substitution of one of the Zn- coordinating cysteines in TbADAT2 expressed in the context of a heterodimer with wild type TbADAT3. AD3 C291A denotes the analogous substitution but in TbADAT3 in the context of wild type TbADAT2. C) Steady-state kinetic analysis of the TbADAT2 C136A mutant by incubation with increasing saturating concentrations of tRNA substrate. The reaction velocity (pmols of inosine formed per minute) was plotted against tRNA substrate concentration. This curve was fit to the Michaelis-Menten equation using SigmaPlot software. The inset shows a similar experiment but with the recombinant wild type enzyme. Each plot represents at least 3 independent experiments.

A B

0.30

1.8

) ) 1.6 0.25 Wild type AD2 C136A 1.4 pmol AD2 C139A pmol 0.20 1.2 AD3 C291A AD3 C294A 1.0 0.15 0.8

0.6 0 mM EDTA 0.10 0.1 mM EDTA 0.4 0.5 mM EDTA 0.05

0.2 1.0 mM EDTA

Inosine produced ( produced Inosine Inosine produced ( produced Inosine 0.0 0.00 0 50 100 150 200 0 50 100 150 200 250 300 Time (min) [Zn] M

C pmol Zn2+/ Enzyme pmol dimer

Wild type 1.99 +/- 0.12 AD2 C136A 0.78 +/- 0.22

ADAT2 2.07 +/- 0.03

Figure 2.6: A single tightly bound zinc ion is sufficient for enzyme activity.

A) Time course deamination reactions as described above (Figure 2.4) were performed using recombinant mutant TbADAT2 C136A (in the context of wild type TbADAT3) in the presence of increasing concentrations of EDTA. After thin-layer chromatography 5’-adenosine and 5’-inosine monophosphate signals were quantitated using ImageQuant software. The amount of inosine produced was plotted over time for each concentration of EDTA. “No EDTA” (filled circle) is a control reaction in the absence of EDTA. 0.1 mM EDTA (open circle), 0.5 mM EDTA (filled triangle) and 1.0 mM EDTA (open triangle) represent reactions in which EDTA was added in 60, 300 and >650 molar excess compared to the enzyme concentration. B) Wild type and mutant recombinant enzymes were incubated as above (Figure 2.4) for 1 hour in the presence of increasing concentration of zinc (ZnCl2). Picomoles of inosine produced were plotted versus zinc concentration (in M). In all cases, the mutant subunit was expressed in the context of its wild type partner. C) Picomoles of Zn2+ per pmol of TbADAT2/3 heterodimer in the recombinant wild type and mutant enzyme were calculated via inductively coupled plasma-emission spectrometry (ICP). ADAT2 denotes recombinant TbADAT2 (homodimer) expressed and purified in the absence of ADAT3.

To test for the possibility of another metal, the mutant and wild type enzymes were also screened for the presence of metals other than zinc. We found no significant measurable levels (beyond buffer background) of any other metals including cobalt and manganese in our preparations (Table 2.3). Because the 6xHis-tag will bind divalent metals based on positive charge, the absence of other metals in the sample implies that the EDTA washing is indeed sufficient to remove any bound metals from the His-tag and that the observed zinc concentration is due to protein binding only. Therefore, the observation that ADAT2 C136A (as heterodimer with ADAT3) retained activity, strongly supports the view that only one zinc ion is necessary and sufficient for catalysis, although two zinc ions are tightly associated with the active wild type enzyme (Fig. 2.5 and 2.6).

2.2.3 TbADAT2 forms inactive homodimers in solution despite coordinating Zn2+

To further assess the role of Zn2+ coordination in enzymatic activity, the Zn2+ content of TbADAT2 expressed in the absence of TbADAT3 was also measured.

TbADAT2 could still bind two moles of Zn2+ per mole of homodimer (Fig. 2.6 C). This is similar to the Zn2+ content of TbADAT2/3 and likewise suggested a multimeric state for

TbADAT2 expressed alone. To explore this idea, TbADAT2 was subjected to size- exclusion chromatography. TbADAT2 has an elution volume corresponding to that of a

50 kDa protein, consistent with a stable homodimer (Fig. 2.3 C and Fig. 2.4); however, no enzymatic activity was observed even after prolonged incubation (>24 h) with substrate (Fig. 2.7). Therefore, while binding the amount of Zn2+ typically found in dimeric deaminases (e.g. bacterial ADATa), TbADAT2 by itself could not catalyze adenosine deamination of its natural substrates (tRNAThr, tRNAVal, etc). Likewise ADAT3

)

pmol 3

1 Wild type ADAT2

( produced Inosine 0 0 50 100 150 200 250 300

Time (min)

Figure 2.7: TbADAT2 in the absence of TbADAT3 has no detectable A to I activity.

A time course deamination reaction in which wild type ADAT2/3 or ADAT2 homodimer was incubated with site-specifically labeled tRNA for increasing amounts of time. Inosine produced (in pmol) was plotted against time. 4 pmol of inosine produced represents 100% adenosine to inosine conversion.

Table 2.3: Metal content in the ADAT2 complex

Wild type and ADAT2 C136A are heterodimers (with ADAT3). ADAT2 expressed in the absence of ADAT3 is a homodimer.

Element Wild type ADAT2 C136A ADAT2

Aluminum ------Boron 0.20 ± 0.05 --- 0.30 ± 0.07 Barium ------Calcium 0.26 ± 0.10 0.20 ± 0.08 0.28 ± 0.10 Cadmium ------Cobalt ------Chromium ------Copper 0.06 ± 0.02 ------Iron ------Lead ------Potassium ------Magnesium ------Manganese ------Molybdenum ------Nickel ------Sodium ------Strontium ------

Numbers reported are the average of three trials +/- standard error. “---” denotes no detection above buffer background.

by itself could not catalyze the reaction (Gerber and Keller 1999; Rubio et. al.

2007)(Rubio et. al. 2008), perhaps hinting at the acute functional interdependence of the two subunits in the heterodimer.

2.2.4 TbADAT2/3 binds one tRNA per heterodimer

Other deaminases, notably ADATa, bind 2 tRNAs per homodimer, 1 per subunit, as shown in the co-crystal of the S. aureus enzyme bound to a tRNAArg stem loop. The model for activity based on a single catalytic Zn2+, as seen in the ADAT2 C136A mutant, suggests that there is only one active site, despite the heterodimeric wild type enzyme coordinating two zinc ions. It is therefore postulated that the heterodimer, differing from other dimeric deaminases, might bind only a single tRNA substrate per catalytic cycle.

To determine the number of tRNAs bound to TbADAT2/3, 32P-labeled tRNAVal was incubated with increasing concentrations of 35S-Met/Cys-labeled TbADAT2/3; this double labeling permitted determination of an accurate stoichiometry in the RNP complex.

Electrophoresis mobility shift assays showed that approximately 1 pmol of TbADAT2/3 binds 1.06 pmol of tRNA (Fig. 2.8). Similar numbers were obtained when the protein concentration was kept constant and the tRNA concentration was varied. This indicates that the heterodimer binds a single tRNA per catalytic cycle. Recombinant TbADAT2 alone was tested in similar binding studies. This protein, however, showed severe defects in substrate binding (Fig. 2.9), partly explaining its lack of enzymatic activity.

2.3 Discussion

Common to all polynucleotide deaminases is the use of a hydrolytic mechanism mediated by zinc found in their active sites. This metal co-factor is coordinated by the evolutionarily conserved motif H(C)…CxxC, a key signature sequence for members of

A C 30 [Protein] 25 20 tRNA-enzyme 15 complex

10 [tRNA] fmol [tRNA] Free tRNA 5 0 0 5 10 15 20 25 30 [ADAT2/3] fmol B 32P in tRNA in 35S in AD2/3 in Ratio of Complex* Complex complex* Complex tRNA to (cpm) (fmol) (cpm) (fmol) heterodimer 0 0 0 0 264 1 137 1 1.09 638 3 304 2 1.19 1562 7 921 7 0.96 4120 18 2034 16 1.14 5982 26 3662 28 0.93

*The specific activities of [32P]-tRNAVal and [35S]-ADAT2/3 were 233,540 and 132,334 cpm/pmol , respectively

Figure 2.8: Wild type TbADAT2/3 binds 1 tRNA per heterodimer.

A) 35S-labeled recombinant TbADAT2/3 was incubated with increasing concentrations of 32P- labeled tRNAVal; protein-tRNA complexes were separated from free tRNA via polyacrylamide gel electrophoresis. The first lane is a control reaction in which the RNA was incubated with buffer only. Arrows indicate where free tRNA versus TbADAT2/3-tRNA complex migrates in the gel. B) Each band from (A) was individually excised from the dried gel, added to scintillation liquid and counted for both 35S and 32P (expressed in counts per minute, cpm). Control lanes with tRNA or protein alone were counted and used to calculate the specific activity of each reactant (not shown). Counts per minute (cpm) of 35S-TbADAT2/3 and 32P-tRNAVal per band are shown. Cpm values were converted to fentomoles using the calculated specific activity of the tRNA or protein as labeled. The last column displays the ratio of tRNA per enzyme (heterodimer). C) The data from (B) was used to plot the tRNA concentration versus the protein concentration in each protein-tRNA complex. This plot yields a slope of 1 which is consistent with 1 tRNA bound per heterodimer.

[Protein] M No protein No

Complex

Free probe

Figure 2.9: TbADAT2 binds tRNA substrate with lower affinity than wild type ADAT2/3

6xHis-tagged TbADAT2 in the absence of ADAT3 was over-expressed in E. coli and purified using Ni2+ affinity (see Materials and Methods Appendix A). Increasing concentrations of ADAT2 Val (0 to 2.5 M) was incubated with end radio-labeled G34 containing transcript tRNA . The “no protein” lane is a mock reaction in which no protein was added to the RNA; this serves as a marker for the migration of free tRNA in the gel. Arrows indicate the free tRNA (“Free probe”) and the protein-tRNA complex (“Complex”).

the CDA superfamily. In addition, a majority of nucleotide and polynucleotide editing deaminases exist in cells as functional multimers, generally homodimers or homotetramers, formed by identical subunits (MacElrevey and Wedekind 2007; Salter et. al. 2009). Regardless of their multimeric state, one Zn2+ ion is bound per subunit and it has been suggested that in many cases each active site acts independently. However, given that each subunit is identical in the multimeric state, it has been difficult to assess what the general contribution of each active site is to enzyme activity and specificity.

Which residues are critical to substrate recognition and activity and to what extent are questions that are especially important with polynucleotide deaminases in that the system must ensure that only the intended nucleotide target is deaminated because rampant deamination could lead to the creation of defective products that could have serious consequences to the health and viability of cells.

Unique among polynucleotide deaminases are the eukaryotic ADAT2/3 enzymes composed of two different subunits that function as heterodimers. In these enzymes, each subunit still presumably coordinates a Zn2+ but, as described earlier, the larger subunit (ADAT3) is missing the key catalytic glutamate and contains numerous additional amino acid substitutions making it distinct from the other subunit (ADAT2) in size and sequence. The naturally occurring asymmetry in the ADAT2/3 heterodimer was exploited to assess the relative contribution of each subunit to catalysis. A cysteine to alanine substitution in one of the key Zn2+-coordinating residues of TbADAT2 (C136A) still maintained a substantial amount of activity. This is surprising given that ADAT2, which contains the full set of residues needed for catalysis, was proposed to be the catalytic subunit of the enzyme (Gerber and Keller 1999). Two different possibilities may explain these results: 1) when C136 is replaced by alanine, TbADAT2 can still

coordinate a Zn2+ via vicinal cysteines or histidines, used as an alternative to the highly conserved Zn2+-coordinating H90…C136xxC139 triad. This would then yield an enzyme that remains active by virtue of still having two Zn2+ ions bound in its active site; or 2) a single Zn2+ is necessary and sufficient for activity. The second possibility is strongly supported by observations obtained from ICP, that the catalytically active C136A mutant only has one bound Zn2+ per heterodimer.

Similar amino acid substitutions of zinc coordinating residues in nucleotide deaminases led to inactive or impaired enzyme activity, but that activity could be rescued by Zn2+ addition, indicating that, despite losing Zn2+ due to the substitution(s), the enzyme maintained a pre-formed active site (Smith et. al. 1994). Here the addition of increasing concentrations of ZnCl2 to the TbADAT2/3 C136A, TbADAT3 C291A, etc. failed to rescue activity, suggesting that the ADAT2/3 active site is not pre-formed but is instead formed during heterodimerization in the presence of zinc, and while the presence of bound zinc does not seem to affect dimerization, bound zinc is needed for the correct conformation of the active site (zinc metalloenzyme active sites reviewed in

Auld 2001). The structural importance of zinc is further corroborated by the experimental observations that overnight incubation with 1,10-phenanthroline, a strong zinc chelator, caused the protein to precipitate out of solution suggesting that zinc is critical for the protein and active site’s structural integrity.

Studies with ScADAT2/3 showed that ADAT2 by itself is inactive; the same is true of TbADAT2. Likewise, ADAT3 in the absence of ADAT2 is also not active (Gerber and Keller 1999; Rubio et. al. 2007). Studies herein show TbADAT2 can form stable homodimers in solution and bind 2 zinc ions per dimer (or 1 Zn2+ per subunit) analogous to the ADATa enzyme. This means that despite dimerizing and binding Zn2+ there is still

interdependence on ADAT3 for activity. Interestingly, the TbADAT2 homodimer could bind tRNA, although very poorly (Fig. 2.9), with a dissociation constant (Kd) in the micromolar range (compared to the nanomolar range of the 2/3 heterodimer). This suggests that its inability to catalyze the reaction is partly due to a substrate binding defect, but it is not clear if lack of activity is also due to improper Zn2+ alignment in its core, whose geometry can be of crucial importance for deaminase activity. These observations indicate that the active site local structure is very sensitive to amino acid substitutions, but importantly the fact that a single Zn2+ is sufficient for activity supports a model by which TbADAT3 contributes to catalysis and is not simply a passive player in the reaction. A corollary of these observations is the fact that TbADAT2/3 contains a single active site and therefore must bind one substrate per heterodimer. Indeed EMSA binding studies show that the stoichiometry of enzyme to substrate is 1:1, clearly supporting this proposal.

How then could the Zn2+ be coordinated in the ADAT2 C136A mutant? Again there are two viable possibilities. One possibility is that Zn2+ coordination most closely resembles what is observed in other deaminases, including the bacterial ADATa, where each subunit binds one Zn2+ (intra-subunit coordination) (Elias and Huang 2005; Kim et. al. 2006; Kuratani et. al. 2005; Losey, Ruthenburg, Verdine 2006) (Fig. 2.10 A, left panel and Fig. 2.11, top panel). In this model, due to the lack of the key catalytic glutamate in

TbADAT3, TbADAT2 would be the only catalytic subunit. A pitfall of this model is that it does not explain how mutations that impair Zn2+-coordination by TbADAT2 can still produce an active enzyme. However, an alternative inter-subunit Zn2+-coordination model for TbADAT2/3 (Fig. 2.10 A, right panel and Fig. 2.11, bottom panel)

A COOH Active site with intra-subunit Active site with alternative inter- (ADAT2) Zn2+ coordination subunit Zn2+ coordination (ADAT3) COOH

HOOC (ADAT3)

COOH B C COOH (ADAT2) (ADAT2)

COOH (ADAT3) Val His Cys Pro Cys

Cys Pro Ala

Pro Cys His

Glu HOOC (ADAT3)

(ADAT2) HOOC

Figure 2.10: Structural modeling supports the possibility of alternative inter-subunit zinc coordination in TbADAT2/3.

A) Using the ADATa (TadA) homodimer (co-crystallized with tRNA anticodon loop) from Staphylococcus aureus (2B3J in the Protein Data Bank) as a modeling template, two ADAT2/3 heterodimer models were built using the FRankenstein’s Monster method (Kosinski et. al. 2003). The model in the left panel represents the heterodimer most like the ADATa homodimer in that each subunit individually coordinates one Zn2+ (intra-subunit coordination). The model in the right panel represents an alternative “swapped” model, which was built by exchanging parts of ADAT2 and ADAT3 containing the HXE and CXXC motifs; this model supports the possibility of inter- subunit Zn-coordination. The models are colored by subunits; ADAT2 in dark blue and ADAT3 in dark pink. Active site residues (His, Cys and Cys) of ADAT2 and ADAT3 are green and red respectively. B) Superposition of the two models in (A) reveals only minor structural difference between the two models. The intra-subunit model (A, left panel) is shown in orange while the inter-subunit “swapped” model (A, right panel) is shown in teal. Also highlighted are the C-termini of each protein subunit (denoted by COOH). C) Superposition of the active sites from the two different models shows the active sites are nearly identical. The panel is colored as in (B). In all models, magenta spheres represent zinc ions.

COOH (ADAT2)

Active site with intra-subunit Zn2+ coordination

HOOC (ADAT3)

COOH (ADAT3)

Active site with alternative inter- subunit Zn2+ coordination

(ADAT2) HOOC

Figure 2.11: Structural model supports the possibility of alternative inter-subunit zinc coordination in TbADAT2/3 (colored by conservation).

This figure shows the same models as explained in figure 2.10. with emphasis placed on the amino acid conservation. The models are colored based on conservation of amino acids: blue is invariant, red is highly variable, yellow and green are intermediate. Magenta spheres represent Z2+. The C-terminal ends (COOH) are labeled for each subunit.

is suggested here. This model is strongly supported by the observation that a single Zn2+ ion is necessary and sufficient for activity in the context of a heterodimer. Three- dimensional structure modeling provides a molecular model for how this inter-subunit

Zn2+-coordination is possible.

Two different three-dimensional models were generated for the TbADAT2/3 heterodimer catalytic domains based on the ADATa (TadA) homodimer using the

FRankenstein’s Monster method (Kosinski et. al. 2003) (Fig. 2.10 A, left panel and Fig.

2.11, top panel). An alternative “swapped” model, built by exchanging parts of ADAT2 and ADAT3 containing the HXE and CXXC motifs to form hybrid active sites, supports, at the very least, the possibility that ADAT3 contributes to catalysis via active site zinc coordination (Fig. 2.10 A, right panel and Fig. 2.11, bottom panel). To highlight the similarity (and likelihood of the swapped model), the two models in Figure 2.10 A were superposed (Fig. 2.10 B), showing negligible differences in the two predicted structures and nearly identical active sites (Fig. 2.10 C).

Although both models are in line with the observed 1:1 stoichiometry for the wild type enzyme, the second inter-subunit coordination model suggests that both subunits partake in catalysis, and that ADAT3 is not simply a structural component as previously speculated. Clearly, the single Zn2+ that remains bound to the ADAT2 C136A mutant adopts the proper geometry and permits proper substrate positioning in the active site.

This is in contrast to TbADAT2 alone, where despite forming a homodimer and binding two Zn2+ ions, it still remains inactive. Admittedly, it is quite puzzling that an alanine substitution of the other TbADAT2 Zn2+ coordinating cysteine (C139A) yields an inactive enzyme. It is possible that this mutation, more so than C136A, affects not only Zn2+- coordination but also local protein structure and perhaps substrate binding. Consistent

with this view, alanine substitutions in either of the two cysteines (C136A or C139A) of

TbADAT2 led to defects in tRNA binding with the defect seen in the TbADAT2 C139A mutant being far greater than that observed in TbADAT2 C136A, a point that will be elaborated further in Chapter 3. Analogous changes in TbADAT3 had negligible effects on substrate binding but deleterious effects on enzyme activity (Ragone, Spears et. al. in press RNA).

Taken together, the observations presented here have broader implications in the evolution of editing deaminases. ADAT2/3s from various organisms resemble cytidine deaminases in their primary sequence, despite catalyzing A to I editing reactions. While it remains feasible that all polynucleotide deaminases evolved from an ancestral cytidine deaminase, it may not be necessarily true that all polynucleotide deaminases are derived via gene duplication of ancestral tRNA enzymes as has been proposed (Gerber and Keller 1999; Schaub and Keller 2002). For example, it has been suggested that ADAT1 (adenosine deaminase for position 37 of eukaryotic tRNAs) evolved from acquisition of an additional motif (motif 3) (Fig. 2.1) by ADAT2. However, the observed sensitivity of TbADAT2 and TbADAT3 to seemingly modest amino acid substitutions near the active site suggest otherwise. Acquisition of motif 3 would have required addition of an average of 60 amino acids between the two Zn2+-coordinating cysteines of ADAT2, which poses a significant structural constraint to this motif acquisition to yield an active enzyme. The structure of ADAR2 also provides a clue to the distinct evolutionary path of these enzymes. ADAR2 and ADAT1 share the common feature of requiring binding of inositol hexakisphospate (IP6) (Macbeth et. al. 2005) for activity. This essential co-factor is bound to these enzymes by evolutionarily conserved residues, none of which are obvious in other deaminases, including ADAT2. Thus, for

ADAT1 to evolve from ADAT2, it would require an additional domain acquisition beyond the 60 amino acid spacer between the active site cysteines, making this evolutionary event even more unfavorable. In fact, phylogenetic comparative analysis of nucleotide and polynucleotide deaminases from a variety of organisms, including single-cell protists

(like trypanosomes), do not support an evolutionary model by which mRNA adenosine deaminases arise from ancestral tRNA anticodon deaminases (Conticello et. al. 2005)

(Fig. 2.12). Given the data presented here, the active site structure of TbADAT2/3 is predictably different from other deaminases in the way it may coordinate its catalytic

Zn2+, whether or not this is true of all ADAT2/3 deaminases will remain an open question and awaits further examination.

In terms of specificity, what could the eukaryotic enzymes possibly gain by incorporating TbADAT3 and adopting a heterodimeric structure? While a homodimer may confer certain functional advantages including cooperativity in substrate recognition, a heterodimer affords a rapid route for functional diversification. Since the heterodimeric eukaryotic deaminases display broader substrate specificity compared to their homodimeric bacterial homologs, it is conceivable that versatility in substrate recognition is the result of gene duplication and divergence of an ancestral ADAT to encode the two subunits in the heterodimeric ADAT2/3. While it is premature to delineate the exact functional inter-dependence between these two subunits, it is evident that fully realizing and preserving the functional gains afforded by each subunit will require obligatory crosstalk between the two. The gradual acquisition of new functions (as in ADAT2/3) through modular additions is not unexpected. However, the likely repositioning of a catalytic metal ion, from independently binding each primordial subunit, to now bridge the two subunits is an elegant solution to generate an efficient and high-fidelity enzyme.

This inter-subunit metal coordination arrangement may best exploit the substrate-binding and catalytic properties of the two subunits while still enabling the evolution of multi- substrate specificity.

C.elegans ADAR ADAR

Figure 2.12: Evolutionary relationships among ADARs, ADATa, and Cytidine deaminases

A Bayesian phylogenetic tree depicting the evolutionary relationships among ADAT1, ADAR, ADAT2 (eukaryotic), ADAT3 (eukaryotic), ADATa (bacterial ADAT), CDA (cytidine deaminase), and dCMP was calculated using protein sequence alignments that correspond to the conserved catalytic deaminase core. This tree does not support the notion that all RNA deaminases (i.e. ADAT1 and ADARs) arise from gene duplication events in an ancestral tRNA deaminase such as ADAT2. Instead the tree suggests that ADARs and ADAT1 form an evolutionary clade distinct from ADATa, ADAT2, and ADAT3. The display is presented using MEGA4 (Tamura et. al. 2007).

Chapter 3:

TbADAT2/3 has Two Distinct tRNA Binding Domains that Act Cooperatively

3.1 Introduction

As discussed previously (chapter 1 and chapter 2), adenosine to inosine deamination in RNA (tRNA, mRNA, miRNA, and rRNA) is an evolutionarily conserved reaction that occurs by hydrolytic deamination mediated by an essential zinc cofactor and a proton shuttling glutamate. When this deamination reaction occurs at the wobble position in tRNAs, inosine effectively allows a single tRNA to decode three different codons for the same amino acid obviating the need for additional tRNAs to be encoded in genomes. Inosine formation at the wobble position is thus essential in bacteria as well as eukaryotes. Although an essential enzyme, there are still significant gaps in the knowledge of ADATs, the enzymes responsible for A to I conversion, especially in terms of how they recognize and bind their tRNA substrate.

Their similarities with cytidine deaminases led Keller and co-workers to propose that evolutionarily all RNA deaminases are derived from an ancestral cytidine deaminase

(Gerber and Keller 1999). Adding credence to this hypothesis, recent collaborative studies from the Alfonzo and Papavasiliou labs showed that the TbADAT2/3 enzyme could perform both types of deaminations, albeit in different substrates: A to I in tRNA and C to U in DNA (Rubio et. al. 2007). Therefore TbADAT2/3 maintains some of

the characteristics of Keller's ancestral deaminase, but it is still not clear how this enzyme achieves substrate recognition or catalytic flexibility. A number of bacterial

ADAT structures (Elias and Huang 2005; Kim et. al. 2006; Kuratani et. al. 2005) and the recent co-crystal of the Staphylococcus aureus ADATa (SaADATa) complexed with the anticodon stem loop (ASL) of tRNAArg has given insights into enzyme-substrate interactions by the bacterial enzyme (Losey, Ruthenburg, Verdine 2006). The SaADATa homodimer binds the ASL via residues within the enzyme's active site while the tRNA backbone contributes minimally to RNA binding. However, the eukaryotic ADAT2/3 requires a full length tRNA for deamination activity suggesting that the substrate recognition in eukaryotes is different and that tRNA tertiary structure likely contributes to binding (Auxilien et. al. 1996).

While a great deal of effort has been invested to characterize bacterial ADATs, little is known about the binding interactions of ADAT2/3 with its tRNA substrates. The bacterial structures can serve as a guide but there are many significant differences between ADATa and ADAT2/3 to consider when examining how ADAT2/3 binds to tRNA. First, ADATa is a homodimer with two complete active sites (Elias and Huang

2005; Wolf, Gerber, Keller 2002). The eukaryotic enzyme, however, is a heterodimer with only one complete active site; there is a second ―active site‖ but it is incomplete by virtue of a naturally missing glutamate in ADAT3 (pseudo-active site). Current studies

(Chap. 2) have shown that the necessary and sufficient active site Zn2+ is coordinated by the typical H…C…C triad but in an atypical way; the coordination of zinc is inter-subunit with the histidine from ADAT2 (H93) and the two cysteines from ADAT3 (C291 and

C294) creating the complete active site (chapter 2, Spears et. al. JBC in press and Fig.

3.1). The role of the second,

Pseudo- Active site active site

90 92 136 139 TbADAT2 N-term HALAHAEFVAVEELLRQATA----GNCGAVSQDLADYVLYVVVEPCIMCAA

Zn2+ Zn2+

TbADAT3 C-term AMSCMVCPEHSVFMDMGNAL------—VERDKCQRTVEKLVFTVPHDLV 294 291 252 Figure 3.1: TbADAT2/3 heterodimer has one active site and one pseudo-active site.

A schematic of TbADAT2/3 highlighting the presence of an active site and a pseudo-active site. Typical active site residues conserved throughout the deaminase families are highlighted in black. Dark gray spheres represent Zn2+ while gray dotted lines highlight the amino acids responsible for zinc coordination. The active site which is made up of ADAT2 H90, E92 and ADAT3 C291, C294 contains the glutamate (ADAT2 E92) essential for deaminase activity while the pseudo-active site contains a valine (ADAT3 V252) instead of glutamate and is thus not a complete active site.

incomplete active site and its importance in the eukaryotic deaminases has yet to be determined but will be focused on in this chapter. Secondly, ADATa recognizes just the tRNA anticodon stem loop and binds two tRNAs, one per active site, while ADAT2/3 recognizes only a full length tRNA and likely binds only one tRNA per catalytic cycle. A single tRNA bound per catalytic cycle has been confirmed in T. brucei ADAT2/3 but whether this holds true for ADAT2/3 in all organisms remains an unanswered question.

In addition to the differences in substrate recognition (anticodon stem-loop vs full length tRNA), ADAT2/3 likely differs from its bacterial counterpart in that the RNA binding domain has to be more relaxed to accommodate seven or eight tRNAs instead of just one.

The focus of this chapter is on investigating how the differences between ADATa and ADAT2/3 affect substrate recognition and binding. The importance of the heterodimeric nature of the protein and how the atypical interplay between the subunits contributes to binding are also explored. The Huang laboratory recently proposed that

ADAT2/3 contains an RNA binding site away from the active site; in silico work in the

Alfonzo laboratory confirms that ADAT2 contains key tRNA binding residues at its C- terminus (KR-domain). Indeed, deletion or substitution of these residues leads to defects in substrate binding and loss of enzymatic activity. Furthermore, these studies show that the pseudo-active site also contributes to binding and acts cooperatively with the KR domain to bind and correctly position the tRNA for catalysis. Since similar motifs are found in nearly all eukaryotic ADAT2/3s but are absent in most bacterial ADATs, the data presented here suggests that these motifs represent a general mode of tRNA recognition and binding in the eukaryotic tRNA deaminase enzymes. The data also support a previous model in which through evolution, and in order to accommodate

many different substrates, binding domains have been appended away from the enzyme's active site providing critical substrate binding functions (Elias and Huang

2005).

3.2 RESULTS

3.2.1 Recombinant TbADAT2/3 stably binds tRNA

As a prelude to identifying regions of TbADAT2/3 important for tRNA binding, an

Electrophoretic Mobility Shift Assay (EMSA) was established by incubating recombinant protein with in vitro-transcribed radioactively labeled tRNAVal (one of ADAT2/3’s natural substrates). Under these conditions a band, which migrated more slowly in the gel compared to an RNA alone control, was observed only in the presence of protein (Fig.

3.2 A, right panel), suggesting formation of a stable complex between the recombinant protein and the tRNA. Additionally, larger complexes, which did not enter the gel, were observed at the highest protein concentrations. A dissociation constant (Kd) of 1.31 ±

0.83 M was estimated for this tRNA. Similar values were obtained even when the larger complexes were included in the calculations. Although this Kd approximates the observed Km of the wild type enzyme for the same tRNA (0.75 M), the Kd was consistently higher than the Km.

With the bacterial ADATa, a stable tRNA-protein complex was only obtained by replacing the target adenosine (A34) by zebularine, a purine analog that resembles the proposed transition state for the deaminase reaction. Presumably, the use of the zebularine-containing tRNA prevents turnover of the substrate, which led to a stable enzyme tRNA-complex. This approach led to the co-crystallization of the bacterial enzyme with RNA (Losey, Ruthenburg, Verdine 2006).

[Protein] M [Protein] M [Protein] M A

complex

free probe Val Val Val tRNA (G34 ) tRNA (I34 ) tRNA (A34 )

1.0 total

0.8 /[tRNA] Val tRNA G 0.6 34 tRNAVal K (μM)

complex Val d tRNA I34 Val tRNA A

0.4 34 G34 containing 0.15 ± 0.02 Protein]

- I34 containing 0.34 ± 0.11 0.2

A34 containing 1.31 ± 0.83 [tRNA

0.0 0.0 0.2 0.4 0.6 0.8 1.0

[Protein] (M)

Figure 3.2: Wild type TbADAT2/3 stably binds tRNA in vitro.

Radiolabeled tRNAVal (8 nM) from T. brucei was incubated with increasing concentrations of recombinant TbADAT2/3 expressed in E. coli and purified by Ni2+-chelate chromatography. A) Representative Electrophoretic Mobility Shift Assay (EMSA) polyacrylamide gel to determine the extent of tRNA binding. Right panel shows TbADAT2/3 binding to an A34-containing tRNA (natural substrate). Middle and left panel show a similar experiment but with either an I34 or G34-containing tRNA. In all panels, the first lane shows a mock reaction in which the tRNA was incubated in binding buffer in the absence of enzyme. Lanes 2-6, tRNA incubated with increasing concentrations of the enzyme (0.052, 0.18, 0.35, 0.65, 1.18 M, respectively). "Free probe" denotes the migration of the unbound tRNA and "complex" denotes the migration of the TbADAT2/3 bound tRNA. B) The reaction products from (A) were used to calculate the fraction of tRNA bound by calculating the percent of the probe shifted divided by the total (bound and unbound) probe in each reaction. These values were plotted against TbADAT2/3 concentration in

M and fitted to a single exponential curve. The dissociation constant (Kd) was calculated by non- linear regression using SigmaPlot software.

Not satisfied with the amount of observed binding to the natural A34-containing tRNA, attention was turned to product mimics. The binding of the T. brucei enzyme to different tRNA derivatives was explored to help identify a more stable complex that could be used to probe potential TbADAT2/3 RNA-binding motifs. Since zebularine is not readily

Val available, stable interactions with the reaction product inosine 34 (I34)-containing tRNA

Val were examined. Inosine is a guanosine analog; therefore, G34-containing tRNA was also tested. Recombinant wild type enzyme formed stable complexes with these two substrates; however, binding was significantly improved in comparison to the A34- containing tRNA, with Kd values of 0.34 ± 0.11 M and 0.15 ± 0.02 M for the I34- and

G34-containing substrates respectively (Fig. 3.2). This binding is also specific for tRNA as shown by competition assays with specific (tRNAVal) and non-specific (T. brucei splice leader RNA) RNAs. A 2-fold molar excess of ―cold‖ tRNAVal was able to out compete radiolabeled tRNA while splice leader RNA did not completely out compete tRNA binding even at a 64-fold molar excess (Fig. 3.3). Together the data lead to the conclusion that

TbADAT2/3 can stably and specifically bind tRNA in vitro.

3.2.2 Single-turnover kinetics validates G34 containing tRNA substrates for binding studies

To validate the use of a product-mimicking tRNA for further binding studies and characterization of ADAT2/3 residues important for binding, enzyme single-turnover kinetic experiments were performed by incubation of an A34-containing tRNA with saturating concentrations of enzyme. This allowed for the calculation of observed rates

(kobs) by plotting the amount of inosine produced in these reactions over time. The data were fit to the equation f=a(1-e-kt) where f and t are inosine formed and time of incubation respectively, a represents inosine produced at the end point of the reaction

A [tRNA Val] [Splice Leader]

complex complex

free probe free probe

1 2 3 4 5 6 7 8 9 10 11 12

100

B tRNA Val 80 Splice leader RNA

% Bound % 40

0 0 20 40 60

Fold excess of competitor

Figure 3.3: Wild-type TbADAT2/3 specifically binds tRNA in vitro.

A) Representative Electrophoretic Mobility Shift Assay (EMSA) to determine the extent of tRNA binding in the presence of either unlabeled tRNA or splice leader RNA used as specific and non- specific substrate competitors, respectively. The left panel shows TbADAT2/3 binding to radioactive tRNAVal in the presence of the same ―cold‖ tRNA. The right panel shows TbADAT2/3 binding to the same radioactively labeled tRNAVal but in the presence of splice leader RNA. Lane 1 shows a mock reaction in which the tRNA probe was incubated in binding buffer in the absence of enzyme and competitor. Lane 2 shows the tRNA probe incubated with wild type enzyme in the absence of any competitors. Lanes 3-7 and 8-12, tRNA incubated with increasing excess of cold competitor (2, 4, 8, 16, 64-fold). "Free probe" denotes the migration of the unbound tRNA, and "complex" denotes the migration of the TbADAT2/3 bound tRNA. B) The reaction products from (A) were used to calculate the fraction of tRNA bound by calculating the percent of the RNA shifted divided by the total (bound and unbound) probe in each reaction. These values were plotted against competitor fold excess and fitted to a single exponential decay curve using SigmaPlot.

A B

1.0 0.16

0.8 0.12

0.6

obs 0.08 k 0.4 0.15 M 0.45 M 0.04 0.2 1.00 M

Fraction Inosine produced Inosine Fraction 5.00 M 10.00 M Kd App = 0.05 +/- 0.01 M 0.0 0.00 0 50 100 150 200 0 2 4 6 8 10

Time (min) [TbADAT2/3] M

Figure 3.4: Kinetic determination of the dissociation constant for TbADAT2/3.

A) Single-turnover assays were performed as described in the materials and methods. For each protein concentration (ranging from 0.15 M to 10 M) fraction of inosine produced was plotted -kt against time and fit to the single exponential equation f = a(1-e ). kobs values were calculated for each protein concentration. B) kobs values were plotted against TbADAT2/3 concentration and fit to a single ligand binding curve; the apparent dissociation constant (Kd App) was calculated by non-linear regression using SigmaPlot software.

and k is kobs (Fig. 3.4 A). The kobs values were subsequently used to calculate an apparent dissociation constant by plotting the kobs values versus protein concentration.

An apparent Kd of 0.05 ± 0.01 M (Fig. 3.4 B) was measured; this is in an accepted range with the Kd measured by EMSA with the G34-containing tRNA (Kd = 0.15 ± 0.02

M). Importantly, the kinetically determined Kd shows about a 25-fold improvement when compared to the Kd calculated by EMSA using the same A34-containing substrate

(compare 1.3 M to 50 nM) (Fig. 3.2 and 3.4). With these data in hand, it is concluded that using a tRNA substrate which mimics the product (ie. G34-containing tRNA) in further EMSA studies is a fair alternative to using the natural A34-containing substrate.

3.2.3 TbADAT2 has a predicted RNA binding motif at its C-terminus important for tRNA binding

The relatively high affinity of the enzyme for the G34-containing tRNA allowed for identification of regions of the enzyme important for RNA binding. Because ADAT2 is has more identity to ADATa than ADAT3 does to ADATa, and because ADAT2 has long been thought to be primary catalytic component of ADAT2/3, ADAT2 was examined for potential tRNA binding residues/domains. Bacterial ADATs can efficiently deaminate

Arg short stem loops corresponding to the ASL of tRNA ACG, the only in vivo substrate for this enzyme (Wolf, Gerber, and Keller 2002). This minimal RNA substrate was, in fact, used in the co-crystalization experiments, which revealed a number of direct contacts between the SaADATa and RNA, including a network of hydrogen bonds that involved all five nucleotides in the anticodon loop and to a lesser degree contacts with specific nucleotides in the stem (Losey, Ruthenburg, Verdine 2006; Wolf, Gerber, Keller 2002).

To investigate the specific contribution of analogous residues in the ADAT2 subunit, the sequences of SaADATa and TbADAT2 were aligned to compare whether those amino

A TbADAT2 MVQDTGKDTNLKGTAEANESVVYCDVFMQAALKEATCALEEGEVPVGCVLVKADSSTAAQ 60 SaADATA ------MTNDIYFMTLAIEEAKKAAQLGEVPIGAIITKDDE----- 35 :. ** *::**. * : ****:*.::.* *. TbADAT2 AQAGDDLALQKLIVARGRNATNRKGHALAHAEFVAVEELLRQATAGTSENIGGGGNCGAV 120 SaADATA ------VIARAHNLRETLQQPTAHAEHIAIER------AAKVLG------67 ::**.:* : :. ****.:*:*. ::: :* TbADAT2 SQDLADYVLYVVVEPCIMCAAMLLYNRVRKVYFGCTNPRFGGNGTVLSVHNSYKGCSGED 180 SaADATA SWRLEGCTLYVTLEPCVMCAGTIVMSRIPRVVYGADDPKGGCSGSLMNLLQQSN---FNH 124 * * . .***.:***:***. :: .*: :* :*. :*: * .*:::.: :. : :. TbADAT2 AALIGYESCGGYRAEEAVVLLQQFYRRENTNAPLGKRKRKDLSVV 225 SaADATA RAIVD----KGVLKEACSTLLTTFFKNLRANKKSTN------156 *::. * * . .** *::. .:* :

ADATa Enterobacter cloacae RQQKKQQKAELKSSGD------168 ADATa Cronobacter turicensis RLEKKALKEAAKTATGTDPAA------196 ADATa Shigella flexneri RQEIKAQKKAQSSTD------178 ADATa Escherichia coli RQEIKAQKKAQSSTD------178 ADATa Salmonella enterica RQEIKALKKADRAEGAGPAV------183 ADATa Citrobacter rodentium RQEIKALKKASSATE------180 ADATa Klebsiella pneumoniae REEKKALKKARAQTGES------180 ADATa Erwinia tasmaniensis RAEKKALRQRQSTDGI------186 ADATa Yersinia pseudotuberculosis REQQKALKQAQRAAEGKL------200 ADATa Photorhabdus asymbiotica REQHKALKAARRQQEENQ------169 ADATa Shewanella baltica REEKKALKQAQKAQQAIE------175 ADAT2 Trypanosoma brucei NTNAPLGKRKRK-----DLSVV------225 ADAT2 Leishmania major NPNAPGHKRRRK------A------281 ADAT2 Entamoeba histolytica NEKAPEPNKKKK-----NEEK------166 ADAT2 Mus musculus NPNAPKSKVRKK-----DCQKS------191 ADAT2 Homo sapiens NPNAPKSKVRKK-----ECQKS------191 ADAT2 Gallus gallus NPNAPKSKVRKK-----DNRK------172 ADAT2 Drosophila melanogaster NPAAPPQAKKKK------160 ADAT2 Caeorhabditis elegans NPFAPPEKRKTKK----PKLDEIK------176 ADAT2 Candida albicans NYKKPIGKYNSKKR---HFANDEE------261

Figure 3.5: Alignment of SaADATa and TbADAT2 suggests differences in tRNA binding.

A) Sequence comparison between the Adenosine Deaminases Acting on tRNA of S. aureus and T. brucei shows that a number of key residues involved in substrate binding by the bacterial enzyme have naturally undergone non-conservative changes in the T. brucei enzyme. The amino acid sequences of T. brucei ADAT2 and S. aureus ADATa were aligned using Clustal W. Conservative changes are denoted by dots ( . and : ) placed under the sequences and identical residues by an asterisk (*). Amino acids involved in RNA interaction in the co-crystal of SaADATa and an ASL (anticodon stem loop) representing he anticodon arm of S. aureus tRNAArg are shown in boldfaced letters. The amino acids in TbADAT2 shown within boxes correspond to the critical binding residues in ADATa and have been mutated to alanine and are shown to play no role in tRNA binding in T. brucei. B) ClustalW alignment of ADATa and ADAT2 C-terminal sequences from different organisms. Highlighted by the grey box is the KR-domain at the C- terminal end of eukaryotic ADAT2s.

acids critical for RNA binding in the bacterial enzyme were conserved in the eukaryotic enzyme (Fig. 3.5 A). This alignment showed that about 1/3 of the key residues in

SaADATa had already undergone non-conservative replacements in TbADAT2 during evolution. For example, the wild type sequence of TbADAT2 already contains naturally occurring alanine and valine replacements at the equivalent positions as the conserved arginine 125 and serine 138 from bacterial deaminases (Fig. 3.5 A). These observations indicate that active site residues important for the SaADATa enzyme to bind RNA may only have a minor contribution in the case of the T. brucei enzyme.

To examine the potential contribution of those residues which were conserved, individual amino acid substitutions were made to the amino acids cooresponding to two of the most critical binding residues in ADATa. The catalytic glutamate 92 and arginine

159 of TbADAT2 (Fig. 3.5 A) were changed to alanine. Each mutant was co-expressed with wild type TbADAT3 in E. coli and shown to still form a heterodimer. The mutants were tested for enzymatic activity as well as binding and as expected the catalytic glutamate mutant abolished enzymatic activity. However, neither mutation had any impact on tRNA binding (Kd = 0.21 ± 0.08 M and 0.16 ± 0.03 M for E92A and R159A, respectively) (Table 3.1). These Kd values are within experimental error of the wild type

Kd (0.15 ± 0.02 M, Table 3.1). Therefore, active-site residues involved in ADATa binding to its substrate play only minor roles (if any) in the T. brucei enzyme. This observation also suggested that if the reason for the apparent low affinity of the wild type enzyme for the A34-containing tRNA was due to the relatively rapid turnover of the substrate into product, then a catalytically defective mutant (where the conserved glutamate has been replaced for alanine), should bind to the A34-containing tRNA (in

EMSAs) with the same affinity seen for the G34-containing tRNA.

Table 3.1: Binding parameters for wild type (TbADAT2/3) and C-terminus mutants of TbADAT2.

Enzyme Kd M Wild type 0.15 ±0.02 ADAT2 C-ter10 1.20 ± 0.43 ADAT2 C-ter5 0.62 ± 0.29 ADAT2 C-ter5A 3.28 ± 0.63 ADAT2 E92A 0.21 ± 0.08 ADAT2 R159A 0.16 ± 0.03 ADAT2 K216A 0.96 ± 0.11 ADAT2 R217A 0.44 ± 0.11 ADAT2 K218A 1.10 ± 0.24 ADAT2 R219A 0.71 ± 0.11

The dissociation constants (Kd) were calculated from EMSA data fitted by non-linear regression. binding data were fitted using the SigmaPlot software.

This possibility was tested by performing an EMSA with mutant ADAT2 E92A; as expected this mutant formed a stable complex with the A34-containing tRNA with a Kd of

0.48 ± 0.13 M is within the range for the calculated Kd for a) the same mutant with the

G34-containing tRNA (0.21 ± 0.08 M, Table 3.1) and b) the wild type enzyme with the

G34 and I34-containing substrates.

Partly due to earlier reports by Grosjean and co-workers on the requirement of a full length tRNA for activity (Auxilien et. al. 1996), Huang and co-workers suggested that eukaryotic deaminases have evolved the ability to utilize many different substrates by acquiring RNA binding motifs that are distinct from active site residues (Elias and Huang

2005). Comparison of the bacterial deaminase sequences to those of the two subunits of the S. cerevisiae deaminase reveal an extension at the C-terminus of ADAT2 and at the

N-terminus of ADAT3, which they suggest may harbor the RNA binding function. A sequence comparison between bacterial ADATa and TbADAT2 identified a stretch of ten amino acids (KRKRKDLSVV) at the C-terminus of TbADAT2, termed here the KR- domain. This domain is found in all of the trypanosomatid ADAT2 protein sequences and generally across eukaryotic ADAT2 sequences but does not exist in most bacterial counterparts (Fig. 3.5 B).

The TbADAT2 sequence was also further analyzed for potential RNA binding residues with the RNABindR program (http://bindr.gdcb.iastate.edu/RNABindR/)

(Terribilini et. al. 2006) and the KR-domain showed high probability for a potential RNA binding domain (Figure 3.6). A series of ADAT2 C-terminal mutants were constructed to determine the role the KR-domain may play in substrate binding and/or deaminase activity.

MVQDTGKDTNLKGTAEANESVVYCDVFMQAALKEATCALEEGEVPVGCVLVKADSSTAAQ ------AQAGDDLALQKLIVARGRNATNRKGHALAHAEFVAVEELLRQATAGTSENIGGGGNCGAV -----+----+----+-++-++++++------SQDLADYVLYVVVEPCIMCAAMLLYNRVRKVYFGCTNPRFGGNGTVLSVHNSYKGCSGED ------++++++------AALIGYESCGGYRAEEAVVLLQQFYRRENTNAPLGKRKRKDLSVV ------+--+++++-----

Figure 3.6: KR-domain is a predicted tRNA binding domain.

The amino acid sequence of TbADAT2 was also analyzed for potential RNA binding domains by the RNAbindR program, which uses a naïve Bayesian classifier for all predictions as described by Terribilini,M. 2006 and 2007. Potential binding domains indentified denoted by (+) signs under the sequence, boldfaced letters denote the KR-domain which has been analyzed further in these studies.

Initially, two mutants were generated: ADAT2 C-ter5 which retained all positively charged residues but bears a deletion of the last five amino acids (DLSVV) and ADAT2

C-ter10 in which the last 10 residues were deleted (including all five charged residues of the KR-domain). Binding studies were performed with these two mutants and indeed the ADAT2 C-ter10 showed a defect in binding and yielded an 8-fold increase in Kd compared to the wild type (Fig. 3.7 A, Table 3.1). The ADAT2 C-ter5 mutant, however, showed a smaller binding defect (Kd= 0.62 ± 0.29). A mutant in which the 5 charged residues were replaced by alanine (ADAT2 C-ter5A) showed the greatest binding defect with a Kd of 3.28 ± 0.63 M (Fig. 3.7 B, Table 3.1) and in turn could not support any detectable enzymatic activity which will be discussed further in the following section (see

3.2.4). This lack of activity is similar to what was observed with the 10-amino acid deletion mutant (ADAT2 C-ter10). The 3-fold increase in Kd of ADAT2 C-ter5A over the

ADAT2 C-ter10 mutant, however, suggests that the string of alanine residues may affect the local structure beyond the binding defect seen with the 10-amino acid deletion.

To study the specific contribution of individual residues further, a series of mutants were generated in which single amino acids of the KR-domain were replaced by alanine residues (Fig. 3.8). The Kd defects ranged in magnitude from a 3- and 4.7-fold increase for ADAT2 R217A and ADAT2 R219A and as high as a 7-fold increase for

ADAT2 K216A and ADAT2 K218A. Interestingly, the lysine to alanine mutations had similar and comparable effects on binding as the 10-amino acid deletion. The arginine for alanine replacements while still affecting Kd, had smaller effects (Table 3.1).

Regardless, these data indicate that the bulk of the binding is likely due to the contribution of the positively charged residues, specifically the lysines in the KR-domain.

A B [Protein] M [Protein] M

complex complex free free probe probe

ADAT2 C-ter10 ADAT2 C-ter5A

C D

0.5 0.5

0.4 0.4

Total Total

0.3 0.3

tRNA tRNA

/ /

0.2 0.2

Complex Complex

0.1 0.1

tRNA tRNA Kd = 1.20 0.43 M Kd = 3.28 0.63 M 0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 0 1 2 3 4 5 6 [Protein] M [Protein] M Figure 3.7: The C-terminus of TbADAT2 is critical for tRNA binding.

Two mutants were generated bearing either a deletion of the last 10 amino acids of TbADAT2 (ADAT2 C-ter10) or a replacement of the 5 positively charged amino acids of the KR-domain by alanines (ADAT2 C-ter5A). These TbADAT2 mutants were co-expressed with wild type TbADAT3 in E. coli and the resulting recombinant proteins (heterodimers) were purified by Ni2+-chelate Val chromatography. These mutants were used in EMSAs. A) Radioactive G34-containing tRNA was incubated with increasing concentrations of recombinant ADAT2 C-ter10/TbADAT3 heterodimer (materials and methods). B) A similar experiment as in (A) but with ADAT2 C- ter5A/TbADAT3. "Free probe" denotes the migration of the unbound tRNA, "complex" denotes the migration of the protein-bound tRNA, as also highlighted by arrows. Single ligand binding curves for for ADAT2 C-ter10 (C) and ADAT2 C-ter5A (D). tRNA bound (complex) was plotted versus protein concentration and data were fit to a single ligand binding curve using SigmaPlot. Kd values as shown were calculated from each respective curve. Each curve represents the average of at least three independent trials. The error bars show the range among the trials.

[Protein] M A B

complex

free probe 0.5 ADAT2 K216A

0.4 Total complex

0.3 tRNA free / probe 0.2

ADAT2 R217A Complex ADAT2 K216A ADAT2 R217A

tRNA 0.1 ADAT2 K218A complex ADAT2 R219A 0.0 free 0.0 0.5 1.0 1.5 2.0 2.5 probe ADAT2 K218A [Protein] M

complex free probe ADAT2 R219A Figure 3.8: The KR-domain of TbADAT2 is critical for tRNA binding.

Single alanine replacement mutants were generated bearing a substitution of each of the positively charged residues of the KR domain for alanine. These TbADAT2 mutants were co- expressed with wild type TbADAT3 in E. coli and the resulting recombinant proteins (heterodimers) were purified by Ni2+-chelate chromatography. These mutants were used in Val EMSAs. A) Radioactive G34-containing tRNA was incubated with increasing concentrations of recombinant KR-domain mutant ADAT2/3 heterodimers. Each mutant is listed below a representative EMSA gel. "Free probe" denotes the migration of the unbound tRNA, "complex" denotes the migration of the protein-bound tRNA. B) Single ligand binding curves for KR-domain mutants. tRNA bound (complex) was plotted versus protein concentration and data were fit to a single ligand binding curve using SigmaPlot. Kd values were calculated from each respective curve. Each curve represents at least 3 independent trials.

Given the behavior observed with mutant ADAT2 C-ter5A and the structural argument made above, it is possible that some of the observed defects are due to more global changes in the overall structure of the enzyme. We have thus analyzed wild type and some of the mutants of TbADAT2/3 by circular dichroism spectroscopy. TbADAT2/3 contains a total of 15 phenylalanines, 15 tyrosines and 6 tryptophans (5 phenylalanines,

9 tyrosines and no tryptophan in TbADAT2 and 10 phenylalanines, 6 tyrosines and 6 tryptophans in TbADAT3). Measurements were taken at wavelengths between 200 and

260 nm. Little differences were observed between the mutants described in table 3.1 and the wild type. The scans showed characteristic minima at 208 and 222 nm (- helices) and at 218 (-sheets) (Fig. 3.9).

The minor differences seen at various points in the scans for the mutants compared to the wild type are within the experimental error of the machine on which the circular dichroism experiments were performed. Each protein sample was scanned at least 3 times and raw data from each scan is averaged to give the data used to create each plot in figure 3.8. As shown, with the exception of E92A at 224nm, the deviations in the mutant scans from the wild type are all within the typical variability for independent scans of the wild type alone indicating that the small differences seen are not statistically significant. In addition, size-exclusion chromatography showed that all of the mutants had identical elution profiles as the wild type, in that they all formed heterodimers with

TbADAT3 (Fig. 3.10). Therefore the KR-domain plays a critical role in substrate binding and the observed differences in binding behavior between the mutants and the wild type are not likely due to indirect effects from major disruptions in the global structure of the enzyme.

1000

)

1 -

-1000

1 - -2000

(deg dm (deg -3000

Wt ADAT2/3 -4000 ADAT2 E92A

Ellipticity ADAT2 R159A ADAT2 C-ter 10 -5000 ADAT2 C-ter C5 ADAT2 C-ter5A -6000 190 200 210 220 230 240 250 260 270

Wavelength (nm)

Figure 3.9: None of the protein variants had drastic effects on the overall structure of TbADAT2/3.

Both wild type recombinant TbADAT2/3 and the various mutants from figures 3.6 and 3.7 were analyzed by circular dichroism. The UV spectra were taken at various wavelengths between 200 and 260 nm. Molar elipticity ([]) was calculated as described in the Materials and Methods and plotted as a function of wavelength. None of the mutants had significant spectral deviations when compared to the wild type (within error of multiple scans of the wild type TbADAT2/3) indicating that no major structural rearrangements had taken place.

A B Mol. wt Void 158 69 17 3 (kDa) 2 Wild type TbADAT2/3 1 0.9

0.8 Vitamin B12 (1.35 kDa) 0 0 3 6 9 12 15 18 21 24 0.7

20 ) 0.6 Myoglobin (17 kDa) 15 ADAT2 mAU C-ter10 0.5 av Ovalbumin (44 kDa) 10 K 0.4 TbADAT2/3 (69 kDa) 5 ADAT2 C-ter10 0.3 (68 kDa) Gamma globulin Abs 280nm ( 280nm Abs 0 (158 kDa) 0 3 6 9 12 15 18 21 24 0.2 ADAT2 C-ter5A (69 kDa) Thyroglobulin 15 0.1 (670 kDa)

ADAT2 10 0.0 C-ter5A 1 10 100 1000 5 Molecular weight (kDa)

0 0 3 6 9 12 15 18 21 24 Retention Volume (ml) Figure 3.10: TbADAT2 C-terminal mutants still heterodimerize with ADAT3.

A) Wild type TbADAT2/3 and the two C-terminal mutants were fractionated by size-exclusion chromatography as indicated in the Materials and Methods section. Both mutants eluted with a size consistent with that calculated for the wild type (~68 kDa), again suggesting that these mutations cause no major effects on the multimeric state of these mutants. B) Kav vs molecular weight standard curve. The standards and their respective molecular weights are as marked (gray diamonds) and labeled in grey. The Wild type and mutant ADAT2/3 enzymes are marked (black circles) and labeled in black.

3.2.4 The C-terminus of TbADAT2 is crucial for enzyme activity

The binding studies established that the computer-predicted motif (KR-domain) indeed plays a crucial role in tRNA binding. However, as explained earlier, the binding studies were performed with a G34-containing tRNA. Thus for the binding data to be meaningful, the effect of the KR-domain mutations on enzymatic activity were tested using the A34-containing tRNA substrate. Recombinant mutant protein was incubated with saturating levels of radio-labeled substrate. Plotting the initial velocity against substrate concentration yielded kinetic parameters that were compared to established parameters for the wild type recombinant TbADAT2/3 (Vmax=0.25 ± 0.06 pmol/min,

-1 Km=0.75 ± 0.11 M and kcat =0.19 ± 0.07 min ) (Fig. 2.2 and Table 3.2). The ADAT2 C- ter5 which retained the 5 charged arginines and lysines had a Km of 0.76 ± 0.21 M

-1 and a kcat of 0.12 ± 0.05 min (Fig 3.11 and Table 3.2), these values are very similar to the wild type enzyme thus it was concluded that these residues are not major contributors to enzyme activity. On the contrary, mutant ADAT2 C-ter10 and ADAT2 C- ter5A as previously mentioned had no detectable enzyme activity (Table 3.2), suggesting that the tRNA binding mediated by the KR-domain may indeed be essential for enzyme activity.

The steady-state kinetic parameters for the single alanine substitution mutants were also determined (Table 3.2). Replacement of the key catalytic glutamate in the

ADAT2 by alanine as expected led to a complete loss of activity. Further analysis of the individual alanine substitutions in the KR-domain showed that all the mutations led to increases in Km with negligible changes in kcat, reinforcing the proposal that the

Table 3.2: Kinetic parameter for wild type and C-terminal mutants of TbADAT2

Enzyme Km kcat kcat/ Km M min-1 min-1/ M wild type 0.75 ± 0.11 0.19 ± 0.07 0.25 ADAT2 C-ter10 ------ADAT2 C-ter5 0.76 ± 0.21 0.14 ± 0.05 0.18 ADAT2 C-ter5A ------ADAT2 E92A ------ADAT2 K216A 2.99 ± 1.50 0.10 ± 0.03 0.03 ADAT2 R217A 2.54 ± 0.79 0.09 ± -0.01 0.04 ADAT2 K218A 2.01 ± 0.15 0.12 ± 0.03 0.06 ADAT2 R219A 2.78 ± 0.28 0.13 ± 0.06 0.05

Specific activities were calculated as described in the ―Materials and Methods‖ (see Appendix A). Kinetic constants were determined according to the Michaelis-Menten equation. Kinetic data were fitted using the SigmaPlot software.

A B [tRNA] [tRNA]

+ 0.4

0.3 /min)

pI 0.2

pmoles (

o 0.1 KM = 0.76 ± 0.21 M pA v -1 kcat= 0.14 ± 0.05 min 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 ADAT2 C-ter5 ADAT2 C-ter10 [tRNA] (M) Figure 3.11: Steady-state kinetic analysis of TbADAT2/3 KR-domain mutants.

A) A representative one-dimensional thin-layer chromatography (TLC) analysis of the reaction products. pA and pI denote the migration of unlabeled 5'-AMP (pA) and 5'-IMP (pI) used as TLC markers and visualized by UV shadowing (not shown). The fraction of pA converted into pI during each reaction was calculated by dividing the amount of radioactive pI produced by the total (pA+pI), this value was then used to calculate the picomoles of 5'-IMP produced. A no enzyme control was routinely used for background subtraction. (+) refers to a reaction in which the radioactive tRNA substrate was incubated with wild type TbADAT2/3 as a positive control. B) The initial velocity (Vo) is plotted as a function of substrate concentration given in M. The data was fitted by non-linear regression to the Michaelis-Menten equation using the SigmaPlot kinetic software.

enzymatic defects observed correlate with binding defects and establishing the KR- domain as a bona fide binding determinant for this enzyme.

3.2.5 Pseudo-active site also contributes to substrate binding

While the KR-domain is critical for binding, it is likely not the only factor necessary for binding because the TbADAT2 homodimer which should behave similarly to ADATa and have two KR-domains still has severe binding defects (Fig. 2.9).

Additionally ADATa and ScADAT2 together could not reconstitute deaminase activity

(Wolf, Gerber, Keller 2002); it is only in the presence of ADAT3 that ADAT2 supports deaminase activity suggesting that ADAT3 or the ADAT2/3 complex further contributes to activity by the addition of a second binding domain. In ADAT2/3 there is only one true active site because ADAT3 is missing the catalytic glutamate. The pseudo-active site is created via inter-subunit zinc-coordination. As illustrated in figure 3.1, the pseudo-active site is comprises of ADAT2 C136, C139 and ADAT3 H252. While the pseudo active site is important for activity, it alone cannot confer activity nor can it replace the catalytic active site (Table 3.3). When amino acid substitutions were made to the catalytic glutamate (ADAT2 E92A) rendering the enzyme inactive, substitutions of ADAT3 (V254E or P235A+V254E) could not rescue the activity. Also noteworthy is the fact that the presence of two active sites—ADAT3 HPV to HAE in the context of a wild type ADAT2— also inactivates the enzyme (Table 3.3); however, other amino acid substitutions of V254

(V254L, V254T, and V254S) were tolerated (Table 3.3). These data together with the data presented in chapter 2 suggest that although the pseudo-active site is not directly involved in catalysis, it is important to the overall deaminase activity of the enzyme.

Table 3.3: The pseudo-active cannot rescue mutations of the active site.

a b b,c Enzyme Km kcat kcat/Km M min-1 min-1/M

Wild type 0.75 ± 0.11 0.19 ± 0.07 0.25

ADAT2 E92A ------

ADAT2 E92A/ ADAT3 V254E ------

ADAT2 E92A/ ADAT3 P253A V254E ------

ADAT3 P253A V254E ------

ADAT3 V254E ------

ADAT3 V254L 2.77 ± 0.58 0.05 ± 0.03 0.02

ADAT3 V254T 2.14 ± 0.79 0.11 ± 0.01 0.05

a Each enzyme is in the context of heterodimer with a wild type partner subunit unless noted. b Values are obtained from at least three independent trials. “---” denotes no activity detected c kcat values are calculated using experimental Vmax values, protein concentrations via Bradford assays, and fraction of active protein.

Kinetic parameters of wild type and mutant TbADAT2/3. In all cases, mutant subunits were co- expressed in the context of a heterodimer with a wild type partner unless otherwise noted (ie. ADAT2 E92A/ADAT3 V254A represents a mutant heterodimer in which ADAT2 and ADAT3 were mutated).

Because the pseudo-active site does not contribute directly to the chemistry of the reaction but is clearly important in overall enzyme function, it could be an additional binding domain. This idea is strengthened by the fact that the pseudo-active site can coordinate zinc and the precedence for zinc metalloenzymes to possess one catalytic and one structural zinc metal ion. To probe the role of the pseudo-active site in binding a series of single and double amino acid substitutions were made to the conserved histidine, proline, and cysteines of both the active and pseudo-active site. The focus on histidine and cysteines is of course because of their role in zinc coordination. Proline was also examined because proline often provides a necessary kink turn in the peptide which can be important for correct protein structure and substrate positioning (Vanhoof et. al. 1995).

Electrophoresis mobility shift assays (EMSAs) support the idea that the pseudo- active site but not the active site contributes to overall substrate binding (Fig. 3.12 and

3.13). Substitution of amino acids important for zinc coordination in the active site (ie

ADAT2 H90 or ADAT3 C291) had little to no effect on binding as seen in Kd values that are within two-fold of the wild type value (compare 0.35 ± 0.11 M to wild type 0.15 ±

0.03 M) (Fig. 3.12). However, substitution of amino acids responsible for zinc coordination of the pseudo active site (ie ADAT3 H252 or ADAT2 C139A) had a three to eight fold increase in Kd implicating this domain in substrate binding (Fig. 3.11). While an increase Kd was not seen as expected for ADAT3 H252, a closer examination of the

ADAT2/3 structural models points to another histidine in ADAT3 (H300) that is could allow for minimal zinc coordination such that the binding site is maintained. As expected the pseudo-active site proline but not the active site proline affects binding (Fig 3.13).

A [Protein] M [Protein] M [Protein] M - - -

complex

free probe

Wild type ADAT3 C291A ADAT2 C139A

Enzyme Kd Fold B M Change

Wild type 0.15 ± 0.03 -- ADAT2 H90A 0.22 ± 0.06 1.5 ADAT3 C291A 0.15 ± 0.03 -- ADAT3 C294A 0.35 ± 0.11 2.3 ADAT3 H252A 0.12 ± 0.02 -- ADAT2 C136A 0.45 ± 0.07 3.0 ADAT2 C139A 1.12 ± 0.40 7.5

Figure 3.12: The pseudo-active site contributes to substrate binding.

Increasing concentrations of wild type or mutant TbADAT2/3 were incubated with tRNAVal. tRNA- protein complexes were separated from free tRNA on a non-denaturing polyacrylamide gel. Shown is a representative gel using the wild type (A, left panel), a active site mutant (ADAT3 C291A; A, middle panel) and a pseudo-active site mutant (ADAT2 C139A; A, right panel). ―- "denotes a negative control lane with RNA alone; this lane is a marker for the migration of free tRNA in the gel. Arrows denote where free RNA and protein-RNA complexes migrate in the gel. (B) Apparent Kd values (calculated from EMSA, see materials and methods) for a series of zinc coordination active and pseudo-active site mutant enzymes. ―Fold change‖ is the relative increase in Kd as compared to the wild type; ―--― denotes same as wild type (no change). Each value represents the average of at least three independent trials. In all panels the mutant subunit is expressed in the context of a heterodimer with the wild type partner.

A [Protein] M [Protein] M [Protein] M - - -

complex

free probe

Wild type ADAT3 P290A ADAT2 P135A

Enzyme Kd Fold M Change

Wild type 0.15 ± 0.03 -- ADAT3 P290A 0.13 ± 0.01 --

ADAT2 P135A 2.37 ± 0.62 15.8

Figure 3.13: The pseudo-active site proline contributes to substrate binding.

EMSAs were performed as in 3.12. Shown is a representative gel using the wild type (A, left panel), a active site mutant (ADAT3 P290; A, middle panel) and a pseudo-active site mutant (ADAT2 P135A; A, right panel). ―-"denotes a negative control lane with RNA alone; this lane is a marker for the migration of free tRNA in the gel. Arrows denote where free RNA and protein-RNA complexes migrate in the gel. (B) Apparent Kd values (calculated from EMSA, see materials and methods) for a series of active and pseudo-active site mutant enzymes. ―Fold change‖ is the relative increase in Kd as compared to the wild type; ―--― denotes same as wild type (no change). Each value represents the average of at least three independent trials. In all panels the mutant subunit is expressed in the context of a heterodimer with the wild type partner.

The proline adjacent to the active site has no direct bearing on binding (Kd = 0.13 ± 0.01) but when the proline adjacent to the pseudo-active site increases the Kd 16-fold (Fig.

3.12). In all cases, concomitant with the decreased binding affinity is loss of or decrease in deaminase activity.

3.2.6 Pseduo-active site and KR-domain act cooperatively to bind single tRNA

Interested in the binding interplay between the KR domain and the pseudo-active site, Hill plots were used to examine potential cooperativity between the two sites. Hill plots are typically used to investigate how the binding of one ligand to a binding site affects the binding of the second, third, etc. ligand to a different binding site. We use Hill plots to explain cooperativity as it applies to the binding of different parts of a single tRNA to the two different binding sites, the KR domain and pseudo active site. In this type of analysis plot slopes greater than one indicates positive cooperativity while slopes equal to or less than one indicate either negative or no cooperation. In this case, cooperativity suggests that binding of the tRNA in one binding domain affects the further positioning and binding in the second domain.

Analysis of the wild type ADAT2/3 enzyme gives a Hill plot slope of 1.6 indicating a positive cooperative interaction (Fig. 3.14 and Table 3.4). KR domain deletions and mutations not only increased the Kd for the tRNA but also affected the cooperativity; the

Hill plot slope of all of the different KR domain mutants decreased below one indicating a loss in positive cooperativity (Fig. 3.14 and Table 3.4). Likewise mutations to the pseudo-active site also showed an increase in Kd concomitant with a loss of cooperativity. As a control Hill plot analyses were also performed for mutations of other amino acids likely not involved in binding such as the active site residues.

[Protein] M [Protein] M A - -

Protein-RNA complex

Free tRNAVal

Wild type KR domain mutant

Kd = 0.15 0.03 M Kd = 1.20 0.33 M

1.0 B

0.5 Enzyme Slope

))

y -

Wild Type 1.63 max

y 0.0 ADAT3 C291A 1.29 /( ADAT2 C-ter10 0.71 ADAT2 C-ter5A 0.90 Log (y Log -0.5 ADAT2 C136A 0.90

-1.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 Log [Protein] Figure 3.14: The pseudo-active site and KR-domain act cooperatively to bind a single tRNA.

A) Increasing concentrations of wild type or mutantTbADAT2/3 were incubated with tRNAVal . tRNA-protein complexes were separated from free tRNA on a non-denaturing polyacrylamide gel. ―-"denotes a negative control lane with RNA alone; this lane is a marker for the migration of free tRNA in the gel. Arrows denote where free RNA and protein-RNA complexes migrate in the gel. B) A Hill plot derived from the data collected during EMSA binding experiments. On the y-axis, y represents the fraction bound. Plots with a slope >1 show cooperativity while <1 has no or negative cooperativity. Although there is only one representative plot for each of four groups (wild type, active site mutants, pseudo-active site mutants, and KR-domain mutants), other mutants in the group follow the same general trend (See Table 3.4).

Table 3.4: The Pseudo-active site and KR domain act cooperatively to bind single tRNA.

Enzyme Kd M Hill Number Wild type 0.15 0.03 1.63 ADAT3 P290A 0.13 0.01 1.30 ADAT3 C291A 0.15 0.02 1.10 ADAT3 C294A 0.35 0.11 1.29 ADAT2 P135A 2.37 0.62 0.87 ADAT2 C136A 0.45 0.07 0.90 ADAT2 C139A 1.12 0.40 0.87 ADAT2 C136A / AD3 H252A 0.61 0.13 0.89 ADAT2 K218A 1.09 0.24 0.90 ADAT2 C-ter5A 3.28 0.63 0.90

Apparent dissociation constants (Kd) and Hill plot slopes (Hill number) for wild type and mutant TbADAT2/3. In all cases, mutant subunits were co-expressed in the context of a heterodimer with a wild type partner unless otherwise noted (ie. ADAT2 E136A/ADAT3 H252A represents a mutant heterodimer in which residues of ADAT2 and ADAT3 were changed).

Amino acid substitutions in the active site did not affect positive cooperativity as all slope values were greater than 1. These data together with the decrease or loss of activity seen in KR-domain and pseudo-active mutants suggest that binding and proper orientation of the tRNA for activity requires both binding sites to act cooperatively.

3.3 Discussion

Mechanistically, whether acting on a riboflavin precursor, an amino acid or nucleotides (free or within RNA and DNA polymers), all deaminases contain active sites with a tightly bound Zn2+ cofactor, which generates a nucleophile by activation of water.

What then distinguishes the different members of this deaminase superfamily is their unique substrate specificity. Free nucleotide deaminases use the 5' and 3' hydroxyls flanking the sugar as key recognition motifs to the exclusion of larger polymers

(Navaratnam et. al. 1998). On the other hand, RNA and DNA deaminases have more open active sites that allow accommodation of these larger substrates. In addition, the presence of a ―flap‖, which closes the active site, plays critical roles in what substrates can be accommodated (Xie et. al. 2004). In the case of free nucleotide deaminases, this flap is short and less flexible than that of polynucleotide deaminases as shown by

Wedekind and co-workers in a comparative analysis of CDD1 (a free nucleotide deaminase) from yeast and APOBEC-1 (Xie et. al. 2004).

The recently solved structure of an active deaminase domain of APOBEC3G (a

C to U DNA deaminase) also showed that in addition (besides the active-site flap) a groove spanning the active site of the enzyme is formed by a set of both charged and aromatic amino acids (Furukawa et. al. 2009; Holden et. al. 2008; Shandilya et. al.

2010). Groove formation for the purpose of substrate orientation is as critical for activity as direct contact of the ssDNA substrate by specific amino acids. Despite these

divergences in substrate recognition, ssDNA-specific deaminases like AID or

APOBEC3G still share a conserved amino acid core with tRNA deaminase (ADATa and

ADAT2/3), including the presence of the two core signature motifs found in all members of the deaminase superfamily: an active site glutamate (serving as a general base for catalysis) and a histidine/cysteine/cysteine triad involved in Zn2+ coordination. This evolutionary conservation has led to the proposal that all deaminases are derived from an ancestral nucleotide deaminase and through evolution a core deaminase domain has been appended to different specificity domains that will then provide binding to different substrates (i.e. riboflavin precursor vs. RNA or DNA).

In the specific case of tRNA deaminases, the process of evolution has taken the use of inosine at the wobble position in tRNAs into separate paths for bacteria and eukarya. In the eukaryal system, seven to eight different tRNAs undergo A to I editing depending on the organism (Gerber and Keller 1999). In bacteria, only tRNAArg undergoes A to I editing, but this single editing event is still essential for translation.

Interestingly, the presence of G34-containing tRNAs in bacteria for the same codons

(which are missing in eukarya) makes possible the limited use of inosine in the bacterial

Arg system, as G34 can decode C and U ending codons. In turn, tRNA is the only tRNA in bacteria for which a G34-containing counterpart is not encoded in the genome (Wolf,

Gerber, Keller 2002). This observation has raised the question of how the eukaryal enzyme manages to accommodate the different substrates while maintaining target site specificity.

Huang and co-workers have proposed that for recognition of seven to eight different substrates, a tRNA binding module has been appended to the eukaryal deaminases away from their active site, which in turn led to accumulation of mutations

resulting in active site relaxation and concomitant multi-substrate specificity (Elias and

Huang 2005). They asserted that likely regions in ADAT2 and ADAT3, that show little sequence conservation with the bacterial RNA deaminases, might in fact harbor the appended RNA binding domain. To test this model and elucidate some of the basis for substrate recognition, the T. brucei tRNA deaminase (TbADAT2/3) and bacterial tRNA deaminases were compared. Given the remarkable catalytic flexibility of the TbADAT2/3

(able to perform both A to I and C to U editing in vitro) (Rubio et. al. 2007), our laboratory has concentrated on utilizing this enzyme to explore its mode of tRNA binding. Work presented here demonstrated that as predicted by Huang and co-workers, the C- terminal end of ADAT2 contains an essential domain for tRNA binding (Ragone and

Spears et. al. in press in RNA) as shown by EMSAs.

Interestingly, in these studies, the recombinant enzyme does not readily bind an

A34-containing tRNA by EMSA; admittedly this observation may appear counter to the behavior of the enzyme, which is able to efficiently catalyze inosine formation in this substrate (its natural substrate). It is possible that this reflects the fact that once catalysis occurs, the enzyme turns over the product making it difficult to catch the enzyme product complex if the assay is performed at optimal temperature (27 oC). However, little binding was observed during EMSA (performed at 4 oC); thus, it may be possible that substrate binding requires some type of conformational change in the enzyme, which may be temperature dependent. Currently, however, we have no evidence for this premise and it will thus remain an open question. Along the same lines, EMSA with the wild type enzyme and the G34 and I34-containing substrates showed the presence of two well- defined complexes at higher protein concentrations (arrows in Fig. 3.2 A). This may reflect either different multimeric forms of the enzyme upon substrate binding or as

mentioned above different conformations of the protein and/or the substrate. In either case, determination of the Kd for either of the shifted bands yielded identical values.

Kinetic measurements yielded an apparent Kd of 50 nM. Although this is in the experimental range of the Kd measured for the G34–containing tRNA (150 nM), it is still 3- fold better which could partly explains how the enzyme avoids competitive inhibition by binding naturally occurring G34-containing iso-acceptors in vivo. It is also possible that in vivo, posttranscriptional modifications may serve as negative determinants for non- productive binding to other tRNAs that at the primary sequence level may resemble its natural substrate. Regardless the KR-domain is critical for tRNA binding.

The KR-domain is formed by a string of positively charged amino acids (arginines and lysines) termed the KR-domain. Due to its overall charge, one could envision possible interactions between the KR-domain and the phosphate backbone of the tRNA as suggested by Grosjean and co-workers for the yeast enzyme (Auxilien et. al. 1996). A similar domain rich in positively charged amino acids is present in ADAR3, an orphan deaminase of unknown function (a homolog of the mRNA deaminases ADAR1 and 2) .

It was recently shown that mutations of the analogous R-domain of ADAR3 specifically impaired ssRNA binding (Chen et al. 2000). These authors proposed that this domain might play a crucial role in recognizing loops often found near the targeted adenosine in mRNA substrates. The KR-domain of ADAT2/3 could serve an analogous binding function but unlike ADAR3 the targeted adenosine is distal to this binding domain.

Notably, binding of TbADAT2 alone to either the A34 or G34-containing tRNA is poor (>3

M) despite still harboring a KR-domain, indicating that the KR-domain is necessary but not sufficient to provide functional binding. This suggests that another significant binding site exists perhaps within ADAT3 or within ADAT2 and ADAT3 interactions.

Indeed, we present evidence supporting a second domain that seems to be critical for binding and perhaps positioning the tRNA in the active site. Amino acid substitutions of residues in this pseudo-active site decrease A to I activity with the largest effect seen on Km suggesting a binding defect. Additionally, EMSA data for these same mutants reveal an average 9-fold increase in Kd (Fig. 3.12 and 3.13). Further analysis of this binding defect in conjunction with earlier studies of the KR-domain shows that these two sites act cooperatively to bind a single tRNA. If residues in either the KR- domain or the pseudo-active site are changed, positive cooperativity is lost and the Kd increases by 4- to 20-fold.

The important role in substrate binding helps explain why a) the KR-domain alone is not enough to confer substrate binding (ie the ADAT2 homodimer shows a severe binding defect) and b) the addition of another complete active site does not enhance activity. Cooperativity between the KR domain and the pseudo-active site can also help explain the earlier observation that only one tRNA binds per heterodimer. The bacterial ADATa homodimer binds two tRNAs per catalytic cycle (shown by the co- crystal structure) or one per active site. In the case of TbADAT2/3 the active and pseudo-active site could not both bind a single tRNA if the pseudo-active site is needed in concert with the KR-domain to bind and correctly position the tRNA in the single active site. The fact that the full length tRNA is the only in vitro substrate also suggests that there are multiple points beyond just the anticodon stem loop that are needed for substrate recognition and that are contacted by the protein. The model proposed is that the tRNA tertiary structure is recognized and bound, perhaps by the KR-domain and it is this initial binding and the binding contributions of the pseudo-active site that allows

for the subsequent binding and positioning of the tRNA into the active site (Fig. 3.15).

ADAT2 ADAT3 ADAT2 ADAT3 ADAT2 ADAT3

Zn Zn Zn

I34

Initial binding of Further substrate Catalysis tRNA to ADAT2/3 accommodation

Figure 3.15: ADAT2/3-tRNA binding model.

The proposed binding mechanism has 2 steps. First, TbADAT2/3 heterodimer binds the tRNA via the KR-domain, a general RNA binding domain. Then with the help of contacts at or near the pseudo-active site the tRNA anticodon loop is correctly positioned into the catalytic active site. Catalysis is dependent on correct positioning of the tRNA.

Chapter 4:

Partial Purification of the Archaeal m1A to m1I tRNA Deaminase Enzyme

4.1 Introduction

Every field has those senior scientists who seem to know everything and for whom junior scientists would do anything to please. In the world of tRNA modifications,

Henri Grosjean is one such mentor. The search for the elusive m1A to m1I enzyme(s) from archaea started as a favor to Henri to find a potentially unique deaminase or deaminase-like protein(s) and has continued as a secondary project in addition to the other works presented earlier. While this project is still a work in progress, significant headway has been made in working out a purification scheme for the m1A to m1I enzyme(s).

1-Methylinosine (m1I) in the TΨC loop is so far only found in tRNAs from archaea

(Grosjean et. al. 1996; Juhling et. al. 2009). This modification occurring at position 57 has been observed across a wide variety of archaea in isoacceptors of every tRNA except histidine and glutamine. In organisms where this editing/modification activity is

1 observed, nearly every tRNA with A57 is modified to m I. tRNAs from Haloferax volcanii,

1 His for example, are always modified to m I with the exception of tRNA which has A57.

1 Similarly, in Halobacterium salinarum all A57-containing tRNAs are also modified to m I

(Juhling et. al. 2009).

These observations lead to interesting questions about the essential nature of the enzyme(s) and how the enzyme(s) responsible for this activity discriminate between substrate and non-substrate tRNAs. Perhaps more interesting is how the enzyme recognizes m1A and if the requirement for m1A substrate means that this enzyme(s) is not like other tRNA deaminases in the metal ion it uses to catalyze the reaction.

m1I is also found in eukaryotes; however, it is located 3’-adjacent to the anticodon (position 37). Also how it is formed differs in eukaryotes as compared to archaea. In eukaryotes, A37 is first deaminated to I37 via hydrolytic deamination catalyzed by ADAT1. This reaction is then followed by a SAM-dependent methyltransferase

1 reaction to create m I37. In archaea, the order of modification is reversed. Grosjean and

1 co-workers showed that in Haloferax volcani A57 is first methylated by TrmI (m A57) and

1 then is further deaminated by an unknown enzyme to create m I57. The importance of

1 this modification is still not clear. However, in the adjacent site m A58 in yeast is critical

Met for tRNA i stability (Anderson et. al. 1998; Anderson, Phan, Hinnebusch 2000). In other

1 hyperthermophilic archaea m A58 is needed for growth at high temperatures (Droogmans

1 et. al. 2003). m A58 also forms a Hoogsteen base pair with U54 in most tRNAs, stabilizing

1 the tertiary structure (Guelorget et. al. 2010). Perhaps the presence of m I57 in archaea

1 1 is also needed for further stability ensuring that m A58 and not m A57 base pairs with U54.

4.2 Results

4.2.1 Thermococcus kodakaraensis can deaminate m1A to m1I

The enzyme responsible for the conversion of m1A to m1I is unknown and in an attempt to identify the enzyme(s), Thermococcus kodakaraensis (Tko) was chosen as the representative archaea to study because 1) the genome has been sequenced and 2) much work from Dr. Tom Santangelo (Reeve laboratory, OSU Dept. of Microbiology) and

Imanaka’s laboratory has established invaluable genetic tools and expression plasmids in this system making gene knockouts and protein over-expression possible (Endoh et. al. 2006; Santangelo, Cubonova, Reeve 2008; Sato et. al. 2005). This organism is a strict anaerobe that grows optimally at 85 oC (Atomi et. al. 2004). With the help of Dr.

Tom Santangelo (Reeve lab) T. kodakaraensis cell pellets from cultures grown in sodium pyruvate based media were obtained.

As a proof-of-principle experiment in vitro transcribed tRNA, TrmI

(methyltransferase) and SAM (S-adenosyl methionine), were incubated in the presence of a crude Tko cell extract for one hour at 50oC (Fig. 4.1). Thin layer chromatography

(TLC) shows the presence of m1I and further validates the order of the pathway in that no I, m1I or m1A were seen in the absence of SAM. Furthermore, there is no m1I in the absence of cell extract (Fig 4.2). Thus, Tko is an appropriate organism to use for searching for the deaminase-like protein.

4.2.2 Bioinformatic searches to identify candidate deaminases

Because the entire Tko genome was sequenced (Fukui et. al. 2005), the first approach taken to find potential candidate deaminase enzymes was a computational.

Searching the Tko genome using BLAST (NCBI) with the ADATa (from bacteria), ADAT2 and ADAT3 (from eukaryotes) or ADAT1 sequences as queries yielded a few potential candidates though the e-scores were quite high (typically no less than 0.01). The best scoring proteins and those appearing most often throughout the searches were annotated as a riboflavin biosynthesis protein and deoxycytidylate deaminase. Both proteins have the signature HAE and PCxxC domains, because of their role in riboflavin biosynthesis

CCA* CCA* CCA* * * * T.ko protein TrmI fraction * * A * *m1A * * m1I ** * * 57 ** * * 57 ** * * 57 ** ** ** * ** SAM * ** * **

Nuclease P1

m1A *pA 1 *pA *pm A57 *pA 1 *pA m I *pA 1 *pA *pm I57 *pA *pA *pA *pA A *pA *pA

Figure 4.1 m1A to m1I activity assay schematic

Asp 32 tRNA GUC was in vitro transcribed in the presence of α P-ATP. This tRNA was gel purified and 1 position 57 was site specifically methylated (m A57) by incubation with TrmI, a SAM dependent methyltransferase. This methylated tRNA was phenol extracted, ethanol precipitated and used as the substrate for all m1A to m1I assays. After incubation with Tko extracts or protein fractions, the tRNA was again phenol extracted, ethanol precipitated and digested to single nucleotide monophosphates using nuclease P1. Digested products were dried, resuspended in ddH2O and nucleotides were separated via one-dimenisonal thin layer chromatography (1D-TLC). Arrows mark A, m1A, and m1I as shown by markers in the TLC schematic.

-SAM +SAM

[Extract] -

m1A

m1I

1 2 3 4 5 6

Figure 4.2 Thermococcus kodakaraensis has m1A to m1I activity

In vitro transcribed radio-labelled tRNAAsp was incubated with Trm I (methyltransferase) and Thermococcus kodakaraensis cell extract in the presence and absence of SAM at 50 oC for one hour. The tRNA was digested to single nucleotides and separated using TLC. Shown are the resulting 1D-TLCs. Lanes 1 and 4 are control reactions in which there is no Tko extract. Lanes 2- 3 and 5-6 have increasing amounts of Tko extract. The presence of m1IA and m1I is SAM dependent and the presence of m1I is m1A dependent. Arrows mark A, m1A, and m1I as shown by nucleotide markers.

only the deoxycytidylate deaminase was cloned and over expressed for purification in E. coli. The deoxycytidylate deaminase protein was successfully expressed and purified but had no m1A to m1I activity in vitro. The de Crecy-Lagard lab also identified another candidate enzyme which was also expressed and purified from E. coli in the Alfonzo lab but also had no in vitro activity. Finally, a recently published cytidine (C) to uridine (U) deaminase (CDAT8) from Methanopyrus kandleri was used as a BLAST query however, it yielded the same riboflavin biosynthesis protein and deoxycytidylate deaminase which was not active in vitro.

Of course, these in vitro studies do not take into account the fact that the potential deaminase-like protein could have a binding partner such as that seen with the

ADAT2/3 heterodimer from eukaryotes. Another point of contention is the differences in growth environment in Tko versus E. coli. Tko requires less common metals for growth, which suggests that perhaps metalloproteins such as ADATa and ADAT2/3 that commonly use zinc in their active site in other organisms might have evolved to utilize different metals for activity in Tko. Tko cells, when grown on pyruvate, absolutely need a source of strontium (SrCl2 ·6H2O) and iron (FeNH4) and removing either from the growth media even though both are added at concentrations less than 100 M (75 M and 26

M, respectively) yields no growth (T. Santangelo, personal communication).

4.2.3 Native protein purification via column chromatography (pilot runs)

As bioinformatic searches continued, simultaneous attempts were made to purify the native enzyme using column chromatography. First, a series of pilot runs were tested to find appropriate columns to use in a final purification scheme. Tko cells grown on pyruvate were harvested in late log phase (OD600 ~0.8) and suspended in lysis buffer (50

mM Tris pH=7.5, 5 mM MgCl2, 4 mM DTT, and 2 mM protease inhibitor cocktail; 250 mM

NaCl was added for storage) at a ratio of 5 ml of buffer per gram of cell pellet. Cell extracts were made by one freeze/thaw cycle followed by sonication and a clarifying spin at 100,000 xg for 30 minutes. Using this extract, different chromatographic columns were tested to find those which yielded the best protein separation while retaining the most m1A to m1I activity. After each test run, the column input (IP), flow through (FT) and elution fractions were tested for activity using the assay described (Fig 4.1).

Three initial columns were tested: Q sepharose, DEAE (Diethylaminoethyl) cellulose and heparin column. In all cases cell extracts were loaded onto the column and proteins were eluted with a linear gradient from low salt (0 M KCl) to high salt (1 M KCl).

Active protein fractions consistently eluted from the Q sepharose column at about 300 mM KCl (Fig 4.3). DEAE similarly produced fractions with m1A to m1I activity at 350 mM

KCl (Fig. 4.3). The heparin column also produced fractions with activity; however, the binding of the protein of interest to this column is limited in that there is always some active protein in the flow-through and initial low salt washes. Nonetheless all three columns can potentially be used in final purification schemes. Additionally, active fractions from heparin were used as input for the Q sepharose column and vice versa. In both cases, m1A to m1I activity was observed in protein peaks from each respective column. DEAE active fractions were also used as input for Q the sepharose column and again activity was retained in the Q sepharose fractions.

Mono S (strong cation exchange column) and HIC (hydrophobic interaction chromatography) columns were also tested. The Mono S while producing a symmetrical peak eluting around 300 mM KCl, the protein(s) of interest

Q Sepharose DEAE Cellulose % Buffer B (1M

800 60 % Buffer B (1M ) 900 100 ) 800 90 700 55

80 mAU 700 mAU 600 50 600 70 60 500 45 500 50 400 40 400

40 300 35 KCl 300 30 KCl 200 200 30

20 )

)

Absorbance ( Absorbance Absorbance ( Absorbance 100 10 100 25 0 0 0 20 0 2 4 6 8 10 12 14 16 18 20 22 24 0 2 4 6 8 10 12 14 16 18 20 22 24 Fraction # Fraction #

m1A

m1I

IP FT 1 3 5 7 9 11 13 15 17 19 21 9 11 13 15 17 19

Fraction # Fraction #

Figure 4.3 Native protein purification column pilots (Q Seph and DEAE)

Tko cell extract was used as input to test the appropriateness of a series of columns to be used for native protein purification of the m1A to m1I enzyme(s). Shown in the top panels are chromatographs from Q sepharose (left) and DEAE cellulose (right). The black trace is the absorbance at 280 nm while the gray trace is % buffer B which contains 1 M KCl. In all cases the starting equilibration buffer (50 mM Tris pH 7.5) has no salt. Highlighted in pink are those fractions shown by TLC to be active. Shown in the bottom panels are representative TLCs. IP denotes input (what was injected into each column), FT denotes column flow through, and numbers denote fraction number as seen in the chromatographs. All chromatographic separations were performed at 4 oC.

100

did not bind to the column as seen by activity in the flow through (Fig. 4.4). A series of 1- ml HIC columns were also tested. The matrices screened include phenyl HP, octyl FF, butyl-S FF, phenyl FF (low substrate), phenyl FF (high substrate), butyl FF, and butyl HP

(where HP stands for high performance and FF stands for fast flow). Of these media, only the phenyl FF elution fractions had activity (Fig. 4.5). Because the active protein fractions coming from Q sepharose, heparin, or DEAE are in 300-500 mM KCl, HIC would make a good secondary column because salt can be easily added to active fractions for binding to the HIC column and the elution from this column effectively exchanges the higher salt buffer for a lower salt buffer.

At different stages during these pilot experiments, samples were applied to a gel filtration column (Superdex 200) to determine the size of the protein(s) involved in activity. Whether high or low salt buffers were used during gel filtration, no protein fraction retained activity. Preliminary data suggested that the activity eluted at a size of about 20 kDa but this could not be confirmed by enzymatic activity. Lack of sizing data suggests that perhaps the deaminase of interest is part of a protein complex whose subunits tend to dissociate during size exclusion.

101

200 100 % % Buffer B (1M ) 180 90 160 80 mAU 140 70 120 60 100 50 80 40

60 30 KCl

Absorbance ( Absorbance 40 20 ) 20 10 0 0 0 2 4 6 8 10 12 14 16 18 20 22 24

Fraction #

m1A

m1I

IP - FT 9 11 13 15

Fraction #

Figure 4.4 Native protein purification secondary column pilots (Mono S)

Tko protein fractions from above were used as input to the Mono S column as above (Fig. 4.3). Although a protein elution peak was observed around 450 mM KCl, none of the fractions had activity. “-“ denotes a negative control reaction in which no extract or protein fraction was added.

102

Q Sepharose active protein

peak

S FF S

Phenyl FF (low) FF Phenyl

Phenyl HP Phenyl OctylFF Butyl (high) FF Phenyl Butyl FF Butyl HP Butyl HIC HIC HIC + - Phenyl Butyl Octyl

m1A Wash 2M KCl m1I

Elution No KCl A

Frac 1 Frac 2

Figure 4.5 Native protein purification column pilots (HIC screen)

Tko protein fractions were used as input to test the appropriateness of a series of HIC columns to be used for native protein purification of the m1A to m1I enzyme(s). Shown in the left panel is the scheme used to quickly screen multiple 1 ml HIC columns. 1 ml of active Q sepharose protein fraction was added to each column. The columns were washed 1 ml of wash buffer (50 mM Tris pH 7.5 + 2 M KCl) (collected as flow through, FT). Low salt elution buffer was used to collect 2x 1 ml factions. In the TLC shown on the right each HIC column screened is listed across the top. For each column, the first lane is fraction 1 and the second lane is fraction 2.

103

4.2.4 Native protein purification scheme

Through use of pilot runs, a purification scheme was determined. Tko cell extract would be applied to the heparin affinity column followed by the anion exchange Q- sepharose column. HIC and mono Q (strong anion exchange) columns would follow.

After each column, specific activity and fold purification were calculated using the established activity assay (Fig. 4.1). Peak protein fractions were also separated using

10% polyacrylamide/ SDS PAGE to analyze sample complexity. The goal of the purification scheme was to achieve sufficient purification (less than 10 bands on a

Coomassie stained SDS-PAGE gel) for mass spectrometry analysis. The first step of the purification scheme can be seen in figure 4.6. The protein(s) eluted as a single peak around 300 mM KCl. The active fractions, which represent an 8-fold purification over the cell extract (Fig. 4.7 and Table 4.1) were pooled and dialyzed to remove the salt. The salt-free heparin pool was injected into the Q sepharose.

From the Q sepharose proteins actively eluted between 180-300 mM KCl (Fig.

4.8). The active protein fraction though active still contained many different proteins; several individual protein bands are observable on a Coomassie stained gel (Fig. 4.9).

This column yielded 2-fold purification step (Table 4.1). The active Q-sepharose fractions were pooled and KCl was added to 2 M final concentration so the sample could be applied to the Phenyl-sepharose FF column. During the scale-up this column did not behave as it did during the initial screens and the protein did not bind to the column as it had in the 1 ml column screen (Fig 4.10). The active fraction was instead in the column flow-through and not bound to the column as in the initial screen. Nonetheless this active flow through fraction still represents about a 3.3-fold purification for a total purification of about 58-fold over the initial cell extract (Fig. 4.11 and Table 4.1).

104

Heparin (Affinity column) % mAU 140 3000

120 2500

100 Col. vol: 1 ml 2000 Input: Tko extract 80 Buffer A: 50 mM Tris, pH 7.5 1500 Buffer B: 50 mM Tris, pH 7.5 + 2M KCl Flow: 0.5 ml/min 60 Frac vol: 1 ml 1000 40

500 20

0 0 1 2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Waste 0.0 10.0 20.0 30.0 40.0 50.0 ml

Figure 4.6 Heparin chromatograph

Tko cell extract was loaded onto a heparin column and eluted with a linear salt gradient from 0 to 2M KCl. The chromatograph from Unicorn (AKTA purifier) is shown. The information contained within the box gives the details of the elution including column volume (Col vol), buffer flow rate (Flow ) and the volume of each fraction collected (Frac vol). The dark blue trace is absorbance (UV), the green trace is % buffer B, and the light blue trace is % conductivity. The pink dotted line is the sample injection point. Shown across the bottom in black is total volume and in red are individual fractions. Highlighted in pink stripes are those fractions shown to have activity by 1D- TLC (see below). All chromatography was performed at 4 oC.

105

Heparin

(Affinity column)

Heparin

Heparin Pool Heparin Marker Extract Crude

75 kDa

m1A 50 kDa 37 kDa

m1I 25 kDa 20 kDa 15 kDa A

IP FT 2 10 12 14 Fraction

Figure 4.7 Heparin fraction analysis via TLC and SDS-PAGE

Peak protein fractions from the heparin column were tested for activity using the assay described. In all cases, reaction volumes were sufficiently large enough to diluted possible inhibitory effects of high salt concentration. In the TLC on the left IP denotes input (sample that was injected into the column), FT denotes column flow through, and numbers denote fraction number as seen in the chromatograph. Activity is seen in fractions 12 and 13. Protein fraction complexity was examined by separating proteins on a 10% Tricine-SDS PAGE gel. Shown on the right is the coomassie blue stained gel. Molecular weight standards (Marker) are as labeled.

106

Table 4.1 Tko native deaminase fold purification The total fold purification for the scheme highlighted above is about 1200-fold.

Column Input Specific activity Fold purification pmol/min/mg

0.26

540 mg Heparin (cell extract) 2.25 8.7

12 mg Q Sepharose (Heparin pool) 4.61 17.4

6.5 mg HIC Phenyl FF (Q Sepharose pool) 15.17 57.4

0.3 mg Mono Q (HIC Flow through) 320 1205

107

Q Sepharose (Anion exchange column) mAU % 700 140

600 120

500 100 Col. vol: 5 ml Input: Heparin pool (12 mg) 400 80 Buffer A: 50 mM Tris, pH 7.5 Buffer B: 50 mM Tris, pH 7.5 + 2M KCl 300 Flow: 5 ml/min 60 Frac vol: 5 ml

200 40

100 20

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 2627 28 29 30 31 32 33 34 35 36 37 39 40 41 42 43 44 45 46 47 48 50 51 52 53 54 55 Waste 0 0 50 100 150 200 250 300ml

Figure 4.8 Q Sepharose chromatograph

Active protein fractions from the heparin column were pooled and buffer was exchange to remove any salt before loading this sample onto the Q sepharose column and eluted with a linear salt gradient. The chromatograph from Unicorn (AKTA purifier) is shown and is as described above.

108

Q Sepharose

(Anion exchange column)

pool

Sepharose

Marker Heparin Pool Heparin Crude Extract Crude Q Q

75 kDa

m1A 50 kDa 37 kDa

m1I 25 kDa 20 kDa 15 kDa A

-- IP FT 41 42 43 44 45 46

Fraction

Figure 4.9 Q Sepharose fraction analysis via TLC and SDS-PAGE

Peak protein fractions from the Q sepharose column were tested for activity using the assay described. In all cases, reaction volumes were sufficiently large enough to diluted possible inhibitory effects of high salt concentration. In the TLC on the left “--" denotes a negative control reaction in which no protein was added. The TLC is otherwise labeled as above. Activity is seen in fractions 41 and 42. Protein fraction complexity was examined by separating proteins on a 10% polyacrylamide gel and comparing the banding pattern seen in peak fractions from the heparin and Q sepharose columns. Shown on the right is the coomassie blue stained gel. Molecular weight standards (Marker) are as labeled.

109

HIC-Phenyl FF (Hydrophobic interaction column) mAU %

100 140

120 80

100 Col. vol: 5 ml 60 Input: Q Sepharose pool + KCl 80 Buffer A: 50 mM Tris, pH 7.5 + 2 M KCl Buffer B: 50 mM Tris, pH 7.5 60 40 Flow: 5 ml/min Frac vol: 5 ml

40 20 20

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 46 47 48 49 50 Waste 0 0 50 100 150 200 250 ml

Figure 4.10 HIC-Phenyl FF chromatograph

Active protein fractions from the Q sepharose column were pooled and salt was added up to 2 M KCl before loading this sample onto the HIC-Phenyl FF column and eluted with a linear decreasing salt gradient. The chromatograph from Unicorn (AKTA purifier) is shown and is as described above. In this case the active fractions were seen in the flow through (highlighted in pink stripes).

110

HIC-Phenyl FF

(Hydrophobic interaction column)

34 50

- -

19 46

pool

Frac Frac

Seph

Marker Q FT HIC HIC wash final HIC HIC

75 kDa m1A 50 kDa m1I 37 kDa

25 kDa 20 kDa A 15 kDa

-- FT 27 28 29 30 46 47 W

Fraction

Figure 4.11 HIC-Phenyl FF fraction analysis via TLC and SDS-PAGE

Peak protein fractions from the HIC-Phenyl column were tested for activity using the assay described. In all cases, reaction volumes were sufficiently large enough to diluted possible inhibitory effects of high salt concentration. The TLC on the left is labeled as above with the addition of W which denotes final wash from the column. Activity was surprisingly seen in the pooled and concentrated flow through fraction. Protein fraction complexity was examined by separating protein pools on a 10% Tricine-SDS PAGE gel and comparing the banding pattern seen in peak fraction pools compared to Q sepharose fraction pool. Shown on the right is the coomassie blue stained gel where “HIC Frac” represents the pooling and concentrating of fractions 19-34 and 46-50. HIC final wash is the final no salt buffer wash of the column. Molecular weight standards (Marker) are as labeled.

111

The last step of this purification scheme was the Mono Q, a strong anion-exchange matrix. The HIC flow-through fractions were pooled and dialyzed to remove the high salt buffer and were then concentrated for further purification. This sample was injected into the Mono Q (Fig. 4.12). Using information gained from the Q-sepharose pilot run, the elution protocol for the mono Q column was optimized for greatest protein separation.

Here again, unfortunately the protein of interest did not bind to the mono Q as expected.

In this case it could be that the sample was not sufficiently dialyzed and thus carry over salt in the sample kept it from binding. Despite not binding to the matrix this purification step still resulted in protein sample with fairly low complexity as can be seen by just a handful of protein bands in a coomassie stained gel. (Fig. 4.13). This step was a 15-fold purification step yielding a total purification of 1200-fold (Table 4.1).

4.2.5 Mass spectrometry analysis

As mentioned earlier, the end goal of the protein purification scheme devised was to generate a sample with low enough complexity that mass spectrometry analysis would be possible. The final mono Q flow through sample yielded not only high specific activity but also showed fairly low complexity in a protein gel. This flow through sample was submitted to the Ohio State University Mass Spectrometry & Proteomics Facility for analysis. The proteins were digested with trypsin and subjected to LC-MS/MS. The peptides obtained from the mass spec were used to search archaeal genomes for protein matches. Figure 4.14 shows the partial list of the most commonly found proteins.

Upon first glance, nothing stands out as a potential deaminase as none of the proteins have the characteristic HAE and PCxxC domains.

112

Mono Q (Strong anion exchange column)

mAU %

140 140

120 120

100 100

Col. vol: 1 ml 80 Input: HIC FT 80 Buffer A: 50 mM Tris, pH 7.5 + 50 mM KCl 60 Buffer B: 50 mM Tris, pH 7.5 + 2 M KCl Flow: 1 ml/min 60 Frac vol: 250 ul 40 40

20 20 0 1 2 3 5 7 911 14 17 20 23 26 29 32 35 38 41 44 47 50 53 56 59 62 65 68 71 74 1 3 5 7 9 11 14 17 20 23 26 29 1 3 5 7 912 15 1821 24 27 30 33 36 39 43 46 4952 55 58 61 64 67 70 74 77 80 Waste 0.0 10.0 20.0 30.0 40.0 50.0 ml

Figure 4.12 Mono Q chromatograph

Active protein fractions from the HIC-Phenyl FF column flow through were pooled and buffer was exchanged to remove all salt before loading this sample onto mono Q column. Protein was eluted with a linear salt gradient. The chromatograph from Unicorn (AKTA purifier) is shown and is as described above. Again the active fractions were seen in the flow through (highlighted in pink stripes). Flow through fractions were pooled and concentrated for activity assays.

113

Mono Q

(Strong anion exchange column)

35 (C) 35 (D) 61

27 (B) 27

- -

31 57

35 27 61

- - -

FT (A) FT

31 23 57

frac frac frac

Mono Q Mono

Neg Cont. Neg

Mono Q Pool Q Mono Pool Q Mono

Mono Q Pool Q Mono

Mono Q Mix Q Mono

Mono Q Mono Marker FT Q Mono Q Mono Q Mono

m1A 75 KDa 50 KDa m1I 37 KDa 25 KDa 20 KDa A 15 KDa 10 KDa

Figure 4.13 Mono Q fraction analysis via TLC and SDS-PAGE

Peak protein fractions from the mono Q column were tested for activity using the assay described. In all cases, reaction volumes were sufficiently large enough to diluted possible inhibitory effects of high salt concentration. Individual fractions had no activity so peak protein fractions were pooled, concentrated, buffer exchanged, and tested for activity. In the TLC on the left each fraction pool is given a letter and a corresponding color; FT (A; red), fractions 23-27 (B; blue), 31-35 (C; green) and 57-61 (D; black). These letters (and colors) are the same as seen in the last three lanes “Mono Q Mix.” Only in the presence of the flow through was activity seen in the mixes. Additionally the pool of 23-27 had a small amount of activity. Protein fraction pool complexity was examined by separating proteins on a 10% Tricine-SDS PAGE gel. Shown on the right is the coomassie blue stained gel of the mono Q fraction pools. Molecular weight standards (Marker) are as labeled. The mono Q FT fraction was submitted for mass spectrometry analysis.

114

Using the sizes of the proteins seen in the coomassie stained gel of the mono Q FT fraction, as a guide, the initial list can be made shorter. Of interest on the shorter list are is an alanyl-tRNA synthetase-like protein and a miaB-like protein (protein that catalyzes the post-transcriptional methylthiolation of N-6-isopentenyladenosine in tRNAs) because of their closely related functions in tRNAs. However, much like the initial attempts to reconstitute the enzyme activity in vitro, the AlaRS-like proteins also not active.

4.3 Discussion

Although the m1A to m1I enzyme(s) still remains elusive, hopes are high that only missing piece at this point is the size of the activity. Working out conditions for obtaining the size via gel filtration will prove most valuable in solving this deaminase puzzle. The mass spec data did not reveal any standout candidates for a deaminase enzyme; however, the protein fraction submitted had robust activity and thus it is likely that one or more proteins on the list from the mass data is the m1A to m1I enzyme. Grosjean and co- workers were able to show that very crude cell extract fractions from H. volcanii had enhanced m1A to m1I activity when added together. Additionally any co-factor that could enhance the activity would have to be a macromolecule greater than 10 kDa (the molecular weight cut off of their dialysis (Grosjean et. al. 1995). There is of course precedence for deaminases needing a functional partner, take ADAT2/3 for example.

Preliminary unpublished work from our laboratory suggests that perhaps editing and modification enzymes form even larger editing complexes in vivo. In light of these new data, it would be interesting to try a pull-down type experiment with TrmI to see if the methyltransferase and the deaminase are a functional complex within the cell. Clearly,

TrmI is able to function on its own but perhaps the deaminase activity requires the presence of another protein as suggested by the data from Grosjean et. al. (1995).

115

Certainly there is still work to be done but the purification scheme presented here will prove successful in identifying the archaeal tRNA deaminase.

116

Figure 4.14 Mass spectrometry analysis of Mono Q FT sample

The flow-through fraction from the mono Q column was submitted for mass spectrometry analysis. The protein sample was digested with trypsin and those peptides were sequenced and used to search genome databases at NCBI. The most common proteins found from those peptides are listed here.

117

Chapter 5:

Concluding remarks

Although ADATs are essential enzymes in bacteria as well in eukaryotes, little was known about the eukaryotic ADAT2/3 before the completion of the studies presented here. A co-crystal structure of the ADATa and the tRNA anticodon stem loop provides clues as to the inner workings of ADAT2/3 but in lieu of having a crystal structure of ADAT2/3, questions about the enzyme must be answered using biochemical approaches and molecular modeling. The interest in TbADAT2/3 is not simply because it is an essential enzyme but because it is the only known deaminase that can perform both A to I and C to U editing in vivo. This unique property suggests that perhaps the

TbADAT2/3 active site is different from that of the bacterial ADATa, and potentially other eukaryotic ADAT2/3 active sites. If it holds true that TbADAT2/3 active site is sufficiently different from the human ADAT2/3, this could provide potential drug targets against T. brucei and other trypanosomes. However, it is also possible that TbADAT2/3 does not represent the exception but instead defines the rules for eukaryotic deaminases. Being the most complete study of a deaminase to date, our data suggests that TbADAT2/3 while being unique in its ability to catalyze A to I and C to U, the inter-subunit active site and cooperative binding sites are common to eukaryotic tRNA deaminase.

118

Using a combination of steady state kinetics, binding studies (via EMSA), and molecular modeling, our studies show that TbADAT2/3 has a unique active site architecture whereby the essential catalytic zinc is coordinated via inter-subunit contacts from ADAT2 and ADAT3. This is unlike the bacterial ADATa and other homodimeric deaminases where each subunit coordinates its own zinc ion. These studies also go on to show that the tRNA binding domain as originally proposed by Huang’s group has moved away from the active site to increase binding flexibility and accommodate more than one tRNA. The KR domain is critical for binding in TbADAT and likely represents a more generalized tRNA binding for all ADAT2/3s because a similar KR domain is found at the C-terminal end of nearly all ADAT2s. Furthermore, the pseudo-active site also contributes to binding and works cooperatively with the KR domain to bind and ensure proper orientation a single tRNA in the active site. Interestingly, preliminary footprinting data suggest that the tRNA is contacted by the enzyme in two major areas: the D-loop and the anticodon stem.

With these data in hand, a model has been suggested whereby ADAT2/3 recognizes the full length tRNA tertiary structure and binds the tRNA via the KR domain.

Further contacts by the pseudo-active site help guide and position the tRNA into the active site for catalysis. However, how the enzyme discriminates between the tertiary structures of substrate versus non-substrate tRNAs in vivo will remain an open question.

Future studies in the Alfonzo laboratory will aim to address this question. Regardless, the studies presented offer a much more complete picture of ADAT2/3 than what was previously known and the molecular models provided go a long way, absent any crystal structures, to help explain and interpret the data presented visually.

119

Appendix A:

Material and Methods

120

A. 1 Plasmid DNA Mini-Prep (From Alfonzo lab 2003 worksheet; Ref. The cloning Manual (Maniatis); Birnboim and Doly (1979); Ish-Horowicz and Burke (1981))

1. Transfer a single colony into 2 ml of 2XYT media (+antibiotic, if needed) in a culture tube.

2. Incubate culture overnight at 37oC w/ shaking Note: Grow for no more than 12-15 hours.

3. Harvest cells by centrifugation at max speed (13.2 x1000 rpm) for 5 minutes at 4oC.

5. Decant supernatant. Spin at desktop and pipet out remaining super. Note: Cell pellets can be frozen at -20oC for later processing.

6. Add 100 l Sol. I (ice cold) to all tubes. Resuspend the pellets by racking or vortexing. Leave the tube @ room temp. 2-3 min.

7. Add 200 l Sol II (room temp) to small group of tubes and invert gently and quickly ~5 times. Repeat with all tubes. Leave on ice 5-10 min.

8. Add 150 l Sol III (room temp) to small group of tubes and invert gently and quickly ~5 times. Repeat with all tubes. Leave on ice 5-10 min.

9. Spin down cellular proteins for 15 min (13.2 x1000 rpm) at 4oC.

10. Pipet 900 l EtOH into clean 1.5 ml eppendorf tubes.

11. Transfer all supernatant (~450 l) into EtOH containing tubes and mix well by inverting.

13. Pellet DNA by centrifugation at max speed (13.2 x1000 rpm) for 30 minutes.

14. Remove and discard supernatant and air dry pellet.

15. Add 50 l ddH2O (TE can also be used) and resuspend pellet by pipeting up and down.

Solution I: 25 mM Tris, pH=8, 10 mM EDTA, pH=8 Solution II: 0.2 N NaOH, 1% SDS Solution III: 3 M KOAc pH=4, 11% glacial acetic acid

121

A.2 Site directed mutagenesis (QuikChange PCR—Stragene)

Site directed mutagenesis was carried out via manufacturer’s protocol.

1. Oligonucleotide primer pairs were created using Stratagene’s guidelines. - Primers were 35-40 nucleotides in length (roughly 15 nt on each side of the mutation) o - Primers had at least 40% GC content with a Tm > 75 C - 3’ and 5’ end of primers were G or C

2. Mix sample reaction(s). -5 l 10x reaction buffer -1 l dsDNA plasmid template (100ng) -1 l 40uM Forward Oligo -1 l 40uM Reverse Oligo -2 l dNTPs (2.5mM each)

-39 l ddH2O Add just before putting the reaction into the thermo cycler. -1 l pfx (a pfuTurbo substitute) DNA polymerase

3. Thermocycling conditions (15-20 cycles): -Denature DNA 94 oC for 30 seconds -Anneal primers 45-55 oC for 30 seconds -Elongation 68 oC for 1 min per kb of DNA template length

-on last cycle end with temperature hold at 10 oC

4. Check efficiency of PCR by electrophoresis -load 10 l of PCR reaction onto a 0.8% agarose gel -run gel at 90 volts for 15 minutes -visualize product using (UVP; ie UV light)

5. Digest parental DNA with DpnI leaving only mutant DNA -1 l 10x NEB Buffer 4 -2 l PCR product

-6.5 l ddH2O -0.5 l DpnI enzyme (New England Biolabs) -Incubate at 37 oC overnight.

6. Transform mutated DNA into E. coli (DH5) for mutant screening. -Recovery of transformation is optional (I typically do no recover cells for this protocol). -Spread transformed E. coli on 2xYT agar plate with appropriate antibiotic.

7. Screen colonies for the correct transformant by sequencing or appropriate restriction digest reaction.

122

A.3 Transformation of E. coli (DH5a and BL-21) with plasmid DNA (constructs for protein over-expression)

1. Thaw frozen competent cells (stored at -80oC) on ice. -Need only 10 l of competent cells

2. Add 0.5 l of DNA to cells.

3. Incubate on ice 30-60 minutes. Gently flick bottom of tube to mix cells and DNA

4. Heat shock cells at 37 oC for 2 minutes (no more than 2 minutes)

5. Incubate cells on ice for 3 minutes.

6. (Optional) Add 250 l of plain 2xYT media to cells and allow cells to recover by placing tube at 37 oC for 30-60 minutes. Then pellet cells by spinning in bench top centrifuge for 2 minutes. Decant ~150 l of media.

7. Spread plate cells on 2xYT agar plates containing the appropriate antibiotic. (Note if you did not recover cells, add 75 l of plain 2xYT to cells prior to spread plating to aid in even spreading. If you recovered cells, the remaining 2xYT media will aid in spreading.)

If transforming DNA (correct, sequenced clone) into BL-21 cells, 3 phase streak for isolation instead of spread plating the cells.

123

A.4 Protein Over-expression in E. coli (BL-21 (RIL))

Day 1: 1. 3-phase streak cells containing protein construct on 2xYT plate with proper antibiotic. Streak from glycerol stock or from a single colony on another plate. -Streak must be less than 1 week old to use for the protein over- expression.

Day 2: 1. Start 20 ml overnight liquid culture from single colony and pre-warm 1.5L of 2xYT media (without antibiotic).

Day 3:

1. Inoculate 1.5L 2xYT (plus appropriate antibiotic) with 10ml of starter culture.

2. Incubate culture at 37 oC with shaking.

3. Check OD600 after 1 hour and note the OD.

4. Allow culture to grow with shaking at 37oC and continuing checking OD every hour until OD600=0.60.9 (Note: It usually takes about 2.5-3 hours to reach the proper OD. Generally after 1 hr OD600=0.10-0.15. After 2 hours OD600=0.40-0.50. After 2.5 hours OD600=0.6-0.7.

5. Quickly cool culture to room temp (by swirling flask over ice) and induce protein expression with IPTG. -Final IPTG concentration will vary depending on your protein and conditions being used but typically I use 0.5 mM final conc.

6. Induce culture at room temp (with shaking) overnight -The induction time and temperature will also vary depending on your expression conditions. Typically the culture is induced at 37 oC or 25 oC for 3-18 hrs.

124

A.5 Native Protein Purification

(Note: Day 1-3 are covered in the protein over-expression protocol)

Day 1:

1. Transform BL21 cells with plasmid containing protein of choice (or streak new plate from glycerol stock).

Day 2:

1. Start 20 ml overnight from fresh plate of BL21 cells and pre-warm media (without antibiotic) at 37oC.

Day 3:

1. Inoculate pre-warmed 1.5 L of 2xYT containing the proper antibiotic w/ 10ml of overnight culture.

2. If you have not yet done so, make a glycerol stock and store at -80oC

3. Induce culture with IPTG when OD600 reaches 0.6-0.9. -Induction time will vary from about 3 hours to overnight. -Induction temperature will be 37 oC or room temp. Day 4:

1. Harvest each 1.5 L prep in 2x 500ml bottles @ 6000 rpm for 5 minutes. You will have to fill and spin bottles multiple times to be able to pellet entire 1.5L prep. -Make 1x Binding Buffer from 8x buffer (in cold room) -Make 1x Wash Buffer from 8x buffer (in cold room)

Note: Be sure to check Buffer pH=8!! Acidic pH will inhibit binding of His tag to Ni2+ beads

**ENTIRE procedure must be done on ICE and gloves must be worn at all times**

2. Suspend each pellet in ~100 ml of 1x binding buffer + 1x P.I. and transfer to 4 x 50 ml conical tubes.

3. Freeze suspended pellets @-80 oC until frozen (can stop procedure here).

4. Thaw cells (Note: If completely frozen solid, this step will take about 2 hours at room temperature).

5. Sonicate sample in 50 ml conical tubes. -Use the microprobe to perform 5 x 30 second sonication.

125

-Do not exceed level 6 on Branson Sonicator if using microprobe.

6. Check sonication efficiency by microscopy -Add 10μl to a slide and cover with coverslip. -Make sure cells are completely lysed.

7. Pool sonicated cell extract into 1 x 250 ml bottle. Pellet cellular debris @ 10,000 rpm for 15 min

8. Pour super in 35 ml centrifuge bottles (you will need 5-6 bottles) and spin again @ 23,000 rpm (~67,000 xg) for 30-45 min.

Note: Balance is critical for high speed spins. Ultracentrifuge can also be used for this step.

9. Pour super from all 5 bottles into 4 x 50 ml clean conical tubes

10. Divide Ni2+ charged beads evenly into all 4 tubes. Let bind for 1 hour with rotation in the cold room.

Note: During this hour make a protein gel and set aside.

11. Balance 50 ml tubes and spin down beads @ 2000-3000 rpm (level 6) for 10 minutes in cold room centrifuge. -Use a 25 ml pipette to PIPETTE off super but DO NOT disturb the beads.

Note: DO NOT POUR off super!

Save the Flow Through in clearly labeled 50 ml tubes (-80 oC) -Set aside 50 l of liquid in a microcentrifuge tube labeled Flow Through (FT)

12. Add 30 ml of 1x Wash buffer to each tube and gently invert 10 times

13. Spin beads + wash @ 2000-3000 rpm for 10 minutes (in cold room) -Use 25 ml pipette to Pipette off super and discard -Save 50 l in a tube labeled Wash 1 (W1)

14. Combine beads into one tube and pour beads into column (Note: Do not pipette beads!). Gently swirl and pour into column.

15. Wash ALL residual beads from tubes with wash buffer and add to column. Continue to wash tubes with wash buffer until little to no beads remain in the tubes (~200 ml). Collect last drips of wash buffer flowing through column as Wash 2 (W2).

126

16. Place collection tubes under column and add 10 ml of Elution Buffer to the column. -Collect 5 x 2ml fractions

17. Clean, strip, and store Ni2+ beads. -Add ~5ml of elution buffer and collect in waste container. -Rinse used bead with 20 ml water. -Strip beads with 3-5 ml of 0.5 M EDTA; rinse with 20 ml of water -Add 20% EtOH to beads and allow a little bit to flow through column. Cap bottom of column and cover top. Store in fridge (at 4oC).

18. Add 20μl of each sample to 5μl of SDS Protein load + ME -FT, W1, W2, E1, E2, E3, E4, E5 -Load gel +Ladder and run at 50 mAmps for 2.25 hrs

19. Pool clean fractions and dialyze overnight at 4oC in 1 L of Storage Buffer (In the case of TbADAT we use 50 mM Hepes 8, 0.1 mM EDTA, 2 mM DTT, 1 mM MgCl2, and 100 mM KCl (optional)).

Generation of Ni2+ charged agarose beads

Before the next protein purification the beads must be charged by addition of Ni2+.

1. Drain 20% EtOH from beads. Discard flow through.

2. Rinse beads by allowing 15-20 ml of ddH2O to flow through beads. Discard flow through.

2+ 3. Charge beads with Ni by allowing 10-15 ml of NiSO4 (saturated solution in ddH2O) to flow through beads. Discard flow through.

4. Rinse beads by allowing 15-20 ml of ddH2O to flow through beads. Discard flow through.

4. Equilibrate beads by allowing 15-20 ml of 1x non-denaturing binding buffer to flow through beads. Discard flow through.

127

A.6 Labeling and purifying protein with 35S-Met/Cys

Make 5x M9 Salts Minimal Media -Mix Salts -64 g Na2HPO4 -15 g KH2PO4 -2.5 g NaCl -5 g NH4Cl -pH to 7.2-7.4 w/ NaOH -Autoclave 20 min -Store @ room temp

1x M9 Minimal Media (For step 1A below) Final Conc. 100mL 1L - 5x M9 minimal media 1x 20mL 200mL -20% Glucose 0.2% 1mL 10mL

-1M MgSO4*7H2O 2 mM 200L 2mL -1M CaCl2*2H2O 100 M 10L 100L -1000x Metals Mix (see below) 1x 100L 1mL

1. Transform plasmid with proper construct for protein expression into BL21 DE3 RIL X, a methionine auxotroph E. coli cell line. Pay attention to the antibiotics. Use these cells for the labeling experiment. Follow the ―Protein over-expression protocol‖ above (A.4) with the following changes:

A. At OD600 0.6-0.8 harvest cells (6000 rpm for 5 minutes) and suspended in freshly prepared 1x M9 minimal medium and 100 M ―cold‖ methionine.

B. Grow cells for an additional 30 minutes at 37 oC before induction.

C. Induce protein expression was with IPTG (0.5 mM final concentration) and 1 mCi of radio label 35S Met/Cys.

128

D. Following 15 hours of induction (at 25 oC), harvest cells and follow protocols A.5 for purification with the following change: freeze thaw cells in 1x binding buffer to lyse. Cannot sonicate due to radiolabel.

-Be sure to keep track of ―hot‖ protein and waste. -Label all tubes, beads, etc for 35S use.

129

A.7 10% Tricine-SDS PAGE

1. Clean all plates and gel casting apparatus.

2. Make separating gel -1.5 ml S mix -2.5 ml 3x Buffer -1.5 ml 50% Glycerol - fill up to 7.5 ml with ddH2O -7.6 l Temed -38 l 10% APS

Pipet 6 ml gel mix into gel cast. Add 0.1% SDS to the top to form a straight front. Let polymerize for 30 minutes.

3. Make stacking gel -500 l S mix -1.5 ml 3x Buffer - fill up to 6 ml with ddH2O -9 l Temed -45 l 10% APS

Remove the SDS. Pipet ~3 ml gel mix into gel cast. Position appropriate comb. Let polymerize for 30 minutes.

4. Run Gel -Anode buffer (Bottom): 1x TrisBuffer -Cathode buffer (Top): 1x SDS-PAGE Buffer

5. Fix gel to dehydrate 1 hour to overnight. Add buffer to just cover gel and incubate at room temperature with gentle agitation. -50% MeOH, 10% acetic acid (v/v)

6. Stain proteins at least 30 minutes. Add buffer to just cover gel and incubate with gentle agitation at room temperature. -0.025% coomassie brilliant blue, 10% acetic acid (w/v)

7. De-stain gel. -10% acetic acid (v/v)

130

A.8 Protein concentration via Bradford reaction (BioRad)

**Turn on the Spec to warm up the lamp while completing the reactions**

1. Dilute the stock Bradford reagent (store at 4oC)

-1 part reagent and 4 parts ddH2O (ie 200 l reagent and 800 l ddH2O) -make 2 ml diluted reagent per 9 reactions (remember 6 reactions will the std curve)

2. Add 200 l diluted Bradford reagent to as many clean, dry cuvettes as needed.

3. Add 5 l of each standard to each respective cuvette (for the standard curve)

-add 5 l of ddH2O to one cuvette to use as a ―blank‖. -0.2, 0.4, 0.6, 0.8, and 1.0 mg/ml are within the linear range of the reaction

4. Add 5 l of each sample to each respective cuvette -If the sample is really concentrated, dilute the sample before adding it the Bradford reagent. Be sure to take your dilution factor into account when calculating final protein concentration.

5. Wait at least 10 minutes (but no more than 1 hour) before taking the OD readings.

6. Set the Spec wavelength to 595nm. Insert the ―blank‖ and press [read blank]. Insert each standard and press [read sample] then insert each sample.

7. Use the standards to create a standard curve that will be used to calculate the sample protein concentration. Take your dilution factor into account when calculating the concentration. (Note: I typically make a couple of dilutions of the protein and average the readings to get a more accurate protein concentration).

131

A. 9 Circular dichroism spectroscopy

1. Contact the person(s) in charge of the CD machine (AVIV 62DS Spectropolarimeter) to schedule time to use it. Typically someone in the Foster lab is in charge of sign up.

- Schedule work time on the machine - Ask about the level of liquid nitrogen and if you need to buy more - Ask about instructions. If you’ve never used the CD machine before, you’ll need to be trained on how to run it successfully

2. Carefully prepare protein samples so all samples have the same concentration (about 1 mg/ml). There are later calculation for correcting for small concentration differences but all concentrations should be as close to the same as possible.

-Use 4 ml 10,000 MWCO concentrator (Millipore) to concentrate protein. The concentration step can also be used to exchange buffer for assay buffer (10 mM Tris, pH 8)

3. Follow instruction given to you to prepare the machine for use.

-bring ddH2O, Kim wipes, waste container, cuvette, loading needle, samples -Frosty liquid nitrogen tank during use is normal

4. Carefully add buffer control sample (10 mM Tris, pH 8) to cuvette using a needle. Do not introduce air bubble into sample in the cuvette because they will mess up your reading. Air bubble can be tapped out if need be but try to avoid having to do this.

5. Take measurements. The output is 3 reads per wavelength and then an average of those three reads. Numbers obtained from the buffer control will be subtracted from all sample reads.

6. Repeat steps 4 and 5 for each protein sample. Carefully flush the cuvette after each sample with copious amounts of ddH2O.

7. Calculate ellipticity values collected directly from the machine. Converted to molar ellipticity using the equation: []=100x/cl, where [] refers to molar ellipticity,  refers to ellipticity, c equals the solute molarity and l equals the path length of the cuvette used.

-Here ―c‖ is the parameter that can be used to correct for small difference in solute molarity.

132

A. 10 Adenosine to inosine activity assay

o 1. Denature T7-transcribed (site specifically radio labeled at A34) tRNA at 70 C for 3 minutes and allow the RNA to refold by slow cooling to room temperature for 2 minutes.

Assemble 10 l reaction(s) on ice -1 l 10x reaction buffer

-1 l Hot tRNA (Val A34) -x l Protein

-8-x l ddH2O

Initial assay protein concentrations used 0, 0.1, 0.5, 1, and 10 g.

2. Incubate reactions at 28 oC for 1, 3, and 15 (overnight) hours. At each time point

add reaction to a labeled tube containing 85 l ddH2O and 50 l Tris-saturated phenol. Vortex phenol mixture.

3. Separate the aqueous and phenol phases by spinning at max speed (13.2K rpm) for 10 minutes in a microcentrifuge. Carefully pipette off the top aqueous phase (contains hot tRNA) and add to a labeled microcentrifuge tube containing 250 l EtOH (200 proof), 5 l 3M NaOAc, and 1 l glycogen.

4. Allow the EtOH precipitation reaction to sit on ice for at least 30 minutes.

5. Precipitate the hot tRNA by spinning the EtOH mixture at max speed (13.2K rpm) for 30 minutes in a microcentrifuge. Carefully remove the EtOH by decanting and/or 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 9 l of 1x Nuclease P1 buffer (as supplied with the enzyme MP Biomedical) and 1l of nuclease P1 to the pellet. Incubate at 37 oC for >5 hours (overnight is best).

7. Dry sample using a speed vac (Savant) on high heat. Suspend digested tRNA

pellet in 3 l ddH2O.

8. Spot 1 l of sample on thin layer chromatography (TLC) plate and develop in an appropriate solvent system. Note: in the case of A to I, Solvent ―C‖ (phosphate buffer, ammonium sulfate, and n-propanol in a 100:60:2 (v:w:v) ratio) is used.

9. Allow the solvent front to migrate to the top of the plate (within 5 mm of the top edge of the plate). Remove the plate from the tank and allow it to air dry. 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. 133

A. 11 Kinetic adenosine to inosine assay

[Hot]= [Cold]=

[Substrate] Rxn Hot Cold Hot Final (M) Vol. (l) RNA RNA RNA RNA Cold RNA ddH2O pmol pmol pmol l l l 1 2 3 4 5 6 7 8 9 10

Kinetic assays were preformed as above (Adenosine to inosine activity protocol) with the following changes:

1. Hold the protein concentration constant and sufficiently low enough to saturate the reaction with tRNA substrate. Note: typically 100-200 ng of protein is added to each reaction.

Assemble 10 l reaction(s) on ice -1 l 10x reaction buffer

-1 l Hot tRNA (Val A34) -1 l Protein (100-200ng) -x l Cold tRNA

-7-x l ddH2O.

2. Incubate reactions at 28 oC for length of time that is within the linear range of the reaction. Note: For most ADAT2/3 mutants 45 to 60 minutes was sufficient. In a few cases reactions were incubated up to 3 hours.

3. Phenol extract, ethanol precipitate, and digest as above.

4. Spot and develop the TLC as as above.

5. Calculate the pmol of inosine formed at each substrate concentration by multiplying the faction of inosine formed (pI/(pI+pA)) by the pmol of substrate used in the reaction. Further calculate reaction velocity by dividing pmol of inosine by time (of the reaction in minutes).

134

6. Plot Velocity (Vo in pmol/min) against substrate concentration ([s] in M) and use SimgaPlot to fit data to Michaelis Menten kinetics and calculate Vmax and KM. kcat can be derived from Vmax by the equation kcat = Vmax/pmol protein.

135

A.12 Single-turnover kinetic assay to determine binding dissociation constant (Kd)

Single-turnover kinetic were preformed as above (Adenosine to inosine activity protocol) with the following changes

1. Incubate T7-transcribed tRNA with a constant saturating concentration of TbADAT2/3 for increasing time (0, 10, 20, 30 ,45, 60, 90, 180 minutes). Repeat the time course for different protein concentrations ranging from 0.05 M to 10 M.

2. At each time point the tRNA was phenol extracted, ethanol precipitated, digested, and spotted as above.

3. Calculate the fraction of inosine produced and plot it against time (in minutes) and fit data to the equation: f=a(1-e-kt)

where f is the fraction of inosine produced, t is time (minutes), a is the fraction of inosine produced at the reaction end point and k is kobs. Calculate kobs values for each protein concentration and plot values against each respective protein concentration. Fit data to a single ligand-binding curve using SigmaPlot kinetic software. Calculate Kd app from that curve.

136

A. 13 1-methyl adenosine (m1A) to 1-methyl inosine (m1I) activity assay

Assay for activity (Thermococcus kodakaraensis extracts)

1. Denature T7-transcribed (radio labeled at every A) tRNA at 70oC for 3 minutes and allow the RNA to refold by slow cooling to room temperature for 2 minutes.

2. Methylate substrate (A57) by incubating tRNA with methyltransferase TrmI.

A. Assemble 100 l reaction(s) on ice. -10 l 10x reaction buffer -50 l Hot tRNA -25 l TrmI -3 l 20 mM SAM

-12 l ddH2O

B. Incubate reactions at 50 oC for 3 hours. Stop reaction by phenol extraction (with 50 l Tris saturated phenol).

C. Separate the aqueous and phenol phases by spinning at max speed (13.2K rpm) for 10 minutes in a microcentrifuge. Carefully pipette off the top aqueous phase (contains hot tRNA) and add to a labeled microcentrifuge tube containing 250 l EtOH (200 proof), 5 l 3M NaOAc, and 1 l glycogen.

D. Allow the EtOH precipitation reaction to sit on ice for at least 30 minutes.

E. Precipitate the hot tRNA by spinning the EtOH mixture at max speed (13.2K rpm) for 30 minutes in a microcentrifuge. Carefully remove the EtOH by decanting and/or pipetting. Allow the remaining pellet to air dry (Note: a Geiger counter can be used to ensure no loss of sample).

F. Resuspend methylated tRNA substrate in 50 l ddH2O.

3. Test cellular extract (column fraction) for activity. 1 o Denature T7-transcribed (methylated m A57 from step 2) tRNA at 70 C for 3 minutes and allow the RNA to refold by slow cooling to room temperature for 2 minutes.

Assemble 10 l reaction(s) on ice -1 l 10x reaction buffer 1 -1 l Hot tRNA (m A57) -x l Protein (cellular extract)

-8-x l ddH2O

4. Incubate reactions at 50 oC for 1 hour.

137

5. Stop reaction by phenol extraction. Add 90 l ddH2O and 50 l Tris-saturated phenol. Vortex.

6. Separate the aqueous and phenol phases by spinning at max speed (13.2K rpm) for 10 minutes in a microcentrifuge. Carefully pipette off the top aqueous phase (contains hot tRNA) and add to a labeled microcentrifuge tube containing 250 l EtOH (200 proof), 5 l 3M NaOAc, and 1 l glycogen.

7. Allow the EtOH precipitation reaction to sit on ice for at least 30 minutes.

8. Precipitate the hot tRNA by spinning the EtOH mixture at max speed (13.2K rpm) for 30 minutes in a microcentrifuge. Carefully remove the EtOH by decanting and/or pipetting. Allow the remaining pellet to air dry (Note: a Geiger counter can be used to ensure no loss of sample).

9. Digest the tRNA to single nucleotides by adding 9 l of 1x Nuclease P1 buffer (as supplied with the enzyme MP Biomedical) and 1l of nuclease P1 to the pellet. Incubate at 37 oC for >5 hours (overnight is best).

10. Dry sample using a speed vac (Savant) on high heat. Suspend digested tRNA

pellet in 3 l ddH2O.

11. Spot 1 l of sample on thin layer chromatography (TLC) plate and develop in an appropriate solvent system. Note: in the case of m1A to m1I, Solvent ―C‖ (phosphate buffer, ammonium sulfate, and n-propanol in a 100:60:2 (v:w:v) ratio) is used.

12. Allow the solvent front to migrate to the top of the plate (within 5 mm of the top edge of the plate). Remove the plate from the tank and allow it to air dry. 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.

13. Compare the results to cold nucleotide markers and/or published TLC maps.

138

A. 14 Electrophoresis mobility shift assay (EMSA) (Bandshift assay)

This protocol is for a 12 reactions (2x6 reactions)-12 lane 6% polyacrylamide gel (see attached organization worksheet).

1. Dilute RNA and Protein sample to appropriate working concentration (see attached worksheet). Spin diluted protein sample for 15 minutes at 13.2k rpm to pellet any precipitated protein.

2. While spinning diluted protein, add appropriate amount of ddH2O (16, 14.75, 13.5, 11, 6, and 0 l) to each tube (on ice). When spin has completed, remove protein (supernatant) to clean microcentrifuge tube.

3. Heat 26 l RNA (0.08 M) at 70oC for 3 minutes. Cool RNA at room temperature for 2 minutes (aids in tRNA refolding).

4. Add 26 l of room temp 10x HKM buffer to the folded tRNA. Gently pipette up and down a few times to mix.

5. Add 4 l of RNA/HKM buffer mix to each tube (on ice). Add appropriate amount of clarified diluted protein to each tube (0, 1.25, 2.5, 5, 10, and 16 l).

6. Incubate reaction on ice for 15 minutes.

7. Add 5 l of 50% Glycerol to each tube. Add 5 l of 10x load dye to tube #1 (no protein control). Pipette gently 3 times and Load all 25 l in gel.

8. Run gel at NO MORE THAN 100 volts for 1.5 hours. Place gel on filter paper and cover with plastic wrap. Dry for 1 hour with heat and cool to room temperature before exposing gel to a phosphorimager screen.

9. Develop gel and quantitate signal using Image Quant software.

10. Calculate percent bound using the equation: Bound = tRNAcomplex/(tRNAfree + tRNAcomplex)

139

6% Acrylamide Non-Denaturing gel mix -6% Acrylamide in 1x NNB (ddH2O)

10x HKM Assay Buffer -500 mM Hepes pH=7.5 (or 8.0), 10 mM MgCl2, and 50 mM KCl

Standard Reaction: -2 l 10x HMK bufferl -2 l tRNA (diluted) -x l Protein

-16-x lddH2O

Example reaction setup:

RNA Used: tRNA Val (G34) Protein Used: TbADAT2/3 wild type Stock [RNA] (M) = 0.2 M Stock [Protein] (g/l) = 0.5 mg/ml Working [RNA] (M) = 0.08 M Working [protein] (g/l) = 0.029 g/l

Rxn # l of ddH2O l of protein 1 16.00 0.00 2 14.75 1.25 3 13.50 2.50 4 11.00 5.00 5 6.00 10.00 6 0.00 16.00

RNA Used: tRNA Val (G34) Protein Used: TbADAT2/3 wild type Stock [RNA] (M) = 0.2M Stock [Protein] (g/l) = 0.5 mg/ml Working [RNA] (M) = 0.08M Working [protein] (g/l) = 0.087 g/l

Rxn # l of ddH2O l of protein 7 16.00 0.00 8 14.75 1.25 9 13.50 2.50 10 11.00 5.00 11 6.00 10.00 12 0.00 16.00

140

References

Agris PF, Vendeix FA, Graham WD. 2007. tRNA's wobble decoding of the genome: 40 years of modification. J Mol Biol 366(1):1-13.

Alfonzo JD, Blanc V, Estevez AM, Rubio MA, Simpson L. 1999. C to U editing of the anticodon of imported mitochondrial tRNA(trp) allows decoding of the UGA stop codon in leishmania tarentolae. EMBO J 18(24):7056-62.

Anderson J, Phan L, Cuesta R, Carlson BA, Pak M, Asano K, Bjork GR, Tamame M, Hinnebusch AG. 1998. The essential Gcd10p-Gcd14p nuclear complex is required for 1-methyladenosine modification and maturation of initiator methionyl-tRNA. Genes Dev 12(23):3650-62.

Anderson J, Phan L, Hinnebusch AG. 2000. The Gcd10p/Gcd14p complex is the essential two-subunit tRNA(1-methyladenosine) methyltransferase of saccharomyces cerevisiae. Proc Natl Acad Sci U S A 97(10):5173-8.

Antes T, Costandy H, Mahendran R, Spottswood M, Miller D. 1998. Insertional editing of mitochondrial tRNAs of physarum polycephalum and didymium nigripes. Mol Cell Biol 18(12):7521-7.

Atomi H, Fukui T, Kanai T, Morikawa M, Imanaka T. 2004. Description of thermococcus kodakaraensis sp. nov., a well studied hyperthermophilic archaeon previously reported as pyrococcus sp. KOD1. Archaea 1(4):263-7.

Auld DS. 2001. Zinc coordination sphere in biochemical zinc sites. Biometals 14(3- 4):271-313.

Auxilien S, Crain PF, Trewyn RW, Grosjean H. 1996. Mechanism, specificity and general properties of the yeast enzyme catalysing the formation of inosine 34 in the anticodon of transfer RNA. J Mol Biol 262(4):437-58.

Bass BL and Weintraub H. 1988. An unwinding activity that covalently modifies its double-stranded RNA substrate. Cell 55(6):1089-98.

Beier H, Lee MC, Sekiya T, Kuchino Y, Nishimura S. 1992. Two nucleotides next to the anticodon of cytoplasmic rat tRNA(asp) are likely generated by RNA editing. Nucleic Acids Res 20(11):2679-83

141

Benne R, Van den Burg J, Brakenhoff JP, Sloof P, Van Boom JH, Tromp MC. 1986. Major transcript of the frameshifted coxII gene from trypanosome mitochondria contains four nucleotides that are not encoded in the DNA. Cell 46(6):819-26.

Betat H, Rammelt C, Morl M. 2010. tRNA nucleotidyltransferases: Ancient catalysts with an unusual mechanism of polymerization. Cell Mol Life Sci 67(9):1447-63.

Betts L, Xiang S, Short SA, Wolfenden R, Carter CW,Jr. 1994. Cytidine deaminase. the 2.3 A crystal structure of an enzyme: Transition-state analog complex. J Mol Biol 235(2):635-56.

Blanc V and Davidson NO. 2010. APOBEC-1-mediated RNA editing. Wiley Interdiscip Rev Syst Biol Med 2(5):594-602.

Blanc V, Henderson JO, Newberry EP, Kennedy S, Luo J, Davidson NO. 2005. Targeted deletion of the murine apobec-1 complementation factor (acf) gene results in embryonic lethality. Mol Cell Biol 25(16):7260-9.

Blum B, Bakalara N, Simpson L. 1990. A model for RNA editing in kinetoplastid mitochondria: "guide" RNA molecules transcribed from maxicircle DNA provide the edited information. Cell 60(2):189-98.

Brusa R, Zimmermann F, Koh DS, Feldmeyer D, Gass P, Seeburg PH, Sprengel R. 1995. Early-onset epilepsy and postnatal lethality associated with an editing- deficient GluR-B allele in mice. Science 270(5242):1677-80.

Campbell DA, Thomas S, Sturm NR. 2003. Transcription in kinetoplastid protozoa: Why be normal? Microbes Infect 5(13):1231-40.

Carter CW,Jr. 1995. The nucleoside deaminases for cytidine and adenosine: Structure, transition state stabilization, mechanism, and evolution. Biochimie 77(1-2):92-8.

Cattaneo R, Kaelin K, Baczko K, Billeter MA. 1989. Measles virus editing provides an additional cysteine-rich protein. Cell 56(5):759-64.

Charriere F, Helgadottir S, Horn EK, Soll D, Schneider A. 2006. Dual targeting of a single tRNA(trp) requires two different tryptophanyl-tRNA synthetases in trypanosoma brucei. Proc Natl Acad Sci U S A 103(18):6847-52.

Clayton CE. 2002. Life without transcriptional control? from fly to man and back again. EMBO J 21(8):1881-8.

Conticello SG, Thomas CJ, Petersen-Mahrt SK, Neuberger MS. 2005. Evolution of the AID/APOBEC family of polynucleotide (deoxy)cytidine deaminases. Mol Biol Evol 22(2):367-77.

142

Crain PF, Alfonzo JD, Rozenski J, Kapushoc ST, McCloskey JA, Simpson L. 2002. 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 8(6):752-61.

Crick FH. 1966. Codon--anticodon pairing: The wobble hypothesis. J Mol Biol 19(2):548- 55.

Czerwoniec A, Dunin-Horkawicz S, Purta E, Kaminska KH, Kasprzak JM, Bujnicki JM, Grosjean H, Rother K. 2009. MODOMICS: A database of RNA modification pathways. 2008 update. Nucleic Acids Res 37(Database issue):D118-21.

Das A and Bellofatto V. 2004. Genetic regulation of protein synthesis in trypanosomes. Curr Mol Med 4(6):577-84.

Droogmans L, Roovers M, Bujnicki JM, Tricot C, Hartsch T, Stalon V, Grosjean H. 2003. Cloning and characterization of tRNA (m1A58) methyltransferase (TrmI) from thermus thermophilus HB27, a protein required for cell growth at extreme temperatures. Nucleic Acids Res 31(8):2148-56.

Elias Y and Huang RH. 2005. Biochemical and structural studies of A-to-I editing by tRNA:A34 deaminases at the wobble position of transfer RNA. Biochemistry 44(36):12057-65.

Endoh T, Kanai T, Sato YT, Liu DV, Yoshikawa K, Atomi H, Imanaka T. 2006. Cell-free protein synthesis at high temperatures using the lysate of a hyperthermophile. J Biotechnol 126(2):186-95.

Fassel VA. 1978. Quantitative elemental analyses by plasma emission spectroscopy. Science 202(4364):183-91.

Fukui T, Atomi H, Kanai T, Matsumi R, Fujiwara S, Imanaka T. 2005. Complete genome sequence of the hyperthermophilic archaeon thermococcus kodakaraensis KOD1 and comparison with pyrococcus genomes. Genome Res 15(3):352-63.

Furukawa A, Nagata T, Matsugami A, Habu Y, Sugiyama R, Hayashi F, Kobayashi N, Yokoyama S, Takaku H, Katahira M. 2009. Structure, interaction and real-time monitoring of the enzymatic reaction of wild-type APOBEC3G. EMBO J 28(4):440- 51.

Gaston KW, Rubio MA, Spears JL, Pastar I, Papavasiliou FN, Alfonzo JD. 2007. C to U editing at position 32 of the anticodon loop precedes tRNA 5' leader removal in trypanosomatids. Nucleic Acids Res 35(20):6740-9.

Gerber AP, Grosjean H, Melcher T, Keller W. 1998. Tad1p, a yeast tRNA-specific adenosine deaminase, is related to the mammalian pre-mRNA editing enzymes ADAR1 and ADAR2. EMBO J 17(16):4780-9.

143

Gerber AP and Keller W. 1999. An adenosine deaminase that generates inosine at the wobble position of tRNAs. Science 286(5442):1146-9.

Gray MW. 1976. O2'-methylinosine, a constituent of the ribosomal RNA of crithidia fasciculata. Nucleic Acids Res 3(4):977-88.

Gray MW. 2003. Diversity and evolution of mitochondrial RNA editing systems. IUBMB Life 55(4-5):227-33.

Gott JM, Somerlot BH, Gray MW. 2010. Two forms of RNA editing are required for tRNA maturation in physarum mitochondria. RNA 16(3):482-8.

Greeve J, Navaratnam N, Scott J. 1991. Characterization of the apolipoprotein B mRNA editing enzyme: No similarity to the proposed mechanism of RNA editing in kinetoplastid protozoa. Nucleic Acids Res 19(13):3569-76.

Grosjean H, Auxilien S, Constantinesco F, Simon C, Corda Y, Becker HF, Foiret D, Morin A, Jin YX, Fournier M, et al. 1996. Enzymatic conversion of adenosine to inosine and to N1-methylinosine in transfer RNAs: A review. Biochimie 78(6):488- 501.

Grosjean H, Constantinesco F, Foiret D, Benachenhou N. 1995. A novel enzymatic pathway leading to 1-methylinosine modification in haloferax volcanii tRNA. Nucleic Acids Res 23(21):4312-9.

Guelorget A, Roovers M, Guerineau V, Barbey C, Li X, Golinelli-Pimpaneau B. 2010. Insights into the hyperthermostability and unusual region-specificity of archaeal pyrococcus abyssi tRNA m1A57/58 methyltransferase. Nucleic Acids Res 38(18):6206-18.

Habig JW, Dale T, Bass BL. 2007. miRNA editing--we should have inosine this coming. Mol Cell 25(6):792-3.

Harris RS, Petersen-Mahrt SK, Neuberger MS. 2002. RNA editing enzyme APOBEC1 and some of its homologs can act as DNA mutators. Mol Cell 10(5):1247-53.

Higuchi M, Single FN, Kohler M, Sommer B, Sprengel R, Seeburg PH. 1993. RNA editing of AMPA receptor subunit GluR-B: A base-paired intron-exon structure determines position and efficiency. Cell 75(7):1361-70.

Hirano K, Young SG, Farese RV,Jr, Ng J, Sande E, Warburton C, Powell-Braxton LM, Davidson NO. 1996. Targeted disruption of the mouse apobec-1 gene abolishes apolipoprotein B mRNA editing and eliminates apolipoprotein B48. J Biol Chem 271(17):9887-90.

144

Holden LG, Prochnow C, Chang YP, Bransteitter R, Chelico L, Sen U, Stevens RC, Goodman MF, Chen XS. 2008. Crystal structure of the anti-viral APOBEC3G catalytic domain and functional implications. Nature 456(7218):121-4.

Holley RW, Everett GA, Madison JT, Zamir A. 1965. Nucleotide sequences in the yeast alanine transfer ribonucleic acid. J Biol Chem 240:2122-8.

Hopper AK, Pai DA, Engelke DR. 2010. Cellular dynamics of tRNAs and their genes. FEBS Lett 584(2):310-7.

Hundley HA and Bass BL. 2010. ADAR editing in double-stranded UTRs and other noncoding RNA sequences. Trends Biochem Sci 35(7):377-83.

Hurst SR, Hough RF, Aruscavage PJ, Bass BL. 1995. Deamination of mammalian glutamate receptor RNA by xenopus dsRNA adenosine deaminase: Similarities to in vivo RNA editing. RNA 1(10):1051-60.

Johnson PF and Abelson J. 1983. The yeast tRNATyr gene intron is essential for correct modification of its tRNA product. Nature 302(5910):681-7.

Juhling F, Morl M, Hartmann RK, Sprinzl M, Stadler PF, Putz J. 2009. tRNAdb 2009: Compilation of tRNA sequences and tRNA genes. Nucleic Acids Res 37(Database issue):D159-62.

Kane JP. 1983. Apolipoprotein B: Structural and metabolic heterogeneity. Annu Rev Physiol 45:637-50.

Kawahara Y, Zinshteyn B, Sethupathy P, Iizasa H, Hatzigeorgiou AG, Nishikura K. 2007. Redirection of silencing targets by adenosine-to-inosine editing of miRNAs. Science 315(5815):1137-40.

Kim J, Malashkevich V, Roday S, Lisbin M, Schramm VL, Almo SC. 2006. Structural and kinetic characterization of escherichia coli TadA, the wobble-specific tRNA deaminase. Biochemistry 45(20):6407-16.

Knight SW and Bass BL. 2002. The role of RNA editing by ADARs in RNAi. Mol Cell 10(4):809-17.

Kosinski J, Cymerman IA, Feder M, Kurowski MA, Sasin JM, Bujnicki JM. 2003. 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 53 Suppl 6:369-79.

Koslowsky DJ, Bhat GJ, Perrollaz AL, Feagin JE, Stuart K. 1990. The MURF3 gene of T. brucei contains multiple domains of extensive editing and is homologous to a subunit of NADH dehydrogenase. Cell 62(5):901-11.

145

Koslowsky DJ, Bhat GJ, Read LK, Stuart K. 1991. Cycles of progressive realignment of gRNA with mRNA in RNA editing. Cell 67(3):537-46.

Kuratani M, Ishii R, Bessho Y, Fukunaga R, Sengoku T, Shirouzu M, Sekine S, Yokoyama S. 2005. Crystal structure of tRNA adenosine deaminase (TadA) from aquifex aeolicus. J Biol Chem 280(16):16002-8.

Lellek H, Kirsten R, Diehl I, Apostel F, Buck F, Greeve J. 2000. Purification and molecular cloning of a novel essential component of the apolipoprotein B mRNA editing enzyme-complex. J Biol Chem 275(26):19848-56.

Lonergan KM and Gray MW. 1993. Editing of transfer RNAs in acanthamoeba castellanii mitochondria. Science 259(5096):812-6.

Losey HC, Ruthenburg AJ, Verdine GL. 2006. Crystal structure of staphylococcus aureus tRNA adenosine deaminase TadA in complex with RNA. Nat Struct Mol Biol 13(2):153-9.

Macbeth MR, Schubert HL, Vandemark AP, Lingam AT, Hill CP, Bass BL. 2005. Inositol hexakisphosphate is bound in the ADAR2 core and required for RNA editing. Science 309(5740):1534-9.

MacElrevey C and Wedekind JE. 2007. Chemistry, phylogeny, and three-dimensional structure of the APOBEC protein family. In: RNA and DNA editing: Molecular mechanisms and their integration into biological systems. Smith HC, editor. Hoboken, NJ, USA: John Wiley & Sons, Inc.

Maas S, Patt S, Schrey M, Rich A. 2001. Underediting of glutamate receptor GluR-B mRNA in malignant gliomas. Proc Natl Acad Sci U S A 98(25):14687-92.

Mahendran R, Spottswood MR, Miller DL. 1991. RNA editing by cytidine insertion in mitochondria of physarum polycephalum. Nature 349(6308):434-8.

Mahendran R, Spottswood MS, Ghate A, Ling ML, Jeng K, Miller DL. 1994. Editing of the mitochondrial small subunit rRNA in physarum polycephalum. EMBO J 13(1):232-40.

Martinez-Calvillo S, Vizuet-de-Rueda JC, Florencio-Martinez LE, Manning-Cela RG, Figueroa-Angulo EE. 2010. Gene expression in trypanosomatid parasites. J Biomed Biotechnol 2010:525241.

McClain WH, Lai LB, Gopalan V. 2010. Trials, travails and triumphs: An account of RNA catalysis in RNase P. J Mol Biol 397(3):627-46.

Mehta A, Kinter MT, Sherman NE, Driscoll DM. 2000. Molecular cloning of apobec-1 complementation factor, a novel RNA-binding protein involved in the editing of apolipoprotein B mRNA. Mol Cell Biol 20(5):1846-54.

146

Morl M, Dorner M, Paabo S. 1995. C to U editing and modifications during the maturation of the mitochondrial tRNA(asp) in marsupials. Nucleic Acids Res 23(17):3380-4.

Morrison JR, Paszty C, Stevens ME, Hughes SD, Forte T, Scott J, Rubin EM. 1996. Apolipoprotein B RNA editing enzyme-deficient mice are viable despite alterations in lipoprotein metabolism. Proc Natl Acad Sci U S A 93(14):7154-9.

Murphy FV,4th and Ramakrishnan V. 2004. Structure of a purine-purine wobble base pair in the decoding center of the ribosome. Nat Struct Mol Biol 11(12):1251-2.

Navaratnam N, Fujino T, Bayliss J, Jarmuz A, How A, Richardson N, Somasekaram A, Bhattacharya S, Carter C, Scott J. 1998. Escherichia coli cytidine deaminase provides a molecular model for ApoB RNA editing and a mechanism for RNA substrate recognition. J Mol Biol 275(4):695-714.

Navaratnam N, Morrison JR, Bhattacharya S, Patel D, Funahashi T, Giannoni F, Teng BB, Davidson NO, Scott J. 1993. The p27 catalytic subunit of the apolipoprotein B mRNA editing enzyme is a cytidine deaminase. J Biol Chem 268(28):20709-12.

Nishikura K. 2010. Functions and regulation of RNA editing by ADAR deaminases. Annu Rev Biochem 79:321-49.

Palenchar JB and Bellofatto V. 2006. Gene transcription in trypanosomes. Mol Biochem Parasitol 146(2):135-41.

Petersen-Mahrt SK and Neuberger MS. 2003. In vitro deamination of cytosine to uracil in single-stranded DNA by apolipoprotein B editing complex catalytic subunit 1 (APOBEC1). J Biol Chem 278(22):19583-6.

Petit V, Guetard D, Renard M, Keriel A, Sitbon M, Wain-Hobson S, Vartanian JP. 2009. Murine APOBEC1 is a powerful mutator of retroviral and cellular RNA in vitro and in vivo. J Mol Biol 385(1):65-78.

Pollard VW, Rohrer SP, Michelotti EF, Hancock K, Hajduk SL. 1990. Organization of minicircle genes for guide RNAs in trypanosoma brucei. Cell 63(4):783-90.

Powell LM, Wallis SC, Pease RJ, Edwards YH, Knott TJ, Scott J. 1987. A novel form of tissue-specific RNA processing produces apolipoprotein-B48 in intestine. Cell 50(6):831-40.

Randau L, Stanley BJ, Kohlway A, Mechta S, Xiong Y, Söll D. 2009. A cytidine deaminase edits C to U in transfer RNAs in archaea. Science 324:657.

Rubio MA, Pastar I, Gaston KW, Ragone FL, Janzen CJ, Cross GA, Papavasiliou FN, Alfonzo JD. 2007. An adenosine-to-inosine tRNA-editing enzyme that can perform C-to-U deamination of DNA. Proc Natl Acad Sci U S A 104(19):7821-6.

147

Rubio MA, Rinehart JJ, Krett B, Duvezin-Caubet S, Reichert AS, Soll D, Alfonzo JD. 2008. Mammalian mitochondria have the innate ability to import tRNAs by a mechanism distinct from protein import. Proc Natl Acad Sci U S A 105(27):9186- 91.

Rubio MA, Ragone FL, Gaston KW, Ibba M, Alfonzo JD. 2006. C to U editing stimulates A to I editing in the anticodon loop of a cytoplasmic threonyl tRNA in trypanosoma brucei. J Biol Chem 281(1):115-20.

Rosenberg BR, Hamilton CE, Mwangi MM, Dewell S, Papavasiliou FN. 2011. Transcriptome-wide sequencing reveals numerous APOBEC1 mRNA-editing targets in transcript 3' UTRs. Nat Struct Mol Biol 18(2):230-6.

Salter JD, Krucinska J, Raina J, Smith HC, Wedekind JE. 2009. A hydrodynamic analysis of APOBEC3G reveals a monomer-dimer-tetramer self-association that has implications for anti-HIV function. Biochemistry 48(45):10685-7.

Santangelo TJ, Cubonova L, Reeve JN. 2008. Shuttle vector expression in thermococcus kodakaraensis: Contributions of cis elements to protein synthesis in a hyperthermophilic archaeon. Appl Environ Microbiol 74(10):3099-104.

Sato T, Fukui T, Atomi H, Imanaka T. 2005. Improved and versatile transformation system allowing multiple genetic manipulations of the hyperthermophilic archaeon thermococcus kodakaraensis. Appl Environ Microbiol 71(7):3889-99.

Scadden AD and Smith CW. 2001. RNAi is antagonized by A-->I hyper-editing. EMBO Rep 2(12):1107-11.

Schaub M and Keller W. 2002. RNA editing by adenosine deaminases generates RNA and protein diversity. Biochimie 84(8):791-803.

Schiffer S, Rosch S, Marchfelder A. 2002. Assigning a function to a conserved group of proteins: The tRNA 3'-processing enzymes. EMBO J 21(11):2769-77.

Schwer B and Stunnenberg HG. 1988. Vaccinia virus late transcripts generated in vitro have a poly(A) head. EMBO J 7(4):1183-90.

Scott J, Pease RJ, Powell LM, Wallis SC, McCarthy BJ, Mahley RW, Levy-Wilson B, Knott TJ. 1987. Apolipoprotein B gene: Plasma cholesterol and atherosclerosis. Biochem Soc Symp 53:155-63.

Severi F, Chicca A, Conticello SG. 2011. Analysis of reptilian APOBEC1 suggests that RNA editing may not be its ancestral function. Mol Biol Evol 28(3):1125-9.

Shah RR, Knott TJ, Legros JE, Navaratnam N, Greeve JC, Scott J. 1991. Sequence requirements for the editing of apolipoprotein B mRNA. J Biol Chem 266(25):16301-4.

148

Shandilya SM, Nalam MN, Nalivaika EA, Gross PJ, Valesano JC, Shindo K, Li M, Munson M, Royer WE, Harjes E, et al. 2010. Crystal structure of the APOBEC3G catalytic domain reveals potential oligomerization interfaces. Structure 18(1):28-38.

Smith AA, Carlow DC, Wolfenden R, Short SA. 1994. Mutations affecting transition-state stabilization by residues coordinating zinc at the active site of cytidine deaminase. Biochemistry 33(21):6468-74.

Sommer B, Kohler M, Sprengel R, Seeburg PH. 1991. RNA editing in brain controls a determinant of ion flow in glutamate-gated channels. Cell 67(1):11-9.

Tamura K, Dudley J, Nei M, Kumar S. 2007. MEGA4: Molecular evolutionary genetics analysis (MEGA) software version 4.0. Mol Biol Evol 24(8):1596-9.

Teng B, Burant CF, Davidson NO. 1993. Molecular cloning of an apolipoprotein B messenger RNA editing protein. Science 260(5115):1816-9.

Terribilini M, Lee JH, Yan C, Jernigan RL, Honavar V, Dobbs D. 2006. Prediction of RNA binding sites in proteins from amino acid sequence. RNA 12(8):1450-62.

Van Horn DJ, Yoo CJ, Xue D, Shi H, Wolin SL. 1997. The la protein in schizosaccharomyces pombe: A conserved yet dispensable phosphoprotein that functions in tRNA maturation. RNA 3(12):1434-43.

Vanhoof G, Goossens F, De Meester I, Hendriks D, Scharpe S. 1995. Proline motifs in peptides and their biological processing. FASEB J 9(9):736-44.

Vogel A, Schilling O, Spath B, Marchfelder A. 2005. The tRNase Z family of proteins: Physiological functions, substrate specificity and structural properties. Biol Chem 386(12):1253-64.

Wohlgamuth-Benedum JM, Rubio MA, Paris Z, Long S, Poliak P, Lukes J, Alfonzo JD. 2009. Thiolation controls cytoplasmic tRNA stability and acts as a negative determinant for tRNA editing in mitochondria. J Biol Chem 284(36):23947-53.

Wolf J, Gerber AP, Keller W. 2002. tadA, an essential tRNA-specific adenosine deaminase from escherichia coli. EMBO J 21(14):3841-51.

Wolfenden R. 1969. On the rate-determining step in the action of adenosine deaminase. Biochemistry 8(6):2409-12.

Xie K, Sowden MP, Dance GS, Torelli AT, Smith HC, Wedekind JE. 2004. The structure of a yeast RNA-editing deaminase provides insight into the fold and function of activation-induced deaminase and APOBEC-1. Proc Natl Acad Sci U S A 101(21):8114-9.

149

Xiong Y and Steitz TA. 2006. A story with a good ending: TRNA 3'-end maturation by CCA-adding enzymes. Curr Opin Struct Biol 16(1):12-7.

Yokoyama S and Nishimura S. 1995. Modified nucleosides and codon recognition. In: tRNA: Structure, biosynthesis and function. Soll D and RajBhandary U, editors. Washington, DC: American Society for Microbiology. 207 p.

Yoo CJ and Wolin SL. 1997. The yeast la protein is required for the 3' endonucleolytic cleavage that matures tRNA precursors. Cell 89(3):393-402.

150