Investigation of Peptidyl-Prolyl Cis/Trans Isomerases in the Virulence of Staphylococcus

Investigation of Peptidyl-prolyl cis/trans isomerases in the virulence of Staphylococcus

aureus

A Dissertation presented to

the faculty of

the College of Arts and Sciences of Ohio University

In partial fulfillment

of the requirements for the degree

Doctor of Philosophy

Rebecca A. Keogh

August 2020

This Dissertation titled

Investigation of Peptidyl-prolyl cis/trans isomerases in the virulence of Staphylococcus

aureus

REBECCA A. KEOGH

has been approved for

the Department of Biological Sciences

and the College of Arts and Sciences by

Ronan K. Carroll

Assistant Professor of Biological Sciences

Florenz Plassmann

Dean, College of Arts and Sciences 3

ABSTRACT

REBECCA A. KEOGH, Doctorate of Philosophy, August 2020, Biological Sciences

Investigation of peptidyl-prolyl cis/trans isomerases in the virulence of Staphylococcus aureus

Director of Dissertation: Ronan K. Carroll

Staphylococcus aureus is a leading cause of both hospital and community- associated infections that can manifest in a wide range of diseases. These diseases range in severity from minor skin and soft tissue infections to life-threatening sepsis, endocarditis and meningitis. Of rising concern is the prevalence of antibiotic resistant S. aureus strains in the population, and the lack of new antibiotics being developed to treat them. A greater understanding of the ability of S. aureus to cause infection is crucial to better inform treatments and combat these antibiotic resistant superbugs. The ability of S. aureus to cause such diverse infections can be attributed to the arsenal of virulence factors produced by the bacterium that work to both evade the human immune system and assist in pathogenesis. Many of these virulence factors have redundant or overlapping functions, and consequently, vaccines and antibiotics designed to target one or a small number of virulence factors have had limited success in combatting infection. An alternative approach to targeting individual factors has been to identify global regulators or systems that affect the expression of numerous virulence factors or virulence factor activity in the cell.

Peptidyl-prolyl cis/trans isomerases (PPIases) are a family of enzymes that have been shown to regulate virulence factor activity and assist in protein secretion in 4 numerous bacteria. In addition, the deletion of certain PPIases has resulted in an attenuation of virulence, demonstrating their importance in infection. The objective of this dissertation was to characterize three PPIase proteins: PrsA, PpiB and trigger factor

(TF) encoded by S. aureus and determine if they contribute to virulence factor production, regulation and disease.

PrsA is a membrane-bound lipoprotein that has been shown to assist in protein secretion and contribute to virulence in the pathogens Listeria monocytogenes and

Streptococcus pyogenes. In S. aureus, PrsA was found to be involved in oxacillin resistance and to have PPIase activity, but little was known about the ability of the protein to contribute to infection. The data in Chapter 3 reveal that PrsA affects the activity of the major hemolytic alpha-toxin (Hla) in S. aureus, and that it interacts with multiple membrane-anchored proteins that contribute to antibiotic resistance. Despite these findings, Chapter 3 also demonstrates that PrsA is dispensable for virulence in a murine sepsis model of infection, a murine abscess model of infection, and in macrophage and nasal epithelial cell infection.

PpiB was characterized as a predicted cytoplasmic protein encoding a putative

PPIase domain in S. aureus. Initial work by our group confirmed the cellular location and

PPIase activity of PpiB and showed that a ppiB mutant had reduced hemolytic and nuclease activity. Chapter 2 of this dissertation identifies a role for PpiB in virulence, and shows that the PPIase activity of PpiB does not contribute to disease. Further characterization of how PpiB contributes to virulence is outlined in Chapter 3, where

PpiB is revealed to affect the activity of numerous virulence factors including the alpha 5 phenol-soluble modulins (αPSMs) and Hla. This project identified the first known PPIase to contribute to S. aureus infection, and identified specific virulence factors regulated by

PpiB.

TF is the best characterized bacterial PPIase that was first identified in

Escherichia coli for its ability to bind to the ribosome and assist in the chaperoning of nascent polypeptides. Numerous studies on Gram-positive pathogens have identified a role for TF in virulence-associated processes. Despite its apparent role in virulence in multiple pathogens, no work was done on S. aureus TF prior to this dissertation. In

Chapter 4, we identify a role for TF in biofilm formation and virulence in S. aureus.

The work presented in this dissertation reveals that S. aureus PPIases act through distinct mechanisms to regulate virulence factors and in some cases, contribute to infection. Identification of PPIases as novel regulators of virulence strengthens our knowledge of S. aureus virulence factor regulation and can be used to better inform future treatments.

DEDICATION

To my parents, Joanie and Paul Keogh. This dissertation would not have been possible

without your endless love, support, encouragement, and supplied wine.

ACKNOWLEDGMENTS

Ronan, there are so many lessons you taught me over the last four years, too many to include here so I will highlight a few that have been particularly impactful. First, I remember interviewing with you and thinking wow, this person doesn’t just want a student, they want me. That feeling of importance is something you have continued to foster throughout my graduate career, and every talk we had along the way helped me fight my crippling imposter syndrome. You assured me I belonged in this community and that was possibly the most important lesson of all. Second, you taught me how to be a great mentor and therefore inspired me to continue a career in academia. You gave me advice to better myself, and you asked me for my advice in return. You made me feel like my voice mattered, and that I was a colleague as well as a mentee. I strive to be that mentor, one who grows along with their students. Third, you taught me how not to do retro-orbital infections.

Next, to the Carroll lab past and present, thank you for accepting me at my best and my worst. For helping me finish the celebratory champagne and blocks of cheese and pretending my eyes weren’t puffy after I cried behind the freezer. Each of you are what made work so special to me. A special thank you to my partner in crime, Rachel Zapf. I couldn’t have asked for a better person to have there with me every step of the way. You were always there to listen, giving incredible advice both in science and in life. I can only hope I gave you the same.

To my other scientific mentors. Clay Caswell, I wouldn’t be in science if you didn’t take a chance on me at Virginia Tech. You taught me what it meant to be a 8 researcher and laid the foundation for my career and my confidence. Jimmy Budnick and

Lauren Sheehan, I am so grateful for your mentorship in the lab and in navigating my career. Also, for the pizza rolls. To my committee, Dr. Erin Murphy, Dr. Donald Holzschu and Dr. Sarah Wyatt. This dissertation would not have been possible without your advice and suggestions for improvement. Thank you for believing in me.

To my family and friends. Graduate school is hard, and I often doubted myself along the way. You never doubted me. Mom and dad, thank you for being my best friends and biggest cheerleaders. In your eyes I can do whatever I set my mind to, and your unwavering support is what gets me through the hardest of times. My fellow grad- students, you made these four years incredible. I appreciate each and every one of your friendships which I will take with me well-beyond Athens.

Finally, a thank you to my funding sources, without whom I would not have a career or a dissertation. Those sources include the National Institute of Allergy and

Infectious Disease, the Bill and Melinda Gates Foundation, the Ohio University Student

Enhancement Award, the College of Arts and Sciences Graduate Student Research Fund, and the Ohio University Donald Clippinger Fellowship.

TABLE OF CONTENTS

Page

Abstract ...... 3 Dedication ...... 6 Acknowledgments...... 7 List of Tables ...... 12 List of Figures ...... 13 Chapter 1: Peptidyl-prolyl cis/trans isomerases (PPIases) in Staphylococcus aureus ..... 15 Staphylococcus aureus ...... 15 Disease Prevalence and Presentation ...... 15 Virulence Factors and Protein Secretion...... 16 PPIases ...... 19 Function and Activity ...... 19 S. aureus PPIases ...... 21 Chapter 2: The Intracellular Cyclophilin PpiB Contributes to the Virulence of Staphylococcus aureus Independently of Its Peptidyl-Prolyl cis/trans Isomerase Activity ...... 26 Abstract ...... 27 Introduction ...... 28 Results ...... 30 PpiB is Required for Virulence in a Murine Abscess Model of Infection ...... 30 Determination of the PPIase Active Site Residues in PpiB ...... 31 Substitution of PPIase Active Sites Eliminates PPIase Activity in PpiB ...... 34 PpiB F64A Demonstrates Reduced Nuclease Refolding Ability ...... 36 PpiB is Expressed in the F64A Strain ...... 37 The PpiB F64A Strain Does Not Demonstrate Reduced Hemolytic Activity ...... 40 The PPIase Activity of PpiB Does Not Contribute to S. aureus Virulence ...... 42 Discussion ...... 44 Materials and Methods ...... 50 Strains and Strain Construction ...... 50 Bacterial Growth Conditions ...... 50 Murine Abscess Model of Infection ...... 51 10

Site-Directed Mutagenesis and Protein Purification ...... 51 Chymotrypsin-Coupled PPIase Activity Assay ...... 52 Nuclease Refolding Assay ...... 53 Western Blotting ...... 53 Hemolysis Assay ...... 54 Ethics Statement...... 54 Chapter 3: Novel Regulation of Alpha-Toxin and the Phenol-Soluble Modulins by Peptidyl-Prolyl cis/trans Isomerase Enzymes in Staphylococcus aureus ...... 57 Abstract ...... 58 Introduction ...... 59 Results ...... 63 PpiB is Required for Virulence in a Abscess and Systemic Model of Infection .. 63 PpiB is Required for Survival Inside Macrophages and Human Nasal-Epithelial Cells ...... 66 Exoproteome Analysis Reveals Greater Alterations in Secreted Protein Abundance in a ΔppiB Mutant than a ΔprsA Mutant ...... 67 A ΔppiB Mutant has Reduced PSM Production ...... 74 The ΔppiB and ΔprsA Mutants Both Display Reduced Hemolysis of Rabbit Erythrocytes ...... 78 In vivo Immunoprecipitation Identifies Greater Abundance of PpiB Target Proteins than PrsA...... 81 Discussion ...... 87 Materials and methods ...... 92 Strains and Strain Construction ...... 92 Bacterial Growth Conditions ...... 92 Murine Abscess Model of Infection ...... 93 Murine Systemic Model of Infection ...... 93 Macrophage Infection and Cell Differentiation ...... 94 Nasal Epithelial Cell Infection ...... 95 Exoproteome Analysis ...... 95 Butanol Extraction of PSMs ...... 97 Human-Erythrocyte Hemolysis Assay ...... 97 Rabbit Erythrocyte Hemolysis Assay ...... 97 Protein Immunoprecipitation Assay ...... 98 11

Reverse Transcriptase-Quantitative PCR (RT-qPCR) ...... 100 Ethics Statement...... 100 Chapter 4: Trigger factor contributes to biofilm formation and virulence in Staphylococcus aureus and cooperates with the cytoplasmic chaperone PpiB...... 103 Abstract ...... 104 Introduction ...... 105 Results ...... 109 A tig Mutant Has Reduced Biofilm Formation in S. aureus...... 109 TF is Required for Virulence in a Murine Systemic Model of Infection ...... 110 TF Does Not Contribute to Protection From Acid Challenge in S. aureus ...... 113 A ppiB/tig Double Mutant Exhibits a Defect in Cell Viability After Acid Challenge ...... 114 The Mutation of tig in a ppiB Mutant Leads to a Further Reduction in Hemolytic Activity ...... 116 There is More PpiB Produced in a tig Mutant ...... 118 Discussion ...... 120 Materials and Methods ...... 123 Strains and Strain Construction ...... 123 Bacterial Growth Conditions ...... 123 Static Biofilm Formation Assay...... 124 Murine Systemic Model of Infection ...... 124 Acid Challenge Assay ...... 125 Cell-Free Hemolysis Assay...... 125 Western Blotting ...... 126 Ethics Statement...... 126 Chapter 5: Conclusions ...... 127 PrsA...... 127 PpiB...... 129 TF and chaperone compensation ...... 133 References ...... 136

LIST OF TABLES

Page

Table 2.1. Strains and plasmids used in this study...... 55 Table 2.2. Oligonucleotides used in this study...... 56 Table 3.1. Proteins with altered abundance in ΔppiB culture supernatants...... 71 Table 3.2. Proteins with altered abundance in ΔprsA culture supernatants...... 74 Table 3.3. Proteins interacting with PrsA...... 84 Table 3.4. Proteins interacting with PpiB...... 85 Table 3.5. Strains and plasmids used in this study...... 101 Table 3.6. Oligonucleotides used in this study...... 102 Table 4.1. Strains and plasmids used in this study...... 123

LIST OF FIGURES

Page

Figure 1.1. Select virulence factors in S. aureus...... 17 Figure 1.2. Genomic analysis of S. aureus PPIase proteins...... 21 Figure 2.1. PpiB is required for virulence in a murine abscess model of infection...... 31 Figure 2.2. Identification of PPIase active site residues in PpiB...... 33 Figure 2.3. Substitution of predicted active site residues eliminates PpiB PPIase activity...... 35 Figure 2.4. PpiB increases the rate of Nuc refolding in the absence of PPIase activity. .. 37 Figure 2.5. Growth curve analysis in TSB of USA300 TCH1516 (Wild type), the ΔppiB mutant, and F64A strain...... 39 Figure 2.6. Coomassie stained gel of intracellular protein samples used in western blot analysis...... 39 Figure 2.7. PpiB expression is comparable in the wild-type and F64A substituted strain...... 40 Figure 2.8. PpiB-dependent hemolytic activity does not require PPIase activity...... 42 Figure 2.9. The PPIase activity of PpiB does not contribute to S. aureus virulence...... 43 Figure 2.10. Multiple sequence alignment of PpiB homologues...... 47 Figure 3.1. PpiB is required for virulence in a murine abscess (A) and systemic (B-I) model of infection...... 65 Figure 3.2. PpiB is required for survival in RPMI2650 nasal epithelial (A,C) and THP-1 macrophage cells (B,D)...... 67 Figure 3.3. Exoproteome analysis reveals greater alterations in secreted protein abundance in a ΔppiB mutant than a ΔprsA mutant...... 70 Figure 3.4. The αPSMs are the primary toxins responsible for human erythrocyte lysis while...... 76 Figure 3.5. Decreased αPSM production in a 횫ppiB mutant...... 77 Figure 3.6. A ppiB and prsA mutant display reduced hemolysis in rabbit blood...... 80 Figure 3.7. RT-qPCR analysis of hla and PSM transcript levels in, W.T.; ΔppiB, and ΔprsA strains...... 81 Figure 3.8. Schematic diagram of proposed mechanisms of PpiB and PrsA...... 91 Figure 4.1. A tig mutant has reduced biofilm formation in S. aureus...... 110 Figure 4.2. TF is required for virulence in a murine systemic model of infection...... 112 Figure 4.3. TF does not contribute to protection from acid challenge in S. aureus...... 114 14

Figure 4.4. A ppiB/tig double mutant has a defect in cell viability after acid challenge. 116 Figure 4.5. The mutation of tig in a ppiB mutant leads to a further reduction in hemolytic activity...... 118 Figure 4.6. There is more PpiB produced in a tig mutant...... 119 Figure 5.1. Schematic for PrsA...... 129 Figure 5.2. Schematic for PpiB...... 133 Figure 5.3. Schematic for TF and compensation...... 135

CHAPTER 1: PEPTIDYL-PROLYL CIS/TRANS ISOMERASES (PPIASES) IN

STAPHYLOCOCCUS AUREUS

Staphylococcus aureus

Disease Prevalence and Presentation

S. aureus is commonly found as a commensal bacterium in the anterior nares of humans, where it lives an asymptomatic lifestyle of adherence. Individuals can be classified into three different categories in regard to S. aureus nasal colonization with

20% being persistent carriers, 30% being intermittent carriers and 50% never having been colonized (1). Those colonized with S. aureus are at a greater risk of developing infection, as the bacteria is opportunistic, and can switch from an asymptomatic lifestyle to one of infection (2).

Given its prevalence in the population as well as its opportunistic potential, it is not surprising that S. aureus is the leading cause of hospital-acquired infections (3). The most common infections caused by S. aureus are skin and soft tissue infections (SSTIs), which account for approximately 10 million outpatient visits and 500,000 hospitalizations in the US every year (7). SSTIs often present as minor, self-limiting boils and abscesses, however, infections can progress into life-threatening exfoliative disease and necrotizing fasciitis (4–6). In addition to the skin, S. aureus can establish infection in virtually any tissue in the human body. It is the leading cause of endocarditis and lower respiratory infections, and is responsible for the second highest number of hospital-related bloodstream infections (7, 8). Infections from S. aureus can also cause meningitis, 16 osteomyelitis, and toxic shock syndrome, demonstrating the incredible diversity of this pathogen in its ability to infect humans.

Of additional concern is the rise of antibiotic resistant strains in the population, including methicillin resistant S. aureus (MRSA), which is categorized as a serious threat by the Center of Disease Control and Prevention (11). While initial drug resistant strains were isolated from hospital-acquired infections (HA-MRSA), population studies have demonstrated that approximately 90% of all MRSA infections are community associated

(CA-MRSA) and present in previously healthy individuals that have not recently been in contact with healthcare settings (12). CA-MRSA isolates are hypervirulent in comparison to HA-MRSA isolates, and can be more easily spread from host to host, making them a high priority for the development of novel effective treatments (13). This dissertation will focus on the USA300 lineage of S. aureus, as it represents the most common CA-MRSA isolates found in North America (14).

Virulence Factors and Protein Secretion

The success of this bacterium as a pathogen can be largely attributed to the vast number of virulence factors it produces and secretes, that aid S. aureus in evading the immune system and damaging the host (7, 15, 16). These virulence factors include toxins which lyse host cells, exoenzymes that degrade host molecules, adhesins that assist in the ability of the bacterium to adhere to the host, immune evasion molecules, and superantigens that elicit a dangerous overactive immune response (8, 17–20). Figure 1.1 groups some of the major virulence factors found in S. aureus based on their function.

These virulence factors have been the subject of numerous studies, with the hopes that 17 abrogating their activity would lead to a less dangerous form of the bacterium that our immune system could more readily clear (21). However, a major challenge of targeting virulence factors is that S. aureus often produces numerous molecules that have redundant or overlapping functions, and can therefore persist and compensate if one is absent. Still, coming up with novel ways to target multiple virulence factors has been a promising strategy for vaccine and therapeutic development which continues to be investigated (22, 23).

Figure 0.1. Select virulence factors in S. aureus. Exoenzymes, including proteases and nuclease can degrade host molecules such as proteins and DNA respectively. Immune evasion factors, including protein A allow the bacteria to escape clearance by the host. Adhesins allow S. aureus to adhere to host cells as well as to form multicellular biofilm communities. Superantigens elicit an overactive immune response and prevent the development of adaptive immunity. Toxins are secreted molecules which lyse cells. 18

One approach to targeting multiple virulence factors is to inhibit their secretion, as each of these proteins must exit the bacterial cell and become anchored in the cell wall, envelope, or released into the host environment to perform their respective functions (7,

24, 25). S. aureus encodes the machinery for numerous major secretion systems: (i) the twin-arginine transport system (TAT), (ii) the type VII secretion system (ESX), (iii), the general secretory secretion system (sec), and (iv) the accessory sec secretion system (26–

28). In addition, S. aureus encodes dedicated transport systems such as ABC-transporters and the Pmt system which translocates the αPSMs (29).

Although there are multiple systems, the overwhelming majority of proteins and virulence factors exit the cell via the sec secretion system. In this system, newly synthesized proteins exit the ribosome and are trafficked through the cytoplasm by a chaperone protein. These proteins are then brought to the SecYEG translocon where they are powered through the secretion channel by the ATPase SecA in an unfolded state (30).

Once outside the cell, proteins must fold into their active confirmation to be fully functional. In many bacteria, the chaperone protein SecB associates with the nascent polypeptides and brings them to the secretion machinery. Interestingly, S. aureus lacks a

SecB homolog and there is no known alternative chaperone that functions in the same way. One approach to targeting the activity of multiple virulence factors would be to identify a novel chaperone involved in this secretion process, with the hope that inhibiting that chaperone would limit the number of virulence factors that get secreted out of the cell. An alternative approach would be to identify a protein that helps virulence factors fold into their active confirmation once outside the cell, and to target that protein 19 so virulence factors have reduced activity. This dissertation focus’ on a family of proteins called peptidyl-prolyl cis/trans isomerases (PPIases) that assist in protein folding after secretion, as well as chaperoning proteins in other bacteria. The overarching goal was to determine if S. aureus produces any functional PPIase proteins and to determine if they are involved in the regulation of virulence factors and ultimately, the virulence of the bacterium.

PPIases

Function and Activity

PPIases are a superfamily of enzymes that catalyze the cis-to-trans isomerization around proline-peptide bonds (31). All amino acids with the exception of proline exist in the trans isomerization state in vivo, due to the steric hindrance of side chains in the cis confirmation. Proline, however, has a unique aromatic structure leading to reduced steric hindrance in the cis form, and thus can exist in either state with the cis confirmation occurring approximately 6.5% of the time (32). Consequently, peptidyl-prolyl bonds have low energetic preference and can slow down the process of protein folding by a factor of

100 (33). PPIase enzymes have been shown to accelerate this isomerization in vitro and in vivo, and thus directly contribute to protein folding and activity (34, 35).

PPIases are ubiquitous in nature and can be distinguished into three subclasses

(cyclophilins, FK506 binding proteins (FKBPs) and parvulins) based on their ability to bind different inhibitors. Members of each class have PPIase activity, however, there is little sequence or structural similarity between classes (35). The first cyclophilin was isolated from porcine kidneys and could be inhibited by the peptide cyclosporine (CsA), a 20 natural compound with immunosuppressant activities (36). This cyclophilin was later found to be homologous to the major human PPIase, cyclophilin A (CypA) (37). The first

FKBP family PPIase was similarly identified based on its binding to the immunosuppressant drug FK506 (38). The FKBP class of proteins was subsequently named for their ability to bind this molecule. Finally, a protein from the parvulin class of

PPIase proteins was first identified in E. coli and it was later found that this class of enzymes could be inhibited by the molecule juglone (39, 40). The respective binding of each of these inhibitors occurs at highly conserved amino acid sequences that comprise the enzymatically active region of these proteins. These regions will henceforth be referred to as PPIase domains.

Interestingly, while numerous PPIases have functional isomerase activity, most reports suggest that the PPIase domain is dispensable for overall protein function (41–

43). This is largely attributed to the modular nature of PPIase proteins. PPIases often possess additional domains with diverse functions (44–46). Most frequently, these domains facilitate protein-protein interactions or give PPIases chaperone activity.

Chaperone proteins have diverse functions in the cell including preventing protein aggregation, assisting in protein trafficking and delivery to secretion machinery, and aiding proteins under stress or starvation conditions (47, 48). The diversity in the function of PPIase proteins as well as their ubiquity in nature have led to a mounting body of work focused on characterizing these proteins. This dissertation will focus on characterizing three distinct PPIases in the bacterium S. aureus. 21

S. aureus PPIases

In bacteria, many PPIases have been characterized for their ability to assist in the folding of virulence factors and contribute to virulence (48–50). Interestingly, while some reports show the PPIase activity is necessary for virulence (51), others demonstrate that it is dispensable, and that other activities such as the ability of these proteins to act as chaperones are what contributes to virulence-related phenotypes (41, 43). Due to the presence of PPIases in all known organisms and their contribution to virulence in other bacteria, we hypothesized that S. aureus would encode PPIases that contribute to virulence. Genomic analysis revealed that S. aureus encodes three PPIase proteins, PrsA,

PpiB, and TF; one belonging to each functional class (Figure 1.2). This dissertation will examine what is known about each of these proteins and how they function in S. aureus.

Figure 0.2. Genomic analysis of S. aureus PPIase proteins. The PPIase domain of each protein is highlighted in black. PrsA also encodes a signal peptide as well as a SurA chaperone domain. TF has a ribosome-binding domain as well as a chaperone domain.

PrsA

PrsA is a membrane-anchored lipoprotein belonging to the parvulin family of

PPIases, which is the best characterized family of bacterial PPIase proteins. While parvulin-family PPIases are found in prokaryotes and eukaryotes, PrsA proteins are found in only Gram-positive bacteria, highlighting their necessity in the Gram-positive extracellular environment (35). PrsA proteins have an N-terminal secretion sequence, a conserved PPIase domain consisting of approximately 100 amino acids, and a C-terminal domain that is highly variable. S. aureus PrsA encodes a putative SurA domain at its C- terminus which is often associated with chaperone activity. In B. subtilis, prsA is essential for growth and the correct folding and activity of secreted proteins (52, 53). In contrast, there are two PrsA-like homologs in L. monocytogenes, PrsA1 and PrsA2, and neither is essential for cell viability. PrsA2 has been well-characterized in L. monocytogenes for its role in virulence, cell-to-cell spread, and the activity of numerous virulence factors including listeriolysin-O (LLO) and a phospholipase (54). There is no known role for

PrsA1 despite its 75% structural similarity to PrsA2. Work by Alonzo et al. demonstrated that the PPIase activity of PrsA2 was required for virulence, but was dispensable for virulence factor activity and they suggested an alternative role for PrsA2 as a chaperone protein (51). These results highlight the modular nature of many PPIase proteins and their diverse functions in bacteria.

S. aureus encodes one PrsA protein which has functional PPIase activity (55).

Numerous studies identify a role for PrsA in the susceptibility of S. aureus to antibiotics

(56, 57), and work by Wiemels et al. showed that PrsA affects the activity of virulence 23 factors such as secreted proteases and phospholipase C (Plc). PrsA was extensively characterized in the methicillin-sensitive S. aureus isolate HG001 where it was found to contribute to virulence, affect protein secretion and regulate the major hemolytic toxins the αPSMs (58). Despite the large body of literature on PrsA in S. aureus, whether it contributes to disease in MRSA isolates was unknown prior to this dissertation. In addition, whether there is a PPIase-independent role for PrsA in S. aureus remains elusive. Chapter 3 will further explore this protein in order to determine its role in S. aureus virulence.

PpiB

PpiB is a cytoplasmic PPIase belonging to the cyclophilin family. Cyclophilins are found in numerous organisms and vary greatly in terms of size, from relatively small single-domain proteins to large multi-domain proteins (59). The role of cyclophilins also varies between organisms with a range of findings showing they can assist in protein folding, contribute to stress tolerance, and contribute to virulence (60). Despite these findings, cyclophilins remain the most understudied class of PPIases in bacteria.

Perhaps the best studied bacterial cyclophilin is PpiB in E. coli, which contributes to biofilm formation as well as cell motility (61). A follow-up study revealed that the

PPIase activity of PpiB was necessary for its biofilm phenotype as well as the ability of the protein to interact with other proteins in the cell. Interestingly, the same study showed that the PPIase activity was dispensable for motility and that the protein contained chaperone activity that was independent of its isomerization activity (41). Collectively, these results suggest that PpiB in E. coli has multiple roles in the cell that are not limited 24 to prolyl-isomerization. Similarly, two cyclophilins encoded by Mycobacterium tuberculosis were found to exhibit chaperone activity in vivo and in vitro, where they could rescue cells from thermal shock as well as during oxidative and acidic stress (62).

Work by Wiemels et al. was the first to establish a role for the S. aureus cyclophilin PpiB in the folding of the secreted virulence factor nuclease (63). This study also demonstrated that PpiB is necessary for the ability of S. aureus to lyse red blood cells, highlighting a potential role in the bacterium’s ability to cause disease. While this study established a connection of PpiB to virulence factors, it did not explore whether

PpiB contributed to disease or whether it had PPIase activity. Chapter 2 of this dissertation is focused on characterizing the PPIase activity of PpiB and determining how the protein contributes to disease. Alternative roles for PpiB in pathogenesis are further explored in Chapters 3 and 4 where results suggests its involvement in protein secretion and chaperone activity.

Trigger factor is a member of the FKBP family of PPIases and is highly conserved in bacteria. It was first identified in E. coli as a cytoplasmic chaperone protein consisting of three domains (64). The N-terminal domain binds the ribosome, and associates with nascent proteins after they are translated. The central portion encodes its

PPIase domain that has been shown to have functional PPIase activity, though Kramer et al. demonstrated that the PPIase activity of TF is dispensable for cytosolic protein folding

(43). Finally, the C-terminal end of TF encodes a SurA-like chaperone domain and helps to traffic nascent proteins through the cytoplasm (46). 25

While numerous papers have studied the structure and folding dynamics of TF in

E. coli, the only studies that have established a role for TF in virulence have been performed in Gram-positive bacteria (35, 64–66). In the oral pathogen Streptococcus mutans, a trigger factor homolog was shown to be a regulator of stress tolerance, biofilm formation, and competence, which are all critical components of the pathogen’s ability to colonize and cause infection (67). In Streptococcus pyogenes, the secretion and processing of the major virulence factor SpeB was shown to be dependent on functional trigger factor in the cell (68). The deletion of tig in Streptococcus suis resulted in the suppression of virulence genes as well as defects in host-cell adherence, hemolysis and resistance to stress (69). Additionally, a tig mutant in L. monocytogenes was defective in bacterial survival in vivo as well as in heat and ethanol stress (70). Most recently, Cohen et al. demonstrated that in addition to being a cytoplasmic chaperone, TF is functional as an anchorless cell-wall protein in S. pneumoniae, and that it contributes to virulence in mice (71). In addition, injection of mice with recombinant TF resulted in a protective immune response, showing its potential as a vaccine candidate (71). Despite the numerous and diverse virulence-related phenotypes that have been identified for tig in other Gram-positive pathogens, no studies have been done on tig in S. aureus. Chapter 4 of this dissertation describes the first phenotypes for a tig mutant in this pathogen.

CHAPTER 2: THE INTRACELLULAR CYCLOPHILIN PPIB CONTRIBUTES TO

THE VIRULENCE OF STAPHYLOCOCCUS AUREUS INDEPENDENTLY OF ITS

PEPTIDYL-PROLYL CIS/TRANS ISOMERASE ACTIVITY

Rebecca A. Keogh, Rachel L. Zapf, Richard E. Wiemels, Marcus A. Wittekind, Ronan K.

Carroll.

2018, doi: 10.1128/IAI.00379-18 27

Abstract

The Staphylococcus aureus cyclophilin PpiB is an intracellular peptidyl prolyl cis/trans isomerase (PPIase) that has previously been shown to contribute to secreted nuclease and hemolytic activity. In this study, we investigate the contribution of PpiB to

S. aureus virulence. Using a murine abscess model of infection, we demonstrate that a ppiB mutant is attenuated for virulence. We go on to investigate the mechanism through which PpiB protein contributes to virulence, in particular the contribution of PpiB PPIase activity. We determine the amino acid residues that are important for PpiB PPIase activity and show that a single amino acid substitution (F64A) completely abrogates

PPIase activity. Using purified PpiB F64A protein in vitro, we show that PPIase activity only partially contributes to Nuc refolding and that PpiB also possesses a PPIase- independent activity. Using allelic exchange, we introduce the F64A substitution onto the S. aureus chromosome, generating a strain that produces enzymatically inactive PpiB.

Analysis of the PpiB F64A strain reveals that PPIase activity is not required for hemolysis of human blood or virulence in a mouse. Together, these results demonstrate that PpiB contributes to S. aureus virulence via a mechanism unrelated to prolyl- isomerase activity.

Introduction

Staphylococcus aureus is a Gram-positive bacterium that colonizes the anterior nares of approximately 30% of the population. In addition, S. aureus is considered an opportunistic pathogen, causing diseases that range in severity from minor skin and soft tissue infections to life-threatening sepsis, endocarditis, and necrotizing fasciitis (72).

One contributing factor for the diversity of diseases caused by S. aureus is the myriad of virulence factors produced by the organism. These include immune evasion proteins, microbial surface components recognizing adhering matrix molecules (MSCRAMMs), toxins, superantigens, and exoenzymes. One such exoenzyme secreted by S. aureus is

Staphylococcal nuclease (Nuc), which plays important roles in immune evasion and biofilm growth (73–78).

Previous work in our lab investigating the process of Nuc secretion identified an intracellular peptidyl-prolyl cis/trans isomerase (PPIase), PpiB, that contributes to Nuc activity (63). PPIase enzymes (also known as foldases) catalyze the cis to trans isomerization of proline peptide bonds, which is often the rate-limiting step in protein folding (79). PpiB is a functional PPIase belonging to the cyclophilin family, and it assists in the refolding of Nuc in vitro. Culture supernatants from a ppiB mutant strain had decreased Nuc activity and also a decrease in hemolytic activity, suggesting that there are additional cellular targets for PpiB within the S. aureus cell. While this work clearly demonstrated a role for PpiB in the secretion of S. aureus virulence factors, it remains unclear if PpiB is contributing to nuclease/hemolytic activity through its PPIase enzymatic activity or via some other function. Furthermore, it is unknown if the reduction 29 in nuclease/hemolytic activity in a ppiB mutant manifests as decreased virulence, although a recent Tn-seq study suggests that a ppiB mutant has decreased fitness in an abscess model of infection (80). In this study, we set out to determine the role of PpiB during infection. Specifically, we investigate if PpiB contributes to virulence via its

PPIase activity. Previous work by Alonzo et. al., studying the PPIase PrsA2 in Listeria monocytogenes, showed that while PrsA2 demonstrates PPIase enzymatic activity, its role in virulence is not entirely dependent on this activity (54). PrsA2 catalytic activity is required for full virulence in vivo, but is not required to restore defects in hemolytic activity, phospholipase activity, and L2 plaque formation (51). These results show that

PPIase enzymes can have additional functions not limited to PPIase activity that contribute to virulence.

Here, we investigate the role of PpiB in S. aureus virulence and show that a

ΔppiB mutant is attenuated in a murine abscess model of infection. We investigate the contribution of PpiB PPIase activity by determining the amino acid residues within PpiB that are necessary for PPIase activity and use this information to construct a strain containing a single substitution (F64A) that abolishes this activity. This strain (which produces an enzymatically inactive form of PpiB) is tested for known PpiB functions including nuclease refolding, hemolytic activity, and virulence, to determine the contribution of PPIase activity to PpiB function. Together, the results demonstrate that the PPIase activity of PpiB partially contributes to Nuc refolding; however, it is not required for hemolytic activity or virulence. These data suggest that PpiB has additional roles in the cell that contribute to the virulence of S. aureus. 30

Results

PpiB is Required for Virulence in a Murine Abscess Model of Infection

Previous work by our group has demonstrated that PpiB, an intracellular cyclophilin family member, contributes to the activity of secreted virulence factors in S. aureus (63). Specifically, reduced nuclease and hemolytic activity were observed in culture supernatants from a ppiB mutant strain. We hypothesized that this reduction in activity would manifest as an attenuation of virulence during infection. Due to the previously demonstrated role for Nuc in avoiding clearance by neutrophil extracellular traps (NETS) (75), we postulated that removal of ppiB (and the subsequent reduction in

Nuc activity (63)) would manifest as increased clearance of the bacteria by the immune system. To test this hypothesis, we employed a murine abscess model of infection to compare the number of wild-type and ppiB mutant bacteria present in abscesses following a 7-day infection. BALB/c mice were injected subcutaneously in the lower right flank with 106 CFU of either wild-type or a ppiB strain and following a 7-day infection period, mice were sacrificed, abscesses were excised, and the number of bacteria present in abscesses determined. The ppiB mutant was significantly attenuated for virulence in the abscess model, with approximately 14-fold less bacteria recovered from abscesses (Figure 2.1). This result supports our hypothesis that reduced virulence factor activity in the ppiB mutant leads to attenuation of virulence in vivo. 31

Figure 2.1. PpiB is required for virulence in a murine abscess model of infection. Female, 6-week old BALB/c mice were injected subcutaneously with wild-type S. aureus or a ΔppiB mutant strain. The infection was allowed to proceed for 7 days. Mice were then sacrificed before abscesses were excised, homogenized, and diluted and plated to enumerate bacteria present in the abscesses. In mice infected with the ΔppiB mutant strain, a significant 14-fold reduction in bacterial numbers was detected. Experiments were performed twice with an n=16 for each strain. The median value for each sample is indicated. Significance was determined by Student’s t test. ** = p < 0.01.

Determination of the PPIase Active Site Residues in PpiB

PpiB is a member of the cyclophilin family of PPIases. In this family of enzymes, six amino acid residues are known to be important for PPIase activity (81). When we aligned the sequence of PpiB to the prototypical cyclophilin family member, human cyclophilin A (82), we observed that all six of these residues (H58, R59, F64, Q114,

F116 and W152) are conserved in PpiB (Figure 2.2A). To investigate if these six residues make up an enzymatic pocket similar to that in cyclophilin A, we used the 32

SWISS-MODEL automated protein structure homology-modelling server (83–85) to predict the three-dimensional structure of PpiB and compare it to the previously published structure of cyclophilin A (Figure 2.2B and 2.2C). The position and coordination of all six amino acids is highly conserved in the PpiB predicted structure.

Based on this analysis, we hypothesized that these six residues comprise the PPIase active site in PpiB. 33

Figure 2.2. Identification of PPIase active site residues in PpiB. (A) Sequence alignment of PpiB and human cyclophilin A. The six active site residues in human cyclophilin A are colored along with their corresponding amino acids in PpiB. (B) Structure of human cyclophilin A (PDB ID: 2A2N). The six active site residues are color coded the same as in panel A above. (C) Predicted structure of PpiB. The three- dimensional structure of PpiB was predicted using the SWISS-MODEL automated protein structure homology-modelling server. The six predicted active site residues are color coded the same as in panels A and B. The 28 amino acid internal loop found in PpiB that is absent in cyclophilin A is indicated (red box). 34

Substitution of PPIase Active Sites Eliminates PPIase Activity in PpiB

We have previously shown that purified PpiB protein (i) is functional as a PPIase, and (ii) assists in the folding of Staphylococcal nuclease. Based on these data and the results from the mouse infection assay (Figure 2.1), we hypothesized that the PPIase activity of PpiB contributes to the virulence of S. aureus by assisting the folding (and hence activation) of secreted virulence factors. To test this hypothesis, we set out to construct a strain of S. aureus that expresses an enzymatically inactive form of PpiB and determine if the loss of PPIase activity leads to an attenuation of virulence similar to that observed in the ppiB mutant.

The first step in this process was to experimentally determine if the six amino acid residues outlined above (H58, R59, F64, Q114, F116, and W152) comprise the PpiB

PPIase active site (36). Site-directed mutagenesis was performed whereby each of the six amino acids was substituted with alanine (H58A, R59A, F64A, Q114A, F116A, and

W152A). The resulting recombinant proteins were purified and used in a chymotrypsin- coupled PPIase activity assay previously used by our group to study PpiB (63). As expected, wild-type PpiB protein exhibits PPIase activity in the oligopeptide cleavage assay (Figure 2.3). Analysis of the PpiB substituted proteins revealed that 4 of the 6 substitutions abrogated PPIase activity. Specifically, the R59A, F64A, Q114A, and

F116A substituted forms of PpiB did not display PPIase activity. The H58A and W152A substituted proteins displayed modest reductions in PPIase activity. The F64A substituted form of PpiB consistently showed no PPIase activity above negative controls 35

(Figure 2.3), therefore we elected to proceed with and further characterize this enzymatically inactive form of PpiB.

Figure 2.3. Substitution of predicted active site residues eliminates PpiB PPIase activity. Purified recombinant PpiB (5 nM) was used in a chymotrypsin-coupled PPIase activity assay. Chymotrypsin only cleaves the substrate peptide when the prolyl peptide bond is in the trans conformation, therefore in the absence of a PPIase (blue line) the reaction proceeds to completion based on the slow endogenous rate of cis-to-trans isomerization of the peptide bond. Addition of PpiB to the reaction results in an increased rate of cleavage consistent with PpiB functioning as a PPIase (red line). The addition of PpiB containing substitutions F116A, Q114A, R59A, or F64A did not increase the rate of reaction above the negative control. Addition of PpiB containing substitutions H58A or W152A resulted in an increase in the rate of cleavage but not to the same extent as wild- type PpiB. Experiments were performed a minimum of three times. Representative data sets are shown.

PpiB F64A Demonstrates Reduced Nuclease Refolding Ability

In vitro studies by our group have demonstrated that PpiB is required for the optimal folding of Nuc (12). To determine if the PPIase activity of PpiB is necessary for

Nuc refolding, we performed a Nuc refolding assay whereby we measured the rate of

Nuc refolding in the presence of native PpiB and compared it to the enzymatically inactive PpiB F64A. Recombinant Nuc protein was purified and denatured in 8 M urea.

Denatured Nuc was allowed to refold by diluting it into urea-free buffer in the absence of a PPIase or in the presence of PpiB or PpiB F64A. The activity of refolded Nuc in each sample was determined by adding an oligonucleotide probe. The probe contains a quencher at the 3’ end and a fluorophore at the 5’ end. When active Nuc is present it digests the probe and a fluorescent signal is detected. As previously shown (63) when native PpiB is added to the reaction, there is increased fluorescence indicative of active refolded Nuc (Figure 2.4). In the absence of PpiB, low levels of fluorescence are detected indicating minimal Nuc activity. Reactions containing PpiB F64A show an intermediate phenotype with fluorescence levels higher than the negative control (no

PpiB) but less than in reactions containing native PpiB. While these data demonstrate that the PPIase activity of PpiB plays a role in Nuc refolding (compare native PpiB to

PpiB F64A), they also show that in the absence of PPIase enzymatic activity PpiB still contributes to Nuc refolding (compare PpiB F64A to no PpiB). Together, these results suggest that PpiB possesses some additional PPIase-independent activity that contributes to Nuc refolding. 37

Figure 2.4. PpiB increases the rate of Nuc refolding in the absence of PPIase activity. Denatured Nuc was diluted 1:40 into denaturant-free buffer containing an oligonucleotide probe. Refolding and subsequent activity of Nuc were visualized by fluorescence resulting from cleavage of the probe. Refolding reactions in the presence of PpiB (red line) demonstrate high levels of nuclease activity while reactions without PpiB (blue line) demonstrate low levels of nuclease activity. Addition of PpiB F64A to the reaction resulted in an increase in nuclease activity (above the negative control), although the increase was less than that observed with wild-type PpiB protein (green line). Experiments were performed a minimum of three times. Representative data are shown.

PpiB is Expressed in the F64A Strain

Having identified an amino acid substitution (F64A) that abrogates PpiB PPIase activity, we next wanted to construct a strain of S. aureus that expresses this enzymatically inactive form of PpiB on the chromosome. To do this, we utilized the allelic exchange plasmid pJB38 (86). Once the chromosomal copy of ppiB had been exchanged for the ppiB F64A allele, we performed growth rate analysis to confirm that there was no difference in growth rate between the wild type, ppiB mutant, and F64A 38 strain (Figure 2.6). A western blot was then performed to ensure the PpiB F64A protein was being expressed at comparable levels to wild-type PpiB. Intracellular protein fractions from overnight cultures of wild-type S. aureus, the ppiB mutant, and PpiB

F64A strains were separated by SDS-PAGE (Figure 2.7) and probed using anti-PpiB antiserum. Results show comparable levels of PpiB protein in the wild-type and F64A strain, and no PpiB in the ppiB mutant (Figure 2.5). While the amount of PpiB is similar in the wild-type and F64A strain, the F64A substituted form of PpiB appears to run at a slightly higher molecular weight than wild-type PpiB. This apparent difference in migration was also visible when the purified recombinant forms of each protein were analyzed by SDS-PAGE and stained with Coomassie blue (data not shown). Although the reason for this difference is unclear, previous studies have demonstrated that single amino acid substitutions can lead to differences in SDS-PAGE migration rates (87). The western blot analysis also revealed the presence of a cross reacting band that migrated at a slightly higher apparent molecular weight than PpiB. Interestingly this band appeared to be reduced in the ppiB mutant, although equal quantities of protein were loaded in each lane (Figure 2.7). We speculate that this band may represent an intracellular protein whose expression is controlled by PpiB in a PPIase independent manner (as it is present at similar levels in the wild type and F64A strain). 39

Figure 2.5. Growth curve analysis in TSB of USA300 TCH1516 (Wild type), the ΔppiB mutant, and F64A strain. No significant differences in growth were observed between the three strains. Data shown is the average of two independent biological replicates.

Figure 2.6. Coomassie stained gel of intracellular protein samples used in western blot analysis.

Figure 2.7. PpiB expression is comparable in the wild-type and F64A substituted strain. A western immunoblot was performed using polyclonal anti-PpiB antisera and whole cell lysates from wild-type S. aureus, the ∆ppiB mutant, and PpiB F64A substituted strains. No PpiB was detected in the ∆ppiB strain while comparable levels of PpiB protein were detected in the wild-type and PpiB F64A substituted strains. A slight difference in apparent molecular weight of the PpiB F64A substituted protein was observed.

The PpiB F64A Strain Does Not Demonstrate Reduced Hemolytic Activity

Previous work in our lab has shown culture supernatants from a ppiB mutant strain have decreased hemolytic activity (12). S. aureus secretes a number of toxins capable of lysing erythrocytes (88). The identity of the PpiB-regulated toxin and the role of PpiB in hemolysis remains unknown. To determine if PPIase activity is required for

PpiB-dependent hemolytic activity and to explore the identity of the hemolysin responsible, we performed erythrocyte lysis assays using both human and rabbit blood with culture supernatants from the wild-type, ppiB mutant, and PpiB F64A strains. A significant decrease in human erythrocyte lysis was observed in ppiB mutant 41 supernatants (as previously reported), however no decrease in hemolytic activity was observed using supernatants from the F64A strain (Figure 2.8A). This result demonstrates that the PPIase activity of PpiB is not required for hemolysis. Interestingly, when the experiments were repeated using rabbit blood, only a small, non-significant reduction in hemolysis was observed in ppiB mutant culture supernatants (Figure 2.8B). This result shows that the PpiB-dependent hemolytic factor is less potent against rabbit erythrocytes than human erythrocytes indicating that alpha toxin (Hla) is not responsible, as it has greater affinity for rabbit erythrocytes than human erythrocytes (18, 89–91). Together these results show that (i) alpha toxin is not the PpiB-dependent hemolysin, and (ii) PpiB- dependent hemolysis is independent of PPIase activity, strongly suggesting that PpiB possesses some additional function.

Figure 2.8. PpiB-dependent hemolytic activity does not require PPIase activity. Erythrocyte lysis assays were performed using S. aureus culture supernatants and whole human blood (A) or rabbit blood (B). A significant decrease in hemolytic activity against human erythrocytes was observed using culture supernatants from a ppiB mutant strain. Culture supernatants from the PpiB F64A strain did not display any significant decrease in hemolytic activity compared to the wild-type strain. Hemolytic activity was greatly reduced in an agrB mutant while a small, non-significant reduction in hemolysis was observed in an hla mutant. (B) Using rabbit blood, a small, non-significant reduction in hemolytic activity was observed in the ppiB mutant while no decrease was observed in the PpiB F64A strain. Hemolytic activity was greatly reduced in both the agrB and hla mutant strains. Hemolysis assays were performed a minimum of 3 times. Hemolytic activity in the wild type strain for each experiment is set to 100% and the relative hemolytic activity of the other strains is indicated as a percentage. The data presented in the average of 4 replicates. Significance was determined by Student’s t test. **** = p < 0.001, * = p < 0.05, ns = not significant.

The PPIase Activity of PpiB Does Not Contribute to S. aureus Virulence

The results outlined in Figure 2.1 show that PpiB is required for virulence in a murine abscess model of infection. To test the hypothesis that PpiB contributes to virulence via its PPIase activity, we repeated the murine abscess infection to compare the virulence of wild-type S. aureus to the ppiB mutant and PPIase inactivated F64A strain.

As previously shown (Figure 2.1), a significant reduction in bacterial numbers was observed in the abscesses of mice infected with the ppiB mutant. In contrast, there was 43 no reduction in bacterial numbers when the F64A strain was used (Figure 2.9). These data clearly show that while PpiB is contributing to virulence in S. aureus, it is not doing so via its PPIase activity. Together, our data show that PpiB has an additional role in the cell not limited to prolyl-isomerization.

Figure 2.9. The PPIase activity of PpiB does not contribute to S. aureus virulence. Groups of eight BALB/c mice were injected subcutaneously with wild-type S. aureus, the ΔppiB mutant, and PpiB F64A strain. The infection was allowed to proceed for 7 days. Mice were then sacrificed before abscesses were excised, homogenized, and diluted and plated to enumerate bacteria present in the abscesses. As previously observed (Figure 2.1), a decrease in the number of bacterial cells was observed in the abscesses of mice infected with the ΔppiB mutant strain. No significant decrease in bacterial numbers was observed in abscesses of mice infected with the PpiB F64A strain. The median value for each sample is indicated. Significance was determined by Student’s t test. * = p < 0.01, ns = not significant.

Discussion

Bacterial PPIase proteins have been extensively studied based on their ability to catalyze the cis-to-trans isomerization around proline-peptide bonds and assist in the folding of secreted virulence factors (54, 92–95). Previous work by our group on the S. aureus cyclophilin PpiB demonstrated that PpiB is a functional PPIase, that it affects the refolding and activity of the virulence factor Nuc, and that it contributes to hemolytic activity (63). Based on these findings, we sought to investigate whether PpiB contributes to virulence in S. aureus and whether its demonstrated PPIase activity (that assists in the folding of at least one of these virulence factors) contributes to disease.

We utilized a murine abscess model of infection to show that a ΔppiB mutant is attenuated compared to wild-type S. aureus. A 14-fold reduction in bacterial numbers was observed in abscesses of mice infected with the ΔppiB mutant. Our initial hypothesis was that this reduction in virulence was due to the loss of PpiB PPIase activity and its role in the secretion and folding of Nuc, as well as a hemolytic toxin. To test this hypothesis, we created a strain (F64A) in which the PPIase activity of PpiB was eliminated and tested it for known PpiB-related phenotypes. Results from in vitro nuclease refolding assays show that PpiB F64A can still assist the refolding of Nuc, although not to the levels observed when using wild-type PpiB protein. This demonstrates that the while PPIase activity contributes to Nuc folding in vitro, PpiB also assists refolding of Nuc in a PPIase-independent manner. Many PPIases also possess chaperone activity (41, 42, 96, 97), therefore it is possible that PpiB may also be functioning as a chaperone. Typically chaperone proteins prevent the aggregation of 45 proteins inside the cell under stressful conditions (47). By binding to Nuc and protecting it under denaturing conditions, PpiB may be acting as a chaperone as well as assisting in the folding of Nuc via its PPIase activity. The idea of PpiB functioning as both a PPIase and chaperone is consistent with recent work by Skagia et al. who demonstrate a similar dual-role for a PpiB homologue in E. coli (41, 61). Deletion of ppiB in E. coli results in increased swimming motility, swarming motility, and biofilm formation. The swarming and biofilm phenotypes could not be complemented by enzymatically inactive PpiB mutants, indicating that these processes require PpiB PPIase activity. However, the swimming phenotype could be restored by an enzymatically inactive form of PpiB, indicating that this process required a PPIase-independent function of the PpiB protein.

Together, these data show that E. coli PpiB can function as a PPIase but also has PPIase- independent activity.

Interestingly, when we examined the predicted three-dimensional structure of

PpiB (Figure 2.2C), we observed a 28 amino acid loop that is present in S. aureus PpiB but absent in human cyclophilin. To investigate if this loop could potentially impart

PPIase-independent activity, we performed a BLAST search using the 28 amino acid sequence (i.e. residues 122 to 149 in PpiB). To identify potentially novel functions for this region we excluded hits to members of the Staphylococceae, and to annotated

PPIases. The resulting matches show that residues 122 to 135 in PpiB have a high degree of homology to proteins in the PIMT family (protein-L-isoaspartate(D-aspartate) O- methyltransferase). Proteins in this family are methyltransferases that catalyze the repair of damaged aspartic acid residues. Although widespread in nature S. aureus does not 46 appear to encode any PMIT family members. A multiple sequence alignment of PpiB homologues shows that the PMIT region is unique to S. aureus PpiB (Figure 2.10) therefore it is tempting to speculate that the PPIase-independent function of PpiB may be mediated via methyltransferase activity and repair of damaged aspartic acid residues.

Work is currently underway in our lab to further explore the role of this PMIT homologous region of PpiB.

Figure 2.10. Multiple sequence alignment of PpiB homologues. PpiB homologues from E. coli, Legionella pneumophila, Streptococcus pneumoniae, Listeria monocytogenes, and Bacillus subtilis were aligned with S. aureus PpiB using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/). The region of PpiB (located withing the 28 amino acid loop) displaying homology to PMIT family proteins is highlighted in red.

To further investigate the role of PpiB in the cell, we performed whole-blood hemolysis assays. While our previous work demonstrated a hemolytic defect in a ppiB mutant, it was unknown which hemolytic toxin PpiB was regulating. Here, we perform whole-blood hemolysis assays in both human and rabbit blood. Interestingly, we 48 observed that the F64A strain does not have a hemolytic defect in either source of blood.

This shows that the PPIase activity of PpiB is not contributing to hemolytic activity.

Furthermore, we demonstrate that while PpiB is contributing to hemolysis in human- blood, it is not contributing to hemolysis in rabbit blood. As it is well-established that the

S. aureus alpha toxin is the main contributor to erythrocyte lysis in rabbit blood, we conclude that PpiB is not affecting alpha toxin production or secretion (89). The dramatic reduction in hemolysis in whole-human blood is suggestive that PpiB may be regulating the α phenol-soluble modulins (αPSMs), as it has been shown that these toxins strongly contribute to human erythrocyte lysis (18, 90, 91). Consequently, we hypothesize that

PpiB is regulating activity of the αPSMs via a PPIase independent mechanism.

Interestingly If this hypothesis is correct, and the PSMs are subject to regulation by PpiB, they would represent the second S. aureus secreted virulence factors that are (i) PpiB- dependent and (ii) are secreted from the cell via non-canonical methods (Nuc being the other). Though secreted by the general secretion system, several aspects of Nuc secretion are atypical. First, it has an unusually long (60 amino acid) secretion signal sequence

(78). Second, it is initially secreted from the cell as a longer, less active form (NucB) that is subsequently proteolytically cleaved into a shorter, more active form (NucA) (98).

Third, there is evidence of sec-independent secretion of Nuc (99). And finally, the 13 amino acid section of NucB that is proteolytically removed (to generate NucA) has been shown to aid in the secretion process via an unknown mechanism (100). The unconventional nature of Nuc secretion suggests additional factors are involved, one of which may be PpiB. In contrast to Nuc, the αPSMs are not secreted via the general 49 secretion pathway. Instead, they are secreted via a dedicated ABC transport system, the

Pmt system (29). While very little is known about this system, it is possible that PpiB could also be involved in this process. Work is ongoing in our lab to determine the molecular interaction partners of PpiB in an effort to better understand how it is involved in the secretion and activity of Nuc and (potentially) the αPSMs, and if additional secreted virulence factors are regulated by PpiB.

Results from the murine abscess infection model suggest that the reduction in nuclease and hemolytic activity in a ppiB mutant (and possibly additional defects) manifest as attenuation of virulence. Importantly, the PpiB F64A strain was not attenuated, clearly demonstrating that the PPIase activity of PpiB is not contributing to virulence. Collectively, the data presented in this study suggest that PpiB has an additional role, not limited to PPIase activity in S. aureus. We hypothesize that PpiB may be acting intracellularly as a chaperone (possibly with methyltransferase activity), or alternatively PpiB may be acting under a novel mechanism not previously described in a bacterial PPIase.

Materials and Methods

Strains and Strain Construction

All bacterial strains and plasmids are listed in Table 1 and oligonucleotides listed in Table 2. A ppiB mutant strain was constructed by allelic exchange using plasmid pJB38 (86). DNA sequences flanking the ppiB gene were amplified using primer pairs

#207/#208 and #209/#210 and cloned into pJB38 to generate plasmid pRKC0212. This plasmid was recombined onto the S. aureus chromosome and then excised to generate a

ppiB deletion strain, according to published protocol (101). The F64A substituted strain was also constructed by allelic exchange using plasmid pJB38. DNA sequences flanking the ppiB gene were amplified using primer pairs #311/#312 and #313/#314 (with genomic DNA as template) and the ppiB gene containing the F64A mutation was amplified using primer pair #315/#316 (and plasmid pRKC0378 as template). The resulting PCR products were ligated together and cloned into pJB38 generating plasmid pRKC0438. This plasmid was recombined onto the S. aureus chromosome and then excised to generate the F64A substituted strain. S. aureus containing a transposon insertion in the hla gene was obtained from the Network on Antimicrobial Resistance in

Staphylococcus aureus (NARSA) transposon mutant library (102) and transduced into S. aureus USA300 TCH1516 using bacteriophage ϕ11.

Bacterial Growth Conditions

S. aureus cultures were routinely grown at 37 °C with shaking in tryptic soy broth

(TSB) and E. coli cultures at 37 °C with shaking in lysogeny broth (LB). Where appropriate antibiotics were used at the following concentrations; chloramphenicol 5 μg 51 ml-1, ampicillin 100 μg ml-1, kanamycin 50 μg ml-1. For comparative analysis of supernatants, S. aureus cultures were synchronized as follows. Overnight starter cultures

(5 ml) of each strain were diluted 1:100 in 10 ml of fresh, pre-warmed TSB, and grown for 3 h to mid-exponential phase. The 3 h, mid-exponential phase cultures were subsequently diluted into 25 ml of fresh TSB at a starting OD600 of 0.05. The cultures were then grown for the time indicated, typically 15 h.

Murine Abscess Model of Infection

A subcutaneous abscess model was chosen to closely mimic skin and soft tissue infections, as described by Malachowa et. al. (103). Cultures of S. aureus were grown in

TSB for 2.5 h until an OD600 of approx. 0.75. Bacterial cells were pelleted by centrifugation and resuspended in sterile PBS for inoculation. Inocula were prepared at

106 CFU/50 μl for injection and confirmed via serial diluting and plating. Six-week-old

BALB/c female mice were shaved on the right flank before being treated with Nair to completely remove hair. Mice were injected in the right flank with 50 μl and infections were allowed to persist for 7 days. Over the course of infection, mice were monitored for irregular activity or distress. After 7 days, mice were euthanized with CO2, abscesses were excised and homogenized. Homogenates were then diluted and plated onto TSB for colony counting.

Site-Directed Mutagenesis and Protein Purification

Oligonucleotide primers were designed to substitute the six predicted PpiB active residues for alanine. The residues were substituted using the following primer pairs:

H58A (#234/#235), R59A (#236/#237), F64A (#238/#239), Q114A (#240/#241), F116A 52

(#242/#243), and W152A (#244/#245) and the DNA was cloned into the pMALc5X vector (New England Biolabs [NEB]) for substitutions. The product was then digested with DpnI enzyme to eliminate any methylated DNA from the parental strain. Digested product was transformed into competent cells and grown overnight at 37oC. Products were sequenced to confirm substitutions to alanine. The resulting plasmids, pRKC0376- pRKC0381, express an N-terminal maltose binding protein (MBP) fusion to the substituted PpiB proteins. The presence of MBP fused to PpiB does not affect the PPIase activity of PpiB (63). The MBP-PpiB fusion proteins were expressed in E. coli as follows. A 100 ml flask of LB was inoculated with 1 ml of an overnight starter culture of

NEB Express cells containing the substituted plasmids (pRKC0376-pRKC0381 depending on the substitution) and grown to an OD600 of 0.6. Expression of the MBP-

PpiB fusions was induced by the addition of isopropyl- β-D-thiogalactopyranoside

(IPTG) (to a final concentration of 0.3 mM), and the culture was grown for an additional

3 h. The cells were harvested, resuspended in 5 ml of column buffer (20 mM Tris-HCl

[pH 7.4], 200 mM NaCl, 1 mM EDTA), sonicated, and centrifuged for 20 min at 20,000 x g. Cell lysates were loaded onto 0.5 ml of amylose resin that had been equilibrated with column buffer, washed two times with 10 ml of column buffer, and eluted in six 0.5 ml fractions with column buffer containing 10 mM maltose. Fractions containing MBP-

PpiB with substitutions were pooled and stored in 40% glycerol.

Chymotrypsin-Coupled PPIase Activity Assay

The chymotrypsin-coupled PPIase activity assay was performed as described previously for PpiB (9). Reaction mixtures containing assay buffer, chymotrypsin, and 53 purified PpiB, PpiB F64A, or water were prepared and stored briefly on ice. The reaction was initiated by adding the mixture to the oligopeptide Suc-AAFP-pNA (Sigma) in a cuvette and absorbance measured at 390 nm using a Genesys 30 spectrophotometer

(Thermo Fisher).

Nuclease Refolding Assay

The nuclease refolding and activity assay was performed as described by Wiemels et al. (9). Recombinant histidine-tagged Nuc protein (Nuc-His6) was purified and denatured in 8 M urea. Denatured Nuc was diluted 1:40 into a reaction mixture (10 nM final concentration) containing buffer (20 mM Tris [pH 8.0], 10 mM CaCl2), the FRET probe (2 μM), and either purified recombinant PpiB (1μM), PpiB F64A (1 μM), or water.

Dilution of the denatured Nuc (and the corresponding dilution of urea) facilitated refolding of the protein. The activity of refolded Nuc was observed by an increase in nuclease activity and cleavage of the oligonucleotide FRET probe. Variations in the rate of Nuc refolding are manifested as variations in the rate of nuclease activity against the

FRET probe over time.

Western Blotting

Intracellular protein samples for western blots were prepared from stationary phase cultures of the wild-type, ΔppiB mutant, and F64A strain as previously described

(104). Samples were separated by SDS-PAGE (Figure 2.7), transferred to PVDF membrane and probed using rabbit polyclonal anti-PpiB antibody that was raised using

MBP-tagged PpiB protein (63). 54

Hemolysis Assay

Cell-free supernatants from 15 h synchronized S. aureus cultures were diluted 1:2 in reaction buffer (40 mM CaCl2, 1.7% NaCl) and incubated at 37°C with 25 μl of whole blood (human or rabbit). Following a 10 min incubation, samples were centrifuged at

5,500 x g, and 100 μl of the supernatant was transferred to a 96-well plate. Erythrocyte lysis was determined by measuring absorbance of the samples at OD543.

Ethics Statement

Human blood was obtained from anonymous donors at Ohio University. All collections, handling, and usage of blood was approved by the Ohio University

Institutional Review Board. Rabbit blood was obtained from BioIVT. 6-week old Balb/C mice were ordered from Envigo and held at the Ohio University Office of Laboratory

Animal Resources. All animal work was done under approval of the Institutional Animal

Care and Use Committee by trained lab personnel.

Table 2.1. Strains and plasmids used in this study. Strain or plasmid Characteristicsα Source S. aureus TCH1516 Community-associated USA300 MRSA isolate (105) RN4220 Restriction deficient transformation recipient (106) RKC0323 TCH1516 ΔppiB This work RKC0505 TCH1516 PpiB F64A This work RKC0165 TCH1516 agrB (17) RKC0183 TCH1516 hla This work NE1354 USA300 JE2 hla::Bursa; hla NARSA transposon (102) mutant E. coli DH5α Cloning strain Invitrogen BL21(DE3)/pLysS Protein expression strain Promega NEB Express Protein expression strain NEB

Plasmids pMK4 Gram-positive shuttle vector (Cmr) (107) pJB38 Temperature-sensitive allelic exchange plasmid (86) (Cmr) pET24d C-terminal His6 tag expression vector Novagen pMALc5X N-terminal MBP fusion expression vector NEB pRKC0276 pET24d nuc (63) pRKC0212 pMALc5X ppiB (63) pRKC0376 pMALc5X ppiB H58A This work pRKC0377 pMALc5X ppiB R59A This work pRKC0378 pMALc5X ppiB F64A This work pRKC0379 pMALc5X ppiB Q114A This work pRKC0380 pMALc5X ppiB F116A This work pRKC0381 pMALc5X ppiB W152A This work pRKC0295 pJB38 containing DNA flanking ppiB gene This work pRKC0438 pJB38 containing ppiB F64A allele This work αCmr, chloramphenicol resistance

Table 2.2. Oligonucleotides used in this study. Name Sequenceα #207 CAAGAGCTCAAACTAGAAAATGACGTTAGCTC #208 CGGACGCGTTATATTCTCCATTCATGTTATGATAC #209 CGGACGCGTTATCTAAACATAATTAACTACCAAC #210 GGGGTACCACAAATCCACATACAATATTG #311 CACCTAACATTGCACCCATAAA #312 GCATGTGTCACAAAATTTTCAA #313 TTCGGTCAAATCATTGATGG #314 AATGGATATGTCACCTTAAACCA #315 ATGGCTAACTATCCACA #316 TTATTCTTCAACATCAATAGATT #234 GGAATCACATTCgctCGTGTCATTAATG #235 CATTAATGACACGagcGAATGTGATTCC #236 GAATCACATTCCACgctGTCATTAATGAC #237 GTCATTAATGACagcGTGGAATGTGATTC #238 GTCATTAATGACGCTATGATTCAAGGT #239 ACCTTGAATCATAGCGTCATTAATGAC #240 ACTAATGGTTCAGCTTTTTTCATTGTT #241 AACAATGAAAAAAGCTGAACCATTAGT #242 GGTTCACAATTTGCTATTGTTCAAATG #243 CATTTGAACAATAGCAAATTGTGAACC #244 GGTGGTACACCAGCTTTAGATCAAAAA #245 TTTTTGATCTAAAGCTGGTGTACCACC FRET probe Cy3-CCCCGGATCCACCCC-BHQ2 αBHQ2, black hole quencher 2.

CHAPTER 3: NOVEL REGULATION OF ALPHA-TOXIN AND THE PHENOL-

SOLUBLE MODULINS BY PEPTIDYL-PROLYL CIS/TRANS ISOMERASE

ENZYMES IN STAPHYLOCOCCUS AUREUS

Rebecca A. Keogh, Rachel L. Zapf, Emily Trzeciak, Gillian G. Null, Richard E.

Wiemels, Ronan K. Carroll

Abstract

Peptidyl-prolyl cis/trans isomerases (PPIases) are enzymes that catalyze the cis- to-trans isomerization around proline bonds, allowing proteins to fold into their correct confirmation. Previously, we identified two PPIase enzymes in Staphylococcus aureus

(PpiB and PrsA) that are involved in the regulation of virulence determinants and have shown that PpiB contributes to S. aureus virulence in a murine abscess model of infection. Here, we further examine the role of these PPIases in S. aureus virulence and, in particular, their regulation of hemolytic toxins. Using murine abscess and systemic models of infection, we show that a ppiB mutant in a USA300 background is attenuated for virulence but that a prsA mutant is not. Deletion of the ppiB gene leads to decreased bacterial survival in macrophages and nasal epithelial cells, while there is no significant difference when prsA is deleted. Analysis of culture supernatants reveals that a ppiB mutant strain has reduced levels of the phenol-soluble modulins and that both ppiB and prsA mutants have reduced alpha-toxin activity. Finally we perform in vivo immunoprecipitation to identify cellular targets of PpiB and PrsA. Results suggest a novel role for PpiB in S. aureus protein secretion. Collectively, our results demonstrate that PpiB and PrsA influence S. aureus toxins via distinct mechanisms, and that PpiB but not PrsA contributes to disease.

Introduction

Staphylococcus aureus is a Gram-positive bacterium that resides in the anterior nares of approximately one-third of the population. Diseases caused by S. aureus range in severity from minor skin and soft tissue infections, to life-threatening infections such as endocarditis, necrotizing fasciitis, and sepsis (72). This incredible diversity in diseases is largely due to the multitude of virulence factors that S. aureus produces, such as exoenzymes that assist in the degradation of host molecules, adhesins that aid in attachment to surfaces, and toxins that lyse host cells (18, 19, 77, 108). Two of the best characterized toxins for their role in infection are α-toxin (encoded by the hla gene) and the phenol-soluble modulins (PSMs) (89, 109).

Hla is a receptor-mediated pore-forming toxin, which binds the sheddase

ADAM10 on the surface of host cells and disrupts their cellular membranes. Hla has been implicated as the primary toxin responsible for the lysis of rabbit erythrocytes, which have high amounts of ADAM10 coating their surface (89, 110, 111). Interestingly, human red blood cells have very little ADAM10 on their surface and consequently, it takes high levels of accumulated Hla to lyse human erythrocytes. Lysis of human erythrocytes is more efficiently accomplished by the alpha phenol-soluble modulins

(αPSMs) (112). S. aureus encodes four αPSMs, each approximately 20-25 amino acids in size, on a single polycistronic transcript, called the αPSM transcript. A fifth αPSM (Hld or the delta-toxin) is encoded within the regulatory RNA molecule RNAIII. Studies on the regulation of both Hla and the αPSMs have been largely centered around the agr system. AgrA has been shown to bind directly to the promoter region of the αPSM 60 transcript where it activates transcription, while hla translation is regulated by the agr effector RNAIII (113). Interestingly, it has also been demonstrated that the αPSMs can regulate Hla production in murine skin and lung models of infection (114), although the exact mechanism is unclear. Investigating the regulation of these toxins will improve our understanding of their relative contribution during human infection and may help in the development of “anti-virulence” approaches to combat infections caused by S. aureus.

Peptidyl-prolyl cis/trans isomerases (PPIases) are a family of enzymes that catalyze the cis-to-trans isomerization of proline peptide bonds. Proline is a unique amino acid in that it can exist in both the cis and trans isomerization state in vivo. Correct protein folding is often not possible when a proline peptide bond is in the incorrect configuration and therefore the isomerization rate of proline peptide bonds can be the rate-limiting step in protein folding (115). PPIase enzymes accelerate this isomerization and therefore assist in the regulation of proteins via a post-translational mechanism.

Numerous studies have identified bacterial PPIases that contribute to virulence (35, 50,

54, 60, 92). In addition to acting as foldases proteins, PPIases often have moonlighting roles and functions not limited to prolyl isomerization in the cell (61, 116, 117). S. aureus encodes three PPIase enzymes; trigger factor (TF), PrsA, and PpiB. Recent work in our lab (and others) has demonstrated that both PrsA and PpiB influence the activity of secreted virulence factors (57, 58, 117).

Initial work in our lab showed that a ΔprsA mutant has reduced phospholipase C

(PI-PLC) activity and decreased protease activity, and that a ΔppiB mutant has reduced hemolysis and nuclease activity (63). A follow-up study demonstrated that PpiB 61 contributes to virulence in a murine abscess model of infection, independently of its

PPIase activity (117). Recently, work by Lin et al. on the methicillin-sensitive S. aureus strain HG001, demonstrated that a ΔprsA mutant had better survival than WT in a murine systemic model of infection (58). They also concluded that there was altered abundance of 67 exoproteins and 163 cell wall associated proteins in a ΔprsA mutant, including the virulence factors surface protein A (SpA), immunodominant staphylococcal antigen B

(IsaB) and the αPSMs.

Due to the contribution of these two PPIases to the regulation of multiple virulence determinants, we hypothesized that PrsA and PpiB would contribute to virulence in a murine systemic model of infection using a community acquired methicillin resistant S. aureus (CA-MRSA) USA300 strain. In this study, we demonstrate that a USA300 ΔppiB mutant is attenuated in both an abscess and systemic model of infection but there is no significant attenuation with a ΔprsA mutant. The same virulence trend was observed during intracellular survival assays using macrophage and human nasal-epithelial cells. To understand the molecular mechanism underlying the attenuation of virulence in the ΔppiB mutant, we examined toxin production by measuring the hemolytic activity of culture supernatants against human and rabbit erythrocytes. We identify a significant reduction in the activity and secretion of the αPSMs in a ΔppiB mutant as well as a reduction in the activity of Hla in both the ΔprsA and ΔppiB mutants.

Immunoprecipitation analysis of PpiB and PrsA suggests a role for PpiB in the Sec secretion pathway and that PrsA is involved in cell-wall processing. Together, these data 62 suggest a role for two PPIase proteins in the regulation of S. aureus hemolytic toxins via two distinct mechanisms.

Results

PpiB is Required for Virulence in a Abscess and Systemic Model of Infection

Previously, we demonstrated that a ΔppiB mutant was attenuated for virulence in an abscess model of infection using the CA-MRSA strain USA300 TCH1516 (117).

Recently, Lin et al. demonstrated increased survival of a ΔprsA mutant in a sepsis model using the methicillin-sensitive S. aureus strain HG001 (58). To better understand the role of each PPIase protein, we performed both localized (abscess) and systemic (sepsis) murine infection models using the USA300 strain TCH1516 and isogenic PPIase deletion mutants. USA300 strains are the most frequent cause of community acquired MRSA infection in North America and are also a frequent cause of hospital associated MRSA

(HA-MRSA).

The murine abscess model of infection was performed as described previously

(103). Briefly, 6-week old BALB/c mice were injected in the lower right flank with 106

CFU of either WT, a ΔppiB mutant or a ΔprsA mutant strain. Following a 7-day infection, animals were sacrificed, abscesses were excised, and number of bacteria present was determined. The ΔppiB mutant displayed a reduced bacterial load in a skin abscess, confirming our previous finding. However, a ΔprsA mutant showed no apparent alteration in recovered bacteria (Figure 3.1A).

We next performed a murine systemic model of infection as described by Alonzo et al. (118). 6-week old BALB/c mice were injected with 107 CFU of each strain via the retro-orbital venous plexus and monitored for 3 days. Three of the eight ΔprsA infected mice died over the course of infection (Figure 3.1B) and the surviving mice were 64 monitored for change in weight (Figure 3.1C). Following 3 days of infection, mice were sacrificed, and the brain, lungs, heart, liver, spleen and kidneys were excised for processing. Each organ was homogenized, diluted, and plated to quantify the number of bacteria per organ. For ΔppiB infected mice, a significant reduction in bacterial burden was identified in the kidneys, heart, spleen, liver, and lungs (Figure 3.1D-I). A non- significant reduction in bacterial burden was also observed in the brain of the ΔppiB infected mice (p = 0.065), and overall a modest non-significant reduction in weight loss was also observed in the ΔppiB infected mice (Figure 3.1I&C). No difference in bacterial burden was observed in any of the organs from mice infected with the ΔprsA mutant

(Figure 3.1D-I). This is in contrast to the findings of Lin et al. which showed increased murine survival when infected with a ΔprsA mutant (58).

Figure 3.1. PpiB is required for virulence in a murine abscess (A) and systemic (B-I) model of infection. (A) Female 6-week old BALB/c mice were injected subcutaneously with wild-type S. aureus, a ΔppiB mutant or a ΔprsA mutant strain. The infection was allowed to proceed for 7 days. Mice were then sacrificed before abscesses were excised, homogenized, and diluted and plated to enumerate bacteria present in the abscesses. (B-H) Female 6-week old BALB/c mice were injected retro orbitally with wild-type S. aureus, a ΔppiB mutant or a ΔprsA mutant strain. The infection was allowed to proceed for 3 days. Mice were then sacrificed before organs were excised, wighed, diluted and plated to enumerate bacteria present in the organs. (B) Kaplan-Meyer curve for mouse survival during 3-day systemic infection. Three mice infected with the ΔprsA mutant strain died. (C) Percent initial body weight during 3-day systemic infection. The reduction in body weight in ΔppiB infected mice was less than that observed in wild type or ΔprsA infected mice, although the difference was not statistically significant. (D-I) CFU/g recovered from kidneys, heart, lung, spleen, liver, and brains of infected mice. In mice infected with the ΔppiB mutant strain, a significant reduction in bacterial numbers was detected in the kidneys, heart, and lungs of infected mice. Experiments were performed with an n=8 for each strain. Error bars represent standard deviation. Significance was determined by Student’s t test for the abscess model of infection (A). *, p< 0.05; and by a Mann- Whitney U-test for systemic organ data *, p< 0.05.

PpiB is Required for Survival Inside Macrophages and Human Nasal-Epithelial Cells

Although long considered an extracellular pathogen, abundant evidence now exists to demonstrate that S. aureus has the ability to reside inside of both professional phagocytic cells and non-professional phagocytic cells during infection. To further characterize the contribution of PrsA and PpiB to virulence, we examined the ability of isogenic mutants to survive in the intracellular environment. These experiments were performed in professional phagocytic cells Tohoku Hospital Pediatrics (THP1) human monocytes/macrophage), and non-professional phagocytic cells Roswell Park Memorial

Institute (RPMI) 2650, a human nasal epithelial cell line). Cells were infected at a MOI of 10 with either WT, the ΔppiB or the ΔprsA mutant strains. The three strains were used to infect THP-1 macrophages and RPMI 2650 cells and the number of intracellular bacteria determined at 2 h and 48 h. The ΔppiB mutant displayed a small but significant decrease (p = 0.0365) in recovery compared to WT in macrophages at 2 h (Figure 3.2B).

The ΔppiB mutant also had a significant decrease (p = 0.0018) in recovery compared to

WT in RPMI 2650 cells at 48 h (Figure 3.2C). These results are consistent with a ΔppiB mutant being attenuated for virulence in murine models of infection. The ΔprsA mutant showed no significant differences in recovery at any of the selected timepoints in both cell types (Figure 3.2A-D). 67

Figure 3.2. PpiB is required for survival in RPMI2650 nasal epithelial (A,C) and THP-1 macrophage cells (B,D). Gentamycin protection assays were performed at an MOI of 10 with WT, the ΔppiB or the ΔprsA mutant strains. After 2 and 48 hours of infection, bacterial cells were diluted and plated to enumerate surviving bacteria. Significance was determined by Student’s t test (A). *, p< 0.05; **, p<0.01.

Exoproteome Analysis Reveals Greater Alterations in Secreted Protein Abundance in a

ΔppiB Mutant than a ΔprsA Mutant

To investigate further why attenuation of virulence was observed in the ΔppiB mutant but not the ΔprsA mutant, we analyzed the secreted proteome of the WT, ΔppiB and ΔprsA mutant strains to identify proteins with altered abundance that may contribute to infection. Secreted proteins were detected via label-free mass spectrometry and data was analyzed by the Scaffold program and compared based on average normalized 68 abundance. Proteins with a minimum of a 2-fold change in abundance when compared to

WT and a p-value <0.05 (based on Student’s t-test) were considered for analysis. Based on these parameters, 86 proteins displayed altered abundance in culture supernatants of the ΔppiB mutant (Table 3.1). 26 were found in greater abundance in the ΔppiB mutant, while 60 were less abundant compared to WT. In contrast, only 16 proteins were identified as having significantly altered abundance in the secreted fraction of a ΔprsA mutant compared to WT. Of these, 6 were found at a higher level and 10 were less abundant in culture supernatants of the ΔprsA mutant (Table 3.2). Of note is that PpiB and PrsA peptides were detected in small amounts at 0.01 and 0.03 respectively in their isogenic mutant strains when compared to WT. We attribute this to artifact based on the technique detecting small peptides and not full-length proteins.

Proteins that made our cutoffs were grouped into categories based on known roles in the cell to identify common pathways or regulons with altered abundance (Figure 3.3).

Notably, 2 of the 16 proteins with altered abundance in a ΔprsA mutant were cell envelope proteins. One of these, penicillin binding protein 2 (Pbp2a), was found in greater abundance in culture supernatants of the ΔprsA mutant. Work by Jousselin et al. demonstrated that S. aureus PrsA was involved in oxacillin resistance via regulation of pbp2a, and that a ΔprsA mutant had a reduced amount of Pbp2a in the cell wall (57). We hypothesize that in the ΔprsA mutant, Pbp2a may have a defect in cell wall anchoring, which could account for the greater levels in the supernatant.

A large number of cytoplasmic proteins had decreased abundance in the exoproteome of a ΔppiB mutant. Interestingly, we did not see the same trend in a ΔprsA 69 mutant. Work by Wang et al. has recently demonstrated that S. aureus secretes numerous cytoplasmic proteins in extracellular vesicles to allow intracellular communication (119).

Notably, they show this secretion is mediated by the αPSMs, which promote the release of extracellular vesicles from the bacterial cell. This finding, coupled with our previous work that shows a defect in human erythrocyte lysis in a ΔppiB mutant (63, 117), led us to hypothesize that PpiB might be regulating secretion of the αPSMs. A reduction in the amount of αPSMs could explain why cytoplasmic proteins were found in less abundance in a ΔppiB mutant. 70

Figure 3.3. Exoproteome analysis reveals greater alterations in secreted protein abundance in a ΔppiB mutant than a ΔprsA mutant. (A) Culture supernatants from a ΔppiB mutant reveal there are 86 proteins with altered abundance when ppiB is deleted. (B) Culture supernatants from a ΔprsA mutant reveal there are 16 proteins with altered abundance when prsA is deleted. Proteins were grouped according to general function and the number of proteins falling into each functional category were plotted.

Table 3.1. Proteins with altered abundance in ΔppiB culture supernatants. Gene Protein Fold Functional Grouping Designation Name Change1

SAUSA300_1533 YdfA 25.51 Conserved hypothetical protein SAUSA300_0062 ArcB 23.74 Amino acid biosynthesis SAUSA300_0962 QoxB 18.74 Energy metabolism SAUSA300_0226 FadB 13.36 Fatty acid and phospholipid metabolism SAUSA300_2581 SasA 7.36 Cell envelope SAUSA300_1305 OdhB 4.93 Energy metabolism SAUSA300_0912 FabI 4.78 Fatty acid and phospholipid metabolism SAUSA300_1594 YajC 4.71 Protein fate SAUSA300_2061 AtpH 3.70 Energy metabolism SAUSA300_2060 AtpA 3.44 Energy metabolism SAUSA300_0547 SdrD 3.37 Cell envelope SAUSA300_2143 3.32 Conserved hypothetical protein SAUSA300_2178 RpoA 3.23 Transcription SAUSA300_2058 AtpD 3.13 Energy metabolism SAUSA300_0527 RpoB 2.87 Transcription SAUSA300_2059 AtpG 2.80 Energy metabolism SAUSA300_0528 RpoC 2.63 Transcription Cellular processes (includes toxins and SAUSA300_0194 MurP 2.49 virulence factors) SAUSA300_1565 2.40 Central intermediary metabolism SAUSA300_1685 2.28 Conserved hypothetical protein SAUSA300_2453 NcaC 2.23 Transport and binding proteins SAUSA300_2440 FnbB 2.17 Cell envelope SAUSA300_0724 2.10 Cell envelope SAUSA300_0514 CysE 2.08 Amino acid biosynthesis SAUSA300_0963 QoxA 2.05 Energy metabolism SAUSA300_2573 IsaB 2.00 Unknown function SAUSA300_1881 GatA 0.50 Protein synthesis SAUSA300_0691 SaeR 0.49 Regulatory functions SAUSA300_2469 SdaAA 0.49 Energy metabolism Purines, pyrimidines, nucleosides, and SAUSA300_0386 Xpt 0.48 nucleotides SAUSA300_1258 0.48 Energy metabolism SAUSA300_1293 LysA 0.48 Amino acid biosynthesis SAUSA300_0325 GcvH 0.48 Energy metabolism SAUSA300_1360 UbiE 0.48 Protein synthesis SAUSA300_0480 Pth 0.48 Protein synthesis Purines, pyrimidines, nucleosides, and SAUSA300_0716 0.48 nucleotides

Table 3.1 Cont. Purines, pyrimidines, nucleosides, and SAUSA300_2066 Upp 0.47 nucleotides Biosynthesis of cofactors, prosthetic SAUSA300_0492 FolP 0.47 groups, and carriers SAUSA300_2076 0.46 Central intermediary metabolism SAUSA300_1640 Icd 0.46 Energy metabolism SAUSA300_0841 0.46 Conserved hypothetical protein SAUSA300_2197 RplP 0.45 Protein synthesis SAUSA300_1443 RluB 0.45 Protein synthesis SAUSA300_1159 NusA 0.45 Transcription SAUSA300_1530 YbeY 0.44 Conserved hypothetical protein Biosynthesis of cofactors, prosthetic SAUSA300_0820 SufS 0.44 groups, and carriers Mobile and extrachromosomal element SAUSA300_1937 0.44 functions SAUSA300_1049 MurI 0.43 Cell envelope SAUSA300_1679 AcsA 0.43 Central intermediary metabolism SAUSA300_1178 RecA 0.42 DNA metabolism Cellular processes (includes toxins and SAUSA300_1269 FemA 0.42 virulence factors) SAUSA300_1882 GatC 0.41 Signal transduction Biosynthesis of cofactors, prosthetic SAUSA300_1614 HemL1 0.41 groups, and carriers SAUSA300_0067 0.41 Unknown function Biosynthesis of cofactors, prosthetic SAUSA300_1634 CoaE 0.40 groups, and carriers SAUSA300_1288 DapA 0.40 Amino acid biosynthesis SAUSA300_1478 0.37 Cell envelope SAUSA300_1285 0.35 Transport and binding proteins Purines, pyrimidines, nucleosides, and SAUSA300_0971 PurL 0.35 nucleotides SAUSA300_0368 RpsR 0.33 Protein synthesis SAUSA300_1357 AroC 0.33 Amino acid biosynthesis SAUSA300_0919 MurE 0.32 Cell envelope SAUSA300_1156 ProS 0.32 Protein synthesis SAUSA300_0753 0.30 Conserved hypothetical protein SAUSA300_0741 UvrB 0.29 DNA metabolism SAUSA300_0692 SaeQ 0.27 Conserved hypothetical protein SAUSA300_1523 0.27 Conserved hypothetical protein Purines, pyrimidines, nucleosides, and SAUSA300_2526 PyrD 0.26 nucleotides SAUSA300_0364 YchF 0.26 Unknown function SAUSA300_1144 TrmFO 0.24 Unknown function 73

Table 3.1 Cont. SAUSA300_1861 0.24 Conserved hypothetical protein SAUSA300_1007 0.24 Unknown function SAUSA300_0329 0.24 Unknown function SAUSA300_0732 0.23 Conserved hypothetical protein SAUSA300_0538 0.23 Energy metabolism SAUSA300_2251 0.22 Energy metabolism Cellular processes (includes toxins and SAUSA300_2025 RsbU 0.22 virulence factors) SAUSA300_2525 0.21 Conserved hypothetical protein SAUSA300_2510 0.20 Conserved hypothetical protein SAUSA300_2312 Mqo 0.18 Energy metabolism SAUSA300_2296 0.17 Unknown function SAUSA300_0737 SecA1 0.10 Protein fate SAUSA300_2477 CidC 0.09 Energy metabolism SAUSA300_1711 PutA 0.05 Energy metabolism SAUSA300_2125 0.05 Transport and binding proteins SAUSA300_0857 PpiB 0.01 Conserved hypothetical protein

1 Fold change is based on comparing abundance of proteins in ΔppiB/WT. Fold change >1 is indicative of an increase in abundance in ΔppiB culture supernatants. Fold change <1 is indicative of an decrease in abundance in ΔppiB culture supernatants.

Table 3.2. Proteins with altered abundance in ΔprsA culture supernatants. Gene Protein Fold Functional Grouping Designation Name Change1 SAUSA300_1018 11.13 Conserved hypothetical protein SAUSA300_0062 ArcB 7.92 Amino acid biosynthesis SAUSA300_2052 2.96 DNA metabolism SAUSA300_1606 2.63 Conserved hypothetical protein SAUSA300_0963 QoxA 2.23 Energy metabolism SAUSA300_1341 Pbp2 2.06 Cell envelope Central intermediary SAUSA300_0318 NanE 0.50 metabolism SAUSA300_1763 EpiP 0.44 Protein fate Mobile and extrachromosomal SAUSA300_1937 0.44 element functions SAUSA300_2082 RpoE 0.42 Transcription SAUSA300_0923 HtrA2 0.38 Protein fate SAUSA300_0279 EsaA 0.37 Cell envelope SAUSA300_2032 KdpC 0.32 Transport and binding proteins Fatty acid and phospholipid SAUSA300_0226 FadB 0.31 metabolism Mobile and extrachromosomal SAUSA300_1934 0.30 element functions SAUSA300_1790 PrsA 0.03 Protein fate 1 Fold change is based on comparing abundance of proteins in ΔprsA/WT. Fold change >1 is indicative of an increase in abundance in ΔprsA culture supernatants. Fold change <1 is indicative of an decrease in abundance in ΔprsA culture supernatants.

A ΔppiB Mutant has Reduced PSM Production

Our previous publications demonstrated that culture supernatants from a ΔppiB mutant were less hemolytic for human blood than supernatants from the wild type (63,

117). Although the reduction in hemolytic activity was clear, the specific toxin responsible for this phenotype was not identified. Based on the results of the exoproteome analysis above, and the facts that the primary toxins responsible for human erythrocyte lysis are the αPSMs and that Hla activity against human erythrocytes has been shown to be negligible (Figure 3.4, (21)), we hypothesized that there is decreased 75 abundance (or activity) of αPSMs in a ΔppiB mutant. To test this hypothesis, we performed a butanol extraction to isolate the αPSM peptides from S. aureus culture supernatants (120). This experiment allows visualization of lysis mediated exclusively via the PSMs, as they are the only toxin that will be extracted from culture supernatants, and thus, no other hemolysins will be present. The resulting extracted peptides were used in a hemolysis assay and analyzed by SDS-PAGE to visualize peptide abundance (Figure

3.5). Extracts from a ΔppiB mutant displayed a modest (1.5-fold) yet significant reduction in hemolysis when compared to WT (Figure 3.5A). No reduction in hemolytic activity was observed in peptide extracts taken from the ΔprsA mutant or a hla mutant

(negative control). SDS-PAGE analysis of extracts revealed a band corresponding to the size to the PSM peptides in each of the samples that was reduced in a ΔppiB mutant.

(Figure 3.5B). Together these results demonstrate a reduction in PSM abundance when ppiB is absent from the cell confirming that PpiB influences PSM production.

Figure 3.4. The αPSMs are the primary toxins responsible for human erythrocyte lysis while. Hla is the primary toxin active against rabbit erythrocytes. (A) A decrease in hemolytic activity against human erythrocytes was observed using culture supernatants from an αPSM mutant strain. (B) A decrease in hemolytic activity against rabbit erythrocytes was observed using culture supernatants from an hla mutant. Significance was determined by Student's t test. **** p < 0.001; ** p < 0.01; ns, not significant.

Figure 3.5. Decreased αPSM production in a 횫ppiB mutant. (A) Hemolytic activity of butanol-extracted peptides. Butanol-extracted samples were resuspended in water and incubated with whole human blood. The deletion of ppiB results in a significant decrease in hemolytic activity in comparison to WT. (B) SDS- PAGE analysis of butanol extracted samples from panel A. PSM levels are reduced in a 횫ppiB mutant. Significance was determined by Student’s t test. *, p < 0.05.

The ΔppiB and ΔprsA Mutants Both Display Reduced Hemolysis of Rabbit Erythrocytes

The αPSMs have been shown to regulate Hla by delaying transcription and subsequently, Hla secretion (114). Consequently, we hypothesized that alpha-toxin activity would also be altered in a ΔppiB mutant. Previously, we demonstrated that the

ΔppiB mutant displayed a slight reduction in hemolytic activity against rabbit erythrocytes, however, results were not deemed significant (117). To better test the activity of Hla in a ΔppiB mutant we repeated the hemolysis assay over a time course of

20 minutes, with measurements taken every 5 minutes to determine the degree of lysis.

Interestingly, culture supernatants from both a ΔppiB and a ΔprsA mutant displayed reduced hemolysis at earlier time points (Figure 3.6A) when compared to wild type. A representative time point at 10 minutes displayed a significant reduction in rabbit erythrocyte lysis in both the ΔppiB and ΔprsA mutants, although not to the levels of a hla mutant strain (Figure 3.6B). At later timepoints (>20 mins) the degree of hemolytic activity in the ΔppiB and ΔprsA mutant samples increased to a level similar to that of the wild type. This result explains our previous observation of a modest, non-significant reduction in hemolytic activity in the ΔppiB mutant (117).

Taken collectively, these data suggest that both PpiB and PrsA are involved in the regulation of Hla. To investigate if this regulation occurred at the level of transcription or post-trancriptionally, RT-qPCR was performed to examine hla and αPSM transcript levels in the WT, ΔppiB and ΔprsA strains. No significant differences were detected in either transcript indicating that the regulation is likely post transcriptional (Figure 3.7).

We postulate that the reduction in Hla in a ΔppiB mutant could be due to the reduction of 79

αPSMs in culture supernatants. This supports the findings of Berube et al. in which they demonstrate that the αPSMs regulate Hla production during infection (114). Since no significant reduction in PSM activity was observed in a ΔprsA mutant, we hypothesize that PrsA is regulating Hla independently of the PSMs. One explanation for why a ΔprsA mutant may have reduced cytolytic activity against rabbit erythrocytes is if the PPIase activity of PrsA is required to help assist in the folding of Hla. If this is the case, deleting prsA could lead to less active Hla in culture supernatants.

Figure 3.6. A ppiB and prsA mutant display reduced hemolysis in rabbit blood. Rabbit erythrocyte lysis assays were performed using S. aureus culture supernatants and whole rabbit blood. (A) A decrease in hemolytic activity against rabbit erythrocytes over time was observed using culture supernatants from ppiB, prsA and hla mutant strains. (B) A representative time point at 10 minutes reveals a significant reduction in hemolysis in the ppiB and prsA mutant strains. The data presented in A and B are the averages of 4 replicates. Significance was determined by Student’s t test. ****, p < 0.001; ***, p < 0.005. 81

Figure 3.7. RT-qPCR analysis of hla and PSM transcript levels in, W.T.; ΔppiB, and ΔprsA strains.

In vivo Immunoprecipitation Identifies Greater Abundance of PpiB Target Proteins than

PrsA

Previous work in our lab has demonstrated that PpiB contributes to S. aureus virulence independently of its PPIase activity, suggesting an alternative function/activity for this protein (117). To further investigate this alternative function, and to elucidate how PpiB and PrsA independently regulate S. aureus toxins, we performed an immunoprecipitation experiment to identify proteins that interact with PpiB and PrsA.

The ppiB gene was amplified, HA-tagged, and cloned into the multicopy plasmid pMK4 under the control of its native promoter (plasmid pRKC0131). To identify proteins interacting with PpiB, pRKC0131 (ppiB-HA) was transduced into the ΔppiB mutant strain, and immunoprecipitation performed using anti-HA magnetic beads. 82

Immunoprecipitation experiments were performed by formaldehyde crosslinking proteins in cultures that had been grown for 3 hours to mid-exponential phase. A negative control strain containing the empty pMK4 vector in the ΔppiB mutant was used for comparison. Magnetic beads containing PpiB-HA (and associated proteins) were collected and analyzed via mass-spectrometry to identify all peptides associated with

PpiB. A similar analysis was performed for PrsA using a His-tagged prsA construct in the same pMK4 plasmid (plasmid pRKC0126). Anti-His magnetic beads were analyzed and compared to an empty vector control in the ΔprsA mutant strain.

Data were analyzed using the Scaffold program and compared via the exclusive spectrum count of each protein identified in the samples. Proteins with a minimum of 3- fold increase in abundance (in the tagged strain compared to the empty vector control) were selected for further analysis. We also excluded any data where the average spectrum count was > 10 for both the overexpresser and empty vector strains. Surprisingly neither the αPSMs nor Hla were identified in the immunoprecipitation assay, suggesting the role of PpiB and PrsA in regulating these toxins is indirect (see below).

As predicted (based on its cellular location), PrsA was found to interact with a large number of cell-wall anchored proteins (Table 3.3). This included the signal peptidase I (SpsB) which assists in the anchoring of cell-wall proteins by cleaving the N- terminal signal peptide in the general secretory pathway. Work by Schallenberger et al. demonstrated that inhibition of SpsB with arylomycin led to increased abundance of the proteins PrsA, HtrA and SAOUHSC_01761 (121). They suggested the reason for increased secretion of these proteins could have been as a response to the inhibition of 83

SpsB (28). Notably, both htrA and SAUSA300_1606 (SAOUHSC_01761 homologue in

TCH1516) both have altered abundance in the secreted fraction of a ΔprsA mutant (Table

3.3). HtrA is a highly conserved protein in Gram-positives where it functions as a chaperone and a protease (122, 123). SAUSA300_1606 is an uncharacterized protein and relatively little is known about it other than its induction during vancomycin treatment

(124). How these proteins might be interacting in the cell remains unknown, but our findings support a connection between the anchoring of PrsA to the cell wall and its connection to antibiotic resistance.

The largest fold enrichment for a protein identified in the PpiB immunoprecipitation analysis (after PpiB itself) was for the Sec translocase subunit

SecA1 (8.5-fold enrichment, Table 3.4). SecA is the ATPase that facilitates protein translocation through the SecYEG translocon, in the general secretory pathway (125). In

Escherichia coli, the chaperone protein SecB transports newly synthesized proteins from the ribosome to SecA for secretion (126). Interestingly, S. aureus does not encode a SecB homologoue, and to date it is unknown if any alternative chaperone substitutes for SecB in the secretion process (125). As previously stated many PPIases also function as chaperone proteins (31, 41, 127). This, in addition to our recent finding that PpiB possesses a PPIase-independent activity (117), and the data shown here of an association between PpiB and SecA, suggests that PpiB may function as a chaperone in the general secretion system. If so, this could account for the reduced activity of secreted virulence factors, such as Hla and nuclease (Nuc), in a ΔppiB mutant (63).

Table 3.3. Proteins interacting with PrsA. Gene Protein Fold Protein Description Designation Name Change1 SAUSA300_1790 PrsA 15.61748634 Foldase protein PrsA SAUSA300_1512 Pbp3 9.75 Penicillin-binding protein 3 D-alanine-activating SAUSA300_0838 DltD 7.25 enzyme/D-alanine-D-alanyl, dltD protein SAUSA300_0958 LcpB 6 Transcriptional regulator Probable cell wall amidase SAUSA300_1588 LytH 6 LytH Probable quinol oxidase SAUSA300_0963 QoxA 5.636363636 subunit 2 Uncharacterized leukocidin- SAUSA300_1974 LukB 4.833333333 like protein 1 SAUSA300_0032 MecA 4.5 Penicillin-binding protein 2 Putative transcriptional SAUSA300_2259 LcpC 4.454545455 regulator SAUSA300_0274 4.285714286 Uncharacterized protein SAUSA300_0419 3.8125 Uncharacterized lipoprotein Iron compound ABC SAUSA300_2136 HtsA 3.705882353 transporter, iron compound- binding protein SAUSA300_1982 GroL 3.702702703 60 kDa chaperonin Probable quinol oxidase SAUSA300_0962 QoxB 3.608695652 subunit 1 Putative phage infection SAUSA300_2578 3.6 protein AcrB/AcrD/AcrF family SAUSA300_2213 3.5 protein SAUSA300_2328 3.333333333 Uncharacterized protein SAUSA300_2092 Dps 3.142857143 General stress protein 20U SAUSA300_2100 3.133333333 Lytic regulatory protein SAUSA300_0992 3.090909091 Putative lipoprotein SAUSA300_2144 3.083333333 Uncharacterized protein SAUSA300_0868 SpsB 3 Signal peptidase I SAUSA300_0279 EsaA ∞ Putative membrane protein N-acetylmuramoyl-L-alanine SAUSA300_2579 LytZ ∞ amidase domain protein 1 Fold change is based on proteins found in association with PrsA in comparison to an empty-vector control. Fold chance >1 indicates greater abundance in PrsA immunoprecipitation samples. Fold change = ∞ indicates protein not detected in negative control.

Table 3.4. Proteins interacting with PpiB. SAUSA300 Gene Protein Fold Protein Description Number Name Change1

Putative peptidyl-prolyl cis-trans SAUSA300_0857 PpiB 31.5 isomerase Protein translocase subunit SecA SAUSA300_0737 SecA1 8.5 1 SAUSA300_1027 RpmF 7.8 50S ribosomal protein L32 Immunoglobulin-binding protein SAUSA300_2364 Sbi 7.5 sbi SAUSA300_1535 RpsU 6.25 30S ribosomal protein S21 SAUSA300_1178 RecA 5.75 Protein RecA SAUSA300_0220 PflB 5.5 Formate acetyltransferase Aerobic glycerol-3-phosphate SAUSA300_1193 GlpD 4.88 dehydrogenase ATP-dependent 6- SAUSA300_1645 PfkA 4.83 phosphofructokinase SAUSA300_0757 Pgk 4.83 Phosphoglycerate kinase SAUSA300_0798 4.63 Lipoprotein Inosine-5’-monophosphate SAUSA300_0388 GuaB 4.5 dehydrogenase SAUSA300_1525 GlyQS 4.18 Glycine--tRNA ligase ATP-dependent zinc SAUSA300_0489 FtsH 3.86 metalloprotease FtsH SAUSA300_2067 GlyA 3.75 Serine hydroxymethyltransferase Aspartyl/glutamyl-tRNA SAUSA300_1880 GatB 3.63 (Asn/Gln) amidotransferase subunit B SAUSA300_0491 CysK 3.63 Cysteine synthase SAUSA300_1150 Tsf 3.62 Elongation factor Ts GMP synthase [glutamine- SAUSA300_0389 GuaA 3.56 hydrolyzing] SAUSA300_1684 3.25 Uncharacterized protein SAUSA300_0533 Tuf 3.09 Elongation factor Tu SAUSA300_0693 SaeP 3.07 Putative lipoprotein 6-phosphogluconate SAUSA300_1459 Gnd 3.06 dehydrogenase, decarboxylating SAUSA300_0496 LysS 3 Lysine--tRNA ligase Ribonucleoside-diphosphate SAUSA300_0716 NrdE ∞ reductase SAUSA300_2251 ∞ Dehydrogenase family protein Glutamyl-tRNA (Gln) SAUSA300_1881 GatA ∞ amidotransferase subunit A 86

SAUSA300_1586 AspS ∞ Aspartate--tRNA ligase Polyribonucleotide SAUSA300_1167 Pnp ∞ nucleotidyltransferase SAUSA300_0009 SerS ∞ Serine--tRNA ligase SAUSA300_1629 ThrS ∞ Threonine--tRNA ligase SAUSA300_2214 FemX ∞ Lipid II:glycine glycyltransferase 1Fold change is based on proteins found in association with PpiB in comparison to an empty-vector control. Fold chance >1 indicates greater abundance in PpiB immunoprecipitation samples. Fold change = ∞ indicates protein not detected in negative control

Discussion

In bacteria, PPIase enzymes are traditionally studied for their roles in protein

(re)folding. In Gram-positive bacteria, parvulin-type PPIases (including PrsA) are often anchored on the external leaflet of the cell membrane, and are thought to fold secreted proteins after they have been translocated out of the cell. S. aureus PrsA appears to function in a similar manner. A prsA mutant strain has decreased proteolytic and phospholipase activity (63), increased sensitivity to -lactam antibiotics (57) and decreased hemolytic activity (this study). We hypothesize that all of these defects arise as a result of defective protein folding in the absence of PrsA. This idea is supported by a number of observations in our proteomic and immunoprecipitation data analysis. First, the number of secreted proteins displaying altered levels in prsA mutant culture supernatants was relatively low (16 proteins, Table 3.2). This suggests that PrsA targets are present in these supernatants, however, they may exhibit reduced activity due to incorrect folding (Figure 3.8A). Second, the majority of proteins found to interact with

PrsA by immunoprecipitation are localized to the cell envelope and play important roles in cell wall synthesis and stability. Of particular interest was the identification of Pbp2a in the immunoprecipitation assay. This result supports the findings of Jousselin et al. and validates the approach used in this study to identify proteins that interact with PrsA (57).

Interestingly, EsaA, a component of the recently identified S. aureus type 7 secretion system, was identified in the immunoprecipitation assay as interacting with PrsA and was also found at lower levels in the culture supernatant. This may indicate a potential role for PrsA in type 7 secretion. 88

An unexpected finding in this study was that PrsA did not contribute to virulence in either a murine abscess or sepsis model of infection. This result conflicts with the results obtained by Lin et al., who demonstrate attenuation of virulence in a HG001 prsA mutant (58). One possible explanation for the discrepancy in results observed is the strain background used in each study. HG001, the strain used by Lin et al., is a methicillin sensitive strain derived from NCTC8325 (with the rsbU gene repaired) (58).

The strain used in this study, TCH1516, is a methicillin resistant USA300 strain.

USA300 strains typically produce high levels of toxins and previously it was shown that a USA300 strain is more hemolytic than HG001 (128). We speculate that, in USA300 strains, the high level of toxin production and overlapping activities of S. aureus toxins may negate the requirement for PrsA activity when it comes to causing disease in a mouse model of infection. In strains with relatively lower levels of toxin production

(such as HG001), the loss of PrsA activity may have a more dramatic effect and result in attenuation. To test this hypothesis, we are currently broadening our investigations and exploring the role of PrsA in a variety of S. aureus strains.

While the role of PrsA in S. aureus appears to be similar to that of homologues in other Gram-positive bacteria (such as Listeria monocytogenes), the mechanism of action of PpiB remains elusive. PpiB clearly plays an important role in infection, as a ppiB mutant is attenuated in abscess and sepsis models of infection, and displays reduced intracellular survival. The activity and abundance of several virulence factors (including nuclease, alpha toxin, and the PSMs) is reduced in ppiB culture supernatants, which explains the attenuation of virulence, however the mechanism through which PpiB 89 functions is not clear. We previously demonstrated that PpiB (i) is found in the cytoplasm, (ii) possesses PPIase activity, and (iii) its PPIase activity was dispensable during infection (63, 117). Taken together, this implies that PpiB functions inside the cell, in a PPIase-independent manner, to influence virulence factor production. Our first indication of a potential biological role for PpiB comes from the secreted proteomic and immunoprecipitation data presented herein. SecA, a component of the general secretion pathway in S. aureus was found to interact with PpiB and was also found at significantly reduced levels in ppiB mutant culture supernatants. In E. coli, during general secretion, a chaperone protein called SecB binds to newly synthesized proteins and delivers them to

SecA for translocation. No SecB homologue has been identified in S. aureus, therefore, based on our data, we hypothesize that one potential role for PpiB could be to functionally compensate for SecB and chaperone proteins prior to secretion (Figure

3.8A). Thus, the loss of PpiB could result in general secretion defects and/or the secretion of misfolded proteins. This might explain why so many proteins had altered abundance in the culture supernatants of ppiB mutant strains compared to prsA mutants

(86 proteins altered in ppiB mutant, 16 in prsA mutant). Another role of chaperone proteins is to prevent the aggregation of misfolded proteins under stress (47). If PpiB is functioning as a chaperone, it could additionally function to protect the αPSM peptides within the S. aureus cell. If this is the case, it could explain why there are less active

αPSMs in culture supernatants of a ΔppiB mutant.

S. aureus, like many other cells, has the ability to produce and secrete extracellular vesicles (EVs) (119). Recently, the formation of EV’s in S. aureus has been 90 shown to be enhanced by the αPSMs (119). EVs are loaded with proteins, including many proteins typically thought of as being cytoplasmic. The presence of EVs in culture supernatants may explain why cytoplasmic proteins are commonly identified during proteomic studies of secreted fractions, which we also observed in our proteomic data set.

Interestingly, of the 86 proteins that were altered in abundance in ppiB mutant supernatants, a large proportion of them are considered cytoplasmic. These proteins were present in wild type supernatants, but decreased in ppiB mutant supernatants. It is possible that PpiB plays a role in EV biogenesis and that in the mutant there is reduced

EV production, which accounts for the variation in cytoplasmic proteins. Since a ppiB mutant produces less αPSMs which are necessary for EV release, it is also possible that the same number of EV’s are being generated in a ΔppiB mutant but they are unable to be released from the cell due to the decrease in αPSMs (Figure 3.8B). We are currently investigating EV production in a ppiB mutant.

In conclusion, while the role of PpiB in pathogenesis is clear, its molecular mechanism of action remains uncertain. The results generated in this study further our understanding of S. aureus virulence factors regulated by PpiB and suggest that PpiB may be functioning in a secretion-related manner. Further studies are being conducted to investigate this novel protein to better understand how it regulates virulence factor production and virulence. 91

Figure 3.8. Schematic diagram of proposed mechanisms of PpiB and PrsA. (A) PpiB functioning as a chaperone to deliver nascent proteins to the Sec secretion machinery. Once secreted, proteins begin to fold and the membrane-anchored PrsA, which acts as a PPIase, assists this process. (B) PpiB positively regulates the αPSMs. When active αPSMs are within the cell, they promote the release of EV’s from the membrane. Image created with BioRender.

Materials and methods

Strains and Strain Construction

All bacterial strains and plasmids are listed in Table 3.5 and oligonucleotides in

Table 3.6. The ΔppiB and ΔprsA mutant strains were constructed via allelic exchange as previously described (63, 101, 117). For overexpresser strains used in the immunoprecipitation, plasmids pRKC0131, pRKC0126 and pMK4 (empty vector) were transduced into RKC0323 (ΔppiB) and RKC0085 (ΔprsA) respectively. A Δαpsm strain was constructed by allelic exchange using plasmid pJB38 (86). DNA sequences flanking the αPSM transcript were amplified using primer pairs 490/644 and 645/493 and a third sequence containing the erythromycin cassette from the bursa aurealis transposon was amplified using primers 638 and 639 (86). These three fragments were cloned to generate plasmid pRKC0674. This plasmid was recombined onto the S. aureus chromosome and excised to make the deletion strain according to previously published protocol (101).

Bacterial Growth Conditions

S. aureus cultures were grown in tryptic soy broth (TSB) shaking at 37°C. Where appropriate, the antibiotic chloramphenicol was used at the concentration of 5 μg ml-1.

For analysis of culture supernatants, S. aureus cultures were synchronized as follows.

Replicate overnight cultures were grown in 5 ml of TSB for 15 hours. The next day, cultures were diluted 1:100 in 10 ml of pre-warmed TSB and grown for 3 hours to mid- exponential phase. The 3-hour cultures were then diluted into 25 ml of TSB in 250 flasks and normalized to an optical density at 600 nm (OD600) of 0.05. Resulting flasks were grown overnight for 15 hours. 93

Murine Abscess Model of Infection

A subcutaneous abscess infection was performed as described by us previously

Keogh et al. (117). Cultures were grown for 2.5 hours in TSB to an OD600 of 0.75.

Resulting bacterial cells were pelleted via centrifugation and resuspended in sterile phosphate-buffered saline (PBS) to prepare an inocula of 106 CFU/50 μl. Each inocula was then confirmed via serial diluting and plating. Six-week old female BALB/c mice were anesthetized by isoflourane inhalation before being shaved at the right flank and treated with Nair to remove fur. Mice were then injected with 50 μl of their respective strains and the infection was allowed to proceed for 7 days. Following 7 days of infection, mice were euthanized with CO2 and abscesses were excised and homogenized.

Homogenates were serial diluted and plated onto TSB to count recovered CFU/ abscess.

Murine Systemic Model of Infection

A systemic model of dissemination was chosen to mimic septic infection and cultures were prepared as previously described by Spaan et al. (88). Cultures were grown for approximately 2.5 hours in TSB to an OD600 of 0.75. Resulting bacterial cells were pelleted via centrifugation and resuspended in sterile phosphate-buffered saline (PBS) to prepare an inocula of 107 CFU/100 μl. 6-week old BALB/c mice were injected retro- orbitally with 100 μl of bacterial cultures and the infection was allowed to proceed for 3 days. Following 3 days of infection, mice were euthanized with CO2 and the brain, lungs, heart, liver, kidneys and spleen were harvested. Each organ was weighed and homogenized before serial dilution and plating was conducted to quantify recovered bacteria/ organ. 94

Macrophage Infection and Cell Differentiation

THP-1 infection assays were performed as outlined in Carroll et al. (104).

Macrophages were seeded at a density of 2 x 105 macrophages/well in a volume of 500 µl of RPMI 1640 with 10% FBS and 1% penicillin/streptomycin. A multiplicity of infection

(MOI) of 10 was used to infect each well (2 x 106 bacteria/well). Strains were prepared by taking 250 µl of overnight culture grown at 37°C and inoculating it into a 250 ml flask containing 25 ml TSB. The bacteria were then grown in a shaking 37°C incubator until their optical density (OD600) reached 0.4 – 0.6. The volume of bacteria needed to perform the infection was then transferred into a 1.5 ml centrifuge tube and pelleted for 20 min at

3000 rpm. The supernatant was removed and discarded. The pellet was then washed with

500 µl of phosphate buffered saline (PBS) and pelleted for 20 min at 3000 rpm. The pellet was resuspended in 20% human serum and 80% PBS. This solution was then incubated in a 37°C water bath for 30 min to opsonize the bacteria. Opsonized bacteria were resuspended into the correct volume of 1640 RPMI medium with 10% FBS.

The macrophages were washed twice with 37°C prewarmed PBS. Each well received 500 µl of the opsonized bacteria resuspended in 1640 RPMI with 10% FBS.

Plates were centrifuged at 1000 rpm for 10 min at room temperature and placed into a

37°C incubator supplemented with 5% CO2 for 60 min. After 60 min, the medium was aspirated, and the wells were washed once with prewarmed PBS as described above. To each well, 500 µl of RPMI 1640 with 10% FBS and 30 µg/ml of gentamicin was added.

Cells were incubated in a 37°C incubator supplemented with 5% CO2 for 60 min. The end of this 60 min incubation marked hour 0 of intracellular. After this incubation, the 95 medium was aspirated and replaced with 500 µl of RPMI 1640 with 10% FBS and 5

µg/ml of gentamicin for the rest of the infection.

Intracellular bacteria were harvested from the macrophages at 2 h and 48 h representing initial and late stages of infection. Medium was aspirated, and the cells were washed twice with prewarmed PBS as described above. A 500 µl aliquot of 0.5% Triton

X-100 in PBS was added to each well to lyse the macrophages and release the intracellular bacteria. The lysates for each time point were serially diluted in PBS in sterile 96 well plates. A volume of 100 µl of each of the diluted lysate was plated on TSA plates. Plates were incubated in a 37°C incubator and colonies were counted to determine the number of recovered intracellular bacteria and calculated to determine CFU/ml.

Nasal Epithelial Cell Infection

Strains were prepared and washed in the same way as described above except the bacteria were not opsonized. After washing the bacterial pellet with PBS, the bacteria were resuspended into the correct volume of EMEM with 10% FBS. RPMI 2650 cells were seeded at the same density and volume as described above except using EMEM with 10% FBS and 1% penicillin/streptomycin. Cells were washed twice with prewarmed

PBS and infected with bacteria at an MOI of 10 (2 x 106 bacteria/well).

Exoproteome Analysis

Bacterial strains were grown in triplicate in 5 ml overnight cultures shaking at

37°C. The next day, cultures were diluted 1:100 in 10 ml of TSB and incubated for 3 hours shaking at 37°C. After 3 hours, the OD600 of each culture was measured and each sample was normalized to an OD600 of 0.05 in 100 ml of TSB. After 15 hours of growth 96 in the flask, samples were split into 2 x 50 ml tubes and centrifuged at 3,000 RPM for 15 min. Cell-free supernatants were harvested and filter sterilized through a 0.45 μm filter disk to ensure all bacterial cells were removed from the sample. 10 ml of TCA was added to the culture supernatants and incubated overnight at 4°C. The following day samples were centrifuged at 11,000 RPM for 10 min and the supernatant was removed. Resulting pellets containing concentrated proteins were washed with ice-cold acetone and sent to the University of Nebraska Lincoln Proteomics Facility for mass spectrometry analysis.

TCA precipitated samples were dissolved in a solution of 7M urea, 2M thiourea, 5 mM DTT, 0.1 M Tris, pH 8 and 1 x PhosStop by gentle shaking for 1.5 h at 24°C. After reduction the samples were alkylated for 30 min using a 3-fold molar excess of iodoacetamide. The protein concentration was determined using the CB-X protein assay kit. 100 μg of protein in 10 μl of the urea solution was diluted and digested with 2 μg trypsin (1:50 enzyme:substrate ratio) for 16 h overnight, then an additional 1 μg of trypsin was added and digestion continued for a further 4 h. 200 ng of each of the 9 samples was run by nanoLC-MS/MS using a 2 h gradient on a 0.075 mm x 250 mm C18

Waters CSH column feeding into a Q-Exactive HF mass spectrometer.

Raw datafiles were loaded directly in to Progenesis QI for proteomics version 2.0 and alignment of the chromatograms showed a ≥95% match. Features (peaks) were extracted across all runs and areas under the peaks calculated. Runs were normalized using the “normalize to all proteins” setting. Data were exported from Progenesis and analyzed using the search engine Mascot (Matrix Science, London, UK; version 2.6.1).

Mascot was set up to search 2 databases: the common contaminants database 97 cRAP_20150130 database (117 entries) and a combined database containing 2 uniprot reference proteomes for S. aureus NCTC8325 & C0673_20170607 (5705 entries) assuming the digestion enzyme trypsin. Mascot was searched with a fragment ion mass tolerance of 0.060 Da and a parent ion tolerance of 10.0 PPM. Oxidation of methionine; deamidated of asparagine and glutamine; and carbamidomethyl of cysteine were specified in Mascot as variable modifications.

Butanol Extraction of PSMs

PSM peptides were isolated as previously described (120). 5 ml of synchronized, cell-free culture supernatants were incubated shaking at 37°C with 3 ml of n-butanol for 2 h. Cultures were then centrifuged and 1 ml of the organic layer was removed for vacuum centrifugation for 12 h at 5,000 rpm.

Human-Erythrocyte Hemolysis Assay

Synchronized bacterial cultures were grown in quadruplicate as described in bacterial growth conditions for 15 h. Samples were centrifuged and cell-free culture supernatants were harvested and diluted 1:2 in reaction buffer containing 40 mM CaCl2, and 1.7% NaCl. 25 µl of whole-human blood was added to samples and they were incubated at 37°C while rotating. After 10 min, samples were centrifuged at 5,500 x g and the resulting supernatant was transferred to a 96-well plate. The degree of erythrocyte lysis was determined by reading sample absorbance at an OD543.

Rabbit Erythrocyte Hemolysis Assay

Bacterial strains were grown in quadruplicate in a 25 ml flask of TSB as described above. Cultures were then centrifuged to obtain the cell-free supernatant. Supernatants 98 were diluted 1:10 before being further diluted 1:2 in reaction buffer containing 40 mM

CaCl2, and 1.7% NaCl and incubated in a 37°C water bath with 125 µl of rabbit blood.

200 µl of samples were removed every 5 min after incubation and centrifuged at 5,500 x g. 100 µl of supernatant was transferred to a 96-well plate and erythrocyte lysis was determined by reading the absorbance of the samples at OD543.

Protein Immunoprecipitation Assay

Duplicate bacterial cultures were inoculated into 100 ml of TSB with appropriate antibiotics in 250 ml flasks for 3 h to mid-exponential phase. Samples were centrifuged at

3,000 rpm for 20 min and pellets were resuspended in sterile PBS. 37% formaldehyde was added to cultures to a final concentration of 1% and a 20 min shaking incubation was performed at room temperature. The reaction was quenched with the addition of glycine to a final concentration of 200 mM. Cells were centrifuged and the remaining pellets were washed with sterile PBS and then re-suspended in sterile water. Cells were then lysed with lysostaphin at a concentration of 10 mg/ml and incubated at 37°C for 30 min.

Following this, cells were treated with DNase I and incubated at 37°C for 30 min.

Resulting lysates were then sonicated at 20 % amplitude, 2 x 15 sec. Samples were then centrifuged at 13,000 rpm for 1 min and supernatants were used for immunoprecipitation.

For immunoprecipitation, anti-HA or anti-His magnetic beads were added to respective samples and incubated at 4°C for 1 h. Beads were collected with a magnetic rack and washed 10 times with 1X PBS. Final washed beads were resuspended in 20 µl of water and sent to the University of Nebraska Lincoln Proteomics Facility for analysis. 99

The 2 magnetic bead samples were resuspended in ammonium bicarbonate containing 5 mM DTT and reduced at 37°C for 1 h. Samples were then alkylated (10 mM IAM for 20 min at 22°C in the dark). The IAM was quenched with a molar equivalent of DTT.

Trypsin was added and digestion carried out overnight at 37°C. The digest was dried down and dissolved in 2.5% acetonitrile, 0.1% formic acid. 5 ul of one digest was run by nanoLC-MS/MS using a 2 h gradient on a 0.075 mm x 250 mm C18 Waters CSH column feeding into a Q-Exactive HF mass spectrometer.

All MS/MS samples were analyzed using Mascot (Matrix Science, London, UK; version 2.6.1). Mascot was set up to search the cRAP_20150130.fasta (117 entries); uniprot_S_aureus_USA300_20170823 database (2607 entries) assuming the digestion enzyme trypsin. Mascot was searched with a fragment ion mass tolerance of 0.060 Da and a parent ion tolerance of 10.0 PPM. Deamidated of asparagine and glutamine, oxidation of methionine and carbamidomethyl of cysteine were specified in Mascot as variable modifications.

Scaffold (version Scaffold_4.8.4, Proteome Software Inc., Portland, OR) was used to validate MS/MS based peptide and protein identifications. Peptide identifications were accepted if they could be established at greater than 80.0% probability by the

Peptide Prophet algorithm (129)with Scaffold delta-mass correction. Protein identifications were accepted if they could be established at greater than 99.0% probability and contained at least 2 identified peptides. Protein probabilities were assigned by the Protein Prophet algorithm(130). Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy 100 the principles of parsimony. Proteins sharing significant peptide evidence were grouped into clusters.

Reverse Transcriptase-Quantitative PCR (RT-qPCR)

RT-qPCR was performed as described previously (131). Briefly, bacterial pellets were collected 6 hours after subculture and total RNA was isolated. Complimentary DNA

(cDNA) was synthesized from 1 µg of total RNA using iScript reverse transcriptase (Bio-

Rad) according to the manufacturer’s directions. The cDNA was diluted 10 times and used in SYBR Green reactions in technical duplicates to analyze the expression of PSM, and hla. Transcription of the housekeeping gene gyrB was used as the endogenous control in each strain. Relative expression was determined by first comparing the amount of each individual gene transcript to gyrB within the same strain, followed by expression of these values in comparison to each respective gene in wild type strain TCH1516.

Ethics Statement

Whole human-blood was isolated from donors in agreement with the Ohio

University Institutional Review Board. Rabbit blood was purchased from Hemostat- laboratories. Six-week-old female BALB/c mice were purchased from Envigo and held at the Ohio University Office of Laboratory Animal Resources. Animal work was performed under approval of the Institutional Animal Care and Use Committee at Ohio

University and performed by trained lab personnel.

101

Table 3.5. Strains and plasmids used in this study. Name Characteristics Source

Strains S. aureus

RN4220 Restriction-defecient transformation recipient (106) TCH1516 Community-associated USA300 MRSA isolate (105) RKC0323 TCH1516 ΔppiB (117) RKC0085 TCH1516 ΔprsA (63) RKC0183 TCH1516 hla::Bursa, hla mutant (21) This RKC0374 TCH1516 ΔppiB pMK4_ppiB-HA work This RKC0536 TCH1516 ΔppiB pMK4 work RKC0283 TCH1516 ΔprsA pMK4_prsA-his (24) RKC0129 TCH1516 ΔprsA pMK4 (24) USA300 LAC isolate cured of plasmids LAC- JE2 (102) p01 and LAC-p03 USA300 JE2 hla::Bursa, hla NTML NE1354 (102) transposon mutant AH1263 USA300 Lac isolate cured of plasmid Lac-p03 (132) RKC0521 AH1263 hla::Bursa, hla mutant (131) This RKC0753 AH1263 ΔαPSMs work

Plasmids pMK4 Gram-positive shuttle vector (Cmr) (107) pMK4_ppiB-HA (vector overexpressing ppiB pRKC0131 (63) with an HA tag) pMK4_prsA-His (vector overexpressing prsA pRKC0126 (63) with a poly-histidine tag) This pRKC0674 pJB38 containing DNA flanking αPSM transcript with ery cassette work

102

Table 3.6. Oligonucleotides used in this study. Name Sequence #0273 GGTGCTGGGCAAATACAAGT (gyrB) #0274 TCCCACACTAAATGGTGCAA (gyrB) #0263 TGCAAATGTTTCGATTGGTC (hla) #0264 CCCCAATTTTGATTCACCATA (hla) #0271 ACAGGAGGACAAAACGATGG (psmα) #0272 CCCTATTGGTATAGTGGCCTGA (psmα) #0490 CAAGACGTCCGTCGGTCTACCTTTCCATGC #0493 GGGGTACCACGTGGCACTTTCCAAAAAC #0638 CCGGAATTCGCTCCTTGGAAGCTGTCAGT #0639 AAAACTGCAGGAAGCAAACTTAAGAGTGTGTTGA #0644 CCGGAATTCGATGTGAGGTGAGTCTTGTTAGTTTG #0645 AAAACTGCAGAGATTACCTCCTTTGCTTATGAGT

103

CHAPTER 4: TRIGGER FACTOR CONTRIBUTES TO BIOFILM FORMATION

AND VIRULENCE IN STAPHYLOCOCCUS AUREUS AND COOPERATES WITH

THE CYTOPLASMIC CHAPERONE PPIB.

Rebecca A. Keogh, Rachel L. Zapf, Emily C. Marino, Gillian G. Null, Donald L.

Holzschu, Ronan K. Carroll

104

Abstract

Peptidyl-prolyl cis/trans isomerases (PPIases) are enzymes that catalyze the cis- to-trans isomerization around proline-peptide bonds. Numerous bacterial PPIases contribute to virulence via their PPIase activity, or by PPIase independent activity in the cell. Staphylococcus aureus encodes three PPIase proteins; PrsA, PpiB and Trigger factor

(TF). Previous work by our group and others has demonstrated a role for both PrsA and

PpiB in the regulation of virulence factors and antibiotic resistance, however, S. aureus

TF remains largely unstudied. Here, for the first time, we identify a role for TF in the virulence of S. aureus and show cooperation between TF and the cytoplasmic PPIase

PpiB. Mutation of the tig gene (encoding TF) leads to a reduction in biofilm development in vitro as well as reduced bacterial recovery from the heart in vivo after systemic infection. Further analysis of a tig mutant demonstrates that while TF is dispensable under acid stress, a ppiB/tig double mutant has a significant decrease in cell viability under acid stress conditions. We go on to demonstrate that a ppiB/tig double mutant has a reduction in hemolytic activity against human erythrocytes, and that this reduction is more severe than in a ppiB mutant alone. Finally, using western immunoblot analysis we show increased levels of PpiB in a tig mutant compared to wild-type S. aureus, suggesting that there is compensation between these proteins in the cell. Together these results identify a role for TF in S. aureus and suggest functional interplay or cooperation between TF and PpiB. 105

Introduction

PPIase proteins are ubiquitous in nature and contribute to virulence in numerous bacteria (68, 92, 94, 117). There are three functional classes of PPIase proteins; cyclophilins, parvulins, and FK506 binding proteins, which share no sequence or structural similarity but all have the ability to accelerate isomerization around proline- peptide bonds (34). Despite this highly conserved enzymatic activity, PPIases often have additional functions in the cell including chaperone activity, regulation of gene expression, and roles in protein secretion (35). Interestingly, the majority of reports demonstrating a role for PPIase proteins in the cell find that elimination or inactivation of

PPIase activity has little to no effect on observable phenotypes (41, 43, 97, 117, 133,

134). PPIase proteins often function as molecular chaperones, helping deliver nascent polypeptides to secretion machinery, preventing protein aggregation in the cell, or protecting proteins under unfavorable conditions (31, 48, 135). While the chaperone activity of PPIase proteins has been studied extensively in other bacteria, there is little known about how PPIases function as chaperones in S. aureus.

S. aureus is a dangerous human pathogen that can cause life-threatening infections in virtually any niche of the human body. The diversity of S. aureus infections can be largely attributed to the vast number of virulence factors produced by the bacterium, which all have to be secreted and properly folded outside of the cell in order to function (7, 8, 108). Previous work by our group sought to investigate the role of three

PPIase proteins in S. aureus; PrsA, PpiB and TF and determine if they contributed to secreted virulence factor activity (63). Data from this work suggested that while all three 106

PPIases were dispensable for growth, only a prsA and a ppiB mutant exhibited phenotypic defects in virulence factor activity assays. A prsA mutant was found to have reduced protease activity and phospholipase activity while a ppiB mutant displayed decreased hemolysis and nuclease activity. While the deletion of either prsA or ppiB had numerous effects, each PPIase appeared to regulate its own subset of virulence factors with no overlap. These data suggested that PrsA and PpiB have distinct and unrelated functions in the cell (63). Further analysis of these proteins by our group found that a ppiB mutant is attenuated in a murine abscess and systemic model of infection and prsA is dispensable (136). We also demonstrated that PpiB contributes to virulence independently of its PPIase activity, most likely through chaperone activity (117). Here, we identify the first documented role for the third PPIase protein, TF, in S. aureus, and show that TF and PpiB cooperate as chaperones to assist S. aureus under stress.

TF is a modular chaperone protein that is found in all bacteria. The protein consists of three domains: (i) an N-terminal ribosome binding domain, (ii) a central peptidyl-prolyl cis/trans isomerase (PPIase) domain, and (iii) a C-terminal chaperone domain (64). The N and C-terminal domains work together to bind nascent polypeptides as they exit the ribosome, and to assist in the chaperoning of proteins in the cytoplasm

(46). The central PPIase domain is a FK506 binding factor domain that catalyzes the cis- to-trans isomerization around proline-peptide bonds, which is often the rate limiting step in protein folding (137). In E. coli, deleting tig had no apparent role in the cell, however, the deletion of both tig and the cytoplasmic chaperone dnaK led to protein aggregation, and had detrimental effects on cell survival at different temperatures (138, 139). 107

Interestingly, while TF has PPIase activity in E. coli, that activity is not necessary for the folding of cytoplasmic proteins (43).

While TF has been characterized extensively in E. coli for its structure and folding kinetics, all previous studies that have demonstrated a role for TF in virulence were performed in Gram-positive bacteria (79). In the oral pathogen Streptococcus mutans, the TF homolog ropA is involved in the bacterium’s ability to tolerate acid and oxidative stress as well as in biofilm formation (67). Deletion of TF homologs in both

Streptococcus suis as well as Listeria monocytogenes resulted in a decrease in stress tolerance as well as attenuation in vivo (69, 70). Additionally, recent work by Cohen et al. demonstrates that Streptococcus pneumoniae TF promotes host-cell adherence, contributes to virulence, and elicits a protective immune response in the host making it a potential candidate for vaccine development (71).

Based on the role of TF in the virulence of multiple Gram-positive pathogens as well as the contribution of two other PPIases to virulence factor activity in S. aureus, we sought to reexamine TF and address whether it contributes to virulence in S. aureus. In this study we demonstrate that S. aureus TF contributes to biofilm formation and is necessary in a murine systemic model of infection in an organ-specific manner. We then show that a double mutant in tig and the cytoplasmic PPIase ppiB has reduced survival under stress conditions, demonstrating possible functional compensation between these two cytoplasmic chaperones. Further analysis reveals that the ppiB/tig double mutant has a greater defect in hemolytic activity than a ppiB single mutant alone. These data suggest that TF functions to compensate in toxin regulation in the absence of ppiB. Finally, 108 western blot analysis shows that there is increased PpiB in a tig mutant in comparison to wild-type. These data are consistent with work by Deuerling et al. as well as Teter et al. which showed cooperation between TF and the chaperone DnaK in E. coli (138, 139).

Collectively, these results demonstrate the first known role for TF in S. aureus virulence and suggest functional overlap for these PPIase proteins via chaperone activity in this bacterium.

109

Results

A tig Mutant Has Reduced Biofilm Formation in S. aureus

Previous work by our group demonstrated that tig was dispensable for growth in rich media (63). This is consistent with data from numerous groups showing that deleting tig has no apparent effect on bacterial cell growth under nutrient-rich conditions (70, 71,

140). However, work by Wen et al. demonstrated that the TF homolog RopA was necessary for biofilm formation in S. mutans (67). To determine whether S. aureus TF was necessary for growth in biofilm, we performed a static biofilm formation assay using crystal violet as previously described (141). Briefly, a 24-well plate was treated with 10% human plasma overnight. The next day, a 1:100 dilution of cells grown in biofilm media were inoculated into each well and grown statically at 37 °C. After 8 hours, wells were washed to remove all non-adherent bacteria and the remaining biomass was stained with

0.05% crystal violet. A significant reduction in biomass was observed in a tig mutant when compared to wild-type after 8 hours of biofilm development (Figure 4.1A).

In order to determine if the reduction in biomass was due to biofilm formation or a growth defect in biofilm media, we performed a growth analysis of a tig mutant over time in both biofilm and planktonic cultures. Interestingly, a reduction in biofilm in a tig mutant was observed as early as 3 hours on a plate (Figure 4.1B) where growth in planktonic cultures of biofilm media showed no significant difference at any time point

(Figure 4.1C). Collectively, these results suggest that TF is dispensable for planktonic growth in biofilm media but is necessary in either adherence or growth on a surface. 110

Figure 4.1. A tig mutant has reduced biofilm formation in S. aureus. (A) Biofilm formation after 8 hours of growth and (B) Static biofilm formation over time. Bacterial strains were grown in TSB supplemented with 1% NaCl before being inoculated into 24-well plates containing 10% human plasma. Biofilm formation was quantified with 0.05% crystal violet staining, followed by extraction with 30% acetic acid and OD543 measurement. Assays were performed using four biological replicates. Student’s t-test was used to determine significance, *= p<0.05. Error bars shown with SEM. (C) Growth of wild-type S. aureus strain USA300 and tig mutant in biofilm media with shaking. No difference in growth rate was observed between the wild-type strain and the tig mutant strain. Data shown are the averages from three independent biological replicate cultures. Error bars represent the standard deviation.

TF is Required for Virulence in a Murine Systemic Model of Infection

Previous work has demonstrated a role for TF in virulence in the Gram-positive pathogens S. suis, L. monocytogenes, and S. pneumoniae (69–71). We hypothesized that the necessity for TF in other pathogens as well as in S. aureus biofilm formation would 111 result in attenuation of a tig mutant in vivo. To test this, we utilized a murine systemic model of infection as previously described (136). Briefly, six-week old female BALB/c mice were injected retro-orbitally via the venous plexus with 107 CFU and monitored over seven days of infection. 50% of mice infected with wild-type S. aureus died during infection, in comparison to only 25% of the tig mutant cohort. Although this difference was not statistically significant due to the relatively small size of the cohorts of mice infected (n=12), it nonetheless suggests that TF may be required for full virulence (Figure

4.2A). After seven days of infection, surviving mice were euthanized, and the kidneys and hearts were harvested to determine the bacterial burden. A significant reduction

(p=0.0295) in recovered bacteria was observed in the heart, and a modest non-statistically significant (p=0.1238) reduction in kidney burden was also observed (Figure 4.2B&C).

Collectively, these data suggest that TF has a role in S. aureus systemic infection, that may be organ specific. 112

Figure 4.2. TF is required for virulence in a murine systemic model of infection. Female 6-week-old BALB/c mice were injected retro-orbitally with WT S. aureus or a tig mutant strain. The infection was allowed to proceed for 7 days. Mice were then sacrificed before organs were excised, weighed, diluted and plated to enumerate bacteria present in the organs. (A) Kaplan–Meyer curve for mouse survival during 7-day systemic infection. Six mice infected with WT S. aureus and three mice infected with the tig mutant died over the course of infection. (B and C) CFU/g recovered from heart and kidneys of infected mice. In mice infected with the tig mutant strain, a significant reduction (p=0.0295) in bacterial numbers was detected in the heart of infected mice. Experiments were performed with an n = 12 for each strain. Error bars represent standard deviation. Significance was determined by Student’s t test ,* p < 0.05. 113

TF Does Not Contribute to Protection From Acid Challenge in S. aureus

We next sought to determine whether TF was contributing to virulence as a chaperone protein. TF is both annotated as a chaperone protein and has been demonstrated to have chaperone activity in E. coli as well as in multiple Gram-positive bacteria (64, 69–71). Chaperone proteins commonly function to help proteins traffic through the cytoplasm or to prevent the aggregation of proteins in unfavorable conditions such as stress or shock (135, 142). In order to determine whether TF possesses canonical chaperone activity in S. aureus, we assessed the ability of a tig mutant to survive acid challenge as described by Belli and Marquis (143). Briefly, WT and tig mutant cultures were grown to mid-exponential phase in tryptic soy-broth (TSB) adjusted to a pH of 7.0 with HCl. Cells were then diluted and plated to calculate initial CFU/ml. Following this, the bacteria were harvested by centrifugation and washed in 0.1 M glycine before being challenged for 1 hour by incubation in 0.1 M glycine at a pH of 2.0. Surviving bacteria were enumerated by diluting and plating and the percent survival was calculated.

Surprisingly, there was no difference in survival between the tig mutant and wild-type S. aureus (Figure 4.3). This result was unexpected, as we hypothesized TF would act as a chaperone to protect S. aureus from acid challenge. However, it is possible that S. aureus

TF protects the cell under different stress conditions such as heat shock or oxidative stress, or that other S. aureus chaperones are compensating for the loss of tig in the cell under acid challenge. 114

Figure 4.3. TF does not contribute to protection from acid challenge in S. aureus. S. aureus strains were grown in TSB until mid- exponential phase (OD600 ~ 0.6), cells were harvested by centrifugation, washed once with 0.1 M glycine buffer, pH 7.0, and then subjected to acid killing at pH 2.0 for 1 h. A tig mutant has no significant difference in survival in comparison to WT. Cells were diluted and plated onto TSB for bacterial enumeration before and after acid killing and results are expressed as perfect survival. The data presented are an average of four biological replicates and significance was determined using Student’s t-test, ns= not significant. Error bars represent standard deviation.

A ppiB/tig Double Mutant Exhibits a Defect in Cell Viability After Acid Challenge

We have previously speculated that the cytoplasmic PPIase PpiB functions as a chaperone protein in S. aureus, due to its role in protein secretion and association with

SecA, the major ATPase of the sec secretion system (136). This speculation coupled with numerous studies showing that TF and the chaperone protein DnaK have overlapping functions in E. coli led us to hypothesize that a double mutant of both tig and ppiB would have detrimental effects on S. aureus cell survival (144). To test this hypothesis, we 115 constructed a ppiB/tig double mutant and again tested the bacteria under acid stress. As previously observed (Figure 4.3), no difference in survival was observed between the WT and tig mutant (Figure 4.4). Similarly, no defect in survival was detected between the WT and ppiB mutant. However, the ppiB/tig double mutant exhibits a pronounced (8-fold) and statistically significant reduction in survival after acid challenge, demonstrating a requirement for at least one of these cytoplasmic PPIases in the cell during unfavorable conditions (Figure 4.4). These data also demonstrate, for the first time, a potential overlap in functions between the two intracellular PPIase proteins in S. aureus.

116

Figure 4.4. A ppiB/tig double mutant has a defect in cell viability after acid challenge. S. aureus strains were grown in TSB until mid-exponential phase (OD600 ~ 0.6), cells were harvested by centrifugation, washed once with 0.1 M glycine buffer, pH 7.0, and then subjected to acid killing at pH 2.0 for 1 h. A ppiB/tig mutant has an 8-fold reduction in survival after acid killing. Cells were diluted and plated onto TSB for bacterial enumeration before and after acid killing and results are expressed as perfect survival. The data presented are an average of four biological replicates and significance was determined using Student’s t-test *, P < 0.05; ns= not significant. Error bars represent standard deviation.

The Mutation of tig in a ppiB Mutant Leads to a Further Reduction in Hemolytic Activity

Previous work by our group demonstrated that a ppiB mutant has decreased hemolytic activity towards human erythrocytes where a tig mutant does not (63, 136). To determine if TF and PpiB have functional overlap/compensation in S. aureus, we compared the hemolytic activity of the ppiB/tig double mutant to the wild type and ppiB and tig single mutants. Whole-human blood hemolysis assays were performed as 117 previously described (131). Briefly, overnight cultures of bacteria were normalized to the same OD600 before being grown for 15 hours. Sterilized culture supernatants were incubated with whole-human blood and lysis of red blood cells was measured by absorbance at 543nm. Consistent with our previously published data the ppiB mutant demonstrated a significant reduction in hemolytic activity, while no reduction was observed in the tig mutant (Figure 4.5A) (63, 136). Interestingly, the ppiB/tig double mutant demonstrated a reduction in hemolysis which was significantly greater than that observed in the ppiB single mutant (Figure 4.5A). These data, showing a further reduction in hemolytic activity when tig is deleted in the ppiB mutant, suggest that TF may be compensating for the loss of ppiB.

118

Figure 4.5. The mutation of tig in a ppiB mutant leads to a further reduction in hemolytic activity. Erythrocyte lysis assays were performed using S. aureus culture supernatants and whole human blood. A significant decrease in hemolytic activity against human erythrocytes was observed using culture supernatants from both a ppiB mutant and a ppiB/tig double mutant strain. Culture supernatants from the ppiB/tig double mutant were significantly reduced in comparison to the ppiB single mutant. Hemolysis assays were performed a minimum of 3 times. The data presented are the averages of 4 replicates. Significance was determined by Student’s t test. **, P < 0.01; *, P < 0.05; ns= not significant.

There is More PpiB Produced in a tig Mutant

To investigate potential functional compensation between PpiB and TF we examined PpiB protein levels in a tig mutant and compared them to WT S. aureus. We performed a western blot using a polyclonal anti-PpiB antibody to determine the amount of PpiB being produced in a tig mutant. Overnight bacterial cultures were normalized to 119 the same OD600 before being grown for 15 hours. Bacterial cells were collected by centrifugation, lysed, and the cytoplasmic fraction was harvested (consistent with our previous findings that PpiB is an intracellular protein (63)). Western blot analysis reveals that there is a greater abundance of PpiB in a tig mutant, compared to WT S. aureus

(Figure 4.6). As a negative control, no PpiB was observed in a ppiB single mutant. These data show that there is increased production of PpiB when TF is absent. We hypothesize that this functional compensation happens both ways, and that further analysis of the ppiB/tig double mutant will provide a more comprehensive understanding of the complex chaperone network in S. aureus.

Ladder WT ΔprsA tig ΔppiB

Figure 4.6. There is more PpiB produced in a tig mutant. A western immunoblot was performed using polyclonal anti-PpiB antisera and whole-cell lysates from wild-type S. aureus, ppiB mutant, prsA mutant and the tig mutant strain. No PpiB was detected in the ppiB mutant strain, while comparable levels of PpiB protein were detected in the wild-type and the prsA mutant strain. An increase in PpiB levels can be observed in the tig mutant.

120

Discussion

TF is a well-characterized PPIase that contributes to virulence in numerous Gram- positive bacteria (69–71). The deletion of a TF in S. mutans had no effect on planktonic growth in optimal conditions but led to significant alterations in biofilm formation (67).

Authors speculate that the differences in biofilm formation may be due to an alteration in surface proteins in a RopA-deficient strain. Our data supports that S. aureus TF may function in a similar manner. A decrease in biofilm formation in a tig mutant could be seen as early as three hours into biofilm development and continued over time, suggesting a potential effect on growth rate (Figure 4.2B) however, when we tested the same strain for planktonic growth in biofilm media, we saw no significant growth defect

(Figure 4.2C). These data are suggestive of a potential role for TF in biofilm adherence, or in binding to human plasma which was utilized to coat wells before biofilm growth.

We next established that a TF mutant contributes to virulence in a systemic model of infection, and that there is a significant defect in bacteria recovered from the heart

(Figure 4.2A 4.and 2B). Infections in the heart such as endocarditis require numerous cell-wall associated adhesion proteins including ClfA, FnBPA, ClfB and SraP for adherence to thrombus (145). Interestingly, these same 4 proteins contribute to biofilm formation (146). ClfA and ClfB are known to assist in biofilm formation due to their ability to bind the human plasma protein fibrinogen (147) where FnBPA contributes to biofilm formation through binding fibronectin (148). SraP has not been well characterized in S. aureus but work by Sanchez et al. demonstrated that an S. aureus sraP mutant forms significantly reduced biofilm in comparison to wild-type (149). We 121 hypothesize that S. aureus TF may be critical for the activity of these surface proteins and therefore critical to biofilm formation and colonization of the heart. Although the mechanism behind this is unclear, it would not be the first case of TF regulating the activity of a cell-wall anchored protein, as in E. coli TF is necessary for maturation of the outer membrane protein OmpA (150).

In addition to establishing a role for TF in biofilm formation and virulence, we go on to demonstrate functional overlap/compensation between TF and the cytoplasmic

PPIase PpiB. Our hypothesis was that TF and PpiB possess chaperone activity which is required to protect the cell under unfavorable conditions, such as acid stress.

Interestingly, while neither single mutant exhibited a significant difference in bacterial survival, a double mutant was more susceptible to killing by HCl (Figure 4.4). These data are suggestive that each of these proteins can help assist S. aureus under stress conditions, and that they may be compensating for one another in their absence. This would not be the first instance of bacterial chaperone proteins exhibiting functional compensation, as the overlapping functions of TF and the chaperone DnaK are well- demonstrated in E. coli (43, 64, 138, 151, 152). Furthermore, this redundancy of function hypothesis is supported by our western blot data showing increased production of PpiB in a tig mutant. We hypothesize that a reciprocal increase in TF production occurs in a ppiB mutant.

To further investigate the interplay between TF and PpiB, we performed a hemolytic activity assay in human blood. Previous work by our group has demonstrated a role for PpiB in the lysis of human red blood cells via regulation of the phenol-soluble 122 modulins (PSMs) (136). We repeated this assay with the ppiB/tig double mutant to determine if there was a greater reduction in hemolysis when tig was absent from the cell.

Interestingly, our results demonstrate that the double mutant is significantly reduced in hemolysis in comparison to the ppiB single mutant. These data suggest that TF can compensate for the loss of ppiB. We do not know the exact mechanism of this compensation or at what level of regulation it may be occurring. To determine this, we performed a western blot and identified greater levels of PpiB being produced in a tig mutant. Whether or not this increased production is happening at the transcriptional or translational level is still being determined.

Taken together, our data provides the first insight into how TF functions in S. aureus biofilm formation and virulence. Our group is currently investigating the mechanism by which TF contributes to virulence through the potential regulation of cell- wall associated proteins. In addition to the first characterization of TF in S. aureus, this is the first report of functional overlap between PPIase proteins in S. aureus. These data open up numerous studies into characterizing this overlap and highlight the importance of functional compensation between proteins in the cell.

123

Materials and Methods

Strains and Strain Construction

All bacterial strains and plasmids used are listed in Table 4.1. The ΔppiB mutant was constructed by allelic exchange as previously described (102, 117). S. aureus containing a transposon insertion in the tig gene was obtained from the Network on

Antimicrobial Resistance in Staphylococcus aureus (NARSA) transposon mutant library

(102) and was transduced into the USA300 wild-type background TCH1516, as well as the ΔppiB mutant using bacteriophage φ11 to generate the tig single mutant and ppiB/tig double mutant respectively.

Table 4.1. Strains and plasmids used in this study. Strain or plasmid Characteristicsα Source S. aureus TCH1516 Community-associated USA300 MRSA isolate (105) RKC0323 TCH1516 ΔppiB (117) RKC0085 TCH1516 ΔprsA (63) RKC0114 TCH1516 tig::Bursa, tig mutant (63) RKC0414 RKC0323 containing tig::Bursa, tig mutant This work

Plasmids pMK4 Gram-positive shuttle vector (Cmr) (86)

αCmr, chloramphenicol resistance

Bacterial Growth Conditions

S. aureus cultures were grown in TSB shaking at 37 °C. When appropriate, antibiotics were added at the following concentrations: chloramphenicol (5 μg mL-1), erythromycin (5 μg mL-1), and lincomycin (25 μg mL-1). E. coli cultures were grown in 124 lysogeny broth (LB) shaking at 37 °C and ampicillin (100 μg mL-1) was added when appropriate. For comparative analysis of S. aureus culture supernatants, 5 mL overnight starter cultures were diluted 1:100 in 10 mL of fresh TSB and grown shaking at 37 °C for

3 h to mid-exponential phase. The resulting 3 h cultures were then diluted into 25 mL of fresh TSB at a starting optical density OD600 of 0.05. Cultures were then grown while shaking at 37 °C for 15 hours.

Static Biofilm Formation Assay

The ability to form biofilms was tested using methods adapted from Marroquin et al (141). Overnight bacterial cultures were grown in TSB supplemented with 0.2 M NaCl.

Cultures were then diluted 1:50 into a 24-well plate containing 10% human plasma and grown for 8 h at 37 °C. After incubation, biofilms were washed twice with sterile PBS before being stained with 0.05% crystal violet for 5 minutes. Biofilms were again washed twice with sterile PBS to remove all non-adherent crystal violet and the resulting biofilms were treated with 30% acetic acid before measuring Abs 595.

Murine Systemic Model of Infection

A systemic model of dissemination was utilized to mimic septic infection as previously described (136). Briefly, cultures were grown for 2.5 h in TSB to an OD600 of

0.75. Resulting bacterial cells were pelleted via centrifugation and resuspended in sterile phosphate-buffered saline (PBS) to prepare an inocula of 107 CFU/100 μL. Six-week-old

BALB/c mice were injected retro-orbitally with 100 μL of bacterial cultures and the infection was allowed to proceed for seven days. Mice were monitored over the course of infection for significant weight loss or morbidity. Following seven days of infection, 125 mice were euthanized with CO2 and the heart and kidneys and spleen were harvested.

Organs were weighed and homogenized before serial dilution and plating was conducted to quantify recovered bacteria/ organ.

Acid Challenge Assay

Acid challenge was performed based on methods adapted from Belli and Marquis

(143). S. aureus strains were grown for 3 h to mid-exponential phase in TSB adjusted to pH 7.0 with HCl. Subsequent cultures were diluted and plated to calculate the CFU/mL before acid challenge. The remainder of cultures were centrifuged at 4,000 g for 10 minutes and washed with 0.1 M glycine buffer at pH 7.0. Cells were then resuspended in

TSB containing 0.1M glycine at a pH of 2.0 adjusted with HCl. Resulting cultures were incubated statically at 37 °C for 1 hour, after which the viable bacteria were again diluted and plated. % survival was calculated using the following equation: ((initial bacteria-final bacteria)/initial bacteria) x 100.

Cell-Free Hemolysis Assay

Hemolysis assays were performed as described by Zapf et al. (131).

Synchronized, cell-free, S. aureus supernatants were diluted 1:2 in reaction buffer (40 mM CaCl2, 1.7% NaCl) and incubated at 37 °C in a tube revolver with 50 μL of whole human blood. Following a 10-min incubation, the samples were centrifuged at 5,500 X g, and 100 μL of the supernatant was transferred to a 96-well plate. The degree of erythrocyte lysis was determined by reading the absorbance of the samples at OD543. 126

Western Blotting

Intracellular protein samples for Western blots were prepared from stationary- phase cultures of the wild-type, ΔppiB mutant, ΔprsA mutant and tig mutant strain as previously described (104). Samples were separated by SDS-PAGE, transferred to a polyvinyldine difluoride (PVDF) membrane, and probed using rabbit polyclonal anti-

PpiB antibody that was raised using a MBP-tagged PpiB protein.

Ethics Statement

Human blood samples were obtained from donors at Ohio University. All collections, handling, and usage of blood was approved by the Ohio University

Institutional Review Board. Whole rabbit blood was purchased from Hemostat

Laboratories. Six-week-old BALB/c mice were ordered from Envigo and held at the Ohio

University Office of Laboratory Animal Resources. All animal work was done by trained lab personnel and approved by the Institutional Animal Care and Use Committee.

127

CHAPTER 5: CONCLUSIONS

The overarching goal of this dissertation was to characterize the three PPIase proteins in S. aureus, and to determine if they contributed to virulence. Preliminary work by our lab identified that S. aureus encodes three PPIase proteins (PrsA, PpiB and TF) but their potential role in virulence was largely unknown. The data generated in this dissertation identify a role for both PpiB and TF in S. aureus virulence and suggest that all three of these proteins act through distinct mechanisms to regulate virulence factor production and activity. In addition, the data presented herein reveal the first known compensation between PPIase proteins in S. aureus.

PrsA

In chapter 3 we examined the contribution of PrsA to virulence in both a skin and soft tissue as well as a systemic model of infection. We hypothesized that PrsA would be necessary for virulence, as previous work by our group found that it contributes to the activity of proteases and phospholipase C in S. aureus (63). Interestingly, our results found that a prsA mutant was not attenuated in either model of infection. We next found that a prsA mutant had decreased lytic activity against rabbit erythrocytes, a process known to be mediated by S. aureus alpha toxin (110). It is surprising that PrsA regulates both proteases and alpha toxin, which are known to contribute to virulence in vivo, but

PrsA does not contribute to virulence in any of the models we tested. However, there are numerous studies that find extracellular proteases are involved in maintaining virulence stability, and that the deletion of certain proteases results in a hypervirulent phenotype

(153). It is therefore possible that the decreased alpha toxin activity, which would 128 normally result in attenuation, is negated by the hypervirulent phenotype found in protease mutants.

Another finding from chapter 3 was that PrsA associates with two penicillin- binding proteins: Pbp3 and PBP2A. These data support work by other groups demonstrating a role for PrsA in antibiotic resistance (57, 127). Jousselin et al. found that

PrsA was able to alter susceptibility of S. aureus to β-lactam antibiotics via regulation of

PBP2A. A prsA mutant has less PBP2A in the membrane but the mRNA levels of pbp2A were unchanged. Collectively, these data support that PrsA is a membrane-anchored

PPIase involved in post-transcriptional regulation in S. aureus.

Recent work by Lin et al. found that PrsA was necessary for adhesion to lung epithelial cells and virulence in the methicillin-sensitive S. aureus strain HG001 (58).

While these results conflict with our findings, we believe this is due to stain specific differences in S. aureus, which displays up to 20% variability in genomic sequence between isolates (154). A known difference between HG001 and the USA300 lineage of

MRSA is that USA300 strains produce more toxins and have greater hemolytic activity

(128). It is possible that the absence of PrsA in our USA300 strain has less impact on virulence because this strain produces such a high level of toxins, some of which are likely to fold correctly even in the absence of PrsA. Additionally, biofilm formation, autolysis, and surface protein expression are known to be highly variable between MSSA and MRSA strains (155–157). These differences highlight the incredible diversity in S. aureus strains and the challenge of combatting infection. Figure 5.1 summarizes some of our findings for PrsA in the USA300 lineage. 129

Figure 5.1. Schematic for PrsA. PrsA is a membrane-bound lipoprotein that assists in extracellular protein folding. PrsA is necessary for the activity of numerous secreted virulence factors such as alpha-toxin, proteases and phospholipase C which utilize the general sec secretion system. Additionally, a prsA mutant has decreased levels of penicillin-binding protein 2a in the cell-wall resulting in altered antibiotic susceptibility.

PpiB

Chapter 2 investigated the cytoplasmic cyclophilin PpiB, and its contribution to disease. Initial work by our lab demonstrated that PpiB is an active PPIase that contributes to folding of the secreted virulence factor nuclease (63). While these data supported a role for PpiB in the regulation of a virulence factor, they did not demonstrate a direct role for PpiB in virulence. This was addressed in chapter 2 with a murine abscess 130 model of infection, which closely mimics skin and soft tissue infections in humans. A ppiB mutant was attenuated after seven days of infection, demonstrating the necessity of

PpiB in virulence. Our initial hypothesis was that the PPIase activity of PpiB was responsible for its role in virulence. Interestingly, the data from this chapter find that the

PPIase activity of PpiB is dispensable for virulence, suggesting for the first time that

PpiB has an additional function in the cell.

One hypothesis is that PpiB possesses chaperone activity like its Gram-negative homolog in E. coli (41). Chaperone proteins perform diverse roles in the cell including assisting in protein folding and preventing the aggregation of misfolded proteins, trafficking proteins through the cytoplasm to secretion machinery, and assisting cytoplasmic proteins in the cell under unfavorable conditions (47). We demonstrate that

PpiB can perform each of these canonical chaperone functions in this dissertation. In chapter 2, the PPIase activity of PpiB was abrogated when certain active site residues were substituted for the amino acid alanine. Analysis of the PPIase inactive protein F64A in a nuclease activity assay showed that this protein was able to assist in the folding of nuclease, even though it lacked PPIase activity. This residual activity demonstrates that

PpiB possesses PPIase independent chaperone activity and can accelerate nuclease folding even without PPIase activity. We go on to demonstrate a role for PpiB in protein secretion in chapter 3, where in vivo pull-down analysis revealed that PpiB associates with SecA, the ATPase that powers the sec secretion system. PpiB is an intracellular protein and does not interact with SecA for its own secretion. We therefore speculate that

PpiB is involved in the delivering of nascent proteins through the cytoplasm to the sec 131 secretion machinery. In E. coli, the cytoplasmic chaperone SecB is known to function to deliver nascent proteins to the sec translocon (30). Notably, secB is absent in S. aureus and there is no known homolog that has a similar function. To further investigate a potential role for PpiB in protein secretion, we perform secreted proteome analysis of a ppiB mutant in comparison to WT in chapter 3. Results demonstrate that the deletion of ppiB results in the altered abundance of 86 proteins in the extracellular environment. This is in contrast to a prsA mutant, which only had 16 proteins with altered abundance.

Collectively, the results presented in chapters 2 and 3 support a PPIase independent role of PpiB as a chaperone protein, that both assists in the protection of proteins such as nuclease and contributes to protein secretion.

In chapter 4, we further investigate the chaperone activity of PpiB by demonstrating that PpiB is necessary for S. aureus survival under unfavorable conditions.

In an acid-shock assay, we show that a ppiB mutant has a modest, although non- significant reduction in survival in comparison to WT. Interestingly, a double mutant of ppiB and the chaperone tig resulted in a significant reduction in S. aureus survival after acid shock. These results demonstrate that both PpiB and TF possess chaperone activity and assist S. aureus in unfavorable conditions. Furthermore, these data suggest that chaperone proteins can compensate for the loss of a single chaperone in S. aureus.

An alternative/additional role for PpiB in S. aureus secretion was speculated in chapter 3, with the finding that many of the proteins found in altered abundance in a ppiB mutant are cytoplasmic in nature. Notably, many of these cytoplasmic proteins were found in reduced abundance outside of the cell when ppiB was deleted. One possible 132 explanation for the reduction in cytoplasmic proteins being found outside of the cell is that a ppiB mutant secretes fewer extracellular vesicles. Extracellular vesicles have recently been shown to carry numerous S. aureus proteins including both cytoplasmic proteins and virulence factors out of the cell and into the extracellular environment (119,

158). While little is currently known about the function of these vesicles in S. aureus, their release from the cell was found to be mediated by the αPSMs (119). Results from chapter 3 demonstrate that a ppiB mutant exhibits a reduction in PSM activity and abundance in comparison to WT. We therefore speculate that the reduction in PSMs results in fewer extracellular vesicles being released in a ppiB mutant, ultimately resulting in fewer cytoplasmic proteins found outside of the cell. Our lab is currently investigating the production of extracellular vesicles in a ppiB mutant.

Collectively, we identify PpiB as a PPIase that contributes to virulence independently of its PPIase activity. We identify that PpiB can act as a chaperone and that PpiB plays a role in the sec secretion pathway. Finally, we speculate that PpiB may be involved in alternative protein secretion via the release of extracellular vesicles. Figure

5.2 summarizes our findings for PpiB. 133

Figure 5.2. Schematic for PpiB. PpiB is an intracellular chaperone that associates with SecA, the major ATPase that powers the sec secretion system. PpiB is required for the activity of numerous virulence factors including nuclease and the αPSMs. In addition, the deletion of ppiB leads to numerous changes in the exoproteome of S. aureus which can be due to alterations in sec secretion or a reduction in αPSM-mediated vesicle release from the cell.

TF and chaperone compensation

Finally, chapter 4 identified the first known role for TF in S. aureus. Previous work by our group found that tig was dispensable for planktonic growth in optimal conditions (63). While these data are important, it is now known that a more common alternative lifestyle for bacteria is in biofilms (147). Biofilms are multicellular communities that can be made up of one or multiple bacterial species that can promote survival in diverse niches. In humans, S. aureus biofilm development on cell tissues or indwelling medical devices can make clearance more difficult or lead to chronic disease 134

(147). In chapter 4, we find that a tig mutant has a defect in biofilm development over time. We then determine that TF contributes to virulence in a murine systemic model of infection. These data provide the first insight into S. aureus TF and its role in infection.

Based on homology to other Gram-positive TF proteins, we next hypothesized that TF would have chaperone activity and be able to assist S. aureus under stress conditions. As stated above, we determined that TF is necessary for survival after acid stress in the absence of ppiB. In addition, we found that mutating tig in a ppiB mutant background led to further reductions in hemolysis than in a ppiB mutant alone. Both results suggest that PpiB and TF have chaperone activity, and that TF may compensate for the loss of ppiB in the cell. Finally, we measured the abundance of PpiB produced in the cell in the absence of tig. Results demonstrate that more PpiB is produced when tig is absent, further confirming there is compensation between these chaperones in S. aureus.

Findings for TF and PpiB are highlighted in figure 5.3. 135

Figure 5.3. Schematic for TF and compensation. TF associates with the ribosome to assist in protein trafficking post-translation. Active TF is necessary for biofilm formation, and TF has functional overlap with the cytoplasmic chaperone PpiB.

Overall, the results presented in this dissertation strengthen our understanding of all three PPIase proteins encoded by S. aureus. We find that each PPIase regulates a number of virulence factors or processes, and additionally, we identify that PPIase proteins can compensate for the loss of each other highlighting the first compensation between these proteins in S. aureus. These results highlight the complex network of virulence factor regulation in S. aureus and provide further insight into post-translational regulation through PPIase and chaperone activity. 136

REFERENCES

1. van Belkum A, Verkaik NJ, de Vogel CP, Boelens HA, Verveer J, Nouwen JL, Verbrugh HA, Wertheim HFL. 2009. Reclassification of Staphylococcus aureus Nasal Carriage Types . J Infect Dis 199:1820–1826. 2. Mehraj J, Witte W, Akmatov MK, Layer F, Werner G, Krause G. 2016. Epidemiology of Staphylococcus aureus Nasal Carriage Patterns in the Community, p. 55–87. In Stadler, M, Dersch, P (eds.), How to Overcome the Antibiotic Crisis : Facts, Challenges, Technologies and Future Perspectives. Springer International Publishing, Cham. 3. Klein E, Smith DL, Laxminarayan R. 2007. Hospitalizations and deaths caused by methicillin-resistant Staphylococcus aureus, United States, 1999-2005. Emerg Infect Dis 13:1840–1846. 4. Tong SYC, Davis JS, Eichenberger E, Holland TL, Fowler VG. 2015. Staphylococcus aureus infections: Epidemiology, pathophysiology, clinical manifestations, and management. Clin Microbiol Rev 28:603–661. 5. Kobayashi SD, Malachowa N, Deleo FR. 2015. Pathogenesis of Staphylococcus aureus abscesses. Am J Pathol 185:1518–1527. 6. Otto M. 2010. Basis of Virulence in Community-Associated Methicillin-Resistant Staphylococcus aureus. Annu Rev Microbiol 64:143–162. 7. Thomer L, Schneewind O, Missiakas D. Pathogenesis of Staphylococcus aureus Bloodstream Infections. 8. Lacey KA, Geoghegan JA, Mcloughlin RM. The Role of Staphylococcus aureus Virulence Factors in Skin Infection and Their Potential as Vaccine Antigens. 9. Guerrero MLF, López JJG, Goyenechea A, Fraile J, De Gó rgolas M. 2009. Endocarditis caused by staphylococcus aureus a reappraisal of the epidemiologic, clinical, and pathologic manifestations with analysis of factors determining outcome. Medicine (Baltimore) 88:1–22. 10. Bartlett JG. 2004. Nosocomial bloodstream infections in US hospitals: Analysis of 24,179 cases from a prospective nationwide surveillance study. Infect Dis Clin Pract 12:376. 11. Frieden T. 2013. Antibiotic resistance threats in the United States. Centers Dis Control Prev 114. 12. Liu C, Graber CJ, Karr M, Diep BA, Basuino L, Schwartz BS, Enright MC, O’Hanlon SJ, Thomas JC, Perdreau‐Remington F, Gordon S, Gunthorpe H, Jacobs R, Jensen P, Leoung G, Rumack JS, Chambers HF. 2008. A Population‐Based Study of the Incidence and Molecular Epidemiology of Methicillin‐Resistant Staphylococcus aureus Disease in San Francisco, 2004–2005 . Clin Infect Dis 137

46:1637–1646. 13. David MZ, Daum RS. 2010. Community-associated methicillin-resistant Staphylococcus aureus: Epidemiology and clinical consequences of an emerging epidemic. Clin Microbiol Rev 23:616–687. 14. Lance R. Thurlow, Gauri S. Joshi and ARR. 2012. Virulence Strategies of the Dominant USA300 Lineage of Community Associated Methicillin Resistant Staphylococcus aureus (CA-MRSA). FEMS Immunlogy Med Microbiol. 15. Rte Becher D, Hempel K, Sievers S, Zü Hlke D, Pané-Farré J, Otto A, Fuchs S, Albrecht D, Rg Bernhardt J, Engelmann S, Vö Lker U, Maarten Van Dijl J, Hecker M. 2009. A Proteomic View of an Important Human Pathogen – Towards the Quantification of the Entire Staphylococcus aureus Proteome. 16. Green ER, Mecsas J. 2015. Bacterial Secretion Systems – An overview. Am Soc Microbiol 4:1–32. 17. Chen C, Krishnan V, Macon K, Manne K, Narayana SVL, Schneewind O. 2013. Secreted Proteases Control Autolysin-mediated Biofilm Growth of Staphylococcus aureus *. Publ JBC Pap Press. 18. Otto M. 2014. Staphylococcus aureus toxins. Curr Opin Microbiol. 19. Williams RJ, Henderson B, Nair SP. Staphylococcus aureus Fibronectin Binding Proteins A and B Possess a Second Fibronectin Binding Region That May Have Biological Relevance to Bone Tissues. 20. O’Halloran DP, Wynne K, Geoghegan JA. 2015. Protein A is released into the Staphylococcus aureus culture supernatant with an unprocessed sorting signal. Infect Immun 83:1598–609. 21. Berube BJ, Wardenburg JB. 2013. Staphylococcus aureus α-toxin: Nearly a century of intrigue. Toxins (Basel) 5:1140–1166. 22. Fowler VG, Allen KB, Moreira ED, Moustafa M, Isgro F, Boucher HW, Corey GR, Carmeli Y, Betts R, Hartzel JS, Chan ISF, McNeely TB, Kartsonis NA, Guris D, Onorato MT, Smugar SS, DiNubile MJ, Sobanjo-ter Meulen A. 2013. Effect of an investigational vaccine for preventing Staphylococcus aureus infections after cardiothoracic surgery: a randomized trial. Jama 309:1368–78. 23. Proctor RA. 2015. Recent developments for Staphylococcus aureus vaccines: Clinical and basic science challenges. Eur Cells Mater 30:315–326. 24. Kusch H, Engelmann S. 2014. Secrets of the secretome in Staphylococcus aureus. Int J Med Microbiol 304:133–141. 25. Schallenberger MA, Niessen S, Shao C, Fowler BJ, Romesberg FE. Type I Signal Peptidase and Protein Secretion in Staphylococcus aureus. 138

26. Goosens VJ, Monteferrante CG, Maarten J, Dijl V. 2014. The Tat system of Gram- positive bacteria ☆. BBA - Mol Cell Res 1843:1698–1706. 27. Freudl R. 2013. Leaving home ain’t easy: Protein export systems in Gram-positive bacteria. Res Microbiol 164. 28. Warne B, Harkins CP, Harris SR, Vatsiou A, Stanley-Wall N, Parkhill J, Peacock SJ, Palmer T, Holden MTG. The Ess/Type VII secretion system of Staphylococcus aureus shows unexpected genetic diversity. 29. Chatterjee SS, Joo H-S, Duong AC, Dieringer TD, Tan VY, Song Y, Fischer ER, Cheung GYC, Li M, Otto M. 2013. Essential Staphylococcus aureus toxin export system. Nat Med 19:364–7. 30. Yamane K, Bunai K, Kakeshita H. Protein Traffic for Secretion and Related Machinery of Bacillus subtilis. 31. Göthel SF, Marahiel MA. 1999. Peptidyl-prolyl cis-trans isomerases, a superfamily of ubiquitous folding catalysts. C Cell Mol Life Sci 55:423–436. 32. Stewart DE, Sarkar A, Wampler JE. 1990. Occurrence and role ofcis peptide bonds in protein structures. J Mol Biol. 33. Brandts JF, Halvorson HR, Brennan M. 1975. Consideration of the Possibility That the Slow Step in Protein Denaturation Reactions is Due to Cis-Trans Isomerism of Proline Residues. Biochemistry. 34. Lang K, Schmid FX, Fischer G. 1987. Catalysis of protein folding by prolyl isomerase. Nature. 35. Unal CM, Steinert M. 2014. Microbial Peptidyl-Prolyl cis/trans Isomerases (PPIases): Virulence Factors and Potential Alternative Drug Targets. Microbiol Mol Biol Rev 78:544–571. 36. Fischer G, Bang H, Mech C. 1984. [Determination of enzymatic catalysis for the cis-trans-isomerization of peptide binding in proline-containing peptides]. Biomed Biochim Acta 43:1101–11. 37. Takahashi N, Hayano T, Suzuki M. 1989. Peptidyl-prolyl cis-trans isomerase is the cyclosporin A-binding protein cyclophilin. Nature. 38. Harding MW, Galat A, Uehling DE, Schreiber SL. 1989. A receptor for the immuno-suppressant FK506 is a cis-trans peptidyl-prolyl isomerase. Nature. 39. Rahfeld JU, Rücknagel KP, Schelbert B, Ludwig B, Hacker J, Mann K, Fischer G. 1994. Confirmation of the existence of a third family among peptidyl-prolyl cis/trans isomerases Amino acid sequence and recombinant production of parvulin. FEBS Lett. 40. Hennig L, Christner C, Kipping M, Schelbert B, Rücknagel KP, Grabley S, 139

Küllertz G, Fischer G. 1998. Selective inactivation of parvulin-like peptidyl-prolyl cis/trans isomerases by juglone. Biochemistry. 41. Skagia A, Vezyri E, Sigala M, Kokkinou A, Karpusas M, Venieraki A, Katinakis P, Dimou M. 2017. Structural and functional analysis of cyclophilin PpiB mutants supports an in vivo function not limited to prolyl isomerization activity. Genes to Cells 22:32–44. 42. Zhang X-C, Wang W-D, Wang J-S, Pan J-C, De La Rosa M. 2013. PPIase independent chaperone-like function of recombinant human Cyclophilin A during arginine kinase refolding. FEBS Lett 587:666–672. 43. Kramer G, Patzelt H, Rauch T, Kurz TA, Vorderwülbecke S, Bukau B, Deuerling E. 2004. Trigger Factor Peptidyl-prolyl cis/trans Isomerase Activity Is Not Essential for the Folding of Cytosolic Proteins in Escherichia coli. J Biol Chem 279:14165–14170. 44. Geitner AJ, Schmid FX. 2012. Combination of the human prolyl isomerase FKBP12 with unrelated chaperone domains leads to chimeric folding enzymes with high activity. J Mol Biol. 45. Jakob RP, Zoldák G, Aumüller T, Schmid FX. 2009. Chaperone domains convert prolyl isomerases into generic catalysts of protein folding. Proc Natl Acad Sci U S A. 46. Merz F, Hoffmann A, Rutkowska A, Zachmann-Brand B, Bukau B, Deuerling E. 2006. The C-terminal domain of Escherichia coli trigger factor represents the central module of its chaperone activity. J Biol Chem 281:31963–31971. 47. Buchner J. 1996. Supervising the fold: functional principles of molecular chaperones. FASEB J 10:10–9. 48. Cahoon LA, Freitag NE, Seveau SM, Plano G. 2014. Listeria monocytogenes virulence factor secretion: don’t leave the cell without a chaperone. 49. Roset MS, Fernández LG, Delvecchio VG, Briones G. Intracellularly Induced Cyclophilins Play an Important Role in Stress Adaptation and Virulence of Brucella abortus. 50. Reffuveille F, Connil N, Sanguinetti M, Posteraro B, Chevalier S, Auffray Y, Rince A. 2012. Involvement of Peptidylprolyl cis/trans Isomerases in Enterococcus faecalis Virulence. 51. Iii FA, Xayarath B, Whisstock JC, Freitag NE. Functional analysis of the Listeria monocytogenes secretion chaperone PrsA2 and its multiple contributions to bacterial virulence. 52. Wahlström E, Vitikainen M, Kontinen VP, Sarvas M. 2003. The extracytoplasmic folding factor PrsA is required for protein secretion only in the presence of the cell wall in Bacillus subtilis. Microbiology 149:569–577. 140

53. Vitikainen M, Pummi T, Airaksinen U, Wahlström E, Wu H, Sarvas M, Kontinen VP. 2001. Quantitation of the capacity of the secretion apparatus and requirement for PrsA in growth and secretion of α-amylase in Bacillus subtilis. J Bacteriol 183:1881–1890. 54. Alonzo F, Port GC, Cao M, Freitag NE. 2009. The posttranslocation chaperone PrsA2 contributes to multiple facets of Listeria monocytogenes pathogenesis. Infect Immun 77:2612–2623. 55. Heikkinen O, Seppala R, Tossavainen H, Heikkinen S, Koskela H, Permi P, Kilpeläinen I. 2009. Solution structure of the parvulin-type PPIase domain of Staphylococcus aureus PrsA – Implications for the catalytic mechanism of parvulins. BMC Struct Biol 9. 56. Roch M, Lelong E, Panasenko OO, Sierra R, Renzoni A, Kelley WL. 2019. Thermosensitive PBP2a requires extracellular folding factors PrsA and HtrA1 for Staphylococcus aureus MRSA β-lactam resistance. Commun Biol 2:1–11. 57. Jousselin A, Manzano C, Biette A, Reed P, Pinho MG, Rosato AE, Kelley WL, Renzoni A. 2015. The Staphylococcus aureus Chaperone PrsA Is a New Auxiliary Factor of Oxacillin Resistance Affecting Penicillin-Binding Protein 2A. 58. Lin M-H, Li C-C, Shu J-C, Chu H-W, Liu C-C, Wu C-C. 2018. Exoproteome Profiling Reveals the Involvement of the Foldase PrsA in the Cell Surface Properties and Pathogenesis of Staphylococcus aureus. Proteomics 18:1700195. 59. Dimou M, Venieraki A, Katinakis P. 123AD. Microbial cyclophilins: specialized functions in virulence and beyond. World J Microbiol Biotechnol 1. 60. Roset MS, Fernández LG, Delvecchio VG, Briones G. 2013. Intracellularly Induced Cyclophilins Play an Important Role in Stress Adaptation and Virulence of Brucella abortus. 61. Skagia A, Zografou C, Vezyri E, Venieraki A, Katinakis P, Dimou M. 2016. Cyclophilin PpiB is involved in motility and biofilm formation via its functional association with certain proteins. Genes to Cells. 62. Pandey S, Sharma A, Tripathi D, Kumar A, Khubaib M, Bhuwan M, Chaudhuri TK, Hasnain SE, Ehtesham NZ. Mycobacterium tuberculosis Peptidyl-Prolyl Isomerases Also Exhibit Chaperone like Activity In-Vitro and In-Vivo. 63. Wiemels RE, Cech SM, Meyer NM, Burke CA, Weiss A, Parks AR, Shaw LN, Carroll RK. 2017. An intracellular peptidyl-prolyl cis/ trans isomerase is required for folding and activity of the Staphylococcus aureus secreted virulence factor nuclease. J Bacteriol 199. 64. Hoffmann A, Bukau B, Kramer G. 2010. Structure and function of the molecular chaperone Trigger Factor. BBA - Mol Cell Res 1803:650–661. 65. Duong F, Eichler J, Price A, Leonard MR, Wickner W. 1997. Biogenesis of the 141

Gram-Negative Review Bacterial Envelope Inner Membrane Translocation Large polar proteins or polar domains of integral membrane proteins cross the membrane through preprotein. Cell 91:567–573. 66. Haldar S, Tapia-Rojo R, Eckels EC, Valle-Orero J, Fernandez JM. 2017. Trigger factor chaperone acts as a mechanical foldase. Nat Commun 8:668. 67. Wen ZT, Suntharaligham P, Cvitkovitch DG, Burne RA. 2005. Trigger factor in Streptococcus mutans is involved in stress tolerance, competence development, and biofilm formation. Infect Immun 73:219–225. 68. Nelson, DC, Garbe, J, Collin, M. 2011. The cysteine proteinase SpeB from Streptococcus pyogenes – a potent modifier of immunologically important host and bacterial proteins . Biol Chem 392. 69. Wu T, Zhao Z, Zhang L, Ma H, Lu K, Ren W, Liu Z, Chang H, Bei W, Qiu Y, Chen H. 2011. Trigger factor of Streptococcus suis is involved in stress tolerance and virulence. Microb Pathog 51:69–76. 70. Bigot A, Botton E, Dubail I, Charbit A. 2006. A homolog of Bacillus subtilis trigger factor in Listeria monocytogenes is involved in stress tolerance and bacterial virulence. Appl Environ Microbiol 72:6623–6631. 71. Cohen A, Troib S, Dotan S, Najmuldeen H, Yesilkaya H, Kushnir T, Shagan M, Portnoi M, Nachmani H, Benisty R, Tal M, Ellis R, Chalifa-Caspi V, Dagan R, Nebenzahl YM. 2019. Streptococcus pneumoniae Cell Wall-Localized Trigger Factor Elicits a Protective Immune Response and Contributes to Bacterial Adhesion to the Host. Sci Rep 9:1–12. 72. Lu T, DeLeo FR. 2015. Pathogenesis of Staphylococcus aureus in Humans, p. 711–748. In Human Emerging and Re-emerging Infections. John Wiley & Sons, Inc., Hoboken, NJ, USA. 73. Schilcher K, Andreoni F, Uchiyama S, Ogawa T, Schuepbach RA, Zinkernagel AS. 2014. Increased neutrophil extracellular trap-mediated Staphylococcus aureus clearance through inhibition of nuclease activity by clindamycin and immunoglobulin. J Infect Dis. 74. Olson ME, Nygaard TK, Ackermann L, Watkins RL, Zurek OW, Pallister KB, Griffith S, Kiedrowski MR, Flack CE, Kavanaugh JS, Kreiswirth BN, Horswill AR, Voyich JM. Staphylococcus aureus Nuclease Is an SaeRS-Dependent Virulence Factor. 75. Thammavongsa V, Missiakas DM, Schneewind O. Staphylococcus aureus degrades neutrophil extracellular traps to promote immune cell death. 76. Moormeier DE, Bose JL, Horswill AR, Bayles KW. Temporal and Stochastic Control of Staphylococcus aureus Biofilm Development. 77. Mann EE, Rice KC, Boles BR, Endres JL, Ranjit D, Chandramohan L, Tsang LH, 142

Smeltzer MS, Horswill AR, Bayles KW. 2009. Modulation of eDNA release and degradation affects Staphylococcus aureus biofilm maturation. PLoS One 4. 78. Kiedrowski MR, Kavanaugh JS, Malone CL, Mootz JM, Voyich JM, Smeltzer MS, Bayles KW, Horswill AR. 2011. Nuclease modulates biofilm formation in community-associated methicillin-resistant staphylococcus aureus. PLoS One 6. 79. Ünal CM, Steinert M. Microbial Peptidyl-Prolyl cis/trans Isomerases (PPIases): Virulence Factors and Potential Alternative Drug Targets. 80. Valentino MD, Foulston L, Sadaka A, Kos VN, Villet RA, Maria JS, Lazinski DW, Camilli A, Walker S, Hooper DC, Gilmore MS. Genes Contributing to Staphylococcus aureus Fitness in Abscess-and Infection-Related Ecologies. 81. Huai Q, Kim H-Y, Liu Y, Zhao Y, Mondragon A, Liu JO, Ke H. Crystal structure of calcineurin–cyclophilin– cyclosporin shows common but distinct recognition of immunophilin–drug complexes. 82. Davis TL, Walker JR, Ouyang H, MacKenzie F, Butler-Cole C, Newman EM, Eisenmesser EZ, Dhe-Paganon S. 2008. The crystal structure of human WD40 repeat-containing peptidylprolyl isomerase (PPWD1). FEBS J 275:2283–2295. 83. Biasini M, Bienert S, Waterhouse A, Arnold K, Studer G, Schmidt T, Kiefer F, Cassarino TG, Bertoni M, Bordoli L, Schwede T. 2014. SWISS-MODEL: Modelling protein tertiary and quaternary structure using evolutionary information. Nucleic Acids Res 42:252–258. 84. Arnold K, Bordoli L, Kopp J, Schwede T. 2006. The SWISS-MODEL workspace: A web-based environment for protein structure homology modelling. Bioinformatics 22:195–201. 85. Benkert P, Biasini M, Schwede T. 2011. Toward the estimation of the absolute quality of individual protein structure models. Bioinformatics 27:343–350. 86. Bose JL, Fey PD, Bayles KW. 2013. Genetic tools to enhance the study of gene function and regulation in Staphylococcus aureus. Appl Environ Microbiol. 87. Noel D, Nikaido K, Ames GFL. 1979. A Single Amino Acid Substitution in a Histidine-Transport Protein Drastically Alters Its Mobility in Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis. Biochemistry 18:4159–4165. 88. Spaan AN, Reyes-Robles T, Badiou C, Cochet S, Boguslawski KM, Yoong P, Day CJ, De Haas CJC, Van Kessel KPM, Vandenesch F, Jennings MP, Le C, Kim V, Colin Y, Van Strijp JAG, Henry T, Torres VJ. 2015. Staphylococcus aureus targets the Duffy antigen receptor for chemokines (DARC) to lyse erythrocytes HHS Public Access. J Cell Host Microbe 18:363–370. 89. Bhakdil S. 1991. Alpha-Toxin of Staphylococcus aureus 55:733–751. 90. Wang R, Braughton KR, Kretschmer D, Bach T-HL, Queck SY, Li M, Kennedy 143

AD, Dorward DW, Klebanoff SJ, Peschel A, DeLeo FR, Otto M. 2007. Identification of novel cytolytic peptides as key virulence determinants for community-associated MRSA. Nat Med 13:1510. 91. Cheung GYC, Duong AC, Otto M. 2012. Direct and synergistic hemolysis caused by Staphylococcus phenol-soluble modulins: implications for diagnosis and pathogenesis. Microbes Infect 14:380–386. 92. Alonzo F, Freitag NE. 2010. Listeria monocytogenes PrsA2 is required for virulence factor secretion and bacterial viability within the host cell cytosol. Infect Immun 78:4944–4957. 93. Ma Y, Bryant AE, Salmi DB, Hayes-Schroer SM, Mcindoo E, Aldape MJ, Stevens DL. 2006. Identification and Characterization of Bicistronic speB and prsA Gene Expression in the Group A Streptococcus. J Bacteriol 188:7626–7634. 94. Hermans PWM, Adrian P V., Albert C, Estevão S, Hoogenboezem T, Luijendijk IHT, Kamphausen T, Hammerschmidt S. 2006. The streptococcal lipoprotein rotamase A (SlrA) is a functional peptidyl-prolyl isomerase involved in pneumococcal colonization. J Biol Chem 281:968–976. 95. Cho K, Arimoto T, Igarashi T, Yamamoto M. 2013. Involvement of lipoprotein PpiA of Streptococcus gordonii in evasion of phagocytosis by macrophages. Mol Oral Microbiol 28:379–391. 96. Chakraborty A, Das I, Datta R, Sen B, Bhattacharyya D, Mandal C, Datta AK. 2002. A single-domain cyclophilin from Leishmania donovani reactivates soluble aggregates of adenosine kinase by isomerase-independent chaperone function. J Biol Chem 277:47451–47460. 97. Chakraborty A, Sen B, Datta R, Datta AK. 2004. Isomerase-Independent Chaperone Function of Cyclophilin Ensures Aggregation Prevention of Adenosine Kinase Both in vitro and under in vivo Conditions †. Biochemistry 43:11862– 11872. 98. Davis A, Moore IB, Parker DS, Taniuchi H. 1977. Nuclease B. A possible precursor of nuclease A, an extracellular nuclease of Staphylococcus aureus. J Biol Chem 252:6544–53. 99. Rauzier J, Reyrat J-M, Recchi C, Gicquel B. 2002. Signal-sequence-independent secretion of the staphylococcal nuclease in Mycobacterium smegmatis. Microbiology 148:529–536. 100. Suciu D, Inouye M. 1996. The 19-residue pro-peptide of staphylococcal nuclease has a profound secretion-enhancing ability in Escherichia coli. Mol Microbiol 21:181–195. 101. Bose JL. 2014. Genetic Manipulation of Staphylococci, p. 101–111. In Methods in molecular biology (Clifton, N.J.). 144

102. Fey PD, Endres JL, Yajjala VK, Widhelm TJ, Boissy RJ, Bose JL, Bayles KW. 2013. A genetic resource for rapid and comprehensive phenotype screening of nonessential Staphylococcus aureus genes. MBio 4:1–8. 103. Malachowa N, Kobayashi SD, Braughton KR, DeLeo FR. 2013. Mouse Model of Staphylococcus aureus Skin Infection, p. 109–116. In . Humana Press, Totowa, NJ. 104. Carroll RK, Rivera FE, Cavaco CK, Johnson GM, Martin D, Shaw LN. 2014. The lone S41 family C-terminal processing protease in Staphylococcus aureus is localized to the cell wall and contributes to virulence. Microbiology 160:1737– 1748. 105. Highlander SK, Hultén KG, Qin X, Jiang H, Yerrapragada S, Mason EO, Shang Y, Williams TM, Fortunov RM, Liu Y, Igboeli O, Petrosino J, Tirumalai M, Uzman A, Fox GE, Cardenas AM, Muzny DM, Hemphill L, Ding Y, Dugan S, Blyth PR, Buhay CJ, Dinh HH, Hawes AC, Holder M, Kovar CL, Lee SL, Liu W, Nazareth L V., Wang Q, Zhou J, Kaplan SL, Weinstock GM. 2007. Subtle genetic changes enhance virulence of methicillin resistant and sensitive Staphylococcus aureus. BMC Microbiol 7:1–14. 106. Kreiswirth BN, Löfdahl S, Betley MJ, O’Reilly M, Schlievert PM, Bergdoll MS, Novick RP. 1983. The toxic shock syndrome exotoxin structural gene is not detectably transmitted by a prophage. Nature 305:709–12. 107. Sullivan MA, Yasbin RE, Young FE. 1984. New shuttle vectors for Bacillus subtilis and Escherichia coli which allow rapid detection of inserted fragments. Gene 29:21–26. 108. Jenkins A, Diep BA, Mai TT, Vo NH, Warrener P, Suzich J, Stover CK, Sellman BR. 2015. Differential Expression and Roles of Staphylococcus aureus Virulence Determinants during Colonization and Disease. MBio 6:e02272-14. 109. Joo HS, Chatterjee SS, Villaruz AE, Dickey SW, Tan VY, Chen Y, Sturdevant DE, Ricklefs SM, Otto M. 2016. Mechanism of gene regulation by a staphylococcus aureus toxin. MBio. 110. Kobayashi SD, Malachowa N, Whitney AR, Braughton KR, Gardner DJ, Long D, Wardenburg JB, Schneewind O, Otto M, DeLeo FR. 2011. Comparative analysis of USA300 virulence determinants in a rabbit model of skin and soft tissue infection. J Infect Dis 204:937–941. 111. Wilke GA, Wardenburg JB. 2010. Role of a disintegrin and metalloprotease 10 in Staphylococcus aureus -hemolysin-mediated cellular injury. Proc Natl Acad Sci 107:13473–13478. 112. Cheung GYC, Joo HS, Chatterjee SS, Otto M. 2014. Phenol-soluble modulins - critical determinants of staphylococcal virulence. FEMS Microbiol Rev 38:698– 719. 145

113. Queck SY, Jameson-Lee M, Villaruz AE, Bach T-HL, Khan BA, Sturdevant DE, Ricklefs SM, Li M, Otto M. 2008. RNAIII-Independent Target Gene Control by the agr Quorum-Sensing System: Insight into the Evolution of Virulence Regulation in Staphylococcus aureus. Mol Cell 32:150–158. 114. Berube BJ, Sampedro GR, Otto M, Wardenburg JB. 2014. The psmα locus regulates production of Staphylococcus aureus alpha-toxin during infection. Infect Immun 82:3350–3358. 115. Kofron JL, Kuzmic P, Kishore V, Colon-Bonilla E, Rich DH. 1991. Determination of kinetic constants for peptidyl prolyl cis-trans isomerases by an improved spectrophotometric assay. Biochemistry 30:6127–6134. 116. Pandey S, Sharma A, Tripathi D, Kumar A, Khubaib M, Bhuwan M, Chaudhuri TK, Hasnain SE, Ehtesham NZ. 2016. Mycobacterium tuberculosis Peptidyl-Prolyl Isomerases Also Exhibit Chaperone like Activity In-Vitro and In-Vivo. 117. Keogh RA, Zapf RL, Wiemels RE, Wittekind MA, Carroll RK. 2018. The Intracellular Cyclophilin PpiB Contributes to the Virulence of Staphylococcus aureus Independently of Its Peptidyl-Prolyl cis/trans Isomerase Activity . Infect Immun 86:1–13. 118. Alonzo III F, Benson MA, Chen J, Novick RP, Shopsin B, Torres VJ. 2012. Staphylococcus aureus leucocidin ED contributes to systemic infection by targeting neutrophils and promoting bacterial growth in vivo. Mol Microbiol 83:423–435. 119. Wang X, Thompson CD, Weidenmaier C, Lee JC. 2018. Release of Staphylococcus aureus extracellular vesicles and their application as a vaccine platform. Nat Commun 9. 120. Otto H-SJ. 2014. The isolation and analysis of phenol-soluble modulins of Staphylococcus epidermidis. Methods Mol Biol 93–100. 121. Schallenberger MA, Niessen S, Shao C, Fowler BJ, Romesberg FE. 2012. Type I signal peptidase and protein secretion in Staphylococcus aureus. J Bacteriol 194:2677–2686. 122. Antelmann H, Darmon E, Noone D, Veening JW, Westers H, Bron S, Kuipers OP, Devine KM, Hecker M, Van Dijl JM. 2003. The extracellular proteome of Bacillus subtilis under secretion stress conditions. Mol Microbiol 49:143–156. 123. Rigoulay C, Entenza JM, Halpern D, Widmer E, Moreillon P, Poquet I, Gruss A. 2005. Comparative Analysis of the Roles of HtrA-Like Surface Proteases in Two Virulent Staphylococcus aureus Strains. Infect Immun 73:563–572. 124. McCallum N, Spehar G, Bischoff M, Berger-Bächi B. 2006. Strain dependence of the cell wall-damage induced stimulon in Staphylococcus aureus. Biochim Biophys Acta - Gen Subj. 146

125. Schneewind O, Missiakas D. 2014. Sec-secretion and sortase-mediated anchoring of proteins in Gram-positive bacteria. Biochim Biophys Acta - Mol Cell Res. 126. Pugsley AP. 1993. The complete general secretory pathway in gram-negative bacteria. Microbiol Rev 57:50–108. 127. Jousselin A, Renzoni A, Andrey DO, Monod A, Lew DP, Kelley WL. 2012. The Posttranslocational Chaperone Lipoprotein PrsA Is Involved in both Glycopeptide and Oxacillin Resistance in Staphylococcus aureus. Antimicrob Agents Chemother 56:3629–3640. 128. Herbert S, Ziebandt AK, Ohlsen K, Schäfer T, Hecker M, Albrecht D, Novick R, Götz F. 2010. Repair of global regulators in Staphylococcus aureus 8325 and comparative analysis with other clinical isolates. Infect Immun 78:2877–2889. 129. Pandey A, Mann M, Aebersold R, Goodlett DR, Link AJ, Eng J, Schieltz DM, Carmack E. 2000. Empirical Statistical Model To Estimate the Accuracy of Peptide Identifications Made by MS/MS and Database Search. Fenyo, D Curr Opin Biotechnol 405:37. 130. Nesvizhskii AI, Keller A, Kolker E, Aebersold R. 2003. A Statistical Model for Identifying Proteins by Tandem Mass Spectrometry. Anal Chem 75:4646–4658. 131. Zapf RL, Wiemels RE, Keogh RA, Holzschu DL, Howell KM, Trzeciak E, Caillet AR, King KA, Selhorst SA, Naldrett MJ, Bose JL, Carroll RK. 2019. The Small RNA Teg41 Regulates Expression of the Alpha Phenol-Soluble Modulins and Is Required for Virulence in Staphylococcus aureus . MBio 10. 132. Boles BR, Thoende M, Roth AJ, Horswill AR. 2010. Identification of genes involved in polysaccharide- independent Staphylococcus aureus biofilm formation. PLoS One 5. 133. Timerman AP, Wiederrecht G, Marcy A, Fleischer S. 1995. Characterization of an exchange reaction between soluble FKBP-12 and the FKBP·Ryanodine receptor complex. Modulation by FKBP mutants deficient in peptidyl-prolyl isomerase activity. J Biol Chem. 134. Behrens S, Maier R, De Cock H, Schmid FX, Gross CA. 2001. The SurA periplasmic PPIase lacking its parvulin domains functions in vivo and has chaperone activity. EMBO J. 135. Wickner S, Maurizi MR, Gottesman S. 1999. Posttranslational quality control: Folding, refolding, and degrading proteins. Science (80- ) 286:1888–1893. 136. Keogh RA, Zapf RL, Trzeciak E, Null GG, Wiemels RE, Carroll RK. 2019. Novel Regulation of Alpha-Toxin and the Phenol-Soluble Modulins by Peptidyl-Prolyl cis / trans Isomerase Enzymes in Staphylococcus aureus 1–22. 137. Hesterkamp T, Bukau B. 1996. Identification of the prolyl isomerase domain of Escherichia coli trigger factor. FEBS Lett 385:67–71. 147

138. Deuerling E, Schulze-Specking A, Tomoyasu T, Mogk A, Bukau B. 1999. Trigger factor and DnaK cooperate in folding of newly synthesized proteins. Nature. 139. Teter SA, Houry WA, Ang D, Tradler T, Rockabrand D, Fischer G, Blum P, Georgopoulos C, Hartl FU. 1999. Polypeptide flux through bacterial Hsp70: DnaK cooperates with trigger factor in chaperoning nascent chains. Cell 97:755–765. 140. Lyon WR, Caparon MG. 2003. Trigger Factor-Mediated Prolyl Isomerization Influences Maturation of the Streptococcus pyogenes Cysteine Protease. J Bacteriol 185:3661–3667. 141. Marroquin S, Gimza B, Tomlinson B, Stein M, Frey A, Keogh RA, Zapf R, Todd DA, Cech NB, Carroll RK, Shaw LN. 2019. MroQ is a Novel Abi-domain Protein That Influences Virulence Gene Expression in Staphylococcus aureus via Modulation of Agr Activity . Infect Immun. 142. Hristozova N, Tompa P, Kovacs D. A Novel Method for Assessing the Chaperone Activity of Proteins. 143. Belli WA, Marquis RE. 1991. Adaptation of Streptococcus mutans and Enterococcus hirae to acid stress in continuous culture. Appl Environ Microbiol. 144. Genevaux P, Keppel F, Langendijk-genevaux PS, Hartl FU, Georgopoulos C. 2004. In vivo analysis of the overlapping functions ofDnaK and trigger factor 5. 145. Foster TJ, Geoghegan JA, Ganesh VK, Diseases I. 2017. Adhesion, invasion and evasion: the many functions of the surface proteins of Staphylococcus aureus 12:49–62. 146. Kwiecinski JM, Crosby HA, Valotteau C, Hippensteel JA, Nayak MK, Chauhan AK, Schmidt EP, Dufrêne YF, Horswill AR. 2019. Staphylococcus aureus adhesion in endovascular infections is controlled by the arlRS–MgrA signaling cascadePLoS Pathogens. 147. Paharik AE, Horswill AR, Roy J, City I. 2016. The Staphylococcal Biofilm: Adhesins, regulation, and host response Alexandra 4:1–48. 148. Mccourt J, O’Halloran DP, Mccarthy H, O’Gara JP, Geoghegan JA. 2014. Fibronectin-binding proteins are required for biofilm formation by community- associated methicillin-resistant Staphylococcus aureus strain LAC. FEMS Microbiol Lett. 149. Sanchez CJ, Shivshankar P, Stol K, Trakhtenbroit S, Sullam PM, Sauer K, Hermans PWM, Orihuela CJ. 2010. The pneumococcal serine-rich repeat protein is an intraspecies bacterial adhesin that promotes bacterial aggregation in Vivo and in biofilms. PLoS Pathog 6:33–34. 150. Crooke E, Wickner W. 1987. Trigger factor: a soluble protein that folds pro- OmpA into a membrane-assembly-competent form. Proc Natl Acad Sci U S A 84:5216–5220. 148

151. Tomoyasu T, Mogk A, Langen H, Goloubinoff P, Bukau B. 2001. Genetic dissection of the roles of chaperones and proteases in protein folding and degradation in the Escherichia coli cytosol. Mol Microbiol 40:397–413. 152. Singh VK, Utaida S, Jackson LS, Jayaswal RK, Wilkinson BJ, Chamberlain NR. 2019. Role for dnaK locus in tolerance of multiple stresses in Staphylococcus aureus 3162–3173. 153. Kolar SL, Antonio Ibarra J, Rivera FE, Mootz JM, Davenport JE, Stevens SM, Horswill AR, Shaw LN. 2013. Extracellular proteases are key mediators of Staphylococcus aureus virulence via the global modulation of virulence- determinant stability. Microbiologyopen 2:18–34. 154. Baba T, Bae T, Schneewind O, Takeuchi F, Hiramatsu K. 2008. Genome sequence of Staphylococcus aureus strain newman and comparative analysis of staphylococcal genomes: Polymorphism and evolution of two major pathogenicity islands. J Bacteriol 190:300–310. 155. Memmi G, Nair DR, Cheung A. 2012. Role of ArLRS in autolysis in methicillin- sensitive and methicillin-resistant Staphylococcus aureus strains. J Bacteriol 194:759–767. 156. Shakeri F, Shojai A, Golalipour M, Alang SR, Vaez H, Ghaemi EA. 2010. Spa diversity among MRSA and MSSA strains of Staphylococcus aureus in north of Iran. Int J Microbiol 2010:12–16. 157. McCarthy H, Rudkin JK, Black NS, Gallagher L, O’Neill E, O’Gara JP. 2015. Methicillin resistance and the biofilm phenotype in staphylococcus aureus. Front Cell Infect Microbiol 5:1–9. 158. Ebner P, Luqman A, Reichert S, Hauf K, Popella P, Forchhammer K, Otto M, Götz F. 2017. Non-classical Protein Excretion Is Boosted by PSMα-Induced Cell Leakage. Cell Rep 20:1278–1286.

! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !

Thesis and Dissertation Services ! !