A Small RNA and DNA Binding Protein Contribute to Biofilm Development in Bartonella Henselae

University of South Florida Scholar Commons

Graduate Theses and Dissertations Graduate School

7-2-2019

A Small RNA and DNA Binding Protein Contribute to Biofilm Development in Bartonella henselae

Udoka Okaro University of South Florida, [email protected]

Follow this and additional works at: https://scholarcommons.usf.edu/etd

Part of the Biochemistry Commons, and the Microbiology Commons

Scholar Commons Citation Okaro, Udoka, "A Small RNA and DNA Binding Protein Contribute to Biofilm Development in Bartonella henselae" (2019). Graduate Theses and Dissertations. https://scholarcommons.usf.edu/etd/7878

This Dissertation is brought to you for free and open access by the Graduate School at Scholar Commons. It has been accepted for inclusion in Graduate Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected]. A Small RNA and DNA Binding Protein Contribute to Biofilm Development in

Bartonella henselae

Udoka Okaro

A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy with a concentration in Allergy and Infectious Diseases Department of Molecular Medicine Morsani College of Medicine University of South Florida

Major Professor: Burt Anderson, Ph.D. Robert Deschenes, Ph.D. Niketa Patel, Ph.D. Lindsey Shaw, Ph.D. Ximing Sun, Ph.D.

Date of Approval: th June 26 , 2019

Keywords: ncRNAs, Transcriptional regulator, Auto-transporter adhesin, Trans-acting RNA.

‘It takes a village to raise a child’

Lately, this African proverb has resonated with me. My life choices have been a culmination of positive interactions with a lot of people. From family to high school teachers,

college professors, career mentors, and more recently graduate school professor.

I would like to thank my advisor, Dr. Burt Anderson, for patiently guiding, teaching and

advising me throughout my time in his lab. Dr. Anderson did not only assume the role of a

professor, but he also mentored, made sure I was focused and that every achievement was directed

towards my long term career goal. His outlook on life and attitude toward science is inspiring. I

will forever be grateful for his counsel, opinion and recommendations.

I appreciate and recognize the time and effort all my committee members put towards my

project, committee meetings, or work in progress update. I would like to thank Dr. Deschenes, Dr.

Shaw, Dr. Patel and Dr. Sun for their contributions toward my projects, publications, and grants.

I would also like to thank Dr. Ghansah, Dr. Seyfang, Dr. Teng, Mr. Batson, and Dr. Combie

for the time and effort they spent writing recommendation letters, inputs towards grant

applications, cheers when I published and support towards scientific networking.

Graduate school is not a walk in the park, but I had a great support system both home and

abroad; Nicole Staviski, Emily Palumbo, Hung Nguyen, Chris Hinojo, and Ahmad Jalloh. We all

started as one cohort, supporting, cheering, studying, crying (well, some of us), celebrating each

other’s achievements. I am grateful for the internet because I have friends (Adaobi, Lota, Nkechi,

Nkem, Uche, Oby) who prayed, cried, and laughed with me and were always available. My friends

held me down emotionally and mentally.

To everyone else, the continuously changing team at the Anderson lab, Shonique, and

Meagan at the Ph.D. office, Michael Ramsamooj and the molecular medicine team, Helena and

Rohini, AMSGS, Amanda and Dr. Cha at the Muma microscopy core, thank you all for

contributing to my growth and development at USF.

To my family, the ones who always stood with me. My Papa was a college professor with a successful celebrated career. He made sure my siblings and I got every resource we needed academically- the power of education as he said, was priceless. To my mum, the woman who bore and raised me, you always supported my move in life, you gave up your career to raise five kids and it is saddening your life was cut short just when it was time to celebrate those kids. This degree is for you: because you always told me I could do anything. To my siblings’ (Somy, Muna, Ugo, and Jide), aunts/uncles, cousins, parents’ in-law, thank you for prayers and emotional support.

Finally, my husband Chino, my no. 1 fan, your support has been my greatest strength

throughout this journey. You were a great audience of one every time I needed to practice, you

agonized with me, advised and encouraged me when I felt I had nothing more to give and most of

all you celebrated all my awards and achievements like it was yours. I could not have asked for a

better life partner. To my two little girls, Bella and Sobi, my unconditional sources of love through

grad school, I pray that someday, when you both grow up, you will always remember that you can

do anything you put your mind to- mummy did it.

Thank God for life, for love, for family and great friends. It truly took a village to raise this

child.

Table of Contents

List of Tables ...... iv

List of Figures ...... v

List of Abbreviations...... vi

Abstract ...... viii

Chapter 1: Introduction ...... 1 1.1. Introduction to Bartonella species ...... 1 1.2. Bartonella henselae ...... 3 1.3. Human disease and treatment ...... 2 1.3.1. Endocarditis ...... 3 1.3.2. Cat scratch disease ...... 4 1.3.3. Bacillary angiomatosis ...... 4 1.3.4. Bacteremia ...... 5 1.3.5. Treatment considerations for Bartonella endocarditis ...... 6 1.4. B.henselae pathogenesis ...... 7 1.5. B.henselae virulence factors ...... 7 1.5.1 VirB/VirD4 type IV secretion system ...... 7 1.5.2 Bartonella adhesin A (BadA) ...... 8 1.5.3 Trw T4SS ...... 9 1.5.4 Other outer membrane proteins ...... 10 1.6 Biofilm regulation ...... 10 1.7 Non-coding RNAs ...... 11 1.8 Riboswitches and RNA thermometers ...... 13 1.9 Transcriptional factors ...... 16 1.10 Summary ...... 17

Chapter 2: B.henselae forms a biofilm ...... 18 2.1. Background ...... 18 2.2 Materials and Methods ...... 19 2.2.1. Bacterial strains...... 19 2.2.2. B.henselae forms a biofilm...... 20 2.2.3. BadA expression on B.henselae constructs...... 21 2.2.4. badA expression within a biofilm...... 21 2.2.5. The role of BadA in biofilm formation...... 22 2.2.6. The viability of cells within a biofilm...... 23 2.2.7. The components of a B.henselae Houston-1 biofilm...... 23 2.2.8. Environmental factors required for optimal biofilm formation ...... 23

i 2.3 Results ...... 24 2.3.1. B.henselae forms a biofilm...... 25 2.3.2. badA is highly expressed within a biofilm ...... 25 2.3.3. BadA plays a significant role in biofilm formation...... 27 2.3.4. badA expressing strains harbor more viable cells within a biofilm...... 29 2.3.5. B.henselae biofilm contains polysaccharides, protein, and eDNA ...... 29 2.3.6. badA expression and biofilm formation are influenced by environmental conditions...... 33 2.4. Conclusion ...... 37

Chapter 3: Elucidating the role of the 3' end of Brt1 ...... 44 3.1. Background ...... 44 3.2. Materials and Methods ...... 46 3.2.1. Bacterial strains ...... 46 3.2.2. Construction of B.henselae Houston-1-GFP strains...... 47 3.2.2.1. Construction of B.henselae Houston-1/pNS2 gfp ...... 47 3.2.2.2. Construction of B.henselae Houston-1/pNS2 brt1Δ3'-gfp ...... 48 3.2.2.3. Construction of B.henselae Houston-1/pNS2 brt1-gfp ...... 48 3.2.3. Fluorescence microscopy ...... 49 3.2.4. Flow cytometry ...... 49 3.2.5. Construction of B.henselae Houston-1 Brt1 Δ3' ...... 50 3.2.6. Biofilm formation in B.henselae Houston-1 Brt1Δ3' ...... 50 3.2.7. brt1, trp1, and badA expression in B.henselae Houston-1 Brt1Δ3' ...... 51 3.3. Results ...... 51 3.3.1. The 3'end of Brt1 acts as a transcriptional terminator...... 51 3.3.2. Biofilm formation is increased in B.henselae Houston-1 brt1Δ3' ...... 53 3.3.3. brt1, trp1, and badA expression in B.henselae Houston-1 brt1Δ3' ...... 54 3.4. Conclusion ...... 55

Chapter 4: Role of Trp1 in biofilm regulation ...... 58 4.1. Background ...... 58 4.2. Materials and Methods ...... 61 4.2.1. Construction and purification of SoluBL21/pET28a Trp1...... 61 4.2.2. Biotinylation, amplification, and purification of the badA promoter region...... 62 4.2.3. Electrophoretic mobility shift assay (EMSA) ...... 62 4.2.4. RT-qPCR of trp1 and badA in a biofilm community...... 63 4.2.5. badA promoter region pulldown ...... 63 4.3. Results ...... 64 4.3.1. Trp1 binds badA promoter region to activate gene transcription ...... 65 4.3.2. trp1 and badA expression is upregulated in a biofilm community ...... 65 4.3.3. Trp3, amongst other genes, also binds the badA promoter region...... 65 4.4. Conclusion ...... 66

ii Chapter 5: Role of Brt1 in biofilm regulation ...... 70 5.1. Background...... 70 5.2. Materials and Methods ...... 71 5.2.1. Construction of B.henselae Houston-1Δbrt1 ...... 71 5.2.2. RNA extraction and RT-qPCR...... 73 5.2.3. Biofilm formation in B.henselae Houston-1Δbrt1 ...... 73 5.2.4. In vitro transcription of Brt1 ...... 73 5.2.5. Brt1 binding partners ...... 73 5.2.6. In vivo expression of the brt, trp and badA system in fleas...... 74 5.3. Results...... 74 5.3.1. Biofilm formation B.henselae Houston-1Δbrt1 ...... 74 5.3.2. trp1 and badA expression are increased in B.henselae Houston- 1Δbrt1 ...... 75 5.3.3. Brt1 functions as a transacting RNA...... 77 5.3.4. B.henselae is able to survive in both cat flea and feces ...... 78 5.3.5. trp1/badA system is not expressed in the cat flea but upregulated in flea feces ...... 81 5.4. Conclusion ...... 81

Chapter 6. Discussion ...... 85 6.1. Potential treatment options/drug targets ...... 90 6.1.1. RNA targets ...... 91 6.1.2 Adhesins ...... 92 6.1.3 Transcription factors ...... 92

References ...... 95

Appendices ...... 111 Appendix A: Copyright permission for Chapter 1 ...... 112 Appendix B: Copyright permission for Figure 1.2...... 113 Appendix C: Copyright permission for Chapter 2 ...... 114

iii List of Tables

Table 1.1. Significant differences between the arthropod vector of B. henselae and the mammalian host ...... 16 Table 2.1. Bacterial strains and primers used for aim 2...... 20

Table 3.1. Bacterial strains and primers used for aim 3...... 46

Table 4.1. Bacterial strains and primers used for aim 4...... 62

Table 4.2. Proteins that bind to the badA promoter region...... 67

Table 5.1. Bacterial strains and primers used for aim 5...... 72

Table 5.2. mRNAs that bind to Brt1...... 78

List of Figures

Figure 1.1: Characteristics of Brt1 and Trp1...... 13

Figure 1.2: Line diagram showing the relationship between brt1 and trp1...... 15

Figure 2.1: Scanning electron micrograph of B. henselae growth and biofilm formation...... 24

Figure 2.2: Growth rates and badA transcript levels in B. henselae constructs...... 26

Figure 2.3: B. henselae biofilm formation using a 96 well e-plate...... 28

Figure 2.4: Scanning electron micrographs (SEM) of B. henselae biofilms...... 30

Figure 2.5: Constituents of a B. henselae biofilm...... 32

Figure 2.6: Optimum temperature for a B. henselae biofilm formation...... 34

Figure 2.7: Optimum pH of a B. henselae biofilm formation...... 36

Figure 3.1: Growth of B. henselae brt1-GFP constructs...... 49

Figure 3.2: Comparison of GFP intensity in Brt1-GFP constructs...... 52

Figure 3.3: Growth and biofilm analysis of B. henselae Houston1/brt1Δ3'...... 54

Figure 3.4: Cell viability and gene expression levels in B. henselae Houston-1/brt1Δ3'...... 55

Figure 4.1: EMSA image for Trp1 and badA promoter region...... 64

Figure 4.2: Gene expression levels in a B. henselae Houston-1 biofilm...... 65

Figure 5.1: Growth curve and biofilm analysis for B. henselae Houston-1/Δbrt1...... 75

Figure 5.2: Cell viability and gene expression levels in B. henselae Houston-1/Δbrt1 biofilm...... 76

Figure 5.3: Bacteria growth in the cat flea (C. felis) and feces ...... 80

Figure 5.4: Gene expression levels in the cat flea (C. felis)...... 81

Figure 6.1: The hypothesized model for biofilm regulation ...... 94

List of Abbreviations

asRNA antisense RNA

BA bacillary angiomatosis

BadA Bartonella adhesin a

BCNE blood culture negative endocarditis

Beps Bartonella effector proteins

Brt Bartonella regulatory transcript

CI cell index

CLSM confocal laser scanning microscopy

CSD cat scratch disease

CV crystal violet

EPS exopolysaccharide/extracellular polymeric substance

FHA filamentous hemagglutinin

HIV human immunodeficiency virus

HMDS hexamethyldisilazane

HTH helix-turn-helix

IGEM immuno-gold electron microscopy

ncRNA noncoding RNA

OMP outer membrane protein

PBS phosphate buffered saline

PI propidium iodide

PLA polylactic acid

PLGA poly lactic-co-glycolic acid

PPRs pattern recognition receptor

QS quorum sensing

RBP RNA binding protein

RBS ribosome binding site

RNAT RNA thermometer

RTCA real-time cell adhesion

SEM scanning electron microscopy sRNA small RNA

TAA trimeric autotransporter adhesin

TEM transmission electron microscope

TF transcription factor

TLR toll-like receptors

TRCF transcription-repair coupling factor

Trp transcriptional regulatory protein

TPM transcript per million bases

UTR untranslated region

XRE xenobiotic response element

vii

Abstract

A biofilm, which is associated with 80% of chronic infections in humans, is formed when

bacteria aggregate, attach to a substrate and secrete a matrix protecting the bacteria from host cell

defenses and antibiotics. Bartonella henselae (B. henselae) is the causative agent of cat scratch disease, persistent bacteremia, and one of the most frequently reported causes of blood-culture negative endocarditis (BCNE) in patients. The ability of B. henselae to adhere to the heart valve, form a biofilm and vegetation to cause endocarditis increases the morbidity and mortality rate in infected patients. The presence of a trimeric autotransporter adhesin (TAA) called Bartonella adhesin A (BadA) has been linked to biofilm formation in B. henselae. BadA is a protein of 3036 amino acids and a member of the TAAs found in Bartonella and other Gram-negative bacteria.

The function of BadA has been studied in vitro and is critical for agglutination, host cell adhesion and activation of a pro-angiogenic host response. However, very little is known about badA gene regulation or the molecular basis of biofilm formation. This work aims to determine whether BadA is necessary for the establishment of biofilms and how the bacteria regulate badA expression.

Using genetic mutations, real-time cell adhesion assay, RT-qPCR, and microscopy, it was shown that BadA is required for biofilm formation. Using an in-frame complete deletion strain of badA, a reduced ability to form a biofilm was observed which was restored in the deletion strain complemented with a partial badA. Analysis of the B. henselae transcriptome shows nine highly transcribed, homologous RNAs, termed Bartonella regulatory transcript (Brt1-9). The Brts are short-sized (<200 nucleotides), highly expressed, and located in an intergenic region indicative of small RNAs (sRNA). The Brts are predicted to form a stable stem and loop structure with a

viii potential terminator/riboswitch region on the 3′ end. Located ~20 nucleotides downstream of each

Brt is a poorly transcribed helix-turn-helix DNA binding protein gene termed transcriptional regulatory protein (trps 1-9). High brt transcription stops just before the start of the trp implicating the 3’ loop of the Brt as a terminating loop. Replacement of the trp with a gfp reporter gene shows that in the absence of the 3′ end of Brt1, gfp is transcribed. Also consistent with our findings, an increase in both the transcription of trp1 and badA and the formation of a biofilm in mutants of the brt1 gene was observed. Furthermore, to determine the role of the Trp in regulating badA, an electrophoretic mobility shift assay was carried out. The data confirms that Trp1 binds the promoter region of badA gene to regulate gene expression. In summary, the brt1/trp1 regulon affects badA transcription and biofilm formation in B. henselae. Understanding the mechanism and condition(s) by which the brt/trp regulatory system regulates badA is a plausible approach to the development of treatments that target the formation of biofilm-related diseases and persistent bacteremia in humans.

Chapter 1: Introduction

Note to Readers: Portions of this chapter are published in “Okaro, U., Addisu, A., Casanas, B.

and Anderson B. Bartonella Species, an Emerging Cause of Blood-Culture-Negative Endocarditis.

Clinical Microbiology Reviews, 2017. 30(3): p. 709-746.”, and has been included with permission

from the publisher.

1.1 Introduction to Bartonella Species

During World War 1, a disease plagued about 1 million soldiers in Europe including US

armies, one-fifth of the German and Austrian troops, and about one-third of the British forces.

Since then, the same outbreak has been reported in the United States, France, and Burundi [1-3].

This plague was called the ‘five-day fever,’ and in 1960, Bartonella quintana was isolated as the

etiologic agent [4]. Bartonella quintana belongs to the Bartonella genus, a group of fastidious,

Gram-negative, facultative intracellular pathogens with a unique intra-erythrocytic lifestyle. Other

species include Bartonella bacilliformis; the agent for Carrion’s disease and Bartonella henselae

(B. henselae); known for causing cat scratch disease (CSD). There are currently about 42 species

and subspecies in the Bartonella genus [5] that have been isolated from a variety of hosts such as

humans, rats, cattle, squirrels, cats, dogs, etc. [6-12]. The presence of Bartonella species in the

blood of these infected animal reservoirs usually does not lead to severe illness. This vast range

of infected animals, therefore, serves as a reservoir for potential zoonotic infection. Of the 42

species, 13 have been shown to infect humans with diseases like blood culture-negative

endocarditis (BCNE), verruga peruana, Oroya fever, lymphadenopathy, CSD, neuroretinitis, etc.

[5, 13-20]. Only two known species of Bartonella; B. bacilliformis and B. quintana, infect humans as their host reservoir. These two species together with B. henselae cause the vast majority of human diseases attributed to Bartonella species [21]. Bartonella is transmitted to mammalian hosts through a variety of blood-sucking arthropods such as fleas, lice, ticks, and sandflies [22]. It

is known to adapt to the gut of the obligate, blood-sucking arthropod vector and to the erythrocytes

of the mammalian host signaling a heme requirement for growth [23, 24]. This requirement for

heme is met in the laboratory through the use of hemoglobin, erythrocytes or hemin added to agar

base and culture media.

Diagnosis of Bartonella infection is confirmed in the lab using PCR detection of the 16S

rRNA or by serological assays to detect specific antibodies in patient sera [14, 25-27]. Giemsa, hematoxylin-eosin and Warthin-starry stains are used for histological detection of the bacterium in infected tissues [28-30].

The genome size for Bartonella ranges from 1.45Mbp to about 2.6 Mbp. The species associated with rodents possess larger genome sizes (from a range of 1.6- 2.6 Mbp) while the

species found in humans, B. bacilliformis and B. quintana have smaller genome size (1.45 - 1.5

Mbp) [31-33]. This genome size difference can be attributed to horizontal gene transfer because,

unlike the human-restricted species, the rodent-associated species carry more genes such as the

Type 4 secretion system (T4SS), trimeric autotransporter adhesin (TAA), and other adhesive genes

(e.g. outer membrane proteins and filamentous hemagglutinin) needed for host adaptability [33].

1.2 B. henselae

B. henselae was first isolated in 1992 from a febrile patient infected with human immunodeficiency virus (HIV) [12]. B. henselae is endemic worldwide and commonly causes long-term bacteremia in domestic and feral cats (Felis cattus). The transmission vector is the cat

2 flea (Ctenocephalides felis) [34, 35] with about 50% of cats having serological evidence of previous or current infection [36]. Transmission to humans occurs indirectly by inoculation of contaminated flea feces through the scratch of a cat [37] and occasionally through a bite [38], but cat-to-cat transmission by the cat flea without direct contact transmission has been demonstrated

[39]. Possible cat-to-cat transmission mechanisms include flea bite and ingestion of fleas and flea feces [40]. There has been no data to confirm flea-to-human transmission to date [34, 38, 41].

Bacteremic healthy cats with B. henselae are associated with bacillary angiomatosis (BA) and CSD in their human contacts. The majority of CSD cases occur in pediatric patients (age 5 –

9) and for people living in the southern United States in general. The estimates predict 22,000 CSD cases in the United States each year, with approximately 2,000 hospitalizations. Most cases are reported between September-January [42]. Some attribute the seasonal prevalence to cats breeding patterns, the peak period of domestic cat adoptions, and the temporal presence of fleas on cats that spread the bacteria among the cat population [42]. Bacteremic cats are more likely to infect fleas resulting in flea vector persistence [43]. B. henselae is detected experimentally in oral swabs from infected cats [38, 44], is capable of replicating in the gut of the cat flea, shedding and surviving in flea feces up to 3 days after exposure [45-47].

1.3 Human Diseases and Treatment

1.3.1 Endocarditis.

Also called infective endocarditis (IE), is an inflammation of the inner lining of the heart.Bacterial endocarditis, the most common type, occurs when bacteria attaches and grows in the heart. B. henselae and B.quintana represent 90% of all cases of endocarditis caused by

Bartonella species [48]. In patients with other bacterial agents endocarditis such as staphylococcal endocarditis, B. henselae may also be involved as a co-infectant [49]. Bartonella endocarditis

3 usually occurs in people with a history of past valve diseases, cat or cat flea exposure [48-50]. B. henselae and B. quintana exhibit tropism for endothelial cells and are able to replicate and persist in endothelial cells.

1.3.2 Cat-scratch disease (CSD).

B. henselae causes CSD in humans and is transmitted by a cat’s scratch, bite or lick [51].

A cat bite or scratch followed by local inflammation 10-14 days later and significant enlargement of regional lymph nodes (lymphadenopathy) are the classic symptoms of CSD. Systemic symptoms like fever and malaise typically develop and could last for several weeks. Most cases are benign and usually resolve spontaneously in immunocompetent persons. Severe complications such as meningitis, osteomyelitis, encephalitis, and endocarditis are also known to occur primarily in immunocompromised individuals [52, 53]. Occuloglandular syndrome (also called Parinaud’s occuloglandular syndrome) is an ocular symptom of CSD of granulomatous conjunctivitis with pre and postauricular lymphadenopathy; an earlier case report highlights an increasing spectrum of ocular involvement by CSD presenting as an optic nerve granuloma [54, 55]. Studies show that more than 50% of domestic cats are Bartonella carriers but seldom show signs or symptoms of infection [41, 56, 57]. The cat flea mainly transmits the bacteria between cats although other arthropods, mainly ticks of the genus Ixodes are implicated as vectors [58].

1.3.3 Bacillary Angiomatosis (BA).

BA is a vascular endothelial proliferative disease that develops as a single or multiple papulonodular cutaneous lesions. The etiologic agents are B. henselae and B. quintana [13, 25,

59]. BA was initially described in patients with HIV but is also described in both immunocompetent hosts [60] and in immunocompromised status such as organ transplant recipients on immunosuppressive therapy. The cutaneous lesions are highly vascular, bruising or

easily bleeding; due to the underlying effect of these species of Bartonella causing abnormal vascular endothelial cell proliferation and neovascularization [61, 62]. Studies have shown multiple mechanisms for this characteristic of Bartonella species causing endothelial and vascular

proliferation. The reported mechanisms of pathogenesis include vascular endothelial growth factor

(VEGF) induced capillary proliferation and mechanisms that inhibit apoptosis of vascular endothelial cells through caspase inactivation and DNA fragmentation [63-65]. The skin lesions may be superficial or deep into the subdermal structures, sometimes even involving the bones.

Regional lymphadenopathy, the involvement of mucous membranes of the mouth, conjunctivae and the gastrointestinal tract including the perianal area has also been described. It is also known that visceral involvement with the liver, spleen, lymph nodes and bone marrow can occur, with other reports documenting isolated visceral role in the absence of skin lesions [66, 67].

1.3.4 Bacteremia.

Invasion of erythrocytes and continued intraerythrocytic presence in their respective reservoir hosts (sometimes referred to as intraerythrocytic bacteremia) is an indication of

inoculation of a reservoir host by Bartonella species [68, 69]. Bartonella species use several

molecular mechanisms for erythrocyte invasion and immune response evasion that would

otherwise cause symptoms in bacteremic hosts [70-73]. Such asymptomatic bacteremia was

reported in various non-mammalian and mammalian hosts including humans [74-77].

Asymptomatic bacteremia can persist from a few weeks to several months in studies conducted in

healthy human volunteers [78]. In humans, a proportion of the erythrocytes, usually no more than

1% is infected by B.quintana, and such low-level bacteremia may persist with little or no

symptoms for months to years [69, 79, 80]. In a B.tribocorum infected rat model, B-cell deficient

rats had more prolonged bacteremia when compared with immunocompetent rats [80, 81]. It is

5 likely that low-level bacteremia may herald the pathogenesis and the clinical development of endocarditis in humans; however, this has not been confirmed [82].

1.3.5 Treatment considerations for Bartonella endocarditis.

Bartonella species can evade the host immune system and resist antimicrobial agents as a result of their pathogenic and virulence mechanisms, particularly their ability to persist in a primary niche [69, 83-86]. Earlier studies have shown that numerous Bartonella species are commonly susceptible to several classes of antibiotics, including beta-lactams, macrolides, and aminoglycosides [86-88]. Despite being susceptible at low minimal inhibitory concentrations

(MIC), subsequent studies and clinical experience have shown that treatment failures for

Bartonella infections are a significant problem [89-91]. Some of the antimicrobial classes tested against Bartonella show only bacteriostatic properties except for aminoglycosides such as gentamycin. In cases of Bartonella endocarditis, the issue of host defense evasion, potential biofilm formation, and antimicrobial resistance is even more crucial as cultures are likely to be negative and diagnosis delayed. Based on experience from the past two decades in the treatment of Bartonella endocarditis, multiple reports and recommendations advocate the use of at least two antibiotics, one of them being an aminoglycoside [92-94]. The recommended duration of therapy is generally for a minimum of 4 weeks in native valve disease and 6 weeks in prosthetic valve endocarditis. Aminoglycosides are recommended at least for the first two weeks of therapy. The duration of aminoglycoside use and the total duration of therapy correlates with clinical outcomes with longer duration conferring mortality benefit [95, 96]. However, in cases where antibiotics fail, surgical resection of the infected heart valve may be required [97].

1.4 B. henselae pathogenesis

The Toll-like receptors (TLR) are a type of pattern recognition receptor (PRRs) which are

used by the innate immune system to detect pathogens. Bartonella can evade immune system recognition because of the reduced endotoxin activity of the lipopolysaccharide in the outer membrane of the bacterium. After inoculation via an intradermal or intravenous route, Bartonella evades the host innate immune system for the primary niche, hypothesized to be a biofilm community and grows within the biofilm [98]. A biofilm is defined as a community of bacteria irreversibly adherent to a surface and embedded in an exopolysaccharide matrix. Biofilms form in five steps, initial adhesion, irreversibly attachment, maturation steps 1 and 2, and dispersion [99].

Growth in a biofilm enables B. henselae to maintain a presence in the flea feces and in the reservoir host (cat) and to evade the host immune system during human infection. Dispersion is a crucial stage as cells are released from the community to colonize new surfaces and form a new community. During a relapse, cells are dispersed from the biofilm to re-infect the blood causing bacteremic relapse [69].

1.5 B. henselae virulence factors

Most Bartonella species possess a VirB/VirD4 T4SS and surface appendages initially described as type IV pili comprised of TAA with the exception of a few which possess one or the other [31, 68, 100, 101]. TAAs are outer membrane proteins found on many other Gram-negative bacteria and shown to be involved in bacterial agglutination as well as facilitating adhesion to extracellular matrix components and host cells. Bartonella species and most Gram-negative bacteria possess TAAs [102-105]. B. henselae possess two dominant virulence genes which facilitate infection: the TAA called Bartonella adhesin A (BadA) and the VirB T4SS [106, 107],

as well as a range of other genes like the T4SS Trw [108], filamentous hemagglutinins [109, 110]

and outer membrane proteins [5].

1.5.1 VirB/VirD4 Type IV secretion system

T4SSs are transporters found in Gram-negative bacteria and are used to translocate virulence factors into host cells [111]. T4SSs are present and also characterized in other Gram- negative bacteria including Helicobacter pylori, Agrobacterium tumefaciens, and Legionella

pneumophila [69, 112-114]. The VirB/VirD4 T4SS in B. henselae is the most characterized T4SS

among the genus as reviewed in Harms et al. 2012 [68]. Schulein and Dehio (2002) [114], reported

that the VirB/VirD4 is used to colonize the vascular endothelium and promotes a VirB/VirD4- dependent inhibition of apoptosis [114]. It is also used to translocate the Bartonella effector proteins (Beps A-G) into the host cell resulting in an invasome mediated uptake, activation of a proangiogenic phenotype and inhibition of apoptosis [115].

1.5.2 Bartonella adhesin A (BadA).

TAAs have been studied extensively in other Gram-negative bacteria like the YadA of

Yersinia enterocolitica, NadA of Neisseria meningitidis, Hia and Hsf of Haemophilus influenzae

[103, 116-119]. BadA is a TAA protein found in B. henselae and is the longest known protein in the TAA family at 328 kDa per monomer and almost 1 million Daltons for the trimer. It forms filaments reported to be as long as 240nm on the surface of B. henselae [102, 120]. As all TAAs,

BadA is characteristically made up of a head domain, a stalk region, and a C-terminal membrane anchor. The BadA head is responsible for auto-agglutination and adherence to the host extracellular matrix (ECM) and the stalk region is required to bind fibronectin [121]. The C- terminal membrane anchors the whole protein to the bacterial outer membrane, oligomerizes the stalk domain and aids movement of the polypeptide chain through the cell membrane [105, 122].

The length and molecular size of the TAA also vary between species and this variation is thought to be related to the size of the neck/stalk region [100, 123, 124]. Trimerization is required to maintain the stability and adhesive property of the protein [118]. BadA is known to aid adherence of bacterial cell to host cell and extracellular matrix protein including fibronectin [125]. BadA is required for secretion of vascular endothelial growth factor (VEGF) from the host cell [125].

Regulation of badA is linked to the BatR/S two-component system, the general stress response system and a family of nine unannotated and highly transcribed RNAs designated as Bartonella regulatory transcript (Brt1-9) found upstream of a transcriptional regulator [110, 126, 127]. In

Quebatte et al. (2009) [110], a ΔbatR mutant showed differential downregulation of badA as time increases post-human endothelial cells infection but overall upregulation of badA over 48 hours when compared with the wildtype. Tu et al. (2016) also showed that the Brt1 RNA and the Trp1 regulatory protein affected badA transcription and biofilm formation negatively and positively respectively. An antisense knockdown of brt1 shows upregulation of badA transcription and biofilm formation and overexpression of the trp1 gene also upregulated badA biofilm dependent formation [128].

1.5.3 T4SS Trw.

T4SS Trw is a second surface localized T4SS acquired during evolution of bartonellae and discovered in B. henselae by Suebert et al. (2003) [108]. It is present in 13 species that adapt to diverse mammalian reservoir hosts and absent from human-specific B. bacilliformis, cat-specific

Bartonella clarridgeiae and the species of the ruminant-specific sub-branch, which all diverted early during the evolution of the bartonellae [129]. The presence and function of flagella and the

Trw T4SS were hypothesized to be mutually exclusive [111]. It is required for intraerythrocytic adhesion, invasion, and survival but dispensable for endothelial cell infection [130, 131].

1.5.4. Other outer membrane proteins

Filamentous hemagglutinin (Fha) is present in eight copies of varying lengths in B.

henselae and is known to mediate adhesion in Gram-negative bacteria [132, 133]. Fhas are

reported to be controlled by BatR, the sensor partner of the two-component system, BatR/S [110].

In addition to BadA and Fhas expressed on the outer membrane of B. henselae, outer membrane

proteins (OMPs) are also found in Bartonella. OMP43, a 43-kDa protein is a major porin protein

reported to be involved in proliferation and binding endothelial cells [134, 135]. In B.quintana, the

variably expressed outer membrane proteins (VOMPS) are reported to be involved in

autoaggregation and collagen binding [136].

1.6 Biofilm Regulation

The ability to adhere and auto-aggregate is an essential aspect of biofilm formation as it enables the bacteria to form microcolonies and ultimately form a biofilm [5]. A biofilm is a community of adherent cells protected by an extra polymeric substance (EPS) and able to survive transient exposure to antimicrobial substances. The ability to survive and tolerate this exposure to antimicrobials relates to bacterial persistence and the relapse of chronic infection after treatment.

Bacteria cells are seeded from a biofilm during the dissemination/dispersion step to form a new biofilm. As previously mentioned, the first step of biofilm formation is surface adhesion and agglutination facilitated by outer membrane proteins. Once bacteria agglutinate, they communicate effectively using chemical signaling molecules. This signaling process called quorum sensing (QS) enables bacteria to track changes in the environment triggering changes in gene expression [137].

Biofilms are typically stable, involved in chronic infection and relapse, and display increased resistance to antibiotics and host immune response. Microbial biofilms are implicated in a variety

of diseases: dental plaque, chronic lung infection in cystic fibrosis patients as well as heart valve vegetations in patients with infective endocarditis as reviewed in Bjarnsholt (2013) [138].

In many Gram-negative bacteria, TAAs are linked to adhesion, the first step in biofilm

formation [139]. B. henselae biofilm growth was first reported by Kyme et al [140], which

describes the aggregates as a phase variation, a property now shown to be attributed to the presence

or absence of BadA on the surface [100]. Growth characteristics were thought to be associated

with variation in genotype and OMP expression. BadA accumulates as a dense surface layer

hairlike structures ( ̴240nm) and its absence prevents auto-agglutination [125]. The role of TAAs

in biofilm formation has been proposed or experimentally demonstrated for a wide range of Gram-

negative bacteria including Acinetobacter baumanii [141], Burkholderia species [142, 143],

Escherichia coli UPEC and EHEC [139, 144] and Salmonella enterica [145]. Previously published

reports by Okaro et al. (2017, 2019) [5, 98] show that BadA is linked to biofilm formation. An in- frame deletion mutant of the badA gene displayed a severely, reduced ability to form biofilms.

Scanning electron micrographs of B. henselae biofilm also show aggregates similar to the

vegetative mass observed in patients with BCNE. It was speculated that biofilms bolster

transmission of the bacteria from the arthropod vector [98]. B. henselae has been shown to grow,

replicate and persist in the cat flea gut, and is excreted in the cat flea feces [35, 47, 146]. A biofilm

EPS is typically made up of eDNA, proteins, polysaccharides and other excreted components

[147].

1.7 Non-coding RNAs

The regulatory circuit for biofilms is initiated once bacteria adhere to a surface. RNAs have

been implicated in the regulation of genes which control biofilm formation in B. henselae and

other Gram-negative bacteria [128, 148, 149]. Since the discovery of the first signaling RNA [150],

RNAs have been shown to work through a variety of approaches like regulating transcription,

translation, DNA silencing, and mRNA stability [151]. Genome-wide studies show that a high percentage of RNA molecules within the bacterial genome are transcribed as non-coding small

RNA (ncRNA) [152].

ncRNAs are small-sized, post-transcriptional regulatory molecules that regulate bacterial physiology like metabolism, communication and biofilm formation [153-155]. ncRNA may function as trans-acting antisense transcript (asRNA) or as cis-acting RNA. ncRNA expression changes in response to environmental conditions such as pH, nutrient availability, antimicrobial peptides and competition with other microbes as reviewed in Ortega et al. (2014) [156].

Bartonella regulatory transcripts 1-9 (Brts 1-9), previously described by Tu et al. (2016)

[128], are a family of highly expressed, multi-copy sRNAs specific to Bartonella species (Fig

1.1c). The Brts are found in a location of the B. henselae genome known to harbor a novel genomic region enriched in repeats and highly expressed genes with proteins of unknown function (Fig

1.2a) [157]. This high repetitive characteristic is indicative of a horizontally acquired gene which may have been duplicated. The Brt RNAs are also highly conserved (Fig 1.1b, 75-100%) and the number of RNAs vary between species [128]. Nucleotide alignment of Brt 1-9 show that nucleotide conservation is maximal at the 3' end of the gene (Fig 1.1b, red box), a location predicted to fold into a highly stable stem and loop structure requiring very low Gibbs energy (Fig

1.1a, red circle). Tu et al. [128] also showed that knocking down the highly expressed brt1 positively affected the expression of badA/BadA involved in biofilm formation. This evidence suggests that a native Brt1, a ncRNA, negatively regulates badA expression.

Fig 1.1. Characteristics of Brt1 and Trp1. A. Predicted secondary structure of Brt1. B. Conservation areas of the Brts (red box shows 3' region). C. RNA-seq analysis showing the reads and direction of the nine Brts. Green indicates transcription (left to right

5’ to 3’) in the direction depicted, red indicates transcription in the reverse direction (right to left as indicated) and yellow indicates

regions with ambiguity (Brt4 and Brt9 have identical sequences). D. The conservation of the Trps. The blue box depicts

conservation around the HTH domain.

1.8 Riboswitches and RNA thermometers.

Riboswitches and RNA thermometers (RNAT) are regulatory cis-encoded sRNAs which fold into intricate structures in response to metabolites or environmental changes (pH, temperature) to modify the expression of a downstream gene [158, 159]. These intricate structures are typically found in the 5' untranslated region (UTR) of the gene which is being regulated (reviewed in Jameel

M. Abduljalil (2018) [159]). Riboswitches can sense metabolites like amino acids, carbohydrates,

13 co-enzymes, hemoglobin, metal ions like magnesium, environmental conditions like pH while

RNATs monitor temperature changes only [160, 161].

Downstream and in the same direction as each brt gene described above, is a gene which codes for a DNA transcription factor called Transcriptional regulatory protein, Trp1-9 (Fig 1.1c).

Like the brts, the trps are also present in multiple copies and conserved at the amino acid level

(65-100%), especially around the characteristic helix-turn-helix xenobiotic response element domain (Fig 1.1d, blue box). Under the conditions used for the RNA-seq in Tu et al.,[128], the trps were minimally transcribed (Fig 1.1c) . Each brt and the accompanying trp are arranged side- by-side with less than 15 nucleotides separating the tentative end of a brt from the open reading frame (ORF) of the corresponding trp gene (Fig 1.2b), thus reducing the possibility of a native trp promoter. In view of the high transcript rate of the brt, the low transcript rate of the downstream trp gene (Fig 1.1c), the close brt/trp gene arrangement without a discernible trp promoter (Fig

1.2b), and the stem and loop structure on the 3' end of the Brt1 RNA (Fig 1.1a- red circle), it is feasible that the 3' loop of Brt1 RNA affects the downstream trp1 transcription by either acting as a terminating or anti - terminating loop.Since riboswitches or RNATs respond to physical or metabolic changes in the environment, the primary differences between the cat flea arthropod vector and the mammalian hosts may signal the activation or repression of the trp gene. Some of the differences between the arthropod flea and mammalian hosts are listed in table 1.1. Both riboswitches and RNATs have been shown to regulate virulence genes including genes that initiate biofilm formation.

Fig 1.2. Diagram showing the relationship between Brt1 and Trp1. A. Location of the nine Brt RNAs on the B. henselae Houston-

1 genome. The red oval indicates the region of the genome containing all nine Brt RNA family members. Genome map reproduced from Omasits et al., 2013 with permission. The sequence of Brt1 and Trp1. Dotted lines represent Brt1 putative promoter, followed by the Brt1 RNA (Red). Red, underlined sequence represents the 3' stem-loop found in Brt1. Black, italicized fonts show the tentative Trp1 ribosome binding site within the intergenic region separating Brt1 and Trp1. The RBS is closely followed by Trp1 start codon (GTG) and coding region (Blue). B. henselae sequence adapted from NCBI Reference Sequence:

NC_005956.1:1438600-1439626.

Table 1.1: Significant differences between the arthropod vector mammalian host of B. henselae.

Factor Arthropod Vector Mammalian system (cat flea) (cat and human) Temperature ̴13-35°C [162, 163] 36.5-37.5°C [164]

Heme 0 mM to 5 mM [165] 0.5 μM (Free heme in the blood) [166]

pH 6.0-6.8 [167] 7.34-7.45 [168]

1.9 Transcription factors

Proteins play a significant role in gene expression either by initiating transcription, translation or modifying expression at the post-translational level. Transcription factors (TF) are

proteins which bind specific DNA sequences to regulate expression of a downstream gene by

either binding the promoter region to successfully initiate gene transcription or as repressor

proteins that inhibit transcription. TFs have been shown to affect biofilm formation by either

directly affecting transcription of genes related to biofilm or indirectly through a secondary

regulatory cascade [169].

Further analysis of the TF gene (trp1) found downstream of brt1 shows that the Trps have

a helix-turn-helix (HTH) domain and are xenobiotic response elements (XRE). XREs are the

second most frequently occurring transcription factors in bacteria and are known to control

metabolic and virulence genes [170, 171]. HTHs are functional motifs capable of binding the major

groove of DNA. Typically the second helix recognizes and binds the sequence while the first helix

stabilizes the protein DNA interaction binding as a dimer [172, 173]. HTH-XRE proteins typically

make specific DNA contacts through their N-terminal domains, but their DNA binding affinity is

modulated by their C-terminal domains [174]. MrpJ, a helix-turn-helix (HTH) xenobiotic response

element (XRE) family transcriptional regulator found in P.mirabilis serves as a switch between

motility and adhesion [175]. BswR, an XRE family transcriptional regulator is involved in the

regulation of P. aeruginosa swarming motility and biofilm formation [176]. Tu et al. [128] showed

that trp1 is a positive regulator of badA. An overexpressing trp1 construct demonstrated increased

badA expression and biofilm formation.

1.10 Summary

Gene expression in bacteria is continuously regulated for overall fitness in any given environment. sRNAs, TF and other regulators may activate or repress a specific gene. Some regulators respond to a specific factor like temperature, while others respond to a plethora of conditions, metabolites, ligands pH, ions, etc. Jointly, both TF and sRNAs regulate genes with a

variety of cellular functions, communicating with genes through a multifaceted network. We

hypothesize that one such network is the brt-trp-badA network which is used to control biofilm formation in B. henselae. As previously explained, B. henselae has to respond to quick changes in the environment (arthropod-mammal) to successfully colonize and live within a host system. It has also been shown that sRNAs and TFs regulate virulence genes including genes responsible for adhesion and biofilm formation by other bacteria. We propose that this novel arrangement between an RNA and a TF affects badA transcription thereby controlling adhesion and biofilm formation in B. henselae. This thesis focuses on the function of the 3' loop of Brt1 as a potential terminator/ anti-terminator to control the transcription of the TF gene downstream-trp1- which is shown to be a positive regulator of badA. Finally, we studied the relationship between Trp1 protein and transcription of the badA gene, a significant virulence gene in B. henselae.

CHAPTER 2: B. henselae requires BadA for optimum biofilm formation

Note to Readers: Portions of this chapter are published in “Okaro, U., et al., The trimeric

autotransporter adhesin BadA is required for in vitro biofilm formation by Bartonella henselae.

npj Biofilms and Microbiomes, 2019. 5(1)” and included with permission from the publisher.

2.1. Background

B. henselae is transmitted to the cat through the cat flea. It is transmitted to humans by infected flea feces through the scratch of a cat. BadA is a TAA implicated as one of the major proteins responsible for attachment and aggregation, the first step of biofilm formation in B. henselae. The expression level of badA have been shown to correlate with biofilm formation [120,

128]. Biofilms are implicated in both parts of the B. henselae life cycle. B. henselae can colonize, replicate and persist in the cat flea. It is excreted in flea feces and detected in both fleas and feces at least 12 days post-infection [177]. Biofilm formation increases not only the persistence of bacteria but also the transmission efficiency from the flea to the vertebrate host. Longer viability, therefore, increases the likelihood of human transmission. Secondly, B. henselae biofilms are an essential component of the vegetative growths observed in the heart valves found in BCNE patients as reviewed in Bjarnsholt (2013) [138]. B. henselae can form a stable biofilm in patients with infective endocarditis and surgery is often required to eradicate the biofilm [138].

Biofilms grow as three-dimensional structures in their natural environment. We used a 3- dimensional nanofibrous scaffold, previously used to grow tumor cells, to model biofilm growth

accurately in vitro [178]. As a more reproducible and information-rich alternative to the traditional crystal violet assay, the xCELLigence real-time cell analysis (RTCA) system was used for monitoring biofilm dynamics continuously throughout the entire assay. In this system, adherence of bacteria to gold microelectrodes embedded in the bottom surface of xCELLigence microplates

(e-plates) impedes the flow of current between electrodes. This impedance signal reported as the cell index (CI), provides a composite assessment of cell number, cell size, and cell substrate attachment [179]. As bacteria grow, the close interaction between adhering cells begins to impede the flow of current. Importantly, neither the gold electrodes nor the weak electric field perturbs bacterial adhesion or growth [180].

In this chapter, we investigate the role of badA expression in biofilm formation and the components and conditions favorable for biofilm formation.

2.2. Materials and Methods

2.2.1. Bacterial Strains.

E.coli strain DH5α (Invitrogen 18258-012), B. henselae Houston-1 [12], B. henselae

Houston-1 ΔbadA [181] and B. henselae Houston-1 ΔbadA/pNS2PTrcbadA [98] were all used for

this study (Table 2.1). E.coli DH5α cells were grown at 37°C on either LB agar or in liquid LB broth. B. henselae was grown on heart infusion agar supplemented with 1% bovine hemoglobin or

liquid Schneiders media (Sigma Aldrich, S9895) supplemented with 10% fetal bovine serum for

three days as described by Riess et al. [182]. Growth conditions were kept at 5% CO2 at 37°. Lima et al. (2014) described B. henselae Houston-1 ΔbadA, a non-polar in-frame deletion mutant of badA [181]. B. henselae Houston-1 ΔbadA/pNS2PTrcbadA, a partial complement of badA

consisting of the N-terminal head, a truncated neck region with only one neck domain, and the C-

19 terminal membrane anchor was described by Okaro et al. (Fig 2.2b) [98]. B. henselae Houston-1

ΔbadA/pNS2PTrcbadA was grown on agar supplemented with kanamycin (50µg/ml).

Table 2.1 Bacterial strains and primers used for aim 2.

Strain/Primer sequence Source/Sequence

B. henselae Houston -1 Regnery et al. 1992 [14]

B. henselae Houston-1 ΔbadA Lima et al. 2014 [181]

B. henselae Houston-1 ΔbadA/pNS2PTrcbadA Okaro et al. 2019 [98]

rpld F ATGCGCATGACTACGAA

rpld R CAGGGGCGAAGATGTTTAAG

badA screen 1F ACGCATGTAGAGAATGGTGA

badA screen 1R CTTCGCATCTTCAAGCACTATCT

2.2.2. B. henselae forms a biofilm

Bacteria were inoculated on either a nitrocellulose membrane or the 3-dimensional nanofibrous scaffold to visualize and quantitate the amount of biofilm formed by B. henselae. The

3D nanofibrous scaffold designated as 3P was produced by electrospinning polylactic-co-glycolic acid (PLGA), polylactic acid (PLA) and mono - methoxypolyethylene glycol (mPEG) block copolymer [178]. The scaffold/membrane was immersed in 100% ethanol for 15 secs, washed with

PBS, transferred to 96 well plates containing 50μl PBS and placed under UV wavelength light for

45 minutes to ensure sterility. The well was washed with 100µl of media, inoculated with 150µl

6 of the bacterial cells (10 ) and grown at 37°C, 5% CO2 for 8 hours. 50µl of media was added to

each well and grown for 24 to 72 hours total. The scaffold/membrane was fixed overnight in a

mixture of 2% paraformaldehyde and 2% glutaraldehyde in 0.2M sodium cacodylate buffer, pH

7.2 with or without 0.15% alcian blue. Samples were washed 2x in 0.2M sodium cacodylate and

post-fixed for 30mins in 1% OsO4. The dehydration steps occurred using ascending alcohol concentration (35% - 5min, 50%-5mins, 75% - 5mins, 90% - 5mins, 100% - 10mins 2x, 50% hexamethyldisilazane (HMDS) + 50% of absolute EtOH- 10mins, Pure HMDS – 10mins).

Samples were air-dried overnight, mounted on adhesive carbon film and coated for 30secs with

Au/Pd (60; 40) at 16.40g/cm and 25mA. Joel JSM6490LV scanning electron microscope operated at 4Kv was used to image the scaffolds and secondary images collected as TIFF/JPEG.

2.2.3. BadA expression in B. henselae constructs.

To confirm that our partial complement expresses BadA on the outer membrane, all three strains were used for immunofluorescence microscopy. Briefly, bacterial cells were grown on a 6 well chamber plate coated with poly-l-lysine overnight to adhere. The wells were washed in PBS and fixed with 4% paraformaldehyde for 20mins. Fixative was washed off, samples were blocked with 5% nonfat milk for 1 hour and washed again. The slide was incubated for 1 hour with 40µl of anti-BadA rabbit antibody (1:200) and 40 µl of a 1:10,000 dilution of fluorophore-conjugated

secondary (Alexa 488 anti-rabbit) subsequently. The slide was washed, mounted and air-dried

overnight. Images were exported as TIFF.

2.2.4. badA expression within a biofilm.

As the expression of badA diminishes after multiple passages, we routinely used B.

henselae Houston-1 from a frozen stock stored at -80°C and the bacteria were discarded after four

passages to prevent loss of adhesin expression [100]. To compare the expression of badA in a

biofilm or in growth as planktonic cells, B. henselae was cultured in Schneiders liquid media at

37°C, 5% CO2 for three days on a six-well polystyrene plate (Corning 3506) with or without mild

shaking at 100rpm. The supernatant was carefully aspirated to prevent biofilm disruption. The

biofilm was gently washed twice with PBS, and the RNA is extracted by directly adding trizol

(Life Technologies, 15-596-026) to the plate. The planktonic growth was pelleted and resuspended

in trizol. 10µg of the resulting RNA was treated with DNase (Thermo Fisher Scientific, AM1907),

and 1µg reverse transcribed to cDNA using the iScript cDNA synthesis kit (BioRad, 1708891).

RT‐qPCR was performed in a total volume of 25 μl which consist of 12.5 μl of the 2X Maxima

SYBR green/fluorescein qPCR kit (Thermo Fisher Scientific, K0241), 300 nmol of the forward

and reverse badA primers (screen 1F and 1R) (Table 2.2), and 2μl cDNA. The 50S ribosomal

protein L4 (rplD), was used as the reference gene for normalization. The reaction conditions were

as follows: a single cycle of 95°C for 3 mins, 40 cycles of 95°C for 10 s, and 60°C for 30 s,

followed by 95°C for 45 s and 55°C for 1 min. Melt curve analysis was used to confirm that primer

dimers were not generated. The comparative CT method was used to analyze data [183]. RT-qPCR

experiments were conducted as independent replicates and mean values compared between groups

using the student’s t-tests. SigmaPlot software (Systat Software, San Jose, CA) was used for

statistical analysis. Differences between groups were statistically different for a P-value <0.05.

2.2.5. The role of BadA in biofilm formation.

The xCELLigence® RTCA (ACEA Bioscience Inc.) was used to measure and monitor cell

adherence. This system measures cell adherence by recording the electrical impedance signal from

adherent cells on the bottom of specialized 96 well e-plates (ACEA, Biosciences Inc.) with gold

micro-electrodes. Maximum CI is achieved when the surface of the microelectrode is covered by cells giving a constant CI. This saturated CI remains constant during biofilm formation but decreases during detachment [179]. 106 cells were seeded onto the plates, and real-time

measurement occurred every 15 mins over 48-72 hours. The growth conditions were kept at 37°C

and 5% CO2. Post incubation, wells were washed with sterile water to remove planktonic cells and stained with 0.1% crystal violet solution for 15mins to determine the amount of biomass in each

well. The wells were decolorized using 30% acetic acid, and the absorbance at 550nm was

measured. For each RTCA experiment, individual strains were seeded in 6 different wells and

averaged. The mean values were compared between groups using the student’s t-test.

2.2.6. The viability of cells within a biofilm

CLSM was used to confirm the components of a B. henselae biofilm. The 96 well e-plates used for real-time monitoring were gently washed to prevent biofilm disruption. Molecular probes:

Film tracer biofilm viability kit (Invitrogen, L10316) containing STYO9 and propidium iodide

(PI) was used to stain wells according to the manufacturer’s protocols. Wells were stained for 30

mins, washed twice with sterile water and imaged immediately after the final wash. Samples were

examined using an Olympus Fluoview FV1000 microscope and image exported as TIFF.

2.2.7. Characterizing the components of a B. henselae Houston-1 biofilm.

The 96 well e-plates used for real-time monitoring were gently washed to prevent biofilm

disruption. Molecular probes capable of binding DNA (Hoechst stain (Thermofisher, 33342)),

proteins (Sypro ruby biofilm matrix stain (Invitrogen, F10318)), and polysaccharides (Wheat germ

agglutinin fluorescein conjugate (Invitrogen, W834)) were used to stain wells according to

manufacturer’s protocols. Wells were stained for 30 mins, washed with sterile water and imaged

as described above.

2.2.8. Environmental factors required for optimal biofilm formation

To compare the rate of biofilm formation and gene expression under conditions that mirror

the arthropod vector and mammalian hosts in vitro, bacteria cells were grown under different

environmental conditions: pH (6.6/7.2), temperature (27°C vs. 37°C), and treatments (proteinase

K, 10µg/ml and 1U/µl DNase1 at time of inoculation or 24 hours after inoculation). The experiment was carried out using the same methods outlined in 2.2.4.

2.3 Results

2.3.1. B. henselae forms a biofilm.

To examine biofilm formation by B. henselae Houston-1, 10^6 bacteria in a 150µl volume

were inoculated into a 96 well polystyrene plate with a nitrocellulose membrane on the bottom.

The membrane was removed after 8 to 72 hours incubation and processed to track biofilm

formation. Scanning electron microscopy (SEM) was used to observe biofilms formed by B. henselae Houston-1. Individual rods were observed to aggregate and form micro-colonies within

8 hours of inoculation (Fig 2.1a). These micro-colonies aggregate to form extensive, defined colonies observed 24 hours post-inoculation (Fig 2.1b). Rapid growth and biofilm are observed 48 hours after inoculation (Fig 2.1c). At 72 hours of growth, EPS surrounds the mature biofilm (Fig

2.1d).

Fig 2.1. Scanning electron micrograph of B. henselae growth and biofilm formation from 8 hours (A), through 72 hours (D). A.

Cells attach within 8 hours, before aggregation and production of micro-colonies (B -24 hours). C. The aggregates undergo

tremendous growth in 48 hours to produce biofilm within 72 hours (D).

2.3.2. badA is highly expressed within a biofilm

The transcription rate of badA in B. henselae grown in a biofilm was compared to

planktonic bacteria grown with mild shaking at 100 rpm to prevent adhesion. Using primers specific to badA (Table 2.1), a significant difference in badA expression was observed between the two growth methods (Fig 2.2a). badA was preferentially expressed in a biofilm community in comparison to planktonic growth. Fig 2.2b shows the functional parts used for B. henselae

Houston-1 ΔbadA and B. henselae Houston-1 ΔbadA/pNS2PTrcbadA constructs. Fig 2.2c shows

that all three strains have no significant difference in growth rates confirming that the presence or

absence of badA does not impact growth. In fig. 2.2d, badA transcription in the wild type and

partial complement are observed. badA is significantly upregulated in the partial complement

because of the strength of the Trc promoter which drives badA expression. Finally, BadA is

expressed in both B. henselae Houston-1 and B. henselae Houston-1 ΔbadA/pNS2PTrcbadA.

Immunofluorescent microscopy using anti-BadA antibody and fluorophore-conjugated secondary antibody on intact bacteria cells shows BadA expression in the BadA expressing strain and we observe no fluorescence in the B. henselae Houston-1 ΔbadA strain (Fig 2.2e). This signifies that

badA/BadA is needed for adhesion of cells: the onset of biofilm formation.

Fig 2.2. Growth rates and badA transcript levels in B. henselae constructs. A. badA transcript level from bacteria grown as a biofilm

or planktonic (P = 0.004). B. The partially complemented clone consist of the head region, membrane anchor and a truncated neck

region. C. Growth curve experiment between strains. D. RT-qPCR using primers specific to the badA head confirming a significant

difference in badA expression between all strains (P≤0.024). E. B. henselae Houston-1∆badA/pNS2P badA expresses BadA on Trc the surface. Confocal microscopy with anti-BadA antibody and fluorophore-conjugated secondary antibody. RT-qPCR bars represent the average of 6 replicates and error bars represent the SEM.

2.3.3. BadA plays a significant role in biofilm formation.

To assess B. henselae biofilm formation in real-time, the xCelligence RTCA system was

employed. All three strains (B. henselae Houston-1, B. henselae Houston-1 ∆badA [181] and B.

henselae Houston-1 ∆badA/pNS2PTrcbadA) adhered to the plate to varying degrees and at differing times (Fig 2.3a-enlarged image). In fig 2.3a, the Houston-1 parental strain adhered more efficiently in the early stages of growth (<20 hours), but the B. henselae Houston-1 ∆badA

displayed a higher cell index (CI) that was statistically different from the B. henselae Houston-1

(p <0.001, students t-test) at the end of the experiment (80 hours). There was no significant

difference between B. henselae Houston-1 ∆badA and B. henselae Houston-1

∆badA/pNS2PTrcbadA (P = 0.163, students t-test) at that same time point prompting the use of

microscopy and traditional endpoint biofilm assays to evaluate the B. henselae Houston-1 ∆badA

cells. Crystal violet (CV) stain on the e-plates post-real-time monitoring, shows that B. henselae

Houston-1 biofilm is statistically different from B. henselae Houston-1 ∆badA (P = 0.004) but not

significantly different from B. henselae Houston-1 ∆badA/pNS2PTrcbadA (P = 0.6) (Fig 2.3b).

Partial complementation of badA in B. henselae Houston-1∆badA/pNS2PTrcbadA using the strong

promoter (Ptrc), resulted in an intermediate level of biofilm formation (Fig. 2.3b). Despite the high

cell index recording from the RTCA system at 80 hours, using the standard CV biofilm method,

B. henselae Houston-1 ∆badA cells did not form as much biofilm compared to the BadA expressing strains. To determine how much of each B. henselae cells did not adhere to the e-plate, the cell density of the supernatant (OD600) was measured (Fig 2.3c). The B. henselae Houston-1 ∆badA

supernatant was the least adherent, indicating that BadA is vital for adherence of bacteria to a

surface. Both the real-time analytical experiment and the CV staining data confirm that BadA is

essential for optimal biofilm formation in B. henselae.

Fig 2.3. B. henselae biofilm formation using a 96 well e-plate. A. Real-time cell index of strains at 37°C and 5% CO2, the magnified area shows 0-20 hours. B. Biomass from B. henselae cells from plate used for graph 2a stained with 0.1% CV. Students t-test: *( P

= 0.004), **(P = 0.6). C. The density of supernatant aspirated from experiment 3a before staining with CV. Bars represent the average of 6 replicates and error bars represent SEM. D. CLSM image of B. henselae biofilm populations using the STYO9/PI live/dead staining after 3 days of incubation. White arrows depict cells with a partially disrupted membrane (yellow cells), white circle depicts the concentration of dead population (red cells).

2.3.4. badA expressing strains harbor more viable cells within a biofilm.

To determine the viability of cells in the B. henselae biofilm, we utilized a two

fluorescence cell viability kit containing SYTO9 green-fluorescent nucleic acid stain and the red-

fluorescent nucleic acid stain, propidium iodide (PI). Confocal laser scanning microscope

(CLSM) was used to investigate the cells stained with SYTO9 and PI. SYTO9 stains nucleic acid

hence it stains both live and dead cells but in the presence of PI, which penetrates disrupted

membranes, the SYTO9 is displaced and the cell fluoresces yellow for partial SYTO9

displacement or red for complete displacement. Both B. henselae Houston-1 and B. henselae

Houston-1 ∆badA/pNS2pTrcbadA cells growing in biofilms show more viable cells exhibiting green fluorescence and few dead cells with yellow fluorescence (Fig. 2.3d, white arrows). In contrast to the BadA expressing strains, the biofilm formed by B. henselae Houston-1∆badA shows a large population of dead cells (Fig. 2.3d, white circle).

2.3.5. B. henselae biofilm contains polysaccharides, protein, and eDNA.

B. henselae Houston-1 was used to characterize the biochemical composition of the biofilm. Scanning electron microscopy (SEM) was employed to examine images of the biofilm produced by B. henselae grown on a 3-dimensional nanofibrous scaffold. In Fig 2.4 (top row, L-

R), SEM images show scaffold only, B. henselae Houston-1, B. henselae Houston-1 ∆badA and

B. henselae Houston-1 ∆badA/pNS2PTrcbadA cultured on the scaffold. As seen on the top row

images L-R, microscopic analysis showed that the wild-type, B. henselae Houston-1 exhibits

massive growth, adhesion, and aggregation in comparison to B. henselae Houston-1 ∆badA. The

B. henselae Houston-1 ∆badA/pNS2PTrcbadA displayed an intermediate level of growth and

aggregation. B. henselae Houston-1 ∆badA shows sparse adhesive and aggregative properties in

the reduced formation of micro-colonies. Using Alcian blue dye dissolved in glutaraldehyde and

paraformaldehyde, the bacterial EPS was preserved to provide a quantitative image of the bacterial biofilm (Fig 2.4, bottom row). The bottom images confirm that B. henselae Houston-1 produced

the most biofilm with a smooth layer covering a dense growth. Although the scaffold (white arrow)

is still visible, it is mostly covered biofilms. In contrast, B. henselae Houston-1 ∆badA

demonstrated a thin layer of biofilm (red arrow) with some bacterial cells bare and unprotected by

a matrix (yellow arrow). The bacteria still show minimal adhesion with reduced biofilm formation

despite a complete deletion of the badA gene B. henselae Houston-1 ∆badA. Finally, B. henselae

Houston-1 ∆badA/pNS2PTrcbadA exhibits an incompletely assembled biofilm showing that complementation with portions of the badA gene can induce limited biofilm formation.

Fig 2.4. Scanning electron micrographs (SEM) images of B. henselae biofilms. Biofilms established on a 3P scaffold after 72 hours

incubation at 37°C and 5% CO2. The top row shows bacteria growth preserved by the addition of fixatives (aldehydes only). Bottom

row: Biofilm produced by each strain: EPS is preserved by the addition of cationic dye, Alcian blue. White arrow depicts bare

scaffold and yellow arrow depicts single bacteria rods not covered by EPS. Red arrow depicts weak biofilm.

We investigated the presence of DNA, proteins, and polysaccharides in a B. henselae

biofilm using a confocal laser scanning microscope. An experiment was performed on the e-plates

used for the real-time cell adhesion monitoring. The plates were stained with dyes which are

capable of binding to polysaccharides (wheat germ agglutinin), protein (sypro ruby), and DNA

(Hoechst). Fig 2.5a shows dyes binding to the biofilm components confirming that a B. henselae

Houston-1 biofilm contains polysaccharides, proteins, and DNA.

To confirm the presence of both protein and eDNA in the biofilm, We measured the ability of the two enzymes, - proteinase K and DNase1-, to inhibit B. henselae biofilm formation. The

RTCA experiment was repeated and cells were treated with proteinase K or DNase1 during inoculation or 24 hours after inoculation. Real-time monitoring of B. henselae Houston-1 cells cultured and treated with 10ug/ml of proteinase K resulted in a significant decrease (40%-66%

inhibition) in biofilm formation depending on treatment time (Fig. 2.5b). Similarly, treatment with

DNase1 (1U/µl) resulted in a slight reduction in B. henselae biofilm (22%-43% inhibition).

Fig 2.5. Constituents of a B. henselae biofilm. A. CLSM images of B. henselae biofilm with WGA fluorescein (polysaccharides),

SYPRO Ruby (protein) and Hoechst (DNA). B. Effects of proteinase K and DNase on a B. henselae biofilm using RTCA show that enzymes added at time of inoculation caused significantly less adhesion and biofilm formation compared to a 24-hour old biofilm.

2.3.6. badA expression and biofilm formation are influenced by environmental conditions.

The response of B. henselae to changes in environmental temperature and pH was investigated. At a lower temperature that is consistent with the C. felis vector in the environment,

the cell index for B. henselae 48 hrs post-inoculation was 15% more than the cell index at 37°C

which correlates with the cat or human host temperature (Fig. 2.6a). We compared the viability of the cells within a biofilm in different temperature using the SYTO9/PI stain. A significant amount

of B. henselae cells are viable when grown at 37°C compared to cells grown at 27°C (Fig 2.6b).

Cells grown at 27°C displayed areas with red/yellow cells (white circle) indicative of disrupted

membranes and reduced viability. badA expression is also insignificantly upregulated at 37°C (P

≤ 0.022) supporting the hypothesis that at 37°C, B. henselae requires BadA to form a biofilm (Fig

2.6c).

Fig 2.6. The optimum temperature for a B. henselae biofilm. badA expression and biofilm formation in B. henselae Houston-1 is susceptible to growth conditions. A: cell adhesion and biofilm formation under different temperatures. B. CLSM image of B. henselae biofilm populations under different temperatures using the STYO9/PI live/dead staining C. badA transcript levels from

bacteria grown at different temperatures (students t-test, * = P ≤ 0.022). RT-qPCR bars represent the average of 6 replicates and

error bars represent the SEM.

The effect of pH on B. henselae biofilm was compared using growth media with a pH of

6.6 and 7.2. At both pHs, B. henselae cells form biofilms (Fig 2.7a). There was no statistical

difference between growth in media of varying pH. However, badA expression statistically differs

at a pH of 7.2 (Fig 2.7b, students t-test, P = 0.007). It is observed that the expression of badA gene

is optimal at the inoculation pH of 7.2, at day 1 and gradually decreases as a biofilm is formed (Fig

2.6c). Interestingly, the pH goes up during the initial stages of biofilm formation (24 hours) and then declines at 48 hours for the BadA expressing stains, B. henselae Houston-1, and B. henselae

Houston-1 ∆badA/pNS2PTrcbadA. However, B. henselae Houston-1 ∆badA maintains a near

slightly basic pH throughout the experiment (Fig. 2.7c).

Fig 2.7. Optimum pH of a B. henselae biofilm. A. Biofilm growth at varying pH. B. badA gene expression under different pH

(students t-test, * = P = 0.007). RT-qPCR bars represent the average of 6 replicates and error bars represent the SEM. C. pH of each strain at inoculation, 24, 48 and 72 hours endpoints. pH values are determined with pH strips (images) and a pH electrode

(actual value).

2.4 Conclusion

BadA is a TAA implicated as one of the major proteins responsible for attachment and

aggregation, the first step of biofilm formation by B. henselae [120]. To determine if B. henselae

forms a biofilm, the bacteria were cultured on 96 well polystyrene plates with cellulose membranes

at the bottom over 72 hours. After observing the presence of biofilms microscopically (Fig. 2.1),

we quantified the expression of badA in a biofilm and compared that to the expression level in

planktonic growth (Fig 2.2a). Using the three constructs described in table 2.1, we investigated the

differences in gene expression, cell viability, biofilm formation, and composition, and

characterized the conditions favorable for biofilm formation in B. henselae.

Given the size of BadA (328 kDa per monomer), numerous attempts to make a full - length

BadA complement proved unsuccessful, as others have also reported [184], so we opted to use a

partial complement bearing the functional domains of the full-length protein (Fig 2.2b). All three

strains showed comparable growth rates (Fig 2.2c), and it was not surprising that B. henselae

Houston-1 ΔbadA/pNS2PTrcbadA shows high levels of badA transcript compared to the wild-type

(Fig 2.2d). Fig. 2.2e shows surface localization of the truncated BadA; the levels of BadA protein

were not proportionally high as the corresponding mRNA observed in fig. 2.2d. Compared to the

native full - length BadA, the truncated protein may be more susceptible to protease degradation

or may be incompletely translocated to the outer membrane.

The microplate assay is one of the most commonly used methods for measuring in vitro

biofilms [185]. While it is inexpensive and straightforward, the data is sensitive to sedimentation

and loosely attached biofilms [186]. To avoid these issues, the xCelligence RTCA was used. High cell index levels of B. henselae Houston-1 and B. henselae Houston-1 ΔbadA/pNS2PTrcbadA were

expected; however, the B. henselae Houston-1 ∆badA cell index was unanticipated because it

lacked badA gene (Fig 2.3a). Growth rate experiments over 96 hours show that all three strains do

not significantly differ (Fig. 2.2c), hence the increased cell impedance from the B. henselae

Houston-1 ∆badA mutant was not due to increased growth rate and did not correlate with our

previous data showing that the mutant produced less biofilm in comparison to the wild-type [5].

At the end of the real-time monitoring, we determined the cell density (OD600) in the supernatant

of each strain. The OD600 recorded by B. henselae Houston-1 ∆badA was twice as high as B. henselae Houston-1 and 10% more than B. henselae Houston-1 ΔbadA/pNS2PTrcbadA (Fig 2.3c),

therefore, the real-time analysis output was either adherent cells not present in a mature biofilm or

sediment from planktonic or dead cells. A crystal violet stain was used on the e-plates post-real-

time monitoring to demonstrate that B. henselae Houston-1 ∆badA formed less biofilm, 13% in

comparison to B. henselae Houston-1, and 20% of the biofilm formed by B. henselae Houston-1

ΔbadA/pNS2PTrcbadA (Fig 2.3b). Despite the complete in-frame deletion of BadA, B. henselae

Houston-1 ∆badA still forms a biofilm that can be attributed to the presence of other outer membrane proteins present on the bacterial surface that function as adhesins. An example is the

outer membrane proteins like Omp43 which bind endothelial cells (EC) [135] and Pap31 which binds fibronectin [187].

Microscopic examination of the biofilm population on the e-plate post-real-time monitoring using a confocal microscope show a profound biofilm surface apparent from the hazy aggregates which give off a green fluorescence in B. henselae Houston-1 (Fig 2.3d). While the majority of the B. henselae Houston-1 cells indicate an intact membrane confirming cell viability within the biofilm, dead or compromised cells with a yellowish fluorescence (white arrows) are observed. B. henselae Houston-1 ΔbadA/pNS2PTrcbadA did not exhibit as much biofilm as the

wild-type strain, but it displayed adhesive and aggregative properties with comparable cell

38 viability to the B. henselae Houston-1. Although the RTCA real-time monitoring records the most impedance from the B. henselae Houston-1 ΔbadA, the viability assay shows that a majority of the bacterial population fluoresces red (Fig. 2.3d, white circle), which suggests that a significant amount of the cells recorded by the gold electrodes are dead cells within the biofilm.

With a scaffold and staining methods previously described in Behnke O and Zelander T,

(1970) [188], we visualized the three - dimensional structures of the biofilms. Bacterial cells grown on the scaffold formed aggregates around the scaffold, similar to those seen in in vivo infective endocarditis. The biofilm formed maintained structural integrity through sample processing. In

Fig 2.4, the top row shows a 3-dimensional image of the scaffold followed by images of bacteria grown on a scaffold. B. henselae Houston-1 and B. henselae Houston-1 ΔbadA/pNS2PTrcbadA both show bacterial aggregates with the B. henselae Houston-1 exhibiting massive growth enveloping the scaffold surface. Growth and adhesion to scaffold branches by B. henselae Houston-1 ΔbadA were reduced compared to the BadA expressing strains. Alcian blue is a cationic dye known to stain and preserve the structure of polysaccharides by binding carboxyl or sulfate groups present in glycosaminoglycan/mucopolysaccharides to form an insoluble complex [188]. B. henselae

Houston-1 shows a smooth solid mass of biofilm covering the scaffold (Fig. 2.4, white arrow) while B. henselae Houston-1 ΔbadA/pNS2PTrcbadA shows less biofilm with more visible scaffolds

(white arrow). In contrast, B. henselae Houston-1 ΔbadA mirrors the sparsity of the aggregation in biofilm production. The ΔbadA strain shows weak biofilm production (red arrow) and visible single rod-shaped cells (yellow arrow) on the surface of the scaffold.

Fluorescent stains were used to identify the components of a B. henselae biofilm. WGA fluorescein is an extensively used lectin which binds N-acetylglucosamine and sialic acid residues to emit green fluorescence. Hoechst is a nucleic acid stain which emits blue fluorescence when

39 bound to A-T rich regions of DNA and Sypro-ruby is a ruthenium-based fluorescent dye which emits red fluorescence when it interacts with essential amino acids like lysine, and histidine [189].

All three stains demonstrate that the B. henselae biofilm is made up of e-DNA, protein, and polysaccharides (Fig 2.5a). As e-DNA and protein would be sensitive to DNase and proteinase K cleavage, cells were treated either during inoculation, or 24 hours post-inoculation. Treatment with either proteinase k or DNase markedly reduced the biofilm (Fig 2.5b). Both treatments at the time of inoculation did not completely inhibit biofilm formation, but biofilm formation was reduced to less than 50%. Maturing biofilms (24 hours) were more resistant to the application of the proteolytic enzyme. We did not expect to observe a complete dispersal because protein interactions with exopolysaccharides and nucleic acid components are involved in a biofilm matrix. Hence we predicted that while proteinase K and DNase will induce an increased inhibitory effect, it would not result in the complete dispersal of the biofilm. The sensitivity of the biofilm to both enzymes confirms the presence of proteinaceous components and extracellular DNA.

We examined the effects of environmental conditions on a B. henselae Houston-1 biofilm formation. One of the main differences between the arthropod vector of B. henselae, C. felis, and the mammalian host is temperature. B. henselae must adapt to the lower temperature while inhabiting the C. felis vector and rapidly adjust to the higher temperature of the vertebrate host such as cats and humans. To emulate the temperature of fleas in the environment, B. henselae was grown at 27°C; a representative temperature reported to model an environment for adult flea activity [190, 191]. It has been speculated that bacterial persistence and colonization in the flea gut is temperature dependent [191]. The effect of temperature on badA/BadA expression and biofilm formation in B. henselae was examined (Fig. 2.6). While there is no statistical difference between cell adhesion and biofilm formation under different temperatures, B. henselae grown at

27°C rapidly forms a biofilm in comparison to B. henselae grown at 37°C (Fig 2.6a). However, at

a temperature consistent with the mammalian host, B. henselae cells within the biofilm are more

viable when compared to lower temperatures growth (Fig. 2.6b). Perhaps, bacterial persistence at

the 27°C temperature is reduced as observed by Schotthoefer et al. [191]. To confirm if bacteria

are unable to persist at 27°C because of differential gene expression, we looked at the expression

of badA at the lower temperature (Fig 2.6c). Previous studies with other bacteria have shown that

expression of some surface proteins is dependent on temperature with lower temperatures resulting

in less adhesin expression [192]. badA was not efficiently expressed at 27°C when compared to

the high expression rate at 37°C (Fig 2.6c), perhaps the reduced rate of badA expression and

adhesion is correlated to the low cell viability observed in Fig 2.6b. Different outer membrane proteins like the filamentous hemagglutinin (Fha) with eight gene copies may compensate for the diminished role of BadA at 27°C. In conditions consistent with mammals, B. henselae transcribes badA efficiently within the first few days of growth as adhesion is needed for optimal aggregation.

Then badA transcription slows down once the biofilm is assembled (Fig 2.6c). At this stage of biofilm development, polysaccharides are produced for the assembly of the extracellular matrix whereas less BadA and other surface adhesins will be needed to aggregate cells within the biofilm

(cohesive force). Differential gene expression has been observed within other bacterial biofilms where the production of surface appendages like flagella has been reported to be reduced in sessile species with an increase in surface proteins used for transportation and excretion of extracellular products [193].

Data from Fig. 2.7a shows that a slightly basic pH (pH 7.2) favors bacterial growth, and badA transcription (Fig 2.7b). Excretion of polysaccharides, one of the components of biofilm

EPS, is sensitive to pH [194]. An alkaline environment favors biofilm development as seen in

Fig.2.7c. The pH of growth media when a biofilm is being formed is slightly alkaline (pH 7.5).

After the biofilm is formed, we observed a decline in media pH for BadA expressing cells possibly

due to the metabolism of amino acids in Schneiders media. It is interesting to note that B. henselae

Houston-1 ΔbadA maintains an alkaline pH throughout the entire growth period (Fig 2.7c). B.

henselae has been shown to catabolize amino acids through the TCA cycle releasing CO2 which

will contribute to an acidic pH. The TCA cycle releases carbon to generate ATP used to provide

the energy needed for a variety of downstream effects like activation of response regulators in a

two-component system. It is possible a change in metabolome products characterize the pH of the

biofilm media. In Pseudomonas fluorescens, planktonic metabolism was characterized by a change

in metabolome products whereas the biofilm bacteria exhibited exopolysaccharide metabolism

[195].

In conclusion, BadA is required for optimal biofilm formation; it plays a significant role in

B. henselae persistence and infection hence, it is likely that B. henselae uses BadA as an adhesin

to attach itself to a wide range of host cells and extracellular matrix proteins. There, the bacteria

form biofilms comprised of polysaccharides, protein, and e-DNA that help the bacteria prevent

phagocytosis and contribute to the persistence. We propose that B. henselae cells are disseminated

from the biofilm to circulate in the bloodstream to continue to spread infection, explaining the

enigma of persistent or relapsing bacteremia in patients infected with B. henselae. BadA plays a

critical role in this process as we have shown in this report and as observed in patients with

infective endocarditis caused by B. henselae [5].

Finally, we present evidence that a higher temperature and slightly basic pH consistent with the mammalian host are optimal for growth, adhesion, and badA expression. Since badA is not

42 expressed efficiently at lower temperatures consistent with adult fleas in the environment, we propose that other outer membrane proteins may be expressed in vivo in the flea.

Chapter 3: Elucidating the role of the 3' end of Brt1

3.1. Background

Gene regulation by transcription termination-antitermination often referred to as

transcription attenuation, is a strategy commonly used by bacteria to sense specific environmental

or metabolic signals and allows RNA polymerase to respond by either terminating transcription or transcribing the downstream genes of an operon [196].

Stem-loops, also called hairpins are base pairing patterns that occur when two regions of the same strand complement each other and base pair to form a double helix with an unpaired loop region. Computer-based analyses are used to search for possible alternative RNA-hairpin structures in the leader sequence preceding a specific gene or operon to predict transcription attenuators [197]. In RNAs, stem-loops can direct RNA folding, protect the messenger RNA

(mRNA) structural stability, provide RNA binding protein recognition sites and serve as a substratum for enzymatic reactions [198]. Stem-loops are used to terminate transcription in prokaryotes using a rho-independent transcription mechanism; a process where stem-loop forms on an mRNA strand, dissociating the RNA polymerase from the DNA template strand [199]. Rho- independent termination is frequently used by riboswitches.

Riboswitches are cis-acting RNA regulatory elements that respond to small molecule metabolites concentration or to environmental changes like pH and temperature (RNA thermometers) to affect gene expression levels. They were first described as transcription attenuators because changes in transcript level affected transcript elongation by the formation of a

terminating/anti-terminating structure. Riboswitches have been shown to regulate gene expression

by using Rho-independent termination [200], sequestering the ribosome binding site (RBS) [201],

as a ribozyme that cleaves itself in the presence of a metabolite [202], regulating adjacent genes

[203], or responding to temperature by melting the stem-loop to allow transcription of a

downstream gene [204]. Riboswitches generally consist of two parts: the aptamer region that binds the ligand and the expression platform that adopts the conformational RNA structure change regulating gene expression.

The RNA thermometer is another method of regulating the expression of a downstream

gene. Temperature-dependent stem-loops/hairpins structures form in the mRNA preventing

transcription. Most thermometers form stem-loops which sequester the ribosome binding site

sequence. Very few signal transduction systems that respond to sudden changes in temperature

can detect actual changes [205], most measure the effects of temperature-induced damage (heat

shock) rather than temperature [206]. RNATs, on the other hand, respond immediately to control

the already existing mRNA translation [160].

As previously discussed, the Brt RNAs all have a stem-loop at the 3' end (Fig 1.1a) with

very low Gibbs free energy that is considered the most favored stable structure. The brts are highly

conserved especially around the 3' region (Fig 1.1b) and are highly transcribed under the conditions

of the RNA-seq (Fig 1.1c) [128]. However, the trps found downstream of each brt were minimally

transcribed. The high brt transcription ends just before the coding region of the downstream trp

gene. Also, the region between the tentative end of the RNA and the beginning of each trp harbors

approximately 15 nucleotides eliminating the probability of a native trp promoter (Fig 1.2b).

Using the rationale that the most expressed Brt, Brt1, may have the most impact on phenotype, we used the first pair of this multicopy gene family, brt1/trp1 to study the role of the

3' end of the Brt1 as a transcript terminator preventing transcription of the downstream trp1.

3.2. Materials and Methods.

3.2.1. Bacterial Strains

E.coli strain DH5α, DH12s, B. henselae Houston-1/pNSbrt1-GFP, B. henselae Houston-

1/pNS2brt1Δ3'-GFP, B. henselae Houston-1/pNS2-GFP and B. henselae Houston-1brt1Δ3' were all used for this study (Table 3.1). E.coli DH5α and DH12s cell were grown at 37°C on either LB agar or liquid LB broth. B. henselae is grown as previously described. All B. henselae Houston-1-

GFP constructs were grown on chocolate agar supplemented with kanamycin (50µg/ml). B. henselae Houston-1brt1Δ3' were grown on chocolate agar.

Table 3.1. Bacterial strains and primers used for aim 3

Primer/plasmid Sequence/purpose

Brt1 prom f GCAGGTCGACATAACTTTCACAGACGAAAACGG

Brt1 prom R GCACGGATCCCGACTTGCTTTTAGTG

Brt1 term R GCACGGATCCTGATGCTTCTTCTGAAGCGA

GFP for TGGCGGATCCGAGGAGAATTAAGCATGCG

GFP rev TGGCTCTAGAGAGGAGAATTAAGCATGCG

Brt1 3' F1 GCAGGGATCCTAGCGATAAAC

Table 3.1 (continued).

Primer/plasmid Sequence/purpose

Brt1 3' R1 AGACCCCGGAATACCAACG

Brt1 3' F2 CGTTGGTATTCCGGGGTCTGTGGGGAATCTTTCAAA

Brt1 3' R2 TCTAGAGGATCCCTTGCCCTCTG

Brt1 F TAGGGCCTGGAAGCTCTAAC

Brt1 R GTCTGGAAGCACTGACTACG

B. henselae Houston- To control for transcription of GFP without a promoter 1/pNS2-GFP

B. henselae Houston- To assess transcription of GFP using a native brt1 promoter 1/pNS2PTrcbrt1-GFP and gene

B. henselae Houston- To assess transcription of GFP using a native brt1 promoter 1/pNS2PTrcbrt1Δ3'-GFP and the brt1 gene without the 3' end of the Brt1.

Restriction sites are italicized.

3.2.2. Construction of B. henselae Houston-1-GFP strains.

3.2.2.1. Construction of B. henselae Houston-1/pNS2 gfp mut 3

A 751nt gfp product was codon-optimized for B. henselae and commercially synthesized by Gene Universal in a pJET plasmid. The gfp was amplified using gfp for and rev primers (table 3.1). The PCR product was cleaned up (NEB T1030s), digested with BamHI and

Xbal enzymes (NEB RO136S, RO145S), and gel purified (NEB T1020s). The purified GFP amplicon was ligated into the pNS2 plasmid using the BamHI and Xbal sites. The plasmids were

transformed into competent DH5α E.coli cells and the resulting colonies selected using 50µg/ml

kanamycin. Colonies are grown in 5ml of LB media supplemented with kanamycin; plasmid was

extracted using a miniprep kit (Promega A1460) and sequence confirmed by Genewiz Universal.

The plasmid was electroporated into B. henselae Houston-1 using the protocol described by Resto-

Ruiz et al. 2000 [207] to produce B. henselae Houston-1 pNS2- gfp mut 3 (Fig 3.1a, Table 3.1).

3.2.2.2. Construction of B. henselae Houston-1/pNS2 brt1Δ3'- gfp mut 3

A 471nt PCR product bearing the Brt1 promoter and 145/196nt of the full-length

brt1 was amplified using Brt1 prom for and Brt1 prom rev primers (Table 3.1). The product was

processed (purified, digested and gel extracted as described above) and ligated in-frame into the

pNS2-PTrc plasmid using the Sall and BamHI sites to remove the Trc promoter. The plasmid was

transformed and selected as described above. Positive colonies were grown in 5ml of LB media

supplemented with kanamycin; plasmid is extracted using and sequence confirmed. Positive

plasmids were electroporated into B. henselae Houston-as described above to produce B. henselae

Houston-1 pNS2-Brt1 promoter-brt1 Δ3'-gfp (Fig 3.1a, Table 3.1).

3.2.2.3. Construction of B. henselae Houston-1/pNS2 brt1- gfp mut 3

In order to generate B. henselae Houston-1/pNS2 brt1-GFP, a 609nt PCR product bearing the brt1 promoter and 196/196nt of the full-length brt1 and the intergenic region between

brt1 and trp1 was amplified using Brt1 prom for and Brt 1 term rev primers (Table 3.1). The

product was processed as described above and ligated in-frame into the pNS2-PTrc plasmid using

the Sall and BamHI sites. The plasmids were transformed, selected, sequence confirmed and

electroporated as described above to produce B. henselae Houston-1 pNS2-Brt1 promoter-brt1-

gfp (Fig 3.1a). All three GFP expressing strains were used for growth rates to determine differences

in growth pattern if any (Fig 3.1b) 48

Fig 3.1: Growth of B. henselae brt1-GFP constructs. A. Diagram of the different constructs of brt-gfp in B. henselae. B. Growth curve experiment using all three Bh-GFP expressing strains.

3.2.3. Fluorescence microscopy

Using CLSM imaging, we monitored GFP expression in B. henselae GFP strains, 10^6

bacteria cells in a 200µl volume were seeded into 8-well chamber slides and grown overnight at

37°C and 5% CO2. Samples were examined next day at 10x magnification using an Olympus

Fluoview FV1000 microscope. Images were exported as TIFF.

3.2.4. Flow Cytometry

B. henselae GFP cells were seeded into a 50ml flask (100,000 cells/flask) for 48 hours.

Cells are collected by gently scraping and washed with PBS. Flow cytometry analysis was

performed with BD FacsCanto II flow cytometer using the 530/30nm detector with laser emitted

at a fixed wavelength of 488nm. A minimum total of 50,000 gated events were collected for all

three samples. Instrument setting was kept the same for all samples and independent experiment.

Flow cytometry analysis was performed three times using a cutoff of 50,000 cells. Data were

analyzed using Facs Diva 6.3.1 software and graphs were made with Sigma plot.

3.2.5. Construction of B. henselae Houston-1 Brt1Δ3'

A deletion mutant of the 3' end of brt1 in B. henselae Houston-1 using the two-step mutagenesis strategy described by Mackichan et al. (2008) was constructed with certain modifications. [208]. To generate two fragments of the gene, B. henselae genomic DNA was used as a template for PCR. Primer pair Brt1 3' F1 and R1 was used to amplify the first fragment containing the promoter region and the brt1 gene without the 3' loop. The second fragment containing the 15 nucleotides intergenic region between brt1 and trp1 and the complete trp1 gene was amplified using primer pair Brt1 3' F2 and R2. The two purified PCR products were used as templates for mega prime PCR using only the primers Brt1 3' F1 and R2. The resulting product was ligated into the “suicide” plasmid pJM05 at the BamHI restriction site, transformed in E.coli

DH12S, verified by plasmid sequencing and then incorporated into B. henselae Houston-1 by transconjugation. Transconjugants were selected by plating on Heart Infusion agar with supplemented with kanamycin (30 μg/mL). Colonies were counter-selected on agar containing

10% sucrose to select for a second cross-over event, resulting in the full-length gene being replaced with the truncated version producing B. henselae Houston-1 Brt1Δ3' (Fig 3.3a). The deletion mutant was confirmed by PCR and RT-PCR to ensure the absence of the full-length brt1.

3.2.6. Biofilm formation in B. henselae Houston-1 Brt1Δ3'

Biofilm was monitored using methods described in Aim 2.2.4. Cells were seeded onto the plates, and real-time measurement occurred every 15 mins over 72 hours. Data was exported as an excel sheet, and Sigmaplot was used for graph and data analysis.

3.2.7. brt1, trp1, and badA expression in B. henselae Houston-1 Brt1Δ3'

To examine the expression of badA, brt1, and trp1 in a B. henselae Houston-1 Brt1Δ3'

culture, cells were cultured on chocolate agar plates at 37°C, 5% CO2 for 3 days. The method described in aim 2.3.3 was used. Briefly, bacteria were collected, gently washed with PBS and

RNA extracted with the trizol reagent. 10µg of the resulting RNA was treated with Turbo DNase and 1µg reverse transcribed to cDNA. RT‐qPCR was performed in a total volume of 25 μl as previously described using primers badA screen 1F and 1R (Table 2.1), trp1 f and r (Table 3.1),

brt1 F and R (Table 3.1). The 50S ribosomal protein L4 (rpld primer f and R: table 2.1), was used

as the reference gene for normalization. The reaction conditions are as previously described.

3.3. Results

3.3.1. The 3'end of Brt1 acts as a transcriptional attenuator.

Expression of GFP under the control of a full-length or truncated Brt1 was monitored using both fluorescence microscopy and flow cytometry. All transformants showed differential gene expression level. The fluorescence of the GFP expressing B. henselae Houston-1 pNS2/brt1 Δ3’-

gfp was stronger for both microscopy (Fig 3.2a) and flow cytometry (Fig 3.2b) in comparison to the fluorescence observed from the control B. henselae Houston-1 pNS2–gfp or the B. henselae

Houston-1 pNS2/brt1-gfp. In the absence of the 3' terminating loop, GFP expression using a flow cytometer was at least 30 times more intense than the Houston-1 pNS2-gfp or the B. henselae

Houston-1 pNS2/brt1-gfp (Fig 3.2b, P= 0.008). Both Houston-1/pNS2-gfp and the B. henselae

Houston-1 pNS2/brt1-gfp exhibited low GFP fluorescence that did not differ statistically (P=

0.119). The presence or absence of the 3' region did not impact growth as growth rate with all three strains resulted in similar curves (Fig 3.1b).

Fig 3.2: Comparison of GFP intensity in Brt1-GFP constructs. A. Confocal microscopy showing B. henselae Houston-1/pNS2-gfp,

B. henselae Houston-1/pNS2-brt1-gfp, B. henselae Houston-1/pNS2-brt1Δ3'-gfp using confocal microscopy. B. Comparison of

GFP intensity in B. henselae Houston-1/pNS2-gfp, B. henselae Houston-1/pNS2-brt1-gfp, B. henselae Houston-1/pNS2-brt1Δ3'- gfp using flow cytometry. Bars represent the average of three independent experiments. Error bars represent standard error of the mean. (Students t-test, * = P = 0.008, ** = P = 0.119).

3.3.2. Biofilm formation increases in B. henselae Houston-1 brt1Δ3'

To assess the ability of B. henselae Houston-1/brt1Δ3' to form biofilms, the xCelligence

RTCA method applied in aim 2.2.4 was used. Fig 3.3c shows the effect of the 3' terminating loop

on biofilm formation. The absence of the loop significantly affected adhesion and biofilm

formation in B. henselae Houston-1/brt1Δ3'. We observed that in the absence of this loop, the

adherent period for B. henselae Houston-1/brt1Δ3' was faster and quicker in comparison to the

time it takes the wildtype B. henselae Houston-1 to adhere (Fig 3.3c, magnified circle). The biofilm

formed between the two strains was significantly different (P = ˂0.001, Mann-Whitney U

Statistic). Fluorescent microscopy was used to evaluate the viability of cells within each biofilm

(Fig 3.4a) and we observed that after 48 hours, the cells within the biofilm are less viable when compared to the Houston-1 cells.

Fig 3.3: Growth and biofilm analysis of B. henselae Houston-1/brt1Δ3‘. A. Construction of B. henselae Houston-1/brt1Δ3‘. B.

Growth curve of Houston-1 vs. Houston-1/brt1Δ3‘. C. Biofilm formation of Houston-1 vs. Houston-1/brt1Δ3‘.

3.3.3. brt1, trp1, and badA expression in B. henselae Houston-1 brt1Δ3'

Figure 3.4b shows that in the absence of the 3' end of Brt1, both trp1 and badA transcription rates are upregulated in comparison to the B. henselae Houston-1. brt1 is still detected in B. henselae Houston-1/brt1Δ3' signifying that the gene is still being transcribed. Since growth curves

between the two strains did not differ (Fig 3.3b), the changes in gene expression was not a factor

of growth. The absence of the 3’ stem-loop affected the transcription of the downstream gene, trp1

and also upregulated badA expression.

Fig. 3.4. Cell viability and gene expression levels in B. henselae Houston-1/brt1Δ3'. A. CLSM image of B. henselae Houston-1 and B. henselae Houston-1/brt1Δ3 biofilm populations using the STYO9/PI live/dead staining. B. Gene expression level in B. henselae Houston-1/brt1Δ3' biofilm show significant upregulation of trp1 and badA (P = 0.001). There was no significant

difference in brt1 expression (** = P = 0.115).

3.4. Conclusion

Genetic adaptation is the cornerstone of fitness and survival, and flexibility in controlling

gene expression permits the bacteria to rapidly express genes required for survival. Regulatory

regions may be present in the 5′ UTR of the mRNA (e.g., riboswitches, thermosensors, and pH

sensors) that they regulate.

The Brts are ncRNAs that are highly expressed in the genome of B. henselae. The Brts was

the 4th highest genome reads by the RNA-seq analysis [157]. Indeed, the Brts have been

characterized as being “located in a plastic, repeat-rich genome region leading to strong

transcription of genes that do not represent a bona fide protein-coding ORF” (Fig 1.2a) [157]. This

characteristic signifies horizontal gene acquisition. Horizontal gene transfer allows bacteria to

acquire and express genes which respond quickly and adapt to a stressed environment promoting

bacteria survival and fitness in a new environment.

Bioinformatic analysis using NCBI BLAST show that the Brts have a high degree of homology and are specific to the Bartonella species only. The number of RNAs varied between species with the human-specific species bearing 0- 2 copies [128]. It is, therefore, possible that B. henselae acquired and duplicated the new region of high repeats to help adapt to new mammalian hosts.

RNA-Seq shows that although the Brts are highly transcribed, the downstream trp genes

are poorly transcribed [128]. The distance between the tentative end of the brts and the start codon of the trps is approximately 15nt (Fig 1.2b and Fig 3.3a). The arrangement suggests that the cis-

acting element in the 3' end of the Brt1 controls transcription of the downstream trp1. Using the

neural network for promoter prediction program [209], we were unable to find a putative promoter

within the 15 nucleotides intergenic region between brt1 and trp1 or an open reading frame on the

Brt1 RNA. Using ARNold [210], Brt1 is predicted to form a rho-independent stem and loop

structure. Perhaps this terminating region is responsible for pre-mature trp transcription. To

confirm this hypothesis, the trp1 gene was replaced with the gfp gene. The brt1 promoter and a full-length or truncated brt1 was used to drive the transcription of gfp using a promoterless plasmid with gfp to serve as a control (Fig 3.1a). To determine if the 3' loop acts as a terminator, a portion of the brt1 bearing a native promoter but without a terminating loop structure was cloned upstream of the gfp gene (Fig 3.1a) and GFP expression in the absence of the 3' end was quantified. All strains were analyzed for growth defect. Fig 3.1b shows that all three strains have no significant difference in growth rate. The bacteria clones were grown on a 6 well plate for 24 hours and

imaged. We observed that in the absence of the 3'end of Brt1, there is GFP transcription and

translation which is otherwise absent in the full-length Brt1 (Fig 3.2a) . We used a flow cytometer

to quantify the intensity of the GFP expression and observed an upregulated GFP expression in

comparison to the full-length Brt1 and the promoterless GFP plasmid. Deletion of the 3' loop

clearly resulted in higher GFP expression (Fig 3.2).

To confirm the effect of this stem-loop on the native trp1 and biofilm formation, the 3' end

of the Brt1 in the B. henselae Houston-1 was deleted to generate B. henselae Houston-1 brt1Δ3'

(Fig 3.3a). Biofilm data indicates that the Δ3' mutant shows a significant increase in time for

adherence, and for biofilm formation in comparison to the B. henselae Houston-1 (Fig 3.3c). The

growth curves for both strains did not show a significant difference (Fig 3.3b), hence the difference

in both aggregation and biofilm formation was not a factor of growth. The confocal analysis using

the live/dead stain previously described shows that cells in the biofilms have a significant

difference in cell viability (Fig 3.4a). Cells in the B. henselae Houston-1 brt1Δ3' biofilm were less

viable as signified by the increased presence of PI. Previously in Chapter two, the absence of badA

was speculated to affect cell metabolism. Bacteria cells regulate checkpoint(s) which control cell

growth and replication rates in a biofilm by using quorum sensing. It is also possible that deleting

the terminating loop affects a checkpoint ultimately leading to badA overexpression and nutrient exhaustion.

Clearly, the data presented here confirm that Brt1 acts as a transcription terminator for trp1.

We speculate that one of the major differences between the arthropod vector and the mammalian host (outlined in table 1.1) may be responsible for the formation of this terminating loop. Further experiments will be carried out to determine if any of the environmental factor(s) is/are responsible for the activation/deactivation of the loop.

Chapter 4: Role of Trp1 in biofilm regulation

4.1. Background

Prokaryotic cell regulation must be highly responsive to dramatic environmental changes to enable bacteria to adapt. Stress signals activate or repress genes controlling the expression of genes that are best suited for the fitness of the cell in the current environment. The most critical control point for gene expression is transcriptional initiation. It is regulated by a combination of factors, including the sequence of DNA and structure that permits promoter recognition, transcription factors, and small molecules [211]. Transcriptional factors (TF)/transcriptional regulators/DNA binding proteins are proteins that bind specific sequences upstream of their target genes and are involved in gene regulation by activating or repressing gene expression [212]. TFs must contain at least one DNA binding domain. Three recurrent DNA-binding motifs have been described: the helix-turn-helix (HTH), the winged-helix (WH) and the β ribbon [213, 214]. The

HTH motif is the most widely used DNA - binding motif in prokaryotes. It consists of two helices, packed at angles of approximately 120°, together with a tight four - residue turn with glycine usually found in the second position. The HTH motif alone is not enough for independent folding, and a third α helix stabilizes the motif as a compact, globular domain. The second helix in the HTH motif is called the 'recognition helix' because it is inserted into the major groove of the DNA and is critical to nucleotide specificity. Most TFs are identified by the presence of a DNA binding domain using sequence searches against protein family databases like NCBI-BLAST [215] and

PFam [216].

XREs (xenobiotic response elements) are a family of TFs and the second most frequently occurring regulator family in bacteria that control various and diverse metabolic functions [217].

The XRE family of transcriptional regulators are a large family of proteins with a DNA -binding helix-turn-helix motif similar to the SinR protein in Bacillus subtilis [218]. XRE-family regulators share an exclusive N-terminal HTH DNA binding domain, while the C-terminal regulatory region is highly variable [219]. In the literature, HTH XRE has been shown to serve as a switch between bacteria motility, adhesion [220], and biofilm formation [173, 176]. Most prokaryotic TFs are homodimers that bind to DNA sites that are palindromic or pseudo palindromic often called DNA recognition sequences, regulatory elements or transcription factor binding sites (TFBS) [221, 222].

TFs may work alone or with other proteins in a complex to promote (activate) or block

(repress) recruitment of RNAP ultimately affecting gene transcription. In the presence of a repressor, TF is inactive but become activated once the repressor is dissociated from target DNA after association with an inducer [223]. On the other hand, transcription factors known as positive regulators or activators, require interaction with effector ligands to function. Repressors and activators are interconvertible depending on the DNA binding position of the promoters [223].

Prokaryotic transcription factors can detect changes in extracellular environmental conditions and/or intracellular metabolic states. Some genes are regulated by 2 or more transcription factors with each transcription factor expressed under different conditions as observed in the stress response genes of E.coli, biofilm and planktonic states of cells [224-226]. Hence in a biofilm, expression rates of TFs will fluctuate depending on their regulatory target(s).

Advances in molecular genetics have shown that while unusual, bacterial genomes may harbor multiple copies of their genes [227]. By remodeling their genomes, bacteria adapt to new environmental niches. Genome sequencing has revealed a prominent role for gene gain and loss in

niche adaptation, specialization, host switching, and other changes in lifestyle. [228]. Multicopy

genes are common on the bacterial chromosomes and while no phenotype has been consistently

associated with multicopy genes, multi copies of rRNA have been linked to high growth rates

[229].

The multicopy trp genes are annotated as genes which code for transcriptional regulators.

They encode the distinguishing HTH-XRE factor. Under the conditions for the RNA seq carried out by Tu et al. (2016) [128], the trps were minimally transcribed. Overexpression of the trp1 gene shows that trp1 is a positive regulator of badA [128].

To confirm the relationship between the badA and Trp1, we employed the use of

bioinformatics, electrophoretic mobility shift assay (EMSA) and a tagged DNA pull-down

accompanied by mass spectrometry. We used both BioPHP [230] and EMBOSS [231] to screen

for the presence of a palindrome or an inverted repeat in the badA promoter region. EMSA is a

rapid, sensitive, qualitative method used to detect protein-nucleic acid interactions [232, 233]. In

the assay, the protein and the DNA components are combined in vitro, and the resulting mixture

is subjected to non-denaturing electrophoresis. The mobility of a protein-nucleic acid complex is

typically less than that of the free nucleic acid. The DNA pulldown product is subjected to

electrophoresis and mass spectrometry (MS) to identify other proteins isolated by a DNA pull-

down strategy from a Houston-1 lysate using a biotinylated badA promoter region.

In this chapter, I hypothesize that one or more of the Trps are the TFs, activators, and/or

repressors responsible for regulating the transcription of badA.

4.2. Materials and Methods

Table 4.1. Bacterial strains and primers used for aim 4.

Plasmid/primer Sequence/purpose

badA Promoter F AGTACACAACAAAAACAGCC

5' Biotin-badA Promoter R (Biotin) GTCTCTTTGATGTGACAG

badA Promoter R GTCTCTTTGATGTGACAG

badA coding region F ACGCATGTAGAGAATGGTGA

badA coding region R CTTCGCATCTTCAAGCACTATCT

5’ biotin-badA coding region R (Biotin) CTTCGCATCTTCAAGCACTATCT

Solu E.coli Expression of His-tagged Trp1

4.2.1. Construction and purification of SoluBL21/pET28a Trp1

The trp1 gene, codon-optimized for B. henselae, was cloned into pET28a plasmid using

the Ndel, and Xhol restriction sites and a 6x‐His affinity tag appended to the N terminus. The

plasmid was transformed into SoluBL21 competent E.coli (Amsbio, C700200). Positive clones were selected with 50ug/ml kanamycin and confirmed by PCR. Solu BL21 cells were grown in

LB media until OD reached 0.4 (OD600nm). Cells were induced with 0.5mM IPTG for 3 hours at

27°C, pelleted and lysed with a cocktail of B-per lysis buffer (ThermoSci, 89821), a protease

inhibitor, KCl, and lysosome. The 6xHis-tagged Trp1 was purified using Ni‐NTA affinity spin

column (ThermoSci, 88225) according to the manufacturer’s protocol. Protein was concentrated

and desalted with a protein concentrator (Thermo Scientific™ Pierce™ PES, 88513) and sample

purity checked with a 4-12% Bis-Tris gel (Invitrogen, NP0332Box) stained with Coomassie. A

western blot was also performed using an anti-his tag primary antibody to confirm the presence of

the tagged protein. Protein concentration was calculated using OD280nm.

4.2.2. Biotinylation, amplification, and purification of the badA promoter region.

A 342bp region consisting of the badA promoter and coding region was amplified using primers badA Promoter F and biotin-badA promoter R (Table 4.1) bearing a standard biotin tag on

3'end. The amplicon was cleaned up (NEB T1030s), gel extracted on a 0.7% agarose gel and purified (NEB T1020s). The unlabeled promoter region used as a control was made with the same primer set without a biotin tag (badA Promoter R: Table 4.1). Our negative control, a 350bp of the

3' coding region of badA was made using the primer set badA coding region F and R (Table 4.1).

All amplicons were run on a 0.7% agarose gel to purify and confirm the absence of primer dimers or contaminating band(s).

4.2.3. Electrophoretic Mobility Shift Assay (EMSA)

20fmol (1nM) of biotin-labeled badA promoter region, 200pmol (10µM) of Trp1 protein,

1x binding buffer, 2.5% glycerol, 5mm MgCl2, 0.05% Np-40, and 50ng/µl of Poly-IDC

(ThermoFisherSci., 20148) was mixed in a total volume of 20µl and incubated for 30 mins at 37°C.

200nM of non-biotin badA promoter region was used as a competitive inhibitor. 5µl of 5x loading

buffer was added, and the reaction was resolved by electrophoresis using a 6% DNA retardation

gel (Invitrogen, EC6365) in 0.5x TBE buffer. The gel was transferred at 380mA for 50 mins onto

a nylon membrane (ThermoFisherSci. Biodyne™ B Nylon Membrane, 0.45 µm, No 77016) using

cooled 0.5x TBE buffer in a circulating water bath. The membrane was cross-linked at 120mJ/cm2

with the Stratagene UV strata linker (auto crosslink function). The membrane was blocked,

washed, equilibrated and exposed according to the manufacturer’s protocol. Bio-rad chemidoc

XRS imaging system was used to detect bands and images exported as TIFF. EMSA experiments

were repeated three times independently to confirm reproducibility.

4.2.4. RT-qPCR of trp1 and badA in a biofilm community.

Using primer pairs badA screen 1F and 1R (Table 2.1) and trp1 F and R (Table 3.1), the expression levels of trp1 and badA in a biofilm was analyzed and compared to bacteria grown in the planktonic state. The bacteria were grown in a 6-well plate for 48 hours, washed gently, and trizol reagent added directly to the biofilm at the bottom of the plate. To mimic planktonic growth, the 6 well plate containing the samples were incubated on a rocker with mild agitation (100rpm) to prevent adhesion. Sample processing and RT-qPCR experiment were performed as described in

Aim 3.2.6.

4.2.5. badA promoter region pulldown.

To identify all proteins which may act as activators or inhibitors of badA transcription, the biotinylated badA promoter region from aim 4.2.2 above was used to perform an in vitro pulldown assay. B. henselae Houston-1 cells were grown as a biofilm in a 50ml flask. The supernatant was removed and adherent cells at the bottom of the flask were lysed using a non-denaturing cell lysis buffer (ThermoSci. 89821) with protease inhibitor. The cells debris was pelleted for 5 mins at

14,000xg, and the lysate supernatant was incubated overnight with the biotinylated DNA at 4°C.

The sample was incubated with streptavidin beads at room temperature for an hour, washed according to the manufacturer’s protocol and eluted in 60ul of the elution buffer present in the kit.

The sample was spun to dry in a vacuum, reconstituted using ultrapure mass spectrometry grade water and analyzed using an LC/MS. Data were analyzed using Proteomics discover (ThermoSci) and SEQUEST [234]. Products were sorted based on peptide-spectrum matches (PSM) and confidence intervals (0.01 for high and 0.05 for medium).

4.3. Results

4.3.1. Trp1 binds badA promoter region to activate gene transcription.

EMSA was carried out to determine the role of the Trp1 protein as a transcriptional factor for badA. The data shows that the purified Trp1 protein binds the 342bp containing the putative promoter region of badA gene (Fig 4.1). A substantial shift in the control DNA band was observed when incubated with the recombinant Trp1 protein. This band shift is suppressed in the presence of a 200-fold excess unlabeled DNA which acts as a competitive inhibitor. To further confirm the specificity of the Trp1 to the putative promoter region, a negative control which consists of badA coding region (3' end of the badA gene) was added to Trp1 alongside the biotin-tagged promoter region and no band displacement was observed. This result suggests that the Trp1 regulates the transcription of badA gene through direct, specific interaction with the badA promoter region.

Fig 4.1. EMSA image of Trp1 and badA putative promoter region. EMSA image confirming that Trp1 binds the promoter region of badA to turn on gene transcription.

4.3.2. trp1 and badA are upregulated in a biofilm community

RT-qPCR using badA and trp1 specific primers was used to access transcription levels

between growth methods. The data shows that both genes are significantly upregulated in a biofilm

community when compared to bacteria grown to mimic planktonic growth (Fig 4.2a). Likewise

looking at the transcription rate of trp1 and badA in a B. henselae Houston-1 biofilm, we observe

that both genes are significantly upregulated to coincide with the first step of biofilm formation-

adhesion on day 1. As the number of days increases, we observe that the transcription rate of both

genes reduces as the biofilm is being formed (Fig 4.2b). This data confirms that both trp1 and

badA are needed for optimal biofilm formation.

Fig 4.2. Gene expression levels in a biofilm. A. Comparison of trp1 and badA expression levels in a biofilm. B. Expression levels of trp1 and badA overtime in a B. henselae Houston-1 biofilm. RT-qPCR bars represent the average of 6 replicates and error bars represent the SEM. (P ≤ 0.001 for both Fig 4.2a and b).

4.3.3. Trp3, amongst other genes, also binds the badA promoter region.

To determine whether badA is capable of being bound and regulated by any other DNA

binding protein, we employed the use of streptavidin pulldown coupled with mass spectrometry.

Data shows that Trp1 is not the only protein which interacts with badA promoter. Using the PSM

and a medium confidence interval cutoff, it was noted that 5 other proteins (Table 4.2) not

including Trp1 are able to bind the promoter region of badA to regulate expression. Trp3, a HTH-

XRE-DNA binding protein and member of the Trp family is also able to bind the badA promoter

region. The other proteins identified could associate either by direct binding or through protein-

protein interaction as in the case of ClpB which is known to help with protein folding/refolding.

Table 4.2. Proteins that bind to the badA promoter region.

Accession Description Locus number PSM number A0A0H3LWB5 Hypothetical prophage protein BH03140 111

A0A0H3LWT1 Transcription-repair-coupling factor BH08750 111

A0A0H3LXU6 Chaperone protein ClpB BH14110 71

A0A0H3LYA4 Transcriptional regulator (Trp3) BH13590 22

A0A0H3LXV5 Uncharacterized protein BH14240 19

4.4. Conclusion

As briefly mentioned in the introduction, TFs recognize palindromic or pseudo palindromic

sites on the promoters of genes they regulate. Gene expression is regulated by TFs that bind the promoter region of a gene to activate/repress gene transcription. Many bacterial transcription factors are dimeric proteins and their transcription factor binding site (TFBS) are generally believed to be palindromic or symmetrical [222, 235]. Bioinformatics analysis using a 342 bp upstream of the badA start codon shows that there are palindromic sequences upstream of the badA start codon. To determine whether Trp1 directly binds badA promoter region, EMSA was

employed. Fig 4.1 confirms that there is a direct interaction between the promoter region of badA

and the Trp1 TF. We observe a single lower band in lane one for the target DNA but on the addition

of Trp1, we observe an increase in band size signaling increased molecular weight from the interaction of badA and Trp1 (Fig 4.1, lane 2). We also observe a competitive interaction with the addition of non-biotinylated target DNA- lane 3. The intensity of the band in column two decreases suggesting biotinylated DNA displacement which is confirmed by the increased intensity of the lower band in lane 3. We also confirmed that this interaction is specific to the promoter region of badA; a negative control using a coding region of the badA gene was unable to bind or compete with badA DNA for Trp1 (lanes 5 and 6).

As confirmed by the EMSA, trp1, binds directly to badA. If the Trp1 is responsible for activating badA transcription and BadA is responsible for the first step in biofilm formation, then an upregulation of Trp1 will lead to upregulation in BadA and biofilm formation. Fig 4.2 shows upregulation of trp1 and badA in a biofilm in comparison to planktonic growth. In 4.2b, we observe a decline in gene transcription level as days increases perhaps signaling the reduced need for trp1 and badA once a biofilm is formed.

To identify other proteins which may regulate expression of badA, the biotinylated promoter region of badA was used to perform a pull-down with cell lysate of B. henselae Houston-

1. The DNA-protein complex was purified using streptavidin beads. While the EMSA was used to confirm that Trp1 binds badA promoter, the pull-down identified other activators/repressors, etc. that may also interact with the promoter region of badA. Trp3 is a transcriptional regulator with an

HTH DNA binding site, hence it is capable of regulating a gene much like the Trp1. Much like

Trp1, the Trp3 belongs to the family of nine transcriptional HTH regulator found in B. henselae and also harbors Brt3 upstream of the start codon. Other proteins pulled down include ClpB, an

ATP-dependent chaperone protein and a part of the stress-induced multi-chaperone system involved in protein metabolism and in the recovery of cells from heat-induced damage [236]. It

unfolds denatured protein aggregates and exposes new binding sites on proteins bound to it contributing to solubilization and refolding of a denatured protein aggregate by DnaK. The

transcription-repair coupling factor (TRCF) in bacteria is a pathway for nucleotide excision repair

(NER) and is also used to dislodge inactive RNA polymerase molecules stalled on the template

DNA.

In general, we confirm that Trp1 interacts directly with the badA promoter region (Fig 4.1).

badA and trp1 transcript levels are significantly upregulated in a biofilm community especially on

the first day of biofilm formation and decreases as biofilms are formed (Fig 2.1, Fig 4.2). Our data

show that the combined approach of capturing DNA affinity and mass-spectrometric protein

sequencing identifies proteins involved in the formation of multi-component complexes on DNA

through direct interaction with specific sequence elements or conformational structures. We were

able to capture proteins which interact with badA in a biofilm community. While we are not

entirely certain as to the functions of each protein, it is particularly interesting to note that Trp3

can also bind the badA promoter region. A previous report linked Trp1 as a positive regulator

[128]. This chapter provides evidence of direct interaction between Trp1 and badA.

Results of this aim shed light on the relationship between Trp1, badA/BadA and biofilm

formation in B. henselae. We speculate that both Trp1 and Trp3 are able to bind the promoter

region of badA as positive regulators to activate gene expression. This interaction may or may not

require co-activators. Bacteria may typically duplicate genes to compensate for a copy loss. The

bacteria will transcribe and translate badA/BadA for adhesion to the surface. Once the bacteria are

firmly attached, the signal for the synthesis of polysaccharides for the production of EPS starts and

biofilm is being formed. EPS synthesis and biofilm formation become a feedback signal to stop transcription of badA. The repressor and/or co-repressor for badA will bind to repress badA and

trp1 genes leading to the decrease observed in figure 4.2b.

Chapter 5: Role of Brt1 in biofilm regulation

5.1. Background.

sRNAs, generally 50–400nts in length, constitute a significant class of post-transcriptional regulators that control gene expression for functions as diverse as membrane homeostasis, carbon metabolism, quorum sensing, biofilm formation, etc. [128, 237]. Trans-encoded sRNAs regulate mRNAs through short, imperfect interactions between bases. These sRNAs base pair at/near the target RBS and block transcription or translation by preventing access to the ribosome. Others base pair at more distant locations and may interfere with ribosome binding by other mechanisms such as increasing ribosome binding by preventing secondary inhibitory structures from forming, decreasing or increasing mRNA stability or affecting protein stability [237, 238]. Regulatory sRNAs usually requires the RNA binding protein, Hfq in Gram-negative bacteria for the function and/or stability of trans-regulated mRNAs [154], but some sRNAs do not require Hfq [239]. sRNAs may also impact translation by functioning as an asRNA.

Antisense RNAs maybe protein-coding or non-protein-coding genes and can regulate the expression of their target genes from transcription and translation to RNA degradation at one or more stages of the gene expression process [240]. Characteristically, they usually lack protein- coding potential with the exception of a few [241]. The untranslated region (UTR) of an asRNA typically overlaps with neighboring genes to control transcription [242]. asRNA may be cis-acting or trans-acting. Cis-acting asRNAs are transcribed from the opposite strand of the target gene at the target gene locus. They have a high degree of complementarity with the target gene and function by blocking access to the RBS or recruiting RNase to degrade a target [242, 243]. The

end result of most cis-acting asRNA is repression [244]. On the other hand, trans-acting RNA is

transcribed from a locus that is distant from the target gene, displays a low degree of

complementarity with the target gene, can target multiple loci, form less stable complexes with

their targeting transcripts, usually encoded in an intergenic region, control translation or

degradation of target mRNA, and sometimes require RNA chaperone protein such as Hfq to

perform their functions [201, 244-246]. asRNAs have been implicated in biofilm formation by

controlling adhesion [247], iron metabolism [248], eDNA and polysaccharide export [249], etc.

The Brts are approximately 200nt in size. Their specific roles are unknown, but they lack

protein-coding potential, are found in an intergenic region, and it has been shown that Brt1 is a

negative regulator of badA and biofilm formation: in an antisense construct of Brt1, badA

expression and biofilm formation is increased suggesting that Brt1 may function in trans to

negatively regulate badA [128]. In this aim, we study the role of Brt1 in badA expression,

regulation, and biofilm formation. We hypothesize that Brt1 may act in trans to down-regulate

badA expression via direct interaction with the badA transcript independent of the downstream trp

expression. We also predict that it may bind other mRNAs and/or protein to affect protein translation.

5.2. Materials and Methods

5.2.1. Construction of B. henselae Houston-1 Δbrt1

A deletion mutant of brt1 in B. henselae Houston-1 was constructed using a two-step

mutagenesis strategy described by Mackichan et al. (2008) [208] and in section 3.2.5. To generate

two fragments of the gene, B. henselae genomic DNA was used as a template and primer pairs

Brt1 KO F1, R1, and Brt1 KO F2, R2 was used (Table 5.1). The resulting product was ligated into

the “suicide” plasmid pJM05 at the BamHI restriction site, transformed into DH12s, confirmed by

71 plasmid sequencing and then incorporated into B. henselae Houston-1 by transconjugation.

Positive clones were selected 30 μg/mL with kanamycin and counter-selected on agar containing

10% sucrose for a second cross-over event, resulting in the full-length gene being replaced with the truncated version to generate B. henselae Houston-1 Δbrt1. The deletion mutant was confirmed by PCR and RT-PCR using Brt1F and R (Table 3.1) to ensure the absence of the full-length brt1.

Table 5.1. Bacterial strains and primers used for aim 5.

Primer/Strain Sequence/purpose

Brt1 KO F1 GCACGGATCCATGAAAGCTCTTTATTCCTTTG

Brt1 KO R1 CAATGCATAGAAAATTG

Brt1 KO F2 GGTGGGGAATCTTTCAAAG

Brt1 KO R2 GCACGGATCCGGGATTCTGATTCAAGTTGA

Hfq BamH1 for GCACGGATCCGTAGAAAGATCACAACACC

Hfq Xhol for GACACTCGAGGTTTATTCAGAGCTTTCACC

Brt1 T7 for GAAATTAATACGACTCACTATAGGTGTTGGT

Brt1 T7 Rev TATAACCGGGTGCGA

B. henselae Houston-1 Δbrt1 To access B. henselae phenotype in the absence of brt1

B. henselae/pNS2PTrc Houston-1/pNS2PTrc control [250]

B. henselae/pNS2T5DsRed2 Red fluorescent protein (RFP)-expressing B. henselae [181]

Restriction sites and T7 promoter are italicized.

5.2.2. RNA extraction and RT-qPCR

We examined the expression of badA, brt1, and trp1 in B. henselae Houston-1Δbrt1.

Bacteria were grown for 3 days as previously described. Primer pairs badA screen 1F and 1R

(Table 2.1), Trp1 F and R (Table 3.1) and Brt1F and R (Table 3.1) were used for this experiment.

The reaction was performed as described in Aim 2.2.3.

5.2.3. Biofilm formation in B. henselae Houston-1 Δbrt1

The ability of B. henselae Houston-1Δbrt1 to form biofilms in comparison to the wild type

B. henselae Houston-1 was analyzed. The RTCA machine was used to measure biofilm formation

in real-time using the protocol outlined in Aim 2.2.2.

5.2.4. In vitro transcription of Brt1

brt1 was amplified using Brt1 T7 for and rev primers (Table 5.1). PCR amplicon was

purified and converted into RNA in vitro (Thermosci K0441). 1µg of DNA was mixed with 4µl

reaction buffer, 2µl transcript aid enzyme and 2µl each of ATP/CTP/GTP and 2µl of a 60:40 ratio

of UTP and Biotin-16-UTP (Roche 50-100-3411) in a total volume of 20µl. A non-biotin Brt1

control was made without Biotin-16-UTP. The reaction was incubated at 37°C for 5 hours, DNase1

was added to remove contaminating DNA followed by phenol:chloroform extraction. RNA was

precipitated and reconstituted in 50ul of DEPC-water. Product was confirmed on a 1% agarose gel

and quantified on Agilent Bioanalyzer.

5.2.5. Brt1 binding partners.

To analyze other binding partners of Brt1, biotinylated Brt1 RNA was incubated for an

hour with cell lysate from B. henselae grown either as a biofilm or planktonic. The protocol was

carried out as described in Aim 4.2.5. The product was prepped for RNA seq using Illumina

stranded mRNA kit (Illumina 15031047) and samples were prepped according to the manufacturer's protocol. Brt 1 pulldown product was analyzed using Illumina MiSeq. Data were

analyzed with Qiagen CLC workbench and quantified using the transcripts per kilobase million

values (TPM) and gene unique reads to normalize for gene length and sequencing depth. Reads were mapped to B. henselae genome (Accession number BX897699) to identify sequences.

5.2.6. In vivo expression of the brt, trp and badA system in fleas.

7 x 108/ml of B. henselae/pNS2T5DsRed2 [181] was inoculated in 10ml of defibrinated

cat blood and fed to 600 cat fleas. On different time points (1, 3, 5, and 10) days post-infection, both fleas and feces are collected either in trizol for RNA extraction or anesthetized by chilling at

−20°C for 5 min and then placed into a drop of PBS for dissection. RNA was extracted from both fleas and feces using the trizol/chloroform protocol described in aim 2.2.3 above. The gut of the anesthetized flea was pulled out using tweezers and fixed in 2% glutaraldehyde, and the fecal matter was also fixed in 2% glutaraldehyde. For fecal biofilm preservation, 0.15% alcian blue was added to fixative. Samples were processed as described in section 2.2.2 and imaged using both

Olympus Fluoview FV1000 confocal microscope and a Joel JSM6490LV scanning electron microscope operated at 22Kv. Secondary images were collected as TIFF/JPEG.

5.3. Results.

5.3.1. Biofilm formation increases in B. henselae Houston-1Δbrt1.

Biofilm formation is significantly upregulated in B. henselae Houston-1Δbrt1. We

evaluated the capability of B. henselae Houston-1/Δbrt1 to forms biofilms using the xCelligence

RTCA. Fig 5.1b shows the effect of deleting brt1 on biofilm formation. The absence of brt1

significantly upregulated adhesion, aggregation and biofilm formation. We observed that the

74 adherent period is faster, much less than the time it took the wildtype, B. henselae Houston-1 to adhere. The biofilm formed between the two strains was significantly different (P = ˂0.001, Mann-

Whitney U). This difference was not attributed to growth rates as both strains have no significant difference in growth rate (Fig 5.1a). Fluorescent microscopy shows that the cells within each biofilm do not maintain the same viability (Fig 5.2a). Much like B. henselae Houston-1brt1Δ3', the B. henselae Houston-1Δbrt1also has a significant population of cells which are either dead or have compromised membranes. In contrast, the wild type biofilm maintains a considerable amount of viable cells.

Fig 5.1: Growth and biofilm analysis of B. henselae Houston-1 Δbrt1. A. Growth curve of Houston-1 vs. Houston-1 Δbrt1. B.

Biofilm formation by Houston-1 vs. Houston-1 Δbrt1.

5.3.2. trp1 and badA expression are increased in B. henselae Houston-1Δbrt1.

Figure 5.2b shows that in the absence of brt1, both trp1 and badA transcription rates are upregulated in comparison to the B. henselae Houston-1. brt1 is barely detected in B. henselae

Houston-1Δbrt1 signifying that the gene transcription was successfully deleted. As the Brts are

quite conserved, we are unable to produce primers which were uniquely specific to Brt1 hence

primers are still capable of binding with limited complementarity to regions in other Brts.

However, irrespective of the ability to bind other Brts, Brt1 transcription was severely attenuated

in comparison to the control wild-type strain. Since growth curves between the two strains did not

significantly differ (Fig 5.1a), gene upregulation was not a factor of growth.

Fig. 5.2. Cell viability and gene expression levels in B. henselae Houston-1 Δbrt1.A. CLSM image of a B. henselae Houston-1 Δbrt1 biofilm population using the STYO9/PI live/dead staining. B.

Expression levels of trp1 and badA overtime in a B. henselae Houston-1 Δbrt1 biofilm in

comparison to B. henselae Houston-1 (Students t-test, P ˂ 0.001). RT-qPCR bars represent the

average of 6 replicates and error bars represent the SEM.

5.3.3. Brt1 may function as a trans-acting RNA.

Using the biotinylated Brt1, we performed a pulldown assay with streptavidin beads using cell lysates of bacteria grown in a biofilm or as planktonic cells. Table 5.2 shows the mRNA identified from the pulldown assay and RNA-Seq. Targets were analyzed using a 21 TPM cutoff.

Brt1 incubated with planktonic cell lysate did not pull down any mRNA. mRNAs pulled down using a biofilm cell lysate were analyzed using unique gene reads and TPM to compensate for large genes like badA. Genes were sorted according to their involvement in cellular processes: stress (blue), surface adhesins and transporters (yellow), RNA biogenesis (green), metabolism

(orange), DNA binding (gray) and proteins of unknown function (no color).

Both BH16000 and mopA are annotated as heat-shock chaperone proteins which help prevent protein misfolding (Table 5.2- blue). BadA, BH13030, BH13140, SecY, FhaB1, and

BH00460 are either surface-associated proteins or autotransporters (yellow highlight). The green rows represent RNA polymerases and enzymes for RNA synthesis and degradation. This is followed by the mRNAs of genes involved in metabolic functions like SucA and SdhA (orange),

DNA binding proteins like Trp 7 and hupB (gray) and proteins of unknown functions (no color).

The control, biotinylated Brt1 incubated with lysate from B. henselae grown as planktonic cells failed to bind any mRNA. All planktonic unique reads mapped as 4 and under.

Table 5.2. mRNAs that bind to Brt1.

5.3.4. B. henselae is able to survive in both cat flea and feces.

Using the B. henselae/pNS2T5DsRed2 which expresses the red fluorescent protein

(DsRed2), bacteria were resuspended in cat blood and fed to cat fleas. Fleas and fecal matter were collected at different time points and observed. Fig 5.3a shows that the bacteria are able to grow and replicate in the cat flea. The intensity of the DsRed protein increased as the days increased signifying that the bacterial load increases in the gut of the flea. Fecal matter confocal image (Fig

5.3b) shows that B. henselae/pNS2T5DsRed2 is excreted in the feces and is able to grow and replicate. The red fluorescence can be seen in the fecal matter on days 5 and 10 post-infection and we do not observe fluorescence in the uninfected feces. Finally, scanning electron micrographs also show that bacterial biofilm, preserved by the addition of alcian blue, is present in the fecal matter. Tiny clusters of biofilms similar to biofilms previously observed in figures 2.1 and 2.4 are observed in fig 5.3c. The control fecal matter does not present biofilms. This data confirms that B. henselae is able to survive and replicate in the cat flea and feces 10 days post-infection.

Fig. 5.3. Bacteria growth in the cat flea (C. felis) and feces. A. Confocal microscopy showing the growth of B. henselae in the gut of C.felis. Scale bars -50µm. B. Confocal microscopy showing the growth of B. henselae in the fecal matter of C.felis. Scale bars -

50µm. C. Scanning micrographs of B. henselae biofilm in the fecal matter of C.felis. Scale bars: top row-20µm, bottom row: -

200µm.

5.3.5 trp1/badA system is not expressed in the cat flea but upregulated in flea feces.

RT-qPCR was performed using RNA extracted either from the flea or flea feces. Fig 5.4

shows that the brt, trp1, and badA are barely detected in the flea. In fig 5.4b, fecal gene expression shows that the three genes are detected however, brt1 maintained a strong, persistent presence from day 2. badA was minimally expressed in the feces after day 2 and we observe an increase in a trp1 from day1 to day 5 without an associated increase in badA. This suggests that Brt1, Trp1, and BadA may not be the system involved in biofilm formation in the arthropod vector.

Fig. 5.4. Gene expression in the cat flea (C. felis) and feces. A. RT-qPCR showing the expression of brt1, trp1, and badA in C.felis.

B. RT-qPCR showing the expression of brt1, trp1, and badA in the feces of C.felis). (P ˂ 0.005 for all t-tests between the blood

RNA and flea or fecal RNA). Bars represent the average of two independent experiments. Error bars represent standard error of

the mean.

5.4. Conclusion

As previously mentioned, trans-encoded RNAs can regulate multiple mRNAs because they

share limited complementarity with targets [246]. Most studied trans-acting RNAs regulate

translation and/or mRNA stability [245]. Some characterized sRNA binds the 5’ UTR of the target

to occlude the RBS and subject the target mRNA for degradation [251]. A few other RNAs activate

expression of target mRNA by binding upstream of the AUG and disrupting the formation of an

inhibitory structure that sequesters the RBS [252].

The Brt RNAs have been previously implicated in the regulation of badA [128]. In this

aim, B. henselae Houston-1 Δbrt1 strain showed an increase in adhesion and biofilm formation

and a significant upregulation for the expression of the trp1 gene and badA expression. This

suggests that Brt1 negatively regulates badA. The Brts/Trps are found in an intergenic region; a

characteristic of genes acquired by horizontal gene transfer.

We also observe that Brt1 may exert trans-acting roles on other mRNA asides from BadA.

Gene expressions differ greatly in biofilm compared to planktonic gene profiles. When bacteria are exposed to higher temperatures, heat shock proteins are induced to cope with damage to proteins. Bacteria are able to sense changes in temperature using nucleic acids (DNA or RNA) or proteins [253]. In table 5.2, we observe that heat shock proteins located in different areas of the genome interact with Brt1 mRNA. We also observe that Brt1 binds BadA mRNA possibly in a repressive manner. B. henselae Houston-1Δbrt1 presents increased badA and biofilm formation and an antisense knockdown of Brt1 also upregulated badA expression [128]. Fha mediates adhesion in Gram-negative bacteria and in B. henselae [110, 133]. SecY, a secretory pathway component is associated with decreased biofilm formation [254, 255] and ion channels typically enable communication between cells in a biofilm [256]. RNA polymerases ropB and rpoC catalyze transcription of DNA to RNA while Ribopolynucleotide nucleotidyltransferase has been

implicated in biofilm formation [257]. SucA, sdhA, and homologs of thiG1 have all been

implicated in biofilm formation: downregulation of the genes severely affected biofilm formation

[258-261]. We observed that control biotinylated Brt1 incubated with planktonic cell lysate did

not bind any of this mRNAs perhaps signifying that the Brt1 RNA is not only expressed under conditions consistent with mammals but also required an initial adhesion to become active.

In vivo expression using B. henselae/pNS2T5DsRed2 confirms that B. henselae is able to

grow and replicate in the gut of the cat flea vector. The fluorescence of the DsRed2 increased as

the days increased suggesting increased bacterial load. The fecal matter of the cat flea was

examined, and B. henselae biofilm was observed in the feces. Confocal microscopy also detected

B. henselae/pNS2T5DsRed2 fluorescence confirming their growth and viability in cat flea feces

as previously published [177]. To confirm the regulatory system used to form biofilm in the cat

flea, RT-qPCR primers for brt1, trp1, and badA were used to detect gene expression. Fig 5.4

confirms that the system is minimally expressed in the cat flea and may not be responsible for

biofilm formation. We noted that B. henselae/pNS2T5DsRed2 used to inoculate the blood

expresses high levels of brt1 but brt1, trp1, and badA expression was downregulated in the flea

despite the increasing presence of the bacteria in the flea gut (Fig 5.3a). However in the flea feces

for day 2, we observe complete repression of the Brt gene, upregulation of Trp1 and a slight

expression of badA. Brt1 expression increases in day three and trp and badA expression reduces

(Fig 5.4b). Also, fig 2.6 shows that badA is significantly downregulated in vitro under conditions

that mimic the arthropod vector. We propose that the brt1/trp1/badA regulatory system is only

used under conditions relative to the mammalian host as supported by the low expression of the

system in fleas. We also propose that this unique regulatory system may be responsive to

environmental conditions in the mammalian host and the initial adhesion stages of a biofilm. This

is supported by the fact that Brt1 did not bind mRNA of cells grown as planktonic cells at 37°C.

To summarize, Brt1 contains a transcript terminator for trp1 but may also act as an asRNA

for degradation of the BadA mRNA. We further propose that it also acts as a trans-regulating RNA

to regulate mRNAs of other outer membrane proteins, stress response, and metabolic genes as well

as genes controlling transcription and RNA stability (rpoB, rpoC, mopA, and pnp) to target mRNAs. While brt/trp/badA genes were detected in the cat feces, the expression levels were significantly lower than the expression levels in mammalian conditions hence it is possible that biofilm formation in the arthropod vector occurs through a different regulatory pathway.

Chapter 6: Discussion

Bacteria are one of the most abundant organisms on earth with current estimates of at least

1030 with ten-fold uncertainty [262, 263]. Bacteria occupy a variety of ecological niches that give

their inhabitants advantages, but a small portion of them can cause diseases [264-266]. Bacterial

survival is dependent upon the ability to sense and rapidly adapt to specific changes within the

immediate environment. Temperature, heme, pH, nutrient source, and concentration of

metabolites, ions, and oxygen are some variable environmental conditions known to influence

bacterial gene expression [98, 267, 268].

Biofilms are considered the default mode of growth for bacteria [269]. Bacteria form

biofilms in response to environmental cues like change in temperature, nutrient availability or as

a defense mechanism from insult by the host cell [270]. Biofilm formation is recognized as an

essential aspect of many, if not most bacterial diseases, including native and prosthetic valve

endocarditis, osteomyelitis, dental caries, otitis media, medical device-related infections, ocular

implant infections, and chronic lung infections in cystic fibrosis patients [271]. In this thesis, we

show that B. henselae is capable of forming a biofilm and that the bacteria expresses genes which contribute to optimal survival and fitness in a stressed environment.

From Fig 2.1, we observed that B. henselae is capable of forming a biofilm within 48 hours starting with adhesion and aggregation. The aggregates form microcolonies, grow and produce biofilms. BadA, a TAA found in B. henselae, is responsible for the adhesion and aggregation phase and we observe that biofilm cells transcribe more badA, unlike the planktonic counterpart (Fig

2.2a). In the absence of badA, B. henselae cells are able to form a weak biofilm compared to badA

expressing strains (Fig 2.3). This is because badA expressing strains tend to go through the

adhesion and aggregation phase much faster to form colonies and eventually biofilms (Fig 2.3;

enlarged image). In Fig 2.3b, we observed that using the standard CV stain, badA strains still have

more biomass, fewer cells floating in the growth media, and maintained viable cells within the

biofilm (Fig 2.3c, Fig 2.3d). In contrast, the B. henselae Houston-1 ∆badA had less biomass post

biofilm assay, more cells in the growth media and dead cells within the weak biofilm (Fig 2d). In

fig. 2.4, we ascertained that both B. henselae Houston-1 and B. henselae Houston-1

∆badA/pNS2PTrcbadA are both able to grow in a 3D fashion wrapping aggregates around the

branches of the scaffold and formed biofilms which enveloped the scaffold. B. henselae Houston-

1 ∆badA, in contrast, lacked the ability to form considerable-sized aggregates with inferior

biofilms. We observed that a B. henselae Houston-1 biofilm contains polysaccharides, protein, and

eDNA (Fig 2.4a) and that treatments with enzymes capable of digesting the major component either reduced the ability of the bacteria for form biofilms or severely affected a mature biofilm

(Fig 2.5b).

Previously, it was briefly discussed that B. henselae has to adapt to the different

environmental condition of both the cat flea and the mammalian vector. Mimicking the conditions of both vector and host in vitro shows that B. henselae favor conditions related to mammalian

hosts. It is able to grow and form a biofilm with viable cells and express badA at a temperature of

37°C and pH 7.2 consistent with mammals. Under conditions relative to the cat flea, it is still able

to grow, form biofilms and express badA but in comparison to mammalian host conditions, this

ability is significantly downregulated.

Biofilm regulation has been linked to sRNAs and proteins which regulate gene transcription and protein translation. Both sRNAs and proteins have been linked to biofilm regulation in B. henselae Houston-1 [98, 128]. RNAs play a pivotal role in bacteria; RNA regulators are ubiquitous, often well conserved and can exceed the number and diversity of protein regulators [237]. Understanding RNA function will elucidate the pathway for bacterial pathogenesis and survival like biofilm formation. Small regulatory RNAs (sRNAs) are now considered to be significant gene expression regulators in bacteria after transcription. Their importance in probably all bacterial species is related to their variety as well as the diversity of the target genes physiological functions.

In chapter three, the impact of Brt1, an sRNA with a 3' terminating loop on trp1, a TF linked as a positive regulator of badA was studied. trp1 was minimally transcribed in B. henselae

Houston-1 under the conditions used for RNA-seq analysis by Tu et al [128]. Trp1 is found 15nt downstream of a highly expressed Brt1 (Fig 1.1a, Fig 1.2b). The proximity of trp1 and brt1 and the lack of an open reading frame/ internal promoter in brt1 dismisses the probability of trp1 bearing a native promoter bolstering the idea that trp1 is transcribed from the brt1 promoter. The evidence that brt1 is highly transcribed and the downstream gene is barely transcribed suggests that the stem-loop at the end of the Brt1 RNA (Fig 1.1a), is responsible for a premature transcript termination preventing the downstream trp gene from being transcribed (Fig 1.1c). To resolve this hypothesis, the brt promoter and gene with and without the 3' stem-loop was cloned upstream of the gfp gene (Fig 3.1a). gfp transcription was driven from the brt1 promoter and we studied the impact of the 3' loop on gfp. In Fig 3.2a, we noticed that in B. henselae Houston-1/pNS2 brt1Δ3'- gfp which lacks the 3' stem-loop, gfp is transcribed and GFP expression is observed microscopically. B. henselae Houston-1/pNS2 brt1-gfp, a clone which harbors the full-length Brt1

terminated the transcription of gfp while the control B. henselae Houston-1/pns-gfp was unable to

transcribe gfp because it lacked a promoter. Using a flow cytometer to confirm Fig 3.2a

microscopy images, we observed that GFP intensity in B. henselae Houston-1/pNS2 brt1Δ3'-gfp

was about 30x more intense than B. henselae Houston-1/pNS2 brt1-gfp and B. henselae Houston-

1/pNS2-gfp. There was no statistical difference between GFP expression in B. henselae Houston-

1/pNS2 brt1-gfp and B. henselae Houston-1/pNS-gfp. To reconfirm this, the 3' region of Brt1 was

deleted to create B. henselae Houston-1 Brt1Δ3' (Fig 3.3a). Deletion of the terminating region

increased aggregation, biofilm formation and an upregulation of the trp1 and badA expression (Fig

3.3c, Fig 3.4b).

The trp1 gene found downstream of this terminating region codes for a HTH-XRE-TF

capable of binding and regulating gene transcription. A previous publication presented evidence that trp1 is a positive regulator of badA; overexpression of trp1 results in an increase in badA transcription [128]. In chapter four, we examined the relationship between trp1 and badA. As trp1 is a TF, we wanted to see if it had a direct relationship with badA as a gene regulator. Trp1 protein was His-tagged and purified using a pET28a plasmid in E.coli. The promoter region of badA was amplified, biotinylated, incubated with Trp1 and subjected to electrophoresis. EMSA, a retardation assay was performed and a shift in mobility was noted. In Fig 4.1, lane 1, the biotinylated badA promoter region migrated quicker in contrast to the slow mobility of the band in lane 2 which contains the biotinylated badA promoter region and Trp1 protein. A discernible shift is observed confirming that badA putative promoter is bound to Trp1 increasing the size of the band. This increase in size causes the band to migrate slowly. In lane 3, in the presence of a competitive inhibitor; the intensity of the migrated band is reduced confirming that the competitive inhibitor, a nonbiotinylated DNA, is able to bind and displace the biotinylated DNA. To confirm that Trp1

is able to bind the promoter region of badA only, a 352 bp amplicon of the 3' coding region of badA was biotinylated (lane 4) and incubated with Trp1 and no band shift is noted (lane 5).

Furthermore, incubation of Trp1, badA promoter region and the biotinylated coding region of badA

did not produce a badA promoter region displacement (lane 6). If Trp1 is responsible for binding

badA, and badA is responsible for biofilm formation, then an increase in both gene expression

should be observed in a biofilm. To further confirm this theory, RT-qPCR data in fig. 4.2 shows that in comparison to planktonic cells, biofilm cells express more trp1 and badA. it is noteworthy

to point out that since trp1/badA is needed for the first stages of biofilm formation, the expression

decreases over time once a biofilm is formed [98]. This is likely because of the anchoring abilities

of the EPS. Once a biofilm is formed, the community uses the EPS to anchor colonies to the surface

eliminating the need for the BadA adhesin. Typically, some genes have more than 1 transcription

factor. Fig 4.3 shows that five other proteins are capable of binding the badA promoter region. Of

notable interest is Trp3; a member of the 9 copies of the trp gene family and a HTH-XRE-TF.

Clpb, a stress-induced multi-chaperone protein that binds possibly to help stabilize interactions

between mRNAs and proteins was also observed. To summarize chapter four, Trp1 is a positive

regulator which binds the badA promoter region to regulate gene expression.

Regulatory RNAs may present as small trans-encoded RNAs (sRNAs) regulating one or

more target genes located elsewhere on the chromosome, or as cis-encoded antisense RNAs

(asRNAs) that overlap and complement their target genes encoded on the opposite genomic locus

DNA strand as reviewed in Sesto et al. (2013) [151]. The regulatory function of non-coding RNAs is often closely linked to RNases activity that either promotes specific transcript cleavage and degradation or prevents RNase transcript recognition and degradation of the transcript increasing the stability of the target RNA in the cell [272]. A knockdown antisense strain of Brt1 showed

increased biofilm formation and upregulation of badA [128]. Deletion of the brt1 presented

increased biofilm formation (Fig 5.1b) and upregulation of trp1 and badA confirming that in B.

henselae Houston-1, brt1 is a negative regulator of trp1 and badA (Fig 5.2b). In contrast to the

wildtype B. henselae Houston-1, the viability of cells within the B. henselae Houston-1Δbrt1

biofilm is severely reduced (Fig 5.2a). We observe that a biotinylated Brt1 pulled down BadA mRNA amongst others. Since badA and Brt1 are located in different areas of the genomes, Brt1 may also function as antisense trans-regulating RNA. It is also noted that Brt1 is able to regulate other genes related to surface adhesion, metabolism, stress response and RNA synthesis. With most of the genes previously implicated in biofilm formation. In summary, Brt1 is a negative regulator which may function in cis to terminate trp1 transcription or in trans to regulate BadA mRNA.

6.1 Potential treatment options/drug targets

Environmental changes influence virulence gene expression, and pathogens are acquiring resistance or developing tolerance to antibiotics. Antibiotic resistance and tolerance in the treatment of bacterial infections is a serious and growing problem. Antibiotic resistance is an acquired genetic change in the bacteria while antibiotic tolerance is the ability of the otherwise susceptible bacteria to survive transient antibiotic exposure typically due to the protection conferred by a biofilm. A recent publication estimates >10,000,000 deaths per year associated with antibiotic-resistant and tolerant infections [273]. Most clinically approved antibiotics target growth/integrity of bacterial cell walls, translation, and replication/segregation of DNA, while

RNAs, transcription process and factors appear to be an underused target. New drug targets are required to help combat antibiotics resistance. Transcription is the process by which the enzyme

RNA polymerase (RNAP) synthesizes RNA from a template DNA. There are currently two

antibiotics that target bacterial RNAP on the market: rifamycin which binds RNAP to hinder the

growth of RNA product [274] and fidaxomicin/lipiarmicin speculated to bind the switch region to

prevent the formation of transcription process [275, 276]. As reviewed by Cong et al. (2016) [277],

bacterial transcription is an excellent target for novel antibacterial development for the following

reasons: (i) transcription is an essential process for cell viability; (ii) bacterial RNAP and the

associated transcription factors are highly conserved, allowing for the potential development of

broad-spectrum anti-transcriptional antibiotics [278]; (iii) eukaryotic RNAP is not similar to the

bacterial homolog at the sequence level suggesting low potential cytotoxicity [279]. This novel

idea disrupts the transcription process and is an untapped area for bacterial disease treatments. All

of the regulatory genes studied here may be used as a potential drug target.

6.1.1. RNA targets

Anti-infective RNA based drugs have gained significant attention mostly because RNAs play multifunctional roles that regulate disease infection and progression at multiple levels. In mammalian cells, antisense oligonucleotides that target mRNA by complementary base pairing recognition have been used to treat familial hypercholesterolemia [280]. Aminoglycoside antibiotics are oligosaccharides containing several ammonium groups and bind the aminoacyl- transfer RNA (tRNA) decoding site (A site) on the 16S rRNA [281, 282]. Indeed, several compounds have been developed that kill bacteria by mimicking ligands for riboswitches controlling essential genes, demonstrating that regulatory RNA elements are druggable targets

[283]. Some antibiotics target pockets that have the ability to bind specific ligands and RNA can fold into 3D structures with binding sites for specific ligands [284]. Literature also shows that the ribosome-antibiotic complex formed for drugs that target the ribosome, bind rRNAs and not ribosomal protein [285]. Since the first discovery of asRNAs in prokaryotes, antisense

oligonucleotides have been used as drug targets when Zamecnik and Stephenson found one that

could inhibit viral replication and protein synthesis in the viral RNA of the Rous sarcoma virus in

1978 [286]. The first FDA approved drug, fomivirsen, was developed for the treatment of

cytomegalovirus retinitis in patients with AIDS.

6.1.2 Adhesins

In light of the resurgence of infectious diseases, membrane proteins are prime candidates

for the development of urgently needed novel anti-infectives [287]. One of the first steps during

bacterial infection is adhesion of the pathogen to host cells. Adhesion interference is, therefore, an effective way of preventing or treating bacterial infections. Small peptides are being developed to target P. aeruginosa adhesion [288, 289].

6.1.3 Transcription factors

TFs are also good integration points for multiple signaling pathways. They control a plethora of pathogenic and virulence factors and genes in bacteria. Recent work in drug discovery has shown that TFs are susceptible to drug inhibition and are being employed to inhibit downstream signaling [290]. While this is shown to lack specificity mostly because of redundancy in the signaling cascade, some have been shown to disrupt protein-protein and protein–DNA interactions in vitro.

In conclusion, we have developed a hypothetical model of biofilm regulation in B. henselae

Houston-1 (Fig 6.1). We speculate that B. henselae Houston-1 cells undergo an initial adherence step within the first 8-24 hours after contact with a surface as supported by fig 2.1a. This initial contact releases a signal sensed by the brt1 TF which binds the brt1 promoter region. The Brt1 and Trp1 are transcribed as a full-length mRNA evidenced by the increased expression of trp1 in

a biofilm in 24-48 hours (Fig 4.2). Trp1 is translated and binds the badA promoter region to activate gene badA transcription (Fig 4.1). BadA expression, in turn, promotes aggregation and an

irreversible surface adhesion (Fig 2.1b-c). Once aggregation is optimal, we speculate that a

different signal, perhaps the auto inducer responsible for quorum sensing may bind to Brt1 to

induce the formation of the 3' stem-loop initiating Trp1 transcript termination. This is supported

by fig.4.2 where we observe a decrease in Trp1 from day 3: from fig 2.1, a biofilm is already

formed by day 3(72 hours). Brt1 is transcribed as a 3' stem-loop RNA capable of exerting trans-

acting RNA abilities on BadA to target the badA mRNA for degradation (Fig 4.2-badA expression decreases) and on other mRNAs required for stress response, RNA stability, EPS metabolism, etc.

(Table 5.2). EPS is synthesized and biofilm is formed. This brt/trp/badA regulation is minimally

functional in the C.felis arthropod vector signifying that perhaps a different outer membrane may

serve as the adhesin for biofilm formation (Fig 5.4). Data presented within this study are the first

to identify and experimentally characterize an adhesin, transcriptional regulator and RNA

responsible for biofilm formation in B. henselae.

Fig 6.1: A hypothesized model for biofilm regulation by B. henselae. B. henselae undergoes an initial adhesion stage which signals for the transcription of Brt1. Brt1 and Trp1 are transcribed from the Brt1 promoter. Trp1 activates badA transcription and BadA helps aggregate and adhere cells to a surface. Once aggregation reaches the optimum level, a feedback loop is generated and a stimulus binds Brt1 causing the 3' loop. This 3' terminating loop prevents Trp1 transcription. Brt1 binds BadA mRNA to degrade it while binding the mRNA of other genes to either degrade or promote stability. We predict that metabolic genes are positively regulated by Brt leading to EPS synthesis and biofilm formation.

References

1. Raoult, D., et al., Outbreak of epidemic typhus associated with trench fever in Burundi. The Lancet, 1998. 352(9125): p. 353-358. 2. Ohl, M.E. and D.H. Spach, Bartonella quintana and urban trench fever. Clin Infect Dis, 2000. 31(1): p. 131-5. 3. Comer, J.A., et al., Antibodies to Bartonella species in inner-city intravenous drug users in Baltimore, Md. Arch Intern Med, 1996. 156(21): p. 2491-5. 4. Vinson, J.W., In vitro cultivation of the rickettsial agent of trench fever. Bulletin of the World Health Organization, 1966. 35(2): p. 155-164. 5. Okaro, U., et al., Bartonella Species, an Emerging Cause of Blood-Culture-Negative Endocarditis. Clinical Microbiology Reviews, 2017. 30(3): p. 709-746. 6. Sato, S., et al., Bartonella jaculi sp. nov., Bartonella callosciuri sp. nov., Bartonella pachyuromydis sp. nov. and Bartonella acomydis sp. nov., isolated from wild Rodentia. Int J Syst Evol Microbiol, 2013. 63(Pt 5): p. 1734-40. 7. Gray, G.C., et al., An epidemic of Oroya fever in the Peruvian Andes. Am J Trop Med Hyg, 1990. 42(3): p. 215-21. 8. Bermond, D., et al., Bartonella birtlesii sp. nov., isolated from small mammals (Apodemus spp.). Int J Syst Evol Microbiol, 2000. 50 Pt 6: p. 1973-9. 9. Renaud Maillard, P.R., Francine Barrat, Corinne Bouillin Danielle Thibault1, Christelle Gandoin, Lénaig Halos1, Christine Demanche, Annie Alliot, Jacques Guillot, Yves Piémont, Henri-Jean Boulouis, Muriel Vayssier-Taussat, Bartonella chomelii sp. nov., isolated from French domestic cattle (Bos taurus). International Journal of Systematic and Evolutionary Microbiology, 2004. 51(1): p. 215-220. 10. Kordick, D.L., et al., Bartonella clarridgeiae, a newly recognized zoonotic pathogen causing inoculation papules, fever, and lymphadenopathy (cat scratch disease). J Clin Microbiol, 1997. 35(7): p. 1813-8. 11. Chomel, B.B., A.C. Wey, and R.W. Kasten, Isolation of Bartonella washoensis from a Dog with Mitral Valve Endocarditis. Journal of Clinical Microbiology, 2003. 41(11): p. 5327-5332. 12. Regnery, R.L., B.E. Anderson, J.E. Clarridge, M. Rodriquez-Barradas, D.C. Jones, and J.H. Carr., Characterization of a novel Rochalimaea species, R. henselae sp. nov., isolated from blood of a febrile, human immunodeficiency virus-positive patient. . J Clin Microbiol, 1992. 30: p. 265-274. 13. Koehler, J.E., et al., Isolation of Rochalimaea species from cutaneous and osseous lesions of bacillary angiomatosis. N Engl J Med, 1992. 327(23): p. 1625-31. 14. Regnery, R.L., et al., Serological response to "Rochalimaea henselae" antigen in suspected cat-scratch disease. Lancet, 1992. 339(8807): p. 1443-5. 15. Anderson, B., et al., Detection of Rochalimaea henselae in cat-scratch disease skin test antigens. J Infect Dis, 1993. 168(4): p. 1034-6. 16. Jeanclaude, D., et al., Bartonella alsatica endocarditis in a French patient in close contact with rabbits. Clinical Microbiology and Infection, 2009. 15, Supplement 2: p. 110-111.

17. Lin, E.Y., et al., Candidatus Bartonella mayotimonensis and endocarditis. Emerg Infect Dis, 2010. 16(3): p. 500-3. 18. Huarcaya, E., et al., Bartonelosis (Carrion's Disease) in the pediatric population of Peru: an overview and update. Brazilian Journal of Infectious Diseases, 2004. 8: p. 331-339. 19. Margileth, A.M. and D.F. Baehren, Chest-wall abscess due to cat-scratch disease (CSD) in an adult with antibodies to Bartonella clarridgeiae: case report and review of the thoracopulmonary manifestations of CSD. Clin Infect Dis, 1998. 27(2): p. 353-7. 20. Kerkhoff, F.T., et al., Demonstration of Bartonella grahamii DNA in ocular fluids of a patient with neuroretinitis. J Clin Microbiol, 1999. 37(12): p. 4034-8. 21. Anderson, B.E. and M.A. Neuman, Bartonella spp. as emerging human pathogens. Clin Microbiol Rev, 1997. 10(2): p. 203-19. 22. HERTIG, M.P., Phlebotomus and Carrion’s Disease. Am. J. Trop. Med., 1942. 23. Roden, J.A., et al., Hemin binding protein C is found in outer membrane vesicles and protects Bartonella henselae against toxic concentrations of hemin. Infect Immun, 2012. 80(3): p. 929-42. 24. Chomel, B.B., et al., Isolation of Bartonella henselae and Two New Bartonella Subspecies, Bartonella koehlerae Subspecies boulouisii subsp. nov. and Bartonella koehlerae Subspecies bothieri subsp. nov. from Free-Ranging Californian Mountain Lions and Bobcats. PLoS ONE, 2016. 11(3): p. e0148299. 25. Relman, D.A., et al., The agent of bacillary angiomatosis. An approach to the identification of uncultured pathogens. N Engl J Med, 1990. 323(23): p. 1573-80. 26. Bergmans, A.M., et al., Pitfalls and fallacies of cat scratch disease serology: evaluation of Bartonella henselae-based indirect fluorescence assay and enzyme-linked immunoassay. J Clin Microbiol, 1997. 35(8): p. 1931-7. 27. Anderson, B., et al., Characterization of a 17-kilodalton antigen of Bartonella henselae reactive with sera from patients with cat scratch disease. J Clin Microbiol, 1995. 33(9): p. 2358-65. 28. Minnick, M.F. and B.E. Anderson, Chapter 105 - Bartonella, in Molecular Medical Microbiology (Second Edition), Y.-W. Tang, et al., Editors. 2015, Academic Press: Boston. p. 1911-1939. 29. Lepidi, H., P.E. Fournier, and D. Raoult, Quantitative analysis of valvular lesions during Bartonella endocarditis. Am J Clin Pathol, 2000. 114(6): p. 880-9. 30. Litwin, C.M., B. Anderson, R. Tsolis, A. Rasley., The Bartonellaceae, Brucellaceae, and Francisellaceae, in Manual of Molecular and Clinical Laboratory Immunology. 2016, ASM Press: Washington, DC. p. 473-481. 31. Engel, P., et al., Parallel evolution of a type IV secretion system in radiating lineages of the host-restricted bacterial pathogen Bartonella. PLoS Genet, 2011. 7(2): p. e1001296. 32. Engel, P. and C. Dehio, Genomics of Host-Restricted Pathogens of the Genus Bartonella. 2009. 33. Berglund, E.C., et al., Run-off replication of host-adaptability genes is associated with gene transfer agents in the genome of mouse-infecting Bartonella grahamii. PLoS Genet, 2009. 5(7): p. e1000546. 34. Chomel, B.B., et al., Experimental transmission of Bartonella henselae by the cat flea. J Clin Microbiol, 1996. 34(8): p. 1952-6. 35. Higgins, J.A., et al., Acquisition of the cat scratch disease agent Bartonella henselae by cat fleas (Siphonaptera:Pulicidae). J Med Entomol, 1996. 33(3): p. 490-5.

36. Karem, K.L., C.D. Paddock, and R.L. Regnery, Bartonella henselae, B. quintana, and B. bacilliformis: historical pathogens of emerging significance. Microbes and Infection, 2000. 2(10): p. 1193-1205. 37. Chomel, B.B., et al., Ecological fitness and strategies of adaptation of Bartonella species to their hosts and vectors. Vet Res, 2009. 40(2): p. 29. 38. Demers, D.M., et al., Cat-scratch disease in Hawaii: etiology and seroepidemiology. J Pediatr, 1995. 127(1): p. 23-6. 39. Chomel, B.B., Cat-scratch disease. Rev Sci Tech, 2000. 19(1): p. 136-50. 40. Foil, L., et al., Experimental infection of domestic cats with Bartonella henselae by inoculation of Ctenocephalides felis (Siphonaptera: Pulicidae) feces. J Med Entomol, 1998. 35(5): p. 625-8. 41. Koehler, J.E., C.A. Glaser, and J.W. Tappero, Rochalimaea henselae infection. A new zoonosis with the domestic cat as reservoir. Jama, 1994. 271(7): p. 531-5. 42. Jackson, L.A., B.A. Perkins, and J.D. Wenger, Cat scratch disease in the United States: an analysis of three national databases. Am J Public Health, 1993. 83(12): p. 1707-11. 43. Gutiérrez, R., Y. Nachum-Biala, and S. Harrus, Relationship between the Presence of Bartonella Species and Bacterial Loads in Cats and Cat Fleas (Ctenocephalides felis) under Natural Conditions. Applied and environmental microbiology, 2015. 81(16): p. 5613-5621. 44. Lappin, M.R. and J. Hawley, Presence of Bartonella species and Rickettsia species DNA in the blood, oral cavity, skin and claw beds of cats in the United States. Veterinary Dermatology, 2009. 20(5-6): p. 509-514. 45. Seki, N., et al., Quantitative analysis of proliferation and excretion of Bartonella quintana in body lice, Pediculus humanus L. Am J Trop Med Hyg, 2007. 77(3): p. 562-6. 46. Fournier, P.E., et al., Experimental model of human body louse infection using green fluorescent protein-expressing Bartonella quintana. Infect Immun, 2001. 69(3): p. 1876-9. 47. Finkelstein, J.L., et al., Studies on the growth of Bartonella henselae in the cat flea (Siphonaptera: Pulicidae). J Med Entomol, 2002. 39(6): p. 915-9. 48. Fournier, P.E., et al., Epidemiologic and clinical characteristics of Bartonella quintana and Bartonella henselae endocarditis: a study of 48 patients. Medicine (Baltimore), 2001. 80(4): p. 245-51. 49. Barbier, F., et al., Bartonella quintana coinfection in Staphylococcus aureus endocarditis: usefulness of screening in high-risk patients? Clin Infect Dis, 2009. 48(9): p. 1332-3. 50. Tattevin, P., et al., Update on blood culture-negative endocarditis. Med Mal Infect, 2015. 45(1-2): p. 1-8. 51. Rolain, J.M., et al., Molecular detection of Bartonella quintana, B. koehlerae, B. henselae, B. clarridgeiae, Rickettsia felis, and Wolbachia pipientis in cat fleas, France. Emerg Infect Dis, 2003. 9(3): p. 338-42. 52. Florin, T.A., T.E. Zaoutis, and L.B. Zaoutis, Beyond cat scratch disease: widening spectrum of Bartonella henselae infection. Pediatrics, 2008. 121(5): p. e1413-25. 53. Nelson, C.A., S. Saha, and P.S. Mead, Cat-Scratch Disease in the United States, 2005- 2013. Emerg Infect Dis, 2016. 22(10): p. 1741-6. 54. Carithers, H.A., Oculoglandular disease of parinaud. A manifestation of cat-scratch disease. Am J Dis Child, 1978. 132(12): p. 1195-200. 55. Aziz, H.A., et al., Cat Scratch Disease: Expanded Spectrum. Ocul Oncol Pathol, 2016. 2(4): p. 246-250.

56. Chomel, B.B., et al., Bartonella henselae prevalence in domestic cats in California: risk factors and association between bacteremia and antibody titers. J Clin Microbiol, 1995. 33(9): p. 2445-50. 57. Bergmans, A.M., et al., Prevalence of Bartonella species in domestic cats in The Netherlands. J Clin Microbiol, 1997. 35(9): p. 2256-61. 58. Hercik, K., et al., Molecular evidence of Bartonella DNA in ixodid ticks in Czechia. Folia Microbiol (Praha), 2007. 52(5): p. 503-9. 59. Spach, D.H., Bacillary angiomatosis. Int J Dermatol, 1992. 31(1): p. 19-24. 60. Cockerell, C.J., et al., Bacillary epithelioid angiomatosis occurring in an immunocompetent individual. Arch Dermatol, 1990. 126(6): p. 787-90. 61. Garcia, F.U., et al., Bartonella bacilliformis stimulates endothelial cells in vitro and is angiogenic in vivo. Am J Pathol, 1990. 136(5): p. 1125-35. 62. Cockerell, C.J. and P.E. LeBoit, Bacillary angiomatosis: a newly characterized, pseudoneoplastic, infectious, cutaneous vascular disorder. J Am Acad Dermatol, 1990. 22(3): p. 501-12. 63. Kempf, V.A., et al., Evidence of a leading role for VEGF in Bartonella henselae-induced endothelial cell proliferations. Cell Microbiol, 2001. 3(9): p. 623-32. 64. Kirby, J.E. and D.M. Nekorchuk, Bartonella-associated endothelial proliferation depends on inhibition of apoptosis. Proc Natl Acad Sci U S A, 2002. 99(7): p. 4656-61. 65. Kirby, J.E., In vitro model of Bartonella henselae-induced angiogenesis. Infect Immun, 2004. 72(12): p. 7315-7. 66. Kemper, C.A., et al., Visceral bacillary epithelioid angiomatosis: possible manifestations of disseminated cat scratch disease in the immunocompromised host: a report of two cases. Am J Med, 1990. 89(2): p. 216-22. 67. LeBoit, P.E., The expanding spectrum of a new disease, bacillary angiomatosis. Arch Dermatol, 1990. 126(6): p. 808-11. 68. Harms, A. and C. Dehio, Intruders below the radar: molecular pathogenesis of Bartonella spp. Clin Microbiol Rev, 2012. 25(1): p. 42-78. 69. Schulein, R., et al., Invasion and persistent intracellular colonization of erythrocytes. A unique parasitic strategy of the emerging pathogen Bartonella. J Exp Med, 2001. 193(9): p. 1077-86. 70. Cartwright, J.L., et al., The IalA invasion gene of Bartonella bacilliformis encodes a (de)nucleoside polyphosphate hydrolase of the MutT motif family and has homologs in other invasive bacteria. Biochem Biophys Res Commun, 1999. 256(3): p. 474-9. 71. Mitchell, S.J. and M.F. Minnick, Characterization of a two-gene locus from Bartonella bacilliformis associated with the ability to invade human erythrocytes. Infect Immun, 1995. 63(4): p. 1552-62. 72. Conyers, G.B. and M.J. Bessman, The gene, ialA, associated with the invasion of human erythrocytes by Bartonella bacilliformis, designates a nudix hydrolase active on dinucleoside 5'-polyphosphates. J Biol Chem, 1999. 274(3): p. 1203-6. 73. Minnick, M.F. and J.M. Battisti, Pestilence, persistence and pathogenicity: infection strategies of Bartonella. Future Microbiol, 2009. 4(6): p. 743-58. 74. Kosoy, M.Y., et al., Isolation of Bartonella spp. from embryos and neonates of naturally infected rodents. J Wildl Dis, 1998. 34(2): p. 305-9. 75. Maggi, R.G., et al., Bartonella henselae in captive and hunter-harvested beluga (Delphinapterus leucas). J Wildl Dis, 2008. 44(4): p. 871-7.

76. Valentine, K.H., et al., Bartonella DNA in loggerhead sea turtles. Emerg Infect Dis, 2007. 13(6): p. 949-50. 77. Chang, C.C., et al., Bartonella spp. isolated from wild and domestic ruminants in North America. Emerg Infect Dis, 2000. 6(3): p. 306-11. 78. Vinson, J.W., G. Varela, and C. Molina-Pasquel, Trench fever. 3. Induction of clinical disease in volunteers inoculated with Rickettsia quintana propagated on blood agar. Am J Trop Med Hyg, 1969. 18(5): p. 713-22. 79. Rolain, J.M., et al., Bartonella quintana in human erythrocytes. Lancet, 2002. 360(9328): p. 226-8. 80. Kordick, D.L. and E.B. Breitschwerdt, Intraerythrocytic presence of Bartonella henselae. J Clin Microbiol, 1995. 33(6): p. 1655-6. 81. Koesling, J., et al., Cutting edge: antibody-mediated cessation of hemotropic infection by the intraerythrocytic mouse pathogen Bartonella grahamii. J Immunol, 2001. 167(1): p. 11-4. 82. Chomel, B.B., et al., Clinical impact of persistent Bartonella bacteremia in humans and animals. Ann N Y Acad Sci, 2003. 990: p. 267-78. 83. Breitschwerdt, E.B. and D.L. Kordick, Bartonella infection in animals: carriership, reservoir potential, pathogenicity, and zoonotic potential for human infection. Clin Microbiol Rev, 2000. 13(3): p. 428-38. 84. Jacomo, V., P.J. Kelly, and D. Raoult, Natural history of Bartonella infections (an exception to Koch's postulate). Clin Diagn Lab Immunol, 2002. 9(1): p. 8-18. 85. Zbinden, R., M. Hochli, and D. Nadal, Intracellular location of Bartonella henselae cocultivated with Vero cells and used for an indirect fluorescent-antibody test. Clin Diagn Lab Immunol, 1995. 2(6): p. 693-5. 86. Rolain, J.M., et al., Culture and antibiotic susceptibility of Bartonella quintana in human erythrocytes. Antimicrob Agents Chemother, 2003. 47(2): p. 614-9. 87. Maurin, M., et al., MICs of 28 antibiotic compounds for 14 Bartonella (formerly Rochalimaea) isolates. Antimicrob Agents Chemother, 1995. 39(11): p. 2387-91. 88. Rolain, J.M., et al., Recommendations for treatment of human infections caused by Bartonella species. Antimicrob Agents Chemother, 2004. 48(6): p. 1921-33. 89. Musso, D., M. Drancourt, and D. Raoult, Lack of bactericidal effect of antibiotics except aminoglycosides on Bartonella (Rochalimaea) henselae. J Antimicrob Chemother, 1995. 36(1): p. 101-8. 90. Ives, T.J., et al., In vitro susceptibilities of Bartonella henselae, B. quintana, B. elizabethae, Rickettsia rickettsii, R. conorii, R. akari, and R. prowazekii to macrolide antibiotics as determined by immunofluorescent-antibody analysis of infected Vero cell monolayers. Antimicrob Agents Chemother, 1997. 41(3): p. 578-82. 91. Rolain, J.M., M. Maurin, and D. Raoult, Bactericidal effect of antibiotics on Bartonella and Brucella spp.: clinical implications. J Antimicrob Chemother, 2000. 46(5): p. 811-4. 92. Raoult, D., et al., Outcome and treatment of Bartonella endocarditis. Arch Intern Med, 2003. 163(2): p. 226-30. 93. Elliott, T.S., et al., Guidelines for the antibiotic treatment of endocarditis in adults: report of the Working Party of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother, 2004. 54(6): p. 971-81.

94. Gould, F.K., et al., Guidelines for the diagnosis and antibiotic treatment of endocarditis in adults: a report of the Working Party of the British Society for Antimicrobial Chemotherapy. J Antimicrob Chemother, 2012. 67(2): p. 269-89. 95. Murdoch, D.R., et al., Clinical presentation, etiology, and outcome of infective endocarditis in the 21st century: the International Collaboration on Endocarditis- Prospective Cohort Study. Arch Intern Med, 2009. 169(5): p. 463-73. 96. Pendle, S., A. Ginn, and J. Iredell, Antimicrobial susceptibility of Bartonella henselae using Etest methodology. J Antimicrob Chemother, 2006. 57(4): p. 761-3. 97. Delahaye, F., et al., Indications and optimal timing for surgery in infective endocarditis. Heart, 2004. 90(6): p. 618. 98. Okaro, U., et al., The trimeric autotransporter adhesin BadA is required for in vitro biofilm formation by Bartonella henselae. npj Biofilms and Microbiomes, 2019. 5(1): p. 10. 99. Monroe, D., Looking for chinks in the armor of bacterial biofilms. PLoS biology, 2007. 5(11): p. e307-e307. 100. Riess, T., et al., Analysis of Bartonella adhesin A expression reveals differences between various B. henselae strains. Infect Immun, 2007. 75(1): p. 35-43. 101. Batterman, H.J., et al., Bartonella henselae and Bartonella quintana adherence to and entry into cultured human epithelial cells. Infect Immun, 1995. 63(11): p. 4553-6. 102. Szczesny, P., et al., Structure of the Head of the Bartonella Adhesin BadA. PLOS Pathogens, 2008. 4(8): p. e1000119. 103. Capecchi, B., et al., Neisseria meningitidis NadA is a new invasin which promotes bacterial adhesion to and penetration into human epithelial cells. Mol Microbiol, 2005. 55(3): p. 687-98. 104. Hoiczyk, E., et al., Structure and sequence analysis of Yersinia YadA and Moraxella UspAs reveal a novel class of adhesins. EMBO J, 2000. 19(22): p. 5989-99. 105. Linke, D., et al., Trimeric autotransporter adhesins: variable structure, common function. Trends Microbiol, 2006. 14(6): p. 264-70. 106. Schmid, M.C., et al., The VirB type IV secretion system of Bartonella henselae mediates invasion, proinflammatory activation and antiapoptotic protection of endothelial cells. Mol Microbiol, 2004. 52(1): p. 81-92. 107. Kaiser, P.O., et al., Bartonella spp.: throwing light on uncommon human infections. Int J Med Microbiol, 2011. 301(1): p. 7-15. 108. Seubert, A., R. Schulein, and C. Dehio, Bacterial persistence within erythrocytes: a unique pathogenic strategy of Bartonella spp. Int J Med Microbiol, 2002. 291(6-7): p. 555-60. 109. Alsmark, C.M., et al., The louse-borne human pathogen Bartonella quintana is a genomic derivative of the zoonotic agent Bartonella henselae. Proc Natl Acad Sci U S A, 2004. 101(26): p. 9716-21. 110. Quebatte, M., et al., The BatR/BatS two-component regulatory system controls the adaptive response of Bartonella henselae during human endothelial cell infection. J Bacteriol, 2010. 192(13): p. 3352-67. 111. Dehio, C., Infection-associated type IV secretion systems of Bartonella and their diverse roles in host cell interaction. Cell Microbiol, 2008. 10(8): p. 1591-8. 112. Backert, S. and M. Clyne, Pathogenesis of Helicobacter pylori infection. Helicobacter, 2011. 16 Suppl 1: p. 19-25.

100

113. Nagai, H. and T. Kubori, Type IVB Secretion Systems of Legionella and Other Gram- Negative Bacteria. Front Microbiol, 2011. 2: p. 136. 114. Schulein, R. and C. Dehio, The VirB/VirD4 type IV secretion system of Bartonella is essential for establishing intraerythrocytic infection. Mol Microbiol, 2002. 46(4): p. 1053-67. 115. Schulein, R., et al., A bipartite signal mediates the transfer of type IV secretion substrates of Bartonella henselae into human cells. Proc Natl Acad Sci U S A, 2005. 102(3): p. 856- 61. 116. El Tahir, Y. and M. Skurnik, YadA, the multifaceted Yersinia adhesin. Int J Med Microbiol, 2001. 291(3): p. 209-18. 117. Singh, B., et al., Haemophilus influenzae surface fibril (Hsf) is a unique twisted hairpin- like trimeric autotransporter. Int J Med Microbiol, 2015. 305(1): p. 27-37. 118. Cotter, S.E., et al., Trimeric autotransporters require trimerization of the passenger domain for stability and adhesive activity. J Bacteriol, 2006. 188(15): p. 5400-7. 119. Surana, N.K., et al., The Haemophilus influenzae Hia autotransporter contains an unusually short trimeric translocator domain. J Biol Chem, 2004. 279(15): p. 14679-85. 120. O'Rourke, F., et al., Adhesins of Bartonella spp. Adv Exp Med Biol, 2011. 715: p. 51-70. 121. Kaiser, P.O., et al., The head of Bartonella adhesin A is crucial for host cell interaction of Bartonella henselae. Cell Microbiol, 2008. 10(11): p. 2223-34. 122. Bialas, N., et al., Bacterial cell surface structures in Yersinia enterocolitica. Arch Immunol Ther Exp (Warsz), 2012. 60(3): p. 199-209. 123. Gilmore, R.D., et al., The Bartonella vinsonii subsp. arupensis Immunodominant Surface Antigen BrpA Gene, Encoding a 382-Kilodalton Protein Composed of Repetitive Sequences, Is a Member of a Multigene Family Conserved among Bartonella Species. Infect Immun, 2005. 73(5): p. 3128-36. 124. Riess, T., et al., Bartonella adhesin a mediates a proangiogenic host cell response. J Exp Med, 2004. 200(10): p. 1267-78. 125. Kaiser, P.O., et al., Analysis of the BadA stalk from Bartonella henselae reveals domain- specific and domain-overlapping functions in the host cell infection process. Cell Microbiol, 2012. 14(2): p. 198-209. 126. Tu, N., et al., A family of genus-specific RNAs in tandem with DNA-binding proteins control expression of the badA major virulence factor gene in Bartonella henselae. Microbiologyopen, 2016. 127. Tu, N., et al., Characterization of the general stress response in Bartonella henselae. Microb Pathog, 2016. 92: p. 1-10. 128. Nhan Tu, R.K.C., Andy Weiss, Lindsey Shaw, Gael Nicolas, Sarah Thomas, Amorce Lima, Udoka Okaro and Burt Anderson, A family of genus-specific RNAs in tandem with DNA-binding proteins control expression of the badA major virulence factor gene in Bartonella henselae. Open Microbilogy, 2016. 129. Saenz, H.L., et al., Genomic analysis of Bartonella identifies type IV secretion systems as host adaptability factors. Nat Genet, 2007. 39(12): p. 1469-76. 130. Dehio, C., Molecular and cellular basis of bartonella pathogenesis. Annu Rev Microbiol, 2004. 58: p. 365-90. 131. Vayssier-Taussat, M., et al., The Trw type IV secretion system of Bartonella mediates host-specific adhesion to erythrocytes. PLoS Pathog, 2010. 6(6): p. e1000946.

101

132. Rojas, C.M., et al., HecA, a member of a class of adhesins produced by diverse pathogenic bacteria, contributes to the attachment, aggregation, epidermal cell killing, and virulence phenotypes of Erwinia chrysanthemi EC16 on Nicotiana clevelandii seedlings. Proc Natl Acad Sci U S A, 2002. 99(20): p. 13142-7. 133. Franz, B. and V.A.J. Kempf, Adhesion and host cell modulation: critical pathogenicity determinants of Bartonella henselae. Parasites & Vectors, 2011. 4: p. 54-54. 134. Kang, J.G., et al., Comparative proteomic analysis of outer membrane protein 43 (omp43)-deficient Bartonella henselae. J Vet Sci, 2018. 19(1): p. 59-70. 135. Burgess, A.W. and B.E. Anderson, Outer membrane proteins of Bartonella henselae and their interaction with human endothelial cells. Microb Pathog, 1998. 25(3): p. 157-64. 136. Zhang, P., et al., A family of variably expressed outer-membrane proteins (Vomp) mediates adhesion and autoaggregation in Bartonella quintana. Proc Natl Acad Sci U S A, 2004. 101(37): p. 13630-5. 137. Donlan, R.M., Biofilms: Microbial Life on Surfaces. Emerg Infect Dis, 2002. 8(9): p. 881-90. 138. Bjarnsholt, T., The role of bacterial biofilms in chronic infections. APMIS, 2013. 121(s136): p. 1-58. 139. Totsika, M., et al., Molecular characterization of the EhaG and UpaG trimeric autotransporter proteins from pathogenic Escherichia coli. Appl Environ Microbiol, 2012. 78(7): p. 2179-89. 140. Kyme, P., B. Dillon, and J. Iredell, Phase variation in Bartonella henselae. Microbiology, 2003. 149(Pt 3): p. 621-9. 141. Bentancor, L.V., et al., Identification of Ata, a multifunctional trimeric autotransporter of Acinetobacter baumannii. J Bacteriol, 2012. 194(15): p. 3950-60. 142. Mil-Homens, D. and A.M. Fialho, Trimeric autotransporter adhesins in members of the Burkholderia cepacia complex: a multifunctional family of proteins implicated in virulence. Front Cell Infect Microbiol, 2011. 1: p. 13. 143. Zimmerman, S.M., et al., The Autotransporter BpaB Contributes to the Virulence of Burkholderia mallei in an Aerosol Model of Infection. PLoS One, 2015. 10(5): p. e0126437. 144. Valle, J., et al., UpaG, a new member of the trimeric autotransporter family of adhesins in uropathogenic Escherichia coli. J Bacteriol, 2008. 190(12): p. 4147-61. 145. Raghunathan, D., et al., SadA, a trimeric autotransporter from Salmonella enterica serovar Typhimurium, can promote biofilm formation and provides limited protection against infection. Infect Immun, 2011. 79(11): p. 4342-52. 146. Bouhsira, E., et al., Ctenocephalides felis an in vitro potential vector for five Bartonella species. Comp Immunol Microbiol Infect Dis, 2013. 36(2): p. 105-11. 147. Sutherland, I.W., The biofilm matrix – an immobilized but dynamic microbial environment. Trends in Microbiology, 2001. 9(5): p. 222-227. 148. Jackson, D.W., et al., Biofilm Formation and Dispersal under the Influence of the Global Regulator CsrA of <em>Escherichia coli</em>. Journal of Bacteriology, 2002. 184(1): p. 290. 149. Papenfort, K., et al., Specific and pleiotropic patterns of mRNA regulation by ArcZ, a conserved, Hfq-dependent small RNA. Molecular Microbiology, 2009. 74(1): p. 139-158. 150. Grundy, F.J. and T.M. Henkin, tRNA as a positive regulator of transcription antitermination in B. subtilis. Cell, 1993. 74(3): p. 475-482.

102

151. Sesto, N., et al., The excludon: a new concept in bacterial antisense RNA-mediated gene regulation. Nature Reviews Microbiology, 2012. 11: p. 75. 152. Sorek, R. and P. Cossart, Prokaryotic transcriptomics: a new view on regulation, physiology and pathogenicity. Nat Rev Genet, 2010. 11(1): p. 9-16. 153. Papenfort, K. and J. Vogel, Small RNA functions in carbon metabolism and virulence of enteric pathogens. Front Cell Infect Microbiol, 2014. 4: p. 91. 154. Lenz, D.H., et al., The small RNA chaperone Hfq and multiple small RNAs control quorum sensing in Vibrio harveyi and Vibrio cholerae. Cell, 2004. 118(1): p. 69-82. 155. Shao, Y., et al., Functional determinants of the quorum-sensing non-coding RNAs and their roles in target regulation. Embo j, 2013. 32(15): p. 2158-71. 156. Ortega, Á.D., et al., Non-coding RNA regulation in pathogenic bacteria located inside eukaryotic cells. Frontiers in Cellular and Infection Microbiology, 2014. 4(162). 157. Omasits, U., et al., Directed shotgun proteomics guided by saturated RNA-seq identifies a complete expressed prokaryotic proteome. Genome Res, 2013. 23(11): p. 1916-27. 158. Artsimovitch, I., A processive riboantiterminator seeks a switch to make biofilms. Molecular microbiology, 2010. 76(3): p. 535-539. 159. Abduljalil, J.M., Bacterial riboswitches and RNA thermometers: Nature and contributions to pathogenesis. Non-coding RNA research, 2018. 3(2): p. 54-63. 160. Kortmann, J. and F. Narberhaus, Bacterial RNA thermometers: molecular zippers and switches. Nature Reviews Microbiology, 2012. 10: p. 255. 161. Atilho, R.M., et al., A bacterial riboswitch class for the thiamin precursor HMP-PP employs a terminator-embedded aptamer. eLife, 2019. 8: p. e45210. 162. MICHAELK.RUST, J.S.A., SomeAbioticFactorsAffectingtheSurvivalof theCatFlea,Ctenocephalidesrelis(Siphonaptera:Pulicidae)1. Environ.Entomo, 1983. 12: p. 490-495. 163. de Silva, A.M. and E. Fikrig, Arthropod- and host-specific gene expression by Borrelia burgdorferi. J Clin Invest, 1997. 99(3): p. 377-9. 164. Hutchison, J.S., et al., Hypothermia therapy after traumatic brain injury in children. N Engl J Med, 2008. 358(23): p. 2447-56. 165. Battisti, J.M., et al., Environmental signals generate a differential and coordinated expression of the heme receptor gene family of Bartonella quintana. Infect Immun, 2006. 74(6): p. 3251-61. 166. Liu, M. and F. Biville, Managing iron supply during the infection cycle of a flea borne pathogen, Bartonella henselae. Frontiers in cellular and infection microbiology, 2013. 3: p. 60-60. 167. Scott, D.W., W.H. Miller, and C.E. Griffin, Chapter 1 - Structure and Function of the Skin, in Muller & Kirk's Small Animal Dermatology (Sixth Edition), D.W. Scott, W.H. Miller, and C.E. Griffin, Editors. 2001, W.B. Saunders: Philadelphia. p. 1-70. 168. Ahn Jin, K., Chapter 27 - Acidosis, in Comprehensive Pediatric Hospital Medicine, L.B. Zaoutis and V.W. Chiang, Editors. 2007, Mosby: Philadelphia. p. 125-132. 169. Fox, E.P. and C.J. Nobile, A sticky situation: untangling the transcriptional network controlling biofilm development in Candida albicans. Transcription, 2012. 3(6): p. 315- 322. 170. Hu, Y., et al., The XRE Family Transcriptional Regulator SrtR in Streptococcus suis Is Involved in Oxidant Tolerance and Virulence. Frontiers in cellular and infection microbiology, 2019. 8: p. 452-452.

103

171. McCallum, N., et al., Transcriptional profiling of XdrA, a new regulator of spa transcription in Staphylococcus aureus. J Bacteriol, 2010. 192(19): p. 5151-64. 172. Wintjens, R. and M. Rooman, Structural classification of HTH DNA-binding domains and protein-DNA interaction modes. J Mol Biol, 1996. 262(2): p. 294-313. 173. Galperin, M.Y., Structural classification of bacterial response regulators: diversity of output domains and domain combinations. Journal of bacteriology, 2006. 188(12): p. 4169-4182. 174. Koudelka, G.B., Recognition of DNA structure by 434 repressor. Nucleic Acids Research, 1998. 26(2): p. 669-675. 175. Li, X., et al., Repression of bacterial motility by a novel fimbrial gene product. The EMBO Journal, 2001. 20(17): p. 4854. 176. Wang, C., et al., BswR controls bacterial motility and biofilm formation in Pseudomonas aeruginosa through modulation of the small RNA rsmZ. Nucleic Acids Res, 2014. 42(7): p. 4563-76. 177. Bouhsira, E., et al., Assessment of persistence of Bartonella henselae in Ctenocephalides felis. Applied and environmental microbiology, 2013. 79(23): p. 7439-7444. 178. Girard, Y.K., et al., A 3D Fibrous Scaffold Inducing Tumoroids: A Platform for Anticancer Drug Development. PLoS ONE, 2013. 8(10): p. e75345. 179. Gutiérrez, D., et al., Monitoring in Real Time the Formation and Removal of Biofilms from Clinical Related Pathogens Using an Impedance-Based Technology. PLoS ONE, 2016. 11(10): p. e0163966. 180. Türker Şener, L., et al., iCELLigence real-time cell analysis system for examining the cytotoxicity of drugs to cancer cell lines. Experimental and therapeutic medicine, 2017. 14(3): p. 1866-1870. 181. Lima, A., et al., Zebrafish embryo model of Bartonella henselae infection. Zebrafish, 2014. 11(5): p. 434-46. 182. Riess, T., et al., Analysis of a novel insect cell culture medium-based growth medium for Bartonella species. Appl Environ Microbiol, 2008. 74(16): p. 5224-7. 183. Schmittgen, T.D. and K.J. Livak, Analyzing real-time PCR data by the comparative C(T) method. Nat Protoc, 2008. 3(6): p. 1101-8. 184. Schmidgen, T., et al., Heterologous expression of Bartonella adhesin A in Escherichia coli by exchange of trimeric autotransporter adhesin domains results in enhanced adhesion properties and a pathogenic phenotype. J Bacteriol, 2014. 196(12): p. 2155-65. 185. O'Toole, G.A., Microtiter dish biofilm formation assay. J Vis Exp, 2011(47). 186. Azeredo, J., et al., Critical review on biofilm methods. Crit Rev Microbiol, 2017. 43(3): p. 313-351. 187. Dabo, S.M., et al., Bartonella henselae Pap31, an extracellular matrix adhesin, binds the fibronectin repeat III13 module. Infect Immun, 2006. 74(5): p. 2513-21. 188. Behnke, O. and T. Zelander, Preservation of intercellular substances by the cationic dye alcian blue in preparative procedures for electron microscopy. J Ultrastruct Res, 1970. 31(5-6): p. 424-8. 189. Simpson, R.J., Fluorescent Staining of Proteins with SYPRO Ruby. CSH Protoc, 2006. 2006(5). 190. de Silva, A.M. and E. Fikrig, Arthropod- and host-specific gene expression by Borrelia burgdorferi. The Journal of clinical investigation, 1997. 99(3): p. 377-379.

104

191. Schotthoefer, A.M., et al., Effects of temperature on the transmission of Yersinia Pestis by the flea, Xenopsylla Cheopis, in the late phase period. Parasites & Vectors, 2011. 4: p. 191-191. 192. Klemm, P., Fimbriae : adhesion, genetics, biogenesis, and vaccines, P. Klemm, Editor. 1994, CRC Press: Boca Raton :. 193. Whiteley, M., et al., Gene expression in Pseudomonas aeruginosa biofilms. Nature, 2001. 413(6858): p. 860-4. 194. Oliveira, R., et al., Polysaccharide production and biofilm formation by Pseudomonas fluorescen: effects of pH and surface material. Colloids and Surfaces B: Biointerfaces, 1994. 2(1): p. 41-46. 195. Booth, S.C., et al., Differences in Metabolism between the Biofilm and Planktonic Response to Metal Stress. Journal of Proteome Research, 2011. 10(7): p. 3190-3199. 196. Merino, E. and C. Yanofsky, Transcription attenuation: a highly conserved regulatory strategy used by bacteria. Trends in Genetics, 2005. 21(5): p. 260-264. 197. Vitreschak, A.G., et al., Attenuation regulation of amino acid biosynthetic operons in proteobacteria: comparative genomics analysis. FEMS Microbiol Lett, 2004. 234(2): p. 357-70. 198. Svoboda, P. and A.D. Cara, Hairpin RNA: a secondary structure of primary importance. Cellular and Molecular Life Sciences CMLS, 2006. 63(7): p. 901-908. 199. Farnham, P.J. and T. Platt, Rho-independent termination: dyad symmetry in DNA causes RNA polymerase to pause during transcription in vitro. Nucleic Acids Res, 1981. 9(3): p. 563-77. 200. Wilson, K.S. and P.H. von Hippel, Transcription termination at intrinsic terminators: the role of the RNA hairpin. Proc Natl Acad Sci U S A, 1995. 92(19): p. 8793-7. 201. Waters, L.S. and G. Storz, Regulatory RNAs in bacteria. Cell, 2009. 136(4): p. 615-28. 202. Winkler, W.C., et al., Control of gene expression by a natural metabolite-responsive ribozyme. Nature, 2004. 428(6980): p. 281-6. 203. Andre, G., et al., S-box and T-box riboswitches and antisense RNA control a sulfur metabolic operon of Clostridium acetobutylicum. Nucleic Acids Res, 2008. 36(18): p. 5955-69. 204. Loh, E., et al., A trans-acting riboswitch controls expression of the virulence regulator PrfA in Listeria monocytogenes. Cell, 2009. 139(4): p. 770-9. 205. Guisbert, E., et al., Convergence of Molecular, Modeling, and Systems Approaches for an Understanding of the Escherichia coli Heat Shock Response. Microbiology and Molecular Biology Reviews, 2008. 72(3): p. 545-554. 206. Klinkert, B. and F. Narberhaus, Microbial thermosensors. Cell Mol Life Sci, 2009. 66(16): p. 2661-76. 207. Resto-Ruiz, S.I., et al., Transcriptional activation of the htrA (High-temperature requirement A) gene from Bartonella henselae. Infect Immun, 2000. 68(10): p. 5970-8. 208. MacKichan, J.K., et al., A SacB mutagenesis strategy reveals that the Bartonella quintana variably expressed outer membrane proteins are required for bloodstream infection of the host. Infect Immun, 2008. 76(2): p. 788-95. 209. Reese, M.G., Application of a time-delay neural network to promoter annotation in the Drosophila melanogaster genome. Comput Chem, 2001. 26(1): p. 51-6. 210. Naville, M., et al., ARNold: a web tool for the prediction of Rho-independent transcription terminators. RNA Biol, 2011. 8(1): p. 11-3.

105

211. Seshasayee, A.S.N., K. Sivaraman, and N.M. Luscombe, An Overview of Prokaryotic Transcription Factors, in A Handbook of Transcription Factors, T.R. Hughes, Editor. 2011, Springer Netherlands: Dordrecht. p. 7-23. 212. Browning, D.F. and S.J. Busby, The regulation of bacterial transcription initiation. Nat Rev Microbiol, 2004. 2(1): p. 57-65. 213. Carl O . Pabo, R.T.S., Transcriptional factors: Structural Families and Principles of DNA Recognition. Ann. Rev. Biochem, 1992. 61: p. 1053-95. 214. Phillips, K. and S.E.V. Phillips, Electrostatic activation of Escherichia coli methionine repressor. Structure, 1994. 2(4): p. 309-316. 215. Altschul, S.F., et al., Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res, 1997. 25(17): p. 3389-402. 216. El-Gebali, S., et al., The Pfam protein families database in 2019. Nucleic Acids Research, 2018. 47(D1): p. D427-D432. 217. Novichkov, P.S., et al., RegPrecise 3.0--a resource for genome-scale exploration of transcriptional regulation in bacteria. BMC Genomics, 2013. 14: p. 745. 218. Gaur, N.K., J. Oppenheim, and I. Smith, The Bacillus subtilis sin gene, a regulator of alternate developmental processes, codes for a DNA-binding protein. Journal of Bacteriology, 1991. 173(2): p. 678-686. 219. Kulinska, A., et al., Genomic and functional characterization of the modular broad-host- range RA3 plasmid, the archetype of the IncU group. Appl Environ Microbiol, 2008. 74(13): p. 4119-32. 220. Li, X., et al., Repression of bacterial motility by a novel fimbrial gene product. The EMBO Journal, 2001. 20(17): p. 4854-4862. 221. Huffman, J.L. and R.G. Brennan, Prokaryotic transcription regulators: more than just the helix-turn-helix motif. Current Opinion in Structural Biology, 2002. 12(1): p. 98-106. 222. Leuze, M.R., et al., Binding Motifs in Bacterial Gene Promoters Modulate Transcriptional Effects of Global Regulators CRP and ArcA. Gene regulation and systems biology, 2012. 6: p. 93-107. 223. Ishihama, A., Prokaryotic genome regulation: multifactor promoters, multitarget regulators and hierarchic networks. FEMS Microbiology Reviews, 2010. 34(5): p. 628- 645. 224. Ishihama, A., The Nucleoid: an Overview. EcoSal Plus, 2009. 3(2). 225. Gama-Castro, S., et al., RegulonDB (version 6.0): gene regulation model of Escherichia coli K-12 beyond transcription, active (experimental) annotated promoters and Textpresso navigation. Nucleic Acids Res, 2008. 36(Database issue): p. D120-4. 226. Salgado, H., et al., RegulonDB (version 4.0): transcriptional regulation, operon organization and growth conditions in Escherichia coli K-12. Nucleic Acids Res, 2004. 32(Database issue): p. D303-6. 227. McTavish, H., et al., Multiple copies of genes coding for electron transport proteins in the bacterium Nitrosomonas europaea. Journal of Bacteriology, 1993. 175(8): p. 2445. 228. Didelot, X., A. Darling, and D. Falush, Inferring genomic flux in bacteria. Genome Res, 2009. 19(2): p. 306-17. 229. Klappenbach, J.A., J.M. Dunbar, and T.M. Schmidt, rRNA operon copy number reflects ecological strategies of bacteria. Applied and environmental microbiology, 2000. 66(4): p. 1328-1333.

106

230. Bikandi, J., San Millán, R., Rementeria, A., and Garaizar, J., n silico analysis of complete bacterial genomes: PCR, AFLP-PCR, and endonuclease restriction. Bioinformatics, 2004. 20: p. 798-9. 231. Rice P., L.I.a.B.A., EMBOSS: The European Molecular Biology Open Software Suite. Trends in Genetics, 2000. 16(6): p. 276-277. 232. Garner, M.M. and A. Revzin, The use of gel electrophoresis to detect and study nucleic acid— protein interactions. Trends in Biochemical Sciences, 1986. 11(10): p. 395-396. 233. Fried, M.G., Measurement of protein-DNA interaction parameters by electrophoresis mobility shift assay. Electrophoresis, 1989. 10(5-6): p. 366-76. 234. Eng, J.K., A.L. McCormack, and J.R. Yates, An approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. J Am Soc Mass Spectrom, 1994. 5(11): p. 976-89. 235. Edwards, A.N., B.R. Anjuwon-Foster, and S.M. McBride, RstA Is a Major Regulator of Clostridioides difficile Toxin Production and Motility. mBio, 2019. 10(2): p. e01991-18. 236. Watanabe, Y.H., M. Takano, and M. Yoshida, ATP binding to nucleotide binding domain (NBD)1 of the ClpB chaperone induces motion of the long coiled-coil, stabilizes the hexamer, and activates NBD2. J Biol Chem, 2005. 280(26): p. 24562-7. 237. Storz, G., J. Vogel, and Karen M. Wassarman, Regulation by Small RNAs in Bacteria: Expanding Frontiers. Molecular Cell, 2011. 43(6): p. 880-891. 238. Wassarman, K.M., 6S RNA: a regulator of transcription. Molecular Microbiology, 2007. 65(6): p. 1425-1431. 239. Song, T., et al., A new Vibrio cholerae sRNA modulates colonization and affects release of outer membrane vesicles. Mol Microbiol, 2008. 70(1): p. 100-11. 240. Pelechano, V. and L.M. Steinmetz, Gene regulation by antisense transcription. Nature Reviews Genetics, 2013. 14: p. 880. 241. Su, W.Y., et al., Bidirectional regulation between WDR83 and its natural antisense transcript DHPS in gastric cancer. Cell Res, 2012. 22(9): p. 1374-89. 242. Georg, J. and W.R. Hess, cis-antisense RNA, another level of gene regulation in bacteria. Microbiology and molecular biology reviews : MMBR, 2011. 75(2): p. 286-300. 243. Kawano, M., L. Aravind, and G. Storz, An antisense RNA controls synthesis of an SOS- induced toxin evolved from an antitoxin. Mol Microbiol, 2007. 64(3): p. 738-54. 244. Saberi, F., et al., Natural antisense RNAs as mRNA regulatory elements in bacteria: a review on function and applications. Cellular & molecular biology letters, 2016. 21: p. 6- 6. 245. Gottesman, S., Micros for microbes: non-coding regulatory RNAs in bacteria. Trends Genet, 2005. 21(7): p. 399-404. 246. Aiba, H., Mechanism of RNA silencing by Hfq-binding small RNAs. Curr Opin Microbiol, 2007. 10(2): p. 134-9. 247. Pichon, C., et al., An in silico model for identification of small RNAs in whole bacterial genomes: characterization of antisense RNAs in pathogenic Escherichia coli and Streptococcus agalactiae strains. Nucleic acids research, 2012. 40(7): p. 2846-2861. 248. Massé, E., et al., Small RNAs controlling iron metabolism. Current Opinion in Microbiology, 2007. 10(2): p. 140-145. 249. Schoenfelder, S.M.K., et al., The small non-coding RNA RsaE influences extracellular matrix composition in Staphylococcus epidermidis biofilm communities. PLOS Pathogens, 2019. 15(3): p. e1007618.

107

250. Gillaspie, D., et al., Plasmid-based system for high-level gene expression and antisense gene knockdown in Bartonella henselae. Appl Environ Microbiol, 2009. 75(16): p. 5434- 6. 251. Sharma, C.M., et al., A small RNA regulates multiple ABC transporter mRNAs by targeting C/A-rich elements inside and upstream of ribosome-binding sites. Genes Dev, 2007. 21(21): p. 2804-17. 252. Hammer, B.K. and B.L. Bassler, Regulatory small RNAs circumvent the conventional quorum sensing pathway in pandemic Vibrio cholerae. Proc Natl Acad Sci U S A, 2007. 104(27): p. 11145-9. 253. Roncarati, D. and V. Scarlato, Regulation of heat-shock genes in bacteria: from signal sensing to gene expression output. FEMS Microbiology Reviews, 2017. 41(4): p. 549- 574. 254. Tenorio, E., et al., Systematic characterization of Escherichia coli genes/ORFs affecting biofilm formation. FEMS Microbiology Letters, 2003. 225(1): p. 107-114. 255. Poquet, I., et al., Clostridium difficile Biofilm: Remodeling Metabolism and Cell Surface to Build a Sparse and Heterogeneously Aggregated Architecture. Frontiers in Microbiology, 2018. 9(2084). 256. Prindle, A., et al., Ion channels enable electrical communication in bacterial communities. Nature, 2015. 527(7576): p. 59-63. 257. Carzaniga, T., et al., The RNA processing enzyme polynucleotide phosphorylase negatively controls biofilm formation by repressing poly-N-acetylglucosamine (PNAG) production in Escherichia coli C. BMC microbiology, 2012. 12: p. 270-270. 258. Kim, M., M. Kim, and K.-s. Kim, YmdB-mediated down-regulation of sucA inhibits biofilm formation and induces apramycin susceptibility in Escherichia coli. Biochemical and Biophysical Research Communications, 2017. 483(1): p. 252-257. 259. Beloin, C., et al., Global impact of mature biofilm lifestyle on Escherichia coli K-12 gene expression. Mol Microbiol, 2004. 51(3): p. 659-74. 260. Gaupp, R., et al., Advantage of upregulation of succinate dehydrogenase in Staphylococcus aureus biofilms. Journal of bacteriology, 2010. 192(9): p. 2385-2394. 261. Herzberg, M., et al., YdgG (TqsA) controls biofilm formation in Escherichia coli K-12 through autoinducer 2 transport. Journal of bacteriology, 2006. 188(2): p. 587-598. 262. Bar-On, Y.M., R. Phillips, and R. Milo, The biomass distribution on Earth. Proc Natl Acad Sci U S A, 2018. 115(25): p. 6506-6511. 263. Magnabosco, C., et al., The biomass and biodiversity of the continental subsurface. Nature Geoscience, 2018. 11(10): p. 707-717. 264. Whitman, W.B., D.C. Coleman, and W.J. Wiebe, Prokaryotes: the unseen majority. Proc Natl Acad Sci U S A, 1998. 95(12): p. 6578-83. 265. Brown, S.P., D.M. Cornforth, and N. Mideo, Evolution of virulence in opportunistic pathogens: generalism, plasticity, and control. Trends Microbiol, 2012. 20(7): p. 336-42. 266. Eckburg, P.B., et al., Diversity of the human intestinal microbial flora. Science, 2005. 308(5728): p. 1635-8. 267. Toyofuku, M., et al., Environmental factors that shape biofilm formation. Biosci Biotechnol Biochem, 2016. 80(1): p. 7-12. 268. Di Bonaventura, G., et al., Effect of environmental factors on biofilm formation by clinical Stenotrophomonas maltophilia isolates. Folia Microbiol (Praha), 2007. 52(1): p. 86-90.

108

269. Jefferson, K.K., What drives bacteria to produce a biofilm? FEMS Microbiology Letters, 2004. 236(2): p. 163-173. 270. Morikawa, M., Beneficial biofilm formation by industrial bacteria Bacillus subtilis and related species. Journal of Bioscience and Bioengineering, 2006. 101(1): p. 1-8. 271. Donlan, R.M. and J.W. Costerton, Biofilms: survival mechanisms of clinically relevant microorganisms. Clin Microbiol Rev, 2002. 15(2): p. 167-93. 272. Arraiano, C.M., et al., The critical role of RNA processing and degradation in the control of gene expression. FEMS Microbiol Rev, 2010. 34(5): p. 883-923. 273. O’Neill, J., Antimicrobial Resistance: Tackling a crisis for the health and wealth of nations. . http://www.jpiamr.eu/wp-content/uploads/2014/12/AMR-Review-Paper- Tackling-a-crisis-for-the-health-and-wealth-of-nations_1-2.pdf. 274. Calvori, C., et al., Effect of rifamycin on protein synthesis. Nature, 1965. 207(995): p. 417-8. 275. Scott, L.J., Fidaxomicin: a review of its use in patients with Clostridium difficile infection. Drugs, 2013. 73(15): p. 1733-47. 276. Srivastava, A., et al., New target for inhibition of bacterial RNA polymerase: 'switch region'. Current opinion in microbiology, 2011. 14(5): p. 532-543. 277. Ma, C., X. Yang, and P.J. Lewis, Bacterial Transcription as a Target for Antibacterial Drug Development. Microbiology and molecular biology reviews : MMBR, 2016. 80(1): p. 139-160. 278. Perez, J.C. and E.A. Groisman, Evolution of transcriptional regulatory circuits in bacteria. Cell, 2009. 138(2): p. 233-44. 279. Chopra, I., Bacterial RNA polymerase: a promising target for the discovery of new antimicrobial agents. Curr Opin Investig Drugs, 2007. 8(8): p. 600-7. 280. Matsui, M. and D.R. Corey, Non-coding RNAs as drug targets. Nature Reviews Drug Discovery, 2016. 16: p. 167. 281. Wright, G.D., A.M. Berghuis, and S. Mobashery, Aminoglycoside antibiotics. Structures, functions, and resistance. Adv Exp Med Biol, 1998. 456: p. 27-69. 282. Moazed, D. and H.F. Noller, Interaction of antibiotics with functional sites in 16S ribosomal RNA. Nature, 1987. 327(6121): p. 389-94. 283. Dersch, P., et al., Roles of Regulatory RNAs for Antibiotic Resistance in Bacteria and Their Potential Value as Novel Drug Targets. Frontiers in Microbiology, 2017. 8(803). 284. Thomas, J.R. and P.J. Hergenrother, Targeting RNA with small molecules. Chem Rev, 2008. 108(4): p. 1171-224. 285. Sutcliffe, J.A., Improving on nature: antibiotics that target the ribosome. Curr Opin Microbiol, 2005. 8(5): p. 534-42. 286. Stephenson, M.L. and P.C. Zamecnik, Inhibition of Rous sarcoma viral RNA translation by a specific oligodeoxyribonucleotide. Proceedings of the National Academy of Sciences of the United States of America, 1978. 75(1): p. 285-288. 287. Norrby, S.R., C.E. Nord, and R. Finch, Lack of development of new antimicrobial drugs: a potential serious threat to public health. Lancet Infect Dis, 2005. 5(2): p. 115-9. 288. Cachia, P.J. and R.S. Hodges, Synthetic peptide vaccine and antibody therapeutic development: prevention and treatment of Pseudomonas aeruginosa. Biopolymers, 2003. 71(2): p. 141-68.

109

289. Sheth, H.B., et al., Development of an anti-adhesive vaccine for Pseudomonas aeruginosa targeting the C-terminal region of the pilin structural protein. Biomed Pept Proteins Nucleic Acids, 1995. 1(3): p. 141-8. 290. McCarthy, M.W., et al., Novel Agents and Drug Targets to Meet the Challenges of Resistant Fungi. The Journal of Infectious Diseases, 2017. 216(suppl_3): p. S474-S483.

110

Appendices

111

Appendix A: Copyright permission for Chapter 1

112

Appendix B: Copyright permission for Figure 1.2

113

Appendix C: Copyright permission for Chapter.2

114