DEAD box protein csha from staphylococcus aureus and RNA dependent RNA polymerases from viruses

Dong-Eun Kim Enzymatic Studies on Nucleic Acid Binding Proteins: DEAD-box Protein CshA from Staphylococcus aureus and RNA-dependent RNA Polymerases from Viruses Submitted by NGUYEN

Dissertation for Degree of Doctor

Supervisor: Prof Dong-Eun Kim

Enzymatic Studies on Nucleic Acid Binding Proteins: DEAD-box Protein

CshA from Staphylococcus aureus and

RNA-dependent RNA Polymerases

from Viruses

Submitted by

NGUYEN THI DIEU HANH

February, 2015

Department of Bioscience and Biotechnology

Graduate School of Konkuk University

Enzymatic Studies on Nucleic Acid Binding Proteins: DEAD-box Protein

CshA from Staphylococcus aureus and

RNA-dependent RNA Polymerases

This certifies that the Dissertation of

NGUYEN THI DIEU HANH is approved.

Approved by Examination Committee:

i

TABLE OF CONTENTS

List of Tables iv

List of Figures iii

List of Scheme vii

Abstract viii

Chapter 1 Characterization of DEAD-box protein CshA from Staphylococcus aureus on RNA substrates 1

1.1 Introduction 1

1.1.1 Staphylococcus aureus 1

1.1.2 DEAD-box protein 3

1.1.3 The role of DEAD-box protein CshA in S aureus 4

1.2 Materials and Methods 9

1.2.1 Preparation of recombinant S aureus CshA 9

1.2.2 Preparation of RNA oligonucleotides 9

1.2.3 RNA-dependent ATP hydrolysis 11

1.2.4 Duplex RNA unwinding assay 11

1.2.5 Ribonuclease assay 12

1.2.6 RNA strand annealing assay 13

1.2.7 RNA strand exchange assay 13

1.3 Results 15

1.3.1 Duplex RNA stimulates ATP hydrolysis by S aureus CshA 15

1.3.2 S aureus CshA is unable to catalyze RNA unwinding 17

1.3.3 S aureus CshA possesses ribonuclease activity 19

1.3.4 S aureus CshA is an endoribonuclease that cleaves ssRNA at preferred

sequences 23

1.3.5 S aureus CshA possesses RNA strand annealing activity 30

1.4 Discussion 36

1.5 Conclusion 41

Chapter 2 Functions of DEAD-box Protein CshA from Staphylococcus aureus on DNA substrates 42

2.1 Introduction 42

2.2 Methods 44

2.2.1 Nucleic acid substrates 44

2.2.2 DNA filter-binding assay 47

2.2.3 ATP hydrolysis 48

2.2.4 Duplex DNA unwinding assay 48

2.2.5 DNA strand exchange assay 49

2.2.6 DNA strand annealing assay 50

2.2.7 Inhibition assay of DNA strand annealing 51

2.3 Results 53

2.3.1 S aureus CshA binds to duplex DNA with overhangs 53

2.3.2 DNA strand exchange is promoted by S aureus CshA 57

2.3.3 S aureus CshA catalyzes strand annealing of complementary ssDNA into dsDNA 67

2.3.4 S aureus CshA anneals partial duplex DNAs into nicked or gapped dsDNA 74

2.3.5 CshA-mediated DNA strand annealing is inhibited by dsDNA and high NaCl concentration 78

iii

2.4 Discussion 82

2.5 Conclusion 87

Chapter 3 Efficient colorimetric assay for RNA synthesis by viral RNA-dependent RNA polymerases using thermostable pyrophosphatase 88

3.1 Introduction 88

3.2 Materials and Methods 91

3.2.1 Preparation of recombinant proteins 91

3.2.2 Pyrophosphate hydrolysis assay 91

3.2.3 Colorimetric assay of RdRp activity 92

3.2.4 Radioactive assay of RdRp activity 92

3.2.5 Inhibition assay of DMUT (3'-deoxy 5-methyl-uridine-5'triphosphate) on RdRp activity 93

3.3 Results and Discussion 93

3.3.1 Conversion pyrophosphate to inorganicphosphate by thermostable pyrophoaphatase 93

3.3.2 Comparison of the colorimetric assays to measure RdRp activity 96

3.3.3 Use of the colorimetric assay to detect RdRp activity from the crude cellular extract 97

3.3.4 Application of the colorimetric assay for RNA synthesis by RdRp to screen inhibitor 101

3.4 Conclusions 103

References 104

List of Tables

Table 1.1 Sequences of single-stranded RNAs used in this study 10

Table 1.2 Structures of double-stranded RNA substrates used in this study 10

Table 2.1 Oligonucleotides used in this study 45

Table 2.2 Structures of DNA substrates 46

v

List of Figure

Figure 1.1 Scheme of different sites of staphylococcal infections and biofilm of

Staphylococcus aureus 2

Figure 1.2 Conversed motifs of DEAD-box proteins 3

Figure 1.3 The current model of Bacillus subtilis and Staphylococcus aureus 5

Figure 1.4 Model of CshA in the degradation of the agr mRNA in the wild type and CshA mutant type 7

Figure 1.5 RNA-dependent ATP hydrolysis by S aureus CshA 16

Figure 1.6 RNA duplex unwinding by Staphylococcus aureus CshA 18

Figure 1.7 CshA degrades the single-stranded RNA region of duplex RNA substrates 21

Figure 1.8 CshA did not possess 5’-exoribonuclease activity 24

Figure 1.9 CshA is an endoribonuclease 26

Figure 1.10 Mapping of RNA cleavage sites on two different single-stranded RNA substrates (R0 and BR0) in RNA degradation by CshA 29

Figure 1.11 RNA strand annealing activity of CshA 32

Figure 1.12 S aureus CshA does not catalyze RNA strand exchange 35

Figure 1.13 Catalytic activities of S aureus CshA on RNA substrates 40

Figure 2.1 DNA binding and ATP hydrolysis by the recombinant CshA from S aureus 55

Figure 2.2 CshA shows DNA strand exchange activity rather than dsDNA unwinding 58

Figure 2.3 DNA strand exchange activity of CshA on various dsDNA substrates 63

Figure 2.4 Assay for the DNA strand exchange activity of CshA with 3′-tail

dsDNA substrate to form a forked dsDNA product 65

Figure 2.5 ssDNA strand annealing by CshA 69 Figure 2.6 ssDNA strand annealing activity of CshA to form diverse dsDNA

products 72

Figure 2.7 Formation of duplex DNAs with gap and nick is promoted by CshA 76 Figure 2.8 Inhibition of CshA-catalyzed DNA strand annealing with dsDNA

competitor and high concentration of salt 80

Figure 2.9 Plausible roles of DNA annealing and DNA strand exchange activities

by CshA in dsDNA break repair 86

Figure 3.1 Colorimetric detection of RNA polymerase activity using thermostable

pyrophosphatase 95

Figure 3.2 Comparison of RdRp activity assay methods between colorimetric assay

and radioactive assay 97

Figure 3.3 The colorimetric assay of FMDV 3Dpol RdRp from the crude cellular

extract 99

Figure 3.4 Inhibition of 3’-deoxy 5-methyl-uridine-5’triphosphate (DMUT) on the

activity of FMDV 3Dpol 102

vii

List of Scheme

Scheme 3.1 Reaction scheme for colorimetric assay of RNA polymerase 90

Abstract

Enzymatic Studies on Nucleic Acid Binding Proteins:

DEAD-box Protein CshA from Staphylococcus aureus

and RNA-dependent RNA Polymerases from Viruses

Nguyen, Thi Dieu Hanh Department of Bioscience and Biotechnology Graduate School of Konkuk University

Nucleic acids and proteins are two of the most important biomolecules in every living organism Interactions between nucleic acids (RNA and DNA) and proteins are required to play important roles in central biological processes, ranging from replication of genome, transcription, translation and recombination, in which the nucleic acid binding proteins utilize nucleic acids as substrates in the enzymatic reactions Based on their functions, nucleic acid binding proteins can be divided into several groups, including polymerases, transcription factors, nucleases and other enzymes and structural proteins In my study, two classes of nucleic acid binding proteins from bacteria and virus were chosen for enzymatic studies; DEAD-

box protein (CshA from Staphylococcus aureus) and viral RNA-dependent RNA

polymerases

ix

In the first chapter, I investigated DEAD-box protein CshA from

Staphylococcus aureus for its unusual enzymatic activities on RNA DEAD-box

proteins are an important class of proteins that are widely distributed in both prokaryotes and eukaryotes These proteins are characterized as putative RNA helicases and are involved in nearly all RNA metabolic processes, including transcription, splicing, RNA transport, ribosome biogenesis, translation, and RNA decay Recently, it has been reported that some DEAD-box proteins are known to be components of the RNA degradosome, albeit it does not cleave RNA directly To date, detailed role of the DEAD-box proteins in the RNA degradation is poorly understood In this study, I demonstrated that the DEAD-box protein CshA from the

vancomycin-resistant Staphylococcus aureus strain Mu50 possesses RNA

endoribonuclease activity and complementary RNA strand annealing activity Despite having RNA-dependent ATPase activity, CshA did not exhibit RNA

helicase activity in vitro Instead, CshA catalyzed the degradation of single-stranded

RNAs at phosphodiester bonds between adenine or uracil and any bases (A-N and U-N), and cytosine and other bases except guanine (C-A/U/C) In addition, I observed that CshA possesses RNA strand annealing activity, which converts complementary single-stranded RNA substrates into double-stranded RNA duplexes Thus, I suggest that the endoribonuclease and RNA strand annealing activities of the DEAD-box protein CshA may contribute to RNA remodeling in the bacterial RNA degradosome To my knowledge, this study is the first to report that a DEAD-box protein from a pathogenic bacterium is implicated in multiple ATP-independent activities on RNA, such as annealing and degradation

In the second chapter, I examined enzymatic activities of CshA on DNA substrates I observed that CshA has two ATP-independent activities; annealing of complementary single-stranded DNA (ssDNA) and strand exchange on short double-stranded DNA (dsDNA) These DNA strand annealing and exchange activities are independent of Mg2+ or ATP binding and hydrolysis CshA binds to dsDNA containing diverse end structures with various affinities: forked dsDNA > tailed dsDNA > blunt-end dsDNA The rate and efficiency of CshA-catalyzed ssDNA annealing and DNA strand exchange is negatively correlated with the binding affinities of CshA to the dsDNA product ssDNA annealing activity with tailed dsDNA substrates as well as versatile DNA strand exchange activity of CshA suggests a possible role in dsDNA break repair processes

In the last chapter, I developed a novel method for assaying the RNA synthesis activities of RNA-dependent RNA polymerase (RdRp) RdRp is another class of nucleic acid binding protein, which is essential for the replication of the RNA genome-containing positive-strand RNA viruses To screen potential drug candidates against RNA viruses, a simple method to detect activity of RdRp is necessary I developed a simple colorimetric assay to quantify the RNA synthesis activity of RdRp by measuring the pyrophosphates released during nascent RNA synthesis RNA polymerase reaction was quenched by heating at 70 °C for 5 min, during which thermostable inorganic pyrophosphatase converted the accumulated pyrophosphates into inorganic phosphates Subsequently, the amount of inorganic phosphate was measured using a color-developing reagent Using RdRps from hepatitis C virus and the foot-and-mouth disease virus, I demonstrated that this

xi

colorimetric assay facilitates the measurement of RNA polymerase activity

Taken together, I have studied and investigated detailed biochemical enzymatic properties of two nucleic acid binding proteins from bacteria and viruses

My studies on enzymatic activities of the DEAD-box protein CshA for DNA and

RNA will contribute for development of antibiotics against S aureus, because the

DEAD-box protein CshA is indispensable for bacterial pathogenesis In addition, I have developed a simple and efficient assay method for detection of viral RdRp activity This assay method will facilitate high-throughput screening of antiviral drug candidates against several pathogenic RNA viruses such hepatitis C virus (HCV) and foot-and-mouth disease virus (FMDV)

Key words: DEAD-box protein, CshA, Staphylococcus aureus,

endoribonuclease, RNA-dependent RNA polymerase, colorimetric assay

Chapter 1 Characterization of DEAD-box protein

CshA from Staphylococcus aureus on RNA substrates

1.1 Introduction

1.1.1 Staphylococcus aureus

Staphylococcus aureus is known as at Gram-positive bacterial pathogens that

causes wide spectrum of diseases from innocuous skin to severe life-threatening

infections S aureus is a prominent infectious bacterium that causes nosocomial and post-surgical wound infections S aureus can cause minor infections as food

poisoning or develop into dangerous diseases such as septicemia, endocarditis,

pneumonia, toxic-shock syndrome and others (Fig 1.1A) These S aureus infections may develop rapidly and are difficult to treat S aureus can produce

biofilm which is an aggregate of bacteria embedded in a self-produced extracellular

matrix The ability of biofilm formation of S aureus offers this bacterium a stable

environment and protects them against extreme conditions as well as antibiotic treatment Therefore, the presence of biofilm is a severe problem in the treatment of

a medical environment, as they can colonize catheters or surgical devices S aureus

infections are the most common cause of hospital-acquired infections, thus the emergence of hypervirulent and multidrug resistant strains represent a real healthcare problem in hospital environments (Fig 1.1)

2

A

B

Figure 1.1 Scheme of different sites of staphylococcal infections (A) and

biofilm (B) of Staphylococcus aureus (Bar, 20μm) [1, 2]

1.1.2 DEAD-box protein

DEAD-box proteins are an important class of proteins that are widely distributed in both prokaryotes and eukaryotes These proteins are characterized as putative RNA helicases and are involved in nearly all RNA metabolic processes, including transcription, splicing, RNA transport, ribosome biogenesis, translation, and RNA decay [3-5] DEAD-box proteins often contain nine conserved amino acid motifs; the DEAD motif itself is composed of four conversed amino acids (Asp-Glu-Ala-Asp) DEAD-box proteins possess numerous RNA-dependent activities such as RNA binding, RNA-dependent ATP hydrolysis, and ATP-dependent RNA unwinding Because of their important roles in RNA metabolism, the functions of diverse DEAD-box proteins in cellular processes have been widely investigated (Fig 1.2)

Figure 1.2 Conversed motifs of DEAD-box proteins [4]

4

1.1.3 The role of DEAD-box protein CshA in S aureus

In S aureus, two open reading frames were identified to encode putative

DEAD-box proteins and one of them is CshA DEAD-box protein CshA is involved

in biofilm formation and cold adaptation An S aureus strain mutant for CshA

displayed a cold-sensitive phenotype, with complete growth inhibition at room temperature [6] Microbial biofilm formation is an important determinant of chronic infection in humans and is involved in a wide variety of staphylococcal infections in the body [7] Biofilm formation increases antibiotic resistance and bacterial growth under extreme conditions such as high temperature, high salt concentration, UV radiation, and acidic conditions [8-10] CshA has also been identified as a potential RNA helicase component of the RNA degradosome in bacteria: CshA interacts with components of the RNA degradosome from the gram-positive model organism

Bacillus subtilis and from S aureus, and the S aureus CshA interacts with

phosphofructokinase, enolase, RNase Y, and RnpA, which is a protein subunit of RNase P (Fig 1.3) However, detailed biochemical characteristics of CshA activity

on RNA substrates are still unknown

A

B

Figure 1.3 The current model of Bacillus subtilis (A) and Staphylococcus

aureus (B) degradosome-like complex [11, 12]

6

The biofilm formation of S aureus offers bacteria almost impossible to eradicate by the immune system and antibiotic treatment The attachment to various surfaces of biofilm depends on the expression of bacterial surface encoded proteins, whereas the dispersal mode of growth that relies on the secretion of proteins such as hemolysins and proteases A quorum-sensing system, agr, was used to regulate the switch from adhesive to dispersal behavior in S aureus [13] and cshA gene encoded DEAD-box CshA protein cshA is genetically upstream of agr and participate into a balance of agr mRNA abundance [13, 14] Model of CshA in the degradation of the agr mRNA showed CshA is able to degrade a significant portion of the agrBDCA mRNAs, therefore, the quorum sensing system is working correctly and only small amounts of RNAIII is produced to stimulate low hemolysis and normal biofilm formation On the contrary, the agrBDCA mRNAs was not degrade correctly in absence of CshA, leading to a much higher level of RNAIII levels, strongly stimulates production of hemolysins and inhibits the production of biofilm components (Fig 1.4)

Figure 1.4 Model of CshA in the degradation of the agr mRNA in the wild

type (A) and CshA mutant type (B) [14]

8

To provide further understanding of the roles of DEAD-box proteins in

nucleic acid metabolism, I investigated a DEAD-box protein from Staphylococcus

aureus strain Mu50 Isolated in 1997, Mu50 was one of the first methicillin-resistant

S aureus strains reported to have reduced susceptibility to vancomycin [15, 16]

Basic Local Alignment Search Tool (BLAST) protein searches of the S aureus

Mu50 genome database have identified two open reading frames (one with 506 amino acids termed CshA and the other with 448 amino acids termed CshB) that encode putative DEAD-box proteins predicted to be ATP-dependent RNA helicases [17, 18]

In this study, I characterized the enzymatic activities of CshA from S aureus

Mu 50 on RNA substrates As a putative ATPase-dependent RNA helicase, CshA was first examined for ATP hydrolysis and duplex RNA unwinding activity with various RNA substrates Surprisingly, CshA exhibited no duplex RNA unwinding activity However, CshA was found to have RNA degradation and RNA annealing activities These unique enzymatic activities may account for the possible roles and functions of CshA in the RNA degradosome complex as well as in RNA metabolic processes in bacteria

1.2 Materials and Methods

1.2.1 Preparation of recombinant S aureus CshA

Recombinant S aureus CshA was provided by Professor Young-Seok Heo,

Department of Chemistry, Konkuk University

1.2.2 Preparation of RNA oligonucleotides

The single-stranded RNA fragments listed in Table 1.1 were chemically synthesized by Cosmo Genetech (Seoul, Korea) Oligonucleotides indicated with an

asterisk (see Table 1.1) were synthesized by in vitro transcription using T7

polymerase, as described previously [19] Some of the RNA oligonucleotides were labeled with 32P at the 5′ end using T4 polynucleotide kinase (10 U, Takara, Tokyo, Japan) and 1 L of [-32P] ATP (3,000 Ci/mmol, GE Healthcare, Piscataway, NJ, USA) in 20 L of reaction buffer containing 50 mM Tris-HCl, pH 8.0, 10 mM MgCl2, and 5 mM dithiothreitol (DTT) at 37 °C for 1 h The labeled RNA strands were subsequently purified via phenol/chloroform extraction and subsequent ethanol precipitation RNA duplexes (shown in Table 1.2) were prepared by annealing two RNA oligonucleotides as follows: the mixture of complementary oligonucleotides was heated at 85 °C for 5 min and then cooled slowly to room temperature

1.2.3 RNA-dependent ATP hydrolysis

CshA (1 M) was added to a reaction mixture containing 50 mM Tris-HCl,

pH 7.5, 25 mM NaCl, 5 mM MgCl2, 2 mM DTT, and 200 M ATP spiked with 0.6

Ci of [-32P] ATP (3,000 Ci/mmol, GE Healthcare) in the presence or absence of 1

M of various RNA substrates (single-stranded RNAs [ssRNAs]: R1, R0, and R4; double-stranded RNAs [dsRNAs]: blunt-ended 47R/47R, 5′-tail 47R/25R2, and 3′-tail 47R/25R1) The reaction mixtures (20 L) were incubated at room temperature for 0.1, 0.5, 1, 2, 5, 10, 15, 20, 30, 45, and 60 min, respectively At each designated time point, the reaction was stopped by the addition of 1 L of 4 M formic acid to each aliquot (1.5 L) of reaction mixture The quenched reaction aliquot (1 L) was blotted on polyethyleneimine cellulose for thin-layer chromatography (Macherey-Nagel, Düren, Germany) and developed in 0.4 M potassium phosphate, pH 3.4 Unreacted ATP and product Pi were separated and quantified using a phosphor

imager and OptiQuant/Cyclone software (Packard Instrument Co.)

1.2.4 Duplex RNA unwinding assay

For the duplex RNA unwinding assay, 100 nM CshA and 10 nM of each 32labeled duplex RNA substrate (blunt-ended 25R/25R, blunt-ended 47R/47R, 3′-tail 47R/25R1, 5′-tail 47R/25R2, two-overhang 47R/25R3; see Table 2) were mixed in a buffer containing 50 mM Tris-HCl (pH 7.5), 25 mM NaCl, 2 mM DTT, 0.15 mg/mL BSA, 5 mM MgCl2, and 1 M trap RNA (unlabeled RNA strand used to anneal to the displaced strand of duplex RNA substrates), in the presence or absence of 5 mM ATP The duplex RNA unwinding reactions were incubated at room temperature for

P-12

15 min, and were stopped by adding an equal volume of the quenching buffer (100

mM EDTA, pH 8.0, 0.4% SDS, 20% glycerol, 0.1% bromophenol blue, and 0.1% xylene cyanol) The dsRNAs and ssRNAs were resolved by native (urea-free) PAGE (15 %) Control size markers corresponding to unwound products were produced by heating each duplex substrate with 600-fold of trap RNA at 95 °C Radioactivity was quantified using a Cyclone phospho imager (PerkinElmer) and analyzed using OptiQuant/Cyclone software (Packard Instrument Co.)

1.2.5 Ribonuclease assay

For the ribonuclease assay, 0.25 M CshA was mixed with 30 nM of 32labeled ssRNA or dsRNA substrates in a buffer containing 50 mM Tris-HCl (pH 7.5), 25 mM NaCl, 2 mM DTT, 0.15 mg/mL BSA, and 5 mM MgCl2 at room temperature After various designated time points, the reaction was stopped by adding of an equal volume of the quenching buffer used in the RNA unwinding assay The quenched reaction mixtures were loaded onto a 15% polyacrylamide/8 M urea denaturing gel and resolved by PAGE Radioactivity was visualized using a Cyclone (PerkinElmer) and analyzed by OptiQuant/Cyclone software Size markers for RNA cleavage was prepared by degrading the ssRNA substrate by alkaline hydrolysis or by using RNase T1, which cleaves ssRNA at guanosine residues [20]

P-For the alkaline hydrolysis of ssRNA, 60 nM of the 5ʹ-end–labeled ssRNA substrate (47-mer) mixed with 25 mg/mL yeast tRNA (Invitrogen, CA, USA) was incubated

in alkaline hydrolysis buffer (Ambion, Austin, TX, USA) to create a pool of RNA ladders which was heated to 95 °C for 10 min before use A guanosine residue-specific size marker was generated by degradation of 5ʹ-labeled ssRNA substrate

with RNase T1 (Ambion) as follows: 0.5 U RNase T1 and 30 nM of the labeled ssRNA substrate were incubated at room temperature for 5 min, and then the reaction was stopped by heating at 85 °C for 5 min.

1.2.6 RNA strand annealing assay

In a standard RNA strand annealing reaction mixture, 1 nM CshA was mixed with 0.1 nM of 32P-labeled ssRNA and 0.25 nM of unlabeled complementary ssRNA in a Mg2+-free buffer containing 50 mM Tris-HCl (pH 7.5), 25 mM NaCl, 2

mM DTT, 10 mM EDTA, and 0.15 mg/mL BSA The reaction mixture was held at room temperature for various incubation times (0.1, 0.5, 1, 2, 5, and 10 min) The reaction was then stopped with the quenching buffer used in the RNA unwinding reaction Annealed products and ssRNA substrates were resolved by nondenaturing PAGE (15%) run at 80 V, detected by using a Cyclone phospho imager (PerkinElmer), and analyzed using OptiQuant/Cyclone software

1.2.7 RNA strand exchange assay

For the RNA strand exchange assay, 1 M CshA was mixed with 10 nM 32labeled duplex RNA (47R/25R3; prepared by annealing unlabeled R0 and labeled R3) and 500 nM of unlabeled ssRNA (R3) in a Mg2+-free buffer containing 50 mM Tris-HCl (pH 7.5), 25 mM NaCl, 2 mM DTT, 10 mM EDTA, and 0.15 mg/mL BSA The reaction mixture was incubated at room temperature for increasing lengths of time (0.1, 1, 5, 10, 20, and 30 min) and then stopped with the quenching buffer used

P-in the RNA unwP-indP-ing reaction The reaction products were resolved on 15% nondenaturing polyacrylamide gels, visualized using a Cyclone phosphor imager

14 (PerkinElmer), and analyzed using OptiQuant/Cyclone software

1.3 Results

1.3.1 Duplex RNA stimulates ATP hydrolysis by S aureus CshA

The S aureus DEAD-box protein CshA is predicted to be an ATP-dependent

RNA helicase It was thus necessary to determine whether CshA contained the presumed ATP hydrolysis activity and whether the type of RNA substrate used would affect the ATP hydrolysis activity To address these questions, ATP hydrolysis

by CshA was examined in the presence or absence of several ssRNAs of different lengths or dsRNAs with different terminal structures Then, the time course of ATP hydrolysis by CshA in the presence of each RNA ligand was quantified (Fig 1.5) The results show that ATP hydrolysis by CshA is significantly stimulated by dsRNA ligands Differences in RNA duplex structures slightly affected the ATPase activity Although both dsRNA ligands (47R/25R1 and 47R/25R2) contained a 25 bp duplex region and a 22 nt single-stranded tail, the position of the single-stranded tail (i.e., at the 5ʹ end or the 3ʹ end of the duplex) affected ATP activity: ATP hydrolysis by CshA was more efficient in the presence of dsRNA with a 5ʹ tail (47R/25R1) than in the presence of dsRNA with a 3ʹ-tail (47R/25R2; 60 min incubation) Interestingly, blunt-ended dsRNA, which is known to be an unfavorable substrate for DEAD-box proteins, enhanced ATP hydrolysis by approximately 6-fold as compared with ATP hydrolysis in the absence of an RNA ligand In contrast, the effect of ssRNAs on ATP hydrolysis by CshA was insignificant, although ATPase activity was slightly stronger in the presence of long ssRNA ligands than in the presence of short ssRNAs This observation could be explained by the formation of hairpin structure

or self-complementarity in long ssRNA substrates

16

Figure 1.5 RNA-dependent ATP hydrolysis by S aureus CshA Time course

of RNA-dependent ATP hydrolysis by CshA in the presence or absence of various RNA ligands ATP (200 M) spiked with [-32P] ATP was incubated with 1 M of CshA for various time points with or without RNA ligand (1 M) at room temperature Products of ATP hydrolysis were analyzed on polyethyleneimine-cellulose thin-layer chromatography and quantified by measuring radioactivity The progress of ATP hydrolysis by CshA is represented in the graph as concentration of ATP hydrolyzed

1.3.2 S aureus CshA is unable to catalyze RNA unwinding

The RNA-dependent ATPase activity of CshA led us to test whether CshA could catalyze RNA duplex unwinding Several types of duplex RNA substrates were prepared (see Table 1.2): blunt-ended 25 bp duplex RNA (25R/25R) and 25 bp duplex RNA with a 22 nt tail at the 3ʹ end (47R/25R1), the 5ʹ end (47R/25R2), or 11

nt tail at both ends (47R/25R3) The unwinding reactions were carried out by incubating CshA (100 nM) with each 32P-labeled duplex RNA substrate (10 nM) supplemented with the trap RNA (1 μM) in the presence or absence of ATP (5 mM)

If CshA possessed RNA helicase activity, 32P-labeled ssRNA should be displaced from duplex RNA in the presence of ATP However, the results on the native gel indicated that CshA was unable to displace the 32P-labeled ssRNA from any type of duplex RNA (Fig 1.6) Control experiments in the absence of CshA or ATP showed

no unwinding, indicating that the 25 bp duplex is not spontaneously unwound under our assay conditions In addition, CshA had no catalytic activity on blunt-ended dsRNA substrate in the presence of ATP However, CshA produced blunt-ended RNA duplexes instead of ssRNA as the unwinding product when incubated with dsRNA substrates containing overhangs (5ʹ, 3ʹ, or both) in the presence or absence

of ATP The fact that the same products were formed in the unwinding reactions regardless of the presence of ATP prompted us to hypothesize that CshA degrades the single-stranded regions in the RNA duplexes and does not affect the blunt ends

18

Figure 1.6 RNA duplex unwinding by Staphylococcus aureus CshA

Representative PAGE image of duplex RNA unwinding by CshA with various RNA substrates Each 32P-labeled duplex RNA substrate (10 nM) and 1 M unlabeled single-stranded RNA as a trap (i.e., R1 for 25R/25R, R1 for 47R/25R1, R2 for 47R/25R2, and R3 for 47R/25R3) were incubated with 0.1 M CshA in the presence or absence of ATP (5 mM) for 15 min at room temperature The reaction products were analyzed with 15% nondenaturing PAGE Asterisks indicate the positions of radiolabels

1.3.3 S aureus CshA possesses ribonuclease activity

The finding that duplex RNAs with overhangs treated with CshA migrated to the same position in the gel as blunt-ended dsRNA led us to investigate whether CshA possessed ribonuclease activity I tested degradation of the same duplex RNA substrates used in the unwinding assay, except that the ssRNA was labeled with 32P

at the 5ʹ end Half of the reaction mixture was loaded on a nondenaturing polyacrylamide gel and the other half on a denaturing polyacrylamide gel If CshA preferentially degraded the single-stranded regions of the dsRNA substrates, the resulting products would be shortened RNA fragments of blunt-ended dsRNA; the nondenaturing PAGE would retain the RNA fragments as duplexes, while the denaturing PAGE would dissociate the duplex RNA fragment into ssRNA components (Fig 1.7A) Analysis using both nondenaturing and denaturing PAGE, shown in Fig 1.7B and 1.7C, respectively, revealed that no RNA degradation products formed when the blunt-ended dsRNA substrate (47R/47R) was used In contrast, when dsRNA substrates containing 5ʹ overhangs (47R/25R2 and 47R/25R3) were incubated with CshA, the protein cleaved the single-stranded region of RNA duplexes to form short RNA fragments (Fig 1.7B and C) Additionally, when RNA duplexes with 3ʹ overhangs (47R/25R1) were used, the size of the reaction products separated on nondenaturing PAGE corresponded to that of the blunt-ended RNA duplex 25R/25R (Fig 1.7B) and was resolved into 25-mer ssRNA on the denaturing polyacrylamide gel (Fig 1.7C) These results demonstrate that CshA can cleave the single-stranded region of an RNA duplex to form blunt-ended dsRNA, but it cannot degrade blunt-ended dsRNA Although the same pattern of RNA degradation was

20 observed in reactions with or without ATP, CshA seemingly displayed enhanced cleavage of RNA substrates in the presence of ATP

Figure 1.7 CshA degrades the single-stranded RNA region of duplex RNA

substrates (A) Schematic diagram of ribonuclease assay with various 32P-labeled RNA duplexes and analysis of degraded RNA products with either nondenaturing or denaturing PAGE Reactions were carried out by incubation of 10 nM 32P-labeled duplex RNA substrates with 0.1 M CshA in the presence or absence of 5 mM ATP

22

for 15 min at room temperature Subsequently, the reactions were divided into 2 parts One part was analyzed on 15% nondenaturing PAGE; the other part was analyzed on 10% denaturing (8 M urea) PAGE (B, C) Representative nondenaturing and denaturing PAGE analysis of degraded RNA products from various RNA duplexes, respectively Asterisks indicate the positions of radiolabels

1.3.4 S aureus CshA is an endoribonuclease that cleaves ssRNA at preferred

sequences

Next, I investigated whether CshA degrades ssRNA by acting as an exoribonuclease or an endoribonuclease As shown above, CshA did not cleave blunt-ended dsRNA; therefore, I performed the RNase reaction with dsRNA substrate containing 5ʹ-labeled overhangs The type of RNase activity could be determined by using denaturing PAGE to analyze the type of degraded RNA products produced (e.g., mononucleotides or short RNA fragments; Fig 1.8A) I carried out the RNase reaction with 5ʹ-overhang 25 bp RNA duplexes containing 22

nt single strands with two different sequences (Fig 1.8B) CshA (250 nM) was incubated with the substrates (30 nM) for various time points, and the resulting reaction products were then analyzed on a denaturing gel The results revealed that the reaction products were not mononucleotides but RNA fragments Moreover, the RNA fragments formed from the degradation of the two different dsRNA substrates were different, although these two substrates contained single stranded regions that were of the same length This observation suggests that CshA cleaves 5ʹ-overhang dsRNA substrates in the middle of the single strand at preferred sites and thus does not have 5ʹ-exoribonuclease activity

24

Figure 1.8 CshA did not possess 5’-exoribonuclease activity (A)

Schematics of the ribonuclease assay on the 32P-labeled double-stranded RNA substrate containing a single-stranded RNA region at the 5ʹ end to test whether CshA contains 5ʹ-exoribonuclease or endoribonuclease activity (B) PAGE analysis

of RNA degradation by CshA with two RNA duplexes with 22 nt 5ʹ-overhangs Sequences of the two duplex RNA substrates used in the ribonuclease assay are shown CshA (0.25 M) was incubated with 32P-labeled double-stranded RNA substrates (30 nM) at room temperature for 0.1, 1, 2, 5, and 10 min; the reaction products were then separated on a 15% denaturing (8 M urea) PAGE gel For comparison as RNA ladders, the respective 5ʹ-end 32

P-labeled single-stranded (ssRNA; 30 nM) was mixed with RNase T1 (0.5 U) for 5 min at room temperature

I then examined whether CshA has 3ʹ-exoribonuclease activity with ssRNA substrates To test this RNase activity, CshA was incubated with 5ʹ-labeled ssRNA substrate for several lengths of time If CshA had 3ʹ-exoribonuclease activity, a set

of RNA fragments differing in size by one nucleotide size would be formed as the incubation time increased (Fig 1.9A) However, the products separated on the denaturing gel were RNA fragments of a fixed size (left panel in Fig 1.9B) For comparison, I also tested the RNase activity of another endoribonuclease, RNase T1, which cleaves single-stranded RNA after guanine residues (right panel in Fig 1.9B) The pattern of RNA cleavage observed with RNase T1 was similar to that obtained with CshA; this result indicates that CshA has endoribonuclease activity rather than 3ʹ-exoribonuclease activity

26

Figure 1.9 CshA is an endoribonuclease (A) A schematic diagram for a test

of the 3ʹ-exoribonuclease activity on 5ʹ-labeled single-stranded RNA substrate Gel pictorials are shown on the denaturing PAGE for reaction products by each enzyme activity (B) PAGE analysis of RNA degradation by CshA with ssRNA substrate (R0, 47-mer) The 5ʹ-labeled ssRNA substrate (30 nM) was incubated with CshA (0.25 M) or RNase T1 (0.5 U) for various incubation time points Reaction products were analyzed on the 15% denaturing (8 M urea) PAGE gel; P* represents 5ʹ-end labeling with 32

P