Bacterial viruses targeting multi- resistant Klebsiella pneumoniae and Escherichia coli

Harald Eriksson

©Harald Eriksson, Stockholm University 2015

ISBN 978-91-7649-123-2

Published papers are reproduced with the permission of the publisher.

Printed in Sweden by Holmbergs, Malmö 2015

Distributor: Department of Molecular Biosciences, the Wenner-Gren Institute

Populärvetenskaplig sammanfattning

Ett växande problem i dagens

samhälle är den ökade resistensen

hos bakterier mot antibiotika,

vilket i det långa loppet kan leda

till att tidigare milda infektioner

blir svårbehandlade och i vissa fall

dödliga. Forskning pågår på flera

fronter för att lösa detta problem;

från att minska användningen av

antibiotika och endast behandla

svåra infektioner, till att ny- eller

vidareutveckla ny antibiotika.

Dessutom sker det forskning på att

utveckla alternativa behandlings-

metoder mot bakteriella

infektioner, som inte är lika tids-

och kostnadsintensiva som

utvecklingen av ny traditionell

antibiotika.

Fokus i denna avhandling har varit

bakteriens naturliga fiende,

bakterieviruset, som i vissa fall

kan nyttjas i terapeutiskt syfte för

avdödandet av bakterier i

infektioner hos människan.

Genom att isolera bakterievirus

från miljöer där det sker en

kontinuerlig kamp om överlevnad

mellan bakterierna och dess virus,

kan vi ta del av den naturliga

genetiska variation som det finns

mellan dessa. Studierna i denna

avhandling har lett till att vi

isolerat ett flertal nya bakterie-

virus, primärt mot multiresistenta

Klebsiella pneumoniae men även

mot Escherichia coli. De isolerade

bakterievirusen har sedan

karaktäriserats, från evolutionära

och taxonomiska aspekter (artikel

I, II), till ingående analys av

arvsmassa, strukturella proteiner

och infektionsförlopp (artikel II).

Vidare har vi utvärderat hur en

sammansatt blandning av bakterie-

virus lyckas avdöda en större

mängd kliniska isolat av multi-

resistenta K. pneumoniae (artikel

III). Slutligen har vi studerat den

biologiska effekten av predation

på den bakteriella värden, genom

att studera ett kliniskt isolat av

multiresistent K. pneumoniae och

deras förmåga att producera

biofilm, deras tillväxthastighet och

virulens (artikel IV) efter att ha

utvecklat resistens mot

antagonistiska bakterievirus.

Abstract

The global increase in antibiotic

resistance levels in bacteria is a

growing concern to our society and

highlights the need for alternative

strategies to combat bacterial

infections. Bacterial viruses

(phages) are the natural predators

of bacteria and are as diverse as

their hosts, but our understanding

of them is limited. The current

levels of knowledge regarding the

role that phage play in the control

of bacterial populations are poor,

despite the use of phage therapy as

a clinical therapy in Eastern

Europe.

The aim of this doctoral thesis

is to increase knowledge of the

diversity and characteristics of

bacterial viruses and to assess their

potential as therapeutic agents

towards multi-resistant bacteria.

Paper I is the product of de novo

sequencing of newly isolated

phages that infect and kill multi-

resistant Klebsiella pneumoniae.

Based on similarities in gene

arrangement, lysis cassette type

and conserved RNA polymerase,

the creation of a new phage genus

within Autographivirinae is

proposed.

Paper II describes the genomic

and proteomic analysis of a phage

of the rare C3 morphotype, a

Podoviridae phage with an

elongated head that uses multi-

resistant Escherichia coli as its

host.

Paper III describes the study of

a pre-made phage cocktail against

125 clinical K. pneumoniae

isolates. The phage cocktail

inhibited the growth of 99 (79 %)

of the bacterial isolates tested. This

study also demonstrates the need

for common methodologies in the

scientific community to determine

how to assess phages that infect

multiple serotypes to avoid false

positive results.

Paper IV studies the effects of

phage predation on bacterial

virulence: phages were first

allowed to prey on a clinical K.

pneumoniae isolate, followed by

the isolation of phage-resistant

bacteria. The phage resistant

bacteria were then assessed for

their growth rate, biofilm

production in vitro. The virulence

of the phage resistant bacteria was

then assessed in Galleria

mellonella. In the single phage

treatments, two out of four phages

showed an increased virulence in

the in G. mellonella, which was

also linked to an increased growth

rate of the phage resistant bacteria.

In multi-phage treatments

however, three out of five phage

cocktails decreased the bacterial

virulence in G. mellonella

compared to an untreated control.

List of papers

This doctoral thesis is based on the papers listed below, referred to in the text

by their associated Roman numerals.

I. Eriksson, H., Maciejewska, B., Latka, A., Majkowska-

Skrobek, G., Hellstrand, M., Melefors, Ö., Wang, JT.,

Kropinski, A.M., Drulis-Kawa, Z. and Nilsson, A.S.

A suggested new bacteriophage genus, “Kp34likevirus”,

within the Autographivirinae subfamily of Podoviridae.

Viruses 2015, 7, 1804-1822; doi:10.3390/v7041804

II. Khan Mirzaei, M., Eriksson, H., Kasuga, K., Haggård-

Ljungqvist, E., Nilsson, A.S.

Genomic and proteomic analysis of ECORp10, a newly

characterized Podoviridae phage with C3 morphology.

PLoS ONE 2014, 9(12): e116294;

doi:10.1371/journal.pone.0116294

III. Eriksson, H., Berta, D., Örmälä-Odegrip, A., Giske, G. C.,

Nilsson, A.S.

A novel phage cocktail inhibiting the growth of 99 β-

lactamase carrying Klebsiella pneumoniae clinical isolates in vitro

Manuscript.

IV. Örmälä-Odegrip, A., Eriksson, H., Mikonranta, L.,

Ruotsalainen, P., Mattila, S., Hoikkala, V., Nilsson, A.S.,

Bamford, J.K.H. and Laakso, J.

Evolution of virulence in Klebsiella pneumoniae treated with

phage cocktails

Manuscript.

Table of contents

Introduction .................................................................................................. 9 Phages ......................................................................................................................... 9

Phage classification .............................................................................................. 9 Caudovirales taxonomy .................................................................................... 10 Phage infection ................................................................................................... 12 Phage evolution .................................................................................................. 13

Phage impact on bacterial hosts ........................................................................... 15 Phage–bacteria coevolution ............................................................................. 15 Bacterial defense systems ................................................................................ 16

Phage therapy .......................................................................................................... 22 The rise of antibiotic resistance ....................................................................... 22 The history of phage therapy ........................................................................... 23 Phage therapy today ......................................................................................... 24 Proteins as antimicrobial compounds ............................................................. 26

Enterobacteriaceae ................................................................................................. 27 Klebsiella spp. ..................................................................................................... 27 Escherichia coli ................................................................................................... 28

Aims.............................................................................................................. 29

Results and discussion ............................................................................. 30 Paper I ....................................................................................................................... 30 Paper II ..................................................................................................................... 32 Paper III .................................................................................................................... 33 Paper IV .................................................................................................................... 34 Concluding remarks and future perspectives ..................................................... 35

Acknowledgments ..................................................................................... 38

References .................................................................................................. 40

Abbreviations

Abi abortive infection

carba carbapenem

Cas CRISPR associated genes

CRISPR Clustered Regularly Interspaced Short Palindromic

Repeats

DNA deoxyribonucleic acid

ECOR Escherichia coli reference collection

ESBL extended spectrum β-lactamase

HGT horizontal gene transfer

Kpn Klebsiella pneumoniae

LO lysis from without

LPS lipopolysaccharide

LIN lysis inhibition

MS mass spectrometry

MOI multiplicity of infection

PEG polyethylene glycol

RNA ribonucleic acid

RM restriction modification

SEM scanning electron microscopy

Sie superinfection exclusion

TA toxin-antitoxin

TEM transmission electron microscopy

9

Introduction

Phages

Bacteriophages, or phages for short, are viruses that infect and propagate

in bacteria. Phages are the most abundant biological entity on the planet, with

estimated amounts of up to 1031 in the biosphere (Hendrix et al., 1999). Out

of all isolated phages, 96 % are tailed phages that belong to the order

Caudovirales (Ackermann, 2001). Phages are ancient molecular machines

that are believed to have existed by the time bacteria, archaea and eukarya

diverged in the tree of life, based on the folding patterns of their proteins

(Ackermann, 1998). They are highly specific for their target bacterial strain;

this specificity is often as narrow as individual serotypes of a bacterial species.

The large biodiversity of viruses is reflected in but also driven by their hosts,

the prokaryotes and archaea. Phages can carry either RNA or DNA genomes

that are either single or double stranded, and they can be enveloped or non-

enveloped and have varied morphologies.

Phage classification

The relatively simple and straight forward method of classifying

prokaryotes by their 16S rRNA (Woese and Fox, 1977) has no analogue in the

phage world (Hendrix et al., 1999). Phages traditionally have been

taxonomically classified based on the visualization of the free virion structure.

With advances in whole genome sequencing, attempts have been made to

classify phages in more detail. The mosaic-like structure of phage genomes

requires the use of more intricate methods to classify them, not only by

comparing the identities of individual genes but also the spatial placement of

the genes in the genome and the gene order (Hendrix et al., 1999).

The crystal structures of phage capsid proteins have revealed a conserved

three dimensional (3-D) structure, even though there is no or very low amino

acid sequence similarity between phage families. These tertiary structures

recur throughout the viral kingdom, e.g., between the capsid folds of

Podoviridae HK97, Myoviridae T4 and even in the eukaryote Herpes simplex

virus (Cardone et al., 2013). This structural conservation of phage proteins is

evident in several parts of the phage virion, from the assembly of the phage

10

head by the capsid proteins to the joining of the head and tail assembly by the

head to tail connector proteins and also in the setup and assembly of phage

tails. The phage tails consist of tape measure proteins, tail tube proteins, tail

terminator proteins and, in the case of Myoviridae, the tail sheath protein. At

the distal end, the host adsorption apparatus can be either a relatively simple

tail spike or a more complex baseplate with tail fibers, or both components

(Veesler and Cambillau, 2011).

The head to tail connector proteins exhibit large size and sequence

variations between phages, as with the 37-kDa Φ29 portal protein compared

to its 83-kDa P22 counterpart. Regardless of size variation, these compared

portal proteins assemble into an overall similar structure (Veesler and

Cambillau, 2011). The similarity of the tertiary structures of these viral

proteins might either be an indication of a common archaea/eukarya ancestry

for all viruses or evolutionary convergence (Veesler and Cambillau, 2011).

Caudovirales taxonomy

The phages of Caudovirales are classified into three distinct families (fig. 1)

based on their tail morphology. While the archetypical phage has an

icosahedral and symmetrical head, there is a variation with an elongated

morphology for the head structure, often referred to by a letter followed by a

numeral; as shown, A represents Myoviridae phages (fig. 1A), B represents

Siphoviridae (fig. 1B) and C is Podoviridae (fig. 1C). The numeral can range

from 1 to 3: 1 is a symmetrical head, 2 an elongated head with a length-to-

width ratio ranging between 1 and 2, and 3 has a length-to-width ratio greater

than 2 (Bradley, 1967).

Podoviridae have a short and non-flexible receptor binding complex for

cell adhesion and DNA injection into the host cell, and their genome is

encoded by double-stranded DNA. Further subdivision of the Podoviridae

family comprises the Autographivirinae and the Picovirinae subfamilies; the

Autographivirinae-type phages carry a gene for a single-subunit RNA

polymerase and thus have the ability to transcribe their own genes, while the

Picovirinae are classified on the basis of their small genome size, special tail

vertex structure and DNA polymerase (Adriaenssens et al., 2015, Lavigne et

al., 2008). From recent classification studies, it has been shown that while

horizontal gene transfer (HGT) mechanisms are a contributing factor to the

genetic diversity and exchange of genetic material between phages,

evolutionary relationships can be readily observed between different phage

subfamilies and genera (Adriaenssens et al., 2015, Lavigne et al., 2009,

Lavigne et al., 2008).

11

The second Caudovirales family are the Siphoviridae, which have a

double-stranded DNA genome, and a long, flexible and non-contractile tail

(Adriaenssens et al., 2015). Approximately 61 % of all isolated and visualized

tailed phages belong to this morphotype (Ackermann, 2001).

The third family of the order Caudovirales is the Myoviridae, whose

members have a complex and contractile tail and encode their genomes using

double-stranded DNA. The Myoviridae adhesion process to the bacterial cell

is a two-step process. Reversible adhesion occurs through interactions

between the long tail fibers of the phage with a bacterial cell receptor, which

in turn prompts a conformational change in the phage that allows for

irreversible binding to the cell through the short tail fibers (Thomassen et al.,

2003). Binding leads to a contraction of the outer tail sheath, which exposes

the inner tube, leading to penetration of the cell wall and injection of the phage

DNA into the cell. There are currently four subfamilies described in the

Myoviridae family: the Tevenvirinae, Spounavirinae, Peduovirinae and

Eucampyvirinae. These subfamilies have been classified and established

based on sequence similarity and morphological features of the contained

phages, but there are few shared characteristics that are conserved throughout

these phage subfamilies (Lavigne et al., 2009, Javed et al., 2014).

Figure 1. Graphical representation of the virion structure of the tailed phages. A) Myoviridae, with elongated head and neck whiskers (phage T4), B) Siphoviridae with flexible tail and tail fibers (phage

T5) and C) Podoviridae with icosahedral head and tail fibers (phage T7).

12

Phage infection

The main concern of phages is temporal: to ensure that their genes are

transcribed in the correct order so that the host cell is not lysed before progeny

virions are assembled. This process is commonly called a genetic cascade,

where each set of genes that gets transcribed during the infection opens up

another set of genes for expression, leading to a step-wise activation of each

set of genes that drives the infection. This is mediated through several

different strategies, from utilizing the host's own transcriptional machinery

with phage-encoded transcription factors, activators, repressors and anti-

terminators or through phage-encoded polymerases that are utilized to drive

the infection and open up the next operon for transcription (Krebs et al., 2012).

The organization of genes into functional modules is a conserved feature

among phages and can often be used to predict functions of de novo sequenced

genes through the placement of their genome (Veesler and Cambillau, 2011).

Phages can exhibit three different types of lifecycles that are dependent on

their genetic heritage: virulent, temperate or persistent (fig. 2). A virulent

phage that infects a bacterial cell progresses through the lytic cycle (fig. 2A,

left side), in which the phage genome is replicated and capsids are assembled,

followed by host cell lysis to release the progeny phages.

A temperate phage can coexist with a bacterial cell via the integration of its

genome either into the host or through a plasmid. When a temperate phage

infects a bacterium, there is a competition between phage expressed proteins

to either repress or initiate the lytic cycle versus the lysogenic cycle (Bertani,

1951). The outcome of the competition results in the phage entering the lytic

cycle or integrating into the bacterial genome. The prophage will be

maintained and inherited as part of the bacterial genome until a de-repression

event. While the factors that cause de-repression of the lysogenic state are

dependent on the prophage in question, DNA damage, heat shock or chemical

Figure 2. A) Phages that have the genetic prerequisites can enter both the lytic and lysogenic cycles (both left and right cycle), while others can only follow the lytic cycle (leftmost cycle). These are called temperate and virulent (or lytic) phages, respectively. B) Persistent infection. Host chromosome depicted in blue, phage genome in red.

13

induction have been identified in phage lambda as causing excision of the

prophage and induction of the lytic pathway (Little and Michalowski, 2010).

Some phages can enter a temporary and quasi-stable state of pseudolysogeny,

sometimes called a carrier state, during cell starvation and nutrient depletion

of the bacterial cell that is being infected (Los and Wegrzyn, 2012). In this

pseudolysogenic state, the infection process will stall, providing the bacterium

more time to live and potentially a more promising environment for the phage

to propagate in the future. This pseudolysogenic state can persist (Cenens et

al., 2013a), however, even after the bacterial cell enters a high-nutrition

environment, and their genomes are asymmetrically inherited in the daughter

cells as demonstrated with the temperate phage P22 in Salmonella

typhimurium (Cenens et al., 2013b).

A third type of life style has been described in certain filamentous phages,

e.g., phage M13, in which the infection is persistent and progeny phages are

continuously released by excretion (fig. 2B). In this type of infection, the cells

remain unharmed albeit with decreased cellular growth (Chopin et al., 2002).

Phage evolution

There are several mechanisms that contribute to the evolution of phages,

such as DNA replication errors, environmental mutagens and HGT-

mechanisms (Abedon, 2009, Duffy et al., 2008). Single point mutations occur

at a low frequency during genome replication. These mutations are then

incorporated into progeny phages, which creates a large pool of diversity

among the phages. Due to the fast generation times of bacteria and phages,

together they constitute a good model system for insight into evolutionary

processes (Abedon, 2009). This natural plasticity of phages to adapt through

single point mutations has been demonstrated using phage T7, which uses

lipopolysaccharides (LPS) as binding receptors. The phage was subjected to

multiple passes through its wild type E. coli host and various LPS mutants,

which led to the expansion of the phage host range across these mutants

through point mutations in its tail fiber gene (Qimron et al., 2006). However,

HGT mechanisms open up more rapid evolutionary forces that operate not

only in the same bacterial species but also over species boundaries (Ochman

et al., 2000).

The integration of DNA into genomes is often facilitated through a process

termed recombination, which is responsible for the exchange or rearrangement

of genetic material. In prokaryotic cells, recombination comprises some of the

DNA repair mechanisms that maintain the integrity of the genome but also

part of the exchange of genetic material through different HGT mechanisms.

In general, two types of recombination have been described: homologous and

14

site-specific. The requirement for homologous recombination is the presence

of two similar or identical nucleotide strands, which can be joined. In E. coli,

the RecF pathway is primarily responsible for repairing single-stranded breaks

but can also repair double-stranded breaks, while the RecBCD pathway is

responsible for repairing double-stranded breaks. In general, for both

pathways, helicases are recruited to the identified DNA break to unwind the

DNA helix and nucleases degrade one of the DNA strands, which in turn

recruits the protein RecA to the free 3´ single-stranded end of the DNA. RecA

then takes part in the joining of a homologous DNA strand and a Holliday

junction is formed, followed by branch migration and strand separation after

the recombination is complete. For a comprehensive review of these

mechanisms, see references (Kowalczykowski et al., 1994, Kowalczykowski,

2000).

Site-specific recombination can be classified into two general categories:

transpositional recombination, which is further subdivided into replicative or

non-replicative transposition, or conservative site-specific recombination.

Transpositional recombination is mediated by a site-specific recombinase

protein designated transposase, which is often encoded in the mobile genetic

element. The transposase, depending on the type, can either insert the mobile

element through excision and integration, utilizing double-stranded breaks of

both the target and recipient sequences, or through replicative recombination,

in which the donor sequence is joined to the recipient, replicated and then

disjoined (Hallet and Sherratt, 1997).

Conservative site-specific recombination is a common method for

temperate phages to integrate themselves into bacterial genomes, in which

site-specific recombination proteins termed integrases mediate the insertion

of the phage genome at specific sites in the bacterial genome. Integrases are

categorized into two families: serine recombinases, which insert the DNA by

creating a double-stranded break, or tyrosine recombinases, which cleave the

DNA in two constitutive steps via single-strand breaks (Hallet and Sherratt,

1997). For reviews of these mechanisms, see references (Hallet and Sherratt,

1997, Groth and Calos, 2004).

15

Phage impact on bacterial hosts

In contrast to their lytic counterparts, temperate phages can confer

advantages via phage-encoded genes that are beneficial to the bacterial host

(Brabban et al., 2005). There are several examples in which the prophage

encodes for exotoxins, which directly promote virulence of the bacterial host

cell. For example, Vibrio cholera lysogenized by the temperate phage CTXφ

introduces cholera toxins ctxA and ctxB (McLeod et al., 2005), which cause

severe diarrhea in the host, increasing the transmission of the bacterial host

into the environment and to other hosts. There are also more discrete methods

to increase virulence gained by infecting phages, such as regulatory factors

that increase the expression of native bacterial virulence genes or phage-

encoded enzymes that change the expression profiles of a bacterium to

increase its virulence (Wagner and Waldor, 2002). Salmonella enterica phage

phage SopEφ encodes a type three secretion system effector protein SopE, that

promotes the invasion capability of its bacterial host (Mirold et al., 1999), and

Vibrio cholera phage CTXφ also encodes for a toxin co-regulated type IV

pilus (Karaolis et al., 1999). Other notable examples of phages that improve

bacterial host fitness include the S. enterica phage Gifsy-2, which encodes a

superoxide dismutase that enables the bacteria to survive the oxidative stress

inside phagocytes by catalyzing innate superoxide and hydrogen peroxide in

the environment into molecular oxygen and hydrogen peroxide. Another

example is the Staphylococcus aureus phage φPVL (Kaneko et al., 1998),

which encodes a human phagocyte cytotoxin that is released into the

surrounding environment and damages phagocytes (Wagner and Waldor,

2002). These examples illustrate the balance and tradeoff between the

decreased fitness of the bacteria via increased genome size by the integrated

prophage and the increased fitness conferred by the phage.

Phage–bacteria coevolution

Bacteria have numerous predators, which include phages, protozoans and

protists. It is estimated that a large portion of bacterial mortality,

approximately 50 %, is due to phage predation (Diaz-Munoz and Koskella,

2014). There are two main explanatory models to describe the complex

relationship between predator and prey, phage and bacteria. The “kill the

winner” hypothesis argues that any abundant host in the environment will

become a target for predation due to its abundance, and subsequently lead to

an increase in the predators of this prey. Thus, any successful phenotype in

the prey would eventually lead to a negative frequency selection of the

population density, where the actual benefits of a phenotype with high fitness

would be selected against when becoming abundant in the biosphere. A

16

possible parallel driving force behind the biodiversity of phages and bacteria

is the niche differentiation hypothesis, where rather than competition between

phages towards the same abundant host, a selective pressure exists to drive

differentiation towards different hosts, thereby eliminating the need for

competition and extinction of the less fit predator (Diaz-Munoz and Koskella,

2014).

Lytic phages prey on susceptible bacteria in an antagonistic manner, where

the phage goal is to produce progeny and the goal of the bacterium is to

propagate. This drives the evolution and diversification of both bacteria and

phage, and each in turn responds to an evolutionary adaptation or

circumvention of the adaptions of the other, which drives biodiversity

(Buckling and Brockhurst, 2012).

Bacterial defense systems

The close and intricate relationship between bacteria and phage is obvious

when investigating bacterial defense against phages and how phages have

counter-evolved against these defenses (Emond et al., 1998). Bacteria have

evolved strategies to counter each step of the phage infection process, from

the first step of cell adhesion to the last step in which progeny are assembled

and released into the environment. This interaction and defense against phage

predation can be divided into three discrete categories: adsorption inhibition,

restriction and abortive infections (Hyman and Abedon, 2010).

Adsorption inhibition

Adsorption prevention is the first line of defense of the bacterial cell where

the phage is prevented from attaching to and injecting its genome into the host

(fig. 3A). Phages have been shown to bind a multitude of different surface

structures of the bacterial cell wall prior to adsorption. The bacterial cell wall

component adhered to is often a vital and conserved part of the bacterial cell,

such as ferrichrome outer membrane transport protein, FhuA, which phage T5

binds to (Flayhan et al., 2012). Phage attachment to the bacterial cell is often

a two-part process, as with the Myoviridae phage T4, where the long tail fibers

interact and reversibly bind with outer membrane porin protein C (ompC) of

the bacterial cell surface, leading to a conformational change in the short tail

fibers of the phage, which irreversibly bind to the core region of the bacterial

LPS layer, leading to injection of its genome into the bacterial cell

(Thomassen et al., 2003).

17

Bacterial surface receptors

The blockage of phage receptors (fig. 3A) not only enables bacteria to be

protected against infection but can also be used by phages to prevent

superinfection in a process called superinfection exclusion (Sie). Sie has been

demonstrated with phage T5, a virulent phage that infects E. coli and expresses

a lipoprotein early during the infection process that blocks the cell receptor

that it originally adhered to, thus hindering superinfection by other T5 phages.

This masking process is also beneficial to the phage after host cell lysis and

progeny release as it stops progeny phages from adhering to lysed cell debris

(Labrie et al., 2010). An example of a competitive inhibitor produced by

bacteria is an antimicrobial protein designated J25, which binds to the iron

transporter channel FhuA in E. coli (Labrie et al., 2010). FhuA is also used by

phages T1, T5 and φ80 as an adsorption receptor with which it competes.

Antimicrobial protein J25 is produced by E. coli in low-nutrition

environments when competition for resources is high to promote its own

growth (Destoumieux-Garzon et al., 2005).

Capsule

Both E. coli and K. pneumoniae can produce an exopolysaccharide capsule

layer around the cell. This capsular layer is often considered a virulence factor

because it prevents the immune system from recognizing immunogenic

components of the bacterial cell wall, as is the case with some strains of E.

coli (Jann and Jann, 1987) and with K. pneumoniae (Podschun and Ullmann,

1998) infections. The capsule also protects the bacterium from phagocytosis.

Phages have evolved the ability to degrade these polysaccharide capsule

layers, as demonstrated by phages K1E and K1-5 (Leiman et al., 2007), which

carry virion-associated glycosidases that enable the phages to degrade and

penetrate the capsular layer to reach the cellular wall (fig. 3B).

18

Figure 3. Bacterial phage defense mechanisms. A) Adsorption inhibition, B) Restriction, C) Abortive infection.

19

Biofilm

The ability to produce an alternative external protective barrier, or biofilm,

is a highly successful defense mechanism for bacteria (Hall-Stoodley et al.,

2004). These biofilms consists of secreted proteins, lipids, DNA and

polysaccharides that mediate the aggregation of bacteria into micro-colonies

with their own microclimates that protect the bacteria from environmental

threats such as phages and antibiotics (Bjarnsholt, 2013). Biofilms are

produced by some bacteria when they cross a threshold in terms of cell

numbers via quorum sensing mechanisms. Through the production of

biofilms, bacteria have been found to survive for prolonged amounts of time

on surfaces such as catheters, artificial implants and contact lenses, which is

problematic in health care settings (Kostakioti et al., 2013).

As with capsules, biofilms are highly efficient in masking bacteria.

Biofilms decrease the amount of active compound that reaches the bacteria

during antibiotic treatments and also prevent phage tail fibers from finding

their surface receptors on the bacterial cell (fig. 3A). Although bacterial cells

remain sensitive to an administered antibiotic, a biofilm hinders successful

antibiotic treatment, as the antibiotic is prevented from being taken up by the

bacteria and metabolized (Bjarnsholt, 2013).

Some phages have evolved a response to biofilms by encoding tail-

associated biofilm degrading enzymes called depolymerases (Cornelissen et

al., 2011). These phage-associated enzymes can disrupt the protective biofilm

that shields the bacterium. Subsequently, by infecting and replicating in host

cells, phages spread their progeny further into the biofilm-shielded

microcolony and step-wise destroy the protective barrier of the biofilm.

Restriction

The second line of bacterial defense is the restriction of invading foreign

DNA. Several types of systems exist that work towards the same goal: the

digestion of foreign DNA and the protection of the host DNA (fig. 3B).

Restriction modification systems

Restriction modification (RM) systems target invading and un-methylated

DNA, while the host DNA is protected through methylation. These two

component systems are divided into different categories dependent on protein

complexity, recognition sites and their mode of action. The type II RM system

is the most disseminated of the groups and has been found in 80 % of all

sequenced bacterial genomes (Wilson and Murray, 1991).

RM systems are composed of the restriction endonuclease (REase) and the

methyltransferase (MTase), which work in concert. The REase specifically

targets and cleaves unmethylated DNA strands at specific nucleotide

20

sequences, while the MTase methylates these same nucleotides. Both proteins

exist in the cell at the same time, albeit at different concentrations, where the

REase is in abundance. Phages have evolved several different strategies to

evade digestion by RM systems, e.g., base modification to prevent

recognition, DNA masking and endonuclease blocking (Samson et al., 2013).

An example of endonuclease blocking occurs with phage T7, which co-injects

its genome with an accessory protein designated Ocr that mimics the structure

of the EcoKI REase-recognized site. Ocr has a higher affinity for the REase,

preventing digestion of the phage genome. This gives the MTase time to

methylate the phage genome, which is then protected from REase digestion,

and the infection can proceed without interruption (Atanasiu et al., 2002).

Subsequent phage progenies will have methylated genomes that will protect

them from similar RM systems until they infect a susceptible bacterial cell

that does not contain an RM system. Progeny phages from these bacterial cells

will not have methylated genomes and will be susceptible to digestion by the

same RM system.

Another escape strategy of note is the lambda-encoded antirestriction

protein Ral, which enhances the MTase activity of the infected cell (Loenen

and Murray, 1986). This changes the balance between the REase and MTase

in favor of the unprotected foreign DNA, which allows the invading and

replicating genome to be methylated and protected from REase degradation in

the cell.

CRISPR-Cas

The CRISPR-Cas system consists of genes encoding protein components,

the Cas proteins, and a nucleotide array of clustered regularly interspaced

short palindromic repeats (CRISPR). In this CRISPR array, short nucleotide

sequences called spacers are stacked between repeats. The spacers have been

acquired from previous invading foreign DNA and confer immunity against

subsequent infection of the same origin.

The CRISPR-Cas system has been denoted as an adaptive immune system

of prokaryotes (Westra et al., 2012a, Wiedenheft et al., 2012) and is a defense

system that works in two steps: an adaption/acquisition phase (Westra et al.,

2012b) in which foreign DNA is identified, cleaved, and inserted into the

CRISPR array and an interference phase (Swarts et al., 2012) in which the

acquired DNA confers immunity against a subsequent infection by a phage

with an identical sequence to the acquired DNA. In the type I CRISPR-Cas

system, the acquired DNA is transcribed and inserted into a ribonucleoprotein

complex called the Cascade complex. This Cascade complex consists of

proteins encoded by the Cas genes (casABCDE), which are assembled

together with transcribed, processed and matured RNA transcripts from the

21

CRISPR array, now called crRNA. It is this ribonucleoprotein complex that

targets specific DNA strands using its antisense RNA strand and then recruits

Cas3 for nucleolytic degradation (Yosef et al., 2012). Three major types of

CRISPR-Cas systems have been described, based on the homology of the Cas

proteins (Fineran and Charpentier, 2012).

Several evasion strategies exist for phages to overcome restriction by the

CRISPR-Cas system, for example through point mutations in the recognized

region. Another evasion strategy that has been described with Pseudomonas

aeruginosa temperate phages is a phage-encoded small protein that disrupts

the Cas and crRNA complex and thus inactivates the entire CRISPR-Cas

defense system (Bondy-Denomy et al., 2013). Interestingly, these genes are

placed in an operon that is associated with late gene expression, during which

the phage capsid genes are transcribed and translated, which is the final stage

of the genetic cascade of the infecting phage. This suggest that the anti-

CRISPR protein is packaged inside the virion and injected with the DNA to

immediately hinder the CRISPR acquisition process during the infection

process.

Abortive infection

Abortive infection (Abi) either drives the infected cell to arrest the phage

infection cycle or to induce bacterial autolysis. This type of phage exclusion

system has mostly been studied in Lactococcus lactis, due to its extensive use

in the dairy fermentation industry, but has also been identified in E. coli

(Chopin et al., 2005). The Abi systems characteristically allow a successful

phage infection to progresses normally until it is suddenly aborted, often just

prior to the release of progeny phages (fig. 3C). Most Abi systems that have

been identified are residents on plasmids; however, there are some exceptions

in which the Abi system is part of the bacterial genome (Samson et al., 2013).

Several types of Abi systems have been identified and characterized, and

most are dependent on a single gene for their mode of action. The targets of

the Abi systems are as diverse as the systems themselves, and hindrance of a

cell’s DNA replication process, destabilization of RNA transcripts and

arresting protein synthesis have been identified (Emond et al., 1998). Some of

these identified Abi systems are constitutively expressed in the cell, while

others are induced only after a phage infection. However, in general, all Abi

gene products are toxic for the cell when artificially induced at high levels

(Chopin et al., 2005). An example of an Abi system is the cryptic phage e14,

which is part of the Escherichia coli K12 genome. E14 encodes for the

superinfection exclusion protein Lit, which arrests all cellular translation upon

T4 infection, aborting the infection and inducing cellular suicide of the

infected cell. This cascade of events is started by the transcription of the major

22

capsid protein of T4, which interacts with the major host translation factor EF-

Tu in its host, which in turn becomes susceptible to proteolytic cleavage by

the metalloprotease Lit. This cascade of events leads to suspension of the

translation of all proteins in the bacterial cell, and cellular death is ensured

(Bingham et al., 2000).

Another type of abortive infection of phages in bacteria is facilitated by

toxin-antitoxin (TA) systems, which are encoded on plasmids that require the

constant expression of both the toxin and the antitoxin in the host cell to ensure

its survival. Phage infection disrupts this homeostasis, and cell death follows.

In a recent study of phage infection and TA system coevolution in Erwinia

carotovora, the ToxIN system was investigated for its mode of action during

phage infection by examining escape mutants of surviving phages (Fineran et

al., 2009). These Abi escape mutant phages were subsequently sequenced, and

an identified mutated sequence was similar to the RNA antitoxin transcript of

the TA system, which mimicked the function of the original antitoxin in the

host cell, thus preventing the inhibition of phage infection (Blower et al.,

2012).

Phage therapy

The rise of antibiotic resistance

The spread of multidrug-resistant bacteria is of increasing concern to both

the medical sector and to society as a whole. Since the 1970s, only four new

classes of antibiotics have been approved, and due to the high cost associated

with the development of new antibiotics, as well as the rapid acquisition of

resistance, few new antibiotics are in the pipeline (Cooper and Shlaes, 2011).

However, over the last couple years, two new classes of antibiotics have been

approved: fidaxomicin, a narrow spectrum antibiotic that has been approved

for the treatment of Gram-positive Clostridium difficile, and bedaquiline, a

narrow spectrum antibiotic against Mycobacterium tuberculosis (Butler et al.,

2013). The acquisition and dissemination of resistance towards novel

antibiotics is a rapid process and can pass between bacterial species (Davies,

1994). For most novel types of antibiotics, resistant bacteria can be isolated

within months if not immediately after introduction. The ease of this rapid

dissemination of resistance between bacterial species is a major health

concern.

This alarming trend is visible everywhere, with a 424 % increase in

reported multi-resistant ESBL-carrying bacterial infections in Sweden during

the years 2007-2014 (Anonymous, 2013b). This type of increase is a

23

worldwide trend, though it occurs at different levels. Countries that have

generous antibiotic prescription routines or over-the-counter distribution of

antibiotics have the highest incidences of multi-resistant bacteria in the world.

Several actions have been launched from government bodies since 2004 to

combat this problem; one example is the EU-wide ban in 2006 of the use of

antibiotics in animal food, where antibiotics were added to promote growth in

livestock (Anonymous, 2005). Another action that has been taken is the

creation of the ReAct network in 2004, a global network of scientists to

promote debate and increase awareness of growing resistance against

antibiotics (Cars and Hällström, 2004). The major focus of all efforts is to

decrease the spread of multi-resistant bacteria by decreasing the usage of

antibiotics in society. This is made possible by discontinuing the use of

antibiotics for minor infections. Due to the pressing need for alternatives to

conventional antibiotics, research into bacteria- and bacteriophage-derived

antimicrobial compounds has been greatly expanded.

Traditionally, ESBL is carried on plasmids and is disseminated among the

Enterobacteriaceae family of prokaryotes, and the conferred resistance has

evolved rapidly over time. Since the isolation of the first ESBL-resistant

isolate, clinical isolates with larger resistance profiles have been

characterized, from the original characterization of resistance against

aminoglycosides in the 1970s, followed by resistance to extended spectrum

cephalosporins and to ceftazidime (Podschun and Ullmann, 1998). Resistant

clinical isolates of K. pneumoniae and E. coli have emerged against what was

previously the last-resort class of antibiotics, the carbapenems. This highlights

the importance of not only decreasing the widespread misuse of antibiotics in

today’s society but also in finding alternative treatment methods for bacterial

infections.

The history of phage therapy

Phage therapy was originally pioneered by Felix d´Herelle (D´Herelle,

1917, Schultz, 1927) as a remedy against dysentery caused by Shigella

infection during 1930s. This sparked great interest in the scientific and

medical community during that time, and many phage mixes and treatment

remedies were launched against various infections and ailments. However, the

lack of scientific understanding of phage biology, the high level of phage

specificity and the discovery of antibiotics led to the quick abandonment of

phage therapy in the West. In the following decades, only a few institutes

continued to research and use phages as a treatment method, most notably the

Eliava Institute in Georgia and the Wroclaw Institute of Experimental Therapy

in Poland (Miedzybrodzki et al., 2012). These institutes are actively involved

24

in the development and administration of phages to patients. However, the

validity of such therapies is still treated with skepticism in the West due to a

lack of adequately controlled clinical trials (Sulakvelidze et al., 2001).

Phage therapy today

Due to the high specificity and narrow host range of phages, treatment is

often more complex when compared to traditional antibiotics. A reason behind

the success of antibiotic treatment is the availability of broad spectrum

antibiotics, which only require the identification of the class of bacterium

before therapy can be administered. As both a blessing and a curse,

bacteriophage specificity requires the identification of the infecting bacterial

species before treatment can be administered. The specificity of

bacteriophages also reduces unwanted damage to the patient microflora during

therapy compared to treatment with antibiotics.

There are several commercial entities currently working within this field,

with phage products for food processing (Intralytix Inc., Micreos BV) and

human phage therapy product investigations (Ampliphi Biosciences corp.,

Novolytics ltd.). Clinical trials for phage therapy are underway; for example,

in 2009 a successful double blind, placebo and randomized phase II clinical

trial was performed using a phage cocktail consisting of six different phages.

This study investigated the safety and efficacy of using phages as a treatment

method for multi-resistant Pseudomonas aeruginosa in chronic otitis patients

(Wright et al., 2009). The treated group showed a 76 % decrease in bacterial

counts, while the placebo group showed a 9 % increase in bacterial cell density

in the place of infection.

A phage therapy study funded by Nestlé has been running extensive trials

of phage cocktails in Bangladesh (Sarker et al., 2012). Nestlé’s main

bacterium of interest is E. coli, which contaminates many sources of drinking

water, causing severe dehydration due to diarrhea in local populations. Safety

trials have been performed in adults, but the project was terminated in 2013

without notice. In 2013, the PhagoBurn project received funding from the

European Union 7th framework program to evaluate phage therapy treatment

of E. coli and P. aeruginosa infections in burn wound infections in clinical

trials (Ravat and Chatard, 2013).

There are solutions to address the narrow host ranges of phages; one is to

isolate phages with broad host ranges that utilize conserved receptors. Another

solution is to combine multiple phages with similar virulence characteristics

but with different specificities for bacterial surface receptors to produce a

phage cocktail. Ideally this cocktail would decrease the risk of bacterial

resistance developing. Bacteria acquire resistance towards bacteriophages

25

rapidly, as they do towards antibiotics, but the ease of isolation of new phages

and the low cost creates a lower threshold for investigating and replacing

phages that have become ineffective against the bacteria.

Treating untyped bacteria using bacteriophages creates novel challenges,

and strategies are to either use a pre-made phage cocktail with phages specific

against a multitude of serotypes of the same bacterial species or to tailor a

phage cocktail selection with several phages specific for the patient serotype

(Pirnay et al.). The first method requires only the isolation and typing of the

infecting species before treatment starts, while the second tailored treatment

requires not only isolation and typing of the bacteria but also susceptibility

testing of phages using an existing phage library. After susceptibility tests

have been performed, a selection of effective phages can be put into a cocktail

and administered to the patient.

Susceptibility tests to ascertain the host ranges of bacterial viruses can give

false positives or negatives for various reasons. One parameter of a phage

cocktail that must be monitored is its multiplicity of infection (MOI). This

number describes the number of phages added per bacterial cell, and thus how

many phages are expected to infect an individual bacterial cell. Problems arise

when using too high a multiplicity of infection, creating phenomena such as

lysis from without (LO) (Abedon, 2011) or lysis inhibition (LIN) (Bode,

1967). Both phenomena have been observed and investigated for phage T4,

one of the most studied phages in the Myoviridae family. LO occurs when a

bacterial cell membrane is destabilized by a multitude of phages adsorbing to

its membrane, with phage-associated and enzymatically active domains that

degrade the integrity of the cell membrane without specificity for the

bacterium (Abedon, 2011). This process leads to the premature lysis of the

host cell with no release of phage progeny.

A passive phage treatment consists of using a high MOI phage cocktail that

utilizes the ability of phages to induce lysis from without, for which a

productive infection is not needed for pathogen clearance and treatment

success. This ability of phages to non-specifically lyse bacteria can be utilized

to create passive treatment regimens, in which the outcome of the treatment is

not dependent on the successful infection and replication of the phages.

LIN has been identified and studied for phage T4 and occurs when a

successful T4 infection is underway, followed by a superinfecting T4. During

the normal progression of a T4 infection, the transmembrane T4 holin is

expressed and accumulates in the inner membrane during the late phase of

infection. It stays dormant while structural proteins are assembled and the

phage genome is packaged into the progeny capsids. At a critical

concentration, the holin oligomerizes and creates pores in the inner membrane

that allow the phage endolysin to escape the cytoplasm and degrade the outer

26

cell membrane, causing cell lysis. LIN has been demonstrated from

superinfecting T4 phages, where the protein RI interacts with the

transmembrane holin and delays oligomerization and bacterial cell lysis. If the

infected cell is challenged by a superinfection phage every 10 minutes, LIN

can be maintained indefinitely (Tran et al., 2005).

Proteins as antimicrobial compounds

Commercial interest in phage therapy is tentative because the use of natural

proteins or viruses are not patentable, and the regulatory framework for

development and testing in clinical trials creates a high threshold for academic

groups to overcome (Verbeken et al., 2012). The main difficulty in using

proteins as antimicrobial agents concerns their possible rapid inactivation

through proteolytic degradation, renal clearance or immunological blockage.

These obstacles can, however, be overcome by screening for inherently robust

and non-immunogenic proteins (Resch et al., 2011). In summary, several

frontiers in bacteria- and phage-encoded antimicrobial proteins exist; for

example, endolysins (Schmelcher et al., 2012, Walmagh et al., 2012), holins

(Ludwig et al., 2008), colicins (Hecht et al., 2012) and other protein products

have been isolated from phages (Liu et al., 2004).

Large substrate screenings are underway to identify compounds with

antimicrobial activity, both originating from phage (Liu et al., 2004) and

bacteria (Haney and Hancock, 2013). Proteins as antimicrobial compounds

have the advantage of being less immunogenic compared to whole phages,

and recombinant proteins are patentable and thus marketable.

Lysins are proteins that are expressed late in phage infections when phage

progeny have already been assembled in the infected cell. The lysins mediate

the degradation of the peptidoglycan layer of the bacterial cell wall, which

eventually leads to cellular lysis. They are effective in lysing Gram-positive

bacteria that have an exposed peptidoglycan layer, while Gram-negative

bacteria are protected due to their peptidoglycan cell wall layer being

sandwiched between an outer and inner cell membrane.

Another example of active research for bacteria-derived antimicrobial

compounds is the investigation of bacteriocins, such as colicins (Hecht et al.,

2012). Colicins are small protein molecules that can kill susceptible bacteria

by creating pores in the cell membrane or by inhibiting peptidoglycan

synthesis, thus hindering cell growth. These proteins are isolated from

bacteria, which encode for these proteins to increase their competitive

advantage in their habitat by inhibiting the growth of closely related bacterial

strains. Work is ongoing to screen, identify and characterize this class of

proteins for possible use as antimicrobial compounds (Hecht et al., 2012).

27

Recent advances in creating a hybrid bacteriocin-lysin show promise, and a

recombinant lysin has been shown to successfully target and kill Yersinia

pestis in vitro (Lukacik et al., 2012).

Enterobacteriaceae

In this thesis, we have isolated bacterial viruses targeting E. coli and K.

pneumoniae, which are both found in the bacterial Enterobacteriaceae family.

E. coli and K. pneumoniae are the most commonly isolated multi-resistant

bacterial species in the European Union; their isolation rates exceed that of

methicillin-resistant Staphylococcus aureus (MRSA) and vancomycin-

resistant enterococcus (VRE) (Anonymous, 2013a). Enterobacteriaceae

constitute a large family of rod-shaped, Gram-negative and facultative

anaerobic bacteria. This family of bacteria includes a wide range of bacterial

species from commensal organisms to potential human pathogens, including

E. coli, Klebsiella, Salmonella, Shigella and Y. pestis. Several bacterial species

of this family live in the intestines of warm-blooded animals, while others live

in soil or water.

Klebsiella spp.

Klebsiella is a rod shaped, non-motile, encapsulated and facultative

anaerobic Gram-negative bacterium. The genus Klebsiella contains species

that have a wide range of habitats, soil and plants, including mammalian

intestine, mucosal surfaces and skin lesions. Most Klebsiella species have the

ability to produce a thick and protective capsule around the bacterial cell,

which protects the cell from phagocytosis and other host defense mechanisms.

This capsule is composed of a diverse range of acidic polysaccharides, of

which 77 different serotypes have been described. Some of these capsular

antigens, termed K-antigens, have been associated with higher pathogenicity

in both animal models and isolated clinical samples (Simoons-Smit et al.,

1984). Serotyping of the LPS layer has revealed eight different compositions

of its outer polysaccharide layer, designated O-antigen, whereas the O1

antigen is the predominant variation in clinical isolates. This highly

immunogenic and endotoxic LPS layer is mostly concealed by the outer

capsule.

K. pneumoniae is an opportunistic and nosocomial pathogen that infects

immunocompromised patients, infants and the elderly. Patients who are

already hospitalized for other afflictions are especially vulnerable to K.

pneumonia infections, and the risk of being colonized is directly correlated

28

with duration of hospital stay (Podschun and Ullmann, 1998). While in normal

situations an infection by K. pneumoniae would be easily treatable, the

increased spread and dissemination of the ESBL resistance plasmid

throughout Enterobacteriaceae is increasing the severity and problems in

treating these infections. This increases the need for alternative treatment

methods and the need to lower the use of antibiotics in minor infections to

decrease the dissemination of antibiotic resistance in bacteria.

Escherichia coli

E. coli is a rod-shaped, Gram-negative and facultative anaerobic bacterium

that is a commensal species often found in the gastrointestinal tract of warm-

blooded animals. The majority of E. coli species are flagellated and are

encapsulated or produce biofilms. Due to its intimate relationship with

humans and our ability to easily cultivate this bacterium, E. coli has been one

of the most well-characterized bacterial species since its identification in

1885. E. coli is the most routinely used laboratory organism, due to its ease of

cultivation, its ability to easily accept foreign DNA and the ease of genetic

manipulation of this bacterial species. There have been described over 145

different O-antigens and over 80 different K-antigens in E. coli species (Davis

et al., 1973). The majority of E. coli species are commensal and assist the host

with the breakdown of food and production of vitamin K (Bentley and

Meganathan, 1982), but several pathogenic serotypes have been described that

cause diarrheal diseases or colonize extra-intestinal sites, such as the urinary

tract (Lai et al., 2013).

29

Aims

The purpose of this thesis project has been to isolate, purify, and

characterize phages and to test different hypotheses regarding phage therapy

efficacy, in vitro and in vivo, for single phages and on phage mixtures. The

long term goal is to establish a phage collection and strategies for curing multi-

resistant Klebsiella pneumoniae infections in humans.

These aims have been realized in the papers of this thesis as follows:

Isolation and characterization of phages against multi-resistant

Klebsiella pneumoniae and Escherichia coli (Paper I and Paper II)

Characterization and annotation of phage genomes to identify

possible toxins (Paper I, II)

Creation of novel phage cocktails and test them against a selection

of Klebsiella pneumoniae clinical isolates (Paper III)

Testing treatment efficacies in vitro and in vivo (Paper IV)

30

Results and discussion

Paper I

A suggested new bacteriophage genus, “Kp34likevirus”, within the

Autographivirinae subfamily of Podoviridae

Paper I is a product of our continuous work to isolate and sequence novel

phages from waste water effluent plants in Stockholm, Sweden. The decreased

cost of whole genome sequencing has resulted in an abundance of phage

genomes being submitted to NCBI GenBank, but there is limited work being

done on the taxonomical classification of these phages. Newly sequenced

phages are often loosely associated into large super-genera based on similarity

to the closest available phage genome, as with several recent cases with

Phikmvlikevirus (Drulis-Kawa et al., 2011, Lin et al., 2014).

There are some ongoing efforts to classify and categorize phages into new

families and genera, for example with Podoviridae (Lavigne et al., 2008),

Myoviridae (Lavigne et al., 2009) and Siphoviridae (Adriaenssens et al.,

2014). Within the Podoviridae subfamily of Autographivirinae there are

currently three established genera: the T7likevirus, Sp6likevirus and

Phikmvlikevirus (Lavigne et al., 2008). The results of this study increased the

taxonomical resolution of the Autographivirinae subfamily of phages. Based

on sequence homology to Klebsiella phages KP34 (Drulis-Kawa et al., 2011),

NTUH-K2044-K1-1 (Lin et al., 2014), and F19 (KF765493), newly isolated

phages SU503 and SU552A were characterized and compared to each other

and the Phikmvlikevirus genus. These five phages demonstrate a high

nucleotide pairwise identity to each other, between 75.1 – 75.9 %, with 29

conserved genes between themselves and similar gene arrangements. In

comparison to members of the taxonomically close Phikmvlikevirus, 18

conserved genes are shared across the genus boundaries. Within the subfamily

of Autographivirinae, a total of nine genes are conserved between the type

phages of the genera T7likevirus, Sp6likevirus, Phikmvlikevirus and the

proposed “Kp34likevirus”. Furthermore, the phages of the proposed phage

genus “Kp34likevirus” show distinct differences compared to the phages in

the closest relative phage genus, the Phikmvlikevirus. The setup and gene

variants of the lysis cassette genes differ between the genera: the KP34-like

31

phages have a unimolecular spanin (u-spanin) and a spanin-holin-endolysin

setup, while the φKMV lysis cassette has the gene arrangement holin-

endolysin-spanin with a dual protein spanin (i-spanin/o-spanin) complex

similar to that of phage lambda.

The Autographivirinae obligatory RNA polymerase also shows divergence

between the KP34-like phages and the Phikmvlikevirus group members. When

comparing RNA polymerase specificity and recognition loops, they are highly

conserved between the member phages of the proposed “Kp34likevirus” and

divergent from the phages in the Phikmlikevirus genus group. Among the

KP34-like phages, the recognition loops are identical, while the specificity

loops show a minor divergence of four amino acids out of 25, just after the

DNA interacting basepairs of the loop. The conservation of the DNA

interacting regions of the RNA polymerase suggests the use of a genus-

specific promoter sequence within the proposed “Kp34likevirus” phages. This

suggests that a conserved genus of phages that is widespread in the biosphere

has been identified.

32

Paper II

Genomic, Proteomic, Morphological, and Phylogenetic Analyses of

vB_EcoP_SU10, a Podoviridae Phage with C3 Morphology

Paper II describes vB_EcoP_SU10 (SU10) as an E. coli phage with the rare

C3 morphology that has an elongated head with a length-to-width ratio of over

2.5. Phages with this rare phenotype have only been fully described in a few

cases (Ackermann, 2001). Morphological analysis using transmission electron

microscopy (TEM), ultra-thin section electron microscopy, and scanning

electron microscopy (SEM) was performed on the phage and on infected

bacterial cultures at different time points in the infection cycle. Thin section

micrographs showed that phage capsids are arranged in a honeycomb-like

structure before and during the DNA packaging process in the host cell. No

relationship between the genome size and head length was observed in

comparison to φEco32 (Pavlova et al., 2012), Serratia marcenses phage

KSP100 (Matsushita et al., 2009), Salmonella enterica phage 7-11 (Kropinski

et al., 2011) and E. coli phage NJ01 (Li et al., 2012), all of which are C3

morphology Podoviridae phages.

Comparison of the tail lengths between free and adsorbed virions revealed

more than a two-fold difference in length. Two possible explanations exist for

this observation: either there is a post-adsorption change in the virion tail

structure or the observed difference is an artifact of the two different

preparation methods for the ultra-thin sectioning TEM and regular TEM.

Genome analysis and annotation of SU10 revealed over 40 % nucleotide

sequence similarity and also a gene arrangement in common with phage

φEco32 (Pavlova et al., 2012). Based on these similarities, a phage-encoded

sigma factor and a sigma-70 inhibitor were identified.

Transcriptional in silico analysis suggested that several structural genes are

transcribed early in the infection process through host sigma-70 promoters.

Ultra-thin sectioning micrographs confirm these structural proteins as early as

five minutes after adsorption to host cells.

Structural proteins were investigated using mass spectrometry (MS) from

mature and PEG-precipitated virions, where 34 proteins out of 125 predicted

coding sequences were identified and confirmed. The phages contain a

ribosomal slippage site that can create an alternative major capsid gene

variant, which has also been confirmed through MS.

Phylogenetic analysis was performed with C1 and C3 Podoviridae phages

that demonstrated sequence similarity with the major capsid and scaffolding

protein of SU10 to investigate the evolutionary divergence of the elongated

head morphology. Based on these phylogenetic analysis, the C1 and C3

morphology phages diverged approximately 280 million years ago and

33

adapted to different bacterial family hosts. The C3 morphology phages infect

members of the Enterobacteriaceae family, while the C1 morphology phages

infect members of Pseudomonadaceae bacteria.

Paper III

A novel phage cocktail inhibiting the growth of 99 β-lactamase-carrying

Klebsiella pneumoniae clinical isolates in vitro

The antibiotic resistance of Enterobacteriaceae is often conferred through

a plasmid-mediated enzyme that has the ability to hydrolyze and inactivate β-

lactam antibiotics, which includes penicillins and third-generation

cephalosporins (Doern, 1995). Additional resistance to carbapenem

antibiotics (ESBLcarba) has emerged (Jacoby, 1997), further reducing the

number of effective treatments.

A phage cocktail consisting of six phages that target K. pneumoniae isolates

was assembled from a collection of 60 isolated phages. The phages were

selected for their ability to prey on 24 extended spectrum β-lactamase (ESBL)-

carrying K. pneumoniae isolates. The cocktail was then assessed against a total

of 125 K. pneumoniae isolates with different resistance profiles, including

carbapenem-hydrolyzing ESBLcarba isolates. As part of this study, host range

screenings of the phage cocktail and the individual phages were performed

using standard materials and methods adapted for antibiotic resistance

screenings in clinical laboratories to determine the ease of adaptation of phage

screenings to these labs. The use of standard protocols and equipment only

required slight adjustments to determine phage susceptibility in clinical

laboratories.

The phage cocktail inhibited the growth of 99 of the 125 (79 %) clinical

isolates. However, two phages out of the six were responsible for the majority

of the bacterial inhibition across the isolates, and productive infection was not

confirmed to be the cause of the bacterial growth inhibition.

34

Paper IV

Evolution of virulence in Klebsiella pneumoniae treated with phage

cocktails

Paper IV investigates the impact of phage predation on bacterial virulence.

While resistance is a major problem with traditional antibiotics, the short

generation times and ongoing co-evolutionary battle between bacteria and

phage have created a large natural reservoir of phages that can be used to

overcome bacterial phage resistance. The assembly of phage cocktails that

adsorb to different bacterial receptors decreases the risk of resistance

developing during treatment (Levin and Bull, 2004, Chan and Abedon, 2012).

Selecting phages that utilize conserved bacterial receptors that are also

virulence factors has previously been shown to attenuate the pathogenicity of

phage-resistant bacteria (Filippov et al., 2011, Laanto et al., 2012).

The study sought to isolate phage-resistant bacterial clones that had grown

resistant to one or more phages and to characterize these bacterial clones in

terms of growth characteristics, biofilm production and virulence in a

eukaryotic host. K. pneumoniae isolate 07RAFM-KPN-524 (CTX-M9 ESBL-

harboring) was isolated from a patient in Sweden in 2007 and chosen as a

routine bacterial host. Four strictly lytic phages that produced halo plaques

were isolated for this study, suggesting the presence of biofilm-degrading

depolymerases (Hughes et al., 1998).

In order to obtain phage-resistant mutants, several microcosms were set up

with combinations of the four phages to facilitate predation on 07RAFM-

KPN-524 continuously for one week. Phage-resistant bacteria were isolated,

growth rate characteristics and biofilm production determined for each of the

bacterial clones. Clones were then used to infect Galleria mellonella, a model

organism for virulence testing in eukaryote infections (McLaughlin et al.,

2014). Three out of the four treatments that contained a single phage showed

an increased bacterial growth rate (phages B, H and K; p = < 0.001), while a

single multi-phage combination (BHX) showed a decrease in maximum

growth rate compared to an untreated control (p < 0.001).

Biofilm production was lowered in the phage resistant bacteria from multi

phage treatments compared to the single phage treatments (p < 0.001).

Individually, two phages decreased biofilm production (phages K and X) in

comparison to the untreated control (p = 0.001 and p = 0.002).

While these results were surprising, and our hypothesis was that phage

predation would decrease the virulence of the bacteria in the in vivo model, it

has been previously reported that increased growth rate could be a plastic

response to phage predation in the bacteria. These results suggest that terminal

35

investment (Poisot, 2012) plays some role in the increased growth rates

observed; however, further investigation is required.

Concluding remarks and future perspectives

The aim of this thesis has been to gain a better understanding and increase

our knowledge of lytic bacteriophages that infect and kill multi-resistant

bacteria and to sequence, annotate and characterize these phages. One goal

has also been to test phage cocktails against a selection of multi-resistant

bacteria. We have isolated bacteriophages against ESBL-carrying

Enterobacteriaceae with a special focus on K. pneumoniae. The long term

goal has been to create a phage cocktail that is effective against a range of K.

pneumoniae isolates that would effectively lyse a large selection of clinical

isolates.

Two phages that were isolated during the study exhibited large sequence

similarities to each other and also to already-sequenced Klebsiella phages

KP34, F19 and NTUH-K2044-K1-1. From our analysis, we suggest the

creation of a new Podoviridae genus called “Kp34likevirus” within the

Autographivirinae subfamily (Paper I). However, in contrast to the member

phages of the proposed “Kp34likevirus” genus, the existing Phikmvlikevirus

member phages φKMV, LKA1, LIMElight and LIMEzero have demonstrated

distinct differences between themselves, and their relationships to each other

should be re-examined (Adriaenssens et al., 2011, Ceyssens et al., 2006).

Based on phylogenetic analysis of the conserved genes of the phages that were

compared in this study, the monophyletic clade of Pseudomonas phages

φKMV, PT2, PT5, MPK6, MPK7, LUZ19, LKD16 and φkF77 comprises

candidates for a study to revise the Phikmvlikevirus genus.

In paper II, we characterized the polyvalent Podoviridae phage

vB_KpnP_SU10 (SU10) with the rare C3 morphology, a phage that infects a

broad range of the E. coli reference collection (ECOR) (Ochman and Selander,

1984) but also two clinical ESBL-harboring E. coli isolates. From our study,

we suspect that the process of phage infection and assembly are not as

straightforward as first thought. SU10 has both host and phage promoters

upstream of some structural genes, for example the capsid gene, scaffold gene

and portal protein gene. The classical model of a phage infection process can

be divided into three parts: 1) transcription of early phage genes and

commandeering of the host transcriptional apparatus, 2) transcription of

middle genes and replication of the phage genome and 3) transcription of late

genes, e.g., structural genes and cell lysis. This set of events suggest that DNA

packaging is one of the last steps of the infection cycle; however, with

36

observations from our in silico analysis of transcriptional starts combined with

ultra-thin TEM micrographs, we observed the assembly of capsids early in the

infection process.

Further studies of the head elongation process and the mechanisms

underlying size determination of the elongated head morphology are needed

to determine if, for example, the scaffolding protein or a unidentified tape

measure protein is responsible for the capsid length.

In paper III, new phages were isolated and assembled into an experimental

phage cocktail, which was assessed for in vitro activity against 125 clinical

ESBL-harboring K. pneumoniae isolates. These isolates were collected

worldwide and included different types of ESBL-resistant strains, including

metallo-β-lactamase ESBLcarba resistant strains. This pilot study was to

determine the feasibility of a pre-made phage therapy cocktail against a wide

selection of unknown isolates. Phages that possessed good growth

characteristics and a wide host range were selected and assembled into a phage

cocktail. Further characterization of the phages in the cocktail is

recommended, such as their average burst size, latent period and adsorption

rate. When compared, these characteristics could improve the selection of

phages for future cocktails. By adding or replacing phages in the current

cocktail, we can extend the host range to include, for example, capsular

serotype K2. Optimization can be performed by both identifying the epitope

that the phage receptor recognizes and by selecting phages that use essential

surface proteins as receptors, thus decreasing the probability of resistance by

receptor mutation/variation in the bacteria. However, to ascertain the

receptors that phages use for adsorption, deletion mutants must be constructed

in a susceptible host or through whole genome sequencing of the wild type

bacterial host and subsequent phage resistant strains.

The phage cocktail could next be evaluated in pre-clinical in vivo trials

using mice in a lung infection model (Debarbieux et al., 2010). However, most

animal models do have drawbacks that must be considered, as the infecting

bacterial isolate must be optimized for disease progression in the animal. This

optimization consists of selecting pathogenic strains of the bacterium that

result in quick disease progression and 100 % lethality. Another drawback of

current phage therapy studies is that phage inoculation occurs immediately

prior to or after the administration of pathogenic bacteria, and thus these

results should be viewed with some caution. These protocols should be

adjusted for more realistic disease progression, i.e., with a slow infection and

no phage treatment until symptoms arise that would simulate a real-world

infection scenario.

Another aspect of phage therapy that must be addressed is the current

legislation that regulates medicinal products, which has been written with

37

traditional antibiotics and medicinal products in mind, and future phage

therapy products and their standing in these regulations must be investigated

(Verbeken et al., 2014). Furthermore, the possibility of using engineered

phages must be addressed, both legally and biologically. The template for this

engineered phage would be a fully characterized, sequenced and annotated

phage, selected for its high burst size and short latency period. By replacing

the recognition region of the tail fibers, one could switch the specificity of the

phage and thus create a fast track for pharmaceutical use, lessening the

legislative burden for phage therapy. By using one model phage for each

bacterial genus, co-evolutionary aspects of the phage bacteria could be

exploited, such as promoter and codon usage adaption. The aspect of

acceptance from society in using genetically modified bacterial viruses as

therapeutic agents must also be considered before their use as medicinal

products.

Paper IV explores the relationship between phages and bacteria and the

effects that phage predation has on the bacterium, whether it be with a single

phage or several phages simultaneously. Interestingly, we discovered that the

selection of the phage used in therapy is important, some of the resulting

phage-resistant bacteria in our study exhibited an increased growth rate and

lethality to the eukaryotic host. To further understand the mechanism

underlying the increased virulence effects in the bacterium, further

characterization of phages as well as phage-bacteria interactions should be

performed. This is in contrast to studies that have shown attenuated bacterial

virulence in phage-resistant bacteria from phages that adsorb using bacterial

receptors that also serve as virulence factors.

Phages H and X in the study (Paper IV) demonstrate two opposite effects

on the host bacterium, and their difference in gene content, receptor usage and

growth characteristics must be investigated to understand and predict which

phages are suitable for phage therapy.

38

Acknowledgments

First and foremost I would like to thank my wife Joanna and my family for

the support and encouragement over these years. I would also like to thank my

supervisor Dr. Anders S. Nilsson for taking me in as PhD student, and my co-

supervisor Professor emerita Elisabeth Haggård-Ljungquist for all her

suggestions, help and support during these years.

For making my time at Stockholm University more enjoyable, I would like

to especially thank Jaroslav Belotserkovsky, Sidney Carter and Bo Lindberg.

Good friends and great colleagues!

I would also like to mention and thank previous members of the group:

Lina Sylwan, Hanna Nilsson, Richard Odegrip, Sridhar Mandali, Marios

Hellstrand, David Berta and Andreas Bertilsson. It wouldn’t have been the

same without you! From the corridor, with shared moments and lunches

together: Asal, Alicia, Ali, Fredrik, Kun, Mattias, Melanie, Mohea, Noushin

and Sara.

For bringing new perspectives to old problems, sharing a lab bench and

projects together, Anni-Maria Örmälä-Odegrip. Your refreshing energy,

optimism and “can do” attitude really made my last year go so much faster!

Many thanks to my international collaborations Andrew Kropinski,

Zuzanna Drulis-Kawa and her group for the joint work with paper I. Thank

you for the great teamwork and collaboration! For assistance with the thesis

preparation I would like to mention Asem Abd El Monem Fatom (figure 1),

Susanna Hua (figure 3), Callum Cooper and American Journal Experts for

editing and proof-reading the thesis.

39

For making my days at Stockholm University more fun and engaging, I

would like to thank the early morning gym group at the SU staff gym: Peter

Reinhed and Wilhelm Widmark for all those mornings together, the afternoon

coffee club: Ainars Bajinskis and Sara Shakeri Manesh for all the nice talks,

shared moments and coffee together, and the technical staff for all the help I

received from Inga-Britt Olausson, Görel Lindberg, Eva Eyton and Eva

Pettersson during these years.

My grateful and cordial thanks go to Stiftelsen Olle Engquist Byggmästare,

for their funding of the project over the years and to the foundation Sven and

Lilly Lawskis fond for their stipend and travelling grants during my first two

years.

40

References

ABEDON, S. T. 2009. Phage evolution and ecology. Adv Appl Microbiol, 67, 1-45.

ABEDON, S. T. 2011. Lysis from without. Bacteriophage, 1, 46-49. ACKERMANN, H. W. 1998. Tailed bacteriophages: The order Caudovirales.

Advances in Virus Research, 51, 135-201. ACKERMANN, H. W. 2001. Frequency of morphological phage descriptions

in the year 2000. Brief review. Arch Virol, 146, 843-57. ADRIAENSSENS, E. M., CEYSSENS, P. J., DUNON, V., ACKERMANN,

H. W., VAN VAERENBERGH, J., MAES, M., DE PROFT, M. & LAVIGNE, R. 2011. Bacteriophages LIMElight and LIMEzero of Pantoea agglomerans, belonging to the "phiKMV-like viruses". Appl Environ Microbiol, 77, 3443-50.

ADRIAENSSENS, E. M., EDWARDS, R., NASH, J. H., MAHADEVAN, P., SETO, D., ACKERMANN, H. W., LAVIGNE, R. & KROPINSKI, A. M. 2014. Integration of genomic and proteomic analyses in the classification of the Siphoviridae family. Virology.

ADRIAENSSENS, E. M., EDWARDS, R., NASH, J. H., MAHADEVAN, P., SETO, D., ACKERMANN, H. W., LAVIGNE, R. & KROPINSKI, A. M. 2015. Integration of genomic and proteomic analyses in the classification of the Siphoviridae family. Virology, 477, 144-54.

ANONYMOUS 2005. Regulation 1831/2003/EC on additives for use in animal nutrition, replacing Directive 70/524/EEC on additives in feeding-stuffs. In: COMMISSION, E. (ed.). European Union.

ANONYMOUS 2013a. Annual Epidemiological Report 2013 Reporting on 2011 surveillance data and 2012 epidemic intelligence data.

http://ecdc.europa.eu/: European Centre for Disease Prevention and Control.

ANONYMOUS. 2013b. Statistics for Extended Spectrum Beta-Lactamase (ESBL) [Online]. Public Health Agency of Sweden: Folkhälsomyndigheten. Available: http://www.folkhalsomyndigheten.se/amnesomraden/statistik-och-undersokningar/sjukdomsstatistik/extended-spectrum-beta-lactamase-esbl.

ATANASIU, C., SU, T. J., STURROCK, S. S. & DRYDEN, D. T. 2002. Interaction of the Ocr gene 0.3 protein of bacteriophage T7 with

41

EcoKI restriction/modification enzyme. Nucleic Acids Res, 30, 3936-44.

BENTLEY, R. & MEGANATHAN, R. 1982. Biosynthesis of vitamin K (menaquinone) in bacteria. Microbiol Rev, 46, 241-80.

BERTANI, G. 1951. Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli. J Bacteriol, 62, 293-300.

BINGHAM, R., EKUNWE, S. I., FALK, S., SNYDER, L. & KLEANTHOUS, C. 2000. The major head protein of bacteriophage T4 binds specifically to elongation factor Tu. J Biol Chem, 275, 23219-26.

BJARNSHOLT, T. 2013. The role of bacterial biofilms in chronic infections. APMIS Suppl, 1-58.

BLOWER, T. R., EVANS, T. J., PRZYBILSKI, R., FINERAN, P. C. & SALMOND, G. P. 2012. Viral evasion of a bacterial suicide system by RNA-based molecular mimicry enables infectious altruism. PLoS Genet, 8, e1003023.

BODE, W. 1967. Lysis inhibition in Escherichia coli infected with bacteriophage T4. J Virol, 1, 948-55.

BONDY-DENOMY, J., PAWLUK, A., MAXWELL, K. L. & DAVIDSON, A. R. 2013. Bacteriophage genes that inactivate the CRISPR/Cas bacterial immune system. Nature, 493, 429-32.

BRABBAN, A. D., HITE, E. & CALLAWAY, T. R. 2005. Evolution of foodborne pathogens via temperate bacteriophage-mediated gene transfer. Foodborne Pathog Dis, 2, 287-303.

BRADLEY, D. E. 1967. Ultrastructure of Bacteriophages and Bacteriocins. Bacteriological Reviews, 31, 230-&.

BUCKLING, A. & BROCKHURST, M. 2012. Bacteria-virus coevolution. Adv Exp Med Biol, 751, 347-70.

BUTLER, M. S., BLASKOVICH, M. A. & COOPER, M. A. 2013. Antibiotics in the clinical pipeline in 2013. J Antibiot (Tokyo), 66, 571-91.

CARDONE, G., HEYMANN, B., CHENG, N., TRUS, B. L. & STEVEN, A. C. 2013. Viral Molecular Machines. In: ROSSMAN, G. M. & RAO, B. V. (eds.) Viral Molecular Machines. SpringerLink.

CARS, O. & HÄLLSTRÖM, N. 2004. ReAct - Action on Antibiotic Resistance [Online]. http://www.reactgroup.org/. Available: http://www.reactgroup.org/uploads/what-we-do/react-activities/pressrelease-a-new-moment-to-tackle.pdf.

CENENS, W., MAKUMI, A., MEBRHATU, M. T., LAVIGNE, R. & AERTSEN, A. 2013a. Phage-host interactions during pseudolysogeny: Lessons from the Pid/dgo interaction. Bacteriophage, 3, e25029.

CENENS, W., MEBRHATU, M. T., MAKUMI, A., CEYSSENS, P. J., LAVIGNE, R., VAN HOUDT, R., TADDEI, F. & AERTSEN, A.

42

2013b. Expression of a novel P22 ORFan gene reveals the phage carrier state in Salmonella typhimurium. PLoS Genet, 9, e1003269.

CEYSSENS, P. J., LAVIGNE, R., MATTHEUS, W., CHIBEU, A., HERTVELDT, K., MAST, J., ROBBEN, J. & VOLCKAERT, G. 2006. Genomic analysis of Pseudomonas aeruginosa phages LKD16 and LKA1: establishment of the phiKMV subgroup within the T7 supergroup. J Bacteriol, 188, 6924-31.

CHAN, B. K. & ABEDON, S. T. 2012. Phage therapy pharmacology phage cocktails. Adv Appl Microbiol, 78, 1-23.

CHOPIN, M. C., CHOPIN, A. & BIDNENKO, E. 2005. Phage abortive infection in lactococci: variations on a theme. Curr Opin Microbiol, 8, 473-9.

CHOPIN, M. C., ROUAULT, A., EHRLICH, S. D. & GAUTIER, M. 2002. Filamentous phage active on the Gram-positive bacterium Propionibacterium freudenreichii. J Bacteriol, 184, 2030-3.

COOPER, M. A. & SHLAES, D. 2011. Fix the antibiotics pipeline. Nature, 472, 32.

CORNELISSEN, A., CEYSSENS, P. J., T'SYEN, J., VAN PRAET, H., NOBEN, J. P., SHABUROVA, O. V., KRYLOV, V. N., VOLCKAERT, G. & LAVIGNE, R. 2011. The T7-related Pseudomonas putida phage phi15 displays virion-associated biofilm degradation properties. PLoS One, 6, e18597.

D´HERELLE, F. 1917. Sur un microbe invisible antagoniste des bacilles dysentériques. Compt. rend. Acad. sci, 165, 373.

DAVIES, J. 1994. Inactivation of antibiotics and the dissemination of resistance genes. Science, 264, 375-82.

DAVIS, B. D., DULBECCO, R., EISEN, H. N., GINSBERG, H. S. & WOOD, B. W. 1973. The enteric baccilli. Microbiology. New York, Evanston & London: Harper & Row.

DEBARBIEUX, L., LEDUC, D., MAURA, D., MORELLO, E., CRISCUOLO, A., GROSSI, O., BALLOY, V. & TOUQUI, L. 2010. Bacteriophages can treat and prevent Pseudomonas aeruginosa lung infections. J Infect Dis, 201, 1096-104.

DESTOUMIEUX-GARZON, D., DUQUESNE, S., PEDUZZI, J., GOULARD, C., DESMADRIL, M., LETELLIER, L., REBUFFAT, S. & BOULANGER, P. 2005. The iron-siderophore transporter FhuA is the receptor for the antimicrobial peptide microcin J25: role of the microcin Val11-Pro16 beta-hairpin region in the recognition mechanism. Biochem J, 389, 869-76.

DIAZ-MUNOZ, S. L. & KOSKELLA, B. 2014. Bacteria-phage interactions in natural environments. Adv Appl Microbiol, 89, 135-83.

DOERN, G. V. 1995. Trends in antimicrobial susceptibility of bacterial pathogens of the respiratory tract. Am J Med, 99, 3S-7S.

DRULIS-KAWA, Z., MACKIEWICZ, P., KESIK-SZELOCH, A., MACIASZCZYK-DZIUBINSKA, E., WEBER-DABROWSKA, B.,

43

DOROTKIEWICZ-JACH, A., AUGUSTYNIAK, D., MAJKOWSKA-SKROBEK, G., BOCER, T., EMPEL, J. & KROPINSKI, A. M. 2011. Isolation and characterisation of KP34--a novel phiKMV-like bacteriophage for Klebsiella pneumoniae. Appl Microbiol Biotechnol, 90, 1333-45.

DUFFY, S., SHACKELTON, L. A. & HOLMES, E. C. 2008. Rates of evolutionary change in viruses: patterns and determinants. Nat Rev Genet, 9, 267-76.

EMOND, E., DION, E., WALKER, S. A., VEDAMUTHU, E. R., KONDO, J. K. & MOINEAU, S. 1998. AbiQ, an abortive infection mechanism from Lactococcus lactis. Appl Environ Microbiol, 64, 4748-56.

FILIPPOV, A. A., SERGUEEV, K. V., HE, Y., HUANG, X. Z., GNADE, B. T., MUELLER, A. J., FERNANDEZ-PRADA, C. M. & NIKOLICH, M. P. 2011. Bacteriophage-resistant mutants in Yersinia pestis: identification of phage receptors and attenuation for mice. PLoS One, 6, e25486.

FINERAN, P. C., BLOWER, T. R., FOULDS, I. J., HUMPHREYS, D. P., LILLEY, K. S. & SALMOND, G. P. 2009. The phage abortive infection system, ToxIN, functions as a protein-RNA toxin-antitoxin pair. Proc Natl Acad Sci U S A, 106, 894-9.

FINERAN, P. C. & CHARPENTIER, E. 2012. Memory of viral infections by CRISPR-Cas adaptive immune systems: acquisition of new information. Virology, 434, 202-9.

FLAYHAN, A., WIEN, F., PATERNOSTRE, M., BOULANGER, P. & BREYTON, C. 2012. New insights into pb5, the receptor binding protein of bacteriophage T5, and its interaction with its Escherichia coli receptor FhuA. Biochimie, 94, 1982-9.

GROTH, A. C. & CALOS, M. P. 2004. Phage integrases: Biology and Applications. Journal of Molecular Biology, 335, 667-678.

HALL-STOODLEY, L., COSTERTON, J. W. & STOODLEY, P. 2004. Bacterial biofilms: from the natural environment to infectious diseases. Nat Rev Microbiol, 2, 95-108.

HALLET, B. & SHERRATT, D. J. 1997. Transposition and site-specific recombination: adapting DNA cut-and-paste mechanisms to a variety of genetic rearrangements. FEMS Microbiol Rev, 21, 157-78.

HANEY, E. F. & HANCOCK, R. B. 2013. Peptide design for antimicrobial and immunomodulatory applications. Biopolymers.

HECHT, O., MACDONALD, C. & MOORE, G. R. 2012. Intrinsically disordered proteins: lessons from colicins. Biochem Soc Trans, 40, 1534-8.

HENDRIX, R. W., SMITH, M. C. M., BURNS, R. N., FORD, M. E. & HATFULL, G. F. 1999. Evolutionary relationships among diverse bacteriophages and prophages: All the world's a phage. Proceedings of the National Academy of Sciences of the United States of America, 96, 2192-2197.

44

HUGHES, K. A., SUTHERLAND, I. W., CLARK, J. & JONES, M. V. 1998. Bacteriophage and associated polysaccharide depolymerases--novel tools for study of bacterial biofilms. J Appl Microbiol, 85, 583-90.

HYMAN, P. & ABEDON, S. T. 2010. Bacteriophage host range and bacterial resistance. Adv Appl Microbiol, 70, 217-48.

JACOBY, G. A. 1997. Extended-spectrum beta-lactamases and other enzymes providing resistance to oxyimino-beta-lactams. Infect Dis Clin North Am, 11, 875-87.

JANN, K. & JANN, B. 1987. Polysaccharide antigens of Escherichia coli. Rev Infect Dis, 9 Suppl 5, S517-26.

JAVED, M. A., ACKERMANN, H. W., AZEREDO, J., CARVALHO, C. M., CONNERTON, I., EVOY, S., HAMMERL, J. A., HERTWIG, S., LAVIGNE, R., SINGH, A., SZYMANSKI, C. M., TIMMS, A. & KROPINSKI, A. M. 2014. A suggested classification for two groups of Campylobacter myoviruses. Arch Virol, 159, 181-90.

KANEKO, J., KIMURA, T., NARITA, S., TOMITA, T. & KAMIO, Y. 1998. Complete nucleotide sequence and molecular characterization of the temperate staphylococcal bacteriophage phiPVL carrying Panton-Valentine leukocidin genes. Gene, 215, 57-67.

KARAOLIS, D. K., SOMARA, S., MANEVAL, D. R., JR., JOHNSON, J. A. & KAPER, J. B. 1999. A bacteriophage encoding a pathogenicity island, a type-IV pilus and a phage receptor in cholera bacteria. Nature, 399, 375-9.

KOSTAKIOTI, M., HADJIFRANGISKOU, M. & HULTGREN, S. J. 2013. Bacterial biofilms: development, dispersal, and therapeutic strategies in the dawn of the postantibiotic era. Cold Spring Harb Perspect Med, 3, a010306.

KOWALCZYKOWSKI, S. C. 2000. Initiation of genetic recombination and recombination-dependent replication. Trends Biochem Sci, 25, 156-65.

KOWALCZYKOWSKI, S. C., DIXON, D. A., EGGLESTON, A. K., LAUDER, S. D. & REHRAUER, W. M. 1994. Biochemistry of homologous recombination in Escherichia coli. Microbiol Rev, 58, 401-65.

KREBS, E. J., GOLDSTEIN, E. S. & KILPATRICK, T. S. 2012. Lewin´s Genes XI. In: KREBS, E. J., GOLDSTEIN, E. S. & KILPATRICK, T. S. (eds.) Lewin´s Genes XI. 11 ed.: Jones & Bartlett learning.

KROPINSKI, A. M., LINGOHR, E. J. & ACKERMANN, H. W. 2011. The genome sequence of enterobacterial phage 7-11, which possesses an unusually elongated head. Arch Virol, 156, 149-51.

LAANTO, E., BAMFORD, J. K., LAAKSO, J. & SUNDBERG, L. R. 2012. Phage-driven loss of virulence in a fish pathogenic bacterium. PLoS One, 7, e53157.

LABRIE, S. J., SAMSON, J. E. & MOINEAU, S. 2010. Bacteriophage resistance mechanisms. Nat Rev Microbiol, 8, 317-27.

45

LAI, Y., ROSENSHINE, I., LEONG, J. M. & FRANKEL, G. 2013. Intimate host attachment: enteropathogenic and enterohaemorrhagic Escherichia coli. Cell Microbiol, 15, 1796-808.

LAVIGNE, R., DARIUS, P., SUMMER, E. J., SETO, D., MAHADEVAN, P., NILSSON, A. S., ACKERMANN, H. W. & KROPINSKI, A. M. 2009. Classification of Myoviridae bacteriophages using protein sequence similarity. BMC Microbiol, 9, 224.

LAVIGNE, R., SETO, D., MAHADEVAN, P., ACKERMANN, H. W. & KROPINSKI, A. M. 2008. Unifying classical and molecular taxonomic classification: analysis of the Podoviridae using BLASTP-based tools. Res Microbiol, 159, 406-14.

LEIMAN, P. G., BATTISTI, A. J., BOWMAN, V. D., STUMMEYER, K., MUHLENHOFF, M., GERARDY-SCHAHN, R., SCHOLL, D. & MOLINEUX, I. J. 2007. The structures of bacteriophages K1E and K1-5 explain processive degradation of polysaccharide capsules and evolution of new host specificities. J Mol Biol, 371, 836-49.

LEVIN, B. R. & BULL, J. J. 2004. Population and evolutionary dynamics of phage therapy. Nat Rev Microbiol, 2, 166-73.

LI, Y., CHEN, M., TANG, F., YAO, H., LU, C. & ZHANG, W. 2012. Complete genome sequence of the novel lytic avian pathogenic coliphage NJ01. J Virol, 86, 13874-5.

LIN, T. L., HSIEH, P. F., HUANG, Y. T., LEE, W. C., TSAI, Y. T., SU, P. A., PAN, Y. J., HSU, C. R., WU, M. C. & WANG, J. T. 2014. Isolation of a bacteriophage and its depolymerase specific for K1 capsule of Klebsiella pneumoniae: implication in typing and treatment. J Infect Dis, 210, 1734-44.

LITTLE, J. W. & MICHALOWSKI, C. B. 2010. Stability and instability in the lysogenic state of phage lambda. J Bacteriol, 192, 6064-76.

LIU, J., DEHBI, M., MOECK, G., ARHIN, F., BAUDA, P., BERGERON, D., CALLEJO, M., FERRETTI, V., HA, N., KWAN, T., MCCARTY, J., SRIKUMAR, R., WILLIAMS, D., WU, J. J., GROS, P., PELLETIER, J. & DUBOW, M. 2004. Antimicrobial drug discovery through bacteriophage genomics. Nat Biotechnol, 22, 185-91.

LOENEN, W. A. & MURRAY, N. E. 1986. Modification enhancement by the restriction alleviation protein (Ral) of bacteriophage lambda. J Mol Biol, 190, 11-22.

LOS, M. & WEGRZYN, G. 2012. Pseudolysogeny. Adv Virus Res, 82, 339-49.

LUDWIG, A., VON RHEIN, C., MISCHKE, A. & BRADE, V. 2008. Release of latent ClyA cytolysin from Escherichia coli mediated by a bacteriophage-associated putative holin (BlyA) from Borrelia burgdorferi. Int J Med Microbiol, 298, 473-81.

LUKACIK, P., BARNARD, T. J. & BUCHANAN, S. K. 2012. Using a bacteriocin structure to engineer a phage lysin that targets Yersinia pestis. Biochem Soc Trans, 40, 1503-6.

46

MATSUSHITA, K., UCHIYAMA, J., KATO, S., UJIHARA, T., HOSHIBA, H., SUGIHARA, S., MURAOKA, A., WAKIGUCHI, H. & MATSUZAKI, S. 2009. Morphological and genetic analysis of three bacteriophages of Serratia marcescens isolated from environmental water. FEMS Microbiol Lett, 291, 201-8.

MCLAUGHLIN, M. M., ADVINCULA, M. R., MALCZYNSKI, M., BARAJAS, G., QI, C. & SCHEETZ, M. H. 2014. Quantifying the clinical virulence of Klebsiella pneumoniae producing carbapenemase Klebsiella pneumoniae with a Galleria mellonella model and a pilot study to translate to patient outcomes. BMC Infect Dis, 14, 31.

MCLEOD, S. M., KIMSEY, H. H., DAVIS, B. M. & WALDOR, M. K. 2005. CTXphi and Vibrio cholerae: exploring a newly recognized type of phage-host cell relationship. Mol Microbiol, 57, 347-56.

MIEDZYBRODZKI, R., BORYSOWSKI, J., WEBER-DABROWSKA, B., FORTUNA, W., LETKIEWICZ, S., SZUFNAROWSKI, K., PAWELCZYK, Z., ROGOZ, P., KLAK, M., WOJTASIK, E. & GORSKI, A. 2012. Clinical aspects of phage therapy. Adv Virus Res, 83, 73-121.

MIROLD, S., RABSCH, W., ROHDE, M., STENDER, S., TSCHAPE, H., RUSSMANN, H., IGWE, E. & HARDT, W. D. 1999. Isolation of a temperate bacteriophage encoding the type III effector protein SopE from an epidemic Salmonella typhimurium strain. Proc Natl Acad Sci U S A, 96, 9845-50.

OCHMAN, H., LAWRENCE, J. G. & GROISMAN, E. A. 2000. Lateral gene transfer and the nature of bacterial innovation. Nature, 405, 299-304.

OCHMAN, H. & SELANDER, R. K. 1984. Standard reference strains of Escherichia coli from natural populations. Journal of Bacteriology, 157, 690-693.

PAVLOVA, O., LAVYSH, D., KLIMUK, E., DJORDJEVIC, M., RAVCHEEV, D. A., GELFAND, M. S., SEVERINOV, K. & AKULENKO, N. 2012. Temporal regulation of gene expression of the Escherichia coli bacteriophage phiEco32. J Mol Biol, 416, 389-99.

PIRNAY, J. P., DE VOS, D., VERBEKEN, G., MERABISHVILI, M., CHANISHVILI, N., VANEECHOUTTE, M., ZIZI, M., LAIRE, G., LAVIGNE, R., HUYS, I., VAN DEN MOOTER, G., BUCKLING, A., DEBARBIEUX, L., POUILLOT, F., AZEREDO, J., KUTTER, E., DUBLANCHET, A., GORSKI, A. & ADAMIA, R. The Phage Therapy Paradigm: Pret-a-Porter or Sur-mesure? Pharm Res, 2010, 10.

PODSCHUN, R. & ULLMANN, U. 1998. Klebsiella spp. as nosocomial pathogens: epidemiology, taxonomy, typing methods, and pathogenicity factors. Clin Microbiol Rev, 11, 589-603.

47

POISOT, T., BELL,T., MARTINEZ, E., GOUGAT-BARBERA, C., HOCHBERG, M.E. 2012. Terminal investment induced by a bacteriophage in a Rhizosphere bacterium. F1000Research, 1.

QIMRON, U., MARINTCHEVA, B., TABOR, S. & RICHARDSON, C. C. 2006. Genomewide screens for Escherichia coli genes affecting growth of T7 bacteriophage. Proc Natl Acad Sci U S A, 103, 19039-44.

RAVAT, F. & CHATARD, D. 2013. Evaluation of Phage Therapy for the Treatment of Escherichia coli and Pseudomonas aeruginosa Wound Infections in Burned Patients (PHAGOBURN) [Online]. http://www.phagoburn.eu/

Available: http://www.phagoburn.eu/. RESCH, G., MOREILLON, P. & FISCHETTI, V. A. 2011. A stable phage

lysin (Cpl-1) dimer with increased antipneumococcal activity and decreased plasma clearance. Int J Antimicrob Agents, 38, 516-21.

SAMSON, J. E., MAGADAN, A. H., SABRI, M. & MOINEAU, S. 2013. Revenge of the phages: defeating bacterial defences. Nat Rev Microbiol, 11, 675-87.

SARKER, S. A., MCCALLIN, S., BARRETTO, C., BERGER, B., PITTET, A. C., SULTANA, S., KRAUSE, L., HUQ, S., BIBILONI, R., BRUTTIN, A., REUTELER, G. & BRUSSOW, H. 2012. Oral T4-like phage cocktail application to healthy adult volunteers from Bangladesh. Virology, 434, 222-32.

SCHMELCHER, M., DONOVAN, D. M. & LOESSNER, M. J. 2012. Bacteriophage endolysins as novel antimicrobials. Future Microbiology, 7, 1147-1171.

SCHULTZ, E. W. 1927. The Bacteriophage: Its Prophylactic and Therapeutic Value: A Review. Cal West Med, 27, 481-7.

SIMOONS-SMIT, A. M., VERWEY-VAN VUGHT, A. M., KANIS, I. Y. & MACLAREN, D. M. 1984. Virulence of Klebsiella strains in experimentally induced skin lesions in the mouse. J Med Microbiol, 17, 67-77.

SULAKVELIDZE, A., ALAVIDZE, Z. & MORRIS, J. G., JR. 2001. Bacteriophage therapy. Antimicrob Agents Chemother, 45, 649-59.

SWARTS, D. C., MOSTERD, C., VAN PASSEL, M. W. & BROUNS, S. J. 2012. CRISPR interference directs strand specific spacer acquisition. PLoS One, 7, e35888.

THOMASSEN, E., GIELEN, G., SCHUTZ, M., SCHOEHN, G., ABRAHAMS, J. P., MILLER, S. & VAN RAAIJ, M. J. 2003. The structure of the receptor-binding domain of the bacteriophage T4 short tail fibre reveals a knitted trimeric metal-binding fold. J Mol Biol, 331, 361-73.

TRAN, T. A., STRUCK, D. K. & YOUNG, R. 2005. Periplasmic domains define holin-antiholin interactions in T4 lysis inhibition. J Bacteriol, 187, 6631-40.

48

WAGNER, P. L. & WALDOR, M. K. 2002. Bacteriophage control of bacterial virulence. Infect Immun, 70, 3985-93.

WALMAGH, M., BOCZKOWSKA, B., GRYMONPREZ, B., BRIERS, Y., DRULIS-KAWA, Z. & LAVIGNE, R. 2012. Characterization of five novel endolysins from Gram-negative infecting bacteriophages. Appl Microbiol Biotechnol.

VEESLER, D. & CAMBILLAU, C. 2011. A common evolutionary origin for tailed-bacteriophage functional modules and bacterial machineries. Microbiology and Molecular Biology Reviews : MMBR, 75, 423-33, first page of table of contents.

VERBEKEN, G., PIRNAY, J. P., DE VOS, D., JENNES, S., ZIZI, M., LAVIGNE, R., CASTEELS, M. & HUYS, I. 2012. Optimizing the European regulatory framework for sustainable bacteriophage therapy in human medicine. Arch Immunol Ther Exp (Warsz), 60, 161-72.

VERBEKEN, G., PIRNAY, J. P., LAVIGNE, R., JENNES, S., DE VOS, D., CASTEELS, M. & HUYS, I. 2014. Call for a dedicated European legal framework for bacteriophage therapy. Arch Immunol Ther Exp (Warsz), 62, 117-29.

WESTRA, E. R., SWARTS, D. C., STAALS, R. H. J., JORE, M. M., BROUNS, S. J. J. & VAN DER OOST, J. 2012a. The CRISPRs, They Are A-Changin': How Prokaryotes Generate Adaptive Immunity. Annual Review of Genetics, Vol 46, 46, 311-339.

WESTRA, E. R., VAN ERP, P. B., KUNNE, T., WONG, S. P., STAALS, R. H., SEEGERS, C. L., BOLLEN, S., JORE, M. M., SEMENOVA, E., SEVERINOV, K., DE VOS, W. M., DAME, R. T., DE VRIES, R., BROUNS, S. J. & VAN DER OOST, J. 2012b. CRISPR immunity relies on the consecutive binding and degradation of negatively supercoiled invader DNA by Cascade and Cas3. Mol Cell, 46, 595-605.

WIEDENHEFT, B., STERNBERG, S. H. & DOUDNA, J. A. 2012. RNA-guided genetic silencing systems in bacteria and archaea. Nature, 482, 331-8.

WILSON, G. G. & MURRAY, N. E. 1991. Restriction and modification systems. Annu Rev Genet, 25, 585-627.

WOESE, C. R. & FOX, G. E. 1977. Phylogenetic structure of the prokaryotic domain: the primary kingdoms. Proc Natl Acad Sci U S A, 74, 5088-90.

WRIGHT, A., HAWKINS, C. H., ANGGARD, E. E. & HARPER, D. R. 2009. A controlled clinical trial of a therapeutic bacteriophage preparation in chronic otitis due to antibiotic-resistant Pseudomonas aeruginosa; a preliminary report of efficacy. Clin Otolaryngol, 34, 349-57.

YOSEF, I., GOREN, M. G. & QIMRON, U. 2012. Proteins and DNA elements essential for the CRISPR adaptation process in Escherichia coli. Nucleic Acids Res, 40, 5569-76.