1.2 an Overview of Phage Therapy

Development of Natural and Engineered Bacteriophages as Antimicrobials

by Robert James Citorik BSc. in Microbiology, University of New Hampshire (2008)

Submitted to the Microbiology Graduate Program in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

at the

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

June 2018

Signature redacted A uthor ...... Microbiology Graduate Program Signature reaacted May 25, 2018 C ertified by ...... Timothy K. Lu Associate Professor of Biological Engineering and Electrical Engineering and Computer Science Thesis Supervisor Signature redacted A ccepted by ...... Kristala L. Jones Prather MASSACHUSMlS INSTrTUTE OF TECHNOWGY rthur D. Little Professor of Chemical Engineering Chair of Microbiology Program JUL 09 2018 LIBRARIES ARCHIVES 2 Development of Natural and Engineered Bacteriophages as Antimicrobials by Robert James Citorik

Submitted to the Microbiology Graduate Program on May 25, 2018, in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Abstract

One of the major public health concerns of the modern day is the emergence and spread of extensively antibiotic-resistant pathogens. We have already seen the arrival of infections caused by bacteria resistant to all available antibiotics in the therapeutic arsenal. In addition, we have learned much of the incredible importance of the microbial communities that cohabit our bodies, and of how perturbations to these communities can lead to long-lasting health effects. Bacteriophages may provide a solution for both of these problems, in that they are narrow-spectrum and can be used to specifically kill target microbes without disrupting whole community structure through off-target effects. Here, various approaches to creating phage- based therapeutics are explored, including the isolation and application of naturally occurring wild-type phages, the conversion of temperate phages to obligately lytic phages to permit their use as a resource in phage therapeutics, and the creation of programmable, sequence-specific antimicrobials through phage-mediated genetic payload delivery. These efforts are expected to contribute to the field by expanding the approaches available to develop next-generation, phage-based antimicrobials.

Thesis Supervisor: Timothy K. Lu Title: Associate Professor of Biological Engineering and Electrical Engineering and Computer Science

3 4 Acknowledgments

I would like to take this opportunity to first express my sincere gratitude to all of the teachers, friends, family, and other role models who have helped and inspired me to complete this leg of my journey. Without all of you, none of this would have been possible.

To Tim Lu, my supervisor and mentor for the past 7 years: I thank you for giving me the opportunity to explore exciting and creative science under your guidance. I recognize that being in this laboratory has been a unique experience, and that many are not given the same freedom and independence to pursue new ideas and explore interesting side projects.

To my thesis committee members, Jim Collins, Eric Alm, and Deb Hung: I am grateful for all of your guidance and expertise in pursuing, as well as finishing, my

PhD research. None of you needed to say yes when I asked for your time in this endeavor, and for this I am very thankful. I hope that our conversations continue beyond my time at MIT.

To my program: I feel extremely lucky to have found the Microbiology Program at MIT, and I thank Alan Grossman for creating this opportunity for such interdisci- plinary research, as well as Bonnielee Whang for helping keep us in line and all those faculty who have helped to organize and lead the way. This unique program allowed me to venture beyond basic science research and enter into the exciting fields of biological engineering and synthetic biology, where I have found an exciting interface in which to build upon my previous research in pathogenic microbiology.

To my wife: Jenna, I would never have survived this experience without your support and understanding. Your effort and commitment to giving your all to your students, as well as your husband, is truly inspiring. This degree belongs to the both of us. I love you with all of my heart, and being able to start forever with you has been an unanticipated highlight of my graduate school tenure. If we can survive being a PhD student at MIT and an elementary school teacher working tireless hours, we can survive anything. On to the next chapter!

5 To my family: I thank my mother and father for instilling in me the drive that has brought me to where I am today, for I am certain that I am here because they cared about my education from the very first day of school. I know the days of getting help with homework are long gone, but those lessons in problem-solving, as well as how to look out for your children, will last a lifetime. I thank my brother and sister, as well as my brother- and sister-in-law, for always being there for me. They are the kind of support system crucial to success in any aspect of life. I also thank my grandparents and the others who have helped teach me lessons far beyond the realm of academia.

To my friends and labmates: I thank you for your day-to-day support, from the little things to the major ones. From bringing joy to mundane tasks to having a drink after impressive failures. You are all the best. I cannot believe the number of lifelong friends I have made by spending too much time together in the lab. To Mark, I thank you for being a simultaneous peer, friend, and mentor. I hope that you have benefited even a fraction as much as I have from our lab bromance.

To Area Four: I thank you for your caffeine. I have seen many baristas come and go during my tenure, but the coffee has stood the test of time.

And to all those not specifically mentioned here: the people who have touched my life from my early years to today are countless, and the journey would never have been the same save for all of these interactions. So a final thank you goes out to the unnamed, whose influence neither of us may even remember, but who have helped me to be where I am today. And with these words, I will continue seeking to leave a positive mark on the world as I go forth and SCIENCE!

6 Contents

1 Introduction 13 1.1 Introduction ...... 13

1.2 An Overview of Phage Therapy ...... 13

1.3 Engineered Phages as Therapeutics and Tools for Health ...... 16

1.4 New Phage Engineering Strategies ...... 20

1.5 Bacteriophages and the Microbiome ...... 22

1.6 Chapter Overviews ...... 24

2 Phages from Without: Isolation and Characterization of Phag E~s

from the Environment 27

2.1 Introduction ...... 28

2.2 Isolation of Novel Bacteriophages ...... 29

2.3 Characterization of Bacteriophages ...... 29

2.4 Murine Gastrointestinal Colonization Model ...... 33 2.5 D iscussion ...... 37 2.6 Experimental Details ...... 39

3 Phages from Within: Utilization of Prophages for Bacterial Target-

ing 43

3.1 Introduction ...... 44

3.2 Discovery and Characterization of <}Kpn852 ...... 47

3.3 Construction of Lytic Phage Derivatives ...... 50 3.4 D iscussion ...... 54

7 3.5 Experimental Details ...... 55

4 Phages born Anew: Using CRISPR Payloads to Create Sequence-

Specific Antimicrobials 59

4.1 Introduction ...... 60

4.2 Transformation Assays for Validation ...... 62 4.3 Cell-Based Delivery ...... 64 4.4 Phage-Based Delivery ...... 66

4.4.1 Toxin-Antitoxin Activation ...... 69

4.5 Targeting Virulence Genes in Galleria mellonella Models of Infection 71

4.6 Population Sculpting ...... 73

4.7 D iscussion ...... 74

4.8 Experimental Details ...... 78

5 Conclusion 87

8 List of Figures

2-1 Plaque assay on propagation hosts ...... 31

2-2 Mutant derivative K6.2 shows probable capsule alteration ...... 31

2-3 Double-agar spotting assay for determining host range of inhibitory activity ...... 32

2-4 Visualization of selected bacteriophages by transmission electron microscopy (TEM ) ...... 34

2-5 Murine gastrointestinal colonization model ...... 36

2-6 Phage therapy reduces K. pneumoniae levels in murine gut ...... 36 2-7 Phage

3-1 Two disparate lifestyles for bacteriophages ...... 44

3-2 Overview of yeast assembly for phage genome reconstruction ..... 46

3-3 K. pneumoniae KPNIH31 harbors a viable prophage ...... 48 3-4 Polishing of the

3-5 Phage

3-6 TEM visualization of K. pneumoniae phage

3-7 Phage

3-8 Overexpression of antirepressor antC in E. coli C-1 prevents lysoge-

nization by phage N15 ...... 52

3-9 Engineering an obligately lytic derivative of temperate phage N15 . . 53

4-1 RGN overview schematic ...... 61

9 4-2 Design and validation of programmable RGN constructs by transfor-

m ation ...... 63

4-3 Characterization of escape mutants that tolerated transformation of a

cytotoxic RGN construct ...... 64

4-4 Mobilizable RGNs can be conjugated into target bacteria for selective

removal of multidrug resistance ...... 65

4-5 RGN constructs delivered via bacteriophage particles (DRGN) exhibit

efficient and specific antimicrobial effects against strains harboring

plasmid or chromosomal target sequences ...... 67

4-6 Characterization of 4RGN-mediated killing of antibiotic-resistant bac-

teria ...... 68

4-7 Treatment of E. coli with 4RGNs induces DNA damage and an SOS

response in cells that possess a cognate target sequence ...... 69

4-8 RGN-mediated targeting of toxin-antitoxin systems can lead to cyto-

toxicity ...... 71

4-9 'IRGN particles elicit sequence-specific toxicity against enterohemor-

rhagic E. coli in vitro and in vivo ...... 72

4-10 Minimum inhibitory concentrations (MICs) ...... 73

4-11 Comparison of 4RGNeae to conventional antibiotic treatment of EHEC-

infected Galleria mellonella larvae ...... 73

4-12 Programmable remodeling of a synthetic microbial consortium ... . 75

4-13 Bacterial strains used in this study ...... 79

11WIMP11- , ' '" I 01WIMIMMIPM, 11,111.11 11WIP-Mr- ______List of Tables

2.1 Isolation of bacteriophages against K. pneumoniae ...... 30

2.2 TEM morphological characterization ...... 35

2.3 Bacterial strains for phage isolations ...... 39

3.1 Infectivity of KPNIH strains by

11 12 Chapter 1

Introduction

Portions of this chapter are adapted from reviews written or co-written by the author [1-3].

1.1 Introduction

In an age when antibiotic-resistant bacterial infections are increasingly common and lead to more treatment failures, new solutions are sorely needed. One promising answer to these mounting concerns is a century-old solution forgotten by Western medicine that employs natural predators of bacteria: enter the bacteriophages.

1.2 An Overview of Phage Therapy

Like many antibiotics, bacteriophages (phages) are naturally occurring and have the capacity to kill bacteria. Unlike antibiotics, however, phages are viruses and so can replicate and amplify within their target bacteria. These 'bacteria eaters' are conceptually similar to human viruses, but can only infect bacterial cells. From a human perspective, phage therapy is a good example of the proverbial saying 'the enemy of my enemy is my friend'. Phages reproduce by infecting bacteria and turning them into phage manufacturing plants, which can prove lethal to the bacterial cell.

Once the progeny have been assembled, the phage initiates enzyme-mediated lysis,

13 Bacteriophages for Human Health rupturing the host bacterium from the inside and releasing the next generation of infectious particles to repeat the cycle.

Phage therapy dates back to the early 20th century, shortly after the independent discovery of bacteriophages by bacteriologists Frederick Twort (1915) [4] and Felix d'Herelle (1917) [5]. It was found that sometimes a phenomenon occurred resulting in zones of clearing on bacteria growing in Petri dishes or a healthy, cloudy culture of bacteria to suddenly lyse and become clear. The lysed culture could be filtered and added to other cultures of the bacteria to cause a repeatable clearing effect. Phages able to lyse cultures of the intestinal pathogen Shigella were found alongside the bacteria in patients with dysentery, with an increase in phage concentration often corresponding to recovery, and so were thought to be part of the natural course of disease [6].

After some reported successes, research and commercial development of phage- based therapies expanded over the following decades, although there was still contro- versy surrounding the nature of the agent or agents causing the bactericidal effects.

An early debate centered on whether phages were indeed of a viral nature or instead were autolytic enzymes produced by the bacteria themselves, and it was not until 1940 that electron micrographs were published visually depicting what was responsible [7].

In the beginning, scientists did not yet fully appreciate the diversity of phages and the specificity of each one to its target. Unlike broad-spectrum antibiotics, which kill bacteria fairly indiscriminately, phages are usually highly specific to a target bacterial species, and often limited to certain strains within a particular species. Early experiments and trials were likely impacted by these and other gaps in knowledge, and choosing the wrong phages or using suboptimal phage preparations to treat a bacterial infection may have led to inconsistencies in treatment outcomes. As a result, the initial excitement associated with phage therapy was short-lived. The appearance of antibiotics in the physician's toolbox around the mid-20th century coincided with a decline in phage therapy for various reasons, including the incredible efficacy of the new drugs, an insufficient understanding of phages and mixed reports of success, and even political motivations.

14 Chapter 1 Citorik Bacteriophages for Human Health

Despite the complicated history leading to phage therapy falling out of favor in the West [8], it has continued to some extent in other parts of the world, primar- ily in Russia and Eastern Europe. The George Eliava Institute of Bacteriophages, Microbiology and Virology in Tbilisi, Georgia, is one of the most well-known of the clinics still specializing in this antibacterial therapy. As well as preparations designed for off-the-shelf use for particular disease indications, the institute has a large bacteriophage collection that can be screened against bacterial pathogens from patient samples to create personalized treatments.

For phage therapy to become accepted in modern Western medicine, there are several challenges that must be addressed, the first of which may be to definitively prove safety and efficacy. Perhaps the first safety trials were performed by d'Herelle, who administered phage to himself as well as to family members, noting no adverse effects from the preparations [8]. As a result of the discontinuation of phage therapy in the first half of the 20th century, safety and efficacy testing did not evolve concomitantly with the more rigorous clinical testing standards, which now include blind or double- blind trials with appropriate controls and strict statistical analyses. Accordingly, the efficacy of phage therapy versus standards of care has never been convincingly demonstrated to Western standards, although renewed efforts are emerging [9, 10].

On this front, the large-scale Phagoburn trial has been funded by the European

Union and sponsored by Pherecydes Pharma. Phagoburn, which focuses on phage treatment of infected burn wounds, was launched in 2013, with the first patients entering into the phase I/II clinical study in 2015. In order to test safety, efficacy and pharmacodynamics, cocktails of around a dozen different phages targeting either

Escherichia coli or Pseudomonas aeruginosa were designed to treat enrolled patients identified as carrying these specific pathogens. Additionally, AmpliPhi Biosciences, working with the University of Adelaide and Flinders University, recently dosed their first patient with phages targeting Staphylococcus aureus in a phase I safety trial at the Queen Elizabeth Hospital in Adelaide, Australia. The design and implementation of these types of trials will undoubtedly prove critical to moving the technology into modern medicine.

Citorik Chapter I 15 Bacteriophages for Human Health

Beyond safety and efficacy trials, potential phage therapies may need to overcome other hurdles, including the release of endotoxin or other toxic bacterial products when infected bacteria rupture. Other issues are the emergence of bacterial resistance to phages, bacterial strain coverage, pharmacokinetics and how to dose a self-replicating agent, as well as the patient immune responses, which may target the phage when administered into the body.

These challenges are already being addressed to some degree, with efforts demonstrating phages designed to kill bacteria with limited endotoxin release [11, 12] and the selection for phages with improved residence time in the body [13]. Bacterial strain coverage can be improved using phage cocktails, which have multiple phages able to target different receptors on a bacterium, which could also decrease the likeli- hood of resistance. Additionally, phages can be naturally evolved to subvert defensive tactics employed by bacteria that have become resistant to infection, meaning that components of a cocktail that are no longer effective could be re-derived instead of shelved alongside defunct antibiotics.

The field of phage therapy is poised very differently now than it was a century ago. The understanding of phage biology that we have today promises to help us to select bacteriophage cocktail components appropriately, produce and purify them according to suitable standards, and design and execute rigorous clinical trials.

1.3 Engineered Phages as Therapeutics and Tools for Health

As an alternative approach to using only naturally derived phages, some engineering efforts in synthetic biology have been made toward adding functions or improving existing phages. For example, Lu and Collins [14] incorporated the gene encoding

DspB, an enzyme that degrades a polysaccharide adhesin implicated in biofilm formation, into an engineered T7 phage. This modified phage effectively cleared F. coli biofilms through cycles of infection, phage-mediated lysis, and release of the recombi-

16 Chapter 1 Citorik Bacteriophages for Human Health nant dispersin enzyme to enzymatically degrade the biofilm material itself and expose protected cells. Additionally, the phage was modified to carry a gene from phage T3 in order to expand its host range and permit infection of the biofilm-forming strain used in the study. However, despite the promise of phage therapeutics, bacteria can display resistance toward phages through innate means, such as restriction-modification systems [15, 16], as well as adaptive means, typified by clustered regularly interspaced short palindromic repeats (CRISPR)-CRISPR-associated (Cas) systems [17]. More- over, mechanisms may emerge in a bacterial population during the course of selective pressure by phages, including phenotypic [18] and genotypic [19] causes of decreased phage adsorption, among others [20]. These hurdles may be tackled through the use of phage cocktails [21], high-throughput phage evolution, or perhaps, given predictable evolutionary pathways, through the rational engineering of phages [22].

In contrast to taking advantage of a phage's natural ability to lyse a target cell, some studies have focused on using virus particles only for their capacity to deliver nucleic acids to target cells. Such an approach was taken by Westwater et al. [23], in which the group utilized the non-lytic, filamentous phage M13 to deliver specialized phagemid DNA in place of the phage genome to target cells. The engineered phagemids (plasmids carrying signals to enable packaging into phage particles) were designed to encode the addiction toxins Gef and ChpBK to elicit destruction of target cells. Hagens and Blisi [11] also applied this toxic payload concept using M13 to deliver genes encoding the restriction enzyme BglII or the A S holin to kill target E. coli by the introduction of double-stranded breaks in the chromosome or the creation of cytoplasmic membrane lesions, respectively. Subsequently, delivery of BglII was used to rescue mice infected with P. aeruginosa by adapting the system with an engineered derivative of the P. aeruginosa filamentous phage Pf3 [12]. These methods also resulted in a marked decrease in release of endotoxin, one of the major concerns with lytic phage therapy [24], as compared to killing via lysis by a lytic phage [11, 12]. More recently, M13-derived particles were used to express a lethal mutant of catabolite activator protein in E. coli 0157:H7, a food-borne pathogen that causes outbreaks of hemorrhagic colitis [25]. Biotechnology companies have also begun to

Citorik Chapter 1 17 Bacteriophages for Human Health

make use of recombinant phage methods, such as virus-like particles that deliver

genes encoding small, acid-soluble proteins to cause toxicity to target cells through non-specific binding to DNA [26].

Rather than encoding killing functions directly within phage particles, Edgar et

al. [27] used phage A as a chassis to generate antibiotic-resensitizing particles through

the delivery of dominant wild-type copies of rpsL and gyrA. The transduction of these

genes into target cells resistant to streptomycin and fluoroquinolones, conferred by

mutations in rpsL and gyrA, respectively, resulted in the production of wild-type

enzymes susceptible to the formerly ineffective drugs. In another demonstration, Lu

and Collins [28] engineered M13 to carry genes encoding transcription factors that

modify the native regulation of bacterial gene networks. Constructs encoding the

LexA3 repressor or SoxR regulator were used to disable the SOS response and DNA

repair or to modulate the response to oxidative stress in target cells, respectively, thus potentiating the toxic effects of antibiotic treatment and even resensitizing a

resistant bacterial strain. A dual-function phage was also created and validated by

using M13 harboring the global regulator csrA, to inhibit biofilm formation and the

associated increase in antibiotic resistance, and the porin ompF, to improve drug

penetration. Examples such as these demonstrate the capacity for bacteriophages

to be engineered as gene delivery devices in order to perturb genetic networks in

bacteria for both research and therapeutic applications. With this approach, one can

alter a gene network at a particular node and observe the qualitative and quantitative

effects in order to better characterize native regulatory systems. As models of the

interactions in complex regulatory webs of pathogens grow increasingly robust, the

ability to know which strands to tug to elicit desired effects may enable rationally

designed novel therapeutics based on predictable behaviors.

The development and improvement of next-generation sequencing technology has

enabled genomic and metagenomic analyses of phage populations [29, 30]. For ex-

ample, sequencing of gut viral metagenomes has implicated phages as reservoirs of

antibiotic-resistance genes [31] and their role in influencing the intestinal microbiota

has been of recent interest [32]. Since bacteriophages must encode mechanisms to

18 Chapter 1 Citorik

IM1,11.1I INIMPITI ""M Bacteriophages for Human Health control their host cells in order to infect, divert cellular resources to propagate, build progeny phages, and, in many cases, lyse their hosts to release new particles, phage genomes constitute a vast library of parts that can be used to manipulate bacteria for study or treatment. On the basis of this concept, Liu and colleagues [33] developed a method for mining such tools to generate novel therapeutics against S. aureus.

Predicted phage open reading frames were cloned with inducible expression into the target strain and screened for growth-inhibitory properties. Identified phage proteins were used to pull bacterial targets out of cell lysates and a library of small molecules was screened to identify inhibitors of the protein-protein interaction, with the hy- pothesis that these molecules might demonstrate similar modulatory action on the host target. In this way, the authors identified novel compounds capable of inhibiting the initiation of bacterial DNA replication in analogy with the phage proteins. Since currently available drugs that target replication only act on topoisomerases, this work demonstrates that mining phage proteins long evolved to inhibit bacterial processes has the potential to expand the antibiotic repertoire by leading us to discover drugs against previously unused targets [33, 34].

In addition to random-discovery screens, phage lysins have been specifically investigated in recent years as potential antimicrobials. These enzymes are employed by bacteriophages to degrade the bacterial cell wall and permit the release of progeny phages [35]. In another functional metagenomic study, phage DNA was isolated from a mixture of feces from nine animal species, cloned into a shotgun library for inducible expression in E. coli, and used in primary and secondary screens to detect lysins from the phage DNA pool [36]. As a discovery tool, a specific lysin from a phage of Bacillus anthracis was used to develop a novel antimicrobial by identifying an enzyme involved in the production of the lysin target and designing a cognate chemical inhibitor [37]. Though lysins are considered useful antimicrobials for Gram- positive pathogens, Gram-negative bacteria possess an outer membrane that prevents access of these extracellular enzymes to the cell wall [38]. To overcome this barrier, a chimeric protein composed of the translocation domain of the Yersinia pestis bacteriocin, pesticin, and the enzymatic domain of lysozyme from the E. coli phage T4

Citorik Chapter 1 19 Bacteriophages for Human Health

was engineered. The hybrid bacteriocin was shown to be active against E. coli and

Y. pestis strains, including those expressing the cognate immunity protein conferring

resistance to unmodified pesticin [39, 40].

Bacteriophages have also been used to implement real-world applications of biosens-

ing [41-45]. In areas from healthcare and hospital surfaces to food preparation and

other industrial processes, methods for the rapid detection of pathogenic organisms

are paramount in preventing disease and avoiding the public relations and finan-

cial burdens of recalling contaminated products. The amount of time necessary for

many conventional detection methods is long due to the requirement for bacterial

enrichment before detection of the few bacteria present in complex samples in order

to achieve sufficient assay sensitivity and specificity [46]. Engineered bacteriophage-

based detectors have the advantage of rapid read-outs, high sensitivity and specificity, and detection of live cells [47]. A common design strategy is the creation of reporter-

based constructs packaged within phage or phage-like particles that infect target cells

and ultimately result in the production of fluorescent, colorimetric, or luminescent

signals. Furthermore, sensor designs can include genetically engineered phage that

express a product causing ice nucleation [48] or that incorporate tags for linking to

detectable elements such as quantum dots [49]. Though most of these examples of

specifically modified phages have been enabled by advancements in engineering and

synthetic biology to achieve real-world applicability, the concept of using natural

phage as sensing tools is not a new one. Phage typing and other techniques have

made use of the narrow host range of phage to identify species or strains of bacteria

based on a target bacteria's ability to bind, propagate, or be lysed by non-engineered viruses [47].

1.4 New Phage Engineering Strategies

Historically, modifications to bacteriophage relied on random mutagenesis or ho-

mologous recombination, both of which are inefficient and necessitate intensive screen-

ing to identify mutants of interest. The relatively large size of most bacteriophage

20 Chapter 1 Citorik

III lI Ill1 ~I 1-'11lis lmIll'ipli ll'MRIIIIIIII Bacteriophages for Human Health

genomes and their inherent toxicity to bacterial hosts has confounded the use of con-

ventional molecular biology techniques for engineering. However, recent synthetic

biology tools have revitalized the ability to make rational additions or modifica-

tions to phage genomes. Among such improvements, the phage defense function

encoded by CRISPR-Cas systems, previously adapted for genome editing [50-55] and

reviewed in [56], has been described for improving recombineering in bacteriophages

by counter-selecting unmodified phages with wild-type target sequences [57]. In vitro

assembly of large constructs has also been made possible with techniques such as

Gibson assembly [58], which enzymatically stitches together DNA fragments with

overlapping homology, thus allowing for insertions of heterologous DNA and site-

directed mutagenesis using PCR. Moreover, transformation of overlapping fragments

into yeast in conjunction with a compatible yeast artificial chromosome leads to in

vivo recombination-based assembly of large constructs [59, 60]. Genomes can be as-

sembled with modifications or be modified post-assembly in yeast, where they are

non-toxic to the host, and then purified and rebooted in bacteria to produce engi-

neered phage progeny. Current DNA synthesis technology, in concert with in vivo

and in vitro recombination, also permits de novo chemical synthesis of bacteriophage

genomes. Smith et al. [61 utilized this approach to synthesize, clone, and produce

infectious particles of the 5386 bp phage

has created the first bacterial cell with a synthetic genome of 1.1 Mb [62]. These

in vitro approaches to creating engineered phage genomes have the potential to be

combined with newer methods in efficiently rebooting phages from genomic DNA, including efficient uptake of full genomes by cell wall deficient L-form bacteria [63]

and completely cell-free synthesis of viable phages [64, 65]. By rendering bacteriophages genetically accessible, synthetic biology can permit more precise studies of their underlying biology and inspire creation of novel therapeutic agents.

Despite advances in rational engineering of bacteriophages, tampering with systems finely tuned by evolution can lead to fitness defects [66]. For example, roughly

30% of the genome of the bacteriophage T7 was refactored, a process whereby genes and their respective regulatory elements were separated into distinct modules to per-

Citorik Chapter 1 21 Bacteriophages for Human Health mit systematic analysis and control [67]. The refactored genome produced viable bacteriophage, albeit with significantly reduced fitness. Multiple rounds of in vitro evolution restored wild-type viability at the expense of some of the design elements, implying that rational design can be coupled with evolution to ensure the creation of robust biological systems [68]. Similarly, the evolutionary stability of a T7 phage engineered to infect encapsulated E. coli by producing a capsule-degrading endosialidase as an exoenzyme was investigated in vitro [69]. While the engineered phage permitted replication in the encapsulated strain, the benefit conferred by endosialidase production was shared by wild-type, non-producing 'cheater' phages, which could quickly outcompete the engineered viruses due to their higher fitness. Although these studies point to the fragility of current synthetic biology efforts, bacteriophage-based systems can serve as an excellent platform to understand the constraints placed on synthetic genetic circuits by evolution and inform future designs.

1.5 Bacteriophages and the Microbiome

Subtractive therapies may employ chemicals, peptides, or even replicating entities to remove bacteria from the gut with varying degrees of specificity. In medicine, this is currently accomplished through the use of antibiotics, which tend to be broad- spectrum in nature, exhibiting activity towards many different bacteria. As a result, treatment of a patient aimed to remove an infectious pathogen also leads to the unintended reduction of other members of the microbiota. This community shifting may cause the patient to become susceptible to other temporary or chronic conditions to which they are normally protected, including antibiotic-associated infections with

C. difficile. The development of targeted antimicrobials, such as bacteriocins and bacteriophages, could yield more effective subtractive therapies.

Bacteriophage therapy is a highly specific method of killing bacteria through the use of natural or engineered viral parasites. The application of phages as antimicrobials has seen a renewed interest with the growing threat of antibiotic-resistant pathogens [9, 70]. Though early phage therapies targeted intestinal pathogens [71],

22 Chapter 1 Citorik Bacteriophages for Human Health clearance issues have recently been reported wherein bacterial and phage populations stably coexist in the murine gut [72, 73]. Knowledge attained from research into the ecology of phages in the gut may be pivotal in determining factors that lead to successful therapies in this complex ecosystem.

A newer focus of the field is to examine the natural role of these viruses in shaping host-associated bacterial populations [32]. Metagenomic research of the human microbiome has described the fecal virome of both healthy and diseased donors. These studies include measuring phage diversity, variability, and stability [74], and analyzing changes associated with diet [75] or inflammatory bowel disease (IBD) [76] in humans, or antibiotic treatment in mice [31]. For example, a study of the viromes of monozygotic twins and their mothers revealed high interpersonal variation of virome composition, but low intrapersonal diversity that was dominated by temperate phages, or those that can exist in a silenced life cycle within bacteria [74]. Diet has been identified as one factor that affects the viral community, as putting human subjects on a controlled diet altered community structure and resulted in a level of convergence for individuals on the same diet [75]. Interestingly, whereas IBD correlates with a reduced level of bacterial diversity, multiple cohorts have revealed a concomitant increase in bacteriophage richness, specifically those belonging to Caudovirales [76]. To explore phage-bacterial host dynamics, gnotobiotic mice were seeded with a defined, 15-member human commensal community and challenged orally with virus-like particles (VLPs) purified from healthy human donors. In this study, the authors made several interesting observations, including an increase in specific phages correlating with a transient decrease in specific bacteria, different phages and bacteria showing non-simultaneous temporal population dynamics, and evidence of phage resistance likely due to ecological, rather than genetic, factors [77]. These types of comparison and challenge studies should continue to yield important information useful for designing phage-based strategies for targeting pathogens or altering microbial communities in the gut, particularly in regards to determining the factors that lead to transient versus stable community rearrangements.

Engineered phages, or those modified to carry additional or alternative functions

Citorik Chapter 1 23 Bacteriophages for Human Health

to those naturally occurring, may prove useful for therapeutics as design can be in-

formed by new knowledge. Recently, it was found that certain phages possess Ig-like

protein domains on their capsids that enhance association with mucus [78]. This or

alternative localization domains may be useful for improving residence time in the gut

or helping concentrate phages to relevant biogeographies. Phage engineering efforts

have included altering host adsorption factors to change host range [79] or encoding

a dispersal enzyme to help break up bacteria in protective biofilms [14]. Phages have

also been used as DNA delivery agents to reverse antibiotic resistance [27, 28] or to

exert broad-spectrum [12, 23, 80] or sequence-specific [81, 82] antimicrobial activity.

Additionally, genome engineering and tools such as CRISPR-Cas [57] and assem-

bly methods including Gibson [58] and yeast [79] assembly should prove useful for

the development of new phages with augmented capabilities in modulating microbial

communities. We believe that the use of natural or engineered phages as therapeutics

for microbiota-related diseases has been understudied relative to the complementary

modality of introducing natural or engineered microbes, and thus represents a fasci-

nating area of investigation.

1.6 Chapter Overviews

This thesis is presented as a series of chapters focusing on different aspects of de-

veloping bacteriophage therapeutics for targeting bacteria. In Chapter 2, the premise

is to isolate novel phages by beginning with a panel of bacteria one wishes to tar-

get. The focus of this chapter is on carbapenem-resistant K. pneumoniae and phages

are collected from unprocessed sewage. These new isolates are characterized for host

range on the panel of bacteria and morphology is assessed through visualization by

transmission electron microscopy (TEM). One bacteriophage is selected for in vivo

testing in a murine gastrointestinal colonization model and shown to elicit a significant

decrease in bacterial load, though clearance was not achieved.

Whereas Chapter 2 focuses on the use of wild-type phage isolates, Chapter 3

introduces the concept of modifying phages for use in therapeutics, specifically utiliz-

24 Chapter 1 Citorik

IiIlll||Nlll'lillil | MMil||||||1|(|1|||||| Bacteriophages for Human Health ing a phage engineering platform to create obligately lytic derivatives from wild-type temperate phages. Prophage prediction is used to mine a panel of K. pneumoniae strains for potential temperate phages residing within the bacterial genomes and phage 4Kpn852 is successfully isolated (formerly referred to as pKPN-852). This temperate phage, induced from host strain KPNIH31, is visualized by TEM, assayed for host range, and genomic DNA is extracted in order to polish the ends of the previously submitted sequence. Though initial efforts encountered roadblocks during the engineering pipeline, a highly related phage of E. coli, phage N15, was targeted for the same modifications planned for TKpn852. A null mutation introduced into the major repressor protein of N15 resulted in the generation of an obligately lytic derivative as evidenced by the clear plaque phenotype. Additionally, the platform was used to interrupt the repressor gene with a fluorescent reporter gene in a simultaneous knock-out/knock-in approach.

In Chapter 4, the creation of modified bacteriophages is shifted from modifications of the phage genome to replacement with an alternative, engineered payload.

In brief, phage M13 is loaded with a DNA construct encoding a CRISPR-Cas system programmed to recognize and cleave undesirable genetic signatures in recipient bacteria. These phages encoding RNA-guided nucleases (4RGNs) are able to elicit specific killing in bacterial strains harboring antibiotic resistance and virulence determinants, while having no ill effects toward similar or even otherwise isogenic strains lacking the determinants. The IRGNs are applied to demonstrate specific remodeling of a three- strain microbial consortium in vitro and shown to significantly improve survival of

Galleria mellonella larvae when challenged with enterohemorrhagic E. coli 0157:H7.

These experiments show the potential of 4RGNs for development as anti-infectives and for manipulating complex microbial communities.

By exploring multiple pathways for the development of phage-based therapeutics, this work seeks to improve the opportunities for its successful translation.

25 Chapter 1 Citorik Chapter I 25 Bacteriophages for Human Health

Chapter 1 Citorik 26 Chapter 1 Citorik Chapter 2

Phages from Without: Isolation and Characterization of Phages from the Environment

Abstract

Bacteriophages represent an alternative class of antimicrobials that may have specific utility against pathogens for which antibiotics have become ineffective. In order to facilitate phage-based approaches to combat multidrug-resistant K. pneumoniae, bacteriophages were isolated from sewage against a small panel of highly resistant bacterial strains. Phage isolates were characterized for morphological traits and the host range of inhibitory effects, indicating a diversity of infectivity towards the different strains. A selected phage against the carbapenem-resistant strain KPNIH1 was assayed in a murine gastrointestinal colonization model and demonstrated a significant reduction in K. pneumoniae load, though total bacterial clearance was not observed.

27 Bacteriophages for Human Health

2.1 Introduction

The Gram-negative bacterium Klebsiella pneumoniae is a silent colonizer of the mammalian gut that, under some conditions, has the potential to escape to other body sites and cause infectious pathologies [83]. The emergence of extensively drug- resistant strains, including at least one clinical case in the U.S. in which the isolate was resistant to all available antibiotic treatments [84], has placed this microbe as an urgent public health concern [85]. Particularly for cases where the current antimicrobial arsenal is dry, research and development of bacteriophage therapies is an important area of exploration for alternative or next-generation therapeutics.

The genomic bases of antibiotic resistance and virulence determinants that allow

K. pneumoniae to be a successful pathogen have been studied in detail [85]. One especially important factor is the polysaccharide capsule that the organism produces to protect itself from immune and other challenges, which has been intimately tied with virulence [86]. Isolates of K. pneumoniae are notorious for variability in capsule composition and synthesis [87, 88], which poses an issue for developing natural immunity, vaccination, and even phage therapies, which are often highly specific to capsule type. Efforts to isolate phages against this organism often identify new capsule types to which these phages are specific [89-91]. In light of this, phage collections capable of targeting as many strains as possible will prove crucial toward the development of phage therapies against K. pneumoniae.

In 2011, an outbreak of carbapenem-resistant K. pneumoniae occurred when an index patient at the National Institutes of Health (NIH) Clinical Center led to the colonization of 18 patients and 6 deaths from bloodstream infections [92]. A collection of these clinical isolates was subjected to whole genome sequencing [93], and a panel of these KPNIH strains was acquired for the development of bacteriophage- based therapeutics. Here, a collection of novel bacteriophages was isolated against these outbreak strains in order to further efforts toward translational phage therapy.

While this chapter focuses on isolating wild-type phages, the subsequent chapter will introduce engineering efforts for K. pneumoniae phage therapeutics. Not covered

28 Chapter 2 Citorik Bacteriophages for Human Health in this work is the potential of capsule-degrading enzymes that can be mined from phages and used as standalone treatments [941, though this represents a potential future direction for developing agents based on the newly isolated phages.

2.2 Isolation of Novel Bacteriophages

Since K. pneumoniae is often found residing within the gastrointestinal tract, sewage was chosen as a potential source for isolating bacteriophages against the organism. In brief, raw sewage from the City of Cambridge was collected and filtered to remove any cells or larger debris. To acquire phages against a particular bacterial strain, filtered source material was first enriched in liquid culture with the target strain to amplify any infectious phages present, then individual phages were detected by spotting samples onto lawns of the enrichment strain and assessing for formation of visible plaques (Fig. 2.1). Phage isolates were serially passaged in order to ensure populations contained single phages. This process was repeated as necessary to attempt to improve strain coverage (Table 2.1).

After serial purification, bacteriophage isolates were spotted onto their respective indicator host strains in order to observe plaque morphology and confirm homogeneity prior to further characterization (Fig. 2-1). Plaque morphology can be an indicator of particular characteristics of a phage, where traits such as larger plaques may indicate a larger burst size (number of phages released from a single cell) and halos of lighter growth around plaques may be evidence of polysaccharide-degrading enzymes capable of altering host outer membrane structures. All phage preparations formed plaques on their respective propagation strains and no contaminants were observed.

2.3 Characterization of Bacteriophages

The majority of bacteriophages tend to be very specialized for the particular hosts with which they have evolved. This is believed to be a major advantage of phage therapy, in that there should be minimal off-target effects when attempting

Citorik Chapter 2 29 Bacteriophages for Human Health

Phage Name Bacteria Source Enrichment Propagation Material Host Host

<}KPNIH27 K. pneumoniae sewage KPNIH27 KPNIH27

<>K6-KpA K. pneumoniae sewage 1000527 K6.2

Table 2.1: Isolation of bacteriophages against K. pneumoniae: A panel of bacterial strains was used as bait to collect new bacteriophages from source material. Filtered sewage was enriched using the indicated enrichment hosts and phages were isolated and purified by serial plaque passaging. In some cases, a strain different from the enrichment strain was found to serve as a better host for the isolate phage and was thus used for continuing the passaging and for propagation, where relevant. to treat a specific infection. On the other hand, however, phages are restricted to infecting only specific isolates of any given target, meaning a single phage is unable to provide comprehensive coverage for all strains of a particular bacterial species. It

30 Chapter 2 Citorik Bacteriophages for Human Health

100 10-1 102 10- 10-4 10- 10- 10-7 100 10-1 102 10- 10-4 10- 10- 10-7 OKPNIH1-1 4K6-1

POW _ w W Q)KPNIH24-29 OK6-KpA W

0KPNIH27 I K6-B

4'KPNIH29-2 I K6.2-KpC1 DKPNIH30 I K6.2-10S

OKPNIH31-A @ @ * 0K6.2-527S

OKPNIH33 ISIS 4VK6-31A

OKp527-3_1 I 1 : 4)K6-31C

4 0Kp527-2 * 9* cOKp527-3_2

4K6.1-527S . 4,Kpn852*

Figure 2-1: Plaque assay on propagation hosts: Bacteriophages isolated from sewage on strains of K. pneumoniae were serially passaged at least 6 times to ensure homogeneity of the population. Lysates generated from purified phage isolates were spotted onto double-agar plates of the strains used for propagation in order to check for contaminants and observe plaque morphology.

is, therefore, imperative to know the host range of a bacteriophage to understand its

effectiveness toward the various strains that may be relevant for a given pathogen.

Kp K6.1 Kp K6.2

Figure 2-2: Mutant derivative K6.2 shows probable capsule alteration: An isolation streak of K. pneumoniae resulted in the appearance of a colony morphotype with a reduction of mucoid appearance versus the parental colonies. The opaque (K6.1) and translucent (K6.2) colonies were restreaked for isolation and used to com- pare the infectivity of various phage isolates.

Citorik Chapter 2 31 J

Bacteriophages for Human Health

K. pneumoniae Strains

o) m r- M~ oC' r-( r.LO UNC) r'

o a~o~ n O~0 O 0C ) O r- - El ".EE* 4KPNIH1-1 EL.a. a.5-a-a.aEpEaE8C ciKPNIH24-2 PKPNIH27 4KPNIH29-2 4'KPNIH30 DKPNIH31-A* 4KPNIH33 U) (DK6.1-527S '5 0 U) OK6-1 U) 0) DK6-KpA - K1, 'U 0. 0K6-B 0 0 0K6.2-KpC1 U EU 0K6.2-10S 4K6.2-527S 4K6-31A UM- "EMEE. ',t' **UEES*EE 4K6-31 C cIKp527-3_1 4Kp527-3_2 MM .UEB r .__ME. 4IKp527-2 4tKpn852

Figure 2-3: Double-agar spotting assay for determining host range of inhibitory activity: Newly isolated bacteriophages were assayed to determine their range of infectivity beyond the initial enrichment and propagation hosts. Double-agar plates were prepared with all bacterial hosts and phage lysates containing around 10 or 10" PFU/mL were spotted on all strains. Spots shown here are representative of the qualitative results of three independent assays. *Phage

To analyze the host range of the bacteriophages isolated above, a spotting assay was performed for each phage across the panel of bacterial isolates. Results indicated

32 Chapter 2 Citorik Bacteriophages for Human Health that the isolates possessed uniqueness in terms of the breadth of strains on which they were capable of forming zones of inhibition or plaques (Fig. 2-3). Of particular interest, the bacterial strain K. pneumoniae K6.2 was isolated from K6 as a colony appearing less opaque than the parental (K6.1) strain (Fig. 2-2). These strains showed a marked difference in both host range and apparent efficiency of bacterial killing as evidence by phage host range assays (Fig. 2-3). Strain K6.2 is anticipated to have a mutation resulting in a reduction or change in capsule production, which seems to render it much more susceptible to infection by all phages that infected both strains, with the exception of the temperate phage

2.4 Murine Gastrointestinal Colonization Model

In collaboration with the Segre Lab at the National Human Genome Research Institute (NHGRI), selected bacteriophages were subjected to initial testing in an established mouse model of gastrointestinal colonization by K. pneumoniae MKP103,

Citorik Chapter 2 33 Bacteriophages for Human Health

Normal Triggered Normal Triggered

I a- 04

(%4

0 CO) 29 CL)

CO) I

C\1 p

CO CJ

Figure 2-4: Visualization of selected bacteriophages by transmission electron microscopy (TEM): Phage samples were loaded onto FCF200-Ni grids and stained with 2% uranyl acetate. Grids were visualized on a Tecnai Spirit Transmission Electron Microscope (FEI) operating at 80 kV.

Citorik Chapter 2 34 Chapter 2 Citorik Bacteriophages for Human Health

Capsid Capsid Tail Tail Contract Predicted Phage Name Length Width Length Width Observed Family

,DKPNIH1-1 115 - 181 20.6 yes Myoviridae 4)K6-1 106 70.4 111 21.7 yes Myoviridae 4KPNIH24-2 94.3 - 114 20.3 yes Myoviridae bKPNIH27 90.3 - 117 20.3 yes Myoviridae 4KPNIH29-2 113 75.5 113 22.3 yes Myoviridae 4KPNIH30 73.9 - 119 19.1 yes Myoviridae

DKPNIH33 66.3 - 162 12.8 no Siphoviridae 4K6-B 60.9 - 167 13.2 no Siphoviridae

Table 2.2: TEM morphological characterization: Ten bacteriophage isolates were subjected to visualization via transmission electron microscopy (TEM) to determine morphological characteristics. Measurements were taken during image acquisition using AMT Capture Engine (AMT, Woburn, MA). Values are given in nm and represent the means of at least ten measurements. a derivative of strain KPNIH1 with blaKPC-3 deleted from plasmid pKpQIL [98]. In this model, mice are first subjected to antibiotic treatment to disrupt the normal microbiota and improve colonization, then the bacteria are introduced via gavage and remain detectable for at least 7 days. Since 4KPNIH1-1 was previously isolated and shown to efficiently infect the parental strain K. pneumoniae KPNIH1, this phage was selected for testing efficacy in vivo. To assay for reduction of colonization by Kp, bacteriophages were delivered by gavage 2 h prior to bacterial administration and again two days later, and bacterial and phage titers were monitored by performing viable counts on fecal samples taken 1, 2, 4, and 7 days post-infection (Fig. 2-5). When administered by gavage, 4KPNIH1-1 treatment was found to significantly decrease the levels of K. pneumoniae MKP103 detectable by viable cell counting at 2 (p<0.002), 4 (p<0.0 2), and 7 (p<0.01) days post-infection (Fig. 2-6). Though a significant decrease was observed in bacterial counts with phage admin-

Citorik Chapter 2 35 Bacteriophages for Human Health A1 -i U__ I 4, I I &I it Antibiotics Cocktail

Figure 2-5: Murine gastrointestinal colonization model: A model for colonization of C57BL/6J mice by K. pneumoniae has been previously established. Mice are given an antibiotics cocktail for 7 days prior to oral gavage of bacteria to facilitate colonization of the gut. For phage therapy experiments, the bacteriophage preparation ("

0 Control * OKPNIHI-1

Figure 2-6: Phage therapy reduces K. pneumoniae levels in murine gut: Bacteriophage

36 Chapter 2 Citorik Bacteriophages for Human Health gavage at 2 days post-infection, it was observed that levels were mostly undetectable by Day 7 (Fig. 2-7). Future efforts should be aimed at determining the cause of this low residence time in order to improve treatment and decolonization.

109-

0 108 IL. 107-

c 105 104 0101 eg sos

10 P' s9: * Q

lime Post-Inoculation

Figure 2-7: Phage

2.5 Discussion

Bacteriophages are a resource that could prove invaluable as therapeutics against

bacteria for which treatment options are rapidly dwindling. Since they are dependent

on their bacterial host(s) for providing the means for propagation, they are highly

evolved to efficiently deliver DNA into host cells. Unlike antibiotics, which require

extensive screening of microbes and compound libraries to uncover even candidate

molecules, bacteriophages can readily be isolated from an appropriate source material

where they would be expected to coexist with their hosts. As such, there is no shortage

of phages, since they are continually evolving to successfully infect and propagate on

continually evolving hosts.

Here, a collection of bacteriophages against carbapenem-resistant K. pneumoniae

was isolated and shown to elicit inhibitory effects on various combinations of the

Citorik Chapter 2 37 Bacteriophages for Human Health bacterial panel. After characterization of these new isolates, one phage was selected for initial experimental treatments aimed at reducing K. pneumoniae colonization of the gut in a murine model. Though a significant reduction was observed, the results seem to imply that further optimization of the phage would be useful.

A potential major limitation of translating phage therapies from the lab into the clinic is the differences that can be observed in infectivity of the phage toward its bacterial host in the lab and in an animal. For example, bacteria in the gut will be in different growth conditions resulting in potentially altered receptor expression or a growth state not permissive to phage infection. One study showed that E. coli phages

T4 and T7 differed in their ability to infect and propagate in vivo, despite both being extremely efficient bacteria killers in vitro [72]. Additionally, studies have shown the potential for stable coexistence of phage and their bacterial hosts, which contrasts with the desired mutual exclusivity for using a phage therapeutic [73]. It has been suggested that some bacteria may be protected from potential phage predators by localization within mucus [99]. The ability to engineer phages, which is the focus of the next two chapters, may permit the optimization of wild-type phage isolates, through methods such as appending mucus localization domains as has been found in the proteins of some phage capsids [78]. Optimization for gut residence time may also be possible through serial evolution of potential phage candidates, as has been performed previously to improve phage residence time in the blood through repeated passaging [13]. Resistance of bacteria to phages will not be discussed in detail here, but phage evolution and cocktail approaches are commonly accepted approaches to decrease the rate of occurrence. The ability to isolate, engineer, and evolve phages is a great benefit of their biological nature as compared to traditional chemotherapeutic antimicrobial approaches.

Citorik Chapter 2 38 Chapter 2 Citorik Bacteriophages for Human Health

2.6 Experimental Details

Strains and Culture Conditions

Unless otherwise noted, bacterial cultures were grown at 37'C with LB medium (BD Difco) containing carbenicillin at 100 pg/mL.

Strain Source Capsule Confidence

KPNIH1 Segre Lab KL107 very high

KPNIH10 Segre Lab KL107 very high

KPNIH24 Segre Lab KL106? good

KPNIH27 Segre Lab KL37+- good

KPNIH29 Segre Lab KL64* very high

KPNIH30 Segre Lab KL106? good

KPNIH31 Segre Lab KL27 very high

KPNIH32 Segre Lab KL107 very high

KPNIH33 Segre Lab KL107 very high

K6 ATCC 700603 KL6* very high

K6.1 K6 derivative ?-

K6.2 K6 derivative ?-

1000527 ATCC BAA-2146 - -

1100975 ATCC BAA-2472 - -

1002565 ATCC BAA-2470 - -

1100770 ATCC BAA-2473 - -

Table 2.3: Bacterial strains for phage isolations: Klebsiella pneumoniae strains used in these studies. Capsule and Confidence are outputs from the Kaptive algorithm for prediction of capsule synthesis loci [1001. This algorithm was only applied to the strains for which genome sequence is available.

39 Chapter 2 Citorik Chapter 2 39 Bacteriophages for Human Health

Bacteriophage Isolation

Since K. pneumoniae is often found residing within the gastrointestinal tract, sewage was chosen as a potential source for isolating bacteriophages against the

organism. In brief, raw sewage was acquired from the City of Cambridge (Port-

land St and Washington St) in collaboration with the MIT Underworlds project

(http://underworlds.mit.edu/) and filtered through a 0.2 Am membrane to remove

any bacterial cells or larger particles. Filtered source material was added to log-

phase cultures of the bacterial strains being used as targets (Table 2.3) and growth

was continued at 37'C for 24-48 h with gentle agitation. After the enrichment period, cultures were harvested by centrifugation and the supernatants filtered through 0.45

pm to remove any remaining living cells from the samples. Filtrates were stored at 40C.

To check for the presence of infectious phages from enrichments, samples were

serially diluted and spotted onto double-agar plates of host bacterial strains. Spots

were allowed to absorb into the agar and plates were wrapped and incubated at 37'C

overnight. Plates were checked for the appearance of plaques, observable as holes in

the bacterial lawn, or any zones of inhibition as indications of activity against the

target strain. For isolation and purification of phages, plaques were picked with a

pipette tip into SM buffer (50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 10 mM MgSO 4 ) and incubated at 55'C for 15-30 min to encourage resuspension into the buffer and

aid in prevention of bacterial contamination from the host lawn. Plaque suspen-

sions were then diluted and spotted onto double-agar plates of the host bacteria, as

above. This serial passaging process was typically performed for 6 consecutive plaque

resuspensions to ensure the presence of a single phage isolate in the final suspensions.

Bacteriophage Propagation

In order to generate higher volume and higher titer phage lysates from phage sam-

ples, final plaque resuspensions or previous lysates were used to infect liquid cultures

of early log-phase bacterial host strains. Infections were performed at 37'C, 250 rpm

40 Chapter 2 Citorik

M Bacteriophages for Human Health and incubations were continued until lysis was observed, as evidenced by a clearing of the liquid culture, or up to overnight for non-clearing cultures. Infected cultures were then harvested by centrifugation, filtered through 0.45 Am, and stored at 4'C.

When higher concentrations of phage stocks were required, lysates were subjected to polyethylene glycol (PEG) precipitation overnight (see Chapter 3) and then pelleted and resuspended in a lower volume of SM Buffer.

Visualization of Phages by TEM

PEG-precipitated phage samples were loaded onto Formvar/carbon film-coated nickel grids (FCF200-Ni, Electron Microscopy Sciences). Negative staining was performed using 2% uranyl acetate. Grids were visualized on a Tecnai Spirit Transmission

Electron Microscope (FEI) operating at 80 kV. For measuring phage aspects, at least

10 phages were used to determine average sizes.

Murine Decolonization Model

The murine model for phage experiments was conducted in collaboration with the

Segre Lab at the National Human Genome Research Institute (NHGRI). To aid in colonization establishment, 8 to 12 week old female C57BL/6J mice (Jackson Labora- tory) were first administered a cocktail of antibiotics (metronidazole 1 g/L, neomycin

1 g/L, vancomycin 0.5 g/L, ampicillin 1 g/L) in drinking water for 1 week prior to bacterial inoculation. Two hours prior to bacterial administration, mice were gavaged with PBS or 1010 PFU of the test phage diluted in PBS. Subsequently, 106 CFU of K. pneumoniae MKP103 was gavaged to establish colonization (Day 0). Fecal samples were harvested at 1, 2, 4, and 7 days post-infection and used to determine bacterial or viral load by homogenizing in PBS and plating dilutions onto LB agar for CFU or onto double-agar plates prepared with MKP103 for PFU. Viable counts for bacteria and phage were expressed as CFU or PFU per gram of fecal material.

Citorik Chapter 2 41 Bacteriophages for Human Health

Data analysis and Statistics

All data were analyzed using Prism 7 for Mac OS X, version 7.Oc (GraphPad Software, San Diego, CA, USA, http://www. graphpad.com/).

Chapter 2 Citorik 42 Chapter 2 Citorik Chapter 3

Phages from Within: Utilization of Prophages for Bacterial Targeting

Abstract

A main selection criterion for traditional therapeutic bacteriophages is an obligately lytic lifestyle. Those phages capable of entering into the quiescent lysogenic cycle may not always kill the target bacterial host and can even facilitate horizontal transfer of nondesirable genetic information, including antibiotic resistance and other virulence determinants. These temperate phages, however, represent a vast reservoir of untapped potential and can often be found residing within bacterial genomes in the rapidly growing resources of metagenomic datasets. Here, a method for engineering bacteriophages is applied toward the conversion of a temperate phage of E. coli into one that is strictly lytic. Additionally, a related phage candidate capable of infecting carbapenem-resistant K. pneumoniae is identified and shown capable of being rebooted from purified genomic DNA in a laboratory cloning strain, though current inefficiencies in the process necessitate further optimization in order to subject the new phage to the yeast-based assembly pipeline for genome engineering in the absence of alternative assembly approaches.

43 Bacteriophages for Human Health

3.1 Introduction

In general, bacteriophages can be split into two groups based on the lifecycle deci- sions that may occur once the genetic material has been ejected into a host bacterial cell (Fig. 3-1). Virulent, or strictly lytic bacteriophages, are so named because they are exclusively programmed to begin the typical route of commandeering host resources to produce progeny virions and ultimately cause lysis of the cell. Temperate, or lysogenic, phages, on the other hand, immediately encounter a binary decision that may result in either this lytic phase or a lysogenic phase, in which the phage genome is forced into dormancy and often integrated into the bacterial genome. A bacterial lysogen subsequently replicates this prophage along with its own chromosome, and daughter cells will each receive a copy. Under a variety of signals, typically those indicating stress to the bacterial cell, a prophage may switch into the lytic phase, excising itself from the genome and replicating progeny phage particles to be released from the cell upon enzyme-mediated lysis.

Lysis

Lysogeny

Figure 3-1: Two disparate lifestyles for bacteriophages: (i) Obligately lytic phages infect a host bacterial cell and immediately begin the programmed propagation that leads to eventual lysis of the host. (ii) Alternatively, temperate phages may enter into a relatively dormant state in which they are maintained with the bacterial genome. (iii) Various environmental triggers may result in de-repression of the prophage that leads to excision and activation of the lytic program.

Within the field of phage therapy, it is generally accepted that being obligately lytic is a desirable characteristic of any candidate phage [71, 101]. A phage entering into lysogeny will not actually kill the infected cell and often provides superimmunity to the bug, meaning immunity against subsequent phage infections of the same or

44 Chapter 3 Citorik Bacteriophages for Human Health related viruses. Additionally, this capacity to integrate into the bacterial genome also carries with it the potential to facilitate horizontal gene transfer. As the result of recombination or inappropriate packaging into progeny virions, phages have been found to carry elements including virulence factors [102] and antibiotic resistance genes [31], and through lysogenization of the bacterial host can confer these undesirable traits to the bacterium. Pathogens including Vibrio cholerae and enterohemorrhagic E. coli may possess their pathogenic potential as the direct result of elements, such as the Ctx and Stx toxins that these organisms produce, respectively, that are encoded within prophages in the bacterial genomes. The potential for these phages to become integrated into the host genome thus is a characteristic best avoided when considering agents for therapeutic purposes. The bacteriophage A, first described by Lederberg in 1951, is an extremely well- studied temperate phage that has had major contributions to the understanding of molecular biology [103] and creation of modern genetic tools [2]. The genetic switch responsible for this phage's lysis vs lysogeny decision has since been described and reviewed in great detail [104]. The robustness of this fate-determining element has even lead to its use in the modern field of synthetic biology, including the creation of engineered E. coli strains with synthetic gene regulatory networks such as a bistable toggle switch [105] or a synthetic oscillator [106]. Though the fate decision of temperate phages involves multiple elements of varying influence operating from the transcriptional to protein activity levels, the ultimate determinant of the decision is the amount of a major repressor protein in the cell, responsible for controlling gene expression levels from the phage genome, which may interact directly or indirectly with one or more antirepressors also encoded by the phage [107]. The discovery of lysogeny determining elements was first made by isolating and studying spontaneous phage mutants that formed clear plaques on a bacterial lawn instead of the normal turbid plaques associated with lysogenic conversion by the parental phage [108]. Within some environments, including the human microbiome, it has been found that temperate phages appear to be more abundant than strictly lytic phages [32]. Additionally, it has been observed that lysogenic phages can naturally evolve into

Citorik Chapter 3 45 Bacteriophages for Human Health strictly lytic derivatives [109]. With these pieces of information, temperate phages represent a major resource for the development of agents that target bacteria within the environments in which they are found. In order to make use of this abundant resource, however, the described negative aspects of temperate phages must first be circumvented. Recently, a method for capturing and modifying bacteriophage genomes was described, which makes use of an orthologous host, the yeast Saccha- romyces cerevisiae, that can be used for efficient assembly of DNA pieces through homologous recombination and stably maintain entire phage genomes that would be toxic within bacteria [59, 79]. This method allows for targeted modifications to be made to the phage genome once it has been stably captured in the yeast strain or preemptively during the capture process (Fig. 3-2).

Extract Amplify Mix w/ Genome P_ YAC tYeast * Assembly Reboot Transform Extract @ *- (0YACJ E. coli EYAC

Figure 3-2: Overview of yeast assembly for phage genome reconstruction: Phage DNA is first acquired to be used as template material (purify from phage or bacterial genome, or perform total or partial synthesis). The genome is amplified in fragments such that each overlaps with the next in order to facilitate assembly, with the ends overlapping with the acceptor arms of a YAC amplicon. Fragments, in addition to the YAC backbone, are pooled and transformed into yeast cells, where they are assembled into a

In this chapter, the approach used for modifying temperate phages for therapeutics is to remove the genetic potential for lysogenization and force the phage's natural lytic

46 Chapter 3 Citorik Bacteriophages for Human Health pathway, though the alternative payload approach featured in the Chapter 4 could also be applied.

3.2 Discovery and Characterization of 4Kpn852

A potential benefit of utilizing prophages for generating targeting agents is that they can be found within the genome of the bacterial lysogen, indicating that the bacterium was, at least at some point, susceptible to this phage [110]. Addition- ally, DNA sequence databases represent an ever-growing resource for the discovery of prophages that may be linked to a particular bacterial host. The PHAge Search Tool (PHAST) is a freely available search algorithm that facilitates prediction of prophage elements and complete prophages from a sequenced genome or other DNA input (http://phast.wishartlab.com/). In order to mine the K. pneumoniae panel for potential prophages, the web-based PHAST tool was used to predict the presence of any hypothetically complete prophages present within the strains, meaning the region contains all expected cornerstone genes (replication, structural, functional) [111]. Of interest, K. pneumoniae KPNIH31, which had been previously sequenced [93], contained a 51.6 kb linear DNA molecule (pKPN-852) that was predicted to be an intact prophage. To explore the viability of the phage, a culture of KPNIH31 was grown to early-log phase and induced with mitomycin C, a DNA damaging agent known to induce prophages [112]. Supernatant from induced culture was harvested by centrifugation and filtration to remove any remaining bacterial cells, then spotted onto double-agar plates of the K. pneumoniae panel. Plaque formation on multiple strains indicated that there indeed was an inducible phage in strain KPNIH31 (Fig. 3-3), which was subsequently named

Citorik Chapter 3 47 Bacteriophages for Human Health ble 3.1). Capsule specificity is not a surprising barrier to host range, and the fact that this phage hit multiple strains from the panel indicates its potential merit as a therapeutic.

100 101 10-2 10- 10-4 10-

Lysate from induction of KPNIH31

Figure 3-3: K. pneumoniae KPNIH31 harbors a viable prophage: When induced by mitomycin C, K. pneumoniae KPNIH31 produces a phage capable of forming plaques, shown here spotted on KPNIH33 as an indicator.

Strain CPS Susceptible Type

KPNIH1 2 Yes KPNIH10 2 Yes KPNIH24 1 No KPNIH27 ? No KPNIH29 ? No KPNIH30 1 No KPNIH31* ? No KPNIH32 2 Yes KPNIH33 2 Yes

Table 3.1: Infectivity of KPNIH strains by 4DKpn852: Temperate phage

In order to characterize the isolated prophage, its genome was first purified by PEG precipitation of phage lysate and nuclease treatment to remove contaminating bacterial gDNA, followed by DNA extraction from viral particles. To complete the

48 Chapter 3 Citorik Bacteriophages for Human Health genome, primers were designed from the existing sequence data and used for additional Sanger sequencing from purified genome. As a result of the predicted lifecycle of this covalently closed, linear dsDNA [114], DNA purified from viral particles permitted access to areas of the genome that were less accessible when the prophage was originally sequenced from within its bacterial host. As a result, 139 bp were added to the beginning of the published sequence and 1,498 bp appended to the end (Fig. 3-4). Studies covering various linear, dsDNA temperate phages have previously compared the genomes of this rare group [115]. Given consistencies in genomic organization, the major repressor from DKpn852 was easily predicted and confirmed by alignment (Fig. 3-5B). Additionally, the 10 bp cos site was predicted and an extremely large tail tip gene consisting of 12,705 bp identified by comparison to 4iKO2, which was previously shown to possess a similarly large (10,302 bp) version of that from phage N15 (3,186 bp) [115].

5.qoo 10.000 1spoo 20.POO 25.ODO 30.900 35.00 40.900 45.900 50900 s3.2s9

I, yNew End (+1,498 bp) New End (+139 bp)

CTTC(G I IGTAGATTTT cos site

Figure 3-4: Polishing of the 4Kpn852 genome: Existing sequence information for pKPN-852 was used to design primers for sequencing holes in the genome of 41)Kpn852. As a result, a total of 1,637 bp was added to the genome. The 10 bp cos site was also identified by comparison to phage 4KO2 and is indicated blown out below the genome.

Phage DKpn852 was next visualized using TEM in order to observe morphological characteristics. This phage was found to possess a long, flexible tail that appeared to be non-contractile, placing it into the Siphoviidae in the company of phages including the classical temperate phage A.

49 Chapter 3 Citorik Chapter 3 49 Bacteriophages for Human Health

2000 4'00 6600 8600 1OoW o 14600 16600 18600 20600 22600 24600 26600 28600 300 32600 34600 36600 38600 40600 42600 400 460C

NIS Virion Genorne (AF064539) R 26'00 46'00 6600 Bdoo 1o6oo i20oo 14600 16600 18600 2D600 22600 24600 266DO 2800 32600 34600 356o0 38600 40600 42600 44600 46600 48600 50600 52600

.h n .: o m1 T--D | if ~rrui~~ ~LJEDonJq~w a ~J2 phiKpn8S2 Virion Genome (no cos) A

1 50 100 150 200 250 300 350 400 450 500 550 609 Consensus inenan mm. =NMI= Identity

ti ll III I 1 1i1g f il lilt Ii I HI 11 1 1 ini I III Neill 1111 1 1tt 111I $ s il II I 1 11 r# 1.Kpn852cB U an 1I IfII a moa soama I I I ill'I 1 l1 1 11 11 1 o I I li fl It 11 1 1 1 8111 I If 1I toEI1 lilt I I il 2.N1ScB aum2n a a a meI a anamMm a' 8 1

Figure 3-5: Phage

3.3 Construction of Lytic Phage Derivatives

In order to create the engineered obligately lytic version of 1DKpn852, primers were designed and utilized to amplify overlapping fragments of the phage genome, with homologous arms being added to facilitate assembly into the YAC backbone for maintenance in S. cerevisiae. Though genomic DNA appeared to be successfully captured in this way, purified

50 Chapter 3 Citorik Bacteriophages for Human Health

50 nm E. coi K. pneumoniae N15 cOKpn852

Figure 3-6: TEM visualization of K. pneumoniae phage DKpn852 and E. coli phage N15: Phage N15 and 4DKpn852 induced from E. coli C-1 and K. pneumoniae KPNIH31, respectively, were precipitated by PEG/NaCl and subjected to negative staining by uranyl acetate (see methods for details). TEM images were captured at 120,000x magnification and scale bar represents 50 nm. of wild-type DNA, this process was quite inefficient and likely explains the inability to achieve success when booting from yeast-extracted YAC DNA (Fig. 3-7A).

A B

0 OKpn852 N15

Figure 3-7: Phage 4Kpn852 genome rebooting is inefficient in E. coli: Bac- teriophage genomic DNA was purified from phage lysates and electroporated into E. coli. After recovery, supernatants were filtered and spot-plated onto double agar plates prepared with (A) K. pneumoniae KPNIH33 or (B) E. coli C-1 indicator hosts for 4Kpn852 or N15, respectively. Plaques were observed for both phages, indicating successful rebooting in E. coli from purified genomic DNA. Arrows indicate the presence of isolated plaques for 4Kpn852.

Given N15 has previously been well-characterized (reviewed in [116]), lytic approaches were undertaken with this phage, which was shown to reboot efficiently in the E. coli host (Fig. 3-7B) and thus predicted to be more amenable to the yeast

Citorik Chapter 3 51 Bacteriophages for Human Health assembly pipeline (Fig. 3-2). It has been shown that the antirepressor AntC inter- acts with the major repressor CB in order to permit expression of genes necessary for lytic development [117]. Indeed, when antC was overexpressed within E. coli C-1, infection with wild-type N15 resulted in the formation of clear plaques, indicating a failure of lysogenization (Fig. 3-8).

E. coli C-1 wild-type

E. coli C-1 pZA31-antC

Figure 3-8: Overexpression of antirepressor antC in E. coli C-1 prevents lysogenization by phage N15: The antirepressor antC was amplified from phage N15 and cloned into a plasmid for constitutive expression. Whereas spotting of N15 onto the wild-type E. coli C-1 host resulted in the characteristic turbid plaques of lysogenic phage infection (top), spotting onto this same host containing the antC overexpression vector resulted in the formation of clear plaques (bottom).

In order to build this phenotype into the phage itself, a null mutation was introduced into cB, preventing the expression of the repressor (Fig. 3-9A-B). To facilitate the engineering approach, wild-type N15 was captured in S. cerevisiae through the yeast assembly method (Fig. 3-2) [79]. Wild-type N15 was shown to successfully reboot from the captured genome when extracted from yeast and transformed into E. coli (Fig. 3-7B). Next, primers were designed to introduce the desired mutation into cB during amplicon generation, with products again assembled in yeast and booted in E. coli. When spotted onto lawns of E. coli C-1, wild-type N15 showed the characteristic turbid plaques of a temperate phage, while the engineered phage yielded only clear plaques (Fig. 3-9D). This result was similarly achieved when using a restriction- based in vitro approach (Fig. 3-9A, right side), though this method depends on the presence of convenient restriction sites within the genome. Additionally, the yeast- based approach allowed the straightforward creation of a strictly lytic reporter phage expressing GFP (Fig. 3-9C-D).

Citorik Chapter 3 52 Chapter 3 Citorik Bacteriophages for Human Health

A C

Ia lb

2a4 42b

.1 10 20 30 40 50 60 AT cATTAA GCTATGAAAACATGA GCTAAC ,ACTGAAA CACUCCCJTTAC AGCTGAAACT AC T T

Figure 3-9: Engineering an obligately lytic derivative of temperate phage N15: E. coli phage N15 was modified in order to prevent lysogenic growth. A: Gen- eral strategy for genome modification and reconstruction. Purified DNA from N15 is self-ligated in- order to facilitate its use as template material. For yeast assembly (left side), overlapping amplicons with or without sequence modifications are generated from the phage genome (la) and transformed into S. cerevisiae for assembly (2a). For the in vitro restriction approach, an amplicon is generated containing the desired mutation(s) and enzymes are used to digest this amplicon as well as the phage genome (1b). Digested phage genomic DNA is ligated with the engineered amplicon to generate the full phage genome containing the cB null mutation (2b). For both approaches, the assembled phage genomes are finally transformed into E. coli (3) to facilitate phage rebooting (4). B: Approach for creating null mutations within the cB coding sequence. Primers used and point mutations generated are indicated below the mutant sequence. C: Illustration of genetic swap used to simultaneously create a reporter phage that is also obligately lytic. D: Spot phenotypes of N15 and mutant derivatives. N15 wild-type phage produces turbid clearings on E. coli C-1 (left) while N15 cB is obligately lytic and produces clear plaques and N15 cB::gfp also produces fluorescence.

Citorik Chapter 3 53 Bacteriophages for Human Health

3.4 Discussion

In the development of bacteriophages for therapeutic purposes, temperate phages are often immediately discarded as undesirable agents as the result of their potential for integrating into a potential target bacterium. At the same time, however, temperate phages in environments like the human microbiome are abundant and can readily be mined from bacterial genome sequences present in publicly available databases. In light of these two facets being at odds, methods for removing the negative aspects from temperate phages could prove incredibly useful. As a proof-of-principle, this work has demonstrated a pipeline for creating obligately lytic phage derivatives by deletion of the major repressor protein in order to deregulate the lytic cycle and prevent entering into a state of lysogeny. Phage N15 was successfully engineered in this manner, and a lytic phenotype was observed as contrasted to the temperate phenotype of the wild-type phage. Initial efforts to use this pipeline on the newly isolated K. pneumoniae phage

Chapter 3 Citorik 54 Chapter 3 Citorik Bacteriophages for Human Health

3.5 Experimental Details

Bacterial Strains and Growth Conditions

Unless otherwise noted, bacterial cultures were grown at 37'C with LB medium (BD Difco), with carbenicillin added for cultures of K. pneumoniae. Where indicated, antibiotics were added to the growth medium to the following final concentrations: 100 pg/mL carbenicillin (Cb) and 25 pg/mL chloramphenicol (Cm). Plasmid pZA31-antC was created by amplifying and cloning antC from the N15 genome into KpnI/BamHI sites in the pZA31 backbone [119] and transformed into E. coli C-1 to create the antirepressor overexpression host. Yeast cultures were grown using YPD with 2% glucose or synthetic dropout medium lacking leucine for maintenance of the pRS415 YAC and incubating at 30 C.

Mitomycin C Induction of Prophages

Bacterial lysogens were grown overnight to saturation and then subcultured between 1:100 and 1:500 into fresh LB broth and incubated at 37'C, 250 rpm until early to mid-log phase (2-3 hours). Induction was performed by adding mitomycin C to 0.5- 1.0 pg/mL and continuing incubation for at least 3 h to overnight. Induced cultures were harvested by centrifugation and filtered through 0.2 or 0.45 Pm membranes. All bacteriophage preparations were stored at 4'C.

Double-Agar Spotting Assay

To prepare double-agar plates, 200-300 pL of saturated bacterial cultures were added to 3.5 mL of molten 0.6% agar LB and immediately mixed and poured onto LB agar plates. Once top agar was solidified, 2.5 pL of sample dilution series (typically 10-fold) in PBS were carefully spotted onto the plate and allowed to absorb into the agar. Plates were wrapped and incubated overnight in the upright position.

Citorik Chapter 3 55 Bacteriophages for Human Health

Visualization of Phages by TEM

PEG-precipitated phage samples (see below) were loaded onto Formvar/carbon

film-coated nickel grids (FCF200-Ni, Electron Microscopy Sciences). Negative stain-

ing was performed using 2% uranyl acetate. Grids were visualized on a Tecnai Spirit

Transmission Electron Microscope (FEI) operating at 80 kV.

DNA Purification and Genome Polishing

Phage genomic DNA was purified from PEG-precipitated lysates. Briefly, phage

lysates were generated from lysogenic strains (as above) and treated with DNase

and RNase to remove contaminating bacterial genomic DNA. NaCI was added to the

lysates at 1-2.5 M final concentration, then PEG-8,000 was added to 10% and samples

were allowed to precipitate overnight at 4'C. The following day, samples were spun

at 12,000 x g for 30 min and the supernatant was aspirated. Pellets were resuspended

with SM Buffer (50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 10 mM MgSO 4 ) to 0.01- 0.1x of the initial lysate volume. Phage suspensions were stored at 4'C. Phage DNA

was extracted using the Zymo Viral Kit (Zymo Research). To finish the ends of

the genome that were not picked up from sequencing the bacterial host, primers for

reading into these regions were designed and submitted with virion-purified template DNA for Sanger sequencing to Quintara Biosciences.

DNA Software Tools

For prophage prediction, bacterial genomes were inputted into the PHAge Search

Tool (PHAST) at http://phast.wishartlab.com/. DNA analyses, including basic

alignments, BLASTs, and Mauve genome alignments, were performed using Geneious R9 (Biomatters).

Yeast Assembly of Phage Genomes

Yeast assembly was adapted from Ando et al. [791. Briefly, primers were designed

to amplify pieces from the phage genome with 30-100 bp of overlap to facilitate

56 Chapter 3 Citorik

Ililill lillI 'r I' I!!|" I|||Hillilliylli Bacteriophages for Human Health

assembly. For mutations in engineered phages, primers were designed to introduce

the changes during the amplification step. All amplifications were carried out using

Kapa HiFi polymerase (Kapa Biosystems).

To prepare competent yeast cells, S. cerevisiae BY4741 was cultured for 24 h

and then subcultured 1:10 into YPD and incubated for 4 h at 30'C. Cells were then washed twice with water and once with 0.1 M lithium acetate before resuspending in

a volume of 0.1 M lithium acetate to result in an approximately 100-fold increase in

cell concentration from the initial harvest. Transformation reactions were performed by combining 50 pL of cells with 50 pL of DNA, 240 pL 50% PEG-3,350, 36 AL 1 M

lithium acetate, and 25 pL of 2 mg/mL boiled salmon sperm DNA. Reactions were incubated for 30 min at 30'C and then 45 min at 42'C. Transformants were selected

by plating reactions onto leucine dropout medium and incubating for 24-72 h at 30'C.

Yeast colonies were restreaked for isolation and grown up in liquid synthetic dropout medium for 48 h. YACs were harvested using the YeaStar Genomic Extraction Kit

(Zymo Research) and electroporated into E. cloni 10G ELITE electrocompetent cells

(Lucigen) for maximal transformation efficiency. The E. coli transformant cultures were recovered for 4 h at 37 C, then supernatants were collected and filtered through

0.2 prm membranes to remove bacterial cells. Viable phages present in filtrates were detected by spotting samples onto double-agar lawns of the corresponding indicator.

Restriction/Ligation of Phage Genomic DNA

Restriction and ligation reactions were performed using enzymes and buffers from

New England Biolabs (NEB). For genome editing of N15, purified viral genome was first self-ligated using T4 ligase and incubating at 16'C for 16 h. XhoI and PvuI were used to digest the self-ligated phage genome and the desired amplicon. Digested amplicon was cleaned up by gel purification using the QIAPrep Spin Miniprep Kit

(Qiagen) and digested N15 was purified using the Zymo Viral Kit (Zymo Research).

Purified digested fragments were ligated with T4 ligase at 16'C for 16 h. Ligations were transformed into E. cloni lOG ELITE electrocompetent cells (Lucigen) and phage detection was performed as above.

Citorik Chapter 3 57 Bacteriophages for Human Health

Citorik Chapter 3 58 Chapter 3 Citorik Chapter 4

Phages born Anew: Using CRISPR Payloads to Create Sequence-Specific Antimicrobials

The following chapter is adapted from a previous publication to which Mark Mimee

and I contributed equally as first authors [82].

Abstract

Current antimicrobial strategies tend to be broad-spectrum in nature, leading to indiscriminate killing of commensal bacteria and accelerated evolution of drug resistance. In contrast, we propose a class of programmable-spectrum antimicrobials whose activity can be rationally customized against specific DNA sequences.

Using CRISPR-Cas technology, we create RNA-guided nucleases (RGNs) and show that highly efficient, autonomous delivery to microbial populations can be achieved through the use of bacteriophage (

We design RGNs that selectively target undesirable genes or polymorphisms, including antibiotic resistance and virulence determinants, in carbapenem-resistant Enter- obacteriaceae and enterohemorrhagic Escherichia coli. Delivery of

59 Bacteriophages for Human Health

enable programmable remodeling of microbiota.

4.1 Introduction

The emergence and proliferation of multidrug-resistant bacterial pathogens has

elicited mounting concern over the dwindling treatment options for these organisms.

Recently, carbapenem-resistant Enterobacteriaceae (CRE), a group of intestinal-

dwelling Gram-negative bacteria known to cause life-threatening opportunistic infec-

tions, were highlighted as one of three most urgent threats among antibiotic-resistant

bacteria [1201. Carbapenems have traditionally been reserved as a last resort treat-

ment for Gram-negative infections, but the spread of extended-spectrum -lactamases

has necessitated the increased usage of carbapenems and favored the emergence of

carbapenem-resistant strains refractory towards most or all current treatment options.

The responsible enzymes, including New-Delhi metallo-#-lactamase 1 (NDM-1), may

confer pan-resistance to -lactam antibiotics and are frequently co-harbored with ad-

ditional resistance determinants on mobile plasmids that facilitate rapid dissemina-

tion within and beyond Enterobacteriaceae [121]. The diversity of multidrug-resistant

bacteria compounds the difficulty of developing conventional treatments that target

pathogens and commensal reservoirs, but avoid non-specific broad-spectrum activity

and undesired pressure on non-target cells.

Here, we introduce an alternative antimicrobial approach that imposes direct evo-

lutionary pressure at the gene level by utilizing efficiently delivered programmable

RNA-guided nucleases (RGNs). We engineered the clustered, regularly interspaced, short palindromic repeats (CRISPR)-CRISPR associated (Cas) system, naturally em-

ployed in bacteria as a defense strategy against mobile elements [17, 122], to effect cell

death or plasmid loss upon detection of genetic signatures associated with virulence

60 Chapter 4 Citorik Bacteriophages for Human Health or antibiotic resistance. The Type II CRISPR-Cas system of Streptococcus pyogenes is an effective, programmable tool for genome editing and gene expression in a wide variety of organisms [51]. The specificity of CRISPR-Cas is dictated by short, spacer sequences flanked by direct repeats encoded in the CRISPR locus, which are tran- scribed and processed into mature guide RNAs (crRNA) [123]. With the aid of a trans-activating small RNA (tracrRNA), crRNAs license the Cas9 endonuclease to introduce double-stranded breaks in target DNA sequences [50, 123]. Through simple modifications of spacers in the CRISPR locus, an RGN can direct cleavage of almost any DNA sequence, with the only design restriction being a requisite NGG motif immediately 3' of the target sequence [50]. By packaging RGNs into bacteriophage particles or harnessing mobilizable plasmids, we implemented conditional-lethality devices with high specificity, modularity and multiplexability against undesired DNA sequences (Fig. 4-1).

crRNA

RNA-guided Nuclease (RGN) @1

Cell Death

Figure 4-1: RGN overview schematic: Bacteriophage-delivered RNA-guided nuclease (RGN) constructs differentially affect host cell physiology in a sequence- dependent manner. If the target sequence is: (i) absent, the RGN exerts no effect; (ii) chromosomal, RGN activity is cytotoxic (Fig. 4-5); (iii) episomal, the RGN leads to either (iiia) cell death (Fig. 4-5) or (iiib) plasmid loss (Fig. 4-8A), depending on the presence or absence of toxin-antitoxin (TA) systems (Fig. 4-8B), respectively.

61 Chapter 4 Citorik Chapter 4 61 Bacteriophages for Human Health

4.2 Transformation Assays for Validation

To establish RGN functionality in mediating sequence-specific cytotoxicity, we designed RGNs to induce double-stranded breaks in blasHv-is or blaNDM-1, which encode extended-spectrum and pan-resistance to #-lactam antibiotics, respectively [124, 125]. Transformation of plasmid-borne RGNs (pRGNs) into E. coli containing a chromosomal copy of these target genes resulted in nearly a thousand-fold decrease in transformation efficiency as compared to wild-type cells lacking the target (Fig. 4-2A). These results corroborate the mutual exclusivity between a functional crRNA and a cognate target locus [126, 127]. Sequence analysis of 30 escape mutants, cells that receive and maintain an RGN plasmid despite the presence of a target sequence, revealed that tolerance was exclusively due to a defective construct, frequently resulting from a spacer deletion within the CRISPR locus (Fig. 4-3). Furthermore, deletion of the tracrRNA as well as inactivation of the RuvC-like nuclease domain of Cas9(DOA) [50] abrogated the loss of transformation efficiency in cells that harbored a target sequence. Thus, a catalytically active endonuclease, tracrRNA and crRNA are necessary and sufficient to mediate sequence-specific cytotoxicity in E. coli (Fig. 4-1). Antibiotic resistance genes often reside on large, multi-copy plasmids capable of autonomous transfer in microbial populations, leading to horizontal dissemination of drug resistance [121]. RGN activity against high-copy plasmids was verified with a GFP-expressing, ColEl-derived vector containing a standard -lactamase selectable marker (pZE-bla7-gfp) [119] or blaNDM-1 (pZE-baNDM-1-gfp)- Vectors bearing this ColEl origin are reported to be present at copy numbers of 50-70 per cell [119]. Transformation of pRGNndm-1, a plasmid-borne RGN targeting blaNDM-1, into cells containing pZE-blaNDM-1-gfp led to a three-logo reduction in transformants retaining carbenicillin resistance, whereas transformation of pRGNndm-1 into cells containing target-free pZE-bla -gfp did not lead to a reduction in resistant transformants (Fig. 4- 2B). The activity of RGNs is therefore sufficient to exclude even high-copy antibiotic resistance plasmids from cells and can re-sensitize a resistant population to antibiotics.

62 Chapter 4 Citorik Bacteriophages for Human Health

Similarly, transformation of pRGNndm-1 into cells possessing pZE-blaNDM-1-gfp led to an approximately thousand-fold decrease in GFP+ cells, but no decrease was found with transformation of pRGNndm-1 into cells possessing pZE-blaz-gfp (Fig. 4-2C).

A 2 A.y

pRON I ------RNA4uIded 3 Nudesse (RON) Cl

WDeath 'V

%*f------10 EMG2 Wr EMG2::NDM-1 EMG2::SHV-18 e pRGNndrn-1 A pRONndrn-1 MrAcfRNA m pRGNshv-18 v pRGNndhn-1 Cas9Wm B 10' C 100- OPRGN RNA-Guldeda

& Resensitizatton I

G 10 pZE-brgp pZE-bI.4.,-gfp pbZ-Op PZE-aIjg.1-9?P 0 pRONndrn-1 U pRONndm r3 pRGNshv-18 M pRGNahv-18

Figure 4-2: Design and validation of programmable RGN constructs by transformation: (A) Plasmids pRGNndm-1, pRGNshv-18, pRGNndm-1 AtracrRNA, and pRGNndm-1 CaSD1OA were transformed into competent wild-type EMG2 (EMG2 WT) as well as otherwise isogenic strains containing chromosomally integrated blaNDM-1 (EMG2::NDM-1) or blasHV-18 (EMG2::SHV-18). Transformants were enumerated on LB+chloramphenicol (Cm) to select for pRGN transformants and to determine transformation efficiencies, which demonstrated the specific incom- patibility of an RGN construct and its cognate protospacer (n=4). (B) Plasmids pRGNndm-1 and pRGNshv-18 were transformed into EMG2 cells containing either pZE-blaNDM--gfp or pZE-blaz-gfp plasmids. Transformants, first selected in appropriate antibiotic media, were enumerated on LB+Cm or LB+Cm+carbenicillin (Cb) agar to calculate the ratio of transformants retaining Cb resistance (CbR) to total transformants. Error bars indicate s.e.m. of three independent experiments, each with three biological replicates (n=9). (C) EMG2 cells containing either pZE-blaz- gfp or pZE-blaNDM--gfp plasmids were transformed with pRGNndm-1 or pRGNshv- 18 plasmids and transformants were selected overnight in LB+chloramphenicol (no antibiotic selection for plasmid maintenance was applied). Plasmid loss was determined by calculating the percentage of GFP-positive cells following gating by forward and side scatter. Error bars indicate s.e.m. of measurements from three independent experiments, each with three biological replicates (n=9).

Citorik Chapter 4 63 Bacteriophages for Human Health

Tn10.10 Insertions (5) Deletion (19) nt Insertion (6) 3279

CRISPR

Mutation Number %

Deletion of spacer and one repeat 19 63.3

tracrRNA Insertion (A at nt140/141) 6 20.0

Transposon Insertion (TnlO.10) 5 16.7

Figure 4-3: Characterization of escape mutants that tolerated transformation of a cytotoxic RGN construct: EMG2::NDM-1 or EMG2::SHV-18 colonies that tolerated transformation of the pRGNndm-1 or pRGNshv-18 plasmids (Fig. 4- 2B) were re-isolated and sequenced to identify escape mutations. Spacer deletion in the CRISPR locus, point mutations in tracrRNA and transposon insertions in cas9 led to pRGN inactivation in successful transformants. Five escape mutants from three independent experiments were sequenced per strain (n=30).

4.3 Cell-Based Delivery

The viability of RGNs for antimicrobial therapy hinges on high-efficiency delivery

of genetic constructs to bacterial cells. We explored two mechanisms of horizontal

gene transfer naturally employed by bacteria to acquire foreign genetic elements: plas-

mid conjugation and viral transduction. Although constrained by requirements for

cell-cell contact, conjugative plasmids often possess wide host ranges and no recipient factors necessary for DNA uptake have been identified [128]. Efficient transfer of

RGNs was achieved using the broad-host-range plasmid R1162 mobilized by E. coli

S17-1, which contains the conjugative machinery of plasmid RP4 integrated into its chromosome. In filter mating experiments, conjugative transfer of RGNs elicited a

40-60-fold reduction in target carbenicillin-resistant recipient cells (Fig. 4-4B). Un- der selection for transconjugants, transfer of RGNs into recipients yielded a 2-3-logio reduction in target cells as compared to controls, suggesting that conjugation effi-

64 Chapter 4 Citorik Bacteriophages for Human Health ciency, as opposed to RGN activity, limits RGN efficacy in this context (Fig. 4-4C). Future work may be necessary to further optimize the efficiency of conjugation-based delivery vehicles for antimicrobials based on mobilizable RGNs.

a Conjugation

*m m m m m m m m o

RGN Donor Recipient

b c 10-- .O

M 101U. U.

10'04

10'2 102. EMG2::NDM-1 EMG2::SHV-18 EMG2::NDM-1 EMG2::SHV-18 0 RGNndm-1 0 RGNshv-18 . RGNndm-1 . RGNshv-18

Figure 4-4: Mobilizable RGNs can be conjugated into target bacteria for selective removal of multidrug resistance: (A) Schematic of mobilizable RGN-mediate cell killing. (B-C) S17-1 Apir donor cells possessing mRGNndm-1 or mRGNshv-18 were mated at a donor:recipient ratio of 340 66:1 for 3 hours with EMG2 recipient cells that contain blaNDM-1 (EMG2::NDM- 1) or blasHv-18 (EMG2::SHV-18) integrated into the chromosome. Cultures were plated on LB+carbenicillin to select for surviving recipient cells (B) and LB+chloramphenicol+carbenicillin to select for transconjugants (C) (chloramphenicol resistance is encoded by the RGN plasmids). (C) Transfer of a mobilizable RGN into cells containing the cognate target sequence (dashed line) reduced the number of viable transconjugants to the limit of detection (*) (100 CFU/mL or 500 CFU/mL for three or six of the biological replicates, respectively) in almost all cases. Error bars indicate s.e.m. of three independent experiments, each with three biological replicates (n=9).

65 Chapter 4 Citorik Chapter 4 65 Bacteriophages for Human Health

4.4 Phage-Based Delivery

Bacteriophages are natural predators of prokaryotes and are highly adept at in- jecting DNA into host cells. To implement phage for RGN delivery, we engineered phagemid vectors by pairing RGN constructs targeting blaNDM- or blaSHv-18 with an fl origin for packaging into M13 particles. Phage-packaged RGNndm-1 (PRGNndm-1) was capable of comprehensively transducing a population of E. coli EMG2 (Fig. 4- 5.A). To test the 4RGNs, we conjugated native plasmids containing blaNDM-1 (pNDM- 1) or blasHv-18 (pSHV-18) from clinical isolates into EMG2. Treatment of the EMG2 pNDM-1 or EMG2 pSHV-18 strains with the cognate 4RGNs resulted in 2-3-logo reductions in viable cells even in the absence of any selection (Fig. 4-5.B). Further- more, DRGNs engendered no toxicity against wild-type EMG2 or EMG2 containing non-cognate plasmids (Fig. 4-5.B). In naturally occurring Type II CRISPR-Cas systems, the CRISPR locus may contain multiple spacers, each of which is processed into independent crRNA molecules that license Cas9 to cleave cognate DNA sequences [123]. To explore the utility of a single 4RGN exhibiting activity against more than one genetic signature, we engineered a construct containing two spacers encoding two different crRNAs for targeting the blaNDM-1 and blasHv-18 resistance genes (4RGNndm-1/shv-18). DRGNndm- 1/shv-18 generated 2-3-logio reductions in viable cells counts of EMG2 pNDM-1 or EMG2 pSHV-18, but not of wild-type EMG2 (Fig. 4-5b). Thus, RGNs may be multiplexed against different genetic signatures, enabling simultaneous targeting of a variety of virulence factors and resistance genes that may exist in microbial populations. In addition to antibiotic-modifying enzymes, such as -lactamases, alterations in host proteins constitute a major antibiotic resistance mechanism [129]. Owing to the specificity of the CRISPR-Cas system in prokaryotes, we suspected RGNs could discriminate between susceptible and resistant strains that differ by a single nucleotide mutation in DNA gyrase (gyrA), which confers resistance to quinolone antibiotics [129]. Indeed, 4RGNgyrAD87G exhibited specific cytotoxicity only towards

66 Chapter 4 Citorik Bacteriophages for Human Health quinolone-resistant E. coli harboring the chromosomal gyrAD87G mutation and not towards otherwise isogenic strains with the wild-type gyrA gene (Fig. 4-5.C). A B 10. 109 JjEXf6Je A& -j 10' S

10' .5 U .210' EaI zL 1W2 U V iArzY~ 0 5 - Total cells > 10- -o- Transductants 10' 1~6 1.~ ii' 10' EMG2 WT EMG2 pNDM-1 EMG2 pSHV-18 Phage concentration (TFU1,ImL) * SM Buffer A 4RGNshv-18 * ORGNndm-1 vORGNndm-1/shv-18 C 101, IN ihJVr'M

=) 10'. U-

106-

105-

10'3 EMG2 WT EMG2 gyrAWG 0 SM Buffer M ORGNndm-1 A oRGNgyrA,,,

Figure 4-5: RGN constructs delivered via bacteriophage particles (

Killing curves revealed that

Citorik Chapter 4 67 Bacteriophages for Human Health tericidal effect achieved by 2-4 hours (Fig. 4-6.A). Moreover, GRGN antimicrobial activity increased with phagemid particle concentration (Fig. 4-6.B).

A B

1_ ~1* .=--- -. - _w _

- 1102

ime (h) M01 SEMG2 SM Buffer pNDM-1 SM Buffer + EMG2 WT + EMG2 gyrA,,

Figure 4-6: Characterization of 4)RGN-mediated killing of antibiotic- resistant bacteria: (A) Time-course treatment of EMG2 WT or EMG2 pNDM-1 with SM buffer, 4RGNndm-1 or <}RGNsh-18 at a multiplicity of infection (MOI) ~20. Data represent the fold change in viable colonies at indicated time points relative to time 0. (B) Dose response curve of EMG2 WT and EMG2 gyrAD87G treated with various concentrations of 4.RGNgyrAD87G for two hours. Data represent fold change in viable colonies relative to SM buffer treated samples. Error bars represent s.e.m. of three independent biological replicates (n=3).

To further characterize the cellular response to RGN-mediated targeting, we assessed treatment of cells possessing a GFP reporter under SOS regulation. E. colh and other bacteria respond to chromosomal double-stranded breaks, including those artificially generated by the meganuclease I-SceI, by tindig DNA repair through the acttieon of the SOS response [130]. We observed a 2.6- or 4.0-fold increase in fluorescence in cells containing the reporter plasmid sa plasmid-borne (blNDM-1) or chromosomal (gyrAD87G) target site, respectively, when treated with the cognate versus non-cognate DRGNs (Fig. 4-7). These results confirm that RGNs can induce DNA damage in target cells and demonstrate that they can be coupled with SOS- based reporters to detect specific genes or sequences, even at the single-nucleotide level.

68 Chapter 4 Citorik Bacteriophages for Human Health

a b EMG2 WT E SM Buffer EMG2 gyrADS 1-* 0 PRGNgyrADS 7G M*'RGNndm-1 C

0 0 Z.5 Z W

' I lip IV W ' eFITC-A FITC-A 0 EMG2 pNDM-1

0 4,000 - 0

I0 LL. .rn.FriMF

10. IV EMG2 WT EMG2 FITC-A pNDM-1 EMG2 gyrAD,7 * SM Buffer C 4'RONndm-1 0 4RGNyrA.

Figure 4-7: Treatment of E. coli with PRGNs induces DNA damage and an SOS response in cells that possess a cognate target sequence: EMG2 wild-type (WT) (A), EMG2 gyrAD87G (B) and EMG2 pNDM-1 (C) containing the pZA3LG reporter plasmid were treated with either SM buffer, 1RGNndm-1 or 4RGNgyrAD87G. GFP expression on pZA3LG is under the control of the SOS- responsive PL(LexO) promoter [131]. Injection of single-stranded phagemid DNA led to a mild induction of the SOS response, whereas RGN activity in cells that possessed a cognate target sequence led to stronger induction of SOS. Histograms were generated by combining data from four biological replicates and are normalized to the mode of the population. (D) Summary of flow cytometry histograms. The arithmetic means of the geometric mean fluorescence of populations in A-C were calculated across four independent biological replicates (n=4). Error bars represent s.e.m.

4.4.1 Toxin-Antitoxin Activation

We were intrigued to observe that targeted cleavage of blaNDM- with DRGNndm-1 in the context of the native plasmid was lethal to host cells, whereas targeted cleavage of the same gene in a standard cloning vector was not ('pNDM-1' versus 'pZA-ndml- gfp', respectively, in Fig. 4-8A). Therefore, we hypothesized that 4RGN-induced plasmid loss in itself does not elicit lethality, but rather results in cytotoxicity via other co-harbored plasmid-borne functions. Toxin-antitoxin systems are components

Citorik Chapter 4 69 Bacteriophages for Human Health of natural plasmids that ensure persistence in bacterial populations by inhibiting growth of daughter cells that fail to inherit episomes following cell division. These addiction modules traditionally consist of a labile antitoxin that quenches the activity of a stable toxin. Owing to the differential stability of these two components, ces- sation of gene expression upon plasmid loss leads to depletion of the antitoxin pool faster than the toxin pool, resulting in de-repression of toxin activity and, ultimately, stasis or programmed cell death [132]. Analysis of the sequenced pSHV-18 plasmid revealed the presence of a unique toxin-antitoxin module, pemIK, which is commonly found among isolates harboring extended-spectrum 6-lactamases [133]. When complemented with the PemI antitoxin expressed constitutively in trans (pZA31-pemI),

Chapter 4 Citorik 70 Chapter 4 Citorik Bacteriophages for Human Health

A B 109 E 106- * AA _io10A AJlr -) 1010

1061

10' 10 10 '

O ~ 104- 10- 10A

pNbM-1 pZA-ndm1-gfp pSHV18 pZA31-gfp pSHV18 pZA31-peml E GRGNndm-1 A ORGNshv-18 6 SM Buffer U ORGNndm-1 A ORGNshv-18 Cb o3 4RGNndm-1 Cb A dRGNshv-18 Cb 0 SM Buffer Cb O 'RGNndm-1 Cb A ORGNshv-18

Figure 4-8: RGN-mediated targeting of toxin-antitoxin systems can lead to cytotoxicity: (A)EMG2 E. coli containing the natural pNDM-1 plasmid or the blaNDM-1 gene in a synthetic expression vector (pZA-ndm1-gfp) were treated with either DRGNndm-1 or GRGNshv-18 at MOI -20 and plated onto both non-selective LB and LB+carbenicillin (Cb) to select for blaNDM-1-containing cells. 4RGNndm- 1 treatment of cells harboring pNDM-1 resulted in a reduction in viability in the absence of selection, whereas 4RGNndm-1 treatment of cells with pZA-ndm1-gfp demonstrated similar cytotoxicity only under selective pressure for maintenance of the pZA-ndm1-gfp plasmid. (B) EMG2 pSHV-18 complemented with the cognate antitoxin (pZA31-pemI) for the PemK toxin or a control vector (pZA31-gfp) was treated with SM buffer, NRGNndm-1 or DRGNshv-18. Cultures were plated on LB and LB+Cb and colonies were enumerated to assess cytotoxicity or plasmid loss.

4.5 Targeting Virulence Genes in Galleria mellonella Models of Infection

To further demonstrate the versatility of RGNs for specifically combating

pathogens, we designed a 4RGN to target intimin, a chromosomally encoded vir-

ulence factor of enterohemorrhagic E. coli 0157:H7 (EHEC) necessary for intestinal

colonization and pathology. Encoded by the eae gene, intimin is a cell-surface adhesin

that mediates intimate attachment to the host epithelium, permitting subsequent dis-

ruption of intestinal tight junctions and effacement of microvilli [134]. Treatment of

EHEC with DRGNeae resulted in a 20-fold reduction in viable cell counts; this cyto-

toxicity was increased an additional 100-fold under kanamycin selection for DRGNeae

transductants (Fig. 4-9A). The increase in cytotoxicity with selection for cells receiv-

Citorik Chapter 4 71 Bacteriophages for Human Health ing the construct implies that the efficacy of 4RGN treatment was limited by delivery in this strain. Furthermore, IRGN treatment was assessed in Galleria mellonella larvae, an infection model that yields virulence data often predictive for higher-order mammals [135]. This model has also been used to evaluate the efficacy of antimicrobials or phage therapy against various Gram-negative, Gram-positive and fungal pathogens [135]. Administration of IRGN to EHEC-infected G. mellonella larvae significantly improved survival over no treatment or an off-target DRGN control (Log- rank test, p<0.001) (Fig. 4-9B). Moreover, 4RGNeae was significantly more effective than chloramphenicol treatment, to which the EHEC strain was resistant (Log-rank test, p<0.05) (Fig. 4-10, Fig. 4-11). Although GRGNeae treatment was inferior to carbenicillin, to which the bacteria were susceptible, these data support RGNs as viable alternatives for cases where bacteria are highly resistant to existing antibiotics. Improvements in delivery efficiency with (RGNeae would be expected to improve treatment efficacy and outcome. a b 1000107 Uninfected

E A 75 LL A A 0 106-

50. o 0 SM Buffer (DRGNeae 72 E oRGNndm-1 U)C 104 0 oRGNndm-lKm 25, > A ORGNsae SM Buffer A ORGNeae Km 0RGNndm-1 103- EMG2 WT EHEC 0 24 Time (h)2

Figure 4-9: PRGN particles elicit sequence-specific toxicity against enterohemorrhagic E. coli in vitro and in vivo: (A) E. coli EMG2 wild-type (WT) cells or ATCC 43888 F' (EHEC) cells were treated with SM buffer, 4RGNndm-1 or DRGNeae at a multiplicity of infection (MOI) ~100 and plated onto LB agar to enumerate total cell number or LB+kanamycin (Km) to select for transductants with IRGNs (n=3). (b) Galleria mellonella larvae were injected with either PBS or approximately 4x105 colony forming units (CFU) of EHEC. Subsequent administration of 4RGNeae at MOI -30 significantly improved survival compared to SM buffer or dIRGNndm-1 treatment (Log-rank test, p<0.001). Survival curves represent an ag- gregate of four independent experiments, each with 20 larvae per treatment group (n=80).

72 Chapter 4 Citorik Bacteriophages for Human Health

Strain AMP CAZ CTX IPM OFX CIP GEN CAR CHL EMG2 Wild-Type 2 0.25 <0.0625 0.25 0.125 50.03125 4 ND ND EMG2 gyrAD87o 4 0.25 50.0625 0.25 0.5 0.125 4 ND ND EMG2 pNDM-1 >64 >64 >64 32 0.125 50.03125 >64 ND ND EMG2 pSHV-18 64 1 0.25 0.25 0.125 50.03125 >64 ND ND ATCC43888 F' (EHEC) 4 ND ND ND ND 50.03125 8 4 >64

Figure 4-10: Minimum inhibitory concentrations (MICs): Strains were assessed for antibiotic resistance according to CLSI protocol and MICs determined (pg/mL) (see methods for details). AMP=ampicillin; CAZ=ceftazidime; CTX=cefotaxime; IPM=imipenem; OFX=ofloxacin; CIP=ciprofloxacin; GEN=gentamicin; CAR=carbenicillin; CHL=chloramphenicol; ND=Not Determined

100. - Uninfected - Carbenicillin

--* 75-

50. -

n a- 0 24 48 72 Time (h)

Figure 4-11: Comparison of 64 pg/mL for carbenicillin and chloramphenicol, respectively (Supplementary Table 1). Administration of

4.6 Population Sculpting

In addition to implementing targeted antimicrobial therapies, RGNs can be used to sculpt the composition of complex bacterial populations (Fig. 4-12). Current

Citorik Chapter 4 73 Bacteriophages for Human Health therapies that use a prebiotic, probiotic or drug to modify the human microbiota have demonstrated potential for alleviating various disease states, but remain poorly characterized in terms of off-target effects and the specific mechanisms by which they act [137]. In concert with the host range of the delivery vehicle, RGN activity can selectively remove bacteria with specific genomic content. This could reduce the prevalence of unwanted genes, including antibiotic resistance and virulence loci, or metabolic pathways from bacterial communities without affecting bystanders. To demonstrate a proof-of-principle for 'bacterial knockdowns' using RGNs, we constructed a synthetic consortium comprised of three phage-susceptible E. coli strains with differential antibiotic resistance profiles. We used 3-lactam-resistant E. coli EMG2 pNDM-1, quinolone-resistant RFS289 (gyrAD87G), and chloramphenicol- resistant CJ236. Application of 4DRGNndm-1 elicited >400-fold killing of EMG2 pNDM-1, while leaving RFS289 and CJ236 cell populations intact. Treatment with

20,000-fold killing of RFS289 without a concomitant reduction in EMG2 pNDM-1 or CJ236 (Fig. 4-12). These results demonstrate that RGNs can selectively knockdown bacteria that contain target DNA sequences while allowing the remaining non-target bacteria to dominate the population. Adapting this approach for tuning endogenous microbiota could be accomplished by deliver- ing RGNs in vivo via broad-host-range phages or phage cocktails, or with conjugative plasmids. An appropriately targeted bacterial knockdown approach could be employed in functional studies of complex microbiota and to complement additive therapies, such as probiotics, for microbiome-associated diseases by clearing specific niches or removing defined genes from bacterial populations.

4.7 Discussion

In light of the rising tide of antibiotic resistance, interest in engineered cellular and viral therapeutics as potential biological solutions to infectious disease has resurged. By repurposing parts developed by nature, synthetic biologists have designed artificial gene circuits for antimalarial production [138], engineered probiotics [139] and phage

74 Chapter 4 Citorik Bacteriophages for Human Health

/ 7cm GNa (Gene cmBjuer GR~ -G

RFS289 2.3x10 (2.OxlO')

CJ236 2.x10 (1.7x10')

CJ2367 RFS289 CJ2367 2.3x10 (1.1x10') 5.3x10' (7.OxlO) 2.9xIO (5.3x10') RFS289 (7.5x10') EMG2 pNDM-1 4.5x10' 5.OO (726x10) EMG2 pNDM-1 5.9x107(j.1x10*)

EMG2 pNDM-1 1.3x10 (1.3x10')

SM Buffer 4RGNndm-1 DRGNgyrA~ 70

Figure 4-12: Programmable remodeling of a synthetic microbial consortium: A synthetic population composed of three different E. coli strains was treated with either SM buffer, 4RGNndm-1, or DRGNgyrAD87G at an MOI ~100 and plated onto LB with chloramphenicol, streptomycin or ofloxacin to enumerate viable cells of E. coli CJ236, EMG2 pNDM-1 or RFS289 strains, respectively. 4FRGNndm-1 targets blaNDM-1 in EMG2 pNDM-1 and 4RGNgyrAD87G targets the gyrAD87G allele in RFS289. Circle area is proportional to total population size and numbers represent viable cell concentrations (CFU/mL) of each strain after the indicated treatment. The s.e.m. based on three independent experiments is indicated in parentheses (n=3). therapeutics to eradicate biofilms [14] or potentiate antibiotic activity [27, 28]. Here, we demonstrate that transmissible CRISPR-Cas systems can act as a platform for programmable antimicrobials that harness site-specific cleavage to induce cytotoxicity, activate toxin-antitoxin systems, resensitize bacterial populations to antibiotics, and sculpt bacterial consortia. This work complements the recent finding that the Vibrio cholerae phage, ICP1, encodes its own CRISPR-Cas system to counteract a host-encoded phage defense locus [140] and that CRISPR-Cas constructs transformed

Citorik Chapter 4 75 Bacteriophages for Human Health

into electrocompetent cell populations are incompatible with cells that contain cognate target sequences [55, 127, 141]. In contrast to these latter studies, we show that CRISPR-Cas technology can be applied in situ for the removal of undesired genes from microbial populations and in vivo to treat infection in the absence of artificial selection. Moreover, we demonstrate that RGNs can be used to artificially activate plasmid-borne toxin-antitoxin systems, which has recently become an attractive antimicrobial strategy [142]. In addition to validating antimicrobial activity, we further demonstrate potential applications of RGNs in the deletion of plasmids from cells or the detection of DNA elements with up to single nucleotide resolution using a DNA-damage-responsive reporter. Since CRISPR-Cas systems are widely conserved in prokaryotes, the development and optimization of novel delivery vehicles will aid in the creation of new RGNs capable of targeting additional strains, including multidrug-resistant pathogens as well as key players in natural microbial communities, such as the human microbiome. Phage- based therapies are dependent on their ability to deliver nucleic acids into bacteria, which can be resisted through a variety of mechanisms [20]. These delivery vehicles can be limited to a subset of bacteria defined by the chosen phage, such that the design of programmable antimicrobials may require additional considerations as to the phage platform chosen. However, rational modification to phage host range through tail fiber alterations [143] or the use of bacteriophage cocktails [24] can mitigate the host range limitations of phage-based therapies. Although the use of bacteriophages in humans has been met with challenges [24], especially in the Western world, a re- naissance in phage-based therapeutics has begun to address these challenges, such as demonstrating safety in humans [144], improving the persistence of phages remaining in circulation by reducing their clearance by the host [13] and minimizing endotoxin release by using non-lytic phage engineered with heterologous kill functions [11]. Additionally, we devised a complementary delivery strategy using a mobilizable broad-host-range system to introduce RGNs to recipient cells via conjugation. The use of conjugative delivery from probiotics into target bacteria would enable a platform where engineered cells could integrate complex environmental cues and execute

76 Chapter 4 Citorik Bacteriophages for Human Health lethal payload delivery, akin to previously described sentinel cells [145]. Future work is needed to improve the efficiency and spectra of delivery strategies for RGNs, which may include broad-host-range bacteriophages and more efficient conjugative strategies, as well as chemical delivery technologies.

Delivery systems which extend to higher organisms could also enable RGNs to modulate the prevalence of specific genes in wild-type populations [146]. Owing to the modularity and simplicity of CRISPR-Cas engineering, libraries of multiplexed

The addition of facile, sequence-informed rational design to a field dominated by time- and cost-intensive screening for broad-spectrum small-molecule antibiotics could have the potential to reinvigorate the dry pipeline of new antimicrobials.

77 Chapter 4 Citorik Chapter 4 77 Bacteriophages for Human Health

4.8 Experimental Details

Strains and Culture Conditions

Unless otherwise noted, bacterial cultures were grown at 37'C with LB medium

(BD Difco). Where indicated, antibiotics were added to the growth medium to the following final concentrations: 100 pg/mL carbenicillin (Cb), 30 pg/mL kanamycin (Km), 25 pg/mL chloramphenicol (Cm), 100 pg/mL streptomycin (Sm), and 150 ng/mL ofloxacin (Ofx).

Strain Construction

E. coli EMG2 SmR was generated by plating an overnight culture of E. coli EMG2 onto LB+Sm. Spontaneous resistant mutants were re-streaked onto LB+Sm and an isolated colony was picked and used as the recipient for conjugation of the multidrug resistance plasmids. Overnight cultures of EMG2 SmR (recipient), E. coli CDC1001728 (donor for pNDM-1) and K. pneumoniae K6 (donor for pSHV-18) were washed in sterile PBS and 100 pL of donor and recipient were spotted onto LB agar plates and incubated at 37'C overnight. Transconjugants were harvested by scraping the cells in 1 mL of sterile PBS and plating onto LB+Sm+Cb.

The chromosomal integrations of the blaNDM-1 and blaSHV-18 -lactamase genes

and generation of EMG2 gyrAD87G were performed by A-Red recombineering using the pSIM9 system [1471.Templates for integration at the non-essential lacZYA locus were generated by amplifying the blaNDM-1 and blasHV-18 genes from lysates of CDC1001728 and K6 using the primers rcD77/78 and rcD73/74, respectively. Tem- plates for construction of EMG2 gyrAD8 7G were obtained by amplifying gyrA from RFS289 using primers mmD155/161. Strains are listed in Figure 4-13.

Plasmid Construction

To generate the RGN plasmids (Fig. 4-2), an intermediate vector pZA-RGNO, which lacks a CRISPR locus, was created. The tracrRNA and PL(TetO-1) promoter

78 Chapter 4 Citorik Bacteriophages for Human Health

Identifier Strain/Plasmid Relevant Features Source/Reference

Bacterial Strains fRC149 Escherichiacoli EMG2 F+ CGSC #4401 fMM28 E. coli CJ236 FA(HindIII ::cat(Tra P~il_gCm ) NEB #E4141S fMM194 E. coi RFS289 F', gyrAim7G (Qfx) CGSC #5742 fMM269 E. coli EMG2:.NDM-1 EMG2 A1acZYA::blaNDWi this study fMM268 E. coli EMG2::SHV-18 EMG2 AlacZYA::blasHv-su this study fMM384 E. coli EMG2 gyrAD.9o7 EMG2 gyrAmG (QfxR') this study fRC275 E. coli EMG2 SmR EMG2 rpSLK43N ( this study fRC301 E. coli DH5aPRO M13cp Chasteen et al 2006 fMM425 E. coli CDC1001728 pNDM-1 ATCC BAA-2469 fMM278 E. coli S17-1 Rir RP4-2-Tc::Mu-Km::Tn7 Simon et al 1983 fMM362 E. coli EMG2 pNDM-1 pNDM-1 from ATCC BAA-2469 this study fMM426 Klebsieila pneumoniae K6 pSHV-18 ATCC #700603 fRC280 E. coli EMG2 pSHV-18 pSHV-18 from ATCC #700603 this study fMM427 E. coli 0157:H7 43888 eae ATCC #43888 fMM428 E. coli 0157:H7 43888 F' F' from CJ236 this study

Figure 4-13: Bacterial strains used in this study. were synthesized (Genewiz) and amplified using primers mmD98/99, cas9 was amplified from pMJ8067 using primers mmD74/75 and the vector backbone was amplified from pZA11G using primers mmD82/83. Each PCR product was purified and ligated by Gibson assembly [58]. To create the final backbone vector for the

RGN plasmids, the pBBR1 origin, chloramphenicol resistance marker, tL17 termi- nator, and CRISPR locus cloning site were amplified from an intermediate vector pBBR1-MCS1-tLl7 using mmD151/154, digested with NheI and Sac-HF, and ligated with pZA-RGNO digested with SacI-HF and AvrII to create pZB-RGNO. Di- gestion of this vector with PstI-HF and XbaI allowed for the insertion of assembled

CRISPR loci. The AtracrRNA pRGNndm-1 plasmid was created by amplification of pRGNndm-1 with mmD162/163, Clal digestion, and self-ligation. The Cas9D1OA mutant plasmids were constructed through site-directed mutagenesis of pRGNndm-1 with primers mmD108/109 and the KAPA HiFi PCR kit (KAPA Biosystems).

The CRISPR loci were constructed through isothermal annealing and ligation of short, single-stranded oligonucleotides (Integrated DNA Technologies). Each spacer and repeat piece was built by a corresponding oligo duplex connected to adjacent pieces by 6 bp overhangs. In addition, the terminal repeats were designed to contain a 17 bp extension comprised of a BsaI restriction site to generate an overhang

Citorik Chapter 4 79 Bacteriophages for Human Health that allowed insertion into the pUC57-Km-crRNA0 backbone vector synthesized by Genewiz.

To assemble the CRISPR loci, 500 pmol of sense and antisense oligos in a given

duplex were annealed by boiling for 10 minutes at 99'C and cooled to room temper-

ature. 300 pmol of each annealed duplex were combined with 15 U of T4 polynu- cleotide kinase (Affymetrix), 400 U of T4 DNA ligase (NEB), T4 ligase buffer (NEB)

and ddH2 0 to a volume of 20 pL. Following incubation at 25'C for 1 hour, the reaction products were purified using a Qiagen QIAquick PCR Purification Kit. Purified products were digested for three hours with BsaI-HF and re-purified using QlAquick.

To prepare the crRNA backbone vector, pUC57-Km-crRNA0 was amplified using

primers mmD104/105, subsequently digested with BbsI to generate compatible over-

hangs, and ligated with the assembled CRISPR loci. Positive clones of the CRISPR

loci were digested from the entry vector using PstI-HF and XbaI and ligated into pZB-RGNO digested with the same enzymes to create the final RGN plasmids.

Phagemid vector pZEf-gfp was created previously by adding the fl origin amplified from the yeast shuttle pRS series [148] into pZE22-gfp12. The RGN constructs

consisting of the genes encoding the tracrRNA, Cas9, and a sequence-targeting cr-

RNA were amplified as a single product from the respective pRGN vectors using

KAPA HiFi polymerase (Kapa Biosystems) with primers rcD169/183 and digested with AvrII and XmaI (New England Biolabs). These inserts were ligated with a

backbone derived from amplifying the kanamycin resistance cassette, ColEl replication origin and the fl origin required for packaging into M13 particles off of pZEf-gfp with primers rcD184/185 and digesting with the same enzymes. Ligated plasmids were transformed into E. coli DH5aPro for sequence verification and plasmid purification.

The pZE-blaNDM- 1-gfp and pZA-blaNDM- 1-gfp vectors were constructed by swapping the antibiotic resistance cassette of the Lutz-Bujard vectors pZE12G and pZA12G12. The blaNDM-1 gene was amplified from a lysate of CDC1001728 using primers mmD8/9 and the PCR product was digested with SacI-HF and XhoI. The

digested product was ligated into the Lutz-Bujard vectors digested with the same

80 Chapter 4 Citorik Bacteriophages for Human Health enzymes. The PemI antitoxin complementation plasmid pZA31-pemI was created by first amplifying the pemI coding sequence from pSHV-18 with mmD253/254. The PCR product was digested with BamHI and KpnI and ligated with the large fragment of a pZA31G digest with the same enzymes. The SOS-responsive pZA3LG reporter plasmid was derived from pZE1LG [131] by swapping the origin of replication and antibiotic resistance marker with pZA31G using AatII and AvrII as restriction enzymes.

Mobilizable RGNs were created by first amplifying the R1162 replication origin and oriT using mmD266/267. The chloramphenicol selection marker and RGN locus were amplified from pRGNndm-1 and pRGNshv-18 with mmD247/248. PCR products were digested with Spel and XmaI, ligated and transformed into E. coli S17-1

Apir to create the donor cells used in matings.

Minimum Inhibitory Concentration (MIC) Determination

MICs were determined by broth microdilution using LB broth according to the CLSI guidelines [136].

Transformation Assays

Overnight cultures were diluted 1:100 in fresh LB and grown to an optical density

(OD6 0 0 ) of approximately 0.3-0.5. Following 15 minutes of incubation on ice, cultured cells were centrifuged at 3,200 x g, and pellets were resuspended in one tenth volume of TSS buffer (LB, 10 % polyethylene glycol, 5 % dimethyl sulfoxide, 50 mM Mg2+ at pH 6.5) [149]. A 100 pL aliquot of cells was incubated with 10 ng of RGN plasmid

DNA. Plasmids were purified from the DH5aPro cloning host using a Qiagen QlAprep

Spin Miniprep Kit and the concentration was determined using a Quant-iT PicoGreen dsDNA Assay Kit (Invitrogen). Following 30 minutes of incubation on ice, cells were heat shocked at 42'C for 30 seconds, returned to ice for 2 minutes and recovered for

1.5 h at 37'C in 1 mL of SOC broth (HiMedia). For the chromosomal target assay,

Citorik Chapter 4 81 Bacteriophages for Human Health serial dilutions of cells were plated on LB+Cm to select for transformants. Plates were incubated overnight at 37'C, and the number of colony forming units (CFU) were enumerated the following day. Transformation efficiency was used to assess whether the given RGN plasmid was toxic to cells and was calculated as the CFU/mL per pg of DNA transformed (Fig. 4-2A).

For the episomal target assay, following recovery, cultures were washed in fresh LB, diluted 1:100 in LB supplemented with chloramphenicol to select for transformants and incubated for 16 h at 37'C. Samples were washed in sterile PBS, serially diluted and plated on LB+Cm and LB+Cm+Cb (Fig. 4-2B) or analyzed by flow cytometry

(Fig. 4-2C). Colonies were enumerated the following day and plasmid loss was inferred by calculating the ratio of CbR CmR CFUs to CmR CFUs.

Overnight cultures of RGN transformants were also diluted 1:100 in sterile PBS, aliquoted in duplicates in a 96-well plate and immediately assayed using a BD LSR-

Fortessa cell analyzer. Cells were consistently gated by side scatter and forward scatter across independent biological replicates. Fluorescence measurements were performed using a 488 nm argon excitation laser. The GFP gate and laser voltages were initially determined using untreated pZE-blaz-gfp and EMG2 cells as positive and negative controls, respectively, and implemented across biological replicates. BD

FACSDIVA software was used for data acquisition and analysis.

Sequence Analysis

Escape mutants from transformation assays were re-isolated by passaging surviving colonies onto LB+Cm+Cb. DNA isolation for escape sequencing analysis was performed by either extracting plasmid DNA from isolated escape mutants using the

Qiagen QlAprep Spin Miniprep Kit or by amplifying the integrated target locus using primers mmD9/234 or mmD3/4 for blaNDM-1 and blasHV-18, respectively. Sequencing was performed by Genewiz using the primers mmD112-115/153 and rcD11 for analysis of the RGN plasmids and mmD3 or mmD234 for examination of the integrated resistance genes.

82 Chapter 4 Citorik Bacteriophages for Human Health

Phagemid Purification

Phagemids encoding the RGNs were purified using the Qiagen QlAprep Spin

Miniprep Kit (Qiagen) and transformed into E. coli DH5aPro along with the m13cp helper plasmid for generation of phagemid-loaded M13 particles [150]. Strains were inoculated and grown overnight in 250 mL LB+Cm+Km to maintain m13cp and the phagemid, respectively. Cells were pelleted and the supernatant fluid containing the phagemid particles was passed through a 0.2 pum filter. For all purifications except the 4)RGNgyrAD8 7G purification for the dose response curve, M13 phagemid particles were precipitated by the addition of 5% polyethylene glycol (PEG-6000) and 0.5

M NaCl and incubation overnight at 4C [23] and pelleted at 12,000 x g for 1 h.

Purified phagemid pellets were resuspended gently in 1/100th volume of SM buffer 0 (50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 10 mM MgSO 4 ) and stored at 4 C. For the

7.0 to solubilize phagemid particles. Tris-HCl [pH 7.5], NaCl and MgSO 4 were added to reconstitute the composition of SM buffer.

Titers were measured by incubating sample dilutions with E. coli EMG2 for 30 minutes and enumerating transductants by plating on LB and LB+Km. Titers were defined in TFU 1oo/mL, which is the concentration of phagemid at which -100% of a recipient population of an equivalent cell concentration would be transduced.

Phagemid Kill Assays

Cultures were inoculated and grown overnight in LB with appropriate antibiotics at 37'C with shaking. The following day, overnights were subcultured 1:100 into 3 mL LB (no antibiotics) and grown at 37'C with shaking until the OD6 00 reached approximately 0.8. Cultures were diluted into LB to 108 CFU/mL for pNDM-1 and pSHV-18 assays (Fig. 4-5b, 4-6a) or 106 CFU/mL for gyrAD87G (Fig. 4-5c) and

Citorik Chapter 4 83 Bacteriophages for Human Health

EHEC assays (Fig. 4-9a) and 245 pL of the suspension was added to 5 pL of purified phagemid stock in a 96-well plate and incubated static at 37'C. The number of viable cells in samples at each interval during the time-course or at 2 h for endpoint assays was determined by serial dilution and spot plating onto LB, LB+Cb, and LB+Km to analyze cytotoxicity, plasmid loss, and phagemid delivery, respectively. Initial suspensions were also diluted and plated onto LB to quantify the initial bacterial inocula. Colonies were enumerated after 8-9 h incubation at 37'C to calculate cell viability (CFU/mL) and averaged over three independent experiments. Non-linear curve fitting of the time-course to an exponential decay curve was performed using

GraphPad Prism.

Galleria mellonella Model

Larvae of the model organism Galleriamellonella [135] were purchased from Van- derhorst Wholesale, Inc. (St. Marys, OH, USA) and received in the final larval instar for survival assays. Larvae were removed from food source, allowed to acclimate for at least 24 h at room temperature in the dark, and used within 4 days of receipt.

For all injections, a KDS100 (KD Scientific) or Pump 11 Elite (Harvard Apparatus) automated syringe pump was set to dispense a 10 pL volume at a flow rate of ~1 pL/s through a 1 mL syringe (BD) and 26G needle (BD). To prepare bacteria for injection, an overnight culture of E. coli 0157:H7 43888 F' was subcultured in Dulbecco's

Modified Eagle Medium (Gibco) for 4 hours at 37'C with shaking until OD600 ~0.6. Cultures were washed twice in PBS and diluted to a concentration of approximately

4x10 5 CFU/mL. In accordance with other studies [152], twenty larvae per treatment group were randomly selected based on size (150-250 mg) and excluded based on poor health as evidenced by limited activity, dark coloration, or reduced turgor prior to experiments. Larvae were delivered injections without blinding of either PBS or bacteria behind the final left proleg. Approximately an hour after the first injection, SM buffer, antibiotic, or <>RGN treatment was administered behind the final right proleg (Fig. 4-9c and 4-11). Larvae were incubated at 37'C and survival was monitored at 12 h intervals for 72 h, with death indicated by lack of movement and

84 Chapter 4 Citorik Bacteriophages for Human Health unresponsiveness to touch [135]. Kaplan-Meier survival curves were generated and analyzed with the log-rank test using GraphPad Prism.

LexA Reporter Assay

Overnight cultures of EMG2 WT, EMG2 pNDM-1 and EMG2 gyrAD87G containing the SOS-responsive reporter plasmid pZA3LG [131] were diluted 1:50 in LB and incubated with either SM buffer,

Bacterial Matings

Donor and recipient strains grown overnight in LB with appropriate antibiotics were diluted 1:100 in fresh media and grown to approximately OD6 0 0 =1. Cells were pelleted, resuspended in sterile PBS and mating pairs were mixed at a donor to recipient ratio of 340 66:1. Mating mixtures were pelleted, resuspended in 20 PL of

PBS and spotted onto nitrocellulose filters placed on LB agar plates. Initial bacterial suspensions were serially diluted and plated on LB agar plates to quantify the initial inocula. Matings proceeded at 370 C for 3 h with a single mixing step. At 90 minutes, mating mixtures were collected by vigorously vortexing the filters in 1 mL sterile

PBS. Cells were pelleted, resuspended in 20 pL PBS and re-seeded onto filters and incubated as above for the remaining 90 minutes. At the end of the 3 h mating, cells were again recovered by vigorously vortexing the filters in 1 mL sterile PBS. Mating mixtures were serially diluted in PBS and plated onto LB+Cb to select for total number of Cb-resistant recipient cells and LB+Cb+Cm to select for transconjugants.

Colonies were enumerated following overnight incubation at 37'C to determine viable cell counts and were averaged over nine independent biological replicates (Fig. 4-4).

85 Chapter 4 Citorik Chapter 4 85 Bacteriophages for Human Health

Synthetic Consortia Remodeling

E. coli CJ236, EMG2 pNDM-1 and RFS289 strains grown overnight in LB with

appropriate antibiotics were diluted 1:100 into fresh LB (no antibiotics) and grown

to OD6 0 0 '0.8. Cultures were seeded into fresh LB such that the initial mixture

contained -1x10 CFU/mL of each strain and 245 pL of the suspension was added to

5 pL of SM buffer or purified

of CJ236, EMG2 pNDM-1 and RFS289, respectively. Samples were then incubated, plated and enumerated as in phagemid kill assays. The composition of the synthetic

ecosystem under each treatment condition was determined by counting viable colonies

on plates selective for each strain as above and data were calculated as viable cell

concentration (CFU/mL) averaged over three biological replicates (Fig. 4-12).

Data Analysis and Statistics

All data were analyzed using GraphPad Prism version 6.0 (GraphPad Software, San Diego, CA, USA, www.graphpad.com).

Chapter 4 Citorik 86 Chapter 4 Citorik Chapter 5

Conclusion

The threat of return to a pre-antibiotic era is one that must be taken seriously.

The global health interest demands research and development of novel modalities for treating bacterial infections in cases where antibiotics are no longer effective, or as a means for preserving the efficacy of the antibiotics arsenal. In addition to being an alternative therapeutic, bacteriophages represent an entirely new approach to anti- infectives by their narrow-spectrum nature. Broad-spectrum antibiotics have their place when the clinical situation demands preemptive action before characterization of the infectious agent, but the use of targeted, narrow-spectrum antimicrobials will help avoid common problems associated with antibiotic usage, including the selection for resistant organisms and dysbioses of human microbial communities. A disrupted gut microbiome is being linked to more and more disease states, including obesity, immune disorders, and psychological and neurological diseases [3]. The risks of chronic antibiotic treatment, in addition to the loss in effectiveness, demand the creation of next-generation smart drugs capable of discriminating between the therapeutic target and commensal bacteria.

Towards the end of improving the creation of phage therapies, this work has en- compassed multiple modalities of developing bacteriophage-based agents for targeting specific bacteria of interest, with a focus on E. coli and K. pneumoniae, two species

87 Bacteriophages for Human Health of enteric bacteria that have emerged as particularly well-equipped superbugs. In the first chapter, the novel isolation of bacteriophages against a panel of target bacteria was undertaken as a means of demonstrating the discovery and application of naturally occurring wild-type bacteria. A murine gut model demonstrated a significant but incomplete effect of treatment on decolonization of K. pneumoniae, indicating the need for further optimization and perhaps implicating a strong need for phage engineering tools and methods, some methods of which are explore in the following two projects.

The next chapter sought to challenge a paradigm of phage therapy that necessi- tates the discounting of temperate phages. As these phages represent a large resource of sequenced viruses, and are very abundant in environments like the gut microbiome, efforts were made to pilot a pipeline for conversion of temperate phages into obligately lytic phages in order to be made available for clinical use. These efforts were successful for engineering an obligately lytic E. coli phage, but were unable to result in production of engineered progeny for a newly characterized phage of K. pen-

moniae, likely due to an inefficiency of rebooting this particular phage with the E.

coli pipeline. Such partial successes lend credence to the need for improving phage engineering methods.

In the final chapter, an engineering approach to alter the payload delivered by a bacteriophage was employed. The CRISPR-Cas system was loaded onto a phagemid vector and delivered into recipient cells via a phage vehicle. These agents represent a novel class of programmable antimicrobials whose activity can be tuned to kill target bacteria based on the presence of predefined genetic signatures, such as antibiotic resistance or virulence determinants.

In summary, bacteriophages have multiple paths into the clinic, including using wild-type or engineered phages, or even developing phage-derived components. Ef- forts to improve the isolation, characterization, engineering, and in vivo efficacy of bacteriophages will undoubtedly prove invaluable in the future of translational phage- based therapeutics. Given the recent emergence of untreatable bacterial infections, this future could not come at a better time.

88 Chapter 5 Citorik Bibliography

1. Citorik, R. J. & Lu, T. K. Coming of Phage. The Biologist 63, 12-15 (2016).

2. Citorik, R. J., Mimee, M. & Lu, T. K. Bacteriophage-based synthetic biology

for the study of infectious diseases. Current opinion in microbiology 19C, 59- 69 (July 2014).

3. Mimee, M., Citorik, R. J. & Lu, T. K. Microbiome therapeutics - Advances

and challenges. Advanced drug delivery reviews 105, 44-54 (Oct. 2016).

4. Twort, F. An investigation on the nature of ultra-microscopic viruses. The Lancet 186, 1241-1243 (Dec. 1915).

5. d'Herelle, F. Sur un microbe invisible antagoniste des bacilles dysenterique. Comptes rendus de l'Acadimie des Sciences 165, 373-375 (1917).

6. Summers, W. C. Bacteriophage therapy. Annual review of microbiology 55, 437-51 (Jan. 2001).

7. Ackermann, H.-W. The first phage electron micrographs. Bacteriophage 1, 225- 227 (July 2011).

8. Summers, W. C. The strange history of phage therapy. Bacteriophage 2, 130- 133 (Apr. 2012).

9. Kingwell, K. Bacteriophage therapies re-enter clinical trials. Nature Reviews Drug Discovery 14, 515-516 (July 2015).

10. Sansom, C. Phage therapy for severe infections tested in the first multicentre

trial. The Lancet. Infectious diseases 15, 1384-5 (Dec. 2015).

89 Bacteriophages for Human Health

11. Hagens, S. & Blksi, U. Genetically modified filamentous phage as bactericidal

agents: a pilot study. Letters in applied microbiology 37, 318-23 (Jan. 2003).

12. Hagens, S., Habel, A., von Ahsen, U., von Gabain, A. & Blhsi, U. Therapy of

experimental pseudomonas infections with a nonreplicating genetically modified phage. Antimicrobial agents and chemotherapy 48, 3817-22 (Oct. 2004).

13. Merril, C. R. et al. Long-circulating bacteriophage as antibacterial agents. Pro-

ceedings of the National Academy of Sciences of the United States of America 93, 3188-3192 (Apr. 1996).

14. Lu, T. K. & Collins, J. J. Dispersing biofilms with engineered enzymatic bacte-

riophage. Proceedings of the National Academy of Sciences of the United States of America 104, 11197-202 (July 2007).

15. Bertani, G. & Weigle, J. J. Host controlled variation in bacterial viruses. Jour- nal of bacteriology 65, 113-21 (Feb. 1953).

16. Tock, M. R. & Dryden, D. T. F. The biology of restriction and anti-restriction. Current opinion in microbiology 8, 466-72 (Aug. 2005).

17. Barrangou, R. et al. CRISPR provides acquired resistance against viruses in prokaryotes. Science (New York, N. Y.) 315, 1709-12 (Mar. 2007).

18. Bull, J. J., Vegge, C. S., Schmerer, M., Chaudhry, W. N. & Levin, B. R. Phe-

notypic resistance and the dynamics of bacterial escape from phage control. PloS one 9, e94690 (Jan. 2014).

19. Manning, P. A., Puspurs, A. & Reeves, P. Outer membrane of Escherichia coli

K-12: isolation of mutants with altered protein 3A by using host range mutants

of bacteriophage K3. Journal of bacteriology 127, 1080-4 (Sept. 1976).

20. Labrie, S. J., Samson, J. E. & Moineau, S. Bacteriophage resistance mecha-

nisms. Nature reviews. Microbiology 8, 317-27 (May 2010).

21. Levin, B. R. & Bull, J. J. Population and evolutionary dynamics of phage

therapy. Nature reviews. Microbiology 2, 166-73 (Feb. 2004).

90 Chapter 5 Citorik Bacteriophages for Human Health

22. Lu, T. K. & Koeris, M. S. The next generation of bacteriophage therapy. Cur-

rent opinion in microbiology 14, 524-31 (Oct. 2011).

23. Westwater, C. et al. Use of genetically engineered phage to deliver antimicrobial

agents to bacteria: an alternative therapy for treatment of bacterial infections.

Antimicrobial agents and chemotherapy 47, 1301-7 (Apr. 2003).

24. Abedon, S. T., Kuhl, S. J., Blasdel, B. G. & Kutter, E. M. Phage treatment of

human infections. Bacteriophage 1, 66-85 (Mar. 2011).

25. Moradpour, Z. et al. Genetically engineered phage harbouring the lethal

catabolite gene activator protein gene with an inducer-independent promoter

for biocontrol of Escherichia coli. FEMS microbiology letters 296, 67-71 (July 2009).

26. Fairhead, H. SASP gene delivery: a novel antibacterial approach. Drug news

perspectives 22, 197-203 (May 2009).

27. Edgar, R., Friedman, N., Molshanski-Mor, S. & Qimron, U. Reversing bacte-

rial resistance to antibiotics by phage-mediated delivery of dominant sensitive

genes. Applied and environmental microbiology 78, 744-51 (Feb. 2012).

28. Lu, T. K. & Collins, J. J. Engineered bacteriophage targeting gene networks

as adjuvants for antibiotic therapy. Proceedings of the National Academy of

Sciences of the United States of America 106, 4629-34 (Mar. 2009).

29. Henn, M. R. et al. Analysis of high-throughput sequencing and annotation

strategies for phage genomes. PloS one 5, e9083 (Feb. 2010).

30. Reyes, A., Semenkovich, N. P., Whiteson, K., Rohwer, F. & Gordon, J. I. Going

viral: next-generation sequencing applied to phage populations in the human

gut. Nature reviews. Microbiology 10, 607-17 (Sept. 2012).

31. Modi, S. R., Lee, H. H., Spina, C. S. & Collins, J. J. Antibiotic treatment ex- pands the resistance reservoir and ecological network of the phage metagenome.

Nature 499, 219-22 (July 2013).

91 Chapter 5 Citorik Chapter 5 91 Bacteriophages for Human Health

32. Mills, S. et al. Movers and shakers: influence of bacteriophages in shaping the mammalian gut microbiota. Gut microbes 4, 4-16 (2013).

33. Liu, J. et al. Antimicrobial drug discovery through bacteriophage genomics. Nature biotechnology 22, 185-91 (Feb. 2004).

34. Walsh, C. Antibiotics: Actions, Origins, Resistance illustrate, 335 (American Society of Microbiology, Washington, D.C., Jan. 2003).

35. Pastagia, M., Schuch, R., Fischetti, V. A. & Huang, D. B. Lysins: the arrival of

pathogen-directed anti-infectives. Journal of medical microbiology 62, 1506-16 (Oct. 2013).

36. Schmitz, J. E., Schuch, R. & Fischetti, V. A. Identifying active phage lysins

through functional viral metagenomics. Applied and environmental microbiology 76, 7181-7 (Nov. 2010).

37. Schuch, R. et al. Use of a bacteriophage lysin to identify a novel target for

antimicrobial development. PloS one 8, e60754 (Jan. 2013).

38. Fischetti, V. A. Exploiting what phage have evolved to control gram-positive pathogens. Bacteriophage 1, 188-194 (July 2011).

39. Lukacik, P. et al. Structural engineering of a phage lysin that targets gram-

negative pathogens. Proceedings of the National Academy of Sciences of the United States of America 109, 9857-62 (June 2012).

40. Patzer, S. I., Albrecht, R., Braun, V. & Zeth, K. Structural and mechanistic

studies of pesticin, a bacterial homolog of phage lysozymes. The Journal of biological chemistry 287, 23381-96 (July 2012).

41. Smartt, A. E. & Ripp, S. Bacteriophage reporter technology for sensing and

detecting microbial targets. Analytical and bioanalytical chemistry 400, 991- 1007 (May 2011).

42. Schofield, D. a., Sharp, N. J. & Westwater, C. Phage-based platforms for

the clinical detection of human bacterial pathogens. Bacteriophage 2, 105- 283 (Apr. 2012).

92 Chapter 5 Citorik Bacteriophages for Human Health

43. Lu, T. K., Bowers, J. & Koeris, M. S. Advancing bacteriophage-based microbial

diagnostics with synthetic biology. Trends in biotechnology 31, 325-7 (June 2013).

44. Singh, A., Poshtiban, S. & Evoy, S. Recent advances in bacteriophage based

biosensors for food-borne pathogen detection. Sensors (Basel, Switzerland) 13, 1763-86 (Jan. 2013).

45. Tawil, N., Sacher, E., Mandeville, R. & Meunier, M. Bacteriophages: biosens-

ing tools for multi-drug resistant pathogens. The Analyst 139, 1224-36 (Mar. 2014).

46. Velusamy, V., Arshak, K., Korostynska, 0., Oliwa, K. & Adley, C. An overview

of foodborne pathogen detection: in the perspective of biosensors. Biotechnology advances 28, 232-54 (2010).

47. Smartt, A. E. et al. Pathogen detection using engineered bacteriophages. An- alytical and bioanalytical chemistry 402, 3127-46 (Apr. 2012).

48. Wolber, P. K. & Green, R. L. Detection of bacteria by transduction of ice

nucleation genes. Trends in biotechnology 8, 276-9 (Oct. 1990).

49. Edgar, R. et al. High-sensitivity bacterial detection using biotin-tagged phage

and quantum-dot nanocomplexes. Proceedings of the National Academy of Sci-

ences of the United States of America 103, 4841-5 (Mar. 2006).

50. Jinek, M. et al. A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science (New York, N. Y.) 337, 816-21 (Aug. 2012).

51. Mali, P. et al. RNA-guided human genome engineering via Cas9. Science (New York, N. Y.) 339, 823-6 (Feb. 2013).

52. Cong, L. et al. Multiplex genome engineering using CRISPR/Cas systems. Science (New York, N. Y.) 339, 819-23 (Feb. 2013).

53. Hwang, W. Y. et al. Efficient genome editing in zebrafish using a CRISPR-Cas

system. Nature biotechnology 31, 227-9 (Mar. 2013).

Citorik Chapter 5 93 Bacteriophages for Human Health

54. Cho, S. W., Kim, S., Kim, J. M. & Kim, J.-S. Targeted genome engineering in human cells with the Cas9 RNA-guided endonuclease. Nature biotechnology 31, 230-2 (Mar. 2013).

55. Jiang, W., Bikard, D., Cox, D., Zhang, F. & Marraffini, L. A. RNA-guided

editing of bacterial genomes using CRISPR-Cas systems. Nature biotechnology 31, 233-9 (Mar. 2013).

56. Mali, P., Esvelt, K. M. & Church, G. M. Cas9 as a versatile tool for engineering biology. Nature methods 10, 957-63 (Oct. 2013).

57. Kiro, R., Shitrit, D. & Qimron, U. Efficient engineering of a bacteriophage

genome using the type I-E CRISPR-Cas system. RNA biology 11, 42-4 (Jan. 2014).

58. Gibson, D. G. et al. Enzymatic assembly of DNA molecules up to several hun- dred kilobases. Nature methods 6, 343-5 (May 2009).

59. Lu, T. K. et al. Recombinant phage and methods 2013.

60. Jaschke, P. R., Lieberman, E. K., Rodriguez, J., Sierra, A. & Endy, D. A fully

decompressed synthetic bacteriophage oX174 genome assembled and archived in yeast. Virology 434, 278-84 (Dec. 2012).

61. Smith, H. 0., Hutchison, C. A., Pfannkoch, C. & Venter, J. C. Generating

a synthetic genome by whole genome assembly: phiX174 bacteriophage from

synthetic oligonucleotides. Proceedings of the National Academy of Sciences of the United States of America 100, 15440-5 (Dec. 2003).

62. Gibson, D. G. et al. Creation of a bacterial cell controlled by a chemically synthesized genome. Science (New York, N. Y.) 329, 52-6 (July 2010).

63. Kilcher, S., Studer, P., Muessner, C., Klumpp, J. & Loessner, M. J. Cross-

genus rebooting of custom-made, synthetic bacteriophage genomes in L-form

bacteria. Proceedings of the National Academy of Sciences of the United States of America 115, 567-572 (Jan. 2018).

94 Chapter 5 Citorik Bacteriophages for Human Health

64. Shin, J., Jardine, P. & Noireaux, V. Genome replication, synthesis, and assem-

bly of the bacteriophage T7 in a single cell-free reaction. ACS synthetic biology 1, 408-13 (Sept. 2012).

65. Rustad, M., Eastlund, A., Marshall, R., Jardine, P. & Noireaux, V. Synthesis

of Infectious Bacteriophages in an E. coli-based Cell-free Expression System.

Journal of visualized experiments : JoVE, 1-9 (Aug. 2017).

66. Sleight, S. C., Bartley, B. A., Lieviant, J. A. & Sauro, H. M. Designing and en-

gineering evolutionary robust genetic circuits. Journal of biological engineering 4, 12 (Nov. 2010).

67. Chan, L. Y., Kosuri, S. & Endy, D. Refactoring bacteriophage T7. Molecular systems biology 1, 2005.0018 (Jan. 2005).

68. Springman, R., Molineux, I. J., Duong, C., Bull, R. J. & Bull, J. J. Evolutionary

stability of a refactored phage genome. ACS synthetic biology 1, 425-30 (Sept. 2012).

69. Gladstone, E. G., Molineux, I. J. & Bull, J. J. Evolutionary principles and syn-

thetic biology: avoiding a molecular tragedy of the commons with an engineered

phage. Journal of biological engineering 6, 13 (Sept. 2012).

70. Reardon, S. Phage therapy gets revitalized. Nature 510, 15-16 (June 2014).

71. Sulakvelidze, A., Alavidze, Z. & Morris, J. G. Bacteriophage therapy. Antimi-

crobial agents and chemotherapy 45, 649-59 (Mar. 2001).

72. Weiss, M. et al. In vivo replication of T4 and T7 bacteriophages in germ-free mice colonized with Escherichia coli. Virology 393, 16-23 (Oct. 2009).

73. Maura, D. et al. Intestinal colonization by enteroaggregative Escherichia coli

supports long-term bacteriophage replication in mice. Environmental microbiology 14, 1844-54 (Aug. 2012).

74. Reyes, A. et al. Viruses in the faecal microbiota of monozygotic twins and their

mothers. Nature 466, 334-8 (July 2010).

Citorik Chapter 5 95 Bacteriophages for Human Health

75. Minot, S. et al. The human gut virome: inter-individual variation and dynamic response to diet. Genome research 21, 1616-25 (Oct. 2011).

76. Norman, J. M. et al. Disease-specific alterations in the enteric virome in in-

flammatory bowel disease. Cell 160, 447-60 (Jan. 2015).

77. Reyes, A., Wu, M., McNulty, N. P., Rohwer, F. L. & Gordon, J. I. Gnotobiotic

mouse model of phage-bacterial host dynamics in the human gut. Proceedings

of the National Academy of Sciences of the United States of America 110, 20236-41 (Dec. 2013).

78. Barr, J. J. et al. Bacteriophage adhering to mucus provide a non-host-derived

immunity. Proceedings of the NationalAcademy of Sciences of the United States of America 110, 10771-6 (June 2013).

79. Ando, H., Lemire, S., Pires, D. P. & Lu, T. K. Engineering Modular Viral

Scaffolds for Targeted Bacterial Population Editing. Cell systems 1, 187-196 (Sept. 2015).

80. Krom, R. J., Bhargava, P., Lobritz, M. a. & Collins, J. J. Engineered Phagemids

for Nonlytic, Targeted Antibacterial Therapies. Nano Letters 15, 4808-4813 (2015).

81. Bikard, D. et al. Exploiting CRISPR-Cas nucleases to produce sequence-specific

antimicrobials. Nature biotechnology, 1-6 (Oct. 2014).

82. Citorik, R. J., Mimee, M. & Lu, T. K. Sequence-specific antimicrobials using

efficiently delivered RNA-guided nucleases. Nature Biotechnology, 1141-1145 (Sept. 2014).

83. Selden, R., Lee, S., Wang, W. L., Bennett, J. V. & Eickhoff, T. C. Nosocomial

klebsiella infections: intestinal colonization as a reservoir. Annals of internal medicine 74, 657-64 (May 1971).

84. De Man, T. J. B. et al. Genomic Analysis of a Pan-Resistant Isolate of Klebsiella

pneumoniae, United States 2016. mBio 9, 1-7 (Apr. 2018).

96 Chapter 5 Citorik Bacteriophages for Human Health

85. Holt, K. E. et al. Genomic analysis of diversity, population structure, viru-

lence, and antimicrobial resistance in Klebsiella pneumoniae, an urgent threat

to public health. Proceedings of the National Academy of Sciences of the United States of America 112, E3574-81 (July 2015).

86. Domenico, P., Diedrich, D. L. & Straus, D. C. Extracellular polysaccharide pro-

duction by Klebsiella pneumoniae and its relationship to virulence. Canadian journal of microbiology 31, 472-8 (May 1985).

87. Shu, H. Y. et al. Genetic diversity of capsular polysaccharide biosynthesis in

Klebsiella pneumoniae clinical isolates. Microbiology 155, 4170-4183 (2009).

88. Pan, Y.-J. et al. Genetic analysis of capsular polysaccharide synthesis gene

clusters in 79 capsular types of Klebsiella spp. Scientific reports 5, 15573 (Oct. 2015).

89. Hsu, C.-R., Lin, T.-L., Pan, Y.-J., Hsieh, P.-F. & Wang, J.-T. Isolation of a

bacteriophage specific for a new capsular type of Klebsiella pneumoniae and

characterization of its polysaccharide depolymerase. PloS one 8, e70092 (2013).

90. Lin, T.-L. et al. Isolation of a bacteriophage and its depolymerase specific for

K1 capsule of Klebsiella pneumoniae: implication in typing and treatment. The Journal of infectious diseases 210, jiu332- (2014).

91. Pan, Y.-J. et al. Klebsiella phage

multiple host capsular types. Journal of virology, JVI.02457-16 (Jan. 2017).

92. Snitkin, E. S. et al. Tracking a Hospital Outbreak of Carbapenem-Resistant

Klebsiella pneumoniae with Whole-Genome Sequencing. Science Translational Medicine 4, 148ra116-148ra116 (2012).

93. Conlan, S. et al. Single-molecule sequencing to track plasmid diversity

of hospital-associated carbapenemase-producing Enterobacteriaceae. Science TranslationalMedicine 6, 1-12 (2014).

Citorik Chapter 5 97 Bacteriophages for Human Health

94. Pires, D. P., Oliveira, H., Melo, L. D. R., Sillankorva, S. & Azeredo, J. Bacteriophage-encoded depolymerases: their diversity and biotechnological ap-

plications. Applied microbiology and biotechnology 100, 2141-51 (Mar. 2016).

95. Ackermann, H.-W. Basic phage electron microscopy. Methods in molecular biology (Clifton, N.J.) 501, 113-26 (2009).

96. Bradley, D. E. Ultrastructure of bacteriophage and bacteriocins. Bacteriological reviews 31, 230-314 (Dec. 1967).

97. Ackermann, H.-W. Phage classification and characterization. Methods in molecular biology (Clifton, N.J.) 501, 127-40 (2009).

98. Ramage, B. et al. Comprehensive arrayed transposon mutant library of Kleb-

siella pneumoniae outbreak strain KPNIH1. Journal of bacteriology 199, 1-9 (July 2017).

99. Chibani-Chennoufi, S., Bruttin, A., Dillmann, M.-L. & Briissow, H. Phage-

host interaction: an ecological perspective. Journal of bacteriology 186, 3677- 86 (June 2004).

100. Wyres, K. L. et al. Identification of Klebsiella capsule synthesis loci from whole

genome data. Microbial genomics 2, e000102 (Dec. 2016).

101. Carlton, R. M. Phage therapy: past history and future prospects. Archivum immunologiae et therapiae experimentalis 47, 267-74 (Jan. 1999).

102. Faruque, S. M. & Mekalanos, J. J. Phage-bacterial interactions in the evolution of toxigenic Vibrio cholerae. Virulence 3, 556-65 (Nov. 2012).

103. Casjens, S. R. & Hendrix, R. W. Bacteriophage lambda: Early pioneer and still relevant. Virology 479-480, 310-30 (May 2015).

104. Oppenheim, A. B., Kobiler, 0., Stavans, J., Court, D. L. & Adhya, S. Switches

in bacteriophage lambda development. Annual review of genetics 39, 409-29 (Jan. 2005).

105. Gardner, T. S., Cantor, C. R. & Collins, J. J. Construction of a genetic toggle switch in Escherichia coli. Nature 403, 339-42 (Jan. 2000).

98 Chapter 5 Citorik Bacteriophages for Human Health

106. Elowitz, M. B. & Leibler, S. A synthetic oscillatory network of transcriptional

regulators. Nature 403, 335-8 (Jan. 2000).

107. Campbell, A. Comparative molecular biology of lambdoid phages. Annual review of microbiology 48, 193-222 (1994).

108. Kaiser, A. D. Mutations in a temperate bacteriophage affecting its ability to lysogenize Escherichia coli. Virology 3, 42-61 (Feb. 1957).

109. Bruttin, A. & Brnssow, H. Site-specific spontaneous deletions in three genome

regions of a temperate Streptococcus thermophilus phage. Virology 219, 96- 104 (May 1996).

110. Canchaya, C., Proux, C., Fournous, G., Bruttin, A. & Briissow, H. Prophage

genomics. Microbiology and molecular biology reviews : MMBR 67, 238-76, table of contents (June 2003).

111. Zhou, Y., Liang, Y., Lynch, K. H., Dennis, J. J. & Wishart, D. S. PHAST: a

fast phage search tool. Nucleic acids research 39, W347-52 (July 2011).

112. Livny, J., Larock, C. N. & Friedman, D. I. Identification and isolation of lyso-

gens with induced prophage. Methods in molecular biology (Clifton, N.J.) 501, 253-65 (2009).

113. Chen, L. et al. Multiplex PCR for identification of two capsular types in

epidemic KPC-producing Klebsiella pneumoniae ST258 strains. Antimicrobial

agents and chemotherapy 58, 1-15 (2014).

114. Mardanov, A. V. & Ravin, N. V. Conversion of linear DNA with hairpin telom-

eres into a circular molecule in the course of phage N15 lytic replication. Journal of molecular biology 391, 261-8 (Aug. 2009).

115. Casjens, S. R. et al. The pKO2 linear plasmid prophage of Klebsiella oxytoca. Journal of bacteriology 186, 1818-32 (Mar. 2004).

116. Ravin, N. V. N15: the linear phage-plasmid. Plasmid 65, 102-9 (Mar. 2011).

Citorik Chapter 5 99 Bacteriophages for Human Health

117. Mardanov, A. V. & Ravin, N. V. The antirepressor needed for induction of linear plasmid-prophage N15 belongs to the SOS regulon. Journal of bacteriology 189, 6333-8 (Sept. 2007).

118. Jiang, W. et al. Cas9-Assisted Targeting of CHromosome segments CATCH enables one-step targeted cloning of large gene clusters. Nature communications 6, 8101 (2015).

119. Lutz, R. & Bujard, H. Independent and tight regulation of transcriptional units in Escherichia coli via the LacR/O, the TetR/O and AraC/I1-12 regulatory elements. Nucleic acids research 25, 1203-10 (Mar. 1997).

120. For Disease Control, C. & Prevention. Antibiotic resistance threats in the United States, 2013 tech. rep. (2013).

121. Nordmann, P., Dortet, L. & Poirel, L. Carbapenem resistance in Enterobac- teriaceae: here is the storm! Trends in molecular medicine 18, 263-72 (May 2012).

122. Garneau, J. E. et al. The CRISPR/Cas bacterial immune system cleaves bacteriophage and plasmid DNA. Nature 468, 67-71 (Nov. 2010).

123. Deltcheva, E. et al. CRISPR RNA maturation by trans-encoded small RNA and host factor RNase III. Nature 471, 602-607 (Mar. 2011).

124. Rasheed, J. K. et al. Characterization of the extended-spectrum beta-lactamase reference strain, Klebsiella pneumoniae K6 (ATCC 700603), which produces the novel enzyme SHV-18. Antimicrobial agents and chemotherapy 44, 2382-8 (Sept. 2000).

125. Rasheed, J. K. et al. New Delhi metallo--lactamase-producing Enterobacteri- aceae, United States. Emerging infectious diseases 19, 870-8 (June 2013).

126. Bikard, D., Hatoum-Aslan, A., Mucida, D. & Marraffini, L. A. CRISPR inter- ference can prevent natural transformation and virulence acquisition during in vivo bacterial infection. Cell host & microbe 12, 177-86 (Aug. 2012).

100 Chapter 5 Citorik Bacteriophages for Human Health

127. Gomaa, A. A. et al. Programmable Removal of Bacterial Strains by Use of

Genome-Targeting CRISPR-Cas Systems. mBio 5 (Jan. 2013).

128. P6rez-Mendoza, D. & de la Cruz, F. Escherichia coli genes affecting recipient

ability in plasmid conjugation: are there any? BMC genomics 10, 71 (Jan. 2009).

129. Jacoby, G. A. Mechanisms of resistance to quinolones. Clinical infectious diseases 41 Suppl 2, S120-6 (July 2005).

130. Pennington, J. M. & Rosenberg, S. M. Spontaneous DNA breakage in single living Escherichia coli cells. Nature genetics 39, 797-802 (June 2007).

131. Dwyer, D. J., Kohanski, M. A., Hayete, B. & Collins, J. J. Gyrase inhibitors

induce an oxidative damage cellular death pathway in Escherichia coli. Mol Syst Biol 3 (Mar. 2007).

132. Hayes, F. Toxins-antitoxins: plasmid maintenance, programmed cell death, and cell cycle arrest. Science (New York, N. Y.) 301, 1496-9 (Sept. 2003).

133. Mnif, B. et al. Molecular characterization of addiction systems of plasmids

encoding extended-spectrum beta-lactamases in Escherichia coli. The Journal of antimicrobialchemotherapy 65, 1599-603 (Aug. 2010).

134. Kaper, J. B., Nataro, J. P. & Mobley, H. L. Pathogenic Escherichia coli. Nature reviews. Microbiology 2, 123-40 (Feb. 2004).

135. Desbois, A. P. & Coote, P. J. Wax moth larva (Galleria mellonella): an in vivo

model for assessing the efficacy of antistaphylococcal agents. The Journal of antimicrobial chemotherapy 66, 1785-90 (Aug. 2011).

136. Clinical and Laboratory Standards Institute. Methods for Dilution Antimi-

crobial Susceptibility Tests for Bacteria That Grow Aerobically; Approved Standard-Seventh Edition 64 (2006).

137. Sonnenburg, J. L. & Fischbach, M. A. Community health care: therapeutic

opportunities in the human microbiome. Science Translational Medicine 3, 78ps12 (Apr. 2011).

Citorik Chapter 5 101 Bacteriophages for Human Health

138. Paddon, C. J. et al. High-level semi-synthetic production of the potent antimalarial artemisinin. Nature 496, 528-32 (Apr. 2013).

139. Duan, F. & March, J. C. Engineered bacterial communication prevents Vib-

rio cholerae virulence in an infant mouse model. Proceedings of the National

Academy of Sciences of the United States of America 107, 11260-4 (June 2010).

140. Seed, K. D., Lazinski, D. W., Calderwood, S. B. & Camilli, A. A bacteriophage

encodes its own CRISPR/Cas adaptive response to evade host innate immunity. Nature 494, 489-91 (Feb. 2013).

141. Vercoe, R. B. et al. Cytotoxic Chromosomal Targeting by CRISPR/Cas Sys-

tems Can Reshape Bacterial Genomes and Expel or Remodel Pathogenicity Islands. PLoS Genet 9, e1003454 (Apr. 2013).

142. Williams, J. J. & Hergenrother, P. J. Artificial activation of toxin-antitoxin

systems as an antibacterial strategy. Trends in microbiology 20, 291-8 (June 2012).

143. Lin, T.-Y. et al. A T3 and T7 recombinant phage acquires efficient adsorption

and a broader host range. PloS one 7, e30954 (Jan. 2012).

144. Bruttin, A. & Briissow, H. Human volunteers receiving Escherichia coli phage

T4 orally: a safety test of phage therapy. Antimicrobial agents and chemotherapy 49, 2874-8 (July 2005).

145. Saeidi, N. et al. Engineering microbes to sense and eradicate Pseudomonas

aeruginosa, a human pathogen. Molecular systems biology 7, 521 (Aug. 2011).

146. Esvelt, K. M., Smidler, A. L., Catteruccia, F. & Church, G. M. Concerning

RNA-guided gene drives for the alteration of wild populations. eLife, e03401 (July 2014).

147. Datta, S., Costantino, N. & Court, D. L. A set of recombineering plasmids for

gram-negative bacteria. Gene 379, 109-15 (Sept. 2006).

102 Chapter 5 Citorik

I ill Bacteriophages for Human Health

148. Sikorski, R. S. & Hieter, P. A system of shuttle vectors and yeast host strains

designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Ge- netics 122, 19-27 (May 1989).

149. Chung, C. T., Niemela, S. L. & Miller, R. H. One-step preparation of compe-

tent Escherichia coli: transformation and storage of bacterial cells in the same

solution. Proceedings of the National Academy of Sciences of the United States of America 86, 2172-5 (Apr. 1989).

150. Chasteen, L., Ayriss, J., Pavlik, P. & Bradbury, A. R. M. Eliminating helper

phage from phage display. Nucleic acids research 34, e145 (Jan. 2006).

151. Dong, D., Sutaria, S., Hwangbo, J. Y. & Chen, P. A simple and rapid method to isolate purer M13 phage by isoelectric precipitation. Applied microbiology and biotechnology 97, 8023-9 (Sept. 2013).

152. Ramarao, N., Nielsen-Leroux, C. & Lereclus, D. The insect Galleria mellonella

as a powerful infection model to investigate bacterial pathogenesis. Journal of

visualized experiments : JoVE, e4392 (Jan. 2012).

Citorik Chapter 5 103