THE PENNSYLVANIA STATE UNIVERSITY SCHREYER HONORS COLLEGE

DEPARTMENT OF CHEMISTRY

INVESTIGATION AND STRUCTURAL ANALYSIS OF INNATE IMMUNE SENSOR PKR ACTIVATION BY THE CLOSTRIDIA BOLTEAE TWISTER RIBOZYME

ANANYA ANMANGANDLA SPRING 2017

A thesis submitted in partial fulfillment of the requirements for a baccalaureate degree in Chemistry with honors in Chemistry

Reviewed and approved* by the following:

Philip C. Bevilacqua Professor of Chemistry Professor of Biochemistry and Molecular Biology Thesis Supervisor

Raymond L. Funk Professor of Chemistry Honors Adviser

David D. Boehr Associate Professor of Chemistry Faculty Reader in Chemistry

* Signatures are on file in the Schreyer Honors College. i

ABSTRACT

The innate immune system is the body’s first line of defense. This system uses a generalized mechanism to differentiate cellular RNA from pathogenic RNA like that of bacteria and viruses. The RNA-activated protein kinase, PKR, binds dsRNA and is a necessary sensor in the innate immune response. PKR is known to recognize long stretches of viral dsRNA and inhibit translation by autophosphorylation and subsequent phosphorylation of initiation factor eIF2α. Recently, PKR has been found to be activated by a wider range of pathogens, including certain bacteria. We studied the interaction between PKR and the self-cleaving, structurally compact C. bolteae twister ribozyme. This RNA is representative of small, structured RNAs that are typically found in bacteria. I found that as 3’ tail length of the RNA increases, PKR activation becomes more potent. Further, I find that in the presence of lowered Mg2+, similar to when bacterial RNA enters the human cell, activation of PKR by the short, compact ribozyme still occurs, albeit two-fold lower, supporting the hypothesis that PKR potentially is activated by this or similar RNAs in vivo. Following these findings, the interaction between PKR and the twister ribozyme was further characterized through PKR footprinting. While no obvious PKR footprint is observed, we show that following PKR addition, tertiary structure of the RNA remains intact, consistent with previous characterizations of PKR’s interactions with functional

RNAs. This may indicate PKR is binding nonspecific helical regions of the RNA. My results represent a model system for similar RNAs in vivo and provide new insights into how PKR acts as an innate immune signaling protein for the presence of bacteria and suggest a reason for the apparent absence of protein-free riboswitches and ribozymes in the human genome.

ii

TABLE OF CONTENTS

LIST OF FIGURES ...... iii

LIST OF TABLES ...... iv

ACKNOWLEDGEMENTS ...... v

Chapter 1 Introduction to Thesis...... 1

1.1 Innate Immunity and Innate Immune Sensors ...... 1 1.2 RNA Structure and Function ...... 4 1.3 Activation of PKR by RNA ...... 6 1.4 References ...... 8

Chapter 2 Activation of PKR by C.bolteae Twister Ribozyme ...... 10

2.1 Abstract ...... 10 2.2 Introduction ...... 11 2.2.1 PKR Activation by Functional RNAs ...... 11 2.2.2 Twister Ribozyme ...... 12 2.3 Materials and Methods ...... 15 2.3.1 RNA Sequence Design ...... 15 2.3.2 RNA Preparation and Purification ...... 15 2.3.3 PKR Activation Assay ...... 16 2.4 Results and Discussion ...... 17 2.4.1 3’-Tail Activation of PKR by the Twister Ribozyme ...... 17 2.4.2 Activation of T+15 Twister Ribozyme at Physiological Mg2+ Level .. 22 2.5 References ...... 24

Chapter 3 PKR Footprinting and Structural Analysis of Twister Ribozyme ...... 25

3.1 Abstract ...... 25 3.2 Introduction ...... 25 Ribonuclease Structure and Mechanism ...... 26 PKR Footprinting of Vc2 Riboswitch and glmS Riboswitch-Ribozyme ...... 27 3.3 Materials and Methods ...... 30 3.3.1 RNA Sequence Design ...... 30 3.3.2 RNA Preparation and Purification ...... 30 3.3.3 RNA Structure Mapping and PKR Footprinting Experiments ...... 31 3.4 Results and Discussion ...... 32 3.5 References ...... 39

Chapter 4 Future Directions ...... 40 iii

Chapter 4 References ...... 43

Appendix A Twister Ribozyme Predicted Secondary Structures ...... 44

Appendix B Chapter 3 Supplementary Figures ...... 45

ACADEMIC VITA ...... 48

iv

LIST OF FIGURES

Figure 1.1 PKR Domain Organization...... 3

Figure 1.2 PKR Translation Inhibition Mechanism...... 3

Figure 1.3 Central Dogma Illustration...... 4

Figure 1.4 Levels of RNA Structure...... 5

Figure 2.1 Secondary Structure Model of Twister Ribozyme ...... 13

Figure 2.2 2.9-Å Crystal Structure of env22 Twister Ribozyme ...... 14

Figure 2.3 Clostridia bolteae Twister Ribozyme Constructs...... 18

Figure 2.4 Increasing 3'-tail length leads to potent activation of PKR...... 21

Figure 2.5 Activation of PKR at Physiological Mg2+Concentration...... 23

Figure 3.1 Footprints of PKR on Functional RNAs...... 28

Figure 3.2 Single Stranded Stacked Loop L3 Cleaved by RNase V1...... 35

Figure 3.3 Summary of Structure Mapping Experiments...... 38

Figure S1 T+15 Structure Mapping and PKR Footprinting...... 45

Figure S2 PKR Comparison...... 46

Figure S3 Ribonuclease Optimization Experiments...... 47

v

LIST OF TABLES

Table 3.1: Nuclease/Ladder Cleavage Sites ...... 33

vi

ACKNOWLEDGEMENTS

First, I would like to express my gratitude to my research mentor Dr. Philip Bevilacqua.

Despite a seemingly impossible schedule, you are always available to answer a quick question.

You have taught me how to think like a scientist, design experiments, and properly read literature. After having taken many classes over the course of college, two of my favorites are still Chem 110H and Chem 572, which I attribute to your ability to engage students and create a fulfilling learning environment. I will always have a fond memory of the Bevilacqua lab, and much of that is due to your mentorship and guidance.

I would like to thank both my graduate mentors, Dr. Chelsea Hull and Kyle Messina.

Chelsea, when I joined the lab I remember feeling intimidated working with someone who had as much knowledge and experience as you did. However, that ended up being a huge blessing. Even early on you pushed me to understand what I was doing and why I was doing it, rather than just giving me directions to follow without thought. You taught me to write detailed outlines of my experiments, which has helped me work and think independently in the lab. I am so grateful for your mentorship and miss seeing you around, but am incredibly happy for you and your new family. Kyle, I have had a great time working with you. You are always available to answer my questions, and I appreciate your detailed comments on all my writing. I am inspired by your work ethic and ability to juggle two fairly different projects so well.

To the rest of the Bevilacqua lab, thank you for allowing me to feel so comfortable and integrated into the lab, despite being an undergraduate. Thank you Dr. James Mayer, for inviting me to work in your lab over the summer of 2016, with no previous knowledge in your research area. That experience taught me so much about myself as a person and scientist. Thank you to vii my committee members Dr. David Boehr and Dr. Raymond Funk for your guidance regarding my thesis composition.

Lastly, I would like to sincerely thank my parents, Uma and Sharath, for their endless support of me and my goals. Your confidence in my abilities has pushed me to reach further and try harder in everything I do. To my sister, Sarayu, thank you for being the only person I know who talks more than I do, you always make me smile and are always available for a conversation. This holds true for my boyfriend, Derek, and all of the wonderful friends I have made over the course of my time at Penn State, who have provided me with many happy memories.

1

Chapter 1

Introduction to Thesis

1.1 Innate Immunity and Innate Immune Sensors

The innate immune system is the body’s first line of defense. It is naturally present in the body, unlike adaptive immunity, and uses a generalized mechanism to differentiate cellular RNA from pathogenic RNA like that of bacteria and viruses.1,2,3 Innate immune sensors recognize pathogen-associated molecular patterns (PAMPs), which are conserved, generalized molecular structures such as glycoproteins, proteoglycans, lipopolysaccharides, or nucleic acid motifs that are commonly found among microbes and essential to their survival or infectivity.4 These sensor proteins are designated as pattern recognition receptors (PRRs), and vary by cell location, binding specificity, and activation of unique downstream signaling pathways. The variety of mechanisms employed by these sensors allow for an immediate response to a range of pathogens.5

Our interest lies in the RNA virus activated PRRs such as toll-like receptors (TLRs), retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs), and nucleotide-binding oligomerization domain-containing (NOD)-like receptors (NLRs).4 TLRs are a family of transmembrane glycoproteins containing a C-terminal signaling domain and N-terminal binding domain that binds either lipids or nucleic acids.5 The C-terminus contains a conserved ~200 amino acid region in its cytoplasmic tails, referred to as the TIR domain. The N-terminal extracellular domain contains multiple leucine-rich repeats that form a horseshoe structure and is known to be the PAMP-binding region.6 Upon ligand binding, TLRs dimerize and undergo a 2 conformational change, promoting recruitment of downstream signaling molecules.7 This eventually leads to the production and release of type I interferons (IFN) to launch an antiviral inflammatory response. There are 10 TLRs present in humans, and five of these are thought to be important in structural recognition of viral RNA and surface glycoproteins.6 TLRs are expressed either at the surface of the cell or the inner endo-lysosomal membranes and are thus essentially ineffective in recognizing intracellular pathogens.8 Examples of PRRs present in the cytosol are discussed below.

RLRs are a family of cytosolic proteins that recognize viral RNAs. They belong to a family of (DExD/H) box helicases with a short conserved amino acid sequence. This family includes the retinoic acid-inducible gene I product (RIG-I), melanoma differentiation-associated antigen 5 (MDA5), and laboratory of genetics and physiology 2 (LGP2). RIG-1 and MDA5 contain an RNA binding C-terminal domain.4 Upon binding, the cysteine-aspartic protease recruiting domain (CARD) in the N-terminus is exposed, enabling interaction with CARD- containing proteins to trigger downstream signaling.9 Due to the lack of a CARD domain in the

N-terminus, LGP2 is thought to be a negative regulator, modulating innate immune response as opposed to acting as a PAMP sensor.4 RLRs share signaling pathways with TLRs, leading to a similar induced gene expression, and eventual antiviral inflammatory response via IFNs.

NLRs, like RLRs, are a family of cytosolic proteins that act as innate immune sensors.

Over 20 have been found in the human genome. These proteins are characterized by three functional domains: the C-terminal leucine rich repeat domain, N-terminal effector binding domain, and central NAIP, CIIT, HET-E, and TP1 (NACHT) domain. Following PAMP recognition by the LRR domain, NLR oligomerization and activation is stimulated via induced 3 proximity, while the effector binding domain, generally consisting of a CARD or pyrin domain, signals downstream events.10

Aside from the PRRs mentioned above, additional cytosolic sensors exist in the cell. The

IFN-induced RNA-activated protein kinase (PKR), which has been extensively studied in the

Bevilacqua lab, recognizes viral RNA and leads to an innate immune response. PKR is a 50-60 kDa protein kinase containing two dsRNA binding motifs (dsRBM), and one kinase domain

(Figure 1.1).

Figure 1.1 PKR Domain Organization. The N-terminal domain consists of two dsRBM binding motifs (yellow and orange) and the C-terminal domain (purple) acts as a kinase. PKR is known to recognize long stretches of viral dsRNA and inhibit translation by autophosphorylation and phosphorylation of Ser51 on initiation factor eIF2α (Figure 1.2).1, 3

Figure 1.2 PKR Translation Inhibition Mechanism. PKR dimerizes on dsRNA, autophosphorylates, and subsequently phosphorylates eIF2α, inhibiting translation initiation The dsRBM interacts non-sequence specifically with the wide, shallow minor groove of A-form

RNA primarily with the 2’-hydroxyls. This domain is quite conserved and present in other dsRNA binding proteins such as Dicer, Drosha, or PACT.11,12 Currently no structure exits of 4 RNA-bound PKR, likely due to the non-specific nature of these interactions. Activation of PKR requires dimerization in the presence of long dsRNA, where generally 16 bp are required for monomeric binding and 30 bp to dimerize and activate.13,14,15 However as PKR dimerization is required for activity, high RNA concentrations turn off PKR activity due to the formation of

PKR/RNA monomers.13 This leads to a bell shaped dependence of PKR activation on RNA concentration. PKR has been shown to be activated by a wide range of pathogens, including certain bacteria, which are discussed further in Section 1.3.16

1.2 RNA Structure and Function

DNA (deoxyribonucleic acid) contains all the genetic information of a given organism, making DNA essential to the existence of life. In order to carry out cellular functions, DNA is transcribed into messenger RNA (mRNA), which is exported into the cytoplasm of the cell and subsequently translated by the ribosome to synthesize necessary proteins that regulate or perform various cellular processes.17 This flow of information is referred to as the central dogma of molecular biology (Figure 1.3).

Figure 1.3 Central Dogma Illustration. DNA is transcribed into RNA, which is then translated into protein 5 Three main RNA types exist within the cell: transfer RNA (tRNA), messenger RNA (mRNA), and ribosomal RNA (rRNA). In recent years, researchers have shown that RNA plays a larger role beyond that of a protein template.

Unlike DNA, RNA is single stranded, giving it the unique ability to fold back onto itself to base pair, creating various structures with unique functions. RNA can be described in terms of primary, secondary, and tertiary structure. The RNA nucleotide sequence represents the primary structure. Secondary structure describes the intramolecular interactions an RNA makes with itself. Examples of secondary structure include hairpin structures, internal bulges, or stem loops.

The tertiary level of RNA structure describes multiple RNA strands or distant regions of RNA strands interacting with each other including, but not limited to pseudoknots, kissing hairpins, or three-way junctions. Examples of each of these levels of structure are illustrated in Figure 1.4 below. Functional RNAs include those with gene regulatory functions such as riboswitches18,19 and microRNAs20,21, and ribozymes, RNAs with catalytic function22,23.

Figure 1.4 Levels of RNA Structure. Diagram showing primary, secondary, and tertiary levels of RNA structure and associated motifs. 6 1.3 Activation of PKR by RNA

As described briefly in Section 1.1, PKR is classically activated by long stretches of dsRNA. However, recently it has been found to activate by multiple RNA motifs including 5’- triphosphorylated single-stranded RNA (5’ppp-ssRNA), secondary structures including bulges, and an array of functional RNAs including the HDV ribozyme, HCV IRES, and HIV-I TAR

RNA.24-28

Through a collaboration between our lab and the Cameron lab, it was found that 5’ppp- ssRNA activates PKR even with very little secondary structure.14, 27 Other types of primary structure modifications were made such as 5’-pp, 5’-OH, and a 7mG cap, but only the 5’ppp- ssRNA version activated PKR.27 However, dsRNA without the 5’-ppp was found to activate

PKR, suggesting two individual methods of PKR activation. Evolutionarily, human RNA does not contain a 5’-ppp that is often found in pathogenic RNA, offering a potential explanation for the differentiation between self and non-self RNA by PKR.

PKR has also been shown to recognize secondary structure beyond perfectly dsRNA. For example, functional viral RNA such as HIV-TAR RNA activates PKR and contains many internal bulges and loops among regions of strictly dsRNA. Work in our lab has shown that PKR is only activated when the viral RNA dimerizes, allowing enough base pairs to accommodate

PKR dimerization. In contrast, the monomeric form of the virus inhibits PKR.29 Another viral

RNA shown to activate PKR is HCV IRES RNA. This RNA contains long, extended bp regions with diverse secondary structure. Our lab found PKR was most potently activated by domain II of the IRES, which is thought to mimic A-form dsRNA resulting in high activation.28

Tertiary structure of RNA has also been shown to participate in PKR regulation, both as an activator and inhibitor. For example, the pseudoknot that forms in the 5’UTR of cellular 7 mRNA for interferon-γ activates PKR. This interferon is responsible for upregulation of PKR in the cell.30 Similarly, tRNA , which is characterized by compact tertiary structure, was shown to potently activate PKR in its native form. However, when primary structure modifications were made, this activity was lost.31 Similar effects were found on a variety of transcripts with a wide variety of modifications.32 In addition RNA modifications increase translation.33 On the other

34 hand, tertiary interactions in viral VAI RNA were shown to inhibit PKR.

In addition to those mentioned above, PKR has been shown to activate in the presence of various functional RNAs, which are discussed in Section 2.2.

Chapter 2 will discuss my experiments utilizing the twister ribozyme as a model system for small structured RNAs and characterizing its interaction with PKR. I found that 3’-tail length of the RNA increased PKR activation. These findings indicate simple structural motifs on small ribozymes can activate PKR. To follow up this study, I wanted to understand the binding interaction between PKR and the twister ribozyme. Therefore, in Chapter 3 I discuss my structure mapping and PKR footprinting experiments where I tried to understand how PKR dimerizes on the small ribozyme, a nontraditional PKR activator. Here, I saw no clear PKR footprint, but show native tertiary structure remains intact. Together, these findings support

PKRs ability to activate in the presence of small structured RNAs in vivo.

8 1.4 References

1. Balachandran, S.; Roberts, P. C.; Brown, L. E.; Truong, H.; Pattnaik, A. K.; Archer, D. R.; Barber, G. N., Essential role for the dsRNA-dependent protein kinase PKR in innate immunity to viral infection. Immunity 2000, 13 (1), 129-41. 2. Nallagatla, S. R.; Toroney, R.; Bevilacqua, P. C., Regulation of innate immunity through RNA structure and the protein kinase PKR. Curr Opin Struct Biol 2011, 21 (1), 119-27. 3. Munir, M.; Berg, M., The multiple faces of proteinkinase R in antiviral defense. Virulence 2013, 4 (1), 85-9. 4. Jensen, S.; Thomsen, A. R., Sensing of RNA viruses: a review of innate immune receptors involved in recognizing RNA virus invasion. J Virol 2012, 86 (6), 2900-10. 5. Akira, S.; Uematsu, S.; Takeuchi, O., Pathogen recognition and innate immunity. Cell 2006, 124 (4), 783-801. 6. Akira, S.; Takeda, K., Toll-like receptor signalling. Nat Rev Immunol 2004, 4 (7), 499-511. 7. O'Neill, L. A.; Bowie, A. G., The family of five: TIR-domain-containing adaptors in Toll-like receptor signalling. Nat Rev Immunol 2007, 7 (5), 353-64. 8. Mogensen, T. H., Pathogen recognition and inflammatory signaling in innate immune defenses. Clin Microbiol Rev 2009, 22 (2), 240-73, Table of Contents. 9. Kato, H.; Takeuchi, O.; Sato, S.; Yoneyama, M.; Yamamoto, M.; Matsui, K.; Uematsu, S.; Jung, A.; Kawai, T.; Ishii, K. J.; Yamaguchi, O.; Otsu, K.; Tsujimura, T.; Koh, C. S.; Reis e Sousa, C.; Matsuura, Y.; Fujita, T.; Akira, S., Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature 2006, 441 (7089), 101-5. 10. Franchi, L.; McDonald, C.; Kanneganti, T. D.; Amer, A.; Nunez, G., Nucleotide-binding oligomerization domain-like receptors: intracellular pattern recognition molecules for pathogen detection and host defense. J Immunol 2006, 177 (6), 3507-13. 11. Stefl, R.; Oberstrass, F. C.; Hood, J. L.; Jourdan, M.; Zimmermann, M.; Skrisovska, L.; Maris, C.; Peng, L.; Hofr, C.; Emeson, R. B.; Allain, F. H., The solution structure of the ADAR2 dsRBM-RNA complex reveals a sequence-specific readout of the minor groove. Cell 2010, 143 (2), 225-37. 12. Wu, H.; Henras, A.; Chanfreau, G.; Feigon, J., Structural basis for recognition of the AGNN tetraloop RNA fold by the double-stranded RNA-binding domain of Rnt1p RNase III. Proceedings of the National Academy of Sciences of the United States of America 2004, 101 (22), 8307-12. 13. Lemaire, P. A.; Anderson, E.; Lary, J.; Cole, J. L., Mechanism of PKR Activation by dsRNA. J. Mol. Biol. 2008, 381 (2), 351-60. 14. Zheng, X.; Bevilacqua, P. C., Activation of the protein kinase PKR by short double-stranded RNAs with single-stranded tails. RNA (New York, N.Y.) 2004, 10 (12), 1934-45. 15. Manche, L.; Green, S. R.; Schmedt, C.; Mathews, M. B., Interactions between double-stranded RNA regulators and the protein kinase DAI. Mol Cell Biol 1992, 12 (11), 5238-48. 16. Nallagatla, S. R.; Toroney, R.; Bevilacqua, P. C., RNA structure and regulation of innate immunity through protein kinase PKR. Curr Opin Struct Biol 2011, 21 (1), 119-27. 17. Crick, F., Central dogma of molecular biology. Nature 1970, 227 (5258), 561-3. 18. Breaker, R. R., Riboswitches and the RNA world. Cold Spring Harbor perspectives in biology 2012, 4 (2). 19. Serganov, A.; Nudler, E., A decade of riboswitches. Cell 2013, 152 (1-2), 17-24. 20. Bartel, D. P., MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004, 116 (2), 281-97. 21. Bartel, D. P., MicroRNAs: target recognition and regulatory functions. Cell 2009, 136 (2), 215- 33. 9 22. Cech, T. R.; Zaug, A. J.; Grabowski, P. J., In vitro splicing of the ribosomal RNA precursor of Tetrahymena: involvement of a guanosine nucleotide in the excision of the intervening sequence. Cell 1981, 27 (3 Pt 2), 487-96. 23. Lilley, D. M., Catalysis by the nucleolytic ribozymes. Biochem. Soc. Trans. 2011, 39 (2), 641-6. 24. Patel, R. C.; Stanton, P.; Sen, G. C., Role of the amino-terminal residues of the interferon-induced protein kinase in its activation by double-stranded RNA and heparin. J. Biol. Chem. 1994, 269 (28), 18593-8. 25. Patel, R. C.; Sen, G. C., PACT, a protein activator of the interferon-induced protein kinase, PKR. EMBO J. 1998, 17 (15), 4379-90. 26. Patel, C. V.; Handy, I.; Goldsmith, T.; Patel, R. C., PACT, a stress-modulated cellular activator of interferon-induced double-stranded RNA-activated protein kinase, PKR. J. Biol. Chem. 2000, 275 (48), 37993-8. 27. Nallagatla, S. R.; Hwang, J.; Toroney, R.; Zheng, X.; Cameron, C. E.; Bevilacqua, P. C., 5'- triphosphate-dependent activation of PKR by RNAs with short stem-loops. Science 2007, 318 (5855), 1455-8. 28. Toroney, R.; Nallagatla, S. R.; Boyer, J. A.; Cameron, C. E.; Bevilacqua, P. C., Regulation of PKR by HCV IRES RNA: importance of domain II and NS5A. J. Mol. Biol. 2010, 400 (3), 393-412. 29. Heinicke, L. A.; Wong, C. J.; Lary, J.; Nallagatla, S. R.; Diegelman-Parente, A.; Zheng, X.; Cole, J. L.; Bevilacqua, P. C., RNA dimerization promotes PKR dimerization and activation. J. Mol. Biol. 2009, 390 (2), 319-38. 30. Ben-Asouli, Y.; Banai, Y.; Pel-Or, Y.; Shir, A.; Kaempfer, R., Human interferon-gamma mRNA autoregulates its translation through a pseudoknot that activates the interferon-inducible protein kinase PKR. Cell 2002, 108 (2), 221-32. 31. Nallagatla, S. R.; Jones, C. N.; Ghosh, S. K.; Sharma, S. D.; Cameron, C. E.; Spremulli, L. L.; Bevilacqua, P. C., Native tertiary structure and nucleoside modifications suppress tRNA's intrinsic ability to activate the innate immune sensor PKR. PLoS One 2013, 8 (3), e57905. 32. Nallagatla, S. R.; Bevilacqua, P. C., Nucleoside modifications modulate activation of the protein kinase PKR in an RNA structure-specific manner. RNA (New York, N.Y.) 2008, 14 (6), 1201-13. 33. Anderson, B. R.; Muramatsu, H.; Nallagatla, S. R.; Bevilacqua, P. C.; Sansing, L. H.; Weissman, D.; Kariko, K., Incorporation of pseudouridine into mRNA enhances translation by diminishing PKR activation. Nucleic Acids Res. 2010, 38 (17), 5884-92. 34. Launer-Felty, K.; Wong, C. J.; Wahid, A. M.; Conn, G. L.; Cole, J. L., Magnesium-dependent interaction of PKR with adenovirus VAI. J. Mol. Biol. 2010, 402 (4), 638-44.

10

Chapter 2

Activation of PKR by C.bolteae Twister Ribozyme

[Published as part of a paper entitled “Bacterial Riboswitches and Ribozymes Potently Activate the Human Innate Immune Sensor PKR” by Chelsea M. Hull, Ananya Anmangandla, and Philip C. Bevilacqua in ACS Chemical Biology 11, 1118-1127 (2016). Only my contributions to this paper are provided in this chapter, written in my own words]

2.1 Abstract

The RNA-activated protein kinase, PKR, binds dsRNA and is a necessary sensor in the innate immune response. In this study, we looked at the interaction between PKR and the structurally compact twister ribozyme. This RNA is representative of small, structured RNAs that are typically found in bacteria. Using PKR activation assays, we found a trend of higher

PKR activation with increasing 3’ tail length. Under typical activation conditions of 4 mM Mg2+

Similar experiments were performed at physiological Mg2+ levels of 0.5 mM Mg2+ where we found activation was still potent, albeit two-fold lower. Our evidence with this RNA indicates that similar RNAs may be capable of PKR activation in vivo. Chapter 3 supplements this study by attempting to footprint PKR onto the twister ribozyme and understand the observed trend in activation increase with 3’-tail extension.

11 2.2 Introduction

2.2.1 PKR Activation by Functional RNAs

The cell contains multiple regulatory mechanisms, generally controlled by protein enzymes. However, a number of these regulatory mechanisms are controlled by functional RNAs including riboswitches and ribozymes. It is hypothesized that these functional RNAs were more abundant in the RNA world era and were phased out as protein enzymes began to dominate cellular activity. Still, many classes of these functional RNAs have been identified over the last few decades. Riboswitches can modulate transcription, splicing, translation or RNA stability by sensing intracellular metabolite concentrations.1 These structured RNAs are usually present in the noncoding regions of bacterial mRNA. Riboswitches are composed of an aptamer and an expression platform which are responsible for metabolite binding and interfacing with the transcriptional or translation machinery, respectively.2 Upon binding of a ligand, usually a metabolite, the riboswitch adopts an alternative fold, regulating downstream processes.1

Ribozymes are functional RNAs capable of catalyzing specific biochemical processes and often times regulate downstream function by self-cleaving. Despite being much simpler in composition than traditional protein enzymes, ribozymes are able to catalyze self-cleavage rapidly and at a specific site. Ribozymes vary in their secondary structures and three-dimensional folds. In general, ribozymes cleave the RNA backbone by a reversible phosphodiester-cleavage reaction, but the exact mechanism differs between ribozymes.3

PKR has been shown to activate in the presence of various riboswitches and ribozymes.

Through work in the Bevilacqua lab, including my work described in Section 2.4, it was found that PKR is activated by bacterial riboswitches and ribozymes.4-5 My graduate mentor, Dr. 12 Chelsea Hull, found for the same ACS Chemical Biology paper, that PKR was potently activated by the cyclic di-GMP riboswitch and glmS riboswitch-ribozyme in comparison to long dsRNA.

She also found that the glmS riboswitch-ribozyme revealed two PKR footprints similar in length to 16 bp RNA, consistent with dimerization on functional bacterial RNAs. Such activation of

PKR by unrelated and varying RNA tertiary structures strongly suggests that specific RNA tertiary structure is not necessary for PKR activation.4 This follows the expectation for an innate immune sensor, which are known to have broad responses, and is consistent with nonspecific

PAMP recognition through dsRBDs.6 My contribution to this study was to investigate whether a small and simple ribozyme, twister, could regulate activation of PKR.

2.2.2 Twister Ribozyme

Self-cleaving ribozymes are noncoding RNAs which undergo cleavage and regulate downstream function under appropriate conditions. Ribozymes are much simpler in composition than traditional enzymes, yet display rates and specificities that approach those of protein enzymes.

In January of 2014 a paper was published in Nature Chemical Biology, discussing the discovery of a self-cleaving ribozyme class called the twister ribozyme.7 Through the use of bioinformatics, the researchers discovered a commonly occurring motif with conserved secondary structure, coining it “twister” (Figure 2.1a). The RNA motif was found in the bacterial class Clostridia as well as a variety of eukaryotic species. Many examples of the twister ribozyme are circularly permutated, meaning the location of 5’ and 3’end varies. These permutations are referred to as Type P1, P3, or P5 configurations (Figure 2.1b, c). The twister 13 ribozyme resembles small riboswitches in size and structural complexity, but is functionally similar and shares a strikingly similar genetic context to the hammerhead ribozyme.

i + ii

Figure 2.1 Secondary Structure Model of Twister Ribozyme. (a) Conserved sequence of twister ribozyme based on 2,690 twister ribozymes in Type P1 configuration. The cleavage site is indicated with a black arrowhead. Gray, black, and red nucleotides indication conservation of at least 75%, 90%, and 97%, respectively, and less conserved nucleotides are indicated by circles. Green shading denotes predicted base pairing, and two predicted psuedoknots are labeled i and ii. R and Y represent purine and pyrimidine, respectively. Parenthetical numbers are the variable lengths that form the stem loop structures where indicated. Circularly permutated Type (b,c) P3 (b) and P5 (c) configurations of twister ribozyme that open at the P3 or P5 stem respectively. Image adapted from Roth, A. et. al. (2014) Nat. Chem. Biol. In order for twister to cleave, Mg2+ cations are required, though the observed lack of divalent cation specificity suggests a structural rather than direct catalytic role.7 The cleavage products produced are a 5’ product containing 2’3’-cyclic phosphate on the 3’ end and a 3’ product with a

5’OH. Mechanistic evidence supports a concerted general acid-base pathway.8 However, the 14 specific mechanism by which this occurs, i.e. specific roles of nucleobases and metal ions, is still disputed. Multiple crystal structures exist of bimolecular versions of several twister ribozymes.

A representative crystal structure of the P3 type env22 twister ribozyme is shown in Figure 2.2.9

While several crystal structures exist, the method of inhibition varies between constructs with some substituting the active site 2’OH with either a 2’H or 2’OCH3 sometimes altering the active site. Additionally, inconsistencies between the structures, like location of Mg2+ ions within the

8-10 structures, have also stymied characterization of the catalytic mechanism.

a b

Figure 2.2 2.9-Å Crystal Structure of env22 Twister Ribozyme. (a) Secondary fold of the env22 twister ribozyme bimolecular construct, which contains a dU5 at the U5-A6 cleavage site. (b) Ribbon view of the 2.9-Å structure of env22 twister ribozyme color-coded as shown in a. Image adapted from Ren, A. et. al. (2014) Nat. Commun.

15 2.3 Materials and Methods

[Adapted from a previous research summary by me titled “Investigating PKR Activation by

Bacterial Ribozymes” Fall 2015]

2.3.1 RNA Sequence Design

Twister RNA constructs (T – T+15) were transcribed from hemi-duplex templates, where the bottom strand is complementary (underlined) to the T7 promoter. Bold G’s are not present in the native transcript and were added to enhance transcription efficiency.

(BS template) T 5’TTCCCACTCTGCATTGATCAGGGCTTGTGACCTGCACCGGCTATAGGCCGGTGGCTGCATTA GGAAGGTATAGTGAGTCGTATTAATTTC 3’ T+5 5’CTCCTTTCCCACTCTGCATTGATCAGGGCTTGTGACCTGCACCGGCTATAGGCCGGTGGCTG CATTAGGAAGGGGCGGTATAGTGAGTCGTATTAATTTC 3’ T+10 5’ATTTCCTCCTTTCCCACTCTGCATTGATCAGGGCTTGTGACCTGCACCGGCTATAGGCCGGTG GCTGCATTAGGAAGGGGCAGCTGGTATAGTGAGTCGTATTAATTTC 3’ T+15 5’CATGGATTTCCTCCTTTCCCACTCTGCATTGATCAGGGCTTGTGACCTGCACCGGCTATAGGC CGGTGGCTGCATTAGGAAGGGGCAGCTGCCCCTGGTATAGTGAGTCGTATTAATTTC 3’

2.3.2 RNA Preparation and Purification

In vitro T7 transcription procedure is as follows. The reaction components for each primer were heated at 37°C for 2 hours in a water bath, then quenched with equal volume 2x formamide buffer (95% formamide, 20mM EDTA). The samples were run on a 10% denaturing

PAGE gel, then RNA bands were located via UV-shadowing. The bands were excised with a razor blade and extracted overnight into 1xTEN250 buffer [10mM Tris (pH 7.5), 1mM EDTA, 16 250 mM NaCl]. RNA was then ethanol precipitated and resuspended in 1xTE buffer [10mM Tris

(pH 7.5), 1mM EDTA] then stored at -20°C until use. The concentration of the RNA was determined via NanoDrop UV-Vis Spectrophotometer. For PKR activation assays, RNA is renatured at 95˚C for 2 min followed by room temperature for 10 min.

2.3.3 PKR Activation Assay

Because PKR is purified from E.coli in its phosphorylated form, dephosphorylation of

WT PKR is done using lambda protein phosphatase (λPP). λPP and PKR are incubated at 30°C for 1 hr, and λPP is then inhibited using 2mM sodium orthovanadate. RNAs were diluted 4-fold from 20 μM stocks using 1xTEN100 buffer [10mM Tris (pH 7.5), 1mM EDTA, 100mM NaCl] and renatured at 90°C for 2 min. Final concentration 0.8 μM PKR was incubated with various

RNAs for 10 min at 30°C. Reactions were in standard PKR activation buffer [20 mM HEPES

(pH 7.5), 4 mM MgCl2, 50 mM KCl] and 1.5 mM DTT, 100 M ATP (Ambion) plus 15 Ci [γ-

32P]-ATP. The samples were fractioned using a 10% Bis-Tris (Novex) SDS PAGE gel, dried and exposed to a phosphorimager screen (Molecular Dynamics). The phosphorylated bands were detected using a Typhoon Phosphorimager and quantified using ImageQuant software. Each assay had a negative control of 1xTEN100 buffer and a ds79 RNA positive control at 0.1μM for normalization.

For assays performed with variable magnesium concentration, Mg2+ was added separately and not in the PKR activation buffer. Following renaturation of RNA, free 10x Mg2+ was added and incubated with the RNA for 5 min at 30˚C. Upon addition of PKR, incubation times were extended to 20 min for all trials to allow sufficient reaction time for lower [Mg2+]. 17 2.4 Results and Discussion

2.4.1 3’-Tail Activation of PKR by the Twister Ribozyme

The purpose of the following experiments was to study the ability of the Clostridia bolteae twister ribozyme to regulate the innate immune sensor PKR. As discussed in Section 2.2, the twister ribozyme is a recently discovered class of self-cleaving ribozymes. The small, 61 nucleotide core, structurally compact nature of the ribozyme would not traditionally be thought to activate PKR. Therefore, we hypothesized there might be an inhibitory effect if only the core was incubated with PKR. However, in nature ribozymes are part of a larger native construct.

When the twister motif was first identified, it was found to be present in many species of eukarya

(fungus, insects, plants, flatworm, cnidarian, fish) and bacteria. However, few mechanistic studies were performed with bacterial twister ribozymes. Additionally, the mechanistic studies performed were done with bimolecular constructs. A mutation study was performed on a bimolecular twister complex derived from Clostridia bolteae.7 Bacteria of the Clostridia class can lead to a number of human diseases including gangrene, food poisoning, tetanus, and botulism.11 Further, the C. bolteae bacterium itself appears to be linked to autism, as it is found in high concentration in the intestinal tracts of autistic children.12

In order to test the active region of the C. bolteae twister ribozyme was identified and additional flanking nucleotides were added to the open stem (+5, +10, +15), resulting in four slightly varying sequences (Figure 2.3). The predicted cleaved and precleaved structures for all

RNAs are shown in Appendix A based on mFold RNA folding prediction. The flanking nucleotides were identified from the native RNA sequence. As PKR is classically known to bind dsRNA of >33 bp, the flanking region was added in anticipation that it might allow 18 determination of a minimum bp requirement for activation. Each RNA (T, T+5, T+10, T+15) was transcribed via in vitro T7 transcription, which is performed in the presence of Mg2+, which also promotes ribozyme activation. Initially, purification following transcription indicated two distinct bands. To help identify this band, a size comparison with known RNA was performed.

Interestingly, the slower band was significantly shorter than the full length predicted constructs by 7-22 nt depending on the construct. This was attributed to self-cleavage, since 7 nucleotides would be cleaved on the T ribozyme construct and remaining constructs similarly lost 7 plus their corresponding 5’extension. We reasoned that the Mg2+ present during transcription was causing the ribozyme to cleave. Therefore, rather than an extended bp region, the constructs differed in the length of their 3’ tail.

Figure 2.3 Clostridia bolteae Twister Ribozyme Constructs. Four sequences were designed containing nucleotides flanking the active region of the ribozyme. In vitro transcription conditions lead to ribozyme self-cleavage and the resulting purified sequences are shown. Structures are based on mFold RNA folding form structure predictor. 19 My graduate mentor, Dr. Chelsea Hull, had previously found that all of the bacterial hairpins she examined required flanking tails to activate PKR, with a preference for a 3’-tail.5, 13

Therefore, even though we had planned to have both the 5’ and 3’ end present and base paired, we still pursued PKR activation by varying 3’-tail length.

Initial experiments studying PKR activation by T – T+15 resulted in inconsistent activation levels. Transcription yields for all four constructs were very low, and required concentration by in vacuuo water removal. However, this caused the relative levels of salt and

EDTA present in the buffer to vary between RNA isolates. Notably, the concentrated EDTA appeared to be chelating Mg2+ required for proper folding of the twister constructs and activation of PKR (data not shown). Therefore, larger transcriptions were performed in order to regulate buffer conditions.

Three trials were performed for each RNA (T – T+15) at larger transcriptions (Figure

2.4). For standard dsRNA, maximum PKR activation occurs at ~0.1 uM RNA concentration.

However, for the C. bolteae twister ribozyme, optimal activation was shown to vary depending on flanking nucleotide length, but generally peaked around 0.6 uM RNA. In comparison with traditional activators of PKR, the twister ribozyme is much more compact and contains only 19 bp, including tertiary interactions, in the core, possibly requiring a higher RNA concentration for activation. We found that increasing the length of the 3’-tail led to higher activation of PKR.

Activation by T and T+5 was very low, below 20% maximum when normalized to 79 bp dsRNA. This indicates that not all RNAs activate PKR. However, modest activation (32%) was observed for T+10 RNA , albeit at a high concentration of RNA (3 uM). T – T+10 all exhibit the expected bell curve and some level of PKR activation, but it is unclear whether the level of activation observed is biologically meaningful. 20 Ultimately, the most significant result is the increase in activation with tail length. The

T+15 construct normalized PKR activation is 78%, which is more than double the activation of

T+10, and a much greater jump when compared to the previous addition of five nucleotides.

Essentially, the first five-nucleotide addition displayed a 1.5-fold increase in activation, followed by a 2-fold increase from T+5 to T+10 and 2.5-fold increase from T+10 to T+15. One explanation for this phenomenon may be the presence of additional base pairing in the T+15 construct in the lower stem region. The additional ds region may facilitate binding and dimerization of PKR, leading to higher activation.

21 a) T T+5

T+10

T+15

100 b) T 90 T+5 80 T+10 T+15 70

60

50

40

30

20

10

0 0.01 0.1 1 10

Figure 2.4 Increasing 3'-tail length leads to potent activation of PKR. (a) Representative gels from PKR activation assay experiments for each of the constructs studied (T-T+15) (b) PKR activation increases with 3’ tail length. Maximum activation is ~80% for T+15. Percent PKR activation was normalized to that of 0.1 μM dsRNA-79 in 4 mM Mg2+ conditions. Data is from an average of 3 trials, error bars indicate standard deviations. 22 2.4.2 Activation of T+15 Twister Ribozyme at Physiological Mg2+ Level

Following the 3’ tail extension experiments described above, a related study was performed with the strongest PKR activation twister construct, T+15. The experiments described in Section 2.4.1 were performed in 4 mM Mg2+, which is standard for PKR experiments and more comparable to the Mg2+ concentration in bacteria (~2 mM).14-15 However, Mg2+ concentration is 0.5 mM in the human cell, so when pathogenic RNAs enter the human cell it could lead to a structural change in the RNA.16 Namely, the RNA may lose some structure in the presence of lowered divalent metal ion concentration. Mg2+ is greatly important in functional

RNA folding and activity, and contributes significantly to the compact nature of the twister ribozyme.7, 14 Activation assays were performed for the T+15 twister ribozyme with 0.5 mM

Mg2+ and we found that activation was still significant, albeit two-fold lower (Figure 2.5).

Maximum activation of PKR decreased ~2-fold from 85% to 52%, and the RNA concentration of maximum activation was observed to shift ~2-fold from 1.5 M to 0.7 M. The RNA concentration value for maximum activation of the T+15 construct at 4 mM Mg2+ and the maximum activation at 4 mM Mg2+ (+ ~5%) is higher than what was observed in the experiments shown in Figure 2.4. We believe these shifts are due to the modified experimental procedure for variable Mg2+ experiments, which specifies a longer incubation period of 20 minutes for both Mg2+ concentrations as opposed to 10 minutes for previous experiments.

Overall, these results show good precedence for PKR activation by small bacterial ribozymes.

23 a)

b)

2-fold

2-fold

Figure 2.5 Activation of PKR at Physiological Mg2+Concentration. (a) Representative gel from activation assay experiments. Percent PKR activation was normalized to that of 0.1 μM dsRNA-79 in 4 mM Mg2+ conditions (b) PKR activation assay data for three trials at 4 mM and 0.5 mM Mg2+. Red arrows depict effect of shifting Mg2+ concentration from 4mM to 0.5 mM. RNA was incubated with PKR for 20 minutes (vs. 10 min in Figure 2.4)

24 2.5 References

1. Zhang, J.; Lau, M. W.; Ferré-D'Amaré, A. R., Ribozymes and Riboswitches: Modulation of RNA Function by Small Molecules. Biochemistry 2010, 49 (43), 9123-31. 2. Garst, A. D.; Edwards, A. L.; Batey, R. T., Riboswitches: Structures and mechanisms. Cold Spring Harbor perspectives in biology 3 (6). 3. Serganov, A.; Patel, D. J., Ribozymes, riboswitches and beyond: regulation of gene expression without proteins. Nature reviews. Genetics 2007, 8 (10), 776-90. 4. Hull, C. M.; Anmangandla, A.; Bevilacqua, P. C., Bacterial Riboswitches and Ribozymes Potently Activate the Human Innate Immune Sensor PKR. ACS chemical biology 2016, 11 (4), 1118-27. 5. Hull, C. M.; Bevilacqua, P. C., Mechanistic Analysis of Activation of the Innate Immune Sensor PKR by Bacterial RNA. J. Mol. Biol. 2015, 427 (22), 3501-15. 6. Tian, B.; Bevilacqua, P. C.; Diegelman-Parente, A.; Mathews, M. B., The double-stranded-RNA- binding motif: interference and much more. Nature reviews. Molecular cell biology 2004, 5 (12), 1013- 23. 7. Roth, A.; Weinberg, Z.; Chen, A. G.; Kim, P. B.; Ames, T. D.; Breaker, R. R., A widespread self- cleaving ribozyme class is revealed by bioinformatics. Nat Chem Biol 2014, 10 (1), 56-60. 8. Liu, Y.; Wilson, T. J.; McPhee, S. A.; Lilley, D. M., Crystal structure and mechanistic investigation of the twister ribozyme. Nat Chem Biol 2014, 10 (9), 739-44. 9. Ren, A.; Košutić, M.; Rajashankar, K. R.; Frener, M.; Santner, T.; Westhof, E.; Micura, R.; Patel, D. J., In-line alignment and Mg(2+) coordination at the cleavage site of the env22 twister ribozyme. Nat Commun 2014, 5, 5534. 10. Eiler, D.; Wang, J.; Steitz, T. A., Structural basis for the fast self-cleavage reaction catalyzed by the twister ribozyme. Proceedings of the National Academy of Sciences of the United States of America 2014, 111 (36), 13028-33. 11. Carol L. Wells, T. D. W., Medical Microbiology. 4 ed.; University of Texas Medical Branch at Galveston: Galveston, Texas, 1996. 12. Pequegnat, B.; Sagermann, M.; Valliani, M.; Toh, M.; Chow, H.; Allen-Vercoe, E.; Monteiro, M. A., A vaccine and diagnostic target for Clostridium bolteae, an autism-associated bacterium. Vaccine 2013, 31 (26), 2787-90. 13. Zheng, X.; Bevilacqua, P. C., Activation of the protein kinase PKR by short double-stranded RNAs with single-stranded tails. RNA (New York, N.Y.) 2004, 10 (12), 1934-45. 14. Truong, D. M.; Sidote, D. J.; Russell, R.; Lambowitz, A. M., Enhanced group II intron retrohoming in magnesium-deficient Escherichia coli via selection of mutations in the ribozyme core. Proceedings of the National Academy of Sciences of the United States of America 2013, 110 (40), E3800- 9. 15. Tyrrell, J.; McGinnis, J. L.; Weeks, K. M.; Pielak, G. J., The cellular environment stabilizes adenine riboswitch RNA structure. Biochemistry 2013, 52 (48), 8777-85. 16. Vella, F., Molecular biology of the cell (third edition): By B Alberts, D Bray, J Lewis, M Raff, K Roberts and J D Watson. pp 1361. Garland Publishing, New York and London. 1994. Biochemical Education 1994, 22 (3), 164-164. 25

Chapter 3

PKR Footprinting and Structural Analysis of Twister Ribozyme

3.1 Abstract

PKR has been shown to activate in the presence of RNAs with a minimum ~20 bp.

Therefore, the twister ribozyme, which has 23 bp, is an unlikely candidate for PKR activation, as it barely meets the base pair requirement. In Chapter 2 I showed extension of the 3’-tail on the twister ribozyme, leads to enhanced PKR activation. I also showed that under physiological

Mg2+ concentrations, activation of the highest activating, T+15 construct is still two-fold lower, but still potent. This study focuses on using structure mapping and PKR footprinting to characterize the structural interaction between PKR and the T+15 twister ribozyme, which has 19 bp, and determine how activation is so potent despite the small, compact nature of this ribozyme.

While no obvious PKR footprint was observed, most of the stem regions match the expected ribozyme structure, and PKR addition does not disrupt native tertiary structure of the ribozyme.

3.2 Introduction

PKR is a dsRBM containing protein, therefore it binds nonspecifically to dsRNA, making it difficult to fooprint. The dsRBM binds dsRNA through a sequence non-specific interaction with 2’OH in the wide, shallow minor groove.1 Ribonuclease (RNase) digestion is a commonly used technique to map the secondary structure of RNA. The most often used ribonucleases for 26 this technique are RNase V1, A, and T1, largely due to ease of availability of these nucleases.

Each cleaves unique nucleotides or regions of secondary structure, allowing for elucidation of the RNA secondary structure. RNase V1 cleaves double stranded regions, RNase A cleaves single stranded U’s and C’s, and RNase T1 cleaves single stranded G’s. When used in conjunction with RNA-binding proteins, information can be deduced about the interaction between the RNA and said protein. This is technique is referred to as protein footprinting.

Ribonuclease Structure and Mechanism

As briefly mentioned above, three major ribonucleases are utilized for RNA digestion:

RNase V1, RNase A, and RNase T1. RNase V1 was first isolated from cobra Naja oxiana venom.

This endoribonuclease is approximately 32 kDa and was shown to hydrolyze double stranded

RNA without any base preference.2-4 In one study, addition of 20 mM EDTA caused the RNase to be greatly inhibited, indicating the necessity for divalent ions for activity. It was also shown that RNase V1 will cleave tertiary interactions, single stranded stacked regions, and non- canonical base pairs, in addition to traditional Watson-Crick base pairs.5 Complete digestion produces oligonucleotides of 2-4 bases terminated at the 5’-phosphate. It is believed that RNase

3 V1 interacts with the minor grove of dsRNA, leading to the observed cleavage pattern.

RNase A, a bovine pancreatic ribonuclease, is an endoribonuclease that cleaves single stranded RNA at the 3´ end of C and U residues. Much like the alkaline breakdown of RNA, the first step of the enzymatic reaction produces a 2’, 3’-cyclic monophosphate. In the second step, the cyclic nucleotide is hydrolyzed to a 3’-nucelotide as opposed to a mixture of 2’ and 3’- nucleotides which is the result of alkaline hydrolysis.6 This breakdown is thought to proceed 27 through a concerted acid-base mechanism. RNase A is small (~13 kDa), very stable, and does not contain a cofactor, making is very useful in protein and nucleic acid research.7

Ribonuclease T1 is a fungal endonuclease that cleaves single stranded RNA after guanine residues. It is similar to RNase A in size and function, but differs in primary sequence and three dimensional structure.8 The RNA breakdown proceeds through the same intermediate and produces the same products as RNase A, but utilizes a triester-like mechanism as opposed to the classical concerted acid-base mechanism.9 These RNases, along with many others, serve as useful tools in RNA structure mapping. The nucleotides cleaved by RNase V1, RNase A, and

RNase T1 are summarized in Table 3.1.

PKR Footprinting of Vc2 Riboswitch and glmS Riboswitch-Ribozyme

Work in our lab performed by Dr. Chelsea Hull showed possible binding modes for PKR on larger functional RNAs such as the Vc2 riboswitch and glmS riboswitch-ribozyme.10

Structure mapping of the Vc2 alone was generally consistent with the available crystal structure of this RNA. Addition of the cyclic-diGMP Vc2 riboswitch ligand led to several structural changes that were consistent with published effects. Upon addition of PKR, it was found that a subset of the strengthened interactions corresponded to the tertiary structure of the riboswitch, suggesting that PKR enhances tertiary structure. While some of the observed footprints are induced upon cdiGMP binding, many are unique to PKR and indicate PKR footprints to the aptamer. Upon addition of PKR, multiple double stranded regions appeared to become protected, suggesting PKR preferably binds a specific helical region. Interestingly, mutation G83C in the

Vc2 aptameric domain that disrupts native RNA tertiary structure does not reveal a distinct 28 footprint, despite potent activation of PKR. This was thought to allow access by PKR to multiple activating helical regions, revealing no distinct footprint.

Similar experiments were performed by Dr. Hull using the glmS riboswitch-ribozyme.

Once again, structure mapping of RNA alone was consistent with the crystal structure, supporting native folding of the riboswitch-ribozyme. Upon addition of PKR, multiple protections were observed in the RNase V1 lanes, supporting PKR’s binding preference for double stranded regions. The positions of protections corresponding to PKR were found on the peripheral double stranded regions of the riboswitch.

16 bp dsRNA Vc2 Riboswitch glmS Riboswitch-Ribozyme

Figure 3.1 Footprints of PKR on Functional RNAs. Footprints of PKR are indicated by purple cylinders and shown on A-form dsRNA (left), the Vc2 riboswitch (middle), and the glmS riboswitch- ribozyme (right). Two footprints are possible for each functional ribozyme, consistent with dimerization. Pink spheres indicate protected nucleotides regardless of PKR addition and black spheres indicate PKR protected nucleotides. Image adapted from Hull, C. et al. (2016) ACS Chem Biol.

29 In each case, two footprints are found on both functional RNAs that are A-form-like and similar in length of 16 bp dsRNA (Figure 3.1). Like what is known for PKR’s interaction with double stranded viral RNA, this suggests PKR dimerizes on functional RNAs and becomes active. As native tertiary structure remains intact in both cases, these results also suggest specific tertiary structure is not needed for PKR activation. Instead, duplexed regions, like those highlighted in Figure 3.1, act as vital elements for PKR activation. 30

3.3 Materials and Methods

3.3.1 RNA Sequence Design

As described previously, twister RNA (T+15) was transcribed from hemi-duplex templates, where the bottom strand is complementary (underlined) to the T7 promoter. Bold G’s are not present in the native transcript and were added to enhance transcription efficiency.

(BS template) T+15 5’CATGGATTTCCTCCTTTCCCACTCTGCATTGATCAGGGCTTGTGACCTGCACCGGCTATAGGC CGGTGGCTGCATTAGGAAGGGGCAGCTGCCCCTGGTATAGTGAGTCGTATTAATTTC 3’

3.3.2 RNA Preparation and Purification

As described previously, in vitro T7 transcription procedure is as follows. The reaction components for each primer were heated at 37°C for 2 hours in a water bath, then quenched with equal volume 2x formamide buffer (95% formamide, 20mM EDTA). The samples were run on a

10% denaturing PAGE gel, then self-cleaved RNA bands were located via UV-shadowing. The bands were excised with a razor blade and extracted overnight into 1xTEN250 buffer [10mM Tris

(pH 7.5), 1mM EDTA, 250 mM NaCl]. RNA was then ethanol precipitated and resuspended in in 1xTE buffer [10mM Tris (pH 7.5), 1mM EDTA] then stored at -20°C until use. The concentration of the RNA was determined using the NanoDrop UV-Vis Spectrophotometer. For

PKR activation assays, RNA is renatured at 95˚C for 2 min followed by room temperature for 10 min.

31

3.3.3 RNA Structure Mapping and PKR Footprinting Experiments

[Method partially adapted from Dr. Chelsea Hull] RNAs were end labeled with [γ-32P]-

ATP using T4 PNK (NEB) and purified via denaturing PAGE. Radiolabeled RNAs were renatured with appropriate amounts of 1xTEN100 at 90°C for 2 min and room temperature for 10 min. RNA was incubated with nonspecific RNA (tRNAphe) and protein (BSA) to minimize background to final concentrations of 130 ng and 1.5 μM, respectively. RNAs were also incubated with increasing concentrations (0, 0.44, 1.75, 7 μM of PKR for 30 min at room temperature RNA was subjected to limited digestion by single stranded specific nucleases

(RNase T1, RNase A) and double stranded specific nuclease (RNase V1) under native conditions in structure buffer [20 mM HEPES (pH 7), 100 mM NaCl, and 4mM MgCl2]. Nuclease experiments were performed at various times and concentrations. To generate a hydrolysis ladder, labeled RNA was incubated in Na2CO3/NaHCO3 and 2 mM EDTA for 8 or 10 min at

90°C. To generate a T1 ladder (all G’s), labeled RNA was incubated under denaturing conditions in 0.1 U/μL RNase T1, 18 mM Na-citrate (pH 3.5), 0.9 mM EDTA, and 6 M urea for 20 min at

50°C. All reactions were quenched using extra denaturing 2xformamide loading buffer [10 mM

EDTA, 93% formamide, 2x BPB, 2x XC, 0.1x TBE, 0.2% SDS, CAPS buffer (pH 11)]. Samples were run on a 12% denaturing PAGE, 8.3 M urea sequencing gel and phosphorylated bands were detected using a Typhoon Phosphorimager and quantified using ImageQuant software.

32 3.4 Results and Discussion

As discussed in Chapter 2, multiple versions of a unimolecular C. bolteae twister ribozyme structure varying in 3’ tail length were tested for activation by RNA-activated protein kinase (PKR). Of these ribozyme variants, the structure with the longest 3’-tail addition, T+15, most potently activated PKR. Therefore, further structural studies were performed to elucidate the reasoning behind the heightened activation and confirm the active site structure remains intact despite the tail addition.

We hypothesized the significantly increased activation for the T+15 construct was due to additional base pairing in the stem. The additional nucleotides could potentially allow the tail to fold back on itself and form three additional base pairs. This hypothesis was supported by mFold secondary structure predictor whereby the lowest free energy secondary structure, with intact psuedoknots, for the T+15 sequence contained three base pairs in the lower stem of the ribozyme.11 The free energy associated with this structure is -23.7 kcal/mol, and although mFold cannot predict pseudoknots, the nucleotides involved in the critical pseudoknot interaction on the twister ribozyme are not tied up in base pairing in this structure. Two additional structures, both with a slightly more favorable free energy of -24.8 kcal/mol, were also proposed but these structures tie up the nucleotides involved in the psuedoknot interaction and presumably would not allow the native structure to form. To confirm the existence of this stem interaction, and to further characterize the specific interaction between PKR and the T+15 twister ribozyme, we performed structure mapping and PKR footprinting experiments. By structure-mapping the

RNA, any changes in structure due to PKR binding might be seen. We utilized ribonuclease digestion in order to study the RNA structure, a summary of each type of digestion is shown in

Table 3.1 below. 33 Table 3.1: Nuclease/Ladder Cleavage Sites

Digestion Type Cleaved Nucleotides OH ladder Every nucleotide

Denaturing T1 All G’s

RNase T1 Single stranded G RNase A Single stranded U & C

RNase V1 Double stranded regions Summary of digestion experiments via RNases or base hydrolysis and corresponding cleaved nucleotides as utilized in structure mapping experiments. The experiments were performed on the T+15 construct and are described in greater detail in the

Materials and Methods above. BSA and tRNA are added to each experiment to prevent potential

RNase contamination from nonspecifically cleaving the T+15 construct. The RNA was tested with and without PKR in order to see any structural changes from the addition of PKR.

Figures for each sequencing gel can be found in Appendix B. The first structure mapping experiment was performed using two PKR concentrations, 7 uM and 3 uM, in order to observe any gradual change in band strength. 7 uM PKR follows standard PKR activation assay conditions. This change would indicate the presence of a footprint (Figure S1). Patterning is consistent with the compact, double stranded nature of the twister RNA, as many bands are visible in the RNase V1 lanes. However, there appeared to be contamination of RNase A in the

RNase V1 lanes. Since RNase A cleaves single stranded regions and V1 cleaves double stranded regions, they should not have similar patterning.

Next I looked at effects of PKR on the cleavage pattern. If PKR disrupts base-pairing interactions, the bands in the V1 lane should lighten in the presence of PKR as the region becomes single stranded. While there appears to be some change in intensity of the bands in

Figure S1, the RNase A contamination makes it unclear whether disruption is actually occurring. 34 This experiment was repeated with new RNase aliquots and is shown in Figure 3.3. Here, the radiolabeled RNA had higher counts, which lead to a large background signal and the PKR lanes all appeared to have high amounts of RNA degradation. Further, enzyme concentration needed to be optimized as RNase T1 and RNase A lanes were under-digested (not shown) while RNase V1 lanes were slightly over-digested.

Further experiments were performed to troubleshoot and optimize enzyme conditions.

First, a new PKR aliquot was tested with the RNA to determine if the previous PKR aliquot had become contaminated. Both the T+15 and V2L aptamer region of the V. cholerae cyclic-di-GMP

(cdiGMP) riboswitch were tested to compare the relative levels of degradation. RNA degradation was markedly less for lanes with lower counts and the PKR aliquot did not seem to affect degradation significantly (Figure S2). From this experiment it was decided that by lowering counts on the sequencing gel, RNase contamination might be negligible.

To optimize enzyme experiments, reactions times and/or concentrations of each RNase was varied to either decrease or increase digestion appropriately. Two different combinations were tried for each RNase (Figure S3). The results appeared more successful than past experiments shown in Figure 3.1 and Figure S1, especially in the RNase V1 lanes.

Unfortunately, all sequencing experiments performed after this point did not yield clear gel images. Months later, the source was determined to be a result of degraded foam within the cassettes used to expose the radioactive gel to the PI screen. Our lab published a communication providing a quick, easy fix for this issue.12 However, because this was resolved months later and in the interim I proceeded to analyze the structure and PKR interaction with the previously collected structure mapping data. 35 Overall, the nuclease digestion results generally match the predicted structure of the

T+15 twister ribozyme (Figure 3.3). Base pairing in the P3 stem of the RNA was consistent with the expected structure, and both tertiary interactions were consistent. This may indicate PKR enforces native tertiary structure, as Dr. Chelsea Hull also found in when PKR footprinting the

Vc2 riboswitch. The T1 pseudoknot is cleaved by RNase V1 and indicated in Figure 3.3. In the

T2 interaction, only the two of the nucleotides (G7, C8) indicated cleavage by RNase V1. The complement nucleotides (G38, C39) were not cleaved at all, so it is not clear whether they are single or double stranded, but allows the possibility this interaction may not be present in the actual structure.

Figure 3.2 Single Stranded Stacked Loop L3 Cleaved by RNase V1. L3 stem loop in T+15 twister ribozyme is readily cleaved by RNase V1, above is a sequence-identical stem loop in a preQ1 riboswitch (Class I) aptamer bound to preQ1 (PDB ID: 2L1V). Nucleobase orientation strongly suggests stacking interaction occur in the L3 single stranded loop and provide an explanation for RNase V1 cleavage in this region. 36 In the L3 stem loop, RNase V1 readily cleaved all of the nucleotides. As RNase V1 cleaves single stranded stacked nucleotides, I used RNA Cosmoss characterization of secondary structure motifs to identify the 5’-CUAUA-3’ loop in another RNA and determine whether π- π stacking could occur. An identical stem loop is found in the preQ1 riboswitch aptamer, and there appears to be significant π- π stacking in the loop (Figure 3.2).13 Therefore, this digestion by

RNase V1 in L3 appears to agree with the proposed structure.

The largest structural inconsistency is in the P5 stem. The proposed base pairing in the P5 stem does not appear possible, since G63 and G64 were cleaved by RNase T1 and therefore single stranded. It is unclear whether wobble base pairs are forming or if enzyme concentration was too high, causing nonspecific cleavage and over-digestion. Because the twister ribozyme is so compact, it is possible the ribonucleases cannot access many areas of the ribozyme. It is also possible the RNA structure is dynamic, as many of the bands in the RNA only lanes are weak

(Figure 3.3). Therefore, it is difficult to gauge whether enzyme concentrations are too high or too low, as many areas of the ribozyme show no cleavage at all. Overall, the structure mapping of the T+15 ribozyme is consistent with the predicted structure, but upon addition of PKR no footprint is observed.

As for indicators of PKR binding, no clear pattern changes could be seen to signify PKR binding. In each case where a gradual lightening or darkening of adjacent lanes could be seen, similar differences are observed in the background control experiments (no nuclease), suggesting the trend is due to RNase contamination in PKR as opposed to changes in secondary structure.

Unfortunately, with these ambiguous results, it is unclear how PKR binds to the T+15 RNA.

However, the low-resolution cleavage of additional ds nucleotides in the P5 stem (C72-A74) may explain the increased PKR activation when compared to the T+10 construct (Figure S1). 37 Further, indications that both pseudoknot tertiary interactions remain intact following PKR addition are consistent with previous findings, which signify PKR tolerates native tertiary structure. In Chapter 4 I discuss future experiments and alternative methods of characterizing the interaction between PKR and the T+15 twister ribozyme. 38

Figure 3.3 Summary of Structure Mapping Experiments. Nucleotides cleaved by RNase T1 or A (green) and nucleotides cleaved by RNase V1 (red) in the absence of PKR are shown in the T+15 RNA sequence. Uncleaved nucleotides are shown in black. Original proposed base pairing in the lower stem is shown in grey. This gel is representative of multiple sequencing experiments, and indicates the highly double stranded nature of the ribozyme. Various regions of the T+15 ribozyme are labeled on the right of the gel. Nucleotides 64-76 are assigned based on Figure S1. 39 3.5 References

1. Tian, B.; Bevilacqua, P. C.; Diegelman-Parente, A.; Mathews, M. B., The double-stranded-RNA- binding motif: interference and much more. Nature reviews. Molecular cell biology 2004, 5 (12), 1013- 23. 2. Auron, P. E.; Weber, L. D.; Rich, A., Comparison of transfer ribonucleic acid structures using cobra venom and S1 endonucleases. Biochemistry 1982, 21 (19), 4700-6. 3. Dhananjaya, B. L.; CJ, D. S., An overview on nucleases (DNase, RNase, and phosphodiesterase) in snake venoms. Biochemistry. Biokhimiia 2010, 75 (1), 1-6. 4. Favorova, O. O.; Fasiolo, F.; Keith, G.; Vassilenko, S. K.; Ebel, J. P., Partial digestion of tRNA- aminoacyl-tRNA synthetase complexes with cobra venom ribonuclease. Biochemistry 1981, 20 (4), 1006- 1011. 5. Lockard, R. E.; Kumar, A., Mapping tRNA structure in solution using double-strand-specific ribonuclease V1 from cobra venom. Nucleic Acids Res. 1981, 9 (19), 5125-40. 6. Markham, R.; Smith, J. D., The structure of ribonucleic acid. I. Cyclic nucleotides produced by ribonuclease and by alkaline hydrolysis. Biochem. J 1952, 52 (4), 552-7. 7. Cuchillo, C. M.; Nogués, M.; Raines, R. T., Bovine Pancreatic Ribonuclease: 50 Years of the First Enzymatic Reaction Mechanism(). Biochemistry 2011, 50 (37), 7835-41. 8. Pace, C. N.; Heinemann, U.; Hahn, U.; Saenger, W., Ribonuclease T1: Structure, Function, and Stability. Angewandte Chemie International Edition in English 1991, 30 (4), 343-360. 9. Loverix, S.; Winqvist, A.; Strömberg, R.; Steyaert, J., Mechanism of RNase T1: concerted triester-like phosphoryl transfer via a catalytic three-centered hydrogen bond. Chemistry & Biology 2000, 7 (8), 651-658. 10. Hull, C. M.; Anmangandla, A.; Bevilacqua, P. C., Bacterial Riboswitches and Ribozymes Potently Activate the Human Innate Immune Sensor PKR. ACS chemical biology 2016, 11 (4), 1118-27. 11. Zuker, M., Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 2003, 31 (13), 3406-15. 12. Bingaman, J. L.; Frankel, E. A.; Hull, C. M.; Leamy, K. A.; Messina, K. J.; Mitchell, D., 3rd; Park, H.; Ritchey, L. E.; Babitzke, P.; Bevilacqua, P. C., Eliminating blurry bands in gels with a simple cost-effective repair to the gel cassette. RNA (New York, N.Y.) 2016, 22 (12), 1929-1930. 13. Kang, M.; Peterson, R.; Feigon, J., Structural Insights into riboswitch control of the biosynthesis of queuosine, a modified nucleotide found in the anticodon of tRNA. Molecular cell 2009, 33 (6), 784-90.

40

Chapter 4

Future Directions

Our goal is to characterize and understand the interaction between PKR and bacterial

RNAs. The system described in Chapters 2 and 3 explores one type of RNA structural motif, a

3’-tail, and the implications of extending it. We have confirmed that PKR is activated due to the presence of a 3’-tail, but further characterization attempts were met with difficulties. Therefore, alternative methods could be used to understand the location of PKR binding and dimerization on the twister ribozyme.

To further understand the results described in Chapter 2, it may be useful to better mimic the cellular environment. For example, using molecular crowders to better mimic the cellular environment may lead to an increase in activation due increased collisions between molecules of interest. Alternatively, we could choose to broaden the experiment and study other bacterial

RNAs under similar conditions to determine the importance of a 3’tail in conjunction with other activating motifs and their relative necessity for activation.

To extend this idea further into a true in vivo system, PKRs preference for various RNAs and their corresponding structural motifs could be probed using CLIP (crosslinking and immunoprecipitation) technique. This method fixes RNA-protein complexes in living cells, and could be used with PKR as the protein target especially in bacterially infected cells.

Recently, my work in the research group has taken a different form. As described previously, RNA (ribonucleic acid) is traditionally known as a template for protein synthesis, but has been shown to be catalytically active like protein enzymes. Therefore, molecular structures 41 of RNA are highly useful in determining the mechanisms behind RNA catalytic function and protein interaction. However, x-ray crystallography, the most commonly used method of RNA structure determination, involves the use of various abiological methods and conditions to crystallize RNA, and may not provide an accurate representation of RNA in vivo. Therefore, our lab developed an idea to use cryo-electron microscopy, which involves the vitrification of samples prior to visualization by electron microscopy, to solve structures of smaller, catalytically active RNAs. This method has been shown to work best with large structures such as viruses and protein complexes. Therefore, my second graduate mentor Kyle Messina and I, have worked to develop an RNA construct that is both catalytically relevant and large enough to be visualized.

Unlike x-ray crystallography, cryo-EM would allow for an accurate in vitro picture of a functionally relevant RNA. Cryo-EM does not require that the RNA be dehydrated or stabilized by copious amounts of metal ions unlike x-ray crystallography. During x-ray crystallography

RNAs often form crystal-crystal contacts that induce the RNA to misfold. The vitrification process of cryo-EM circumvents this issue entirely by flash freezing the samples trapping the molecules in an in vitro-like and biologically relevant active state. Lastly, crystals of ribozymes in the pre-cleaved state require an inhibitory modification, such as a deoxy in place of the nucleophilic hydroxyl, to prevent self-cleavage during the week or longer crystal growth period.

However, such substitutions can prevent native, catalytically relevant interactions. Since cryo-

EM freezes the samples almost instantly, we can use the wild-type ribozyme with little to no modifications eliminating structural perturbations.

If we are able to successfully solve a structure of a ribozyme using cryo-EM, we would like to expand the technique to study other systems including PKR bound RNA. Despite decades of interest in PKR’s structure and function, no crystal structures exist of RNA bound PKR due to 42 its nonspecific dsRBM-dsRNA interaction and difficulty to crystallize. Hence, little is known about PKR’s domain interactions and relative orientations.1-3 Therefore, developing a high resolution structure of PKR and PKR bound RNA would provide critical information about the mechanism of PKR binding and activation.

43 Chapter 4 References

1. Dar, A. C.; Dever, T. E.; Sicheri, F., Higher-order substrate recognition of eIF2alpha by the RNA-dependent protein kinase PKR. Cell 2005, 122 (6), 887-900. 2. Nanduri, S.; Carpick, B. W.; Yang, Y.; Williams, B. R.; Qin, J., Structure of the double-stranded RNA-binding domain of the protein kinase PKR reveals the molecular basis of its dsRNA-mediated activation. EMBO J. 1998, 17 (18), 5458-65. 3. Nanduri, S.; Rahman, F.; Williams, B. R.; Qin, J., A dynamically tuned double-stranded RNA binding mechanism for the activation of antiviral kinase PKR. EMBO J. 2000, 19 (20), 5567-74.

44

Appendix A

Twister Ribozyme Predicted Secondary Structures

T T+5 T+10 T+15

Precleaved

AC A A C G U C U A G C A G C GC A U CG U C C G C G GG UG A CC U AC C C A A GA U C G G C U A A G A A 5’ G U

Cleaved G G G A A 3’

Secondary structures of precleaved and cleaved version of T-T+15 twister ribozyme. Cleavage sites are indicated with a black arrow. Structures were determined using mFold. “Cleaved” constructs match those in Figure 2.3. Bold C’s were added in error in an attempt to enhance transcription yields, C’s should have been added to the DNA primer instead than the G’s that were included. 45 Appendix B

Chapter 3 Supplementary Figures

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

23

Figure S1 T+15 Structure 22 21 Mapping and PKR Footprinting. Multiple ribonuclease digestion experiments were performed with 3uM and 7uM PKR addition. Contamination of RNase A is visible in RNase V1 lanes. High levels of background cleavage are visible around nt 21-23, and these lanes were discounted in structure mapping. 46

13 14 15 16 17 18

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

Figure S2 PKR Comparison. RNA concentration and PKR aliquots were varied to determine whether RNase contamination was present in utilized sample. 0.5x Twister refers to 2-fold dilution of 0.6 uM T+15 RNA. T+15 twister RNA samples were compared to V2L riboswitch. Changing PKR aliquot (old to new) did not appear to be an issue, and the original RNA (0.6 uM) concentration utilized appeared appropriate. 47

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Figure S3 Ribonuclease Optimization Experiments. Various concentrations, temperatures and times were tested to determine optimal conditions for enzyme digestion experiments. Red boxes indicate chosen conditions for future structure mapping and PKR footprinting experiments. All trials are RNA only and do not contain PKR.

ACADEMIC VITA

ANANYA ANMANGANDLA [email protected] Local Address Permanent Address 121 W. Fairmount Ave., Apt. 13 2574 Muirfield Way State College, PA 16801 Lansdale, PA 19446 Mobile: (215) 510-5339 Home: (215) 699-2470

EDUCATION The Pennsylvania State University, University Park, PA Anticipated Graduation: May 2017 B.S. in Chemistry Eberly College of Science

Relevant Courses: Biological Chemistry Nucleic Acids Chemistry Organic Chemistry Inorganic Chemistry Transition Metal Chemistry Analytical Chemistry

SKILLS - Proficient in analysis and operation of various spectroscopic techniques including IR, Mass Spectroscopy, 1H NMR, 13C NMR, and 2D NMR as well as EPR and Mössbauer Spectroscopy - Experienced with denaturing acrylamide and protein gel electrophoresis - Practiced in basic laboratory techniques such as titration, dilution, pipetting, etc. - Experienced with nitrogen glove boxes and Schlenk line technique - Radionuclide certified with experience using 32P ATP for experimentation and phosphorimager gel analysis

RELEVANT EXPERIENCE Undergraduate Research Assistant August 2014 – Present Bevilacqua Research Group – Pennsylvania State University University Park, PA - Perform independent biochemistry research studying the interaction between the innate immune sensor PKR and small RNAs, specifically ribozymes and development of a new method of RNA structure determination via cryogenic electron microscopy - Recipient of 3M Fellowship to conduct undergraduate research (Summer 2015) Teaching Assistant – Instrument Room September 2016 – Present Pennsylvania State University University Park, PA - Assist students taking organic chemistry lab with NMR, IR, GC, and UV/Vis data collection and analysis - Troubleshoot basic instrument errors, such as problematic background spectra, improper shimming, etc. CENTC Undergraduate Summer Research Program June 2016 – August 2016 Mayer Research Group – Yale University New Haven, CT - Performed independent research studying various nanoparticles as hydrogen acceptors for a ruthenium-based dehydrogenation catalyst - The Center for Enabling New Technologies through Catalysis (CENTC) is a National Science Foundation funded phase II center for chemical innovation Teaching Assistant - Fundamentals of Organic Chemistry Lab January 2016 – May 2016 Pennsylvania State University University Park, PA - Demonstrated proper laboratory etiquette, graded all formal reports and pre-lab assignments, and provided a positive learning environment for 14 students - Taught students how to perform techniques such as column chromatography, TLC, steam distillation, acid/base extraction, and hot recrystallization

- Assisted all experimental organic chemistry course students with various instrumentation techniques and subsequent analysis Transition Metal Chemistry – Honors Option September 2015 – December 2015 Pennsylvania State University University Park, PA - Performed 20 hours of bioinorganic laboratory work studying the Class 1c ribonucleotide reductase (RNR) - Utilized techniques such as stopped-flow, freeze-quench, EPR spectroscopy, and Mossbauer Spectroscopy Exam Grader – CHEM 450, 212, 110H Spring 2017, Fall 2014 Pennsylvania State University University Park, PA

PUBLICATIONS  Hull, C. M., Anmangandla, A., Bevilacqua, P.C., Bacterial Riboswitches and Ribozymes Potently Activate the Human Innate Immune Sensor PKR. ACS Chem. Biol. 2016, 11, 1118-1127. POSTER PRESENTATIONS  Anmangandla, A., Nanoparticle Assisted Catalytic Alcohol Dehydrogenation in an Aqueous Environment; Center for Enabling New Catalysis Annual Meeting, Seattle, WA 2016  Anmangandla, A., Activation of the Innate Immune Sensor PKR by C.bolteae Twister Ribozyme; Eberly College of Science Undergraduate Research Expedition, University Park, PA 2016 Research Experience for Undergraduates Summer Symposium, University Park, PA 2015 Central PA ACS Undergraduate Research Symposium, University Park, PA 2015

ACTIVITIES Science LionPride Member 2013 – 2015 - Facilitated science-specific tours for Eberly College of Science - Assisted in science related events such as STEM days, Exploration U; Member of 2014 CI Knoll Committee Hospitality THON Committee September 2014 - February 2015 - Assisted with all food and beverage related jobs (dancer meals, stocking fridges, concessions, etc.) during Penn State’s Dance MaraTHON raising awareness for pediatric cancer - Attended and assisted in events prior to THON (Family Carnival, 5K, etc.) Chemistry 297A – Service Learning August 2013 - December 2013 - Performed simple chemistry demonstrations for local Elementary and Middle School Events - Prepared for and attended Haunted-U, Exploration-U, and Discovery-U camps

HONORS/AWARDS Joseph A. Dixon Memorial Scholarship in Chemistry 2016-present - “To provide recognition and financial assistance to outstanding undergraduate students enrolled or planning to enroll in the Eberly College of Science at Penn State University Park who are majoring in or planning to major in Chemistry, or successor degree program(s).” 3M Undergraduate Research Fellowship May 2015 – July 2015 - Recipient of undergraduate research fellowship to conduct research at Penn State for 10 weeks in the summer of 2015 The Marion J. Eyster Scholarship Fund 2015-2016 - “To recognize and aid superior full-time junior or senior students enrolled in the College of Science with a major in Chemistry.” Eberly College of Science Undergraduate Research Support Fall 2014, Fall 2015 - Submitted a research proposal and awarded $500 for annual research expenses - Required to participate in Eberly College of Science Undergraduate Research Expedition Dalton Undergraduate Student Research Fund for Women in Sciences 2014-2016 - Recognized as an outstanding woman in science and given the opportunity to meet with Dr. Barbara Dalton, Vice-President Venture Capital, Pfizer Inc. and discuss my research goals/progress